The effect of general practice team composition and climate on staff and patient experiences: a systematic review

Ruth Abrams; Bridget Jones; John Campbell; Simon de Lusignan; Stephen Peckham; Heather Gage

doi:10.3399/BJGPO.2023.0111

Article Figures & Data

Figures

Tables

Download figure
Open in new tab
Download powerpoint
Figure 1. Conceptual framework
Download figure
Open in new tab
Download powerpoint
Figure 2. Integrated PRISMA diagram

Tables

Figures

View popup

Table 1. Summary of review methodology

Review methodology
Search	Databases	· MEDLINE, Embase, Cochrane Library, CINAHL, PsycINFO, and Web of Science
The four search strategies can be found in Supplementary Boxes S1–S4. The associated PRISMA diagrams are in Supplementary Figures S1–S4	Terms	· Terms related to primary health care (such as family practice and general practice) and teams (including, but not limited to, staff, interprofessional, interdisciplinary, and multidisciplinary).
	Limiters search stage 1	· Date: January 2015 to December 2021 · Language: English only, because of resource limitations · Countries: where systems of health care were comparable to the UK, such as Canada, New Zealand, and Australia, and excluding studies set in low- and middle-income countries (because of different levels of resources and priorities), and in the US (because of its heterogenous system of provision, dominance of private insurance funding, and lack of universal coverage).
	Search stage 2	· US only. Studies set in US integrated care systems (that align enrolled patients with primary healthcare practitioners and use gatekeeping to specialist services) added because search 1 returned four US articles (despite the country filters) that were considered relevant. Search 2 was same as search 1 in all other respects.
	Search stage 3	· Targeted search, including additional keywords that searches 1 and 2 had identified as potentially relevant, including: ’characteristics’ or ‘structure’ or ‘ratio’ or ‘size’.
	Search stage 4	· Date range extended back to January 2012 because searches 1–3 had identified relevant earlier articles outside the original search dates. Stage 4 was in 2 steps: search criteria 1 and 2 combined, and search criteria 3.
Screening	Titles and abstracts, followed by full text	· Screening was undertaken independently by two reviewers (RA and HG). Differences were discussed to determine consensus; a third reviewer (BJ) was asked to adjudicate three articles.
	Inclusion criteria	· Empirical analysis of team composition (structure) OR climate (relational processes) as the primary focus, AND staff outcomes (including job satisfaction, wellbeing, stress, or burnout) OR patient outcomes (including experience, satisfaction, or clinical effectiveness/utilisation). · Multidisciplinary team working (such as ≥2 different roles/skills). · Evidence on team composition (structure) that relates to staff ratios, grades, and profession.¹¹ · Evidence on team climate that relates to relational processes of team working including discussion of shared perceptions of organisational policies, practices, and procedures.¹¹
	Exclusion criteria	· Studies evaluating single roles (for example, nurses and pharmacists) or single patient groups/conditions (for example, diabetes) because they did not represent the full range of general practice service delivery.⁶⁰ · Articles reporting change in skill mix due to task reassignment among existing team members (for example, substitution and delegation) because this was not considered to be a change in team composition.⁶¹ There is already a large and growing body of evidence on the effects of task reassignment.^62–65 · Non-empirical, non-peer reviewed, grey literature, and dissertations. · Set in: secondary care, hospitals, outpatient/non-primary ambulatory care, hospices, or long-term care or home-care services.
	Quality assessment	· Two reviewers (RA and BJ) carried out independent quality assessment of all included studies using the Mixed Methods Assessment Tool (MMAT).⁶⁶ Articles were scored (1 = high quality and well reported; 2 = good quality; and 3 = lower quality or badly reported but still relevant) so that assessments of the reviewers could be compared. In line with MMAT guidance, no studies deemed of low quality were excluded.
Data extraction	· Characteristics of included studies (bibliographic details, country of study, setting, sample/population, data and methods, variables, outcomes, and study limitations) were collated into a Microsoft Excel table (Supplementary Table S1).
Analysis	· Texts of included articles were added to NVivo (version 12), coded, and synthesised into a thematic structure consistent with the conceptual framework and research questions.⁶⁷ Themes were discussed with team members to corroborate findings.

View popup

Table 2. Summary of findings

	Outcomes for staff The most frequently researched outcome for staff (five studies) was burnout, measured as emotional exhaustion; the predictors of work satisfaction were explored in one study.	Outcomes for patients Patient outcomes were predominantly indicators of quality of care (four studies); hospital use and all-cause mortality (indicators of the clinical effectiveness in primary health care) were used in two studies.
Team composition Mix of skills and staff characteristics (for example, team size, disciplinary mix, and provider age and sex)	One study found emotional exhaustion or burnout was lower when physicians account for a higher proportion of the total team FTE. This study recorded higher burnout than other studies (85%) and no difference between rates for physicians and other clinicians (nurse practitioners and physician assistants).¹⁹ Other studies reported lower rates of burnout (30%–60%) with physicians (especially residents) at higher risk than other clinical and support staff.^21,22,26 Two studies reported higher burnout among female physicians^19,20 and non-physician clinicians.¹⁹ One study reported no association between team size and structure (family physician plus one or two or three other roles) and emotional exhaustion or burnout.²⁰	Patients in practices with a predominance of female physicians reported better continuity, comprehensiveness, and responsiveness of care, and more counselling and screening, although these associations were confounded by the younger average age of female doctors. In adjusted analysis, the only significant difference from male predominant practices was worse access in female predominant practices, which was explained in terms of higher part-time working by female physicians.²⁸ Hospital readmission rates were found to be lower in the panels of female clinicians, although that finding did not extend to index hospital admissions or ED visits. Panels of physicians had the lowest ED visits (versus nurse practitioners and physician assistants). The strongest predictors of higher hospital use were panel complexity and less time in clinical practice (attributed to less clinical acumen and lower risk tolerance).¹⁸ This study also found that hospital use was not associated with the proportion of care FTE that was physicians.
Team climate The ‘relational process of teamworking’,¹¹ variously measured (for example, work environment, staff stability, delegation, leadership, team effectiveness, team functioning, team dynamics, and workload)	There is consistent evidence that a range of factors synonymous with good working environments and team dynamics reduce the risks of emotional exhaustion and burnout, including: team stability, staff feeling and acting like a team, having control over workload, participating in decision making, and working at the top of their competencies.^20–22,26 Staff and skill shortages were identified as catalysts to burnout.^20–22 One study concluded that ‘culture trumps structure’; a poor team culture (as measured by the Team Climate Inventory) could override the effects of a stable team environment and have a negative effect on emotional exhaustion.²⁶ Clinician satisfaction was associated positively with team dynamics, but through the mediation of patient care coordination: clinicians were found to derive satisfaction from better patient care coordination, which the researchers associated with good team dynamics.²⁵ In another study, more than one-third of physicians reporting burnout were also satisfied.²⁰	Higher workload and staff insufficiency were significantly associated with more complaints, less clinician time with patients, and lower patient-reported quality of care, with diminishing benefits observed from adding extra staff above VHA-recommended levels because of coordination problems and ‘social loafing’.^23,24 Better team functioning was associated with reduced hospital admissions (vulnerable patients) and lower all-cause mortality (all patients, not vulnerable). Greater emotional exhaustion was associated with lower ambulatory care sensitive admissions; staff sufficiency was associated with higher all-cause admissions.²⁷ Better relational climate and cohesion of the work group was associated with improved quality of care.^23,24 Team climate was found to mitigate the adverse effects of high workload on patient outcomes. While workload negatively affected quality of care if the relational climate was poor, a strong relational climate can protect against poor quality of care if the workload is high.²⁴ Team dynamics were found to be strongly positively associated with physician-rated patient care coordination, which, as noted above, in turn mediated a strong positive association between team dynamics and clinical work satisfaction.²⁵

ED = emergency department. FTE = full-time equivalent. VHA = Veterans Health Administration

Supplementary Data

RA2_10.3399BJGPO.2023.0111_supp.pdf -
Supplementary material is not copyedited or typeset, and is published as supplied by the author(s). The author(s) retain(s) responsibility for its accuracy.