Context Transitioning primary care physicians from fee-for-service to capitation or blended payments and incorporating team-based care are widely regarded as desirable transformations for patient-centred medical homes, but there are few studies evaluating the effects of these changes.

Objective To assess the effects of capitation payment and team-based care on chronic disease management and prevention by evaluating a large-scale transition of primary care physicians to medical homes in Ontario.

Design We conducted a longitudinal population-wide study using routinely collected administrative data. We used Poisson regression models to examine the association between type of medical home and diabetes and cancer screening in 2011. We calculated outcomes for each fiscal year between 2001 and 2011 and used a fitted non-linear model to compare changes in outcomes between type of medical home over time.

Participants All Ontario patients enrolled to a medical home in 2011 (N = 10 675 480).

Intervention Patient enrolment in team-based capitation, non-team capitation, or enhanced fee-for-service medical homes as of March 31, 2011.

Main outcome measures Proportion of eligible patients receiving testing for diabetes and screening for cervical, breast, and colorectal cancer.

Results Patients in team-based capitation were more likely to receive recommended testing for diabetes in 2011 than were patients in non-team capitation or enhanced fee-for-service (40%, 36%, and 32%, respectively) and this was true even after adjustment for patient and physician characteristics (relative risk [95% CI]) compared with enhanced fee-for-service: team-based capitation 1.22 [1.18 to 1.25]; non-team capitation 1.10 [1.07 to 1.14]). Patients in team-based capitation experienced the greatest improvement in recommended testing for diabetes over time, followed by patients in non-team capitation (absolute difference in improvement [95% CI]) compared with enhanced fee-for-service: team-based capitation 10.6% [7.9% to 13.2%]; non-team capitation 4.1% [1.5% to 6.8%]). Patients in team-based capitation experienced the greatest improvement in cervical cancer screening but there were no significant differences in change over time between medical homes for breast and colorectal cancer.

Conclusion The shift to capitation payment and the addition of team-based care in Ontario were associated with moderate improvements in diabetes care but the effect on cancer screening was less clear.

Context Transitional shelters provide safe and temporary accommodation for women in crisis. In 2011 and 2012, shelters across Canada reported more than 60 000 admissions of women. Many of these women have suffered interpersonal violence, complex trauma, and considerable financial burden. They are at disproportionate risk of health deterioration. To date, little is known about the health care experiences of Canadian women living in shelters.

Objective To explore the health care experiences of women living in Vancouver’s transitional shelters; specifically, to determine how health care is accessed, to identify specific barriers to access, and to examine how the health care system could better serve these women.

Design A qualitative study using focus groups.

Participants Sixteen women residing in 2 transitional shelters located in Vancouver, BC.

Instrument Three focus groups were conducted to explore the health care experiences of the study participants. Interviews were audiorecorded and transcribed verbatim. NVivo software was used for data analysis to help identify units of meaning for key themes.

Findings Six main themes were identified: Women in transition want family doctors; transportation is a barrier to care; relocation inhibits access; out-of-pocket costs are unaffordable; access to specialists is limited without a GP; and continuity of care is highly valued.

Conclusion Conventional health care services are failing women living in transitional shelters, leaving them doctorless and unengaged. While they most often use walk-in clinics, this fragmented care does not meet their complex needs. Given their histories of trauma and crisis, continuity of mental and physical health care is essential. Family physicians should create specific plans to connect with and care for these vulnerable women.

Context The Canadian Primary Care Sentinel Surveillance Network (CPCSSN) is a multidisease surveillance system that collects quarterly electronic medical record (EMR) health data from more than 800 primary care practices across Canada. The data extracted are standardized and used to inform quality monitoring at the practice level, and to conduct epidemiologic research using the large data set.

Objective Assess the feasibility of using CPCSSN to obtain chart data required for use in an intervention, the Cardiovascular Health Awareness Program (CHAP).

Design A pilot evaluation of preintervention and postintervention data extractions.

Participants Six urban primary care practice sites (3 community health centres, 3 interprofessional capitation-based practices) and 48 primary care providers in eastern Ontario participating in CHAP.

Intervention CPCSSN was used to replace direct chart extraction for 4 purposes: 1) identify eligible patients (CPCSSN produced a list of all individuals aged older than 65 years and their coordinates, which was used to automate CHAP recruitment letters); 2) evaluate participation bias (comparison of participants’ and nonparticipants’ baseline data [eg, age, sex, diagnoses, risk factors, attachment to practice]); 3) calculate risk scores (lipid profile and diagnoses of hypertension and diabetes were added to CHAP session data to calculate the Framingham and CANRISK scores of participants); 4) assess the intervention’s effects (12 months post CHAP, modifiable risk factors [blood pressure, lipids profile, body mass index, and waist circumference] will be extracted to assess changes from baseline).

Results CPCSSN identified 3084 seniors; primary care providers excluded 934 (deceased, too frail, or had dementia). Participation bias was assessed successfully. Lipid data to calculate Framingham scores was available for 94% and 62% of participants in practices with established and new (1 to 2 years) EMR systems, respectively. Effects of the intervention will be assessed in 1 year.

Discussion and conclusion The cost of chart extraction often means studies relying on clinical data are not feasibility. The growing penetration of EMRs opens the opportunity to use automated processes for data extraction. However, the lack of standards across EMR systems means that one approach cannot be used across practices. The existing processes for data extraction and standardization developed by CPCSSN for 12 EMR vendors can be used to provide accurate data efficiently for studies requiring clinical data extraction.

Context Depression is associated with substantial disability but less is known about the contribution of anxiety symptoms to disability.

Objective To examine the disability profiles pertaining to depressive and anxious symptomatology (DAS).

Design Data were drawn from the Dialogue Project, an observational study conducted in 64 primary care clinics in Quebec.

Participants The sample of 14 833 adults was recruited in primary care clinics’ waiting rooms when consulting a general practitioner for oneself.

Instruments The Hospital Anxiety and Depression Scale (HADS) was used to assess DAS. The World Health Organization Disability Schedule (WHODAS, short 12-item version) was used to assess disability in 6 domains: mobility, life activities, cognition, social participation, self-care, and getting along.

Main outcome measures Respondents were divided into 4 groups according to their HADS scores: anxious subjects (n = 4667), depressive subjects (n = 2304), anxious-depressive subjects (n = 1785), and control subjects (n = 9543).

Results Preliminary results indicate that depressive subjects, compared with anxious subjects, presented higher levels of disability for all 6 domains, and anxious-depressive subjects presented the worst loss of functionality. For the social participation domain, multiple logistic regression models showed that anxious subjects (odds ration [OR] = 4.10, 95% CI 3.72 to 4.52), depressive subjects (OR = 7.50, 95% CI 5.95 to 9.46), and anxious-depressive subjects (OR = 18.62, 95% CI 15.76 to 22.01) were more likely to present disability than control subjects were.

Conclusion Depressive symptoms were associated with more disability than anxiety, but comorbid depressive-anxious symptomatology significantly increased disability levels in all domains. Anxious symptoms are also associated with a higher level of disability than the absence of DAS. Recognizing that any type of DAS likely causes disability will further the efforts to improve diagnosis and treatment of common mental disorders in primary care.

Context The British Columbia Ministry of Health’s Clinical Care Management initiative was used as a case study to better understand large-scale change within BC’s health system. Eleven guidelines have been introduced into hospitals across all BC health authorities in an attempt to improve evidence-based clinical practice. Fifteen additional Clinical Care Management guidelines are to be implemented over the next 3 years, with many priorities touching on family medicine.

Objective Using a complex system framework, this study examined mechanisms that enable and constrain the implementation of clinical guidelines across various clinical settings.

Design Qualitative descriptive.

Participants Purposively selected respondents in each health authority; 3 purposively selected guidelines.

Intervention Researchers applied a general model of complex adaptive systems plus 2 specific conceptual frameworks (realist evaluation and system dynamics mapping) to define and study enablers and constraints. Findings from focus group sessions and interviews with clinicians, executives, managers, and board members were validated through an online survey.

Findings The functional themes for managing large-scale clinical change included creating a context to prepare clinicians for health system transformation initiatives; promoting shared clinical leadership; strengthening knowledge management, strategic communications, and opportunities for networking; and clearing pathways through the complexity of a multilevel, dynamic system.

Discussion Recommended action-oriented strategies included engaging local champions; supporting local adaptation for implementation of clinical guidelines; strengthening local teams to guide implementation; reducing change fatigue; ensuring adequate resources; providing consistent communication, especially for front-line care providers; and supporting local teams to demonstrate the clinical value of the guidelines to their colleagues.

Conclusion Bringing a complex systems perspective to clinical guideline implementation resulted in a clear understanding of the challenges involved in large-scale change.

Context Among patients taking warfarin, lower socioeconomic status is associated with poorer control of anticoagulation. However, the extent to which socioeconomic status influences the risk of hemorrhage is unknown.

Objective To examine the extent to which socioeconomic status influences the risk of hemorrhage in older individuals newly commencing warfarin therapy for atrial fibrillation.

Design Population-based cohort study using health care administrative data.

Participants Patients aged 66 years or older with atrial fibrillation who commenced warfarin therapy between April 1, 1997, and November 30, 2011, in Ontario. Validated ICD-9 and ICD-10 codes were used to identify a diagnosis of atrial fibrillation. We used neighbourhood-level income quintiles as a measure of socioeconomic status.

Main outcome measures The primary outcome was an emergency department visit or hospitalization for hemorrhage and the secondary outcome was fatal hemorrhage. Validated ICD-9 and ICD-10 codes were used to identify a diagnosis of hemorrhage.

Results We studied 166 742 older patients with atrial fibrillation who newly commenced warfarin therapy. Of these, 16 371 (9.8%) were hospitalized for hemorrhage during a median follow-up of 369 (interquartile range 102 to 865) days. After extensive multivariable adjustment using Cox proportional hazards regression, we found that those in the lowest income quintile faced an increased risk of hospitalization for hemorrhage relative to those in the highest income quintile (adjusted hazard ratio 1.18, 95% CI 1.12 to 1.23). Similarly, the risk of fatal hemorrhage (n = 1802) was increased in the lowest relative to the highest income quintile (adjusted hazard ratio 1.28, 95% CI 1.11 to 1.48).

Conclusion Among older individuals receiving warfarin therapy for atrial fibrillation, lower socioeconomic status is a risk factor for hemorrhage and hemorrhage-related mortality. This factor should be carefully considered by clinicians when initiating and monitoring warfarin therapy.

Context For opioid-dependent patients not benefiting from conventional treatments (ie, oral methadone), evidence suggests that supervised injectable medications are effective.

Objective This baseline analysis aims to describe participants’ characteristics at entry into a study comparing injectable diacetylmorphine and hydromorphone and factors independently associated with previous access to methadone at high doses.

Design SALOME is a phase III, randomized, double-blind controlled trial comparing injectable diacetylmorphine and hydromorphone.

Setting Vancouver, BC.

Participants A total of 202 chronic, opioid-dependent current injection opioid users who had at least 1 previous episode of opioid maintenance treatment.

Main outcome measures An analysis has been done to determine characteristics associated with reaching a stable dose of oral methadone in the 5 years prior to trial recruitment.

Results These data emphasize that study participants were in need of alternative treatments at the time of enrolment and fit the profile of patients to whom supervised injectable treatment should be offered. Implications of previous methadone treatment experience in these individuals who might be eligible for injectable opioid assisted treatment will be reviewed.

Conclusion For opioid-dependent patients not benefiting from conventional treatments (ie, oral methadone), evidence suggests that supervised injectable medications are effective.

Context Little is known about the health status of Canadian inmates in provincial detention centres or about their access to health care in the community. American and European data show that the health of the incarcerated is poor, with a disproportionate burden of mental illness, infectious diseases, chronic diseases, and premature mortality. Recently released individuals in the United States and Europe have been shown to have poor access to primary care.

Objective To describe the primary care experience of adult inmates at the Hamilton-Wentworth Detention Centre in Ontario.

Design Written surveys were distributed to blocks of men and women at the Hamilton-Wentworth Detention Centre. Inmates were allowed to complete surveys in their cells at any time. Completed surveys were placed in sealed envelopes and returned to the nurse manager.

Participants Adult men and women with working knowledge of English were included in the study. A total of 419 surveys were distributed and 125 were returned (30%). Of the respondents, 17% indicated they were female; 80% indicated they were male. The mean age was 36 years.

Instrument A 2-page survey with multiple choice questions and space for extended written responses.

Results and findings Among respondents, 38% (P < .001) report that they do not have a family doctor or primary care provider, compared with 15% of the Canadian population. Those with no family doctor noted several barriers, including difficulty finding a physician who is accepting new patients; difficulty navigating the health care system; and frequent incarcerations. Among respondents, 43% reported having had unmet health care needs in the year before coming to jail, compared with 8.8% of the Canadian population. Barriers included previous experience of stigma in the health care system and inaccessible clinic locations and schedules. Participants reported an average of 2.1 emergency department visits in the year before incarceration, compared with 0.29 visits per year for the Canadian population.

Conclusion Those incarcerated at the Hamilton-Wentworth Detention Centre report having poorer access to primary care and higher unmet health care needs than average Canadians. Incarceration could provide an opportunity to connect inmates with primary care and thus improve health care and health.

Context Competition for seats in Canadian medical schools has led many Canadians to undertake medical education abroad. Systematic barriers make it necessary for internationally educated physicians (IEPs) hoping to practise in Canada to complete postgraduate residencies. To do so, they must transition into new medical education systems. The transitional experiences of IEPs are not well understood

Objective To better understand the transitional experiences of IEPs.

Design This exploratory phenomenological qualitative study sought the perspectives of 20 recent graduates from the University of Manitoba family medicine residency program.

Participants Most participants were Canadians studying abroad. Participants were drawn from 5 recent cohorts of residents entering training between 2008 and 2012. Recruitment was intentional, with all identified IEPs from these cohorts invited to participate. A modest snowball effect was noted.

Results Twenty-one percent of potential participants took part. Participant interviews were inductively coded, employing initial and emotion coding. More than 70 codes were identified, grouped into 10 categories, which led to the identification of 2 main themes: the clinical practice gap and the stigmatization of IEPs. The development of the clinical practice gap is attributed to curricular differences in the timing of graduated clinical responsibility between Canadian and international medical educational systems. Participants perceived that they were stigmatized. They described being labeled, singled out, rejected, and mistreated. They recognized that being Canadian afforded them some invisibility as IEPs, although they could be revealed at any time. They further perceived that those for whom English was a second language or who were from visible minorities could not be invisible. Participants described the presence of dual resident hierarchies, by program of training (Royal College vs College of Family Physicians of Canada) and by country of training (Canada vs international countries). As IEP family medicine residents, they were assigned low status in these hierarchies. Their experiences are conceptualized as stigmatization.

Conclusion Many questions remain. What led to the development of this stigmatization? Does the clinical practice gap contribute to perpetuation of this stigmatization? Could pre-residency training mitigate the clinical practice gap? What are the learning needs specific to Canadians studying abroad? These findings are relevant to the work of program directors and curriculum directors in family medicine.

Context Continuity of care (CoC) is a critical aspect of practice in generalist disciplines and across several dimensions of clinical care. To date, there have been several interventions at the level of residency training to inculcate CoC. Their mechanisms and effects on the formation of a caring physician identity is unknown. In this critical narrative review, we analyzed the existing literature to understand the pedagogic rationales and effectiveness of these interventions.

Methods We searched the existing medical education literature via MEDLINE, EMBASE, and CINAHL for English-language articles since 1975 that described or evaluated CoC interventions at the level of postgraduate training. We excluded commentaries and nonpostgraduate papers. We analyzed the extracted literature for themes across interventions, for the proposed pedagogic mechanisms or rationales, and for evidence of effectiveness.

Results Twenty-two papers dealt with CoC at the postgraduate level but only 15 described interventions, with the remaining describing satisfaction with training. Studies describing interventions have focused on organizational changes to the structure of residency programs. Examples of such interventions included a longitudinal family medicine component, a resident patient panel, and increasing dedicated time for practice in family medicine settings. There is a lack of rigorous evaluation of most interventions. Our analysis suggests that the pedagogic mechanisms are not clearly articulated in the literature. Furthermore, current structural interventions do not conceptualize continuity of care as a part of the identity of the physician.

Conclusion Pedagogic evidence suggests that to be effective, training must help integrate the practice of CoC as an essential part of physician identity. Current interventions in the literature do not describe a programmatic approach to integrating CoC into emerging physician identity and have not been rigorously evaluated for outcomes. Adopting an identity formation framework can help training become more relevant and transferrable to future practice, and suggest appropriate outcomes for evaluation.

Context Relational continuity is defined as the “ongoing therapeutic relationship between a patient and one or more providers.” It has been found to be a predictor of positive outcomes for both patients and physicians. Given this, teaching and modeling relational continuity in family medicine residency training is highly important. The challenge lies in how to measure relational continuity for residents in training programs.

Objective To determine to what extent the electronic medical record (EMR) is a useful tool in understanding relational continuity experiences of family medicine residents. To determine relational continuity experiences of family medicine residents at the end of a 2-year residency and the factors influencing the opportunities for developing relational continuity between residents and patients.

Design Mixed methods.

Participants Purposive sample of residents (n = 5) and preceptors (n = 5) at an academic teaching clinic.

Instruments Electronic medical record data was used to measure relational continuity. A constructivist grounded theory approach was used to conduct semistructured interviews with all participants, and constant comparative analysis was used to determine factors that influence relational continuity for residents.

Main outcome measures Quantitative: Relational continuity, as determined by the Usual Provider Continuity (UPC) index, a commonly used measure of continuity of care (ratio of patient visits to usual care provider). Qualitative: Themes emerging from interviews.

Results and findings Residents were found to have relatively low relational continuity. For most residents, 49.3% to 67.5% of patients were in the UPC index of 0 to 0.24, and a low proportion of patients were seen at a UPC index of 0.50 or greater. Qualitative findings resulted in a model of relational continuity, where continuity of care in family medicine, relational continuity with patients, therapeutic alliance, and ownership all interacted and were affected positively or adversely by program and systems factors, patient factors, and attitudes and experiences of both the resident and the preceptor.

Conclusion Electronic medical record data provide a good picture of continuity rates. However, the factors that affect relational continuity can only be found by interviewing residents and preceptors. While most residents valued relational continuity, factors such as patient schedules and program or system factors made it difficult to establish high rates of continuity with patients.

Context Research skills empower family physicians to produce and appraise new knowledge and to evaluate improvements in their clinical work. The College of Family Physicians of Canada’s Section of Researchers has developed a blueprint for family medicine research success with the vision that research is a core component of training, scholarship, and clinical practice. This context is the impetus for developing a program for building research culture and capacity in the Department of Family Medicine at Western University. This study represented the first step in the design of our research capacity development initiative.

Objective To describe facilitators, barriers, and strategies for clinical faculty to conduct research.

Methods In this descriptive qualitative study, key informant interviews were conducted with 10 members of the department with different roles: clinician-teachers, researchers, administrative leaders, and research staff. Interviews were audiorecorded and transcribed verbatim. Interviews were analyzed individually by 2 team members, who identified themes, and then reviewed by the entire team, who further elaborated and refined the themes.

Findings Findings were organized around individual, organizational, and external factors that acted as facilitators and barriers to clinical faculty research, and around the success and limitations of research capacity development initiatives already in place. Individual themes centred on motivation and the knowledge, skills, and confidence necessary for clinical teachers to conduct research. Organizational themes included the need for role clarity, collaborations, and organizational support. External themes focused on organizations both within and outside the university setting, including the College of Family Physicians of Canada.

Conclusion These results are being used to design a strategy for strengthening research capacity among clinical faculty in the Department of Family Medicine at Western University. The ultimate goal is to enhance research culture, as measured by increased research interest, skills and confidence, and institutional support; and to increase research capacity, measured by publications and presentations. These findings will be of interest to other departments of family medicine in Canada interested in building research culture and capacity.

Context Continuity of care in primary health care (PHC) has been linked to reductions in emergency department use and avoidable hospitalizations.

Objective To determine if patient continuity with family physicians in 2 provinces affected emergency department use for family practice sensitive conditions and hospitalizations for ambulatory care sensitive conditions.

Design This study uses patient-level physician billing data, hospital discharge data, and emergency department data to look at the relationship between continuity of care with a family medicine physician (ie, relational continuity) and the use of avoidable hospital services in Saskatchewan and Alberta. Six years of patient-level physician billing data in Alberta and Saskatchewan were linked to the Discharge Abstract Database and the National Ambulatory Care Reporting System to investigate the effects of continuity of PHC on health service utilization. To measure continuity of care we used the Usual Provider Continuity index to assess the level of patient attachment to their most common PHC provider. Regression analyses were conducted to identify any association between continuity of care and emergency department use for family practice sensitive conditions and hospitalizations for ambulatory care sensitive conditions. Patients and provider characteristics were examined for possible associations with continuity.

Results In line with the literature, increased relational continuity of care was found to decrease the use of avoidable hospital services.

Conclusion The report reinforces the importance of encouraging the general public, family physicians, other PHC providers, and policy makers to support improved relational continuity of care. Looking for opportunities to increase continuity of care within primary health care could improve patient-provider relationships, improve the identification and management of chronic illnesses, and decrease unnecessary hospital use.

Context Canadians visit the emergency department (ED) frequently, at times for minor medical problems that might be more appropriately treated elsewhere. Many jurisdictions are evaluating strategies to reduce avoidable ED visits, and identifying and quantifying these visits can help efforts to improve appropriateness of care.

Objective This study from the Canadian Institute for Health Information examines ED utilization to identify visits that might be more appropriately managed at a family physician’s office or clinic, known as family practice sensitive conditions (FPSCs).

Design Emergency department visits for primarily low acuity reasons where patients were not admitted to inpatient beds were calculated and compared to rates of visits for other reasons. Possible explanatory factors including diagnosis and acuity, time of visit, length of stay, and patient age, socioeconomic status, and rurality were examined.

Target population Included in the analysis are all unscheduled visits to EDs (excluding urgent care centres) in 2013 to 2014 where patients were discharged home.

Instrument This study uses data from the Canadian Institute for Health Information’s National Ambulatory Care Reporting System. The following jurisdictions submitted ED data with complete ICD-10-CA diagnoses in 2013 and 2014 and are included in the results: Ontario (all facilities); Alberta (all facilities); Nova Scotia (5 facilities); Saskatchewan (4 facilities); Prince Edward Island (1 facility); and Yukon (1 facility).

Results Overall, 1 in 5 nonadmitted ED visits were for FPSCs, totaling more than 1.4 million visits in 2013 and 2014. The most common reasons for these visits included upper respiratory infections (13%), antibiotic therapies (13%), sore throats (8%), ear infections (7%), and care following surgery such as dressing changes and removal of stitches (5%). More than one-third (35%) of nonadmitted visits for children were for FPSCs, compared with only 12% for patients age 85 and older. Additionally, 32% of nonadmitted ED visits among rural-dwelling patients were for FPSCs, versus 17% among urban-dwelling patients.

Conclusion Providing care for patients presenting with FPSCs in settings such as doctors’ offices and clinics might improve continuity of care and the patient experience, and allow ED resources to be focused on those who more appropriately require them.

Context Polypharmacy is a recognized source of harm for frail elders, yet many experimental approaches to try to reduce the numbers of medications in nursing homes have failed. Some studies suggest that a focus on disease management, rather than on frailty-appropriate treatments, might be contributing to polypharmacy. A provincial pharmacy data analysis published in September 2014 suggested that prescribing continues to rise for elderly patients in nursing homes.

Objective To describe prevalence of polypharmacy for a representative group of frail elders in nursing homes in British Columbia and their associated patient characteristics; and to determine if there are potential associations between hypertension and diabetes surrogate markers and numbers of medications prescribed.

Design Cross-sectional survey.

Participants A total of 213 randomly selected nursing home patients from 6 nursing homes in BC’s lower mainland with a total population of 950 patients and with a highly regarded physician and pharmacist care model in British Columbia’s lower mainland.

Intervention Patient data collated from the pharmacy database, patient chart, and local administrative database.

Main outcome measures Mean number of medications prescribed, systolic blood pressure, HbA_1c level, hospital visits for each patient, and demographic characteristics.

Results The mean (SD) number of medications prescribed was 7.5 (3.4). Among the patients in the study, 70.4% had a diagnosis of hypertension and 100% had a measured systolic blood pressure, mean (SD) 126 (18) mm Hg. Only 27% had a diagnosis of diabetes, mean (SD) HbA_1c level of 6.5 (1.19). The mean (SD) number of hospital visits in the preceding year for each patient was 0.7 (1.5). A diagnosis of hypertension was associated with more medications (P = .04), and increasing age was associated with fewer (P = .002). No association was found between the number of medications and a diagnosis of diabetes or dementia, nor with length of stay, prescribing physician, or number of hospital visits.

Conclusion The number of medications prescribed to frail elders remains high. Chronic disease treatment might be contributing to this large number. In the case of hypertension, the mean systolic blood pressure was surprisingly low and relaxation of surrogate targets might result in fewer medications. Future efforts to address polypharmacy might have more success if they directly address appropriate drug therapy for frail elders with chronic diseases.

Context Our patients tell us they are using marijuana for period symptoms but there is no reported evidence for gynecologic indications. There is evidence for cannabinoid receptors in the uterus. Before starting clinical trials, we need to know how marijuana is being used.

Objective To understand how women have been using marijuana for relieving the pain and other symptoms of menstrual periods.

Methods Questionnaires were provided online and on paper to marijuana vendors in retail shops in Vancouver, and to women recruited through the shops, by word of mouth, from advocacy groups, and using a classified advertisement website (Craigslist). The vendors were asked if they had recommended marijuana for period symptoms and how it was used. The women’s questionnaire asked about demographic characteristics plus whether the respondent had (current or past) pain with her periods, whether she had ever used marijuana for her period pain, and, if so, how she used it (smoked, eaten, etc), and if she had experienced any benefit.

Results There were 192 out of 201 usable questionnaires received from women and 9 from vendors. The mean age of respondents was 30.9 years with a range of 18 to 62 years. The worst period pain reported was a mean of 8.3 out of 10 and usual period pain was 5.9 out of 10. Women also complained of headaches (66.7%), nausea (54.2%), diarrhea (50.5%), feeling angry (27.1%), dizziness (66.1%), feeling sad (66.1%), fever or chills (25.5%), weakness (56.3%), and vomiting (20.3%). For period symptoms, 170 women (88.5%) said they had used marijuana and 152 (89.4%) said it helped; 153 women (79.7%) said they knew other women who used marijuana for periods. Eight of the nine vendors had been recommending marijuana and all said it helped their clients. Women had been using it in many forms; however, mostly they smoked or ate it. Specific recommendations from the women included using high-CBD (cannabidiol), low-THC (delta-9-tetrahydrocannabinol) forms and titrating doses so that it allowed women to function.

Conclusion Many women are using marijuana successfully to relieve the symptoms of menstrual periods. Clinical trials are needed to assess if low-THC, high-CBD products are helpful, because they might have fewer side effects and risks.

Context The Chinook Primary Care Network is a leader in adopting elements of the Patient’s Medical Home model of primary care in Alberta. Chinook practices still face various challenges in Patient’s Medical Home transformations.

Objective To determine whether and how a formal cognitive and organizational science method can help practices with change management efforts.

Design Prospective multiple case study.

Participants Physicians and staff of 3 Chinook Primary Care Network practices.

Intervention Cognitive task analysis (CTA) is a family of tools developed to understand and improve the work of teams of experts, and support knowledge translation. It is based on understanding and improving “macrocognition” skills (coordination, problem detection and monitoring, sense making, planning, and managing uncertainty and irregularity) and “mental models” (beliefs about the nature of the team’s work, why activities are done, what actions will produce what results, etc). Cognitive task analysis has been widely successful in aviation, the nuclear industry, intensive care units, and the military. Practice facilitators trained in CTA visited each practice for 2 days to interview physicians and staff and make observations. The team met regularly for several weeks after each site visit to code and abstract the interview and observation data, develop profiles of each macrocognitive skill, and understand each practice’s mental model.

Main outcome measures A CTA report was generated for each practice, addressing their macrocognition skills and mental model of practice. Rich descriptions of the CTA findings and how the practices make use of them were the outcomes.

Results No significant macrocognitive skills deficits were identified. Practices had distinctive patterns of how they used the skills, which were emphasized, and how much demand their work routines placed on each skill. The 3 practices had 3 different mental models of practice, emphasizing continuous quality improvement, highly integrated teamwork, or physician autonomy.

Discussion The CTA reports made practices explicitly aware of trade-offs they were making implicitly, and allowed them to consider alternatives. The different mental models had differing implications for what changes practices would pursue, and how they would affect the practices.

Conclusion Primary care practice, like other forms of expertise-intensive knowledge work, can benefit from formal cognitive analytic methods.

Context Owing to earlier detection and better treatments, the number of cancer survivors has risen substantially in the past 3 decades. In addition to higher risk of recurrence or second malignancy, more than 50% of survivors are afflicted with late effects. Cancer survivors visit a wide variety of specialists during treatment, yet 75% of them also see their FPs, and this proportion does not decrease after treatment. Upon treatment completion, FPs are expected to play an integral role in cancer survivors’ care. However, literature reveals low confidence levels among FPs with regard to survivorship care. Furthermore, considerable deficits currently exist in the transition of care to FPs from other specialists.

Objective In 2014, we conducted a survey targeting FPs that aimed to identify their needs and perceived barriers in providing optimal care to cancer survivors.

Design A brief survey containing Likert-type and short-answer questions was designed and distributed to FPs practising within academic, public, and private settings in Montreal, Que. Based on a targeted FP population estimated at 500, sample size was set at 75. Survey design aimed to identify specific outcome measures including current needs and perceived barriers to cancer survivor care. Data analysis comprised descriptive statistics, an open-coding approach to identifying main themes, and Student t tests. Ethics approval was obtained.

Results Of 165 surveys distributed, 93 were completed, for a 56.3% response rate. Among respondents, 94% were already caring for cancer survivors, and of these 73% expressed a willingness to take on new cancer survivors. A statistically significant number of FPs disagreed more than agreed to “receiving summary reports” (P = .001), as well as for “familiar with guidelines” (P = .01). The biggest barriers to cancer survivor care were lack of communication with specialists (53%), and lack of knowledge of survivorship care (36%).

Conclusion Our findings demonstrate the need to improve communication with specialists, as well as to educate FPs about survivorship. In response, a Mainpro-accredited pilot workshop was created and has thus far been delivered to 123 FPs at 4 sites. Preliminary findings indicate high relevance of pilot content (95%).

Context Although case management (CM) is increasingly implemented to address the complex needs of vulnerable clients, few studies have examined patients’ and FPs’ experiences. The aim of this study was to examine the experience of people with chronic diseases, frequent users of health care services, and FPs who participated in a CM intervention.

Design A descriptive qualitative approach.

Setting Four family medicine groups in the Saguenay region of Quebec.

Participants There was a total of 45 participants: people with chronic diseases (n = 25), who are frequent users of health care services, and FPs (n = 20) who participated in a CM intervention by a nurse in primary care.

Methods Data were collected through in-depth interviews (frequent users) and 4 focus groups (FPs). Thematic analysis of the verbatim transcripts used the 6 dimensions of service integration proposed by the National Collaboration for Integrated Care and Support: 1) consideration of patient and family needs; 2) communication with the patient and between providers; 3) access to information; 4) involvement in decision making; 5) care planning; and 6) transitions between various health professionals.

Results Patients confirmed that the CM nurse was usually their privileged contact with primary care. She or he actively involved them in developing and carrying out their individualized services plan with other health care partners. Patients and FPs believed that patients’ needs were taken into consideration. They thought that the case manager facilitated communication and coordination with and among health care partners as well as better access to relevant information, but a few patients and FPs had not fully understood CM at the beginning of the intervention. Some FPs would have wished to have had more interactions with the case manager. A few people thought that this implies too many appointments. Patients and FPs agreed on the fact that patients were actively involved in decision making. The individualized services plan was considered helpful to improve transitions between services.

Conclusion The experience of people with chronic diseases, who are frequent users of health care services, and FPs who participated in a CM intervention in primary care was, overall, positive, regarding care integration.

Context The increasing selective serotonin reuptake inhibitor prescription numbers in the population are being driven in part by continuation of medication after acute treatment as maintenance therapy to prevent recurrence. Most depression is treated in the primary care setting. There is no evidence from randomized controlled trials of the effectiveness of this long-term maintenance treatment in primary care patients.

Objective To assess the effectiveness of long-term maintenance treatment with fluoxetine for prevention of depression recurrence in primary care; and to compare outcomes at 18 months for those who trialed discontinuation compared with those who did not.

Design Multicentre double-blinded randomized controlled trial.

Intervention Ongoing maintenance fluoxetine treatment compared with tapered withdrawal to placebo.

Participants Participants identified from electronic medical records’ prescribing data had initial eligibility screening by their family doctors: they were aged 18 to 75 years, had a historical diagnosis of depression, and were taking fluoxetine as maintenance treatment to prevent recurrence for at least 15 months.

Follow-up Follow-up occurred at weeks 1, 2, and 4, and months 3, 6, 9, 12, 15, and 18 to monitor recurrence, side effects, withdrawal effects, other reasons for exiting treatment, and general functioning.

Main outcome measures The primary outcome was recurrence of depression over 18 months. Secondary outcomes included comparison of all outcomes at 18 months for those who trialed withdrawal with those who did not.

Results Among identified patients, 33% (419 of 1273) responded to the invitation to participate. Of these, 156 were ineligible or did not consent and 263 were randomized into taper and continuation arms. There were 30 (23.3%) depression recurrences in the taper arm and 14 (10.5%) in the continuation arm (absolute difference 12.8%; 96% CI 3.4% to 2.3 %; P = .005). For every 17 patients taking maintenance antidepressant medication, 1 was unable to discontinue because of withdrawal symptoms. At 18 months, 47% of the taper group and 4% of the continuation group were no longer taking antidepressants (absolute difference 43%; 96% CI 33% to 53%).

Conclusion The absolute benefit (12.8%) is similar to that for acute treatment. Most patients taking maintenance antidepressant medication experienced no benefit over 18 months. It seems reasonable to present these data to patients and offer a trial of discontinuation: the number needed to trial for 1 patient to successfully discontinue medication at 18 months is 2. There was no difference in outcome measures at 18 months for those who trialed discontinuation.

Context The College of Family Physicians of Canada (CFPC) is responsible for setting standards for accreditation and the curriculum that individual family medicine residency training programs must follow. There are no specific criteria for patient safety and quality improvement content at this level.

Objective To conduct a thematic analysis of the CFPC’s current documentation for accreditation (The Red Book) and family medicine curriculum (Triple C reports 1 and 2) with respect to patient safety competencies as outlined by the Canadian Patient Safety Institute (CPSI).

Methods Thematic analysis of the family medicine competencies was conducted as outlined by the Triple C curriculum (using the 4 principles of family medicine, the CanMEDS–Family Medicine framework, and an in-depth curriculum guide for implementation), as well as by overarching Red Book standards for accreditation and enhanced skills training programs, with respect to the 6 CPSI patient safety competencies. The CPSI patient safety principles were used as nodes to which the CFPC documentation was coded by 2 researchers. Researchers met to discuss any discrepancies in coding until consensus was reached.

Results Across the CFPC documentation, the most commonly found patient safety competencies were “work in teams” (n = 24 coding references) and “create a culture of patient safety” (n = 23 coding references). The least commonly found patient safety competencies were “optimize factors” (n = 1 coding references) and “recognize, respond to, and disclose adverse events” (n = 1 coding references). The patient safety competency of “manage safety risks” was also not common (n = 5 coding references).

Conclusion A relevant framework for patient safety and quality improvement education is needed for family medicine residency training programs across Canada. This framework should focus on optimizing human factors and disclosing near misses or adverse events, as well as on managing patient safety risks in primary care. This is to ensure that programs address patient safety across the full continuum of care, including community-based care. Standards for accreditation and the curriculum from the CFPC could help to create a patient safety culture in family medicine training. Future research should include a needs assessment of family medicine residency training programs with regard to patient safety and quality improvement content.

Context Access to mental health services during pregnancy is most commonly mobilized through formal mental health screening. However, despite recommendations by international organizations, fewer than 20% of prenatal care providers routinely conduct mental health screening. A prominent barrier is providers’ perceived concern regarding the acceptability of screening and its results to women.

Objective To describe pregnant women’s views of mental health screening, including screening methods and personal and system-related barriers and facilitators that influence their responses to provider-initiated mental health screening.

Design Multisite, cross-sectional survey.

Setting Five maternity clinics and prenatal classes at 2 community hospitals in Edmonton and Medicine Hat, Alta.

Participants All pregnant English-speaking women presenting for care in the 7 sites were recruited consecutively.

Intervention Self-report questionnaire completed on computer tablet.

Results Of 500 women approached, 460 (92%) participated. Overall, 98% were very (75%) or somewhat (23%) comfortable with being asked about their mood during pregnancy. Women were most comfortable with paper-based (> 90% very or somewhat comfortable) and computer-based (> 83%) screening that was coupled with discussion of results with their provider. Comfort levels were similar whether the screening was completed in the clinic or at home. Telephone-based screening by a nurse was least favoured (62% very or somewhat comfortable). Important barriers to disclosure included others’ normalizing emotions, desiring to handle their mood on their own, preferring to discuss feelings with significant others, and not knowing what emotions were “normal.” Important facilitators included characteristics of the care provider (sensitive, interested), reassurance that emotional health was part of normal care, hearing that other women have emotional issues in the perinatal period, and reassurance that help (in addition to medications) is available.

Conclusion Women are receptive to various screening approaches. Efforts to minimize barriers to and enhance facilitators of screening are key strategies in implementing universal, routine perinatal mental health screening.

Context Selecting the most appropriate candidates for residency training is crucial. A poor match between a resident and a specialty or program can result in the need for interventions that are costly, time intensive, and stressful for both the resident and the faculty member. Developing better methods for selecting residents should result in a better match between residents and programs.

Objective In this study, we triangulated evidence from different sources to develop evidence-based interview questions for use in the Canadian Resident Matching Service selection process.

Design Literature review, data-mining.

Methods In Phase 1, a literature review was carried out to determine common factors found among residents who encounter difficulty in training. In Phase 2, files of past residents in difficulty were examined to look for common indicators that could be targeted through interviews. In Phase 3, evidence was triangulated to develop evidence-based interview questions. In Phase 4, an expert panel reviewed the questions multiple times until consensus was reached.

Results The literature review revealed that deficiencies in professionalism, resiliency, and problem-solving approaches were common factors among residents who encountered difficulty in training. The file review confirmed these areas, as well as sincere interest in the specialty. Key papers from the literature review were re-examined to determine if there were validated questions to ask of candidates. Once essential aspects of key factors were identified, a preliminary list of questions was generated. The expert panel reviewed the questions, and questions went through multiple iterations before finalized forms. Last, a scoring schematic for the individual questions was determined, which included specific elements to watch for in responses. Informally, faculty reported that they were pleased with the new questions; answers given by candidates were more informative than previously. Additionally, scores for candidates showed greater range than in previous Canadian Resident Matching Service rankings, indicating that the new interview questions distinguished between candidates better than previous questions did.

Conclusion While much effort is required to develop evidence-based interview questions, initial results suggest that the new questions better distinguish between candidates. Long-term evaluation will include determining how many residents encounter difficulty in training.

Context Despite best efforts in selection, 6% to 10% of accepted residents in family medicine experience difficulties. Interventions can be time-consuming and expensive. One strategy to reduce the number of interventions needed is to improve the resident selection process, ensuring that the best match possible is being made between resident and program.

Objective The Canadian Resident Matching Service (CaRMS) applications of past residents who encountered difficulty were examined to determine if there were common indicators of difficulty. The goal was to use these data to develop an evidence-based CaRMS file review process.

Design Secondary data analysis, literature review.

Methods Program directors identified 30 residents who encountered difficulty (“cases”). Matched controls who did not encounter difficulty (“controls”) were also identified.

Main outcome measures All CaRMS application materials were extracted from resident files (de-identified and coded). Each component of the CaRMS application was analyzed, with a focus on differences within the application elements between the residents who encountered difficulty and those residents who did not. Factors were triangulated with existing findings in the literature.

Results and findings Multiple indicators were identified (commonalities among cases, notable differences between cases and controls); several reflected findings in the literature. In the curriculum vitae section, cases did not have much variety in their work or volunteer roles and had few leadership positions. For clinical electives, total time was less of an indicator than was the degree to which applicants described their electives. Cases gave vague descriptions, while controls gave detailed descriptions. For reference letters, there were certain words that occurred with much greater frequency for the controls, such as “exceptional” and “well-rounded.” For the personal letter, cases were more likely to be vague, especially about specialty choice, and have inconsistencies within the personal letter. Controls had highly descriptive personal letters that emphasized and gave supporting evidence for specialty choice. The new file review form was piloted with 10 faculty reviewers and their current residents. Greater agreement was seen than with the previous form.

Conclusion The new evidence-based form shows promising results in pilot use. Long-term evaluation of the 2015 cohort will indicate if this process is valid as a better way to select residents.

Context Remediation of residents who encounter difficulty is costly and time intensive, more so when problems have been occurring for an extended time. It is also stressful for family medicine clinical faculty. Improvement in identification of residents in difficulty is one of the goals of competency-based assessment, and was one of the objectives when we implemented the Competency-Based Achievement System (CBAS) in our residency program in 2009. Through improved detection of residents who are struggling, we can improve training outcomes.

Objective To determine if there has been a change to rotation flag patterns for residents experiencing difficulty before and after implementation of CBAS.

Design Secondary data analysis (descriptive comparative cohort study) using archived resident data files (3 years before CBAS implementation and 3 years after CBAS implementation).

Setting Large Canadian family medicine residency program.

Participants Archived files were examined for all residents in the 2005 to 2007 and 2009 to 2011 entry cohorts (N = 393).

Main and secondary outcome measures Files were searched for evidence of flags on in-training rotation evaluations and summative progress reports.

Results Overall, the total number of in-training rotation evaluations with flags has decreased since implementation of CBAS (40% in 2005 to 17% in 2011); as well, there has been a steady increase in flagged residents who are remediated (from a low of 4% in 2006 to 23% in 2011). Total remediations have decreased from 8% in 2005 to 4% in 2011. Rotations where residents are likely to be flagged have mostly remained constant (family medicine, obstetrics, pediatrics, critical care, internal medicine), with the exceptions of general surgery and orthopedics.

Conclusion After implementation of CBAS, residents in difficulty appear to be detected earlier and problems are addressed more quickly. This might be owing to CBAS resulting in more detailed comments from the preceptors, leading to resolution of existing issues more quickly by making the resident aware of their deficiencies and providing resources to overcome those difficulties. Overall, implementation of CBAS appears to have increased the likelihood that residents encountering difficulty obtain the attention they require earlier in their training, resulting in fewer flags on rotations.

Context Perinatal opioid use is highly prevalent in primary care and has been linked to adverse neonatal outcomes. Therefore, integrated care including methadone maintenance treatment (MMT) is the standard of care for the management of opioid use disorders during pregnancy. The most important complication of perinatal MMT is neonatal abstinence syndrome. Neonatal abstinence syndrome is an acute, time-limited consequence of MMT that has been extensively studied. However, long-term developmental outcomes of children with in utero methadone exposure are not well known.

Objective To describe developmental outcomes of a Canadian cohort of children with antenatal methadone exposure.

Design Retrospective chart review

Participants Opioid-dependent pregnant women who attended an integrated primary care–based program and their infants who were born between 2009 and 2011 and followed for 2 years at the Neonatal Follow-up Clinic at St Joseph’s Health Centre in Toronto, Ont, were included in this study.

Main outcome measures Maternal and neonatal demographic characteristics and developmental outcomes

Results Twenty-three maternal-newborn dyads met inclusion criteria. Most mothers had a history of prescription opioid use disorder requiring MMT. The mean birth parameters were as follows: birth weight 2881 g, length 49 cm, and head circumference 34 cm. Almost half of these neonates were in part breastfed, and 87% were discharged home in the care of a biological parent. Fourteen of these children attended the clinic for developmental assessments with 9 lost to follow-up. The overall prevalence of gross motor abnormalities was 36.3% before 12 months and decreased to 12.5% thereafter. Fine motor and communication delays were observed after 1 year of age in 25% and 37.5% of infants, respectively. Abnormal social development was rare.

Conclusion Our study demonstrated that gross motor delays seemed to resolve after 1 year of age, with communication and fine motor delays becoming more prevalent on subsequent assessments. These results also emphasize the importance of early screening and intervention programs for this vulnerable population. Primary care providers play a critical role in supporting families with parental substance use disorders to optimize the environment in which these children are raised and to positively influence their development.

Context Ethnic minorities are less likely to access clinical prevention services, but there is limited evidence about how to facilitate uptake for these populations.

Objective The purpose of this study was to examine the barriers and facilitators to clinical prevention uptake for the South Asian population in Surrey, BC, and to identify effective methods to increase the use of services.

Design The study used an inductive, qualitative design based on a series of focus groups with members of the South Asian community. Transcripts were analyzed using thematic data coding, and responses were compared based on clinical prevention topic area and group characteristics.

Participants A total of 62 adults between the ages of 40 and 80 participated in 1 of 8 focus groups in health care and community settings.

Instrument The focus group guide for the study covered a total of 10 priority clinical prevention topics and each group discussed 3 of these topics. In order to facilitate open discussion participants were asked to respond to questions based on a scenario about a fictitious person who learns about a clinical prevention topic and then chooses not to access it. Participants also completed a brief demographic interview.

Findings The findings from this study can be classified into 3 levels of recommended changes. At the micro level, themes included barriers to effective communication with care providers and logistical concerns. At the meso level, participants suggested exploring different delivery models and locations for services. At the macro level the discussion focused on the creation of stronger policies in the interest of the public good.

Conclusion Based on these findings we concluded that barriers to culturally accessible and acceptable care occur at multiple levels of the health system. These findings can be used to inform how existing and emerging programs, educational materials, and services can be tailored to meet the values and needs of South Asians in order to increase the uptake of clinical prevention services for this population.

Context Falling as an older adult is a common and complex phenomenon with potentially serious consequences; however, there has been a relatively low uptake and acceptance of fall prevention strategies in this population. Clinical approaches are needed to identify factors that might facilitate or impede older adults’ acceptance of risk status and need for intervention.

Objective We hypothesize that a narrative approach to interviewing older people who have fallen will reveal important gaps compared with clinical history taking that could better inform collaborative clinical decision making.

Design A series of comparative case studies based on a convenience sample of 15 older adults referred to a falls prevention clinic was used, each including the patient narrative elicited by a member of the research team; clinical history generated by the falls clinic team; and clinicians’ impressions elicited through team interviews. We compared these 3 components within and across cases using thematic analysis to generate an enriched understanding of each participant’s experience of falling. Strategies used to improve trustworthiness included field notes, verbatim transcription, member checking, audit trail of analysis meetings, presentation of preliminary findings to multidisciplinary audiences, and external review.

Findings Three key concepts emerged from patient narratives: perception of the cause of the fall (internal vs external locus of control); the significance of the fall; and perceived need for and benefits or risks of action. Interviews with the falls clinic team identified instances of salient information in patient narratives that was missed by clinical history taking or noted but excluded from the clinical report for various reasons. Whether the missing narrative information was the result of failing to elicit it or opting to not report it, falls team clinicians confirmed that the inclusion of this missed information might have resulted in a more holistic understanding of patient context and facilitated collaborative management, particularly in patients not readily engaging with fall prevention strategies.

Conclusion Important insights into older adults’ experiences with having fallen were revealed using a narrative approach. In concert with clinical history taking, a narrative approach could be of particular use with patients who do not readily accept interventions based on traditional falls risk assessments.

Context Despite evidence demonstrating differences in cancer survival between provinces, and between regions within Ontario, the reasons for these differences are not fully understood. Previous research has suggested that diagnostic or treatment delay might be a contributing factor; however, little is known about geographic differences in diagnostic and treatment delay, especially in the Canadian context.

Objective To examine urban and rural differences in the time intervals between the onset of first symptoms and cancer diagnosis and treatment.

Design This work was part of the International Cancer Benchmarking Partnership, an international collaboration involving 9 countries. It comprised 5 separate areas of research (modules), each examining reasons for observed differences in cancer survival. The focus of this research is module 4, which was a cross-sectional survey that examined the underlying causes of cancer diagnostic and treatment delay.

Participants In Ontario, adults aged 40 years and older diagnosed with breast, colorectal, lung, or ovarian cancers were identified through the population-based cancer registry. Eligible individuals were first contacted by Cancer Care Ontario for consent to be contacted by International Cancer Benchmarking Partnership research staff. Those who consented were mailed a survey between May 2014 and October 2015.

Instrument The survey examined the pathway to diagnosis (eg, screening, emergency department, or primary care presentation); patient, diagnostic, and treatment time intervals; perceived health status; treatments for comorbidities; and sociodemographic factors.

Outcome measures The primary outcome is the patient (onset of symptoms to presentation to health care provider), diagnosis (presentation to health care provider to cancer diagnosis), treatment (cancer diagnosis to start of treatment), and total (onset of symptoms to start of treatment) time intervals. The predictor variable is urban or rural status (defined using the 2011 Canadian census).

Results Recruitment will be completed by October 2015. To date, 2510 individuals have consented (breast [n = 1205], colorectal [n = 745], lung [n = 475], or ovary [n = 85]), of which 88.5% have returned a completed survey (breast [n = 1091], colorectal [n = 634], lung [n = 402], ovary [n = 79]). Differences in the patient, diagnostic, and treatment time intervals by urban versus rural status, by pathway to diagnosis, and by cancer type will be presented.

Conclusion The results will inform Cancer Care Ontario’s strategy to improve the diagnosis and treatment time intervals, thereby potentially improving cancer survival.

Context Intrauterine device (IUD) insertion has been traditionally taught at scheduled training sessions or during rotations at contraception teaching clinics. The limited number of skilled faculty who can teach procedural skills, and issues of time and access, have made it difficult for residents to learn this procedure.

Objective To determine how effective the use of video-module instruction is as an alternative procedural skill teaching method to deliver IUD insertion training for family medicine residents.

Design This randomized, 2-group (n = 39) pretest-posttest (written) experimental research design focused on the comparison of residents’ posttest performance based on the traditional instructional (designated site demonstration) and video-module approaches to teaching the IUD insertion procedure.

Participants Incoming 2012 first-year family medicine residents (n = 39), 53.8% (n = 21) of whom were male and 46.2% (n = 18) of whom were female. Exclusion criteria included previous IUD insertion during clinical training or previous medical practice that involved IUD insertion (eg, international medical graduates).

Intervention Exposure to traditional instruction or video-module approaches to teaching the IUD insertion procedure.

Main outcome measures Assessment of knowledge and performance of the procedure.

Results Although the traditional and intervention instructional groups both improved significantly in their clinical knowledge scores from pretest to posttest, there was found to be no difference between the groups on the written examinations. On the posttest IUD insertion performance assessment, residents in the video-module intervention group performed significantly better than those in the traditional instructional group (P < .05, mean effect size difference Cohen d = 0.75). The internal reliability (Cronbach α) of the 27-item performance checklist was α = 0.77.

Conclusion Both the traditional and video-module instructional methods were found to be reliable and valid approaches for the teaching and learning of the IUD insertion procedure. The results provide support for the use of a video-module teaching session as a potential method to supplement or enhance IUD insertion procedural skill training. The ability to access this method of instruction online makes this approach highly feasible for residents. In addition, the performance assessment and patient interview checklists that were developed as part of the video-module instructional method might prove to be useful to guide and evaluate the IUD insertion procedures with actual patients.

Context BETTER (Building on Existing Tools to Improve Chronic Disease Prevention and Screening in Primary Care) is a chronic disease prevention and screening (CDPS) program that has been implemented in various primary care settings in Newfoundland and Labrador. The program aims to improve CDPS for cardiovascular disease, diabetes, cancer, and associated lifestyle factors in patients aged 40 to 65 years. The key component of BETTER is the prevention practitioner (PP)—a health care professional with specialized skills in CDPS who structures and complements physicians’ CDPS practices in the primary care setting. Some physicians involved in BETTER were sceptical of the benefits of having a PP.

Objective To explore patients’ perspectives on their visits with PPs in 3 primary care settings in Newfoundland and Labrador.

Methods Of 154 patients who received 1 or more prevention visits with a PP, 90 volunteered to provide written feedback. Besides their demographic characteristics, patients were asked what they liked about their visits, what they would have liked to have been different about their visits, and to provide other comments. We employed qualitative description to analyze the data.

Findings We identified 4 main themes regarding patients’ perspectives on their visits with a PP. 1) Value of the visit: All 90 patients who provided feedback unanimously reported their visits with the PP to be positive and appreciated having a PP as part of the health care team. 2) Visit characteristics: Patients valued the focus on health (as opposed to disease) and that the visit was personalized, comprehensive, and motivating. 3) Characteristics of the PP: The patients saw their PPs as key to their positive experience with the program. Patients characterized the PPs as compassionate, attentive, thorough, knowledgeable, motivating, and supportive. 4) Access concerns: Patients perceived their previous access to preventive care to be limited and expressed concerns about the termination of the program.

Conclusion Patients’ views that a PP should be an available resource within primary care settings were a contrast to the perception of some physicians involved in BETTER. Our findings might help inform stakeholders in primary care settings as they consider implementing programs to enhance their current CDPS practices.

Objective To assess the effectiveness of the iOphthalmology mobile application for enhancing patients’ knowledge of and satisfaction with cataract surgery at the Kensington Eye Institute, a high-volume Canadian cataract centre in Toronto, Ont.

Design Cross-sectional study.

Participants Seventy-seven patients who were undergoing cataract surgery at the Kensington Eye Institute for the first time with no previous ocular pathology.

Intervention To improve patient knowledge of surgical procedures, we developed the iOphthalmology mobile application in 2014. This application is currently available for no cost on the Google Play digital distribution platform on Android devices, including smartphones and tablets. This application was used by cataract surgery patients with the goal of increasing patient knowledge of cataract surgery and increasing patient satisfaction.

Main outcome measures Following patient consent, a questionnaire was administered to assess the following: knowledge of cataract surgery, perceived outcome of surgery, decisional conflict level, uncertainty level, sex, occupation, age, highest education level, Ontario Health Insurance Plan status, and literacy status. After using the application, overall satisfaction of all patients was measured.

Results Both decisional conflict scores and cataract surgery knowledge scores significantly improved after using the application (mean difference 7.1 [P = .02] and 3.4 [P < .01], respectively). Multiple regression analysis revealed being illiterate, decisional conflict level, and female sex to be critical factors in the degree to which the application increased patients’ knowledge of cataract surgery (β = 3.1, P < .003 for literacy level, β = 2.0, P = .01 for decisional conflict, and β = −0.9, P < .01 for female sex). A total of 71 of 77 (92.2%) patients were satisfied with the mobile application. Patient satisfaction score was significantly correlated with the overall change in patients’ knowledge of cataract surgery score (correlation coefficient 0.31, P = .01).

Conclusion The iOphthalmology mobile application for Android devices is effective in enhancing patients’ knowledge regarding cataract surgery and in decreasing their anxiety in their decision to proceed with cataract extraction surgery. This open-source mobile application can potentially benefit patients who might have restricted access to health care, including patients in countries with low literacy rates. The use of this application before cataract surgery can help to alleviate the worldwide burden of cataract blindness.

Context In Canada, chronic hepatitis B infection is largely a disease of immigrants from endemic countries, with approximately 600 000 cases, based on a 6% rate in immigrants. Among untreated chronic hepatitis B patients, 20% to 25% can progress to cirrhosis, chronic liver failure, and hepatocellular carcinoma.

Objective To evaluate chronic hepatitis B management and monitoring in a primary care practice with a high prevalence of cases from Asian immigrants. The project identified HBsAg–positive patients and assessed care according to Canadian hepatitis B guidelines for monitoring.

Design Retrospective chart review from a primary care office with predominantly immigrant patients from Asia.

Setting Toronto, Ont.

Participants A total of 2100 patients were screened. We identified 97 (4.6%) patients with HBsAg–positive status. All patients were Asian immigrants.

Methods The study assessed the data from laboratory results in ALT, HBeAg status, HBV DNA viral load, platelet counts, abdominal ultrasound, and other data for liver cirrhosis assessment. Patients were assigned to 1 of 3 groups. The “red” high-risk group included high-viral-load patients for therapy consideration. The “yellow” missing-laboratory-results group included patients missing at least 1 of the results; and the “green” group was the low-viral-load group.

Results Among these 97 patients with hepatitis B, 33.0% of the patients were in the “red” high-viral-load group, with only 11.3% receiving treatment for hepatitis B, and 51.5% of the patients were in the “yellow” missing-laboratory-results group, missing 1 of HBeAg, HBV DNA, ALT, or platelet results; 42.3% of patients were not undergoing regular abdominal ultrasound, with more than 12 months’ interval. Only 23.7% of patients were in the “green” low-viral-load group.

Conclusion This project has identified that a considerable care gap exists in chronic hepatitis B monitoring. The Canadian hepatitis B consensus guidelines (2012) recommend close regular monitoring of chronic hepatitis B. More resources for hepatitis B care from federal and provincial governments are needed owing to the expected increase of hepatitis B from immigrants. As with treating other chronic diseases, primary care physicians should have more support to establish a vigilant monitoring system for chronic hepatitis B to reduce the rates of cirrhosis and liver cancer in future.

Context Chronic low back pain is one of the most common presenting complaints in the physician’s office. Treatment is often challenging and recovery depends on various factors, often resulting in considerable investments of time and resources.

Objective To assess which Web-based interventions aimed at chronic low back pain are of benefit to patients?

Methods Randomized controlled trials (RCTs) studying Web-based interventions directed at adults with chronic low back pain were reviewed. Retrospective studies, narrative reviews, non-randomized trials, and observational studies were excluded. Electronic databases and bibliographies were searched.

Results Eight unique RCTs were identified (n = 1414). The number of patients randomized in each trial ranged from 51 to 580. Four trials studied online cognitive behaviourial therapy (CBT), 2 studied moderated discussion through e-mail or online chat, and 2 studied other Web-based interventions with interactive features. Use of CBT was associated with reduced catastrophization among patients. Mixed results were reported with regards to reduction in pain levels and disability. One study that measured health care utilization reported reduced utilization with the use of moderated e-mail discussion.

Conclusion Limited data are available regarding effective Web-based interventions to improve outcomes for patients with chronic low back pain. Eight RCTs with small sample sizes were identified in this review. Online CBT appears to show some promise in terms of reducing catastrophization and improving patient attitudes. Further research in this area with larger-scale studies focusing on appropriate outcomes appears to be a priority.

Context Substance use disorders (SUDs) are a considerable cause of morbidity, mortality, and health care utilization. Few patients are in treatment, largely as a result of poor access to and retention in specialized addiction treatment programs. Primary care might be the solution: outcomes are as good as those in specialized care, retention is better, and capacity is greater. However, because of a lack of training and support, few primary care providers (PCPs) screen and appropriately care for patients with SUDs. As well, PCPs continue to inappropriately prescribe high doses of opioids, increasing the risk of harm (including addiction) to patients. Although there is substantial evidence that shared-care models between PCPs and addiction physicians are the solution to service and knowledge gaps, this model is rarely used in addiction medicine in Canada. We are conducting a program evaluation of a shared-care addiction medicine pilot, SUN:SHARE.

Objective To determine if the PCPs participating in the SUN:SHARE pilot improve their management of SUDs.

Setting The three participating sites are an inner-city community health centre, an inner-city family health team, and a community family health team affiliated with an academic centre.

Intervention In the intervention, addiction physicians facilitate addiction education sessions, assess patients with SUDs, and have case discussions with PCPs at the community sites. The addiction physicians also provide urgent telephone and e-mail consultations with PCPs.

Methods We will survey PCPs at baseline and at 6 months to look for reported changes in knowledge and behaviour (counseling, prescribing, and referrals to ancillary services). Additionally, we will conduct patient and provider satisfaction interviews, and track how frequently the telephone and e-mail consultation services were used. We will use each site’s electronic medical record to track participating PCPs’ prescribing patterns at baseline, 6 months, and 12 months. We will look for changes in the number of prescriptions for medications to treat alcohol use disorders (naltrexone, acamprosate, and disulfiram) and opioid use disorders (buprenorphine-naloxone) and in high-dose opioid prescribing (above 200 mg morphine equivalent per day).

Context Several studies have shown clinical differences in the evolution of schizophrenia and the response to antipsychotics (AP) according to patients’ age and sex. However, few clinical guidelines take this into account in their pharmacologic recommendations.

Objective To describe, in a real-world context, the use of APs (oral or long-acting injectable), according to patients’ age and sex, among those living with schizophrenia.

Methods This is a retrospective cohort study using medical administrative databases available in Quebec. The study cohort consists of all adult patients living in the province of Quebec, diagnosed with schizophrenia, admissible to the public drug insurance plan, and initiating an AP (with a clearance period of 1 year) between January 1998 and December 2006.

Results A total of 16 225 patients (46.2% women) met the inclusion criteria. The median age at AP initiation differed considerably between women (49 years) and men (38 years). The first claimed AP was mainly prescribed by psychiatrists (women 56.7%, men 61.0%, decreasing with age) and general practitioners (women 30.6%, men 27.3%, increasing with age). Moreover, 92.3% of women and 89.2% of men received an AP in monotherapy, 77.9% from a second-generation AP (women 77.2%, men 78.5%), with a slight difference in the molecule used between women and men (olanzapine 42.3% vs 47.9%; risperidone 45.6% vs 39.3%; quetiapine 10.6% vs 10.1%; clozapine 1.5% vs 2.7%). Since long-acting injectable second-generation APs were not available before 2004, only 7.0% of patients taking monotherapy used this type of AP (women 6.7%, men 7.3%). As many as 11.7% of patients did not pursue their treatment after the first month.

Conclusion This study reveals statistically significant differences according to age and sex regarding AP utilization in a “real-world” setting. Ongoing studies will evaluate if these differences in AP treatment lead to different outcomes.

Introduction The high rates of “unplanned” hospital admissions are an increasing source of pressure on health system resources. Ambulatory care sensitive conditions (ACSCs), such as heart failure, diabetes, and chronic obstructive pulmonary disease (COPD), are conditions for which hospitalization could be avoidable with ambulatory care management. Chronic obstructive pulmonary disease is the main cause of emergency admissions and hospitalizations in Canada, with a mean length of stay of 10 days and one of the highest rates of rehospitalization.

Objective To develop a predictive model for future readmissions of patients with COPD, based on the analysis of their hospital episode of care (EOC), and to identify factors or missed opportunities that can be modified to reduce the readmission rate.

Methods This project is a secondary analysis of hospital data for a cohort of patients hospitalized between April 2012 and March 2013 with a primary diagnosis of COPD (ICD-10 J40-J44, J47). The hospital EOC starts at the patient admission (index admission) and ends at hospital discharge, passing through temporal succession of treatments. The main outcomes are the time until the next readmission (COPD and all cause), the number of readmissions, and the total number of days in hospital during the year after the index admission. Independent variables will be analyzed according to the following temporal sequence: 1) status before admission (eg, medical history, comorbidity); 2) status at admission (eg, pulmonary condition); 3) status and care during hospitalization (eg, pharmacological treatments); and 4) status and prescribed treatments at hospital discharge (eg, respiratory conditions, pharmacologic prescription, pulmonary rehabilitation). Data mining algorithms and improved survival analysis with adjusted Cox regression will be used to detect sequential patterns of predictive variables of readmission.

Anticipated results and conclusion To our knowledge, introducing sequential patterns of predictive variables on EOC is a new methodologic approach to identifying factors of rehospitalization for an ACSC. A validated methodology for COPD could be then transferred to other ACSCs, such as heart failure and diabetes, and might improve the identification of “missed opportunities” to reduce rehospitalization rates.

Context Asthma and chronic obstructive pulmonary disease (COPD) are among the most common respiratory illnesses encountered in primary care. Although patients often present with similar symptoms, it is important to distinguish between the 2 diseases because their pathophysiology and management are very different. Simple spirometry is an objective pulmonary function test that can support a diagnosis of asthma or COPD, but recent evidence has demonstrated that patients with moderate to very severe COPD also meet spirometric criteria for asthma diagnosis. Spirometry interpretation algorithms (SIAs) should conform to diagnostic guidelines while acknowledging this overlap and prompting the clinician to consider further clinical assessment where such an overlap exists.

Objective To determine the variations in SIAs that exist in the published literature and online sources.

Design MEDLINE, EMBASE, and mainstream search engines were used to identify all SIA-related material dating back to 1990. Keywords used were spirometry, algorithm, algorism, and algorhythm.

Results Overall, 17 out of 27 SIAs are unable to serve well as stand-alone documents. Twenty-four SIAs lack a logic string for postbronchodilator forced expiratory volume in 1 second to forced vital capacity ratio, potentially impeding COPD diagnosis. Ten SIA rely solely on postbronchodilator improvement to distinguish asthma from COPD. Twenty-three SIAs lack a prompt for bronchodilator challenge when the forced expiratory volume in 1 second to forced vital capacity ratio is normal. Four SIAs unnecessarily recommend testing the diffusing capacity of the lungs for carbon monoxide to confirm COPD diagnosis.

Conclusion Overall, 25 out of 27 SIAs feature variations that might lead to disease misclassification. Further studies are needed to confirm whether this link truly exists. This study points to the need to minimize SIA variability, including the need for a standardized approach to the spirometric diagnosis of asthma and COPD.

Context In 2000, patients with advanced heart failure (HF) accounted for the second-highest number of hospital days in Canada. The condition is responsible for a high burden of hospitalization for symptoms that are currently difficult to manage in the ambulatory setting and associated with high mortality. We propose that providing home-based palliative care (PC) for these patients can meaningfully reduce the number of hospital visits and days in hospital, with the ultimate goal of improving patients’ overall satisfaction with care and quality of life.

Methods In a quality improvement analysis of health care utilization of a sample of patients with advanced HF in an urban tertiary hospital, we measured and compared the rate of clinic visits, emergency department (ED) visits, hospital admissions, and days in hospital before and after implementing home-based PC.

Results A total of 32 patients with HF (69% female, mean age of 83.6 years) were seen in the home-based PC program during a 2-year period. In a preintervention and postintervention comparison, meaningful reductions were seen in rates of clinic visits (mean of 0.3 visits/month and 0.1 visits/month, respectively), ED visits (mean of 1 ED visit/month and 0.2 ED visits/month, respectively), hospital admissions (mean of 1.5 admissions/month and 0.2 admissions/month, respectively), and days in hospital (mean of 12.5 days/month and 4 days/month, respectively). Median length of stay on the home-based PC program was 2.73 months (range 0.1 to 25.6). Once in the home-based PC program, patients received an average of 2.18 visits a month from a PC physician.

Conclusion Home-based care of patients with advanced HF can lead to decreased health care use, including a decreased number of clinic visits, ED visits, and hospitalizations. In turn, we hope it can reduce both patient and system burden and result in greater health-related quality of life for patients with advanced HF.

Context Cardiovascular disease (CVD) is the world’s leading cause of death. In Canada, it is estimated that CVD is the cause of more than 69 500 deaths annually. Substantial evidence from both animal and human studies now links circadian desynchronization and sleep disturbance with adverse health outcomes, including metabolic and cardiovascular dysfunction. Disruption of the circadian system is also associated with increased incidence and severity of intermediate CVD risk factors such as obesity, diabetes, hypertension, and hyperlipidemia. Given the prevalence of chronic circadian disruption (eg, shift work, electric lighting, and social jet lag), it is crucial to examine current literature pertaining to CVD and circadian disruption, and make a case for their acknowledgment in the primary care setting.

Design We present the results from a systematic mixed studies review incorporating quantitative, qualitative, and mixed methods studies, using both animal models and human subjects.

Objective To describe current research linking circadian disruption with CVD and its intermediate risk factors in terms of themes covered, methodology, discipline, intended audience, and the extent to which studies refer to primary care as a potential field of application.

Findings A preliminary literature search suggests a strong, likely causal, link between chronic circadian disruption and CVD. We identified only a small number of articles that discussed the implications of these findings in primary health care. By far most relevant articles we identified were published in basic research (mostly animal studies) and non–family medicine specialist medical journals. Several papers recommended the potential value of chronopharmacotherapy (drug timing) for CVD treatment, but only a limited number of studies were designed to specifically address the effects of such an approach.

Conclusion Based on preliminary findings, we consider chronic circadian disruption a potential modifiable risk factor for CVD. Our results suggest the need to clearly target the primary health care provider audience regarding the causes and consequences of circadian disruption. We recommend that future studies address the effect of the application of circadian concepts in preventive and therapeutic interventions for CVD.

Context and objective Good health begins even before conception. Recommendations from the No Time to Wait: The Healthy Kids Strategy suggest that optimizing patients’ health before conceiving will improve their children’s chances of enjoying good health throughout their lives. In response to these recommendations, the Ontario Ministry of Health and Long-Term Care engaged the Centre for Effective Practice (CEP), in partnership with the Ontario College of Family Physicians, to develop and disseminate the Preconception Health Care Tool (PHCT). The tool is designed to improve maternal and infant health outcomes in primary care by guiding providers’ discussions of health promotion and illness prevention strategies with all patients of reproductive age.

Design The tool was developed using CEP’s integrated knowledge translation approach. A working group comprising a primary care nurse practitioner, a family physician, and CEP staff conducted a comprehensive search, review, and appraisal of existing clinical evidence. Through this process the working group integrated all available evidence and resources into one tool. The tool was tested and refined based on feedback from individuals and organizations with expertise in maternal and infant care, and from providers who practise comprehensive family medicine.

Results The tool presents information for providers to use over a series of visits with their patients of reproductive age to encourage patients to develop a reproductive life plan; assess and optimize patients’ preconception physical and mental health; optimize chronic medical conditions before conception; choose safe medications for women who might become pregnant; and counsel on lifestyle habits (nutrition, physical activity, alcohol, tobacco, and other substances). Current patient and provider resources are included throughout the tool. Preliminary data on dissemination have been collected and analyzed. Since the launch of the PHCT website in March 2015, it has received more than 2100 views. Additional dissemination data will be presented and discussed.

Conclusion The PHCT might help standardize primary care providers’ approach to preconception health care with all patients of reproductive age, and provide guidance and resources to both providers and patients.

Context The World Health Organization recommends exclusive breastfeeding up to 6 months of age, with continued breastfeeding, along with appropriate complementary foods, up to 2 years of age or beyond, owing to breastfeeding’s considerable health benefits. Breast reduction is a common plastic surgery performed in Canada and the United States. The effects of this surgery on breastfeeding are currently unclear in the literature.

Objective To determine the effects of breast reduction surgery on breastfeeding.

Design Systematic review and meta-analysis.

Participants Studies were included if they provided both the absolute number and the percentage of women successful at breastfeeding or lactation following breast reduction surgery.

Methods A medical librarian conducted the literature search. Two independent reviewers reviewed each paper. References of papers meeting the inclusion criteria were reviewed for completeness. Within the search, 2 systematic reviews were identified; references of these were also reviewed. For each accepted study, 2 independent data extractors captured data on the profession of the study authors, study characteristics, type of breast reduction surgery, patient satisfaction with surgery, preoperative discussion on breastfeeding, and number successful at breastfeeding.

Main outcome measures Number of weeks of exclusive or any breastfeeding, percentage of patients satisfied with surgery, percentage of patients with whom breastfeeding was discussed preoperatively, and breastfeeding outcomes by surgery type.

Results Preliminary results of the literature search yielded 495 unique papers; 384 were excluded based on the title or abstract and 60 were excluded after full-paper review, leaving 51 papers for data extraction. A further 15 papers were accepted following reference review. The accepted papers were published between 1946 and 2014, and the main outcome in approximately 40% of the papers was breastfeeding postsurgery.

Conclusion The final results will provide guidance to plastic surgeons as well as general practitioners on counseling women about breastfeeding before and after surgery, and on whether additional breastfeeding support should be provided to these women.

Context Visits to the GP play an important role in the promotion of Canadian children’s health, yet research regarding children’s perspectives on GP visits is scarce. Greater understanding of children’s medical fears is essential to informing interventions for managing procedural distress and pain during GP visits. The current study aims to understand children’s fears specific to GP visits, including fear of medical procedures (eg, needles), fear of experiencing pain, fear of illness, and fear of feeling uninformed about upcoming medical procedures. Equally important, we asked children to describe their positive experiences at GP visits so these can be increased.

Objective To obtain children’s perspectives on GP visits to determine how these experiences can be improved. In particular, this study focused on children’s reports of their positive and negative experiences with GP visits, as well as children’s desire for increased information and involvement in their health care.

Design This is a qualitative, descriptive study. Consistent with the principles of patient-centred care, which emphasize the importance of actively involving patients in their own medical care, this study involved interviewing children directly.

Method A total of 167 participants (sufficient to detect medium effect size at P = .80, α = .05) in grades 2 to 4 (ages 7 to 10 years, mean of 8.07 years) underwent structured interviews regarding their perspectives on GP visits.

Anticipated results All study data has been collected. Participant interview responses will undergo content analysis, which is the systematic and objective description of spoken communication. Participant responses will be coded into the following general categories: what children like or dislike about GP visits, how GP visits can be improved, children’s medical fears, and how to involve children in conversations about their health care.

Conclusion By obtaining children’s perspectives directly, the current study can provide information and guidance for improving GP visits for Canadian children through the lens of patient-centred care. This includes promoting positive experiences at GP visits, managing procedural pain and fear, and increasing preparatory information provided to children about medical procedures. Involving children in their primary health care might help increase children’s understanding of health concepts and prepare them for making medical decisions in adulthood.

Context Medical education is increasingly recognizing and addressing the influence of the pharmaceutical industry on academic physicians and trainees. Among many changes medical institutions have made to try to eliminate directing biased information toward trainees is a restriction on the use of trade names in formal lectures and teaching settings. It is unknown whether there has been a similar trend toward gene