|| Checking for direct PDF access through Ovid
It is debatable whether the body mass index (BMI) value that is associated with the lowest all-cause mortality has increased over time. Such an increase might indicate that the WHO BMI categories for normal weight and overweight need to be revised over time. This study assessed whether the BMI associated with the lowest all-cause mortality increased over time among US adults in the National Health Interview Survey.This survey with up to 25 years of follow-up included 901 197 participants, aged 20 to 79 years with baseline BMI between 18.5 and 39.9 kg m-2 and known mortality status between the original survey dates and 31 December 2011. The BMI values associated with the lowest mortality were estimated for different survey time periods: 1986-1989, 1990-1994, 1995-1999, 2000-2004 and 2000-2009.Among 901 197 participants, 130 495 died during over 14.5 million person-years of follow-up. There was a U-shaped BMI-mortality association in all the survey periods. When we used the original follow-up durations, the BMI associated with the lowest mortality increased monotonically from 23.9 kg m-2 in 1986-1989 to 28.6 kg m-2 in 2005-2009. When using a fixed follow-up duration of 5, 10, 15 or 20 years, we found no monotonic increasing pattern in the BMI associated with the lowest mortality. With duration of 15 years, the estimates of BMI associated with lowest mortality appeared to be relatively stable over time among both the general population and never smokers free from cardiovascular disease and cancer at baseline.With a fixed long-term follow-up duration, the BMI value associated with the lowest mortality remains relatively stable over time.