Descriptive study of interprofessional collaboration between physicians and osteopaths for the pediatric population in Quebec, Canada

This survey, conducted between September and November 2014, was the first (quantitative) phase of a larger sequential mixed method study aimed at improving understanding of IPC between physicians and non-physician osteopaths working with pediatric patients. Postal questionnaire were sent to all family physicians with a pediatric population and pediatricians without a subspecialty in Quebec, according to Scott’s MD Select 2013 directory as well as all members of Ostéopathie Québec (largest professional association in the province). A total of 2802 questionnaires were initially mailed to family physicians (N = 1327), pediatricians (N = 738) and osteopaths (N = 737). Assuming a 20% response rate for physicians/pediatricians and 40% for osteopaths, it is estimated that errors would be limited to 5% and 6% respectively, 19 times out of 20.

Efforts to maximize participation in the study included personalized mailing of questionnaires. The survey was also promoted on the Ostéopathie Québec website, and the first author attended various medical and osteopathic events in fall 2014. A reminder postcard was sent two weeks after the initial mailing and a second questionnaire two weeks later. Study aims were clearly described at the beginning of the questionnaire. Completing and returning the questionnaire anonymously indicate informed consent from participants. The study was approved by the Centre hospitalier universitaire de Sherbrooke ethics committee for health research on humans (#14–115).

The physician and osteopath versions of the survey questionnaire were developed using a three-stage iterative process comprising developmental stage, question testing stage, and pilot stage [18].

The initial version of the questionnaire was based on a literature review of existing questionnaires regarding IPC between conventional medicine and CAM practitioners. It was divided into three categories according to the study’s conceptual model, the Chiropractor-physician model of collaborative practice of Mior et al., (2010): practice parameters, communication, and care delivery parameters including regulation [14]. We considered existing questionnaires on physicians’ [19] or other health professionals’ [20] knowledge of osteopathy, continuing education and sources of information about osteopathy [20] or CAM for the pediatric population [21], experiences of collaboration with CAM practitioners [21], referrals [19, 20, 22], communication and professional relationships [20], and sociodemographic variables related to collaboration [19‐22]. This first version was pretested with two physicians, one pediatrician, three osteopaths, and an expert on IPC and questionnaire development. The modified version was piloted (procedure and duration) in individual face-to-face cognitive interviews with two physicians, two pediatricians and four osteopaths. It took 8 to 10 minutes to complete the survey. The final version of the questionnaires covered: a) knowledge about osteopathic practice parameters including 10 questions regarding general aspects, 10 questions concerning belief in the active role of osteopathy for specific pediatric conditions, and one question regarding sources of information about osteopathy (physician version only); b) communication aspects including interpersonal relationships, referrals, and communication methods (7 questions); c) influence of regulation of osteopathy on IPC (1 question); and d) sociodemographic data: gender, years of experience, type of practice, discipline, personal consultation of an osteopath (physician version), and presence of a physician in the working environment (osteopath version). Qualitative comments could be added by participants at the end of the questionnaire.

The number of correct answers out of 10 was computed for the questions on the general aspects of practice parameters. Similarly, the total number of positive answers for the 10 questions on belief in the role of osteopathy for specific pediatric conditions was calculated. Means and standard deviations were generated. Other quantitative data from the questionnaires regarding communication, influence of regulation, and sociodemographic information were analyzed using descriptive statistics (frequencies and percentages).

Physicians correctly answered 7.2 (SD 1.8) questions out of 10 on average concerning general knowledge about the practice of osteopathy (Table 2). Level of osteopathic education and regulation status of the profession were the questions with the lowest percentage of correct answers. However, almost all physicians agreed that being a physical therapist is not a prerequisite for being an osteopath in Quebec and that osteopaths evaluate and mobilize all body tissues, not just vertebrae. There were 6.8 (SD 1.91) positive answers for belief in the role of osteopathy for specific pediatric conditions (Table 2). A high percentage of physicians think that osteopathy has a role in addressing musculoskeletal pain as well as torticollis and plagiocephaly. When asked what their sources of information about osteopathy are (multiple answers were allowed), more than half of the physicians mentioned asking information directly to an osteopath (52%), followed by patients (51%), and to other health professionals (21%). Information was also obtained by personal searches (17%), from scientific articles (10%), and during continuing education sessions (8%).

Table 3 outlines communication aspects for each type of practitioner. More than one third of respondents had a professional relationship with the other practitioner. Nearly half the physicians referred pediatric patients to osteopaths at least once a month. The large majority of respondents (81% of physicians and 85% of osteopaths) said that communication was required for common patients and that their preferred communication methods were letters or verbal communication through the patient.

The majority of physicians (72%) and osteopaths (62%) said they would be moderately or greatly influenced by the upcoming government regulation of osteopathy and the creation of university-based osteopathic programs in Quebec. Eighteen percent of physicians and 22% of osteopaths would not be very influenced and respectively 10% and 16% said they would not be influenced at all by regulation, either because they already collaborated or because they had no interest in physician-osteopath collaboration (comments on the survey).

Variables statistically associated with osteopathic referrals by physicians at the bivariate level were: gender (female), profession (family physician), general knowledge about practice parameters, direct sources of information about osteopathy, belief in the active role of osteopathy for pediatric conditions, community practice, personal consultation of an osteopath, and having a professional relationship. These last four variables remained in the final stepwise backward logistic regression model (Table 4).

In the final linear regression model F (7247) = 9.117 (p < 0.001), 20.5% of the variance in the number of correct answers (knowledge) concerning general osteopathic practice parameters was explained by having a direct source of information about osteopathy (osteopath or other health professional; p = 0.009) and the number of positive answers for belief in the appropriateness of osteopathy for pediatric conditions (p < 0.001) while considering all the other bivariate associated variables in the model (professional relationship, personal consultation with an osteopath, osteopathic referrals, importance of having an osteopath among collaborators (quite or very important versus not important), and gender).

The main aim of this study was to describe the current situation in Quebec respecting interprofessional collaboration between physicians and non-physician osteopaths working with pediatric patients. We found that some collaboration already exists, including referrals and professional relationships between these practitioners, but other factors might improve this collaboration. Having a professional relationship or having personally used osteopathic services were the factors most strongly associated with osteopathic referrals. It is known that consulting a CAM practitioner or having used CAM personally influences CAM referrals for the pediatric population [22, 23]. Indeed, as noted by Mior and collaborators (2010) in the early stages of implementing collaborative care between conventional and CAM practitioners, informal social meetings or events could foster the personal relationships that help to develop mutual respect and trust and break down barriers to communication.

The majority of the survey respondents acknowledged the importance of communication, including exchanging information about common patients. The preferred communication methods were verbal via the patients or in writing for practitioners. In a previous study, physicians who recommended that parents of young patients consult with a CAM practitioner reported that sustained two-way communication with CAM practitioners was rare [24]. In the present study, one third of respondents provided parents with a written referral, regardless of profession. In their cross-sectional study on communication regarding pediatric patients, Ben-Arye and collaborators (2007) found that, in addition to communicating clinical information and increasing the willingness to respond to the initial referral, the exchange of referral letters positively impacted the development of collaboration between physicians and CAM practitioners [25]. Letters about common patients should include conventional/CAM diagnoses using jargon-free terminology, possible conventional/CAM treatment interactions, and treatment plan and goals [26]. Since disclosure and overall discussion about CAM use during physician/patient encounters is reportedly limited [5, 27], especially for the pediatric population, verbal exchanges via the patient may not guarantee efficient communication between practitioners compared to exchanging letters.

Communication is also an important catalyst for other determinants of collaboration, such as sharing and developing mutual trust and respect [14]. Mutual respect implies knowledge and recognition of the complementarity of other professionals’ contributions [12]. Pediatricians who have a high level of knowledge about CAM are more likely to recommend CAM and discuss CAM with parents [23]. In this study, half the physicians asked an osteopath directly for information while about 20% asked other professionals such as colleagues about the role of osteopathy for pediatric patients. Having such direct sources of information and believing osteopathy to be appropriate for certain pediatric conditions are associated with physicians’ greater knowledge of general osteopathic practice parameters. More knowledge about practice parameters may lead to safer, appropriate two-way referrals. However, very few physicians said they had access to continuing education sessions about osteopathy, suggesting that formal education and information transmission concerning osteopathy are rare and could be improved.

Less than half the physicians knew that non-physician osteopaths are not yet regulated in Quebec, suggesting some confusion about the current status of osteopathic practice. However, the majority of respondents, both physicians and osteopaths, expected to be positively influenced by professional regulation and osteopathic university-based education. IPC between conventional and CAM practitioners is known to be affected by systemic determinants such as regulation [16]. For example, practice conditions governed by Colleges of Physicians and Surgeons were found to be a significant barrier to IPC [15, 28] and, even when referral to unregulated CAM is allowed, physicians must consider patients’ needs in light of medical, legal and ethical issues [29]. A positive view of the upcoming regulation and the creation of osteopathic university-based programs in Quebec might improve interactions between practitioners and increase interprofessional education for future practitioners. More interactions during the training phase might in turn diminish current power disparities observed in practice [30].

This study has some limitations. First, the low response rate, especially for physicians, may limit the generalizability of the results. Although the response rate is typical for this type of survey [3], lack of time, low perceived relevance of the study topic or receiving more surveys a week might explain the physicians’ low response rate [31] compared to the osteopaths. Nearly half the physicians had personally used osteopathic services, which may suggest that those respondents already had a positive attitude toward osteopathy. Since the respondents appeared to be open to and in favour of collaboration, the results likely underestimate the challenges facing collaboration between physicians and osteopaths. To deepen our understanding of IPC between these practitioners, some of the results of this survey are being further explored with qualitative data from purposeful sampling of physicians and osteopaths (phase 2 of the study).