Cross-National Prevalence of Risk of Depression Among Older Adults: Are There Gender and Age Differences?

Introduction: Depression among older adults is becoming one of the most pressing issues in public health, yet it remains one of the less recognized problems and is one of the leading causes of disability in this age group. Among older adults, depression often goes unrecognized. Although depression has been widely studied, there are still several unanswered questions regarding country/context differences, gender, age, and cohort differences, as most studies have been conducted in specific contexts. This study aims to examine cross-national gender gaps in the prevalence of depression risk from 2000 to 2020, controlling for age and cohort effects.

Methods: A random-effects model with robust standard errors was conducted to examine depression cases and predict probabilities. Our analysis involved a sample of 248,559 individuals aged 50 and older from 35 countries with at least 3 waves of reported measures, spanning the years 2000 to 2020. The data sources are: Chilean Social Protection Survey, China Health and Retirement Longitudinal Study (CHARLS), Health and Retirement Study in the United States (HRS), Mexican Health and Aging Study, Survey of Health, Ageing and Retirement in Europe (SHARE; Austria, Belgium, Czech Republic, Denmark, Estonia, France, Germany, Israel, Italy, Netherlands, Poland, Slovenia, Spain, Sweden, Switzerland). Our variable of interest is risk of depression, we used cut-points for risk of depression based on the depression scales used in the different surveys. For instance, HRS uses the CESD-8 scale developed by the Center for Epidemiologic Studies, which contains 8 items – did you feel depressed, that everything you did was an effort, sleep is restless, happy, lonely, enjoy life and sad. In SHARE data, researchers used the EURO-D scale developed by a European consortium. This scale includes 12 items that asked about depression, pessimism, suicidality, guilt, sleep quality, interest, irritability, appetite, fatigue, concentration (on reading or entertainment), enjoyment, and tearfulness measured as binary variables (1=yes; 0=no). Clinically significant depression was assessed with a score of 4+. The same procedure will be used for other surveys, based on the scales they used. For instance, China uses the CESD-10 scale measured in a 4-point Likert scale as used in HRS, while Chile uses the GDS-15, which considers 15 depressive symptoms measured as binary variables (1=yes;0=no) with a cut point for risk of depression of 5 or higher. We explored gender differences over time, and cohort effects.

Results: Throughout the examined period, the average predicted probability of depression was 24.01% for males and 35.70% for females. This translates to a consistent gender gap of 11.69 percentage points. Notably, while the gender gap in depression probability remained stable over time, both males and females exhibited an upward trend in depression probability during this period – about 3% and 5% respectively.