0
Caring for the Uninsured and Underinsured |

Hunger in an Adult Patient Population FREE

Karin Nelson, MD; Margaret E. Brown, MS; Nicole Lurie, MD, MSPH
[+] Author Affiliations

From the Departments of Medicine, Hennepin County Medical Center, Minneapolis, Minn (Dr Nelson) and the University of Minnesota Medical School—Minneapolis (Dr Lurie); and the Institute for Health Services Research, University of Minnesota School of Public Health, Minneapolis (Ms Brown and Dr Lurie).


JAMA. 1998;279(15):1211-1214. doi:10.1001/jama.279.15.1211.
Text Size: A A A
Published online

Context.— Although clinical observations suggest that some patients experience hunger and food insecurity, there are limited data on the prevalence of hunger in adult patients.

Objective.— To determine the prevalence of hunger and food insecurity in adult patients at an urban county hospital.

Design.— Cross-sectional survey conducted in 1997.

Patients.— The primary survey included all patients aged 18 years or older who were admitted to the medicine, surgery, and neurology services during a 2-week period, and all patients who attended the hospital's general medicine clinic during 1 week. A second survey included primary care patients who received insulin from the hospital pharmacy during a 1-month period.

Main Outcome Measures.— Rates of hunger and food insecurity.

Results.— Of 709 eligible patients, 567 (participation rate, 80%) were interviewed in either the clinic (n=281) or hospital (n=286). An additional 170 patients who received insulin were interviewed by telephone (response rate, 75%). Of the primary sample, 68 (12%) respondents reported not having enough food, 75 (13%) reported not eating for an entire day, and 77 (14%) reported going hungry but not eating because they could not afford food. A total of 222 (40%) had received food stamps in the previous year and of those, 113 (50%) had their food stamps reduced or eliminated. Recipients whose food stamps had been eliminated or reduced were more likely to report not having enough food (18% vs 13%,P=.006), not eating for a whole day (20% vs 16%, P=.01), going hungry but not eating (20% vs 16%, P=.08), and cutting down on the size of meals or skipping meals (33% vs 27%,P=.01). In multivariate analysis, independent predictors of hunger included an annual income of less than $10000 (odds ratio [OR], 7.55; 95% CI, 3.01-18.92), drug use (OR, 3.56; 95% CI, 1.46-8.66), and a reduction in food stamp benefits (OR, 1.73; 95% CI, 1.01-2.96). Predictors of food insecurity included an annual income of less than $10000 (OR, 4.12; 95% confidence interval [CI], 1.98-8.58), drug use (OR, 2.11; 95% CI, 1.66-5.08), and a reduction in food stamps (OR, 2.02; 95% CI, 1.23-3.32). In addition, 103 (61%) patients in the sample of diabetics reported hypoglycemic reactions; 32 (31%) of these were attributed to inability to afford food.

Conclusion.— Hunger and food insecurity are common among patients seeking care at an urban county hospital.

A NUMBER of factors not directly related to the delivery of medical care have been shown to affect health status, including education,1,2 race,3 literacy,4 and socioeconomic status.5,6 Malnutrition has also been documented to have adverse health effects.7 While studies of malnutrition demonstrate low birth weight and decreased growth in children,8 adults who are poorly nourished are at increased risk of infections and vitamin deficiency disorders, have more major medical complications, and generate more costs per diagnosis related group.9 Few studies10 have examined rates of hunger in adult patients and, to our knowledge, there are few recent US data on the impact of hunger on health.

During the past year, we have observed increasing numbers of patients in our practice setting who lack money to buy food. This is particularly troubling given the current implementation of welfare reform, in which nearly half of the cost savings come from reductions in food and nutrition programs.11 We undertook this study to determine the prevalence of hunger and to assess the level of food adequacy in adult patients who seek care at a public teaching hospital.

This study was approved by the human subjects review committee at Hennepin County Medical Center, a 435-bed public teaching hospital in Minneapolis, Minn. All patients verbally consented to participate. Two related studies were performed. In the first study, we conducted in-person interviews with all patients aged 18 years and older who were admitted to the medicine, surgery, and neurology services during a 2-week period (April 13, 1997, through April 26, 1997) and all patients who attended the hospital's general medicine clinic during 1 week (June 9, 1997, through June 13, 1997). We used telephone follow-up to contact patients who were not available for interview while they were in the hospital. We excluded patients who lived in a facility that provided meals (such as a nursing home or a group home), were too sick or confused to participate, or spoke neither English nor Spanish. The second study was a telephone survey of all patients with diabetes who received primary care through the hospital and who received insulin from the hospital pharmacy during May 1997.

The survey items included 8 questions which are part of previously validated measures of hunger and food insecurity12,13 (Table 1). Measures of food insecurity (which often precedes overt hunger) include whether patients worried about whether their food supply would run out, were concerned that the food they had would not last until they had money to buy more,14,15 whether patients delayed paying a bill to buy food, had received emergency food from a food shelter, or went to a soup kitchen. Hunger-related items included not having enough food to eat, skipping or decreasing the size of meals, or not eating for an entire day (or going hungry but not eating) because of inability to afford food. Questions about hunger and food insecurity were asked in relation to the previous 12 months and the previous 30 days. Additional survey items obtained by patient self-report included demographic variables (annual household income, race, and level of education), lifestyle habits (including the use of tobacco, alcohol, and illicit drugs), presence and type of health insurance, reduction or elimination of food stamp benefits during the prior year, and health status. Patients in the diabetic group were asked additional questions about episodes of hypoglycemia and its relationship to being able to afford food.

We examined frequency distributions for key variables and assessed the association of sociodemographic variables, benefit reductions, and lifestyle habits with hunger and food insecurity using χ2 tests. Independent risk factors for food insecurity and hunger were assessed using logistic regression. For these analyses, we grouped patients by age as follows: younger than 45 years, 45 to 65 years, and older than 65 years. We defined food insecurity as a positive response to any of the scaled questions regarding not having enough food, worry that food would run out, or reports that food did not last. Each question regarding hunger was analyzed separately. We present results from the primary study, and highlight differences in the diabetic sample, when present.

Of 709 patients who were eligible to participate, we interviewed 281 outpatients and 286 inpatients (response rate, 80%). Twenty-three (3%) refused to participate in the survey and 119 (17%) were discharged from either the clinic or hospital before they could be interviewed and could not be contacted by telephone. Of the 228 patients with insulin-dependent diabetes who were eligible for the study, we interviewed 170 (response rate, 75%). Patient characteristics for the primary sample are displayed in Table 2. There were no significant demographic differences between patients in the primary sample and patients in the sample of those with diabetes (data not shown). The mean age of our sample was 47 years (SD=16), range from 18 to 92 years. Approximately half of the primary patient sample had an annual income of less than $10000, 11% reported being homeless for some period during the previous year, and 20% reported having to live with friends or relatives because they could not afford housing on their own. The inpatients were more likely than outpatients to have an annual income greater than $25000 (27% vs 12%,P=<.001), were older (mean age, 48 vs 45 years, P=.02) and were more likely to be white (57% vs 44%, P=.003).

Table Graphic Jump LocationTable 2.—Patient Characteristics and the Prevalence of Hunger (n=567)

Patients reported high levels of food insecurity and hunger during the previous year (Table 2). Thirty-five percent reported worrying that their food would run out, 28% reported that their food supply would not last until they had money to buy more, 28% reported putting off paying a bill to buy food, 27% reported receiving emergency food during the past year, and 13% reported obtaining food at a soup kitchen.

A smaller but substantial number reported problems with food adequacy. One quarter (24%) of our patient sample reported decreasing the size of meals or skipping meals because they could not afford food, 12% admitted to not having enough food available, 13% reported not eating for an entire day, and 14% reported going hungry but not eating because they could not afford food. Over one third of those who reported having an inadequate food supply said it had occurred at least once a month in the preceding year. Regarding food quality, 19% reported eating no fruits or vegetables in a 2-day period and 40% reported 3 or fewer servings for the same period. Compared with those not reporting hunger, people who reported not eating for an entire day were more likely to have an annual income of less than $10000 (60% vs 42%, P=<.001), to have had their food stamps eliminated during the past year (38% vs 18%,P=.04), and to report drug use (25% vs 13%,P=.05) (Table 3). Compared with those not reporting hunger, patients with incomes greater than $10000 reported similar proportions of hunger and food insecurity.

A total of 222 (40%) of respondents had received food stamps during the previous year and of those, 113 (50%) had their food stamps reduced or eliminated (Table 4). In a bivariate analysis, patients who had food stamps reduced or eliminated were significantly more likely to report food insecurity and hunger than those whose food stamp allotment remained stable or who had never received food stamps. Half of the recipients with food stamp reductions reported worrying that their food would run out (53% vs 41%, P<.001) or that their food did not last (45% vs 34%, P<.001), and had obtained emergency food (48% vs 37%, P<.001).

Table Graphic Jump LocationTable 4.—Effect of Change in Food Aid*

Similar rates of food insecurity (25%) and hunger (13%) were reported in the sample of patients with diabetes. In addition, 103 (61%) reported having experienced hypoglycemic reactions in the previous year, and 32 patients (31%) attributed the hypoglycemic reactions to being unable to afford food. Furthermore, of patients reporting hypoglycemic reactions secondary to being unable to afford food, 8 (26%) reported syncope and required treatment in the emergency department or were hospitalized as a consequence. Eight percent of the sample (n=14) reported they had decreased or stopped taking their insulin because they did not have enough to eat.

In logistic regression analysis, independent predictors of food insecurity, as measured by the presence of any positive response to a food insecurity item, included annual income of less than $10000, reduction in food stamps, and drug use. Independent predictors of hunger were the same: an annual income of less than $10000, illicit drug use, and a reduction in food stamps (Table 5). Similar results (not shown) were obtained for the 3 other hunger-related items (ie, going hungry but not eating, not having enough food, and not eating for a whole day) and for patients in the diabetic sample.

Table Graphic Jump LocationTable 5.—Predictors of Food Insecurity and Hunger

Our study demonstrates that hunger and food insecurity are prevalent problems among patients seeking inpatient and outpatient care at an urban county hospital. We suspect that the actual rates may be higher than reported herein as some people may not admit to food inadequacy even if the problem exists. Furthermore, a substantial portion of the patients we could not locate to interview were homeless or lacked a telephone, suggesting that they had financial problems that are associated with hunger and food insecurity. A previous study of medical patients10 reported a similar prevalence of hunger, but that study did not involve systematic sampling and did not use validated measures to determine hunger or food insecurity. Available estimates suggest that approximately 30 million people cannot obtain enough food to meet their daily needs.8,16 It is noteworthy that the prevalence of this social problem is similar to that of common medical conditions such as hypertension, diabetes, or heart disease.17

Although many social and economic factors including social class, race, education, and literacy affect health status, the impact of hunger has not been adequately studied. Hunger and food insecurity have the potential to affect health on many levels. Although malnutrition has been documented to have adverse health outcomes,8 hunger and food insecurity can precede the medical aspects of malnutrition. Because physiologic needs such as hunger take precedence over other daily activities,18 the need to cope with an inadequate food supply could be expected to change health-related behavior and priorities regarding medical problems. Medication adherence and diet compliance might not be a high priority for patients with limited access to an adequate food supply.

Hunger and food adequacy are linked to social and economic factors. In our study, patients with an annual income of less than $10000 were more likely to report food insecurity and hunger. People with low incomes typically spend a greater percentage of their income on food,19 suggesting that a small change in income could have a greater effect on the ability to obtain a nutritionally adequate diet. The small percentage of our sample (5%) who reported illicit drug use were more likely to report hunger, underscoring the tenuous economic and social circumstance of drug users. In our study, alcohol use and cigarette use were not associated with hunger in either bivariate or multivariate analyses.

Our data suggest a protective effect of the food stamp program. Patients whose food stamp benefits were eliminated or reduced were more likely to report hunger and food insecurity. Even though the Personal Responsibility and Work Opportunity Reconciliation Act11 took effect in January 1998, some food stamp cutbacks were implemented at the local level more than a year before that date. We have no data on the specific circumstances that led to food stamp reductions among individual patients. More than 100 patients (19%) in our study who reported incomes of less than $10000 per year did not receive food stamps in the preceding year, reflecting national data that half of the households living in poverty do not receive food stamps.20 The food stamp program is the United States' largest provider of food aid and is a major target of welfare reform.11 Half of the cost savings in the current welfare reform legislation are expected to come from decreased federal funding for food and nutrition programs, with an estimated decrease of $26.2 billion over the next 4 years.11

Our study has several limitations: the information was based on self-report and the sample was drawn from patients who sought medical care at a single institution. Recipients whose food stamps were reduced or eliminated may have been more likely to report hunger because they were more aware of the issue, although we tried to minimize this reporting bias by asking questions about hunger and food insecurity before asking about food stamps. Thus, the prevalence of hunger may not reflect that of the general population or of other populations of individuals seeking medical care. We know of no reason to suspect that food adequacy is substantially better among other public hospital populations.

While it was beyond the scope of our cross-sectional study to document direct clinical effects of hunger, such as weight loss or decreases in serum albumin levels, the data regarding the associations of hypoglycemia with hunger and food insecurity provide a concerning link to health outcomes. In fact, we first undertook this study because of clinical observations of inadequate food supply in some of our diabetic patients, leading to discontinuation of insulin and hospitalization for ketoacidosis. While we do not know how prevalent such direct health effects of hunger might be, we urge policymakers to consider these data as they make decisions regarding funding for food and nutrition programs. Clinicians caring for a similar patient population should ask about food adequacy, as hunger may affect clinical decision making and health status for individual patients.

Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986.  N Engl J Med.1993;329:103-109.
Pincus T, Callahan LF. Associations of low formal education level and poor health status: behavioral, in addition to demographic and medical, explanations?  J Clin Epidemiol.1994;47:355-361.
Guralnik JM, Land KC, Blazer D, Fillenbaum GG, Branch LG. Educational status and active life expectancy among older blacks and whites.  N Engl J Med.1993;329:110-116.
Williams MV, Parker RM, Baker DW.  et al.  Inadequate functional health literacy among patients at two public hospitals.  JAMA.1995;274:1677-1682.
Epstein AM, Stern RS, Tognetti J.  et al.  The association of patients' socioeconomic characteristics with the length of hospital stay and hospital charges within diagnosis-related groups.  N Engl J Med.1988;318:1579-1585.
Davey-Smith A, Hart C, Blame D, Gillis C, Hawthorne V. Lifetime socioeconomic position and mortality: prospective observational study.  BMJ.1997;314:547-552.
Steffee WP. Malnutrition in hospitalized patients.  JAMA.1980;244:2630-2635.
Physician Task Force on Hunger in America.  Hunger in America: The Growing Epidemic.  Middletown, Conn: Wesleyan University Press; 1985.
Roubenoff R, Roubenoff RA, Preto J, Balke CW. Malnutrition among hospitalized patients: a problem of physician awareness.  Arch Intern Med.1987;147:1462-1465.
Rosenberg E, Bernabo L. Hunger: a hospital survey.  Soc Work Health Care.1992;16:83-95.
Jones JY, Richardson J. Federal Food Programs: Legislation in the 104th Congress: Congressional Research Service Report for Congress.  Washington, DC: Congressional Research Service; October 29, 1996. Document 96-861 ENR.
Kendall A, Olson CM, Frongillo EA. Validation of the Radimer/Cornell Measures of Hunger and Food Insecurity.  J Nutr.1995;125:2793-2801.
Hamilton WL, Cook JT, Thompson WW.  et al.  Measuring Food Security in the United States, Household Food Security in the United States, Summary Report of the Food Security Measurement Project.  Alexandria, Va: USDA Food and Consumer Service; September 1997.
Cohen BE. Food security and hunger policy for the 1990's.  Nutr Today.1990;25:23-27.
Campbell CC. Food insecurity: a nutritional outcome or a predictor variable?  J Nutr.1991;121:408-415.
Uvin P. The state of world hunger.  Nutr Rev.1994;52:151-161.
Edmund EJ, Gillum BS. 1994 Summary: National Hospital Discharge Survey.  Hyattsville, Md: National Center for Health Statistics; 1996. Advance Data From Vital and Health Statistics, No. 278.
Maslow AH. Motivation and Personality.  New York, NY: Harper & Row; 1970.
Kinsey JD. Food and families' socioeconomic status.  J Nutr.1994;124:1878S-1885S.
Brown JL, Allen D. Hunger in America.  Annu Rev Public Health.1988;9:503-526.

Figures

Tables

Table Graphic Jump LocationTable 2.—Patient Characteristics and the Prevalence of Hunger (n=567)
Table Graphic Jump LocationTable 4.—Effect of Change in Food Aid*
Table Graphic Jump LocationTable 5.—Predictors of Food Insecurity and Hunger

References

Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986.  N Engl J Med.1993;329:103-109.
Pincus T, Callahan LF. Associations of low formal education level and poor health status: behavioral, in addition to demographic and medical, explanations?  J Clin Epidemiol.1994;47:355-361.
Guralnik JM, Land KC, Blazer D, Fillenbaum GG, Branch LG. Educational status and active life expectancy among older blacks and whites.  N Engl J Med.1993;329:110-116.
Williams MV, Parker RM, Baker DW.  et al.  Inadequate functional health literacy among patients at two public hospitals.  JAMA.1995;274:1677-1682.
Epstein AM, Stern RS, Tognetti J.  et al.  The association of patients' socioeconomic characteristics with the length of hospital stay and hospital charges within diagnosis-related groups.  N Engl J Med.1988;318:1579-1585.
Davey-Smith A, Hart C, Blame D, Gillis C, Hawthorne V. Lifetime socioeconomic position and mortality: prospective observational study.  BMJ.1997;314:547-552.
Steffee WP. Malnutrition in hospitalized patients.  JAMA.1980;244:2630-2635.
Physician Task Force on Hunger in America.  Hunger in America: The Growing Epidemic.  Middletown, Conn: Wesleyan University Press; 1985.
Roubenoff R, Roubenoff RA, Preto J, Balke CW. Malnutrition among hospitalized patients: a problem of physician awareness.  Arch Intern Med.1987;147:1462-1465.
Rosenberg E, Bernabo L. Hunger: a hospital survey.  Soc Work Health Care.1992;16:83-95.
Jones JY, Richardson J. Federal Food Programs: Legislation in the 104th Congress: Congressional Research Service Report for Congress.  Washington, DC: Congressional Research Service; October 29, 1996. Document 96-861 ENR.
Kendall A, Olson CM, Frongillo EA. Validation of the Radimer/Cornell Measures of Hunger and Food Insecurity.  J Nutr.1995;125:2793-2801.
Hamilton WL, Cook JT, Thompson WW.  et al.  Measuring Food Security in the United States, Household Food Security in the United States, Summary Report of the Food Security Measurement Project.  Alexandria, Va: USDA Food and Consumer Service; September 1997.
Cohen BE. Food security and hunger policy for the 1990's.  Nutr Today.1990;25:23-27.
Campbell CC. Food insecurity: a nutritional outcome or a predictor variable?  J Nutr.1991;121:408-415.
Uvin P. The state of world hunger.  Nutr Rev.1994;52:151-161.
Edmund EJ, Gillum BS. 1994 Summary: National Hospital Discharge Survey.  Hyattsville, Md: National Center for Health Statistics; 1996. Advance Data From Vital and Health Statistics, No. 278.
Maslow AH. Motivation and Personality.  New York, NY: Harper & Row; 1970.
Kinsey JD. Food and families' socioeconomic status.  J Nutr.1994;124:1878S-1885S.
Brown JL, Allen D. Hunger in America.  Annu Rev Public Health.1988;9:503-526.
CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 60

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Topics
PubMed Articles