To estimate the effect of various non-pharmacological lifestyle interventions on blood pressure.

MEDLINE (1966 to April 1995) was searched using keywords including 'prevention', 'dietary change', 'exercise', 'weight loss' and 'blood pressure control'. This was supplemented by examination of the reference list of each of the RCTs identified, consultation with experts in the field, and checking citations.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) with follow-up of 26 weeks or more.

Specific interventions included in the review

Salt reduction, weight reduction, stress management, exercise and reduction of alcohol consumption. Trials of secondary prevention were excluded.

Participants included in the review

Studies involving children or only adults aged less than 40 were excluded. Both hypertensive and normotensive patients were included. No further details of participants were given.

Outcomes assessed in the review

Change in blood pressure.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.

The authors do not state that they assessed validity.

The authors do not state how the data were extracted for the review, or how many of the reviewers performed the data extraction. Data were extracted on sample size, net change in systolic blood pressure and net change in diastolic blood pressured.

How were the studies combined?

Blood pressure (bp) changes were assessed as net changes (i.e. intervention group difference minus control group difference). Net changes for continuous variables were pooled using fixed-effect models.

How were differences between studies investigated?

The authors do not state how any differences between the studies were investigated.

Salt reduction: n=8.

Weight reduction: n=8.

Reducing stress: n=8.

Exercise: n=8.

Alcohol: n=1.

Some RCTs combined interventions and so these may be represented in more than one intervention grouping. It is not clear how many individual studies were included.

The effect of multiple risk factor interventions were investigated in an earlier review (see Other Publications of Related Interest no.1).

Salt reduction (n=8).

In general, the level of reduction of blood pressure was related to the success in reducing salt intake as measured by urinary excretion of sodium. Among Hypertensives (n=6):systolic BP pooled net change = -2.9 (95% CI: -5.8, 0.0); diastolic BP pooled net change = -2.1 (95% CI: -4.0, -0.1).

Normotensives (n=2): systolic BP pooled net change = -1.3 (95% CI: -2.7, 0.1); diastolic BP pooled net change -0.8 (95% CI: -1.8, 0.2).

Weight reduction (n=8).

The importance of considering associated salt intake, alcohol restriction and exercise was not considered in these trials.

Hypertensives (n=4): systolic BP pooled net change = -5.2 (95% CI: -8.3, -2.0); diastolic BP pooled net change = -5.2 (-6.9, -3.4).

Normotensives (n=2): systolic BP pooled net change = -2.8 (95% CI: -3.9, -1.8); diastolic BP pooled net change = -2.3 (-3.2, -1.4).

Reducing stress (n=8).

There was considerable heterogeneity of effects between trials: two trials conducted by the same group and using a meditation intervention reported falls in blood pressure substantially greater than other trials. Due to the significant heterogeneity pooled results are not presented here (they are presented in the paper). Net change in systolic BP ranged from -8.7 to +3.5, net changes in diastolic BP ranged from -11.1 to +0.9.

Exercise (n=8).

One trial did not report blood pressures but stated that no reductions were observed. In two factorial design trials examining exercise and dietary changes it was not possible to examine the main effects of the exercise intervention.

Hypertensives (n=3): systolic BP pooled net change = -0.8 (95% CI: -5.9, 4.2); diastolic BP pooled net change = -3.7 (-7.4, 0.0).

Normotensives (n=3): systolic BP pooled net change = -0.7 (95% CI: -2.8, 1.5); diastolic BP pooled net change = 0.3 (-1.6, 1.1).

Alcohol (n=1).

Men: systolic BP net change = -2.1 (no further details reported). Weight change (potentially associated) was not reported.

The RCT evidence to support commonly recommended lifestyle modifications for control of blood pressure was surprisingly inadequate. The estimates of net changes are probably overestimates of the effects that might be achieved by non-pharmacological interventions.

A reasonable review of the area. The search was limited to one database, although this was supplemented by examination of reference lists and consultations with experts, thus some important publications may have been missed. Very few methodological details, such as the number of reviewers involved in study selection and data extraction, were provided, and few study details were presented in the tables and text. It is therefore difficult to assess whether the study has been carried out using appropriate methodology. The reliability of the results is not clear since study quality was not assessed. However, as only RCTs were included this is not as much of a problem as if study designs with more potential for bias had been used. The statistical pooling appears mostly appropriate although heterogeneity was not formally assessed and so it is not possible to assess whether it was appropriate to pool data. The authors conclusions and recommendations follow from the results presented but should be interpreted with some caution due to the limitations highlighted above.

Practice: The authors do not state any implications for practice.

Research: The authors state that 'there is a need for large-scale long duration trials of non-pharmacological interventions in both hypertensive patients and normotensive people to determine effect sizes more accurately'.

Health Education Authority, London.

1. Ebrahim S, Davey Smith G. A systematic review and meta-analysis of randomised controlled trials of health promotion for prevention of coronary heart disease in adults. BMJ 1997;314:1666-74. This additional published commentary may also be of interest. Jackson R. The evidence to support commonly recommended life-style modifications for control of blood pressure is weak. Weight reduction stands out as a simple but potentially useful intervention. Evidence-based Healthcare 1999;3:81.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.