Food Insecurity: Screening -- Children, adolescents, and adults

Importance

According to the US Department of Agriculture (USDA) Economic Research Service’s Current Population Survey, 12.8% of households experienced food insecurity in 2022, with 7.7% of households experiencing low food security and 5.1% experiencing very low food security.¹ Nearly one-third of households with incomes below the federal poverty threshold are food insecure. Food insecurity is one among a multitude of medical, psychological, and social conditions common among economically disadvantaged households. In both children and adults, experiencing food insecurity is associated with negative effects on health outcomes.^2,3

USPSTF Assessment of Magnitude of Net Benefit

The US Preventive Services Task Force (USPSTF) concludes that the evidence is insufficient to assess the balance of benefits and harms of screening for food insecurity on health outcomes in the primary care setting. There is limited evidence on the health outcome related to screening for and interventions to address food insecurity in the primary care setting, and the balance of benefits and harms cannot be determined; thus, the USPSTF cannot make a recommendation for or against screening.

See the Table for more information on the USPSTF recommendation rationale and assessment. For more details on the methods the USPSTF uses to determine the net benefit, see the USPSTF Procedure Manual.⁴

Table. Summary of USPSTF Rationale

Abbreviation: USPSTF, US Preventive Services Task Force.

Practice Considerations

Patient Population Under Consideration

This recommendation applies to all children, adolescents, and adults.

Definitions

Food insecurity is considered a social risk factor; it is a measurable and modifiable social and economic condition influenced by broader social and structural determinants of health and measured at the household or individual level.³ The Centers for Medicare & Medicaid Services has identified 5 core social risk factors for which community services are helpful: food insecurity, housing instability, transportation problems, utility help needs, and interpersonal safety.⁵ Social risk factors often overlap. For example, food insecurity is associated with other factors such as housing instability and transportation issues. A related term, “social need,” refers to the desire for assistance in addressing a social risk factor and reflects an individual’s priorities.⁶

Food insecurity is defined as an economic and social condition of perceived limited or uncertain access to sufficient amounts of nutritious food needed for an active and healthy life.⁷ Food insecurity can be long term, temporary, episodic (eg, running out of money at the end of the month), or related to a specific event (eg, loss of employment). It can also differ among household members, such as when parents or caregivers go without food to maintain their children’s diet. Food insecurity differs from hunger, which is generally defined as an individual-level physiological state that may result from food insecurity.⁸

Individuals may respond to food insecurity by changing their preferred variety of foods, food sources (eg, grocery stores, convenience stores, or food pantries), the amount of food they consume, or going without food.^9,10 The USDA provides clarification on both food insecurity and food security. Regarding food insecurity, “very low food security” is when some household members reduce their food intake because of an inability to afford enough food, and “low food security” is when at some point during the year, households reduced the quality, variety, or desirability of their diets but maintained normal eating patterns.¹¹ Food security is also divided into 2 categories: “marginal food security,” in which households had problems or anxiety accessing food at times but did not substantially alter the quality, variety, or quantity of their intake, and “high food security,” in which households had no problems or anxiety about consistently accessing adequate food.¹¹

Populations at Risk

Poverty, due to unemployment or other conditions, is the primary cause of food insecurity. Studies have reported that certain populations, such as households with children (particularly those with children younger than 6 years); households headed by a single caregiver; households of those with incomes below 185% of the poverty level; older adults; individuals with disabilities; and veterans are at higher risk for food insecurity.³ Having a chronic medical condition is both a risk factor for and associated with food insecurity. Persons living in low-income neighborhoods are at increased risk of food insecurity for several reasons, including limited access to healthy foods, fewer full-service grocery stores, and limited public and private transportation options.^12-14

There are significant disparities in food insecurity by race and ethnicity. According to data from the USDA, in 2022, 20% of Hispanic households, 22% of non-Hispanic Black households, and 9% of White households experienced food insecurity.¹ Recent estimates for Native American/Alaska Native populations are not available, but in the decade leading up to a 2010 report, annual estimates ranged between 20% and 30%.¹⁵

Screening Tests

There are limited data on how screening tools are used in clinical practice. Multiple social risk factors are often assessed via multidomain tools (eg, Well Child Care, Evaluation, Community Resources, Advocacy, Referral, Education [WE CARE]; Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences [PRAPARE]) rather than screening for food insecurity alone.^3,16 The most frequently used and studied single-domain screening tool for food insecurity is the 2-item Hunger Vital Sign tool. The first question asks respondents if during the past 12 months, they (1) worried about running out of food and not having money to buy more, and (2) experienced running out of food and not having money to buy more.^17-19 It is derived from, and validated against, the USDA Household Food Security Survey.¹⁹

Treatment or Interventions

Interventions applicable to health care settings include food prescription programs; food boxes or pantries provided in, or linked to, clinics; and medically tailored meals.^20,21 Another common form of primary care–based intervention is care coordination linking patients with local, state, or federal programs. The majority of these programs, such as food pantries, free lunch programs, the Special Supplemental Nutrition Program, or the Special Supplemental Nutrition Program for Women, Infants, and Children are delivered in community settings rather than through primary care.

Suggestions for Practice Regarding the I Statement

Potential Preventable Burden

Living in a household with food insecurity as a child is associated with obesity, asthma, mental health conditions, and worse oral health.^2,22 Experiences of hunger due to food insecurity during childhood have also been associated with poor health later in life.^23,24 Adults with food insecurity often experience chronic diseases such as obesity, diabetes, hypertension, and cardiovascular disease. The exact reasons for these connections are unclear, but evidence suggests that stress, challenges managing existing health conditions, dietary changes toward less healthy options, and changes in gut bacteria could all play a role.³

Potential Harms

No studies reported on the harms of screening.^25,26 Potential harms include stigma, fear of involvement in the legal/justice system (eg, disclosing food insecurity, prompting Child Protective Services involvement), and other privacy concerns.

Current Practice

Several studies indicate that clinicians and patients believe that social risks and/or needs, including food insecurity, should be addressed in health care settings. However, estimates of screening for food insecurity vary considerably across providers (between 25% and 100%).²⁷ The screening tool, setting, type of visit, format/mode of delivery, and timing of screening varies across practice settings.²⁸ Additionally, some health care systems use screening tools to measure prevalence and better understand the needs of their population, while others screen and follow with referrals, clinical care, and additional services.²⁰

While it is known that food and nutritional security change over time, there is limited research on the appropriate screening interval for food insecurity. Most screening tools do not assess the duration or periodicity of food insecurity and therefore cannot assess whether it is temporary.²⁹ One study of food insecurity over the first years of the COVID-19 pandemic found that children of caregivers who received general social needs screenings and relevant referrals every 6 months had greater improvements in social needs than those who received annual well-child screenings and referrals.³⁰

Other Related USPSTF Recommendations

The USPSTF has recommendation statements related to other social drivers of health, including screening for intimate partner violence, elder abuse, and abuse of vulnerable adults and the primary prevention of child maltreatment.^31,32

Supporting Evidence

Scope of Review

The USPSTF commissioned a systematic review to evaluate the benefits and harms of screening for food insecurity in the health care setting. To be included in the review, studies had to have occurred in a health care setting and include screening conducted in a clinical setting or identified through a health care delivery or payment system and interventions or programs integrated into, associated with, or referred from health care.^3,33 Studies limited to participants who were undergoing cancer treatment, had other acute medical or psychiatric conditions, or had severe malnutrition or known nutritional deficiencies were excluded from the review.^3,33

Accuracy of Screening Tests and Risk Assessment

Ten fair-quality studies with a total of 123,886 participants assessed 1-item (1 study), 2-item (8 studies), or 6-item (1 study) screening tools for performance.^3,33 All were derived from the USDA’s 18-item Household Food Security Survey. Study settings included primary care (4 studies), primary care plus emergency department (2 studies), emergency department (1 study), and dental clinic (1 study).^3,33 The remaining 2 studies analyzed US Census data. Five studies included only adult participants, 4 studied families with children, and 1 studied teenagers and young adults (aged 15 to 25 years). Eight studies only administered the reference standard and then assessed agreement between item subsets and the full version.^3,33

For the 2-item screening tool, the sensitivity was typically above 95% and specificity was above 82% when the screening tool was fully embedded in the reference standard (in most cases, the 18- or 6-item version of the Household Food Security Survey) or administered within the same larger questionnaire.^3,33 However, the lack of independent administration of the screening tool likely overestimated the performance of the screening tool under usual clinical use.^3,33 The 1-item screening tool had the lowest accuracy, with a sensitivity of 59% and specificity of 87%, but it was also the only study of the screening tool in which it was administered independently from the reference standard.^3,33

Benefits of Screening and Interventions

The USPSTF identified 1 fair-quality randomized clinical trial (n = 789) examining the benefits of screening for food insecurity in families with children younger than 6 years.^3,33 Parents in the intervention group were screened for child maltreatment risk factors (intimate partner violence, parent depression, substance misuse, and use of corporal punishment), which included food insecurity. Families screening positive for food insecurity were given information on how to access federal and local food-related assistance. It found no difference in the percentage reporting food insecurity after 6 months between the intervention group (29.6% with food insecurity) and usual care (29.8% with food insecurity).³⁴

The USPSTF reviewed 2 fair-quality studies (n = 220) examining the benefits of health care–related interventions on food security, intermediate outcomes, and health outcomes.^3,33 The first was a randomized crossover trial (n = 44) that included home delivery of medically tailored meals to patients with diabetes for 12 weeks.³⁵ At the end of the study period (24 weeks), the intervention was associated with reduced food insecurity (41.9% with food insecurity while receiving meals vs 61.5% while not receiving meals; P = .05), improved mental health quality of life (4.7-point change from baseline while receiving meals vs 0.8-point change while not receiving meals), and improved diet quality (eg, the 100-point total Healthy Eating Index score improved by 14.1 points while receiving meals compared with baseline but declined by 17.3 points while not receiving meals). However, there was no impact on other quality-of-life measures; physiologic measures of blood pressure, lipid levels, or glucose levels; or cost-related medication underuse.^3,33,35

The second fair-quality study (n = 176) was a nonrandomized study of interventions comparing children in families who participated in a mobile food pantry with a propensity score–matched cohort of pediatric patients from the same neighborhood as the mobile pantry participants or from nearby neighborhoods not offering a mobile pantry program.³⁶ At 6 months, the study found a smaller increase in body mass index among those who participated in the mobile food pantry. It did not report between-group differences in food insecurity but reported a 1-point reduction (4.3 to 3.3) on a 6-point food insecurity scale among participants in the intervention group.

Several other studies (27 studies) were found that did not contribute to the USPSTF assessment of the evidence because they were of poor quality due to a high risk of bias for the outcomes of interest. Many were pre-post studies or were designed for other purposes (eg, primary interest in other outcomes or comparative effectiveness).^3,33

Harms of Screening and Treatment

The USPTF found no eligible studies on the harms of screening.^3,33 Only 1 study of interventions, the trial of home-delivered meals (n = 44), provided data on intervention harms. Harm was observed for only 1 participant who experienced gastrointestinal distress.³⁵

Response to Public Comments

A draft version of this recommendation statement was posted for public comment on the USPSTF website from June 25, 2024, to July 22, 2024. Some comments expressed that the evidence reviewed by the USPSTF was too narrow in scope and should have included interventions outside the primary care setting and evaluated programs such as the Special Supplemental Nutrition Program for Women, Infants, and Children and the Supplemental Nutrition Assistance Program. USPSTF recommendations focus on services that can be provided in or referred from the primary care setting. The USPSTF did not review evidence on federal programs because these are not under the auspices of health care systems. The USPSTF did include studies of interventions to help facilitate enrollment in these programs. Several commenters asked that the USPSTF include evidence from a broader range of studies (eg, qualitative studies or studies without comparison groups). Some commenters also noted that, given the importance of food, it may be unethical to conduct trials that include a control or comparison group. The USPSTF reviews evidence from multiple, well-conducted studies to adequately determine the benefits and harms of preventive services. For this topic, the USPSTF also included nonrandomized studies and quasi-experimental studies in its review. The USPSTF understands the necessity of food and the complexities involved in conducting and evaluating studies on food insecurity. It is currently studying how to best address this and other types of social risk topics (see the Other Considerations section). Additionally, the USPSTF has a separate commentary addressing this challenging topic.³⁷

Last, some commenters expressed concern that the I statement would be interpreted as a recommendation against screening or that interventions to address food insecurity are ineffective. The USPSTF wishes to clarify that the I statement is a determination that the evidence is insufficient to assess the balance of benefits and harms of screening for food insecurity and is neither a recommendation for nor against screening. Clinicians should continue to use their clinical judgment to determine if screening is appropriate for individual patients.

Other Considerations

The USPSTF recognizes the important connection between a person’s social and economic condition and their health and continues to explore how to best incorporate social needs into its recommendation process. Given the potential evidence base and the feasibility of primary care–based interventions, screening for food insecurity was chosen as an initial focus for a recommendation about a social risk. To assess the balance of benefits and harms, the USPSTF needs evidence that screening for food insecurity leads to health benefits. In the absence of this type of direct evidence, the USPSTF looks for evidence linking studies that show screening tests are accurate with evidence showing that interventions reduce food insecurity and improve health outcomes. However, as the USPSTF reviewed this topic, there were several aspects specific to screening for social risk that complicated its assessment of the balance of benefits and harms.

First, there are limitations to food insecurity interventions that can be provided or referred from the primary care setting. As a social condition, food insecurity requires direct intervention on the social circumstances (ie, poverty) that cause it. Primary care–based and primary care–referrable interventions generally offer only limited “treatment” of food insecurity through interventions that provide access to community resources, or limited access to food without changing a person’s social circumstance (eg, income).

Second, addressing social risk factors in primary care may not have the same direct relationship with health outcomes that is seen with health-focused screenings or interventions. Social risk factors such as food insecurity rarely exist in isolation, and screening for or intervening in one social risk may not completely address the impact on health outcomes.

Third, social determinants may also shape a patient’s situation in a manner that affects their ability to effectively receive other clinical preventive or primary care services, indirectly affecting a variety of health outcomes. For example, a patient’s food insecurity status may be important to identify when providing primary care management of diabetes, separately from providing resources for food insecurity directly. Identifying and addressing a social risk such as food insecurity in the primary care setting may also be necessary to help a patient attend to other preventive care needs.

Fourth, social needs (factors with which patients would like assistance), as opposed to social risks identified by screening, may complicate the direct benefit of screening. For example, a clinician may identify a patient with food insecurity, but if the patient or clinician prioritizes other concerns, the effect of screening and subsequent interventions may be limited.⁶

Last, the USPSTF’s determination was partly based on the small number of acceptable studies on the benefits and harms of food insecurity interventions in the primary care setting. It is reasonable, however, to recognize that providing adequate food and nutrition to individuals without consistent access to it due to poverty is a crucial component of overall health. Given these points, the USPSTF is further exploring how its established methods for assessing the benefits and harms of a preventive intervention can be most effectively used for food insecurity and perhaps other social determinants of health topics as well.

Recommendations of Others

Several organizations provide guidance on screening and interventions for food insecurity. The American Academy of Family Physicians recommends that family physicians use a social risk screening instrument that includes food insecurity.³⁸ The American Academy of Nutrition and Dietetics,³⁹ the American Academy of Pediatrics,⁴⁰ and the American Diabetes Association⁴¹ suggest using the 2-item Hunger Vital Sign screening tool to assess the possibility of food insecurity. The American College of Cardiology/American Heart Association,⁴² American College of Physicians,⁴³ and the American College of Obstetricians and Gynecologists⁴⁴ recommend screening but do not endorse a specific screening tool. AARP has developed a resource guide and toolkit for implementing food insecurity screening and referral for older patients in primary care.⁴⁵

JAMA Patient Page: Screening for Food Insecurity