Association Between Loneliness and Cannabis Consumption Among Older Adults: Insights from a Nationwide Canadian Survey Amid the COVID-19 Pandemic

Introduction

Since the onset of the COVID-19 pandemic in early 2020, the sharp rise in loneliness among older adults has emerged as a significant public health issue in Canada and globally [1, 2]. Compared to younger individuals, older adults face a higher risk of loneliness, largely due to factors like reduced mobility associated with aging, a shrinking social network, and chronic health conditions that limit their ability to engage socially [3, 4]. Throughout the pandemic, measures such as physical distancing, limited access to public services, and the disproportionate burden of COVID-19 on older populations have negatively impacted the daily lives of older Canadians. Numerous studies conducted during this period have documented increased loneliness in this demographic along with its associated risk factors [1]. Nevertheless, the connection between loneliness and cannabis use among older adults has received less attention than other subjects like multimorbidity or gender, both in Canada and in other regions where cannabis is legalized, including some U.S. states.

Cannabis use among older adults has become a critical public health concern for two primary reasons: older adults have shown the largest rise in cannabis consumption, and they are particularly susceptible to the potential negative health and wellbeing consequences of cannabis use [5]. In Canada, cannabis use among older adults increased notably during the pandemic [6]. However, many cannabis-related studies tend to focus on specific populations—such as young adults, people with post-traumatic stress disorder (PTSD) or other chronic illnesses, or the general population—without specifically including or highlighting older adults [7]. This gap restricts the applicability of findings to the aging population. Furthermore, research on cannabis use predominantly concentrates on its physical and mental health impacts, while social wellbeing outcomes like loneliness remain less explored [8]. Thus, it is important to investigate the link between loneliness and cannabis use among older adults during the COVID-19 pandemic.

Loneliness among older adults

Loneliness describes the distress or uncomfortable feelings people experience when their social needs are unmet or when there is a mismatch between the social relationships they desire and those they currently have [3, 9]. Unlike social isolation, which is objectively measured by factors such as the size of one’s social network or level of social participation, loneliness focuses on subjective perceptions of social connection [10]. It is a common issue among older adults, with research indicating that between one-fifth and two-fifths of this population experience loneliness ranging from occasional to severe [11]. For example, findings from the Canadian Longitudinal Study on Aging (CLSA) by Raina et al. (2018) revealed that approximately 25% of older women and 20% of older men reported feelings of loneliness at least some of the time [4]. Certain groups of older adults, including those residing in isolated regions, caregivers, and individuals managing multiple chronic conditions, report even greater levels of loneliness [1]. Loneliness represents a critical factor affecting healthy aging, as it is linked to faster biological aging, increased illness and mortality rates, cognitive decline, poorer mental health, and reduced quality of life [9, 12]. Therefore, understanding and addressing its risk factors in older adults is essential.

The COVID-19 pandemic has exacerbated feelings of loneliness in older adults due to fears of contracting the virus and the implementation of preventive measures like physical distancing [2]. For instance, a Canadian survey conducted in May 2020 found that over 43% of older adults experienced loneliness at least some of the time [13]. Similarly, a study from the UK reported that 51% of older adults felt lonely early in the pandemic [14]. Moreover, a comprehensive review of 33 studies highlighted a significant increase in loneliness throughout the pandemic period [15]. The detrimental effects of loneliness on the health and wellbeing of older adults have been observed both before and during the pandemic. Research by Wister et al. (2022) showed that loneliness prior to the pandemic, as well as its increase during the pandemic, significantly predicted depressive symptoms among older Canadians [16]. Additional studies also found that loneliness during the pandemic was linked to problems such as insomnia [17] and alcohol misuse [18] in older populations.

Cannabis use and wellbeing among older adults

Since the 2018 legalization of cannabis in Canada, usage rates among older Canadians have steadily climbed. Data from The Commonwealth Fund’s 2020 International Health Policy survey show that nearly 18% of Canadians aged 50 to 64 and 8% of those aged 65 and over reported cannabis consumption in the previous year [19]. Notably, older adults have emerged as the fastest-growing demographic for cannabis use in both Canada and the U.S. [20, 21]. This upward trend has been linked to the legalization itself, shifting social attitudes—especially among baby boomers—greater availability, and the expanding use of cannabis for medical purposes among seniors [8]. Following the outbreak of COVID-19, cannabis consumption rose considerably among older adults, albeit less dramatically than among younger populations [21, 22]. According to Statistics Canada (2022), around 20% of cannabis users aged 50 to 64, and 22% of users aged 65 and above, reported increasing their use during the pandemic, compared to a 34% rise in cannabis use among previous users overall [6].

The effects of cannabis use on older adults’ wellbeing are complex and multifaceted. On the positive side, cannabis has been shown to help relieve chronic pain and ease symptoms of stress, anxiety, and depression [23–25]. For instance, during the pandemic, cancer survivors used cannabis to manage anxiety and improve sleep [26], while some older adults reported using it to boost mood or calm anxiety [27]. Nonetheless, these benefits appear to be mostly short-term, highlighting the necessity to further study the potential long-term consequences. Research increasingly links cannabis use in older adults with multimorbidity, harmful behaviors such as heavy alcohol use, a higher risk of falls and injuries, and increased visits to emergency departments [23, 28]. Evidence from the pandemic period reinforces concerns about cannabis’s negative impact, particularly for long-term users. For example, a study by Murkar et al. (2022) showed that increased cannabis use during the pandemic correlated with elevated depressive symptoms in people with PTSD [7]. Additionally, cannabis users tend to experience more COVID-19 hospitalizations and higher mortality rates related to the virus [29]. The established association between cannabis use and respiratory conditions such as Chronic obstructive pulmonary disease further aggravates risks for older adults infected with COVID-19 [30, 31]. Overall, cannabis use contributes to heightened health risks among older adults, both before and throughout the pandemic. The conflicting evidence concerning its advantages and dangers underlines the urgent need for further research in this area.

Loneliness and cannabis use

Compared to other aspects of health and wellbeing, the relationship between loneliness and cannabis use has been less thoroughly explored. Previous research does support a link between these two factors. Although there is a possibility that cannabis use and various health or wellbeing conditions—including loneliness—may influence each other in both directions, most studies emphasize identifying common reasons why people use cannabis [25]. From this viewpoint, cannabis use is often considered a coping mechanism for loneliness. During the COVID-19 pandemic, loneliness was one of the primary drivers behind increased cannabis consumption among Canadians, along with other motives such as stress, boredom, relaxation, and managing emotions like tension, anger, or sadness [6, 32]. For example, a study focusing on young adults who used cannabis in the U.S. found that loneliness was the leading factor for their increased use during the pandemic [33]. Additionally, social isolation and living alone—conditions closely linked to loneliness—have also been associated with greater cannabis use throughout the pandemic [34]. Nonetheless, some longitudinal studies in the U.S., including one that collected data across four waves between 2020 and 2021 [35] and another among American veterans during the pandemic [36], did not find conclusive evidence that feeling lonelier directly caused an increase in cannabis use. These mixed findings illustrate a complex and sometimes contradictory relationship between cannabis use and loneliness both before and during the pandemic.

On the other hand, some research argues that cannabis use may actually increase loneliness. Several explanations support this perspective. One is that cannabis users, like other substance users, often face stigma linked to their substance use, which can significantly contribute to feelings of loneliness [37]. For instance, among people with substance use disorders, stigma related to drug addiction is a major predictor of loneliness [38]. Frank et al. (2013) noted that regular cannabis users sometimes avoid consuming cannabis in the presence of family or at social gatherings, conscious of potential judgment and social distancing between users and non-users [39]. While this does not necessarily prevent them from engaging socially altogether, it may reduce their participation in many social opportunities. Furthermore, individuals who use cannabis regularly or have cannabis use disorder often experience difficulties regulating their emotions, which can extend the duration of loneliness beyond typical levels [40]. According to Cadigan et al. (2023), young adults with higher patterns of loneliness during the pandemic were more likely to have used cannabis in the month before being surveyed, compared to those with lower loneliness trajectories [40]. Moreover, loneliness may be a key mechanism explaining how cannabis use elevates the risk of mental health disorders such as depression or increases suicide risk [41]. Therefore, a deeper understanding of how cannabis use influences loneliness could improve insight into the negative health consequences associated with cannabis.

Overall, cannabis use among older adults remains understudied, particularly in terms of how it affects loneliness. This study therefore aims to explore the connection between cannabis use and loneliness among older Canadians during the pandemic.

Materials and Methods

This research used data from the Canadian Perspectives Survey Series (CPSS) 6 conducted in 2021, titled Substance Use and Stigma During the Pandemic [42]. The CPSS, initiated in early 2020, is a collection of surveys aimed at capturing information on the knowledge, attitudes, and behaviors of Canadians living in all ten provinces. Specifically, Series 6 focused on gathering data about alcohol and drug use, lifestyle habits, social activities, and COVID-19’s impact among Canadians aged 15 and older. In January 2021, a total of 3,941 respondents completed this survey, accounting for approximately 54.4% of the original CPSS panel. For this analysis, a subgroup of 2,020 individuals aged 55 years or older was selected as the study sample.

Dependent variable

The primary outcome measured was loneliness, assessed through an adapted version of the 3-item UCLA loneliness scale [43]. This scale asks participants to rate their experiences related to: (1) feeling a lack of companionship, (2) feeling excluded, and (3) feeling isolated from others. While the original instrument employs a 3-point Likert scale (Hardly ever, Some of the time, Often) and has demonstrated reliability and validity for older adult populations in Canada and internationally [16, 44], the CPSS 6 utilized a 5-point Likert scale ranging from 1 (never) to 5 (always) for these same items. The internal consistency of the scale in this sample was strong, as reflected by a Cronbach’s Alpha of 0.826. Loneliness scores were computed by summing responses to the three items, yielding a range from 5 to 15, where higher values correspond to greater feelings of loneliness.

Independent variables

Two key independent variables were examined: cannabis use frequency and changes in cannabis use. Participants reported their cannabis use frequency over the past thirty days by choosing from seven response options, including never used, no use in the past thirty days, use on one day, 2–3 days, or daily use. These responses were later grouped into three categories for analysis: Never used cannabis, No use in the past 30 days, and Used cannabis in the past thirty days. Additionally, cannabis users were asked to compare their current use to their habits before the COVID-19 pandemic, indicating whether their use had increased, decreased, or remained stable. To include non-users, this variable was recoded into four categories: Never used, Increased use, Decreased or stable use (due to the small number reporting decreased use), and Not stated, the latter capturing about 7% of missing responses.

Control variables

To adjust for confounding factors, a range of socio-demographic and social variables were included in the analysis [1]. Age was categorized into three groups: 55–64 years, 65–74 years, and 75 years or older. Sex was recorded as Male or Female based on sex at birth. Marital status was originally measured with 5 categories but was simplified into Married/partnered and Single (including those not married, separated, or widowed). Education levels were grouped into four tiers: less than high school, high school diploma, some post-secondary education, and university diploma or degree. Employment status was dichotomized as either working or not working. Geographic location was classified by population density into Rural or Urban. Birthplace was categorized as Canada or Other countries. Additionally, three social interaction variables closely tied to loneliness were controlled for: household size, number of close friends and relatives, and social participation. Household size was divided into living alone, living with one other person, or living with three or more individuals. The number of close relatives and friends was measured using five categories: none, one to two, three to five, six to nine, and ten or more. Social participation was indicated by whether respondents were members of any group or association (yes/no).

Additionally, three variables related to substance use were incorporated into the analysis: attitudes toward substance use, alcohol consumption status, and opioid use status. Attitudes toward substance use were measured by combining scores from four statements: “People who have a problem with alcohol or drugs should feel embarrassed or uncomfortable telling friends or family,” “Society should be more accepting of people with alcohol or drug problems,” “People with alcohol or drug problems should feel embarrassed or uncomfortable seeking help or treatment,” and “I have sympathy for people who misuse alcohol or drugs” (with the second and fourth items reverse scored). The total attitude score was calculated by summing these four responses, yielding a Cronbach’s Alpha of 0.640, where higher scores reflect more positive attitudes toward substance use. Alcohol consumption was categorized into three groups: never drank alcohol, drank alcohol in the past 30 days, and did not drink in the past 30 days. Similarly, participants reported their opioid use status (prescribed or non-prescribed) within three categories: never used opioids, used opioids in the past 30 days, and no opioid use in the past 30 days.

Data analytic procedure

Data analysis was conducted using SPSS version 28. Initially, descriptive statistics were calculated to summarize the characteristics of the selected sample, as presented in Table 1. Subsequently, group comparisons were made based on participants’ frequency of cannabis use, examining differences in socio-demographic factors, social interaction variables, and attitudes toward substance use, alcohol, and opioid use; these results are shown in Table 2. Lastly, linear regression models were applied to explore the relationships between loneliness and both cannabis use and changes in cannabis use. In Table 3, Model 1 includes all control variables alongside cannabis use status, whereas Model 2 incorporates all controls with the variable reflecting changes in cannabis use. Following Statistics Canada’s guidelines, all analyses were weighted using both sampling and standardized weights.

Notes: SD = standard deviation.

Note: *p < .05; **p < .01; ***p < .001.

Note: Reference categories are listed in parentheses.

* p < .05, ** p < .01, *** p < .001