Physicians commonly express concern that tight control of blood pressure (BP) may increase risk for hypotension, falls and fractures. However, data to support this are sparse. The objective was to compare the rates of falls and non-spine fractures in the ACCORD Blood Pressure Trial, in which 4733 patients with type 2 diabetes (T2DM) were randomly assigned to standard (STD) or intensive (INT) systolic BP goals: 130–139 vs <120 mm Hg. The mean BP at baseline was 139/76; after the first year of the trial, mean BP was 119/64 in the INT group and 134/71 in the STD group. A subgroup (N = 3282) was followed annually for self-reported falls and fractures (fx). Fx were confirmed by central blinded review of medical records and radiology reports. Fall rates were compared using negative binomial regression models and fx risk was compared using Cox proportional hazards models. Mean follow-up was 4.9 years. At enrollment, the mean age of the subjects was 62 years, 44% were female, 31% were minorities and 34% had cardiovascular disease. The rate of falls was 70.0/100 person-years. There were 273 participants with at least one confirmed non-spine fx (including 19 hip, 63 ankle, 29 foot, 34 humerus, and 25 wrist fx.) Compared to the STD group, the RR for falls in the INT group was 0.81 (95% CI 0.53–1.25, P = 0.34); this did not differ in subgroups defined by baseline characteristics (age, sex, ethnicity, diabetes treatments, or co-morbidities). The non-spine fx rate was significantly lower in the INT than in the STD group (HR 0.78, 95% CI 0.62–0.99, P = 0.046). Although none of the individual fx differed significantly between the treatment groups, all HRs were <1 and there was a trend toward fewer ankle fx in the INT group (HR 0.46, 95% CI 0.21–1.02, P = 0.054).
Conclusions: In older patients with T2DM, intensive control of systolic BP did not result in an increased rate of falls. Fewer INT group participants developed non-spine fx; the mechanism for this finding will be explored in future analyses.