Skip navigation

Treatment of hypertension in the elderly

Bandolier is keen on numbers-needed-to-treat (NNTs) as a simple way of describing complicated information. There are, however, times when examining the complicated detail is as important as looking at the simple take home message. Such is the case with a paper on hypertension in the elderly [1].

Take home message

For readers whose minds feel stretched enough already, the message of this excellent systematic review and meta-analysis is that treatment of hypertension in elderly people is a treatment with "strong, consistent and convincing" evidence of effectiveness. Only eighteen subjects have to be treated to prevent one vascular (cerebrovascular or cardiovascular) event (NNT = 18).

Systematic review

This review [1] involved 13 trials of 16,500 elderly patients aged 60 years or older, lasting at least one year and which evaluated effects of drug treatment on morbidity or mortality outcomes.

Coronary heart disease mortality included fatal myocardial infarctions and sudden or rapid cardiac death; morbidity included non fatal infarctions. Cerebrovascular mortality and morbidity included fatal and non fatal strokes respectively. Cardiovascular mortality summed coronary heart disease and cerebrovascular disease, but also included aneurysms, congestive heart failure and transient ischaemic attacks.

A number of drug treatments were used, including thiazides, reserpine, atenolol, ß-blockers and methyldopa.


Results are presented in diagrams which include odds ratios and NNTs - an at-a-glance review of the evidence.

There is also an interesting split in the analysis, comparing recent, large, high-quality trials with all trials to show the effect of quality. The high quality trials tended to produce more positive results. A summary of the results is shown in the box.

The overall NNT to prevent one cardiovascular event was 18 (95% CI 14 - 25) patients to be treated for five years. To prevent coronary heart disease the NNT was 61 (39 - 141) and to prevent cerebrovascular disease the NNT was 43 (31 - 69).

Stretching the mind

The real Mind-stretcher here is that the authors go on to extend their observations to younger and middle-aged subjects, where the results are not so positive. However, the comparison and discussion raises the possibility that long-term cumulative benefits in younger persons with greater remaining life expectancy and fewer competing risks may exceed those in older persons.

For those treating patients with hypertension, this paper is very much worth reading.


  1. CD Mulrow, JA Cornell, CR Herrera, A Kadri, L Farnett, C Aguilar. Hypertension in the elderly. Implications and generalizability of randomized trials. Journal of the American medical Association 1994 272: 1932-8.

previous or next story in this issue