Although the prevalence of obesity continues to increase in Switzerland, the latest figures suggest a slowdown in the rate of increase. In order to elucidate whether this could be the onset of a trend reversal, we analyzed cross-sectional data by birth cohort. We assessed the prevalence of overweight+ (BMI > or =25 kg/m(2)) and obesity (BMI > or =30 kg/m(2)) in six population surveys with self-reported height and weight values (Switzerland, N = 68,829, 1982-2007, men (45%) and women (55%), aged 20-84 years) by 10-year birth cohorts (from the decade 1910-1919 through to 1970-1979). We found that increases in the prevalence of overweight+ and obesity occurred mainly in the cohort born 1930 to 1939, and again in the cohorts born 1960 to 1979. The accelerated increase in the prevalence of overweight+ in the youngest birth cohort and the lower prevalence in the oldest birth cohorts suggest that the current slowdown seen in Switzerland may not herald the onset of a trend reversal. As this example shows, simple comparisons of prevalence rates over time could provide a misleading picture of actual trends. Birth cohort analysis may offer a valuable alternative.