The associations among depression, cardiovascular disease, and cognitive impairment are well known. Inflammation is increasingly recognized as playing an important role as well. However, the nature of their relationships and which may actually cause the other is not well understood. This article reviews studies over the past year that link depression with dementia and vascular disease. Desirable directions for future work are also explored.

DEPRESSION AND ALZHEIMER DISEASE

Several studies have shown significant correlations between depression and the risk of developing Alzheimer disease; the frequency of depressive episodes appears to be an important factor. Despite the risk relationship, however, depression and Alzheimer disease may not share a common pathology.

No shared pathology

Wilson and colleagues¹ analyzed data from the Chicago Health and Aging Project, a longitudinal cohort study of risk factors for Alzheimer disease that involved two groups of people aged 65 years and older; one group was composed of people who developed dementia during the study, while the other group had already developed dementia or had some degree of cognitive impairment. The investigators reasoned that if pathologic changes are common to depression and dementia, then there would be evidence of change in depressive symptoms along with the progression of dementia. They found only a barely perceptible increase in depressive symptoms among people who developed Alzheimer disease and concluded that there is no shared pathology between depression and Alzheimer disease.

Degree of depression signals risk

Depression has been associated with nearly double the risk of developing dementia and Alzheimer disease. Saczynski et al² evaluated 949 people in the Framingham study, mean age 79 years, using the 60-point Center for Epidemiologic Studies Depression Scale (depression defined as > 16 points). Individuals who had depression at baseline were 1.7 times more likely to develop dementia over the 17-year evaluation period. Results were similar when adjusted for major vascular risk factors (smoking, diabetes, hypertension, and cardiovascular disease). The correlation was slightly lower but still significant when subjects taking antidepressant medications were included in the depressed group.

The study also found a continuous relationship between the level of depression and the likelihood of developing dementia and Alzheimer disease: for every 10-point increase on the depression scale, the risk of developing dementia increased by nearly 50%. This study supports depression as a risk factor for dementia. One could also argue that depression as a simple prodrome to dementia seems unlikely because of the long followup between baseline assessment and the development of dementia.

Multiple episodes of depression increase risk

Dotson et al³ analyzed data from 1,239 older adults from the Baltimore Longitudinal Study of Aging who did not have depression, dementia, or mild cognitive impairment at baseline. Every 1 to 2 years for about 25 years, cognitive status and mood of the subjects were evaluated. About 10% of the participants developed dementia during the course of the study. Of those who developed dementia, 35% had at least one episode of depression; among those who did not develop dementia, only 23% had a depressive episode. Findings were similar when investigators controlled for vascular risks and vascular dementia.

One episode of depression was associated with an 87% increase in risk of dementia; at least two episodes of depression more than doubled the risk (108%). Overall, each episode of depression conferred an additional 14% risk of developing dementia. Among subjects who had had two or more episodes of depression, the median age of developing dementia was 85 years versus 95 years for those without an episode of depression.

This study had the advantages of being prospective for both depression and dementia and of having a long followup period. A dose-effect relationship was observed, with the “dose” being the number of depressive episodes (rather than severity of depression). Because the definition of a depressive episode included subsyndromal depression (not likely to meet the criteria of clinical depression, but still clinically significant), the findings suggest that even minor depression increases the risk of dementia.

Rosenberg et al⁴ found depression to be associated with cognitive impairment in their evaluation of 436 women in their 70s; the women, who were participants in the Women’s Health and Aging Study, were evaluated for up to 9 years. To be included in the evaluation, subjects needed a Mini-Mental State Examination score of at least 24 points (out of 30 possible) and could not be impaired in more than one basic functional capacity: mobility and exercise tolerance, upper extremity, higher functioning (eg, shopping), and basic self-care). Baseline depressive symptoms were measured using the Geriatric Depression Scale.

Cognitive testing included Hopkins Verbal Learning Tests (for immediate and delayed word recall) and Trail Making Tests (for psychomotor speed and executive functioning). Those who were not impaired (ie, having a cognitive test score below the 10th percentile for age-adjusted norms) were followed with up to six examinations over the next 9 years.

Baseline depression was found to be highly associated with incident impairment in all cognitive areas, especially in executive functioning. For every point increase in the depression scale, a 6% to 7% increase was found in the annual risk of impairment in each cognitive domain.

DEPRESSION AND VASCULAR DISEASE LINKED

It is somewhat easier to assess the relationship between depression and vascular disease than between depression and cognitive impairment because of the availability of objective measures of cardiovascular function.

The International Stroke Study (INTERSTROKE),⁵ a case-control study in 22 countries with 3,000 cases of stroke and 3,000 age-, gender-, and ethnicity-matched controls, found nine risk factors that were correlated with 90% of ischemic stroke cases. Depression, with an odds ratio of 1.86, was found to be a more significant risk factor than physical activity, diet, or heavy drinking.

Paranthaman et al⁶ evaluated a number of measures of arterial anatomy and function in 25 subjects with depressive disorder and in 21 nondepressed controls (mean age, 72 years). They found that depressed subjects had significantly higher mean carotid intima media thickness, reduced dilation in response to acetyl choline in preconstricted small arteries, and more dilated Virchow-Robin spaces in the basal ganglia observed on magnetic resonance imaging. This study provides evidence that arterial structure and function may mediate the relationship between depression and vascular disease.

FUTURE DIRECTIONS

Future studies into depression as a risk factor should use very well-characterized cohorts that are controlled for blood pressure and other vascular risk factors. Finding biomarkers for depression would be useful, permitting its detection earlier and with more certainty than is now possible. Prospective studies to evaluate the relationship of depression to cognitive impairment and dementia are needed that start earlier than in middle or old age. The key question that needs study is whether treatment of depression can actually change the onset of cognitive impairment, Alzheimer disease, and vascular disease.

The associations among depression, cardiovascular disease, and cognitive impairment are well known. Inflammation is increasingly recognized as playing an important role as well. However, the nature of their relationships and which may actually cause the other is not well understood. This article reviews studies over the past year that link depression with dementia and vascular disease. Desirable directions for future work are also explored.

DEPRESSION AND ALZHEIMER DISEASE

Several studies have shown significant correlations between depression and the risk of developing Alzheimer disease; the frequency of depressive episodes appears to be an important factor. Despite the risk relationship, however, depression and Alzheimer disease may not share a common pathology.

No shared pathology

Wilson and colleagues¹ analyzed data from the Chicago Health and Aging Project, a longitudinal cohort study of risk factors for Alzheimer disease that involved two groups of people aged 65 years and older; one group was composed of people who developed dementia during the study, while the other group had already developed dementia or had some degree of cognitive impairment. The investigators reasoned that if pathologic changes are common to depression and dementia, then there would be evidence of change in depressive symptoms along with the progression of dementia. They found only a barely perceptible increase in depressive symptoms among people who developed Alzheimer disease and concluded that there is no shared pathology between depression and Alzheimer disease.

Degree of depression signals risk

Depression has been associated with nearly double the risk of developing dementia and Alzheimer disease. Saczynski et al² evaluated 949 people in the Framingham study, mean age 79 years, using the 60-point Center for Epidemiologic Studies Depression Scale (depression defined as > 16 points). Individuals who had depression at baseline were 1.7 times more likely to develop dementia over the 17-year evaluation period. Results were similar when adjusted for major vascular risk factors (smoking, diabetes, hypertension, and cardiovascular disease). The correlation was slightly lower but still significant when subjects taking antidepressant medications were included in the depressed group.

The study also found a continuous relationship between the level of depression and the likelihood of developing dementia and Alzheimer disease: for every 10-point increase on the depression scale, the risk of developing dementia increased by nearly 50%. This study supports depression as a risk factor for dementia. One could also argue that depression as a simple prodrome to dementia seems unlikely because of the long followup between baseline assessment and the development of dementia.

Multiple episodes of depression increase risk

Dotson et al³ analyzed data from 1,239 older adults from the Baltimore Longitudinal Study of Aging who did not have depression, dementia, or mild cognitive impairment at baseline. Every 1 to 2 years for about 25 years, cognitive status and mood of the subjects were evaluated. About 10% of the participants developed dementia during the course of the study. Of those who developed dementia, 35% had at least one episode of depression; among those who did not develop dementia, only 23% had a depressive episode. Findings were similar when investigators controlled for vascular risks and vascular dementia.

One episode of depression was associated with an 87% increase in risk of dementia; at least two episodes of depression more than doubled the risk (108%). Overall, each episode of depression conferred an additional 14% risk of developing dementia. Among subjects who had had two or more episodes of depression, the median age of developing dementia was 85 years versus 95 years for those without an episode of depression.

This study had the advantages of being prospective for both depression and dementia and of having a long followup period. A dose-effect relationship was observed, with the “dose” being the number of depressive episodes (rather than severity of depression). Because the definition of a depressive episode included subsyndromal depression (not likely to meet the criteria of clinical depression, but still clinically significant), the findings suggest that even minor depression increases the risk of dementia.

Rosenberg et al⁴ found depression to be associated with cognitive impairment in their evaluation of 436 women in their 70s; the women, who were participants in the Women’s Health and Aging Study, were evaluated for up to 9 years. To be included in the evaluation, subjects needed a Mini-Mental State Examination score of at least 24 points (out of 30 possible) and could not be impaired in more than one basic functional capacity: mobility and exercise tolerance, upper extremity, higher functioning (eg, shopping), and basic self-care). Baseline depressive symptoms were measured using the Geriatric Depression Scale.

Cognitive testing included Hopkins Verbal Learning Tests (for immediate and delayed word recall) and Trail Making Tests (for psychomotor speed and executive functioning). Those who were not impaired (ie, having a cognitive test score below the 10th percentile for age-adjusted norms) were followed with up to six examinations over the next 9 years.

Baseline depression was found to be highly associated with incident impairment in all cognitive areas, especially in executive functioning. For every point increase in the depression scale, a 6% to 7% increase was found in the annual risk of impairment in each cognitive domain.

DEPRESSION AND VASCULAR DISEASE LINKED

It is somewhat easier to assess the relationship between depression and vascular disease than between depression and cognitive impairment because of the availability of objective measures of cardiovascular function.

The International Stroke Study (INTERSTROKE),⁵ a case-control study in 22 countries with 3,000 cases of stroke and 3,000 age-, gender-, and ethnicity-matched controls, found nine risk factors that were correlated with 90% of ischemic stroke cases. Depression, with an odds ratio of 1.86, was found to be a more significant risk factor than physical activity, diet, or heavy drinking.

Paranthaman et al⁶ evaluated a number of measures of arterial anatomy and function in 25 subjects with depressive disorder and in 21 nondepressed controls (mean age, 72 years). They found that depressed subjects had significantly higher mean carotid intima media thickness, reduced dilation in response to acetyl choline in preconstricted small arteries, and more dilated Virchow-Robin spaces in the basal ganglia observed on magnetic resonance imaging. This study provides evidence that arterial structure and function may mediate the relationship between depression and vascular disease.

FUTURE DIRECTIONS

Future studies into depression as a risk factor should use very well-characterized cohorts that are controlled for blood pressure and other vascular risk factors. Finding biomarkers for depression would be useful, permitting its detection earlier and with more certainty than is now possible. Prospective studies to evaluate the relationship of depression to cognitive impairment and dementia are needed that start earlier than in middle or old age. The key question that needs study is whether treatment of depression can actually change the onset of cognitive impairment, Alzheimer disease, and vascular disease.

The associations among depression, cardiovascular disease, and cognitive impairment are well known. Inflammation is increasingly recognized as playing an important role as well. However, the nature of their relationships and which may actually cause the other is not well understood. This article reviews studies over the past year that link depression with dementia and vascular disease. Desirable directions for future work are also explored.

DEPRESSION AND ALZHEIMER DISEASE

Several studies have shown significant correlations between depression and the risk of developing Alzheimer disease; the frequency of depressive episodes appears to be an important factor. Despite the risk relationship, however, depression and Alzheimer disease may not share a common pathology.

No shared pathology

Wilson and colleagues¹ analyzed data from the Chicago Health and Aging Project, a longitudinal cohort study of risk factors for Alzheimer disease that involved two groups of people aged 65 years and older; one group was composed of people who developed dementia during the study, while the other group had already developed dementia or had some degree of cognitive impairment. The investigators reasoned that if pathologic changes are common to depression and dementia, then there would be evidence of change in depressive symptoms along with the progression of dementia. They found only a barely perceptible increase in depressive symptoms among people who developed Alzheimer disease and concluded that there is no shared pathology between depression and Alzheimer disease.

Degree of depression signals risk

Depression has been associated with nearly double the risk of developing dementia and Alzheimer disease. Saczynski et al² evaluated 949 people in the Framingham study, mean age 79 years, using the 60-point Center for Epidemiologic Studies Depression Scale (depression defined as > 16 points). Individuals who had depression at baseline were 1.7 times more likely to develop dementia over the 17-year evaluation period. Results were similar when adjusted for major vascular risk factors (smoking, diabetes, hypertension, and cardiovascular disease). The correlation was slightly lower but still significant when subjects taking antidepressant medications were included in the depressed group.

The study also found a continuous relationship between the level of depression and the likelihood of developing dementia and Alzheimer disease: for every 10-point increase on the depression scale, the risk of developing dementia increased by nearly 50%. This study supports depression as a risk factor for dementia. One could also argue that depression as a simple prodrome to dementia seems unlikely because of the long followup between baseline assessment and the development of dementia.

Multiple episodes of depression increase risk

Dotson et al³ analyzed data from 1,239 older adults from the Baltimore Longitudinal Study of Aging who did not have depression, dementia, or mild cognitive impairment at baseline. Every 1 to 2 years for about 25 years, cognitive status and mood of the subjects were evaluated. About 10% of the participants developed dementia during the course of the study. Of those who developed dementia, 35% had at least one episode of depression; among those who did not develop dementia, only 23% had a depressive episode. Findings were similar when investigators controlled for vascular risks and vascular dementia.

One episode of depression was associated with an 87% increase in risk of dementia; at least two episodes of depression more than doubled the risk (108%). Overall, each episode of depression conferred an additional 14% risk of developing dementia. Among subjects who had had two or more episodes of depression, the median age of developing dementia was 85 years versus 95 years for those without an episode of depression.

This study had the advantages of being prospective for both depression and dementia and of having a long followup period. A dose-effect relationship was observed, with the “dose” being the number of depressive episodes (rather than severity of depression). Because the definition of a depressive episode included subsyndromal depression (not likely to meet the criteria of clinical depression, but still clinically significant), the findings suggest that even minor depression increases the risk of dementia.

Rosenberg et al⁴ found depression to be associated with cognitive impairment in their evaluation of 436 women in their 70s; the women, who were participants in the Women’s Health and Aging Study, were evaluated for up to 9 years. To be included in the evaluation, subjects needed a Mini-Mental State Examination score of at least 24 points (out of 30 possible) and could not be impaired in more than one basic functional capacity: mobility and exercise tolerance, upper extremity, higher functioning (eg, shopping), and basic self-care). Baseline depressive symptoms were measured using the Geriatric Depression Scale.

Cognitive testing included Hopkins Verbal Learning Tests (for immediate and delayed word recall) and Trail Making Tests (for psychomotor speed and executive functioning). Those who were not impaired (ie, having a cognitive test score below the 10th percentile for age-adjusted norms) were followed with up to six examinations over the next 9 years.

Baseline depression was found to be highly associated with incident impairment in all cognitive areas, especially in executive functioning. For every point increase in the depression scale, a 6% to 7% increase was found in the annual risk of impairment in each cognitive domain.

DEPRESSION AND VASCULAR DISEASE LINKED

It is somewhat easier to assess the relationship between depression and vascular disease than between depression and cognitive impairment because of the availability of objective measures of cardiovascular function.

The International Stroke Study (INTERSTROKE),⁵ a case-control study in 22 countries with 3,000 cases of stroke and 3,000 age-, gender-, and ethnicity-matched controls, found nine risk factors that were correlated with 90% of ischemic stroke cases. Depression, with an odds ratio of 1.86, was found to be a more significant risk factor than physical activity, diet, or heavy drinking.

Paranthaman et al⁶ evaluated a number of measures of arterial anatomy and function in 25 subjects with depressive disorder and in 21 nondepressed controls (mean age, 72 years). They found that depressed subjects had significantly higher mean carotid intima media thickness, reduced dilation in response to acetyl choline in preconstricted small arteries, and more dilated Virchow-Robin spaces in the basal ganglia observed on magnetic resonance imaging. This study provides evidence that arterial structure and function may mediate the relationship between depression and vascular disease.

FUTURE DIRECTIONS

Future studies into depression as a risk factor should use very well-characterized cohorts that are controlled for blood pressure and other vascular risk factors. Finding biomarkers for depression would be useful, permitting its detection earlier and with more certainty than is now possible. Prospective studies to evaluate the relationship of depression to cognitive impairment and dementia are needed that start earlier than in middle or old age. The key question that needs study is whether treatment of depression can actually change the onset of cognitive impairment, Alzheimer disease, and vascular disease.

The relationships among inflammation, Alzheimer disease, and depression have been the subject of recent research at several centers. Alzheimer disease and depression appear to be linked by several genetic and inflammatory processes, although the exact nature of the relationship is not clearly understood. The two disorders also share risk factors for vascular disease. This article reviews the current state of knowledge about inflammation and its implications for Alzheimer disease and depression, and it presents recent findings from the Texas Alzheimer’s Research Consortium, which assessed an array of inflammatory markers in a cohort of patients with Alzheimer disease.

INFLAMMATION MAY MEDIATE DEPRESSION, COGNITIVE DECLINE, AND DEMENTIA

Alzheimer disease and depression share several vascular disease risk factors and appear to be linked through complex and integrated processes. The link may be mediated by long-term inflammatory processes. Hypothalamic-pituitary-adrenal (HPA) axis dysfunction, chronic inflammation, and a deficit in neurotrophin signaling all may play roles in the pathogenesis of depression and Alzheimer disease.¹ Excessive release of glucocorticoids subsequent to HPA-axis dysfunction in chronic depression appears to damage the hippocampus: hippocampal atrophy is a feature in both depression and dementia, and recurrent depression is associated with greater atrophy. The direction of influence—whether depression leads to the factors that increase the risk of Alzheimer disease or the other way around—remains a controversial topic.

Symptoms of depression tend to appear early in Alzheimer disease and increase as dementia progresses to moderate severity. In advanced dementia, depression symptoms tend to decline, although this may reflect the difficulty in assessing depression at advanced stages of dementia.²

Numerous reports have linked inflammation to cognitive dysfunction or decline, as well as to the development of Alzheimer disease.^3–5 Evidence suggests that inflammation is a key mediator between cardiovascular risk factors and Alzheimer disease, although this is also still controversial.

FINDINGS FROM THE TEXAS ALZHEIMER’S RESEARCH CONSORTIUM

The Texas Alzheimer’s Research Consortium, composed of five medical centers, is pursuing a longitudinal, multi-institutional study of Alzheimer disease. The group recently published the results of a study assessing whether inflammatory markers were over- or underexpressed in patients with Alzheimer disease, and whether biomarkers could predict Alzheimer disease status and the age at onset of the disease.⁴ The analysis included 197 patients with Alzheimer disease and 203 control subjects. The evaluation consisted of cognitive assessment, DNA analysis for human genome-wide association studies, and protein microarray analysis from blood. Cardiovascular risk factors were also measured, including serum lipids and blood factors for diabetes risk. The goal was to better understand the pathophysiology of cognitive decline and predict conversion of mild cognitive impairment to Alzheimer disease.

Significant differences were found in the study groups. For example, the median age in the Alzheimer group was significantly higher than in controls (79 vs 70 years, P < .0001), an issue that is being addressed as subjects are replaced due to attrition. The median educational level was higher in the control group (14 vs 16 years, P < .0001) than in the Alzheimer group. Subjects in the Alzheimer group were significantly more likely (P < .001) to carry at least one copy of the APOE ε4 allele.

Degree of inflammation also correlated with Mini-Mental State Examination (MMSE) scores. Subjects with a high inflammatory score had a more accelerated decline in MMSE scores over a 3-year period than those with a low inflammatory score. The association was significant, although not as dramatic as the association between inflammation and age at onset of Alzheimer disease.

The investigators concluded that their findings, while considered preliminary, suggest the existence of an inflammatory endophenotype associated with Alzheimer disease. The findings need to be validated in other populations, including ethnic groups other than Caucasian. The Consortium also will evaluate whether inflammatory biomarkers are associated with progression of mild cognitive impairment to Alzheimer disease.

Inflammation has a mixed association with depression

In a study whose results are not yet published, the Texas Consortium also examined the association between inflammatory markers and depression. Four subscales of depression were used, derived from the Geriatric Depression Scale (GDS) 30: dysphoria (consisting of 11 items), meaninglessness (seven items), apathy and withdrawal (six items), and cognitive impairment (six items).⁵

The GDS30 results as a whole suggested a trend toward an association between depression and inflammatory biomarkers, but the association was not significant. When the results were examined by subscale, however, striking differences were found between Alzheimer patients and the control group. For example, apathy was significantly associated with the C-reactive protein level, and the assocation was much stronger in patients with Alzheimer disease than in controls. Further, the association of apathy with C-reactive protein level was more significant in women than in men.

Other associations were found between several of the inflammatory and antiinflammatory cytokines and the various subscales; the relationship between inflammatory factors and depression appears to be complex and often gender-specific.

Inflammation-depression link is suggestive, not linear

Despite the relationships suggested by the data, no simple linear relationship was identified to indicate that more inflammation leads to more depression in Alzheimer disease. The relationship between inflammation and depression in Alzheimer disease appears to involve a complex interplay between many physiologic processes.

The effect of inflammation also varies with gender and with cognitive impairment. The mechanism that underlies these relationships remains to be determined and will be the focus of further studies with the Texas Alzheimer’s Research Consortium.

The relationships among inflammation, Alzheimer disease, and depression have been the subject of recent research at several centers. Alzheimer disease and depression appear to be linked by several genetic and inflammatory processes, although the exact nature of the relationship is not clearly understood. The two disorders also share risk factors for vascular disease. This article reviews the current state of knowledge about inflammation and its implications for Alzheimer disease and depression, and it presents recent findings from the Texas Alzheimer’s Research Consortium, which assessed an array of inflammatory markers in a cohort of patients with Alzheimer disease.

INFLAMMATION MAY MEDIATE DEPRESSION, COGNITIVE DECLINE, AND DEMENTIA

Alzheimer disease and depression share several vascular disease risk factors and appear to be linked through complex and integrated processes. The link may be mediated by long-term inflammatory processes. Hypothalamic-pituitary-adrenal (HPA) axis dysfunction, chronic inflammation, and a deficit in neurotrophin signaling all may play roles in the pathogenesis of depression and Alzheimer disease.¹ Excessive release of glucocorticoids subsequent to HPA-axis dysfunction in chronic depression appears to damage the hippocampus: hippocampal atrophy is a feature in both depression and dementia, and recurrent depression is associated with greater atrophy. The direction of influence—whether depression leads to the factors that increase the risk of Alzheimer disease or the other way around—remains a controversial topic.

Symptoms of depression tend to appear early in Alzheimer disease and increase as dementia progresses to moderate severity. In advanced dementia, depression symptoms tend to decline, although this may reflect the difficulty in assessing depression at advanced stages of dementia.²

Numerous reports have linked inflammation to cognitive dysfunction or decline, as well as to the development of Alzheimer disease.^3–5 Evidence suggests that inflammation is a key mediator between cardiovascular risk factors and Alzheimer disease, although this is also still controversial.

FINDINGS FROM THE TEXAS ALZHEIMER’S RESEARCH CONSORTIUM

The Texas Alzheimer’s Research Consortium, composed of five medical centers, is pursuing a longitudinal, multi-institutional study of Alzheimer disease. The group recently published the results of a study assessing whether inflammatory markers were over- or underexpressed in patients with Alzheimer disease, and whether biomarkers could predict Alzheimer disease status and the age at onset of the disease.⁴ The analysis included 197 patients with Alzheimer disease and 203 control subjects. The evaluation consisted of cognitive assessment, DNA analysis for human genome-wide association studies, and protein microarray analysis from blood. Cardiovascular risk factors were also measured, including serum lipids and blood factors for diabetes risk. The goal was to better understand the pathophysiology of cognitive decline and predict conversion of mild cognitive impairment to Alzheimer disease.

Significant differences were found in the study groups. For example, the median age in the Alzheimer group was significantly higher than in controls (79 vs 70 years, P < .0001), an issue that is being addressed as subjects are replaced due to attrition. The median educational level was higher in the control group (14 vs 16 years, P < .0001) than in the Alzheimer group. Subjects in the Alzheimer group were significantly more likely (P < .001) to carry at least one copy of the APOE ε4 allele.

Degree of inflammation also correlated with Mini-Mental State Examination (MMSE) scores. Subjects with a high inflammatory score had a more accelerated decline in MMSE scores over a 3-year period than those with a low inflammatory score. The association was significant, although not as dramatic as the association between inflammation and age at onset of Alzheimer disease.

The investigators concluded that their findings, while considered preliminary, suggest the existence of an inflammatory endophenotype associated with Alzheimer disease. The findings need to be validated in other populations, including ethnic groups other than Caucasian. The Consortium also will evaluate whether inflammatory biomarkers are associated with progression of mild cognitive impairment to Alzheimer disease.

Inflammation has a mixed association with depression

In a study whose results are not yet published, the Texas Consortium also examined the association between inflammatory markers and depression. Four subscales of depression were used, derived from the Geriatric Depression Scale (GDS) 30: dysphoria (consisting of 11 items), meaninglessness (seven items), apathy and withdrawal (six items), and cognitive impairment (six items).⁵

The GDS30 results as a whole suggested a trend toward an association between depression and inflammatory biomarkers, but the association was not significant. When the results were examined by subscale, however, striking differences were found between Alzheimer patients and the control group. For example, apathy was significantly associated with the C-reactive protein level, and the assocation was much stronger in patients with Alzheimer disease than in controls. Further, the association of apathy with C-reactive protein level was more significant in women than in men.

Other associations were found between several of the inflammatory and antiinflammatory cytokines and the various subscales; the relationship between inflammatory factors and depression appears to be complex and often gender-specific.

Inflammation-depression link is suggestive, not linear

Despite the relationships suggested by the data, no simple linear relationship was identified to indicate that more inflammation leads to more depression in Alzheimer disease. The relationship between inflammation and depression in Alzheimer disease appears to involve a complex interplay between many physiologic processes.

The effect of inflammation also varies with gender and with cognitive impairment. The mechanism that underlies these relationships remains to be determined and will be the focus of further studies with the Texas Alzheimer’s Research Consortium.

The relationships among inflammation, Alzheimer disease, and depression have been the subject of recent research at several centers. Alzheimer disease and depression appear to be linked by several genetic and inflammatory processes, although the exact nature of the relationship is not clearly understood. The two disorders also share risk factors for vascular disease. This article reviews the current state of knowledge about inflammation and its implications for Alzheimer disease and depression, and it presents recent findings from the Texas Alzheimer’s Research Consortium, which assessed an array of inflammatory markers in a cohort of patients with Alzheimer disease.

INFLAMMATION MAY MEDIATE DEPRESSION, COGNITIVE DECLINE, AND DEMENTIA

Alzheimer disease and depression share several vascular disease risk factors and appear to be linked through complex and integrated processes. The link may be mediated by long-term inflammatory processes. Hypothalamic-pituitary-adrenal (HPA) axis dysfunction, chronic inflammation, and a deficit in neurotrophin signaling all may play roles in the pathogenesis of depression and Alzheimer disease.¹ Excessive release of glucocorticoids subsequent to HPA-axis dysfunction in chronic depression appears to damage the hippocampus: hippocampal atrophy is a feature in both depression and dementia, and recurrent depression is associated with greater atrophy. The direction of influence—whether depression leads to the factors that increase the risk of Alzheimer disease or the other way around—remains a controversial topic.

Symptoms of depression tend to appear early in Alzheimer disease and increase as dementia progresses to moderate severity. In advanced dementia, depression symptoms tend to decline, although this may reflect the difficulty in assessing depression at advanced stages of dementia.²

Numerous reports have linked inflammation to cognitive dysfunction or decline, as well as to the development of Alzheimer disease.^3–5 Evidence suggests that inflammation is a key mediator between cardiovascular risk factors and Alzheimer disease, although this is also still controversial.

FINDINGS FROM THE TEXAS ALZHEIMER’S RESEARCH CONSORTIUM

The Texas Alzheimer’s Research Consortium, composed of five medical centers, is pursuing a longitudinal, multi-institutional study of Alzheimer disease. The group recently published the results of a study assessing whether inflammatory markers were over- or underexpressed in patients with Alzheimer disease, and whether biomarkers could predict Alzheimer disease status and the age at onset of the disease.⁴ The analysis included 197 patients with Alzheimer disease and 203 control subjects. The evaluation consisted of cognitive assessment, DNA analysis for human genome-wide association studies, and protein microarray analysis from blood. Cardiovascular risk factors were also measured, including serum lipids and blood factors for diabetes risk. The goal was to better understand the pathophysiology of cognitive decline and predict conversion of mild cognitive impairment to Alzheimer disease.

Significant differences were found in the study groups. For example, the median age in the Alzheimer group was significantly higher than in controls (79 vs 70 years, P < .0001), an issue that is being addressed as subjects are replaced due to attrition. The median educational level was higher in the control group (14 vs 16 years, P < .0001) than in the Alzheimer group. Subjects in the Alzheimer group were significantly more likely (P < .001) to carry at least one copy of the APOE ε4 allele.

Degree of inflammation also correlated with Mini-Mental State Examination (MMSE) scores. Subjects with a high inflammatory score had a more accelerated decline in MMSE scores over a 3-year period than those with a low inflammatory score. The association was significant, although not as dramatic as the association between inflammation and age at onset of Alzheimer disease.

The investigators concluded that their findings, while considered preliminary, suggest the existence of an inflammatory endophenotype associated with Alzheimer disease. The findings need to be validated in other populations, including ethnic groups other than Caucasian. The Consortium also will evaluate whether inflammatory biomarkers are associated with progression of mild cognitive impairment to Alzheimer disease.

Inflammation has a mixed association with depression

In a study whose results are not yet published, the Texas Consortium also examined the association between inflammatory markers and depression. Four subscales of depression were used, derived from the Geriatric Depression Scale (GDS) 30: dysphoria (consisting of 11 items), meaninglessness (seven items), apathy and withdrawal (six items), and cognitive impairment (six items).⁵

The GDS30 results as a whole suggested a trend toward an association between depression and inflammatory biomarkers, but the association was not significant. When the results were examined by subscale, however, striking differences were found between Alzheimer patients and the control group. For example, apathy was significantly associated with the C-reactive protein level, and the assocation was much stronger in patients with Alzheimer disease than in controls. Further, the association of apathy with C-reactive protein level was more significant in women than in men.

Other associations were found between several of the inflammatory and antiinflammatory cytokines and the various subscales; the relationship between inflammatory factors and depression appears to be complex and often gender-specific.

Inflammation-depression link is suggestive, not linear

Despite the relationships suggested by the data, no simple linear relationship was identified to indicate that more inflammation leads to more depression in Alzheimer disease. The relationship between inflammation and depression in Alzheimer disease appears to involve a complex interplay between many physiologic processes.

The effect of inflammation also varies with gender and with cognitive impairment. The mechanism that underlies these relationships remains to be determined and will be the focus of further studies with the Texas Alzheimer’s Research Consortium.

Alzheimer disease (AD) is a progressive, irreversible, neurodegenerative disease that affects more than 5.3 million people in the United States.¹ This number is significantly higher than the previous estimate of 4.5 million and is projected to increase sharply to nearly 8 million by 2030.¹ At present, the few agents that are approved by the US Food and Drug Administration for treatment of AD have demonstrated only modest effects in modifying clinical symptoms for relatively short periods; none has shown a clear effect on disease progression. New therapeutic approaches are desperately needed.

VASCULAR DISEASE AND ALZHEIMER DISEASE

Although AD is classified as a neurodegenerative dementia, there is epidemiologic and pathologic evidence of an association with vascular risk factors and vascular disease.^2–6 Vascular disease appears to lower the threshold for the clinical presentation of dementia at a given level of AD-related pathology.⁷ The possible association of AD with vascular disease suggests that there are important pathogenic mechanisms common to both AD and vascular disease. For example, there is increasing evidence that perturbations in cerebral vascular structure and function occur in AD.⁸

It has been suggested that cerebral hypoperfusion/hypoxia triggers hypometabolic, cognitive, and degenerative changes in the brain and contributes to the pathologic processes of AD.⁹ A study by Roher and colleagues reveals an association between severe circle of Willis atherosclerosis and sporadic AD.¹⁰ These observations suggest that atherosclerosis-induced brain hypoperfusion contributes to the clinical and pathologic manifestations of AD.

Hypoxia is also known to stimulate angiogenesis, especially via upregulation of hypoxia-inducible genes such as vascular endothelial growth factor (VEGF).^11,12 VEGF, a critical mediator of angiogenesis, is present in the AD brain in the walls of intra-parenchymal vessels, in diffuse perivascular deposits, and in clusters of reactive astrocytes.¹³ In addition, intrathecal levels of VEGF in AD are related to clinical severity and intrathecal levels of amyloid-beta (Aβ).¹⁴ Emerging data support the idea that factors and processes characteristic of angiogenesis are found in the AD brain.^15,16

ENDOTHELIAL ACTIVATION AND ANGIOGENESIS

The angiogenic process is complex and involves several discrete steps, such as endothelial activation, extracellular matrix degradation, proliferation and migration of endothelial cells, and morphologic differentiation of endothelial cells to form tubes. Stimuli known to initiate angiogenesis, including hypoxia, inflammation, and mechanical factors such as shear stress and stretch,²³ either directly or indirectly activate endothelial cells. Activated endothelial cells elaborate adhesion molecules, cytokines and chemokines, growth factors, vasoactive molecules, major histocompatibility complex molecules, procoagulant and anticoagulant moieties, and a variety of other gene products with biologic activity.²⁴ The activated endothelium exerts direct local effects by producing at least 20 paracrine factors that act on adjacent cells.²⁵

ANGIOGENIC SIGNALING MECHANISMS IN BRAIN MICROVESSELS

Signaling mechanisms that have been identified as important to endothelial cell viability and angiogenesis include PI3K/Akt, p38 kinase, ERK, and JNK. In this regard, intracellular Aβ accumulation is toxic to endothelial cells and decreases PI3K/Akt.²⁶ Extracellular Aβ peptides decrease phosphorylation and thus activation of ERK and p38 kinase.²⁶ VEGF promotes endothelial survival, proliferation, and migration through numerous pathways, including activation of ERK, p38 kinase, JNK, and Rho GTPase family members.²³

VASCULAR ACTIVATION IN ALZHEIMER DISEASE

Despite increases in several proangiogenic factors in the AD brain, evidence for increased vascularity in AD is lacking. On the contrary, it has been suggested that the angiogenic process is delayed or impaired in aged tissues, with several studies showing decreased microvascular density in the AD brain.^30–33 Paris et al showed that wild-type Aβ peptides have antiangiogenic effects in vitro and in vivo.³⁴

How can the data showing antiangiogenic effects of Aβ be reconciled with the presence or expression of a large number of proangiogenic proteins by brain microvessels in AD? These conflicting observations suggest an imbalance between proangiogenic and antiangiogeneic processes in the AD brain.

Preliminary experiments in our laboratory show that pharmacologic blockade of vascular activation improves cognitive function in an animal model of AD. Thus, “vascular activation” could be a novel, unexplored therapeutic target in AD.

Acknowledgment

The authors gratefully acknowledge the secretarial assistance of Terri Stahl.

Alzheimer disease (AD) is a progressive, irreversible, neurodegenerative disease that affects more than 5.3 million people in the United States.¹ This number is significantly higher than the previous estimate of 4.5 million and is projected to increase sharply to nearly 8 million by 2030.¹ At present, the few agents that are approved by the US Food and Drug Administration for treatment of AD have demonstrated only modest effects in modifying clinical symptoms for relatively short periods; none has shown a clear effect on disease progression. New therapeutic approaches are desperately needed.

VASCULAR DISEASE AND ALZHEIMER DISEASE

Although AD is classified as a neurodegenerative dementia, there is epidemiologic and pathologic evidence of an association with vascular risk factors and vascular disease.^2–6 Vascular disease appears to lower the threshold for the clinical presentation of dementia at a given level of AD-related pathology.⁷ The possible association of AD with vascular disease suggests that there are important pathogenic mechanisms common to both AD and vascular disease. For example, there is increasing evidence that perturbations in cerebral vascular structure and function occur in AD.⁸

It has been suggested that cerebral hypoperfusion/hypoxia triggers hypometabolic, cognitive, and degenerative changes in the brain and contributes to the pathologic processes of AD.⁹ A study by Roher and colleagues reveals an association between severe circle of Willis atherosclerosis and sporadic AD.¹⁰ These observations suggest that atherosclerosis-induced brain hypoperfusion contributes to the clinical and pathologic manifestations of AD.

Hypoxia is also known to stimulate angiogenesis, especially via upregulation of hypoxia-inducible genes such as vascular endothelial growth factor (VEGF).^11,12 VEGF, a critical mediator of angiogenesis, is present in the AD brain in the walls of intra-parenchymal vessels, in diffuse perivascular deposits, and in clusters of reactive astrocytes.¹³ In addition, intrathecal levels of VEGF in AD are related to clinical severity and intrathecal levels of amyloid-beta (Aβ).¹⁴ Emerging data support the idea that factors and processes characteristic of angiogenesis are found in the AD brain.^15,16

ENDOTHELIAL ACTIVATION AND ANGIOGENESIS

The angiogenic process is complex and involves several discrete steps, such as endothelial activation, extracellular matrix degradation, proliferation and migration of endothelial cells, and morphologic differentiation of endothelial cells to form tubes. Stimuli known to initiate angiogenesis, including hypoxia, inflammation, and mechanical factors such as shear stress and stretch,²³ either directly or indirectly activate endothelial cells. Activated endothelial cells elaborate adhesion molecules, cytokines and chemokines, growth factors, vasoactive molecules, major histocompatibility complex molecules, procoagulant and anticoagulant moieties, and a variety of other gene products with biologic activity.²⁴ The activated endothelium exerts direct local effects by producing at least 20 paracrine factors that act on adjacent cells.²⁵

ANGIOGENIC SIGNALING MECHANISMS IN BRAIN MICROVESSELS

Signaling mechanisms that have been identified as important to endothelial cell viability and angiogenesis include PI3K/Akt, p38 kinase, ERK, and JNK. In this regard, intracellular Aβ accumulation is toxic to endothelial cells and decreases PI3K/Akt.²⁶ Extracellular Aβ peptides decrease phosphorylation and thus activation of ERK and p38 kinase.²⁶ VEGF promotes endothelial survival, proliferation, and migration through numerous pathways, including activation of ERK, p38 kinase, JNK, and Rho GTPase family members.²³

VASCULAR ACTIVATION IN ALZHEIMER DISEASE

Despite increases in several proangiogenic factors in the AD brain, evidence for increased vascularity in AD is lacking. On the contrary, it has been suggested that the angiogenic process is delayed or impaired in aged tissues, with several studies showing decreased microvascular density in the AD brain.^30–33 Paris et al showed that wild-type Aβ peptides have antiangiogenic effects in vitro and in vivo.³⁴

How can the data showing antiangiogenic effects of Aβ be reconciled with the presence or expression of a large number of proangiogenic proteins by brain microvessels in AD? These conflicting observations suggest an imbalance between proangiogenic and antiangiogeneic processes in the AD brain.

Preliminary experiments in our laboratory show that pharmacologic blockade of vascular activation improves cognitive function in an animal model of AD. Thus, “vascular activation” could be a novel, unexplored therapeutic target in AD.

Acknowledgment

The authors gratefully acknowledge the secretarial assistance of Terri Stahl.

Alzheimer disease (AD) is a progressive, irreversible, neurodegenerative disease that affects more than 5.3 million people in the United States.¹ This number is significantly higher than the previous estimate of 4.5 million and is projected to increase sharply to nearly 8 million by 2030.¹ At present, the few agents that are approved by the US Food and Drug Administration for treatment of AD have demonstrated only modest effects in modifying clinical symptoms for relatively short periods; none has shown a clear effect on disease progression. New therapeutic approaches are desperately needed.

VASCULAR DISEASE AND ALZHEIMER DISEASE

Although AD is classified as a neurodegenerative dementia, there is epidemiologic and pathologic evidence of an association with vascular risk factors and vascular disease.^2–6 Vascular disease appears to lower the threshold for the clinical presentation of dementia at a given level of AD-related pathology.⁷ The possible association of AD with vascular disease suggests that there are important pathogenic mechanisms common to both AD and vascular disease. For example, there is increasing evidence that perturbations in cerebral vascular structure and function occur in AD.⁸

It has been suggested that cerebral hypoperfusion/hypoxia triggers hypometabolic, cognitive, and degenerative changes in the brain and contributes to the pathologic processes of AD.⁹ A study by Roher and colleagues reveals an association between severe circle of Willis atherosclerosis and sporadic AD.¹⁰ These observations suggest that atherosclerosis-induced brain hypoperfusion contributes to the clinical and pathologic manifestations of AD.

Hypoxia is also known to stimulate angiogenesis, especially via upregulation of hypoxia-inducible genes such as vascular endothelial growth factor (VEGF).^11,12 VEGF, a critical mediator of angiogenesis, is present in the AD brain in the walls of intra-parenchymal vessels, in diffuse perivascular deposits, and in clusters of reactive astrocytes.¹³ In addition, intrathecal levels of VEGF in AD are related to clinical severity and intrathecal levels of amyloid-beta (Aβ).¹⁴ Emerging data support the idea that factors and processes characteristic of angiogenesis are found in the AD brain.^15,16

ENDOTHELIAL ACTIVATION AND ANGIOGENESIS

The angiogenic process is complex and involves several discrete steps, such as endothelial activation, extracellular matrix degradation, proliferation and migration of endothelial cells, and morphologic differentiation of endothelial cells to form tubes. Stimuli known to initiate angiogenesis, including hypoxia, inflammation, and mechanical factors such as shear stress and stretch,²³ either directly or indirectly activate endothelial cells. Activated endothelial cells elaborate adhesion molecules, cytokines and chemokines, growth factors, vasoactive molecules, major histocompatibility complex molecules, procoagulant and anticoagulant moieties, and a variety of other gene products with biologic activity.²⁴ The activated endothelium exerts direct local effects by producing at least 20 paracrine factors that act on adjacent cells.²⁵

ANGIOGENIC SIGNALING MECHANISMS IN BRAIN MICROVESSELS

Signaling mechanisms that have been identified as important to endothelial cell viability and angiogenesis include PI3K/Akt, p38 kinase, ERK, and JNK. In this regard, intracellular Aβ accumulation is toxic to endothelial cells and decreases PI3K/Akt.²⁶ Extracellular Aβ peptides decrease phosphorylation and thus activation of ERK and p38 kinase.²⁶ VEGF promotes endothelial survival, proliferation, and migration through numerous pathways, including activation of ERK, p38 kinase, JNK, and Rho GTPase family members.²³

VASCULAR ACTIVATION IN ALZHEIMER DISEASE

Despite increases in several proangiogenic factors in the AD brain, evidence for increased vascularity in AD is lacking. On the contrary, it has been suggested that the angiogenic process is delayed or impaired in aged tissues, with several studies showing decreased microvascular density in the AD brain.^30–33 Paris et al showed that wild-type Aβ peptides have antiangiogenic effects in vitro and in vivo.³⁴

How can the data showing antiangiogenic effects of Aβ be reconciled with the presence or expression of a large number of proangiogenic proteins by brain microvessels in AD? These conflicting observations suggest an imbalance between proangiogenic and antiangiogeneic processes in the AD brain.

Preliminary experiments in our laboratory show that pharmacologic blockade of vascular activation improves cognitive function in an animal model of AD. Thus, “vascular activation” could be a novel, unexplored therapeutic target in AD.

Acknowledgment

The authors gratefully acknowledge the secretarial assistance of Terri Stahl.

The number of people in the United States who spend a significant part of each week working as unpaid caregivers is considerable, and the toll exacted for such work is high. Understanding the profi le of the caregiver, the nature of the duties performed, the stress imposed by such duties, and the consequences of the stress can assist the clinician in recognizing the caregiver in need of intervention.

A PROFILE OF THE CAREGIVER

A recent survey estimated that more than 65 million Americans provide unpaid assistance annually to older adults with disabilities.¹ The value of that labor has been estimated at $306 billion annually, or nearly double the combined cost of home health care and nursing home care.^2,3

The typical caregiver is a woman, about 48 years old, with some college education, who spends 20 hours or more each week providing unpaid care to someone aged 50 years or older.¹ The recipients of care often have long-term physical disabilities; mental confusion or emotional problems frequently complicate care.

PSYCHOLOGIC AND PHYSICAL COSTS

Caregiving may take a toll on the caregiver in a variety of ways: behavioral, in the form of alcohol or substance use⁴; psychologic, in the form of depression or other mental health problems⁵; and physical, in the form of chronic health conditions and impaired immune response.⁶ About three-fifths of caregivers report fair or poor health, compared with one-third of noncaregivers, and caregivers have approximately twice as many chronic conditions, such as heart disease, cancer, arthritis, and diabetes, compared with noncaregivers.^2,7 Caregiving also exacts a financial toll, as employees who are caregivers cost their employers $13.4 billion more per year in health care expenditures.⁸ In addition, absenteeism, workday interruptions, and shifts from full-time to part-time work by caregivers cost businesses between $17.1 and $33.6 billion per year.⁹

The cost of caregiving is higher for women, who exhibit higher levels of anxiety and depression and lower levels of subjective well-being, life satisfaction, and physical health.^10,11 The stress of caregiving has also been identified as a risk factor for morbidity among older (66 to 96 years old) caregivers, who have a 63% greater mortality than noncaregivers of the same age.¹²

PSYCHOSOCIAL STRESS, UNHEALTHY BEHAVIORS, AND ILLNESS ARE LINKED

Psychosocial stress is a predictor of disease and can lead to unhealthy behaviors such as smoking, substance abuse, overeating, poor nutrition, and a sedentary lifestyle; these, in turn, can lead to physical and psychiatric illness. Behaviors adopted initially as coping skills may persist to become chronic, thereby promoting either continued wellness (in the case of healthy coping behaviors) or worsening levels of illness (in the case of unhealthy coping behaviors).

McEwen and Gianaros¹³ have suggested that these stress mechanisms arise from patterns of communication between the brain and the autonomic, cardiovascular, and immune systems, which mutually influenceone another. These so-called bidirectional stress processes affect cognition, experience, and behavior.

An integrated model of stress that maps the bidirectional causal pathways among psychosocial stressors, resulting unhealthy behaviors, and illness is needed. Although the steps from unhealthy behaviors to illness are fairly well understood, the links from psychosocial stress, such as those exhibited by caregivers, to unhealthy behaviors are not as clear. Several mediators are under study:

• Personality mediators can be either ameliorative (resilience, self-confi dence, self-control, optimism, high self-esteem, a sense of mastery, and finding meaning in life) or exacerbating (neuroticism and inhibition, which together form the so-called type D personality).
• Environmental mediators include social support, financial support, a history of a significant life change, and trauma early in life, which may increase one’s subsequent vulnerability to unhealthy behaviors.
• Biologic mediators may include prolonged sympathetic activation and enhanced platelet activation, caused by increased levels of depression and anxiety in chronically stressed caregivers.¹⁴

IMPLICATIONS FOR INTERVENTION

A significant percentage of caregivers do not need a clinician’s intervention to help them cope with stress or unhealthy coping skills. Among caregivers aged 50 years or older, 47% indicated in a recent study that the burden of caregiving is low (ie, 1 or 2 on a 5-point scale).¹ Those who respond to stressors as challenges rather than threats tend to be resilient people who exert control over their lives, often through meditation or similar techniques, and have a strong social support network. Many report that caregiving provides them with an opportunity to act in accordance with their values and feel helpful rather than helpless.

Cognitive-behavioral interventions to alleviate stress-related symptoms appear to be more effective if offered as individual rather than group therapy. Teaching caregivers effective coping strategies, rather than merely providing social support, has been shown to improve caregiver psychologic health.¹⁵ Chief among the goals of intervention should be to alter brain function and instill optimism, a sense of control and self-esteem.¹³

The number of people in the United States who spend a significant part of each week working as unpaid caregivers is considerable, and the toll exacted for such work is high. Understanding the profi le of the caregiver, the nature of the duties performed, the stress imposed by such duties, and the consequences of the stress can assist the clinician in recognizing the caregiver in need of intervention.

A PROFILE OF THE CAREGIVER

A recent survey estimated that more than 65 million Americans provide unpaid assistance annually to older adults with disabilities.¹ The value of that labor has been estimated at $306 billion annually, or nearly double the combined cost of home health care and nursing home care.^2,3

The typical caregiver is a woman, about 48 years old, with some college education, who spends 20 hours or more each week providing unpaid care to someone aged 50 years or older.¹ The recipients of care often have long-term physical disabilities; mental confusion or emotional problems frequently complicate care.

PSYCHOLOGIC AND PHYSICAL COSTS

Caregiving may take a toll on the caregiver in a variety of ways: behavioral, in the form of alcohol or substance use⁴; psychologic, in the form of depression or other mental health problems⁵; and physical, in the form of chronic health conditions and impaired immune response.⁶ About three-fifths of caregivers report fair or poor health, compared with one-third of noncaregivers, and caregivers have approximately twice as many chronic conditions, such as heart disease, cancer, arthritis, and diabetes, compared with noncaregivers.^2,7 Caregiving also exacts a financial toll, as employees who are caregivers cost their employers $13.4 billion more per year in health care expenditures.⁸ In addition, absenteeism, workday interruptions, and shifts from full-time to part-time work by caregivers cost businesses between $17.1 and $33.6 billion per year.⁹

The cost of caregiving is higher for women, who exhibit higher levels of anxiety and depression and lower levels of subjective well-being, life satisfaction, and physical health.^10,11 The stress of caregiving has also been identified as a risk factor for morbidity among older (66 to 96 years old) caregivers, who have a 63% greater mortality than noncaregivers of the same age.¹²

PSYCHOSOCIAL STRESS, UNHEALTHY BEHAVIORS, AND ILLNESS ARE LINKED

Psychosocial stress is a predictor of disease and can lead to unhealthy behaviors such as smoking, substance abuse, overeating, poor nutrition, and a sedentary lifestyle; these, in turn, can lead to physical and psychiatric illness. Behaviors adopted initially as coping skills may persist to become chronic, thereby promoting either continued wellness (in the case of healthy coping behaviors) or worsening levels of illness (in the case of unhealthy coping behaviors).

McEwen and Gianaros¹³ have suggested that these stress mechanisms arise from patterns of communication between the brain and the autonomic, cardiovascular, and immune systems, which mutually influenceone another. These so-called bidirectional stress processes affect cognition, experience, and behavior.

An integrated model of stress that maps the bidirectional causal pathways among psychosocial stressors, resulting unhealthy behaviors, and illness is needed. Although the steps from unhealthy behaviors to illness are fairly well understood, the links from psychosocial stress, such as those exhibited by caregivers, to unhealthy behaviors are not as clear. Several mediators are under study:

• Personality mediators can be either ameliorative (resilience, self-confi dence, self-control, optimism, high self-esteem, a sense of mastery, and finding meaning in life) or exacerbating (neuroticism and inhibition, which together form the so-called type D personality).
• Environmental mediators include social support, financial support, a history of a significant life change, and trauma early in life, which may increase one’s subsequent vulnerability to unhealthy behaviors.
• Biologic mediators may include prolonged sympathetic activation and enhanced platelet activation, caused by increased levels of depression and anxiety in chronically stressed caregivers.¹⁴

IMPLICATIONS FOR INTERVENTION

A significant percentage of caregivers do not need a clinician’s intervention to help them cope with stress or unhealthy coping skills. Among caregivers aged 50 years or older, 47% indicated in a recent study that the burden of caregiving is low (ie, 1 or 2 on a 5-point scale).¹ Those who respond to stressors as challenges rather than threats tend to be resilient people who exert control over their lives, often through meditation or similar techniques, and have a strong social support network. Many report that caregiving provides them with an opportunity to act in accordance with their values and feel helpful rather than helpless.

Cognitive-behavioral interventions to alleviate stress-related symptoms appear to be more effective if offered as individual rather than group therapy. Teaching caregivers effective coping strategies, rather than merely providing social support, has been shown to improve caregiver psychologic health.¹⁵ Chief among the goals of intervention should be to alter brain function and instill optimism, a sense of control and self-esteem.¹³

The number of people in the United States who spend a significant part of each week working as unpaid caregivers is considerable, and the toll exacted for such work is high. Understanding the profi le of the caregiver, the nature of the duties performed, the stress imposed by such duties, and the consequences of the stress can assist the clinician in recognizing the caregiver in need of intervention.

A PROFILE OF THE CAREGIVER

A recent survey estimated that more than 65 million Americans provide unpaid assistance annually to older adults with disabilities.¹ The value of that labor has been estimated at $306 billion annually, or nearly double the combined cost of home health care and nursing home care.^2,3

The typical caregiver is a woman, about 48 years old, with some college education, who spends 20 hours or more each week providing unpaid care to someone aged 50 years or older.¹ The recipients of care often have long-term physical disabilities; mental confusion or emotional problems frequently complicate care.

PSYCHOLOGIC AND PHYSICAL COSTS

Caregiving may take a toll on the caregiver in a variety of ways: behavioral, in the form of alcohol or substance use⁴; psychologic, in the form of depression or other mental health problems⁵; and physical, in the form of chronic health conditions and impaired immune response.⁶ About three-fifths of caregivers report fair or poor health, compared with one-third of noncaregivers, and caregivers have approximately twice as many chronic conditions, such as heart disease, cancer, arthritis, and diabetes, compared with noncaregivers.^2,7 Caregiving also exacts a financial toll, as employees who are caregivers cost their employers $13.4 billion more per year in health care expenditures.⁸ In addition, absenteeism, workday interruptions, and shifts from full-time to part-time work by caregivers cost businesses between $17.1 and $33.6 billion per year.⁹

The cost of caregiving is higher for women, who exhibit higher levels of anxiety and depression and lower levels of subjective well-being, life satisfaction, and physical health.^10,11 The stress of caregiving has also been identified as a risk factor for morbidity among older (66 to 96 years old) caregivers, who have a 63% greater mortality than noncaregivers of the same age.¹²

PSYCHOSOCIAL STRESS, UNHEALTHY BEHAVIORS, AND ILLNESS ARE LINKED

Psychosocial stress is a predictor of disease and can lead to unhealthy behaviors such as smoking, substance abuse, overeating, poor nutrition, and a sedentary lifestyle; these, in turn, can lead to physical and psychiatric illness. Behaviors adopted initially as coping skills may persist to become chronic, thereby promoting either continued wellness (in the case of healthy coping behaviors) or worsening levels of illness (in the case of unhealthy coping behaviors).

McEwen and Gianaros¹³ have suggested that these stress mechanisms arise from patterns of communication between the brain and the autonomic, cardiovascular, and immune systems, which mutually influenceone another. These so-called bidirectional stress processes affect cognition, experience, and behavior.

An integrated model of stress that maps the bidirectional causal pathways among psychosocial stressors, resulting unhealthy behaviors, and illness is needed. Although the steps from unhealthy behaviors to illness are fairly well understood, the links from psychosocial stress, such as those exhibited by caregivers, to unhealthy behaviors are not as clear. Several mediators are under study:

• Personality mediators can be either ameliorative (resilience, self-confi dence, self-control, optimism, high self-esteem, a sense of mastery, and finding meaning in life) or exacerbating (neuroticism and inhibition, which together form the so-called type D personality).
• Environmental mediators include social support, financial support, a history of a significant life change, and trauma early in life, which may increase one’s subsequent vulnerability to unhealthy behaviors.
• Biologic mediators may include prolonged sympathetic activation and enhanced platelet activation, caused by increased levels of depression and anxiety in chronically stressed caregivers.¹⁴

IMPLICATIONS FOR INTERVENTION

A significant percentage of caregivers do not need a clinician’s intervention to help them cope with stress or unhealthy coping skills. Among caregivers aged 50 years or older, 47% indicated in a recent study that the burden of caregiving is low (ie, 1 or 2 on a 5-point scale).¹ Those who respond to stressors as challenges rather than threats tend to be resilient people who exert control over their lives, often through meditation or similar techniques, and have a strong social support network. Many report that caregiving provides them with an opportunity to act in accordance with their values and feel helpful rather than helpless.

Cognitive-behavioral interventions to alleviate stress-related symptoms appear to be more effective if offered as individual rather than group therapy. Teaching caregivers effective coping strategies, rather than merely providing social support, has been shown to improve caregiver psychologic health.¹⁵ Chief among the goals of intervention should be to alter brain function and instill optimism, a sense of control and self-esteem.¹³

The traditional paradigm for cardiac care has emphasized the use of technology to treat disease. Our focus on technologies such as echocardiography, advanced imaging instrumentation, and cardiac catheterization mirrors the preoccupation of society as a whole with technologic advances.

Attention has only recently been given to the patient’s emotional experience and how this might relate to outcomes, recovery, and healing. An expanded paradigm of cardiac care incorporates pain relief, emotional support, spiritual healing, and a caring environment. These elements of patient-centered care aim to relieve stress and anxiety in order to achieve a better clinical outcome.

PATIENT-CENTERED CARE

The importance of patient-centered care is illustrated by the results of a 2007 survey in which 41% of patients cited elements of the patient experience as factors that most influenced their choice of hospital.¹ Accepted wisdom on patient choice has historically centered on medical factors such as clinical reputation, physician recommendations, and hospital location, each of which was cited by 18% to 21% of the patients surveyed. Elements of the patient experience cited in the study include stress-reducing factors such as the appearance of the room, ease of scheduling, an environment that supports family needs, convenience and comfort of common areas, on-time performance, and simple registration procedures.

Székely et al² found in a 4-year followup study that high levels of preoperative anxiety predicted greater mortality and cardiovascular morbidity following cardiac surgery. In a study by Tully et al,³ preoperative anxiety was also predictive of hospital readmission following cardiac surgery. Preoperative stress and anxiety are reliable predictors of postoperative distress.⁴

The variety and relative efficacy of interventions to reduce stress and anxiety are not well studied. Voss et al⁵ showed that cardiac surgery patients who were played soothing music experienced significantly reduced anxiety, pain, pain distress, and length of hospital stay. One Cleveland Clinic study of massage therapy, however, was unable to demonstrate a statistically significant therapeutic benefit, despite patient satisfaction with the therapy.⁶

THE ADVENT OF HEALING SERVICES

Identifying patients who exhibit significant preoperative stress and providing, as part of an expanded cardiac care paradigm, emotional care both pre- and postoperatively may ameliorate clinical outcomes. As such, the Heart and Vascular Institute at the Cleveland Clinic formed a healing services division, based on the concept that healing is more than simply physical recovery from a particular procedure. The division’s mission statement is: “To enhance the patient experience by promoting healing through a comprehensive set of coordinated services addressing the holistic needs of the patient.”

At the Cleveland Clinic, healing services are now integrated with standard services to enhance the cardiac care paradigm. Our standard medical services focus on areas of communication and pain control, both of which affect anxiety and stress. The need for enhanced communication is significant: 75% of patients admitted to a Chicago hospital were unable to name a single doctor assigned to their care, and of the remaining 25%, only 40% of responders were correct.⁷

It is worth noting that communicating more information to a patient is not necessarily better. Patients given detailed preoperative information about their disease and the potential complications of their cardiac surgery had levels of preoperative, perioperative, and postoperative stress, anxiety, and depression similar to those who received routine medical information.^8,9 On the other hand, patients desire information about their postoperative plan of care while they are experiencing it, and value communication with physicians, nurses, healing services personnel, and other caregivers when it is presented in a calm and forthright manner. Communications should emphasize that the entire clinical team is there to help the patient get better.

THE FIFTH VITAL SIGN

Pain control is an aspect of care that was long ignored. The goal of the pain control task force at the Cleveland Clinic is the development of effective, efficient, and compassionate pain management.

The fifth vital sign, one that escapes the electronic medical record, can be addressed by this question: “How are you feeling?” Treating pain will reduce stress and anxiety. Before surgery, pain management priorities are discussed with patients, and at each daily encounter the goal is to set, refine, and exceed expectations for pain control through discussion and frequent pain assessments.

Reducing anxiety and stress is the goal of both standard care services and healing services, resulting in more satisfied patients with better clinical outcomes.

CASE: “YOU AND THE TEAM MADE ME GET OUTOF BED AND MOVE FORWARD”

Bobbi is a 78-year-old woman who was initially recovering well following cardiac surgery, including valve surgery, but had to return to the intensive care unit, which is difficult for patients. She was subsequently returned to the floor but was reluctant to walk and progressed slowly, despite normal electrocardiogram, radiographs, and blood panel results. We discovered that her husband was in hospice care in another state, causing Bobbi anxiety as she expressed concern over being her husband’s caregiver while being weakened physically herself. She was fearful of moving forward and her recovery stalled.

The primary care nurse referred her to the healing services team. The healing services team provided support for her anxiety and stress, and reviewed options for managing her husband’s care. She participated in Reiki, spiritual support, and social work services. During her admission her husband died, so the team provided appropriate support.

When asked about her experience upon leavingthe hospital, Bobbi did not mention her surgeon or the success of her heart valve procedure, but commented instead on the healing services team that enabled her to get through the experience.

The traditional paradigm for cardiac care has emphasized the use of technology to treat disease. Our focus on technologies such as echocardiography, advanced imaging instrumentation, and cardiac catheterization mirrors the preoccupation of society as a whole with technologic advances.

Attention has only recently been given to the patient’s emotional experience and how this might relate to outcomes, recovery, and healing. An expanded paradigm of cardiac care incorporates pain relief, emotional support, spiritual healing, and a caring environment. These elements of patient-centered care aim to relieve stress and anxiety in order to achieve a better clinical outcome.

PATIENT-CENTERED CARE

The importance of patient-centered care is illustrated by the results of a 2007 survey in which 41% of patients cited elements of the patient experience as factors that most influenced their choice of hospital.¹ Accepted wisdom on patient choice has historically centered on medical factors such as clinical reputation, physician recommendations, and hospital location, each of which was cited by 18% to 21% of the patients surveyed. Elements of the patient experience cited in the study include stress-reducing factors such as the appearance of the room, ease of scheduling, an environment that supports family needs, convenience and comfort of common areas, on-time performance, and simple registration procedures.

Székely et al² found in a 4-year followup study that high levels of preoperative anxiety predicted greater mortality and cardiovascular morbidity following cardiac surgery. In a study by Tully et al,³ preoperative anxiety was also predictive of hospital readmission following cardiac surgery. Preoperative stress and anxiety are reliable predictors of postoperative distress.⁴

The variety and relative efficacy of interventions to reduce stress and anxiety are not well studied. Voss et al⁵ showed that cardiac surgery patients who were played soothing music experienced significantly reduced anxiety, pain, pain distress, and length of hospital stay. One Cleveland Clinic study of massage therapy, however, was unable to demonstrate a statistically significant therapeutic benefit, despite patient satisfaction with the therapy.⁶

THE ADVENT OF HEALING SERVICES

Identifying patients who exhibit significant preoperative stress and providing, as part of an expanded cardiac care paradigm, emotional care both pre- and postoperatively may ameliorate clinical outcomes. As such, the Heart and Vascular Institute at the Cleveland Clinic formed a healing services division, based on the concept that healing is more than simply physical recovery from a particular procedure. The division’s mission statement is: “To enhance the patient experience by promoting healing through a comprehensive set of coordinated services addressing the holistic needs of the patient.”

At the Cleveland Clinic, healing services are now integrated with standard services to enhance the cardiac care paradigm. Our standard medical services focus on areas of communication and pain control, both of which affect anxiety and stress. The need for enhanced communication is significant: 75% of patients admitted to a Chicago hospital were unable to name a single doctor assigned to their care, and of the remaining 25%, only 40% of responders were correct.⁷

It is worth noting that communicating more information to a patient is not necessarily better. Patients given detailed preoperative information about their disease and the potential complications of their cardiac surgery had levels of preoperative, perioperative, and postoperative stress, anxiety, and depression similar to those who received routine medical information.^8,9 On the other hand, patients desire information about their postoperative plan of care while they are experiencing it, and value communication with physicians, nurses, healing services personnel, and other caregivers when it is presented in a calm and forthright manner. Communications should emphasize that the entire clinical team is there to help the patient get better.

THE FIFTH VITAL SIGN

Pain control is an aspect of care that was long ignored. The goal of the pain control task force at the Cleveland Clinic is the development of effective, efficient, and compassionate pain management.

The fifth vital sign, one that escapes the electronic medical record, can be addressed by this question: “How are you feeling?” Treating pain will reduce stress and anxiety. Before surgery, pain management priorities are discussed with patients, and at each daily encounter the goal is to set, refine, and exceed expectations for pain control through discussion and frequent pain assessments.

Reducing anxiety and stress is the goal of both standard care services and healing services, resulting in more satisfied patients with better clinical outcomes.

CASE: “YOU AND THE TEAM MADE ME GET OUTOF BED AND MOVE FORWARD”

Bobbi is a 78-year-old woman who was initially recovering well following cardiac surgery, including valve surgery, but had to return to the intensive care unit, which is difficult for patients. She was subsequently returned to the floor but was reluctant to walk and progressed slowly, despite normal electrocardiogram, radiographs, and blood panel results. We discovered that her husband was in hospice care in another state, causing Bobbi anxiety as she expressed concern over being her husband’s caregiver while being weakened physically herself. She was fearful of moving forward and her recovery stalled.

The primary care nurse referred her to the healing services team. The healing services team provided support for her anxiety and stress, and reviewed options for managing her husband’s care. She participated in Reiki, spiritual support, and social work services. During her admission her husband died, so the team provided appropriate support.

When asked about her experience upon leavingthe hospital, Bobbi did not mention her surgeon or the success of her heart valve procedure, but commented instead on the healing services team that enabled her to get through the experience.

The traditional paradigm for cardiac care has emphasized the use of technology to treat disease. Our focus on technologies such as echocardiography, advanced imaging instrumentation, and cardiac catheterization mirrors the preoccupation of society as a whole with technologic advances.

Attention has only recently been given to the patient’s emotional experience and how this might relate to outcomes, recovery, and healing. An expanded paradigm of cardiac care incorporates pain relief, emotional support, spiritual healing, and a caring environment. These elements of patient-centered care aim to relieve stress and anxiety in order to achieve a better clinical outcome.

PATIENT-CENTERED CARE

The importance of patient-centered care is illustrated by the results of a 2007 survey in which 41% of patients cited elements of the patient experience as factors that most influenced their choice of hospital.¹ Accepted wisdom on patient choice has historically centered on medical factors such as clinical reputation, physician recommendations, and hospital location, each of which was cited by 18% to 21% of the patients surveyed. Elements of the patient experience cited in the study include stress-reducing factors such as the appearance of the room, ease of scheduling, an environment that supports family needs, convenience and comfort of common areas, on-time performance, and simple registration procedures.

Székely et al² found in a 4-year followup study that high levels of preoperative anxiety predicted greater mortality and cardiovascular morbidity following cardiac surgery. In a study by Tully et al,³ preoperative anxiety was also predictive of hospital readmission following cardiac surgery. Preoperative stress and anxiety are reliable predictors of postoperative distress.⁴

The variety and relative efficacy of interventions to reduce stress and anxiety are not well studied. Voss et al⁵ showed that cardiac surgery patients who were played soothing music experienced significantly reduced anxiety, pain, pain distress, and length of hospital stay. One Cleveland Clinic study of massage therapy, however, was unable to demonstrate a statistically significant therapeutic benefit, despite patient satisfaction with the therapy.⁶

THE ADVENT OF HEALING SERVICES

Identifying patients who exhibit significant preoperative stress and providing, as part of an expanded cardiac care paradigm, emotional care both pre- and postoperatively may ameliorate clinical outcomes. As such, the Heart and Vascular Institute at the Cleveland Clinic formed a healing services division, based on the concept that healing is more than simply physical recovery from a particular procedure. The division’s mission statement is: “To enhance the patient experience by promoting healing through a comprehensive set of coordinated services addressing the holistic needs of the patient.”

At the Cleveland Clinic, healing services are now integrated with standard services to enhance the cardiac care paradigm. Our standard medical services focus on areas of communication and pain control, both of which affect anxiety and stress. The need for enhanced communication is significant: 75% of patients admitted to a Chicago hospital were unable to name a single doctor assigned to their care, and of the remaining 25%, only 40% of responders were correct.⁷

It is worth noting that communicating more information to a patient is not necessarily better. Patients given detailed preoperative information about their disease and the potential complications of their cardiac surgery had levels of preoperative, perioperative, and postoperative stress, anxiety, and depression similar to those who received routine medical information.^8,9 On the other hand, patients desire information about their postoperative plan of care while they are experiencing it, and value communication with physicians, nurses, healing services personnel, and other caregivers when it is presented in a calm and forthright manner. Communications should emphasize that the entire clinical team is there to help the patient get better.

THE FIFTH VITAL SIGN

Pain control is an aspect of care that was long ignored. The goal of the pain control task force at the Cleveland Clinic is the development of effective, efficient, and compassionate pain management.

The fifth vital sign, one that escapes the electronic medical record, can be addressed by this question: “How are you feeling?” Treating pain will reduce stress and anxiety. Before surgery, pain management priorities are discussed with patients, and at each daily encounter the goal is to set, refine, and exceed expectations for pain control through discussion and frequent pain assessments.

Reducing anxiety and stress is the goal of both standard care services and healing services, resulting in more satisfied patients with better clinical outcomes.

CASE: “YOU AND THE TEAM MADE ME GET OUTOF BED AND MOVE FORWARD”

Bobbi is a 78-year-old woman who was initially recovering well following cardiac surgery, including valve surgery, but had to return to the intensive care unit, which is difficult for patients. She was subsequently returned to the floor but was reluctant to walk and progressed slowly, despite normal electrocardiogram, radiographs, and blood panel results. We discovered that her husband was in hospice care in another state, causing Bobbi anxiety as she expressed concern over being her husband’s caregiver while being weakened physically herself. She was fearful of moving forward and her recovery stalled.

The primary care nurse referred her to the healing services team. The healing services team provided support for her anxiety and stress, and reviewed options for managing her husband’s care. She participated in Reiki, spiritual support, and social work services. During her admission her husband died, so the team provided appropriate support.

When asked about her experience upon leavingthe hospital, Bobbi did not mention her surgeon or the success of her heart valve procedure, but commented instead on the healing services team that enabled her to get through the experience.

The impact of clinician stress on the health care system is significant. It can adversely affect the patient experience, compromise patient safety, hinder the delivery of care in a manner that is inconsistent with producing quality outcomes, and increase the overall cost of care.

CLINICIAN STRESS IS PREVALENT

Models of health care that restore human interaction are desperately needed. Clinicians today are overwhelmed by performance assessments that are based on length of stay, use of evidence-based medication regimens, and morbidity and mortality outcomes. Yet clinicians have few opportunities to establish more than cursory relationships with their patients—relationships that would permit better understanding of patients’ emotional well-being and that would optimize the overall healing experience.

Shanafelt et al¹ surveyed 7,905 surgeons and found that clinician stress is pervasive: 64% indicated that their work schedule left inadequate time for their personal or family life, 40% reported burnout, and 30% screened positive for symptoms of depression. Another survey of 763 practicing physicians in California found that 53% reported moderate to severe levels of stress.² Nonphysician clinicians have significant levels of stress as well, with one survey of nurses finding that, of those who quit the profession, 26% cited stress as the cause.³

THE EFFECT OF CLINICIAN STRESSON QUALITY OF CARE

In the Shanafelt et al study, high levels of emotional exhaustion correlated positively with major medical errors over the previous 3 months.¹ Nearly 9% of the surgeons surveyed reported making a stress-related major medical mistake in the past 3 months; among those surgeons with high levels of emotional exhaustion, that figure was nearly 15%. This study also found that every 1-point increase in the emotional exhaustion scale (range, 0 to 54) was associated with a 5% increase in the likelihood of reporting a medical error.¹

In a study of internal medicine residents, fatigue and distress were associated with medical errors, which were reported by 39% of respondents.⁴

STRESS AND COMMUNICATION

Stress can damage the physician-nurse relationship, with a significant impact not only on clinicians, but also on delivery of care. The associated breakdowns in communication can negatively affect several areas, including critical care transitions and timely delivery of care. Stress also affects morale, job satisfaction, and job retention.⁵

ADDRESSING THE IMPACT OF CLINICIAN STRESS

The traditional response to complaints registered by patients has been behavioral coaching, disruptive-behavior programs, and the punitive use of satisfaction metrics, which are incorporated into the physician’s annual evaluation. These approaches do little to address the cause of the stress and can inculcate cynicism instead.

A more useful approach is to define and strive for an optimal working environment for clinicians, thereby promoting an enhanced patient experience. This approach attempts to restore balance to both the business and art of medicine and may incorporate biofeedback and other healing services to clinicians as tools to minimize and manage stress.

The business of medicine may be restored by enhancing the culture and climate of the hospital, improving communication and collaboration, reducing administrative tasks, restoring authority and autonomy, and eliminating punitive practices. The art of medicine may be restored by valuing the sacred relationship between clinician and patient, learning to listen more carefully to the patient, creating better healing environments, providing emotional support, and supporting caregivers.

The impact of clinician stress on the health care system is significant. It can adversely affect the patient experience, compromise patient safety, hinder the delivery of care in a manner that is inconsistent with producing quality outcomes, and increase the overall cost of care.

CLINICIAN STRESS IS PREVALENT

Models of health care that restore human interaction are desperately needed. Clinicians today are overwhelmed by performance assessments that are based on length of stay, use of evidence-based medication regimens, and morbidity and mortality outcomes. Yet clinicians have few opportunities to establish more than cursory relationships with their patients—relationships that would permit better understanding of patients’ emotional well-being and that would optimize the overall healing experience.

Shanafelt et al¹ surveyed 7,905 surgeons and found that clinician stress is pervasive: 64% indicated that their work schedule left inadequate time for their personal or family life, 40% reported burnout, and 30% screened positive for symptoms of depression. Another survey of 763 practicing physicians in California found that 53% reported moderate to severe levels of stress.² Nonphysician clinicians have significant levels of stress as well, with one survey of nurses finding that, of those who quit the profession, 26% cited stress as the cause.³

THE EFFECT OF CLINICIAN STRESSON QUALITY OF CARE

In the Shanafelt et al study, high levels of emotional exhaustion correlated positively with major medical errors over the previous 3 months.¹ Nearly 9% of the surgeons surveyed reported making a stress-related major medical mistake in the past 3 months; among those surgeons with high levels of emotional exhaustion, that figure was nearly 15%. This study also found that every 1-point increase in the emotional exhaustion scale (range, 0 to 54) was associated with a 5% increase in the likelihood of reporting a medical error.¹

In a study of internal medicine residents, fatigue and distress were associated with medical errors, which were reported by 39% of respondents.⁴

STRESS AND COMMUNICATION

Stress can damage the physician-nurse relationship, with a significant impact not only on clinicians, but also on delivery of care. The associated breakdowns in communication can negatively affect several areas, including critical care transitions and timely delivery of care. Stress also affects morale, job satisfaction, and job retention.⁵

ADDRESSING THE IMPACT OF CLINICIAN STRESS

The traditional response to complaints registered by patients has been behavioral coaching, disruptive-behavior programs, and the punitive use of satisfaction metrics, which are incorporated into the physician’s annual evaluation. These approaches do little to address the cause of the stress and can inculcate cynicism instead.

A more useful approach is to define and strive for an optimal working environment for clinicians, thereby promoting an enhanced patient experience. This approach attempts to restore balance to both the business and art of medicine and may incorporate biofeedback and other healing services to clinicians as tools to minimize and manage stress.

The business of medicine may be restored by enhancing the culture and climate of the hospital, improving communication and collaboration, reducing administrative tasks, restoring authority and autonomy, and eliminating punitive practices. The art of medicine may be restored by valuing the sacred relationship between clinician and patient, learning to listen more carefully to the patient, creating better healing environments, providing emotional support, and supporting caregivers.

The impact of clinician stress on the health care system is significant. It can adversely affect the patient experience, compromise patient safety, hinder the delivery of care in a manner that is inconsistent with producing quality outcomes, and increase the overall cost of care.

CLINICIAN STRESS IS PREVALENT

Models of health care that restore human interaction are desperately needed. Clinicians today are overwhelmed by performance assessments that are based on length of stay, use of evidence-based medication regimens, and morbidity and mortality outcomes. Yet clinicians have few opportunities to establish more than cursory relationships with their patients—relationships that would permit better understanding of patients’ emotional well-being and that would optimize the overall healing experience.

Shanafelt et al¹ surveyed 7,905 surgeons and found that clinician stress is pervasive: 64% indicated that their work schedule left inadequate time for their personal or family life, 40% reported burnout, and 30% screened positive for symptoms of depression. Another survey of 763 practicing physicians in California found that 53% reported moderate to severe levels of stress.² Nonphysician clinicians have significant levels of stress as well, with one survey of nurses finding that, of those who quit the profession, 26% cited stress as the cause.³

THE EFFECT OF CLINICIAN STRESSON QUALITY OF CARE

In the Shanafelt et al study, high levels of emotional exhaustion correlated positively with major medical errors over the previous 3 months.¹ Nearly 9% of the surgeons surveyed reported making a stress-related major medical mistake in the past 3 months; among those surgeons with high levels of emotional exhaustion, that figure was nearly 15%. This study also found that every 1-point increase in the emotional exhaustion scale (range, 0 to 54) was associated with a 5% increase in the likelihood of reporting a medical error.¹

In a study of internal medicine residents, fatigue and distress were associated with medical errors, which were reported by 39% of respondents.⁴

STRESS AND COMMUNICATION

Stress can damage the physician-nurse relationship, with a significant impact not only on clinicians, but also on delivery of care. The associated breakdowns in communication can negatively affect several areas, including critical care transitions and timely delivery of care. Stress also affects morale, job satisfaction, and job retention.⁵

ADDRESSING THE IMPACT OF CLINICIAN STRESS

The traditional response to complaints registered by patients has been behavioral coaching, disruptive-behavior programs, and the punitive use of satisfaction metrics, which are incorporated into the physician’s annual evaluation. These approaches do little to address the cause of the stress and can inculcate cynicism instead.

A more useful approach is to define and strive for an optimal working environment for clinicians, thereby promoting an enhanced patient experience. This approach attempts to restore balance to both the business and art of medicine and may incorporate biofeedback and other healing services to clinicians as tools to minimize and manage stress.

The business of medicine may be restored by enhancing the culture and climate of the hospital, improving communication and collaboration, reducing administrative tasks, restoring authority and autonomy, and eliminating punitive practices. The art of medicine may be restored by valuing the sacred relationship between clinician and patient, learning to listen more carefully to the patient, creating better healing environments, providing emotional support, and supporting caregivers.

Traditionally, biofeedback was considered to be a stress management technique that targeted sympathetic nervous system (SNS) overdrive with an adrenal medullary system backup. Recent advances in autonomic physiology, however, have clarified that except in extreme situations, the SNS is not the key factor in day-to-day stress. Rather, the parasympathetic branch of the autonomic nervous system appears to be a more likely candidate for mediating routine stress because, unlike the SNS, which has slow-acting neurotransmitters (ie, catecholamines), the parasympathetic nervous system has the fast-acting transmitter acetylcholine.

VAGAL WITHDRAWAL: AN ALTERNATIVE TO SYMPATHETIC ACTIVATION

Porges¹ first proposed the concept of vagal withdrawal as an indicator of stress and stress vulnerability; this contrasts with the idea that the stress response is a consequence of sympathetic activation and the hypothalamic-pituitary-adrenal axis response. In the vagal withdrawal model, the response to stress is stabilization of the sympathetic system followed by termination of parasympathetic activity, manifested as cardiac acceleration.

Respiratory sinus arrhythmia (RSA), or the variability in heart rate as it synchronizes with breathing, is considered an index of parasympathetic tone. In the laboratory, slow atropine infusion produces a transient paradoxical vagomimetic effect characterized by an initial increase in RSA, followed by a flattening and then a rise in the heart rate.² This phenomenon has been measured in people during times of routine stress, such as when worrying about being late for an appointment. In such individuals, biofeedback training can result in recovery of normal RSA shortly after an episode of anxiety.

Historically, the focus of biofeedback was to cultivate low arousal, presumably reducing SNS activity, through the use of finger temperature, skin conductance training, and profound muscle relaxation. More sophisticated ways to look at both branches of the autonomic nervous system have since emerged that allow for sampling of the beat-by-beat changes in heart rate.

HEART RATE VARIABILITY BIOFEEDBACK

The concept of modifying the respiration rate (paced breathing) originated some 2,500 years ago as a component of meditation. It is being revisited today in the form of heart rate variability (HRV) biofeedback training, which is being used as a stress-management tool and a method to correct disorders in which autonomic regulation is thought to be important. HRV biofeedback involves training to increase the amplitude of HRV rhythms and thus improve autonomic homeostasis.

Normal HRV has a pattern of overlapping oscillatory frequency components, including:

• a high-frequency rhythm, 0.15 to 0.4 Hz, which is the RSA;
• a low-frequency rhythm, 0.05 to 0.15 Hz, associated with blood pressure oscillations; and
• a very-low-frequency rhythm, 0.005 to 0.05 Hz, which may regulate vascular tone and body temperature.

The goal of HRV biofeedback is to achieve respiratory rates at which resonance occurs between cardiac rhythms associated with respiration (RSA, or high-frequency oscillations) and those caused by baroreflex activity (low-frequency oscillations).

Spectral analysis has demonstrated that nearly all of the activity with HRV biofeedback occurs at a low-frequency band. The reason is that activity in the low-frequency band is related more to baroreflex activity than to HRV compared with other ranges of frequency. Breathing rates that correspond to baroreflex effects, called resonance frequency breathing, represent resonance in the cardiovascular system. Several devices are available whose mechanisms are based on the concept of achieving resonance frequency breathing. One such device is a slow-breathing monitor (Resp-e-rate) that has been approved by the US Food and Drug Administration for the adjunctive treatment of hypertension.

Improved HRV may suggest an improved risk status: Kleiger et al⁴ found that the relative risk of mortality was 5.3 times greater for people with SDNN of less than 50 msec compared with those whose SDNN was greater than 100 msec. In Del Pozo’s study, eight of 30 patients in the intervention group achieved an SDNN of greater than 50 msec (vs 0 at pretreatment) compared with three of 31 controls (vs two at pretreatment).³ As an additional benefit of HRV biofeedback, patients in the intervention group who entered the study with hypertension all became normotensive.

In a meta-analysis, van Dixhoorn and White⁵ found fewer cardiac events, fewer episodes of angina, and less occurrence of arrhythmia and exercise-induced ischemia from intensive supervised relaxation therapy in patients with ischemic heart disease. Improvements in scales of depression and anxiety were also observed with relaxation therapy.

Other studies have shown biofeedback to have beneficial effects based on the Posttraumatic Stress Disorder Checklist, the Hamilton Depression Rating Scale, and, in patients with mild to moderate heartfailure, the 6-minute walk test.^6–8

The proposed mechanism for the beneficial effects of biofeedback found in clinical trials is improvement in baroreflex function, producing greater reflex efficiency and improved modulation of autonomic activity.

CONCLUSION

A shift in emphasis to vagal withdrawal has led to new forms of biofeedback that probably potentiate many of the same mechanisms thought to be present in Eastern practices such as yoga and tai chi. Results from small-scale trials have been promising for HRV biofeedback as a means of modifying responses to stress and promoting homeostatic processes that reduce the intensity of symptoms and improve surrogate markers associated with a number of disorders.

Traditionally, biofeedback was considered to be a stress management technique that targeted sympathetic nervous system (SNS) overdrive with an adrenal medullary system backup. Recent advances in autonomic physiology, however, have clarified that except in extreme situations, the SNS is not the key factor in day-to-day stress. Rather, the parasympathetic branch of the autonomic nervous system appears to be a more likely candidate for mediating routine stress because, unlike the SNS, which has slow-acting neurotransmitters (ie, catecholamines), the parasympathetic nervous system has the fast-acting transmitter acetylcholine.

VAGAL WITHDRAWAL: AN ALTERNATIVE TO SYMPATHETIC ACTIVATION

Porges¹ first proposed the concept of vagal withdrawal as an indicator of stress and stress vulnerability; this contrasts with the idea that the stress response is a consequence of sympathetic activation and the hypothalamic-pituitary-adrenal axis response. In the vagal withdrawal model, the response to stress is stabilization of the sympathetic system followed by termination of parasympathetic activity, manifested as cardiac acceleration.

Respiratory sinus arrhythmia (RSA), or the variability in heart rate as it synchronizes with breathing, is considered an index of parasympathetic tone. In the laboratory, slow atropine infusion produces a transient paradoxical vagomimetic effect characterized by an initial increase in RSA, followed by a flattening and then a rise in the heart rate.² This phenomenon has been measured in people during times of routine stress, such as when worrying about being late for an appointment. In such individuals, biofeedback training can result in recovery of normal RSA shortly after an episode of anxiety.

Historically, the focus of biofeedback was to cultivate low arousal, presumably reducing SNS activity, through the use of finger temperature, skin conductance training, and profound muscle relaxation. More sophisticated ways to look at both branches of the autonomic nervous system have since emerged that allow for sampling of the beat-by-beat changes in heart rate.

HEART RATE VARIABILITY BIOFEEDBACK

The concept of modifying the respiration rate (paced breathing) originated some 2,500 years ago as a component of meditation. It is being revisited today in the form of heart rate variability (HRV) biofeedback training, which is being used as a stress-management tool and a method to correct disorders in which autonomic regulation is thought to be important. HRV biofeedback involves training to increase the amplitude of HRV rhythms and thus improve autonomic homeostasis.

Normal HRV has a pattern of overlapping oscillatory frequency components, including:

• a high-frequency rhythm, 0.15 to 0.4 Hz, which is the RSA;
• a low-frequency rhythm, 0.05 to 0.15 Hz, associated with blood pressure oscillations; and
• a very-low-frequency rhythm, 0.005 to 0.05 Hz, which may regulate vascular tone and body temperature.

The goal of HRV biofeedback is to achieve respiratory rates at which resonance occurs between cardiac rhythms associated with respiration (RSA, or high-frequency oscillations) and those caused by baroreflex activity (low-frequency oscillations).

Spectral analysis has demonstrated that nearly all of the activity with HRV biofeedback occurs at a low-frequency band. The reason is that activity in the low-frequency band is related more to baroreflex activity than to HRV compared with other ranges of frequency. Breathing rates that correspond to baroreflex effects, called resonance frequency breathing, represent resonance in the cardiovascular system. Several devices are available whose mechanisms are based on the concept of achieving resonance frequency breathing. One such device is a slow-breathing monitor (Resp-e-rate) that has been approved by the US Food and Drug Administration for the adjunctive treatment of hypertension.

Improved HRV may suggest an improved risk status: Kleiger et al⁴ found that the relative risk of mortality was 5.3 times greater for people with SDNN of less than 50 msec compared with those whose SDNN was greater than 100 msec. In Del Pozo’s study, eight of 30 patients in the intervention group achieved an SDNN of greater than 50 msec (vs 0 at pretreatment) compared with three of 31 controls (vs two at pretreatment).³ As an additional benefit of HRV biofeedback, patients in the intervention group who entered the study with hypertension all became normotensive.

In a meta-analysis, van Dixhoorn and White⁵ found fewer cardiac events, fewer episodes of angina, and less occurrence of arrhythmia and exercise-induced ischemia from intensive supervised relaxation therapy in patients with ischemic heart disease. Improvements in scales of depression and anxiety were also observed with relaxation therapy.

Other studies have shown biofeedback to have beneficial effects based on the Posttraumatic Stress Disorder Checklist, the Hamilton Depression Rating Scale, and, in patients with mild to moderate heartfailure, the 6-minute walk test.^6–8

The proposed mechanism for the beneficial effects of biofeedback found in clinical trials is improvement in baroreflex function, producing greater reflex efficiency and improved modulation of autonomic activity.

CONCLUSION

A shift in emphasis to vagal withdrawal has led to new forms of biofeedback that probably potentiate many of the same mechanisms thought to be present in Eastern practices such as yoga and tai chi. Results from small-scale trials have been promising for HRV biofeedback as a means of modifying responses to stress and promoting homeostatic processes that reduce the intensity of symptoms and improve surrogate markers associated with a number of disorders.

Traditionally, biofeedback was considered to be a stress management technique that targeted sympathetic nervous system (SNS) overdrive with an adrenal medullary system backup. Recent advances in autonomic physiology, however, have clarified that except in extreme situations, the SNS is not the key factor in day-to-day stress. Rather, the parasympathetic branch of the autonomic nervous system appears to be a more likely candidate for mediating routine stress because, unlike the SNS, which has slow-acting neurotransmitters (ie, catecholamines), the parasympathetic nervous system has the fast-acting transmitter acetylcholine.

VAGAL WITHDRAWAL: AN ALTERNATIVE TO SYMPATHETIC ACTIVATION

Porges¹ first proposed the concept of vagal withdrawal as an indicator of stress and stress vulnerability; this contrasts with the idea that the stress response is a consequence of sympathetic activation and the hypothalamic-pituitary-adrenal axis response. In the vagal withdrawal model, the response to stress is stabilization of the sympathetic system followed by termination of parasympathetic activity, manifested as cardiac acceleration.

Respiratory sinus arrhythmia (RSA), or the variability in heart rate as it synchronizes with breathing, is considered an index of parasympathetic tone. In the laboratory, slow atropine infusion produces a transient paradoxical vagomimetic effect characterized by an initial increase in RSA, followed by a flattening and then a rise in the heart rate.² This phenomenon has been measured in people during times of routine stress, such as when worrying about being late for an appointment. In such individuals, biofeedback training can result in recovery of normal RSA shortly after an episode of anxiety.

Historically, the focus of biofeedback was to cultivate low arousal, presumably reducing SNS activity, through the use of finger temperature, skin conductance training, and profound muscle relaxation. More sophisticated ways to look at both branches of the autonomic nervous system have since emerged that allow for sampling of the beat-by-beat changes in heart rate.

HEART RATE VARIABILITY BIOFEEDBACK

The concept of modifying the respiration rate (paced breathing) originated some 2,500 years ago as a component of meditation. It is being revisited today in the form of heart rate variability (HRV) biofeedback training, which is being used as a stress-management tool and a method to correct disorders in which autonomic regulation is thought to be important. HRV biofeedback involves training to increase the amplitude of HRV rhythms and thus improve autonomic homeostasis.

Normal HRV has a pattern of overlapping oscillatory frequency components, including:

• a high-frequency rhythm, 0.15 to 0.4 Hz, which is the RSA;
• a low-frequency rhythm, 0.05 to 0.15 Hz, associated with blood pressure oscillations; and
• a very-low-frequency rhythm, 0.005 to 0.05 Hz, which may regulate vascular tone and body temperature.

The goal of HRV biofeedback is to achieve respiratory rates at which resonance occurs between cardiac rhythms associated with respiration (RSA, or high-frequency oscillations) and those caused by baroreflex activity (low-frequency oscillations).

Spectral analysis has demonstrated that nearly all of the activity with HRV biofeedback occurs at a low-frequency band. The reason is that activity in the low-frequency band is related more to baroreflex activity than to HRV compared with other ranges of frequency. Breathing rates that correspond to baroreflex effects, called resonance frequency breathing, represent resonance in the cardiovascular system. Several devices are available whose mechanisms are based on the concept of achieving resonance frequency breathing. One such device is a slow-breathing monitor (Resp-e-rate) that has been approved by the US Food and Drug Administration for the adjunctive treatment of hypertension.

Improved HRV may suggest an improved risk status: Kleiger et al⁴ found that the relative risk of mortality was 5.3 times greater for people with SDNN of less than 50 msec compared with those whose SDNN was greater than 100 msec. In Del Pozo’s study, eight of 30 patients in the intervention group achieved an SDNN of greater than 50 msec (vs 0 at pretreatment) compared with three of 31 controls (vs two at pretreatment).³ As an additional benefit of HRV biofeedback, patients in the intervention group who entered the study with hypertension all became normotensive.

In a meta-analysis, van Dixhoorn and White⁵ found fewer cardiac events, fewer episodes of angina, and less occurrence of arrhythmia and exercise-induced ischemia from intensive supervised relaxation therapy in patients with ischemic heart disease. Improvements in scales of depression and anxiety were also observed with relaxation therapy.

Other studies have shown biofeedback to have beneficial effects based on the Posttraumatic Stress Disorder Checklist, the Hamilton Depression Rating Scale, and, in patients with mild to moderate heartfailure, the 6-minute walk test.^6–8

The proposed mechanism for the beneficial effects of biofeedback found in clinical trials is improvement in baroreflex function, producing greater reflex efficiency and improved modulation of autonomic activity.

CONCLUSION

A shift in emphasis to vagal withdrawal has led to new forms of biofeedback that probably potentiate many of the same mechanisms thought to be present in Eastern practices such as yoga and tai chi. Results from small-scale trials have been promising for HRV biofeedback as a means of modifying responses to stress and promoting homeostatic processes that reduce the intensity of symptoms and improve surrogate markers associated with a number of disorders.

Posttraumatic stress disorder (PTSD) is a severe anxiety disorder whose symptoms emerge following exposure to extreme stress, such as those encountered in the battlefield or as a result of sexual abuse or natural disasters. The ability to employ coping mechanisms affects the disorder’s presentation as well as the frequency, intensity, and duration of the symptoms. The “Wounded Warrior” program at East Carolina University (Greenville, NC) was developed to promote the functional independence of US Marines, including those with PTSD.

STRESS RESPONSE: INTERACTION OF THE BRAIN AND IMMUNE SYSTEM

Walter Cannon coined the “flight or fight” response to stress in the early 20th century, in which he emphasized the importance of the parasympathetic system.¹ In 1988, Folkow clarified the description as an immune response to stress.² The stress response is now understood to be a neuroendocrine function that includes a feedback loop between the hypothalamus and the pituitary and adrenal glands; stimulation of the hypothalamus promotes secretion of corticotropin-releasing hormone (CRH) into the hypophyseal portal system, which supplies the anterior pituitarywith blood. CRH stimulates the secretion of adrenocorticotropic hormone into the bloodstream by the pituitary, prompting the adrenal glands to release the stress hormone cortisol.

Cortisol mobilizes the body’s defenses to meet the challenge of an adverse situation. It modulates the stress response by inhibiting the further release of CRH by the hypothalamus. Cortisol thus protects healthy cells and tissues by inhibiting an overreaction from the immune system. Without this protective effect, the interaction between the brain and the immune system can become dysregulated, increasing the risk of immune disorders.

THE CENTRAL AUTONOMIC NETWORK

The central nervous system that regulates the overall balance of the autonomic nervous system (ANS) has been called the central autonomic network (CAN).³ The CAN helps control executive, social, affective, attentional, and motivational functions. Therefore, the old paradigm of simply decreasing hyperarousal of the ANS to treat negative affective states and dispositions is inadequate. Instead, restoring the appropriate relationship between the ANS and the central nervous system is the aim behind interventions to treat PTSD.

Autonomic, cognitive, and affective functions assist humans in maintaining balance when confronted with external challenges. The CAN controls inhibitory or negative processes that permit specific behavior and redeploy resources needed elsewhere. When negative circuits are compromised, positive circuits develop, resulting in hypervigilance, the symptoms of which can be devastating and, if not ameliorated, can develop into permanent conditions. In one study,Vietnam veterans with PTSD had an 8% reduction in the volume of their right hippocampus compared with veterans without PTSD. Another study calculated a 26% reduction in the left hippocampus and a 22% reduction in the right in veterans with the most severe PTSD compared with veterans who were in combat but had no PTSD symptoms.⁴

A common subcortical neural system regulates defensive behavior, including autonomic, emotional, and cognitive behavior. When the prefrontal cortex is taken “off line” for whatever reason, parasympathetic inhibitory action is withdrawn, and relative sympathetic dominance, associated with defense, occurs.

CONFRONTING HYPERAROUSAL

The question then arises of how to train the ANS to avoid hypervigilance. Growing evidence supports the use of heart rate variability as a predictor of hypervigilance and inefficient allocation of attentional and cognitive resources.

The overall objective of heart rate variability training is to decrease ANS hyperarousal and to improve its balance. “Wounded warriors” learn to control ANS responses to stress-producing stimuli (eg, thoughts, memories, and images associated with combat). The goal of training is to decrease arousal and maintain ANS balance for increasing lengths of time.

Once it was observed that alpha waves were dysfunctional in vulnerable populations, protocols were developed to train alpha and theta waves as a method of improving function. Peniston and colleagues^5–9 showed that increased alpha and theta brain wave production resulted in normalized personality measures and prolonged the period of time before relapse in alcoholics. This protocol has also shown efficacy as an intervention in depression and PTSD.

BIOFEEDBACK TRAINING PROGRAM

The US Department of Defense is studying a combination of central nervous system biofeedback with ANS biofeedback, with the goal of restoring and maintaining tone between the systems.

The training program used in the study lasts 1 month, and starts with a session for preassessment, 16 biofeedback sessions (four per week), a postprogram evaluation, and a 3-month followup. Each week, participants are exposed to stress-producing stimuli that increase in intensity:

• Week 1: Stroop Color Word Test, math stressor, talk stressor/everyday events
• Week 2: Talk stressor, combat experiences
• Week 3: Images and sounds of combat
• Week 4: Virtual Baghdad or Afghanistan (virtual reality exposure)

Each biofeedback session consists of 5 minutes of baseline evaluation; 5 minutes in which the veteran is subjected to the weekly stressor; 20 minutes of heart rate variability and neurofeedback training; 5 more minutes of training with the weekly stressor; 20 more minutes of heart rate variability and neurofeedback training; and finally 5 minutes of recovery.

SUMMARY

Dysfunction in the balance of both the ANS and central nervous system is associated with symptoms of PTSD in combat veterans. Methods that are designed to restore balance in these systems are needed to ameliorate these symptoms. Biofeedback and neurofeedback are safe methods with which to achieve these goals.

Posttraumatic stress disorder (PTSD) is a severe anxiety disorder whose symptoms emerge following exposure to extreme stress, such as those encountered in the battlefield or as a result of sexual abuse or natural disasters. The ability to employ coping mechanisms affects the disorder’s presentation as well as the frequency, intensity, and duration of the symptoms. The “Wounded Warrior” program at East Carolina University (Greenville, NC) was developed to promote the functional independence of US Marines, including those with PTSD.

STRESS RESPONSE: INTERACTION OF THE BRAIN AND IMMUNE SYSTEM

Walter Cannon coined the “flight or fight” response to stress in the early 20th century, in which he emphasized the importance of the parasympathetic system.¹ In 1988, Folkow clarified the description as an immune response to stress.² The stress response is now understood to be a neuroendocrine function that includes a feedback loop between the hypothalamus and the pituitary and adrenal glands; stimulation of the hypothalamus promotes secretion of corticotropin-releasing hormone (CRH) into the hypophyseal portal system, which supplies the anterior pituitarywith blood. CRH stimulates the secretion of adrenocorticotropic hormone into the bloodstream by the pituitary, prompting the adrenal glands to release the stress hormone cortisol.

Cortisol mobilizes the body’s defenses to meet the challenge of an adverse situation. It modulates the stress response by inhibiting the further release of CRH by the hypothalamus. Cortisol thus protects healthy cells and tissues by inhibiting an overreaction from the immune system. Without this protective effect, the interaction between the brain and the immune system can become dysregulated, increasing the risk of immune disorders.

THE CENTRAL AUTONOMIC NETWORK

The central nervous system that regulates the overall balance of the autonomic nervous system (ANS) has been called the central autonomic network (CAN).³ The CAN helps control executive, social, affective, attentional, and motivational functions. Therefore, the old paradigm of simply decreasing hyperarousal of the ANS to treat negative affective states and dispositions is inadequate. Instead, restoring the appropriate relationship between the ANS and the central nervous system is the aim behind interventions to treat PTSD.

Autonomic, cognitive, and affective functions assist humans in maintaining balance when confronted with external challenges. The CAN controls inhibitory or negative processes that permit specific behavior and redeploy resources needed elsewhere. When negative circuits are compromised, positive circuits develop, resulting in hypervigilance, the symptoms of which can be devastating and, if not ameliorated, can develop into permanent conditions. In one study,Vietnam veterans with PTSD had an 8% reduction in the volume of their right hippocampus compared with veterans without PTSD. Another study calculated a 26% reduction in the left hippocampus and a 22% reduction in the right in veterans with the most severe PTSD compared with veterans who were in combat but had no PTSD symptoms.⁴

A common subcortical neural system regulates defensive behavior, including autonomic, emotional, and cognitive behavior. When the prefrontal cortex is taken “off line” for whatever reason, parasympathetic inhibitory action is withdrawn, and relative sympathetic dominance, associated with defense, occurs.

CONFRONTING HYPERAROUSAL

The question then arises of how to train the ANS to avoid hypervigilance. Growing evidence supports the use of heart rate variability as a predictor of hypervigilance and inefficient allocation of attentional and cognitive resources.

The overall objective of heart rate variability training is to decrease ANS hyperarousal and to improve its balance. “Wounded warriors” learn to control ANS responses to stress-producing stimuli (eg, thoughts, memories, and images associated with combat). The goal of training is to decrease arousal and maintain ANS balance for increasing lengths of time.

Once it was observed that alpha waves were dysfunctional in vulnerable populations, protocols were developed to train alpha and theta waves as a method of improving function. Peniston and colleagues^5–9 showed that increased alpha and theta brain wave production resulted in normalized personality measures and prolonged the period of time before relapse in alcoholics. This protocol has also shown efficacy as an intervention in depression and PTSD.

BIOFEEDBACK TRAINING PROGRAM

The US Department of Defense is studying a combination of central nervous system biofeedback with ANS biofeedback, with the goal of restoring and maintaining tone between the systems.

The training program used in the study lasts 1 month, and starts with a session for preassessment, 16 biofeedback sessions (four per week), a postprogram evaluation, and a 3-month followup. Each week, participants are exposed to stress-producing stimuli that increase in intensity:

• Week 1: Stroop Color Word Test, math stressor, talk stressor/everyday events
• Week 2: Talk stressor, combat experiences
• Week 3: Images and sounds of combat
• Week 4: Virtual Baghdad or Afghanistan (virtual reality exposure)

Each biofeedback session consists of 5 minutes of baseline evaluation; 5 minutes in which the veteran is subjected to the weekly stressor; 20 minutes of heart rate variability and neurofeedback training; 5 more minutes of training with the weekly stressor; 20 more minutes of heart rate variability and neurofeedback training; and finally 5 minutes of recovery.

SUMMARY

Dysfunction in the balance of both the ANS and central nervous system is associated with symptoms of PTSD in combat veterans. Methods that are designed to restore balance in these systems are needed to ameliorate these symptoms. Biofeedback and neurofeedback are safe methods with which to achieve these goals.

Posttraumatic stress disorder (PTSD) is a severe anxiety disorder whose symptoms emerge following exposure to extreme stress, such as those encountered in the battlefield or as a result of sexual abuse or natural disasters. The ability to employ coping mechanisms affects the disorder’s presentation as well as the frequency, intensity, and duration of the symptoms. The “Wounded Warrior” program at East Carolina University (Greenville, NC) was developed to promote the functional independence of US Marines, including those with PTSD.

STRESS RESPONSE: INTERACTION OF THE BRAIN AND IMMUNE SYSTEM

Walter Cannon coined the “flight or fight” response to stress in the early 20th century, in which he emphasized the importance of the parasympathetic system.¹ In 1988, Folkow clarified the description as an immune response to stress.² The stress response is now understood to be a neuroendocrine function that includes a feedback loop between the hypothalamus and the pituitary and adrenal glands; stimulation of the hypothalamus promotes secretion of corticotropin-releasing hormone (CRH) into the hypophyseal portal system, which supplies the anterior pituitarywith blood. CRH stimulates the secretion of adrenocorticotropic hormone into the bloodstream by the pituitary, prompting the adrenal glands to release the stress hormone cortisol.

Cortisol mobilizes the body’s defenses to meet the challenge of an adverse situation. It modulates the stress response by inhibiting the further release of CRH by the hypothalamus. Cortisol thus protects healthy cells and tissues by inhibiting an overreaction from the immune system. Without this protective effect, the interaction between the brain and the immune system can become dysregulated, increasing the risk of immune disorders.

THE CENTRAL AUTONOMIC NETWORK

The central nervous system that regulates the overall balance of the autonomic nervous system (ANS) has been called the central autonomic network (CAN).³ The CAN helps control executive, social, affective, attentional, and motivational functions. Therefore, the old paradigm of simply decreasing hyperarousal of the ANS to treat negative affective states and dispositions is inadequate. Instead, restoring the appropriate relationship between the ANS and the central nervous system is the aim behind interventions to treat PTSD.

Autonomic, cognitive, and affective functions assist humans in maintaining balance when confronted with external challenges. The CAN controls inhibitory or negative processes that permit specific behavior and redeploy resources needed elsewhere. When negative circuits are compromised, positive circuits develop, resulting in hypervigilance, the symptoms of which can be devastating and, if not ameliorated, can develop into permanent conditions. In one study,Vietnam veterans with PTSD had an 8% reduction in the volume of their right hippocampus compared with veterans without PTSD. Another study calculated a 26% reduction in the left hippocampus and a 22% reduction in the right in veterans with the most severe PTSD compared with veterans who were in combat but had no PTSD symptoms.⁴

A common subcortical neural system regulates defensive behavior, including autonomic, emotional, and cognitive behavior. When the prefrontal cortex is taken “off line” for whatever reason, parasympathetic inhibitory action is withdrawn, and relative sympathetic dominance, associated with defense, occurs.

CONFRONTING HYPERAROUSAL

The question then arises of how to train the ANS to avoid hypervigilance. Growing evidence supports the use of heart rate variability as a predictor of hypervigilance and inefficient allocation of attentional and cognitive resources.

The overall objective of heart rate variability training is to decrease ANS hyperarousal and to improve its balance. “Wounded warriors” learn to control ANS responses to stress-producing stimuli (eg, thoughts, memories, and images associated with combat). The goal of training is to decrease arousal and maintain ANS balance for increasing lengths of time.

Once it was observed that alpha waves were dysfunctional in vulnerable populations, protocols were developed to train alpha and theta waves as a method of improving function. Peniston and colleagues^5–9 showed that increased alpha and theta brain wave production resulted in normalized personality measures and prolonged the period of time before relapse in alcoholics. This protocol has also shown efficacy as an intervention in depression and PTSD.

BIOFEEDBACK TRAINING PROGRAM

The US Department of Defense is studying a combination of central nervous system biofeedback with ANS biofeedback, with the goal of restoring and maintaining tone between the systems.

The training program used in the study lasts 1 month, and starts with a session for preassessment, 16 biofeedback sessions (four per week), a postprogram evaluation, and a 3-month followup. Each week, participants are exposed to stress-producing stimuli that increase in intensity:

• Week 1: Stroop Color Word Test, math stressor, talk stressor/everyday events
• Week 2: Talk stressor, combat experiences
• Week 3: Images and sounds of combat
• Week 4: Virtual Baghdad or Afghanistan (virtual reality exposure)

Each biofeedback session consists of 5 minutes of baseline evaluation; 5 minutes in which the veteran is subjected to the weekly stressor; 20 minutes of heart rate variability and neurofeedback training; 5 more minutes of training with the weekly stressor; 20 more minutes of heart rate variability and neurofeedback training; and finally 5 minutes of recovery.

SUMMARY

Dysfunction in the balance of both the ANS and central nervous system is associated with symptoms of PTSD in combat veterans. Methods that are designed to restore balance in these systems are needed to ameliorate these symptoms. Biofeedback and neurofeedback are safe methods with which to achieve these goals.

Question from audience: Why does the Cleveland Clinic start its healing services program preoperatively rather than postoperatively?

Dr. Gillinov: We have a fairly well defined preoperative set of medical tests, and during this process nurses present patients with materials that explain the experience, and nurses and doctors make themselves available in special classes to answer patients’ questions. In doing so, we have increasingly identified patients preoperatively who have stress or problems.

Last week I saw a woman who had a leaking mitral valve, but her symptoms were out of proportion to her disease. She had loss of energy and appetite, and she wasn’t eating much. She was depressed and our team picked that up. She actually never had to undergo surgery. We referred her to a psychologist and, according to her son, she started to feel better. By starting preoperatively, we’re sometimes able to pick out things that we should treat instead of heart disease.

We also provide guided imagery and massage preoperatively.

Dr. Duffy: Healing services is on standing preoperative orders at the hospital. The team goes in proactively and asks, “In addition to your open heart surgery on Wednesday, is there anything we can do to support your emotional and spiritual journey here today?”

Terminology also matters. The term “healing services” is a safe umbrella under which we include biofeedback as one of the services, but it encompasses pastoral care, hospice care, and palliative care. The way it’s integrated into a care model is important. If it’s reserved for end of life, it might be viewed as defective or as a death sentence, so we want the healing services team to be proactive.

Question from audience: How does the primary care physician fit into all of this? I believe that if the physicians in the hospital want to gain patient confidence, they’ll show that they’re communicating well with the primary care physician.

Dr. Gevirtz: The primary care physicians are incredibly open to this idea. They have 12 minutes to deal with people with fibromyalgia, irritable bowel syndrome, chronic pain, noncardiac chest pain, etc. What are they going to do in 12 minutes? They’re grateful if they have a handoff, especially if it’s in the Clinic itself.

Question from audience: Are there any thoughts on making biofeedback part of general training rather than using it just for patients who’ve already experienced trauma?

Dr. Gevirtz: We did a study in which we showed that a biofeedback technician in the primary care setting saved the health maintenance system quite a lot of money, but the administration couldn’t decide whose territory to take to give us an office, so it ended the program.

Dr. Russoniello: How we enable greater access to our intervention is an important question. I see people quit the program if they can’t get access to biofeedback. In an effort to enhance compliance, we’ve incorporated biofeedback into video games, working with a couple of private companies to develop them.The idea is that persons playing the video game can accrue points to enhance their overall score if they perform paced breathing or some other form of biofeedback. Early indications from focus groups are that people will like this.

We have already shown in randomized controlled clinical studies of depression and anxiety that certain video games can improve mood and decrease stress.There is a big movement to get products in people’s hands to help them manage their health.

Question from audience: How much overlap is there between biofeedback methodologies—enhancing heart rate variability, vagal withdrawal, neurofeedback, and electroencephalographic feedback—in the systems you’re targeting and what are the unique contributions of each?

Dr. Gevirtz: We follow a stepped-care model. We start with the simplest and move on to the more complicated technologies. Two published studies with long-term followup showed the effectiveness of a learned breathing technique in alleviating noncardiac chest pain. Simple biofeedback wasn’t even needed. Three years later, the patients were better than they were at the end of the actual training. If you can do it simply, then you do it, and if it doesn’t work, then move on to more and more complicated techniques, with neurofeedback being the last resort.

Question from audience: Has anybody measured the physical impact of stimulating multiple systems on the study subject? In other words, can it be damaging to overstimulate these systems at the same time?

Dr. Gevirtz: We’ve been trying to do that. Recurrent abdominal pain or functional abdominal pain is the most common complaint to pediatric gastroenterologists. We have 1,800 patients a year who make it to the children’s hospital level with this complaint. These are kids who are suffering with very great pain and we we’re pretty sure it’s an autonomically mediated kind of phenomenon. We’re able to measure vagal activity in these kids in ambulatory settings at school and have found very little vagal activity before treatment. After training, they were able to restore vagal activity, and it correlated at the level of 0.63 with a reduction of symptoms. I think it’s important to try to tie the physiology to symptoms. It’s not always easy to do but we’re trying.

Question from audience: I’d like to pick up on two topics that Dr. Duffy raised: the business of medicine and the proposal for informed hope rather than an informed consent before surgery. Something that I see with patients and families at times is this magical expectation promoted by the business side that medicine can do these amazing and wonderful things and doesn’t have any sort of weaknesses. I wonder what role unrealistic expectations promoted by the media, advertising, and others may play in the stress of patients, caregivers, and physicians who need to try to meet the expectations of infallible medicine?

Dr. Duffy: We’ve spun so far the other way with our advanced technology that we’ve lost the human side, especially the concept of a relationship and giving people hope even though they have a terminal condition. It’s a balance between the art and the business of medicine. It’s about setting realistic expectations and realistic hope.

Question from audience: Why does the Cleveland Clinic start its healing services program preoperatively rather than postoperatively?

Dr. Gillinov: We have a fairly well defined preoperative set of medical tests, and during this process nurses present patients with materials that explain the experience, and nurses and doctors make themselves available in special classes to answer patients’ questions. In doing so, we have increasingly identified patients preoperatively who have stress or problems.

Last week I saw a woman who had a leaking mitral valve, but her symptoms were out of proportion to her disease. She had loss of energy and appetite, and she wasn’t eating much. She was depressed and our team picked that up. She actually never had to undergo surgery. We referred her to a psychologist and, according to her son, she started to feel better. By starting preoperatively, we’re sometimes able to pick out things that we should treat instead of heart disease.

We also provide guided imagery and massage preoperatively.

Dr. Duffy: Healing services is on standing preoperative orders at the hospital. The team goes in proactively and asks, “In addition to your open heart surgery on Wednesday, is there anything we can do to support your emotional and spiritual journey here today?”

Terminology also matters. The term “healing services” is a safe umbrella under which we include biofeedback as one of the services, but it encompasses pastoral care, hospice care, and palliative care. The way it’s integrated into a care model is important. If it’s reserved for end of life, it might be viewed as defective or as a death sentence, so we want the healing services team to be proactive.

Question from audience: How does the primary care physician fit into all of this? I believe that if the physicians in the hospital want to gain patient confidence, they’ll show that they’re communicating well with the primary care physician.

Dr. Gevirtz: The primary care physicians are incredibly open to this idea. They have 12 minutes to deal with people with fibromyalgia, irritable bowel syndrome, chronic pain, noncardiac chest pain, etc. What are they going to do in 12 minutes? They’re grateful if they have a handoff, especially if it’s in the Clinic itself.

Question from audience: Are there any thoughts on making biofeedback part of general training rather than using it just for patients who’ve already experienced trauma?

Dr. Gevirtz: We did a study in which we showed that a biofeedback technician in the primary care setting saved the health maintenance system quite a lot of money, but the administration couldn’t decide whose territory to take to give us an office, so it ended the program.

Dr. Russoniello: How we enable greater access to our intervention is an important question. I see people quit the program if they can’t get access to biofeedback. In an effort to enhance compliance, we’ve incorporated biofeedback into video games, working with a couple of private companies to develop them.The idea is that persons playing the video game can accrue points to enhance their overall score if they perform paced breathing or some other form of biofeedback. Early indications from focus groups are that people will like this.

We have already shown in randomized controlled clinical studies of depression and anxiety that certain video games can improve mood and decrease stress.There is a big movement to get products in people’s hands to help them manage their health.

Question from audience: How much overlap is there between biofeedback methodologies—enhancing heart rate variability, vagal withdrawal, neurofeedback, and electroencephalographic feedback—in the systems you’re targeting and what are the unique contributions of each?

Dr. Gevirtz: We follow a stepped-care model. We start with the simplest and move on to the more complicated technologies. Two published studies with long-term followup showed the effectiveness of a learned breathing technique in alleviating noncardiac chest pain. Simple biofeedback wasn’t even needed. Three years later, the patients were better than they were at the end of the actual training. If you can do it simply, then you do it, and if it doesn’t work, then move on to more and more complicated techniques, with neurofeedback being the last resort.

Question from audience: Has anybody measured the physical impact of stimulating multiple systems on the study subject? In other words, can it be damaging to overstimulate these systems at the same time?

Dr. Gevirtz: We’ve been trying to do that. Recurrent abdominal pain or functional abdominal pain is the most common complaint to pediatric gastroenterologists. We have 1,800 patients a year who make it to the children’s hospital level with this complaint. These are kids who are suffering with very great pain and we we’re pretty sure it’s an autonomically mediated kind of phenomenon. We’re able to measure vagal activity in these kids in ambulatory settings at school and have found very little vagal activity before treatment. After training, they were able to restore vagal activity, and it correlated at the level of 0.63 with a reduction of symptoms. I think it’s important to try to tie the physiology to symptoms. It’s not always easy to do but we’re trying.

Question from audience: I’d like to pick up on two topics that Dr. Duffy raised: the business of medicine and the proposal for informed hope rather than an informed consent before surgery. Something that I see with patients and families at times is this magical expectation promoted by the business side that medicine can do these amazing and wonderful things and doesn’t have any sort of weaknesses. I wonder what role unrealistic expectations promoted by the media, advertising, and others may play in the stress of patients, caregivers, and physicians who need to try to meet the expectations of infallible medicine?

Dr. Duffy: We’ve spun so far the other way with our advanced technology that we’ve lost the human side, especially the concept of a relationship and giving people hope even though they have a terminal condition. It’s a balance between the art and the business of medicine. It’s about setting realistic expectations and realistic hope.

Question from audience: Why does the Cleveland Clinic start its healing services program preoperatively rather than postoperatively?

Dr. Gillinov: We have a fairly well defined preoperative set of medical tests, and during this process nurses present patients with materials that explain the experience, and nurses and doctors make themselves available in special classes to answer patients’ questions. In doing so, we have increasingly identified patients preoperatively who have stress or problems.

Last week I saw a woman who had a leaking mitral valve, but her symptoms were out of proportion to her disease. She had loss of energy and appetite, and she wasn’t eating much. She was depressed and our team picked that up. She actually never had to undergo surgery. We referred her to a psychologist and, according to her son, she started to feel better. By starting preoperatively, we’re sometimes able to pick out things that we should treat instead of heart disease.

We also provide guided imagery and massage preoperatively.

Dr. Duffy: Healing services is on standing preoperative orders at the hospital. The team goes in proactively and asks, “In addition to your open heart surgery on Wednesday, is there anything we can do to support your emotional and spiritual journey here today?”

Terminology also matters. The term “healing services” is a safe umbrella under which we include biofeedback as one of the services, but it encompasses pastoral care, hospice care, and palliative care. The way it’s integrated into a care model is important. If it’s reserved for end of life, it might be viewed as defective or as a death sentence, so we want the healing services team to be proactive.

Question from audience: How does the primary care physician fit into all of this? I believe that if the physicians in the hospital want to gain patient confidence, they’ll show that they’re communicating well with the primary care physician.

Dr. Gevirtz: The primary care physicians are incredibly open to this idea. They have 12 minutes to deal with people with fibromyalgia, irritable bowel syndrome, chronic pain, noncardiac chest pain, etc. What are they going to do in 12 minutes? They’re grateful if they have a handoff, especially if it’s in the Clinic itself.

Question from audience: Are there any thoughts on making biofeedback part of general training rather than using it just for patients who’ve already experienced trauma?

Dr. Gevirtz: We did a study in which we showed that a biofeedback technician in the primary care setting saved the health maintenance system quite a lot of money, but the administration couldn’t decide whose territory to take to give us an office, so it ended the program.

Dr. Russoniello: How we enable greater access to our intervention is an important question. I see people quit the program if they can’t get access to biofeedback. In an effort to enhance compliance, we’ve incorporated biofeedback into video games, working with a couple of private companies to develop them.The idea is that persons playing the video game can accrue points to enhance their overall score if they perform paced breathing or some other form of biofeedback. Early indications from focus groups are that people will like this.

We have already shown in randomized controlled clinical studies of depression and anxiety that certain video games can improve mood and decrease stress.There is a big movement to get products in people’s hands to help them manage their health.

Question from audience: How much overlap is there between biofeedback methodologies—enhancing heart rate variability, vagal withdrawal, neurofeedback, and electroencephalographic feedback—in the systems you’re targeting and what are the unique contributions of each?

Dr. Gevirtz: We follow a stepped-care model. We start with the simplest and move on to the more complicated technologies. Two published studies with long-term followup showed the effectiveness of a learned breathing technique in alleviating noncardiac chest pain. Simple biofeedback wasn’t even needed. Three years later, the patients were better than they were at the end of the actual training. If you can do it simply, then you do it, and if it doesn’t work, then move on to more and more complicated techniques, with neurofeedback being the last resort.

Question from audience: Has anybody measured the physical impact of stimulating multiple systems on the study subject? In other words, can it be damaging to overstimulate these systems at the same time?

Dr. Gevirtz: We’ve been trying to do that. Recurrent abdominal pain or functional abdominal pain is the most common complaint to pediatric gastroenterologists. We have 1,800 patients a year who make it to the children’s hospital level with this complaint. These are kids who are suffering with very great pain and we we’re pretty sure it’s an autonomically mediated kind of phenomenon. We’re able to measure vagal activity in these kids in ambulatory settings at school and have found very little vagal activity before treatment. After training, they were able to restore vagal activity, and it correlated at the level of 0.63 with a reduction of symptoms. I think it’s important to try to tie the physiology to symptoms. It’s not always easy to do but we’re trying.

Question from audience: I’d like to pick up on two topics that Dr. Duffy raised: the business of medicine and the proposal for informed hope rather than an informed consent before surgery. Something that I see with patients and families at times is this magical expectation promoted by the business side that medicine can do these amazing and wonderful things and doesn’t have any sort of weaknesses. I wonder what role unrealistic expectations promoted by the media, advertising, and others may play in the stress of patients, caregivers, and physicians who need to try to meet the expectations of infallible medicine?

Dr. Duffy: We’ve spun so far the other way with our advanced technology that we’ve lost the human side, especially the concept of a relationship and giving people hope even though they have a terminal condition. It’s a balance between the art and the business of medicine. It’s about setting realistic expectations and realistic hope.

The effect of emotion on the heart is not confined to depression, but extends to a variety of mental states; as William Harvey described in 1628, “A mental disturbance provoking pain, excessive joy, hope or anxiety extends to the heart, where it affects its temper and rate, impairing general nutrition and vigor.”

In going beyond the well-established role of depression as a risk factor for heart disease, 2010 delivered several important publications recognizing anxiety, anger, and other forms of distress as key factors in the etiology of coronary heart disease (CHD). Other papers of merit elucidated new and overlooked insights into the pathways linking psychosocial stress and cardiovascular risk, and also considered psychologic states that appear to promote healthy functioning.

IMPACT OF NEGATIVE EMOTIONS ON RISK OF INCIDENT CORONARY HEART DISEASE

In a meta-analysis of 20 prospective studies that included 249,846 persons with a mean follow-up of 11.2 years, Roest et al¹ examined the impact of anxiety characterized by the presence of anxiety symptoms or a diagnosis of anxiety disorder on incident CHD. Most of the studies adjusted for a broad array of relevant potential confounders. Findings suggest the presence of anxiety increases the risk of incident CHD by 26% (P P = .003).

In a meta-analysis of 25 prospective studies of 7,160 persons with a mean follow-up exceeding 10 years, Chida and Steptoe² found that anger increased the risk of incident CHD by 19%, after adjustment for standard coronary risk factors. The effect was less stable than that associated with anxiety and depression, and when stratified by gender, the harmful effects of anger were more evident in men than in women. The effect of anger was attenuated when controlling for behavioral covariates. The association between anger and CHD did not hold for all ways of measuring anger, which suggests that the type of anger or the ability to regulate anger may be relevant to the relationship.

A study that did account for the type of anger expression on the risk of incident CHD was conducted by Davidson and Mostofsky.³ The independent effect of three distinct types of anger expression (constructive anger, destructive anger justification, and destructive anger rumination) on 10-year incident CHD was examined, controlling for other psychosocial factors. In men, higher scores for constructive anger were associated with a lower rate of CHD; in both men and women, higher scores for destructive anger justification were associated with an increased risk of CHD.

Insights gained from these studies are as follows:

• The impact of anxiety appears to be comparable to depression, and the effects of anxiety and depression are largely independent.
• If anxiety and depression co-occur, the effect on CHD is synergistic.
• The effects of anger are less clear; its impact may be independent of or dependent on other forms of psychologic distress.
• Distress in general appears to serve as a signal that something is wrong and needs to be addressed. If ignored, it may become chronic and unremitting; because symptoms of distress may lead to systemic dysregulation and increased CHD risk, they may indicate the need for increased surveillance and intervention.

WHY FOCUS ON THE BIOLOGY OF EMOTIONS?

A clear biologic explanation for the influence of emotional factors on physical health would serve to assuage skeptics who doubt that such a link exists or who attribute a common underlying genetic trait to both negative affect and heart disease. Further, focusing on the biology may help answer key questions with respect to emotions and disease processes: What is the damage incurred by negative emotional states and is it reversible? Can compensatory pathways be activated to bypass the mechanisms causing damage or slow the progression of disease?

Cardiac response to worry and stress

In one study attempting to shed light on relevant emotion-related biologic process, the prolonged physiologic effects of worry were examined. Worry episodes and stressful events were recorded hourly along with ambulatory heart rate and heart rate variability in 73 teachers for 4 days.⁴ Autonomic activity, as reflected by a concurrent elevation in heart rate and a decrease in heart rate variability, was increased up to 2 hours after a worry episode. The findings also suggested that the prolonged cardiac effects of separate worry episodes were independent.

Another study sought to determine whether heightened reactivity or delayed recovery to acute stress increases risk of cardiovascular disease.⁵ This meta-analysis included 36 studies to assess whether acute cardiovascular response to various laboratory stressors (ie, cognitive tasks, stress interviews, public speaking). Findings indicated that heightened cardiovascular reactivity was associated with worse cardiovascular outcomes, such as incident hypertension, coronary calcification, carotid intima-media thickness, and cardiovascular events over time.

Role of aldosterone overlooked

Reprinted from Neuroscience and Biobehavioral Reviews (Kubzansky LD, et al. Aldosterone: a forgotten mediator of the relationship between psychological stress and heart disease. Neurosci Biobehav Rev 2010; 34:80–86), © 2010, with permission from Elsevier.

WHY CONSIDER RESILIENCE?

Because the absence of a deficit is not the same as the presence of an asset, greater insight into dysfunction may be gained by explicitly considering what promotes healthy functioning. Ameliorating distress has proven difficult; so, in studying resilience (including the ability to regulate affect), new targets for prevention and intervention may be identified. Although no meta-analysis of resilience factors has been published to date owing to the paucity of data, the studies that have been performed are generally rigorous and have demonstrated consistent findings.

For example, one prospective, well-controlled study of 1,739 men and women demonstrated a protective effect of positive affect (as ascertained by structured interview) against 10-year incident CHD.⁶ The risk of fatal or nonfatal ischemic heart disease events was reduced by 22% (P = .02) for each 1-point increase in positive affect, even after controlling for depression and negative emotions.

Reprinted, with permission, from Archives of General Psychiatry (Kubzansky LD, et al. Arch Gen Psychiatry 2011; 68:400–408), Copyright © 2011 American Medical Association. All rights reserved.

Biology of resilience: Counteracting cellular damage

Genomic changes can be induced by the relaxation response, as evidenced by the differential gene expression profiles of long-term daily practitioners of relaxation (ie, meditation, yoga), short-term (8-week) practitioners of relaxation, and healthy controls.⁸ Alterations in cellular metabolism, oxidative phosphorylation, and generation of reactive oxygen species that counteract proinflammatory responses, indicative of an adaptive response, were observed in both groups that practiced relaxation.

FUTURE DIRECTIONS

Whether and how the sources and effects of psychosocial stress and response to treatment differ across men and women deserves closer examination. A review by Low et al⁹ summarizes the current state of knowledge with respect to psychosocial factors and heart disease in women, noting that the sources of stress associated with increased CHD risk differ across men and women; psychosocial risk factors like depression and anxiety appear to increase risk for both men and women; work-related stress has larger effects in men while stress related to relationships and family responsibilities appear to have larger effects in women.

Although responses to psychosocial stress are not clearly different between men and women, intervention targeted at reducing distress is much less effective in reducing the risk of adverse events in women versus men. The mechanism to explain this difference in effectiveness of intervention urgently requires further exploration.

In conducting this work, several factors are important. The best time to intervene to reduce psychosocial distress is unknown; a key consideration will be, what is the best etiologic window for intervention? Perhaps a life-course approach that targets individuals with chronically high levels of emotional distress who also have multiple coronary risk factors, and that enhances their capacity to regulate emotions would prove superior to waiting until late in the disease process.

Another area that may prove fruitful is to consider in more depth the biology of the placebo effect and whether and how it may inform our understanding of resilience.

More generally, considering why interventions seem to influence outcomes so differently across men and women, applying a life course approach to determine the best etiologic window for prevention and intervention strategies, and conducting a more in-depth exploration of the biology of resilience may lead to improved capacity for population-based approaches to reducing the burden of CHD.

The effect of emotion on the heart is not confined to depression, but extends to a variety of mental states; as William Harvey described in 1628, “A mental disturbance provoking pain, excessive joy, hope or anxiety extends to the heart, where it affects its temper and rate, impairing general nutrition and vigor.”

In going beyond the well-established role of depression as a risk factor for heart disease, 2010 delivered several important publications recognizing anxiety, anger, and other forms of distress as key factors in the etiology of coronary heart disease (CHD). Other papers of merit elucidated new and overlooked insights into the pathways linking psychosocial stress and cardiovascular risk, and also considered psychologic states that appear to promote healthy functioning.

IMPACT OF NEGATIVE EMOTIONS ON RISK OF INCIDENT CORONARY HEART DISEASE

In a meta-analysis of 20 prospective studies that included 249,846 persons with a mean follow-up of 11.2 years, Roest et al¹ examined the impact of anxiety characterized by the presence of anxiety symptoms or a diagnosis of anxiety disorder on incident CHD. Most of the studies adjusted for a broad array of relevant potential confounders. Findings suggest the presence of anxiety increases the risk of incident CHD by 26% (P P = .003).

In a meta-analysis of 25 prospective studies of 7,160 persons with a mean follow-up exceeding 10 years, Chida and Steptoe² found that anger increased the risk of incident CHD by 19%, after adjustment for standard coronary risk factors. The effect was less stable than that associated with anxiety and depression, and when stratified by gender, the harmful effects of anger were more evident in men than in women. The effect of anger was attenuated when controlling for behavioral covariates. The association between anger and CHD did not hold for all ways of measuring anger, which suggests that the type of anger or the ability to regulate anger may be relevant to the relationship.

A study that did account for the type of anger expression on the risk of incident CHD was conducted by Davidson and Mostofsky.³ The independent effect of three distinct types of anger expression (constructive anger, destructive anger justification, and destructive anger rumination) on 10-year incident CHD was examined, controlling for other psychosocial factors. In men, higher scores for constructive anger were associated with a lower rate of CHD; in both men and women, higher scores for destructive anger justification were associated with an increased risk of CHD.

Insights gained from these studies are as follows:

• The impact of anxiety appears to be comparable to depression, and the effects of anxiety and depression are largely independent.
• If anxiety and depression co-occur, the effect on CHD is synergistic.
• The effects of anger are less clear; its impact may be independent of or dependent on other forms of psychologic distress.
• Distress in general appears to serve as a signal that something is wrong and needs to be addressed. If ignored, it may become chronic and unremitting; because symptoms of distress may lead to systemic dysregulation and increased CHD risk, they may indicate the need for increased surveillance and intervention.

WHY FOCUS ON THE BIOLOGY OF EMOTIONS?

A clear biologic explanation for the influence of emotional factors on physical health would serve to assuage skeptics who doubt that such a link exists or who attribute a common underlying genetic trait to both negative affect and heart disease. Further, focusing on the biology may help answer key questions with respect to emotions and disease processes: What is the damage incurred by negative emotional states and is it reversible? Can compensatory pathways be activated to bypass the mechanisms causing damage or slow the progression of disease?

Cardiac response to worry and stress

In one study attempting to shed light on relevant emotion-related biologic process, the prolonged physiologic effects of worry were examined. Worry episodes and stressful events were recorded hourly along with ambulatory heart rate and heart rate variability in 73 teachers for 4 days.⁴ Autonomic activity, as reflected by a concurrent elevation in heart rate and a decrease in heart rate variability, was increased up to 2 hours after a worry episode. The findings also suggested that the prolonged cardiac effects of separate worry episodes were independent.

Another study sought to determine whether heightened reactivity or delayed recovery to acute stress increases risk of cardiovascular disease.⁵ This meta-analysis included 36 studies to assess whether acute cardiovascular response to various laboratory stressors (ie, cognitive tasks, stress interviews, public speaking). Findings indicated that heightened cardiovascular reactivity was associated with worse cardiovascular outcomes, such as incident hypertension, coronary calcification, carotid intima-media thickness, and cardiovascular events over time.

Role of aldosterone overlooked

Reprinted from Neuroscience and Biobehavioral Reviews (Kubzansky LD, et al. Aldosterone: a forgotten mediator of the relationship between psychological stress and heart disease. Neurosci Biobehav Rev 2010; 34:80–86), © 2010, with permission from Elsevier.

WHY CONSIDER RESILIENCE?

Because the absence of a deficit is not the same as the presence of an asset, greater insight into dysfunction may be gained by explicitly considering what promotes healthy functioning. Ameliorating distress has proven difficult; so, in studying resilience (including the ability to regulate affect), new targets for prevention and intervention may be identified. Although no meta-analysis of resilience factors has been published to date owing to the paucity of data, the studies that have been performed are generally rigorous and have demonstrated consistent findings.

For example, one prospective, well-controlled study of 1,739 men and women demonstrated a protective effect of positive affect (as ascertained by structured interview) against 10-year incident CHD.⁶ The risk of fatal or nonfatal ischemic heart disease events was reduced by 22% (P = .02) for each 1-point increase in positive affect, even after controlling for depression and negative emotions.

Reprinted, with permission, from Archives of General Psychiatry (Kubzansky LD, et al. Arch Gen Psychiatry 2011; 68:400–408), Copyright © 2011 American Medical Association. All rights reserved.

Biology of resilience: Counteracting cellular damage

Genomic changes can be induced by the relaxation response, as evidenced by the differential gene expression profiles of long-term daily practitioners of relaxation (ie, meditation, yoga), short-term (8-week) practitioners of relaxation, and healthy controls.⁸ Alterations in cellular metabolism, oxidative phosphorylation, and generation of reactive oxygen species that counteract proinflammatory responses, indicative of an adaptive response, were observed in both groups that practiced relaxation.

FUTURE DIRECTIONS

Whether and how the sources and effects of psychosocial stress and response to treatment differ across men and women deserves closer examination. A review by Low et al⁹ summarizes the current state of knowledge with respect to psychosocial factors and heart disease in women, noting that the sources of stress associated with increased CHD risk differ across men and women; psychosocial risk factors like depression and anxiety appear to increase risk for both men and women; work-related stress has larger effects in men while stress related to relationships and family responsibilities appear to have larger effects in women.

Although responses to psychosocial stress are not clearly different between men and women, intervention targeted at reducing distress is much less effective in reducing the risk of adverse events in women versus men. The mechanism to explain this difference in effectiveness of intervention urgently requires further exploration.

In conducting this work, several factors are important. The best time to intervene to reduce psychosocial distress is unknown; a key consideration will be, what is the best etiologic window for intervention? Perhaps a life-course approach that targets individuals with chronically high levels of emotional distress who also have multiple coronary risk factors, and that enhances their capacity to regulate emotions would prove superior to waiting until late in the disease process.

Another area that may prove fruitful is to consider in more depth the biology of the placebo effect and whether and how it may inform our understanding of resilience.

More generally, considering why interventions seem to influence outcomes so differently across men and women, applying a life course approach to determine the best etiologic window for prevention and intervention strategies, and conducting a more in-depth exploration of the biology of resilience may lead to improved capacity for population-based approaches to reducing the burden of CHD.

The effect of emotion on the heart is not confined to depression, but extends to a variety of mental states; as William Harvey described in 1628, “A mental disturbance provoking pain, excessive joy, hope or anxiety extends to the heart, where it affects its temper and rate, impairing general nutrition and vigor.”

In going beyond the well-established role of depression as a risk factor for heart disease, 2010 delivered several important publications recognizing anxiety, anger, and other forms of distress as key factors in the etiology of coronary heart disease (CHD). Other papers of merit elucidated new and overlooked insights into the pathways linking psychosocial stress and cardiovascular risk, and also considered psychologic states that appear to promote healthy functioning.

IMPACT OF NEGATIVE EMOTIONS ON RISK OF INCIDENT CORONARY HEART DISEASE

In a meta-analysis of 20 prospective studies that included 249,846 persons with a mean follow-up of 11.2 years, Roest et al¹ examined the impact of anxiety characterized by the presence of anxiety symptoms or a diagnosis of anxiety disorder on incident CHD. Most of the studies adjusted for a broad array of relevant potential confounders. Findings suggest the presence of anxiety increases the risk of incident CHD by 26% (P P = .003).

In a meta-analysis of 25 prospective studies of 7,160 persons with a mean follow-up exceeding 10 years, Chida and Steptoe² found that anger increased the risk of incident CHD by 19%, after adjustment for standard coronary risk factors. The effect was less stable than that associated with anxiety and depression, and when stratified by gender, the harmful effects of anger were more evident in men than in women. The effect of anger was attenuated when controlling for behavioral covariates. The association between anger and CHD did not hold for all ways of measuring anger, which suggests that the type of anger or the ability to regulate anger may be relevant to the relationship.

A study that did account for the type of anger expression on the risk of incident CHD was conducted by Davidson and Mostofsky.³ The independent effect of three distinct types of anger expression (constructive anger, destructive anger justification, and destructive anger rumination) on 10-year incident CHD was examined, controlling for other psychosocial factors. In men, higher scores for constructive anger were associated with a lower rate of CHD; in both men and women, higher scores for destructive anger justification were associated with an increased risk of CHD.

Insights gained from these studies are as follows:

• The impact of anxiety appears to be comparable to depression, and the effects of anxiety and depression are largely independent.
• If anxiety and depression co-occur, the effect on CHD is synergistic.
• The effects of anger are less clear; its impact may be independent of or dependent on other forms of psychologic distress.
• Distress in general appears to serve as a signal that something is wrong and needs to be addressed. If ignored, it may become chronic and unremitting; because symptoms of distress may lead to systemic dysregulation and increased CHD risk, they may indicate the need for increased surveillance and intervention.

WHY FOCUS ON THE BIOLOGY OF EMOTIONS?

A clear biologic explanation for the influence of emotional factors on physical health would serve to assuage skeptics who doubt that such a link exists or who attribute a common underlying genetic trait to both negative affect and heart disease. Further, focusing on the biology may help answer key questions with respect to emotions and disease processes: What is the damage incurred by negative emotional states and is it reversible? Can compensatory pathways be activated to bypass the mechanisms causing damage or slow the progression of disease?

Cardiac response to worry and stress

In one study attempting to shed light on relevant emotion-related biologic process, the prolonged physiologic effects of worry were examined. Worry episodes and stressful events were recorded hourly along with ambulatory heart rate and heart rate variability in 73 teachers for 4 days.⁴ Autonomic activity, as reflected by a concurrent elevation in heart rate and a decrease in heart rate variability, was increased up to 2 hours after a worry episode. The findings also suggested that the prolonged cardiac effects of separate worry episodes were independent.

Another study sought to determine whether heightened reactivity or delayed recovery to acute stress increases risk of cardiovascular disease.⁵ This meta-analysis included 36 studies to assess whether acute cardiovascular response to various laboratory stressors (ie, cognitive tasks, stress interviews, public speaking). Findings indicated that heightened cardiovascular reactivity was associated with worse cardiovascular outcomes, such as incident hypertension, coronary calcification, carotid intima-media thickness, and cardiovascular events over time.

Role of aldosterone overlooked

Reprinted from Neuroscience and Biobehavioral Reviews (Kubzansky LD, et al. Aldosterone: a forgotten mediator of the relationship between psychological stress and heart disease. Neurosci Biobehav Rev 2010; 34:80–86), © 2010, with permission from Elsevier.

WHY CONSIDER RESILIENCE?

Because the absence of a deficit is not the same as the presence of an asset, greater insight into dysfunction may be gained by explicitly considering what promotes healthy functioning. Ameliorating distress has proven difficult; so, in studying resilience (including the ability to regulate affect), new targets for prevention and intervention may be identified. Although no meta-analysis of resilience factors has been published to date owing to the paucity of data, the studies that have been performed are generally rigorous and have demonstrated consistent findings.

For example, one prospective, well-controlled study of 1,739 men and women demonstrated a protective effect of positive affect (as ascertained by structured interview) against 10-year incident CHD.⁶ The risk of fatal or nonfatal ischemic heart disease events was reduced by 22% (P = .02) for each 1-point increase in positive affect, even after controlling for depression and negative emotions.

Reprinted, with permission, from Archives of General Psychiatry (Kubzansky LD, et al. Arch Gen Psychiatry 2011; 68:400–408), Copyright © 2011 American Medical Association. All rights reserved.

Biology of resilience: Counteracting cellular damage

Genomic changes can be induced by the relaxation response, as evidenced by the differential gene expression profiles of long-term daily practitioners of relaxation (ie, meditation, yoga), short-term (8-week) practitioners of relaxation, and healthy controls.⁸ Alterations in cellular metabolism, oxidative phosphorylation, and generation of reactive oxygen species that counteract proinflammatory responses, indicative of an adaptive response, were observed in both groups that practiced relaxation.

FUTURE DIRECTIONS

Whether and how the sources and effects of psychosocial stress and response to treatment differ across men and women deserves closer examination. A review by Low et al⁹ summarizes the current state of knowledge with respect to psychosocial factors and heart disease in women, noting that the sources of stress associated with increased CHD risk differ across men and women; psychosocial risk factors like depression and anxiety appear to increase risk for both men and women; work-related stress has larger effects in men while stress related to relationships and family responsibilities appear to have larger effects in women.

Although responses to psychosocial stress are not clearly different between men and women, intervention targeted at reducing distress is much less effective in reducing the risk of adverse events in women versus men. The mechanism to explain this difference in effectiveness of intervention urgently requires further exploration.

In conducting this work, several factors are important. The best time to intervene to reduce psychosocial distress is unknown; a key consideration will be, what is the best etiologic window for intervention? Perhaps a life-course approach that targets individuals with chronically high levels of emotional distress who also have multiple coronary risk factors, and that enhances their capacity to regulate emotions would prove superior to waiting until late in the disease process.

Another area that may prove fruitful is to consider in more depth the biology of the placebo effect and whether and how it may inform our understanding of resilience.

More generally, considering why interventions seem to influence outcomes so differently across men and women, applying a life course approach to determine the best etiologic window for prevention and intervention strategies, and conducting a more in-depth exploration of the biology of resilience may lead to improved capacity for population-based approaches to reducing the burden of CHD.