A cluster-randomized trial of 33,995 adults with uncontrolled hypertension in rural China found that intensive, non-physician-led blood pressure treatment reduced the risk of all-cause dementia by 15 percent over four years, according to results published in Nature Medicine on April 21, 2025. In the China Rural Hypertension Control Project Phase 3 (CRHCP-3), the risk ratio for dementia was 0.85 (95 percent confidence interval, 0.76 to 0.95; P = 0.0035) in villages assigned to the intervention compared with villages assigned to usual care.

The trial randomized 163 villages, comprising 17,407 people, to an intervention arm and 163 villages, comprising 16,588 people, to usual care. In the intervention villages, trained non-physician community health workers, not doctors, started and adjusted blood pressure medication using a simple stepped protocol aimed at getting systolic pressure below 130 mm Hg and diastolic pressure below 80 mm Hg, with a physician supervising from a distance. Over 48 months, systolic pressure fell 22.0 mm Hg more, and diastolic pressure fell 9.3 mm Hg more, in the intervention group than in usual care, the researchers reported.

The trial was open-label: participants and the health workers treating them knew who was assigned to which group, though the researchers say the staff assessing for dementia were kept blinded to group assignment. Independent commentary gathered by the Alzheimer's research news site Alzforum noted that medication adherence differed sharply between arms, 88.0 percent in the intervention group versus 66.4 percent in usual care, partly because intervention participants received discounted or free medication, a gap that could account for some of the measured benefit. The same trial also reported fewer cases of mild cognitive impairment without dementia in the intervention group and no increase in serious adverse events; serious adverse events were in fact less frequent in the intervention villages, with a risk ratio of 0.94.

The result adds to a mixed evidence base on blood pressure and dementia risk that includes the earlier SPRINT MIND trial in the United States, which found a reduction in mild cognitive impairment but not a statistically significant reduction in dementia itself. CRHCP-3 is now among the largest and longest trials to report a significant dementia benefit, and its authors argue the model, using non-physician workers in a low-resource, real-world primary care setting, could be scaled elsewhere.