Primary care practice characteristics associated with team functioning in primary care settings in Canada: A practice-based cross-sectional survey

ABSTRACT

Team-based care is recognized as a foundational building block of high-performing primary care. The purpose of this study was to identify primary care practice characteristics associated with team functioning and examine whether there is relationship between team composition or size and team functioning. We sought to answer the following research questions: (1) are primary care practice characteristics associated with team functioning; and (2) does team composition or size influence team functioning. This cross-sectional correlational study was conducted in Fraser East, British Columbia, Eastern Ontario Health Unit, Ontario and Central Zone, Nova Scotia in Canada. Data were collected from primary care practices using an organization survey and the Team Climate Inventory (TCI) as a measure team functioning. The independent variables of interest were: physicians’ payment model, internal clinic meetings to discuss clinical issues, care coordination through informal and ad hoc exchange, care coordination through electronic medical records and sharing clinic mission, values and objectives among health professionals. Potentially confounding variables were as follows: team size, composition, and practice panel size. A total of 63 practices were included in these analyses. The overall mean score of team climate was 73 (SD: 10.75) out of 100. Regression analyses showed that care coordination through human interaction and sharing the practice’s mission, values, and objectives among health professionals were positively associated with higher functioning teams. Care coordination through electronic medical records and larger team size were negatively associated with team climate. This study provides baseline data on what practice characteristics are associated with highly functioning teams in Canada.

KEYWORDS:

• Team climate inventory
• primary care
• organizational factors and team composition

Introduction

Team-based care is recognized as a foundational building block of high-performing primary care (Bodenheimer et al., 2014). It is fundamental to chronic disease management (Beaulieu et al., 2014; J. D. Mitchell et al., 2019). Team-based care is referred as two or more health professionals working collaboratively to provide comprehensive, continuous, and coordinated primary care services (MacPhee et al., 2020). It is an approach to care that offers patients access to a range of clinicians from different disciplines including family physicians, registered nurses, psychologists, pharmacists, social workers, occupational therapists, and others as needed (Helfrich et al., 2014; Wagner et al., 2017).

Team-based primary care is better equipped to provide comprehensive care for patients with complex chronic diseases, particularly for coordinating and integrating their care, which can result in better health outcomes (Metusela et al., 2020; Samuelson et al., 2012). Past work suggests that team-based care is associated with improved clinical outcomes (Levengood et al., 2019; Proia et al., 2014), reduced health-care costs and health services utilization such as hospitalization and emergency admissions (Levis-Peralta et al., 2020; Mateo-Abad et al., 2020; Metusela et al., 2020), better patient experiences with care (Khan et al., 2021) and better provider well-being, job satisfaction, and productivity (Levis-Peralta et al., 2020; Meyers et al., 2019; J. D. Mitchell et al., 2019). Implementation of team-based primary care has, for several decades, been a focus of primary care transformation (MacPhee et al., 2020; Misfeldt et al., 2017; Peckham, Ho, Marchildon et al., 2018). It remains an essential element of patient-centered medical home construction in the United States and Canada (Bodenheimer & Sinsky, 2014; J. D. Mitchell et al., 2019; Howard et al., 2011).

Background

In Canada, where delivery of primary care is a provincial or territorial responsibility, most care is delivered through privately owned fee-for-service practices (Peckham, Ho, Marchildon et al., 2018). Team-based care is more nascent and has varied in implementation with regard to health-care financing, human resources, and health-care needs of populations (Misfeldt et al., 2017; Peckham, Ho, Marchildon et al., 2018; Wranik et al., 2019). There has been implementation of Family Health Teams and Family Medicine Groups occurred in Ontario, Family Medicine Groups in Quebec (Wranik et al., 2019), Primary Care Networks in British Columbia and Alberta as well as group practices and primary care teams in Nova Scotia (Peckham, Ho, Marchildon et al., 2018; Wranik et al., 2019) and My Health Teams in Manitoba (Bobbette et al., 2021). Teams differ in terms of number, size, organization, composition, governance, and health-care services offered to the population (Bobbette et al., 2021).

While team-based care continues to be implemented across Canada, less is known about how teams function or what practice characteristics foster better team functioning. Team functioning refers to interpersonal processes and actions that occur within the teams when members are carrying out their activities to achieve shared goals (Kilpatrick et al., 2019; MacPhee et al., 2020). Some past work suggests that practice characteristics play an important role in how teams function (Bobbette et al., 2021; Harris et al., 2016; Khan et al., 2021; Lemieux-Charles & McGuire, 2006; McNaughton et al., 2021; Valaitis, Wong, et al., 2020). Examples of characteristics influencing primary care team functioning include use of electronic medical records/health information technology, communication through team meetings and huddles (Fiscella et al., 2017; McNaughton et al., 2021; Sørensen et al., 2018), shared vision, and goals (Fiscella et al., 2017; McNaughton et al., 2021) and type of physician payment model (Pype et al., 2018; Russell et al., 2018). An earlier review conducted Gocan et al. (2014) found similar determinants of effective team functioning including communication, adequate funding, remuneration, incentives, electronic medical records, human resources, clarity of vision, roles, and responsibility.

However, most of these studies are qualitative where quantitative studies to support these theoretical relationships are rare. The few available quantitative studies are limited by relatively small sample sizes (Brown et al., 2015; Howard et al., 2011). Moreover, little is known about what practice characteristics facilitate team increased functioning specifically in primary care. The purpose of this study is to identify primary care practice characteristics associated with team functioning and examine whether there is relationship between team composition or size and team functioning. We sought to answer the following research questions: (1) are primary care practice characteristics associated with team functioning? H0: Primary care practice characteristics such as communication and electronic medical records is associated with team functioning; and (2) does team composition or size influence team functioning? H0: Team composition or size is associated team functioning.

Methods

Design

This study is a part of a larger mixed-methods research program, TRANSFORMATION, which sought to improve the science and reporting of primary care performance. Organizational, provider, and patient-level data were collected from primary care practices in Fraser East, British Columbia (BC), Eastern Ontario Health Unit, Ontario (ON) and Central Zone, Nova Scotia (NS). (Wong et al., 2018). Case studies and deliberative dialogs (day-long discussions) were also conducted in each region (Martin-Misener et al., 2019; Wong et al., 2018). For the purpose of this paper, we used a cross-sectional correlational study design to examine the association between primary care practice characteristics and the team functioning as measured by the Team Climate Inventory Tool (Anderson & West, 1998)

Participants

Once physicians within teams agreed to participate, we collected organizational characteristics and team functioning data from practice managers and staff, respectively. Details about recruitment are reported elsewhere (Martin-Misener et al., 2019; Wong et al., 2018). A total of 87 practices (Fraser East, BC: 22; Eastern Ontario, ON: 26 and Central Zone, NS: 39) and a total of 1158 staff took part in the TRANSFORMATION study. Primary care practice leads completed organizational surveys while all staff (clinicians and office staff) working in participating practices were asked to complete the team climate inventory survey. Practice leads, who were defined by each practice, were either the senior administer or lead physician for each practice. Data collection occurred between 2014 and 2016 using REDCap (Research Electronic Data Capture) and with paper surveys.

Instruments

The data were collected using two surveys. The organizational survey was developed on previous foundational work (Levesque et al., 2012) and the research team’s previous contribution to the development of Canadian Institute of Health Information PHC surveys (Wong et al., 2013). The organizational survey had six sections and 77 questions (Appendix 1). In this study, our independent variables of interest were as follows: physicians’ payment model, internal clinic meetings to discuss clinical issues, care coordination through informal and adhoc exchange, care coordination through electronic medical records and sharing clinic mission, values, and objectives among health professionals (Appendix 1). We also included potentially confounding variables of team size (≤5; 6–10 and ≥11) and composition, and practice panel size (≤3000 patients, 3001–10000 and ˃10,000 patients).

The Team Climate Inventory (TCI) instrument was used to measure team functioning, the dependent variable of interest. The TCI was chosen because it is a widely used validated tool and was adopted by consensus among 12 Canadian Institutes of Health Research-funded Community-Based Primary Health-Care Innovation Teams (Charif et al., 2018) where it was found to best measure team elements that could be linked to better outcomes (Beaulieu et al., 2013). The 19-item TCI is defined as “a team’s shared perceptions of organizational policies, practices, procedures” (Anderson & West, 1998). The TCI measures group climate, functioning, and innovation. There is an overall score and four sub-dimension scores: (1) Participative safety-acknowledges that trust is essential for members’ involvement; (2) Vision or Team objectives-refers to valued outcomes and a common higher goal as motivating factors; (3) Task orientation-refers to a shared concern for excellence; and (4) Support for innovation-the expectation of and support for the introduction of new ways of doing things (Beaulieu et al., 2014; Appendix 2). It has been validated in Canada, in English and French with Cronbach’s alpha of four sub-dimensions ranging from 0.81 to 0.86 with overall Cronbach’s alpha of 0.95 (Beaulieu et al., 2014). Both English and French versions were used in this study. The overall TCI practice scores were considered the outcome variable of interest.

Data analysis

The practices who had the value of < 0.24 were considered to having enough responses and included in the analysis. The value was obtained by (N-n)/Nn, where “N” is the total number of staff working in the practice and “n” indicating the total number of staff who responded to the TCI survey (Dawson, 2003). For example, in a practice of 10 providers, at least three respondents were needed (0.23) to calculate the practice TCI score. TCI scores were only calculated for the subjects who answered at least 60% of the items that constituted the score. An imputation process was performed for missing data by imputing the mean of the individual respondents on the same subdimension. In order to ease interpretability, the TCI scores were transformed to 0–100 where an increased score indicated more of the concept.

We used frequencies and percentages to summarize categorical variables and means and standard deviations to summarize numerical variables. Chi-square test was used to test whether covariates differed across regions. One-way between subjects Analysis of Variance (ANOVA) was used to assess whether TCI scores differed across regions and whether practice characteristics were associated with the TCI scores. Intraclass correlations (ICC) were calculated to measure the proportion of team climate variation that could be explained by region differences. Findings of ICC (.00) indicated no variations in team climate that could be explained by the region-level differences; therefore, we did not use hierarchical linear modeling (HLM) to capture the region differences; rather regression analysis was conducted to identify primary care characteristics associated with the overall TCI score. The hierarchical multiple linear regression analysis was applied because it allowed us to identify factors (practice characteristics) that have an influence on outcome and determine the extent to which these characteristics predict the outcome herein team functioning as measured by team climate inventory (Petrocelli, 2003). Data analysis was performed using Statistical Package for the Social Sciences (SPSS) 21 (Corporation, 1989).The statistical significance level was set at < .05 (2-sided).

Ethical considerations

Approval of research was guaranteed from Institutional review boards of Fraser Health, University of British Columbia (H13-01237), Ottawa Health Sciences Network (20,140,458–01 H), Bruyère Continuing Care (M16-14-029) and the Nova Scotia Health Authority (NSHA REB ROMEO#:1,017,461). Other ethical principles were observed throughout the study process.

Results

Sixty-three of 87 practices provided sufficient data where a practice TCI score could be calculated. There were a total of 566 of 889 (64%) staff who completed the TCI. Table 1 describes characteristics of all primary care practices that participated in the TRANSFORMATION study.

Table 1. Organizational characteristics of all participating primary care practices (n= 87).

	Fraser East-BC (n=22, 25%)		Eastern Ontario Health Unit-ON (n=26, 30%)		Central Zone-NS (n=39, 45%)		Total (n=87, 100%)
Practice characteristics	n	%	n	%	n	%	n	%	p-value
Clinic internal meetings to discuss…clinical issues:									.651
At least weekly	2	9	8	32	8	22	18	22
Monthly	12	55	9	36	16	44	37	45
Other	5	23	4	16	6	17	15	18
Never	3	14	4	16	6	17	13	16
Care coordination through informal or adhoc									.006
exchange**:
Regularly	17	90	21	91	19	56	57	75
Sometimes	2	11	2	9	7	21	11	15
Rarely	0	0	0	0	8	24	8	11
Care coordinated through electronic medical records (EMRs)**:									.006
Regularly	16	84	14	67	16	49	46	63
Sometimes	2	11	7	33	7	21	16	22
Rarely	1	5	0	0	10	30	11	15
Clinic professionals share clinic’ mission, values and objectives:									.728
Totally/partly disagree	3	14	1	5	4	11	8	10
Partly agree	7	33	10	46	12	32	29	36
Totally agree	11	52	11	50	22	58	44	54
Payment model for physician**:									<.001
Fee-for-service	21	96	1	4	28	78	50	60
Capitation or roster	0	0	17	65	0	0	17	20
Salary	0	0	4	15	5	14	9	11
Blended model	1	5	1	4	1	3	3	4
Other	0	0	3	12	2	6	5	6
Team size*:									.013
≤5	2	9	10	39	10	26	22	25
6-10	9	41	1	4	7	18	17	20
≥11	11	50	15	58	22	56	48	55
Patients registered in clinic*:									.024
≤3000	3	16	12	46	13	38	28	35
3001-10,000 patients	12	63	10	39	8	24	30	38
˃10,000 patients	4	21	4	15	13	38	21	27

*p < 0.05, **p < 0.01; chi-square test was used to test differences of practices characteristics across regions. Clinic internal meeting to discuss clinical issues missing data: 4(5%); care coordination through informal or ad hoc exchange_missing data or not applicable:11(13%); care coordination through EMR_missing data or not applicable: 14(16%) and sharing clinic’s mission, values and objectives among health professionals_missing data or nor applicable: 6(7%)

Practice characteristics. Sixteen percent of participating practices reported they never had internal meetings to discuss clinical issues. One in five practices reported meeting “at least weekly” and another 45% reported meeting monthly. Three-quarters (75%) of practices reported regularly coordinating care through informal or ad hoc exchange. Practices in Central Zone, reported significantly lower (56% versus 90% Fraser East or 91% Eastern Ontario; p < .001) regular care coordination through informal or ad hoc exchange than the other two regions. Care coordinated through electronic medical records was significantly higher (p < .01) for practices in Fraser East (84%) followed by Eastern Ontario (67%) and Central Zone (49%). Fifty-four percent of practices reported they totally agreed that their professional shared the mission, values, and objectives. The majority (60%) of physicians were remunerated through fee-for-service; almost all (96%) in Fraser East and 78% of those in Central Zone. Practices in Eastern Ontario reported being paid mostly (65%) through a capitation/roster payment model (Table 1). The majority (38%, p < .05) delivered care to 3,001–10,000 patients across all three regions. The majority (55%) of participating practices had large teams (≥ 11 professionals); Fraser East had few practices (9%) with a small team size (≤5), followed by Central Zone (26%) and Eastern Ontario (39%; Table 1).

There were a total of 889 participants working in the 63 practices included in this analysis. These included family physicians (n = 261), ranging from 1 to 14 per practice, nurse practitioners (n = 30), registered nurses (n = 59), pharmacists (n = 4), physiotherapists (n = 4), psychologists (n = 14), chiropractors (n = 4), dietitians (n = 16), social workers (n = 13), occupational therapist (n = 1), physician assistant (n = 1), respiratory therapists (n = 2), pyscho-geriatrician (n = 1), other staff (n = 48) and administrative staff (n = 431). Among the 566 who completed the TCI, 37% self-identified as family physicians (n = 209), 20% as other health professionals (n = 114) and 43% were administrative staff (n = 243).

Around one-third of the practices were comprised of family physicians, registered nurses, and other health professionals (Table 2). Eastern Ontario had more practices (48%) with teams composed of general practitioners, registered nurses, and other health professionals while Fraser Easter has more practices with only general practitioners (50%) and more practices (35%) with multidisciplinary teams without registered nurses (Table 2).

Table 2. Team composition and team climate inventory scores across regions (n = 63).

Characteristics	Regions						Overall		p-value
	Fraser East, BC (n=20)		Eastern Ontario, ON (n=21)		Central Zone, NS (n=22)
Team composition*	n	%	n	%	n	%	n	%
GPs only	10	50	6	29	6	30	22	35
GPs and RNs only	2	10	4	19	3	15	9	14
GPs, RNs and others	1	5	10	48	7	35	18	29	.035
GPs+others excluding RNs	7	35	1	5	4	20	12	19
Team climate inventory score	M	SD	M	SD	M	SD	M	SD
Participative safety	77	13	74	11	76	11	75	11	.790
Team objectives	70	12	75	12	71	9	72	12	.231
Task orientation	72	12	72	13	69	11	71	11	.662
Support for new ideas	72	12	74	14	70	11	72	12	.526
and innovation
Overall TCI	73	11	74	12	72	10	73	11	.789

M: Mean; SD: Standard deviation; *: Statistically significant at p < .05; **: Statistically significant at p < .01; ***: Statistically significant at p <.001; GPs: General practitioners; RNs: registered nurses.

Team Functioning. The overall mean score of the team climate inventory was 73 on a scale of 0–100 and team functioning in primary care was similar across regions. Although not statistically significant, the team objective score was lower (M= 69.6, SD = 12.3) in Fraser East, compared to Eastern Ontario (M= 75.4, SD = 12.2). Task orientation (M= 69.1, SD = 10.6), supportive new ideas and innovation scores (M= 69.6, SD = 11.0) were lower in Central Zone, compared to other regions (Table 2).

In Ontario, the TCI scores were lower in practices with a team of 6–10 professionals (M = 59.7, SD = 0.0) while in Central Zone and East Fraser, the TCI score was lower in practices with large teams. Although not statistically significant, the practices with teams composed of family physicians, registered nurses and other health professionals had lower TCI scores in Eastern Ontario and Central Zone compared to teams composed of family physicians and registered nurses only (Table 3).

Table 3. Team functioning scores overall and by team size and team composition for eligible practices (n = 63).

	TCI mean score (SD)						Sig.
Variables	Fraser East, BC (n = 20)		Eastern Ontario, ON (n = 21)		Central Zone, NS (n = 22)
Team Size^a	M	SD	M	SD	M	SD	NS
≤5	77	0	80	14	78	9
6–10	75	4	60	0	78	2
≥11	71	15	71	9	69	9
Team composition^a
GPs only	71	7	75	8	74	9	NS
GPs+RNs only	78	11	82	20	78	8
GPs+ RN and others	72	0	69	9	65	11
GPs+others excluding RNs	73	11	90	0	74	9

NS: Non statistically significant; ^a: ANOVA; M: Mean; SD: Standard deviation

The bivariate analyses suggest a significant (p < .001) positive association between participants’ responses about sharing the practice’s mission, values, and objectives among health professionals and the overall TCI score and subdimensions. The results indicate an association between team composition and TCI score with teams composed of family physicians, registered nurses and others having lower TCI scores compared to the teams of family physicians and registered nurses only (p< .05; Table 4). The results of the one-way ANOVA test showed a significant (p <.05) negative association between team size and overall TCI score. Practices with larger team size (≥11) have lower team climate score (M= 69.9, SD = 10.6) compared to practices with small team size (M= 79.2, SD = 11.8), specifically, practices that have larger number of general practitioners have lower team climate score, with practices of ≥6 general practitioners having lower team climate score (M= 68.5, SD = 7.6) compared with practices that have 1–2 general practitioners (M= 77.4, SD = 10.1). Number of administrative staff or other health professionals was not associated with TCI score (Table 4).

Table 4. Bivariate analysis results between practice characteristics and team functioning in primary care (n = 63).

	TCI-overall
Factors	M	SD	F	P-value	η^2
Frequency of internal clinic meetings to discuss…Clinical issues			1.76	.17	.082
At least weekly	76	9
Monthly	70	10
Other	76	12
Never	74	14
Care coordination through informal or adhoc exchange			.09	.91	.003
	73	12
Regularly	75	9
Sometimes	72	6
Rarely
Care coordinated through electronic medical records (EMRs)			1.49	.23	.054
RegularlySometimes Rarely	74 68 73	11 11 6
Clinic professionals share clinic’ mission, values and objectives.			6.28*	.001**	.255
Totally disagree	57	16
Partly disagree	53	7
Partly agree	73	8
Totally agree	75	10
			.92	.46	.060
Payment model for physicians
Fee-for-service	74	10
Capitation or roster Salary	74 72	12 12
Blended model	65	14
Other	63	10
Team size			4.32	.02	.126
≤5	79	12
6-10	74	6
≥ 11	70	11
Patients registered in clinic			4.41*	.02*	.132
≤ 3000 patients	78	10
3001-10000 patients	71	11
˃10,000 patients	68	10
GPs			3.78*	.03*	.112
1-2	77	10
3-5	71	11
≥6	69	8
Administrative staff			2.78	.07	.085
1-3	76	11
4-8	73	11
≥ 9	68	9
Other health professionals			.48	.62	.025
1-2	75	16
3-5	72	9
>=6	71	8
Team composition			2.82*	.05*	.129
GPs only	73	8
GPs+RNs only	80	13
GPs +RNs and others	68	9
GPs+others excluding RNs	75	14

Note: *: P < .05; **: p < .01; ***: p < .001. Care coordination through informal or adhoc exchange missing responses: n = 6. Care coordinated through electronic medical records (EMRs) missing responses: n = 7. Clinic professionals share clinic mission, values, and objectives: n = 4. Patients registered in clinic missing responses: n = 2. Post hoc comparisons using the Tukey HSD test indicated that the mean score of overall TCI for totally agree was significantly higher than totally disagree and partly disagree. The mean score of TCI for partly agree was significantly higher than totally disagree and partly disagree. Post hoc comparisons indicated that larger team size had a significantly lower overall TCI score compared to small team size. M= mean; SD =standard deviation; TCI = Team Climate Inventory. The mean score of TCI for teams of GPs+RNs only was significantly higher compared to the teams composed of GPs+RNs and other health professionals

Table 5 shows the correlations between independent, confounding, and dependent variables. Higher team functioning was significantly related to smaller team size, smaller patient panel, and greater alignment of staff sharing the clinic’s mission, values, and objectives.

Table 5. Correlation matrix of independent and dependent variables.

Variables	1	2	3	4	5	6	7	8
(1) Physician payment model	1
(2) Team size	.117	1
(3) Patients registered in clinic/ patient panel	−.065	.557**	1
(4) Clinic professionals share clinic’s mission, values and objective.	−.002	−.013	−.139	1
(5) Clinic internal meeting to discuss clinical issues	.011	.138	.203	.055	1
(6) Care coordination through informal or adhoc exchange	.221	.116	−.047	−.167	.262*	1
(7) Care coordination through electronic medical records	.252*	.170	.084	.076	.067	.428***	1
(8) Overall team functioning	−.209	−.354**	−.353**	.436**	−.086	−.003	.107	1

*: Significance at p <.05; ** p <.01; *** p < .001 (two-tailed).

The hierarchical multiple linear regression shows that higher scores of higher team functioning remains significantly associated with practices sharing the same mission, values, and objectives and smaller team size. Respondents in practices with total and partial agreement on sharing practice’s mission, values, and objectives among health professionals demonstrated ≥15 points higher on their TCI scores compared to total disagreement on scale of 0–100. Higher team functioning is also significantly related to regular care coordination through informal meeting or ad hoc exchange was also associated with higher team functioning. Practices with regular care coordination through informal meetings or ad hoc exchanges demonstrated 11 points higher of TCI score compared to those with never coordinating care with informal meeting or ad hoc exchange. Care coordination through EMRs and team size ≥ 11 were associated with lower team functioning (Table 6). Respondents in practices with regular and sometimes care coordination through EMRs demonstrated 9 points lower of TCI score compared to those which never coordinate care through EMR. Practices with large team size (≥11) had 11 points of TCI score lower compared to practices with small team size (≤ 5).

Table 6. Primary care practice predictors of team functioning in primary care-hierarchical regression analysis results (n= 63).

Variables		R^2		Coefficients		95% CI for B		Sig.
		R^2	∆R^2	B	SE B	Lower	Upper
Step 1		.152						.045
	Constant			79.399
	Team size 6–10			−3.104	4.429	−11.969	5.761	.486
	Team size ≥11			−5.658	4.320	−14.305	2.989	.195
	Panel size_3001-10,000			−3.439	3.709	−10.863	3.984	.358
	Panel size >10,000			−5.795	4.340	−14.481	2.892	.187
Step 2		.570	.418					.002
	Constant			67.251		56.241	78.261	<.001
	Team size 6–10			−6.930	4.643	−16.287	2.427	.143
	Team size ≥11			−10.764	4.406	−19.643	−1.886	.019
	Panel size_3001-10,000			−.228	3.441	−7.163	6.708	.948
	Panel size >10,000			.055	4.242	−8.494	8.603	.990
	Fee-for services			2.436	4.221	−6.072	10.944	.567
	Payment capitation			3.884	4.549	−5.374	13.143	.402
	Payment-blended			−3.344	7.506	−18.471	11.782	.658
	Payment-others			−6.942	6.053	−19.141	5.257	.258
	At least weekly clinic meetings			−.586	5.495	−11.660	10.498	.916
	Monthly clinic meeting			−5.993	4.423	−14.906	2.921	.182
	Other meeting frequencies			−1.675	5.262	−12.280	8.930	.752
	Regular care coordination through informal and adhoc exchange			10.763	4.743	1.204	20.322	.028
	Care coordination through informal and adhoc exchange-sometimes			12.257	6.200	−.239	24.753	.05
	Regular coordinated care through EMR			−9.147	4.261	−17.734	−.560	<.037
	Sometimes coordinated care through EMR			−19.436	5.132	−29.778	−9.094	<.001
	Extent of sharing clinic mission, values and objectives among professionals-Partly disagree			3.772	8.796	−13.954	21.499	.670
	Extent of sharing clinic mission, values and objectives among professionals-Partly agree			15.172	5.184	4.723	25.620	.005
	Extent of sharing clinic mission, values and objectives among professionals-Totally agree			17.346	4.806	7.659	27.033	.001

R² = .152 for step 1, ∆R² = .418 for step 2. * p < .005; **p <.01; ***p <.001

Physician’s payment model reference (salary); frequency of clinic meetings to discuss clinical issues_reference (never); care coordination frequency through informal and adhoc exchange_reference (rarely); care coordinated through Electronic Medical Records (EMR)_reference (rarely), Extent of sharing clinic mission, values and objectives among health professionals_reference (Totally disagree); Team size) _reference (team size ≤5) and panel size-reference (≤3000 patients)

In step 1, team size and panel size significantly contributed to the model and explained 15% of the total variance in team climate scores. In step 2, physician payment model, sharing clinic’s mission, values and objectives among health professionals, clinic internal meetings to discuss clinical issues, care coordination through informal and ad hoc exchange and care coordination through electronic medical records significantly contributed additional 42% to the total variance in team climate scores. Altogether, seven variables (team size, patient panel, physician payment model, sharing clinic mission, values, and objectives among clinic health professionals, clinic internal meeting to discuss clinical issues, care coordination through informal or ad hoc exchange and care coordination through electronic medical records) explained 57% of the total variance in team climate inventory scores. The final model indicated that sharing clinic mission, values and objectives among clinic health professionals, care coordination through informal and adhoc exchange were significant variables associated with higher team functioning while team size and care coordination through electronic medical records were significant variables associated with lower team functioning (Table 6).

Discussion

This is the largest and geographically diverse quantitative study of team functioning in primary care practices across Canada. Our findings suggest that high-functioning teams have some common practice characteristics including shared mission, values, and objectives of the practice, regular care coordination through informal meetings or ad hoc exchanges and team sizes with five or fewer staff.

Past work byAl-Sayah et al. (2014) and a systematic review by Wranik et al. (2019) also found that sharing practice’s mission, values, and objectives/goals among health professionals positively influences team functioning in primary care. Alignment of common practice goals means that team members engage in an open communication, value each other’s professional roles and skills, and respect, trust, and support each other (Royal College of Physicians, 2017).

It is not surprising that regular informal communication and interactions between team members around care coordination increase team functioning. Our quantitative work aligns with past qualitative studies that suggest a positive influence of practice team meetings on team functioning (Sørensen et al., 2018; Sullivan et al., 2018; Valaitis, Wong, et al., 2020). Team meetings provide a safe environment for members to share and exchange information regarding patient care, care coordination and clinical decision-making and enhance team functioning (Mayo & Woolley, 2016; R. Valaitis et al., 2020). Informal and ad hoc exchanges between team members enhance their familiarity with each other and promote mutual respect that can result in team cohesiveness, psychological safety, and trust relationships for a better team climate (Harris et al., 2016; Mayo & Woolley, 2016;). Trust, in turn, makes the development of shared goals possible (Harris et al., 2016).

Higher participative safety scores, as observed in this study might contribute to higher team functioning since psychological safety influences participation in shared clinical decision-making (Mayo & Woolley, 2016). When teams mostly communicate through their EMRs, team functioning was lower. Past work has found that EMRs facilitate communication between team members, sharing patient’s clinical information and care coordination (Gocan et al., 2014; O’Malley et al., 2015; R. Valaitis et al., 2020) and are associated with higher team functioning in primary care (Howard et al., 2011). Yet, as teams become larger and if EMRs are not effectively configured or implemented, their use may interfere with or replace a minimum productive amount of face-to-face interactions and communication between team members (Mundt et al., 2016). While whole bodies of work are further developed on improving complex human-facing design issues given our computationally intensive systems (Designing For People, 2021), more work is needed on how EMRs can help primary care teams function better. Given the recent pivot to virtual care, this finding is especially important to further identify whether it is quantity or quality of face-to-face interactions that improve team function.

Relative to the smallest teams (≤5), teams with 11 or more members were associated with a lower TCI score which is consistent with past work (Khan et al., 2021; Proudfoot et al., 2007), including a systematic review (Mulvale et al., 2016) where a negative association was found between team size and team climate in primary care practices. Whilst diverse professionals are expected to mobilize their skills, knowledge, and competences supportive for innovation in patient care, a larger team size experiences challenges in communication, interactions, and information exchange (Hysong et al., 2019; Zijl et al., 2021) which might complicate team functioning, potentially explaining this finding. This finding may also be influenced by the increased TCI scores associated with increasing informal care coordination. The larger the team, the more difficult it would be to interact with most team members involved in a patient’s care, thus requiring more EMR communication. Future research is needed to determine the optimal team size for effective team functioning and quality of care (Devlin et al., 2013). Partial or total agreement with sharing the practice’s values was positively associated with a higher TCI score which is consistent with the extant literature.

We found that other practice characteristics, however, were not found to be significantly associated with overall team climate, including physician payment model. Multiple qualitative studies highlight the negative impact of the physician fee for service payment model on team functioning in primary care (Freund et al., 2015; Russell et al., 2018). Some researchers have suggested that that the fee-for-service payment model prevents physicians from collaborating with other professionals (Pype et al., 2018; Russell et al., 2018), limiting opportunities for interactions between team members (Pype et al., 2018). Our work suggests regardless of physician payment model, there are more important practice characteristics influencing team functioning.

Limitations

Our work is limited in that we completed a cross-sectional analysis where we could assess associations but not examine causality. Another limitation includes the complexity in measuring fractional full-time equivalents (FTEs) in considering counts of professionals. More work is needed to understand staffing models in terms of FTEs and how these may impact team climate. Given that the sample completing the TCI was composed mostly of family physicians and administrators, the perspectives of other health clinicians may be insufficiently represented. While data were conducted 5 years ago, the results are still relevant in the current context because team-based care remains nascent in Canada (Khan et al., 2021). New teams are still being formed in primary care settings (Sibbald et al., 2021). Organization of primary care services delivery has not dramatically changed; therefore, the results provide insights on the slow adoption of team-based care and how it functions in primary care. More work is needed on examining whether there are cut-points that can differentiate high-, middle-, and low-functioning teams. The results of this study could contribute in evidence synthesis in the future regarding team-based primary care.

Conclusion

This study presents the largest collection of data on team functioning in primary care in Canada. These results contribute to a growing body of work to understand and plan for improving team functioning within primary care settings. The results showed that care coordination through human interaction and sharing the practice’s mission, values, and objectives among health professionals were positively associated with higher functioning teams. Care coordination through electronic medical records and larger team size were negatively associated with team functioning. Establishing strategies that can foster care coordination through informal or ad hoc exchange between care providers and shared mission, values, and objectives among professionals could facilitate optimization of the team-based care. Interventions that optimize team size and ensure that EMRs are appropriately utilized and enhanced could support effective team communication and information sharing.

This study provides contribution to interprofessional primary care team literature by identifying organizational factors that influence effective team functioning in primary care and suggesting further research to close knowledge gap. Policy-makers and practitioners could use these findings to develop interventions that target these characteristics to support effective team functioning in primary care settings.

Authors’ contribution

STW, RMM, WH, WW, FB, KMM, SJ: Contributed to the conceptualization of the study, study design, supervise data collection, data analysis, interpretation of results, drafting the manuscript, critically revising the manuscript and approved the submission of the manuscript.

IN, JB: Data analysis, interpretation of the results, drafting the manuscript, revising the manuscript and approved the submission of the manuscript.

Additional information

Ndateba, I: RN MN, MPH, PhD student at University of British Columbia School of Nursing and Centre for Health Services and Policy Research

Wong, S.T: RN PhD: Professor at University of British Columbia School of Nursing and Centre for Health Services and Policy

Johnston, S: LLM MD FCFP: Associate Professor at University of Ottawa, Department of Family Medicine

Martin-Misener, R: RN PhD: Professor at Dalhousie University School of Nursing

Hogg, W: MD FCFP: Professor at University of Ottawa, Department of Family Medicine

Wodchis, W.: PhD: Professor at University of Toronto, Institute of Health Policy, Management and Evaluation

Burge, F.: MD FCFP: Professor at University of Dalhousie University, Faculty of Family Medicine

Beaumier, J: RN, MPH: University of British Columbia School of Nursing and Centre for Health Services and Policy

McGrail, KM: MPH PhD: Professor University of British Columbia School of Population and Public Health, and Centre for Health Services and Policy Research.

Acknowledgments

We thank all participants for accepting to participate in this study. We also thank partners who provide valuable contribution in the development and conduct the research. The study was funded by the Canadian Institute for Health Research (CIHR) and the Michael Smith Foundation for Health Research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/13561820.2022.2099359

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.