Alcohol and other drug policies for health care workplaces: evaluation of the content, quality and implementability

Alcohol and other drug (AOD) policies represent a key strategy for preventing and managing employees' substance use at the workplace. Policies are used to clarify rules and organizational expectations concerning acceptable employee behavior as well as the use and availability of alcohol and drugs at work and in work-related functions (Alfred et al., 2021; Pidd et al., 2016). Substance use disorder (SUD) is a cluster of physiological, behavioral and cognitive symptoms associated with the continued use of substances despite substance-related problems (American Psychiatric Association, 2022). It can be detected based on changes in work attendance and cognitive, social and physical performance (Cadiz et al., 2015b). It affects professionals' well-being, competency and physical or emotional condition. Effects can manifest as personality changes, confusion, unprofessional communication, underperformance, isolation and endangerment of patient safety (Rice, 2023). However, the shame and stigma around SUD may encourage those experiencing it to conceal their condition and avoid seeking treatment (Fetterhoff, 2023; Srivastava, 2018).

International Labor Organization (ILO) and European Agency for Safety and Health at Work (EU-OSHA) provide recommendations for the content and implementation of AOD policies at workplaces (Gábor and Kudász, 2022; ILO, 2003). They regard AOD policies as a part of workplace health promotion (WHP). Because WHP operates at multiple levels and provides a holistic perspective on health promotion, it can be structured using the socio-ecological model (SEM), as refined by McLeroy (1988), as the framework for prevention and health promotion in five levels of analysis. For the approach to alcohol prevention in the workplace, Sundqvist et al. (2025) suggests a four-level analysis: intrapersonal (individual characteristics, attitudes and behaviors), interpersonal (e.g. relationships between colleagues, supervisors and teams), organizational (policies, guidelines, culture and norms) and societal level (cultural contexts that shape individuals' beliefs and behaviors) (Sundqvist et al., 2025).

The Luxembourg Declaration (ENWHP, 2018) defines the WHP as the joint efforts of employers, employees and society to improve the health and well-being at work. It is achieved by fostering a collaborative work culture, organizing work in a health-supportive way, making healthy choices easier, encouraging employees' active participation and supporting their personal development. The WHP is based on collaboration across sectors and disciplines and requires commitment from all stakeholders, integration of all organizational levels, systematic planning and evaluation and balanced approach combining individual- and environment- focused measures (ENWHP, 2018). At present, workplaces face higher levels of stress and mental health challenges, alongside increased digitalization, hybrid working and an ageing workforce. These developments demand greater attention to protective factors and health potentials than before (ENWHP, 2024).

According to global and European level recommendations (Gábor and Kudász, 2022; ILO, 2003) AOD policies should cover SUD prevention, early identification and intervention, counselling and referral to the treatment, rehabilitation and support for employees with SUD in returning to the work. AOD policies are also recommended to support supervisors' (Griffith et al., 2021) and work communities' awareness and understanding of AOD-related harms and their ability to intervene (Alfred et al., 2021). The integrative review of Alfred et al. (2021) revealed that 40% of workplaces did not have AOD policies and underlined need for studies on reasons behind it.

Previous research has identified organizational, management, and individual-level factors that contribute to the successful implementation of AOD policies. According to the reviews of Alfred et al. (2021) and Giezek et al. (2026), AOD policies can reduce occupational safety risks if they are adopted at all levels of the organization and integrated into daily operational activities. AOD policies have been found to be associated with lower levels of drug use and misuse of the prescription medications (Hoopsick and Samad, 2024), and when combined with employee support services, with lower rates illicit drug use (Oh, 2023). Comprehensive AOD policies (Cooper and Bixler, 2021; Park and Minnick, 2024; Pidd et al., 2016), which include components such as prevention, education, alcohol and drug testing, support services and clear disciplinary measures, have been found to significantly reduce various forms of SUD across most employment sectors. They also increase awareness of the alcohol policy and employee assistance services (Elling et al., 2023; Pidd et al., 2018), as well as the likelihood of help-seeking and positive attitudes toward AOD policies among employees (Alfred et al., 2021). However, these findings of reduced consumption should be interpreted with caution, as studies using randomized controlled trial (RCT) and trial designs (Elling et al., 2023; Pidd et al., 2018) show opposite results, and consumption is typically self-reported (Alfred et al., 2021). Furthermore, Pidd et al. (2016) and Alfred et al. (2021) found that AOD policies were not significantly associated with work absences due to AOD, working under the influence or consuming AOD at work. These findings highlight the need for further intervention studies with longer follow-up periods, as well as studies focused on the development and implementation of comprehensive programs for employees. According to the review (Giezek et al., 2026), effective prevention for alcohol dependence should include brief motivational interventions and use of modern technologies such as mobile applications and online self-monitoring programs, as well as employer and supervisor support, accessibility and anonymity of the individualized and tailored programs and economic incentives.

Collaboration between middle and lower-level managers and their employees has been found to improve the dissemination of AOD policies and, subsequently, their implementation through organization (Elling et al., 2022b). Managers' knowledge of AOD policy and higher number of employees are strongly associated with early alcohol interventions (Alfred et al., 2021; Sundqvist et al., 2025). Having only a policy statement is not efficient; however clear guidance from supervisors on how to act on colleagues' harmful alcohol use increases perceived responsibility to intervene (Sundqvist et al., 2025).

Deficiencies in clear guidelines and lack of enforcement of AOD policies have been associated with higher levels of alcohol consumption among employees who have easy access to substances and in workplaces where the norms regarding substance use are liberal (Giezek et al., 2026). The dissemination and implementation of AOD policies have been found to be impeded by managers' uncertainty about the process and by the involvement of multiple hierarchical levels. Furthermore, limited time, scarce resources, lack of communication between stakeholders, lack of motivation and unawareness of responsibilities also act as barriers to dissemination (Elling et al., 2022b).

The work community plays an important role in identifying SUD (Salani et al., 2022) and unsafe practices resulting from substance misuse (Blair et al., 2016, 2021). However, although members of work communities tend to be willing to identify cases of SUD, they lack confidence in doing so (Trinkoff et al., 2021). Some consider intervention futile or fear the risks it may present (DesRoches, 2010), leading to cases going unrecorded (Weenink et al., 2015). Supervisors have a duty to intervene (Garcia, 2023), but require access to training in addressing SUD-related risks, minimizing associated stigma (Roche et al., 2019) and providing support (Alfred et al., 2021).

This study focuses on AOD policies in healthcare workplaces. While prevalence of the risky substance use with severe outcomes has been found to be higher in other industries than health care, one of the largest groups of workers exhibiting this risk is found within healthcare sector (Di Censo et al., 2025). Nurses and physicians constitute the largest group of health care professionals, and their work environment predisposes them to risk of substance misuse. Nurses who have easy access to medicines may divert prescription drugs (Kim et al., 2024; Trinkoff et al., 2022a) and physicians may self-prescribe addictive medications (Hartnett et al., 2020). Nurses tend to misuse alcohol and opioids (Cares et al., 2015; Mumba et al., 2019), while physicians most commonly misuse alcohol and sedative, hypnotic or anxiolytic substances (Geuijen et al., 2023). The prevalence of SUD among nurses has been reported to be comparable to that of the general population (Trinkoff et al., 2022b) and that of physicians (Merlo et al., 2022), but it varies between studies based on the method of estimation (Mumba et al., 2019; Wilson et al., 2022), the practice environment (Trinkoff et al., 2022b) and the substances under consideration (Edvardsen et al., 2014).

Previous studies on AOD policies have mainly focused on the manufacturing industry and the general working population. Although, some of these studies include healthcare sector, their results are not addressed separately (Park and Minnick, 2024; Pidd et al., 2016; Roche et al., 2023; Sundqvist et al., 2025), indicating a lack of research focusing specifically on AOD policies in health care settings. Research on AOD policies in healthcare workplaces is important on two levels (Cardno, 2018; Dalglish et al., 2020). First, information is needed on how policy content corresponds to global and European recommendations, and second, on their implementability at organizational level: do managers receive effective and clear operational models for policy implementation to safeguard patients, staff, the organization and the community equally?

The aim of this study was to describe and evaluate AOD policies for health care workplaces by answering two central research questions:

1. How comprehensive are the contents of AOD policies?

2. What is the quality and implementability of AOD policies?

A deductive document analysis (Moilanen et al., 2022) of AOD policies, following the principles of general policy analysis (Cardno, 2018; Dalglish et al., 2020), and including systematic gray literature searches (Godin et al., 2015) was performed. The first step in the analytical process was to define the research question based on preliminary literature searches, which led to the identification of a knowledge gap relating to the description and evaluation of the AOD policies in the health care context.

The second phase of the document analysis process (Cardno, 2018; Dalglish et al., 2020; Moilanen et al., 2022) involved identifying and collecting data (Figure 1). Because AOD policies in health care are generally publicly available in organization- or system-level documents, we searched the gray literature systematically (Godin et al., 2015) using four search strategies with pre-defined inclusion and exclusion criteria (Page et al., 2021). The first search strategy included six international search tools or databases: (1) The Gray Matters – A Practical Search Tool for Evidence-Based Medicine, (2) The Healthcare Management Information Consortium (HMIC) database, (3) OpenGrey, (4) PsycEXTRA, (5) Wonder and (6) Mednar. The Finnish Medic database was also used. The second search strategy was based on iterative customized Google searches. Two search queries were formulated in English to obtain AOD policies outside Finland, and Finnish translations of them to identify policies from Finnish sources. In both cases, the first 100 results were screened for relevant hits. All webpages linked within the retrieved documents were screened to determine the context of the policies in each organization's procedures. The third search strategy involved browsing selected websites (Godin et al., 2015) of three nursing associations and 16 university hospitals in Scandinavia and of 21 Finnish healthcare organizations. The fourth search strategy relied on expert consultation (Godin et al., 2015). The researcher (KL) contacted people in charge from 19 selected health care organizations whose AOD policies did not appear to be publicly available and asked them to provide information.

The third and fourth phases of the document analysis involved selecting and testing tools (Cardno, 2018; Dalglish et al., 2020; Moilanen et al., 2022). No relevant tool for evaluating deductively comprehensiveness of AOD policies was available, so we developed a new one, Alcohol and Other Drug Policy Evaluation (AODPE) tool (Table 1). It is based on AOD policy recommendations from three sources: the International Labor Office's policy statement (12 items) (ILO, 2003, 39–41), the European Occupational Safety and Health Agency's basic principles for AOD policies of the European Agency for Safety and Health at Work (14 items) (Gábor and Kudász, 2022) and the Model for AOD policy of the Finnish Institute of Occupational Health (five items) (FIOH, 2024). The contents of the recommendations were extracted, and based on their similarities and differences, they were categorized in 13 topics and sixty (n = 60) criteria. The fulfillment of each criterion was scored as “fulfilled” (yes), “not fulfilled” (no), or “not applicable” (n/a). The AODPE tool was pilot tested (Moilanen et al., 2022) by applying it to seven AOD policies, leading to the modification of 11 of the initial criteria. The development and pilot testing of the tool were performed by two researchers working collaboratively (AH-L, KL). The constant comparison method was used in all phases and changes were made only after reaching consensus.

To evaluate the quality and implementability of AOD policies, we used The Appraisal of Guidelines for Research and Evaluation–Health Systems (AGREE-HS) tool (Brouwers et al., 2019). This tool examines five (n = 5) quality domains that each have definitions and quality criteria (Table 2). Policies are assessed in each domain using a 7-point scale ranging from 1 (lowest quality) to 7 (highest quality). The use of the AGREE-HS tool was tested on two policies to verify that the items could be applied and interpreted similarly. The use of this tool was piloted and tested by two researchers (AH-L, KL).

In fifth phase of the document analysis (Cardno, 2018; Dalglish et al., 2020; Moilanen et al., 2022), policy documents were exported to an Excel spreadsheet listing their titles, dates and sources. Policies were then read several times to get an overview of their content, after which they were evaluated with the two tools described above. Evaluation with the AODPE tool generated a score for each of its 13 topics and a summed score (expressed as a percentage) indicating overall comprehensiveness of the policy. Overall scores of <50%, 50–75% and >75% were classified as poor, moderate and good, respectively. Each topic (n = 13) was assessed separately based on the fulfillment of its subordinate criteria (min 1, max. 16) and is reported as both a number (n) and a percentage (%), indicating the number and proportion of policies satisfying that criterion (Table 1).

In AGREE-HS tool (Table 2) the maximum and minimum possible scores for each item were seven and one, respectively. The scores for the individual items were then summed up to give an overall score for each policy, for which the maximum and minimum values were 35 and 5, respectively. Finally, the overall score was expressed as a percentage by comparing it to the maximum possible overall score. The final scores are not weighted. However, in accordance with the suggestion in the AGREE-HS manual (AGREE-HS Research Team, 2018), scores of <30%, 30%–70% and >70% were considered indicative of low, moderate and high quality, respectively. Two researchers (AH-L, KL) independently scored the items, and the resulting scores were compared. Differences were then resolved through joint discussion to reach consensus on the final scores.

The data consisted of twenty-five AOD policies from Australia, Canada, Finland, the United Kingdom and the USA. Twenty-two of the policies were created for territorial social and health care organizations and three for hospitals. Policies were found via Google searches (n = 9), targeted website analysis (n = 1) and expert consultation (n = 15). Nineteen were published between the years 2021 and 2024, four between the years 2015 and 2020, and two were undated despite being in current use.

Collectively, the policies (n = 25) comprised 344 pages of text and diagrams, with the lengths of individual policies ranging from six to 39 pages (Mean = 13.8). Eleven policies included appendices; the number per policy ranged from 1 to 10 (mean 3.4) and the total number of appendices was 40. The appendices included instructions on how to act when an employee is intoxicated at work or impaired due to suspected drug use, as well as operating protocols. Other material in appendices included forms for discussions, treatment requests, treatment guidance agreements or plans, warning forms, information on drug testing and details of relevant laws and agreements.

Based on the comprehensiveness evaluation, four of the AOD policies were of poor level (satisfying fewer than 50% of the AODPE criteria; min 22, max 29) while 21 were of moderate level (satisfying 50–75% of the criteria; min 30, max 44). The topics with the highest average percentage of satisfied criteria across all policies were coverage, confidentiality and means from SUD identification to treatment referral (including prevention/risk assessment and assessing/counseling/treatment). The topics with the lowest proportions of satisfied criteria were policy development and monitoring/reviewing (Table 1). All policies claimed to provide models or guidance for dealing with substance related harms in the workplace, promoting employees' health and ability to work, preventing their impairment and supporting a healthy and safe workplace. Three policies also claimed to support work productivity, and two to minimize costs resulting from employees' substance abuse.

Over half (n = 16) of the policies covered preventing the harms of substance abuse and/or gambling to problem identification, early intervention, referral to treatment, follow-up and statements of policy violations. The other policies lacked at least some of these phases. Policies commonly lacked clear or adequate descriptions of employment and advancement opportunities for employees who had completed treatment and rehabilitation.

Almost all (n = 24) of the policies covered all organizational members, and over half (n = 16) non-permanent stakeholders such as students (n = 10), contractors (n = 8), temporary workers (n = 5), civil servants (n = 4) and volunteers (n = 6) and/or visitors (n = 2). Additionally, one covered patients and five relatives. Twenty-one of the policies were applicable to remote work or employees on call, and ten to employees who were on business trips or attending remote meetings and educational events.

All policies specified what substances they cover. These were generally alcohol, drugs, illegal drugs, volatile mind- or function-altering and controlled substances and anabolic steroids and prescription and over the counter medicines. Ten policies provided more detailed definitions of them. Nine policies also covered tobacco and nicotine products, five included gambling and two addressed web addiction.

Most of the policies (n = 22) addressed substance use in the workplace, some of them using the expression “zero tolerance”. Over half (n = 14) clearly forbade bringing substances into the workplace and possessing or selling substances there. Eight prohibited the use of intoxicating substances during working hours. Six policies explicitly permitted the use of legal medications/drugs as well as prescribed and over the counter medications with appropriate supervision and consideration of their impact on work performance. Almost half (n = 11) of the policies provided guidance on alcohol intake in the organization's facilities or during events: five prohibited the provision of alcohol even as a raffle prize, while the rest permitted deviations during special events. Nine policies recommended serving non-alcoholic drinks during organizational events.

One-third (n = 9) of the policies outlined a value basis for AOD work. It included open discussions concerning alcohol and drug abuse, the courage to raise concerns at an early stage, joint responsibility and care for employees with abuse problems. Principles commonly expressed were: (1) no one should be allowed to work under the influence of substances or while hung over, and (2) every employee has primary responsibility for their own work ability and the implementation of the policy. Several policies stated that identified substance abuse should be dealt with promptly, effectively and confidentially in an equal, fair and supportive manner. Every phase of the subsequent process should be thoroughly documented, and the problems should be considered objectively, discreetly and in accordance with jointly agreed procedures regardless of the personal features of the employee or the substances that were used. Early intervention was portrayed as caring and withdrawing from it as negligence.

Nearly one-third (n = 8) of the policies specifically highlighted the need for an employee-centered approach when managing employee SUD. This included respectful encounters and acknowledgement that such employees may not recognize their situation and may experience suffering. Five policies described SUD as an illness or a condition that needs treatment and from which one may recover. Employees should not be left alone, and discussions with them should be conducted in a supportive, non-judgmental manner to minimize the likelihood of concealment. It was also stated that every employee has the right to get help, and that every phase of the policy was intended to improve and restore the employee's health and ability to work.

Twenty of the policies stated that personal information pertaining to the employee's situation and referral to treatment or occupational health care (OHC) visits should be handled confidentially. Eighteen of the policies discussed relevant data protection regulations, and nine stated that no personal information should be disclosed to outsiders without the permission of the individual concerned.

Almost all (n = 23) of the policies defined goals for AOD-related work as aims in line with the content and rationale of the policy. These were described in broad terms and lacked concreteness and scheduling. Goals were to support employees in avoiding risky use of alcohol, identify substance abuse, create a culture where staff feels confident in seeking advice and support and to raise awareness of substance abuse and its effects on work performance. None of the policies included specific measures to achieve these goals or described procedures for evaluating progress towards the goals.

Stakeholders' roles, responsibilities and rights (Table 3) in the SUD intervention process were described with varying exactness. The most frequently discussed roles were those of supervisors and OHC personnel, whereas the least frequently mentioned (n = 3) was that of substance abuse liaison. Eighteen of the policies stated that employees with SUD have the right to demonstrate sobriety with testing, to have a support person present at hearings, to be referred for treatment negotiation and to be heard before termination of employment. Six policies discussed supervisors' rights, including the right to discuss substance misuse with employees and to be informed about whether a treatment and rehabilitation plan has been implemented in accordance with the agreement.

Almost all of the policies (n = 23) mentioned that employers have a responsibility to inform and educate staff about the AOD policy, but only five described this responsibility in detail. In relation to the treatment process, most of the policies described the roles of the employee (n = 24), supervisor (n = 22) and OHC personnel (n = 20), highlighting employees' self-referral, voluntariness, activity and commitment. The supervisor's and OHC's role were described in terms of concrete responsibilities. Fourteen policies mentioned representatives of occupational health and safety, professional organizations and unions, staff, relatives, friends and shop stewards as potential support persons without providing role descriptions. The role of the occupational safety and health (OSH) was described in ten policies, mainly in relation to employees' requests for treatment negotiation (Table 3).

Almost all (n = 23) of the policies highlighted identifying risk signs in employees' behavior and work performance as the first step in prevention. The most detailed policy provided a list of key signs including physical, psychological and social indicators; the policy with the briefest description only a single sentence discussing impaired performance. The majority (n = 19) of the policies detailed the situations in which testing for alcohol and/or drugs is permitted, supporting their positioning with reference to legislation. These included situations in which an employee was applying for a position, starting work, suspected of drug use, undergoing an assessment of work ability or having work tasks changed during rehabilitation. Alcohol testing using a breathalyzer was permitted in acute situations at the workplace, in OHC or by another medical implementer. Over half (n = 16) of the policies discussed preventing harm caused by SUD ranging from extensive lists of procedures to individual statements. The most frequently mentioned preventive actions were education, coaching and fostering an open and friendly discussion atmosphere when addressing SUD-related issues. Almost all (n = 24) of the policies provided guidance on SUD identification, including first-stage action instructions for different stakeholders, as well as specific models and processes for early intervention. The briefest discussion of early intervention stated only that action must be initiated when SUD is detected. Over half (n = 17) of the policies described how to consider the financial consequences such as changes in income and social benefits, as well as additional treatment costs for employees undergoing SUD treatment.

Half of the policies (n = 14) suggested support for daily work and joint meetings with stakeholders to discuss treatment progress for employees undergoing SUD treatment and rehabilitation. Employers were seen as the main stakeholders responsible for organizing support in re-entering work. Some policies (n = 9) stated that an employee's tasks and duties should be re-assessed when they return to their post. Additionally, some policies discussed prospects of promotion and alternative employment for such employees. Two policies highlighted the need for individualized support in the event of relapse. Statements concerning employment and advancement opportunities for workers with previous SUD or who had undergone successful counselling and treatment were scarce.

The work community was expected to accept employees undergoing treatment as equals, support their coping and recovery, and act with propriety. Work community support was also considered important in nine policies when employees re-enter to work. Almost half of the policies (n = 11) underlined that the supervisor, OHC or other relevant stakeholders should organize the handling of the employee's SUD in the work community. In five policies, the employee's supervisor was named as responsible for providing support, and in two, debriefing the work community (Table 3). The importance of securing the employee's consent when handling their return to work and informing the work community was also emphasized.

All policies justified disciplinary actions as the consequences of violating the policy. More than half of the policies (n = 16) provided detailed disciplinary procedures including discussion with a supervisor, notification or written warnings and immediate termination of the employment contract. Referral to treatment was mentioned but not defined as a disciplinary action. Thirteen policies provided guidance on notifying regulatory bodies of employment termination due to SUD.

Half of the policies (n = 12) described monitoring and evaluating policy implementation and seven updating the policy's content as a duty of the employer or its representatives (Table 3). Five policies stated that OHC personnel were responsible for reporting occupational data; three stated that OSH personnel should oversee procedures for preventing harm caused by substance abuse and two stated that human resources (HR) personnel were responsible for managing data concerning substance misuse. Four policies included annual reviewing while two for every two or three years, including partial updating of occupations subject to drug testing.

Over half of the policies (n = 14) were explicitly linked to occupational safety and health care, workplace health promotion and/or HR policies. Twelve of the policies stated management's commitment to the policy and indicated that it was based on joint negotiation between the employer and employees, a management group or regional authorities. Concrete descriptions of the processes through which the policies were collaboratively developed were not provided; however, three of the policies mentioned the involvement of OHC and/or OSH.

Evaluations using the AGREE-HS tool indicated that the AOD policies had quality scores in the moderate range (30–53%). Seven had scores of 30–39%, 17 had scores of 40–50% and one scored 53%. The item with the highest scores was the description of the policy's rationale (Topic). The items with the next highest scores were policy operationalization (Recommendations), followed by precise and concrete specification of implementation procedures (Implementability). The items with the lowest scores related to the evidence base of the policies (Methods). Although the AOD policies were based on legislation, recommendations and instructions, none were described as evidence-based. The item with the lowest score was Participants, for which most policies provided no information whatsoever (Table 2).

This study provides insights into the comprehensiveness, quality and implementability of AOD policies in healthcare, showing that they all were, at best, on moderate level. In addition, a new AODPE tool for evaluating the comprehensiveness of policies was developed (FIOH, 2024; Gábor and Kudász, 2022; ILO, 2003). The policies revealed clear deficiencies across two evaluation levels (Cardno, 2018; Dalglish et al., 2020). The policies did not fully correspond global (ILO, 2003), European (Gábor and Kudász, 2022) or national (FIOH, 2024) recommendations, and they did not provide enough concrete advice to managers for policy implementation. These deficiencies may lead to ineffective and variable implementation and create inequalities in the treatment of employees with SUD (Geuijen et al., 2023).

Results showed that the policies generally did not provide comprehensive coverage of the AOD process, from the identification of problems to returning to work, nor did they include specific measures to achieve or evaluate their goals. These content-related deficiencies may hinder the effective implementation of the policies (Alfred et al., 2021; Cooper and Bixler, 2021; Park and Minnick, 2024; Pidd et al., 2016). Lack of precision in the policies increases uncertainty among managers and employees regarding appropriate procedures, as well as unawareness of roles and responsibilities of relevant stakeholders (Elling et al., 2022a). Furthermore, the prevention of SUD was described in an abstract and narrow manner, and one-third of the policies did not address it at all. This finding is alarming, as it does not align with the current paradigm. ILO (2003) has already emphasized few decades ago that AOD policy paradigm should shift toward prevention rather than reactive approach.

The weak role of prevention was also evident in the consideration of tobacco and nicotine products. Only one-third of the AOD policies addressed them. These products are also addictive in nature, pose a significant health risk and can lead to illnesses that reduce work ability (Le Foll et al., 2022). In healthcare, all AOD policies should therefore ensure smoke-free environments for both patients and employees (Frazer et al., 2016). In recent years, new addictive behaviors at workplaces have emerged such as online gambling and excessive use of internet and social media (ENWHP, 2024). These issues should also be more thoroughly acknowledged when updating AOD policies. These shortcomings in the policies may be due to a limited understanding of organizational responsibilities in SUD prevention or an insufficient knowledge base for prevention. Most of the policies described identification procedures and risk signs, including changes in the employees' behavior and work performance due to SUD. However, only one document discussed diversion of controlled substances, even though this is commonly associated with nurses' SUD (New, 2014) and can have serious consequences for patient safety and the organization (Clark et al., 2022).

Supervisors were named as being primarily responsible for policy implementation, and their roles were described in more detail. This is necessary to support supervisors in their demanding processes of identifying and intervening in SUD (Alfred et al., 2021; Elling et al., 2022a; Sundqvist et al., 2025). Comparatively, few acknowledged the role of work communities, although previous studies have shown that work communities play key roles in identifying SUD and in the process of re-entry. Unfortunately, the members of these communities are often uncertain about intervention strategies and consider SUD difficult to discuss. Therefore, both supervisor and work community training and education should be provided regularly (Cadiz et al., 2012; Trinkoff et al., 2021). Training and education have been found to improve the ability of work communities to identify SUD risk signs (New, 2014; Rice, 2023), change attitudes toward SUD and strengthen related skills, confidence and knowledge (Cadiz et al., 2012; Cadiz et al., 2015a; Elling et al., 2022a; Pidd et al., 2018). It is also notable that only ten policies included students, even though research has shown that SUD often begins during studentship (Molloy, 2024; Strobbe and Crowley, 2017). In future, more attention should be paid to students in AOD policies to ensure early intervention and prevent later escalation of SUD (Ayala et al., 2017; Ruth-Sahd and Schneider, 2022).

Only nine policies considered return-to-work support measures. This is not consistent with existing guidelines (FIOH, 2024; Gábor and Kudász, 2022) and warrants greater attention in future. The use of alcohol and drugs is a sensitive, complex and stigmatized issue, which may hinder re-entry to work after SUD treatment. SUD affects not only employees who experience it directly but also community members, creating a need for support in managing its consequences. However, support provided by and to work communities must align with the wishes of the person with SUD. Regularly scheduled continuing education on SUD has also been recommended to change workplace culture and reduce the stigmatization of SUD (Matthias-Anderson and Yurkovich, 2016).

All organizational policies are value-based documents. Human, democratic and economic values should be stated explicitly to ensure that everyone in organization is treated equally (Cardno, 2018). This study showed that such explicit value statements were missing from most AOD policies. Only a few policies had an employee-centered approach; five treated SUD as an illness and one-third highlighted principles of empathetic treatment and support for employees. In the future, value base should be explicitly described in every AOD policy. More attention should be paid to ethical principles when designing AOD policies because of the sensitive nature of SUD.

The policies generally provided limited descriptions of the procedures used in their development and the organizations' commitment to them, and there was no clear indication of strong collaboration with employees, their representatives or subject expert stakeholders (FIOH, 2024; Gábor and Kudász, 2022; ILO, 2003) highlighting the need for further development. Transparency in the policy development could be enhanced by carefully documenting joint negotiations among all stakeholders and the processes used to ensure management commitment. Management commitment strengthens the credibility of the policy and support responsibility in policy implementation (Alfred et al., 2021; Sundqvist et al., 2025). Employers should be responsible for updating policies to ensure that they remain relevant (FIOH, 2024), but this was not evident in this study. Employees should also be responsive to feedback, to organizational changes and to AOD-related needs, and policies should be revised accordingly. It may be useful for organizations to evaluate the adoption and suitability of their policies in daily practice, for example, by surveys or interviews (Cardno, 2018; Dalglish et al., 2020).

To ensure a preventive approach in workplace AOD policies (Gábor and Kudász, 2022; ILO, 2003), policies in health care should be more aligned with holistic and person-centered workplace health promotion (WHP) (ENWHP, 2018). In healthcare, employees' personal, professional and contextual factors at multiple levels have been found to contribute to SUD (Mercer et al., 2023; Ross et al., 2018). To address these in the design and implementation of the comprehensive preventive AOD policy, the socio-ecological model (SEM) (McLeroy et al., 1988) may be a useful framework to apply (Sundqvist et al., 2025).

The results of this study indicate that AOD policies were not based on current research evidence. Furthermore, they did not include descriptions of the effectiveness and costs of recommended methods (AGREE-HS Research Team, 2018). These findings highlight the need to strengthen the theoretical foundations of the policies in line with recommendations (Gábor and Kudász, 2022; ILO, 2003) and earlier literature (Alfred et al., 2021; Pidd et al., 2016). AOD policy should incorporate evidence-based tools for prevention, intervention and referral to the treatment, and these should be reported transparently.

The newly developed AODPE tool had good feasibility, and it may be useful in guiding AOD policy development in health care organizations. In future, AODPE tool should be tested with larger data sets to confirm its structural validity and condense its content.

The rigor of the analysis in the sixth, integrated, phase (Cardno, 2018; Dalglish et al., 2020; Moilanen et al., 2022) was ensured by assessing both data bias (relating to both purpose and selection) and author bias (in data interpretation, evaluation and reporting). The selection of data for inclusion was strengthened by performing a systematic search using four search strategies and defining clear inclusion and exclusion criteria. Despite this, no international AOD policies were found via gray literature databases or from targeted websites, suggesting that AOD policy documents have poor public availability and are for organizational use only. Mainstream databases (e.g. PubMed and CINAHL) were not used in the search strategy because they rarely include policy texts (Godin et al., 2015). The AOD policies that were retrieved originated from a relatively limited range of countries. This may be due to the use of English and Finnish as search languages. The AOD policies selected for inclusion were not created for research purposes, but their content was consistent with the aims of research questions of the study.

A limitation of this work is that it used a tool that was made for this study and has not yet been validated. However, its content validity is likely to be high because it was based on relevant global, regional and national authorities' guidance on AOD policy development (FIOH, 2024; Gábor and Kudász, 2022). The ILO (2003) guidelines can be considered old but are the latest available version according to the WHO library and were thus considered relevant. It should be noted that after being developed, the tool was piloted by analyzing seven AOD policy texts and refined on the basis of this preliminary analysis. Policy quality and implementability were assessed using the validated AGREE-HS tool (Brouwers et al., 2019), which was piloted by applying it to two AOD policies. The risk of interpretation bias was minimized by the researchers' dialogue and by their understanding of the health care context. Furthermore, the policy assessments were performed by two researchers independently and double-checked.

The seventh phase of the integrated document analysis procedure (Cardno, 2018; Dalglish et al., 2020; Moilanen et al., 2022) involves verifying compliance with research ethics. This was done by strictly following ethical research principles (ALLEA, 2023). The data examined in this work did not require an ethical review statement from a human sciences ethics committee. Some policy texts were obtained by contacting experts; this process was subject to organizational approval at multiple levels. Research permission was needed for the use of two policies. All policies (n = 25) were processed only by two researchers (AH-L, KL) and were coded for handling and reported anonymously to protect stakeholders' privacy (Moilanen et al., 2022).

Assessments of existing AOD policies using two tools revealed several developmental issues relating to their content, quality and implementability. Closer attention to the preparation of such policies is needed to strengthen their contributions to organizational guidance and their alignment with national and global guidelines. In particular, there is a need for clear descriptions of the policy development processes and its stakeholders, engagement of leaders, more concrete statements and a clearly articulated evidence base. Policies should have comprehensive content covering everything from SUD prevention to work re-entry for employees after treatment and should acknowledge the values and ethical principles of care as well as the challenging situation of employees with SUD. Additionally, the role of the work community should be clearly stated and merits further study. Future research should focus on identifying cost-effective interventions for dealing with SUD and ways in which they can be implemented and evaluated in the health care workplaces. There is also a need to study design processes for AOD policies and ways of supporting them. Evidence-based and regular awareness activities should be developed and studied to support the implementation of workplace AOD policy and SUD prevention in health care workplaces.