Anthony J. Viera, MD, MPH

The dream of family practice began more than 6 decades ago with a movement toward personal physicians who have “… the feeling of warm personal regard and concern of doctor for patient, the feeling that the doctor treats people, not illnesses ….” The personal family physician helps patients “… not because of the interesting medical problems they may present but because they are human beings in need of help.”¹ One of the most influential founders of family medicine, Dr. Gayle Stephens, expounded on this idea in a series of essays that tapped into the intellectual, philosophical, historical, and moral underpinnings of our discipline.²

Following the dream and the birth of family medicine—like any organization—its lifecycle can be envisioned as proceeding through the rest of the 7 stages of organizational life (TABLE).³ Now allow me to give you some numbers. There are more than 118,000 family physicians in the United States, 784 family medicine residencies filled by 4530 medical school graduates, more than 150 departments of family medicine, multiple national family medicine organizations, and even a World Organization of Family Doctors.^4,5 The American Board of Family Medicine is the second largest medical specialty board in the country. Family doctors make up nearly 40% of our total primary care workforce.⁶ We launched the venture, got organized, made it. We are an institution.

This final issue of The Journal of Family Practice marks the end of an era of nearly 50 years of publication.

The threat at the institution stage is that we are on the precipice of “closing in.” Many factors are driving this stage: commoditization in health care, market influences and competition for patients, alternative primary care models, erosion of the patient-physician relationship (partly driven by technology), narrowing scope of care, clinician burnout, and the challenges of implementing value-based care, to name a few. You see what comes next in the TABLE.³ The good news is that there is an alternative to the “natural” progression to the ending stage: the path of renewal.³

In the lifecycle of an organization, the path of renewal starts the cycle anew, with dreaming the dream. I recently had the opportunity to visit Singapore to learn about their health system. Singapore is one of the wealthiest countries in the world. I was impressed with their many innovations, including technological ones, as well as new models of care. However, I was most impressed that the country is betting big on family medicine. Their Ministry of Health has launched an initiative they are calling Healthier SG.⁷ The goal is for “all Singaporeans to have a trusted and lifelong relationship with [their] family doctor.” Their dream is to bring personal doctoring to everyone in the country to make Singapore healthier.

While their path of renewal is occurring halfway around the world, here at home, our path of renewal has been ignited over the past several years by the work of the Robert Graham Center; the Keystone Conferences; the American Board of Family Medicine; and the National Academies of Science, Engineering, and Medicine, among others.^8-11 These organizations are aligning around re-centering on ­patient-clinician relationships, measuring what is important, care by interprofessional teams, payment reform, professionalism, health equity, improved information technology, and adherence to the best available evidence. We are working toward the solution shop as opposed to the production line.¹² We are indeed dreaming a new dream.

While I write about this renewal, I close with an ending. This is the final issue of The Journal of Family Practice. It marks the end of an era of nearly 50 years of publication. The Journal of Family Practice has left a lasting mark, providing generations of clinicians with evidence-based, practical guidance to help care for patients as well as serving as an important venue for scholarly work by the family medicine community. Although I have had the privilege of serving the discipline as an editor-in-chief for only a brief time, I am grateful I had the opportunity. Most of all, I appreciate being on the journey of family medicine with you, renewing the dream together.

The references for this Editorial are available in the online version of the article at www.mdedge.com/familymedicine.

The dream of family practice began more than 6 decades ago with a movement toward personal physicians who have “… the feeling of warm personal regard and concern of doctor for patient, the feeling that the doctor treats people, not illnesses ….” The personal family physician helps patients “… not because of the interesting medical problems they may present but because they are human beings in need of help.”¹ One of the most influential founders of family medicine, Dr. Gayle Stephens, expounded on this idea in a series of essays that tapped into the intellectual, philosophical, historical, and moral underpinnings of our discipline.²

Following the dream and the birth of family medicine—like any organization—its lifecycle can be envisioned as proceeding through the rest of the 7 stages of organizational life (TABLE).³ Now allow me to give you some numbers. There are more than 118,000 family physicians in the United States, 784 family medicine residencies filled by 4530 medical school graduates, more than 150 departments of family medicine, multiple national family medicine organizations, and even a World Organization of Family Doctors.^4,5 The American Board of Family Medicine is the second largest medical specialty board in the country. Family doctors make up nearly 40% of our total primary care workforce.⁶ We launched the venture, got organized, made it. We are an institution.

This final issue of The Journal of Family Practice marks the end of an era of nearly 50 years of publication.

The threat at the institution stage is that we are on the precipice of “closing in.” Many factors are driving this stage: commoditization in health care, market influences and competition for patients, alternative primary care models, erosion of the patient-physician relationship (partly driven by technology), narrowing scope of care, clinician burnout, and the challenges of implementing value-based care, to name a few. You see what comes next in the TABLE.³ The good news is that there is an alternative to the “natural” progression to the ending stage: the path of renewal.³

In the lifecycle of an organization, the path of renewal starts the cycle anew, with dreaming the dream. I recently had the opportunity to visit Singapore to learn about their health system. Singapore is one of the wealthiest countries in the world. I was impressed with their many innovations, including technological ones, as well as new models of care. However, I was most impressed that the country is betting big on family medicine. Their Ministry of Health has launched an initiative they are calling Healthier SG.⁷ The goal is for “all Singaporeans to have a trusted and lifelong relationship with [their] family doctor.” Their dream is to bring personal doctoring to everyone in the country to make Singapore healthier.

While their path of renewal is occurring halfway around the world, here at home, our path of renewal has been ignited over the past several years by the work of the Robert Graham Center; the Keystone Conferences; the American Board of Family Medicine; and the National Academies of Science, Engineering, and Medicine, among others.^8-11 These organizations are aligning around re-centering on ­patient-clinician relationships, measuring what is important, care by interprofessional teams, payment reform, professionalism, health equity, improved information technology, and adherence to the best available evidence. We are working toward the solution shop as opposed to the production line.¹² We are indeed dreaming a new dream.

While I write about this renewal, I close with an ending. This is the final issue of The Journal of Family Practice. It marks the end of an era of nearly 50 years of publication. The Journal of Family Practice has left a lasting mark, providing generations of clinicians with evidence-based, practical guidance to help care for patients as well as serving as an important venue for scholarly work by the family medicine community. Although I have had the privilege of serving the discipline as an editor-in-chief for only a brief time, I am grateful I had the opportunity. Most of all, I appreciate being on the journey of family medicine with you, renewing the dream together.

The references for this Editorial are available in the online version of the article at www.mdedge.com/familymedicine.

The dream of family practice began more than 6 decades ago with a movement toward personal physicians who have “… the feeling of warm personal regard and concern of doctor for patient, the feeling that the doctor treats people, not illnesses ….” The personal family physician helps patients “… not because of the interesting medical problems they may present but because they are human beings in need of help.”¹ One of the most influential founders of family medicine, Dr. Gayle Stephens, expounded on this idea in a series of essays that tapped into the intellectual, philosophical, historical, and moral underpinnings of our discipline.²

Following the dream and the birth of family medicine—like any organization—its lifecycle can be envisioned as proceeding through the rest of the 7 stages of organizational life (TABLE).³ Now allow me to give you some numbers. There are more than 118,000 family physicians in the United States, 784 family medicine residencies filled by 4530 medical school graduates, more than 150 departments of family medicine, multiple national family medicine organizations, and even a World Organization of Family Doctors.^4,5 The American Board of Family Medicine is the second largest medical specialty board in the country. Family doctors make up nearly 40% of our total primary care workforce.⁶ We launched the venture, got organized, made it. We are an institution.

This final issue of The Journal of Family Practice marks the end of an era of nearly 50 years of publication.

The threat at the institution stage is that we are on the precipice of “closing in.” Many factors are driving this stage: commoditization in health care, market influences and competition for patients, alternative primary care models, erosion of the patient-physician relationship (partly driven by technology), narrowing scope of care, clinician burnout, and the challenges of implementing value-based care, to name a few. You see what comes next in the TABLE.³ The good news is that there is an alternative to the “natural” progression to the ending stage: the path of renewal.³

In the lifecycle of an organization, the path of renewal starts the cycle anew, with dreaming the dream. I recently had the opportunity to visit Singapore to learn about their health system. Singapore is one of the wealthiest countries in the world. I was impressed with their many innovations, including technological ones, as well as new models of care. However, I was most impressed that the country is betting big on family medicine. Their Ministry of Health has launched an initiative they are calling Healthier SG.⁷ The goal is for “all Singaporeans to have a trusted and lifelong relationship with [their] family doctor.” Their dream is to bring personal doctoring to everyone in the country to make Singapore healthier.

While their path of renewal is occurring halfway around the world, here at home, our path of renewal has been ignited over the past several years by the work of the Robert Graham Center; the Keystone Conferences; the American Board of Family Medicine; and the National Academies of Science, Engineering, and Medicine, among others.^8-11 These organizations are aligning around re-centering on ­patient-clinician relationships, measuring what is important, care by interprofessional teams, payment reform, professionalism, health equity, improved information technology, and adherence to the best available evidence. We are working toward the solution shop as opposed to the production line.¹² We are indeed dreaming a new dream.

While I write about this renewal, I close with an ending. This is the final issue of The Journal of Family Practice. It marks the end of an era of nearly 50 years of publication. The Journal of Family Practice has left a lasting mark, providing generations of clinicians with evidence-based, practical guidance to help care for patients as well as serving as an important venue for scholarly work by the family medicine community. Although I have had the privilege of serving the discipline as an editor-in-chief for only a brief time, I am grateful I had the opportunity. Most of all, I appreciate being on the journey of family medicine with you, renewing the dream together.

The references for this Editorial are available in the online version of the article at www.mdedge.com/familymedicine.

The World Health Organization (WHO) recently released its inaugural report on the devastating global effects of hypertension, including recommendations for combatting this “silent killer.”¹ Notable in the 276-page report is the emphasis on improving access to antihypertensive medications, in part through team-based care and simple evidence-based protocols. This strategy is not surprising given that in clinical medicine we focus on the “high-risk” strategy for prevention­—ie, identify people at increased risk for an adverse health outcome (in this case, cardiovascular disease events) and offer them medication to reduce that risk.²

Should we replace even a small amount of the sodium in processed foods with potassium?

As part of the high-risk strategy, we also counsel at the individual level about lifestyle modifications—but unfortunately, we tend not to get very far. Given the substantial evidence demonstrating its benefits, a low-sodium DASH (Dietary Approaches to Stop Hypertension) eating plan is one of the lifestyle recommendations we make for our patients with hypertension.^3,4 The DASH part of the diet involves getting our patients to eat more fruits, vegetables, and whole grains and limit sugar and saturated fats. To achieve the low-sodium part, we might counsel against added table salt, but mostly we discourage consumption of canned and other foods that are commercially processed, packaged, and prepared, because that’s the source of more than 70% of our sodium intake.⁵ It’s not difficult to understand why real-world uptake of the low-sodium DASH eating plan is low.⁶

This issue of The Journal of Family Practice features a PURL that supports a much more prominent role for salt substitutes in our counseling recommendations.⁷ Potassium­-enriched salt substitutes not only lower blood pressure (BP) but also reduce the risk for cardiovascular events and death.⁸ They are widely available, and while more expensive per ounce than regular salt (sodium chloride), are still affordable.

Still, encouraging salt substitution with one patient at a time is relying on the high-risk strategy, with its inherently limited potential.² An alternative is the population strategy. For hypertension, that would mean doing something for the entire population that would lead to a downward shift in the distribution of BP.² The shift does not have to be large. We’ve known for more than 3 decades that just a 2–mm Hg reduction in the population’s average systolic BP would reduce stroke mortality by about 6%, coronary heart disease mortality by 4%, and total mortality by 3%.⁹ A 5–mm Hg reduction more than doubles those benefits. We are talking about tens of thousands fewer patients with heart disease and stroke each year and billions of dollars in health care cost savings.

Reducing our nation’s sodium intake, a quintessential population approach, has proven difficult. Our average daily sodium intake is about 3600 mg.¹⁰ Guidance on sodium reduction from the US Food and Drug Administration (targeted to industry) has aimed to reduce Americans’ average sodium intake to 3000 mg/d over the short term, fully acknowledging that the recommended sodium limit is 2300 mg/d.¹¹ We’ve got a long way to go.

Might salt substitution at the population level be a way to simultaneously reduce our sodium intake and increase our potassium intake?¹² The closest I found to a population­wide substitution study was a cluster randomized trial conducted in 6 villages in Peru.¹³ In a stepped-wedge design, households had 25% of their regular salt replaced with potassium salt. Small shops, bakeries, community kitchens, and food vendors also had salt replacement. The intention-to-treat analysis showed a small reduction in systolic BP (1.3 mm Hg) among those with hypertension at baseline (n = 428) and a 51% reduced incidence of developing hypertension among the other 1891 participants over the 4673 ­person-years of follow-up.

I found this study interesting and its results compelling, leading me to wonder: In the United States, where most of our sodium comes from the food industry, should we replace even a small amount of the sodium in processed foods with potassium? We’re not getting there with DASH alone. 

The World Health Organization (WHO) recently released its inaugural report on the devastating global effects of hypertension, including recommendations for combatting this “silent killer.”¹ Notable in the 276-page report is the emphasis on improving access to antihypertensive medications, in part through team-based care and simple evidence-based protocols. This strategy is not surprising given that in clinical medicine we focus on the “high-risk” strategy for prevention­—ie, identify people at increased risk for an adverse health outcome (in this case, cardiovascular disease events) and offer them medication to reduce that risk.²

Should we replace even a small amount of the sodium in processed foods with potassium?

As part of the high-risk strategy, we also counsel at the individual level about lifestyle modifications—but unfortunately, we tend not to get very far. Given the substantial evidence demonstrating its benefits, a low-sodium DASH (Dietary Approaches to Stop Hypertension) eating plan is one of the lifestyle recommendations we make for our patients with hypertension.^3,4 The DASH part of the diet involves getting our patients to eat more fruits, vegetables, and whole grains and limit sugar and saturated fats. To achieve the low-sodium part, we might counsel against added table salt, but mostly we discourage consumption of canned and other foods that are commercially processed, packaged, and prepared, because that’s the source of more than 70% of our sodium intake.⁵ It’s not difficult to understand why real-world uptake of the low-sodium DASH eating plan is low.⁶

This issue of The Journal of Family Practice features a PURL that supports a much more prominent role for salt substitutes in our counseling recommendations.⁷ Potassium­-enriched salt substitutes not only lower blood pressure (BP) but also reduce the risk for cardiovascular events and death.⁸ They are widely available, and while more expensive per ounce than regular salt (sodium chloride), are still affordable.

Still, encouraging salt substitution with one patient at a time is relying on the high-risk strategy, with its inherently limited potential.² An alternative is the population strategy. For hypertension, that would mean doing something for the entire population that would lead to a downward shift in the distribution of BP.² The shift does not have to be large. We’ve known for more than 3 decades that just a 2–mm Hg reduction in the population’s average systolic BP would reduce stroke mortality by about 6%, coronary heart disease mortality by 4%, and total mortality by 3%.⁹ A 5–mm Hg reduction more than doubles those benefits. We are talking about tens of thousands fewer patients with heart disease and stroke each year and billions of dollars in health care cost savings.

Reducing our nation’s sodium intake, a quintessential population approach, has proven difficult. Our average daily sodium intake is about 3600 mg.¹⁰ Guidance on sodium reduction from the US Food and Drug Administration (targeted to industry) has aimed to reduce Americans’ average sodium intake to 3000 mg/d over the short term, fully acknowledging that the recommended sodium limit is 2300 mg/d.¹¹ We’ve got a long way to go.

Might salt substitution at the population level be a way to simultaneously reduce our sodium intake and increase our potassium intake?¹² The closest I found to a population­wide substitution study was a cluster randomized trial conducted in 6 villages in Peru.¹³ In a stepped-wedge design, households had 25% of their regular salt replaced with potassium salt. Small shops, bakeries, community kitchens, and food vendors also had salt replacement. The intention-to-treat analysis showed a small reduction in systolic BP (1.3 mm Hg) among those with hypertension at baseline (n = 428) and a 51% reduced incidence of developing hypertension among the other 1891 participants over the 4673 ­person-years of follow-up.

I found this study interesting and its results compelling, leading me to wonder: In the United States, where most of our sodium comes from the food industry, should we replace even a small amount of the sodium in processed foods with potassium? We’re not getting there with DASH alone. 

The World Health Organization (WHO) recently released its inaugural report on the devastating global effects of hypertension, including recommendations for combatting this “silent killer.”¹ Notable in the 276-page report is the emphasis on improving access to antihypertensive medications, in part through team-based care and simple evidence-based protocols. This strategy is not surprising given that in clinical medicine we focus on the “high-risk” strategy for prevention­—ie, identify people at increased risk for an adverse health outcome (in this case, cardiovascular disease events) and offer them medication to reduce that risk.²

Should we replace even a small amount of the sodium in processed foods with potassium?

As part of the high-risk strategy, we also counsel at the individual level about lifestyle modifications—but unfortunately, we tend not to get very far. Given the substantial evidence demonstrating its benefits, a low-sodium DASH (Dietary Approaches to Stop Hypertension) eating plan is one of the lifestyle recommendations we make for our patients with hypertension.^3,4 The DASH part of the diet involves getting our patients to eat more fruits, vegetables, and whole grains and limit sugar and saturated fats. To achieve the low-sodium part, we might counsel against added table salt, but mostly we discourage consumption of canned and other foods that are commercially processed, packaged, and prepared, because that’s the source of more than 70% of our sodium intake.⁵ It’s not difficult to understand why real-world uptake of the low-sodium DASH eating plan is low.⁶

This issue of The Journal of Family Practice features a PURL that supports a much more prominent role for salt substitutes in our counseling recommendations.⁷ Potassium­-enriched salt substitutes not only lower blood pressure (BP) but also reduce the risk for cardiovascular events and death.⁸ They are widely available, and while more expensive per ounce than regular salt (sodium chloride), are still affordable.

Still, encouraging salt substitution with one patient at a time is relying on the high-risk strategy, with its inherently limited potential.² An alternative is the population strategy. For hypertension, that would mean doing something for the entire population that would lead to a downward shift in the distribution of BP.² The shift does not have to be large. We’ve known for more than 3 decades that just a 2–mm Hg reduction in the population’s average systolic BP would reduce stroke mortality by about 6%, coronary heart disease mortality by 4%, and total mortality by 3%.⁹ A 5–mm Hg reduction more than doubles those benefits. We are talking about tens of thousands fewer patients with heart disease and stroke each year and billions of dollars in health care cost savings.

Reducing our nation’s sodium intake, a quintessential population approach, has proven difficult. Our average daily sodium intake is about 3600 mg.¹⁰ Guidance on sodium reduction from the US Food and Drug Administration (targeted to industry) has aimed to reduce Americans’ average sodium intake to 3000 mg/d over the short term, fully acknowledging that the recommended sodium limit is 2300 mg/d.¹¹ We’ve got a long way to go.

Might salt substitution at the population level be a way to simultaneously reduce our sodium intake and increase our potassium intake?¹² The closest I found to a population­wide substitution study was a cluster randomized trial conducted in 6 villages in Peru.¹³ In a stepped-wedge design, households had 25% of their regular salt replaced with potassium salt. Small shops, bakeries, community kitchens, and food vendors also had salt replacement. The intention-to-treat analysis showed a small reduction in systolic BP (1.3 mm Hg) among those with hypertension at baseline (n = 428) and a 51% reduced incidence of developing hypertension among the other 1891 participants over the 4673 ­person-years of follow-up.

I found this study interesting and its results compelling, leading me to wonder: In the United States, where most of our sodium comes from the food industry, should we replace even a small amount of the sodium in processed foods with potassium? We’re not getting there with DASH alone. 

Changing behavior is hard. And at nearly every clinical encounter, we counsel/encourage/remind/help (choose a verb) our patients to make a change—to do something hard. We tell them they need to increase their physical activity, get more sleep, or alter their eating habits. We know that if they make the needed changes, they can improve their health and possibly lengthen their lives. But we also know (from the systematic reviews the US Preventive Services Task Force [USPSTF] uses to make its recommendations) that brief counseling in our offices is largely ineffective unless we connect patients to resources to support the recommended change.

As examples, the USPSTF currently recommends the following (both grade “B”):

• offer or refer adults with cardiovascular disease risk factors to behavioral counseling interventions to promote a healthy diet and physical activity.¹
• offer or refer adults with a body mass index of 30 or higher to intensive, multicomponent behavioral interventions.²

This 2-step rule is tech-free and can be applied by patients in a few seconds to make healthier food choices.

To support our patients when making recommendations such as these, we might refer them to a dietitian for intensive counseling and meal-planning guidance. The American Diabetes Association says that patients seeking to manage their diabetes and prediabetes “can start by working with a registered dietitian nutritionist … to make an eating plan that works for [them].”³ However, this kind of resource is unavailable to many of our patients.

So what else can we do?

We can help patients decide what to buy in the grocery aisle. Nutrition labels are useful, but they are limited by their complexity and requisite level of health literacy.⁴ Even the concept of “calories” is not so intuitive. This challenge with interpreting calories led me (in some of my prior work) to explore a potentially more useful approach: conveying calorie information as physical activity equivalents.⁵

In this issue of The Journal of Family Practice, Dong and colleagues present their findings on whether a simple equation (the Altman Rule) that uses information on nutrition labels may be a reasonable proxy for an even more difficult concept—­glycemic load.⁶ The idea is that consumers (eg, patients with diabetes) can use this rule to help them in their decision-making at the grocery store (or the convenience store or gas station, for that matter, where the high-glycemic-load carbohydrates may be even more tempting). The 2-step rule is tech-free and can be applied in a few seconds. Their research demonstrated that the rule is a reasonable proxy for glycemic load for packaged carbohydrates (eg, chips, cereals, crackers, granola bars). Caveats acknowledged, foods that meet the rule are likely to be healthier choices.

Looking ahead, I would like to see whether counseling patients about the Altman Rule leads to their use of it, and how that translates into healthier eating, lower A1C, and ideally better health. For now, the Altman Rule is worth learning about. It may serve as another tool that you can use to support your patients when you ask them to do the hard work of making healthier food choices. 

Changing behavior is hard. And at nearly every clinical encounter, we counsel/encourage/remind/help (choose a verb) our patients to make a change—to do something hard. We tell them they need to increase their physical activity, get more sleep, or alter their eating habits. We know that if they make the needed changes, they can improve their health and possibly lengthen their lives. But we also know (from the systematic reviews the US Preventive Services Task Force [USPSTF] uses to make its recommendations) that brief counseling in our offices is largely ineffective unless we connect patients to resources to support the recommended change.

As examples, the USPSTF currently recommends the following (both grade “B”):

• offer or refer adults with cardiovascular disease risk factors to behavioral counseling interventions to promote a healthy diet and physical activity.¹
• offer or refer adults with a body mass index of 30 or higher to intensive, multicomponent behavioral interventions.²

This 2-step rule is tech-free and can be applied by patients in a few seconds to make healthier food choices.

To support our patients when making recommendations such as these, we might refer them to a dietitian for intensive counseling and meal-planning guidance. The American Diabetes Association says that patients seeking to manage their diabetes and prediabetes “can start by working with a registered dietitian nutritionist … to make an eating plan that works for [them].”³ However, this kind of resource is unavailable to many of our patients.

So what else can we do?

We can help patients decide what to buy in the grocery aisle. Nutrition labels are useful, but they are limited by their complexity and requisite level of health literacy.⁴ Even the concept of “calories” is not so intuitive. This challenge with interpreting calories led me (in some of my prior work) to explore a potentially more useful approach: conveying calorie information as physical activity equivalents.⁵

In this issue of The Journal of Family Practice, Dong and colleagues present their findings on whether a simple equation (the Altman Rule) that uses information on nutrition labels may be a reasonable proxy for an even more difficult concept—­glycemic load.⁶ The idea is that consumers (eg, patients with diabetes) can use this rule to help them in their decision-making at the grocery store (or the convenience store or gas station, for that matter, where the high-glycemic-load carbohydrates may be even more tempting). The 2-step rule is tech-free and can be applied in a few seconds. Their research demonstrated that the rule is a reasonable proxy for glycemic load for packaged carbohydrates (eg, chips, cereals, crackers, granola bars). Caveats acknowledged, foods that meet the rule are likely to be healthier choices.

Looking ahead, I would like to see whether counseling patients about the Altman Rule leads to their use of it, and how that translates into healthier eating, lower A1C, and ideally better health. For now, the Altman Rule is worth learning about. It may serve as another tool that you can use to support your patients when you ask them to do the hard work of making healthier food choices. 

Changing behavior is hard. And at nearly every clinical encounter, we counsel/encourage/remind/help (choose a verb) our patients to make a change—to do something hard. We tell them they need to increase their physical activity, get more sleep, or alter their eating habits. We know that if they make the needed changes, they can improve their health and possibly lengthen their lives. But we also know (from the systematic reviews the US Preventive Services Task Force [USPSTF] uses to make its recommendations) that brief counseling in our offices is largely ineffective unless we connect patients to resources to support the recommended change.

As examples, the USPSTF currently recommends the following (both grade “B”):

• offer or refer adults with cardiovascular disease risk factors to behavioral counseling interventions to promote a healthy diet and physical activity.¹
• offer or refer adults with a body mass index of 30 or higher to intensive, multicomponent behavioral interventions.²

This 2-step rule is tech-free and can be applied by patients in a few seconds to make healthier food choices.

To support our patients when making recommendations such as these, we might refer them to a dietitian for intensive counseling and meal-planning guidance. The American Diabetes Association says that patients seeking to manage their diabetes and prediabetes “can start by working with a registered dietitian nutritionist … to make an eating plan that works for [them].”³ However, this kind of resource is unavailable to many of our patients.

So what else can we do?

We can help patients decide what to buy in the grocery aisle. Nutrition labels are useful, but they are limited by their complexity and requisite level of health literacy.⁴ Even the concept of “calories” is not so intuitive. This challenge with interpreting calories led me (in some of my prior work) to explore a potentially more useful approach: conveying calorie information as physical activity equivalents.⁵

In this issue of The Journal of Family Practice, Dong and colleagues present their findings on whether a simple equation (the Altman Rule) that uses information on nutrition labels may be a reasonable proxy for an even more difficult concept—­glycemic load.⁶ The idea is that consumers (eg, patients with diabetes) can use this rule to help them in their decision-making at the grocery store (or the convenience store or gas station, for that matter, where the high-glycemic-load carbohydrates may be even more tempting). The 2-step rule is tech-free and can be applied in a few seconds. Their research demonstrated that the rule is a reasonable proxy for glycemic load for packaged carbohydrates (eg, chips, cereals, crackers, granola bars). Caveats acknowledged, foods that meet the rule are likely to be healthier choices.

Looking ahead, I would like to see whether counseling patients about the Altman Rule leads to their use of it, and how that translates into healthier eating, lower A1C, and ideally better health. For now, the Altman Rule is worth learning about. It may serve as another tool that you can use to support your patients when you ask them to do the hard work of making healthier food choices. 

Whenever someone asks me what I do, I happily reply, “I’m just a simple country doctor.” That is, in part, why I am honored to be granted the opportunity to serve as editor-in-chief of The Journal of Family Practice (JFP). As our late colleague Dr. John Hickner noted in his first JFP editorial, he and the 2 editors-in-chief before him (Drs. Jeff Susman and Mark Ebell) were also of the small-town family doc tradition.¹

My goal as this journal’s editorin-chief will be to continue its high academic standing while maintaining its utility for busy clinicians.

My small-town roots trace back to rural South Carolina. I am a first-generation college student and attended medical school on a Navy Health Professions Scholarship. After completing my residency training, I had the privilege of serving for 5 years in the Navy (2 of those years were overseas), where I practiced and taught full-scope family medicine. I saw patients of all ages, attended deliveries, and provided inpatient hospital care, as well as performed a full range of procedures and tests, including colposcopies, skin procedures, vasectomies, flexible sigmoidoscopies, and exercise treadmill testing.

Following military service and completion of a 2-year fellowship and Master of Public Health degree (while working nights at a rural emergency department), I began work at the University of North Carolina at Chapel Hill. I had the good fortune of spending the next 11 years as a faculty member there, where I advanced my research and teaching career. In 2017, I was named the Chair of Family Medicine and Community Health at Duke University School of Medicine, where I continue to have an active outpatient practice.

My experiences have shaped my belief that it is critical that family medicine maintain its presence (and advance its prominence) both in our communities and at our large academic medicine centers, championing service to rural areas, promoting health equity, and advocating for the importance of high-quality primary care delivery and training. No matter where we are, our work is valuable, and we make a difference. Like my predecessors, I have a love of evidence-based medicine. I also have a love of writing, which I can trace back to my days as an intern. I am excited to be able to apply what I have learned over the years to help maintain the rigor, practicality, and relevance of JFP while simultaneously helping to nurture new authors and peer reviewers.

My goal as this journal’s editor-in-chief will be to continue its high academic standing while maintaining its utility for busy clinicians. The provision of evidence-based clinical review articles that are succinct and practical, along with departments (eg, Photo Rounds, Behavioral Health Consult, Practice Alert, PURLs), will remain the journal’s major focus. Within this framework, I also want to share the best evidence and ideas on other aspects of practicing medicine, such as quality improvement, population health, and health equity. I’ll be looking to increase recruitment and mentorship of authors from diverse backgrounds, including those historically ­underrepresented in medicine.

I look forward to working with the editorial board, associate and assistant editors, and staff of JFP to serve the diverse interests and needs of our readers. To that end, we’ll be looking for your guidance. How else can JFP help you in your day-to-day practice? Please let us know your ideas. Drop us a line at [email protected].

Finally, please join me in thanking Drs. Henry Barry and Kate Rowland for all of their work this past year in keeping JFP going strong!

Whenever someone asks me what I do, I happily reply, “I’m just a simple country doctor.” That is, in part, why I am honored to be granted the opportunity to serve as editor-in-chief of The Journal of Family Practice (JFP). As our late colleague Dr. John Hickner noted in his first JFP editorial, he and the 2 editors-in-chief before him (Drs. Jeff Susman and Mark Ebell) were also of the small-town family doc tradition.¹

My goal as this journal’s editorin-chief will be to continue its high academic standing while maintaining its utility for busy clinicians.

My small-town roots trace back to rural South Carolina. I am a first-generation college student and attended medical school on a Navy Health Professions Scholarship. After completing my residency training, I had the privilege of serving for 5 years in the Navy (2 of those years were overseas), where I practiced and taught full-scope family medicine. I saw patients of all ages, attended deliveries, and provided inpatient hospital care, as well as performed a full range of procedures and tests, including colposcopies, skin procedures, vasectomies, flexible sigmoidoscopies, and exercise treadmill testing.

Following military service and completion of a 2-year fellowship and Master of Public Health degree (while working nights at a rural emergency department), I began work at the University of North Carolina at Chapel Hill. I had the good fortune of spending the next 11 years as a faculty member there, where I advanced my research and teaching career. In 2017, I was named the Chair of Family Medicine and Community Health at Duke University School of Medicine, where I continue to have an active outpatient practice.

My experiences have shaped my belief that it is critical that family medicine maintain its presence (and advance its prominence) both in our communities and at our large academic medicine centers, championing service to rural areas, promoting health equity, and advocating for the importance of high-quality primary care delivery and training. No matter where we are, our work is valuable, and we make a difference. Like my predecessors, I have a love of evidence-based medicine. I also have a love of writing, which I can trace back to my days as an intern. I am excited to be able to apply what I have learned over the years to help maintain the rigor, practicality, and relevance of JFP while simultaneously helping to nurture new authors and peer reviewers.

My goal as this journal’s editor-in-chief will be to continue its high academic standing while maintaining its utility for busy clinicians. The provision of evidence-based clinical review articles that are succinct and practical, along with departments (eg, Photo Rounds, Behavioral Health Consult, Practice Alert, PURLs), will remain the journal’s major focus. Within this framework, I also want to share the best evidence and ideas on other aspects of practicing medicine, such as quality improvement, population health, and health equity. I’ll be looking to increase recruitment and mentorship of authors from diverse backgrounds, including those historically ­underrepresented in medicine.

I look forward to working with the editorial board, associate and assistant editors, and staff of JFP to serve the diverse interests and needs of our readers. To that end, we’ll be looking for your guidance. How else can JFP help you in your day-to-day practice? Please let us know your ideas. Drop us a line at [email protected].

Finally, please join me in thanking Drs. Henry Barry and Kate Rowland for all of their work this past year in keeping JFP going strong!

Whenever someone asks me what I do, I happily reply, “I’m just a simple country doctor.” That is, in part, why I am honored to be granted the opportunity to serve as editor-in-chief of The Journal of Family Practice (JFP). As our late colleague Dr. John Hickner noted in his first JFP editorial, he and the 2 editors-in-chief before him (Drs. Jeff Susman and Mark Ebell) were also of the small-town family doc tradition.¹

My goal as this journal’s editorin-chief will be to continue its high academic standing while maintaining its utility for busy clinicians.

My small-town roots trace back to rural South Carolina. I am a first-generation college student and attended medical school on a Navy Health Professions Scholarship. After completing my residency training, I had the privilege of serving for 5 years in the Navy (2 of those years were overseas), where I practiced and taught full-scope family medicine. I saw patients of all ages, attended deliveries, and provided inpatient hospital care, as well as performed a full range of procedures and tests, including colposcopies, skin procedures, vasectomies, flexible sigmoidoscopies, and exercise treadmill testing.

Following military service and completion of a 2-year fellowship and Master of Public Health degree (while working nights at a rural emergency department), I began work at the University of North Carolina at Chapel Hill. I had the good fortune of spending the next 11 years as a faculty member there, where I advanced my research and teaching career. In 2017, I was named the Chair of Family Medicine and Community Health at Duke University School of Medicine, where I continue to have an active outpatient practice.

My experiences have shaped my belief that it is critical that family medicine maintain its presence (and advance its prominence) both in our communities and at our large academic medicine centers, championing service to rural areas, promoting health equity, and advocating for the importance of high-quality primary care delivery and training. No matter where we are, our work is valuable, and we make a difference. Like my predecessors, I have a love of evidence-based medicine. I also have a love of writing, which I can trace back to my days as an intern. I am excited to be able to apply what I have learned over the years to help maintain the rigor, practicality, and relevance of JFP while simultaneously helping to nurture new authors and peer reviewers.

My goal as this journal’s editor-in-chief will be to continue its high academic standing while maintaining its utility for busy clinicians. The provision of evidence-based clinical review articles that are succinct and practical, along with departments (eg, Photo Rounds, Behavioral Health Consult, Practice Alert, PURLs), will remain the journal’s major focus. Within this framework, I also want to share the best evidence and ideas on other aspects of practicing medicine, such as quality improvement, population health, and health equity. I’ll be looking to increase recruitment and mentorship of authors from diverse backgrounds, including those historically ­underrepresented in medicine.

I look forward to working with the editorial board, associate and assistant editors, and staff of JFP to serve the diverse interests and needs of our readers. To that end, we’ll be looking for your guidance. How else can JFP help you in your day-to-day practice? Please let us know your ideas. Drop us a line at [email protected].

Finally, please join me in thanking Drs. Henry Barry and Kate Rowland for all of their work this past year in keeping JFP going strong!

Normal blood pressure (BP) is defined as systolic BP (SBP) < 120 mm Hg and diastolic BP (DBP) < 80 mm Hg.¹ The thresholds for hypertension (HTN) are shown in TABLE 1.¹ These thresholds must be met on at least 2 separate occasions to merit a diagnosis of HTN.¹

Given the high prevalence of HTN and its associated comorbidities, the US Preventive Services Task Force (USPSTF) recently reaffirmed its recommendation that every adult be screened for HTN, regardless of risk factors.² Patients 40 years of age and older and those with risk factors (obesity, family history of HTN, diabetes) should have their BP checked at least annually. Individuals ages 18 to 39 years without risk factors who are initially normotensive should be rescreened within 3 to 5 years.²

Patients are most commonly screened for HTN in the outpatient setting. However, office BP measurements may be inaccurate and are of limited diagnostic utility when taken as a single reading.^1,3,4 As will be described later, office BP measurements are subject to multiple sources of error that can result in a mean underestimation of 24  mm Hg to a mean overestimation of 33 mm Hg for SBP, and a mean underestimation of 14  mm Hg to a mean overestimation of 23 mm Hg for DBP.⁴

Differences to this degree between true BP and measured BP can have important implications for the diagnosis, surveillance, and management of HTN. To diminish this potential for error, the American Heart Association HTN guideline and USPSTF recommendation advise clinicians to obtain out-of-office BP measurements to confirm a diagnosis of HTN before initiating treatment.^1,2 The preferred methods for out-of-office BP assessment are home BP monitoring (HBPM) and 24-hour ambulatory BP monitoring (ABPM).

Limitations of office BP measurement

Multiple sources of error can lead to wide variability in the measurement of office BP, whether taken via the traditional sphygmomanometer auscultatory approach or with an oscillometric monitor.^1,4 Measurement error can be patient related (eg, talking during the reading, or eating or using tobacco prior to measurement), device related (eg, device has not been calibrated or validated), or procedure related (eg, miscuffing, improper patient positioning).

Although use of validated oscillometric monitors eliminates some sources of error such as terminal digit bias, rapid cuff deflation, and missed Korotkoff sounds, their use does not eliminate other sources of error. For example, a patient’s use of tobacco 30 to 60 minutes prior to measurement can raise SBP by 2.8 to 25 mm Hg and DBP 2 to 18 mm Hg.⁴ Having a full bladder can elevate SBP by 4.2 to 33 mm Hg and DBP by 2.8 to 18.5 mm Hg.⁴ If the patient is talking during measurement, is crossing one leg over the opposite knee, or has an unsupported arm below the level of the heart, SBP and DBP can rise, respectively, by an estimated mean 2 to 23 mm Hg and 2 to 14 mm Hg.⁴

Although many sources of BP measurement error can be reduced or eliminated through standardization of technique across office staff, some sources of inaccuracy will persist. Even if all variables are optimized, relying solely on office BP monitoring will still misclassify BP phenotypes, which require out-of-office BP assessments.^1,3FIGURE 1 reviews key tips for maximizing the accuracy of BP measurement, regardless of where the measurement is done.

Continue to: Automated office BP

Automated office BP (AOBP) lessens some of the limitations inherent with the traditional sphygmomanometer auscultatory and single-measurement oscillometric devices. AOBP combines oscillometric technology with the capacity to record multiple BP readings within a single activation, thereby providing an average of these readings.¹ The total time required for AOBP is 4 to 6 minutes, including a brief rest period before the measurement starts. Studies have reported comparable readings between staff-attended and unattended AOBP, which is an encouraging way to eliminate some measurement error (eg, talking with the patient) and to improve efficiency.^5,6

Waiting several minutes per patient to record BP may not be practical in a busy office setting and may require an alteration of workflow. There is a paucity of literature evaluating practice realities, which makes it difficult to know how many patients are getting their BP checked in this manner. Several studies have shown that BP measured with AOBP is closer to awake out-of-office BP as measured with ABPM (discussed in a bit),^5-8 largely through mitigation of white-coat effect. Canada now recommends AOBP as the preferred method for diagnosing HTN and monitoring BP.⁹

Home blood pressure monitoring

HBPM refers to individuals measuring their own BP at home. It is important to remember this definition, as the term is sometimes applied to a patient’s BP measured at home by an observer or to an individual taking their own BP outside of the home (kiosk, pharmacy, at work). The short-term reproducibility of mean BP with HBPM is high. The test-retest correlations of HBPM range from 0.70 to 0.84 mm Hg for mean SBP, and from 0.57 to 0.83 mm Hg for mean DBP.^10-13 In contrast to 24-hour ABPM, HBPM is better tolerated, cheaper, and more widely available.^14,15

There is strong evidence that HBPM adds value over and above office measurements in predicting end-organ damage and cardiovascular disease (CVD) outcomes, and it has a stronger relationship with CVD risk than office BP.¹ Compared with office BP measurement, HBPM is a better predictor of echocardiographic left ventricular mass index, urinary albumin-to-creatinine ratio, proteinuria, silent cerebrovascular disease, nonfatal cardiovascular outcomes, cardiovascular mortality, and all-cause mortality.^15,16 There is no strong evidence demonstrating the superiority of HBPM over ABPM, or vice versa, for predicting CVD events or mortality.¹⁷ Both ABPM and HBPM have important roles in out-of-office monitoring (FIGURE 2³).

Clinical indications for HBPM

HBPM can facilitate diagnosis of white-coat HTN or effect (if already on BP-lowering medication) as well as masked uncontrolled HTN and masked HTN. Importantly, masked HTN is associated with nearly the same risk of target organ damage and cardiovascular events as sustained HTN. In one meta-analysis the overall adjusted hazard ratio for CVD events was 2.00 (95% CI, 1.58-2.52) for masked HTN and 2.28 (95% CI, 1.87-2.78) for sustained HTN, compared with normotensive individuals.¹⁸ Other studies support these results, demonstrating that masked HTN confers risk similar to sustained HTN.^19,20

Even treated subjects with masked uncontrolled HTN (normal office and high home BP) have higher CVD risk, likely due to undertreatment given lower BP in the office setting. Among 1451 treated patients in a large cohort study who were followed for a median of 8.3 years, CVD was higher in those with masked uncontrolled HTN (adjusted hazard ratio = 1.76; 95% CI, 1.23-2.53) compared to treated controlled patients (normal office and home BP).²¹

Home BP monitoring can reveal masked hypertension, which confers risk for endorgan damage similar to that of sustained hypertension.

HBPM also can be used to monitor BP levels over time, to increase patient involvement in chronic disease management, and to improve adherence with medications. Since 2008, several meta-analyses have been published showing improved BP control when HBPM is combined with other interventions and patient education.^22-25 Particularly relevant in the age of increased telehealth, several meta-analyses demonstrate improvement in BP control when HBPM is combined with web- or phone-based support, systematic medication titration, patient education, and provider counseling.^22-25 A comprehensive systematic review found HBPM with this kind of ongoing support (compared with usual care) led to clinic SBP reductions of 3.2 mm Hg (95% CI, 1.6-4.9) at 12 months.²²

Continue to: HBPM nuts and bolts

When using HBPM to obtain a BP average either for confirming a diagnosis or assessing HTN control, patients should be instructed to record their BP measurements twice in the morning and twice at night for a minimum of 3 days (ie, 12 readings).^26,27 For each monitoring period, both SBP and DBP readings should be recorded, although protocols differ as to whether to discard the initial reading of each day, or the entire first day of readings.^26-29 Consecutive days of monitoring are preferred, although nonconsecutive days also are likely to provide valid data. Once BP stabilizes, monitoring 1 to 3 days a week is likely sufficient.

Most guidelines cite a mean BP of ≥ 135/85 mm Hg as the indication of high BP on HBPM.^1,28,29 This value corresponds to an office BP average of 140/90 mm Hg. TABLE 2¹ shows the comparison of home, ambulatory, and office BP thresholds.

Device selection and validation

As with any BP device, validation and proper technique are important. Recommend only upper-arm cuff devices that have passed validation protocols.³⁰ To eliminate the burden on patients to accurately record and store their BP readings, and to eliminate this step as a source of bias, additionally recommend devices with built-in memory. Although easy-to-use wrist and finger monitors have become popular, there are important limitations in terms of accurate positioning and a lack of validated protocols.^31,32

The brachial artery is still the recommended measurement location, unless otherwise precluded due to arm size (the largest size for most validated upper-arm cuffs is 42 cm), patient discomfort, medical contraindication (eg, lymphedema), or immobility (eg, due to injury). Arm size limitation is particularly important as obesity rates continue to rise. Data from the National Health and Nutrition Examination Survey indicate that 52% of men and 38% of women with HTN need a different cuff size than the US standard.³³ If the brachial artery is not an option, there are no definitive data to recommend finger over wrist devices, as both are limited by lack of validated protocols.

The website www.stridebp.org maintains a current list of validated and preferred BP devices, and is supported by the European Society of Hypertension, the International Society of Hypertension, and the World Hypertension League. There are more than 4000 devices on the global market, but only 8% have been validated according to StrideBP.

Advances in HBPM that offset previous limitations

The usefulness of HBPM depends on patient factors such as a commitment to monitoring, applying standardized technique, and accurately recording measurements. Discuss these matters with patients before recommending HBPM. Until recently, HBPM devices could not measure BP during sleep. However, a device that assesses BP during sleep has now come on the US market, with preliminary data suggesting the BP measurements are similar to those obtained with ABPM.³⁴ Advances in device memory and data storage and increased availability of electronic health record connection continue to improve the standardization and reliability of HBPM. In fact, there is a growing list of electronic health portals that can be synced with apps for direct transfer of HBPM data.

Ambulatory blood pressure monitoring

ABPM involves wearing a small device connected to an arm BP cuff that measures BP at pre-programmed intervals over a 24-hour period, during sleep and wakefulness. ABPM is the standard against which HBPM and office BP are compared.^1-3

Continue to: Clinical indications for ABPM

Compared with office-based BP measurements, ABPM has a stronger positive correlation with clinical CVD outcomes and HTN-related organ damage.¹ ABPM has the advantage of being able to provide a large number of measurements over the course of a patient’s daily activities, including sleep. It is useful to evaluate for a wide spectrum of hypertensive or hypotensive patterns, including nocturnal, postprandial, and drug-related patterns. ABPM also is used to assess for white-coat HTN and masked HTN.¹

Among these BP phenotypes, an estimated 15% to 30% of adults in the United States exhibit white-coat HTN.¹ Most evidence suggests that white-coat HTN confers similar cardiovascular risk as normotension, and it therefore does not require treatment.³⁵ Confirming this diagnosis saves the individual and the health care system the cost of unnecessary diagnosis and treatment.

A home monitor that assesses sleep BP is available in some US markets, with data showing its sleep measurements are similar to those obtained with ambulatory BP monitoring.

One cost-effectiveness study using ABPM for annual screening with subsequent treatment for those confirmed to be hypertensive found that ABPM reduced treatment-years by correctly identifying white-coat HTN, and also delayed treatment for those who would eventually develop HTN with advancing age.³⁶ The estimates in savings were 3% to 14% for total cost of care for hypertension and 10% to 23% reduction in treatment days.³⁶ An Australian study showed similar cost reductions.³⁷ A more recent analysis demonstrated that compared with clinic BP measurement alone, incorporation of ABPM is associated with lifetime cost-savings ranging from $77 to $5013, depending on the age and sex of the patients modeled.³⁸

ABPM can also be used to rule out white-coat effect in patients being evaluated for resistant HTN. Several studies demonstrate that among patients with apparent resistant HTN, approximately one-third have controlled BP when assessed by ABPM.^39-41 Thus, it is recommended to conduct an out-of-office BP assessment in patients with apparent resistant HTN prior to adding another medication.⁴¹Twelve percent of US adults have masked HTN.⁴² As described earlier, these patients, unrecognized without out-of-office BP assessment, are twice as likely to experience a CVD event compared with normotensive patients.^1,42,43

ABPM nuts and bolts

ABPM devices are typically worn for 24 hours and with little interruption to daily routines. Prior to BP capture, the device will alert the patient to ensure the patient’s arm can be held still while the BP measurement is being captured.⁴⁴ At the completion of 24 hours, specific software uses the stored data to calculate the BP and heart rate averages, as well as minimums and maximums throughout the monitoring period. Clinical decision-making should be driven by the average BP measurements during times of sleep and wakefulness.^1,14,44FIGURE 3 is an example of output from an ABPM session. TABLE 3^1,44 offers a comparison of HBPM and ABPM.

Limitations of ABPM

While ABPM has been designed to be almost effortless to use, some may find it inconvenient to wear. The repeated cuff inflations can cause discomfort or bruising, and the device can interfere with sleep.⁴⁵ Inconsistent or incorrect wear of ABPM can diminish the quality of BP measurements, which can potentially affect interpretation and subsequent clinical decision-making. Therefore, consider the likelihood of correct and complete usage before ordering ABPM for your patient. Such deliberation is particularly relevant when there is concern for BP phenotypes such as nocturnal nondipping (failure of BP to fall appropriately during sleep) and postprandial HTN and hypotension.

Conduct out-of-office BP assessment of apparent resistant hypertension before adding another medication.

Trained personnel are needed to oversee coordination of the ABPM service within the clinic and to educate patients about proper wear. Additionally, ABPM has not been widely used in US clinical practices to date, in part because this diagnostic strategy is not favorably reimbursed. Based on geographic region, Medicare currently pays between $56 and $122 per 24-hour ABPM session, and only for suspected white-coat HTN.³⁸ Discrepancies remain between commercial and Medicaid/Medicare coverage.⁴⁴

Continue to: Other modes of monitoring BP

Other modes of monitoring BP

The COVID pandemic has changed health care in many ways, including the frequency of in-person visits. As clinics come to rely more on virtual visits and telehealth, accurate monitoring of out-of-office BP has become more important. Kiosks and smart technology offer the opportunity to supplement traditional in-office BP readings. Kiosks are commonly found in pharmacies and grocery stores. These stations facilitate BP monitoring, as long as the device is appropriately validated and calibrated. Unfortunately, most kiosks have only one cuff size that is too small for many US adults, and some do not have a back support.^46,47 Additionally, despite US Food and Drug Administration clearance, many kiosks do not have validated protocols, and the reproducibility of kiosk-measured BP is questionable.^46,47

Mobile health technology is increasingly being examined as an effective means of providing health information, support, and management in chronic disease. Smartphone technology, wearable sensors, and cuffless BP monitors offer promise for providing BP data in more convenient ways. However, as with kiosk devices, very few of these have been validated, and several have been shown to have poor accuracy compared with oscillometric devices.^48-50 For these reasons, kiosk and smart technology for BP monitoring are not recommended at this time, unless no alternatives are available to the patient.

CORRESPONDENCE
Anthony J. Viera, MD, Department of Family Medicine and Community Health, Duke University School of Medicine, 2200 West Main Street, Suite 400, Durham, NC 27705; [email protected]

Normal blood pressure (BP) is defined as systolic BP (SBP) < 120 mm Hg and diastolic BP (DBP) < 80 mm Hg.¹ The thresholds for hypertension (HTN) are shown in TABLE 1.¹ These thresholds must be met on at least 2 separate occasions to merit a diagnosis of HTN.¹

Given the high prevalence of HTN and its associated comorbidities, the US Preventive Services Task Force (USPSTF) recently reaffirmed its recommendation that every adult be screened for HTN, regardless of risk factors.² Patients 40 years of age and older and those with risk factors (obesity, family history of HTN, diabetes) should have their BP checked at least annually. Individuals ages 18 to 39 years without risk factors who are initially normotensive should be rescreened within 3 to 5 years.²

Patients are most commonly screened for HTN in the outpatient setting. However, office BP measurements may be inaccurate and are of limited diagnostic utility when taken as a single reading.^1,3,4 As will be described later, office BP measurements are subject to multiple sources of error that can result in a mean underestimation of 24  mm Hg to a mean overestimation of 33 mm Hg for SBP, and a mean underestimation of 14  mm Hg to a mean overestimation of 23 mm Hg for DBP.⁴

Differences to this degree between true BP and measured BP can have important implications for the diagnosis, surveillance, and management of HTN. To diminish this potential for error, the American Heart Association HTN guideline and USPSTF recommendation advise clinicians to obtain out-of-office BP measurements to confirm a diagnosis of HTN before initiating treatment.^1,2 The preferred methods for out-of-office BP assessment are home BP monitoring (HBPM) and 24-hour ambulatory BP monitoring (ABPM).

Limitations of office BP measurement

Multiple sources of error can lead to wide variability in the measurement of office BP, whether taken via the traditional sphygmomanometer auscultatory approach or with an oscillometric monitor.^1,4 Measurement error can be patient related (eg, talking during the reading, or eating or using tobacco prior to measurement), device related (eg, device has not been calibrated or validated), or procedure related (eg, miscuffing, improper patient positioning).

Although use of validated oscillometric monitors eliminates some sources of error such as terminal digit bias, rapid cuff deflation, and missed Korotkoff sounds, their use does not eliminate other sources of error. For example, a patient’s use of tobacco 30 to 60 minutes prior to measurement can raise SBP by 2.8 to 25 mm Hg and DBP 2 to 18 mm Hg.⁴ Having a full bladder can elevate SBP by 4.2 to 33 mm Hg and DBP by 2.8 to 18.5 mm Hg.⁴ If the patient is talking during measurement, is crossing one leg over the opposite knee, or has an unsupported arm below the level of the heart, SBP and DBP can rise, respectively, by an estimated mean 2 to 23 mm Hg and 2 to 14 mm Hg.⁴

Although many sources of BP measurement error can be reduced or eliminated through standardization of technique across office staff, some sources of inaccuracy will persist. Even if all variables are optimized, relying solely on office BP monitoring will still misclassify BP phenotypes, which require out-of-office BP assessments.^1,3FIGURE 1 reviews key tips for maximizing the accuracy of BP measurement, regardless of where the measurement is done.

Continue to: Automated office BP

Automated office BP (AOBP) lessens some of the limitations inherent with the traditional sphygmomanometer auscultatory and single-measurement oscillometric devices. AOBP combines oscillometric technology with the capacity to record multiple BP readings within a single activation, thereby providing an average of these readings.¹ The total time required for AOBP is 4 to 6 minutes, including a brief rest period before the measurement starts. Studies have reported comparable readings between staff-attended and unattended AOBP, which is an encouraging way to eliminate some measurement error (eg, talking with the patient) and to improve efficiency.^5,6

Waiting several minutes per patient to record BP may not be practical in a busy office setting and may require an alteration of workflow. There is a paucity of literature evaluating practice realities, which makes it difficult to know how many patients are getting their BP checked in this manner. Several studies have shown that BP measured with AOBP is closer to awake out-of-office BP as measured with ABPM (discussed in a bit),^5-8 largely through mitigation of white-coat effect. Canada now recommends AOBP as the preferred method for diagnosing HTN and monitoring BP.⁹

Home blood pressure monitoring

HBPM refers to individuals measuring their own BP at home. It is important to remember this definition, as the term is sometimes applied to a patient’s BP measured at home by an observer or to an individual taking their own BP outside of the home (kiosk, pharmacy, at work). The short-term reproducibility of mean BP with HBPM is high. The test-retest correlations of HBPM range from 0.70 to 0.84 mm Hg for mean SBP, and from 0.57 to 0.83 mm Hg for mean DBP.^10-13 In contrast to 24-hour ABPM, HBPM is better tolerated, cheaper, and more widely available.^14,15

There is strong evidence that HBPM adds value over and above office measurements in predicting end-organ damage and cardiovascular disease (CVD) outcomes, and it has a stronger relationship with CVD risk than office BP.¹ Compared with office BP measurement, HBPM is a better predictor of echocardiographic left ventricular mass index, urinary albumin-to-creatinine ratio, proteinuria, silent cerebrovascular disease, nonfatal cardiovascular outcomes, cardiovascular mortality, and all-cause mortality.^15,16 There is no strong evidence demonstrating the superiority of HBPM over ABPM, or vice versa, for predicting CVD events or mortality.¹⁷ Both ABPM and HBPM have important roles in out-of-office monitoring (FIGURE 2³).

Clinical indications for HBPM

HBPM can facilitate diagnosis of white-coat HTN or effect (if already on BP-lowering medication) as well as masked uncontrolled HTN and masked HTN. Importantly, masked HTN is associated with nearly the same risk of target organ damage and cardiovascular events as sustained HTN. In one meta-analysis the overall adjusted hazard ratio for CVD events was 2.00 (95% CI, 1.58-2.52) for masked HTN and 2.28 (95% CI, 1.87-2.78) for sustained HTN, compared with normotensive individuals.¹⁸ Other studies support these results, demonstrating that masked HTN confers risk similar to sustained HTN.^19,20

Even treated subjects with masked uncontrolled HTN (normal office and high home BP) have higher CVD risk, likely due to undertreatment given lower BP in the office setting. Among 1451 treated patients in a large cohort study who were followed for a median of 8.3 years, CVD was higher in those with masked uncontrolled HTN (adjusted hazard ratio = 1.76; 95% CI, 1.23-2.53) compared to treated controlled patients (normal office and home BP).²¹

Home BP monitoring can reveal masked hypertension, which confers risk for endorgan damage similar to that of sustained hypertension.

HBPM also can be used to monitor BP levels over time, to increase patient involvement in chronic disease management, and to improve adherence with medications. Since 2008, several meta-analyses have been published showing improved BP control when HBPM is combined with other interventions and patient education.^22-25 Particularly relevant in the age of increased telehealth, several meta-analyses demonstrate improvement in BP control when HBPM is combined with web- or phone-based support, systematic medication titration, patient education, and provider counseling.^22-25 A comprehensive systematic review found HBPM with this kind of ongoing support (compared with usual care) led to clinic SBP reductions of 3.2 mm Hg (95% CI, 1.6-4.9) at 12 months.²²

Continue to: HBPM nuts and bolts

When using HBPM to obtain a BP average either for confirming a diagnosis or assessing HTN control, patients should be instructed to record their BP measurements twice in the morning and twice at night for a minimum of 3 days (ie, 12 readings).^26,27 For each monitoring period, both SBP and DBP readings should be recorded, although protocols differ as to whether to discard the initial reading of each day, or the entire first day of readings.^26-29 Consecutive days of monitoring are preferred, although nonconsecutive days also are likely to provide valid data. Once BP stabilizes, monitoring 1 to 3 days a week is likely sufficient.

Most guidelines cite a mean BP of ≥ 135/85 mm Hg as the indication of high BP on HBPM.^1,28,29 This value corresponds to an office BP average of 140/90 mm Hg. TABLE 2¹ shows the comparison of home, ambulatory, and office BP thresholds.

Device selection and validation

As with any BP device, validation and proper technique are important. Recommend only upper-arm cuff devices that have passed validation protocols.³⁰ To eliminate the burden on patients to accurately record and store their BP readings, and to eliminate this step as a source of bias, additionally recommend devices with built-in memory. Although easy-to-use wrist and finger monitors have become popular, there are important limitations in terms of accurate positioning and a lack of validated protocols.^31,32

The brachial artery is still the recommended measurement location, unless otherwise precluded due to arm size (the largest size for most validated upper-arm cuffs is 42 cm), patient discomfort, medical contraindication (eg, lymphedema), or immobility (eg, due to injury). Arm size limitation is particularly important as obesity rates continue to rise. Data from the National Health and Nutrition Examination Survey indicate that 52% of men and 38% of women with HTN need a different cuff size than the US standard.³³ If the brachial artery is not an option, there are no definitive data to recommend finger over wrist devices, as both are limited by lack of validated protocols.

The website www.stridebp.org maintains a current list of validated and preferred BP devices, and is supported by the European Society of Hypertension, the International Society of Hypertension, and the World Hypertension League. There are more than 4000 devices on the global market, but only 8% have been validated according to StrideBP.

Advances in HBPM that offset previous limitations

The usefulness of HBPM depends on patient factors such as a commitment to monitoring, applying standardized technique, and accurately recording measurements. Discuss these matters with patients before recommending HBPM. Until recently, HBPM devices could not measure BP during sleep. However, a device that assesses BP during sleep has now come on the US market, with preliminary data suggesting the BP measurements are similar to those obtained with ABPM.³⁴ Advances in device memory and data storage and increased availability of electronic health record connection continue to improve the standardization and reliability of HBPM. In fact, there is a growing list of electronic health portals that can be synced with apps for direct transfer of HBPM data.

Ambulatory blood pressure monitoring

ABPM involves wearing a small device connected to an arm BP cuff that measures BP at pre-programmed intervals over a 24-hour period, during sleep and wakefulness. ABPM is the standard against which HBPM and office BP are compared.^1-3

Continue to: Clinical indications for ABPM

Compared with office-based BP measurements, ABPM has a stronger positive correlation with clinical CVD outcomes and HTN-related organ damage.¹ ABPM has the advantage of being able to provide a large number of measurements over the course of a patient’s daily activities, including sleep. It is useful to evaluate for a wide spectrum of hypertensive or hypotensive patterns, including nocturnal, postprandial, and drug-related patterns. ABPM also is used to assess for white-coat HTN and masked HTN.¹

Among these BP phenotypes, an estimated 15% to 30% of adults in the United States exhibit white-coat HTN.¹ Most evidence suggests that white-coat HTN confers similar cardiovascular risk as normotension, and it therefore does not require treatment.³⁵ Confirming this diagnosis saves the individual and the health care system the cost of unnecessary diagnosis and treatment.

A home monitor that assesses sleep BP is available in some US markets, with data showing its sleep measurements are similar to those obtained with ambulatory BP monitoring.

One cost-effectiveness study using ABPM for annual screening with subsequent treatment for those confirmed to be hypertensive found that ABPM reduced treatment-years by correctly identifying white-coat HTN, and also delayed treatment for those who would eventually develop HTN with advancing age.³⁶ The estimates in savings were 3% to 14% for total cost of care for hypertension and 10% to 23% reduction in treatment days.³⁶ An Australian study showed similar cost reductions.³⁷ A more recent analysis demonstrated that compared with clinic BP measurement alone, incorporation of ABPM is associated with lifetime cost-savings ranging from $77 to $5013, depending on the age and sex of the patients modeled.³⁸

ABPM can also be used to rule out white-coat effect in patients being evaluated for resistant HTN. Several studies demonstrate that among patients with apparent resistant HTN, approximately one-third have controlled BP when assessed by ABPM.^39-41 Thus, it is recommended to conduct an out-of-office BP assessment in patients with apparent resistant HTN prior to adding another medication.⁴¹Twelve percent of US adults have masked HTN.⁴² As described earlier, these patients, unrecognized without out-of-office BP assessment, are twice as likely to experience a CVD event compared with normotensive patients.^1,42,43

ABPM nuts and bolts

ABPM devices are typically worn for 24 hours and with little interruption to daily routines. Prior to BP capture, the device will alert the patient to ensure the patient’s arm can be held still while the BP measurement is being captured.⁴⁴ At the completion of 24 hours, specific software uses the stored data to calculate the BP and heart rate averages, as well as minimums and maximums throughout the monitoring period. Clinical decision-making should be driven by the average BP measurements during times of sleep and wakefulness.^1,14,44FIGURE 3 is an example of output from an ABPM session. TABLE 3^1,44 offers a comparison of HBPM and ABPM.

Limitations of ABPM

While ABPM has been designed to be almost effortless to use, some may find it inconvenient to wear. The repeated cuff inflations can cause discomfort or bruising, and the device can interfere with sleep.⁴⁵ Inconsistent or incorrect wear of ABPM can diminish the quality of BP measurements, which can potentially affect interpretation and subsequent clinical decision-making. Therefore, consider the likelihood of correct and complete usage before ordering ABPM for your patient. Such deliberation is particularly relevant when there is concern for BP phenotypes such as nocturnal nondipping (failure of BP to fall appropriately during sleep) and postprandial HTN and hypotension.

Conduct out-of-office BP assessment of apparent resistant hypertension before adding another medication.

Trained personnel are needed to oversee coordination of the ABPM service within the clinic and to educate patients about proper wear. Additionally, ABPM has not been widely used in US clinical practices to date, in part because this diagnostic strategy is not favorably reimbursed. Based on geographic region, Medicare currently pays between $56 and $122 per 24-hour ABPM session, and only for suspected white-coat HTN.³⁸ Discrepancies remain between commercial and Medicaid/Medicare coverage.⁴⁴

Continue to: Other modes of monitoring BP

Other modes of monitoring BP

The COVID pandemic has changed health care in many ways, including the frequency of in-person visits. As clinics come to rely more on virtual visits and telehealth, accurate monitoring of out-of-office BP has become more important. Kiosks and smart technology offer the opportunity to supplement traditional in-office BP readings. Kiosks are commonly found in pharmacies and grocery stores. These stations facilitate BP monitoring, as long as the device is appropriately validated and calibrated. Unfortunately, most kiosks have only one cuff size that is too small for many US adults, and some do not have a back support.^46,47 Additionally, despite US Food and Drug Administration clearance, many kiosks do not have validated protocols, and the reproducibility of kiosk-measured BP is questionable.^46,47

Mobile health technology is increasingly being examined as an effective means of providing health information, support, and management in chronic disease. Smartphone technology, wearable sensors, and cuffless BP monitors offer promise for providing BP data in more convenient ways. However, as with kiosk devices, very few of these have been validated, and several have been shown to have poor accuracy compared with oscillometric devices.^48-50 For these reasons, kiosk and smart technology for BP monitoring are not recommended at this time, unless no alternatives are available to the patient.

CORRESPONDENCE
Anthony J. Viera, MD, Department of Family Medicine and Community Health, Duke University School of Medicine, 2200 West Main Street, Suite 400, Durham, NC 27705; [email protected]

Normal blood pressure (BP) is defined as systolic BP (SBP) < 120 mm Hg and diastolic BP (DBP) < 80 mm Hg.¹ The thresholds for hypertension (HTN) are shown in TABLE 1.¹ These thresholds must be met on at least 2 separate occasions to merit a diagnosis of HTN.¹

Given the high prevalence of HTN and its associated comorbidities, the US Preventive Services Task Force (USPSTF) recently reaffirmed its recommendation that every adult be screened for HTN, regardless of risk factors.² Patients 40 years of age and older and those with risk factors (obesity, family history of HTN, diabetes) should have their BP checked at least annually. Individuals ages 18 to 39 years without risk factors who are initially normotensive should be rescreened within 3 to 5 years.²

Patients are most commonly screened for HTN in the outpatient setting. However, office BP measurements may be inaccurate and are of limited diagnostic utility when taken as a single reading.^1,3,4 As will be described later, office BP measurements are subject to multiple sources of error that can result in a mean underestimation of 24  mm Hg to a mean overestimation of 33 mm Hg for SBP, and a mean underestimation of 14  mm Hg to a mean overestimation of 23 mm Hg for DBP.⁴

Differences to this degree between true BP and measured BP can have important implications for the diagnosis, surveillance, and management of HTN. To diminish this potential for error, the American Heart Association HTN guideline and USPSTF recommendation advise clinicians to obtain out-of-office BP measurements to confirm a diagnosis of HTN before initiating treatment.^1,2 The preferred methods for out-of-office BP assessment are home BP monitoring (HBPM) and 24-hour ambulatory BP monitoring (ABPM).

Limitations of office BP measurement

Multiple sources of error can lead to wide variability in the measurement of office BP, whether taken via the traditional sphygmomanometer auscultatory approach or with an oscillometric monitor.^1,4 Measurement error can be patient related (eg, talking during the reading, or eating or using tobacco prior to measurement), device related (eg, device has not been calibrated or validated), or procedure related (eg, miscuffing, improper patient positioning).

Although use of validated oscillometric monitors eliminates some sources of error such as terminal digit bias, rapid cuff deflation, and missed Korotkoff sounds, their use does not eliminate other sources of error. For example, a patient’s use of tobacco 30 to 60 minutes prior to measurement can raise SBP by 2.8 to 25 mm Hg and DBP 2 to 18 mm Hg.⁴ Having a full bladder can elevate SBP by 4.2 to 33 mm Hg and DBP by 2.8 to 18.5 mm Hg.⁴ If the patient is talking during measurement, is crossing one leg over the opposite knee, or has an unsupported arm below the level of the heart, SBP and DBP can rise, respectively, by an estimated mean 2 to 23 mm Hg and 2 to 14 mm Hg.⁴

Although many sources of BP measurement error can be reduced or eliminated through standardization of technique across office staff, some sources of inaccuracy will persist. Even if all variables are optimized, relying solely on office BP monitoring will still misclassify BP phenotypes, which require out-of-office BP assessments.^1,3FIGURE 1 reviews key tips for maximizing the accuracy of BP measurement, regardless of where the measurement is done.

Continue to: Automated office BP

Automated office BP (AOBP) lessens some of the limitations inherent with the traditional sphygmomanometer auscultatory and single-measurement oscillometric devices. AOBP combines oscillometric technology with the capacity to record multiple BP readings within a single activation, thereby providing an average of these readings.¹ The total time required for AOBP is 4 to 6 minutes, including a brief rest period before the measurement starts. Studies have reported comparable readings between staff-attended and unattended AOBP, which is an encouraging way to eliminate some measurement error (eg, talking with the patient) and to improve efficiency.^5,6

Waiting several minutes per patient to record BP may not be practical in a busy office setting and may require an alteration of workflow. There is a paucity of literature evaluating practice realities, which makes it difficult to know how many patients are getting their BP checked in this manner. Several studies have shown that BP measured with AOBP is closer to awake out-of-office BP as measured with ABPM (discussed in a bit),^5-8 largely through mitigation of white-coat effect. Canada now recommends AOBP as the preferred method for diagnosing HTN and monitoring BP.⁹

Home blood pressure monitoring

HBPM refers to individuals measuring their own BP at home. It is important to remember this definition, as the term is sometimes applied to a patient’s BP measured at home by an observer or to an individual taking their own BP outside of the home (kiosk, pharmacy, at work). The short-term reproducibility of mean BP with HBPM is high. The test-retest correlations of HBPM range from 0.70 to 0.84 mm Hg for mean SBP, and from 0.57 to 0.83 mm Hg for mean DBP.^10-13 In contrast to 24-hour ABPM, HBPM is better tolerated, cheaper, and more widely available.^14,15

There is strong evidence that HBPM adds value over and above office measurements in predicting end-organ damage and cardiovascular disease (CVD) outcomes, and it has a stronger relationship with CVD risk than office BP.¹ Compared with office BP measurement, HBPM is a better predictor of echocardiographic left ventricular mass index, urinary albumin-to-creatinine ratio, proteinuria, silent cerebrovascular disease, nonfatal cardiovascular outcomes, cardiovascular mortality, and all-cause mortality.^15,16 There is no strong evidence demonstrating the superiority of HBPM over ABPM, or vice versa, for predicting CVD events or mortality.¹⁷ Both ABPM and HBPM have important roles in out-of-office monitoring (FIGURE 2³).

Clinical indications for HBPM

HBPM can facilitate diagnosis of white-coat HTN or effect (if already on BP-lowering medication) as well as masked uncontrolled HTN and masked HTN. Importantly, masked HTN is associated with nearly the same risk of target organ damage and cardiovascular events as sustained HTN. In one meta-analysis the overall adjusted hazard ratio for CVD events was 2.00 (95% CI, 1.58-2.52) for masked HTN and 2.28 (95% CI, 1.87-2.78) for sustained HTN, compared with normotensive individuals.¹⁸ Other studies support these results, demonstrating that masked HTN confers risk similar to sustained HTN.^19,20

Even treated subjects with masked uncontrolled HTN (normal office and high home BP) have higher CVD risk, likely due to undertreatment given lower BP in the office setting. Among 1451 treated patients in a large cohort study who were followed for a median of 8.3 years, CVD was higher in those with masked uncontrolled HTN (adjusted hazard ratio = 1.76; 95% CI, 1.23-2.53) compared to treated controlled patients (normal office and home BP).²¹

Home BP monitoring can reveal masked hypertension, which confers risk for endorgan damage similar to that of sustained hypertension.

HBPM also can be used to monitor BP levels over time, to increase patient involvement in chronic disease management, and to improve adherence with medications. Since 2008, several meta-analyses have been published showing improved BP control when HBPM is combined with other interventions and patient education.^22-25 Particularly relevant in the age of increased telehealth, several meta-analyses demonstrate improvement in BP control when HBPM is combined with web- or phone-based support, systematic medication titration, patient education, and provider counseling.^22-25 A comprehensive systematic review found HBPM with this kind of ongoing support (compared with usual care) led to clinic SBP reductions of 3.2 mm Hg (95% CI, 1.6-4.9) at 12 months.²²

Continue to: HBPM nuts and bolts

When using HBPM to obtain a BP average either for confirming a diagnosis or assessing HTN control, patients should be instructed to record their BP measurements twice in the morning and twice at night for a minimum of 3 days (ie, 12 readings).^26,27 For each monitoring period, both SBP and DBP readings should be recorded, although protocols differ as to whether to discard the initial reading of each day, or the entire first day of readings.^26-29 Consecutive days of monitoring are preferred, although nonconsecutive days also are likely to provide valid data. Once BP stabilizes, monitoring 1 to 3 days a week is likely sufficient.

Most guidelines cite a mean BP of ≥ 135/85 mm Hg as the indication of high BP on HBPM.^1,28,29 This value corresponds to an office BP average of 140/90 mm Hg. TABLE 2¹ shows the comparison of home, ambulatory, and office BP thresholds.

Device selection and validation

As with any BP device, validation and proper technique are important. Recommend only upper-arm cuff devices that have passed validation protocols.³⁰ To eliminate the burden on patients to accurately record and store their BP readings, and to eliminate this step as a source of bias, additionally recommend devices with built-in memory. Although easy-to-use wrist and finger monitors have become popular, there are important limitations in terms of accurate positioning and a lack of validated protocols.^31,32

The brachial artery is still the recommended measurement location, unless otherwise precluded due to arm size (the largest size for most validated upper-arm cuffs is 42 cm), patient discomfort, medical contraindication (eg, lymphedema), or immobility (eg, due to injury). Arm size limitation is particularly important as obesity rates continue to rise. Data from the National Health and Nutrition Examination Survey indicate that 52% of men and 38% of women with HTN need a different cuff size than the US standard.³³ If the brachial artery is not an option, there are no definitive data to recommend finger over wrist devices, as both are limited by lack of validated protocols.

The website www.stridebp.org maintains a current list of validated and preferred BP devices, and is supported by the European Society of Hypertension, the International Society of Hypertension, and the World Hypertension League. There are more than 4000 devices on the global market, but only 8% have been validated according to StrideBP.

Advances in HBPM that offset previous limitations

The usefulness of HBPM depends on patient factors such as a commitment to monitoring, applying standardized technique, and accurately recording measurements. Discuss these matters with patients before recommending HBPM. Until recently, HBPM devices could not measure BP during sleep. However, a device that assesses BP during sleep has now come on the US market, with preliminary data suggesting the BP measurements are similar to those obtained with ABPM.³⁴ Advances in device memory and data storage and increased availability of electronic health record connection continue to improve the standardization and reliability of HBPM. In fact, there is a growing list of electronic health portals that can be synced with apps for direct transfer of HBPM data.

Ambulatory blood pressure monitoring

ABPM involves wearing a small device connected to an arm BP cuff that measures BP at pre-programmed intervals over a 24-hour period, during sleep and wakefulness. ABPM is the standard against which HBPM and office BP are compared.^1-3

Continue to: Clinical indications for ABPM

Compared with office-based BP measurements, ABPM has a stronger positive correlation with clinical CVD outcomes and HTN-related organ damage.¹ ABPM has the advantage of being able to provide a large number of measurements over the course of a patient’s daily activities, including sleep. It is useful to evaluate for a wide spectrum of hypertensive or hypotensive patterns, including nocturnal, postprandial, and drug-related patterns. ABPM also is used to assess for white-coat HTN and masked HTN.¹

Among these BP phenotypes, an estimated 15% to 30% of adults in the United States exhibit white-coat HTN.¹ Most evidence suggests that white-coat HTN confers similar cardiovascular risk as normotension, and it therefore does not require treatment.³⁵ Confirming this diagnosis saves the individual and the health care system the cost of unnecessary diagnosis and treatment.

A home monitor that assesses sleep BP is available in some US markets, with data showing its sleep measurements are similar to those obtained with ambulatory BP monitoring.

One cost-effectiveness study using ABPM for annual screening with subsequent treatment for those confirmed to be hypertensive found that ABPM reduced treatment-years by correctly identifying white-coat HTN, and also delayed treatment for those who would eventually develop HTN with advancing age.³⁶ The estimates in savings were 3% to 14% for total cost of care for hypertension and 10% to 23% reduction in treatment days.³⁶ An Australian study showed similar cost reductions.³⁷ A more recent analysis demonstrated that compared with clinic BP measurement alone, incorporation of ABPM is associated with lifetime cost-savings ranging from $77 to $5013, depending on the age and sex of the patients modeled.³⁸

ABPM can also be used to rule out white-coat effect in patients being evaluated for resistant HTN. Several studies demonstrate that among patients with apparent resistant HTN, approximately one-third have controlled BP when assessed by ABPM.^39-41 Thus, it is recommended to conduct an out-of-office BP assessment in patients with apparent resistant HTN prior to adding another medication.⁴¹Twelve percent of US adults have masked HTN.⁴² As described earlier, these patients, unrecognized without out-of-office BP assessment, are twice as likely to experience a CVD event compared with normotensive patients.^1,42,43

ABPM nuts and bolts

ABPM devices are typically worn for 24 hours and with little interruption to daily routines. Prior to BP capture, the device will alert the patient to ensure the patient’s arm can be held still while the BP measurement is being captured.⁴⁴ At the completion of 24 hours, specific software uses the stored data to calculate the BP and heart rate averages, as well as minimums and maximums throughout the monitoring period. Clinical decision-making should be driven by the average BP measurements during times of sleep and wakefulness.^1,14,44FIGURE 3 is an example of output from an ABPM session. TABLE 3^1,44 offers a comparison of HBPM and ABPM.

Limitations of ABPM

While ABPM has been designed to be almost effortless to use, some may find it inconvenient to wear. The repeated cuff inflations can cause discomfort or bruising, and the device can interfere with sleep.⁴⁵ Inconsistent or incorrect wear of ABPM can diminish the quality of BP measurements, which can potentially affect interpretation and subsequent clinical decision-making. Therefore, consider the likelihood of correct and complete usage before ordering ABPM for your patient. Such deliberation is particularly relevant when there is concern for BP phenotypes such as nocturnal nondipping (failure of BP to fall appropriately during sleep) and postprandial HTN and hypotension.

Conduct out-of-office BP assessment of apparent resistant hypertension before adding another medication.

Trained personnel are needed to oversee coordination of the ABPM service within the clinic and to educate patients about proper wear. Additionally, ABPM has not been widely used in US clinical practices to date, in part because this diagnostic strategy is not favorably reimbursed. Based on geographic region, Medicare currently pays between $56 and $122 per 24-hour ABPM session, and only for suspected white-coat HTN.³⁸ Discrepancies remain between commercial and Medicaid/Medicare coverage.⁴⁴

Continue to: Other modes of monitoring BP

Other modes of monitoring BP

The COVID pandemic has changed health care in many ways, including the frequency of in-person visits. As clinics come to rely more on virtual visits and telehealth, accurate monitoring of out-of-office BP has become more important. Kiosks and smart technology offer the opportunity to supplement traditional in-office BP readings. Kiosks are commonly found in pharmacies and grocery stores. These stations facilitate BP monitoring, as long as the device is appropriately validated and calibrated. Unfortunately, most kiosks have only one cuff size that is too small for many US adults, and some do not have a back support.^46,47 Additionally, despite US Food and Drug Administration clearance, many kiosks do not have validated protocols, and the reproducibility of kiosk-measured BP is questionable.^46,47

Mobile health technology is increasingly being examined as an effective means of providing health information, support, and management in chronic disease. Smartphone technology, wearable sensors, and cuffless BP monitors offer promise for providing BP data in more convenient ways. However, as with kiosk devices, very few of these have been validated, and several have been shown to have poor accuracy compared with oscillometric devices.^48-50 For these reasons, kiosk and smart technology for BP monitoring are not recommended at this time, unless no alternatives are available to the patient.

CORRESPONDENCE
Anthony J. Viera, MD, Department of Family Medicine and Community Health, Duke University School of Medicine, 2200 West Main Street, Suite 400, Durham, NC 27705; [email protected]

The traditional clinical trial, designed to test whether a new treatment is better than a placebo or another active treatment, is known as a “superiority” trial—although rarely labeled as such. In contrast, the goal of a noninferiority trial is simply to demonstrate that a new treatment is not substantially less effective than the standard therapy.

Such trials are useful when a new therapy is thought to be safer, easier to administer, or less costly than the existing treatment, but not necessarily more effective. And, because it would be unethical to randomize patients with a serious condition for which there already is an effective treatment to placebo, a noninferiority trial is another means of determining if the new treatment is effective.

Noninferiority trials have unique design features and methodology and require a different analysis than traditional superiority trials. Yet many physicians know far less about them; many investigators appear to be less than proficient, as well. A review of 116 noninferiority trials and 46 equivalence trials found that only 20% fulfilled generally accepted quality criteria.¹ To improve the quality of noninferiority trials, the CONSORT (Consolidated Standards of Reporting Trials) Group has published a checklist for trial design and reporting standards.^2,3 Based on this checklist, we came up with 7 key questions to consider when evaluating a noninferiority trial. In the pages that follow, you’ll also find an at-a-glance guide (TABLE) and a methodology review using a hypothetical case (page E7).

1. Is a noninferiority trial appropriate?

The introduction to a noninferiority trial should provide the rationale for this design and the absence of a placebo control group. Look for a review of the evidence of the efficacy of the reference treatment that placebo-controlled trials have revealed, along with the effect size. The advantages of the new treatment over the standard treatment—eg, fewer adverse effects, easier administration, or lower cost—should be discussed, as well.

In the Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY)—a prominent noninferiority trial—investigators compared the standard anticoagulant (warfarin) for patients with atrial fibrillation (AF) at risk of stroke with a new agent, dabigatran.⁴ In the methods section of the abstract and the statistical analysis section of the main body, the authors clearly indicated that this was a noninferiority trial. They began by referring to the existing evidence of warfarin’s effectiveness, then detailed the qualities that make warfarin cumbersome to use, including the need for frequent laboratory monitoring. This was followed by evidence that many patients stop taking warfarin and that even for those who persist with treatment, adequate anticoagulation is difficult to maintain.

The authors went on to state that because dabigatran requires no long-term monitoring, it is easier to use. Therefore, if dabigatran could be shown to be no worse than warfarin in preventing strokes, it would be a reasonable alternative, leaving no doubt that this was an appropriate noninferiority trial.

2. Is the noninferiority margin based on clinical judgment and statistical reasoning?

The noninferiority margin should be based on clinical judgment as to how effective a new treatment must be in order to be declared not clinically inferior to the standard treatment. This can be based on several factors, including the severity of the outcome and the expected advantages of the new treatment. The margin should also take into account the size of the standard treatment’s effect vs placebo. In RELY, for example, the authors noted that the noninferiority margin was based on the desire to preserve at least 50% of the lower limit of the confidence interval (CI) of warfarin’s estimated effect; this was done using data from a previously published meta-analysis of 6 trials comparing warfarin with placebo for stroke prevention in patients with AF.^4-6

3. Are the hypothesis and statistical analysis formulated correctly?

The clinical hypothesis in a noninferiority trial is that the new treatment is not worse than the standard treatment by a prespecified margin; therefore, the statistical null hypothesis to be tested is that the new treatment is worse than the reference treatment by more than that margin. Rejecting a true null hypothesis (for example, because the P value is <.05) is known as a type l error. In this setting, making a type I error would mean accepting a new treatment that is truly worse than the standard by at least the specified margin. Failure to reject a false null hypothesis is known as a type II error, which in this case would mean failing to identify a new treatment that is truly noninferior to the standard.⁷

In RE-LY, the authors stated that the upper limit of the one-sided 97.5% CI for the relative risk of a stroke with dabigatran vs warfarin had to fall below 1.46.⁴ (This is the same as testing the null hypothesis that the hazard ratio is ≥1.46.) Thus, the hypothesis was formulated correctly.

4. Is the sample size appropriate and justified?

The sample size in a noninferiority trial should provide high power to reject the null hypothesis that the difference (or relative risk) between groups is equal to or greater than the noninferiority margin under some clinically meaningful assumption about the true difference (or absolute risk reduction) between groups. A true difference of 0 (or a relative risk of 1) is typically assumed for sample size calculation. However, assuming that the new treatment is truly slightly better or slightly worse than the standard may be clinically appropriate in some cases. This would indicate a need for a smaller or larger sample size, respectively, than that required under the usual assumption of no difference.

When the justification for the sample size in a noninferiority trial is not provided or the number of participants is based on an inappropriate approach (eg, using superiority trial calculations for a noninferiority trial), questions about the quality of the trial arise. The primary concern is whether the noninferiority margin was actually selected before the trial began, as it should have been. And if the researchers used overly optimistic assumptions about the efficacy of the new treatment relative to the standard therapy, the failure to rule out the margin could be misleading. (As with superiority trials that fail to reject the null hypothesis, post hoc power calculations should be avoided.) After the study has ended, the resulting CIs should be used to evaluate whether the study was large enough to adequately assess the relative effectiveness of the treatments.

The RE-LY trial calculated the sample size that was expected to provide 84% power to rule out the prespecified hazard ratio of 1.46, assuming a true event rate of 1.6% per year (presumably for both groups), a recruitment period of 2 years, and at least one year of follow-up. The sample size was subsequently increased from 15,000 to 18,000 to maintain power in case of a low event rate.^4,5

5. Is the noninferiority trial as similar as possible to the trial(s) comparing the standard treatment with placebo?

Characteristics of participants, setting, reference treatment, and outcomes used in a noninferiority trial should be as close as possible to those in the trial(s) comparing the treatment with placebo. This is known as the constancy assumption, and it is key to researchers’ ability to draw a conclusion about noninferiority.

The trials used to calculate the noninferiority margin and the RE-LY trial itself involved similar populations of patients with AF, and the outcome (stroke) was similar.

6. Is a per protocol analysis reported in the results?

In randomized controlled superiority trials, the participants should be analyzed in the groups to which they were originally allocated, regardless of whether they adhered to treatment during the entire follow-up period. Such intention-to-treat (ITT) analysis is important because it provides a more conservative estimate of treatment effect—taking into account that some people who are offered treatment will not accept it and others will discontinue treatment. An ITT analysis therefore tends to minimize treatment effects compared with a “per protocol” analysis, in which participants are analyzed according to the treatment they actually received and are often removed from the analysis if they discontinue or do not adhere to treatment.

Intention-to-treat analysis is important because it provides a more conservative estimate of treatment effect.In noninferiority trials, if patients in the intervention group cross over to the standard treatment group or those in the standard treatment group have poor adherence, an ITT analysis can increase the risk of wrongly claiming noninferiority.⁷ Therefore, a per protocol analysis should be included—and indeed may be preferable.

In RE-LY, ITT analyses were reported, and complete follow-up data were available for 99.9% of patients. However, the rates of treatment discontinuation at one year were about 15% for those on dabigatran and 10% for the warfarin group, and 21% and 17%, respectively, at 2 years.^4,5 If the new treatment were truly less efficacious than the standard treatment, these moderate discontinuation rates could lead to more similar rates of stroke in the 2 groups than would be expected with higher continuation rates, biasing results towards the alternative of noninferiority. Although the original publication of trial results did not include a per protocol analysis, the RE-LY authors later reported that a per protocol analysis yielded similar results to the ITT analysis.

7. Are the overall design and execution of the trial high quality?

Because a poor quality noninferiority trial can appear to demonstrate noninferiority, looking at such studies critically is crucial. Appropriate randomization, concealed allocation, masking, and careful attention to participant flow must all be assessed.^2,3

To continue with our example, the RE-LY trial was well conducted. Randomization was performed centrally via an automated telephone system and 2 doses of dabigatran were administered in a masked fashion, while warfarin was open-label. Remarkably, follow-up was achieved for 99.9% of participants over a median of 2 years, and independent adjudicators masked to treatment group assessed outcomes.^4,5

CORRESPONDENCE
Anne Mounsey, MD, UNC Chapel Hill Department of Family Medicine, 590 Manning Drive, CB 7595, Chapel Hill, NC 27590; [email protected]

The traditional clinical trial, designed to test whether a new treatment is better than a placebo or another active treatment, is known as a “superiority” trial—although rarely labeled as such. In contrast, the goal of a noninferiority trial is simply to demonstrate that a new treatment is not substantially less effective than the standard therapy.

Such trials are useful when a new therapy is thought to be safer, easier to administer, or less costly than the existing treatment, but not necessarily more effective. And, because it would be unethical to randomize patients with a serious condition for which there already is an effective treatment to placebo, a noninferiority trial is another means of determining if the new treatment is effective.

Noninferiority trials have unique design features and methodology and require a different analysis than traditional superiority trials. Yet many physicians know far less about them; many investigators appear to be less than proficient, as well. A review of 116 noninferiority trials and 46 equivalence trials found that only 20% fulfilled generally accepted quality criteria.¹ To improve the quality of noninferiority trials, the CONSORT (Consolidated Standards of Reporting Trials) Group has published a checklist for trial design and reporting standards.^2,3 Based on this checklist, we came up with 7 key questions to consider when evaluating a noninferiority trial. In the pages that follow, you’ll also find an at-a-glance guide (TABLE) and a methodology review using a hypothetical case (page E7).

1. Is a noninferiority trial appropriate?

The introduction to a noninferiority trial should provide the rationale for this design and the absence of a placebo control group. Look for a review of the evidence of the efficacy of the reference treatment that placebo-controlled trials have revealed, along with the effect size. The advantages of the new treatment over the standard treatment—eg, fewer adverse effects, easier administration, or lower cost—should be discussed, as well.

In the Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY)—a prominent noninferiority trial—investigators compared the standard anticoagulant (warfarin) for patients with atrial fibrillation (AF) at risk of stroke with a new agent, dabigatran.⁴ In the methods section of the abstract and the statistical analysis section of the main body, the authors clearly indicated that this was a noninferiority trial. They began by referring to the existing evidence of warfarin’s effectiveness, then detailed the qualities that make warfarin cumbersome to use, including the need for frequent laboratory monitoring. This was followed by evidence that many patients stop taking warfarin and that even for those who persist with treatment, adequate anticoagulation is difficult to maintain.

The authors went on to state that because dabigatran requires no long-term monitoring, it is easier to use. Therefore, if dabigatran could be shown to be no worse than warfarin in preventing strokes, it would be a reasonable alternative, leaving no doubt that this was an appropriate noninferiority trial.

2. Is the noninferiority margin based on clinical judgment and statistical reasoning?

The noninferiority margin should be based on clinical judgment as to how effective a new treatment must be in order to be declared not clinically inferior to the standard treatment. This can be based on several factors, including the severity of the outcome and the expected advantages of the new treatment. The margin should also take into account the size of the standard treatment’s effect vs placebo. In RELY, for example, the authors noted that the noninferiority margin was based on the desire to preserve at least 50% of the lower limit of the confidence interval (CI) of warfarin’s estimated effect; this was done using data from a previously published meta-analysis of 6 trials comparing warfarin with placebo for stroke prevention in patients with AF.^4-6

3. Are the hypothesis and statistical analysis formulated correctly?

The clinical hypothesis in a noninferiority trial is that the new treatment is not worse than the standard treatment by a prespecified margin; therefore, the statistical null hypothesis to be tested is that the new treatment is worse than the reference treatment by more than that margin. Rejecting a true null hypothesis (for example, because the P value is <.05) is known as a type l error. In this setting, making a type I error would mean accepting a new treatment that is truly worse than the standard by at least the specified margin. Failure to reject a false null hypothesis is known as a type II error, which in this case would mean failing to identify a new treatment that is truly noninferior to the standard.⁷

In RE-LY, the authors stated that the upper limit of the one-sided 97.5% CI for the relative risk of a stroke with dabigatran vs warfarin had to fall below 1.46.⁴ (This is the same as testing the null hypothesis that the hazard ratio is ≥1.46.) Thus, the hypothesis was formulated correctly.

4. Is the sample size appropriate and justified?

The sample size in a noninferiority trial should provide high power to reject the null hypothesis that the difference (or relative risk) between groups is equal to or greater than the noninferiority margin under some clinically meaningful assumption about the true difference (or absolute risk reduction) between groups. A true difference of 0 (or a relative risk of 1) is typically assumed for sample size calculation. However, assuming that the new treatment is truly slightly better or slightly worse than the standard may be clinically appropriate in some cases. This would indicate a need for a smaller or larger sample size, respectively, than that required under the usual assumption of no difference.

When the justification for the sample size in a noninferiority trial is not provided or the number of participants is based on an inappropriate approach (eg, using superiority trial calculations for a noninferiority trial), questions about the quality of the trial arise. The primary concern is whether the noninferiority margin was actually selected before the trial began, as it should have been. And if the researchers used overly optimistic assumptions about the efficacy of the new treatment relative to the standard therapy, the failure to rule out the margin could be misleading. (As with superiority trials that fail to reject the null hypothesis, post hoc power calculations should be avoided.) After the study has ended, the resulting CIs should be used to evaluate whether the study was large enough to adequately assess the relative effectiveness of the treatments.

The RE-LY trial calculated the sample size that was expected to provide 84% power to rule out the prespecified hazard ratio of 1.46, assuming a true event rate of 1.6% per year (presumably for both groups), a recruitment period of 2 years, and at least one year of follow-up. The sample size was subsequently increased from 15,000 to 18,000 to maintain power in case of a low event rate.^4,5

5. Is the noninferiority trial as similar as possible to the trial(s) comparing the standard treatment with placebo?

Characteristics of participants, setting, reference treatment, and outcomes used in a noninferiority trial should be as close as possible to those in the trial(s) comparing the treatment with placebo. This is known as the constancy assumption, and it is key to researchers’ ability to draw a conclusion about noninferiority.

The trials used to calculate the noninferiority margin and the RE-LY trial itself involved similar populations of patients with AF, and the outcome (stroke) was similar.

6. Is a per protocol analysis reported in the results?

In randomized controlled superiority trials, the participants should be analyzed in the groups to which they were originally allocated, regardless of whether they adhered to treatment during the entire follow-up period. Such intention-to-treat (ITT) analysis is important because it provides a more conservative estimate of treatment effect—taking into account that some people who are offered treatment will not accept it and others will discontinue treatment. An ITT analysis therefore tends to minimize treatment effects compared with a “per protocol” analysis, in which participants are analyzed according to the treatment they actually received and are often removed from the analysis if they discontinue or do not adhere to treatment.

Intention-to-treat analysis is important because it provides a more conservative estimate of treatment effect.In noninferiority trials, if patients in the intervention group cross over to the standard treatment group or those in the standard treatment group have poor adherence, an ITT analysis can increase the risk of wrongly claiming noninferiority.⁷ Therefore, a per protocol analysis should be included—and indeed may be preferable.

In RE-LY, ITT analyses were reported, and complete follow-up data were available for 99.9% of patients. However, the rates of treatment discontinuation at one year were about 15% for those on dabigatran and 10% for the warfarin group, and 21% and 17%, respectively, at 2 years.^4,5 If the new treatment were truly less efficacious than the standard treatment, these moderate discontinuation rates could lead to more similar rates of stroke in the 2 groups than would be expected with higher continuation rates, biasing results towards the alternative of noninferiority. Although the original publication of trial results did not include a per protocol analysis, the RE-LY authors later reported that a per protocol analysis yielded similar results to the ITT analysis.

7. Are the overall design and execution of the trial high quality?

Because a poor quality noninferiority trial can appear to demonstrate noninferiority, looking at such studies critically is crucial. Appropriate randomization, concealed allocation, masking, and careful attention to participant flow must all be assessed.^2,3

To continue with our example, the RE-LY trial was well conducted. Randomization was performed centrally via an automated telephone system and 2 doses of dabigatran were administered in a masked fashion, while warfarin was open-label. Remarkably, follow-up was achieved for 99.9% of participants over a median of 2 years, and independent adjudicators masked to treatment group assessed outcomes.^4,5

CORRESPONDENCE
Anne Mounsey, MD, UNC Chapel Hill Department of Family Medicine, 590 Manning Drive, CB 7595, Chapel Hill, NC 27590; [email protected]

The traditional clinical trial, designed to test whether a new treatment is better than a placebo or another active treatment, is known as a “superiority” trial—although rarely labeled as such. In contrast, the goal of a noninferiority trial is simply to demonstrate that a new treatment is not substantially less effective than the standard therapy.

Such trials are useful when a new therapy is thought to be safer, easier to administer, or less costly than the existing treatment, but not necessarily more effective. And, because it would be unethical to randomize patients with a serious condition for which there already is an effective treatment to placebo, a noninferiority trial is another means of determining if the new treatment is effective.

Noninferiority trials have unique design features and methodology and require a different analysis than traditional superiority trials. Yet many physicians know far less about them; many investigators appear to be less than proficient, as well. A review of 116 noninferiority trials and 46 equivalence trials found that only 20% fulfilled generally accepted quality criteria.¹ To improve the quality of noninferiority trials, the CONSORT (Consolidated Standards of Reporting Trials) Group has published a checklist for trial design and reporting standards.^2,3 Based on this checklist, we came up with 7 key questions to consider when evaluating a noninferiority trial. In the pages that follow, you’ll also find an at-a-glance guide (TABLE) and a methodology review using a hypothetical case (page E7).

1. Is a noninferiority trial appropriate?

The introduction to a noninferiority trial should provide the rationale for this design and the absence of a placebo control group. Look for a review of the evidence of the efficacy of the reference treatment that placebo-controlled trials have revealed, along with the effect size. The advantages of the new treatment over the standard treatment—eg, fewer adverse effects, easier administration, or lower cost—should be discussed, as well.

In the Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY)—a prominent noninferiority trial—investigators compared the standard anticoagulant (warfarin) for patients with atrial fibrillation (AF) at risk of stroke with a new agent, dabigatran.⁴ In the methods section of the abstract and the statistical analysis section of the main body, the authors clearly indicated that this was a noninferiority trial. They began by referring to the existing evidence of warfarin’s effectiveness, then detailed the qualities that make warfarin cumbersome to use, including the need for frequent laboratory monitoring. This was followed by evidence that many patients stop taking warfarin and that even for those who persist with treatment, adequate anticoagulation is difficult to maintain.

The authors went on to state that because dabigatran requires no long-term monitoring, it is easier to use. Therefore, if dabigatran could be shown to be no worse than warfarin in preventing strokes, it would be a reasonable alternative, leaving no doubt that this was an appropriate noninferiority trial.

2. Is the noninferiority margin based on clinical judgment and statistical reasoning?

The noninferiority margin should be based on clinical judgment as to how effective a new treatment must be in order to be declared not clinically inferior to the standard treatment. This can be based on several factors, including the severity of the outcome and the expected advantages of the new treatment. The margin should also take into account the size of the standard treatment’s effect vs placebo. In RELY, for example, the authors noted that the noninferiority margin was based on the desire to preserve at least 50% of the lower limit of the confidence interval (CI) of warfarin’s estimated effect; this was done using data from a previously published meta-analysis of 6 trials comparing warfarin with placebo for stroke prevention in patients with AF.^4-6

3. Are the hypothesis and statistical analysis formulated correctly?

The clinical hypothesis in a noninferiority trial is that the new treatment is not worse than the standard treatment by a prespecified margin; therefore, the statistical null hypothesis to be tested is that the new treatment is worse than the reference treatment by more than that margin. Rejecting a true null hypothesis (for example, because the P value is <.05) is known as a type l error. In this setting, making a type I error would mean accepting a new treatment that is truly worse than the standard by at least the specified margin. Failure to reject a false null hypothesis is known as a type II error, which in this case would mean failing to identify a new treatment that is truly noninferior to the standard.⁷

In RE-LY, the authors stated that the upper limit of the one-sided 97.5% CI for the relative risk of a stroke with dabigatran vs warfarin had to fall below 1.46.⁴ (This is the same as testing the null hypothesis that the hazard ratio is ≥1.46.) Thus, the hypothesis was formulated correctly.

4. Is the sample size appropriate and justified?

The sample size in a noninferiority trial should provide high power to reject the null hypothesis that the difference (or relative risk) between groups is equal to or greater than the noninferiority margin under some clinically meaningful assumption about the true difference (or absolute risk reduction) between groups. A true difference of 0 (or a relative risk of 1) is typically assumed for sample size calculation. However, assuming that the new treatment is truly slightly better or slightly worse than the standard may be clinically appropriate in some cases. This would indicate a need for a smaller or larger sample size, respectively, than that required under the usual assumption of no difference.

When the justification for the sample size in a noninferiority trial is not provided or the number of participants is based on an inappropriate approach (eg, using superiority trial calculations for a noninferiority trial), questions about the quality of the trial arise. The primary concern is whether the noninferiority margin was actually selected before the trial began, as it should have been. And if the researchers used overly optimistic assumptions about the efficacy of the new treatment relative to the standard therapy, the failure to rule out the margin could be misleading. (As with superiority trials that fail to reject the null hypothesis, post hoc power calculations should be avoided.) After the study has ended, the resulting CIs should be used to evaluate whether the study was large enough to adequately assess the relative effectiveness of the treatments.

The RE-LY trial calculated the sample size that was expected to provide 84% power to rule out the prespecified hazard ratio of 1.46, assuming a true event rate of 1.6% per year (presumably for both groups), a recruitment period of 2 years, and at least one year of follow-up. The sample size was subsequently increased from 15,000 to 18,000 to maintain power in case of a low event rate.^4,5

5. Is the noninferiority trial as similar as possible to the trial(s) comparing the standard treatment with placebo?

Characteristics of participants, setting, reference treatment, and outcomes used in a noninferiority trial should be as close as possible to those in the trial(s) comparing the treatment with placebo. This is known as the constancy assumption, and it is key to researchers’ ability to draw a conclusion about noninferiority.

The trials used to calculate the noninferiority margin and the RE-LY trial itself involved similar populations of patients with AF, and the outcome (stroke) was similar.

6. Is a per protocol analysis reported in the results?

In randomized controlled superiority trials, the participants should be analyzed in the groups to which they were originally allocated, regardless of whether they adhered to treatment during the entire follow-up period. Such intention-to-treat (ITT) analysis is important because it provides a more conservative estimate of treatment effect—taking into account that some people who are offered treatment will not accept it and others will discontinue treatment. An ITT analysis therefore tends to minimize treatment effects compared with a “per protocol” analysis, in which participants are analyzed according to the treatment they actually received and are often removed from the analysis if they discontinue or do not adhere to treatment.

Intention-to-treat analysis is important because it provides a more conservative estimate of treatment effect.In noninferiority trials, if patients in the intervention group cross over to the standard treatment group or those in the standard treatment group have poor adherence, an ITT analysis can increase the risk of wrongly claiming noninferiority.⁷ Therefore, a per protocol analysis should be included—and indeed may be preferable.

In RE-LY, ITT analyses were reported, and complete follow-up data were available for 99.9% of patients. However, the rates of treatment discontinuation at one year were about 15% for those on dabigatran and 10% for the warfarin group, and 21% and 17%, respectively, at 2 years.^4,5 If the new treatment were truly less efficacious than the standard treatment, these moderate discontinuation rates could lead to more similar rates of stroke in the 2 groups than would be expected with higher continuation rates, biasing results towards the alternative of noninferiority. Although the original publication of trial results did not include a per protocol analysis, the RE-LY authors later reported that a per protocol analysis yielded similar results to the ITT analysis.

7. Are the overall design and execution of the trial high quality?

Because a poor quality noninferiority trial can appear to demonstrate noninferiority, looking at such studies critically is crucial. Appropriate randomization, concealed allocation, masking, and careful attention to participant flow must all be assessed.^2,3

To continue with our example, the RE-LY trial was well conducted. Randomization was performed centrally via an automated telephone system and 2 doses of dabigatran were administered in a masked fashion, while warfarin was open-label. Remarkably, follow-up was achieved for 99.9% of participants over a median of 2 years, and independent adjudicators masked to treatment group assessed outcomes.^4,5

CORRESPONDENCE
Anne Mounsey, MD, UNC Chapel Hill Department of Family Medicine, 590 Manning Drive, CB 7595, Chapel Hill, NC 27590; [email protected]