Abstract
Objective:
The U.S. population is aging and increasing numbers of older adults are using cannabis. Cognitive decline is common in older age and subjective memory complaints (SMC) have been associated with increased risk for dementia. While residual cognitive effects of cannabis at younger ages are well understood, the links between cannabis use and cognition in older adults is less clear. The present study represents the first population level analysis of cannabis use and SMC in older adults in the U.S.
Method:
We used the National Survey of Drug Use and Health (NSDUH) dataset to evaluate SMC in respondents over age 50 (N=26,399) according to past-year cannabis use.
Results:
Results revealed that 13.2% (95%CI: 11.5%−15.0%) of those who reported cannabis use also reported SMC, compared to 6.4% (95%CI: 6.1%−6.8%) among individuals with no cannabis use. Logistic regression revealed a two-fold increase (OR=2.21, 95%CI: 1.88–2.60) of reporting SMC in respondents who had used cannabis in the past year, which was attenuated (OR=1.38, 95%CI: 1.10–1.72) when controlling for additional factors. Other covariates, including physical health conditions, misuse of other substances, and mental illness also significantly contributed to SMC outcomes.
Conclusions:
Cannabis use represents a modifiable lifestyle factor that has potential for both risk and protective properties that may impact the trajectory of cognitive decline in older age. These hypothesis generating results are important for characterizing and contextualizing population-level trends related to cannabis use and SMC in older adults.
Keywords: cognition, marijuana, representative survey, aging, dementia, substance use
Introduction
The US population is aging, with over 35.4% of adults now over 50 compared to 27.3% just two decades ago (US Census Bureau, 2001, 2019). Cannabis use among older adults in the US is also increasing. Between 2016 and 2018, cannabis use in adults aged 55 and older increased by 40%, with twice as many men reporting use compared to women; for adults aged 65–69, past year use increased from 4.3% to 8.2% in men and from 2.1% to 3.8% in women (Maxwell et al., 2021). This increasing use of cannabis is likely due to multiple factors, including expanded legalization of cannabis for medical use, legalization of cannabis for recreational use, and/or decriminalization of cannabis in multiple states (Kaskie et al., 2017). Among older adults currently using cannabis, about half initiated use after the age of 30, most reported infrequent use (less than once every 10 days) over the past year (Blazer & Wu, 2009), and 90% denied current emotional or functional problems (Black & Joseph, 2014), despite roughly 20% reporting a history of treatment-seeking for a non-cannabis substance use disorder (SUD) (Kaskie et al., 2017; Wu & Blazer, 2011).
Cognitive decline is well-established with normal aging, and subjective memory complaints (SMC) among older adults are associated with a two-fold increase in dementia risk (Mitchell et al., 2014). However, the potential mechanisms of this association are not well understood and may reflect an array of factors, including increased health anxiety, mood symptoms, white matter lesions, temporal atrophy, cerebral hypometabolism, and neurodegenerative biomarkers (for a review of findings, see Mitchell et al., 2014). Although some older adults maintain strong memory abilities into very old age, almost half of those over age 85 will experience dementia, and 1 in 3 older adults will die with dementia. Age, family history, and genetic factors (e.g., Apolipoprotein E ε4 allele) represent the greatest predictors of dementia, each of which is nonmodifiable (Baumgart et al., 2015). However, several health and lifestyle factors (e.g., cardiovascular health, obesity, smoking status, physical activity, diet, alcohol use, social and cognitive engagement, education, sleep, and mental health) have also been linked to cognitive decline and overall risk for dementia and are potentially modifiable (Baumgart et al., 2015). Cannabis use is one such modifiable lifestyle factor; however, the links between cannabis use and cognitive functioning in older adulthood are not well understood.
Cannabis use has well characterized acute effects on cognition, including impairments to attention, executive functioning, learning, and memory (Campeny et al., 2020; Crane et al., 2013; Crean et al., 2011; Ranganathan & D’Souza, 2006; Volkow et al., 2016). However, many of these effects resolve with abstinence and the residual effects of cannabis use on cognition remain an area of ongoing debate (Crean et al., 2011; Meier et al., 2018; Ross et al., 2020). The most heavily researched cannabinoids contained in cannabis (Δ9-tetrahydrocannabinol [THC] and cannabidiol [CBD]) function as exogenous ligands for CB1 and CB2 receptors in the central and peripheral nervous systems and have been associated with differential and interacting neuropsychological effects (Abate et al., 2021; Aso & Ferrer, 2014; Chayasirisobhon, 2019; Englund et al., 2013; Scott et al., 2019; Weinstein & Sznitman, 2020). THC has notable psychoactive properties, including deficits in cognition, increased anxiety, and inducing psychotic experiences in vulnerable individuals (Cohen et al., 2019; Englund et al., 2013; Scott et al., 2019), while also possessing neuroprotective properties such as enhancing cholinergic transmission and inhibiting amyloid-beta aggregation (Abate et al., 2021; Aso & Ferrer, 2014; Weinstein & Sznitman, 2020). In contrast, CBD has been associated with enhanced learning, reduced anxiety, and inhibition of psychotic processes, while also providing anti-inflammatory and anti-oxidative properties (Chayasirisobhon, 2019; Lucas et al., 2018; Vacaflor et al., 2020). Most of the existing research on cannabis use and cognition has focused on adolescents and young adult participants (Abate et al., 2021; Lisdahl et al., 2021) and considerably less is known about cannabis use in later life stages. Furthermore, existing studies with adults are often limited by a lack of representation for older adults and comparisons between those with heavy chronic use and those without use, thus failing to account for differential effects associated with milder cannabis use in older adults (Scott et al., 2019; Volkow et al., 2016; Weinstein & Sznitman, 2020).
Despite these limitations, three reviews have summarized findings on cannabis use and cognition in samples that include older adults. One recent review of cannabis use in older adults (Vacaflor et al., 2020) found that 17.5% of participants across seven studies (three prospective observational studies, one retrospective survey, and three double-blind randomized controlled trials) reported SMC, although study authors concluded that low-dose, short-term medical cannabis use was generally well tolerated in older adults and did not confer significant risk for adverse cognitive outcomes. These findings were similar to a review of recreational and medical cannabis use in middle to older adulthood (Scott et al., 2019) that found modest reductions in cognitive performance associated with higher doses and heavier lifetime use of cannabis, although negative cognitive effects were less evident in older adults using medical cannabis. Another review (Weinstein & Sznitman, 2020) acknowledged potential risks for cognitive decline in older adults who use cannabis, but the authors concluded that the animal literature and a small number of experimental studies in older adults using medical cannabis indicate that cannabis use may be associated with better cognition. Research continues to investigate the potential therapeutic application of cannabis for neurological disorders in older adults, including Alzheimer’s disease, although the beneficial and adverse effects in human trials is not clear (Abate et al., 2021; Aso & Ferrer, 2014; Bosnjak Kuharic et al., 2021; Cohen et al., 2019). To better understand the associations between cannabis use and cognitive deficits in older adults, representative population-based analyses of SMC and cannabis use patterns are needed. To our knowledge, no population-level analyses of cannabis use and SMC in older adults have been conducted.
To address this gap in the literature, we used the National Survey of Drug Use and Health (NSDUH) dataset to evaluate rates of SMC in older adult respondents (age 50+) according to past year use of cannabis. The NSDUH, a nationally representative cross-sectional survey in the U.S. that assesses drug use and related health concerns, uses a multistage area probability sampling design covering all 50 states, surveying non-institutionalized individuals ages 12 years and older.
Method
Participants and procedures
The present study used data from the 2017–2019 NSDUH surveys and included adults ages 50 years and older (n = 26,399). Interviewers administered study questions using computer-assisted personal interviewing and audio computer-assisted self-interviewing, which provides respondents with privacy to answer potentially sensitive questions such as those related to substance use. Respondents were compensated with $30. Further information regarding survey methods have been reported elsewhere (Substance Abuse and Mental Health Services Administration, 2014). The RTI Institutional Review Board (IRB) approved the data collection procedures in compliance with the Helsinki Declaration, and this secondary data analysis was considered exempt by the University of Michigan IRB.
Measures
Respondents reported their use of cannabis, including (a) any use in the past year, (b) any use in the past month, (c) number of days using cannabis in the past year, and (d) number of days using cannabis in the past month. Respondents reported their use of alcohol, tobacco, cocaine, heroin, hallucinogens, inhalants, methamphetamine, pain relievers, tranquilizers, stimulants, and sedatives; substances used as prescribed by a medical provider were not included in these counts. Alcohol and tobacco were evaluated individually as binary variables. Use of the remaining substances was combined as a new binary variable (Other Substance Use) given the very low base rate (0.9%) of respondents endorsing use of more than one of these substances. Respondents were asked about select health conditions experienced in the past year (i.e., asthma, cancer, heart condition, and hypertension) and these were included as a binary variable (Health Conditions). To control for mental health factors that may impact cognition, we used the NSDUH Mental Health Module, which is a predictive classification variable based on responses to a 12-item questionnaire of past-year and past-month psychiatric symptoms derived from the NSDUH 2012 dataset. Respondents were classified in the NSDUH dataset as having no mental illness, mild mental illness, and moderate to severe mental illness. Regarding SMC, respondents were asked, “Because of a physical, mental or emotional condition, do you have serious difficulty concentrating, remembering, or making decisions?” and responses were recorded a binary variable. Demographic characteristics included age grouping (50–64, 65+), sex (male, female), race/ethnicity (non-Hispanic white, all others), domestic status (married, all others), education (greater than 12 years of education, 12 years or less), and population density (greater than one million persons, less than one million persons, others). Categorization of variables reflected NSDUH-suggested categories to allow for comparison with other publications using NSDUH data. See Supplementary Table 1 for a list of variable names. Further information about NSDUH measures and coding are available online (Center for Behavioral Health Statistics and Quality, 2021).
Data analyses
NSDUH-derived variables were utilized for stratification, clustering, and weighting to ensure that findings are representative of the general non-institutionalized U.S. population. Reported percentages represent weighted statistics to reflect U.S. population estimates. The sample was divided according to use of cannabis over the past year. Bivariate comparisons of demographic and clinical characteristics were assessed utilizing likelihood ratio chi-square tests for categorical variables and t-tests for continuous variables. Multivariable logistic regression was used to examine the association of demographic (age, sex, race/ethnicity, domestic status, education, and population density) and clinical characteristics (alcohol, tobacco, other substance use, physical health conditions, and mental illness) in people who used cannabis compared to those who did not. All data analysis was performed using SAS v 13.1 (SAS Institute, n.d.).
Results
Between 2017 and 2019, 26,399 respondents aged 50 and older completed the survey, 45.2% (95%CI: 44.4% - 45.9%) of whom were over age 65. Approximately 8.2% (95%CI: 7.9% - 8.6%) of the total sample over age 50 reported using cannabis in the past year. Differences by sex revealed that 60.5% (95%CI: 57.9% - 63.1%) of cannabis users were male and that, overall, 10.7% of male and 6.1% of female respondents used cannabis in the past year. See Table 1 for additional data on respondent demographics and clinical characteristics. The average person who used cannabis consumed cannabis 2.26 (95%CI: 2.11 – 2.40) days per week over the past year. Overall, 7.0% (95%CI: 6.7% - 7.3%) of survey respondents reported SMC. Among respondents, 13.2% (95%CI: 11.5%−15.0%) of those who reported cannabis use in the past year also reported SMC, compared to 6.4% (95%CI: 6.1%−6.8%) among those individuals with no cannabis use. Directionally consistent effects were obtained when evaluating associations with past-month cannabis use (see Table 2).
Table 1.
Respondent demographics and clinical characteristics in adults aged 50+.
| Overall | Used cannabis in the past year | |||
|---|---|---|---|---|
| Yes | No | |||
| Raw N, weighted % | 26399 | 2286 (8.2%) | 24113 (91.8%) | |
| Age | Raw N | Weighted % (95CI) | Raw N / Weighted % (95CI) | Raw N / Weighted % (95CI) |
| 50–64 | 14815 | 54.8% (54.1% - 55.6%) | 1764 76.4% (74.1% - 78.7%) A |
13051 52.9% (52.1% - 53.7%) |
| 65+ | 11584 | 45.2% (44.4% - 45.9%) | 522 23.6% (21.3% - 25.9%) |
11062 47.1% (46.3% - 47.9%) |
| Sex | ||||
| Male | 12097 | 46.7% (46.1% - 47.3%) | 1361 60.5% (57.9% - 63.1%) A |
10736 45.5% (44.8% - 46.1%) |
| Female | 14302 | 53.3% (52.6% - 53.9%) | 925 39.5% (36.9% - 42.1%) |
13377 54.5% (53.8% - 55.2%) |
| Race/Ethnicity | ||||
| Non-Hispanic White | 19167 | 71.8% (70.9% - 72.8%) | 1751 77.6% (75.4% - 79.8%) A |
17416 71.3% (70.4% - 72.3%) |
| All others | 7232 | 28.1% (27.2% - 29.1%) | 535 22.4% (20.2% - 24.6%) |
6697 28.7% (27.7% - 29.6%) |
| Domestic status | ||||
| Married | 15649 | 61.2% (60.1% - 62.2%) | 1064 48.9% (46.1% - 51.6%) A |
14585 62.3% (61.2% - 63.4%) |
| All others | 10750 | 38.8% (37.8% - 39.9%) | 1222 51.1% (48.4% - 53.9%) |
9528 37.7% (36.6% - 38.8%) |
| Education | ||||
| Any post-H.S./GED | 15734 | 61.1% (60.3% - 62.0%) | 1439 65.0% (62.4% - 67.5%) A |
14295 60.8% (59.5% - 61.7%) |
| H.S./GED or less | 10665 | 38.9% (38.0% - 39.7%) | 847 35.0% (32.5% - 37.6%) |
9818 39.2% (38.3% - 40.1%) |
| Population Density | ||||
| ≥ 1 million persons | 10518 | 51.1% (50.1% - 52.0%) | 959 53.1% (49.8% - 56.3%) NS |
9559 50.9% (49.9% - 51.9%) |
| < 1million persons | 13159 | 41.9% (40.9% - 42.9%) | 113 41.1% (38.3% - 43.9%) |
12026 42.0% (40.8% - 43.1%) |
| segment not a CBSA | 2722 | 7.0% (6.5% - 7.6%) | 114 5.8% (4.5% - 7.1%) |
2528 7.2% (6.6% - 7.7%) |
| Alcohol use (past year) | ||||
| Yes | 16449 | 62.9% (62.2% - 63.4%) | 1943 86.1% (84.4% - 87.9%) A |
14506 60.8% (60.0% - 61.5%) |
| No | 9950 | 37.1% (36.4% - 37.8%) | 343 13.9% (12.1% - 15.6%) |
9607 39.2% (38.5% - 40.0%) |
| Tobacco use (past year) | ||||
| Yes | 5776 | 20.7% (20.0% - 21.4%) | 1161 50.6% (46.7% - 53.3%) |
4615 18.1% (17.5% - 18.8%) |
| No | 20623 | 79.3% (78.6% - 80.0%) | 1125 49.4% (47.9% - 52.1%) A |
19498 81.8% (81.2% - 82.5%) |
| Other substance use (past year) | ||||
| Yes | 1132 | 4.3% (3.9% - 4.7%) | 442 19.3% (17.1% - 21.5%) |
690 2.9% (2.6% - 3.2%) |
| No | 25267 | 95.7% (95.3% - 96.1%) | 1844 80.7% (78.4% - 82.9%) A |
23423 97.1% (96.8% - 97.4%) |
| Physical health conditions (past year) | ||||
| Yes | 10307 | 39.2% (38.6% - 39.8%) | 753 34.2% (31.7% - 36.6%) A |
9554 39.7% (39.0% - 40.3%) |
| No | 16092 | 60.8% (60.2% - 61.4%) | 1533 65.8% (63.4% - 68.2%) |
14559 60.3% (59.7% - 61.0%) |
| Mental Illness (past year) | ||||
| None | 22534 | 85.9% (85.4% - 86.4%) | 1661 71.9% (69.2% - 74.6%) |
20873 87.2% (86.7% - 87.6%) |
| Mild | 2086 | 7.7% (7.3% - 8.1%) | 271 12.6% (10.4% - 14.8%) A |
1815 7.3% (6.9% - 7.7%) |
| Mod/Sev | 1779 | 6.3% (5.9% - 6.7%) | 354 15.5% (13.3% - 17.7%) A |
1425 5.5% (5.2% - 5.9%) |
| Subjective Memory Complaint | ||||
| Yes | 1933 | 7.0% (6.7% - 7.3%) | 308 13.2% (11.5% - 15.0%) A |
1625 6.4% (6.1% - 6.8%) |
| No | 24344 | 93.0% (92.6% - 93.3%) | 1966 86.8% (85.0% - 88.5%) |
22378 93.6% (93.2% - 93.9%) |
| Cannabis use (past month) | ||||
| Yes | 1528 | 5.3% (5.0% - 5.6%) | 1528 64.8% (62.4% - 67.2%)* |
NA |
| No | 24871 | 94.7% (94.3% - 95.0%) | 758 35.2% (32.8% - 37.5%) |
NA |
Note:
p≤0.005;
not significant at 0.05;
not tested; Physical health conditions include asthma, cancer, heart condition, and hypertension; Mental illness in the past year according to NSDUH designations.
Table 2.
Cannabis use in the past year and past month according to report of subjective memory complaints (SMC) in adults aged 50+.
| SMC | ||
|---|---|---|
| Yes | No | |
| Cannabis use in past year | ||
| Yes | 308 15.6% (13.7% – 17.5%) |
1966 7.8% (7.3% – 8.1%) |
| No | 1625 84.4% (82.5% – 86.3%) |
22378 92.3% (91.9% – 92.7%) |
| Cannabis in past month | ||
| Yes | 211 9.8% (8.4% – 11.4%) |
1308 5.0% (4.6% – 5.4%) |
| No | 1722 90.1% (88.6% – 91.6%) |
23036 95.0% (94.6% – 95.3%) |
Note. SMC=subjective memory complaints.
The association between cannabis use and SMC was further evaluated using a logistic regression, which revealed an unadjusted model odds ratio (OR) of 2.21 (95%CI: 1.88–2.60), indicating greater odds of SMC in those with past-year cannabis use compared to those with no cannabis use. An adjusted model controlling for age, sex, race, domestic status, education, population density, alcohol use, tobacco use, other substance use, health conditions, and mental illness, yielded an adjusted OR of 1.38 (95%CI: 1.10–1.72), indicating greater likelihood of SMC associated with past-year cannabis use (see Table 3). Other measured variables significantly contributed to the likelihood of SMC, including reduced risk of SMC associated with higher levels of education (OR=0.58, 95%CI: 0.51–0.67), being married (OR=0.76, 95%CI: 0.68–0.86), and use of alcohol in the past year (OR=0.65, 95%CI: 0.56–0.75). Risk of SMC increased with having a physical health condition (OR=1.29, 95%CI: 1.12–1.48), use of other substances in the past year (OR=1.52, 95%CI: 1.21–1.92), or mental illness (OR=6.05, 95%CI: 5.03–7.28 for mild severity; OR=20.6, 95%CI: 17.8–23.9 for moderate or severe severity).
Table 3.
Logistic regression of past-year cannabis use on subjective memory complaints (SMC) in adults aged 50+.
| Unadjusted Logistic Model | β (se) | Odds ratio (95% CI) |
|---|---|---|
| Cannabis use in past year | 0.40 (0.04) | 2.21 (1.88 – 2.60) |
| No use cannabis in past year | (referent) | |
| Adjusted Model* | β (se) | Odds ratio (95% CI) |
| Age group NS | ||
| 50–64 | (referent) | -- |
| 65+ | 0.05 (0.03) | 1.10 (0.98 – 1.25) |
| Sex NS | ||
| Male | −0.05 (0.03) | 0.90 (0.78 – 1.03) |
| Female | (referent) | -- |
| Race group NS | ||
| Non-Hispanic White | 0.02 (0.04) | 1.03 (0.89 – 1.20) |
| All others | (referent) | -- |
| Domestic status A | ||
| Married | −0.13 (0.03) | 0.76 (0.68 – 0.86) |
| All others | (referent) | -- |
| Education A | ||
| > 12 years | −0.27 (0.03) | 0.58 (0.51 – 0.67) |
| ≤ 12 years | (referent) | -- |
| Population Density NS | ||
| ≤ 1 million persons | (referent) | -- |
| < 1million persons | −0.04 (0.05) | 1.02 (0.88 – 1.18) |
| segment not a CBSA | 0.09 (0.07) | 1.16 (0.93 – 1.44) |
| Alcohol use (past year) A | ||
| Yes | −0.22 (0.04) | 0.65 (0.56 – 0.75) |
| No | (referent) | -- |
| Tobacco use (past year) NS | ||
| Yes | 0.06 (0.03) | 1.13 (0.98 – 1.30) |
| No | (referent) | -- |
| Other substance use (past year) A | ||
| Yes | 0.21 (0.06) | 1.52 (1.21 – 1.92) |
| No | (referent) | -- |
| Physical health conditions (past year) A | ||
| Yes | 0.13 (0.03) | 1.29 (1.12 – 1.48) |
| No | (referent) | -- |
| Mental Illness (past year) A | ||
| None | (referent) | -- |
| Mild | 0.19 (0.06) | 6.05 (5.03 – 7.28) |
| Moderate/Severe | 1.42 (0.05) | 20.6 (17.8 – 23.9) |
| Cannabis use (past year) B | ||
| Yes | 0.16 (0.06) | 1.38 (1.10 – 1.72) |
| No | (referent) | -- |
Note.
p≤0.005;
p≤0.01;
not significant at 0.05; SMC=subjective memory complaints; CBSA=core based statistical areas; Physical health conditions include asthma, cancer, heart condition, and hypertension; Mental illness in the past year according to NSDUH designations.
When a parallel set of analyses were conducted using past-month cannabis use, the unadjusted model revealed significant OR of 2.08 (95%CI: 1.69–2.56) and the adjusted model revealed directionally consistent but nonsignificant effects (OR=1.15, 95%CI: 0.90–1.49). ORs across covariates were otherwise highly similar to those previously presented, although tobacco use became a significant finding (OR=1.16, 95%CI: 1.01–1.34). See Table 4 for details. We also conducted a dose-response analysis among respondents reporting past-year cannabis use (n=2,286) according to frequency of past-year cannabis use predicting SMC. These results were nonsignificant, with OR of 1.001 (95%CI: 0.999–1.002).
Table 4.
Logistic regression of past-month cannabis use on subjective memory complaints (SMC) in adults aged 50+.
| Unadjusted Logistic Model | β (se) | Odds ratio (95% CI) |
|---|---|---|
| Cannabis use in past month | 0.37 (0.05) | 2.08 (1.69 – 2.56) |
| No use cannabis in past month | (referent) | |
| Adjusted Model | β (se) | Odds ratio (95% CI) |
| Age group NS | ||
| 50–64 | (referent) | -- |
| 65+ | 0.04 (0.03) | 1.09 (0.97 – 1.23) |
| Sex NS | ||
| Male | −0.05 (0.03) | 0.91 (0.79 – 1.05) |
| Female | (referent) | -- |
| Race group NS | ||
| Non-Hispanic White | 0.02 (0.04) | 1.04 (0.90 – 1.20) |
| All others | (referent) | -- |
| Domestic status A | ||
| Married | −0.14 (0.03) | 0.76 (0.67 – 0.85) |
| All others | (referent) | -- |
| Education A | ||
| > 12 years | −0.27 (0.03) | 0.58 (0.51 – 0.67) |
| ≤ 12 years | (referent) | -- |
| Population Density NS | ||
| ≥ 1 million persons | (referent) | -- |
| < 1million persons | −0.04 (0.05) | 1.01 (0.87 – 1.18) |
| segment not a CBSA | 0.09 (0.07) | 1.15 (0.93 – 1.43) |
| Alcohol use (past year) A | ||
| Yes | −0.20 (0.04) | 0.66 (0.57 – 0.77) |
| No | (referent) | -- |
| Tobacco use (past year) C | ||
| Yes | 0.08 (0.04) | 1.16 (1.01 – 1.34) |
| No | (referent) | -- |
| Other substance use (past year) A | ||
| Yes | 0.23 (0.06) | 1.59 (1.27 – 1.99) |
| No | (referent) | -- |
| Physical health conditions (past year) A | ||
| Yes | 0.12 (0.03) | 1.28 (1.11 – 1.47 |
| No | (referent) | -- |
| Mental Illness (past year) A | ||
| None | (referent) | -- |
| Mild | 0.19 (0.06) | 6.11 (5.07 – 7.36) |
| Moderate/Severe | 1.42 (0.05) | 20.91 (17.99 – 24.29) |
| Cannabis use (past month) NS | ||
| Yes | 0.07 (0.06) | 1.15 (0.90 – 1.49) |
| No | (referent) | -- |
Note.
p≤0.005;
p≤0.01;
p≤0.05;
not significant at 0.05; SMC=subjective memory complaints; CBSA=core based statistical areas; Physical health conditions include asthma, cancer, heart condition, and hypertension; Mental illness in the past year according to NSDUH designations.
Discussion
Among a representative sample of U.S. adults over age 50, our results reveal that approximately 8.2% have used cannabis at least once in the past year, and that cannabis use is associated with increased risk of SMC. This base-rate of cannabis use reflects a continued increase among older adult respondents from the NSDUH studies since 2006 (Han et al., 2017) and is comparable to other recent surveys (Maxwell et al., 2021) that reported 8.3% of men and 3.9% of women used cannabis in the past month. Thirteen percent of those who used cannabis in our sample reported SMC, slightly lower than the 17.5% reported in the review by Vacaflor, et al. (2020). We found a two-fold increase (OR=2.21) in likelihood of reporting SMC in respondents who had used cannabis in the past year, which was attenuated (OR=1.38) when controlling for demographic, health conditions, and psychiatric factors. However, this association between cannabis use and SMC became nonsignificant when evaluating respondents who had used cannabis in the past month, when acute and post-acute cognitive effects would be more likely to emerge, and a dose-response relationship between cannabis use and SMC was also nonsignificant. To the best of our knowledge, this is the first population level analysis of associations between cannabis use and SMC in older adults.
While cannabis use and acute cognitive deficits in samples of adolescents and young adults is well established, residual effects are mixed and few studies have examined these associations in older adults. Studies including middle and older adult participants have lacked well controlled, longitudinal, performance-based assessment of cognition and have been limited by small sample sizes and wide age bands that are not specific to older adults, thus diluting potential differential effects of cannabis use in older adulthood specifically (Scott et al., 2019; Vacaflor et al., 2020; Volkow et al., 2016; Weinstein & Sznitman, 2020). Despite these limitations, results of available studies tend to show improved executive functioning with medical cannabis (Bar-Sela et al., 2019; Gruber et al., 2016; Sagar et al., 2021; Scott et al., 2019) and more mixed findings for those who use cannabis recreationally (i.e., lower verbal memory and processing speed in individuals who use recreational cannabis compared to those who do not, but between-group differences in change over time or dose-dependent associations were not evident) (Auer et al., 2016; Dregan & Gulliford, 2012; McKetin et al., 2016; Weinstein & Sznitman, 2020).
The present analyses reveal a relationship between past-year cannabis use and SMC, but this association is complicated by several considerations, and the potential mechanisms underlying this relationship remain unclear. In our analyses, the increased risk of SMC in those who used cannabis in the past year was comparable to the contributions made by comorbid health conditions, such as asthma, cancer, heart conditions, or hypertension, each independent risk factors for SMC. Similarly, misuse of other substances (e.g., cocaine, heroin, inhalants, methamphetamine, etc.) also conferred comparable levels of risk for SMC. However, mental illness revealed a notably stronger effect on SMC, with the greatest risk observed in moderate to severe mental illness, consistent with the clinical literature showing large effects of mental illness on SMC (Keefe & Fenton, 2007; McCleery & Nuechterlein, 2019; Rock et al., 2014). But even after accounting for these and other covariates, past-year cannabis use remained a significant predictor of SMC. Given the higher rates of cannabis use among individuals with chronic health conditions, polysubstance use, and mental illness, these concurrent findings are not surprising, although it is important to underscore that multiple potential causal pathways may exist across these factors and more tightly controlled experimental methods are needed to clarify causal mechanisms underlying the associations between cannabis use and SMC in older adults. Moreover, recent use of cannabis (in the past month) did not show a statistically significant association with SMC, and no dose-response relationship between cannabis and SMC could be established in our analyses. In contrast, being married, having higher levels of education, and consuming alcohol in the past year were associated with small to moderate reductions in risk for SMC, consistent the clinical literature identifying marriage (Sommerlad et al., 2018), education (Stern, 2009; Tucker, 2012), and light to moderate alcohol consumption (Mukamal et al., 2003; Ruitenberg et al., 2002; Topiwala & Ebmeier, 2018) as protective factors for cognitive aging. Taken together, these findings provide important context for understanding the relationship between cannabis use and SMC in older adults and underscore the importance of population-level research to better understand the causal mechanisms underlying this relationship.
Several limitations of our study must also be considered. First, we were limited by the variables available from the NSDUH dataset. Specifically, the item used to identify respondents with SMC is conditioned upon “a physical, mental or emotional condition” impacting cognition, and it is not possible to delineate the factors that may contribute to endorsement of SMC. Additionally, NSDUH does not have objective data regarding the patterns of cannabis use, motivation for use, cannabis strain, cannabinoid profile, or route of administration. Such variables are important factors for better understanding the impacts of cannabis use in older adults and the relative risk and protective factors associated with use patterns and cannabinoid profiles on cognition (Sagar et al., 2021; Scott et al., 2019). Finally, our study did not include objective measures of memory functioning, and SMC is not reliably related to neuropsychological test performance, especially in those with mental illness (Carter et al., 2003; Hohman et al., 2011; Richardson-Vejlgaard et al., 2009). Performance-based neuropsychological testing is older adults is essential to better understand the relationships between cannabis use and cognition in this population.
A clearer understanding of the relationship between cannabis use and cognitive functioning in older adults would aid in differential diagnosis by helping to reduce heterogeneity in clinical phenotypes (e.g., mild cognitive impairment) of dementia. Furthermore, the clinical relevance of cannabis use in older adults presenting with SMC will become increasingly prevalent as cannabis use trends in this demographic continue to expand. Interventions geared toward comprehensive management of modifiable risk factors have demonstrated promising results in reducing the incidence of dementia (e.g., Rosenberg et al., 2019), although a full appreciation of cannabis use as a risk and/or protective factor in older adults remains unclear due to a lack of research in this population. Future research should include prospective, performance-based evaluation of cognitive functioning in older adults using cannabis. Particular attention should also be given to a careful assessment of use patterns, cannabis characteristics, and dose-response relationships. As the population ages and cannabis use increases, it will be important to understand the relationship between cannabis use and objective and subjective measure of cognitive functioning.
In sum, our findings reveal an increased rate of SMC in older adults with past-year cannabis use compared to those without past-year cannabis use. Risks of SMC associated with cannabis use were comparable to physical health problems and misuse of other substances among older adults, but notably less impactful than the presence and severity of comorbid mental illness. While the relationship between cannabis use and SMC remained significant despite accounting for multiple demographic and behavioral factors with known effects on cognition, the association between cannabis use and SMC became weaker and nonsignificant when evaluating more recent use periods (one month), and no dose-response relationship between cannabis use and rates of SMC was detected. As the first population-level analysis of cannabis use and SMC in older adults in the US, these hypothesis generating results help to characterize and contextualize the relationship between cannabis use and SMC in this population and further research is necessary to better understand the mechanisms that may underly these findings.
Supplementary Material
Funding support:
This work was supported by the Veterans Affairs Health Services Research and Development (VA HSR&D) Research Career Scientist Award Number RCS 19-333. Support to Dr. Hampstead from NIA R35AG072262 (effort) is acknowledged. The views expressed in this report are those of the authors, and do not necessarily represent those of the Department of Veterans Affairs or the United States Government. All authors had full access to all the data, and Dr. Ilgen takes responsibility for the integrity of the data and the accuracy of the data analysis.
Footnotes
COI: None.
References
- Abate G, Uberti D, & Tambaro S (2021). Potential and limits of cannabinoids in Alzheimer’s disease therapy. Biology, 10(6), 542. 10.3390/biology10060542 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aso E, & Ferrer I (2014). Cannabinoids for treatment of Alzheimer’s disease: Moving toward the clinic. Frontiers in Pharmacology, 5. 10.3389/fphar.2014.00037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Auer R, Vittinghoff E, Yaffe K, Künzi A, Kertesz SG, Levine DA, Albanese E, Whitmer RA, Jacobs DR, Sidney S, Glymour MM, & Pletcher MJ (2016). Association between lifetime marijuana use and cognitive function in middle age: The Coronary Artery Risk Development in Young Adults (CARDIA) study. JAMA Internal Medicine, 176(3), 352. 10.1001/jamainternmed.2015.7841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bar-Sela G, Tauber D, Mitnik I, Sheinman-Yuffe H, Bishara-Frolova T, & Aharon-Peretz J (2019). Cannabis-related cognitive impairment: A prospective evaluation of possible influences on patients with cancer during chemotherapy treatment as a pilot study. Anti-Cancer Drugs, 30(1), 91–97. 10.1097/CAD.0000000000000685 [DOI] [PubMed] [Google Scholar]
- Baumgart M, Snyder HM, Carrillo MC, Fazio S, Kim H, & Johns H (2015). Summary of the evidence on modifiable risk factors for cognitive decline and dementia: A population‐based perspective. Alzheimer’s & Dementia, 11(6), 718–726. 10.1016/j.jalz.2015.05.016 [DOI] [PubMed] [Google Scholar]
- Black P, & Joseph LJ (2014). Still dazed and confused: Midlife marijuana use by the baby boom generation. Deviant Behavior, 35(10), 822–841. 10.1080/01639625.2014.889994 [DOI] [Google Scholar]
- Blazer DG, & Wu L-T (2009). Nonprescription use of pain relievers by middle-aged and elderly community-living adults: National survey on drug use and health: Nonprescription pain reliever use in late life. Journal of the American Geriatrics Society, 57(7), 1252–1257. 10.1111/j.1532-5415.2009.02306.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bosnjak Kuharic D, Markovic D, Brkovic T, Jeric Kegalj M, Rubic Z, Vuica Vukasovic A, Jeroncic A, & Puljak L (2021). Cannabinoids for the treatment of dementia. Cochrane Database of Systematic Reviews, 2021(9). 10.1002/14651858.CD012820.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Campeny E, López-Pelayo H, Nutt D, Blithikioti C, Oliveras C, Nuño L, Maldonado R, Florez G, Arias F, Fernández-Artamendi S, Villalbí JR, Sellarès J, Ballbè M, Rehm J, Balcells-Olivero MM, & Gual A (2020). The blind men and the elephant: Systematic review of systematic reviews of cannabis use related health harms. European Neuropsychopharmacology, 33, 1–35. 10.1016/j.euroneuro.2020.02.003 [DOI] [PubMed] [Google Scholar]
- Carter SL, Rourke SB, Murji S, Shore D, & Rourke BP (2003). Cognitive complaints, depression, medical symptoms, and their association with neuropsychological functioning in HIV infection: A structural equation model analysis. Neuropsychology, 17(3), 410–419. 10.1037/0894-4105.17.3.410 [DOI] [PubMed] [Google Scholar]
- Center for Behavioral Health Statistics and Quality. (2021). 2020 National Survey on Drug Use and Health (NSDUH): Methodological summary and definitions. Substance Abuse and Mental Health Services Administration. https://www.samhsa.gov/data/ [Google Scholar]
- Chayasirisobhon S (2019). Cannabis and neuropsychiatric disorders: An updated review. 28(2), 13. [PubMed] [Google Scholar]
- Cohen K, Weizman A, & Weinstein A (2019). Positive and negative effects of cannabis and cannabinoids on health. Clinical Pharmacology & Therapeutics, 105(5), 1139–1147. 10.1002/cpt.1381 [DOI] [PubMed] [Google Scholar]
- Crane NA, Schuster RM, Fusar-Poli P, & Gonzalez R (2013). Effects of cannabis on neurocognitive functioning: Recent advances, neurodevelopmental influences, and sex differences. Neuropsychology Review, 23(2), 117–137. 10.1007/s11065-012-9222-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Crean RD, Crane NA, & Mason BJ (2011). An evidence-based review of acute and long-term effects of cannabis use on executive cognitive functions. Journal of Addiction Medicine, 5(1), 1–8. 10.1097/ADM.0b013e31820c23fa [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dregan A, & Gulliford MC (2012). Is illicit drug use harmful to cognitive functioning in the midadult years? A cohort-based investigation. American Journal of Epidemiology, 175(3), 218–227. 10.1093/aje/kwr315 [DOI] [PubMed] [Google Scholar]
- Englund A, Morrison PD, Nottage J, Hague D, Kane F, Bonaccorso S, Stone JM, Reichenberg A, Brenneisen R, Holt D, Feilding A, Walker L, Murray RM, & Kapur S (2013). Cannabidiol inhibits THC-elicited paranoid symptoms and hippocampal-dependent memory impairment. Journal of Psychopharmacology, 27(1), 19–27. 10.1177/0269881112460109 [DOI] [PubMed] [Google Scholar]
- Gruber SA, Sagar KA, Dahlgren MK, Racine MT, Smith RT, & Lukas SE (2016). Splendor in the grass? A pilot study assessing the impact of medical marijuana on executive function. Frontiers in Pharmacology, 7. 10.3389/fphar.2016.00355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Han BH, Sherman S, Mauro PM, Martins SS, Rotenberg J, & Palamar JJ (2017). Demographic trends among older cannabis users in the United States, 2006–13. Addiction, 112(3), 516–525. 10.1111/add.13670 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hohman TJ, Beason-Held LL, & Resnick SM (2011). Cognitive complaints, depressive symptoms, and cognitive impairment: Are they related? Journal of the American Geriatrics Society, 59(10), 1908–1912. 10.1111/j.1532-5415.2011.03589.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaskie B, Ayyagari P, Milavetz G, Shane D, & Arora K (2017). The increasing use of cannabis among older americans: A public health crisis or viable policy alternative? The Gerontologist, gnw166. 10.1093/geront/gnw166 [DOI] [PubMed] [Google Scholar]
- Keefe RSE, & Fenton WS (2007). How should DSM-V criteria for schizophrenia include cognitive impairment? Schizophrenia Bulletin, 33(4), 912–920. 10.1093/schbul/sbm046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lisdahl KM, Filbey FM, & Gruber SA (2021). Introduction to jins special issue: Clarifying the complexities of cannabis and cognition. Journal of the International Neuropsychological Society, 27(6), 515–519. 10.1017/S1355617721000874 [DOI] [Google Scholar]
- Lucas CJ, Galettis P, & Schneider J (2018). The pharmacokinetics and the pharmacodynamics of cannabinoids. British Journal of Clinical Pharmacology, 84(11), 2477–2482. 10.1111/bcp.13710 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Maxwell CJ, Jesdale BM, & Lapane KL (2021). Recent trends in cannabis use in older americans. Annals of Internal Medicine, 174(1), 133–135. 10.7326/M20-0863 [DOI] [PubMed] [Google Scholar]
- McCleery A, & Nuechterlein KH (2019). Cognitive impairment in psychotic illness: Prevalence, profile of impairment, developmental course, and treatment considerations. Dialogues in Clinical Neuroscience, 21(3), 239–248. 10.31887/DCNS.2019.21.3/amccleery [DOI] [PMC free article] [PubMed] [Google Scholar]
- McKetin R, Parasu P, Cherbuin N, Eramudugolla R, & Anstey KJ (2016). A longitudinal examination of the relationship between cannabis use and cognitive function in mid-life adults. Drug and Alcohol Dependence, 169, 134–140. 10.1016/j.drugalcdep.2016.10.022 [DOI] [PubMed] [Google Scholar]
- Meier MH, Caspi A, Danese A, Fisher HL, Houts R, Arseneault L, & Moffitt TE (2018). Associations between adolescent cannabis use and neuropsychological decline: A longitudinal co-twin control study: Adolescent cannabis use and IQ decline. Addiction, 113(2), 257–265. 10.1111/add.13946 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, & Stubbs B (2014). Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: Meta-analysis. Acta Psychiatrica Scandinavica, 130(6), 439–451. 10.1111/acps.12336 [DOI] [PubMed] [Google Scholar]
- Mukamal KJ, Kuller LH, Fitzpatrick AL, Longstreth WT J., Mittleman MA, & Siscovick DS (2003). Prospective study of alcohol consumption and risk of dementia in older adults. JAMA, 289(11), 1405–1413. 10.1001/jama.289.11.1405 [DOI] [PubMed] [Google Scholar]
- Ranganathan M, & D’Souza DC (2006). The acute effects of cannabinoids on memory in humans: A review. Psychopharmacology, 188(4), 425–444. 10.1007/s00213-006-0508-y [DOI] [PubMed] [Google Scholar]
- Richardson-Vejlgaard R, Dawes S, Heaton RK, & Bell MD (2009). Validity of cognitive complaints in substance-abusing patients and non-clinical controls: The Patient’s Assessment of Own Functioning Inventory (PAOFI). Psychiatry Research, 169(1), 70–74. 10.1016/j.psychres.2008.06.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rock PL, Roiser JP, Riedel WJ, & Blackwell AD (2014). Cognitive impairment in depression: A systematic review and meta-analysis. Psychological Medicine, 44(10), 2029–2040. 10.1017/S0033291713002535 [DOI] [PubMed] [Google Scholar]
- Rosenberg A, Mangialasche F, Ngandu T, Solomon A, & Kivipelto M (2019). Multidomain interventions to prevent cognitive impairment, Alzheimer’s disease, and dementia: From FINGER to World-Wide FINGERS. The Journal of Prevention of Alzheimer’s Disease, 1–8. 10.14283/jpad.2019.41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ross JM, Ellingson JM, Rhee SH, Hewitt JK, Corley RP, Lessem JM, & Friedman NP (2020). Investigating the causal effect of cannabis use on cognitive function with a quasi-experimental co-twin design. Drug and Alcohol Dependence, 206, 107712. 10.1016/j.drugalcdep.2019.107712 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ruitenberg A, van Swieten JC, Witteman JC, Mehta KM, van Duijn CM, Hofman A, & Breteler MM (2002). Alcohol consumption and risk of dementia: The Rotterdam Study. The Lancet, 359(9303), 281–286. 10.1016/S0140-6736(02)07493-7 [DOI] [PubMed] [Google Scholar]
- Sagar KA, Dahlgren MK, Lambros AM, Smith RT, El-Abboud C, & Gruber SA (2021). An observational, longitudinal study of cognition in medical cannabis patients over the course of 12 months of treatment: Preliminary results. Journal of the International Neuropsychological Society, 27(6), 648–660. 10.1017/S1355617721000114 [DOI] [PubMed] [Google Scholar]
- SAS Institute. (n.d.). SAS (13.1).
- Scott EP, Brennan E, & Benitez A (2019). A systematic review of the neurocognitive effects of cannabis use in older adults. Current Addiction Reports, 6(4), 443–455. 10.1007/s40429-019-00285-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sommerlad A, Ruegger J, Singh-Manoux A, Lewis G, & Livingston G (2018). Marriage and risk of dementia: Systematic review and meta-analysis of observational studies. Journal of Neurology, Neurosurgery & Psychiatry, 89(3), 231–238. 10.1136/jnnp-2017-316274 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stern Y (2009). Cognitive reserve. Neuropsychologia, 47(10), 2015–2028. 10.1016/j.neuropsychologia.2009.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Substance Abuse and Mental Health Services Administration. (2014). Results from the 2013 national survey on drug use and health: Summary of national findings (NSDUH Series H-48, HHS Publication No. (SMA) 14–4863; p. 184). Substance Abuse and Mental Health Services Administration. [Google Scholar]
- Topiwala A, & Ebmeier KP (2018). Effects of drinking on late-life brain and cognition. Evidence-Based Mental Health, 21(1), 12–15. 10.1136/eb-2017-102820 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tucker AM (2012). Cognitive reserve in aging. 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- US Census Bureau. (2001, October 3). Population by age, sex, race, and Hispanic or Latino origin for the United States: 2000. Census.Gov. https://www.census.gov/data/tables/2000/dec/phc-t-09.html [Google Scholar]
- US Census Bureau. (2019). Age and sex composition in the United States: 2019. Census.Gov. https://www.census.gov/data/tables/2019/demo/age-and-sex/2019-age-sex-composition.html [Google Scholar]
- Vacaflor BE, Beauchet O, Jarvis GE, Schiavetto A, & Rej S (2020). Mental health and cognition in older cannabis users: A review. Canadian Geriatrics Journal, 23(3), 237–244. 10.5770/cgj.23.399 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Volkow ND, Swanson JM, Evins AE, DeLisi LE, Meier MH, Gonzalez R, Bloomfield MAP, Curran HV, & Baler R (2016). Effects of cannabis use on human behavior, including cognition, motivation, and psychosis: A review. JAMA Psychiatry, 73(3), 292. 10.1001/jamapsychiatry.2015.3278 [DOI] [PubMed] [Google Scholar]
- Weinstein G, & Sznitman SR (2020). The implications of late-life cannabis use on brain health: A mapping review and implications for future research. Ageing Research Reviews, 59, 101041. 10.1016/j.arr.2020.101041 [DOI] [PubMed] [Google Scholar]
- Wu L-T, & Blazer DG (2011). Illicit and nonmedical drug use among older adults: A review. Journal of Aging and Health, 23(3), 481–504. 10.1177/0898264310386224 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.