User login
Several depression screening tools work equally well
ABSTRACT
BACKGROUND: Depressive disorders are common in primary care, but the optimal approach for diagnosis remains controversial. This information summary compared various depression case-finding instruments suitable for the office setting.
POPULATION STUDIED: This summary included 28 studies of depression case finding instruments with more than 25,000 screened patients drawn from Veterans Affairs clinics, academic practices, health maintenance organizations, and community practices. Given the variety of study settings, the results of the study are probably generalizable to the average family practice. More information about participants’ age, sex, ethnicity, symptom severity, and comorbid conditions would allow better assessment of the performance of case-finding instruments in specific groups of patients of particular importance in primary care such as women, patients of color, and the chronically medically ill.
STUDY DESIGN AND VALIDITY: The investigators summarized information from studies that evaluated case-finding instruments in the primary care setting. Using terms featuring depressive disorder or depression, the authors searched MEDLINE, a specialized depression trial registry, and bibliographies of selected articles for English-language literature published from 1970 to 2000. They included studies with at least 100 subjects, that used instruments with a depression-specific component and low literacy requirements and complexity, and that used a standard interview to make an independent and masked criterion-based diagnosis of depression. Two independent reviewers abstracted the studies and evaluated study quality. If necessary, original authors were contacted for additional information. Established cut points were used to calculate average likelihood ratios, weighted for study precision and 2-phase assessment if applicable. A summary effectiveness score was used to assess for heterogeneity.
OUTCOMES MEASURED: The major outcome was the average positive and negative likelihood ratios for each case-finding instrument. Outcomes important in primary care that were not addressed included clinician and patient satisfaction and effects of screening on office flow and patient outcomes.
RESULTS: The 28 published studies evaluated 11 different instruments, including the Beck Depression Inventory, the Zung Self-Assessment Depression Scale, and the Primary Care Evaluation of Mental Disorders. Of these, 15 studies met quality assessment standards and were included in the final analysis. With regard to instrument performance in detecting depression, the median positive likelihood ratio was 3.3 (range, 2.3–12.2), and the median negative likelihood ratio was 0.19 (range, 0.14–0.35). No significant difference was found between screens. However, 4 instruments, including the Beck, demonstrated statistically significant variation across studies, which the authors attributed to either variability of study population or design.
This study provides fair evidence that depression case-finding instruments perform similarly and fairly well in detecting and ruling out depression with a wide variety of outpatients. Clinicians should thus choose a specific case-finding instrument based on other characteristics, such as ease of use, response format, and the need to screen for other psychiatric diseases. The Patient Health Questionnaire best meets these criteria, with only 9 questions, a simple format and modules for other illnesses. For screening, the single question, “Have you felt depressed or sad much of the time in the last year?” performs well. It should be kept in mind, however, that this study did not address the impact of using case-finding screens on patient flow, the effectiveness of case finding on outcomes from depressive disease, or whether routine screening for depression is merited.
ABSTRACT
BACKGROUND: Depressive disorders are common in primary care, but the optimal approach for diagnosis remains controversial. This information summary compared various depression case-finding instruments suitable for the office setting.
POPULATION STUDIED: This summary included 28 studies of depression case finding instruments with more than 25,000 screened patients drawn from Veterans Affairs clinics, academic practices, health maintenance organizations, and community practices. Given the variety of study settings, the results of the study are probably generalizable to the average family practice. More information about participants’ age, sex, ethnicity, symptom severity, and comorbid conditions would allow better assessment of the performance of case-finding instruments in specific groups of patients of particular importance in primary care such as women, patients of color, and the chronically medically ill.
STUDY DESIGN AND VALIDITY: The investigators summarized information from studies that evaluated case-finding instruments in the primary care setting. Using terms featuring depressive disorder or depression, the authors searched MEDLINE, a specialized depression trial registry, and bibliographies of selected articles for English-language literature published from 1970 to 2000. They included studies with at least 100 subjects, that used instruments with a depression-specific component and low literacy requirements and complexity, and that used a standard interview to make an independent and masked criterion-based diagnosis of depression. Two independent reviewers abstracted the studies and evaluated study quality. If necessary, original authors were contacted for additional information. Established cut points were used to calculate average likelihood ratios, weighted for study precision and 2-phase assessment if applicable. A summary effectiveness score was used to assess for heterogeneity.
OUTCOMES MEASURED: The major outcome was the average positive and negative likelihood ratios for each case-finding instrument. Outcomes important in primary care that were not addressed included clinician and patient satisfaction and effects of screening on office flow and patient outcomes.
RESULTS: The 28 published studies evaluated 11 different instruments, including the Beck Depression Inventory, the Zung Self-Assessment Depression Scale, and the Primary Care Evaluation of Mental Disorders. Of these, 15 studies met quality assessment standards and were included in the final analysis. With regard to instrument performance in detecting depression, the median positive likelihood ratio was 3.3 (range, 2.3–12.2), and the median negative likelihood ratio was 0.19 (range, 0.14–0.35). No significant difference was found between screens. However, 4 instruments, including the Beck, demonstrated statistically significant variation across studies, which the authors attributed to either variability of study population or design.
This study provides fair evidence that depression case-finding instruments perform similarly and fairly well in detecting and ruling out depression with a wide variety of outpatients. Clinicians should thus choose a specific case-finding instrument based on other characteristics, such as ease of use, response format, and the need to screen for other psychiatric diseases. The Patient Health Questionnaire best meets these criteria, with only 9 questions, a simple format and modules for other illnesses. For screening, the single question, “Have you felt depressed or sad much of the time in the last year?” performs well. It should be kept in mind, however, that this study did not address the impact of using case-finding screens on patient flow, the effectiveness of case finding on outcomes from depressive disease, or whether routine screening for depression is merited.
ABSTRACT
BACKGROUND: Depressive disorders are common in primary care, but the optimal approach for diagnosis remains controversial. This information summary compared various depression case-finding instruments suitable for the office setting.
POPULATION STUDIED: This summary included 28 studies of depression case finding instruments with more than 25,000 screened patients drawn from Veterans Affairs clinics, academic practices, health maintenance organizations, and community practices. Given the variety of study settings, the results of the study are probably generalizable to the average family practice. More information about participants’ age, sex, ethnicity, symptom severity, and comorbid conditions would allow better assessment of the performance of case-finding instruments in specific groups of patients of particular importance in primary care such as women, patients of color, and the chronically medically ill.
STUDY DESIGN AND VALIDITY: The investigators summarized information from studies that evaluated case-finding instruments in the primary care setting. Using terms featuring depressive disorder or depression, the authors searched MEDLINE, a specialized depression trial registry, and bibliographies of selected articles for English-language literature published from 1970 to 2000. They included studies with at least 100 subjects, that used instruments with a depression-specific component and low literacy requirements and complexity, and that used a standard interview to make an independent and masked criterion-based diagnosis of depression. Two independent reviewers abstracted the studies and evaluated study quality. If necessary, original authors were contacted for additional information. Established cut points were used to calculate average likelihood ratios, weighted for study precision and 2-phase assessment if applicable. A summary effectiveness score was used to assess for heterogeneity.
OUTCOMES MEASURED: The major outcome was the average positive and negative likelihood ratios for each case-finding instrument. Outcomes important in primary care that were not addressed included clinician and patient satisfaction and effects of screening on office flow and patient outcomes.
RESULTS: The 28 published studies evaluated 11 different instruments, including the Beck Depression Inventory, the Zung Self-Assessment Depression Scale, and the Primary Care Evaluation of Mental Disorders. Of these, 15 studies met quality assessment standards and were included in the final analysis. With regard to instrument performance in detecting depression, the median positive likelihood ratio was 3.3 (range, 2.3–12.2), and the median negative likelihood ratio was 0.19 (range, 0.14–0.35). No significant difference was found between screens. However, 4 instruments, including the Beck, demonstrated statistically significant variation across studies, which the authors attributed to either variability of study population or design.
This study provides fair evidence that depression case-finding instruments perform similarly and fairly well in detecting and ruling out depression with a wide variety of outpatients. Clinicians should thus choose a specific case-finding instrument based on other characteristics, such as ease of use, response format, and the need to screen for other psychiatric diseases. The Patient Health Questionnaire best meets these criteria, with only 9 questions, a simple format and modules for other illnesses. For screening, the single question, “Have you felt depressed or sad much of the time in the last year?” performs well. It should be kept in mind, however, that this study did not address the impact of using case-finding screens on patient flow, the effectiveness of case finding on outcomes from depressive disease, or whether routine screening for depression is merited.