2011 NAPCRG Annual Meeting Presentations with DFM Faculty Members as Presenters, Co-Presenters or Co-Authors

12 Sessions

P046 R U 4 PC?: Texting and Feedback on Primary Care During Medical School
Andrea Wudyka, MD; Diane Levine, MD; Joel Heidelbaugh, MD; Elie Mulhem, MD; Victoria Neale, PhD; Andrea Wendling, MD; Christopher Morley, PhD; Julie Phillips, MD, MPH

Context: Medical students are often asked which field they plan to enter, and teaching physicians routinely offer feedback on this complex and important decision. Additional data regarding this aspect of the "hidden curriculum" are needed to increase knowledge to be used to attract and retain student interest in primary care. Objective: To gain understanding of incidence and balance of positive and negative feedback on primary care as a career choice, and secondarily, to determine efficacy of texting as a data collection method for medical students. Human Subjects Review: WSU IRB approval for pilot portion (10/2010); expecting multicenter IRB approval before 11/2011. Design: Multicenter observational prospective trial. Cohorts of medical students from 4 universities will collect data for a 30-day period by sending a text message whenever a comment regarding primary care is heard in the context of formal educational activities, and will complete pre/post surveys regarding demographics, career plans, and attitudes toward primary care. Setting: Pre-doctoral curriculum at four medical schools. Participants: Medical students at participating universities (Wayne State, Michigan State, University of Michigan, SUNY). Main and Secondary Outcome Measures: Raw number of negative and positive messages heard; ratio of positive-to-negative comments; effects of student characteristics upon messages received; themes derived from qualitative content of messages; compliance/satisfaction with text messaging. Anticipated Results: Based upon pilot data, we anticipate most feedback to be reported during third year; balance of positive to negative feedback will vary by several factors. Conclusions: Attracting and retaining interest in primary care is essential in medical education today. Aspects of the "hidden curriculum" have been described as important in shaping career decisions of medical students, and feedback from teaching physicians has been named as an area in need of further work. This study aims to contribute concrete data to increase understanding of this piece of medical education.

P064 Factors Influencing Participation in Ilinet: A Public Health/Primary Care Surveillance Program for Influenza
Christopher Morley, PhD, MA, CAS; Buckley Brian; Natalie Jones, BS; John Epling, MD, MSEd; Donna Gowie, AAS

Context: The U.S. Centers for Disease Control (CDC) operates the Outpatient Influenza-like Illness Surveillance Network (ILINet), a collaborative effort with states and primary care providers who regularly report cases of influenze-like illness on a weekly basis. While ILINet in New York State has always maintained an adequate proportion of ILINet providers to population, gaps in geographic and demographic coverage are present. Human Subjects Review: The study was granted an exemption from review by the SUNY Upstate Medical University IRB. Design: Cross-sectional survey of physicians. Setting: Statewide influenza surveillance program, relying upon physician volunteers. Patients or Other Participants: Initial cohort of physician respondents recruited SALT-Net practice-based research network and through the ILINet provider roster.Additional respondents will be sought through partnerships with other practice-based research networks. Main and Secondary Outcome Measures: Initial results were analyzed descriptively, and a bivariate logistic regression model was created with "Participation in ILINet " serving as the outcome variable. Qualitative comments were also collected. Anticipated Results: We expect additional responses to increase the significance of our findings from the initial survey dissemination, to which 86 providers responded. Endorsement of the validity of ILI surveillance was the most significant predictor of participation (O.R.=9.926, p=.004). Years in practice (O.R.=1.097, p=.018) and possession of a master's degree (O.R.=4.346, p=.049) were also significant factors. A time usage and compensation factor was nominally significant (O.R.=1.868, p=.051). Qualitative comments generally endorsed ILI surveillance as a good method of influenza monitoring, with some citing ease of use as an incentive. Conclusions: We expect to identify factors which incentivize participation in surveillance programs like ILINet, as well as barriers to physician participation.

P187 Antepartum Depression Screening Effectiveness in Indentifying At-risk Women
Kristin Rutherford, MS, FNP-C; Christopher Morley, PhD, MA, CAS; Abigail Beaudette; Virginia Young, MLS

Objective: To determine the effectiveness of antepartum depression screening of pregnant women in identifying those at risk for postpartum depression. Design: A systematic search of the current literature on antepartum screening for postpartum depression. A medical librarian conducted searches in Medline, PubMed CINAHL, Cochrane, PsycINFO, and Scopus. Search terms in the databases which utilize MeSH subject headings (Cochrane, Medline, PubMed, PsycINFO) were "Depression, Postpartum" [Mesh], "Depression, Postpartum/prevention and control" [Mesh], "Prenatal care" [Mesh] combined with the keywords "screening" or "diagnosis." In the databases which do not use MeSH subject headings (CINAHL and Scopus) the keywords used were "prenatal care" and "postpartum depression" and "screening." Only English language studies were selected. Outcomes: ROC curve and summary statistics (AUC) for sensitivity & specificity of tests, particulary Edinburgh Postnatal Depression Scale. Qualitative assessment of the findings will be presented as well. Early/Expected Results: Preliminary findings suggest antepartum depression screening does in fact identify women at risk for developing postpartum depression. Multiple studies showed that screening women in the antepartum period is effective in detecting depression during pregnancy. Screening women during the antepartum period leads to early identification and treatment which in turn leads to less postpartum depression. Preliminary Conclusions: Antepartum depression screening of women to identify risk for postpartum depression appears to be effective and relatively efficient and easy to conduct, when using standardized instruments such as the Edinburgh Postnatal Depression Scale.

OP17 Impact of Prior Authorization Requirements in Primacy Care: A Salt-net Study
John Epling, MD, MSEd; Christopher Morley, PhD; Kathleen Barzee, MPH

Context: Insurance companies increasingly require prior authorization for various health-related services prescribed by primary care providers. Objective: What is the economic burden of insurance company prior authorization requirements on primary care practices in New York? Design: Pilot concurrent self-report single cohort observational study. Setting: Primary care practices in the Central and Northern New York region who are members of SALT-Net, a practice-based research network. Participants: Providers, nurses, and clerical staff in nine primary care practices. Methods: Baseline practice characteristics (setting, size, staffing, payer mix) were obtained from each practice. Each participant (provider, nurse, clerical) completed a data card for 25 consecutive prior authorization requests, or consecutive requests for one month, whichever occurred sooner, from insurance companies for a range of services. Data items collected include object of request, provider type, category of insurance, and total number of minutes spent on that request. Where ingle requests required activity by multiple participants, each participant recorded their time individually. Data collection occurred during Fall, 2010. Outcome Measures: Average hours per week for prior authorization requests were calculated for all practices, and stratified by recorder type, request type and insurance type. These figures were combined with US Bureau of Labor Statistics data for New York State to generate an average staffing cost per practice required to deal with prior authorization requests. Results: Roughly 434 individual authorization requests were recorded by participants in nine practices. Initial analyses indicate that practices spent an average of 2.8 staff hours per week with prior authorization requests, at an average staffing cost to the practice of $71 per week. Clerical staff and RNs spent the most time on these requests (1.3 and 0.7 hours/week respectively). Most time was spent on radiologic studies (1.4 hours/week), and medications (1.0 hours/week). Commercial insurance company requirements took the most time (1.7 hours/week), followed by Medicaid (0.4 hours/week) and commercial Pharmacy Benefits Managers (0.3 hours/week). Limitations to these results include incomplete reporting, estimations based on average cost data and relatively short data collection windows. Conclusions: Prior authorization requirements have definable monetary costs to primary care practices.

P220 Predictors of Interest in Future Global Health Practice Among First and Second Year Medical Students
Christopher Morley, PhD, MA, CAS; Ana Villarreal, BA; Carrie Roseamelia, MA

Objective: To identify predictive factors for interest in global health practice among early medical students. Design: Exploratory, descriptive study using a prospective survey. Setting: Year 1-2 Clinical Skills course at an allopathic medical school. Participants: We surveyed all first- and second-year students at the beginning of the 2010 academic year, as they entered a required two-year clinical skills course to measure attitudes toward careers in global health, rural practice, in primary care, and in other settings. The survey included Likert-scaled questions asking about future specialty and practice context preferences, attitudes toward global health practice, primary care, care for the underserved, and geographical location (i.e. rural vs. urban, etc.) of future practice, along with basic demographics. Outcome Measures: Principal Components Analysis (PCA) was utilized to identify 15 factors within the survey. Factor loadings were then modeled via stepwise OLS regression as predictors of interest in future global health practice. All statistical analyses were conducted in SPSS v.18.0.3. Results: We obtained a strong response rate (155/160 MSIs, and 141/160 MSIIs; 92.5% response rate overall). Interest in Urban Care (?=0.32 p<.001), Underserved Care (?=0.32 p<.001), Surgery (?=0.2 p<.001), Patient Communication (?=0.17 p=.002), Preventive Medicine (?=0.16 p=.003), Primary Care (?=0.15 p=.007), and in Advanced Training Opportunities (?=0.14 p=.013) each were significantly predictive of interest in future global health practice. Conversely, concerns about job security and income (?=-0.24 p<.001), as well as interest in Neurology and Psychiatry (?=-0.21 p<.001), were negatively associated with interest in future global health practice. Conclusions: Administrators of global health electives and educational programs may consider these factors in targeting students for recruitment. Conversely, programs may consider introducing global health curricular elements in order to attract students who are interested in primary care, preventive medicine, and in various other forms of underserved care.

P221 Predictors of Medical Student Interest in Rural Practice
Christopher Morley, PhD, MA, CAS; Carrie Roseamelia, MA; Andrea Manyon, MD

INTRODUCTION: Interest in a primary care career and rural upbringing have been shown to be predictors of future medical practice in rural areas. As part of a dedicated expansion of a Family Medicine-Based Rural Medical Scholars Program, we explored these factors and others in a cohort of MSIs and MSIIs as they began a clinical skills training program at SUNY Upstate Medical Unvierisity. Methods: We surveyed all first- and second-year students at SUNY Upstate Medical University at the beginning of academic year 2010, as they entered a required two-year clinical skills course to measure attitudes toward careers in rural practice, in primary care, and in other settings. The survey included Likert-scaled questions asking about future specialty and practice context preferences, attitudes toward primary care, care for the underserved, and geographical location (i.e. rural vs. urban, etc.) of future practice, along with basic demographics. We examined associations between interest in rural practice and predictor variables via OLS regression in SPSS v.18. RESULTS: We obtained a strong response rate (155/160 MSIs, and 141/160 MSIIs; 92.5% response rate overall). Rural origin measured by RUCA code (?=.105, p=.001) and early interest in a primary care career (?=.224, p<.001) were strongly associated with interest in rural practice. Additionally, African-American race (?=-.095, p=.016), concern about educational debt (?=.105, p=.001), and level of concern about the prestige of chosen specialty were inversely associated with interest in eventual rural practice (?=-.141, p=.003). Discussion: Consistent with existing literature, rural origin and interest in primary care were associated with eventual rural practice. These interests may be mitigated by concern over concerns about debt load and prestige.

P225 Evaluation of a Curriculum on Care for Patients With Disabilities: Effects on Medical Student Attitudes and Comfort Levels
Andrew Symons, MD, MS; Christopher Morley, PhD, MA, CAS; Denise McGuigan, MSEd; Elie Akl, MD, PhD

Objective: To evaluate the effect of a new curriculum teaching medical students to care for patients with disabilities. Design: Comparable cohort design, pre/post test. Setting: Two regionally proximal, comparable medical schools. School A implemented the new curriculum, incorporating it into required clinical skills training in the first two years of medical school, and School B maintained an existing curriculum. Participants: Medical students enrolled in required clinical skills courses in Fall, 2008 (pre-test). Students participating in Family Medicine clerkships were invited to respond to the post-test in the AY 2010-2011. Data were collected anonymously. Outcomes: Pre-test was used to establish group similarities, and post-test means were compared between groups via one-way Analysis of Variance (ANOVA). Results: There were virtually no statistical differences between the two student cohorts, in terms of age, gender distribution, or in responses to 18 pre-test and vignette-reaction questions, with two exceptions: students at the School B were in slightly higher agreement that they would "feel comfortable living next door to a person with an intellectual disability that lives by himself " (p=.049); and students at School A indicated that they had experiences similar to a clinical vignette presented in the survey and a slightly higher rate of agreement (p<.001). However, mean responses were statistically different on 15 items, including 13 at a p value of .001 or below. Out of the 15 categories where statistical differences were observed, 13 indicated an improvement in the intervention group relative to the control group, whereas 2 items seemed to be rated more favorably by controls. Conclusions: The intervention appears to have dramatically improved the attitudes and comfort levels of medical students in the provision of care for patients with disabilities. Additional follow-up studies include implementation of the curriculum at a second institution, and factor analysis of the instruments used.

BP24 Analysis of Narrative Responses to a Post-ISPE Self-assessment Completed by Family Medicine Clerkship Students
Christopher Morley, PhD, MA, CAS; Carin McAbee, BS; Andrea Manyon, MD

Objective: To assess psychometric, linguistic and qualitative differences in self-assessment comments between high and low scorers on a biopsychosocial Integrated Standardized Patient Examination (ISPE) that combined assessment of medical knowledge with communication, counseling, and clinical interviewing skills. Design: Descriptive, exploratory, mixed qualitative/quantitative study. Setting: Standardized Patient (SP) examination within a Family Medicine Clerkship. Participants: Third and fourth year medical students. Instruments: 1) A self-assessment module completed by students upon completion of their ISPE session containing qualitative comments about what they did well with, what they could improve upon, and an overall assessment of the SP's understanding. 2) The text of these comments were analyzed using the Linguistic Information and Word Count 007 (LIWC2007) software package, which utilizes a validated dictionary to assess narrative passages on 65 linguistic and psychological domains, based upon percentage of words used that are identified with each domain. 3) Students were graded using a 27-question instrument scoring instrument. 4) Group means on the 65 LIWC-inferred dimensions were compared between high scorers (+1 SD from mean; n=49) and low scorers (-1 SD from mean; n=45) via t-tests in SPSS v18.0.3. Outcomes: Statistically significant difference between high and low scorers on any of 65 measured domains. Results: High scorers had statistically significant larger scores in 11 domains, including overall word count (p=.009) and six other linguistic measures, as well as in four psychological domains: Anxiety (p=.043), words that ascribe causality (p=.010), words that indicate inhibition (p=.045), and words that are indicative of perceptual processes in action (p=.050). Conclusions: High and low scorers appear to be systematically different in a number of ways. Further research into whether such differences are due to temporary factors, such as differential responses towards tests or towards the FM clerkship, or whether such differences are more embedded, is warranted.

ET32 Medical Student Experiences With a New Integrated Standardized Patient Examination: Qualitative Analysis of Focus Group Data
Jennifer R. Flad, PhD; Christopher Morley, PhD, MA, CAS

Objective: To qualitatively gauge student reactions to a new, complex standardized patient examination as a follow-up to previously-reported mixed method evaluation results. Design: Focus groups held before and after SP examination. Transcripts of FG interviews were analyzed using immersion/crystallization methods. Setting: Single-station standardized patient examination featuring a complex patient, integrating biological, psychological, and social factors into one encounter, at the end of a Family Medicine Clerkship. Participants: Medical Students (MSIII's & IV's) enrolled in FM clerkship (N=41). Outcome Measures: Emerging themes from student/research focus group dialogues. Results: Three primary themes emerged from the focus group transcripts. 1. "Readiness for the SP encounter " (highly variable prior clinical experiences; lack of training in translating medical jargon into vernacular speech; a feeling among some but not all that the examination did not resemble experiences in the preceptor's office). 2. "Focus on the Grade " was evident (anxiety about video recording of the encounter, and about unpredictable SP's). Frustration about lack of feedback was also apparent. 3. "Scope of Practice. " Some students challenged conceptions of physician roles in helping patients address psychological and social confounders of health status. In spite of these challenges, many student comments served to validate and endorse the experience. Conclusions: Student responses indicate a lack of focus on biopsychosocial elements in their medical school training. A lack of opportunities to learn to explain diagnoses to patients in nontechnical language was noted as a deficiency in prior training. When students were asked how they learned to give a diagnosis and communicate with patients, many claimed this was done by observing a physician, and revealed a lack of standardization in their understanding of the role of physicians in these encounters. Also, communication-intensive SP-based examinations would benefit by the addition of prompt feedback with each student on strengths and weaknesses, create a learning opportunity.

P384 Health Ramifications of Displacement: Undocumented Migrants, Refugees, IDP's, and Stateless Persons
Lindsey Kingston, PhD; Christopher Morley, PhD, MA, CAS

Objective: To expand upon previous work focused upon stateless populations and health. Because globalized factors related to migration and health rights are often interconnected, this review addresses the following: (1) Undocumented immigrants, who are sometimes referred to as "economic refugees "; (2) Refugees, or people who have been forced to flee their homes and cross international borders; (3) Internally-displaced persons (IDPs), or people who have fled their homes but remain in their home country; and (4) Stateless persons who do not have legal nationality in any nation-state. In addition to providing a definitional background and illustrative examples for each group, this review also outlines the health challenges associated with each case and provides recommendations for protecting the "right to health. " Design: Narrative Review of the Literature. Results: International human rights frameworks outline a universal "right to health, " yet undocumented immigrants and displaced persons face unique health challenges that threaten this basic right. This threat is twofold: First, many are exposed to health risks due to their particular circumstances. Trauma associated with forced displacement and prolonged periods spent in camps, for instance, have negative impacts on the physical and mental health of refugees and internally displaced persons (IDPs). Second, these groups are often denied the medical services necessary for prevention and treatment. For example, undocumented immigrants and the stateless are often denied access to health care such as essential medicines and reproductive services. Conclusions: In order to protect the "right to health " of undocumented immigrants and the displaced, the medical community must be more aware of the specific needs and challenges associated with these groups. This is particularly important for those who interact with such groups, including primary, emergency, and public health practitioners.

DB34 Evaluation of a Chronic Care Model for Diabetes in a Rural Family Practice Network
Christopher Morley, PhD, MA, CAS; Alevtina Durmashkina, MD, PhD; Douglas Rahner, MD; Tricia Lyman, BA; Mary Alm, BS

Objective: Evaluation of clinical outcomes in diabetes patients following implementation of a chronic care treatment model in a 5-practice network. Design: Single-arm pre/post evaluation. Setting: Five-practice family medicine network in rural New York State. Participants: Patients with Diabetes Type I and Type II (N=432). Outcomes: Change in HbA1c, glucose, lipids, creatinine levels, BMI and BP measurements. Data were analyzed using the Statistical Package for Social Sciences (SPSS 18.0.0). Descriptive statistics, including means, standard deviations, and frequencies were calculated. Non-parametric paired t-tests were used to evaluate the differences between pre-test scores and post-test scores following the intervention. Spearman correlations were used to evaluate relationships of changes in HbA1c, glucose, lipids and creatinine levels as well as BMI and BP measurements following the intervention. Results: 392 individuals completed all pre-tests and 401 individual completed all post-tests, providing anthropometric measurements, blood samples for HbA1c, glucose and lipid profile. At pre-test, 161 participants (38.7%) had a higher than recommended HbA1c level (7.0%). There was no significant decrease in BMI, mean glucose and HbA1c levels in the general cohort. However, the subset of participants who had the pre-test HbA1c >7% had significant decreases in mean HbA1c from 8.72% to 8.3%, (P = 0.007), and those with high glucose (>126 mg/dL) had a mean decrease in glucose from 189 to 167.9 mg/dL (p<0.001). There was also significant improvement in lipid parameters of individuals who had clinically high cholesterol, triglycerides, and low HDL as well as improvements in BP. Conclusions: Although there was no control arm in this evaluation, the pre-test measurements were taken in patients who were already diagnosed and under a usual care standard of treatment. The improvements seen following implementation of the Chronic Care Model intervention at each site are therefore indicative of efficacy of this approach.

WS44 Qualitative Research Methods and Design: A Basic Primer/Review
Christopher Morley, PhD, MA, CAS; Jennifer Flad, PhD; Lindsey Kingston, PhD; Carrie Roseamelia, MA

Objectives of the session: When conducting and reporting primary care qualitative research, essential methodological concepts like sampling strategy, validity, sample characteristics, and coding methods are just as important to consider and address as in quantitative work. However, many qualitatively-based conference and journal submissions suffer from a lack of rigor. Issues include nebulous results, failure to present clear research questions, or disconnectedness between research questions, methods, and results. Content of the Session: This session will cover the basic elements of research design and reporting from a qualitative perspective, including: 1) Hypothesis generation and testing in qualitative research ; 2) Forming a Research Question, and selection of an appropriate method to answer it ; 3) Sampling - methods, sample characteristics, representativeness; 4) Methodological approach - Grounded Theory, Phenomenology, Institutional Ethnography, etc.; 5) Evaluating and analyzing the data - coding/theming 6) Reporting findings; 7) A quick review of software that can make analysis easier, or which allows for novel mixing of methods. Method and Extent of Audience Participation: The format will include a brief panel presentation, hand-out of take-home materials, a break-out session with participants practicing interviewing and coding techniques in small groups, and an open question-and-answer session with the four panel members. Prerequisite Knowledge: Participants should have a basic understanding of what qualitative research is. It may be beneficial for participants to have a project in mind, but this is not required.