Does patient self-management education of primary care professionals improve patient outcomes: a systematic review

Two specialist subject librarians assisted in the development of the search strategy designed to identify internationally recognised terminology in peer-reviewed journals. Full details of this strategy are available in the published protocol [28]. Six databases were searched - Cochrane Library, PubMed, ERIC, EMBASE, CINAHL and PsycINFO - in addition to Web searches, Hand searches and Bibliographies. Articles published in advance of September 1st, 2013 were included in the review, with the search conducted by GD and PP. The full search terms and sample search are shown in ‘Additional file 1’. The timeframe was deliberately chosen in order to coincide with the inclusion of the concept of patient empowerment in WONCA Europe’s definition of general practice. It was also the first work package of a larger project. It is intended to repeat the systematic review in 2018.

Studies with the following designs were included: systematic reviews, randomized controlled trials (RCTs), controlled clinical trials, interrupted time series, and controlled before and after studies.

Participants were physicians in primary care settings, other clinicians in primary care settings and patients 18+ years with chronic conditions in primary care settings. Included interventions had an educational focus designed to train primary care clinicians to support patient self-management. This review was concerned with all chronic conditions as they occur generically in the primary care setting, rather than focusing on any one specific chronic condition. Only articles including reference to patient outcomes, measured using validated measurement scales, were included. The primary patient outcome was change in patients’ self-management behaviours; the secondary outcomes were changes in physical health measures, health behaviours including medical adherence and compliance, service utilisation, psychological health, psycho-social function (e.g. Quality of Life, SF36, SF12) physical functioning and knowledge.

The eligibility of studies was determined using the inclusion and exclusion criteria listed in the registered proposal and shown in Table 1.

All abstracts were reviewed using the RefWorks package to categorise the abstracts identified by the search. The initial review of abstracts was undertaken by SB with 10% of same re-checked by AR. The full text articles of all those considered to be of possible relevance to the systematic review were read independently by SB, JG and CC and categorised using the same exclusion reasons. Disagreements were reviewed by AR. The final list of full text articles were then reviewed by JG to confirm relevance. The quality assessment and extraction of thematic content of the final list of articles applicable to the systematic review question were considered by CC and AR.

We assessed risk of bias and overall quality of individual studies using the Quality of Assessment Tool for Quantitative Studies [29] (Tables 2, 3 and 4). For each study, reviewers rated six components (selection bias, study design, confounders, blinding, data collection methods, and withdrawals and dropouts) leading to an overall methodological quality rating for each study of strong, moderate, or weak, with strong quality indicating a low risk of bias. Reviewers resolved rating disagreements through discussion.

We performed a narrative data synthesis as the clinical heterogeneity and differences in outcomes in the two studies meant meta-analysis would have been inappropriate.

Overall 7533 abstracts were reviewed following removal of duplicates from the database and hand/web searches; of these, 43 full text articles were retrieved and read (Fig. 1). Following this second stage review, only two articles that reported patient outcomes were included in the systematic review.

The two papers were both RCTs of educational interventions in primary care health professionals and examined their impact on patient outcome measures (Table 5).

The primary outcome of this review is the effectiveness of educational interventions in terms of patient outcomes. Our results show that education and skills training of primary care health professionals may improve patient performance of self-management activities, improve patient lifestyle behaviours, and patient self-efficacy in making behaviour changes. These changes were associated with improved perception of quality of life.

Becker et al. [30] compared outcomes of three groups of patients with low back pain from one control group of several practices and two intervention groups. The control group comprised of 43 practices with 410 patients and received printed educational information alone. One intervention group received education on guideline implementation (GI) alone (37 practices, 479 patients) and the other intervention group received GI plus training in motivational interviewing techniques (MIT) (38 practices, 70 practice nurses, 489 patients). Both intervention group patients showed significant improvements in functional capacity at six months though there was no difference at 12 months. Patients of the clinicians in the motivational interviewing group showed significant improvement in functional capacity after six months with a greater number of patients reporting pain free days (less permanent pain) after six months and after 12 months. These same patients also demonstrated a significant improvement in quality of life scale, having received one session from a practice nurse trained in motivational interviewing. The control group demonstrated no improvement in functional capacity, no perceived improvement in quality of life and they reported greater levels of permanent pain compared to both intervention groups.

Rubak et al. [31] included 65 GPs from 48 practices and all GPs attended a half day course on intensive treatment of type-2 diabetes. Practices were then randomised into control and intervention groups and the intervention group had an additional one and a half day residential course for 29 GPs from 21 practices (137 patients) on motivational interviewing techniques. The intervention practices also had a half day follow up course on MIT, twice during the following one year. The patients of the intervention group clinicians were shown to be more motivated to change behaviours and significantly more autonomous in their choice of actions leading to behavioural changes than patients of the control clinicians. The patients of the intervention clinicians also reported having received significantly more specific advice from their GP regarding diet, exercise and self-control of diabetes, and they also reported receiving significantly more counselling re smoking cessation. Intervention group patients reported an increase in their awareness of the importance of taking control of their own specific risk factors for diabetes. This study also reported a significant improvement in self-efficacy between pre-training and post training for all primary care health professionals undergoing motivational interviewing training.

The key finding of this systematic review is the scarcity of studies that assess the impact on patient outcomes of training primary care clinicians in patient self-management of chronic conditions. This was surprising given that patient self-management is a core element of person-centered healthcare in family practice and given the volume of published material on patient self-management.

This review shows that when health professionals undergo training in empowering patients for self-management of chronic conditions, it is possible to achieve improvement in patients’ self-efficacy, autonomy and motivation to change, functional capacity, pain free days and quality of life.

One study [30] demonstrated improvement in functional capacity, quality of life scale and a greater number of pain free days reported by patients after six months and after 12 months among those whose primary care clinicians had been given training in motivational interviewing techniques. A second study [31] showed that patients of the health professionals who participated in specific training programmes were more motivated to change behaviours and were more autonomous in their choice of behavioural changes compared to a control group. These patients were significantly more aware of the importance of controlling their diabetes for specific factors, and had a higher level of perception of having received specific advice from their GP on healthy behaviour changes.

Despite much literature on patient self-management in chronic disease, focussing on whether training health professionals regarding patient self-management improves patient outcomes, resulted in only two articles being eligible for inclusion in this systematic review. We are aware that some additional studies have been published more recently on self-management that were outside the chosen search period of this review. We plan to update this review in 2018. However, we feel it is important to publish the findings of this first phase review to highlight that, despite a vast volume of literature on one topic, evidence of impact on patient outcomes is largely lacking during the study period despite the publicity and interest in patient empowerment for self-management in the years leading up to its official inclusion in the European definition in 2011.

The small number of studies included and the range of outcome measures therein made concrete conclusions impossible, both papers describe positive outcomes from teaching motivational interviewing skills to clinicians, but we do not yet know if other approaches would be equally or more effective. The two studies did not report effect sizes, further complicating the interpretation of results. A further limitation is that only articles in English were included, based on available resources.

A total of 1643 patients, 191 clinicians and 164 practices were involved in these studies in two European countries, however further research on this topic is also needed to clarify if other factors are effective in improving patient outcomes other than those involving time constrained clinicians in general practice.

This review has given us concrete evidence of the lack of studies in the English language on improving patient outcomes through training primary care clinicians in patient self-management. Previous studies focus on patient education, group discussion among patients, shared experiences and unstructured acquisition of knowledge during clinical encounters or through leaflets and brochures [16, 25, 32, 33] rather than on assessing the specific effect of specific clinician training on patient outcomes in this setting. Primary care professionals have a longitudinal relationship with patients in the patient’s own community, are accessible to patients, and though contacts are intermittent, there is coordination and continuity of care. This review suggests that teaching motivational interviewing skills to health professionals in primary care may improve self-efficacy and quality of life in their patients, compared to those patients of clinicians who did not participate in this training. It suggests that the addition of motivational interviewing techniques to usual care may have added benefit for patients over usual care in the primary care setting, however further research is needed to identify if other educational interventions or skills are useful.

Further research is needed to distil the specific techniques to empower patients for self-management [34]; to explore and define the various aspects of the concept of patient empowerment [35]; and the variety of approaches that can be taken by primary care physicians to support patients to self monitor and make decisions about their chronic condition. We also need to identify and address potential barriers for self-management in patients [36].

Patients with chronic conditions interact over time with many professionals in primary and secondary care. There are many other interventions that may help to improve patient self-management, (for example group education, health coaching, telemedicine, e-health, media led interventions, voluntary associations, sports organisations and community group activities). Additional factors having a potential impact on successful outcomes include patient preferences for individual or group interventions, and patients’ values, goals, level of education and literacy. Patients have increasingly easy and direct access to online resources. We need research to guide both clinicians and patients to know which methods are best used in which settings [37], and which methods are not suited to particular settings. Careful designs and methods need to be used in future studies to assess the impact of such factors on measuring the outcomes of patient self-management including the Hawthorne effect and language bias. Appropriate assessment tools are required taking account of the integrative model of change appropriate in different settings [38].

Studies are also needed on evidence for the feasibility of training health professionals in patient self-management approaches including efficient use of resources such as time, people and finance, as research into the value and effectiveness of the various methods that can be used to “empower patients” is still in its infancy [37].

We also need to promote the inclusion of validated scales and instruments in future research for measurement and comparison of patient outcomes in chronic conditions.