EVIDENCE FOR A POSITIVE LINEAR RELATION BETWEEN BLOOD-PRESSURE AND MORTALITY IN ELDERLY PEOPLE

Many studies of blood pressure in the elderly have found higher death rates in groups with the lowest blood pressure than in those with inte rmediate values. In a large community study, we examined whether these findings are real or artifacts of short follow-up, co-morbidity, or l ow blood pressure in people near death. In 1982-83, we assessed drug u se, medical history, disability, physical function, and blood pressure in 3657 residents of East Boston, Massachusetts, aged 65 and older. W e identified all deaths (1709) up to 1992 and followed up survivors fo r an average of 10.5 (range 9.5-11.0) years. After adjustment for conf ounding variables (including frailty and disorders such as congestive heart failure and myocardial infarction) and exclusion of deaths withi n the first 3 years of follow-up, higher systolic pressure predicted l inear increases in cardiovascular (p<0.0001) and total (p<0.0007) mort ality. Higher diastolic pressure predicted increases in cardiovascular (p=0.006) but not total (p=0.48) mortality. These results differed fr om those for the first 3 years, during which groups with the lowest sy stolic and diastolic pressures had the highest death rates. In the lon g term, lower blood pressure in old age, as in middle age, is associat ed with better survival. Short-term findings may differ because of ass ociations of co-morbidity and frailty with blood pressure near death. Overall, the findings support recommendations to treat high blood pres sure in elderly people.