Total and cause-specific mortality before and after the onset of the Greek economic crisis: an interrupted time-series analysis.

BACKGROUND Greece was one of the countries hit the hardest by the 2008 financial crisis in Europe. Yet, evidence on the effect of the crisis on total and cause-specific mortality remains unclear. We explored whether the economic crisis affected the trend of overall and cause-specific mortality rates. METHODS We used regional panel data from the Hellenic Statistical Authority to assess mortality trends by age, sex, region, and cause in Greece between January, 2001, and December, 2013. We used Eurostat data to calculate monthly age-standardised mortality rates per 100 000 inhabitants for each region. Data were divided into two subperiods: before the crisis (January, 2001, to August, 2008) and after the onset of the crisis (September, 2008, to December, 2013). We tested for changes in the slope of mortality by doing an interrupted time-series analysis. FINDINGS Overall mortality continued to decline after the onset of the financial crisis (-0·065, 95% CI -0·080 to -0·049), but at a slower pace than before the crisis (-0·13, -0·15 to -0·10; trend difference 0·062, 95% CI 0·041 to 0·083; p<0·0001). The trend difference was more evident for females (0·087, 95% CI 0·064-0·11; p<0·0001) than for males (0·040, 0·013-0·066; p=0·007). Those aged at least 75 years experienced more negative effects (trend difference 0·056, 95% CI 0·042 to 0·071; p<0·0001) than did those aged 20-34 years, in whom mortality trends improved (-0·0074, -0·0089 to -0·0059; p<0·0001). Deaths by diseases of the circulatory system declined more slowly after the onset of compared with before the crisis (trend difference 0·043, 95% CI 0·024 to 0·063; p<0·0001), whereas deaths from vehicular accidents declined faster (-0·0062, -0·0090 to -0·0033; p<0·0001), most prominently among men aged 20-34 years (-0·0065, -0·0085 to -0·0044; p<0·0001). Conversely, deaths from suicides (trend difference 0·0021, 95% CI 0·00092-0·0033; p=0·002), diseases of the nervous system (0·0036, 0·0016-0·0056; p=0·002), and mental health problems (0·00073, 0·000047-0·0014 p=0·038) increased after the onset of the crisis. Also, deaths due to adverse events during medical treatment increased significantly after the onset of the crisis (trend difference 0·0020, 95% CI 0·0012-0·0028; p<0·0001). By comparing the expected values of the period after the onset of the crisis with extrapolated values based on the period before the crisis, we estimate that an extra 242 deaths per month occurred after the onset of the crisis. INTERPRETATION Mortality trends have been interrupted after the onset of compared with before the crisis, but changes vary by age, sex, and cause of death. The increase in deaths due to adverse events during medical treatment might reflect the effects of deterioration in quality of care during economic recessions. FUNDING None.