Abstract
Introduction
Among individuals with low income, medication cost is a well-established barrier to medication adherence. Spending less on basic needs to pay for medication is a particularly concerning cost-coping strategy and may be associated with worse health outcomes. The aims of this study were (1) to describe the demographic and health status characteristics of those who report spending less on basic needs to pay for medication, and (2) to understand the associated psychosocial and financial challenges of individuals who make these choices.
Methods
We administered a survey to primarily low-income adults (n=270) in St. Louis, Missouri as part of a larger study from 2016–2018. Logistic regression was used to model the odds of reporting spending less on basic needs in order to pay for medication.
Results
Spending less on basic needs to pay for medication was significantly more likely in individuals with fair or poor health status (OR = 2.00; 95% CI, 1.03 to 3.93), greater number of chronic conditions (OR = 1.21 per additional 1 condition, 95% CI, 1.06 to 1.40), greater medication expenditure (OR = 1.41 per additional $100; 95% CI, 1.11 to 1.89), and difficulty paying bills (OR = 1.43; 95% CI, 1.11 to 1.85). Individuals who spent less on basic needs were less likely to be fully adherent to their medication regimen (OR = 0.38; 95% CI, 0.18 to 0.77).
Conclusions
Screening for unmet basic needs and offering referrals to social safety net programs in the primary care setting may help patients achieve sustainable medication adherence.
Introduction
Cost-related nonadherence (CRN) is a serious public health concern. It is estimated that approximately 50% of patients may not be taking their medications as prescribed; cost is one of the most common reasons for non-adherence (1, 2). CRN is associated with worse patient outcomes, unnecessary hospitalizations, as well as billions of dollars of avoidable healthcare costs (3, 4). Particularly among people with low income, cost is a well-established barrier to medication adherence (5).
Many people with low income are unable to adhere to medication regimens, and also face difficulties in meeting basic needs such as food, clothing, housing, and transportation. Previous studies have found that inability to meet basic needs is associated with lower medication adherence (6–9). Several reasons may explain this association: some individuals have no money for either basic necessities or medication; others with slightly greater means may spend less on medication in an attempt to meet food needs (9–11); conversely, other patients spend less on basic needs to afford medications (12–14). Individuals spending less on basic needs to afford medications were more likely to have a greater number of prescriptions and greater out-of-pocket medication costs; spending less on basic needs is also associated with a lower likelihood of actual medication adherence (12). Previous nationwide studies found that redirecting personal spending from basic needs to healthcare was more common among female, non-white, younger, and low-income individuals (13, 15).
The tradeoff of various expenses can be a daily reality with serious consequences for people with low income and/or those who are uninsured (16). One study found that elderly women who were at risk of food insufficiency (i.e., had to choose between food and either medication or paying bills) tended to have lower intake of energy, protein, and micronutrients (17). Another study of adults with HIV found that both spending less on basic needs to pay for medical care and the reverse (spending less on medical care to afford basic needs) were associated with greater number of visits to the emergency department (18). Indeed, a tradeoff in either direction can compromise a person’s overall well-being. However, unlike individuals who spend less on medication, those who spend less on basic needs may go unnoticed by a physician focused on assessing adherence to treatment. It is important to understand which populations are at risk for making this tradeoff, and how healthcare providers and policymakers can best assist these individuals.
Little research to date has focused on individuals who choose medication over basic needs. As such, the primary aim of this analysis was to describe the demographic characteristics, health status, and health insurance status of those who spend less on basic needs to pay for medication. Our secondary aims were to assess (1) the association between monthly medication expenses and a decision to spend less on basic needs, as well as (2) the associated psychosocial and financial challenges of individuals who make these choices.
Methods
We conducted a mixed-methods study (2016–2018) to assess cost-coping behaviors and cost-related medication non-adherence in low-income individuals with chronic disease. The parent study included a comprehensive quantitative survey completed by all participants and qualitative semi-structured interviews completed by a subset. This analysis reports on the survey data; limited data from the qualitative interviews and quantitative survey have been presented elsewhere (19, 20). Cognitive interviews were used to refine the survey before administration to study participants. Participants were recruited from multiple sites throughout St. Louis, Missouri, including the waiting rooms at four sites of a Federally Qualified Health Center, an academic medical center, and local newspaper advertisements. Some participants learned about the study by word-of-mouth from previous participants. Participants completed a 45–60 minute interviewer-administered cross-sectional survey that included two open-ended, audio-recorded questions.
Eligible participants were English-speaking adults aged 35–80 who had at least one chronic condition and at least one prescription for this diagnosis, whether or not the prescription was currently being filled. Trained research staff verbally administered the survey to participants in-person and entered responses into REDCap in real time. Written informed consent was obtained from all participants, and participants received a $50 gift card upon survey completion. After completion of data collection, a summary of findings was distributed to all participants who indicated a desire to receive them. All study materials and procedures were approved by the Washington University Human Research Protection Office on March 9, 2016.
Measures
Outcome measure. A single item asked participants, “Have you spent less on basic needs in order to pay for medication?” (yes/no). Participants answering “yes” were then asked which basic needs they had spent less on (food/groceries; utilities; gas/transportation; housing/rent; other medical expenses; other).
Socio-demographic measures included gender, age (continuous), race/ethnicity (Non-Hispanic black vs. other), education level (no high school (HS) diploma; high school graduate or GED; some college or technical school with HS diploma; college graduate or greater), health insurance (either (1) Medicare, private or dual Medicaid and Medicare; (2) Medicaid alone, local coverage program, or VA; or (3) uninsured), prescription drug coverage (covered for past 12 months vs. not), employment (either (1) employed, out of work for < 1 year, or homemaker; or (2) out of work for > 1 year, unable to work, disabled, or retired), monthly income (<$400; $400–799; $800–1199; >$1200), and number of dependents (0; 1; 2 or more).
Health measures included self-reported overall health (excellent/good vs. fair/poor), number of chronic conditions, number of medications, and monthly medication expenditure (including cost of unfilled medications; in hundreds of dollars).
Medication adherence measure. Participants were asked to report all prescriptions. Responses included some PRN (as needed) and over-the-counter medications. For each prescription, participants were asked, “How often do you fill it?”; response options were “usually fill on time”, “sometimes delay filling”, or “never filled”. Participants were also asked, “How often do you miss a dose or take less than prescribed?”; response options were “usually take as directed”, “sometimes miss or reduce”, or “often miss or reduce”. Adherence was defined as “usually” taking all medications on time and at full dosage; anything less was classified as “less than adherent”.
Covariates
Basic needs and financial problems measures. Our measure of unmet needs over the past 12 months included food, gas/transport, housing/rent, utilities, and medical expenses. Our measure of unmet needs at present included food, transport, housing, furniture, and difficulty paying bills. One point was assigned for each category of unmet need. For each aggregate measure, a score ranging from 0 (least unmet need) to 5 (most) was computed by summing across all categories within the measure. Additional measures addressing basic needs identified participants who were eligible for or receiving food stamps, or had visited a food pantry in the last 30 days. We asked participants to rate the quality and quantity of food in their household (enough quantity and good quality; enough quantity but lacking in quality; sometimes/often not enough to eat). Housing-related measures included number of moves in the past 3 months (0 vs. 1+) and a binary measure reflecting current homelessness or living in transitional housing or a shelter.
Financial status measures included not having a bank account and not having enough money to make ends meet. “Ability to make ends meet” is a measure that has previously been used to distinguish varying financial situations within low-income populations (21). Difficulty paying bills assessed whether the respondent ever “had a bill that was past due”, “paid the minimum on a utility bill”, “paid a utility bill late”, or “paid rent or mortgage late”. We also assessed financial difficulties including having a disconnected phone, any disconnected utility, a notice from a collection agency, ever having filed for bankruptcy, and inability to afford a minor emergency. For each aggregate measure, a score was computed by assigning one point for each problem and summing across all problems within the measure.
Psychosocial effects of financial problems were assessed with three items that asked how often participants worry about making ends meet, how often financial worries interfere with sleep or cause physical symptoms, and how often financial problems interfere with or limit relationships with other people. Response options were “never”, “rarely”, “sometimes”, “often”, and “always”.
Statistical Analysis
Descriptive statistics were first calculated for our outcome variable and all correlates. We used logistic regression to model the odds of spending less on basic needs to pay for medication. Predictors that were significantly associated with the outcome in bivariate models were included in the final multivariable model. However, several constructs were assessed with multiple different measures (e.g., comorbidity burden as the number of chronic conditions vs. number of medications; unmet basic needs at present vs. past 12 months; ability to pay bills vs. resulting financial difficulties; psychosocial effects of financial problems as effect on relationships vs. financial worry vs. physical effects of worry). In these cases, to avoid multi-collinearity in the regression model, we chose only one measure to include in the final multivariable model. If only one measure was significantly associated with the outcome in bivariate models, then that measure was chosen; otherwise, we chose the measure more conceptually linked to spending less on basic needs. We further assessed the bivariate relationship between potentially correlated constructs (number of chronic conditions and monthly medication expenses; difficulty paying bills and not enough money to make ends meet) to avoid multi-collinearity. We also evaluated a potential interaction term between monthly medication expenses and medication adherence. All statistical tests were two-sided. Statistical significance was assessed at α=0.05. Statistical analyses were performed in R version 3.6.1.
Results
Participant socio-demographic and health characteristics.
Of the 274 recruited participants, 270 were ultimately included in the analytic sample (three no longer met age requirements when re-checked prior to survey administration and one did not complete the survey). Of these 270 individuals, 59.6% were female, 81.9% identified as non-Hispanic black, and 68.2% reported a monthly income of less than $1,200 (Table 1). Sixty-eight percent of respondents had continuous health insurance for the past year, and the same proportion reported continuous past-year prescription drug coverage. On average, participants had been diagnosed with 4.8 chronic conditions and prescribed 6.1 medications. Average monthly medication expenditure was $82, although this amount was highly variable. Many participants (38.5%) reported asking their provider to switch their prescriptions to generic, and 95.2% of participants reported that their requests were successful.
Table 1.
Sample characteristics of respondents who had vs. had not spent less on basic needs in order to pay for medication (St. Louis, Missouri; 2016–2018), n=270
| Total | Not Spending Less on Basic Needs
(n = 144) |
Spending Less on Basic Needs (n = 126) |
|
|---|---|---|---|
|
| |||
| N (%) | N (%) | N (%) | |
|
| |||
| Gender | |||
| Female | 161 (59.6) | 78 (54.2) | 83 (65.9) |
| Race/ethnicity (ref: Non-Hispanic white/other) | |||
| Non-Hispanic black | 221 (81.9) | 121 (84.0) | 100 (79.4) |
| Education level | |||
| No high school (HS) diploma | 57 (21.1) | 37 (25.7) | 20 (15.9) |
| High school graduate or GED | 101 (37.4) | 49 (34.0) | 52 (41.3) |
| Some college/technical school (with HS diploma) | 68 (25.2) | 35 (24.3) | 33 (26.2) |
| College graduate or greater | 44 (16.3) | 23 (16.0) | 21 (16.7) |
| Employment status (ref: Employed; out of work < 1 year; homemaker) | |||
| Out of work > 1 year; unable to work; disabled; Retired | 169 (62.8) | 91 (63.2) | 78 (62.4) |
| Monthly income | |||
| < $400/month | 46 (17.2) | 24 (16.9) | 22 (17.6) |
| $400–$799/month | 77 (28.8) | 49 (34.5) | 28 (22.4) |
| $800–$1,199/month | 59 (22.1) | 26 (18.3) | 33 (26.4) |
| > $1,200/month | 85 (31.8) | 43 (30.3) | 42 (33.6) |
| Number of dependents | |||
| 0 | 184 (68.1) | 96 (66.7) | 88 (69.8) |
| 1 | 38 (14.1) | 22 (15.3) | 16 (12.7) |
| 2+ | 48 (17.8) | 26 (18.1) | 22 (17.5) |
| Health insurance status | |||
| Private; Medicare; dual | 119 (44.1) | 65 (45.1) | 54 (42.9) |
| Medicaid; local coverage program; Veteran’s Health benefits | 115 (42.6) | 64 (44.4) | 51 (40.5) |
| Uninsured | 36 (13.3) | 15 (10.4) | 21 (16.7) |
| Prescription drug coverage (ref: Full year) | |||
| Less than full year | 87 (32.2) | 38 (26.4) | 49 (38.9) |
| Self-reported overall health (ref: Excellent/Good) | |||
| Fair/Poor | 180 (66.7) | 84 (58.3) | 96 (76.2) |
| Medication adherence (ref: Adherent) | |||
| Less than adherent | 183 (68.0) | 79 (54.9) | 104 (83.2) |
|
| |||
| Mean ± SD | Mean ± SD | Mean ± SD | |
|
| |||
| Age | 55.2 ± 9.6 | 54.9 ± 10.3 | 55.5 ± 8.9 |
| Number of chronic conditions | 4.8 ± 2.5 | 4.1 ± 2.2 | 5.5 ± 2.6 |
| Number of medications | 6.1 ± 3.9 | 5.3 ± 3.7 | 7.0 ± 3.8 |
| Monthly medication expenditure ($) | 82 ± 152 | 47 ± 122 | 113 ± 179 |
Note: Due to rounding, not all percentages may add to exactly 100.
Spending less on basic needs to pay for medication.
Forty-seven percent of respondents reported spending less on at least one basic need to pay for medication, and 32% of respondents reported spending less on more than one basic need. The most common basic need participants spent less on was food (80.2%), followed by transportation (49.2%), utilities (47.6%), housing (27.8%), and other medical expenses (23.0%). Other basic needs that participants reported spending less on included household, leisure, child-related or personal expenses, and debt payments. Of the 47% of individuals who reported spending less on basic needs, only 16.8% reported full medication adherence.
Psychosocial effects of financial problems.
Many respondents reported at least sometimes experiencing: worry about being able to make their monthly living expenses (80.8%), difficulty sleeping or other physical effects due to financial worries (64.5%), and strained interpersonal relationships due to financial problems (61.8%).
Bivariate associations with spending less on basic needs.
Spending less on basic needs to pay for medication was more likely among females and those with a high school education (compared to no high school diploma) (Table 2). It was also more common among individuals who did not have prescription drug coverage for all of the preceding 12 months, as well as those who had worse overall self-reported health, a greater number of chronic disease diagnoses, a greater number of medications, or greater medication expenditure (Table 2). Spending less on basic needs was also associated with decreased odds of medication adherence and greater number of unmet basic needs over the past 12 months, but not unmet basic needs at present (Table 2). Eligibility for or receiving food stamps and visiting a food pantry were not associated with spending less on basic needs to pay for medications, but self-reported inability to afford a sufficient amount of food perceived as high-quality was (Table 2). Finally, not having enough money to make ends meet, difficulty paying bills, financial problems, and all three measures of the psychosocial effects of financial problems were associated with spending less on basic needs to afford medications (Table 2).
Table 2.
Bivariate associations with spending less on basic needs in order to pay for medication (St. Louis, Missouri; 2016–2018), n=270.
| Variable | Odds Ratio | 95% Confidence Interval | P-value |
|---|---|---|---|
|
| |||
| Gender: Male | 1.00 | referent | |
| Female | 1.63 | (1.00, 2.69) | 0.05 |
| Age | 1.01 | (0.98, 1.03) | 0.57 |
| Race/ethnicity: Non-Hispanic black | 1.00 | referent | |
| Non-Hispanic white/other | 1.37 | (0.74, 2.56) | 0.32 |
| Education level: no HS diploma | 1.00 | referent | |
| High school graduate or GED | 1.96 | (1.01, 3.88) | 0.05 |
| Some college/technical school (with HS diploma) | 1.74 | (0.85, 3.63) | 0.13 |
| College graduate or greater | 1.69 | (0.76, 3.80) | 0.20 |
| Employment status: employed; out of work < 1 year; homemaker | 1.00 | referent | |
| Out of work > 1 year; unable to work; disabled; retired | 0.97 | (0.59, 1.59) | 0.89 |
| Monthly income: < $400/month | 1.00 | referent | |
| $400–$799/month | 0.62 | (0.30, 1.31) | 0.21 |
| $800–$1,199/month | 1.38 | (0.64, 3.02) | 0.41 |
| > $1,200/month | 1.07 | (0.52, 2.19) | 0.86 |
| Number of dependents: 0 | 1.00 | referent | |
| 1 | 0.79 | (0.39, 1.60) | 0.52 |
| 2+ | 0.92 | (0.49, 1.74) | 0.81 |
| Health insurance status: Private; Medicare; dual | 1.00 | referent | |
| Medicaid; local coverage program; Veteran’s Health benefits | 0.96 | (0.57, 1.61) | 0.87 |
| Uninsured | 1.69 | (0.80, 3.64) | 0.18 |
| Prescription drug coverage: Full year | 1.00 | referent | |
| Less than full year | 1.78 | (1.06, 2.98) | 0.03 |
| Self-reported overall health: Excellent/Good | 1.00 | referent | |
| Fair/Poor | 2.29 | (1.36, 3.91) | <0.01 |
| Number of chronic conditions | 1.27 | (1.15, 1.43) | <0.01 |
| Number of medications | 1.13 | (1.06, 1.21) | <0.01 |
| Monthly medication expenditure (hundreds of dollars) | 1.66 | (1.29, 2.24) | <0.01 |
| Medication adherence | 0.25 | (0.14, 0.43) | <0.01 |
| Unmet basic needs over the past 12 months | 1.31 | (1.11, 1.54) | <0.01 |
| Unmet basic needs at present | 0.82 | (0.64, 1.05) | 0.12 |
| Receiving/eligible for food stamps | 0.89 | (0.54, 1.46) | 0.64 |
| Visited food pantry in last 30 days | 1.13 | (0.70, 1.84) | 0.62 |
| Characterization of food at home: enough quantity, good quality | 1.00 | referent | |
| Enough food, but lacking in quality | 4.40 | (2.37, 8.44) | <0.01 |
| Sometimes/often not enough to eat | 4.35 | (2.21, 8.82) | <0.01 |
| Number of moves in past 3 months: 0 | 1.00 | referent | |
| 1+ | 0.94 | (0.49, 1.78) | 0.85 |
| Transitional housing, shelter, or homeless | 1.11 | (0.57, 2.15) | 0.77 |
| No bank account (savings or checking) | 0.89 | (0.55, 1.44) | 0.64 |
| Not enough money to make ends meet | 2.72 | (1.67, 4.48) | <0.01 |
| Difficulty paying bills | 1.51 | (1.24, 1.85) | <0.01 |
| Financial problems | 1.48 | (1.24, 1.79) | <0.01 |
| Worry about making monthly living expenses (at least sometimes vs. rarely or never) | 4.16 | (2.10, 8.90) | <0.01 |
| Unable to sleep well or other physical effects of financial worry (at least sometimes vs. rarely or never) | 3.93 | (2.30, 6.88) | <0.01 |
| Financial problems interfere with or limit relationships (at least sometimes vs. rarely or never) | 2.67 | (1.60, 4.50) | <0.01 |
Most demographic characteristics (gender, age, race, employment, income, and number of dependents) were not significantly associated with spending less on basic needs. Lack of a bank account, living in transitional housing or a shelter, at least one move in the past three months, and homelessness were also not associated with spending less on basic needs to pay for medication.
Multivariable model.
Spending less on basic needs to afford medications was significantly more likely in individuals with fair or poor health status (OR = 2.00; 95% CI, 1.03 to 3.93), greater number of chronic conditions (OR = 1.21 per additional 1 condition, 95% CI, 1.06 to 1.40), greater medication expenditure (OR = 1.41 per $100; 95% CI, 1.11 to 1.89), and difficulty paying bills (OR = 1.43; 95% CI, 1.11 to 1.85) (Table 3). Individuals who spent less on basic needs were less likely to be fully adherent to their medication regimen (OR = 0.38; 95% CI, 0.18 to 0.77), controlling for all other covariates. All other variables (education level, prescription drug coverage, number of unmet basic needs over the past 12 months, inability to afford foods perceived as high-quality, not enough money to make ends meet, and worry about living expenses) showed no statistically significant association in the adjusted model. Using alternate measures for certain constructs (comorbidity burden, financial consequences, psychosocial effects of financial problems) did not change statistically significant results.
Table 3.
Multivariable logistic regression model examining characteristics of individuals and associations with spending less on basic needs to pay for medication (St. Louis, Missouri; 2016–2018), n = 270.
| Variable | Odds Ratio | 95% Confidence Interval | P-value |
|---|---|---|---|
|
| |||
| Education level: no HS diploma | 1.00 | Referent | |
| High school graduate or GED | 1.79 | (0.77, 4.26) | 0.18 |
| Some college/technical school (with HS diploma) | 1.09 | (0.43, 2.78) | 0.86 |
| College graduate or greater | 0.99 | (0.34, 2.90) | 0.99 |
| Prescription drug coverage: Full year | 1.00 | Referent | |
| Less than full year | 1.36 | (0.70, 2.65) | 0.36 |
| Self-reported health status: Excellent/Good | 1.00 | Referent | |
| Fair/Poor | 2.00 | (1.03, 3.93) | 0.04 |
| Number of chronic conditions | 1.21 | (1.06, 1.40) | <0.01 |
| Monthly medication expenses (hundreds of dollars) | 1.41 | (1.11, 1.89) | <0.01 |
| Medication adherence | 0.38 | (0.18, 0.77) | <0.01 |
| Unmet basic needs over past 12 months | 1.08 | (0.87, 1.35) | 0.49 |
| Characterization of food in house: Enough quantity, good quality | 1.00 | Referent | |
| Enough quantity, lacking in quality | 2.00 | (0.91, 4.44) | 0.08 |
| Sometimes/often lacking in quantity | 1.68 | (0.70, 4.09) | 0.25 |
| Not enough money to make ends meet | 1.36 | (0.70, 2.67) | 0.36 |
| Difficulty paying bills | 1.43 | (1.11, 1.85) | <0.01 |
| Worry about making monthly living expenses (at least sometimes vs. rarely or never) | 1.09 | (0.39, 3.18) | 0.88 |
Discussion
We studied the demographic and health characteristics of individuals who spend less on basic needs to pay for medication, as well as the related financial and psychosocial challenges of individuals who make this tradeoff. Participants who reported poor overall health and those who had more chronic conditions and greater medication expenditure were more likely to spend less on basic needs to afford medications. Doing so was also associated with a lower likelihood of medication adherence and greater likelihood of financial problems. Further, the strong bivariate associations between spending less on basic needs and each of the psychosocial measures suggests that individuals who spend less on basic needs may experience mental and physical effects from financial worries.
We found that lower income or education levels were not significantly associated with reduced spending on basic needs to afford medications. The majority of participants in this study would be considered “low-income”, but non-income resources (e.g., food stamps or access to food pantries, housing support) can heavily impact individual financial situations (22). Thus, medication costs and ability to meet basic needs likely characterize their day-to-day financial status better than typical measures like income.
Previous research has found that a greater number of prescriptions and more expensive medications are associated with spending less on basic needs in order to afford those medications (13, 15); our study supports those findings. Perceived poor health or serious chronic co-morbidities may prompt individuals to prioritize medication over basic needs. However, it is also possible that other factors drive the decision to spend less on basic needs, and that poorer self-reported health status and greater number of chronic conditions are the results rather than the causes. Depending on the type and cost of medication, the health consequences of food insecurity may outweigh the benefits of increased medication spending. A short report published on this data highlighted this counterproductive tradeoff, demonstrating that individuals with diabetes were significantly more likely than others with chronic disease to spend less on basic needs, mostly food, to afford medication (20). Further, decreased food security may immediately impact health, whereas many long-term preventive medications do not reap benefits until years later. Physicians must openly communicate with patients about their financial situation and ideally, make referrals to medication and basic needs assistance programs. Such assistance programs often require lengthy waits (19); in the meantime, physicians can help patients prioritize spending in a manner most conducive to good health, as medications may not always be the most urgent need.
Zullig et al. noted that multiple cost-coping strategies, including reducing spending on basic needs, requesting less expensive medicine, and borrowing money to afford medications, were all associated with non-adherence (12). Our results on spending less on basic needs were in line with these findings. Individuals with fewer resources may be continually and purposefully alternating where to spend limited resources, unable to fully meet basic needs or achieve complete adherence. Patients may be making significant efforts to adhere to medical guidelines, even if they are unsuccessful (23). Adherence is not a binary affair (“adherent” vs. “non-adherent”), but rather a continuum that reflects not only the medication’s effectiveness, side effects, and patient’s willingness to take the medication, but also a patient’s resources and tradeoffs between competing financial priorities (23, 24). This challenge of “competing priorities” has been documented across diverse low-income populations and for various healthcare-related behaviors (18, 25, 26), and was highlighted in the qualitative interviews we carried out with survey participants (19). Participants explained how limited resources constantly force them to make undesirable tradeoffs, large and small. Often, participants sacrificed their time or exchanged an immediate benefit for a larger future reward. In other cases, participants were forced to forgo one immediate need to meet another more pressing need. Spending less on basic needs to afford medications—the outcome of our present analysis—is in fact one of a broad spectrum of tradeoffs that participants employed (19).
There is some evidence that housing and food instability are associated with a shift from utilization of primary care to more emergency department and hospital care (27). Individuals who cannot afford basic needs may spend less on chronic disease management—including medications and primary care visits—to try to allocate resources for basic needs. Some participants in our study recognized that this may cause health conditions to worsen over time, such that balancing health and basic needs becomes even more difficult (19). Conversely, our study included many people with unmet basic needs who made the opposite tradeoff, namely, spending less on basic needs to afford medications. This too may lead to negative outcomes.
Our findings reinforce the importance of ongoing efforts in primary care to address social determinants of health and unmet basic needs. Health Leads (28) and other social prescription programs (29) in healthcare systems are beginning to address unmet basic needs by connecting patients to community resources such as food pantries and utility assistance programs. Such promising solutions may free up patients’ resources and help patients achieve sustainable medication adherence. Currently, these programs are far from universal. Promisingly, most participants in our study reported being comfortable talking about the cost of medication with their doctor and other members of the healthcare team (20). While we recognize the time constraints of a busy primary care practice, our findings support the implementation of a seamlessly integrated screening and referral processes to address basic needs and medication cost in the primary care setting.
Our study is one of the first to examine the tradeoff between basic needs and medication in a very low-SES population with diverse chronic health conditions. Our broad range of health, socio-demographic, and financial measures enabled us to characterize those who reduce spending on basic needs to pay for medication. We collected adherence data separately for each medication, which we believe is more accurate than global adherence measures for individuals with large numbers of chronic conditions and associated prescriptions. However, our study has several limitations. Our study is based on a sample from a single metropolitan area, and included few Hispanics, Asians, or Native Americans. Thus, our results may not be generalizable to all demographic subgroups. Many participants did not know whether all their medications were generic or branded; hence, we were unable to accurately analyze potential savings from switching to generic medications. Spending less on basic needs to pay for medication was determined by a single yes/no question with no specified time frame within a rather long survey, which may affect response bias. Further, the cross-sectional design prevented determination of causality, and a one-time quantitative assessment cannot easily characterize the cyclical and inter-related nature of spending less in one area to afford another among resource-limited adults. Understanding an individual’s decisions over time may be difficult; however, at both the clinical and population levels, a multi-faceted approach is needed to increase access to both basic needs and medical care.
This research adds to extensive prior work on medication adherence and begins to characterize individuals who reduce spending on basic needs to afford medications. The findings are highly relevant for clinicians who treat low-SES patients with multiple chronic diseases and related prescription medications. Further research is needed to understand how patients prioritize competing financial pressures and how clinicians can support them to improve medication adherence, ensure adequate support for basic needs, and ultimately improve health outcomes.
Supplementary Material
Acknowledgments
We would like to acknowledge all study participants and our community partner, Affinia Healthcare. We would also like to thank Miquela Ibrao, MPH, MSW, Lindsey Manshack, MPH, Natasan McCray, MHA, Amanda Lee, MPH, Samantha Greaney, and Jessica Hao, MD for coordinating and conducting patient surveys.
Funding statement:
This research was supported by NIH grant #R01MD010445 (PI: James). KWR received support from the DeNardo Scholars Fellowship, through the Washington University School of Medicine Office of Medical Student Research and Summer Research Program. AM was also supported by research grants R01CA201429 and R01DK115916. CJH is currently supported by research grant K23HD096204 and was previously funded through KL2TR002346. Use of REDCap was supported by a Clinical and Translational Science Award (CTSA) Grant [UL1 TR000448] and Siteman Comprehensive Cancer Center and NCI Cancer Center Support Grant P30 CA091842. The content is solely the responsibility of the authors and does not necessarily represent the official view of the NIH. The funding sources had no role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the article for publication.
Footnotes
Conflicting and Competing Interests: CJH has received grant funding from the National Institutes of Health for investigator salary and conduct of research.
References
- 1.Brown MT, Bussell JK. Medication Adherence: WHO Cares? Mayo Clinic Proceedings. 2011;86(4):304–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kennedy J, Tuleu I, Mackay K. Unfilled prescriptions of medicare beneficiaries: prevalence, reasons, and types of medicines prescribed. J Manag Care Pharm. 2008;14(6):553–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Osterberg L, Blaschke T. Adherence to Medication. New England Journal of Medicine. 2005;353(5):487–97. [DOI] [PubMed] [Google Scholar]
- 4.Iuga AO, McGuire MJ. Adherence and health care costs. Risk management and healthcare policy. 2014;7:35–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gupta S, McColl MA, Guilcher SJ, Smith K. Cost-related nonadherence to prescription medications in Canada: a scoping review. Patient preference and adherence. 2018;12:1699–715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Surratt HL, O’Grady CL, Levi-Minzi MA, Kurtz SP. Medication adherence challenges among HIV positive substance abusers: the role of food and housing insecurity. AIDS Care. 2015;27(3):307–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cornelius T, Jones M, Merly C, Welles B, Kalichman MO, Kalichman SC. Impact of food, housing, and transportation insecurity on ART adherence: a hierarchical resources approach. AIDS Care. 2017;29(4):449–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Singer AW, Weiser SD, McCoy SI. Does Food Insecurity Undermine Adherence to Antiretroviral Therapy? A Systematic Review. AIDS and Behavior. 2015;19(8):1510–26. [DOI] [PubMed] [Google Scholar]
- 9.Berkowitz SA, Seligman HK, Choudhry NK. Treat or Eat: Food Insecurity, Cost-related Medication Underuse, and Unmet Needs. The American Journal of Medicine. 2014;127(4):303–10.e3. [DOI] [PubMed] [Google Scholar]
- 10.Pooler JA, Srinivasan M. Association Between Supplemental Nutrition Assistance Program Participation and Cost-Related Medication Nonadherence Among Older Adults With Diabetes. JAMA Internal Medicine. 2019;179(1):63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Patel MR, Piette JD, Resnicow K, Kowalski-Dobson T, Heisler M. Social Determinants of Health, Cost-related Nonadherence, and Cost-reducing Behaviors Among Adults With Diabetes: Findings From the National Health Interview Survey. Medical care. 2016;54(8):796–803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zullig LL, Peppercorn JM, Schrag D, Jr DHT, Lu Y, Samsa G, et al. Financial Distress, Use of Cost-Coping Strategies, and Adherence to Prescription Medication Among Patients With Cancer. Journal of Oncology Practice. 2013;9(6S):60s–3s. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Heisler M, Wagner TH, Piette JD. Patient Strategies to Cope with High Prescription Medication Costs: Who is Cutting Back on Necessities, Increasing Debt, or Underusing Medications? Journal of Behavioral Medicine. 2005;28(1):43–51. [DOI] [PubMed] [Google Scholar]
- 14.Madden JM, Graves AJ, Zhang F, Adams AS, Briesacher BA, Ross-Degnan D, et al. Cost-Related Medication Nonadherence and Spending on Basic Needs Following Implementation of Medicare Part D. JAMA. 2008;299(16):1922–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Piette JD, Heisler M, Wagner TH. Problems Paying Out-of-Pocket Medication Costs Among Older Adults With Diabetes. Diabetes Care. 2004;27(2):384–91. [DOI] [PubMed] [Google Scholar]
- 16.Cunningham P, Miller C, Cassil A. Living on the edge: health care expenses strain family budgets. Washington, DC: Center for Studying Health System Change; 2008. [PubMed] [Google Scholar]
- 17.Sharkey JR. Risk and Presence of Food Insufficiency Are Associated with Low Nutrient Intakes and Multimorbidity among Homebound Older Women Who Receive Home-Delivered Meals. The Journal of Nutrition. 2003;133(11):3485–91. [DOI] [PubMed] [Google Scholar]
- 18.Cunningham WE, Andersen RM, Katz MH, Stein MD, Turner BJ, Crystal S, et al. The Impact of Competing Subsistence Needs and Barriers on Access to Medical Care for Persons with Human Immunodeficiency Virus Receiving Care in the United States. Medical Care. 1999;37(12):1270–81. [DOI] [PubMed] [Google Scholar]
- 19.Lee AA, James AS, Hunleth JM. Waiting for care: Chronic illness and health system uncertainties in the United States. Social Science & Medicine. 2020;264:113296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Herrick CJ, Humble S, Hollar L, Chang S-H, Hunleth J, McQueen A, et al. Cost-Related Medication Non-adherence, Cost Coping Behaviors, and Cost Conversations Among Individuals with and Without Diabetes. Journal of General Internal Medicine. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tucker-Seeley RD, Harley AE, Stoddard AM, Sorensen GG. Financial Hardship and Self-Rated Health Among Low-Income Housing Residents. Health Education & Behavior. 2013;40(4):442–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.McKenzie HJ, McKay FH. Food as a discretionary item: the impact of welfare payment changes on low-income single mother’s food choices and strategies. Journal of Poverty and Social Justice. 2017;25(1):35–48. [Google Scholar]
- 23.Hunleth JM, Steinmetz EK, McQueen A, James AS. Beyond Adherence: Health Care Disparities and the Struggle to Get Screened for Colon Cancer. Qualitative health research. 2016;26(1):17–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Fix GM, Hyde JK, Bolton RE, Parker VA, Dvorin K, Wu J, et al. The moral discourse of HIV providers within their organizational context: An ethnographic case study. Patient Education and Counseling. 2018;101(12):2226–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gelberg L, Gallagher TC, Andersen RM, Koegel P. Competing priorities as a barrier to medical care among homeless adults in Los Angeles. American Journal of Public Health. 1997;87(2):217–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Dong KR, Must A, Tang AM, Beckwith CG, Stopka TJ. Competing priorities that rival health in adults on probation in Rhode Island: substance use recovery, employment, housing, and food intake. BMC Public Health. 2018;18(1):289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kushel MB, Gupta R, Gee L, Haas JS. Housing instability and food insecurity as barriers to health care among low-income americans. Journal of General Internal Medicine. 2006;21(1):71–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Berkowitz SA, Hulberg AC, Standish S, Reznor G, Atlas SJ. Addressing Unmet Basic Resource Needs as Part of Chronic Cardiometabolic Disease Management. JAMA Internal Medicine. 2017;177(2):244–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ridberg RA, Bell JF, Merritt KE, Harris DM, Young HM, Tancredi DJ. A Pediatric Fruit and Vegetable Prescription Program Increases Food Security in Low-Income Households. Journal of Nutrition Education and Behavior. 2019;51(2):224–30.e1. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.