Older persons using alcohol and other drugs, compared to younger people, are more susceptible to harms related to substance use due to age-related physiological changes and use of prescription drugs that may interact with alcohol and other drugs and lead to unplanned hospitalization (1–5). Detection of problematic substance use in older persons, however, can be challenging due to comorbid health conditions which mask effects of substance use; hence, practitioners may attribute symptoms to other health conditions and age-related physiological declines (6–8). Some clinicians are uncomfortable asking older patients about substance use (8–11). As a result, older persons might not be offered early intervention and treatment until coexisting health and psychosocial problems worsen.

Comorbid somatic and mental health disorders are common among older persons with chronic problematic substance use leading to resource-intensive old-age care, hospitalization, and post-discharge specialized care (7, 12–17). In Sweden, adults aged 50 years or older account for 66% of those hospitalized with alcohol use disorder and 20% with drug use disorder (18). The Swedish Association of Local Authorities and Regions (SALAR) estimates that one-third of psychiatric-related hospitalizations were among adults aged 65 years or older (19). More than 90% of inpatient stays due to psychiatric and SUD diagnoses among older patients, moreover, were unplanned and cost between US $3,000 and $6,300 (increasing cost with age) (19).

The Swedish substance use and mental health services are decentralized to regional and municipal services. The Social Services Act mandates that municipal social services provide substance use assessments and psychosocial services for treatment of problematic substance use (20, 21). The Health and Medical Services Act, however, assigns the diagnosis of SUD and mental disorders to the regional healthcare systems which provide low threshold services (such as needle exchange programs) and medical treatment for SUD, psychiatric and other comorbid conditions. The separate legal frameworks governing the social services, and the healthcare system complicate sharing patient data. Many older service users in Sweden have multiple hospital admissions prior to receiving an assessment for substance use severity and planned treatment services (12). Because of the fragmentation of substance use, mental and physical health services between municipal and regional authorities, those assessed for substance use severity may not receive the treatments they need leading to multiple unplanned hospitalizations. The Swedish government is investigating the transfer of substance use treatment from municipal social services to regional healthcare services (22). Knowledge about how often, how long and how early older persons are hospitalized after contact with municipal social services for assessment of substance use severity can inform the development of new policies for coordination of care among older persons.

This study examines the relationships between hospitalization and diagnoses of mental health and substance use disorders (at hospital admission) among older persons, aged 50+ years. An analysis linked multiple Swedish registries and assessed associations with 12-month cumulative hospitalized days, hospital admission rate and days to first hospital admissions following an assessment for substance use severity.

Between 2003 and 2017, 2,664 of 3,624 (73.5%) individuals were hospitalized one or more times following an index ASI assessment for substance use severity. See Table 1 for the characteristics of the cohort. About 54.5% (n = 1,976) were admitted with SUD diagnoses, 10.9% (n = 395) with dual diagnoses and 12.0% (n = 434) with psychiatric diagnoses. Furthermore, 14.2% (n = 514) were hospitalized with other diagnoses. Those who were hospitalized, compared to those not hospitalized, were older, not married, had lower employment rates, less education, and more history of hospitalization (see Table 1).

When limiting the analysis to the 12 months post ASI assessment, two in five participants (42.7%; n = 1,549) were hospitalized at least once during the 12 months post ASI assessment. Of those, 73% (n = 1,130) were hospitalized with SUD, 10.8% (n = 168) with dual diagnoses and 7.4% (n = 115) with other psychiatric diagnoses. About 20% (n = 310) were admitted with diagnoses other than psychiatric and/or SUD conditions.

The retrospective cohort study of older persons using alcohol and other drugs investigated the association of SUD, psychiatric and dual diagnoses with three outcome measures: 12-month cumulative hospitalized days, hospital admission rate, and time to first hospitalization. During the study period, almost three-fourth were hospitalized following their initial ASI assessment at municipal social services. More than half were hospitalized with SUD and about one-fifth with either non-SUD related psychiatric conditions or dual diagnosis. While the analysis did not consider what type of treatments service users were provided, the high proportion of hospitalization implies that for some their substance use related treatment needs might not have been met. These findings aligned with previous reports and suggested that medical and psychiatric comorbidities are prevalent among older persons with substance use problem leading to increased healthcare utilization (13, 15, 40–42).

While some older persons assessed for substance use severity might not fulfill the SUD diagnosis criteria, previous studies suggest that using even a small quantity of substance might harm older persons especially those with comorbidities and complex care needs (13, 43, 44). Additionally, co-occurrence of psychiatric diagnosis among older persons with substance use disorder is associated with medical multimorbidity, lower medication adherence, severe medication interactions, repeated hospital admission and longer hospital stay (13, 43, 45–47). Our analysis suggests SUD, psychiatric and dual diagnoses are associated with longer cumulative hospitalized days, higher hospitalization rate, and shorter number of days between initial ASI assessment and first hospitalization. For cumulative hospitalized days and hospital admission rate, the partial effects of SUD and/or psychiatric diagnoses increased with age supporting previous studies which found those aging with problematic substance use have higher healthcare utilization (13, 16, 41, 48, 49). For those who were hospitalized following ASI assessment, however, the partial effects of SUD and/or psychiatric diagnoses on the number of days to first hospitalization did not vary across age. This might be explained by the fact that the analysis included only those hospitalized, who were older and more likely to have a prior inpatient stay than those not hospitalized.

While the current study does not investigate the effect of contextual factors such as coordination of care for substance use and mental health, analysis from The Swedish Agency for Health and Care Services Analysis reported that the majority of service users were not satisfied with the coordination of their treatments within health and social services, and many were forced to coordinate their own treatments between service providers (50). Moreover, official data suggests that more than 90% of hospital admissions due to psychiatric and SUD diagnoses among older patients are unplanned (19). The current study found that most older persons assessed for substance use severity were later hospitalized with SUD and/or psychiatric diagnoses.

In addition to SUD treatment, municipal social services also provide old-age care. While this double responsibility theoretically makes social services ideal treatment providers for older persons with substance use problem, almost three-fifth of the municipalities in Sweden lack any coordination strategy between old-age care and substance use services, and only 11% of municipal old-age care services use standardized instrument to screen for SUD (17, 51). Those providing old-age care either lack the competence or the resources to identify and treat substance use disorder among older service users (52, 53). Moreover, only 40% of municipalities have joint agreements with regional entities to coordinate social and health care services (51) despite legislation stipulating the need for a coordinated individual plan for those in need of care from both social services and healthcare services such as those with dual diagnoses (50). The Swedish government is investigating the transfer of addiction treatment from social services to healthcare services (21, 22). Integrating substance use and mental health services into primary care, training multidisciplinary care providers on old-age problematic substance use, streamlining data sharing within and between health and social care services could result in earlier identification and treatment of substance use disorder and comorbid conditions among older persons (6, 45, 54–59).

Almost 75% of older persons who were assessed for substance use severity were later hospitalized due to any cause. Substance use disorders, dual diagnosis and other mental disorders were the main reasons for hospitalization and were all associated with longer 12-month cumulative hospitalized days, higher admission rate and shorter time to hospitalization. These effects, except for time to hospitalization, become even more pronounced with increasing age. Older persons often have multiple care needs from social services, primary care, and specialist healthcare services. Sensitizing service providers to late life substance use and sharing data across the care continuum could provide multiple points of contact to early identify and to provide timely and equitable care and to reduce the risk of hospitalizations among older persons with substance use problem.

The datasets presented in this article are not readily available because of legal and ethical restrictions. Codes for data preparation and analyses, and Stata log files can be shared through correspondence with WJ (wossenseged.jemberie@umu.se). ASI data are available via the Swedish municipalities. Hospital discharge data are available via the National Board of Health and Welfare. Crime statistics data on persons found guilty of offences are available via the Swedish National Council for Crime Prevention. Requests to access the datasets should be directed to Socialstyrelsen, patientregistret@socialstyrelsen.se; Brottsförebyggande rådet, statistik@bra.se.

The studies involving human participants were reviewed and approved by the Ethical Review Authority in Umeå, Sweden (DNR: 2016/504-31; amendment 2020-06233). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

WJ: conceptualization (lead), methodology (lead), formal analysis (lead), writing—original draft (lead), writing—review and editing (lead), visualization (lead), and funding acquisition (equal). MP: conceptualization (supporting), writing—review and editing (supporting), supervision (lead), validation (equal), and project administration (supporting). DM: conceptualization (supporting), writing—review and editing (equal), supervision (supporting), and validation (equal). LL: funding acquisition (lead), project administration (lead), writing—review and editing (supporting), validation (equal), and supervision (supporting). All authors reviewed and approved submission of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.