Skip to main content
NCBI home page
American Journal of Public Health logoLink to American Journal of Public Health
. 2016 Mar;106(3):443–448. doi: 10.2105/AJPH.2015.302990

Massachusetts Inpatient Medicaid Cost Response to Increased Supplemental Nutrition Assistance Program Benefits

Rajan Anthony Sonik 1,✉
PMCID: PMC4880217  PMID: 26794167

Abstract

Objectives. To investigate the impact of an increase in Supplemental Nutrition Assistance Program (SNAP) benefits on Medicaid costs and use in Massachusetts.

Methods. Using single and multigroup interrupted time series models, I examined the effect of an April 2009 increase in SNAP benefits on inpatient Medicaid cost and use patterns. I analyzed monthly Medicaid discharge data from 2006 to 2012 collected by the Massachusetts Center for Health Information and Analysis.

Results. Inpatient costs for the overall Massachusetts Medicaid population grew by 0.55 percentage points per month (P < .001) before the SNAP increase. After the increase, cost growth fell by 73% to 0.15 percentage points per month (–0.40; P = .003). Compared with the overall Medicaid population, cost growth for people with the selected chronic illnesses was significantly greater before the SNAP increase, as was the decline in growth afterward. Reduced hospital admissions after the SNAP increase drove the cost declines.

Conclusions. Medicaid cost growth fell in Massachusetts after SNAP benefits increased, especially for people with chronic illnesses with high sensitivity to food insecurity.

People in 1 in 7 US households experience food insecurity,1 meaning they experience hunger, insufficient food, or concerns about having enough food.2 Food insecurity is associated with a large set of health problems for both children and adults.3–11 People with chronic illnesses, who usually have heightened costs of care, are both more likely to experience food insecurity12–14 and more vulnerable to its health effects.15 Understanding the contribution of this problem to health care utilization and costs, however, has been difficult with available data.16

One approach to doing so is to look at the inverse question; that is, examining the effects of alleviating food insecurity. The Supplemental Nutrition Assistance Program (SNAP; formerly the Food Stamps Program) is the largest domestic antihunger program.17 Despite some methodological difficulties,18,19 SNAP has been shown to at least partially alleviate food insecurity for recipients,20,21 making it a good candidate for study. Unfortunately, changes to the program usually occur slowly, limiting the ability to detect broad changes in food insecurity levels and their potential impact on health. However, as part of the American Recovery and Reinvestment Act, maximum monthly SNAP allotments to beneficiaries were temporarily increased by 13.6% from April 2009 to October 2013.22 This larger change to the status quo provided a unique opportunity to examine the effects of alleviating food insecurity on recipients.

I analyzed the relationship between the timing of the increase in SNAP benefits and patterns of health care utilization and costs. Massachusetts provided an ideal setting for this investigation. The Massachusetts Center for Health Information and Analysis collects detailed information on each admission to every hospital in the state, allowing analyses of both the general Medicaid population and those with chronic illnesses. Moreover, the state expanded Medicaid coverage as part of its health reform law in 2006, well before the April 2009 SNAP increase, and it passed a wide-ranging payment reform law that came into effect in 2012, well after the SNAP increase. Between these reforms was a time when economic growth was stagnant for low-income populations, even after the formal end of the Great Recession.23 This provided an ample study period from October 2006 to August 2012.

METHODS

Data provided by the Massachusetts Center for Health Information and Analysis included information on all Medicaid and Medicaid managed care admissions to public and private Massachusetts hospitals for 71 months, from October 2006 to August 2012. Excluding certain nonroutine complex transfers, the final data set included information on 422 565 admissions. On the basis of the diagnostic-related group listed for each admission, I created a subgroup of 19 122 admissions of people with chronic health conditions. This approach allowed me to compare utilization and cost patterns between this subgroup and the general Medicaid population before and after the SNAP increase.

Research on the susceptibility to food insecurity for people with specific conditions is lacking. Although some studies have looked at associations between food insecurity and specific conditions,24,25 these have not looked longitudinally at the effects of changes in food security status on disease outcomes. I therefore consulted 4 physician colleagues caring for patients with complex conditions (2 subspecialists and 2 primary care physicians) to develop a list of conditions that practitioners saw as particularly sensitive to food insecurity. The conditions needed to be severe enough and common enough to result in enough admissions to Massachusetts hospitals to provide adequate numbers for the analyses (i.e., multiple admissions across the state per month). They also had to be not so common and varied in severity as to mask any potential effect. Cardiovascular diseases were discussed but excluded for this latter reason. Ultimately, 6 conditions were selected for the study that all providers agreed should be included: sickle cell disease, diabetes, malnutrition (and failure to thrive), cystic fibrosis, asthma, and inflammatory bowel disease. Poor nutrition is associated with worse outcomes for each of these diseases,26–31 further supporting the selections.

Measures

Collapsing the admissions data by month provided total monthly inpatient Medicaid costs for the state, which was the primary outcome variable. This approach also provided a breakdown of variables for the utilization-related components of cost (monthly admissions and average length of stay [LOS] per admission) and the health care inflation component (average cost per day of admission). I ran separate models with each of these components as outcome variables to determine which, if any, drove cost changes. I measured all dependent variables as percentage point changes from the base month, October 2006, to make comparisons across groups readily interpretable. For control variables, I calculated the average age of patients admitted each month, as well as the percentage of monthly admissions for which patients were female, non-Hispanic White, non-Hispanic Black, Hispanic, homeless, on a Medicaid managed care plan, and identified as an emergency admission. Bivariate regressions showed significant associations between monthly costs and each of these covariates, warranting their inclusion in the final models.

Statistical Analysis

With these monthly data and using the control variables, I conducted single and multigroup interrupted time series analyses. Single interrupted time series models test for changes in slope (trends over time) and intercept (immediate changes in outcome levels) associated with the timing of an intervention.32–34 Multigroup interrupted time series models additionally test for differences between study groups in either of these changes.32–34

I modeled the effects of the SNAP increase for the overall Medicaid population in single interrupted time series models, and using multigroup models, I compared these results to those for people with chronic illnesses. For the multigroup analyses, I first compared trends for sickle cell disease to the overall Medicaid population because of the severity of sickle cell disease, its high sensitivity to stressors such as food insecurity, and its high costs of care.35 I then grouped the other diseases with sickle cell disease in steps to increase the underlying sample size for each month of data. I first combined diabetes and malnutrition (and failure to thrive) with sickle cell disease because of their more direct relationship with nutritional status. Finally, I grouped all 6 chronic illnesses together. The statistical models were as follows:

  • (1) Single: Yt = β0 + β1(months)t + β2(post_SNAP↑_binary)t + β3(post_months)t + β4–10 (controls) + et

  • (2) Multigroup: Yt = β0 + β1(months)t + β2(post_SNAP↑_binary)t + β3(post_months)t + β4(chronic_binary) + β5(chronic_binary)(months) + β6(chronic_binary)(post_SNAP↑_binary) + β7(chronic_binary)(post_months) + β8–14(controls) + et

“SNAP↑” refers to the April 2009 SNAP increase, and the “chronic_binary” interaction terms should be interpreted as differences from the results for the overall Medicaid population. Because the “post_” terms already represent differences, the “chronic_binary” × “post_” terms should be interpreted as one would interpret difference-in-differences.

To adjust for potential autocorrelation, as recommended when conducting interrupted time series analyses,32–34 I allowed up to 1 year (12 months) of lag to be considered in the autocorrelated error structure. I used Stata version 13.0 (StataCorp LP, College Station, TX) and the ITSA package34 for the analyses. The ITSA package uses Newey-West SEs, which address both autocorrelation and heteroskedasticity.34

RESULTS

For the overall Medicaid population, monthly inpatient costs grew by 0.55 percentage points per month (P < .001) before the SNAP increase, adjusting for controls. After the SNAP increase, the rate of growth was 0.15 percentage points per month (–0.40; P = .003; Table 1; Figure 1), signifying a 73% decline. The change in intercept was not significant, meaning there was not a significant immediate change in costs within the first month of the SNAP increase.

TABLE 1—

Trends in Monthly Inpatient Medicaid Costs, Admissions, Average LOS per Admission, and Average Cost per Day of Admission Before and After the SNAP Increase of April 2009: Massachusetts, 2006–2012

Variable Total Costs: % Change From October 2006 Costs in $, b (Newey-West SE) Total Admissions: % Change From October 2006 No. of Admissions, b (Newey-West SE) Average LOS per Admission: % Change From Average October 2006 Stay in Days, b (Newey-West SE) Average Cost per Day of Admission: % Change From Average October 2006 Costs in $, b (Newey-West SE)
All Medicaid, single group interrupted time series models
 Slope before SNAP increase 0.55*** (0.15) 0.26* (0.12) −0.17 (0.14) 0.50*** (0.08)
 Immediate change after SNAP increase 0.36 (1.49) 0.56 (2.72) −2.94* (1.27) 3.15 (3.77)
 Change in slope after SNAP increase −0.40** (0.13) −0.44** (0.13) 0.07 (0.12) 0.04 (0.10)
All Medicaid vs sickle cell disease, multigroup interrupted time series models
 All Medicaid: slope before SNAP increase 0.70*** (0.15) 0.27* (0.13) −0.17 (0.09) 0.63*** (0.07)
 All Medicaid: immediate change after SNAP increase −3.19 (3.76) −0.30 (3.29) −5.15** (1.81) 5.16 (3.18)
 All Medicaid: change in slope after SNAP increase −0.57*** (0.16) −0.48** (0.16) 0.07 (0.09) −0.07 (0.13)
 Difference in slope before SNAP increase 3.48*** (0.84) 2.80*** (0.63) 0.71*** (0.13) −0.21 (0.23)
 Difference in immediate change after SNAP increase −20.56 (19.69) −8.76 (20.25) −3.49 (6.94) −5.86 (8.17)
 Difference in change in slope after SNAP increase −3.68* (1.59) −3.18** (1.19) −0.22 (0.25) −0.32 (0.31)
All Medicaid vs sickle cell disease, diabetes, and malnutrition (and failure to thrive), multigroup interrupted time series models
 All Medicaid: slope before SNAP increase 0.95*** (0.21) 0.31 (0.18) −0.08 (0.10) 0.63*** (0.10)
 All Medicaid: immediate change after SNAP increase 1.07 (2.66) 0.34 (2.47) −3.46* (1.33) 6.16 (3.28)
 All Medicaid: change in slope after SNAP increase −0.77*** (0.22) −0.50** (0.18) −0.02 (0.09) −0.07 (0.15)
 Difference in slope before SNAP increase 1.03*** (0.24) 1.59*** (0.29) −0.03 (0.18) −0.25 (0.13)
 Difference in immediate change after SNAP increase −19.85** (6.10) −17.21* (6.94) −0.93 (3.30) −2.22 (5.87)
 Difference in change in slope after SNAP increase −1.03** (0.31) −1.63*** (0.33) 0.20 (0.21) −0.12 (0.20)
All Medicaid vs all 6 chronic illnesses, multigroup interrupted time series models
 All Medicaid: slope before SNAP increase 1.12** (0.38) 0.49 (0.28) −0.04 (0.11) 0.64*** (0.11)
 All Medicaid: immediate change after SNAP increase 5.23 (4.35) 3.86 (3.20) −4.80** (1.65) 6.90 (3.54)
 All Medicaid: change in slope after SNAP increase −0.88* (0.35) −0.64** (0.23) −0.02 (0.10) −0.09 (0.15)
 Difference in slope before SNAP increase 1.22** (0.38) 0.82** (0.29) −0.00 (0.10) 0.15 (0.09)
 Difference in immediate change after SNAP increase −6.27 (9.74) −0.93 (6.02) −5.19 (3.08) 1.92 (5.36)
 Difference in change in slope after SNAP increase −1.23** (0.43) −1.13** (0.34) 0.22 (0.15) −0.32 (0.21)
Open in a new tab

Note. LOS = length of stay; SNAP = Supplemental Nutrition Assistance Program. All dependent variables were measured as the percentage change from the value in the base month, October 2006. Slopes thus indicate the percentage point change per month in the relevant outcome variable. “SNAP increase” refers to the April 2009 increase in SNAP benefits. “Difference” refers to the value for the chronic illness (or group of illnesses) minus the value for the overall population. The 6 chronic illnesses are sickle cell disease, diabetes, malnutrition (and failure to thrive), cystic fibrosis, asthma, and inflammatory bowel disease.

*P < .05; **P < .01; ***P < .001.

FIGURE 1—

FIGURE 1—

Open in a new tab

Inpatient Medicaid Cost Trends Before and After the April 2009 Supplemental Nutrition Assistance Program (SNAP) Increase: Massachusetts, 2006–2012

Note. The fitted line from a model without controls is shown to allow 2-dimensional presentation.

I then modeled the other dependent variables—admissions, LOS per admission, and cost per day of admission (Table 1). For monthly admissions, a growth rate of 0.26 percentage points per month before the SNAP increase changed to a decrease of 0.18 percentage points per month afterward (–0.44; P = .002); the change in intercept was not significant. The change in trend for LOS per admission was not significant, but there was an immediate drop of 2.94 percentage points during the first month of the SNAP increase (P = .02). For cost per day of admission, there were no significant changes after the SNAP increase, either immediately or over time.

In the first multigroup model, monthly costs for people with sickle cell disease initially grew by 4.18 percentage points per month, compared with 0.70 percentage points per month for the overall Medicaid population (+3.48; P < .001; Table 1; note that the values for the overall Medicaid population differ slightly between multigroup models and from the single model because they are being adjusted for different chronic illnesses). After the SNAP increase, the growth rate of costs for sickle cell disease fell by 4.25 percentage points per month, compared with a fall of 0.57 percentage points per month for the overall Medicaid population (–3.68; P = .02; Table 1). In other words, costs for sickle cell disease grew about 6 times as fast before the SNAP increase, and this growth rate fell about 7 and a half times as much afterward. The immediate change in costs did not differ significantly between the groups.

Combining data from sickle cell disease with data from diabetes and malnutrition (and failure to thrive) admissions also produced statistically significant pre–post differences in cost growth and decline compared with the overall Medicaid population (Table 1). Additionally, the immediate drop in costs for the group of chronic illnesses was significantly different from the change for the overall population (–19.85 percentage points; P = .002; Table 1). Finally, adding cystic fibrosis, asthma, and inflammatory bowel disease to the group of chronic illnesses also produced significant pre–post trend differences, although the difference in the immediate change in costs was again nonsignificant (Table 1).

I conducted multigroup models comparing patterns in admissions, LOS per admission, and cost per day of admission as well. For all 3 chronic illness comparison groups, admissions for people with chronic illnesses grew significantly faster before the SNAP increase, and admissions growth dropped significantly more afterward (Table 1; Figure 2). Also, in all 3 comparisons, changes in LOS per admission and cost per day of admission after the SNAP increase were not significantly different between the chronic illness population and the general Medicaid group (Table 1).

FIGURE 2—

FIGURE 2—

Open in a new tab

Overall vs Chronic Illness Medicaid Admissions Before and After the April 2009 Supplemental Nutrition Assistance Program (SNAP) Increase: Massachusetts, 2006–2012

Note. Chronic illnesses in the model were sickle cell disease, diabetes, and malnutrition (and failure to thrive). Fitted lines from a model without controls are shown to allow 2-dimensional presentation.

DISCUSSION

Inpatient Medicaid cost growth declined significantly in Massachusetts after the SNAP increase. This decline in cost trends after the SNAP increase appears to have been driven by reduced hospital admissions and to some extent reduced LOS per admission, rather than by patterns in the cost per day of admission (i.e., health care inflation). This finding implies that health improvements drove the cost outcomes. Furthermore, hospital admissions and costs for the population of people with the selected chronic illnesses rose more sharply during the recession before the SNAP increase and fell more sharply after the increase compared with the overall Medicaid population. This is consistent with the idea that patients with these conditions are not only more susceptible to food insecurity but also particularly responsive to its alleviation. It also suggests that heightened food insecurity was an important factor in the rising health care costs before the SNAP increase.

Researchers previously examining the effect of food insecurity on health care costs have developed estimates on the basis of known associations between food insecurity and specific health conditions and between these conditions and increased utilization.36 My findings extend these types of analyses with data from actual admissions. Importantly, my results imply not only that increasing food insecurity raises health care costs but also that alleviating food insecurity can reduce costs.

Limitations

This study has several potential limitations. First, it is restricted to inpatient stays in Massachusetts, and therefore its findings might not generalize to other states or hospital settings. Second, because of the relatively limited literature available on the effects of food insecurity on people with specific conditions, the process of consulting physicians to select the conditions for the analyses was necessarily somewhat informal. Still, the consensus among the practitioners and previous findings showing associations between poor nutrition and more severe outcomes for each disease raised my confidence in the selection of conditions. Third, the study is limited by the fact that not all Medicaid beneficiaries receive SNAP. However, the overlap between Medicaid and SNAP eligibility is considerable.37

Furthermore, the multigroup interrupted time series models are particularly strong.32–34 Potential confounders would have had to occur at the same time as the SNAP increase, and they would have had to cause different effects on the cost and utilization patterns of people with chronic illnesses than on people in the general Medicaid population. Finally, the SNAP increase took place during a complex economic downturn and amid a host of efforts to revive the economy. The strengths of the multigroup models again buffer against this concern, however. Additionally, during the study period, the SNAP increase was the only immediate large-scale intervention that was both directed at low-income populations and sustained continuously once it began.

Policy Implications

My findings indicate that the costs of the SNAP increase were partially offset by Medicaid savings, especially for beneficiaries with chronic illnesses, who typically have elevated costs of care. Also, because of the link between additional SNAP benefits and reduced hospital admissions, it appears that the allotment amounts before the SNAP increase may not have been sufficient to fully alleviate food insecurity and its associated health effects. This is particularly relevant because the SNAP increase under the American Recovery and Reinvestment Act expired in October 2013 and because Congress has repeatedly proposed more cuts. Moreover, my findings suggest, at least in part, that the additional SNAP benefits in Massachusetts were directed at reductions in food insecurity that improved health. This is important information for ongoing policy debates regarding the appropriate levels of SNAP funding and competing claims about how beneficiaries use their benefits.

More broadly, food security is recognized as a key determinant of health.38 Because of the extraordinarily high costs of health disparities39,40 and because of the general recognition that addressing the social determinants of health will be critical to reducing health disparities,38 the effect of alleviating social ills such as hunger on health care outcomes and costs warrants further research.

ACKNOWLEDGMENTS

I am funded through an Agency for Healthcare Research and Quality T32 Institutional Health Services Research Training Program grant.

I thank Jon Chilingerian, Andrew Wilson, and Dominic Hodgkin of Brandeis University, Philippa Sprinz of Hasbro Children’s Hospital, and Lara Bishay of Boston Children’s Hospital for their thoughtful advice.

HUMAN PARTICIPANT PROTECTION

This study was approved by the Brandeis University institutional review board.

Footnotes

See also Galea and Vaughan, p. 394.

REFERENCES

  • 1.Coleman-Jenson A, Gregory C. Key statistics and graphics. 2015. Available at: http://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-us/key-statistics-graphics.aspx. Accessed August 3, 2015.
  • 2.Bickel GW, Nord M, Price C, Hamilton W, Cook J. Guide to measuring household food security in the United States. 2000. Available at: http://www.fns.usda.gov/sites/default/files/FSGuide.pdf. Accessed August 3, 2015.
  • 3.Alaimo K, Olsen CM, Frongillo EA., Jr Food insufficiency and American school-aged children’s cognitive, academic, and psychosocial development. Pediatrics. 2001;108(1):44–53. [PubMed] [Google Scholar]
  • 4.Bronte-Tinkew J, Zaslow M, Capps R, Horowitz A, McNamara M. Food insecurity works through depression, parenting, and infant feeding to influence overweight and health in toddlers. J Nutr. 2007;137(9):2160–2165. doi: 10.1093/jn/137.9.2160. [DOI] [PubMed] [Google Scholar]
  • 5.Casey PH, Szeto KL, Robbins JM et al. Child health-related quality of life and household food security. Arch Pediatr Adolesc Med. 2005;159(1):51–56. doi: 10.1001/archpedi.159.1.51. [DOI] [PubMed] [Google Scholar]
  • 6.Cook JT, Frank DA, Levenson SM et al. Child food insecurity increases risks posed by household food insecurity to young children’s health. J Nutr. 2006;136(4):1073–1076. doi: 10.1093/jn/136.4.1073. [DOI] [PubMed] [Google Scholar]
  • 7.Ding M, Keiley MK, Garza KB, Duffy PA, Zizza CA. Food insecurity is associated with poor sleep outcomes among US adults. J Nutr. 2015;145(3):615–621. doi: 10.3945/jn.114.199919. [DOI] [PubMed] [Google Scholar]
  • 8.Kleinman RE, Murphy JM, Little M et al. Hunger in children in the United States: potential behavioral and emotional correlates. Pediatrics. 1998;101(1):E3. doi: 10.1542/peds.101.1.e3. [DOI] [PubMed] [Google Scholar]
  • 9.Leung CW, Epel ES, Willett WC, Rimm EB, Laraia BA. Household food insecurity is positively associated with depression among low-income supplemental nutrition assistance program participants and income-eligible nonparticipants. J Nutr. 2015;145(3):622–627. doi: 10.3945/jn.114.199414. [DOI] [PubMed] [Google Scholar]
  • 10.Park CY, Eicher-Miller HA. Iron deficiency is associated with food insecurity in pregnant females in the United States: National Health and Nutrition Examination Survey 1999–2010. J Acad Nutr Diet. 2014;114(12):1967–1973. doi: 10.1016/j.jand.2014.04.025. [DOI] [PubMed] [Google Scholar]
  • 11.Whitaker RC, Phillips SM, Orzol SM. Food insecurity and the risks of depression and anxiety in mothers and behavior problems in their preschool-aged children. Pediatrics. 2006;118(3):e859–e868. doi: 10.1542/peds.2006-0239. [DOI] [PubMed] [Google Scholar]
  • 12.Bartfeld J, Dunifon R. State-level predictors of food insecurity and hunger among households with children. Available at: http://naldc.nal.usda.gov/download/32791/PDF. Accessed August 3, 2015.
  • 13.Huang J, Guo B, Kim Y. Food insecurity and disability: do economic resources matter? Soc Sci Res. 2010;39(1):111–124. [Google Scholar]
  • 14.Nord M. Characteristics of low-income households with very low food security: an analysis of the USDA GPRA food security indicator. Available at: http://www.ers.usda.gov/media/196569/eib25_1_.pdf. Accessed August 3, 2015.
  • 15.Coleman-Jensen A, Nord M. Food insecurity among households with working-age adults with disabilities. Available at: http://www.ers.usda.gov/media/980690/err_144.pdf. Accessed August 3, 2015.
  • 16.Lee JS. Food insecurity and healthcare costs: research strategies using local, state, and national data sources for older adults. Adv Nutr. 2013;4(1):42–50. doi: 10.3945/an.112.003194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.US Department of Agriculture. Supplemental Nutrition Assistance Program (SNAP) Available at: http://www.fns.usda.gov/snap/supplemental-nutrition-assistance-program-snap. Accessed August 3, 2015.
  • 18.Gregory C, Rabbitt MP, Ribar DC. The Supplemental Nutrition Assistance Program and food insecurity. Available at: http://uknowledge.uky.edu/cgi/viewcontent.cgi?article=1010&context=ukcpr_papers. Accessed August 3, 2015.
  • 19.Caswell JA, Yaktine AL, editors. Supplemental Nutrition Assistance Program: Examining the Evidence to Define Benefit Adequacy. Washington, DC: National Academy Press; 2013. [PubMed] [Google Scholar]
  • 20.Mabli J, Ohls J. Supplemental Nutrition Assistance Program participation is associated with an increase in household food security in a national evaluation. J Nutr. 2015;145(2):344–351. doi: 10.3945/jn.114.198697. [DOI] [PubMed] [Google Scholar]
  • 21.Nord M, Golla AM. Does SNAP decrease food insecurity: untangling the self-selection effect. 2009. Available at: http://www.ers.usda.gov/media/184824/err85_1_.pdf. Accessed August 3, 2015.
  • 22.Dean S, Rosenbaum D. SNAP benefits will be cut for nearly all participants in November 2013. Available at: http://www.cbpp.org/sites/default/files/atoms/files/2-8-13fa.pdf. Accessed August 3, 2015.
  • 23.Fry R, Kochchar R. America’s wealth gap between middle-income and upper-income families is widest on record. 2014. Available at: http://www.pewresearch.org/fact-tank/2014/12/17/wealth-gap-upper-middle-income. Accessed August 3, 2015.
  • 24.Seligman HK, Laraia BA, Kushel MB. Food insecurity is associated with chronic disease among low-income NHANES participants. J Nutr. 2010;140(2):304–310. doi: 10.3945/jn.109.112573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Skalicky A, Meyers AF, Adams WG, Yang Z, Cook JT, Frank DA. Child food insecurity and iron deficiency anemia in low-income infants and toddlers in the United States. Matern Child Health J. 2006;10(2):177–185. doi: 10.1007/s10995-005-0036-0. [DOI] [PubMed] [Google Scholar]
  • 26.Reid M. Nutrition and sickle cell disease. C R Biol. 2013;336(3):159–163. doi: 10.1016/j.crvi.2012.09.007. [DOI] [PubMed] [Google Scholar]
  • 27.Gau BR, Chen HY, Hung SY et al. The impact of nutritional status on treatment outcomes of patients with limb-threatening diabetic foot ulcers. J Diabetes Complications. 2015;29(7):882–886. doi: 10.1016/j.jdiacomp.2015.09.011. [DOI] [PubMed] [Google Scholar]
  • 28.Higgins PA, Daly BJ. Adult failure to thrive in the older rehabilitation patient. Rehabil Nurs. 2005;30(4):152–159. doi: 10.1002/j.2048-7940.2005.tb00100.x. [DOI] [PubMed] [Google Scholar]
  • 29.Hankard R, Munck A, Navarro J. Nutrition and growth in cystic fibrosis. Horm Res. 2002;58(suppl 1):16–20. doi: 10.1159/000064763. [DOI] [PubMed] [Google Scholar]
  • 30.Kim JH, Ellwood PE, Asher MI. Diet and asthma: looking back, moving forward. Respir Res. 2009;10(1):49–55. doi: 10.1186/1465-9921-10-49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Wiskin AE, Wootton SA, Beattie RM. Nutrition issues in pediatric Crohn’s disease. Nutr Clin Pract. 2007;22(2):214–222. doi: 10.1177/0115426507022002214. [DOI] [PubMed] [Google Scholar]
  • 32.Shadish WR, Cook TD, Campbell DT. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Boston, MA: Houghton Mifflin; 2002. [Google Scholar]
  • 33.Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27(4):299–309. doi: 10.1046/j.1365-2710.2002.00430.x. [DOI] [PubMed] [Google Scholar]
  • 34.Linden A. Conducting interrupted time series analysis for single and multiple group comparisons. Stata J. 2015;15(2):480–500. [Google Scholar]
  • 35.Kauf TL, Coates TD, Huazhi L, Mody-Patel N, Hartzema AG. The cost of health care for children and adults with sickle cell disease. Am J Hematol. 2009;84(6):323–327. doi: 10.1002/ajh.21408. [DOI] [PubMed] [Google Scholar]
  • 36.Shepard DS, Setren E, Cooper D. Hunger in America: suffering we all pay for. 2011. Available at: https://www.americanprogress.org/wp-content/uploads/issues/2011/10/pdf/hunger_paper.pdf. Accessed August 3, 2015.
  • 37.Dorn S, Wheaton L, Johnson P, Dubay L. Using SNAP to establish, verify and renew Medicaid eligibility. Available at: http://www.urban.org/sites/default/files/alfresco/publication-pdfs/412808-Using-SNAP-Receipt-to-Establish-Verify-and-Renew-Medicaid.PDF. Accessed August 3, 2015.
  • 38.Woolf SH, Braveman P. Where health disparities begin: the role of social and economic determinants—and why current policies may make matters worse. Health Aff (Millwood) 2011;30(10):1852–1859. doi: 10.1377/hlthaff.2011.0685. [DOI] [PubMed] [Google Scholar]
  • 39.LaVeist TA, Gaskin D, Richard P. Estimating the economic burden of racial health inequalities in the United States. Int J Health Serv. 2011;41(2):231–238. doi: 10.2190/HS.41.2.c. [DOI] [PubMed] [Google Scholar]
  • 40.Schoeni RF, Dow WH, Miller WD, Pamuk ER. The economic value of improving the health of disadvantaged Americans. Am J Prev Med. 2011;40(1 suppl 1):S67–S72. doi: 10.1016/j.amepre.2010.09.032. [DOI] [PubMed] [Google Scholar]

Articles from American Journal of Public Health are provided here courtesy of American Public Health Association

RESOURCES