Whitney Robinson

Overview:

I design studies and interpret results from big datasets to figure out why rates of health and disease vary across groups.  More generally, I love thinking up new ways to draw causal inference from observational data.

Non-cancerous gynecologic conditions, like endometriosis, fibroids, and PCOS, are my particular focus.  They affect the well-being of a huge number of people but can be difficult to diagnose and treat.  And treatment often involves complex trade-offs regarding quality of life, fertility, and permanence of treatment effectiveness.  I typically investigate these questions using health care data that were not originally intended for research, such as electronic health records, state physician licensing data, etc.

As a Faculty Epidemiologist in the OB-GYN Department, I genuinely love seeing others thrive in their research.  I have a gift for identifying where people can add the most value on a research team. 

Positions:

Instructor in the Department of Obstetrics and Gynecology

Community Outreach & Education Division
School of Medicine

Member of the Duke Cancer Institute

Duke Cancer Institute
School of Medicine

Education:

Ph.D. 2008

University of North Carolina - Chapel Hill

Publications:

Invited Commentary: What Social Epidemiology Brings to the Table-Reconciling Social Epidemiology and Causal Inference.

In response to the Galea and Hernán article, "Win-Win: Reconciling Social Epidemiology and Causal Inference" (Am J Epidemiol. 2020;189(XX):XXXX-XXXX), we offer a definition of social epidemiology. We then argue that methodological challenges most salient to social epidemiology have not been adequately addressed in quantitative causal inference, that identifying causes is a worthy scientific goal, and that quantitative causal inference can learn from social epidemiology's methodological innovations. Finally, we make 3 recommendations for quantitative causal inference.
Authors
Robinson, WR; Bailey, ZD
MLA Citation
Robinson, Whitney R., and Zinzi D. Bailey. “Invited Commentary: What Social Epidemiology Brings to the Table-Reconciling Social Epidemiology and Causal Inference.Am J Epidemiol, vol. 189, no. 3, Mar. 2020, pp. 171–74. Pubmed, doi:10.1093/aje/kwz197.
URI
https://scholars.duke.edu/individual/pub1501107
PMID
31566211
Source
pubmed
Published In
American Journal of Epidemiology
Volume
189
Published Date
Start Page
171
End Page
174
DOI
10.1093/aje/kwz197

Maternal Dietary Patterns are Associated with Lower Levels of Cardiometabolic Markers during Pregnancy.

BACKGROUND: Elevated levels of cardiometabolic markers are characteristic of normal pregnancy, however, insulin resistance and increased glucose, triglyceride, and cholesterol levels can adversely influence maternal and child health. Diet is a modifiable behaviour that could have significant impact on maternal cardiometabolic levels during pregnancy. We investigated the association between dietary patterns and cardiometabolic markers (glucose, insulin, insulin resistance (HOMA-IR), triglycerides, and cholesterol) during pregnancy. METHODS: Data from the Pregnancy, Infection, and Nutrition prospective cohort study (2000-05) was used (n = 513). Diet was assessed using a food frequency questionnaire. Dietary patterns were derived using latent class analysis (LCA) and the Dietary Approaches to Stop Hypertension (DASH) diet. Linear regression was used to examine the dietary patterns-cardiometabolic markers association during pregnancy. RESULTS: Three dietary patterns evolved from the LCA characterised by high intakes of: (1) hamburgers, hot dogs, bacon, French fries, fried chicken, white bread, and soft drinks; (2) some vegetables, fruit juice, refined grains, mixed dishes, processed meat, and empty calorie foods; and (3) fruits, vegetables, whole grains, low-fat dairy, breakfast bars, and water. After adjustment for potential confounders including prepregnancy body mass index, a diet consistent with Latent Class 3 was negatively associated with maternal insulin (μU/mL: β = -0.12; 95% CI -0.23, -0.01) and HOMA-IR (β = -0.13; 95% CI -0.25, -0.00). Additionally, DASH scores within Tertile 3 (higher dietary quality) were also negatively associated with maternal triglycerides (mg/dL). CONCLUSIONS: The study findings suggest an association between maternal dietary patterns and several cardiometabolic markers during pregnancy.
Authors
Martin, CL; Siega-Riz, AM; Sotres-Alvarez, D; Robinson, WR; Daniels, JL; Perrin, EM; Stuebe, AM
MLA Citation
Martin, Chantel L., et al. “Maternal Dietary Patterns are Associated with Lower Levels of Cardiometabolic Markers during Pregnancy.Paediatr Perinat Epidemiol, vol. 30, no. 3, May 2016, pp. 246–55. Pubmed, doi:10.1111/ppe.12279.
URI
https://scholars.duke.edu/individual/pub1313348
PMID
26848932
Source
pubmed
Published In
Paediatr Perinat Epidemiol
Volume
30
Published Date
Start Page
246
End Page
255
DOI
10.1111/ppe.12279

Sleep duration and obesity among adolescents transitioning to adulthood: do results differ by sex?

OBJECTIVES: To examine the association between short sleep duration and obesity among adolescents (mean age 16 years) transitioning into young adulthood (mean age 21 years) in the National Longitudinal Study of Adolescent Health (N = 10,076). STUDY DESIGN: Self-reported sleep duration was categorized as <6, 6-8, or >8 hours. Obesity status, using measured height and weight, was defined as body mass index ≥95th percentile in adolescence and body mass index ≥30 kg/m(2) in young adulthood. RESULTS: Cross-sectionally, short sleep duration was associated with obesity in adolescent males (prevalence ratio 1.8 [95% CI, 1.3-2.4]) but not in females (prevalence ratio 1.0 [95% CI, 0.7-1.4]). In longitudinal analyses, short sleep duration in adolescence was associated with incident obesity in both males and females (risk ratio 1.2 [95% CI, 1.0-1.6]) in young adulthood. No interactions by sex were noted. CONCLUSIONS: Analyzing the association of sleep duration and obesity longitudinally resolved sex discrepancies observed in earlier cross-sectional analyses. Optimizing sleep duration during adolescence may be an effective intervention to prevent excess weight gain in young adults.
Authors
Suglia, SF; Kara, S; Robinson, WR
MLA Citation
Suglia, Shakira F., et al. “Sleep duration and obesity among adolescents transitioning to adulthood: do results differ by sex?J Pediatr, vol. 165, no. 4, Oct. 2014, pp. 750–54. Pubmed, doi:10.1016/j.jpeds.2014.06.052.
URI
https://scholars.duke.edu/individual/pub1501083
PMID
25066064
Source
pubmed
Published In
J Pediatr
Volume
165
Published Date
Start Page
750
End Page
754
DOI
10.1016/j.jpeds.2014.06.052

Systematic review of prostate cancer's association with body size in childhood and young adulthood.

Body size relatively early in life may influence men's later rate of prostate cancer. We searched for published, English-language studies of the association between prostate cancer incidence or mortality and body size between ages 5 and 29 years. We summarized analyses of childhood body size, early-adult waist circumference, and early-adult body mass index (BMI). Most study designs were case-control or retrospective cohort studies, in which body size was self-reported and recalled. The few studies of childhood body size and early-adult waist circumference indicated null or weak associations. The results from studies of early-adult BMI were heterogeneous (p = 0.04) and showed evidence of funnel plot asymmetry. The random-effects rate ratio (RR) was 1.06 (95% confidence interval [CI]: 0.99, 1.14) per five-unit increase in BMI. Studies using measured (as opposed to self-recalled) height and weight (n = 3) tended to produce stronger associations: fixed-effects summary RR = 1.22 (1.06, 1.39). The same was true for studies that did not adjust for later-life BMI (n = 13): fixed-effects RR = 1.13 (1.06, 1.21). Examining only analyses of advanced or high-grade prostate cancers, results were heterogeneous (p < 0.01). The random-effects summary RR per five units of BMI was 1.01 (95% CI: 0.89, 1.15). In all the reviewed studies, the vast majority of men were of normal weight in childhood and early adulthood. Few studies presented data describing the association between prostate cancer and obesity (e.g., early-adult BMI > or = 30). The exact relationships between early-life body size and prostate cancer remain unclear but appear to be weak.
Authors
Robinson, WR; Poole, C; Godley, PA
MLA Citation
Robinson, Whitney R., et al. “Systematic review of prostate cancer's association with body size in childhood and young adulthood.Cancer Causes Control, vol. 19, no. 8, Oct. 2008, pp. 793–803. Pubmed, doi:10.1007/s10552-008-9142-9.
URI
https://scholars.duke.edu/individual/pub1501070
PMID
18347923
Source
pubmed
Published In
Cancer Causes Control
Volume
19
Published Date
Start Page
793
End Page
803
DOI
10.1007/s10552-008-9142-9

Neighborhood characteristics associated with park use and park-based physical activity among children in low-income diverse neighborhoods in New York City.

Urban parks provide spaces and facilities for children's physical activity (PA) and can be a free resource in low-income communities. This study examined whether neighborhood characteristics were associated with children's park use and park-based moderate-to-vigorous PA (MVPA) in low-income diverse communities and how associations differed between ethnic groups. Data on park visits and MVPA came from 16,402 children 5-10-years old directly observed using the System for Observing Play and Recreation in Communities in 20 parks in low-income neighborhoods with majority Latino or Asian populations in New York City. Neighborhood characteristics included land use mix (LUM), street audits, crime rates, and an area deprivation index. We employed Poisson and negative binomial models to estimate effects of neighborhood-level variables on the number of children observed in parks and engaging in MVPA, overall and by ethnicity. Results for Asian, Latino, and African American children indicated that higher levels of LUM and pedestrian-friendly streets were associated with greater numbers of children in parks and higher MVPA across all three groups. For Asian and Latino children only, quality of environment was positively associated with MVPA, whereas level of deprivation and crime rates in the surrounding neighborhood were negatively associated with children's park-based MVPA. In contrast, a park's access to public transportation was negatively associated with number of all children observed and engaging in MVPA. Study findings suggest that park-based MVPA interventions can be informed by understanding how neighborhood characteristics facilitate and constrain park use and park-based MVPA.
Authors
Huang, J-H; Hipp, JA; Marquet, O; Alberico, C; Fry, D; Mazak, E; Lovasi, GS; Robinson, WR; Floyd, MF
MLA Citation
Huang, Jing-Huei, et al. “Neighborhood characteristics associated with park use and park-based physical activity among children in low-income diverse neighborhoods in New York City.Prev Med, vol. 131, Feb. 2020, p. 105948. Pubmed, doi:10.1016/j.ypmed.2019.105948.
URI
https://scholars.duke.edu/individual/pub1501086
PMID
31836479
Source
pubmed
Published In
Prev Med
Volume
131
Published Date
Start Page
105948
DOI
10.1016/j.ypmed.2019.105948