To review the medical and nursing research literature addressing quality of life (QOL) in adult cardiac patients, and to identify patterns, trends, gaps, and future directions for QOL research in this population.

Medical and nursing databases on Bibliographical Retrieval System, Dialog and MEDLARS were searched from 1987 to 1991 using the following terms: 'QOL', 'intervention' and 'cardiac problem'. Bibliographies of published studies and eight relevant journals were handsearched. Unpublished reports were excluded.

Study designs of evaluations included in the review

Quasi-experimental or experimental designs were included.

Specific interventions included in the review

Pharmacological, mechanical, coronary artery bypass grafting, nursing, acupuncture, behavioural modification, and pharmacological treatment combined with behavioural modification.

Participants included in the review

Adult cardiac patients with a mean age of 57 years; 68% of the patients were male. Medical diagnoses were as follows: hypertension (54%), heart failure (19%), dysrhythmia (10%), angina pectoris (10%) and myocardial infarction (8%).

Outcomes assessed in the review

QOL, usually measured as symptoms; also as physiological distress or well-being, physical functioning or satisfaction, health perception or satisfaction, social functioning or satisfaction, role functioning or satisfaction, and cognitive functioning. Various instruments were used.

How were decisions on the relevance of primary studies made?

The identified titles and abstracts were screened by two researchers. Each retrieved report was reviewed again by two researchers, and was only included in the review if both reviewers were in agreement.

A quality of study rating scale, based on guidelines published by Cooper (see Other Publications of Related Interest no.1) and Smith and Stullenbarger (see Other Publications of Related Interest no.2), was applied. Studies were not excluded on the basis of quality. The methodological characteristics of the studies were coded by one investigator. It is unclear whether the coding was checked by a second reviewer.

A generic coding form was used to record data from the studies. Inter-coder reliability was tested. Any discrepancies between reviewers were resolved through discussion.

Statistical results from primary studies were transformed into z-scores (the standard normal deviate) to assess whether effects were significant, and the d-index (measured in standard deviation units), which provides a measure of the magnitude of the treatment effect.

How were the studies combined?

The oerall average treatment effect was calculated by summing z-scores from individual studies and dividing this by the square root of the total number of studies. Individual d-values were weighted according to sample size and combined to establish the overall magnitude of treatment effect. The U3 statistic was calculated to demonstrate the extent of nonoverlap between scores of treatment and control groups. The binomial effect size display (BESD) was calculated to make the estimate effect more interpretable.

How were differences between studies investigated?

A chi-squared homogeneity analysis was carried out.

Eighty-four studies involving 40,132 participants were included.

Average study quality was rated as moderate to good. There was significant variation in effect size among included studies. Eight potential moderators of overall treatment effect were identified. Overall, there was a small but significant positive effect of treatment on QOL (weighted d-value=0.31). The BESD showed that participants in the experimental groups benefited by 15% with treatment. However, the overall U3 score of 0.62 indicates that 38% of participants in control groups also sustained QOL benefits. No negative effect of treatment was found for any diagnostic category. It was estimated that 16,766 studies showing no treatment effect would be needed to overturn the findings of this meta-analysis.

The results of this meta-analysis suggest a small but significant effect of treatment on QOL, regardless of the form of treatment.

The research question is clear, as are the inclusion and exclusion criteria for studies. More details concerning the databases searched, and criteria for assessing individual study quality would have been useful. There are few details of individual studies, and the usefulness of combining data from studies involving diverse interventions is questionable.

The authors suggest that future research in this area should address the issues of defining and measuring QOL more assiduously, and also examine the problems of data collection in these studies. Detailed profiles of study participants should be provided by trialists.

American Heart Association, Alabama Affiliate, grant number AL-G-920039.

1. Cooper HM. Integrating research: a guide for literature reviews. 2nd ed. Newbury Park (CA): Sage; 1989. 2. Smith MC, Stullenbarger E. A prototype for integrative review and meta-analysis of nursing research. J Adv Nurs 1991;16:1271-83.

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.