Accounting for the Hierarchical Structure in Veterans Health Administration Data: Differences in Healthcare Utilization between Men and Women Veterans


  • Heather G. Allore Department of Internal Medicine, Yale School of Medicine, USA
  • Yuming Ning Department of Internal Medicine, Yale School of Medicine, USA
  • Cynthia A. Brandt VA Connecticut Healthcare System, USA
  • Joseph L. Goulet Department of Psychiatry, Yale School of Medicine, USA



Hierarchical Logistics Models, Random Effects, GLIMMIX, GENMOD, Generalized Estimating Equations, Gender Differences, Veterans


Women currently constitute 15% of active United States of America military service personnel, and this proportion is expected to double in the next 5 years. Previous research has shown that healthcare utilization and costs differ in women US Veterans Health Administration (VA) patients compared to men. However, none have accounted for the potential effects of clustering on their estimates of healthcare utilization. US Women Veterans are more likely to serve in specific military branches (e.g. Army), components (e.g. National Guard), and ranks (e.g. officer) than men. These factors may confer different risk and protection that can affect subsequent healthcare needs. Our study investigates the effects of accounting for the hierarchical structure of data on estimates of the association between gender and VA healthcare utilization. The sample consisted of data on 406,406 Veterans obtained from VA’s Operation Enduring Freedom/ Operation Iraqi Freedom roster provided by Defense Manpower Data Center — Contingency Tracking System Deployment File. We compared three statistical models, ordinary, fixed and random effects hierarchical logistic regression, in order to assess the association of gender with healthcare utilization, controlling for branch of service, component, rank, age, race, and marital status. Gender was associated with utilization in ordinary logistic and, but not in fixed effects hierarchical logistic or random effects hierarchical logistic regression models. This points out that incomplete inference could be drawn by ignoring the military structure that may influence combat exposure and subsequent healthcare needs. Researchers should consider modeling VA data using methods that account for the potential clustering effect of hierarchy.

Author Biographies

Heather G. Allore, Department of Internal Medicine, Yale School of Medicine, USA

Department of Internal Medicine

Yuming Ning, Department of Internal Medicine, Yale School of Medicine, USA

Department of Internal Medicine

Cynthia A. Brandt, VA Connecticut Healthcare System, USA

Department of Emergency Medicine

Joseph L. Goulet, Department of Psychiatry, Yale School of Medicine, USA

Yale School of Medicine


Yano EM, Frayne SM. Health and health care of women veterans and women in the military: research informing evidence-based practice and policy. Womens Health Issues 2011; 21(4 Suppl): S64-6. DOI:

Leslie DL, Goulet J, Skanderson M, Mattocks K, Haskell S, Brandt C. VA health care utilization and costs among male and female veterans in the year after service in Afghanistan and Iraq. Mil Med 2011; 176(3): 265-9. DOI:

Actuary Oot: Women Veterans: Past, present, and future. Department of Veterans Affairs, Office of Policy and Planning. (Administration V ed). Washington, DC 2007.

Frayne SM, Yu W, Yano EM, Ananth L, Iqbal S, Thrailkill A, Phibbs CS. Gender and use of care: planning for tomorrow's Veterans Health Administration. J Womens Health (Larchmt) 2007; 16: 1188-99. DOI:

Haskell SG, Brandt CA, Krebs EE, Skanderson M, Kerns RD, Goulet JL. Pain among Veterans of Operations Enduring Freedom and Iraqi Freedom: do women and men differ? Pain Med 2009; 10: 1167-73. DOI:

Frayne SM, Parker VA, Christiansen CL, Loveland S, Seaver MR, Kazis LE, et al. Health status among 28,000 women veterans. The VA Women's Health Program Evaluation Project. J Gen Intern Med 2006; 21(Suppl 3): S40-6. DOI:

Borrero S, Kwoh CK, Sartorius J, Ibrahim SA. Brief report: Gender and total knee/hip arthroplasty utilization rate in the VA system. J Gen Intern Med 2006; 21(Suppl 3): S54-57. DOI:

Kaur S, Stechuchak KM, Coffman CJ, Allen KD, Bastian LA. Gender differences in health care utilization among veterans with chronic pain. J Gen Intern Med 2007; 22: 228-33. DOI:

Washington DL, Caffrey C, Goldzweig C, Simon B, Yano EM. Availability of comprehensive women's health care through Department of Veterans Affairs Medical Center. Womens Health Issues 2003; 13: 50-54. DOI:

Yano EM, Bastian LA, Frayne SM, Howell AL, Lipson LR, McGlynn G, et al. Toward a VA Women's Health Research Agenda: setting evidence-based priorities to improve the health and health care of women veterans. J Gen Intern Med 2006; 21(Suppl 3): S93-101. DOI:

Kreft IGG, de Leeuw J. Introducing Multilevel Modeling. London: Sage 1998. DOI:

Demidenko E. Mixed Models: Theory and Applications. Hoboken, NJ.: John Wiley & Sons, Inc 2004. DOI:

Goldstein H. Multilevel Statistical Models. 3rd ed. London: Edward Arnold 2003.

Hox J. Multi-Level Analysis. London: Lawrence Erlbaum Associates 2002.

Austin PC, Tu JV, Alter DA. Comparing hierarchical modeling with traditional logistic regression analysis among patients hospitalized with acute myocardial infarction: should we be analyzing cardiovascular outcomes data differently? Am Heart J 2003; 145: 27-35. DOI:

Bowen ME, Gonzalez HM. Racial/ethnic differences in the relationship between the use of health care services and functional disability: the health and retirement study (1992-2004). Gerontologist 2008; 48: 659-67. DOI:

Tuerk PW, Mueller M, Egede LE. Estimating physician effects on glycemic control in the treatment of diabetes: methods, effects sizes, and implications for treatment policy. Diabetes Care 2008; 31: 869-73. DOI:

SAS/STAT 9.2 Production GENMOD Procedure for Windows. Cary, NC: SAS Institute Inc. 2009.

Austin PC, Goel V, van Walraven C. An introduction to multilevel regression models. Can J Public Health 2001; 92: 150-54. DOI:

Liang KY, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika 1986; 73: 13-22. DOI:

SAS Institute. SAS/STAT 9.2 Production GLIMMIX Procedure for Windows. Cary, NC:SAS Institute Inc. 2009.

Little RC, Milliken GA, Stroup WW, et al. SAS System for Mixed Models. Cary, NC:SAS Institute Inc. 1996.




How to Cite

Allore, H. G., Ning, Y., Brandt, C. A., & Goulet, J. L. (2013). Accounting for the Hierarchical Structure in Veterans Health Administration Data: Differences in Healthcare Utilization between Men and Women Veterans. International Journal of Statistics in Medical Research, 2(2), 94–103.



General Articles