Biostatistics and Research Epidemiology/Cancer Epidemiology Center for Health Services Research

Biostatistics and Research Epidemiology/Cancer Epidemiology

1. Senior Research Staff

Gwen Alexander, Ph.D. Andrea Cassidy, Ph.D. Christine Cole Johnson, Ph.D., Head Cancer Epidemiology George Divine, Ph.D. Sharon Hensley-Alford Christine Joseph, Ph.D. Lois Lamerato, Ph.D.

Mei Lu, Ph.D. Edward Peterson, Ph.D., Chair Benjamin Rybicki, Ph.D. Lonni Schultz, Ph.D. James Yang, Ph.D. Ganesa Wegienka, Ph.D. Kimberly Woodcroft, Ph.D.

2. Research Summaries

Principal Investigator: Cole-Johnson, Christine, Ph.D.
Effects of Tailoring Depth in Web-based Smoking Interventions (NIH R01 CA101843, Subcontract)

Henry Ford Health System epidemiologists and biostatisticians have been collaborating for over a decade with Wayne State University toxicologists and environmental health scientists in the "Environmental Health Sciences Center in Molecular and Cellular Toxicology with Human Applications". This EHS Center, which is now in its eleventh year of funding by NIEHS, is composed of numerous programs that bring basic, clinical, and population scientists together in an effort to convert mechanistic research into an effective understanding of disease etiology and progression, as well as prevention and treatment, and to increase the degree to which population studies are poorly supported by mechanistic research. To accomplish these objectives, the Center has developed three Research Cores, four Facility Cores and a Community Outreach and Education Program. One of the three research cores is housed at Henry Ford, the Environmental Epidemiology Research Core, directed by Drs. Christine Johnson and Edward Peterson. The Environmental Epidemiology Research Core's mission is to foster the conduct of multi-disciplinary research in the EHS Center relating to environmental exposures and how such factors interrelate to lifestyle characteristics, human biological mechanisms and systems, and genetic susceptibility to affect human health and quality of life. The focus in the Core is in three disease areas: allergy and asthma, cancer, and neurological diseases. Core members have subsidized access to state-of-the-art technology in the Center’s Facility Cores, including a cell culture core, imaging and cytometry, microarray and bioinformatics (a core supported by HFHS biostatisticians), and high throughput proteomics and genomics cores. Environmental Epidemiology Research Core members are active in the Community Outreach and Education Program, assisting with the development of a web-based instructional program on asthma for middle school age students and giving lectures to high school teachers in a summer training program.

Principal Investigator: Cole-Johnson, Christine, Ph.D.
Endotoxin Exposure Alers Anti-Tetanus IgE in Infants (NIH R21 AI059415, Subcontract)

Recent studies have suggested that high levels of endotoxin exposure during infancy are associated with a reduced risk of subsequent allergic sensitivity. Learning more about the relationship between environmental endotoxin exposure and subsequent allergic disease is potentially important given the increasing prevalence of allergic diseases in the United States. The goal of this proposal is to evaluate the hypothesis that high levels of environmental endotoxin exposure will be associated with reduced anti-tetanus IgE responses following routine immunizations with tetanus toxoid in infancy. This study will utilize the structure of the ongoing WHEALS Study (AI/ES 50681, a study to evaluate the hygiene hypothesis in a multi-racial birth cohort of 3000 children in and around Detroit, MI). In home endotoxin assessments will be increased from 1 to 5 locations (the floor beside the infant's crib, the infant's bed, the parent's bed, the floor beside the parent's bed, and the floor in the living/family room) during both the 1 and 6 month home visits. Endotoxin will be measured using a commercial kinetic Limulus assay. Relationships between endotoxin measurements by location, month of year, and household characteristics (e.g., animals in the home) will be examined. These analyses will provide a better understanding of variations in endotoxin exposure within homes and allow an assessment of whether measurements from certain locations are more closely related to altered IgE production. All infants in this cohort are expected to receive routine immunizations with DTaP (diphtheria, tetanus and acellular pertussis) vaccine at 2,4, 6-7 and 15-18 months of age. In addition to measuring IgE specific for common allergens at 6 and 24 months of age, IgE specific for tetanus toxoid will be measured in the same blood samples using the commercial Pharmacia CAP assay. The anti-tetanus IgE measurements will be analyzed to learn whether IgE production is influenced by endotoxin exposure. Important advantages of measuring anti-tetanus IgE are the likelihood that most infants will receive all immunizations with identical doses at the same ages. The well standardized tetanus immunizations contrast with the highly variable and difficult to measure exposures to natural allergens. Anti-tetanus IgE responses are also likely to be measurable much earlier in infancy than are responses to other allergens. Finally, the designs of WHEALS and this application will allow us to compare the effects of endotoxin exposure on the expression of multiple cytokines related to the development of allergic disease.

Principal Investigator: Cole-Johnson, Christine, Ph.D.
CRN3-Infrastructure (NIH U19CA079689-09, Subcontract)

The CRN Infrastructure activities will work towards: 1) Further developing human and data resources and pursue projects related to our research theme of improving the effectiveness of cancer prevention and treatment through research that identifies system, provider, treatment, and patient factors affecting outcomes; 2) Providing more concerted training and mentoring to junior investigators at CRN sites and investigators at CRN sites and affiliated academic institutions; 3) Accelerating the development of the Virtual Data Warehouse, natural language processing and other standardized approaches to using the informatics resources of CRN delivery systems for research, and coordinate these efforts with the NCI’s Biomedical Informatics Grid (caBIG™); 4) Increasing collaboration between CRN sites and investigators at the NCI as well as in cancer centers and other academic centers; 5) Building a productive research program studying the diffusion of innovations in cancer care; and, 6) Studying and implementing strategies for increasing accrual in clinical trials.

Principal Investigator: Hensley Alford, Sharon, Ph.D.
The HMO Cancer Research Network: "Multiplex" Supplement (NIH U19CA079689-08, Subcontract)

Leaders in genome science and medicine have claimed that the development of genetic tests for susceptibility to common diseases will revolutionize preventive medicine.  Currently, these tests are being marketed directly to consumers without established clinical utility or validity.  The purpose of this research is to use a prototypic genetic susceptibility test (hereafter called the Multiplex Genetic Test) based on the state-of-the-science to better address fundamental, as yet unanswered, questions about this technology including: what do health-care consumers consider to be important risks and benefits of such testing and how do they respond to being offered multiplex testing and related test results.  We are conducting a descriptive study in which all individuals who are approached will be offered Multiplex Genetic Testing.  The test will be offered to a selected sample of 25-40 year old healthy volunteers from an HMO population in Detroit, Michigan (Henry Ford Health System).  The primary outcomes of interest are what proportion of individuals decide to be tested and what are the information-seeking behaviors that occur throughout the process of considering and undergoing genetic testing. We will also specifically explore who is most interested in Multiplex Genetic Testing, individuals’ reasons for and against getting tested, and participants’ understanding of and psycho-social responses to test results.  Outcomes will be measured using 1) a baseline assessment, 2) qualitative and quantitative questionnaires at various stages in the recruitment, consent, and results reporting stages of the study, and 3) a post-feedback follow-up assessment 6 months after testing.  Additionally, we will consider demographic and health-care utilization in the analyses.  Baseline assessment data will be used to evaluate the association between demographics, perceived family history, attitudes and beliefs about genetics and research participation, and participants’ willingness to be tested.  Health care utilization records will be used to determine whether Multiplex Genetic Testing motivated any changes in health-care use.   Our poster presentation will cover the studies main objectives and review the current status of the study including accomplishments and challenges.

Principal Investigator: Joseph, Christine, Ph.D.
A Tailored Asthma Education Program for Urban Teens II (NIH, R01HL068971-05A1)

Asthma death rates for teenagers 15-19 years of age are higher than that of younger children, despite a higher prevalence in the latter group. The small number of publications on urban adolescents suggests that inadequate asthma management plays a significant role in these grim statistics; however, this age group has been difficult to reach, in terms of both connecting and convincing. Consequently, there are few asthma management programs targeting urban, high school students with asthma. We have developed a unique, multi-media, web-based program (Puff City I) to motivate teens to change negative behaviors related to asthma management. Puff City I software uses "tailoring" in conjunction with theoretical models of behavior change, to personalize health messages according to the beliefs, attitudes, and barriers of students with asthma. A school-based approach ensures accessibility. An asthma counselor responds to referrals generate by the software, providing assistance in addressing barriers to effective asthma management. A randomized trial to evaluate Puff City I in Detroit high schools (98% African American) is in its final stages. Results of initial analyses demonstrate short-term behavior changes, in addition to fewer Emergency Department (ED) visits (p=0.03) and hospitalizations (p=0.007), and higher scores for quality of life (p=0.02) when compared to controls. As with any research trial, not all students improved, some relapsed, and a substantial number did not participate. To maximize effectiveness of the program, we propose to use Puff City I data, including teen sociodemographics, motivation, and reported barriers, to make refinements that would target program failures resulting in a new version or Puff City II. In addition, we will test new theory- and empirically-based approaches to recruiting urban high school students into this and other randomized trials. Not only would the latter be useful to research, in terms of generalizability, but will also yield valuable information about attracting this population to health-related interventions. We will evaluate Puff City II in another randomized trial in 9 Detroit high schools. Dissemination of this unique tool will also be explored as part of this proposal. Our collaborators include the University of Michigan Center for Health Communications Research, Detroit Public Schools, and the Detroit Department of Health and Wellness Promotion (formerly Detroit Health Department). The result of this proposal will be a powerful tool for changing negative behaviors related to asthma management in a traditionally hard-to-reach population that experiences high asthma morbidity and mortality.

Principal Investigator: Neslund-Dudas, Christine, M.A.
Residential Segregation, Housing Status and Prostate Cancer in African American and White Men (DOD, W81XWH-07-1-0252)We hypothesize, that race-based residential segregation leads to disparities in both area (census tract/block group) and individual physical and social housing conditions that dispose African-American men to differential environmental conditions that lead to excesses in biological damage, increasing risk for prostate cancer, earlier age of prostate cancer onset, and worse prostate cancer outcomes.  Specifically, we aim to determine whether selected area housing and individual housing status are associated with prostate cancer risk, age at diagnosis, and tumor aggressiveness,  and whether housing status is associated with observed racial differences in these prostate cancer outcomes.  We also aim to determine, through the use of factor analysis, whether area housing and individual housing status, is associated with prostate cancer risk, age at diagnosis, and tumor aggressiveness, through “latent factors” that include diet, physical activity, and genetic polymorphisms and whether those “latent factors” differ by race.  Finally, to begin to test biological pathways through which housing status may impact prostate health outcomes; specifically, whether housing status is associated with markers of DNA damage (polycyclic aromatic hydrocarbons DNA-adducts (PAH)) and DNA stability (telomere content) in prostate tumor tissue and tumor-adjacent normal tissue of African-American and white cases.

Principal Investigator: Rybicki, Benjamin Ph.D.
A Nested Case-Control Study of Prostate Carcinogenesis (NIH, 2R01ES011126-06A2)

Prostate Cancer is a slow growing disease that likely involves a series of environmental insults resulting in accumulated DMA damage eventually leading to overt carcinogenesis. DNA adducts are one of the few biomarkers for exposures directly related to cancer that can be quantified in human cells and a reliable measure of biologically effective dose for known carcinogens such as polycyclic aromatic hydrocarbons (PAH) and 2-amino-1-methyl-6-phenylimidazo[4, 5-b]pyridine (PhIP). Epigenetic markers are emerging as important in determining the extent of prostate carcinogenesis. Recent studies suggest that DNA adduct formation and aberrant gene promoter hypermethylation may be related elements in environmentally-induced carcinogenesis. Most research done with respect to DNA adducts, promoter hypermethylation and prostate cancer has focused on cells harvested from patients with prostate cancer or pre-malignant lesions. While these studies have been instructive, a clearer picture of the interconnection and risk associated with DNA adduct formation and epigenetic changes in prostate can only be gained from studies of prostate tissue captured before the onset of disease. At the Henry Ford Health System, we have characterized and have access to a racially diverse cohort of over five thousand men without prostate cancer from whom benign prostate specimens were surgically removed between 1990 and 2002. We plan to expand this cohort through 2006, and will follow-up cohort members for incident prostate cancer diagnoses through 2010 to achieve a desired study sample size of 800 matched case-control pairs. Building on findings from our initial funding period that characterized determinants of PAH- and PhlP-DNA adducts in the prostate cells of men with prostate cancer, in this competing continuation we seek to better understand the temporal relationship between DNA adducts and other epigenetic changes in the benign prostate and later prostate cancer development. To achieve this objective, we plan to conduct a nested case-control study of prostate cancer that will: 1) determine whether PAH- and PhlP-DNA adducts are predictive of later prostate cancer development after adjusting for other possible confounders; 2) determine in a multivariable model how aberrant gene promoter DNA methylation affects the association between PAH-and PhlP-DNA adducts and prostate cancer; and 3) determine whether DNA adducts in the benign prostate are associated with the level of expression of the p53 and p21waf/cip1 tumor suppressor genes in prostate tumors of men who develop prostate cancer.

Principal Investigator: Rybicki, Benjamin Ph.D.
Admixture Mapping of Sarcoidosis Genes in African Americans (NIH, 1 R01 HL092576-01A2)

Sarcoidosis, a multiorgan granulomatous inflammatory disease, likely results from an exaggerated T cell response to an airborne antigen.  A genetic predisposition to sarcoidosis has long been posited, and independent genome scans in German and African-American affected sib pair samples suggest that multiple genes are involved.  African-Americans are more commonly and severely affected by sarcoidosis, which imply that genes of African ancestry play a significant role in the disease etiology and pathogenesis.  Recent characterization of ancestry informative markers across the genome now makes it feasible to scan the genome for disease genes linked to ancestry in African-American populations. As a research group that has extensively studied the genetic susceptibility of sarcoidosis in African Americans, we have accumulated DMA samples for 1,302 African-American sarcoidosis cases. Many of these cases have participated in one of three previous NIH-funded studies, two family studies and one case-control, that provide a wealth of clinical and epidemiologic data in addition to a DNA sample. From these three studies, we also have DNA and epidemiologic data on 695 African Americans without sarcoidosis who will serve as a control sample. Using these samples, we propose a mapping by admixture linkage disequilibrium (MALD) study to identify sarcoidosis genes linked to African ancestry. The study will involve a multi-staged genome-wide scan targeting specifically those genes of African origin in African Americans that predispose to sarcoidosis susceptibility and radiographically persistent disease. We plan to first screen the genome using a set of 1,536 SNP markers evenly spaced approximately 1.9 cM throughout the genome that are highly informative European - African ancestry differences. In the second stage, we will triple density genotype ancestry informative markers to increase statistical confidence in the results and refine the positions. We will then move to a targeted haplotype-based association study in the most interesting regions. Once we have narrowed the associated genomic areas to specific genes or areas within specific genes, we will sequence the areas that have the highest probability of harboring causal variant(s). In addition, to better understand how putative candidate genes we identify act in sarcoidosis causal pathways involving environmental inciting agents, we will utilize comparable environmental data collected across the three study samples to test for gene-environment interaction. Our proposed study has the potential to uncover genes of modest effect not easily detectable by linkage and may in some instances actually be more statistically powerful than traditional case-control association methods.

Principal Investigator: Wegienka, Ganesa Rebecca Ph.D.
Epidemiology of Regulatory T Cells in Pregnancy and Childhood Atopic Diseases (NIH 1R21AI069271-01A1)

Childhood allergic diseases are a growing epidemic and are associated with considerable morbidity and mortality. Until now, most research has focused on postpartum exposures that may in part cause childhood allergic diseases. Unexplained findings related to birth order and recent findings on immunity in pregnancy have recently also pushed in utero programming (prenatal exposures) to the forefront in studies of causation. Thirty percent of cases of childhood allergic diseases have been attributed to a child's birth order. But, the underlying mechanism cannot be explained by the hygiene hypothesis, because there is little evidence of covariation of infection rates with birth order. Allergic status has been characterized by the T-helper1/T- helper2 ratio (Th1/Th2 ratio) skewed toward a Th2 cell-cytokine predominance, with the role of Th1 in the allergic response not yet defined. Pregnancy has also been assumed to be a Th2 dominant process. Thus pregnancy is likely a critical component in the causal web of childhood allergic diseases. Preliminary data indicate that T regulatory cells (Tregs), which suppress allogenic responses against the fetus in mice and humans, increase in successful pregnancy and decrease, but remain above prepregnancy levels, during the postpartum. Recent research has indicated that Tregs can control both Th1 and Th2 responses. Although a mode of tolerance transference from mother to fetus has not yet been identified, this could explain the effect of birth order on subsequent allergic risk that has been seen. Hence the question: what is the role of maternal Tregs during pregnancy in the risk of childhood allergic diseases? As part of an ongoing NIH- funded study, a cohort of pregnant women is being recruited for longitudinal study of their children at Henry Ford Health System to study the role of early life exposures in the development of childhood allergic disease (postnatal programming). Using a subset of 180 mother-child pairs from this ongoing cohort study, the following specific aims will be studied to detail the role of prenatal programming in childhood allergic disease: 1. Determine the relationships between maternal Tregs (CD4+CD25+CTLA4+ and CD4+CD25+FOXP3+ cells - counts and percentages) during pregnancy and: Tregs in their child's blood at delivery (cord), and 6 and 12 months; IgE in their child's blood at delivery (cord), and 6 and 12 months; and maternal IgE during pregnancy; 2. Quantify the within-woman change in Tregs from pregnancy (second trimester) throughout the postpartum period (1, 6 and 12 months postpartum); and 3. Determine the relationship between pregnancy history (prior pregnancy intervals and birth outcomes) and: maternal Tregs during pregnancy and 1, 6 and 12 months postpartum; and the child's Tregs at delivery (cord blood) and 6 and 12 months.

Principal Investigator: Wegienka, Ganesa Rebecca Ph.D.
Regulatory T Cells in Gestation and Childhood Allergic Disease (NIH 1K01AI070606-01A2)

This Mentored Research Scientist Development Award in Epidemiology and Outcomes Research (K01) will allow me to transition from reproductive epidemiology to childhood allergic disease (CAD) epidemiology research under the guidance of a successful CAD epidemiology research group. Through this proposed research and training program, I will develop the skills needed to become an independent researcher in the field of CAD. CAD are a growing epidemic and are associated with mortality and considerable morbidity. Thirty percent of cases have been attributed to the children's birth order. But, the underlying mechanism cannot be explained by the hygiene hypothesis, because there is little evidence of covariation of infection rates with birth order. Allergic status has been characterized by the T-helper1/T-helper2 ratio (Th1/Th2 ratio) skewed toward a Th2 cell-cytokine predominance, with the role of Th1 in the allergic response not yet defined. Pregnancy is assumed to be a Th2 dominant process. Thus pregnancy is likely a critical component in the causal web of CAD. T regulatory cells (Tregs), which suppress allogenic responses against the fetus in mice and humans, increase in successful pregnancy and decrease, but remain above pre-pregnancy levels, during the postpartum. Research has indicated that Tregs can control both Th1 and Th2 responses. Although a mode of tolerance transference from mother to fetus has not yet been identified, this could explain the observed association between birth order and subsequent CAD risk. Hence the question: what is the role of maternal Tregs during pregnancy in the risk of CAD? As part of an ongoing NIH-funded study, a birth cohort is being recruited for longitudinal study in the Detroit area to study early life exposures in the development of CAD. Using a subset of 225 mother-child pairs from this cohort, we will study the following hypotheses (Tregs: CD4+CD25+FOXP3+ and CD4+CD25+CTLA4+). Our hypotheses are: 1) More prior live births and shorter pregnancy intervals will be predictive of: a. Higher maternal Tregs during pregnancy and at 1, 6 and 12 months postpartum; b. Lower maternal prenatal IgE; and 2) Higher maternal Tregs during pregnancy are predictive of: a. Higher Tregs in their child's blood at delivery (cord), 6 and 12 months and 2 years; b. Lower IgE in their child's blood at delivery (cord), 6 and 12 months and 2 years; c. Reduced risk of their child having a positive skin prick test for common aeroallergens and food allergens at age 2 years.

Center for Health Services Research

Top

1. Senior Research Staff

Jennifer Elston Lafata, Ph.D.
David Nerenz, Ph.D., Director
Manel Pladevall, M.D.

2. Research Summaries

Principal Investigator: Elston Lafata, Jennifer, Ph.D.
Physician Recommendation and Colorectal Cancer Screening (NCI 1R01CA112379-01A1)

Data from our own efforts and those of others indicate that different physicians discuss different things when recommending cancer screening. The U.S. Preventive Services Task Force (USPSTF) while not endorsing a specific style of physician-patient interaction, recently advocated for the use of shared decision-making when making preventive service recommendations to individual patients. We propose to use a mixed-method approach that includes both qualitative and quantitative data collection and analyses to understand the use and utility of different aspects of shared decision-making when physicians recommend CRC screening in primary care. First, we will use direct observation and audio-recording of routine health maintenance visits (n=900) to characterize the nature and content of CRC screening discussions in primary care (Aim 1). Results from these qualitative efforts will be used to derive quantitative measures characterizing physician-patient discussions of CRC screening in primary care. In the second stage, we will join these qualitatively derived measures characterizing physician-patient discussions of CRC screening with automated claims/laboratory data to determine their relationship to 12-month post-visit CRC screening use (Aim 2). A combined pre- and post-visit patient survey will allow us to assess the concordance of patient preferences for screening modality, information, and participation in decision-making with what occurs in the observed visit. We will then be able to evaluate the effect of these concordances on 12-month post-visit CRC screening use (Aim 3). We will use the Henry Ford Health System (HFHS), a large integrated delivery system serving Detroit and its surrounding suburbs, as the setting for these efforts. The diversity and size of the HFHS primary care physician and patient populations combined with the comprehensive, existing automated data systems makes this an ideal environment in which to conduct the proposed work. By joining detailed accounts of the contribution of both physicians and patients to CRC screening discussions in the real world of primary care with subsequent CRC screening use and patient reported preferences, results from the proposed project will inform a new generation of interventions aimed at improving CRC screening participation. They will provide valuable information for training both physicians and patients how to effectively discuss CRC screening as well as facilitate the development of decision aides and policies for diverse populations.

Principal Investigator: Elston Lafata, Jennifer, Ph.D.
The Impact of Diabetes Control on Employment and Work Productivity (NIH 1R56DK074717-01A1)Diabetes is a leading cause of mortality and morbidity.  The prevalence of diagnosed diabetes has increased from 4.9% in 1990 to 7.9% in 2001 in the United States (1).  With the aging of the population, changing ethnic makeup, and increasing obesity, the prevalence of type 2 (non insulin-dependent) diabetes is expected to increase even further (2,3), so that up to 1 in 3 people born in the year 2000 will develop diabetes in their lifetime (4).  Diabetes is thus a major focus of health economics research.  While many studies examine the direct (medical) costs of diabetes, far fewer studies examine how diabetes affects labor market outcomes, such as disability, workforce participation, and absenteeism. It has been estimated that foregone productivity due to diabetes is substantial, at nearly $40 billion in 2002 (5).  However, there is little research on how much of this lost productivity is potentially preventable via optimal diabetes management.  Indeed, although an extensive literature has shown that successful blood glucose, lipid and blood pressure management can prevent or delay the onset and progression of diabetic complications, we know little about how this translates into effects on labor market outcomes among patients with diabetes, and nothing about their impact on future labor market outcomes such as labor force participation and work productivity.  We therefore propose a longitudinal study to examine the effects of levels of clinical risk factors on labor market outcomes among individuals with diabetes who are initially in the labor force (defined as being employed or looking for work).  Our study has the following specific aims: 1) Using the automated data sources of an integrated delivery system to establish a cohort of patients with diabetes and to illustrate the feasibility of administering a mixed-mode, baseline survey to these patients at the time of HbA1c testing;  2)          Evaluate the cross-sectional relationships between risk factor (i.e., glycemic and lipid) levels and work productivity (defined broadly to include average weekly hours worked, work-loss days (absenteeism), and health-related work limitations at baseline. Also evaluate the associations between historical diabetes risk factor levels (available through automated data systems) and baseline work productivity.  3) Conduct a one-year, follow-up patient survey to demonstrate the feasibility of linking longitudinal patient follow ups to HbA1c testing receipt and to examine the impact of baseline risk factor levels on employment and work productivity at follow-up.  Our proposed project uses data from both existing clinical records and new participant surveys.  We will use comprehensive automated data systems (medical records, administrative data, and laboratory data) available within a large multi-specialty medical group practice in southeast Michigan to identify a cohort of health plan members with diabetes in the labor force and their glycemic, lipid, and blood pressure levels.  These patients will be surveyed at baseline and at 1-year follow-up to monitor their labor market outcomes and to provide preliminary data illustrating the feasibility of a multi-year longitudinal study.  Principal Investigator: Nerenz, David, Ph.D.
RE-Engineering the Clinical Research Enterprise: Feasibility of Integrating and Expanding Clinical Research (NIH, Subcontract)

The project's primary aim is to develop the Coordinated Clinical Studies Network (CCSN), a research facility for clinical and health services research that builds on the current capacity of the HMO Research Network (HMORN).  The HMORN is a national network of leading research centers committed to undertaking collaborative, public-domain research program designed to improve the practice of health care in the United States.  HMORN member sites are uniquely positioned to answer the translational research questions posed by Haynes [1999]: 1) does an intervention work?  2) will it work in practice?  and, 3) how much will it cost?  The infrastructure being developed will increase HMORN sites' ability to respond to questions of national research interest, increase the pace that efficacious interventions are moved into practice, and improve research applicability across the diversity of real-world health care delivery systems.  To that end, the project involves adding to existing capacity in the areas of project planning and development, project implementation, systematized data collection and monitoring, project closeout and post-study surveillance, and dissemination including translation where appropriate.  The specific focus of work at Henry Ford has been the development of a procedures manual for multi-site cluster randomized trials.

Principal Investigator: Pladevall, Manel, M.D.
The Clinical Effectiveness of Pharmacy Adherence Information For Diabetes Control (NIDDK 2R01 DK064695-04A1)

Nonadherence to medications is common among patients with diabetes and contributes to suboptimal control of glycemic and lipid plasma levels. Adherence is not routinely measured in clinical practice because no valid, feasible methods have been readily available. The lack of medication adherence information contributes to clinician failure to identify and address patient nonadherence and to clinical inertia and poor health outcomes. Existing electronic prescribing systems hold the potential to display medication adherence information. We propose a 2-arm cluster-randomized trial to test the effectiveness of providing PCPs with both adherence measurements and an adherence clinic to improve adherence to diabetic and lipid-lowering drugs. This adherence clinic will consist of a pharmacist and nurse trained in motivational interviewing (Ml) techniques to improve adherence to medications. Adherence indices will be generated by linking e- prescribing information with pharmacy data. The trial will be conducted among 1,436 patients with diabetes and poor blood glycemic and/or lipid control. Primary care providers will be randomized to one of the following two study arms: 1) usual care - PCPs will write prescriptions electronically but will not be provided patient adherence information or Ml support; and 2) intervention - PCPs will be provided adherence information and adherence prompts electronically plus physicians and patients will receive support from an adherence clinic. Our intervention uses as theoretical behavioral framework elements of the Chronic Care Model, Self-Determination Theory, and the Health Belief Model. The study will use qualitative methods to guide intervention design and implementation and will include both process evaluation and treatment fidelity measures. The intervention will be tailored to patients' adherence and goal levels. Outcomes will include adherence to diabetes and lipid-lowering medications; cholesterol and glycated hemoglobin plasma levels (primary outcome); patients and providers' acceptance and satisfaction; and cardiovascular morbidity- mortality (exploratory outcome). The study will also evaluate the cost effectiveness of the intervention. Patients will be followed for 36 months. Analyses will control for cluster randomization effects. The introduction of sustainable medication adherence monitoring in clinical practice holds great potential to improve health outcomes among patients with diabetes.