Background

Healthcare professionals are important contributors to healthcare quality and patient safety, but their performance does not always follow recommended clinical practice. There are many approachesto influencing practice among healthcare professionals. These approaches include audit and feedback, reminders, educational materials, educational outreach visits, educational meetings or conferences, use of local opinion leaders, financial incentives, and organisational interventions. In this review, we evaluated the effectiveness of patient‐mediated interventions. These interventionsare aimed at changing the performance of healthcare professionals through interactions with patients, or through information provided by or to patients. Examples of patient‐mediated interventionsinclude 1) patient‐reported health information, 2) patientinformation, 3) patient education, 4) patient feedback about clinicalpractice, 5) patient decision aids, 6) patients, or patientrepresentatives, being members of a committee or board, and 7)patient‐led training or education of healthcare professionals.

Objectives

To assess the effectiveness of patient‐mediated interventions on healthcare professionals’ performance (adherence to clinicalpractice guidelines or recommendations for clinical practice).

Search methods

We searched MEDLINE, Ovid in March 2018, Cochrane Central Register of Controlled Trials (CENTRAL) in March 2017, and ClinicalTrials.gov and the International Clinical Trials Registry (ICTRP) in September 2017, and OpenGrey, the Grey Literature Report and Google Scholar in October 2017. We also screened the reference lists of included studies and conducted cited reference searches for all included studies in October 2017.

Selection criteria

Randomised studies comparing patient‐mediated interventions toeither usual care or other interventions to improve professionalpractice.

Data collection and analysis

Two review authors independently assessed studies for inclusion, extracted data and assessed risk of bias. We calculated the risk ratio (RR) for dichotomous outcomes using Mantel‐Haenszel statistics and the random‐effects model. For continuous outcomes, we calculated the mean difference (MD) using inverse variance statistics. Two review authors independently assessed the certainty of the evidence (GRADE).

Main results

We included 25 studies with a total of 12,268 patients. The number of healthcare professionals included in the studies ranged from 12 to167 where this was reported. The included studies evaluated four types of patient‐mediated interventions: 1) patient‐reported health information interventions (for instance information obtained from patients about patients’ own health, concerns or needs before a clinical encounter), 2) patient information interventions (for instance, where patients are informed about, or reminded to attend recommended care), 3) patient education interventions (intended toincrease patients’ knowledge about their condition and options of care, for instance), and 4) patient decision aids (where the patient is provided with information about treatment options including risks and benefits). For each type of patient‐mediated intervention a separate meta‐analysis was produced.

Patient‐reported health information interventions probably improvehealthcare professionals’ adherence to recommended clinicalpractice (moderate‐certainty evidence). We found that for every 100 patients consulted or treated, 26 (95% CI 23 to 30) are in accordance with recommended clinical practice compared to 17 per 100 in the comparison group (no intervention or usual care). We are uncertain about the effect of patient‐reported health information interventionson desirable patient health outcomes and patient satisfaction (very low‐certainty evidence). Undesirable patient health outcomes and adverse events were not reported in the included studies and resource use was poorly reported.

Patient information interventions may improve healthcare professionals’ adherence to recommended clinical practice (low‐certainty evidence). We found that for every 100 patients consulted or treated, 32 (95% CI 24 to 42) are in accordance with recommended clinical practice compared to 20 per 100 in the comparison group (no intervention or usual care). Patient information interventions may have little or no effect on desirable patient health outcomes andpatient satisfaction (low‐certainty evidence). We are uncertain about the effect of patient information interventions on undesirablepatient health outcomes because the certainty of the evidence is very low. Adverse events and resource use were not reported in the included studies.

Patient education interventions probably improve healthcare professionals’ adherence to recommended clinical practice(moderate‐certainty evidence). We found that for every 100 patients consulted or treated, 46 (95% CI 39 to 54) are in accordance with recommended clinical practice compared to 35 per 100 in the comparison group (no intervention or usual care). Patient educationinterventions may slightly increase the number of patients with desirable health outcomes (low‐certainty evidence). Undesirablepatient health outcomes, patient satisfaction, adverse events and resource use were not reported in the included studies.

Patient decision aid interventions may have little or no effect on healthcare professionals’ adherence to recommended clinicalpractice (low‐certainty evidence). We found that for every 100 patients consulted or treated, 32 (95% CI 24 to 43) are in accordance with recommended clinical practice compared to 37 per 100 in the comparison group (usual care). Patient health outcomes, patientsatisfaction, adverse events and resource use were not reported in the included studies.

Authors’ conclusions

We found that two types of patient‐mediated interventions, patient‐reported health information and patient education, probablyimprove professional practice by increasing healthcare professionals’ adherence to recommended clinical practice(moderate‐certainty evidence). We consider the effect to be small tomoderate. Other patient‐mediated interventions, such as patientinformation may also improve professional practice (low‐certainty evidence). Patient decision aids may make little or no difference tothe number of healthcare professionals’ adhering to recommended clinical practice (low‐certainty evidence).

The impact of these interventions on patient health and satisfaction, adverse events and resource use, is more uncertain mostly due tovery low certainty evidence or lack of evidence.