Login Search
Printed from: http://www.napcrg.org/Conferences/AnnualMeeting/AbstractSearch

2014 NAPCRG Annual Meeting

Search by:

Viewing sessions 1 through 20 of 863 sessions

Critical appraisal approaches used in systematic reviews: A meta-narrative synthesis protocol

Quan Nha Hong; Pierre Pluye, MD, PhD


Context: Critical appraisal of included studies is a core step of systematic reviews. A variety of tools have been developed to formalize the appraisal process, and ensure that it is done in a systematic and reproducible manner. To select a critical appraisal tool for conducting a systematic review, researchers face two main challenges. First, more than 250 critical appraisal tools have been reviewed, and there is no clear consensus on the most appropriate tool. Second, different types of systematic reviews have been developed (e.g., reviews of qualitative research studies), and there is a lack of understanding of which critical appraisal approach should be used for each type. Objectives: Identify and compare the critical appraisal approaches undertaken in systematic reviews. More specifically, five research questions will be addressed: What are the critical appraisal approaches used in systematic reviews? What are the key concepts and theories underlying each approach? What are the strengths and limitations of these approaches? What are the differences and similarities between the approaches? What approaches are associated with each type of systematic review (i.e., quantitative, qualitative, and mixed studies reviews)? Design: Meta-narrative synthesis. Results and conclusion: The results will provide invaluable information to researchers to choose the most appropriate critical appraisal approach for their systematic reviews. For each appraisal approach, a storyline of the origin, conceptual and theoretical basis, and strengths and limitations will be yielded. In addition, standards for conducting meta-narrative synthesis were recently published, and still few researchers have used this method. We will contribute to its development by keeping a detailed log of the review process and reflecting on the challenges encountered.

A Trial of a Multi-Step Intervention to Improve the Referral Process, Based on Results of a Failure Modes and Effects Analysis

Caroline Honsa, MD; Neha Sachdev, MD


Context: An effective referral process is essential to providing patients access to specialists. In many primary care offices, the setting in which most referrals are ordered, there is potential to improve the process. A Failure Modes and Effects Analysis (FMEA) was performed using data from cardiology referrals placed by primary care providers (PCPs) at Erie Family Health Center to Northwestern Memorial Hospital in order to identify the failure modes with the greatest frequency and severity. Objective: To test the following hypothesis: a multi-step intervention targeted at high priority failures within the referral process is effective in improving the proportion of successfully scheduled cardiology visits and subsequent PCP review of specialist recommendations. Human Subjects Review: In process. Design: Interventional clinical trial. Setting: One community, ambulatory, primary care office on the near west side of Chicago, referring patients to Northwestern, a large academic institution. Participants: Adult Erie patients (67% best served in Spanish, 37% uninsured). Intervention: 1) Erie IT revises the referral order form to allow referral processors easy access to all required information; 2) Northwestern ensures that bilingual schedulers are available to serve Spanish-speaking patients; 3) Schedulers send a copy of the referral order form to the cardiologist; 4) Erie medical records staff directs the completed consult note to PCP. Main outcome measures: 1) Improvement in repeat FMEA results; 2) Proportion of patients scheduled in cardiology within a month of referral processing; 3) Proportion of consult notes received by referring PCPs within a month of the visit. Anticipated Results: Improvement in the referrals process as outlined above. Conclusions: Failures in the referral process are common, limiting access to specialty care and hindering communication between providers. This study aims to improve the referral process with interventions targeting high priority failures identified by an FMEA analysis. This approach may represent a model for all health systems striving to improve their referrals process.

Costs of Hypertension Detection Within the NHS Health Check Programme Compared to Opportunistic Detection

Christopher Elles Clark, PhD; Jayne Fordham; Maria Greenwood; Suzanne H Richards; John L. Campbell, MD


The NHS Health Check Programme (NHSHC) in England offers 5 yearly assessments to adults free of vascular disease. The Programme’s evidence base has been questioned, and its cost-effectiveness model assumes 75% uptake. Costs to primary care in identifying new cases of hypertension are unknown.

Objective: To estimate costs per new diagnosis of hypertension detected by the NHSHC Programme compared to opportunistic diagnosis.

Design: Prospective audit with cost estimates.

Setting: One rural practice in Devon, England

Patients: Eligible adults aged 40 to 74 years, identified from the practice database.

A 30 minute nurse-led appointment, delivering targeted health advice based on lifestyle history and clinical measurements. Blood pressure (BP) was recorded as the mean of three readings with an automated sphygmomanometer. Patients with mean blood pressure >140/90 were followed up by nurses and doctors according to NICE hypertension diagnostic guidelines. We used published UK consultation costs to estimate costs incurred in confirming a new diagnosis of hypertension, and compared these with estimated costs of detecting the same cases with an opportunistic approach.

Main outcome measures: Costs incurred in confirming new diagnoses of hypertension.

In the first six months 198 patients were invited, 61 (31%) attended, and 19 (31%) required follow up of initial elevated BP; of these five (26%) did not re-attend and hypertension was confirmed in four (21%). Cost per diagnosis through NHSHCs was £674.59 ($1,133.32); an additional cost of £381.30 ($640.58) per case compared to opportunistic detection.

Practice NHSHC uptake is low, but consistent with county-wide uptake (34%). Costs of diagnosing hypertension with NHSHCs are double those for an opportunistic approach. Replication of this finding from larger numbers of practices may challenge one of the cost-effectiveness assumptions of the NHSHC Programme. Primary care resources may be better utilised with opportunistic approaches to the detection of hypertension.

PR1 Designing, Conducting and Reporting Mixed Studies Reviews: Introduction, Review Questions and Synthesis Designs

Pierre Pluye, MD, PhD; Isabelle Vedel, MD, PhD; Quan Nha Hong

11/21/14 9:00 AM - 5:00 PM Wilder

Mixed studies reviews include qualitative, quantitative and mixed methods studies. This form of literature review is becoming popular as it provides a rich understanding of complex health interventions and programs. OBJECTIVES After attending this workshop, participants will: (i) understand the basics of designing mixed studies reviews, (ii) develop rigorous mixed studies review questions appropriate to their topics, (iii) identify appropriate resources for designing such reviews, (iv) distinguish the types of synthesis designs, (v) choose the synthesis design appropriate to their questions, and (vi) apply a template for reporting their mixed studies review. CONTENT Morning Session • First, workshop leaders will provide an overview of mixed studies reviews followed by an open discussion. The overview will cover specific aspects of mixed studies reviews with respect to the traditional seven steps of systematic reviews: question, eligibility criteria, sources, identification, selection, appraisal, and synthesis. • Second, in small groups, participants will formulate a mixed studies review question on their topic of interest and discuss with the other participants for refinement. • Third, workshop leaders will provide two major types of synthesis designs (sequential designs) and examples to stimulate discussion. Afternoon Session • First, workshop leaders will provide two other major types of synthesis designs (convergent designs) and examples to stimulate discussion. • Second, in small groups, participants will choose one synthesis design that best fits their review questions. • The workshop will end with a mini “poster session” where participants will present their work to their peers and provide feedback to other participants. PREREQUISITE KNOWLEDGE Previous research experience with at least qualitative and quantitative research, or mixed methods research, is advisable. The workshop material is available in a wiki-toolkit designed by the workshop leaders (http://toolkit4mixedstudiesreviews.pbworks.com).

PR2 International Collaboration in Innovating Health Systems

Chris van Weel, MD, PhD; Claire Jackson; Deborah Turnbull; Robert L. Phillips Jr, MD, MSPH; Terry Findlay; Robyn Tamblyn; Felicity A Goodyear-Smith, MD, MBChB; Cindy Lam; Anne MacFarlane; Barbara van der Linden; Evelyn M van Weel-Baumgarten; Emma Whitehead

11/21/14 9:00 AM - 5:00 PM O'Neill

Background and aim Health systems differ around the world but face comparable problems in their performance: the challenge to meet ever more complex health needs of diverse populations; the limitations of the disease model; rising health care costs with limited returns in outcome. The differences in health systems present at the same time a natural variation to study and compare the impact of interventions to innovate the responsiveness of health systems and improve population health. This is the rationale to pursue international collaboration and this is the mission of the International Implementation Research Network in Primary Care (IIRNPC) that provides the faculty of this preconference. Expected Outcomes The preconference is to strengthen international collaboration in implementation research and to plan concerted actions. It provides cutting-edge international experiences in implementation of systems’ change, for researchers, (primary health care) professionals, policy makers and patients/service users. It provides opportunities to identify best practices and models of success for implementation research. Target group The preconference aims to bring together primary care implementation researchers; policy makers and research funders; primary care professionals; patients-health care users. Program The Progam consists of four sections in which the participants engage with leading experts in the field to discuss the priorities for international primary care collaboration in strengthening national and regional health systems: 1. Showcases of innovation in the delivery of primary care (care for patients with complex diabetes mellitus; engaging the most difficult to reach – (illegal) migrants – in primary care). 2. Funding of international studies in implementation of primary care: the priorities of funding agencies (Commonwealth Fund, Canadian Institute Health Research, AHRQ, Health Research The Netherlands). 3. Analysing health care systems and how international collaboration can support national (primary) health care development (analysis of the Mexican and Hong Kong health care systems) 4. International networking and collaboration for the innovation of healthcare delivery: planning further concerted actions. Methods The pre-conference will use a number of methods to engage and inform the participants: Short presentations, round table, panel - and plenary discussions, small group discussions, SWOT analysis, priority setting.

PR3 Realist Methodology for Primary Healthcare Research: What Works, for Whom, Under What Circumstances and How?

Justin Jagosh; Paula Bush; Jon Salsberg, MA; Geoff Wong

11/21/14 1:30 PM - 4:30 PM Empire/Hudson 

Realist methodology is a theory-driven approach to understanding the complex array of factors that produce intended and unintended outcomes in health and social interventions. In this pre-conference workshop, participants will explore the advantages and challenges of using the methodology in evaluating or synthesizing evidence in primary healthcare - to move past the question of ‘was it successful?’ to better understanding how, for whom, and under what circumstances interventions produce their particular outcomes.

P175 An Exploratory Study of the Incentives and Disincentives Influencing Quality of Care for Depression and Anxiety in Ontario Family Health Teams: Physicians’ Perspectives

Rachelle Ashcroft, PhD

11/22/14 9:00 AM - 10:40 AM Carnegie/Lyceum

Context This poster presentation presents findings of a qualitative pilot study that examines incentives and disincentives for the treatment of depression and anxiety in Ontario Family Health Teams. There is widespread support that treatment for depression and anxiety be included in primary care settings like Ontario’s Family Health Teams. Despite benefits, the inclusion of care for depression and anxiety in primary care settings - including Family Health Teams - remains modest. This poster reports on findings of a qualitative pilot study investigating incentives and disincentives for the treatment of depression and anxiety in Ontario’s Family Health Teams. Participants identified disincentives that act as barriers, and incentives that help to implement, quality treatment of depression and anxiety in a collaborative primary care setting. Objective The purpose of this study is to identify the disincentives acting as barriers, and incentives that help implement, quality treatment of depression and anxiety in the Family Health Team collaborative care setting. The research question guiding this study is: What are the incentives and disincentives for treating depression and anxiety in Ontario Family Health Teams. Design This project takes an exploratory qualitative approach to identify physicians’ perspectives of the incentives and disincentives for treating depression and anxiety in Ontario Family Health Teams. Participants The sample for this study is primary care physicians in Family Health Teams located in Toronto, Ontario. Semi-structured interviews were conducted with ten primary care physicians from July – August, 2013. Results Findings include ten key themes that may influence quality care of depression and anxiety: clinical indicators, physician attributes, education and training, physician remuneration, patient attributes, physician-patient relationship, psychiatry, interdisciplinary health professionals in Family Health Teams, community based resources, and leadership. Conclusion Findings contribute to knowledge that helps to strengthen collaborative strategies in primary care for the treatment of depression and anxiety. Understanding that there are different types of incentives and disincentives that influence treatment for anxiety and depression in primary care may help service planners who are trying to promote improved mental health care.

P101 Assessing Statin Guidelines With Benefits Harm Trade-Off Interviews

Bruce Barrett, MD, PhD; Jason A. Ricco, MD, MPH; Margaret Wallace, MSc, PharmD; David Kiefer; David Rakel, MD

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Introduction: The 2013 guidelines from the American Heart Association and American College of Cardiology endorse statins when the 10-year cardiovascular event (heart attack + stroke) risk is as low as 7.5%. This represents a major change from previous guidelines, which endorsed treatment at ≥20% risk. Best evidence suggests that statins might benefit patients by reducing 10-year event by 21% (ie. from 20% to about 15.8%, or from 7.5% risk to about 5.9%). Potential harms of statins include myopathy, hepatotoxicity or hyperglycemia, as well as costs and inconveniences of doctor visits, blood draws, and taking a pill each day. Objectives: To: 1) better understand how patients value potential benefits and harms of statins, 2) estimate degree-of-benefit required to justify statins for individuals and populations, and 3) assess how this relates to current guidelines and practice. Methods: Benefit harm trade-off interview methods will use evidence-based treatment scenarios to assess how much cardiovascular event risk reduction is needed to justify potential harms. These methods first describe expected benefits and harms in understandable language, then vary the degree of primary benefit until the respondent changes orientation from “Yes that strategy seems worthwhile for me” to “No that strategy does not seem worthwhile.” This patient-level metric is known as “smallest worthwhile effect” or “sufficiently important difference” (SIDi) which we have defined as “the smallest amount of patient-valued benefit that an intervention would need in order to justify associated costs, risks and other harms.” Averaging across individuals provides estimates of population-level (SIDpop) degree-of-benefit-required, both in terms of group norms and between-person variability. Conclusions: Clinical guidelines aiming for uniform implementation of standardized practice may be misguided, as they neglect the heterogeneity of patient-level values and preferences. Health policy, including clinical guidelines, should take account of the diversity of patients’ health values as they relate to medical decision-making.

P102 Relationship Between Grief and Chronic Pain in Continuity of Care Patient: A Pilot Study

Elise AG Duwe

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: The turn toward patient-centered care, while imperative, has further marginalized patients that present a problem to the system, such as chronic pain sufferers. A deeper understanding of the lived experience for patients labeled in the system as problematic is necessary. Objective: To achieve a greater understanding of the causes of pain disparities and issues related to chronic pain care, in particular links between grief experienced and unexplained pain. Design: Life history interview with survey component. A pilot study for dissertation. Setting: Given the extended engagement, the study takes place in the community, at a primary care clinic, and during hospitalized care and outpatient procedures. Patient: The patient is a 75-year old woman with multiple medical problems, including fibromyalgia, managed by her primary care provider. She has lost four of her six children and survived multiple abusive relationships. Main and Secondary Outcome Measures: McGill pain questionnaire (Melzack, 1975); Historical Loss Scale (Whitbeck et al., 2004), Texas Revised Inventory of Grief (Faschingbauer, 1981), and Stressful Live Events Screening Questionnaire (Goodman et al., 1998) to measure trauma; Resilience Scale (Wagnild & Young, 1987). Results: The patient did not conceive of her physical pain as related to her grief—a separate pain—until gaining from the McGill pain questionnaire similar words to describe both physical pain and grief. Through connecting her pain and grief, the patient now feels empowered to work through her grief in order to reduce her physical pain. Conclusions: The conclusions open toward more questions regarding how people embody social suffering and how to foster healing through addressing suffering embodied as very real and very felt chronic pain. This pilot informs the dissertation, which seeks to understand the experience of chronic pain, including causal attributions and healing practices, for American Indians/Alaska Natives living off-reservation.

P103 Patient Centered Approaches to Brokering Substance Abuse Treatment Services in Primary Care: The Untapped Potential of Harm Reduction Models

Heather Sophia Lee; Shawna V. Hudson, PhD; Denalee OMalley, MSW

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: Primary care plays a pivotal role in the identification and treatment of key medical issues, yet substance abuse issues are often unaddressed. To develop substance abuse approaches that are patient centered in primary care, an understanding of harm reduction strategies, which prioritize patient engagement over the outcome of drug abstinence, is needed. Objective: The objective of this paper is to explore the way in which practitioners define and use harm reduction strategies in two urban, community based settings and describe their potential application in primary care. Design: A qualitative, in-depth interview study of 18 staff members (i.e., case managers, social workers, psychologists, and psychiatrists) at two harm reduction programs with a grounded theory analysis were conducted. Setting: Interviews were conducted at an urban, community-based, drop-in center for homeless active substance users and at a private practice site. Intervention/instrument (as pertinent): Interview guides were designed to elicit descriptions of the implementation of the harm reduction model and experiences in the implementation of these programs. Interviews were digitally recorded and transcribed. Results: Harm reduction workers articulated harm reduction in nuanced ways which clustered into two categories. It is characterized by 1) a low-threshold service delivery model (i.e., where anyone desiring services may access it regardless of their commitment to abstinence) and 2) an attitudinal stance of the provider toward the consumer which is compassionate, non-judgmental, meeting the client/patient “where they are”. Conclusions: Harm Reduction offers primary care providers a framework for the engagement of patients in early intervention around substance use issues which may greatly reduce the stigmatization of substance users in this setting increasing access to healthcare for patients, particularly those in Federally Qualified Health Centers.

P104 A Comparison of Patient Recall and Analysis of Verbal Exchanges During PC Encounters Using MEDICODE

Marie-Thérèse MT Lussier, MD, MSc; Claude Richard

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: In order for patients to adhere to advice, provided by primary care (PC) physicians, they need to recall it afterwards. However, several studies have shown that most patients do not fully understand or memorize it. Objective: To describe patient recall about exchanges on hypertension, diabetes, cholesterol and their related medications during PC consultations. Design: Secondary analysis of communication and patient recall data derived from an RCT. Sites: 9 PC clinics, Ontario (Canada) Participants: 221 adults with one or more of three chronic problems (hypertension, diabetes, and cholesterol) and 19 PC providers. Intervention: Educational training to increase patient participation. Instruments: MEDICODE, a validated coding instrument, was used to code the verbal exchanges on the selected chronic problems and related medications. Patients completed an exit questionnaire on their recall of discussions during the audiotaped encounter. Outcomes: Recall was evaluated as a % agreement by comparing discussions coded with MEDICODE to patients' responses on the recall questionnaire. Results: The following are descriptive results for all the participants pooled together since we found no significant differences between treatment groups. There are four possible situations when comparing the discussions coded from audiotapes with patient reports of discussions on the recall questionnaire: Medicode present, Patient missing; Medicode present, Patient present; Medicode missing, Patient present. Medicode missing, Patient missing. The rate of accurate recall (Medicode present-patient present) for all medications (Rx) taken together is 49,9%, 52,6% for hypertension Rx, 43,8% for diabetes Rx and 52,3% for cholesterol Rx. The rate of accurate recall for all problems together is 57,1%, 72,7% for hypertension, 53,9% for diabetes and 44,5% for cholesterol. Conclusion: The rate of accurate recall varies according to the various chronic problems and treatments discussed. These results are more encouraging that results previously published. Different factors to explain these findings will be presented at the time of the conference.

P105 Development and Validation of a Scale of Perception of Origin of Medically Unexplained Physical Symptoms

Jose Ramirez; Monica Morales-Ramirez; Teofilo Garza-Elizondo; Marco V Gómez-Meza; Oralia Castillo-Guzman; Raul F Gutierrez-Herrera; Azucena M Rodriguez-Gonzalez; Natalia Loskutova

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context. There are no reported scales that reveal the patient´s perception of the origin of Medically Unexplained Physical Symptoms (MUS). According to Goldberg´s Reallocation technique, the treating physician needs to establish an association between the symptom and the stressful event as a therapeutic approach strategy. Objective. Develop and validate a scale to measure the attribution of the origin of symptoms of patients with MUS. Design. Comparative cross-sectional study. Site. Outpatient clinic. Patients. We included 400 patients older than 18 years in two groups: a) 200 diagnosed with MUS and b) 200 with an established organic diagnosis. We excluded patients in psychiatric or psychosocial treatment with a family doctor and with substance abuse. Instrument. A 18-item Likert scale of "Perception of the origin of the symptom" with content validity generated by interview with 25 psychiatric patients and 25 patients from other specialties and a consensus of 7 specialists in psychiatry, psychology and family medicine was designed. Construct validity was determined by principal component analysis. The project was approved by the ethics and research committee of the university center. Results. The scale with 18, 15 and 12 items was analyzed. The total variance explained was greater with 12 items (55.06), with 48 of 66 statistically significant correlations. Bartlett´s sphericity contrast was also highly significant (c2 = 1720.4, df = 66, p <.001), KMO index = .899; Cronbach's alpha was .953. Conclusions. The 12-item scale of psychosocial perception is suitable to be used as an instrument to facilitate the clinical management of MUS psychometric properties. Studies are needed to verify its usefulness.

P106 Effectiveness of an Intervention to Improve Collaboration Between Family Physicians and the Oncology Team in the Follow-Up of Lung Cancer Patients

Lucie Vézina; Michele Aubin, MD, PhD; René Verreault; Lise Fillion; Éveline Hudon; Sébastien Simard; André Tourigny; Serge Dumont; Serge Daneault; Yves Lacasse; Audrey Samson

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: During the cancer treatment phase, family physicians (FP) are often left out of the follow-up of their patients and communication with the oncology team (OT) is frequently suboptimal. No formal protocol of information exchange between primary and secondary/tertiary care is systematically used in usual practice in Quebec (Canada). Objective: To assess the effectiveness of an intervention on FPs’ perceived continuity of care and collaboration with the OT. Design: Randomized controlled trial. Participants: 152 community-based FPs (86.4% participation rate) of patients recently diagnosed with lung cancer exposed (N=78) or not (N=74) to the intervention and surveyed at baseline and at the end of the study (T2). Intervention: 1) systematic appointments with FP; 2) transmission to FP of standardized summaries of patients’ condition; 3) transmission to the OT of patients’ information resulting from FP visits; 4) priority access to FP for cancer patients. Main outcome measures: Perception of continuity of care and of collaboration between FPs and the OT. Results: Following the implementation of the intervention, FPs from the experimental group reported a better perception of collaboration with the OT (p=0.0006). A larger proportion of this group indicated being involved in shared care (57% vs 36%) and fewer in sequential care (9% vs 26%), compared to the control group (p=0.01). In both groups, FPs involved in shared care reported less frequently problems of informational continuity (p=0.008). A slight reduction in the proportion of FPs who perceived problems of informational and management continuity was observed at T2, but this improvement was found in both groups (p>0.05). Conclusion: This intervention seems to positively influence FPs’ perception of collaboration with the OT in the follow-up of their patients with lung cancer. However, its effect appears modest particularly on informational continuity. Further strategies should be developed to strengthen FPs and OT collaboration in cancer care.

P107 First Nations, Inuit, and Métis Experiences of Cancer and Healing: Toward a Dialogue Between Patient and Health Care Professional Perspectives

Chad Hammond; Roanne Thomas

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: First Nations, Inuit, and Métis (FNIM) peoples may have distinct experiences of cancer that require attention to a wide range of factors (e.g., cultural understandings of healing, social determinants of health, access and continuity of care, etc.). The current research in progress explores the complex challenges and opportunities for improving care (including primary care) among FNIM peoples. Objectives: 1) Develop knowledge of FNIM cancer survivorship within Canada; 2) Identify barriers and facilitators of healing from the perspectives of FNIM cancer patients and health care professionals; 3) Work with community partners and stakeholders to translate this knowledge into culturally safe health services. Design: My postdoctoral research project builds upon a “parent” study of FNIM women’s cancer survivorship through arts-based methodologies including photovoice and story journaling. The findings from these photos and stories will be the basis for semi-structured interviews with health care professionals who work with FNIM peoples with cancer. The current study uses critical phenomenology, an approach that situates experiences of illness and healing within larger social processes and power dynamics in health care settings. Setting: Participants will be recruited from across Canada. Participants: 20 health care professionals from primary care and oncology who work regularly with FNIM patients. Anticipated Results: Patients and health care professionals will provide important insights into the current status of FNIM cancer care. Based on these insights, knowledge translation strategies that honour these diverse perspectives and needs will be developed. Conclusions: A report from this research will be disseminated to participants and project collaborators for their feedback on the next steps toward translating this knowledge into health services.

P108 The Role of Primary Care in Diagnosis of Cancer via Emergency Presentation: A Qualitative Synthesis of Significant Event Audits

Elizabeth Mitchell, PhD, BA; Greg Rubin; Una Macleod, PhD, FRCGP

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

CONTEXT: Patients diagnosed with cancer in the context of an emergency presentation to hospital have been shown to have poorer outcomes. It is assumed that such patients present to the Emergency Department without seeing their GP, but little work has been done to identify primary care involvement prior to hospital attendance. Significant Event Audit (SEA) is a quality improvement technique widely used within UK Family Practice. OBJECTIVE: To gain insights into the primary care diagnostic pathway for patients diagnosed with cancer as a result of an emergency presentation, through secondary analysis of SEA documents. DESIGN: Participating family practices were asked to identify their last patient diagnosed as an emergency and to complete an SEA report. Documented accounts were synthesised and analysed using qualitative methods, including development of an Interpretative Matrix to facilitate identification and interpretation of common and diverse aspects of presenting features and pathways of care. SETTING: UK Family Practice. PARTICIPANTS: General practitioners in London and the north of England. RESULTS: SEAs for 122 patients were analysed. In most cases, patients had contact with their practice before diagnosis, primarily in the period immediately prior to admission. In only ten cases could we establish that there had been no input from primary care. Review of those accounts where there was protracted primary care contact generally demonstrated complexity, often related to elderly patients. Practitioners identified learning points around presentation and diagnosis, consultation activity and safety-netting, system issues and communication, co-existing patient factors, and referral guidelines. CONCLUSIONS: We have demonstrated extensive primary care input into the pathway to diagnosis of this patient group. These findings are timely in view of recent work suggesting that a quarter of all cancers are diagnosed through emergency presentation, and contradict the prevailing view that many patients are reluctant to consult their GP when they become ill.

P109 US-Mexico Border Primary Care Attendees’ Attitudes Towards Self-Sampling for HPV Infection for Cervical Cancer Screening

Eribeth K. Penaranda, MD; Jennifer C. Molokwu, MD, MPH; Navkiran Shokar, MA, MD, MPH

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: Women living on the US-Mexico border have one of the highest mortality rates for cervical cancer in the US. Testing for Human Papilloma Virus (HPV) infection has been approved by the FDA for primary screening for cervical cancer. HPV self-sampling has been proposed as an alternative to physician testing for s cervical cancer screening; however, there is lack of information about the feasibility of using this test among women in the US. Objective: To describe US-Mexico border women’s attitudes, preferences and intention to use self-sampling for HPV after receiving a promotora (community health worker)-led educational intervention about self-sampling and completing self-sampling for HPV. Design: cross-sectional study. Setting: Two University- based Family Medicine clinics in El Paso, Texas. Participants: 100 women aged 30 – 65. Intervention: An educational intervention delivered by a lay health worker targeting low literate Spanish or English speaking women. Topics addressed include information about HPV and self-sampling, test accuracy and instructions on how to collect an HPV sample. Following the intervention, participants will perform HPV self-sampling using an FDA-approved assay. Outcome: Participants’ knowledge, beliefs, attitudes and intentions to use self-sampling will be assessed with a quantitative survey. Women’s attitudes towards each test determined by constructs from the Health Belief Model such as benefits, barriers, susceptibility, and self-efficacy will be measured on a Likert scale and scored. Analysis: We will describe the sample and distribution of responses on attitudes, test characteristics, test preferences, and intention to use self-sampling using descriptive statistics. Test characteristics will be rated on a five-point Likert scale and the scores on the self-sampling and Pap smear will be compared with McNemar’s test. Logistic regression will be used to determine if there are socio-demographic or attitudinal factors related to test preference. Results and Conclusions: Pending. 20 participants have completed self-sampling for HPV infection.

P110 Comparison of Group vs Individual Education to Promote Cervical Cancer Screening Uptake: A Randomised Controlled Trial

Navkiran Shokar, MA, MD, MPH; Theresa Byrd; Alok Dwivedi; Jessica Calderon-Mora; Jennifer C. Molokwu, MD, MPH; Eribeth K. Penaranda, MD; Rebekah Salaiz

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Background: In Texas, Hispanic women have the highest incidence of cervical cancer and are 1.9 times more likely to die from it than non-Hispanic Whites; mortality rates are highest for Hispanic women on the US-Mexico border. De Casa en Casa is a community-based cervical cancer screening program providing outreach, education, no-cost screening and navigation services to eligible women in two US-Mexico border counties. Although group education has the potential for cost and time efficiencies, it is unclear whether it is as effective as individual education in improving cervical cancer screening. Objective: To compare the effectiveness of group vs individual education in promoting cervical cancer screening. Design: Randomized controlled trial Participants: Inclusion criteria: 21-65 year old uninsured women due for cervical cancer screening recruited for the program from community and clinic sites in two predominantly Hispanic Texas border counties. Intervention: Consenting program participants will be randomly assigned to group or individual (control) education and will complete a baseline, post-education and 4 month follow up validated survey with items covering sociodemographics, cervical cancer knowledge and beliefs and screening status. Main outcome measures: Uptake of cervical cancer screening at 4 months post intervention by self report and program database. Analysis: Assuming a control group screening rate of 75%, a total sample size of 150 per group achieves more than 80% power to detect equivalence of the two intervention groups. The margin of equivalence, given in terms of the difference, extends from -0.15 to 0.15. The calculation is based on two one sided z-tests with an alpha of 0.05. A Z-test will compare differences in screening proportions between groups. The 95% confidence interval of differences in screening proportions will be compared with equivalence margins. Multiple linear regression will be performed to determine significant changes in scores by intervention groups after adjusting for other cofactors.

P111 Feasibility Trial of a Mammography Decision Support and Navigation Intervention for Formerly Homeless Women With Serious Mental Illness

Lara Carson Weinstein, MD, MPH; Katelyn Hurley, MPH; Marianna LaNoue, PhD; Randa D. Sifri, MD; Ronald Myers

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: Women with serious mental illness (SMI) are at increased risk for breast cancer, have limited access to breast cancer screening, have lower rates of mammography, and, have higher rates of breast cancer morbidity and mortality, as compared to the general population. Objective: To determine the feasibility and acceptability of a computer based decision support and tailored navigation intervention to encourage formerly homeless women with SMI to receive a mammogram. Design: one group pre-post study design Setting: 2 supportive housing programs for formerly homeless people with SMI Participants: Up to 30 formerly homeless women with SMI age 40 and over are currently being recruited Intervention: The intervention consists of 2 phases: a decision support phase and a navigation phase. In the decision support phase, participants meet with a trained decision counselor for a 60-90 minute computer-based educational and decision support session. At the conclusion of the session, the decision counselor contacts the participants and their caseworkers on a monthly basis to assist them in logistical issues related to scheduling and going to a mammogram. Outcome Measures: The primary outcome measure for the study is receipt of a mammogram. Secondary outcome measures include breast cancer screening knowledge and attitudes and decisional conflict. Process measures will also be analyzed. Results: We hypothesize that the intervention will increase mammography participation, improve breast cancer screening knowledge and attitudes, and reduce decisional conflict Conclusions: Behavioral interventions used to increase breast cancer screening among women in the general population, such has reminder letters, telephone counseling, or provider prompts, have limited utility among formerly homeless women with SMI because of problems receiving mail, maintaining consistent phone numbers and navigating the healthcare system. There is a need to identify an effective intervention strategy to facilitate mammography screening in formerly homeless women with SMI.

P112 Markers of Atherosclerosis in Relation to Presence and Progression of Knee Osteoarthritis: The Rotterdam Study

Theun Antonius Hoeven, MD, MSc; Sita Bierma-Zeinstra; Oscar Franco; Albert Hofman; Arfan Ikram; Maryam Kavousi; Joyce van Meurs; Patrick J.E. Bindels, MD, PhD

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context Several observational studies have found an association between atherosclerosis and osteoarthritis (OA) of the hands and knees, predominantly among women. However, the reported cross-sectional associations were modest in effect size and generally attenuated after adjustment for cardiovascular risk factors. Hence, it remains unclear whether atherosclerosis and OA are related or whether they are simply independent disorders with shared risk factors. Objectives To investigate the association between markers of atherosclerosis and presence and progression of knee osteoarthritis. Design Prospective population-based cohort study. Setting The general population of a suburb in the city of Rotterdam, The Netherlands. Participants Markers of atherosclerosis included coronary artery calcification (CAC), and plasma levels of CD40L, VCAM-1, and VEG-f. CAC data was available for 1,669 participants, CD40L, VCAM-1, and VEG-f for 975. Radiographs of the knee were scored with the Kellgren & Lawrence score for osteoarthritis at baseline and follow-up (average follow-up time 4.5 years (±0.5)). Main outcome measures Multivariate logistic regression models were used to calculate odds ratios (95% confidence intervals). Results: Mean age was 73.1 (±7.5) years. Within the study population, 18% had radiographic knee OA, (11% of men, and 23% of women). CAC and VEG-f were not associated with prevalent knee OA. Only among women, CD40L (adjusted odds ratio (aOR) 1.3 (1.1 to 1.6)) and VCAM-1 (aOR 1.3 (1.1 to 1.6)) were associated with prevalent knee OA. No associations with progression were found in women. In men, too few progressors were available to assess associations. Conclusions: In this population-based study, CAC and VEG-f were not associated with presence or progression of knee OA. Only among women, plasma levels of CD40L and VCAM-1 were higher in individuals with knee OA compared to those without knee OA. This might suggest an association between atherosclerosis and knee OA through low-grade systemic inflammation in women.

P113 Findings From Five Years of Electrocardiograph Screening for Division I Student Athletes

Andrew Pasternak, MD, MS; Colin Fuller; Carol Scott; Cheryl Hug-English

11/22/14 9:30 AM - 10:40 AM Astor Ballroom

Context: Screening young athletes by electrocardiogram (ECG) for underlying cardiac issues continues to be controversial in the USA. While sudden cardiac death garners much attention, there is concern about the number of false positives and subsquent testing with ECG screening. Objective : Determine the percentage of athletes with an abnormal ECG at at Division 1 University and address feasibility issues related to a screening program. Design : Prospective descriptive cohort study of incoming student athletes Setting: University of Nevada-Reno Sports Medicine Program Patients: 874 student athletes from June 2008- September 2013 Interventions: Screening ECG performed at the time of their pre-participation exam (PPE) interpreted by board-certified cardiologists with additional expertise in ECG screening. Electrocardiograms were retrospectively re-evaluated in 2014 by a single cardiologist. Outcomes : Abnormal ECG findings warranting further cardiac evaluation. An abnormal ECG was defined using Italian criteria from 2008-July 2011 and Stanford criteria from Aug 2011-2013 Results- 82 (9.3%) athletes were found to have abnormal electrocardiograms requiring additional evaluation. Of these, two athletes (one male, one female) were diagnosed with hypertrophic cardiomyopathy excluding them from competition and requiring cardiology intervention. During the study phase using Italian criteria, 55/523 athletes (10.5%) required further testing versus 27/351 (7.6%) during the Stanford criteria phase. Applying Stanford criteria to athletes initially analyzed using the Italian criteria would have reduced the number of athletes needing further evaluation from 55 to 33. In retrospective analysis, 18 athletes were found who met Stanford criteria for further testing but were missed at the time of their PPE. Conclusions: One in 437 athletes were found to have potentially life threatening abnormalities and not cleared for competition. Applying Stanford criteria lowered the false positive rate compared to Italian criteria. Successful screening programs will require physicians with rigorous training specific to the interpretation of young athletes’ electrocardiograms.