• Physician Personal Characteristics and Obesity Treatment Plan Study

      Schottelkorb, Marisa Jo; The University of Arizona College of Medicine - Phoenix; Appelhans, Brad (The University of Arizona., 2011-03)
      Obesity is a very common finding in the primary care physician's office; however, physicians do not always treat obese patients according to guidelines. This investigation examined the possible relationships between a physician's personal weight history, anti-obesity bias, and the proposed treatment plan for an obese, hypertensive patient. Fifty-five Family Medicine, twenty-one Internal Medicine, and five “Other” Family or Internal Medicine Subspecialty physician participants viewed a video featuring either a hypothetical female or male patient with stage I hypertension and obesity (with a BMI of 32.) Of the total 81 physician participants, only 33 completed the entire study. Of these, 79.4% had attempted weight loss in their lifetime, with 55.9% having attempted weight loss in the past year. Median physician BMI was 25, with a small standard deviation of 2.7. Mean obesity bias score was 50.6 (out of possible 117, the maximum value reflecting the most anti-obese attitudes,) with a standard deviation of 19.7. Among all 81 participants, weight loss was in the three highest-ranked treatment recommendations in 90.9% of physician participants. Only 21.2% recommended all four JNC- 6 recommended components of lifestyle change. 93.9% recommended the initiation of a drug during this diagnostic visit. Statistical significance was not achieved in relating physician personal characteristics such as weight history to treatment recommendations due to the low number of participants completing all study measures.
    • Security and Privacy in Radiology

      Workie, Betre; The University of Arizona College of Medicine - Phoenix; Krupinski, Elizabeth (The University of Arizona., 2011-10-28)
      Radiology is one of the most high-tech fields of medicine. The digitization of medical information including radiographic data has led to improved efficiency and productivity but it has also presented new challenges in the area of privacy and security. In this study we have tried to answer some questions regarding how secure the current electronic radiology systems are and what individual and systemic factors affect the security and privacy of confidential patient data. A total of 77 radiologists and radiology residents participated in an online survey which included questions on physical security, computer systems security, and training and prevention. Since there was no objective way to measure overall security, we have used the overall security grade given by survey takers as our outcome, the dependent variable. Multiple regression analysis and ANOVA (Analysis of Variance) analyses were conducted. The regression analysis, with few exceptions, showed that only two variables contributed significantly to the final security grade. One of them was about unauthorized persons gaining access to the radiology facility, and the second one was about locking computer screens when temporarily away from a work station. A separate analysis was done using data for physical security, 4 computer systems security, training and prevention, and other security/privacy questions. Security threat level to current radiology systems is very low; however, the preparedness of these facilities to protect their infrastructure from future attacks is not adequate and there is room for improvement. As far as keeping confidential patient information private, most institutions seem to be doing a good job.
    • Specificity of Enzyme Immunoassay for Serologic Coccidioidomycosis Diagnosis Compared to Immunodiffusion

      Petein, Nathalie; The University of Arizona College of Medicine - Phoenix (The University of Arizona., 2011-03)
      BACKGROUND: Serologic testing for coccidioidomycosis challenges clinicians due to conflicting small studies regarding the sensitivity and specificity of newer enzyme immunoassay (EIA) tests and the lack of a true gold standard diagnostic test for comparison. METHODS: We analyzed all Lab Corp coccidioidomycosis serological test results from February 2008 through February 2009 and calculated the sensitivity, specificity, and positive/negative predictive values of EIA immunoglobulin (Ig)M and IgG. Immunodiffusion IgM and IgG (ID), complement fixation titers (CF), and tissue/culture diagnosis were used as tests for comparison. The comparison test (CT) was considered positive if any comparison test was positive the day of EIA collection or if tissue/culture diagnosis occurred during the time period. Cases required EIA IgM and IgG and ≥ 2 comparison tests performed the same day for inclusion. Medical records associated with positive EIA and negative comparison test results were reviewed for coccidioidomycosis symptoms, physician diagnosis, and subsequent positive comparison test results. Sensitivity, specificity, and predictive values were calculated, including those with subsequent positive comparison test results. RESULTS: A total of 1445 laboratory test sets were identified. EIA sensitivity and specificity were 83.8% and 92.6%, respectively. Positive and negative predictive values were 61.5% and 97.6%, respectively. Of 94 “false positive” EIA results, 92 (97.9%) were associated with documented coccidioidomycosis symptoms and 81% with coccidioidomycosis physician diagnosis. CONCLUSION: Based on the largest study of sensitivity and specificity calculated from laboratory surveillance data, EIA sensitivity and specificity for coccidioidomycosis diagnosis are lower than previously reported using only coccidioidomycosis laboratory tests as a comparison. However, association of “false positive” EIA results with coccidioidomycosis symptoms and physician diagnosis suggests that ID and CF laboratory tests alone are not a sufficient confirmation test for diagnosis.
    • The Test Tube Baby: Out of Reach or Out of Luck? A Retrospective Look at the Impact of Basal FSH and Age on In Vitro Fertilization Success in a Clinic Operating Without Laboratory Value Thresholds or Age Limits?

      Padrnos, Leslie; The University of Arizona College of Medicine - Phoenix; McGaughey, Robert (The University of Arizona., 2011-03)
      Objective: To assess the impact of age and FSH on IVF outcomes in an assisted reproductive technology clinic that does not have treatment thresholds based on age or laboratory FSH values. Design: Retrospective cohort study Setting: The Arizona Center For Fertility Studies in Phoenix, AZ Patient(s): Women who sought fertility treatment (with the exclusion of patients using donor or frozen oocytes) ages 18-50, representing a total of 1388 IVF cycles Intervention(s): IVF using nondonor embryos Main Outcome Measure(s): Live-birth rate per cycle started Result(s): A total of 1388 IVF cycles with autologous oocytes were analyzed to determine the impact of basal FSH and age on therapy outcomes. The pregnancy rates for individuals 18-34 years old were not significantly different and ranged from 41.1% to 34%. Pregnancy rates for individuals aged 35-39 years old exhibited a reduced pregnancy rate that ranged from 24.7% to 19.8%. For the eldest patients, a significant reduction in pregnancy rates was demonstrated with patients over the age of 40 having a pregnancy rate of 14.3%, and for those 41 years old and 42 and older having pregnancy rates of 7% and 6% respectively. The live birth rate also mirrored this trend with the youngest age group having a live birth rate of 38.9% and the eldest group of individuals aged 42-50 years having a live birth rate of 3.4%. While increasing FSH levels were associated with reduced numbers of oocytes retrieved and transferred during the IVF procedures, there was no statistically significant reduction in pregnancy rate or live birth rate in those with elevated basal FSH levels. Conclusion: The data analysis revealed that increasing age in this population does correlate with decreasing successful outcomes in IVF. At ages 36 and 40 years, there are significant reductions in pregnancy rate. At ages 38 and 40, there are significant reductions in live birth rate. Interestingly, there were no significant differences in pregnancy rate or live birth rate based on basal FSH level.