Department of Biostatistics, Epidemiology and Informatics
The faculty members of our Biostatistics, Epidemiology and Informatics Divisions discover, implement and disseminate novel methods in the three disciplines; the Epidemiology Division also embraces a broad array of clinical specialties. We are one of the nation’s only school-of-medicine departments that span such breadth. And we are the largest basic-science department within the highly rated Perelman School of Medicine (PSOM).
PublicationGuatemala & the University of Pennsylvania: Meeting in the Middle(2013-01-01) Barg, Frances K; Weiss, Eve; Branas, CharlesThe GUATEMALA-PENN relationship is a partnership that is borne from more than a century of research, service and scholarship. Our goal, with this book and with our programs, is to grow this partnership based upon the reciprocal needs of our Guatemalan stakeholders and the mission of the university. Guatemala is only one country away from the US. Our shared heritage and joint economic and social interests make it critical to foster a strong and mutually beneficial relationship between our two countries. Our connections have evolved over time in response to new knowledge gained about each other. This book is a testament to our collective past, present and future. For more information or to donate visit: http://www.med.upenn.edu/globalhealth/guatemalapartners.shtml PublicationPerceived Effectiveness of Acupuncture: Findings From the National Health Interview Survey(2008-12-01) LaRiccia, Patrick J; McMurphy, Suzanne; Gallo, Joseph J; Xie, Daiwei; Branas, CharlesBackground: Knowledge of perceived benefit from acupuncture treatment is important to predict who is using acupuncture, to inform physicians of the possible benefits of acupuncture, to determine where rigorous research should be focused, and to help policy makers predict future demand. Objectives: To determine the proportions of survey respondents who reported perceived effectiveness of acupuncture treatment for specific conditions; and to determine the association of specified demographic variables with perceived effectiveness. Design, Setting, and Participants: Publicly available data from 31,044 noninstitutionalized US adults who participated in the 2002 National Health Interview Survey, a cross-sectional in-home computer-assisted interview. Outcome Measures: The proportions of participants reporting “any help” and “great help” for perceived effectiveness across conditions treated and per condition treated; and the association of the subgroups within age, sex, and body mass index (BMI), along with the subgroups Asian race and Chinese ethnicity, with “any help” of acupuncture treatment across conditions treated. Results: Of the 1,274 respondents who reported having consulted an acupuncturist, 329 had used acupuncture in the last 12 months; 276 had used acupuncture to treat a specific condition. A total of 86% of respondents reported being helped by acupuncture, while 45% reported being greatly helped across conditions treated. In an examination of specific conditions treated, the range for “any help” was 98% to 67%, and 98% to 7% in the “great help” outcome. Older age and obesity were negatively associated with perceived effectiveness statistically but not clinically. Asian race and Chinese ethnicity were not statistically significantly associated with perceived effectiveness. Conclusions: Acupuncture is perceived to be effective by most respondents who used it to treat a specific condition. Older age and obesity are negatively associated with perceived effectiveness, but not at a clinically significant level. A larger sample of Asian and Chinese subgroups is needed to determine if there is an association of these subgroups with perceived benefit. PublicationTrivalent Influenza Vaccine Adverse Event Analysis Based On MedDRA System Organ Classes Using VAERS Data(2015-01-01) Tao, Cui; Du, Jingcheng; Cai, Yi; Chen, YongWe studied serious reports following influnza vaccine from VAERS database in year 2011. Our statistical analyses revealed differences of reactions among different age groups and between genders. The results may lead to additional studies to uncover factors contributing to the individual differences in susceptibility to influenza infection. Publicationmmeta: An R Package for Multivariate Meta-Analysis(2014-01-01) Luo, Sheng; Chen, Yong; Su, Xiao; Chu, HaitaoThis paper describes the core features of the R package mmeta, which implements the exact posterior inference of odds ratio, relative risk, and risk difference given either a single 2 x 2 table or multiple 2 x 2 tables when the risks within the same study are independent or correlated. PublicationAssociation Between β-Genus Human Papillomavirus and Cutaneous Squamous Cell Carcinoma in Immunocompetent Individuals—A Meta-analysis(2016-12-01) Chahoud, Jad; Semaan, Adele; Chen, YongIMPORTANCE Existing epidemiological evidence remains controversial regarding the association between β-genus human papillomavirus (β-HPV) and cutaneous squamous cell carcinoma (cSCC) in immunocompetent individuals. OBJECTIVE We aimed to clarify this association and evaluate type-specific β-HPV involvement. DATA SOURCES We performed a systematic literature search of MEDLINE and EMBASE for studies in humans through June 18, 2014, with no restriction on publication date or language. The following search terms were used: “human papillomavirus” and “cutaneous squamous cell carcinoma or skin squamous cell carcinoma or cSCC or nonmelanoma skin neoplasms.” STUDY SELECTION Articles were independently assessed by 2 reviewers. We only included case-control or cohort studies, in immunocompetent individuals, that calculated the odds ratio (OR) for cSCC associated with overall and type-specific β-HPV. DATA EXTRACTION AND SYNTHESIS We first assessed the heterogeneity among study-specific ORs using the Q statistic and I 2 statistic. Then, we used the random-effects model to obtain the overall OR and its 95% CI for all studies as well as for each type of HPV. We also tested and corrected for publication bias by 3 funnel plot–based methods. The quality of each study was assessed with the Newcastle Ottawa Scale. MAIN OUTCOMES AND MEASURES Pooled ORs and 95% CIs for overall β-HPV and HPV types 5, 8, 15, 17, 20, 24, 36, and 38 association with skin biopsy proven cSCC. RESULTS Seventy-nine articles were assessed for eligibility; 14 studies met inclusion criteria for the meta-analysis and included 3112 adult immunocompetent study participants with cSCC and 6020 controls. For all detection methods, the overall association between β-HPV and cSCC was significant with an adjusted pooled OR (95% CI) of 1.42 (1.18-1.72). As for the type-specific analysis, types 5, 8, 15, 17, 20, 24, 36, and 38 showed a significant association with adjusted pooled ORs (95% CIs) of 1.4 (1.18-1.66), 1.39 (1.16-1.66), 1.25 (1.04-1.50), 1.34 (1.19-1.52), 1.38 (1.21-1.59), 1.26 (1.09-1.44), 1.23 (1.01-1.50), and 1.37 (1.13-1.67) respectively. Our subgroup analysis in studies using only serology for HPV detection showed a significant association between overall β-HPV and HPV subtypes 5, 8, 17, 20, 24, and 38 with an increased risk of cSCC development. CONCLUSIONS AND RELEVANCE This study serves as added evidence supporting β-HPV as a risk factor for cSCC in healthy individuals. The subgroup analysis highlights this significant association for HPV 5, 8, 17, 20, and 38, which may help to direct future prevention efforts PublicationMethods for Linking Community Views to Measureable Outcomes in a Youth Violence Prevention Program(2012-01-01) McDonald, Catherine C; Richmond, Therese S; Guerra, Terry; Thomas, Nicole A; Walker, Alia; Branas, Charles; TenHave, Thomas; Vaughn, Nicole A; Leff, Stephen S; Hausman, Alice JBackground: All parties in community–academic partnerships have a vested interest prevention program success. Markers of success that reflect community’s experiences of programmatic prevention success are not always measurable, but critically speak to community-defined needs. Objective: The purpose of this manuscript was to (1) describe our systematic process for linking locally relevant community views (community-defined indicators) to measurable outcomes in the context of a youth violence prevention program and (2) discuss lessons learned, next steps, and recommendations for others trying to replicate a similar process. Methods: A research team composed of both academic and community researchers conducted a systematic process of matching community-defined indicators of youth violence prevention programmatic success to standardized youth survey items being administered in the course of a program evaluation. The research team of three community partners and five academic partners considered 43 community-defined indicators and 208 items from the youth surveys being utilized within the context of a community-based aggression prevention program. At the end of the matching process, 92 youth survey items were identified and agreed upon as potential matches to 11 of the community-defined indicators. Conclusions: We applied rigorous action steps to match community-defined indicators to survey data collected in the youth violence prevention intervention. We learned important lessons that inform recommendations for others interested in such endeavors. The process used to derive and assess community-defined indicators of success emphasized the principles of community-based participatory research (CBPR) and use of existing and available data to reduce participant burden. PublicationNonresectable Hepatocellular Carcinoma: Long-term Toxicity in Patients Treated with Transarterial Chemoembolization - Single-Center Experience(2008-10-01) Buijs, Manon; Vossen, Josephina A; Frangakis, Constantine; Hong, Kelvin; Georgiades, Christos S; Chen, Yong; Liapi, Eleni; Geschwind, Jean-François HPurpose: To determine the toxicity profile of transarterial chemoembolization (TACE) at 6 months and 1 year after treatment in patients with hepatocellular carcinoma (HCC) in a standardized oncology protocol so that TACE could be compared with systemic chemotherapeutic regimens for liver cancer. Materials and Methods: The study was authorized by the institutional review board. Between January 2002 and January 2007, 190 patients (155 men, 35 women; median age, 65 years; age range, 18 – 84 years) with HCC who underwent TACE treatment were identified from a prospectively collected database. Clinical records of complete blood cell counts and chemical profiles at baseline and at 6 and 12 months after treatment were studied retrospectively. Toxicity was graded according to the common terminology criteria for adverse events (CTCAE). A transition (survival) analysis perspective was used to estimate the distribution of toxicity grades. Patient survival from the first TACE session was calculated with Kaplan-Meier analysis. Results: Grade 3 or 4 toxicity 6 and 12 months, respectively, after treatment included leukocytopenia (7% and 19%); anemia (9% and 19%); thromobocytopenia (13% and 23%); prolonged activated partial thromboplastin time (8% and 18%); elevated aspartate aminotransferase (15% and 18%), alanine aminotransferase (10% and 18%), and alkaline phosphatase (8% and 18%) levels; hypoalbuminemia (10% and 19%); hyperbilirubinemia (10% and 22%); and alopecia (18%). The cumulative survival rate was 58% at 1 year, 39% at 2 years, and 29% at 3 years. These toxicity rates were considerably lower than those reported after treatment with currently used systemic chemotherapeutic agents. Conclusion: Study results show that TACE has a favorable long-term toxicity profile in patients with HCC. Data clearly support the role of TACE in the treatment of patients with nonresectable HHC. PublicationExamination of Carotid Arteries with Quantitative Color Doppler Flow Imaging(1994-02-01) Branas, Charles; Weingarten, Michael S; Czeredarczuk, Michael; Schafer, Paula FThe results of conventional duplex scanning were compared with QCDFI. A total of 224 consecutive patients comprising 442 unilateral carotid systems were examined by conventional duplex techniques. MPSV, as determined by QCDFI, were recorded for each of the 442 carotid segments and grouped according to the previously determined degrees of stenosis. The predictive value of QCDFI was confirmed by angiography with an overall accuracy of 91%. Results obtained by duplex scanning correlated with angiography 89% of the time. Based on QCDFI data, a scale to grade carotid stenosis was developed. Publication"Sleep Disparity" in the Population: Poor Sleep Quality is Strongly Associated with Poverty and Ethnicity(2010-08-11) Patel, Nirav P; Grandner, Michael A; Xie, Dawei; Branas, Charles; Gooneratne, NalakaBackground: Little is known about the social determinants of sleep attainment. This study examines the relationship of race/ethnicity, socio-economic status (SES) and other factors upon sleep quality. Methods: A cross-sectional survey of 9,714 randomly selected subjects was used to explore sleep quality obtained by self-report, in relation to socioeconomic factors including poverty, employment status, and education level. The primary outcome was poor sleep quality. Data were collected by the Philadelphia Health Management Corporation. Results: Significant differences were observed in the outcome for race/ethnicity (African-American and Latino versus White: unadjusted OR = 1.59, 95% CI 1.24-2.05 and OR = 1.65, 95% CI 1.37-1.98, respectively) and income (below poverty threshold, unadjusted OR = 2.84, 95%CI 2.41-3.35). In multivariable modeling, health indicators significantly influenced sleep quality most prominently in poor individuals. After adjusting for socioeconomic factors (education, employment) and health indicators, the association of income and poor sleep quality diminished, but still persisted in poor Whites while it was no longer significant in poor African-Americans (adjusted OR = 1.95, 95% CI 1.47-2.58 versus OR = 1.16, 95% CI 0.87-1.54, respectively). Post-college education (adjusted OR = 0.47, 95% CI 0.31-0.71) protected against poor sleep. Conclusions: A "sleep disparity" exists in the study population: poor sleep quality is strongly associated with poverty and race. Factors such as employment, education and health status, amongst others, significantly mediated this effect only in poor subjects, suggesting a differential vulnerability to these factors in poor relative to non-poor individuals in the context of sleep quality. Consideration of this could help optimize targeted interventions in certain groups and subsequently reduce the adverse societal effects of poor sleep. PublicationAutomated Evaluation of Medical Software Usage: Algorithm and Statistical Analyses(2015-01-01) Cao, Ming; Chen, Yong; Zhu, Min; Zhang, JiajieEvaluating the correctness of medical software usage is critically important in healthcare system management. Turf is a software that can effectively collect interactions between user and computer. In this paper, we propose an algorithm to compare the recorded human-computer interaction events with a predefined path. Based on the pass/fail results, statistical analysis methods are proposed for two applications: to identify training effects and to compare products of the same functionality.