The province of Alberta in Canada is home to 3.7 million people, spread across some 661,000 square kilometers. In 2009, Alberta's nine independent health authorities and three provincial entities were merged and a new entity - Alberta Health Services (AHS) - was formed to co-ordinate the delivery of health services across the province. The task was substantial, with the benefits expected to be just as big.

AHS Director of Knowledge Management (KM) - Capacity Building, Dr. Cathie Scott and KM Consultant, Laura Lagendyk are leading a large multidisciplinary mixed methods research project to understand different models of care and how they support collaboration among primary healthcare professionals. The team is using qualitative data analysis software NVivo 8 to deliver immediate outcomes to primary care networks (PCNs) in the field, as well as facilitating the overarching corporate goals of the five-year project.

Primary health care reform and CoMPaIR 

Primary healthcare (PHC) systems are undergoing significant change both across Canada and throughout the world. New models of providing PHC are being explored, but differ according to their context. The one constant is their end goal. Prevention is a key objective of primary health care, and it involves treating short-term health problems, managing long-term health conditions such as diabetes, and encouraging learning to avoid injury and illness.

While there have been many iterations of change within Alberta’s primary healthcare system the most recent reforms began in 2003. Shortly afterward the CoMPaIR project was funded by the Canadian Health Services Research Foundation (CHSRF), the Alberta Heritage Foundation for Medical Research (AHFMR) and the Calgary Health Region. CoMPaIR is an acronym for Contexts and Models in Primary Healthcare and their impact on Interprofessional Relationships. The project's primary brief was to understand the interprofessional collaborative models that healthcare professionals were using to provide care in new and innovative ways. Underlying this program of research is a focus on developing innovative strategies for making better use of evidence to inform practice.

"There has been a substantial shift in the way front-line health care services are delivered in Canada. Over the past six years primary care networks (PCNs) – where groups of family doctors work with the AHS and other health care professionals to coordinate patient care - have come to the fore," Dr. Scott explained.

"While the initiatives put into practice across each province are different, they all share a common feature – practitioners are being asked to work in new ways across professional boundaries. The CoMPaIR project provides a unique opportunity for Alberta's health care professionals to share their own work and learn about the models being used in other sectors too."

Keeping pace with the evolution of the industry 

The research project was designed in three phases: Phase One focused on three primary care networks in Calgary; Phase Two is provincial in scope; and Phase Three adds a national component.

"One of the challenges for us was that the project was already underway when the announcement was made that there would be fundamental changes to Alberta's health care system," Ms Lagendyk said.

"But what that has also meant is that we're actually studying evolving contexts that will influence how the health care industry will move forward in Alberta. That's challenging, but exciting too."

Choosing qualitative data analysis software 

While CoMPaIR involves quantitative research methods in the form of survey work and network analysis, it is largely a qualitative study. The project team's methods have been primarily interview-based, however literature reviews and observation methods have also been used.

Dr. Scott and Ms. Lagendyk have both used qualitative data analysis software in the past, and chose QSR International's NVivo 8 software for use on CoMPaIR.

"With CoMPaIR, we've done some really in-depth data collection around the context, model and outcomes that the different primary care networks have used and why," Dr. Scott said.

"This comes in the form of field notes, individual and focus group interview data, and observation data from different settings such as meetings and clinics. It's all imported into NVivo."

While NVivo 8 can work with many forms of qualitative data, including audio and video, the CoMPaIR team currently transcribes and imports all of their project data in textual formats. Each of the primary care networks is assigned as a 'case' within the software. The data is organized – according to its content and themes – in a hierarchical structure (known as a ‘tree node’ structure).

"We started with a fairly comprehensive tree that’s morphed as we’ve gone along. We've added and combined 'nodes' [or themes] as we get a little more concrete about which areas of the primary care models we’re going to compare and contrast when we put our reports back together," Ms. Lagendyk explained.

"One of the biggest benefits of NVivo is being able to easily classify the data in different ways, including grouping sets of data within NVivo. For instance, we grouped all of the interviews, references and content relating to physicians in one set; all of the nursing materials in another, and so on. Then being able to tease that apart through the matrix structure or any of the queries that you build in is a huge plus."

"NVivo lets us be quite experimental and innovative in our approach. We’re trying to play with the case approach and pull that data together in different ways, while involving different stakeholders in setting the direction for our work," Dr. Scott added.

Delivering immediate benefit in a long term project 

One of the innovative ways that the CoMPaIR research team is using NVivo 8 is to deliver ongoing insights that can be applied immediately, in what is undeniably a long-term project.

"One of the ways we've been working with NVivo is through a team based approach to a number of smaller sub-projects within CoMPaIR. We'll partner with the research members of the team as well as the actual people within the networks to isolate and examine a topic," Dr. Scott said.

"For instance, Laura was working with two primary care networks that took very different approaches to chronic disease management. One had created a stand-alone clinic, and the other was developing a community of practice involving people who were doing hands-on care in diabetes.

"We already had interviews from people within each of those chronic disease clinics and we analyzed them as a mini-project to look at each model and its advantages and disadvantages, and shared those insights."

Dr. Scott and Ms. Lagendyk are also working in a similar way to share immediate insights with health professionals using CoMPaIR's newsletter, ‘The Echo’. The insights shared in the newsletter may be the result of specific sub-projects or based on the research team's observations of knowledge gaps or specific stakeholder wants and needs.

"It's about picking up on those interesting insights or questions that we’re seeing when we’re doing data collection," said Ms. Lagendyk.

"The first one we did was in response to a lot of people within primary care networks asking about how to create policy - they had very little experience with that kind of thing. So the students that were working with us pulled some literature and quotes out of our NVivo database and did a short, two-pager about policy.

"Another was around one of the small sub-projects we completed with social workers. We did targeted interviews with a few social workers and then combined that with the results of our network analysis and released a short paper on what social workers actually do within this primary care network. That was in direct response to the health care professionals in the network, who wanted us to work with them on articulating their roles.

"NVivo 8 has helped us to deliver those immediate results that can be shared across our networks – and that work will still contribute to the larger study. The fact that we can use NVivo 8 to code data and look at it as a stand-alone project and then also seamlessly merge it in with the rest of our data is a huge advantage."

Facilitating team work 

Today the CoMPaIR project team involves no less than 20 researchers and field-based practitioners. Dr. Scott said whereas on other research projects, all team members often code and work within NVivo 8 together, potential conflicts of interest prevent this from happening on their project. However the use of software is still helping to facilitate outcomes across the team.

"Our team isn't made up of just academics, but practice based people too," she said.

"We have specialists across diverse areas including patient family centered care, public engagement, health policy, workforce optimization, economics, knowledge management, as well as physicians and other health professionals who are embedded within primary care teams. Because some people are actually mentioned by others in the interviews, we haven’t, for ethical reasons, opened up the database to the whole team. We’ve had to come up with a strategy that allows each researcher to get what they need out of the data without actually being in the data."

In response, Dr. Scott and Ms. Lagendyk moved away from whole-of-team engagement and instead allowed smaller groups to have ownership over their focus areas through partnering them in NVivo.

"Each of those key members now frames the questions that they want to ask the data, sends those to us and we use NVivo to pull out content that may address their questions," Dr. Scott explained.

"We then have discussions with that team member about what they’re seeing and what’s been pulled out, and how it helps them understand their particular area of interest.

"It has increased our team's level of engagement and facilitates far better outcomes than trying to work across such a large team using manual methods. Especially for this disbursed team, which spans several countries, it's a new way of approaching research, which is quite exciting."

Data sharing and portability 

Dr. Scott also pointed to the value of having a defined analysis structure when introducing new team members to an established project.

"We’re looking at the prospect of bringing in another person to manage the project," she said.

"Even if we didn’t hand them the whole NVivo 8 database straight away, it seems much easier to be able to give them our tree structure and get them starting to think about the ways that we’ve been looking at the data.

"Rather than me doing three or four major brain dumps and having to pick up where I left off, they would have a well established database and documented analysis across several years as their starting point."

Delivering on the project as a whole 

The CoMPaIR project runs until the end of 2010. At its conclusion the research team will share what they've learnt about the models of primary healthcare across Canada. The results will contribute to an in-depth understanding of the influencing factors in the development of primary health care models. Specifically, Dr. Scott and Ms. Lagendyk's research will help health bodies like the AHS understand what's needed to make these systems work, for the good of both the healthcare professionals and the communities they support.

About Alberta Health Services 

Alberta Health Services (AHS) is tasked with co-ordinating the delivery of health supports and services across the Alberta province. The organization brings together 12 formerly separate health entities in the province: nine geographically based health authorities (Chinook Health, Palliser Health Region, Calgary Health Region, David Thompson Health Region, East Central Health, Capital Health, Aspen Regional Health, Peace Country Health and Northern Lights Health Region) and three provincial entities working specifically in the areas of mental health (Alberta Mental Health Board), addictions (Alberta Alcohol and Drug Abuse Commission) and cancer (Alberta Cancer Board). It also includes Alberta's ground ambulance service. The AHS workforce is made up of 90,000 caring and dedicated individuals. Information: www.albertahealthservices.ca

About Dr. Cathie Scott and Ms. Laura Lagendyk

Dr. Cathie Scott, Director, Capacity Building, AHS leads the development and evaluation of processes to build knowledge management capacity within strategic priority areas and across the organization. Her team will support development of capacity for evidence-informed decision-making primarily through project-based learning. Cathie is also the lead for AHS’s commitment to follow through with next steps for SEARCH Canada’s intellectual assets, working in collaboration with other partner organizations. Cathie completed a postdoctoral fellowship focused on knowledge exchange, holds a PhD and MSc in Community Health Sciences, and a BSc in Physical Therapy. Cathie has strong ties with the academic communities.

Laura Lagendyk has over 10 years experience in academic and applied health research and evaluation and has worked on projects in primary, acute and tertiary care settings. With a Master’s degree in Health Research and a background in science and social work, she continues to further her skills through workshop presentation, group facilitation, and conflict management education. She has a keen interest in working with teams interested in building their capacity to produce and use evidence in innovative projects of direct relevance to their work environment.