Potential for alcohol and drug interactions in older adults: evidence from the Irish longitudinal study on ageing

Older adults are susceptible to adverse effects from the concomitant use of prescription medications and alcohol, in part, because of changes in absorption, distribution and metabolism of alcohol and other medications with age [1]. In addition to these physiological changes older adults have reduced homeostatic capacity, and often take multiple alcohol interactive (AI) medications [2‐4]. The major adverse clinical outcomes of medication-alcohol interactions include raising blood alcohol levels, altering the metabolism of drugs, liver toxicity, gastrointestinal inflammation and bleeding, sedation, disulfiram-like-reactions, and interference with the effectiveness of medications and exacerbating their adverse effects [5, 6]. For example, alcohol potentiates the sedative effects of benzodiazepines, antidepressants (e.g. tricyclics), anithisamines, muscle relaxants and opioids [6], giving rise to falls, car accidents and death [7, 8]. Nonsteroidal anti-inflammatory drugs (NSAIDs), when combined with alcohol, can increase an older patients risk for gastrointestinal bleeding [6]. A systematic review has also indicated that alcohol consumption exacerbates certain chronic conditions, such as liver disease, diabetes mellitus, gastritis and other gastrointestinal conditions, gout, depression, and breast cancer [9].

Although older adults typically report lower levels of heavy drinking [5, 10, 11], many continue to engage in patterns of alcohol consumption that exceed current guidelines for older adults [11‐15]. Despite being susceptible to adverse effects from the concomitant use of AI medications and alcohol, few studies have considered the potential for alcohol-drug interactions in older adults [16, 17]. One study found that 10% of older US adults were at risk for potential adverse effects due to concomitant use of alcohol and AI medications or conditions known to be exacerbated by alcohol, this estimate increased to 26% among current drinkers [18]. Another US study of community dwelling older adults identified 21.5% of current drinkers as at-risk drinkers due to their alcohol use combined with select comorbidities and 21.2% at-risk due to their alcohol use combined with AI medications [12]. While these studies provide prevalence estimates of the potential risk for adverse effects due to concomitant use, specific patterns of concomitant use are lacking [17]. Only one study to date has considered the prevalence of alcohol use by therapeutic class, with the most common combination of alcohol and AI medications occuring with NSAIDs (20.2%), antihistamines (20.1%) and anti-hypertensives (19.8%) [17]. These results relate to older US adults, which may differ from European countries, as Europe reports the highest per-capita alcohol consumption in the world with alcohol related deaths among older Europeans increasing considerably over the past ten years [19]. The purpose of this study is to estimate the prevalence of AI medication exposure and concomitant alcohol use by therapeutic class in a large, nationally representative population sample of older Irish adults. Additional objectives include: (i) the estimation of the prevalence of concomitant alcohol use with specific conditions known to deteriorate with alcohol consumption [9]; and (ii) the identification of sociodemographic and health factors associated with concomitant use of AI medications and alcohol.

In this large, population-based sample of older people, 72% were exposed to at least one AI medication, with exposure varying by therapeutic class. Of those exposed to AI medications, 60% reported concomitant alcohol use, increasing their risk for alcohol related adverse reactions. This is in contrast to a large community based US study in adults of similar ages in which 20% of those on AI medications reported alcohol use [17]. The prevalence of alcohol use across each of the therapeutic classes was also larger in this study relative to the US study [17]. For example, we found that 53.5% of people using CNS agents concurrently drink alcohol, compared to 20.2% of older American adults. This may be due to a lower prevalence of self- reported alcohol consumption in the US study, 20.3% compared to 62.3% of older Irish adults. It is plausible that a higher proportion of older Irish adults drink alcohol, thereby increasing their potential risk for alcohol related adverse reactions. Furthermore, 8% of the current sample, (12.7% of current drinkers), were identified as problem drinkers on the CAGE screening questionnaire, which is higher than previous studies of older people [13, 19]. Some of these differences may be explained by our younger cohort (60 years and older), however this is unlikely to account for the total variation as, although the prevalence of alcohol consumption declined with age, it remained relatively high among our older age-groups.

Although exposure to AI medications varied by therapeutic class, an examination of concurrent alcohol use suggests consistency across therapeutic classes. Concurrent alcohol use ranged between approximately 50-60% for most therapeutic classes, with the exception of muscle relaxants (80%). Similarly, heavy drinking was identified in approximately one in five older adults reporting concurrent use of cardiovascular, blood and anti-diabetic agents. The concurrent use of psychotropics with heavy alcohol consumption is of particular concern, as they may cause the most dangerous alcohol related adverse drug reactions [8]. Sixteen percent of older adults taking CNS agents were identified as heavy drinkers; 14.8% of people taking opioid analgesics reported heavy drinking. Similarly, 13% of people taking anti-epileptics, antipsychotics or hypnotics reported heavy drinking, with almost one in five of those taking antidepressants also reporting heavy drinking. This is consistent with other studies, which suggest that concurrent use of psychotropic drugs and alcohol has become more prevalent in older adults [32‐34]. However, this may be an artefact of increased use of psychotropic medications with age. We also found a high prevalence of alcohol consumption among people with conditions known to be exacerbated by alcohol consumption.

Although we are unable to ascertain the reasons behind alcohol consumption, particularly heavy consumption, in combination with AI medications, previous research has shown that many older adults are often unaware of the potential risk [35]. This may be compounded by the fact that older people are less likely to reveal a history of excessive alcohol consumption [36], and healthcare workers have a lower degree of suspicion when assessing older people [37]. In light of these findings, general practitioners should take the opportunity to review their patients’ alcohol consumption prior to prescribing AI medications. Knowing how much alcohol their older patients are drinking would facilitate an effective assessment of risk, and identify those patients in need of counselling in relation to the safe use of alcohol and medications. Counselling patients on long-term AI medications is particularly important, as it may be easier to abstain from alcohol if concomitant medications are for defined short time periods compared to long-term medications. Pharmacists are also in a position to educate patients in relation to the potential risk associated with alcohol consumption and AI medication at the point of dispensing AI medications.

Limited clinican time for alcohol screening highlights the need to prioritise high-risk groups. Findings from the present study suggest that younger, better educated men, with a history of smoking and living in a city are at greatest risk. One positive finding was a 9% reduction in the odds of alcohol consumption among heavy drinkers with each additional AI drug taken. This is consistent with the US study [17]. In contrast, those with increasing numbers of co-morbitities, known to worsen with alcohol, were significantly more likely to engage in heavy drinking. This may represent a coping strategy.

When interpreting our findings, the following limitations should be considered. A number of medical conditions, and drinking frequency and quantity were based on self-report, which may result in misclassification bias. However, evidence suggests that self-reported alcohol consumption is both reliable and valid [38]. Furthermore, although this study identifies a number of risk factors that increase older peoples’ risk of exposure to alcohol-drug interactions, we do not have information on the drug strength, dose or duration or on the consequences or severity of these interactions. Further study of this cohort is necessary to quantify adverse outcomes associated with concomitant use of AI medications and alcohol.

Furthermore, the finding that alcohol consumption declines with age may represent a survival bias, with heavy drinkers either dying or stopping their alcohol at an earlier age. Future longitiudinal analysis is necessary to determine whether this age related decline remains with advancing age, as research has found that the average amount of alcohol consumed by older people who continue to drink does not change over time [39].

In conclusion, these findings indicate that many Irish older adults drink in excess of the recommended limits for their age, with many using alcohol in combination with alcohol interactive medications or conditions known to be exacerbated with alcohol intake. As the proportion and age of the older population continues to increase, the absolute number using AI medications will also increase. This increase in the prevalence of AI medications, combined with increased use of alcohol suggests the proportion of older adults at risk of alcohol related adverse drug events will increase in future years [17]. These findings highlight the importance of patient education to ensure older adults understand the potential risks associated with their combined alcohol and medication use. Prescribers should also be aware of the potential interactions, and screen patients for alcohol use and provide warnings to minimize patient risk.