Skip to main content

Observational Studies

Our investigators are working on a number of ongoing projects; explore their current efforts below.


Coronary Artery Risk Development in Young Adults

Northwestern Principal Investigator: Donald Lloyd-Jones, MD, ScM

CARDIA is a study examining the development and determinants of clinical and subclinical cardiovascular disease and their risk factors. It began in 1985-1986 with a group of 5,115 black and white men and women, ages 18-30 years, in four sites located throughout the U.S., including Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; and Oakland, California.

They were asked to participate in nine follow-up examinations over the course of 30 years. While the specific aims of each examination have varied, data have been collected on a variety of factors believed to be related to heart disease. These include conditions with clear links to heart disease, such as blood pressure, cholesterol and other lipids, as well as glucose. Data have also been collected on physical measurements such as weight and body composition, in addition to lifestyle factors such as dietary and exercise patterns, tobacco and alcohol use, behavioral and psychological variables, medical and family history and other variables (e.g., insulin).

In addition, subclinical atherosclerosis and myocardial remodeling have been measured via echocardiography, chest and abdominal CT scans and carotid ultrasounds at various time points. A brain MRI was performed on a subset of participants during the last two exams. The CARDIA study is continuing to collect information from participants on a semi-annual basis with the goal of continuing the clinical exams every five years.

Longitudinal Epigenomic Study of Cardiovascular Health Metrics and Cardiac Structure and Function in CARDIA

Northwestern Principal Investigators: Lifang Hou, MD, PhD, and Donald Lloyd-Jones, MD, ScM

This study is nested in CARDIA and profiled DNA methylation measured in blood samples collected from a random subset of 1,000 CARDIA participants at study years (Y) 15 and 20, corresponding to mean subject ages of 40 and 45 respectively. These data currently support a wide variety of projects, examining the molecular effects of exposures and behaviors in early life that are known to affect cardiovascular health (CVH) as well as the molecular antecedents of detectable cardiovascular disease (CVD). But the principal objective of this study is to comprehensively examine CVH throughout youth and middle age, link it to middle-age epigenetic changes and link those changes in turn to subclinical CVD as it develops in mid-life and older age. This study is also comprehensively examining epigenome-wide associations as well as newer concepts involving epigenetic dysregulation such as repetitive DNA methylation at specific regions and loci and epigenetic aging.


Chicago Healthy Eating Environments and Resources Study

Northwestern Principal Investigator: Kiarri Kershaw, PhD, MPH

Improving neighborhood access to healthy foods is a priority for public health departments across the country as part of an effort to give communities the resources needed to make positive changes in their eating behaviors that improve their overall health and lower obesity. However, recent research suggests improving healthy food availability alone may not be sufficient to improve eating behaviors or reduce obesity risk. Further work is needed to understand how individual-environment interactions influence eating decisions.

To meet this need, we developed CHEERS. Between September 2016 and October 2017, we collected data about eating behaviors and food availability in four neighborhoods in the city of Chicago to better understand how women aged 18-44 years use their environment, available resources and each other to make eating decisions that influence overall health and obesity. Our study is important because a better understanding of how people interact with their environments and how these interactions influence health will help guide the development of more effective policies and interventions to promote healthful behavior change.

We chose to focus on women because they are typically responsible for the food preparation and purchasing for their families and thus represent a critical target for developing interventions to improve eating behaviors and lower obesity. We selected this age group because it is a critical period of increased weight due to various factors, including post-pregnancy weight retention and declining muscle mass and muscle strength. We collected information directly from women about eating and food shopping behaviors. We also measured height, weight, body composition and blood pressure and collected dried blood spots.

While distance to the nearest grocery store may be one factor that influences food choice, other factors such as price of groceries and placement or advertising of healthy vs. unhealthy food items may be even more important. In response, our research team members went to the neighborhood stores where participants report shopping to collect information about availability of foods, pricing and product placement within the stores. We anticipate that CHEERS will represent the first step in a larger research agenda to enable Chicago families to make healthier eating decisions and reduce obesity and cardiovascular disease risk.


Chicago Heart Association Detection Project in Industry Study

Northwestern Principal Investigators: Norrina Allen, PhD, and Donald Lloyd-Jones, MD, ScM

The CHA study is a longitudinal prospective investigation of 39,522 men and women (Asian, Black, Hispanic, White and Other) aged 18-59 years at time of screening in 1967-1973 from 84 cooperating companies and organizations in the Chicago area. In the CHA study, screening involved collection of demographic, medical history and medical treatment data by questionnaire; measurement of height, weight, heart rate, casual blood pressure and ECG; collection of blood for measurement of serum total cholesterol and plasma glucose 1 hour after a 50 gram oral glucose load. Data from this study have resided in the Department of Preventive Medicine since the study’s inception. 

Currently the CHA cohort has been followed for over 40 years since the baseline examination in 1967-1973. Multiple methods of mortality ascertainment have been successfully used to determine the vital status of CHA participants. Since 1979, names, dates of birth and social security numbers of members of the CHA cohort have routinely been submitted to the National Death Index (and later NDIPlus) to determine vital status. The NDI followup has been augmented with additional methods, including routine mailings to participants, submission of data to the Social Security Administration, period submissions to Equifax, Inc., mailings to employers, inquiry by telephone and neighborhood contact. In addition, linked Medicare files and National Death Index from 1984 through 2015 are available through existing grants. 

 Epigenomic Biomarkers of HIV-Associated Cancers in Nigeria

An NCI-funded U54 consortium

Northwestern Principal Investigators: Lifang Hou, MD, PhD, and Robert Murphy, MD

Epigenomic Biomarkers of HIV-Associated Cancers in Nigeria is a joint U.S. and Nigerian consortium initiated in 2018 to study the epigenetic signatures of the two most common HIV-associated cancers in Nigeria (liver and cervical cancer) in 600 Nigerians. HIV’s effects on the immune system promote viral co-infections such as hepatitis and the human papilloma virus (HPV) greatly increasing cancer risk in HIV-infected individuals. The consortium’s two research projects seek to identify DNA methylation signatures in blood that are specific to liver and cervical cancers. DNA methylation can improve our understanding of the role of these infections in cancer development, offer earlier diagnosis of premalignant and early-stage cancers and potentially lead to effective new strategies for cancer prevention, diagnosis and treatment. This consortium also provides training and mentoring opportunities for Nigerian investigators in molecular epidemiology studies and especially epigenetics.


Methylomics of Prenatal Gestational Diabetes Mellitus

Northwestern Principal Investigator: Lifang Hou, MD, PhD

The Tianjin GDM Observational Study was established in 1999 in Tianjin, the fourth-largest city in China. After enrolling nearly 1,200 pregnant women, half of whom were diagnosed with GDM from 2005-2009, data was collected on maternal and child characteristics at birth and, later, blood samples to collect genetic data from mothers and children. This project added epigenetic DNA methylation data for all children in the study, and in so doing seeks to study the molecular effects of being exposed to GDM in utero. We also seek to use DNA methylation to profile risks of obesity and cardiometabolic phenotypes among children born to GDM mothers and to measure the efficacy of early-childhood lifestyle interventions aimed at preventing GDM-related metabolic sequelae later in life. These findings may provide novel tools to guide early childhood interventions through early biomarker monitoring and ultimately to implement personalized preventive measures among children exposed to GDM and others at elevated risk of obesity.


Hyperglycemia and Adverse Pregnancy Outcome Study

Northwestern Principal Investigator: Boyd Metzger, MD

HAPO was an observational study that addressed the hypothesis that hyperglycemia in pregnancy, less severe than overt diabetes, is independently associated with increased risk of adverse maternal and neonatal outcomes. HAPO recruited about 25,000 pregnant women in 15 field centers in nine countries from 2000 to 2006. The HAPO Study examined glucose tolerance at 24-32 weeks gestation. Other maternal data included demographics, blood pressure, height, weight, smoking and alcohol use during pregnancy and prenatal, perinatal and postnatal course of care.

Adverse outcomes were primary Cesarean delivery, birthweight > 90th percentile, clinical neonatal hypoglycemia, hyperinsulinemia defined as cord C-peptide > 90th percentile. Additional outcomes were neonatal percent fat > 90th percentile, neonatal sum of skinfolds > 90th percentile, preterm delivery (< 37 weeks gestation), preeclampsia, shoulder dystocia/birth injury and admission to neonatal intensive care unit or hyperbilirubinemia. In HAPO, higher levels of maternal glucose were independently associated with increased frequency of birthweight, cord serum C-peptide and infant adiposity above the 90th percentile.

Based upon these results, a consensus panel formulated the new International Association of Diabetes in Pregnancy Study Groups criteria for the diagnosis of gestational diabetes mellitus (GDM) that have been adopted by the World Health Organization.

Hyperglycemia and Adverse Pregnancy Outcome Follow-Up Study (HAPO FUS)

Northwestern Principal Investigator: Boyd Metzger, MD

Offspring of mothers with pre-existing diabetes mellitus or GDM have an increased risk of obesity in childhood. Moreover, GDM is associated with an increased maternal risk of type 2 diabetes. What has not been established is the risk of childhood obesity and metabolic disorders or maternal risk of disorders of glucose metabolism (diabetes, impaired fasting glucose, impaired glucose tolerance) and other cardiovascular risk factors (dyslipidemia, abdominal adiposity, elevated blood pressure) along the continuum of maternal glucose to levels diagnostic of diabetes. The HAPO FUS was designed to address whether hyperglycemia in pregnancy, less severe than overt DM, is independently associated with increased risk of adverse childhood and maternal outcomes 10 to 14 years later.

Building on the unique HAPO study resources, the HAPO FUS obtained measures of adiposity, glucose, insulin sensitivity and secretion, lipid profile, blood pressure and inflammation in more than 4,800 HAPO mother-offspring (ages 10-14) pairs from multiple ethnic/race groups in 10 of the 15 HAPO field centers at a study visit from 2013 2016. The primary predictor was maternal glycemia during pregnancy. Primary outcomes were childhood adiposity and maternal metabolic disorders 10 to 14 years later. Other childhood outcomes included glycemia, insulin sensitivity and secretion, lipids, blood pressure and inflammation. Other maternal outcomes included measures of cardiovascular risk.

HAPO FUS demonstrated that maternal GDM is associated with higher adiposity in children at ages 10 to 14, including obesity, percent body fat and sum of skinfolds. In addition, maternal GDM is associated with higher frequency of maternal metabolic disorders including Type 2 diabetes and impaired glucose tolerance 10 to 14 years following pregnancy. Additional analyses investigating associations of measures of neonatal adiposity and hyperinsulinemia with measures of adiposity, as well as glycemia, insulin sensitivity and secretion, lipids, blood pressure and inflammation in offspring at ages 10 to 14 are currently underway.


Impact of Chemotherapy on Activity Patterns in Breast Cancer Patients Undergoing Chemotherapy

Northwestern Principal Investigator: Siobhan Phillips, PhD

This is an observational study that uses three 10-day ecological momentary assessment “bursts” at the beginning, middle and end of chemotherapy to examine potential clinical (e.g., treatment dosage, treatment adherence), behavioral (e.g., self-efficacy, motivation) and psychosocial (e.g., fatigue, pain) determinants and outcomes of activity and sleep pattern changes during chemotherapy for breast cancer. Participants wear accelerometers 24/7 to objectively measure physical activity and sleep and complete three short questionnaires throughout the day about their motivation and symptoms during each 10-day burst. We are recruiting through two Northwestern affiliated practices: Prentice Women’s Hospital and Hematology Oncology Associates. This study will help to understand how physical activity changes during treatment and potential implications for treatment outcomes (i.e., adverse events, adherence). Funding: Lynn Sage Breast Cancer Research Foundation. Project status: Analyzing and publishing data.


International Study of Macro-and Micro-Nutrients

Northwestern Principal Investigators: Linda Van Horn, PhD, RD, and Jeremiah Stamler, MD

INTERMAP is a basic epidemiological investigation of relationships between micro- and macronutrient intakes, timed 24-hour urinary biochemistry and metabolic phenotypes and BP. Data was originally collected from 1997-1999 on 4,630 women and men ages 40-59 (including metabolomics data) from diverse population samples in the U.S. (N=2,164), China (PRC, N=832), Japan (N=1,138) and the UK (N=496) at four visits.

Standardized quality-controlled data collection included eight BP measurements (two per visit); four interviewer-administered in-depth multi-pass 24-hour dietary recalls providing data on daily intake of 83 nutrients/person; questionnaire data on occupation/education/other sociodemographic traits, past and family medical history, medication use, special diet by type, dietary supplements, smoking, alcohol use (obtained with each of the four 24-hour dietary recalls, plus two histories of daily intake during the preceding seven days), data on other possible confounders, including for women on parity/menopausal status/estrogen use; height and weight measurement at two visits; two timed 24-hour urine collections on average three weeks apart, the first between visits 1 and 2, the second between visits 3 and 4. Urine specimens are stored in the London Metabolomics Laboratory, Imperial College London in -80°C freezers. The London Laboratory completed urinary metabolome-wide H NMR spectroscopic profiling of urine specimens from both first and second 24-hour urine collections.

The general Aim of the study is to advance knowledge of biochemical pathways involved in BP regulation by identifying urinary metabolic phenotypes associated with the direct effect on BP of high Na intake and inverse BP effect (BP reduction) of the DASH/OmniHeart-protein eating pattern. To achieve this aim, we will identify and quantify the key metabolites related to these contrasting influences on BP and will use state-of-the-art chemometrics, statistical spectroscopy, computational networks and pathway modeling tools to identify and map de novo pathways associated with Na and DASH/OmniHeart-protein and BP. We will then test and validate in external data the INTERMAP derived metabolites and pathways using available dietary and urinary data from the OmniHeart Trial, Urinary Sodium Study (USS) and follow-up data in the INTERMAP China Prospective (ICP) Study. The ultimate goal is to develop more focused and effective strategies for population-wide BP lowering through improved non-pharmacologic approaches (primarily nutritional) and to identify potential new targets for drug treatment.


The Lifetime Risk Pooling Project

Project manager: John Wilkins, MD

The LRPP dataset was created to describe the development of cardiovascular disease risk across the adult life course. In order to generate long-term risk estimates across different race and sex groups and across different the full spectrum of cardiovascular risk factors, we pooled individual-level data from 20 U.S. population-based CVD cohort studies. In total, the dataset contains data from over 275,000 unique individuals and in excess of 3,800,000 person-years of followup. All of the studies included in the dataset performed in-person assessment of participant characteristics and cardiovascular risk factors (e.g., height, weight, blood pressure, cholesterol). Similarly, all included studies contain CVD endpoints and vital status. Given the inherent heterogeneity in some variables across studies (e.g., dietary patterns), extensive harmonization of data has been performed as well. The LRPP dataset is continuously updated as new data become available. Thus, the dataset contains contemporary epidemiological data on cardiovascular exposures and outcomes. 

The LRPP dataset has led to the discovery and dissemination of novel risk estimates, novel metrics for risk communication, enhanced understanding of compression of morbidity and novel insights into extreme or unusual phenotypic characteristics. To date, the LRPP has published in excess of 25 publications in NEJM, JAMA, JAMA Cardiology, Circulation, JACC, JAHA, Diabetes Care and IJE. Collaboration is highly encouraged; for more information please contact, Wilkins at   


Multi-Ethnic Study of Atherosclerosis

Northwestern Principal Investigator: Norrina Bai Allen, PhD

MESA is a study of the characteristics of subclinical cardiovascular disease (disease detected non-invasively before it has produced clinical signs and symptoms) and the risk factors that predict progression to clinical cardiovascular disease or progression of the subclinical disease. MESA scientists study a diverse, population-based sample of 6,814 asymptomatic men and women ages 45-84 from six field centers across the United States, including Northwestern University's Department of Preventive Medicine.

Approximately 38 percent of the MESA participants are white, 28 percent African-American, 22 percent Hispanic and 12 percent Asian, predominantly of Chinese descent. The first examination took place over two years, from July 2000 to July 2002. It was followed by three examination periods that were 17 to20 months in length and a fifth exam April 2010 to January 2012. The sixth exam began in 2016 and is currently being conducted. As part of the in-person examination, a variety of physical assessments and questionnaires are completed, including medical history, physical function, physical activity, anthropometry, blood pressure, current medications, echocardiogram, spirometry, cognitive function and blood collection to determine cardiovascular biomarkers and genetic information. In addition, MESA participants are invited to participate in a clinical trial, INVITE, testing the characteristics associated with vitamin D levels.  

The MESA study continues to make important contributions to our scientific understanding of disease. To date, MESA has published over 1,100 papers.

 Needs Assessment for Physical Activity Promotion and Sedentary Behavior Reduction Interventions Among Breast Cancer Survivors

Northwestern Principal Investigator: Siobhan Phillips, PhD

This is a cross-sectional survey that was conducted in the summer of 2015 to examine survivors’ (n=279) interest and preferences for technology-supported lifestyle interventions. We also explored potential social cognitive determinants of sedentary behavior. Funding: NUCATs pilot voucher. Project status: Analyzing and publishing data.


Women's Health Initiative

Principal Investigator: Linda Van Horn, PhD, RD

WHI is a long-term national health study focused on strategies for preventing heart disease, breast and colorectal cancer and osteoporotic fractures in postmenopausal women. Launched in 1993, the WHI enrolled 161,808 women ages 50-79 into one or more randomized clinical trials, testing the health effects of hormone therapy (HT), dietary modification (DM) and/or calcium and Vitamin D supplementation (CaD) or to an Observational Study (OS). At the end of the initial study period in 2005, WHI Extension Studies (2005-2010, 2010-2020) continued follow up with all women who consented.  

This groundbreaking study changed the way healthcare providers prevent and treat some of the major diseases impacting postmenopausal women. It has been estimated that results from the WHI Hormone Trials have saved $35.2 billion in direct medical costs in the U.S. alone. To date, WHI has published over 1,400 articles and approved and funded 289 ancillary studies. 

In 2015, WHI was awarded additional funding to continue follow-up of participants through 2020. Visit the WHI website for more information.

Follow Preventive Medicine on