Evidence Table 4.3. Assessment of Study Quality in Studies of Treatment of Post Myocardial Infarction Depression (Question 4)

Article IDRepresentativenessBias and ConflictDescription Therapy and ManagementDescription ProtocolOutcomes and Follow-UpStatistical AnalysesConflicts
Dracup, 1991 35%33%34%34%34%34%34%
Crowe, 1996 80%38%59%48%53%51%52%
Frasure-Smith, 1997 85%88%86%87%87%87%87%
Roose, 1998 70%58%64%61%63%62%62%
Johnston, 1999 60%88%74%81%77%79%78%
McFarlane, 2001 75%88%81%84%83%84%83%
Berkman, 2003 100%63%81%72%77%74%75%
Swenson, 2003 90%100%95%98%96%97%97%
Brown, 1993 70%75%73%74%73%73%73%
Taylor, 1997 70%50%60%55%58%56%57%
Strik, 2000 85%83%84%84%84%84%84%
Glassman, 2002 95%88%91%89%90%90%90%

Representativeness: Percentage score was based on a total maximum score of 10 points. This included assessment of how well the study described the study setting and population (2 points), inclusion/exclusion criteria (2 points), non participating patients (2 points), patient characteristics at enrollment (2 patients), and whether the study used a consecutive series or randomly selected sample (2 patients).

Bias and Confounding: Percentage score was based on a total maximum score of 6 points. This included assessment of whether patients were randomly assigned to study groups (2 points), wether patient groups had important differences in patient characteristics (2 points), and whether there was blinding of test interpretation (2 patients)

Description of Therapy: Percentage score was based on a total maximum score of 10 points. This included assessment of how well the study described details of the cardiac therapy given (2 points), whether the study described details of the psychiatric treatment given (2 points), whether there was adequate description of other treatments given (2 points), whether description of the flow of participants was adequate (2 points), and whether assessment of adherence to therapy was given.

Description of Assessment: Percentage score was based on a total maximum score of 4 points. This included assessment of the description of methods used for initial diagnosis of depression (2 points), and the description of the interpretation criteria for a diagnosis of depression (2 points).

Outcomes and Follow up: Percentage score was based on a total maximum score of 12 points. This included assessment of whether the study reported numbers or reasons for withdrawals or those lost to follow up (2 points), the percentage of patients withdrawn or lost to follow up (2 points), whether cardiac outcome measures were defined (2 points), whether depression outcome measures were defined (2 points), whether the same tools for diagnosing depression were used for baseline and follow up (2 points), whether the length of follow-up was planned (2 points).

Statistical Analyses: Percentage score was based on a total maximum score of 8 points. This included assessment of whether statistical tests were clearly identified (2 points), whether loss to follow up was handled appropriately (2 points), whether adjustment was made for confounding (2 points), and whether confidence intervals were reported (2 patients).

Conflict of Interest: Percentage score was based on a total maximum score of 2 points. This involved determination of whether the study identified the sources of funding and involvement of the funding agency.

From: Appendix G. Evidence Tables

Cover of Post-Myocardial Infarction Depression
Post-Myocardial Infarction Depression.
Evidence Reports/Technology Assessments, No. 123.
Bush DE, Ziegelstein RC, Patel UV, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.