Interventions for improving patients' trust in doctors and groups of doctors

Abstract

Background: Trust is a fundamental component of the patient-doctor relationship and is associated with increased satisfaction, adherence to treatment, and continuity of care. Our 2006 review found little evidence that interventions improve patients' trust in their doctor; therefore an updated search was required to find out if there is further evidence of the effects of interventions that may improve trust in doctors or groups of doctors. Objectives: To update our earlier review assessing the effects of interventions intended to improve patients' trust in doctors or a group of doctors. Search methods: In 2003 we searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library), MEDLINE, EMBASE, Health Star, PsycINFO, CINAHL, LILACS, African Trials Register, African Health Anthology, Dissertation Abstracts International and the bibliographies of studies selected for inclusion. We also contacted researchers active in the field. We updated and re-ran the searches on available original databases (Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library issue 2, 2013), MEDLINE (OvidSP), EMBASE (OvidSP), PsycINFO (OvidSP), CINAHL (Ebsco)) as well as Proquest Dissertations and Current Contents for the period 2003 to 18 March 2013. Selection criteria: Randomised controlled trials (RCTs), quasi-randomised controlled trials, controlled before and after studies, and interrupted time series of interventions (informative, educational, behavioural, organisational) directed at doctors or patients (or carers) where trust was assessed as a primary or secondary outcome. Data collection and analysis: Two review authors independently extracted data and assessed the risk of bias of included studies. Where mentioned, we extracted data on adverse effects. We synthesised data narratively. Main results: We included 10 randomised controlled trials (including 7 new trials) involving 11,063 patients. These studies were all undertaken in North America, and all but two involved primary care. As expected, there was considerable heterogeneity between the studies. Interventions were of three main types; three employed additional physician training, four were education for patients and three provided additional information about doctors in terms of financial incentives or consulting style. Additionally, several different measures of trust were employed. The studies gave conflicting results. Trials showing a small but statistically-significant increase in trust included: a trial of physician disclosure of financial incentives; a trial of providing choice of physician based on concordance between patient and physician beliefs about care; a trial of group visits for new inductees into a Health Maintenance Organisation; a trial of training oncologists in communication skills; and a trial of group visits for diabetic patients. However, trust was not affected in a subsequent larger trial of group visits for uninsured people with diabetes, nor with a decision aid for helping choose statins, another trial of disclosure of financial incentives or specifically training doctors to increase trust or cultural competence. There was no evidence of harm from any of the studies. Authors' conclusions: Overall, there remains insufficient evidence to conclude that any intervention may increase or decrease trust in doctors. This may be due in part to the sensitivity of trust instruments, and a ceiling effect, as trust in doctors is generally high. It may be that current measures of trust are insufficiently sensitive. Further trials are required to explore the impact of doctors' specific training or the use of a patient-centred or decision-sharing approach on patients' trust, especially in the areas of healthcare provider choice, and induction into healthcare organisation. International trials would be of particular benefit. The review was constrained by the lack of consistency between trust measurements, timeframes and populations.