Abstract
Background:
Older adults, an increasingly diverse segment of the United States population, are a priority population for prescription painkiller misuse. This study documents trends and correlates of prescription painkiller misuse among Hispanic and non-Hispanic adults ages 50 and older.
Methods:
A secondary analysis of adults 50 years and older across 5 cohorts using the 2015–2019 National Survey on Drug Use and Health (unweighted n=16,181, 8.5% Hispanic, and 54% female). Logistic regression modeling with complex survey design was used to examine trends in prescription painkiller misuse.
Results:
Over time, the prevalence of past year painkiller misuse significantly decreased for Hispanic respondents (56.1% relative decrease, p=0.02); elevated proportions were observed across strata of demographic characteristics.
Conclusions:
Variability in the prevalence of painkiller misuse may be explained by demographic characteristics. Further, these results emphasize the importance of addressing comorbid recreational marijuana use when designing interventions to address painkiller misuse for older adults.
Keywords: prescription drug misuse, NSDUH, alcohol, cannabis, tobacco
Background and Significance
The number of older adults in the United States (U.S.) is growing, with the population aged 65 and over projected to grow from 56.1 million (17% of the population) in 2020 to 94.7 million (23% of the population) by 2060 1. Ethnic minority groups, such as Hispanic-Americans, are expected to more than double over the next four decades 1. Identifying health trends across Hispanic and non-Hispanic aging adults is critical to addressing health disparities and promoting equity as the older adult population diversifies.
Focusing on ethnic trends is essential for several reasons. First, racial/ethnic minority older adults are less likely to utilize health care resources, more likely to report poor or fair health, and are more likely to experience financial hardship2–potential risk factors for prescription drug misuse. Hispanic populations, for instance, may encounter barriers such as language difficulties, immigration-related stress, and limited culturally competent healthcare resources. These factors can contribute to differences in substance use patterns and the effectiveness of prevention and intervention efforts. Second, existing research has identified significant gaps in our understanding of how painkiller misuse impacts different ethnic groups, particularly older adults.
Hispanic adults, in particular, represent a priority population for prescription painkiller misuse given historically elevated rates in the general population compared to their non-Hispanic White counterparts (14.1% vs 10%) and the presence of additional risk factors for misuse. 3 However, recent evidence suggest the gap in painkiller misuse rates may be narrowing4 between Hispanic and non-Hispanic adults. Research is needed to confirm this trend among older adults, and to identify correlates associated with shifting patterns of use. Moreover, previous research indicates that substance use behaviors and their correlates can differ significantly between ethnic groups. For example, studies have shown that non-Hispanic white individuals generally engage in prescription drug misuse more often than non-Hispanic black and Hispanic individuals. However, these patterns can vary based on age, socioeconomic status, and other demographic factors. Documenting trends across strata of these demographic factors is critical for developing effective public health policies and interventions that are tailored to the diverse experiences of older adults from different ethnic backgrounds.
Prescription drug misuse (taking of a medication in a manner or dose other than prescribed) is a continued public health concern in the U.S., particularly among older adults.5 While painkillers can provide effective pain relief when used appropriately—especially for those experiencing chronic pain related to arthritis, cancer, or other conditions—they also carry the risk of misuse and addiction.
Aging adults are a priority population for prescription painkiller misuse, specifically, due to several factors beyond age-related physiological changes, presence of multiple chronic health conditions, and a higher likelihood of being prescribed multiple medications. 6 For example, chronic pain is more prevalent among older adults, leading to higher rates of painkiller prescriptions. The persistent nature of chronic pain can result in prolonged use of painkillers, raising the likelihood of tolerance, dependence, and misuse. Additionally, cognitive decline, which is more common in older adults, can impair an individual’s ability to adhere to prescribed regimens, increasing the risk of misuse either accidentally or as an attempt to manage pain more effectively.
Social issues also play a critical role in the vulnerability of older adults to painkiller misuse. Older adults may experience social isolation, loneliness, and depression, which can contribute to the misuse of prescription medications as a coping mechanism. Furthermore, this demographic may have limited access to alternative pain management strategies, such as physical therapy or mental health services, making them more reliant on pharmacological interventions. Economic influences are also significant, as older adults, particularly those on fixed incomes, may struggle with the cost of medications. This can lead to behaviors such as medication sharing or using leftover prescriptions from past treatments, both of which can contribute to misuse. In addition, there is often a lack of targeted education and resources for older adults regarding the risks of painkiller misuse and the importance of adhering to prescribed regimens. Health care providers may also under-recognize or under-address substance misuse in this demographic due to stereotypes about drug misuse being less common among older adults, leading to insufficient screening and intervention.
The present analysis includes adults ages 50 years and older for several reasons. First, the data structure of the National Survey of Drug Use and Health categorizes respondent ages and the categories 50–64 years and 65 and older were used for analysis. Second, the Older American’s Act defines older adults as ages 60 and older—sooner than the traditional cutoff of 65 years old as an “older adult.” Second, those ages 50 and older should be considered as an “aging population” of potential concern as these individuals will eventually transition into retirement and classify as older adults. Finally, the rate of emergency department visits by older adults ages 55 and older has increased over time and is associated with opioid misuse 7.
The goal of the present study was to (1) document trends in prescription painkiller misuse over time among adults ages 50 and older, (2) explore potential ethnic differences in prescription painkiller misuse trends, and (3) examine correlates associated with past year prescription painkiller misuse. The results of this epidemiological analysis will identify specific risk factors and patterns that are crucial for designing prevention and intervention programs. Findings can also inform healthcare providers and policymakers about the necessity of screening for prescription misuse in routine clinical practice, particularly among older adults across sociodemographic strata. Finally, understanding the demographic and socio-economic characteristics associated with misuse can help in the creation of targeted educational campaigns and support services that address the specific challenges faced by older adults.
Methods
Data Source
We performed a secondary analysis of adults 50 years and older across 5 cohorts (2015–2019) of the National Survey on Drug Use and Health (NSDUH), a cross-sectional nationally representative survey of noninstitutionalized individuals in the United States. Alterations to the prescription drug misuse items prior to 2015 precluded comparisons to earlier years.8 These data are publicly accessible and available online via the Substance Abuse and Mental Health Data Archive (https://www.datafiles.samhsa.gov/). The analytic sample was restricted to observations indicating past year use of pain killers. This secondary analysis was exempt from review by the University of Nevada, Las Vegas institutional review board (IRB# UNLV-2022-385).
Measures
Outcomes of interest.
Prevalence of past-year pain killer misuse across cohorts was assessed using the NSDUH variable PRNMYR. Respondents were provided a definition of prescription drug misuse as “use in any way not directed by a doctor, including use without a prescription of one’s own medication; use in greater amounts, more often, or longer than told to take a drug; or use in any other way not directed by a doctor.” They were then asked about misuse of different prescription drugs, including prescription pain killers.
Correlates of interest.
First, we examined the prevalence of our outcomes by survey year modeled as a continuous variable (2015–2019). Second, we estimated the prevalence of outcomes stratified by ethnicity (Hispanic and non-Hispanic). Based on the results of the initial analysis, we subsequently assessed the prevalence of our outcomes stratified by each level of age (50–64 years, 65+ years), sex (female, male), educational attainment (high school or less, some college, college graduate), household income ($49,999 or less and $50,000 or more), insurance status (no, yes), chronic disease (0 or 1 and 2 or more), and past year recreational substance use (alcohol, marijuana, tobacco).
Analytic Plan
We calculated the relative change in prevalence of past year painkiller misuse between 2015 and 2019 in the entire population of adults ages 50 and older then by ethnicity (Hispanic vs. non-Hispanic). The decision to group non-Hispanic white and non-Hispanic black respondents together was based on several considerations. First, our primary objective was to explore broad trends in painkiller misuse between Hispanic and non-Hispanic populations. Second, the sample sizes for non-Hispanic white and non-Hispanic black respondents, when analyzed separately within specific socio-demographic strata (e.g., age, education level, income, insurance status), were insufficient to provide stable estimates for some of the subgroups. Using logistic regression, we estimated whether there was a log-linear association between pain killer misuse and time. Based on initial results, we further assessed associations between pain killer misuse and time across levels of covariates for Hispanic respondents only. Stratified logistic regression models were used to assess relationships between demographic and substance use variables with experiences of one or more past year painkiller consequences. We used sample weights (provided by National Survey on Drug Use and Health) to account for complex survey design, selection probability, nonresponse, and population distribution. Imputed variables were used, where possible, as provided in the NSDUH dataset.
Results
The analytic sample of adults ages 50 and older reporting past year painkiller use had an unweighted n = 16,181. The weighted data was 8.5% Hispanic, 54% female, and 6.4% reported past year painkiller misuse. Between 2015 and 2019, the prevalence of past year painkiller misuse did not change significantly in the overall population (4.84% relative increase, p=0.86). However, as seen in Figure 1 prevalence rates significantly decreased for Hispanic respondents (56.1% relative decrease, p=0.02) but not for their non-Hispanic counterparts (17.9% relative increase, p=0.52).
Panel Figure 1.
Trends in prevalence of pain killer misuse among those reporting past year use by ethnicity (Hispanic, solid line; non-Hispanic, dashed line), 2015–2019. Panel shows trends for (a) all adults ages 50 and older, (b) adults ages 50–64, and (c) adults ages 65 and older.
Among Hispanic respondents (Table 1), there were significant decreases among those ages 50–64 (61.7% relative decrease, p=0.04), those reporting some college or an associate’s degree (85.8% relative decrease, p=0.045), those earning $49,999 or less annually (56.3% relative decrease, p=0.03), and those with current insurance (62% relative decrease, p=0.02). No differences were observed for either sex. Decreased prevalence of past year painkiller misuse was associated with diagnoses of two or more chronic diseases (69.2% relative decrease, p=0.03), and reporting no past year use of alcohol (58.9% relative decrease, p=0.03), marijuana (60.5% relative decrease, p=0.01), or tobacco (54.7% relative decrease, p=0.02).
Table 1.
Trends in Weighted Prevalence of Past-Year Pain Killer Misuse by Sociodemographic, Chronic Disease, Health Care Use, and Substance Use Characteristics Among Hispanic Adults 50 Years and Older in the United States Reporting Past Year Painkiller Use (unweighted N=1,280), 2015–2019.
| Variable | 2015 | 2016 | 2017 | 2018 | 2019 | Relative % change | p | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
||||||||||||
| P (%) | 95% CI | P (%) | 95% CI | P (%) | 95% CI | P (%) | 95% CI | P (%) | 95% CI | |||
|
|
||||||||||||
| Age | ||||||||||||
| 50–64 | 12.0 | (7.5–18.8) | 9.8 | (5.3–17.4) | 7.5 | (4.0–13.7) | 6.0 | (2.2–15.3) | 4.6 | (1.7–12.3) | −61.7 | 0.044 |
| 65+ | 12.7 | (7.0–22.1) | 5.4 | (1.0–25.8) | 8.0 | (3.6–17.3) | 6.3 | (1.8–19.5) | 6.5 | (2.5–16.6) | −48.8 | 0.213 |
| Sex | ||||||||||||
| Female | 9.1 | (5.3–15.1) | 6.9 | (3.3–14.2) | 8.6 | (4.4–16.2) | 3.9 | (1.5–9.7) | 3.8 | (1.3–10.5) | −58.24 | 0.068 |
| Male | 15.9 | (9.2–26.1) | 10.1 | (4.3–21.8) | 6.8 | (2.9–14.8) | 8.4 | (2.5–24.6) | 7.4 | (3.2–16.2) | −53.46 | 0.158 |
| Education level | ||||||||||||
| HS or less | 12.1 | (7.8–16.3) | 10.0 | (4.4–21.3) | 5.6 | (2.6–11.8) | 4.8 | (1.1–19.1) | 7.1 | (3.5–16.6) | −41.32 | 0.144 |
| Some college | 12.0 | (6.0–22.5) | 7.7 | (3.0–18.6) | 11.6 | (4.8–25.6) | 9.9 | (3.5–25.2) | 1.7 | (0.4–7.6) | −85.83 | 0.045 |
| College graduate | 13.5 | (5.3–30.4) | 4.2 | (1.0–16.6) | 7.7 | (2.2–24.1) | 3.8 | (0.7–18.1) | 3.6 | (0.7–16.0) | −73.33 | 0.232 |
| Income | ||||||||||||
| <$49,999 | 17.4 | (12.1–24.4) | 11.7 | (6.2–21.2) | 6.8 | (3.6–12.3) | 7.0 | (2.3–19.6) | 7.6 | (3.7–15.2) | −56.32 | 0.019 |
| >$50,000 | 2.9 | (0.8–9.8) | 2.3 | (0.1–10.4) | 9.2 | (3.8–20.5) | 4.5 | (1.5–12.8) | 2.7 | (0.5–12.6) | −6.90 | 0.95 |
| Insured | ||||||||||||
| Yes | 12.1 | (8.4–17.2) | 7.7 | (4.1–13.9) | 8.2 | (4.9–13.4) | 5.7 | (1.9–15.8) | 4.6 | (2.1–9.6) | −62.0 | 0.018 |
| No | 13.9 | (4.6–35.1) | 13.6 | (4.2–36.0) | −- | −- | 10.0 | (1.9–29.2) | 14.7 | (4.4–39.0) | 5.8 | 0.881 |
| Chronic disease | ||||||||||||
| 0–1 | 13.0 | (8.9=18.4) | 8.9 | (4.7–16.1) | 10.9 | (6.6–17.5) | 7.7 | (3.0–18.7) | 5.9 | (2.7–12.6) | −54.6 | 0.058 |
| 2 or more | 10.4 | (4.5–22.2) | 6.8 | (2.4–17.5) | 1.0 | (0.2–3.8) | 0.4 | (0.0–2.7) | 3.2 | (0.8–12.1) | −69.2 | 0.035 |
| Past yr substance use | ||||||||||||
| No alcohol | 9.3 | (5.0–16.5) | 11.7 | (5.5–23.1) | 6.0 | (2.5–13.8) | 3.4 | (1.0–12.4) | 4.8 | (1.1–8.4) | −58.97 | 0.031 |
| Alcohol | 14.3 | (8.6–22.9) | 5.6 | (2.5–11.9) | 8.9 | (4.6–16.4) | 7.7 | (3.0–18.3) | 6.5 | (2.6–15.4) | −54.55 | 0.178 |
| No marijuana | 11.9 | (8.4–16.5) | 9.2 | (5.0–16.1) | 7.3 | (4.3–12.1) | 5.0 | (1.8–13.0) | 4.7 | (2.0–10.7) | −60.50 | 0.01 |
| Marijuana | 17.2 | (5.8–41.3) | 2.0 | (0.2–16.0) | 12.8 | (2.8–42.5) | 16.7 | (6.3–37.4) | 11.3 | (3.8–28.4) | −34.30 | 0.778 |
|
| ||||||||||||
| No tobacco | 11.7 | (8.2–16.5) | 7.6 | (3.4–16.1) | 6.9 | (3.8–12.4) | 4.6 | (1.9–10.8) | 5.4 | (2.6–10.7) | −54.70 | 0.015 |
| Tobacco | 14.5 | (6.1–30.8) | 11.5 | (4.9–24.4) | 10.1 | (3.4–26.5) | 11.9 | (2.6–40.3) | 5.2 | (1.5–17.1) | −64.14 | 0.39 |
Note: P=prevalence, 95% CI=95% confidence interval. Bold values represent statistically significant differences (p<0.05). Dashed lines represent suppressed estimates due to sparse data.
Discussion
The present study investigated trends in past year painkiller misuse and associated risk factors among a representative sample of U.S. adults ages 50 and older. In the overall sample, the prevalence of past year painkiller misuse did not significantly change between 2015 and 2019 (4.84% relative increase) consistent with research on prescription drug misuse in the older adult population.9 The adoption of prescribing limits in 2016–2017 may partially explain the decline in prescription painkiller use among Hispanic respondents to comparable levels with their non-Hispanic peers as evidenced in another population level study.10
However, the analysis uncovered significant differences in prevalence estimates for Hispanic versus non-Hispanic respondents. Specifically, while the prevalence of past year painkiller misuse marginally increased for non-Hispanic respondents, it significantly decreased by 56.1% (relative decrease) for Hispanic respondents, particularly among those ages 50–64, with some college or an associate’s degree, earning $49,999 or less annually, and reporting current insurance. Those reporting no insurance had moderately higher estimates of misuse over time. Our findings are consistent with previous studies that have highlighted the variability in prescription drug misuse trends among older adults across different ethnic groups and socio-demographic factors.9 Hispanic populations often exhibit lower overall rates of misuse compared to non-Hispanic populations,11 which may be influenced by cultural factors and community support, for example. The significant decrease in painkiller misuse among older Hispanic respondents observed here may reflect successful outreach and intervention programs tailored to this community, underscoring the role of culturally sensitive prevention strategies. Furthermore, the higher misuse estimates among uninsured respondents underscore the critical need for improving access to healthcare and substance misuse treatment services. These findings suggest that public health efforts should prioritize creating accessible, culturally competent programs that address the specific needs of older Hispanic populations to effectively mitigate the risk of painkiller misuse.
It is worth noting that the analysis demonstrated that decreased prevalence of past year painkiller misuse was associated with being diagnosed with two or more chronic diseases. Older adults diagnosed with multiple chronic diseases may be under more rigorous medical supervision, which can contribute to a reduction in the misuse of prescription medications. Chronic disease management often involves regular monitoring by healthcare providers, adherence to multiple medication regimens, and frequent medical visits. These factors can enhance medication adherence and reduce the likelihood of misuse. Additionally, patients with multiple chronic conditions might be more aware of the potential risks of medication misuse due to increased healthcare education and counseling. Additionally, individuals with multiple chronic conditions may be more likely to experience adverse effects from misuse due to their complex health profiles, leading to a greater awareness and caution in using painkillers as prescribed. This heightened vigilance can be a deterrent against misuse. These findings highlight the importance of integrated care approaches that address both pain management and chronic disease management simultaneously. By fostering a holistic approach to healthcare that includes patient education, regular monitoring, and comprehensive care plans, healthcare providers can help mitigate the risk of painkiller misuse among older adults with chronic conditions.
Furthermore, the absence of recreational substance use in the past year as a protective factor for painkiller misuse aligns with existing literature suggesting that polysubstance use is a significant risk factor for prescription drug misuse. Individuals who do not engage in recreational substance use may have a lower overall propensity for substance misuse, including the misuse of prescription painkillers. Reporting past year marijuana use, however, had a strong association with elevated past year painkiller misuse over time. This is especially relevant as more states legalize cannabis for recreational or medicinal use and recent data indicate that positive perceptions, and use, of cannabis is increasing among US older adults 12. Further, a recent study 13 of older adults indicated that cannabis use was mainly for medical purposes to treat pain, sleep disturbance, anxiety, or depression. Together, these results emphasize the importance of addressing comorbid recreational marijuana use when designing interventions to address painkiller misuse for older adults.
These results should be interpreted in light of the following limitations. First, estimates from the NSDUH may overestimate substance use behavior and underrepresent minority groups 14. Given the estimated rapid growth of other minority older adult populations (such as Asian Americans), purposive sampling methods are needed to strengthen the diversity of surveillance datasets such as the NSDUH. Relatedly, future research should further investigate the distinct patterns of prescription painkiller misuse within these ethnic subgroups to further tailor health education and intervention strategies. Second, due to the cross-sectional nature of this analysis we cannot draw cause and effect conclusions of our observed associations. Third, due to small cell sizes we had to collapse some indicators to retain statistical power, which reduced the specificity of our findings (for example, dichotomizing income categories for analysis). Finally, these data are drawn from self-report measures and are vulnerable to recall and social desirability bias.
Conclusions
Older adults represent a priority population for prescription drug misuse as they are particularly susceptible to adverse effects resulting from polypharmacy, cognitive decline, and experiences of chronic pain.15 There is a need for further investigation into reasons for recreational cannabis use and prescription painkiller misuse—such as chronic pain, access to care, or personal history. Such efforts may help to improve the overall health and wellbeing of the growing diverse U.S. older adult population.
References
- 1.Vespa J, Medina L, Armstrong DM. Population Estimates and Projections. US Department of Commerce, Economics and Statistics Administration, US Census Bureau; 2018:1–15. https://www.census.gov/content/dam/Census/library/publications/2020/demo/p25-1144.pdf [Google Scholar]
- 2.Kirzinger A, Neuman T, Cubanski J, Brodie M. Data Note: Prescription Drugs and Older Adults. KFF. Published 2019. Accessed May 22, 2023. https://www.kff.org/health-reform/issue-brief/data-note-prescription-drugs-and-older-adults/
- 3.Cano M Prescription opioid misuse among U.S. Hispanics. Addict Behav. 2019;98:106021. doi: 10.1016/j.addbeh.2019.06.010 [DOI] [PubMed] [Google Scholar]
- 4.Center for Behavioral Health Statistics and Quality. Racial/Ethnic Differences in Substance Use, Substance Use Disorders, and Substance Use Treatment Utilization among People Aged 12 or Older (2015–2019).; 2021. [Google Scholar]
- 5.Schepis TS, Klare DL, Ford JA, McCabe SE. Prescription Drug Misuse: Taking a Lifespan Perspective. Subst Abuse Res Treat. 2020;14:117822182090935. doi: 10.1177/1178221820909352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ford JA, Schepis TS, McCabe SE. Poly-prescription drug misuse across the life course: Prevalence and correlates across different adult age cohorts in the U.S. Int J Drug Policy. 2021;88:103017. doi: 10.1016/j.drugpo.2020.103017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Carter MW, Yang BK, Davenport M, Kabel A. Increasing Rates of Opioid Misuse Among Older Adults Visiting Emergency Departments. Innov Aging. 2019;3(1):igz002. doi: 10.1093/geroni/igz002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Center for Behavioral Health Statistics and Quality. 2002–2019 National Survey on Drug Use and Health Public Use File Codebook. Substance Abuse and Mental Health Services Administration; 2020. [PubMed] [Google Scholar]
- 9.Schepis TS, McCabe SE, Ford JA. Recent trends in prescription drug misuse in the United States by age, race/ethnicity, and sex. Am J Addict. 2022;31(5):396–402. doi: 10.1111/ajad.13289 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cho G, Chang VW. Trends in Prescription Opioid and Nonopioid Analgesic Use by Race, 1996–2017. Am J Prev Med. 2022;62(3):422–426. doi: 10.1016/j.amepre.2021.08.016 [DOI] [PubMed] [Google Scholar]
- 11.Schuler MS, Schell TL, Wong EC. Racial/ethnic differences in prescription opioid misuse and heroin use among a national sample, 1999–2018. Drug Alcohol Depend. 2021;221:108588. doi: 10.1016/j.drugalcdep.2021.108588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Han BH, Funk-White M, Ko R, Al-Rousan T, Palamar JJ. Decreasing perceived risk associated with regular cannabis use among older adults in the United States from 2015 to 2019. J Am Geriatr Soc. 2021;69(9):2591–2597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yang KH, Kaufmann CN, Nafsu R, et al. Cannabis: An Emerging Treatment for Common Symptoms in Older Adults. J Am Geriatr Soc. 2021;69(1):91–97. doi: 10.1111/jgs.16833 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Grigsby TJ, Howard K, Howard JT. Comparison of Past Year Substance Use Estimates by Age, Sex, and Race/Ethnicity Between Two Representative Samples of the U.S. Adult Population. Popul Res Policy Rev. 2022;41:401–416. doi: 10.1007/s11113-021-09645-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gy Oh, Abner EL Fardo DW, Freeman PR Moga DC. Patterns and predictors of chronic opioid use in older adults: A retrospective cohort study. PLOS ONE. 2019;14(1):e0210341. doi: 10.1371/journal.pone.0210341 [DOI] [PMC free article] [PubMed] [Google Scholar]