2009 Summer Student Research and Clinical Assistantship Program Faculty Research Abstracts

1. Treatment of Opioid Dependence in Drug Court
2. Psychosocial Predictors of Common Cold Outcomes - Study 1
3. Benefit Harm Tradeoffs Assess Sufficiently Important Difference - Study 2
4. Gender and Authorship of Publications in Family Medicine Journals
5. Metabolic Measures of Recent Hmong Immigrants
6. Geographical Distribution of Bronchitis vs. URI Diagnosis: Eastern Wisconsin
7. The Effect of Prolotherapy on Subjects with Osteoarthritic Knee Pain: A Qualitative Study
8. Loss to Follow-up of Patients with Abnormal Cervical Cytology in Two Family Medicine Residency Clinics
9. Technology and Aging: A Pilot Study on the use of Technology by Patients over 65 Years of Age
10. African-American Infant Mortality Investigation
11. Implementing a Pediatric Developmental Screen In Two Family Medicine Clinics
12. Group Prenatal Visits at Wingra Family Medical Center: A Pilot Study
13. Nipple Shield Survey
14. The Reliability and Responsiveness of the SANE Knee Scale in a Population of Adults Undergoing Knee Surgery

1. Title: Treatment of Opioid Dependence in Drug Court

Faculty:Randall Brown, M.D.
Location: Madison

Rationale: Substance use adversely affects individual and population health and public safety. Contact with the criminal justice system potentially represents a critical point for drug involved offenders to access treatment and prevention services. Treatment and service provision under community supervision are known to be more likely than incarceration to affect positive change in terms of reduced substance use, reduced HIV risk behaviors, and increments in community reintegration. Health services (other than substance abuse treatment) to this vulnerable population, with a complex set of health care needs and potentially detrimental effects upon public health and safety, remain understudied. Little is known about the potential role for or impact of primary care services for drug-involved offenders. This could be an important resource for drug courts in areas that do not have specialist facilities. We propose a pilot study of health care access models for drug-involved offenders using 2 clinical settings (primary care vs. drug treatment specialist center) as the conduits for care access and referral.

Hypotheses: 1) Subjects receiving primary care treatment will have reduced emergency room and hospital visits and, 2) Reductions in opioid use and risky HIV behaviors will be comparable for primary care-referred versus specialty care-referred (methadone treatment facility) offenders.

Aims/design: All subjects will receive supervision as usual through the Dane County Drug Treatment Court (DTC). Active participation will be 6 months. The project will randomly assign 15 opioid-dependent participants in drug court to one of three study arms: (1) primary care with Suboxone pharmacotherapy, (2) specialized methadone treatment with Suboxone, or (3) specialized methadone treatment plus methadone pharmacotherapy. Outcomes will include services uptake, treatment adherence, health services utilization, urine drug testing, HIV risk behaviors, and quality of life.

2. Title: Psychosocial Predictors of Common Cold Outcomes - Study 1

Faculty: Bruce Barrett, M.D./Ph.D.
Location: Madison

Background: Little is know about potential differences in how people experience acute upper respiratory viral infections (eg. common cold). We have very recently completed a randomized trial with N=719 participants. That trial tested two kinds of "placebo effects": 1) those attributable to placebo pills, and 2) those attributable to patient-oriented clinical interaction. The resulting data set is the best of its kind, and will allow groundbreaking assessment of potential relationships among psychosocial attributes and common cold outcomes.

Specific aims: To assess possible influence of demographic attributes (gender, age, ethnicity, education) and psychological tendencies (perceived stress, optimism, sociability) on common cold outcomes. These include severity and duration of illness, as assessed by the Wisconsin Upper Respiratory Symptom Survey, and two laboratory-measured biomarkers: IL-8 (inflammatory cytokine) and nasal neutrophil counts.

Research questions:
a. Does baseline psychological health (stress, optimism, social support) predict subsequent outcomes?
b. Do baseline sociodemographic indicators (age, race, education, income) predict subsequent outcomes?
c. Do men and women experience colds differently? Are inflammatory biomarkers the same for both sexes?
d. Do psychosocial attributes mediate cold expression the same way in men and women?
e. What aspects of the doctor patient interaction influence patients' illness experience?

Methods: The data has been collected, entered, and cleaned. The first step in analysis will be to look at the central tendency and dispersion of all variables, using simple tables and graphics. The second step will be to assess bivariate relationships using scatterplots and correlation matrices. Where associations appear to exist, multivariate analyses will be conducted to assess and possibly control for covariate influence.

Products: With the summer medical student's assistance, we will draft, refine, and edit and submit one or two manuscripts to one or more peer-reviewed journals.

3. Title: Benefit Harm Tradeoffs Assess Sufficiently Important Difference - Study 2

Faculty: Bruce Barrett, M.D./Ph.D.
Location: Madison

Background: Sufficiently important difference (SID) is the smallest benefit that an intervention would require in order to justify costs and risks. SID is the quantity of benefit for which clinical trials should be designed and interpreted. Benefit harm trade-off interviews (in-person and telephone) have assessed SID in terms of magnitude of reduction of illness duration and overall severity of common cold. The current project focuses on the potential influence that side effects might have on medical decision-making.

Methods: We used evidence-based simple-language scenarios to assess the benefit harm trade-off for four common cold treatments: vitamin C, the herbal medicine echinacea, zinc lozenges, and the unlicensed antiviral pleconaril. Treatment scenarios and potential benefits were kept constant. Risks and severities of potential side effects were systematically varied, aiming to determine the point at which the interviewee felt that potential benefits equaled potential harms. That "tipping point" represents the SID.

Progress to date: Some N=230 interviews have been conducted, and the data has been entered and is ready for analysis. Next steps include data analysis and write-up of a publishable manuscript.

Specific aims: The summer medical student would assist with basic data analysis (simple graphics, tables) and with writing a manuscript to describe our findings. The student could also assist with the next phases of SID research, including design of structured interviews, and construction and submission of research proposals.

4. Title: Gender and Authorship of Publications in Family Medicine Journals

Faculty: Sarina Schrager, MD, MS
Location: Madison

Background: Despite increasing numbers of women attending medical school and completing residencies, women continue to lag behind men in academic achievement. In 2005, women comprised only 15% of all full professors and 11% of all department chairs. (Hamel, 2006) Women also author original research less often than men. (Jagsi, 2006). Studies in other specialties and among major journals support this discrepancy, though no research in family medicine has been performed. Since writing articles is an integral part of academic medicine, this may be one reason that women do not advance as much as men in the academic medical community.

Methods: This study will examine the 5 main family medicine journals over the past 3 years and determine the gender of authors of original research, review papers, EBM reviews, and letters to the editor. The student will review the journals, set up a database, and aid in the interpretation of results.

5. Title: Metabolic Measures of Recent Hmong Immigrants

Faculty: Gregory Cowan, Ph.D.
Location: Wausau

Background: Changes in metabolic risk factors often accompany immigrants' change from agrarian to westernized diet and lifestyle. Our relatively high fat, low fiber diet results in increased rates of obesity, hyperlipidemia, and hypertension. Incidence of these conditions in existing U.S. Hmong populations is well documented. Yang and Mills (2008) assert that "low socioeconomic status and rapid urbanizarion may have resulted in a shift from a high energy expenditure lifestyle and high fiber diets to a sedentary lifestyle with high saturated food diets, which may be detrimental to the health of many Hmong."

Methods: Existing UW Wausau Family Medicine records include new immigrant physical examination data from the last group of Hmong immigrants who arrived from 2004 through 2006. We will examine clinic charts to record fasting laboratory tests of lipids and glucose. We will compare these data to similar data from a convenience sample, matched for gender and age, of our Hmong patients who have resided in the U.S. for a number of years. These data will provide a picture of the changes in health risk factors associated with a change from traditional to western diet and lifestyle.

6. Title: Geographical Distribution of Bronchitis vs. URI Diagnosis: Eastern Wisconsin

Faculty: Dennis J. Baumgardner, MD
Location: Center for Urban Population Health/Aurora UW Medical Group, Milwaukee

Background: Antibiotic resistance is an important and ongoing health problem. A major component is the prescription of antibiotics for predominantly non-bacterial infections such as acute bronchitis. One study has revealed that clinician use of the term "chest cold" rather than "bronchitis" when a patient presents for an acute respiratory infection may improve patient satisfaction when an antibiotic is not prescribed. In addition, a study has shown no broad regional geographic differences in antibiotic prescribing for colds and bronchitis, but did reveal that pediatricians were less likely to prescribe antibiotics in these circumstances than other primary care clinicians. We propose a geographic information system (GIS) analysis of the distribution of acute bronchitis diagnosis versus diagnosis of cold/URI in Eastern Wisconsin (ages 8-49; N=2,200) including demographic variables of age, gender and race/ethnicity of the patient, season, and type of clinician specialty.

Methods: Our research hypothesis is that the ratio of bronchitis to cold/URI diagnosis and antibiotic prescription will vary by geographic region likely on the basis of individual or group physician custom. Clinic locations will be geocoded, mapped and analyzed using Map Marker Plus and Arc-GIS, and SaTScan or CrimeStat. Statistical analysis will be performed using a chi-squared test (or Fisher exact test, if required) for categorical variables and T test for normally distributed continuous variables and a non-parametric test for ordinal or non-normally distributed continuous variables. Multivariate analysis will be performed using regression models entering significant (p<0.05) or borderline significant (p<0.15) variables identified by univariate analysis.

7. Title: The Effect of Prolotherapy on Subjects with Osteoarthritic Knee Pain: A Qualitative Study

Faculty: David Rabago, MD
Location: Madison

Background: Prolotherapy is a rapidly growing injection-based complementary and alternative therapy for chronic musculoskeletal pain. The technique involves the local injection of solutions (proliferants) at the bone/ligament junction and in joint spaces; the suggested mechanism of action involves an inflammatory response and subsequent anabolic reaction to strengthen local soft tissue or cartilage. Prolotherapy has strong RCT evidence for overuse tendon disorders (tendinopathy); ours is the first rigorous study of its use for knee osteoarthritis.

Methods: Our group currently has an NIH grant to investigate the effect of prolotherapy on osteoarthritic knee pain in a three-arm RCT and 2 prospective case series. Blinded prolotherapy and sham injection subjects are currently out-performing physical therapy in this study. Knowledge of subjective patient experience with this therapy will be critical to our understanding of the clinical utility of prolotherapy. Twenty-five qualitative interviews will be complete and transcribed by June 2009. The student in this study will analyze existing qualitative data with study personnel using a rubric published in prior studies by the PI. The student will gain an understanding of 1) qualitative methodology, 2) the daily conduct of an RCT and 3) prolotherapy, a promising intervention for common, debilitating musculoskeletal pain. The PI is very interested in working with students; the prior 4 SSRCA, Shapiro and undergraduate students mentored by Dr. Rabago have had successful experiences that led to peer-reviewed publications and abstract presentations in professional meetings.

8. Title: Loss to Follow-up of Patients with Abnormal Cervical Cytology in Two Family Medicine Residency Clinics

Faculty: Ann Evensen, MD and Helen Luce, DO
Location: Verona and Wausau

Background: The inherent turnover of residents physicians in training programs may increase the number of patients with abnormal cervical cytology who have inadequate follow-up. Residency programs typically serve communities who are at increased risk for both cervical cancer and poor cervical cancer screening practices including women who smoke and those of low socioeconomic status. The high spontaneous regression rate of low grade cervical cytology abnormalities may decrease the number of poor outcomes from lack of follow-up. However, the magnitude of the loss to follow-up problem is not known.

Methods: The project will be a retrospective chart review to determine the number of patients with abnormalities on cervical screening who were appropriately monitored over the 15-24 months following their original screening. "Appropriate monitoring" will be defined by the American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines in effect at the time of the original screening.

9. Title: Technology and Aging: A Pilot Study on the use of Technology by Patients over 65 Years of Age.

Faculty: Eric D. Miller, M.D. and Bert C. Callahan, M.D.
Location: Beaver Dam

Background: Annual public healthcare spending in the U.S. has grown 2.7 percentage points faster than the per capita Gross Domestic Product, twice the rate of the developed country average. Staffing shortages exist in long-term care facilities nationwide and Medicare and Medicaid reimbursements fall short of covering expenses of care. Under our current paradigm of delivery of healthcare, the coming surge in need for healthcare in our increasingly large aging population will overwhelm our resources and create a national healthcare crisis. Innovative new approaches to providing health care are needed. The goal of health care delivery is to provide safe, effective, evidence-based care, in a manner that is cost effective and preserves the dignity and integrity of the patient, and the doctor-patient relationship. Keeping patients in their home, if done safely, can "decentralize" care, reduce the need for expensive long-term care, and reduce the overall cost demands on the U.S. health care system. Application of technology in novel ways has the potential to meet these goals.

Methods: This Pilot Study aims to enlist a group of patients on whom to prospectively and randomly study the effects of a simple monitoring system on their health and well being, as well as the effect on their family members and their personal physician.

10. Title: African-American Infant Mortality Investigation

Faculty: Lee Dresang, MD;Laura Berghahn, MD;Gloria Sarto, MD;Tom Schlenker, MD
Location: Madison, WI

Background: The United States continues to have one of the highest infant mortality rates of industrialized countries. Most notable is the high rate among African Americans. The US infant mortality rate was 6.78 infant deaths per 1,000 live births in 2004; the black infant mortality rate (BIMR) was 13.25. The Wisconsin 2000-2004 infant mortality rate for all births was 6.7, while the BIMR was 17.6. In contrast to Wisconsin and national statistics, the Dane County BIMR decreased since the 1990s by approximately 70% to 6.4 and has been sustained for 6 years. Preconceptual care and interconceptual care have been linked to improved birth outcomes. Interconceptual care is a subset of preconceptual care addressing care between pregnancies. For many pregnancy complications, the leading risk factor is a history of that complication in a previous pregnancy. Interconceptual care is an opportunity to reduce recurrence risk. Addressing bacterial vaginosis, birth spacing, chronic disease control, stress and substance abuse may improve outcomes. Number of visits, model of care and continuity of care will be assessed.

Methods: A retrospective cohort study will look at clinic records for African American births in 1992-1997 and 2002-2007 to see if a change in preconceptual and interconceptual care is associated with the drop in infant mortality in Madison. The student researcher will help with data collection from clinic records for African American babies who delivered at both Meriter and St. Mary's Hospitals. Once collected, data will be analyzed with the aid of UW biostatisticians and submitted for publication.

11. Title: Implementing a Pediatric Developmental Screen In Two Family Medicine Clinics

Faculty: Beth Potter, M.D and Melissa Stiles
Location: Madison

Background: Family physicians provide care to patients across the lifespan including infants and young children. One of the important roles of the physician is to identify developmental delays in children. It has been well documented that children with developmental delays benefit from early interventions. Studies show that relying on clinical judgment misses up to 70% of children with developmental disabilities. The Ages and Stages Questionnaire is a validated screening tool for developmental delays in children from infancy to age 5. This tool has been shown to identify 70% to 80% of children with developmental delays. Wingra and Belleville Clinics are part of the UW Department of Family Medicine. Each clinic serves a unique population. Wingra serves an urban underserved population with a large Spanish speaking patient base. Belleville serves a rural population. Currently neither clinic has a formal developmental screening process.

Methods: The focus of this project would involve gathering background data on physician knowledge and skills regarding developmental screening, training physicians and staff about the Ages and Stages Questionnaire, and implementing a clinic-wide process for incorporating the developmental screening into everyday clinical care.

12. Title: Group Prenatal Visits at Wingra Family Medical Center: A Pilot Study

Faculty Beth Potter, MD. And Suhani Bora, MD
Location: Madison

Background: Group prenatal care has been shown to improve perinatal outcomes by reducing rates of preterm births, improving prenatal knowledge and satisfaction with care, as well as increasing rates of breastfeeding initiation. The current model of prenatal care consists of fifteen minute appointments with a primary care provider throughout the duration of the pregnancy. The prenatal group visit model extends visits to two hours and allows for education and counseling on topics such as labor and delivery, breastfeeding, proper nutrition, and parenting that require significant counseling in a very limited amount of time while trying to also facilitate access to community resources. Group visits also promote social networking.

Methods: This pilot study will focus on the patient population of Wingra Clinic, a largely underserved population. Patients with newly diagnosed pregnancies will be offered the option of traditional prenatal care or the option of prenatal group visits. Participants in the group prenatal visits will continue to be followed by their primary care provider by alternating group visits with individual visits throughout their pregnancy. The pilot study will involve administering surveys to patients who choose either option throughout their pregnancy. Our aim is to assess the feasibility, knowledge, attitudes, and barriers to group prenatal visits so that we can implement successful group prenatal visits at Wingra Clinic in the near future.

13. Title: Nipple Shield Survey

Faculty: Anne Eglash, MD
Location: Mt. Horeb

Background: Nipple shields are small latex covers that women use over their nipples to help babies latch onto the breast. Despite concerns about a decade ago that using a nipple shield can decrease one's milk supply, nipple shields have gained renewed popularity based on very few studies suggesting that perhaps nipple shields are safe. There have been many reports by patients and health professionals that nipple shields are being used in inappropriate situations, leading to more problems than would be present without nipple shield use.

Methods: The goal of this project is to survey a group of physician breastfeeding specialists in the Academy of Breastfeeding Medicine, to find out what criteria they typically use when recommending a nipple shield, and in what situations they avoid the shield.

14. Title: The Reliability and Responsiveness of the SANE Knee Scale in a Population of Adults Undergoing Knee Surgery

Faculty: John J. Wilson, MD, Tim McGuine, PhD, LAT
Location: Madison and Verona

Background: There are many patient-reported knee-rating scales available for the evaluation of patients' function. Most patient reported knee outcome scales include many questions with numerous subscales, and can be a patient burden. A Single Assessment Numeric Evaluation (SANE) knee scale, consisting of one question, exists, but reliability and responsiveness have not been evaluated compared to other commonly used knee scales.

Purpose: To determine the reliability and responsiveness of a SANE knee scale in adult patients undergoing knee surgery.

Methods: Subjects: Adult patients seen in the UW Health Sports Medicine Center who undergo knee surgery. Each subject completes a series of self-report questionnaires designed to assess their knee function at three separate, distinct intervals. The questionnaires include: 1) The 2000 IKDC Functional Knee Scale, 2) SF-12 Global Quality of Life Assessment 3) Single Assessment Numeric Evaluation (SANE). Subjects will complete the questionnaires 1) at their initial clinic visit, 2) 3 - 5 days after initial visit (via mail or in the clinic), 3) at their first post-operative visit, and 4) at their final scheduled clinic visit (usually 3 - 6 months) following surgery.

Analyses: Reliability of the instruments will be assessed by estimating the intraclass correlation coefficient (ICC) and the associated 95% confidence intervals (CI) for the scores at a subjects' initial and second administration of the surveys. Responsiveness will be assessed estimating the intraclass correlation coefficient (ICC) and the associated 95% confidence interval (CI) for the subjects' initial, third, and final scores. Convergent validity will be assessed by comparing the SANE and 2000 IKDC scores.