Internal Medicine Resident McGill University Montréal, Nova Scotia, Canada
Background: A key element of patient-centered care and collaborative decision-making is accurate patient awareness of their medical diagnoses. This study aimed to assess the agreement between patient-reported and chart-documented diagnoses across various cardiovascular and metabolic conditions to identify gaps in patient awareness.
METHODS AND RESULTS: This study used data from the Courtois Signature Program. Adults (≥18 years) with cardiovascular conditions were recruited from the McGill University Health Centre. Participants completed a questionnaire reporting only medically confirmed diagnoses. A medical archivist independently extracted the chart data for each patient. Diagnoses were coded and analyzed using Cohen’s Kappa (κ) to assess agreement between patient-reported and chart-documented conditions. Analyses included both the entire sample and a subset with the presumed diagnosis, defined as documentation in either the medical chart or self-reported.
A total of 190 participants were recruited. Participants (mean age 61.9 ± 10.9 years) were predominantly male (58%), White (78%), and born in developed countries (78%), with post-secondary education (74%). Agreement varied by condition. Fair agreement was observed for angina (κ=0.246), dyslipidemia (κ=0.355), congestive heart failure (CHF; κ=0.393), and implantable cardioverter defibrillator (ICD; κ=0.222). Moderate agreement was found for coronary artery disease (CAD; κ=0.482), myocardial infarction (MI; κ=0.506), atrial fibrillation (AF; κ=0.499), hypertension (κ=0.401), and percutaneous coronary intervention (PCI; κ=0.461). Substantial agreement was observed for diabetes mellitus (DM; κ=0.803) and coronary artery bypass grafting (CABG; κ=0.716). All kappa values were statistically significant (p < 0.001). The highest agreement was for diabetes mellitus, reflecting strong concordance between the patient survey and chart documentation. In contrast, among those with the presumed diagnosis, agreement dropped notably. Negative kappa values, indicating systematic discordance between self-reported and documented diagnoses, were observed for angina (36 diagnosed, 19.4% agreement; κ=-0.526), hypertension (100 diagnosed, 46.0% agreement; κ=-0.353), and dyslipidemia (115 diagnosed, 47.0% agreement; κ=-0.324), all p< 0.001. Disagreement was also observed for CAD, MI, AF, CHF, ICD, and PCI (κ=-0.155 to -0.474; p< 0.05). The lowest agreement was seen for angina.
Conclusion: While patient knowledge aligns well with chart documents for conditions such as diabetes and procedures like CABG, there are notable differences for others, especially among diagnosed individuals. These findings underscore the need to improve patient education regarding their diagnoses. Further research is needed to identify optimal strategies for doing so.
