Improving Collaboration between Public Health and Family Health Teams in Ontario

Abstract

Objectives:

To identify and explore areas where responsibilities may overlap between family health teams (FHTs) and public health units (PHUs); to identify facilitators or barriers to collaboration; and to identify priority areas for increased collaboration.

Design and context:

Cross-sectional mixed-methods study of FHTs and PHUs in Ontario, Canada, consisting of a postal survey, key informant interviews and a roundtable meeting.

Results:

The survey response rate was 46%. Direct client-based services such as giving immunizations, promoting prenatal health and nutrition, and counselling related to smoking cessation were identified as the top three areas of perceived overlap. The greatest interest in collaboration was expressed in the areas of emergency planning and preparedness, immunization, and prenatal health and nutrition. Good communication with a clear understanding of roles and functions was the most important facilitator, and lack of resources and absence of a clear provincial mandate and direction to collaborate were identified as significant barriers.

Conclusions:

Small, simple client-based projects of interest to both kinds of organization would be the best way to move forward in the short term. Improving communication between FHTs and PHUs, understanding of roles and functions, the use of shared or interoperable information systems and greater clarity from government on the ways in which these two key sectors of the healthcare system are intended to work together were identified as important for the success of increased collaboration.

Both public health and primary care are undergoing significant renewal and re-organization in Ontario. In public health, the challenges were brought to the forefront by the 2003 SARS outbreak. Changes implemented in the wake of the Naylor and Walker reports have resulted in significant investments, revised program standards and the development of a provincial agency to provide scientific and technical advice for health protection and promotion (Naylor et al. 2003; OAHPP 2011; Walker et al. 2003). In primary care, Ontario has moved significantly away from traditional, individual physician fee-for-service–based practices to a variety of patient enrolment models (PEMs) and has introduced multidisciplinary family health teams (FHTs) (OMA 2008). FHTs are multidisciplinary group practices that typically include family doctors, nurses, nurse practitioners and one or more other allied health providers, most commonly pharmacists, social workers or dietitians. Most patients in Ontario are now served by practices in one of the PEMs, with 20%–25% being enrolled in FHTs. FHTs are being encouraged to use population health approaches to program planning and delivery, approaches that have historically been limited to use by public health agencies or large regional health services delivery agencies. One result of these changes is potential overlap in program areas, such as health promotion and population health assessment, that each of these groups might consider part of its responsibilities. There are apparent opportunities for collaboration and more efficient service delivery in areas of overlap, yet the degree of collaboration between FHTs and local and provincial public health units (PHUs) or programs is currently fairly limited.

Rowan and colleagues (2007) describe a number of models of integrating primary healthcare (PHC) and public health that have been proposed and tested in a number of Western countries, including Canada. These models include full integration of public health and PHC functions in a single health authority with a shared governance structure and funding pool, focused collaborations on specific chronic disease prevention initiatives, the use of practice facilitation to improve quality of care delivery in disease prevention or infection control practices, secondment of public health staff to primary care delivery sites and application of the Community-Oriented Primary Care Model (COPC), which involves targeted delivery of a comprehensive set of preventive and primary care interventions to a priority population. As FHTs form and begin to develop, they are evolving business plans that include meeting key quality targets in health promotion and disease prevention activities. They are also grappling with the need to develop evidence-based approaches to determining the health status and health needs in the communities they serve. As a result, interest in building closer links with public health is growing. A recent report by the Perth District Health Unit (2006) summarizes the issues and direction in its health unit, which is interested in pursuing a COPC model in partnership with local FHTs. Other PHU–FHT partnerships are setting different priorities, and in many areas, the linkages are not yet clearly defined. Valaitis and colleagues (2009) have recently reviewed the status of collaboration between primary care and public health in Ontario and found that for the most part, these sectors continue to function independently of each other. They also found that recent changes to both sectors have the potential to enhance the degree of collaboration that exists.

This study was designed to improve our understanding of the ways in which FHTs and local PHUs might work more closely together. Specific objectives included identifying the areas in which there might be overlapping mandates between FHTs and local PHUs; FHT priorities and programs that would assist local health units in improving population health outcomes in their region (and vice versa); current or proposed areas of integration between FHTs and public health, and any facilitators or barriers to collaborations that have been experienced to date; and additional opportunities for collaboration between FHTs and PHUs.

Methods

This study used methodological triangulation including a survey, key informant interviews and a facilitated roundtable meeting to identify patterns of responses and themes. All aspects of the study received ethics approval from the Queen's University Health Sciences Research Ethics Board.

Survey design and distribution

Survey content was developed by scanning existing websites of PHUs and FHTs in Ontario for information about collaborative initiatives underway. Additionally, a list of programs and services offered by the organizations was generated as the basis for a comprehensive list of services used in the survey. All respondents had the opportunity to respond to any item on the list even if was not listed on their organization's site, and an open-ended question was included to capture any missed areas. Draft versions of the survey tool were reviewed by a small number of practising family physicians and public health professionals. As the groups could not agree on a common format for the survey, two versions covering the same material were produced, one for FHTs and one for PHUs. While most content was identical, the PHU version gave the opportunity to offer additional detail on which aspect of the service in question respondents provided (they were asked if a given program/service was considered Health Promotion, Health Protection, Disease Prevention, Clinical Counselling or Clinical Management, or some combination of these).

Surveys were mailed to all known FHTs (n=134) and PHUs (n=36) in Ontario (total N=170) in the fall of 2008. FHT surveys were directed to the lead physician, and PHU surveys to the medical officer of health. A reminder letter was mailed several weeks after the initial mailing to all PHUs and FHTs who had not responded. For each of the 40 possible programs and services included on the questionnaire, respondents were asked whether their organization provides that program/service and whether they felt there was overlap between PHUs and FHTs regarding that program/service. They were also asked to rate, on a 1 to 5 scale, whether their organization had an interest in integration (1 = very low priority, 5 = very high priority). Finally, they were given an opportunity to list other existing/planned linkages and to elaborate on factors that they felt promoted or impeded PHU/FHT integration. Data were entered and analyzed in SPSS v. 17.0.

Interviews

To elicit more detailed data on how PHUs and FHTs could contribute to each other in terms of integrated service provision, factors that promote or impede integration and perceived priority areas for integration, nine interviews were conducted with representatives from FHTs (n=4), PHUs (n=4) and a researcher with experience examining healthcare collaboration (n=1). All were conducted via telephone, and the average interview length was 26.2 minutes (ranging from 16.6 to 31.4 minutes). Interviews were recorded and transcribed verbatim. Transcripts were entered into NVivo v. 8 for analysis. Sample transcripts were reviewed for themes by a group of four investigators that included both primary care and public health perspectives, and a coding scheme was developed. The remaining transcripts were then reviewed by two investigators.

Roundtable meeting

In an effort to distill and build consensus on priority areas and key issues regarding integration, a one-day meeting was held that included participants from various primary care and public health organizations and that included policy makers, front-line workers and researchers. Invitations were sent to organizations, which then designated a delegate. Organizations were selected based on a number of factors including geographic location, level of responsibility (local/provincial), type of organization (government, local PHU, professional association, FHT) and focus (service delivery vs. policy and planning). A small number of researchers known to be active in this field were invited directly. The purpose of the meeting was to (a) identify barriers and obstacles to collaboration between public health and primary care, (b) generate a list of the most suitable programs/areas for collaboration, (c) develop strategies for effective collaboration and (d) identify key structures/processes that should be in place to facilitate collaboration. This meeting was run by a trained facilitator from the Queen's Executive Decision Centre, who utilized a modified nominal group technique (Jones and Hunter 1995; Lloyd-Jones et al. 1999) within an electronic meeting system approach. This approach has the advantages of allowing “groups to rapidly accelerate idea generation and consensus building. It consists of a network of laptops accessing software designed to support idea generation, idea consolidation, idea evaluation and planning. … [It] supports but does not replace verbal interaction” (Masotti et al. 2009). This approach democratizes a meeting by levelling the influence of differing personalities. Ideas are generated by the group, similar ideas are combined to reduce the number of possible options, and these options are voted on anonymously to produce ranked lists of priority/top choices.

Results

Surveys and interviews

The survey response rate was 46% and evenly distributed, as 46% of FHTs responded (n=61) compared to 47% of PHUs (n=17). We would also like to note that while just under half of all health units participated in this study, those that did included larger health units that represent a substantial proportion of the provincial population. Overall, PHUs indicated that only one-third or less of the services they deliver were client-based services delivered in clinical settings (32% for clinical counselling, 22% for clinical management), while the rest were population-based health promotion and disease prevention strategies. This situation differs from that of FHTs, where nearly all services are delivered to clients in clinics.

Table 1 describes the service areas where there was the greatest overlap in service provision. All areas are included for which at least 50% of respondents in each category (PHU and FHT) indicated providing services. Respondents were asked to identify those areas in which they perceived the greatest overlap between PHUs and FHTs.

TABLE 1.

Services provided by both family health teams (FHTs) and public health units (PHUs)

Program/Service	Total (%) (95% CI) N=78	FHT (%) (95%CI) n=61	PHU (%) (95%CI) n=17
Immunization	100 (95–100)*	100 (94–100)*	100 (80–100)*
Chronic disease prevention	96 (89–99)	100 (94–100)*	82 (56–96)
Smoking cessation	95 (87–99)	98 (91–100)	82 (56–96)
Nutrition	94 (86–98)	92 (82–97)	100 (80–100)*
Sexual health counselling/Assessment/Treatment	92 (84–97)	90 (80–96)	100 (80–100)*
Cancer screening	92 (84–97)	98 (91–100)	71 (44–90)
Prenatal health/Nutrition	91 (84–97)	89 (81–97)	100 (80–100)*
Child development	90 (83–97)	87 (79–95)	100 (80–100)*
Vaccination education	86 (78–94)	84 (75–93)	94 (83–100)
STI prevention	86 (78–94)	82 (72–92)	100 (80–100)*
Infectious disease prevention/Treatment	79 (70–88)	74 (63–85)	100 (80–100)*
Infection control	79 (70–88)	73 (62–84)	100 (80–100)*
Postpartum health/Support	77 (68–86)	70 (58–81)	100 (80–100)*
Physical activity promotion	74 (64–84)	67 (55–79)	100 (80–100)*
Addictions prevention	69 (59–79)	62 (50–74)	94 (83–100)
Emergency planning	68 (58–78)	59 (47–71)	100 (80–100)*

^*1-sided 97.5% CI.

Table 2 presents those areas where at least 50% of respondents felt that there was overlap in service provision. The areas that show the largest discrepancy in the percentage of PHUs and FHTs reporting overlap were in the topic areas of addictions assessment/treatment/referral (difference = 37%), ongoing primary care (34%), cancer screening (33%), adolescent services (27%), family services (24%) and mental health (24%). These are areas where differing perceptions exist such that one type of organization believes there is overlap between PHUs and FHTs in terms of delivering this program or service, while the other type of organization does not believe there is overlap. Respondents were also asked to indicate their level of interest in improving integration for specific programs/services on a 1 to 5 scale (1 = very low priority, 5 = very high priority). Table 3 presents the areas where both FHTs and PHUs rated their level of interest as 3.00 or greater.

TABLE 2.

Perceived overlap in service provision

Program/Service	Total (%) (95% CI) N=78	FHT (%) (95%CI) n=61	PHU (%) (95%CI) n=17
Immunization	77 (66–86)	75 (63–86)	78 (50–93)
Prenatal health and nutrition	70 (59–80)	67 (54–79)	76 (50–93)
Smoking cessation	68 (56–78)	67 (54–79)	71 (44–89)
Vaccination education	65 (54–76)	61 (47–73)	82 (57–96)
Child development	63 (51–74)	62 (49–74)	65 (38–86)
Sexual health counselling/Assessment/Treatment	61 (50–72)	57 (45–69)	76 (50–93)
STI prevention	58 (47–69)	58 (46–70)	59 (36–82)
Chronic disease prevention	54 (43–65)	51 (38–64)	67 (45–89)

TABLE 3.

Interest in improving integration

Program/Service	Overall N=78	FHT n=61	PHU n=17
Emergency planning	3.85	3.89	3.67
Immunization	3.81	3.80	3.88
Prenatal health/Nutrition	3.75	3.75	3.73
Postpartum health/Support	3.68	3.85	3.00
Sexual health counselling/Assessment/Treatment	3.60	3.65	3.40
Smoking cessation	3.57	3.67	3.14
STI prevention	3.55	3.69	3.00
Chronic disease prevention	3.46	3.45	3.50
Child development	3.39	3.31	3.71
Immunization outreach clinics	3.34	3.33	3.40
Vaccination education	3.32	3.33	3.27
Infectious disease surveillance	3.19	3.25	3.00
Travel health	3.10	3.13	3.00

Mean scores on a 5-point scale where 1 = low interest and 5 = high interest.

The nature of existing or planned collaborations, such as the type of collaboration or program, were explored through open-ended questions. Follow-up, in-depth interviews with selected survey respondents who agreed to be interviewed also explored these themes, with a focus on existing collaborations. These text and interview responses were coded into themes. The most common linkage involved enhanced communication. Communication included such activities as meetings between representatives of the PHUs and FHTs, information sessions, advisory committees involving individuals from both groups, fax updates and working meals:

We did initiate some meetings with family health teams to explore how we can collaborate, and they at the same time asked for public health input pretty well on all of their advisory committees. (PHU informant)

Regarding specific programmatic areas, prenatal/infant/child/parenting programs were most often mentioned, with immunizations and smoking cessation programs close behind:

We started with an enhanced 18-month visit, trying to get every child to have had one done in our community. (FHT informant)

Pandemic planning (10 mentions) was also commonly the subject of PHU/FHT collaboration. This may have been influenced by concurrent H1N1 pandemic activity.

Interviewees were also asked to comment on the barriers to and facilitators of collaboration. Funding (structure and amount) and not understanding each other's organizations were the two main challenges noted:

Finances. Because it always comes up, well, who's paying for what, and who has got the funding. (FHT informant)

I don't think there's good understanding of each other's mandates. (PHU informant)

The importance of having a person or persons to drive the integrative efforts was emphasized:

I think there has to be a person or persons that are in a role to do that [communication]. (FHT informant)

… having one or two champions, preferably two, but at least one who sees the vision and drives it forward …. (PHU informant)

This connection could take the form of a dedicated liaison, committed leadership or some sort of champion for the cause. Additionally, it was frequently noted that communication and information sharing and knowledge of each other's role made working together much easier.

Roundtable meeting

The roundtable meeting brought together representatives from FHTs (n=10) and PHUs (n=8), policy makers (n=3), researchers (n=3) and the Ontario College of Family Physicians (n=1). A total of 20 different organizations were represented. Tables 4 and 5 summarize the results of the process undertaken to identify key areas for improving collaboration.

TABLE 4.

Top five areas for improving collaboration: Short term

Rank	Area	Votes
1	Emergency planning and preparedness	14
2	Improved communication/Start discussions on collaborating	12
3	Immunization	10
4	Healthy Children Healthy Babies	8
5	Chronic disease management and prevention*	6

“Short term” defined as achievable in under 5 years.

^*Was also ranked as a long-term priority.

TABLE 5.

Top five areas for improving collaboration: Long term

Rank	Area	Votes
1	Surveillance using electronic health records in primary care	14
2	Co-ordinated data collection and data sharing (IT)	12
3	Chronic disease management and prevention*	10
4	Mental health	9
5	Research	8

“Long term” defined as >5 years.

^*Also ranked as short term.

Participants first brainstormed ideas, then grouped similar concepts to produce a list of suggestions. For both tables 4 and 5, 15 items were on each initial list. Each participant who voted (n=19) picked his or her top four suggestions. With a few exceptions that dealt with high-level policy-related issues, the output of this process closely parallels the findings from the survey and interviews. This is likely because the survey was directed to local PHUs and FHTs, while policy makers and other provincial-level organizations were also represented at the meeting. The same process was carried out to identify barriers and facilitators to collaboration. This process confirmed the findings of the interviews regarding key barriers (lack of funding/resources, funding models, lack of understanding of each other's organizations) and facilitators (establishing formal routes of communication, increasing informal networking, improving understanding of roles and mandates).

Two key barriers not previously identified emerged during this process: the lack of a strategic plan at the provincial level and the lack of linkages between information technology (IT) systems of the two sectors. In terms of facilitating future collaborations, two additional concepts were ranked highly: starting with simple projects that meet the mandates of both organizations, and conducting joint needs assessments at the community level. A final wrap-up discussion on the barriers that needed to be addressed noted that improving the interoperability and exchange of data between PHU and FHT IT systems, and greater role clarity from policy makers at higher levels, were essential for long-term improvements in collaboration.

The example of immunization records was used to illustrate an area where both public health and PHC require access to patient-level information and where ability to share data electronically should be a top priority. Identification of a “champion,” or at least a designated individual in each organization to serve as the liaison between organizations, was also suggested by a number of participants, and was consistent with suggestions from the interviews.

Discussion

Public health and PHC are two key sectors of the healthcare system in many jurisdictions that have many shared objectives, but that often do not effectively collaborate (Martin-Misener and Valaitis 2008). This situation may be partly attributed to the different service delivery models adopted by the two sectors, whereby most primary care services are clinic-based and targeted at individual clients, whereas many public health sector services use population-based strategies. Lack of a shared professional culture and differences in perspectives and priorities can result in situations in which each sector is concerned that closer collaborations may erode their own focus or place at risk their ability to attract and control funding from government. Ontario has seen reforms in both sectors, with changes in the primary care sector, which have been significantly greater than those in public health, resulting in a renewed interest in enhancing collaboration.

Our results show that both local public health organizations and family health teams share an interest in improving the degree to which they collaborate. In addition, they largely identify the same areas in which improving collaboration is a priority. These include client-based service delivery, where public health strategies most closely approximate a clinical model of service. Such collaborations, however, are more likely to succeed with improved communication and understanding of the roles and mandates of both organizations. While financial resources were not identified as being required to improve collaboration, lack of resources for strategies – such as time for joint meetings, shared planning processes and integrated IT systems – was clearly identified as a barrier to moving forward. Many respondents reported that it would be of benefit to have one or more individuals in a position dedicated to act as a bridge between the two organizations to plan complementary service delivery and to work towards avoiding duplication. There was strong interest in IT links between the FHTs and PHUs; these were viewed as potentially improving efficiency in managing immunization records, reporting communicable disease and animal bite incidents, and so on, and in conducting population health assessments.

The results of this study also suggest that PHC and public health organizations in Ontario lack clear direction from the Ministry of Health and Long-Term Care regarding opportunities and expectations for collaboration between FHTs and PHUs, as well as dedicated funding to support collaboration. Incentives for meeting collaborative goals may assist in renewing efforts to improve collaboration between these two key sectors of the publicly funded healthcare system in Ontario. Funding models that support the initiation of such collaborations, and sustain those that are successful, should be available to both FHTs and PHUs so that efforts can be initiated by either organization.

Limitations

There are some limitations to this study. First, while the response rate was in the usual range for postal surveys of healthcare providers, it was still only 46%, which introduces the possibility of response bias. If respondents represent those people with higher levels of interest in the topic of collaboration than non-respondents, then the results of this survey will overestimate the degree of interest in improving collaboration. Similarly, key informants and workshop participants were not randomly selected and may represent a group of professionals with a higher-than-average interest in this area. We attempted to limit this impact by inviting a range of key organizations to send a senior delegate of their choice rather than specifying directly who in the organization would have to attend the meeting. In addition, this study surveyed only one model of primary care organization in Ontario, family health teams. While it is likely that some of the findings apply more broadly to other models of primary care delivery both in Ontario and elsewhere, the generalizability of the results will be somewhat limited by this focus. As noted in the methods section, the divergent opinions of PHC and public health on even the meaning and structure of some of the survey questions is also important to consider. Primary care–based respondents preferred an “all or none” question in which a service was either provided or not, while public health respondents emphasized defining those aspects of a service that were delivered. As a result, both the reported overlap in programs and the perceived overlap in activity may represent complementary activities rather than duplication of services.

Conclusions

Both PHUs and FHTs share an interest in developing opportunities to collaborate. Program areas ripe for collaboration or integration are client-based services related to immunization and prenatal/infant/child/parenting (e.g., Healthy Babies Healthy Children), as well as emergency planning. PHUs and FHTs that establish lines of communication and a dedicated liaison position are likely to be more successful in collaborating. Shared information systems are also seen as being of benefit. Simple projects that meet the mandate of both organizations, and in which successes and efficiencies come easily, would make a good starting point for any new collaborative efforts. Stronger policies from government on the expectations for collaboration, and areas in which collaboration is a priority, are required. Provision of funding opportunities or financial incentives to support expected collaborations is seen as being important for their long-term success.