The Impact of the Nurses’ Health Study on Population Health: Prevention, Translation, and Control

Abstract

Objectives. To summarize the overall impact of the Nurses’ Health Study (NHS) over the past 40 years on the health of populations through its contributions on prevention, translation, and control.

Methods. We performed a narrative review of the findings of the NHS, NHS II, and NHS3 between 1976 and 2016.

Results. The NHS has generated significant findings about the associations between (1) smoking and type 2 diabetes, cardiovascular diseases, colorectal and pancreatic cancer, psoriasis, multiple sclerosis, and eye diseases; (2) physical activity and cardiovascular diseases, breast cancer, psoriasis, and neurodegeneration; (3) obesity and cardiovascular diseases, numerous cancer sites, psoriasis, multiple sclerosis, kidney stones, and eye diseases; (4) oral contraceptives and cardiovascular disease, melanoma, and breast, colorectal, and ovarian cancer; (5) hormone therapy and cardiovascular diseases, breast and endometrial cancer, and neurodegeneration; (6) endogenous hormones and breast cancer; (7) dietary factors and type 2 diabetes, cardiovascular diseases, breast and pancreatic cancer, non-Hodgkin’s lymphoma, neurodegeneration, multiple sclerosis, kidney stones, and eye diseases; and (8) sleep and shift work and chronic diseases.

Conclusions. The NHS findings have influenced public health policy and practice both locally and globally to improve women’s health.

As noted in the individual articles in this series, the Nurses’ Health Study (NHS) has evolved to address a broad range of lifestyle factors through questionnaires and biomarkers,¹ including diet, hormones, and trace elements. The study has also expanded from its initial funding to address breast cancer to include many other chronic conditions: type 2 diabetes, cardiovascular disease, fractures, rheumatological conditions, eye diseases, and other endpoints of interest in women’s health. The NHS has informed an array of research areas including studies on obesity, kidney stones, skin conditions, less-common cancers, psychological factors (e.g., depression), and neurodegenerative diseases.

This expansion adds to the return on investment building the rich NHS cohort data through funding by the National Cancer Institute.² It also requires a continuity of funding to support the breadth and validity of health outcomes. To enable these efforts, the NHS has experimented with diverse innovations in infrastructure to maximize effectiveness and cost-efficiency, with many of these logistical innovations becoming common practice for other large epidemiological studies.³ One of the most significant contributions is the use of optically scanned questionnaires and linkage to the National Death Index to complement active follow-up. However, among the most significant innovations introduced in NHS was the repeated assessment of habitual diet. The NHS has made key contributions in nutritional epidemiology research with the creation and validation of the Harvard Semiquantitative Food Frequency Questionnaire.⁴ This improved efficiency has enabled the NHS to create an exceptional, large database of comprehensive, long-term, multidimensional information.

The use of data from prospective cohort studies such as the NHS has previously been addressed with regard to the overall impact of the National Cancer Institute–funded research program in cancer epidemiology.⁵ That report focused primarily on the phases of the discovery, development, and delivery paradigm of cancer research. It also emphasized the importance of research findings from cohort studies and the need for their continued support. The existing National Cancer Institute–funded prospective cohorts, like the NHS, continue to provide key data that guide public health and clinical practice across many chronic conditions.² Evidence from cohort studies can help explain the etiology of disease with fewer sources of bias then other etiological designs. The broader contributions of cohorts in advancing our understandings of lifestyle and prevention of chronic conditions have been thoroughly summarized.^6,7

This article summarizes some of the distinct contributions of the NHS and how the cohort has adapted to changing public health issues. We used the framework of discovery, development, and delivery applied to epidemiology.⁵ More extensive details are presented in the individual updates in this series.

DISCOVERY (DISEASE ETIOLOGY)

The purpose of epidemiological studies relating to discovery focuses on explaining the etiology of diseases and health conditions through hypothesis testing and identification of new risk factors.⁵ The NHS is an example of a cohort that has sustained remarkable scientific productivity in the past 40 years, including more than 1200 publications that have substantially influenced prevention recommendations by many organizations, including the American Cancer Society, American Heart Association, Dietary Guidelines for Americans, and the World Health Organization.³ The cohort follow-up has supported analysis of traditional risk factors through repeated measurements including change in weight^8,9 time since quitting smoking,¹⁰ and change in diet, as well as contributed to the discovery and examination of additional risk factors associated with diet, physical activity, other lifestyle factors, biochemical pathways, and genetic data. In sum, such long follow-up allows the study and detailed analysis of the links between long-term or life-long exposures and diseases with very long lag times.

With regard to endogenous hormones, the NHS cohort, along with other cohorts, confirmed that circulating levels of estrogens and androgens were significantly associated with a higher risk of breast cancer before age 50 years,¹¹ and higher postmenopausal estrogen, androgen, and prolactin levels were related to estrogen receptor–positive breast cancer.¹²

Investigators have used the NHS to examine the role of specific biomarkers in disease risk. For example, the cohort identified novel biomarkers for incident type 2 diabetes, including adipokines, inflammatory cytokines, nutrition metabolites, and environmental pollutants.^13–17 Investigators also documented that higher levels of plasma carotenoids (present in fruits and vegetables, including alpha-carotene, beta-carotene, and lycopene) are inversely associated with breast cancer risk¹⁸ and total folate intake 12 to 16 years before diagnosis is associated with reduced risk of colorectal cancer.^19,20

The NHS cohort has also contributed to advances in genomic research and the investigation of genetics with regard to disease risk.²¹ Genetic data have led to the identification of more than 90 common risk loci for breast cancer; the variants at these loci explain approximately 16% of familial risk of breast cancer.²² For colorectal cancer patients with PIK3CA mutations, those who used aspirin regularly had improved cancer-specific survival and overall survival; however, among patients with wild-type PIK3CA, regular use of aspirin was not associated with cancer-specific or overall survival.²³ These genetic findings suggest that there should be greater importance placed on targeted interventions.

At a broader societal level, with NHS participants extending across the United States, opportunities have emerged for relating environmental exposures to health outcomes, particularly with regard to obesity²⁴ and type 2 diabetes.¹⁶ Investigators have also assessed exposure in relation to built environment physical activity and body mass index (weight in kilograms divided by the square of height in meters).²⁵ Together, these examples demonstrate that individual biomarker level measures, behavior measures, and exposures at the group level (to built environment and air pollution) can all be integrated to refine our understanding of disease etiology.

DEVELOPMENT (THE BASIS FOR CONTROL MEASURES)

Epidemiological findings are a critical component of established approaches to the assessment of disease causality. These findings aim to provide a scientific basis for developing control measures and prevention strategies for groups and populations at risk and to develop needed public health measures and practices.⁵ The NHS cohort has contributed to continued risk assessments and recognized additional exposure associations that help inform development of health recommendations.

The cohort has demonstrated the public health impact and potential for prevention by summarizing evidence on risk factors. Approximately 90% of type 2 diabetes cases may be prevented by diet and lifestyle modifications alone.²⁶ In addition, about 80% of CHD incidence could be prevented by avoidance of smoking, consuming a healthful diet, engaging in moderate to vigorous physical activity for at least 30 minutes most days, and consuming alcohol moderately (half to 1 drink per day).²⁷

Studies also reported findings relating to cancer risks that are associated with lifestyle choices earlier in life. Breast cancer risk was confirmed to be associated with alcohol consumption in early and later adult life, even at low levels of consumption.²⁸ The risk of premenopausal breast cancer among women is higher when paired with greater consumption of red meat in high school²⁹ but risk is lower among women with higher intakes of fiber and fruit during adolescence.^30,31 The NHS has also identified novel lifestyle, dietary, environmental, serological, and genetic risk factors with regard to less-common cancers (endometrial, ovarian, pancreatic, and hematological). In addition, the cohort has helped clarify the importance of timing of exposure across the life course, such as earlier or later-in-life body size, to those risk associations and assessed heterogeneity or lack thereof, in etiological associations across discrete tumor subtypes.³²

The expanding scope of the cohort has allowed investigators to explore the relations between lifestyle and neurodegeneration and associated diseases. Results demonstrated that greater intake of antioxidants,³³ higher nut intake,³⁴ and following the Mediterranean diet³⁵ are all associated with higher cognitive functioning.

The NHS has contributed to continued risk factor assessments over extended periods of time. Repeated assessments of exposures over time have included weight, smoking, and dietary patterns. For example, the NHS cohort has confirmed that excess adiposity is the strongest risk factor for type 2 diabetes,³⁶ and weight across the life course and obesity are strongly but variably associated with risk of cardiovascular disease,^9,37 breast cancer,⁸ endometrial cancer,^38,39 and pancreatic cancer.⁴⁰

Risk prediction models also integrate risk factors to guide stratification and prioritizing of risk-reduction strategies. The NHS data are the basis for models of cardiovascular disease,⁴¹ melanoma, and cancer of the breast,⁴² ovary, and colorectum.⁴³

In the box on the next page, we list many of the significant findings from the NHS that pertain to the association of lifestyle, behavioral, and dietary risk factors, and environment with risk of specific diseases.

Important Outcomes From the Nurses’ Health Study and Associated Significant Findings

Outcome	Significant Findings
Related to smoking
Type 2 diabetes	Passive and active smoking is associated with increased type 2 diabetes risk.
Cardiovascular disease	Smoking is associated with CVD in women.
Colorectal cancer	Cigarette smoking is associated with increased risk of colorectal adenoma and colorectal cancer.
Pancreatic cancer	Cigarette smoking is an important pancreatic cancer risk factor.
Psoriasis	Smoking is significantly associated with increased psoriasis risk.
Multiple sclerosis	Current cigarette smokers have increased risk of MS compared with never smokers.
Eye diseases	Smoking is associated with increased risk of cataracts and age-related macular degeneration.
Related to physical activity
Type 2 diabetes	Sedentary behavior increased type 2 diabetes risk and moderate- to high-intensity exercise lowered type 2 diabetes risk.
Cardiovascular disease	Moderate-intensity physical activity is associated with lower risk of CHD.
Breast cancer	Participating in ≥ 7 h moderate to vigorous physical activity per week decreases risk of breast cancer.
Psoriasis	Lack of physical activity is a major risk factor for psoriasis.
Neurodegeneration	Higher levels of physical activity are associated with better cognitive performance.
Associated with obesity
Type 2 diabetes	Excess adiposity is the strongest type 2 diabetes risk factor.
Cardiovascular disease	Even moderate weight gain since age 18 y is associated with subsequent risk of CHD incidence and CVD mortality.
Breast cancer	Short-term weight gain is associated with increased breast cancer risk that was strongest for premenopausal women.
Endometrial cancer	Obesity accounts for approximately 40% of incident endometrial cancer cases.
Pancreatic cancer	Overweight or inactive women have positive associations with pancreatic cancer risk.
Non-Hodgkin’s lymphoma	Adiposity in young adulthood, adolescence, and childhood is strongly associated with non-Hodgkin’s lymphoma risk.
Psoriasis	Overall and central obesity is associated with increased risk of psoriasis.
Multiple sclerosis	Individuals who are obese have greater risk of MS.
Kidney stones	Obesity and weight gain are both associated with higher risk of kidney stone formation.
Eye diseases	Obesity and high BMI are associated with cataracts.
Related to oral contraceptives
Cardiovascular disease	Current OC use is associated with higher risk of CVD, primarily among women who are smokers and those with hypertension.
Cancer	Mixed effects among current OC users suggesting a higher risk of melanoma and breast cancer, and a lower risk of colorectal and ovarian cancer.
Related to postmenopausal hormone therapy
Cardiovascular disease	Current HT use is generally associated with lower risk of total CHD and nonfatal myocardial infarction.
Breast cancer	Current combined used of estrogen and progestin is associated with increased risk of breast cancer.
Endometrial cancer	Postmenopausal estrogen use is one of the best-established risk factors for endometrial cancer.
Neurodegeneration	Past or current HT users have significantly worse rates of decline in global cognition.
Related to endogenous hormones
Breast cancer	Circulating levels of estrogens and androgens are significantly positively associated with risk of breast cancer.
Related to dietary factors
Type 2 diabetes	Dietary patterns that increase intake of fruits, vegetables, whole grains, and legumes, while decreasing intake of red meats, refined sugars, and sugar-sweetened beverages decrease risk of type-2 diabetes.
Cardiovascular disease	Diet is an important determinant of CVD risk, and trans-fatty acids are strongly associated with CHD risk.
Breast cancer	Dietary pattern characterized by higher intake of fruits, vegetables, whole grains, low-fat dairy, fish, and poultry decreased risk of breast cancer.
Pancreatic cancer	There are positive associations between pancreatic cancer risk and intake of fructose and sugar-sweetened soft drinks and an inverse association between vitamin D and pancreatic cancer risk.
Non-Hodgkin’s lymphoma	There is a significant increased risk of non-Hodgkin’s lymphoma associated with intake of trans fat and red meat.
Neurodegeneration	Greater intake of antioxidants, higher nut intake, and following the Mediterranean diet all are associated with higher cognitive functioning.
Multiple sclerosis	Vitamin D intake may be associated with reduced MS risk.
Kidney stones	DASH diet (high in fruits and vegetables, moderate in low-fat dairy products, and low in red and processed meats) contributes to kidney stone prevention.
Eye diseases	Maintaining a healthy and well-balanced diet helps prevent cataracts, age-related macular degeneration, and primary open-angle glaucoma.
Related to environment
Obesity	Higher levels of bisphenol A and phthalates (byproducts of plastics and other consumer goods) are associated with weight gain. Those living in higher-density counties (i.e., lower sprawl) had lower BMI and higher physical activity.
Type 2 diabetes	Higher urinary levels of persistent organic pollutants and bisphenol A and phthalates are significantly associated with higher type 2 diabetes risk.
Squamous cell carcinoma	Exposure to the sun leading to sunburn, particularly at early ages, increases the risk of incident squamous cell carcinoma.
Related to sleep and shift work
Type 2 diabetes	Too long and too short duration of sleep and decreased quality of sleep increase risk of type 2 diabetes.
Cardiovascular disease	Shift work and not sleeping the optimal 8 h a day is associated with increased risk of CHD.
Breast cancer	There is a positive association between number of years working night shifts and risk of breast cancer.
Colorectal cancer	Higher consumption of animal fat (and processed meats) is associated with higher colon cancer risk, whereas increased consumption of fiber, folate, and vitamin D decrease colon cancer risk.

Note. BMI = body mass index (kg/m²); CHD = coronary heart disease; CVD = cardiovascular disease; DASH = dietary approaches to stop hypertension; HT = hormone therapy; MS = multiple sclerosis; OC = oral contraceptives.

DELIVERY (IMPLEMENTATION OF FINDINGS)

The impact of the NHS extends to the implementation of findings by the public, clinicians, health practitioners, policymakers, industry, and others. Specifically, the cohort has allowed an expanded scope of study that continues to inform policy and practice.

Many findings from the NHS have contributed to public health recommendations summarized in World Health Organization, World Cancer Research Fund, and various reports from the US Surgeon General.⁴⁴ For example, the Food and Drug Administration acted, in part, on evidence from the NHS relating dietary trans-fat consumption to both heart disease and diabetes risk. Removal of trans-fat in Canada and the United States has resulted in decreased incidence of heart disease and diabetes.^45,46 Estimates suggest that this population benefit is a decrease by more than 12% in the number of incident cases of diabetes and a more than 8% decrease in CVD.

Nurses’ HealthStudy research on the benefits of physical activity on disease prevention and premature mortality contributed to the evidence base for the 2008 Physical Activity Guidelines for Americans.⁴⁷ Research over more than 20 years on use of postmenopausal hormone therapy was also key in shifting the discussion about hormone therapy safety and appropriateness of use. The data published in 1995 showed that breast cancer risk increased with increasing duration of hormone therapy use, particularly for combination estrogen-plus-progestin. With this finding (later confirmed by the Women’s Health Initiative randomized trial), the focus of hormone therapy use moved from the long-held stance that “ever use” was safe, to a more appropriate discussion about “duration of use” and the risks and benefits of hormone therapy.

Studies of survivorship after cancer diagnosis and treatment have been broadened to address physical activity and diet changes, showing that higher levels of physical activity reduce risk of recurrence and death among women with breast and colon cancer,⁴⁸ and contribute to guidelines for cancer survivors.⁴⁹

EXPANDING SCOPE OF STUDY

The NHS continues to evolve and explore new areas of etiological and translational research. The cohort is expanding to assess innovations in use of mobile technology as well as working to link information regarding NHS participants’ health and health care utilization.³ Repeated blood collections allow for analysis of change in markers and change in risk to parallel studies of change in adherence to diet guidelines and risk of disease. In addition, with collection of repeated exposure data both before and after cancer diagnosis, the cohort can evaluate when components of lifestyle are important to survival during the disease process, and offer key findings with tangible clinical implications. The availability of data on health and health-related quality of life before and after cancer diagnosis allows for valuable insights into the causes and consequences of cancer on health and well-being.

Recently, the NHS has applied metabolomics to the biology of cancer, potentially uncovering novel pathways in etiology and survival, as well as new targets for intervention.⁵⁰ These metabolomic measures are being expanded to assess health implications in the cohort with regard to hypertension, chronic obstructive pulmonary disease and asthma, and fertility, as well as more detailed data on the impact of alcohol use and disease outcomes.

In addition to further analyzing these aspects of health, the NHS is expanding participant recruitment. Since 1976 when the NHS began, the demographics of nursing have greatly changed with increasing proportions of minorities and men joining the profession. As such, recruitment extended in 2012 to prioritize enrollment of minority participants and in January 2015 the NHS3 began recruiting men as primary study participants.³

CONCLUSIONS

Collectively, the NHS has contributed substantially to the understanding of numerous health- and disease-related outcomes in women. Many of these contributions have been made possible through external collaborations with researchers across the globe including a number of pooling projects that combine individual participant data to reduce heterogeneity in approaches to analysis and also allow for the study of rarer endpoints that are not frequent enough for individual cohorts to publish robust findings.⁵¹ These pooled analyses reduce publication bias for these endpoints. This collaborative expansion continues to evolve and inform guidelines and future research projects.

The NHS has helped refine methods for conduct of prospective cohort studies including measures of exposures, data analysis, statistical methods, and approaches to linking of data sources to inform urban design, air pollution guidelines, and other exposure measures that are now more readily available to investigators. The scope of health conditions being investigated and documented in the cohort all add to the sustained value of the NHS as summarized in this supplement. This study serves as a model prospective cohort study with repeated measurement that is having an impact on public health policy and practice both locally and globally.