ABSTRACT
Background
Food insecurity is a social determinant of health associated with cognitive impairments in older adults and people living with HIV (PLWH). Few studies have examined this relation longitudinally, and no studies have explored how the frequency of food insecurity over time may impact cognitive impairment.
Objective
This study aimed to examine the impact of food insecurity on cognitive impairment over a 2-y follow-up period in a cohort of people living with and without HIV.
Methods
This was a 2-y longitudinal analysis of primarily economically disadvantaged, middle-aged, Black, and Hispanic participants from the Miami Adult Studies on HIV (MASH) cohort. Food insecurity was assessed with the USDA Household Food Security Module at baseline and 12- and 24-mo follow-ups. Food insecurity in all 3 assessments was considered persistent food insecurity. Cognitive impairment was assessed with the Mini-Mental State Examination. Statistical analyses consisted of logistic regressions.
Results
A total of 394 participants (247 HIV positive) with 2-y follow-up data were included in this analysis. At baseline, 104 (26.4%) were food-insecure and 58 (14.7%) had cognitive impairment. Very low food security was associated with cognitive impairment at baseline (OR: 3.23; 95% CI: 1.08, 9.65). PLWH not virally suppressed had higher risk for cognitive impairment compared with HIV-uninfected participants (OR: 2.87; 95% CI: 1.15, 7.18). Additionally, baseline food insecurity (OR: 2.28; 95% CI: 1.08, 4.81) and the frequency of food insecurity over time (OR: 1.50 per year; 95% CI: 1.08, 2.10), particularly persistent food insecurity (OR: 3.69; 95% CI: 1.15, 11.83), were associated with cognitive impairment at 2-y follow-up; the results were consistent after excluding cognitively impaired participants at baseline.
Conclusions
Food insecurity is a significant risk factor for cognitive impairment, particularly among individuals who experience food insecurity frequently or persistently. Screening for food insecurity and interventions to secure access to sufficient, nutritious foods may help delay cognitive decline among socioeconomically disadvantaged individuals.
Keywords: food insecurity, cognitive function, HIV, substance abuse, vulnerable populations
Introduction
Food insecurity (FI) refers to a lack of dependable access to sufficient and nutritious food for an active and healthy life. FI is the result of limited resources and is associated with poverty, unemployment, and high housing costs (1). As such, FI is a socioeconomic condition that impacts 13.7 million (10.5%) households in the United States (2). It is often episodic, triggered by unemployment, inflation, food prices, or unforeseen costs (3, 4). Yet, for many households, FI is a frequent or persistent problem that occurs for an average of 7 mo out of the year (5). Tightly related to sociodemographic factors, FI is disproportionately prevalent among minorities (6) and other marginalized groups, including people living with HIV (PLWH) (7, 8).
In the United States, adults from food-insecure households are at increased risk for chronic diseases and mental health conditions (9, 10). FI has been associated with cognitive deficits among US older adults (11–13). Fewer studies have examined this association among middle-aged adults, but Wong et al. (14) showed an association between very low food security (VLFS) and cognitive decline among US adults aged 40–75 y. Furthermore, neurocognitive disorders are found in ∼20–50% of PLWH (15) and persist despite long-term viral suppression (16). We have previously shown that VLFS was associated with poorer mental health quality of life in PLWH (17). Hobkirk et al. (18) showed that FI was associated with cognitive impairments among PLWH but not HIV-uninfected participants. Although the findings are generally consistent, the existing evidence is limited by heterogeneity in samples, FI measures, and time frames (19). Few longitudinal studies are available, and none have applied repeated FI measurements to determine the effects of frequency and duration of FI on cognitive impairment.
FI and HIV are thought to have a bidirectional relation, as living with HIV reinforces FI and, in turn, FI may promote immunodeficiency and HIV disease progression (20, 21). Indeed, FI is associated with poorer adherence to antiretroviral therapy (ART), lower odds for viral suppression, and lower CD4+ cell counts in PLWH (22–24). Moreover, PLWH have increased nutritional needs, particularly as the HIV disease progresses (25, 26). FI can contribute to inflammation in PLWH (27), a key mechanism in HIV-associated neurocognitive disorders (28). Therefore, via behavioral and biological mechanisms alike, FI has the potential of contributing to or exacerbating cognitive dysfunction in PLWH.
To the best of our knowledge, no studies have examined the longitudinal relation between FI and cognitive impairment among PLWH and people living without HIV, considering the impact of the frequency of FI over time. The aim of this study was to determine whether FI is associated with cognitive impairment cross-sectionally and after a 2-y follow-up. We hypothesized that 1) FI would be independently associated with increased risk for cognitive impairment, 2) the risk for cognitive impairment would correlate with the severity and frequency of FI, and 3) the relation between FI and cognitive impairment would be stronger among PLWH than HIV-uninfected persons.
Methods
This was a longitudinal analysis of data collected between October 2016 and February 2020 from the ongoing Miami Adult Studies on HIV (MASH) cohort (NIDA grant U01-DA040381). Participants are ≥40 y of age, hepatitis B negative, and have documented HIV and hepatitis C virus (HCV) status (positive or negative). For this analysis, we used data collected at baseline and 12- and 24-mo follow-ups, excluding participants who were unable to complete cognitive testing (e.g., unable to read or write, n = 20) or were missing crucial data (n = 9). All participants provided written consent for participation in this study and release of medical records. The protocol for this study was approved by the Florida International University Institutional Review Board.
Food security status
FI was determined with the Household Food Security Module, which assesses a respondent's perceived food sufficiency and adequacy during the past 12 mo (29). Based on the number of affirmative responses, participants were classified as food-secure (score 0−2) or food-insecure (score ≥3). FI was further explored by levels of severity: full (score of 0), marginal (score of 1−2), low (score of 3−7 for households with children and score of 3−5 without children), and VLFS (score ≥8 with children and ≥6 without children). Additionally, we examined the frequency of FI over the 2-y period by using the total sum of time points with self-reported FI. FI (past 12 mo) was measured at baseline and at 12- and 24-mo follow-ups; thus, participants could report FI up to 3 times, equivalent to 3 y of FI. Reporting FI at all times was considered persistent FI.
Cognitive function
Cognitive function was assessed with the Mini-Mental State Examination (MMSE), using scores ≤24 out of 30 as the cutoff for impairment (30). The MMSE is a measure of global cognitive function and the most widely used screening tool for cognitive impairment, with a sensitivity and specificity of 81% and 89%, respectively, for the detection of dementia (31).
Additional covariates
FI has been associated with several conditions that may have an impact on cognitive function, including obesity, diabetes (9, 10), liver disease (32, 33), and mental health problems, such as depression (9) and substance abuse (34–37). Therefore, we considered these risk factors as potential confounders.
Sociodemographic characteristics were self-reported. Substance use was determined by self-report (past 30 d) and urine toxicology (American Bio Medica, Kinderhook, NY, USA). Hazardous alcohol consumption was determined with scores ≥8 in the Alcohol Use Disorders Identification Test (AUDIT) (38). Anthropometric measurements, vital signs, and fasting blood samples were obtained at all study visits. Obesity was defined as a BMI (kg/m2) ≥30, hyperglycemia as fasting blood glucose ≥100 mg/dL, and hypertension as ≥130 mm Hg systolic or ≥85 mm Hg diastolic blood pressure. Liver fibrosis was determined with the Fibrosis-4 Index (FIB-4), using a cutoff of 1.45 (39). Depression symptomatology was assessed with the Center for Epidemiological Studies–Depression Scale (CES-D), with scores ≥16 used to classify the presence of depressive symptoms (40). HIV viral loads and CD4 cell counts were obtained from medical records. HIV viral suppression was defined as having a plasma HIV RNA <200 copies/mL.
Statistical analyses
Descriptive statistics consisted of chi-square tests for categorical variables and t tests or 1-factor ANOVA for continuous variables, reported as n (%) or means ± SDs. Comparison of distributions between baseline and 2-y follow-up consisted of paired-samples t test and McNemar's test for continuous and categorical variables, respectively. The primary exposure of interest was FI and the primary outcome was cognitive impairment. Separate models were used to examine FI by levels of severity: 1) using the traditional classification of FI (low food security and VLFS) versus food-secure (full and marginal food security), 2) comparing marginal-to-very-low food security with full food security, and 3) comparing each of marginal, low, and VLFS with full food security. First, a cross-sectional analysis consisted of binary logistic regressions for cognitive impairment at baseline. Second, we performed binary logistic regressions for cognitive impairment at 2-y follow-up, controlling for baseline MMSE scores. In these analyses, we tested the effect of FI at baseline and the effect of the frequency of FI. We additionally performed multivariable regressions adjusting for sociodemographic characteristics (sex, race/ethnicity, income, household size), depressive symptoms, obesity, hyperglycemia, hypertension, liver fibrosis, and HIV and HCV infections. We tested for interaction effects between FI and covariates; these were removed from models if nonsignificant. Results are reported as ORs and 95% CIs and considered statistically significant at a 2-tailed P < 0.05. The data analysis for this article were generated using SAS software, version 9.4 (SAS Institute, Inc.).
Results
Population characteristics
A total of 394 participants with 24-mo follow-up data were included in this analysis. In the ongoing MASH cohort, the overall loss to follow-up has been ˜3% per year. The study population, as shown in Table 1, is largely comprised of economically disadvantaged (all <200% of the federal poverty level), middle-aged, Black, and Hispanic individuals. At baseline, 104 (26.4%) participants were food-insecure, with low and VLFS in 52 (13.2%) and 52 (13.2%), respectively. An additional 64 (16.2%) participants reported marginal food security. At 2-y follow-up, 104 (26.4%) were food-insecure, with marginal, low, and VLFS in 35 (8.9%), 50 (12.7%), and 54 (13.7%), respectively. Compared with food-secure participants, those who reported FI had fewer household members (P = 0.003), were more likely to use cocaine (P = 0.002), and to report depressive symptoms (63.5 vs. 32.1%; P < 0.0001). No differences were observed on metabolic parameters (BMI, obesity, hyperglycemia, and hypertension) or liver fibrosis. A comparison of sample characteristics by levels of FI can be found in Supplemental Table 1.
TABLE 1.
Population characteristics at baseline1
| Parameter | Total (n = 394) | Food-secure (n = 290) | Food-insecure (n = 104) | P 2 |
|---|---|---|---|---|
| Age, y | 53.9 ± 7.9 | 53.9 ± 8.1 | 54.2 ± 7.0 | 0.72 |
| Sex (male), n (%) | 211 (53.6) | 155 (53.5) | 56 (53.9) | 0.94 |
| Race/ethnicity, n (%) | ||||
| Black non-Hispanic | 283 (71.8) | 218 (75.2) | 65 (62.5) | 0.07 |
| White Hispanic | 55 (14.0) | 37 (12.8) | 18 (17.3) | |
| White non-Hispanic | 25 (6.4) | 17 (5.9) | 8 (7.7) | |
| Multiracial/other | 31 (7.9) | 18 (6.2) | 13 (12.5) | |
| Income (below poverty), n (%) | 309 (78.4) | 227 (78.3) | 82 (78.9) | 0.90 |
| Household size, n | 1.9 ± 1.3 | 2.0 ± 1.4 | 1.6 ± 1.0 | 0.003 |
| Education, n (%) | ||||
| Less than high school | 175 (44.4) | 131 (45.2) | 44 (42.3) | 0.82 |
| High school or GED | 12 (31.7) | 92 (31.7) | 33 (31.7) | |
| More than high school | 94 (23.9) | 67 (23.1) | 27 (26.0) | |
| Substances, n (%) | ||||
| AUDIT >8 | 100 (25.4) | 72 (24.8) | 28 (26.9) | 0.67 |
| Smoker (tobacco) | 212 (53.8) | 151 (52.1) | 61 (58.7) | 0.25 |
| Cannabis | 119 (30.2) | 84 (29.0) | 35 (33.7) | 0.37 |
| Cocaine | 159 (40.4) | 104 (35.9) | 55 (52.9) | 0.002 |
| Opioids | 40 (10.2) | 27 (9.3) | 13 (12.5) | 0.36 |
| Metabolic | ||||
| BMI, kg/m2 | 30.0 ± 7.1 | 30.1 ± 7.3 | 29.7 ± 6.3 | 0.65 |
| Obesity, n (%) | 175 (44.4) | 129 (44.5) | 46 (44.2) | 0.96 |
| Hyperglycemia, n (%) | 90 (22.8) | 66 (22.8) | 24 (23.1) | 0.95 |
| Hypertension, n (%) | 210 (53.4) | 151 (52.3) | 59 (56.7) | 0.43 |
| Hepatic | ||||
| Liver fibrosis (FIB-4 >1.45), n (%) | 112 (28.4) | 82 (29.3) | 30 (28.9) | 0.91 |
| HIV, n (%) | 247 (62.7) | 190 (65.5) | 57 (54.8) | 0.053 |
| On ART, n (%) | 246 (99.6) | 190 (100.0) | 56 (98.3) | 0.07 |
| HIV RNA <200 copies/mL, n(%) | 214 (86.6) | 168 (88.4) | 46 (80.7) | 0.13 |
| CD4 cells/mL3 | 602 ± 343 | 605 ± 352 | 593 ± 315 | 0.82 |
| HCV, n (%) | 43 (10.9) | 31 (10.7) | 12 (11.5) | 0.81 |
| HIV/HCV coinfection, n (%) | 21 (5.3) | 15 (5.2) | 6 (5.8) | 0.82 |
| Depressive symptoms (CES-D >16) , n (%) | 159 (40.4) | 93 (32.1) | 66 (63.5) | <0.0001 |
Values are n (%) or means ± SDs. AUDIT, Alcohol Use Disorders Identification Test; ART, antiretroviral therapy; CES-D, Center for Epidemiological Studies–Depression Scale; FIB-4, Fibrosis-4 Index; GED, General Educational Development; HCV, hepatitis C virus.
Chi-square tests for categorical factors and t test for continuous factors.
CD4+ cell count data on n = 232.
Food-insecure participants tended to be less likely infected with HIV than food-secure participants (54.8 vs. 65.5%, respectively; P = 0.053); conversely, fewer PLWH tended to report FI than HIV-uninfected participants (23.1 vs. 32.0%, respectively; P = 0.053). Of the 247 (62.7%) PLWH, 246 (99.6%) were receiving ART and 214 (86.6%) were virally suppressed. Also, 43 (10.9%) participants were infected with HCV, including 21 (5.3%) who were coinfected with both HIV and HCV. No participants acquired HIV or HCV infection throughout the study period.
Prevalence of cognitive impairment
At baseline, 58 (14.7%) participants had cognitive impairment, including 39 food-secure and 19 food-insecure participants. At 2-y follow-up, 49 (12.4%) participants had cognitive impairment, including 20 participants who did not have cognitive impairment at baseline. However, overall, there was no significant change in MMSE scores [t(393) = 1.89, P = 0.06] or in the frequency of cognitive impairment [χ2(1) = 1.65, P = 0.20] between baseline and 2-y follow-up.
Cognitive impairment at baseline
Table 2 shows the results of logistic regressions for cognitive impairment at baseline. Overall, FI was not associated with cognitive impairment. However, compared with full food security, VLFS was associated with 3.23 (95% CI: 1.08, 9.65; P = 0.04) times the risk for cognitive impairment. Similar results were obtained after adjusting for covariates.
TABLE 2.
Logistic regressions for cognitive impairment (MMSE ≤24) at baseline1
| Model and parameter | Crude OR (95% CI) | P | Adjusted OR (95% CI)2 | P |
|---|---|---|---|---|
| Model 1 | ||||
| Food insecurity3 | 1.59 (0.85, 2.97) | 0.15 | 1.56 (0.78, 3.14) | 0.21 |
| Model 2 | ||||
| Full vs. marginal, low, and very low food security | 1.11 (0.63, 1.95) | 0.72 | 1.05 (0.56, 1.98) | 0.89 |
| Model 3 | ||||
| Levels of food security | ||||
| Full | Reference | Reference | ||
| Marginal | 1.47 (0.59, 3.65) | 0.41 | 1.59 (0.61, 4.14) | 0.34 |
| Low | 1.45 (0.47, 4.50) | 0.52 | 1.29 (0.39, 4.32) | 0.68 |
| Very low | 3.23 (1.08, 9.65) | 0.04 | 4.16 (1.28, 13.55) | 0.02 |
n = 394. HCV, hepatitis C virus; MMSE, Mini-Mental State Examination.
Adjusted for age, years of education, race/ethnicity, sex, income, household size, obesity, hyperglycemia, hypertension, depressive symptoms, liver fibrosis, and HIV/HCV infection.
Food insecure (low and very low food security) vs. food secure (full and marginal food security).
In addition to these results, hypertension was associated with increased risk for cognitive impairment (OR: 1.80; 95% CI: 1.01, 3.22; P = 0.048) and there was a marginal association with liver fibrosis (OR: 1.67; 95% CI: 0.93, 2.99; P = 0.08). Depressive symptoms (OR: 1.35; 95% CI: 0.77, 2.36; P = 0.3), BMI (OR: 0.99; 95% CI: 0.95, 1.03; P = 0.7), obesity (OR: 0.94; 95% CI: 0.54, 1.65; P = 0.8), hyperglycemia (OR: 0.67; 95% CI: 0.32, 1.38; P = 0.3), HIV (OR: 1.52; 95% CI: 0.83, 2.79; P = 0.2), and HCV (OR: 1.63; 95% CI: 0.74, 3.61: P = 0.2) were not significantly associated with cognitive impairment. We tested for an FI–HIV interaction, but the effect was not significant (P = 0.92). On the other hand, the risk for impairment was significantly higher among PLWH who were not virally suppressed (OR: 2.87, 95% CI: 1.15, 7.18; P = 0.02) compared with HIV-uninfected participants.
Cognitive impairment at 2-y follow-up
We performed logistic regressions for cognitive impairment at 2-y follow-up, controlling for MMSE scores at baseline (Table 3). Baseline FI was associated with cognitive impairment at 2-y follow-up (OR: 2.28; 95% CI: 1.08, 4.81; P = 0.03). The association was also significant when comparing marginal-to-VLFS with full food security (OR: 2.47; 95% CI: 1.2, 5.10; P = 0.01). Additionally, for every instance that participants reported FI (baseline and at 12- and 24-mo follow-ups), the odds for cognitive impairment at follow-up increased 1.5 times (95% CI: 1.08, 2.10; P = 0.02). In particular, persistent FI—reporting FI throughout the entire study—was significantly associated with increased risk for cognitive impairment compared with those who never reported FI (OR: 4.18; 95% CI: 1.29, 13.59; P = 0.02). These relations remained significant in the fully adjusted multivariable model. None of the covariates were significantly associated with cognitive impairment at 2-y follow-up. There was no significant interaction effect between FI and HIV (P = 0.23 for FI at baseline; P = 0.58 for FI frequency).
TABLE 3.
Logistic regressions for cognitive impairment (MMSE ≤24) at 2-y follow-up1
| Model and category | OR (95% CI)2 | P | Adjusted OR (95% CI)3 | P |
|---|---|---|---|---|
| Model 1 | ||||
| FI at baseline4 | 2.28 (1.08, 4.81) | 0.03 | 4.21 (1.66, 10.68) | 0.002 |
| Model 2 | ||||
| Full vs. marginal, low, and very low food security at baseline | 2.47 (1.20, 5.10) | 0.01 | 4.41 (1.88, 10.35) | 0.001 |
| Model 3 | ||||
| Levels of food security at baseline | ||||
| Full | Reference | Reference | ||
| Marginal | 2.01 (0.86, 4.70) | 0.11 | 2.31 (0.92, 5.80) | 0.08 |
| Low | 2.80 (1.21, 6.49) | 0.02 | 3.34 (1.32, 8.43) | 0.01 |
| Very low | 3.02 (1.23, 7.41) | 0.02 | 4.62 (1.67, 12.80) | 0.003 |
| Model 4 | ||||
| Frequency of FI, per year | 1.50 (1.08, 2.10) | 0.02 | 1.85 (1.23, 2.77) | 0.003 |
| Model 5 | ||||
| 0 y of FI | Reference | Reference | ||
| 1 y of FI | 1.62 (0.61, 4.30) | 0.33 | 2.37 (0.82, 6.89) | 0.11 |
| 2 y of FI | 2.10 (0.85, 5.16) | 0.11 | 3.23 (1.06, 9.86) | 0.04 |
| 3 y of FI | 3.69 (1.15, 11.83) | 0.03 | 6.77 (1.72, 26.68) | 0.006 |
n = 394. FI, food insecurity; HCV, hepatitis C virus; MMSE, Mini-Mental State Examination.
Adjusted for MMSE scores at baseline.
Adjusted for age, years of education, MMSE scores at baseline, race/ethnicity, sex, income, household size, obesity, hyperglycemia, hypertension, depressive symptoms, liver fibrosis, and HIV/HCV infection.
Food insecure (low and very low food security) vs. food secure (full and marginal food security).
To validate these findings, we repeated the analysis among participants who were not cognitively impaired at baseline (n = 336), shown in Supplemental Table 2. The results remained consistent. Baseline FI was associated with 2.32 (95% CI: 0.91, 5.93; P = 0.08) times the odds for cognitive impairment. Likewise, having marginal-to-VLFS was significantly associated with increased risk for cognitive impairment at 2-y follow-up compared with full food security (OR: 4.24; 95% CI: 1.48, 12.10; P = 0.007). The frequency of FI was associated with 1.53 (95% CI: 1.01, 2.32; P = 0.047) times the odds for impairment for every year that participants experienced FI, and persistent FI was associated with 4.14 (95% CI: 1.09, 15.70; P = 0.04) times the odds for impairment compared with no FI.
Discussion
This study of the MASH cohort found that FI is a predictor of cognitive impairment, both at baseline and after a 2-y follow-up. This is the first study to establish an association between the frequency of FI, particularly persistent FI, and cognitive decline over time. Furthermore, our findings suggest that even marginal food security is a risk factor for cognitive decline, despite it being traditionally classified as food secure. FI was also associated with depressive symptoms and cocaine use, but these did not seem to affect cognitive impairment in this study. While FI did not show an impact on HIV treatment and viral load, PLWH who were not virally suppressed showed an increased risk for cognitive impairment at baseline compared with HIV-uninfected individuals. These findings are highly relevant, as FI is intrinsically linked to mental health and substance abuse (41), and can promote depression (9) and HIV disease progression (23, 24). It is possible that these factors, when compounded, may have additive or exponential effects on cognitive outcomes.
Both at baseline and at 2-y follow-up, VLFS more than tripled the odds for cognitive impairment compared with full food security. This finding is similar to those previously reported among homeless older adults (42) and middle-to-older-aged Hispanics (43). Two studies have also reported associations between FI and cognitive impairment using NHANES data from older adults (11, 13). Studies that have performed cognitive batteries suggest that FI mostly affects executive functions, such as processing speed, sustained attention, verbal fluency, working memory, and immediate learning ability (13, 14, 43).
Over the course of 2 y, 20 participants developed cognitive impairment, but the overall prevalence of cognitive impairment in the sample did not change. This may be due to fluctuations in cognitive function as seen in dementia (44, 45) and HIV-associated neurocognitive disorders (46). Nonetheless, similar to the findings in an HIV-uninfected cohort by Wong et al. (14), FI at baseline was associated with cognitive impairment at 2-y follow-up and the greatest impact was seen in association with VLFS. However, we also found that any level of FI at baseline, including marginal food security, was associated with cognitive impairment at follow-up. Moreover, when we excluded participants with cognitive impairment at baseline, even marginal food security was independently associated with cognitive impairment at follow-up compared with full food security. To the best of our knowledge, we are the first to show this association. In addition, we are the first to show an association between the frequency of FI, particularly persistent FI, and cognitive impairment; those who consistently experienced FI (reporting past 12-mo FI at baseline and at the 2 yearly follow-ups, equivalent to 3 years of FI) had the highest risk for cognitive impairment at 2-y follow-up. In the Women's Interagency HIV Study, persistent FI (defined as 2 consecutive 6-mo intervals of FI) was associated with increased risk for depression and poor mental well-being (47).
Despite the high prevalence of FI and cognitive impairments among PLWH, only 2 cross-sectional studies have previously examined how FI may contribute to cognitive impairment in PLWH. Using the Montreal Cognitive Assessment (MoCA), Hessol et al. (48) were unable to find a significant association between FI and cognitive impairment, possibly related to the high prevalence of impairment in that sample. In contrast, Hobkirk et al. (18) found a significant interaction effect between HIV and FI on the neurocognitive performance of 61 PLWH and 36 HIV-uninfected middle-aged adults. Among PLWH, FI was associated with significantly higher deficits in the domains of speed of information processing, learning, and motor function, but not memory. There were no significant differences in domain deficit scores among HIV-uninfected participants, possibly due to the small sample. Our findings support the hypothesis that FI may promote cognitive impairment in PLWH, independently of adherence to ART and HIV viral load, but we were unable to detect a differential effect between those living with or without HIV possibly due to the vast majority (87%) being virally suppressed. Nonetheless, in our study, PLWH who were not virally suppressed were at a significantly increased risk for cognitive impairment compared with HIV-uninfected participants. More sensitive neurocognitive testing, such as that performed by Hobkirk et al., with larger sample sizes may be able to identify HIV-associated neurocognitive dysfunctions altered by FI.
Interestingly, FI was associated with depressive symptoms and cocaine use, which can contribute to cognitive dysfunction (49, 50). Notably, 40% of all participants, including 64% of food-insecure participants in this study, reported depressive symptoms, which suggests a markedly higher prevalence of depression in this cohort than seen in other US population–based studies of food-insecure individuals (51, 52). It is possible that compounding factors in this population, such as minority status, poverty, FI, chronic disease burden, substance-use disorders, and HIV-related stigma, among others, may contribute to the high prevalence of depressive symptoms in this vulnerable population. In addition, FI was associated with cocaine use, which has been associated with impaired processing speed and executive functioning (53). Chronic substance abuse can lead to long-lasting cognitive impairments (50) and may have significant interactions with FI and mental health (41). For example, cocaine use may contribute to FI by affecting impulse control and reward-based decision making (54). Interestingly, neurocognitive impairment may play a role in poorer ART adherence among PLWH who use cocaine (55), thereby furthering the risk for cognitive dysfunction.
It is worth mentioning that while in this study FI was not associated with liver fibrosis (measured with the FIB-4), this association has been made by Golovaty et al. (32) using NHANES data. We have also found FI to be associated with advanced liver fibrosis when measured with magnetic resonance elastography (33); however, this assessment was not available for all the participants in this analysis. In this study, we observed a trend for liver fibrosis in association with cognitive impairment. Notably, liver disease is a major cause of morbidity and mortality in PLWH (56) and has been associated with cognitive impairments in PLWH (57, 58).
Although the exact mechanisms remain unknown, there are several potential ways by which FI may contribute to cognitive impairment. In resource-rich settings, FI is associated with maladaptive eating behaviors and poor diet quality, although overall energy intake may not be affected (59). Consequently, FI may result in deficiencies of essential nutrients for brain and cognitive function, such as B-vitamins and n–3 PUFAs, as well as dietary patterns that promote neuroinflammatory processes (60, 61). Our finding that persistent FI was associated with cognitive decline, for example, may be due in part to long-term nutritional inadequacies. A “neuroprotective dietary pattern,” consisting of antioxidant- and polyphenol-rich foods including fruits, vegetables, monounsaturated fats, and n–3 fatty acids (i.e., Mediterranean diet) (61, 62), may be inaccessible to people who experience FI. FI can also contribute to cognitive dysfunction through comorbidities directly affected by diet, such as cardiovascular disease and diabetes. These conditions may be aggravated by poor disease self-management as a result of monetary constraints, such as having to choose between food or medication. Our findings related to FI persistence may be particularly relevant, as persistent FI has been associated with increased cost-related medication nonadherence among older adults (63). Interestingly, FI has also been associated with shorter leukocyte telomere length in US adults ages 25−45 y, suggesting that FI may advance aging (64).
Nevertheless, FI is a modifiable risk factor for cognitive impairment. Thus, improving access to sufficient nutritious foods and improving dietary patterns among economically disadvantaged households may allow for lifestyle changes that improve nutritional status and reduce comorbidities, ultimately delaying cognitive decline. In the United States, several food-assistance programs have been implemented to reduce FI, the largest of which is the Supplemental Nutrition Assistance Program (SNAP), also known as food stamps. Although most eligible individuals participate in the program (84% in 2017) (65), many SNAP recipients remain food-insecure even after receiving benefits, as these are often insufficient to relieve monetary constraints (66). Furthermore, the current criteria for eligibility prevent many individuals from obtaining SNAP benefits (67). Therefore, there is a need for improved interventions to secure access to sufficient nutritious foods among socioeconomically disadvantaged groups. Given the consistent findings linking FI to cognitive function, food-assistance programs should account for cognitive impairments when planning benefits. Indeed, cognitive impairment has the potential to exacerbate FI by decreasing functional abilities (68), thereby limiting work opportunities and income, the ability to navigate processes to apply for food-assistance programs, and the ability for disease self-management. However, the effect of cognitive impairments on food security status is an interesting area for future research that has not yet been explored.
This study was conducted on a subset of MASH cohort participants who had completed 2-y follow-up assessments, as data collection was interrupted by the coronavirus disease 2019 (COVID-19) pandemic and any data collected after the pandemic might not be comparable to data collected prior to the pandemic. Nonetheless, the present study consists of a large number of participants with sufficient power for our analyses, as evidenced by the significant findings. Another limitation of this study is the use of the MMSE, a screener that has low sensitivity for identifying mild HIV-associated cognitive disorders. However, in light of our findings, FI may be a clinically relevant risk factor for severe cognitive impairments and dementia. Longitudinal studies using more sensitive neuropsychological batteries and repeated measures of food security will help further our understanding of the long-term effects of FI, including its severity, frequency, and persistence, on cognitive decline. Neuroimaging studies are needed to identify specific brain structures altered by FI. Potential mediators, such as dietary patterns, vascular risk factors, psychiatric comorbidities, and chronic stress, should be examined.
In conclusion, FI is a significant risk factor for cognitive impairment, particularly among individuals who experience FI frequently or persistently. Attention should be paid to PLWH, a population that is disproportionately affected by FI and cognitive dysfunction. Screening for FI and interventions to secure access to sufficient nutritious foods may help delay cognitive decline among socioeconomically disadvantaged individuals.
Supplementary Material
Acknowledgments
The authors’ responsibilities were as follows—MKB, AC, and SSM: designed the research; JAT: conducted the research; JAT and TL: analyzed data; JAT, CSM, and MKB: wrote the manuscript; KES: critically revised the manuscript; MKB: had primary responsibility for final content; and all authors: read and approved the final manuscript.
Notes
This work was supported by the National Institute on Drug Abuse at the National Institutes of Health under grant U01-DA040381.
Author disclosures: The authors report no conflicts of interest.
Supplemental Tables 1 and 2 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/jn/.
Abbreviations used: ART, antiretroviral therapy; FI, food insecurity; FIB-4, Fibrosis-4 Index; HCV, hepatitis C virus; MASH, Miami Adult Studies on HIV; MMSE, Mini-Mental State Examination; PLWH, people living with HIV; SNAP, Supplemental Nutrition Assistance Program; VLFS, very low food security.
Contributor Information
Javier A Tamargo, Department of Dietetics and Nutrition, Florida International University, Miami, FL, USA.
Christina S Meade, Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, NC, USA.
Adriana Campa, Department of Dietetics and Nutrition, Florida International University, Miami, FL, USA.
Sabrina S Martinez, Department of Dietetics and Nutrition, Florida International University, Miami, FL, USA.
Tan Li, Department of Biostatistics, Florida International University, Miami, FL, USA.
Kenneth E Sherman, Division of Digestive Diseases, University of Cincinnati College of Medicine, Cincinnati, OH, USA.
Marianna K Baum, Department of Dietetics and Nutrition, Florida International University, Miami, FL, USA.
Data Availability
Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.
References
- 1. Holben DH, Marshall MB. Position of the Academy of Nutrition and Dietetics: food insecurity in the United States. J Acad Nutr Diet. 2017;117(12):1991–2002. [DOI] [PubMed] [Google Scholar]
- 2. Coleman-Jensen A, Rabbitt MP, Gregory CA, Singh A. Household food security in the United States in 2019. [Internet]. US Department of Agriculture, Economic Research Service; 2020; [cited 2020 Oct 15]. Available from: https://www.ers.usda.gov/publications/pub-details/?pubid=99281. [Google Scholar]
- 3. Tarasuk VS. Household food insecurity with hunger is associated with women's food intakes, health and household circumstances. J Nutr. 2001;131(10):2670–6. [DOI] [PubMed] [Google Scholar]
- 4. Nord M, Coleman-Jensen A, Gregory C. Prevalence of U.S. food insecurity is related to changes in unemployment, inflation, and the price of food. [Internet]. USDA, Economic Research Service; 2014; [cited 2020 Oct 15]. Available from: https://www.ers.usda.gov/publications/pub-details/?pubid=45216. [Google Scholar]
- 5. USDA Economic Research Service . Frequency of food insecurity. [Internet]. USDA, Economic Research Service; 2019[cited 2020 Oct 15]. Available from: https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-us/frequency-of-food-insecurity/. [Google Scholar]
- 6. Centers for Disease Control and Prevention . Estimated HIV incidence and prevalence in the United States, 2014–2018. [Internet]. Centers for Disease Control and Prevention; 2020; [cited 2020 Aug 3]. Available from: http://www.cdc.gov/hiv/library/reports/hiv-surveillance.html. [Google Scholar]
- 7. Kalichman SC, Hernandez D, Cherry C, Kalichman MO, Washington C, Grebler T. Food insecurity and other poverty indicators among people living with HIV/AIDS: effects on treatment and health outcomes. J Community Health. 2014;39(6):1133–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Spinelli MA, Frongillo EA, Sheira LA, Palar K, Tien PC, Wilson T, Merenstein D, Cohen M, Adedimeji A, Wentz Eet al. Food insecurity is associated with poor HIV outcomes among women in the United States. AIDS Behav. 2017;21(12):3473–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Gundersen C, Ziliak JP. Food insecurity and health outcomes. Health Aff. 2015;34(11):1830–9. [DOI] [PubMed] [Google Scholar]
- 10. Gregory CA, Coleman-Jensen A. Food insecurity, chronic disease, and health among working-age adults, ERR-235. [Internet]. US Department of Agriculture, Economic Research Service; 2017; [cited 2020 Aug 3]. Available from: https://purl.fdlp.gov/GPO/gpo83486. [Google Scholar]
- 11. Frith E, Loprinzi PD. Food insecurity and cognitive function in older adults: brief report. Clin Nutr. 2018;37(5):1765–8. [DOI] [PubMed] [Google Scholar]
- 12. Millan MJ, Agid Y, Brüne M, Bullmore ET, Carter CS, Clayton NS, Connor R, Davis S, Deakin B, DeRubeis RJet al. Cognitive dysfunction in psychiatric disorders: characteristics, causes and the quest for improved therapy. Nat Rev Drug Discov. 2012;11(2):141–68. [DOI] [PubMed] [Google Scholar]
- 13. Portela-Parra ET, Leung CW. Food insecurity is associated with lower cognitive functioning in a national sample of older adults. J Nutr. 2019;149(10):1812–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wong JC, Scott T, Wilde P, Li YG, Tucker KL, Gao X. Food insecurity is associated with subsequent cognitive decline in the Boston Puerto Rican Health Study. J Nutr. 2016;146(9):1740–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Saloner R, Cysique LA. HIV-associated neurocognitive disorders: a global perspective. J Int Neuropsychol Soc. 2017;23(9-10):860–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Simioni S, Cavassini M, Annoni JM, Abraham AR, Bourquin I, Schiffer V, Calmy A, Chave JP, Giacobini E, Hirschel Bet al. Cognitive dysfunction in HIV patients despite long-standing suppression of viremia. AIDS. 2009;24(9):1243–50. [DOI] [PubMed] [Google Scholar]
- 17. Hatsu I, Hade E, Campa A. Food security status is related to mental health quality of life among persons living with HIV. AIDS Behav. 2017;21(3):745–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Hobkirk AL, Towe SL, Patel P, Meade CS. Food insecurity is associated with cognitive deficits among HIV-positive, but not HIV-negative, individuals in a United States sample. AIDS Behav. 2017;21(3):783–91. [DOI] [PubMed] [Google Scholar]
- 19. Na M, Dou N, Ji N, Xie D, Huang J, Tucker KL, Gao X. Food insecurity and cognitive function in middle to older adulthood: a systematic review. Adv Nutr. 2020;11(3):667–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Frega R, Duffy F, Rawat R, Grede N. Food insecurity in the context of HIV/AIDS: a framework for a new era of programming. Food Nutr Bull. 2010;31(4):S292–312. [PubMed] [Google Scholar]
- 21. Weiser SD, Young SL, Cohen CR, Kushel MB, Tsai AC, Tien PC, Hatcher AM, Frongillo EA, Bangsberg DR. Conceptual framework for understanding the bidirectional links between food insecurity and HIV/AIDS. Am J Clin Nutr. 2011;94(6):1729S–39S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Singer AW, Weiser SD, McCoy SI. Does food insecurity undermine adherence to antiretroviral therapy? A systematic review. AIDS Behav. 2015;19(8):1510–26. [DOI] [PubMed] [Google Scholar]
- 23. Aibibula W, Cox J, Hamelin AM, McLinden T, Klein MB, Brassard P. Association between food insecurity and HIV viral suppression: a systematic review and meta-analysis. AIDS Behav. 2017;21(3):754–65. [DOI] [PubMed] [Google Scholar]
- 24. Aibibula W, Cox J, Hamelin AM, Mamiya H, Klein MB, Brassard P. Food insecurity and low CD4 count among HIV-infected people: a systematic review and meta-analysis. AIDS Care. 2016;28(12):1577–85. [DOI] [PubMed] [Google Scholar]
- 25.World Health Organization. Nutrient requirements for people living with HIV/AIDS: report of a technical consultation. [Internet]. Geneva (Switzerland): World Health Organization; 2003; [cited 2020 Aug 3]. Available from: https://apps.who.int/iris/handle/10665/42853. [Google Scholar]
- 26. Coyne-Meyers K, Trombley LE. A review of nutrition in human immunodeficiency virus infection in the era of highly active antiretroviral therapy. Nutr Clin Pract. 2004;19(4):340–55. [DOI] [PubMed] [Google Scholar]
- 27. Leddy AM, Roque A, Sheira LA, Frongillo EA, Landay AL, Adedimeji AA, Wilson TE, Merenstein D, Wentz E, Adimora AAet al. Food insecurity is associated with inflammation among women living with HIV. J Infect Dis. 2019;219(3):429–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Alford K, Vera JH. Cognitive impairment in people living with HIV in the ART era: a review. Br Med Bull. 2018;127(1):55–68. [DOI] [PubMed] [Google Scholar]
- 29. Bickel G, Nord M, Price C, Hamilton W, Cook J. Guide to measuring household food security, revised 2000. [Internet]. Alexandria (VA): US Department of Agriculture, Food and Nutrition Service; 2000; [cited 2020 Aug 3]. Available from: https://www.fns.usda.gov/guide-measuring-household-food-security-revised-2000. [Google Scholar]
- 30. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”: A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–98. [DOI] [PubMed] [Google Scholar]
- 31. Tsoi KK, Chan JY, Hirai HW, Wong SY, Kwok TC. Cognitive tests to detect dementia: a systematic review and meta-analysis. JAMA Intern Med. 2015;175(9):1450–8. [DOI] [PubMed] [Google Scholar]
- 32. Golovaty I, Tien PC, Price JC, Sheira L, Seligman H, Weiser SD. Food insecurity may be an independent risk factor associated with nonalcoholic fatty liver disease among low-income adults in the United States. J Nutr. 2020;150(1):91–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Tamargo JA, Sherman KE, Campa A, Martinez SS, Hernandez J, Teeman C, Mandler RN, Chen J, Ehman RL, Baum MK. Food insecurity is associated with magnetic resonance-determined nonalcoholic fatty liver and liver fibrosis in low-income middle-aged adults with and without HIV. Am J Clin Nutr. 2021. doi: 10.1093/ajcn/nqaa362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Baer TE, Scherer EA, Fleegler EW, Hassan A. Food insecurity and the burden of health-related social problems in an urban youth population. J Adolesc Health. 2015;57(6):601–7. [DOI] [PubMed] [Google Scholar]
- 35. Bergmans RS, Coughlin L, Wilson T, Malecki K. Cross-sectional associations of food insecurity with smoking cigarettes and heavy alcohol use in a population-based sample of adults. Drug Alcohol Depend. 2019;205:107646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Campa A, Yang Z, Lai S, Xue L, Phillips JC, Sales S, Page JB, Baum MK. HIV-related wasting in HIV-infected drug users in the era of highly active antiretroviral therapy. Clin Infect Dis. 2005;41(8):1179–85. [DOI] [PubMed] [Google Scholar]
- 37. Whittle HJ, Sheira LA, Frongillo EA, Palar K, Cohen J, Merenstein D, Wilson TE, Adedimeji A, Cohen MH, Adimora AAet al. Longitudinal associations between food insecurity and substance use in a cohort of women with or at risk for HIV in the United States. Addiction. 2019;114(1):127–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The Alcohol Use Disorders Identification Test: guidelines for use in primary care. [Internet]. Department of Mental Health and Substance Dependence; 2001; [cited 2020 Aug 3]. Report no.: WHO/MSD/MSB/01.6a. Available from: https://www.who.int/publications-detail/audit-the-alcohol-use-disorders-identification-test-guidelines-for-use-in-primary-health-care. [Google Scholar]
- 39. Sterling RK, Lissen E, Clumeck N, Sola R, Correa MC, Montaner J, M SS, Torriani FJ, Dieterich DT, Thomas DLet al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43(6):1317–25. [DOI] [PubMed] [Google Scholar]
- 40. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385–401. [Google Scholar]
- 41. Davison KM, Holloway C, Gondara L, Hatcher AS. Independent associations and effect modification between lifetime substance use and recent mood disorder diagnosis with household food insecurity. PLoS One. 2018;13(1):e0191072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Tong M, Tieu L, Lee CT, Ponath C, Guzman D, Kushel M. Factors associated with food insecurity among older homeless adults: results from the HOPE HOME study. J Public Health (Oxf). 2019;41(2):240–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Gao X, Scott T, Falcon LM, Wilde PE, Tucker KL. Food insecurity and cognitive function in Puerto Rican adults. Am J Clin Nutr. 2009;89(4):1197–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Mainland BJ, Herrmann N, Mallya S, Fiocco AJ, Sin GL, Shulman KI, Ornstein TJ. Cognitive fluctuations and cognitive test performance among institutionalized persons with dementia. Am J Alzheimers Dis Other Demen. 2017;32(7):393–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Escandon A, Al-Hammadi N, Galvin JE. Effect of cognitive fluctuation on neuropsychological performance in aging and dementia. Neurology. 2010;74(3):210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Valcour VG. HIV, aging, and cognition: emerging issues. Top Antivir Med. 2013;21(3):119–23. [PMC free article] [PubMed] [Google Scholar]
- 47. Tuthill EL, Sheira LA, Palar K, Frongillo EA, Wilson TE, Adedimeji A, Merenstein D, Cohen MH, Wentz EL, Adimora AAet al. Persistent food insecurity is associated with adverse mental health among women living with or at risk of HIV in the United States. J Nutr. 2019;149(2):240–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Hessol NA, Zepf R, Zobell E, Weiser SD, John MD. Food insecurity and aging outcomes in older adults living with HIV. AIDS Behav. 2017;21(12):3506–14. [DOI] [PubMed] [Google Scholar]
- 49. Perini G, Cotta Ramusino M, Sinforiani E, Bernini S, Petrachi R, Costa A. Cognitive impairment in depression: recent advances and novel treatments. Neuropsychiatr Dis Treat. 2019;15:1249–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Gould TJ. Addiction and cognition. Addict Sci Clin Pract. 2010;5(2):4–14. [PMC free article] [PubMed] [Google Scholar]
- 51. 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):622–7. [DOI] [PubMed] [Google Scholar]
- 52. Montgomery J, Lu J, Ratliff S, Mezuk B. Food insecurity and depression among adults with diabetes: results from the National Health and Nutrition Examination Survey (NHANES). Diabetes Educ. 2017;43(3):260–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Meade CS, Towe SL, Skalski LM, Robertson KR. Independent effects of HIV infection and cocaine dependence on neurocognitive impairment in a community sample living in the southern United States. Drug Alcohol Depend. 2015;149:128–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Hobkirk AL, Bell RP, Utevsky AV, Huettel S, Meade CS. Reward and executive control network resting-state functional connectivity is associated with impulsivity during reward-based decision making for cocaine users. Drug Alcohol Depend. 2019;194:32–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Meade CS, Conn NA, Skalski LM, Safren SA. Neurocognitive impairment and medication adherence in HIV patients with and without cocaine dependence. J Behav Med. 2011;34(2):128–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Sherman KE, Peters MG, Thomas DL. HIV and the liver. Top Antivir Med. 2019;27(3):101–10. [PMC free article] [PubMed] [Google Scholar]
- 57. Ciccarelli N, Fabbiani M, Brita AC, De Marco R, Grima P, Gagliardini R, Borghetti A, Cauda R, Di Giambenedetto S. Liver fibrosis is associated with cognitive impairment in people living with HIV. Infection. 2019;47(4):589–93. [DOI] [PubMed] [Google Scholar]
- 58. Valcour VG, Rubin LH, Obasi MU, Maki PM, Peters MG, Levin S, Crystal HA, Young MA, Mack WJ, Cohen MHet al. Liver fibrosis linked to cognitive performance in HIV and hepatitis C. JAIDS. 2016;72(3):266–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Morales ME, Berkowitz SA. The relationship between food insecurity, dietary patterns, and obesity. Curr Nutr Rep. 2016;5(1):54–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60. Spencer SJ, Korosi A, Layé S, Shukitt-Hale B, Barrientos RM. Food for thought: how nutrition impacts cognition and emotion. NPJ Sci Food. 2017;1(1):7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61. Scarmeas N, Anastasiou CA, Yannakoulia M. Nutrition and prevention of cognitive impairment. Lancet Neurology. 2018;17(11):1006–15. [DOI] [PubMed] [Google Scholar]
- 62. Abbatecola AM, Russo M, Barbieri M. Dietary patterns and cognition in older persons. Curr Opin Clin Nutr Metab Care. 2018;21(1):10–3. [DOI] [PubMed] [Google Scholar]
- 63. Sattler EL, Lee JS. Persistent food insecurity is associated with higher levels of cost-related medication nonadherence in low-income older adults. J Nutr Gerontol Geriatr. 2013;32(1):41–58. [DOI] [PubMed] [Google Scholar]
- 64. Mazidi M, Kengne AP, Vatanparast H. Food security and leukocyte telomere length in adult Americans. Oxid Med Cell Longev. 2017;2017:1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65. Vigil A. Trends in Supplemental Nutrition Assistance Program participation rates: fiscal year 2010 to fiscal year 2017. [Internet]. USDA, Food and Nutrition Service; 2019[cited 2020 Aug 3]. Available from: https://www.fns.usda.gov/snap/trends-supplemental-nutrition-assistance-program-participation-rates-fiscal-year-2010. [Google Scholar]
- 66. Carlson S. More adequate SNAP benefits would help millions of participants better afford food. [Internet]. Center on Budget and Policy Priorities, 2019; [cited 2020 Aug 3]. Available from: https://www.cbpp.org/research/food-assistance/more-adequate-snap-benefits-would-help-millions-of-participants-better. [Google Scholar]
- 67. Keith-Jennings B, Llobrera J, Dean S. Links of the Supplemental Nutrition Assistance Program with food insecurity, poverty, and health: evidence and potential. Am J Public Health. 2019;109(12):1636–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68. Lindbergh CA, Dishman RK, Miller LS. Functional disability in mild cognitive impairment: a systematic review and meta-analysis. Neuropsychol Rev. 2016;26(2):129–59. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.