Number of studies
86 randomised trials (involving more than 20,209 participants)
Compared with usual care, decision aids:
Stacey et al.
·improved knowledge and accuracy of risk perceptions;
·reduced the proportion of people who were passive in decision-making;
·resulted in a higher proportion of patients achieving decisions informed and consistent with their values (when decision aids included an explicit values clarification component);
·reduced the number of people remaining undecided;
·reduced decisional conflict;
·decreased the choice of major elective surgery in favour of conservative options.
Decision aids have no adverse effects on satisfaction but further research is needed to clarify their effect on adherence to chosen option, patient-practitioner communication, cost-effectiveness and use with developing or lower literacy populations.
Personalised Risk Communication
22 randomised trials
There was weak evidence, consistent with a small effect, that personalised risk communication (whether written, spoken or visually presented) increases uptake of screening tests.
Edwards et al.
Communication before Consultations
33 randomised trials (involving 8244 participants)
Compared with a control, communication before consultations increased question asking during consultations. They may also increase patient participation in consultation and improve patient satisfaction.
Kinnersley et al.
Both coaching and written material interventions produced similar effects on question asking, but coaching produced a larger increase in patient satisfaction.
Overall the benefits of ‘communication before consultations’ interventions were minor.
Interactive Health Communication Applications (IHCAs) (2 reviews)
Murray et al.
24 randomised trials (involving 3739 participants)
IHCAs had a significant positive effect on knowledge, social support and clinical outcomes.
Bailey et al.
15 randomised trials (involving 3917 participants)
Positive effects of IHCAs on knowledge, safer sex self-efficacy and intentions and sexual health behavior were found.
Comment: Data were insufficient for meta-analysis of biological outcomes or analysis of cost- effectiveness and thus, the effects on these outcome categories remain unknown.
Interventions to Enhance Medication Adherence
78 randomised trials
Mixed effects were observed for short term and long- term medication adherence.
Haynes et al.
Some, but not all, of the simple interventions, such as counselling, written information and personal phone calls, were effective with people on short-term medication treatments.
The picture for the effectiveness of interventions for longer-term treatments was mixed; few interventions showed promise and those that were effective were complex and multifaceted in nature.
30 randomised trials (involving 4691 participants)
Contracts were shown to ‘potentially’ improve patient adherence (as applied to diagnostic procedures, therapeutic regimens, and/or a health promotion or illness prevention initiative).
Bosch-Capblanch et al.
Comment: The result above is based on only half of the included studies; the effects were not detected over longer periods.
New Methods of Communication
Marteau et al.
13 randomised trials
Little or no effect was shown with respect to smoking cessation or increasing physical activity. A small effect was shown for changing diet.
(on communicating DNA-based disease risk estimates)
The intervention showed potential for altering intentions to change behaviour (in six non-clinical analogue studies).
Comment: The authors concluded that given the small number of trials in this area, more research involving ‘better-quality RCTs’ is needed before recommending application in practice.
Hollands et al.
9 randomised trials (involving 1371 participants)
Overall, results were mixed:
·a positive effect was found for smoking cessation (three trials);
·a positive effect was found for skin examination behaviour (one trial);
(providing visual feedback on medical imaging results)
·no effect was found for change in physical activity (one trial).
Comment: The authors concluded that due to the small number of trials and the mixed results found, the effectiveness of communicating medical imaging results to change health behaviour is largely unknown and thus, its application in practice is not yet recommended.
Nicolson et al.
25 randomised trials (involving 4788 participants)
Written material significantly improved knowledge of medicines in six of twelve trials. In three of these six trials recall of side effects also improved, but medicines recall significantly improved in only a minority of trials (one of four).
The results for attitudinal and behavioural outcomes were mixed.
Comment: Overall, the authors concluded the combined evidence from this review is not sufficient to say whether written medicines information is effective in changing behaviours related to medicine taking.
Self Management Programmes
Foster et al.
17 randomised trials (involving 7442 participants)
Small (clinically insignificant) short-term improvements in pain, disability, fatigue and depression were found.
(Self management programmes run by lay people)
Positive effects on confidence to manage and self- rated health were also found.
There was no effect on quality of life or use of health services.