[Skip to Content]
[Skip to Content Landing]
Invited Commentary
August 2015

Risk for Type 2 Diabetes Mellitus: Person, Place, and Precision Prevention

Author Affiliations
  • 1Center for Health and Community, University of California, San Francisco
JAMA Intern Med. 2015;175(8):1321-1322. doi:10.1001/jamainternmed.2015.2701

Unprecedented increases in obesity in the United States have contributed to greater prevalence of diseases, such as type 2 diabetes mellitus (T2DM), which impair the quality of life and reduce the longevity of affected individuals, create demands on already-strained health care delivery systems, and generate greater health care costs. Observations gleaned from earlier successes in reducing the rates of smoking and smoking-related diseases can inform efforts to reverse this trend. However, smoking reductions were not accomplished primarily by health care activities. Although interventions in primary care settings played an important role,1 other actions, ranging from media campaigns to policies involving advertising bans, taxation, and smoke-free areas, were critical to changing the dynamics of cigarette use. The behavioral causes of obesity—diet and exercise—are even more strongly rooted in factors outside the health care system. Traditional medical treatments alone cannot substantially lower the prevalence and impact of obesity without changes in the obesogenic environment. Addressing environmental causes recasts diet and exercise as behaviors that are not only a function of individual choice and will power but that are strongly shaped by the resources and obstacles encountered in the environments in which behaviors are enacted.2 The neighborhood effects measured by Christine and colleagues3 documented that individuals residing in neighborhoods marked by limited resources for healthy eating and physical activity (PA) are at higher risk for being diagnosed with T2DM. Based on a rigorous prospective, longitudinal design, their research substantiates the claim that the physical and social contexts of neighborhood environments matter for disease onset.

Although we do not yet know what elements of neighborhoods are most essential for generating better health, these researchers provide important clues about which elements have an effect and for whom. One notable finding was that T2DM onset was not predicted by objective measures of the physical environment (eg, geographic information system–based counts of the density of food stores and commercial recreational facilities) but by shared perceptions of people in the community regarding the availability of healthy food and nearby places to be active. This result is consistent with those of other studies that found relatively weak associations between the proximity of supermarkets and outcomes such as body mass index (BMI) and fruit and vegetable consumption.4 Unless the available resources are widely known and viewed as accessible, “if you build it, they will come” may not apply.

A second informative finding was that environmental resources were differentially related to T2DM risk for individuals with higher vs lower incomes. Low-income participants living in areas viewed as having less social cohesion and being less safe for walking were more likely to develop T2DM. Such views were less relevant for high-income individuals. The latter may not need to worry about the social climate of their neighborhoods in relation to PA. Financial resources enable them to join fitness centers that provide safe spaces for exercise, whatever their neighborhood environment is like. In contrast, lower-income individuals have fewer options for exercise. They may not be able to afford gym fees, and concern about the safety of their area may inhibit outdoor activity. Low-income individuals residing in neighborhoods with less social cohesion may also encounter more threat and conflict, and the resulting chronic stress could independently increase T2DM risk through effects on sleep, eating, and neuroendocrine response.5 Although safety and social cohesion more strongly predicted T2DM onset for low- than high-income individuals, resources for PA were more strongly linked to T2DM among high- than low-income individuals. This finding emerged from the researchers’ use of a geographic information system–based measure of resources that included only commercial establishments. The cost of using these facilities may have rendered them inaccessible to those with less income, making their physical proximity irrelevant. Access to free or low-cost gyms, parks, swimming pools, and safe places to walk or run would probably be more likely to contribute to increasing PA among low-income individuals than proximity to commercial establishments.

Overall, measures of the food environment were less strongly related to T2DM onset than were PA resource measures. This result could reflect measurement differences, but diet might be harder to modify via neighborhood resources than exercise. Food preferences and practices may be established earlier in life and shaped by other factors, such as the relative cost of healthy vs unhealthy foods, even if both are physically proximal.

In sum, the findings by Christine et al3 point to the impact of perceived neighborhood resources. Having markets and recreational facilities located nearby may be necessary but not sufficient to enable healthy behaviors. Building more facilities in neighborhoods that lack them is a component of an overall strategy to address the national rise in obesity, but this strategy needs to be informed by an understanding of when such facilities are actually used and the characteristics of the individuals who use them. In brief, the risk for T2DM is a combination of both person and place, and our national strategies need to understand and intervene across these levels.

A multilevel approach that encompasses an understanding of person and place is equally pertinent for providing more effective health care and could build on the conceptual framework underlying precision medicine.6 Precision medicine is based on taxonomies of disease that begin with genetic risk and incorporate subsequent levels (eg, epigenome, metabolome) that theoretically extend to nonbiological factors captured in the exposome. To date, precision medicine has focused largely on the genomics of tumors in relation to cancer treatment. Incorporating data on psychosocial, behavioral, and environmental factors should become more compelling when applied to T2DM and other chronic conditions.

If precision medicine is to fulfill its potential to improve health, it needs to encompass the entire range of determinants of disease onset and progression, ranging from a person’s genetic code to his or her zip code. The moment right now is opportune for the development of a strategy for expanding consideration of data from higher levels of aggregation. Unsustainably high health care costs, an aging population, and projected increases in obesity-related diseases call for more effective prevention efforts. New payment models, such as Accountable Care Organizations, in which health care organizations share in the savings they achieve by providing efficient, high-quality services, are creating incentives for clinicians and health systems to prevent as well as treat disease. To succeed, they will need to address the most important determinants of the health of the populations they serve, and precision medicine will need to encompass precision prevention.

Information regarding a patient’s social and environmental context may become more readily available to clinicians and health systems. A recent Institute of Medicine report7 recommended that, among other measures of social and behavioral determinants of health, personal financial resources and residential address should be routinely collected in all electronic health records. Such information could improve risk stratification of patients and strengthen shared decision making regarding suggested regimens, including those designed to manage or reduce weight and/or increase exercise.

Partnerships between health care systems, which can capture biological, psychosocial, and demographic characteristics of patients, and public health and city planners, who understand neighborhood characteristics, will be needed to quell the rising tide of T2DM and related chronic conditions and their consequent health care costs. Including this information in an expanded vision of precision medicine that encompasses prevention could provide new strategies for assessing and intervening on the entire range of modifiable health determinants.

Back to top
Article Information

Corresponding Author: Nancy E. Adler, PhD, Center for Health and Community, University of California, San Francisco, 3333 California St, Ste 465, San Francisco, CA 94143 (nancy.adler@ucsf.edu).

Published Online: June 29, 2015. doi:10.1001/jamainternmed.2015.2701.

Conflict of Interest Disclosures: None reported.

Additional Contributions: Maria Glymour, ScD, Department of Epidemiology and Biostatistics, University of California, San Francisco (UCSF), and Laura Gottlieb, MD, MPH, Department of Family and Community Medicine, UCSF, provided thoughtful input to this commentary. They did not receive any compensation for this contribution.

Rigotti  NA.  Strategies to help a smoker who is struggling to quit.  JAMA. 2012;308(15):1573-1580.PubMedGoogle ScholarCrossref
Adler  NE, Stewart  J.  Reducing obesity: motivating action while not blaming the victim.  Milbank Q. 2009;87(1):49-70.PubMedGoogle ScholarCrossref
Christine  PJ, Auchincloss  AH, Bertoni  AG,  et al.  Longitudinal associations between neighborhood physical and social environments and incident type 2 diabetes mellitus: the Multi-Ethnic Study of Atherosclerosis (MESA) [published online June 29, 2015].  JAMA Intern Med. doi:10.1001/jamainternmed.2015.2691.Google Scholar
Cummins  S, Flint  E, Matthews  SA.  New neighborhood grocery store increased awareness of food access but did not alter dietary habits or obesity.  Health Aff (Millwood). 2014;33(2):283-291.PubMedGoogle ScholarCrossref
Kelly  SJ, Ismail  M.  Stress and type 2 diabetes: a review of how stress contributes to the development of type 2 diabetes.  Annu Rev Public Health. 2015;36:441-462.PubMedGoogle ScholarCrossref
Committee on a Framework for Development of a New Taxonomy for Disease, Board on Life Sciences, Division on Earth and Life Studies, National Research Council. Total Precision Medicine: Building a Knowledge Network for Biomedical Research and a New Taxonomy of Disease. Washington, DC: National Academies Press; 2011.
Institute of Medicine.  Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: National Academies Press; 2014.