Risk Factors for Depressive Symptoms in Korean Adult Stroke Survivors: The Korea National Health and Nutrition Examination Survey IV-VII (2007-2018)

Sex Differences in Depressive Symptoms in 1308 Patients Post-Stroke at Entry to Cardiac Rehabilitation

PURPOSE

The objective of this study was to determine whether a sex difference exists in the prevalence of post-stroke depressive symptoms (PSDS) at entry to cardiac rehabilitation (CR) and to determine the correlates of PSDS in all patients, and in women and men separately.

METHODS

People post-stroke at entry to CR from database records (2006-2017) were included. Bivariate analyses identified PSDS correlates (≥16 on the Center for Epidemiologic Studies Depression Scale) in all patients and women and men separately.

RESULTS

Patients (n = 1308, 28.9% women), mean age of 63.9 ± 12.9 yr, were 24.2 ± 9.9 mo post-stroke at CR entry. Among all patients, 30.0% had PSDS. A greater proportion of women than men had PSDS (38.6 vs 26.6%; P < .001). Correlates of PSDS in all patients were sex (women) (OR = 1.6: 95% CI, 1.14-2.12), being unemployed, ≤60 yr old, prescribed antidepressant medication, having lower cardiorespiratory fitness (peak oxygen uptake [V̇ o2peak ]), chronic obstructive pulmonary disease (COPD), higher body mass index (BMI), no transient ischemic attack, and longer time from stroke to CR entry (>12 mo). Correlates in women were being obese (BMI ≥ 30), 51-70 yr old, prescribed antidepressant medication, and not married. Correlates in men were being ≤60 yr old, unemployed, prescribed antidepressant medication, having lower V̇ o2peak , sleep apnea, COPD, and no hypertension.

CONCLUSION

Women were disproportionately affected by PSDS at entry to CR in bivariate and multivariable analyses. Women and men had mostly unique correlates of PSDS, indicating tailored strategies to address PSDS are required. PSDS disproportionately affected patients with longer delay to CR entry, suggesting efforts should target timely referral to facilitate earlier and repeated assessments and management.

Measurement invariance of the patient health questionnaire-9 depression scale in a nationally representative population-based sample

The Patient Health Questionnaire-9 (PHQ-9) is widely used to measure the severity of depressive symptoms and to screen for depressive disorder, but its measurement invariance has received little research attention. The aim of this study was to assess the measurement invariance of the PHQ-9 across various sociodemographic and medical-condition groups. The structural validity and internal consistency of the PHQ-9 were also assessed as the prerequisite properties for measurement invariance. This study was conducted using data from the Korea National Health and Nutrition Examination Survey. The included participants comprised 5,347 people older than 19 years. Exploratory graph analysis (EGA) and confirmatory factor analysis (CFA) were performed to determine structural validity, and the omega coefficient (ω ) was used to determine internal consistency. Measurement invariance (configural, metric, and scalar invariance) was evaluated using multigroup confirmatory factor analysis (MGCFA). The single structural model of the PHQ-9 that was validated by EGA was also satisfactory with fit indices of χ2 = 770.765 (p < 0.001), CFI = 0.944, SRMR = 0.040, and RMSEA = 0.076 (90% CI = 0.072-0.081). The ω of the PHQ-9 was 0.812, implying satisfactory internal consistency. The one-factor PHQ-9 had equivalent overall structure, factor loadings, and item intercepts across age groups, suggesting invariance across ages. Partial scalar invariance was demonstrated across sex and marital-status groups. Partial metric and scalar invariance were supported across education groups. Scalar invariance was supported among all of the medical-condition (hypertension, diabetes, cancer, arthritis, asthma, and heart disease) groups. Overall, the measurement invariance of the one-factor PHQ-9 was empirically supported across sociodemographic and medical-condition groups. The PHQ-9 can be reliably used to compare the severity of depressive symptoms across these groups in research and practice.