Interprofessional collaborative practice within cancer teams: Translating evidence into action. A mixed methods study protocol

Most cancer and palliative/end-of-life programs propose interprofessional collaboration as a key modality for improving quality of care [1‐5]. The need for greater collaboration is being driven by the same pressures as those driving the cancer services transformation agenda: the pressure for timely access to care, lack of continuity in care, needs unmet by current services, demand for supportive care and dearth of health human resources. To grapple with these issues, the regional cancer network in Montérégie, a region in Quebec, Canada, has implemented a program designed to expand existing interprofessional collaboration among nurses, doctors, and other care providers (pharmacists, nutritionists and social workers) working on cancer teams.

The development of the 'Psychosocial oncology: Building interprofessional capacity to improve cancer care across the continuum' program (referred to here as the POBC³) was a nurse-led interdisciplinary project that was funded by the Canadian Partnership against Cancer. A summary of the program components is presented in Table 1. This program, which is related to recommendations made in the Registered Nurses' Association of Ontario (RNAO) Best Practice Guidelines [6, 7] and other sources of evidence [8‐10], represents a unique knowledge transfer initiative enabling local cancer team members to experiment with EIPCP. We will use this program to systematically examine the factors and processes involved in the adoption of evidence-based recommendations and their adaptation into practices.

We define EIPCP as a transformative model for cancer services delivery that engages care providers in the 'process of working together to build consensus on common goals, approaches and outcomes. It requires an understanding of own [sic] and others' roles, mutual respect among participants, commitment to common goals, shared decision making, effective communication, relationships and accountability for both the goals and team members' [6]. EIPCP entails proactive strategies that make care providers aware of evidence-based recommendations and facilitate the translation of this knowledge into day-to-day practice, as a basis for quality-of-care improvement.

The gap between research evidence on interprofessional collaboration and practice is wide, well documented [8, 11], and troubling, especially in cancer services where the cancer crisis jeopardizes the ability of health systems to respond to patients' needs [1]. Even though collaboration benefits users, providers, and organizations [12‐14], many professionals only pay lip service to the premise of collaborative practice [15‐17]. Previous studies have emphasized key enablers of and barriers to interprofessional collaboration: a lack of consensus about terminology, the need for interprofessional collaboration initiatives to have champions and external support, sensitivity to the effects of profession-related cultures, and the logistics of implementation [18]. Other barriers include structural issues such as competition between professionals, and conceptual problems such as a lack of understanding of mutual roles and a lack of experience or training in interdisciplinary collaboration among providers [14]. Less is known about how evidence-based recommendations could be adopted and adapted by care providers to overcome those barriers. Moreover, the preferences of those using such services with regard to the ways professionals work together and share the clinical information are poorly understood and are understudied.

Multiple, interacting conditions pose a challenge to the utilization of research findings. For practical reasons, these conditions could be grouped under six main elements as proposed in the comprehensive Ottawa Model of Research Use (OMRU) [19]. This model is an interdisciplinary framework presenting the utilization of research evidence as a dynamic process based on multiple, interrelated decisions and actions. It has provided guidance for numerous studies [20‐22]. The OMRU points to the importance of assessing barriers to the translation of knowledge into action at three levels: the characteristics of the recommendations made in the guidelines (e.g., perceive usefulness, fit with current practice, norms/values), the characteristics of the professionals involved (e.g., awareness, attitudes, knowledge/skills, concerns, current practice) and the characteristics of the practice environment (e.g., users' preferences, work pressure, competing demands/time) (Figure 1).

The literature on innovation and change in health services [23‐29] and RNAO documentation [6, 30] will also serve as conceptual background for our study. Our literature review led us to factor in the following four points: the novelty of EIPCP should be considered from the perspectives of the individuals who experiment with new ways of doing things; the players in the field have their own interpretations of evidence and their own definitions of the weaknesses in their practices--they therefore have intuitive ideas of what should be done to improve care delivery, but it is difficult for them to drive change within the context of their day-to-day activities; the mobilization of multiple actors with different areas of expertise and resources around a specific project focuses their efforts and provides a better chance of success; and the evaluation of an innovation contributes to its success--it allows the professionals involved in reflective activities to develop their receptive capacity through doing. Overall, the literature on innovation was used to define more clearly the research aim and specific objectives. It will also support data analysis, which will consist of matching empirically observed EIPCP translation events to the theoretically predicted elements identified earlier.

This study aims to examine the uptake of evidence-based recommendations from best practice guidelines intended to enhance interprofessional collaborative practices within cancer teams. In this study, care providers are those directly or indirectly (clinicians, managers, decision-makers responsible for governance) involved along the cancer trajectory, including community/home care, specialized hospital and ambulatory cancer services, as well as palliative/end-of-life care.

More specifically, our study objectives are as follows: to assess how professional knowledge, beliefs and the practice environment support or impede the adoption of EIPCP; to assess how patients' knowledge, beliefs and needs influence this adoption process; and to describe the impact of an educational workshop and mentoring program on the uptake and sustainability of EIPCP over a six- to eight-month period.

The planned study constitutes a practical trial, defined as a trial designed to provide comprehensive information that is grounded in real-world healthcare dynamics [31]. An exploratory mixed methods design will be used [32]. This design will involve concurrent quantitative and qualitative data collection and analysis. The quantitative and qualitative results will then be merged to provide complementary perspectives. The mixed methods design will in turn allow us to form a more complete picture of the interrelated contextual elements and the associated individual characteristics that determine EIPCP than use of a single method would allow (Figure 2) [33].

The participants will be selected from two groups. The care provider group (PG) will consist of professionals, managers, and decision-makers responsible for governance at the regional level who are directly or indirectly involved in the POBC³ improvement initiative. The user group (UG) will consist of patients/caregivers using services at one of four care settings in the Montérégie cancer network that are involved in the POBC³. Inclusion criteria for the PG will include being: a key stakeholder knowledgeable about POBC³ planning, implementation and/or evaluation, and a participant in the workshop and/or mentoring activities of the POBC³. Inclusion criteria for the UG will include being 18 years of age or older, able to understand and communicate in French, and a current user of cancer services at one of the participating care settings. The potential PG participants will be recruited during the POBC³ workshop. They will be asked if they agree to be contacted for a study on EIPCP and to provide their contact information. For the UG group, field professionals will be asked to identify potential participants for interviews using a purposive sampling strategy aimed at maximum variability (e.g., age, gender, tumour site, stage on the cancer trajectory) [34]. For users interested in participating, a nurse who has extensive experience with cancer patients will make the initial contact to obtain their informed consent and conduct the interview.

To meet objective one, quantitative data will be collected from professionals using an adapted cancer care version of the survey questionnaire from Davies' study on collaborative maternity care [35]. This questionnaire was constructed on the basis of the literature, which suggested that the following issues should be considered in developing a survey tool in this field: the concepts that should be demonstrated in a collaborative practice and the importance of the components of a collaborative practice model. The questionnaire also includes the Attitude toward Health Care Teams Scale [36], which contains two subscales: quality of care/process (14 items, Cronbach's alpha = 0.83) and physician centrality (six items, Cronbach's alpha = 0.68). Finally, a subscale (five items) taken from the Interprofessional Collaboration Questionnaire [37] will be used to assess the intensity of collaborative practices. These questionnaires provide operational measures that will be used to assess potential adopters' knowledge, attitudes, beliefs and current practice as described in the OMRU. To meet objective three, which focuses on the sustainability of practice over time, the questionnaire will be administered at the end of the POBC³ two-day workshop (T1) and six to eight months later after a period of mentoring (T2). This time frame was adopted first because the POBC³ is currently underway and pre/post-measurements are therefore not possible. Second, it was important to ensure a minimal sample size at T2 by taking potential workforce turnover into account.

Concurrently with T2, we will collect qualitative data to deepen our understanding of service users' perspectives (objective two), the contextual factors and the processes determining EIPCP patterns. In-depth interviews (60 minutes) will be conducted to gain understanding of the experience, the challenges, and the insights of both service users (n = 16) and care providers (n = 24). We will use a systematic interview guide adapted from Edwards' [38] and Peterson's [17] previous works on best practice guidelines implementation. The interviews will be audio-taped. With assistance from the care settings, we will purposefully select archival material that provides records of the EIPCP process. By way of example, documents will include reports, protocols and procedures, aggregated and non-identifiable patient reports, information sheets, minutes of meetings, and other relevant print material. No information from individual patient records will be collected.

Approximately 100 professionals and managers are involved in POBC³ activities. Based on our previous studies [39, 40], we anticipate a participation rate of 80% for our planned survey, leaving us with an a priori sample size of 80. The Power Analysis and Sample Size (PASS 2008) software was used to determine the sample size required for testing differences between groups. Assuming a type-one error rate of 5%, a minimal estimated sample size of 51 participants per group will give 80% power to detect a medium effect size using Cohen's guidelines [41]. The sample size for interviews was determined on the basis of Guest's and his colleagues' experiment, which demonstrated that data saturation occurs primarily after 12 interviews (UG) [42], and taking into account multiple investigation sites (PG).

Quantitative data will be managed using the SPSS. We will conduct an assessment of the validity of the questionnaire with our sample using principal component analysis and analysis of internal consistency [43, 44]. We will then generate descriptive statistics (frequency, mean, and SD) and perform comparative analyses between groups (using the t- test and Mann-Whitney test) to identify PG (T1-T2) differences (p < 0.05, IC 95%) [44]. Qualitative data from interview transcripts and documents will be managed using a formal database using QSR NVivo [45]. Thematic content analysis, using a theoretical orientation strategy, will guide the open-ended coding procedure in order to identify, classify, and reduce data and to build a descriptive matrix [46, 47]. The initial coding labels will be established by building on the elements of the OMRU. Table 2 presents a list of these elements and short definitions that describe the coding labels. In order to monitor collaborative practice change following the POBC³, we will focus on operational processes (how team members provide care), relational processes (how team members communicate), and adaptive processes (how team members use evidence to enhance collaborative practice) [48].

The trustworthiness of our research data and analysis will be ascertained using Miles and Huberman's criteria of credibility/validity, confirmability/objectivity, and transferability [46]. Given that our study will be an ex-post intervention study [49] and that the sampling method constitutes one of its limitations, the internal validity will be increased by use of a theoretical framework, validated investigation tools previously used by best practice guideline expert researchers, systematic data collection and analysis methods, as well as triangulation [50]. An audit trail of the entire research process will be kept [51].

The study has been approved by the Centre de Recherche de l'Hôpital Charles LeMoyne Ethics Board (ref. number AA-HCLM-09-034).