Hypertension: AI-assisted follow-up appears to improve blood pressure without isolating the algorithm’s role

Artificial intelligence-assisted support appears to improve blood pressure control in older adults with hypertension. In a randomized trial involving 200 patients, outcomes after one year were better than with usual care. But the intervention combines monitoring, alerts and proactive interventions: the available abstract does not allow the benefit to be attributed to AI alone.

An older person measures their blood pressure with a blood pressure monitor beside a digital monitoring interface.
Gadgets & TechEditorial illustration, not study data.

Enhanced support compared with usual care

The researchers randomly assigned participants to two groups of 100 patients. The control group received usual care. The other group received AI-assisted health management, including smart monitoring, early alerts and proactive interventions.

The main outcomes were systolic and diastolic blood pressure, along with the proportion of patients whose blood pressure remained below the thresholds of 140/90 mmHg. Follow-up lasted one year. The abstract does not specify the tools used, the frequency of contact or the contribution of healthcare professionals.

Favorable blood pressure results

After one year, the reported blood pressure control rate reached 88.66% in the AI-assisted group, compared with 64.89% with usual care. Mean systolic blood pressure was 128.49 mmHg versus 135.38 mmHg, and diastolic blood pressure was 77.49 mmHg versus 81.84 mmHg.

The abstract also reports better results for several health behaviors, knowledge about hypertension, self-management, beliefs about medication, quality of life and satisfaction. However, it does not provide the magnitude of these differences, making it impossible to assess their practical significance.

A result for the whole package, not AI in isolation

Randomization strengthens the value of the comparison. However, it does not distinguish the effect of the algorithm from that of more active support. This trial evaluates a package of services, not an AI application used on its own.

The results concern blood pressure and self-management indicators. The abstract does not report outcomes for heart attacks, strokes or mortality. Nor does it allow these observations to be generalized to all commercial hypertension monitoring tools.

Limitations to keep in mind

Only the supplied PubMed abstract was reviewed, not the full text. Losses to follow-up, denominators for control rates, medication adjustments, adverse effects and confidence intervals are not specified in this excerpt. Funding and conflicts of interest were not verified.

What we can take away

This trial supports the potential value of enhanced AI-assisted follow-up in older adults with hypertension. It demonstrates neither that AI alone is the effective component nor that a consumer tool would reproduce these results.

Sources and provenance

Partager cet article

X f in

An error or a source to report? Contact the editorial team.