Understanding How and Why Alcohol Interventions Prevent and Reduce Problematic Alcohol Consumption among Older Adults: A Systematic Review

Problematic alcohol use has been increasing in older adults (55+) in recent decades. Many of the effective interventions that are available to prevent or reduce the negative effects of alcohol consumption are aimed at adults in general. It is unclear whether these interventions also work for older adults. The objective of this review was to understand how (i.e., which elements), in which context, and why (which mechanisms) interventions are successful in preventing or reducing (problematic) alcohol consumption among older adults. A systematic review of articles published between 2000 and 2022 was performed using PubMed, PsycINFO, Web of Science and CHINAHL. Realist evaluation was used to analyze the data. We found 61 studies on interventions aimed at preventing or reducing problematic alcohol use. Most of the interventions were not specifically designed for older adults but also included older adults. The findings of the current study highlight three major effective elements of interventions: (1) providing information on the consequences of alcohol consumption; (2) being in contact with others and communicating with them about (alcohol) problems; and (3) personalized feedback about drinking behavior. Two of these elements were also used in the interventions especially designed for older adults. Being in contact with others and communicating with them about (alcohol) problems is an important element to pay attention to for developers of alcohol interventions for older adults because loneliness is a problem for this age group and there is a relationship between the use of alcohol and loneliness.


Introduction
The World Health Organization has identified alcohol-related harm among older adults as an increasing concern [1]. Researchers in biology often define old age as starting at the chronological age of 55+ because at that age changes in body systems become more evident [2,3].
Over recent decades, alcohol use among older people has increased in several countries, including Spain, the United States, and The Netherlands [4][5][6]. One reason for this is that older adults experience more freedom; that is, they have more time for leisure activities, such as attending social gatherings and participating in clubs, many of which routinely involve alcohol consumption [7,8]. Another reason is that some people use alcohol as a coping strategy to overcome negative changes in physical health and mental health that come with ageing [9][10][11], including increased loneliness and social isolation [12], unemployment and economic downturns [13]. In developing countries, alcohol use has been increasing in line with economic development and global marketing. The increased availability and affordability of alcohol among lower socio-economic groups has also played a role [1,14]. For developing countries, this is in line with increasing economic development, global marketing and the greater availability and affordability of alcohol among lower socio-economic groups [1,14].
The consumption of alcohol, even in small amounts, can cause greater harm among older than younger adults. Alcohol can accelerate and aggravate the onset of conditions associated with aging (e.g., falling hazards [15], cognitive impairment [16] and/or sleep disturbance [17][18][19]). Older adults often receive medication for these conditions. The use of alcohol alongside prescription medications, such as benzodiazepines for insomnia [20], leads to negative interactions, particularly because older adults metabolize and excrete alcohol more slowly [21,22]. The combination of alcohol and prescription medicines could lead to increasing alcohol levels in the blood, reducing the efficacy of medication and exacerbating its side effects [9,23].
The development of effective interventions to prevent or reduce alcohol use in older adults is crucial, not only because of the problems that alcohol consumption causes for older individuals, but also because of the increase in the number of older people. It is expected that, in 2050, one in six people worldwide will be aged 65 years or over [24]. As the population of older adults increases, so will the number of older people who have alcohol-related problems. First, prevention is needed because of the problems that alcohol causes in this group; second, the group of older adults is expanding, resulting in more alcohol-related problems.
Numerous effective interventions have been developed to prevent or reduce alcohol consumption, for example interventions carried out by general practitioners, brief interventions [25,26], psychosocial interventions (e.g., motivational interviewing [27]) and e-health interventions (e.g., web-based interventions [28] and smart phone interventions [29]). However, many of these interventions are aimed at adults in general and not specifically at older adults. It is unclear whether these interventions also work for older adults. Older adults were raised in a different period and may have different norms and values regarding drinking alcohol than young adults [30,31].
Two recent reviews on alcohol consumption among older adults have indicated that interventions to prevent or reduce the negative effects of alcohol consumption in older adults specifically are limited in number. Armstrong-Moore et al. [32] found seven interventions, of which five resulted in alcohol reduction. Kelly et al. [33] identified thirteen studies, of which six reduced alcohol consumption. Most effective interventions include elements of (brief) motivational interventions, (brief) advice or personalized reports on risks and problems. Moreover, it is known only whether the interventions are effective and not which elements of the intervention lead to this outcome or in which context and by which mechanisms. With this information more targeted interventions for alcohol use among older adults could be developed.

Objectives
To date, no overview is available showing how interventions to prevent or reduce alcohol use in older adults work and which are successful for older adults specifically. Therefore, we performed a literature review, following a realist approach, on interventions for (older) adults and extracted the elements of the interventions that were effective. When possible, we also explained why these elements were effective. To understand why an (element of an) intervention leads to the desired outcome, it is also important to understand the context in which the intervention is offered to the target group. Therefore, we also took the context into account. The objective of this review is to understand how (which elements of interventions), in which context and why (by which mechanisms) interventions are successful in preventing or reducing (problematic) alcohol consumption among older adults.

Realist Evaluation Approach
This literature review was informed by the realist evaluation approach. A realist evaluation describes not only the intervention and its outcome (O) but also the context (C) and the underlying mechanism (M) [34]. The context includes elements such as the organizational context, participant features, staffing, and geographical and historical context. Mechanisms are a combination of recourses offered by the (social) program or intervention and human understanding and/or responses to that recourse. Mechanisms are not directly observable and include preferences, reasoning, norms or collective beliefs. Outcomes include changes to people and to their lives, but also include other kinds of alterations (e.g., in organizations, workers or governments) [35]. In the current study, the interventions are the programs that help older adults to prevent or reduce their alcohol consumption successfully, the context is operationalized as the way in which the intervention is offered to the target group (e.g., digitally, by phone, in-person, individually or in a group setting), the mechanisms are the reasons why elements of the interventions work and the outcome is the prevention or reduction of alcohol consumption.

Search Strategy
A literature review of peer-reviewed articles published between 2000 and 2020 was performed in April 2020 and updated in February 2022 using PsycINFO, Web of Science (WOS), PubMed and CINAHL (see Table 1). This time span was chosen because the focus on older adults in relation to alcohol issues dates back to the beginning of this century [36]. A combination of five groups of keywords was used to search the databases. These groups of keywords consisted of search terms from all four databases: PsycINFO (thesaurus), Web of Science (no special terms), PubMed (MeSH terms) and CINAHL (heading terms). In addition, synonyms and free text words were used. Four search strings were formed based on the objectives of this review. Due to the scarcity of studies specifically about older adults, we chose to include two groups: a very wide range, which includes older adults (18+), and a specific age range, which only consists of older adults (55+). Table 2 provides an overview of the groups and keywords used. Table 3 provides a summary of the search questions. The review is reported according to PRISMA [37]. Peer-reviewed, publication years 2000-2022, English, Dutch 1 1 No limits for age were provided. Therefore, the word group "Not child" was used. 2 The only age limit that was provided was 19+.

Inclusion Criteria
The studies were selected with the following inclusion criteria: (1) studies that focused on interventions for outpatients to prevent or reduce (problematic) alcohol consumption and that mentioned effective elements; (2) the target group of the studies consisted of people aged 18 years or older; (3) peer-reviewed empirical studies published in English after 2000 and available in full text; and (4) studies conducted in Western high-income countries (e.g., Europe, North America, Australia and New Zealand). The exclusion criteria were the following: (1) studies aimed at inpatients; and (2) studies with a very specific target group (i.e., pregnant women, veterans (due to the specific approach of this group, often carried out by the military), ethnic minorities, students, people with an IQ lower than 85 or the forensic target group).

Study Selection
Based on the inclusion and exclusion criteria, titles, abstracts and full-text articles were screened by the first author (JB). The last author (AR) also screened 20% of all the records (titles, abstracts and full texts). The deviation was less than 10% in all three phases. In the case that JB had doubts about other articles than these 20%, AR was consulted. See Figure 1 for the flowchart of the selection process. Table 2. Groups of key words.

Group 1 Alcohol
alcohol or "alcohol consumption" Group 2 Older adults elder or elderly or senior or old or pension or retire or retirement or "later life" or geriatric or geriatrics or "older adults" or ageing or aging or gerontology or aged

Group 4 Prevention
prevent or prevention or preventing

Group 5 Not child
NOT child or "young adult" or teenage or adolescent Table 3. Search strings.

Search Questions Groups of Keywords
What are the effective elements of interventions for the general population with regard to reducing (problematic) alcohol consumption? 1 (title) and 3 (title) not 5 (only for WOS (topic) and CINAHL (title)) What are the effective elements of interventions for older adults (55+) with regard to reducing (problematic) alcohol consumption? 1 (title) and 2 (title) and 3 (PsycINFO (abstract), WOS (topic), PubMed (title/abstract), CINAHL (abstract)) What are the effective elements of interventions for the general population with regard to the prevention of (problematic) alcohol consumption?

Data Extraction and Analysis
A data extraction form was used, specifying the following information: author(s), title, publication year, study methodology, setting, participants and objective of the study (prevention or reduction), effective elements of the intervention, context, mechanisms and outcome (CMO). Data were extracted by JB and AR independently and discussed thereafter until consensus was reached. A realist evaluation approach was adopted to identify CMO configurations in each study where possible. These configurations described how contextual factors and mechanisms (human responses to elements of interventions or prevention strategies) led to the desired outcomes (prevention or reduction of alcohol consumption). For each study, one or more elements of interventions and/or one or more CMO configurations were drafted. The analyses were performed by JB and AR, focusing on the patterns across elements of interventions and the CMO configurations. The quality of the included studies was assessed using the Mixed Methods Appraisal Tool (MMAT Tool) [38]. The tool includes two screening questions and 19 items for appraising the methodological quality of five categories of studies: qualitative studies, RCTs, non-randomized studies, quantitative descriptive studies and mixed-methods studies. Each study category consists of five items. Each item is rated on a categorical scale (yes, no and cannot tell). The number of items rated "yes" is counted to provide an overall score (0 is low; 5 is high). The appraisal of all the included articles was performed independently by two researchers (JB and AR), and the results were compared; when inconsistencies were apparent, they were discussed until consensus was reached. Studies with a low MMAT score (2 or lower) were only used to support the results found in studies with an MMAT score of 3 or higher.

Study Selection and Characteristics
We included 61 articles in our review. The characteristics of each included paper are presented in Table 4. A total of 33 studies were quantitative and randomized, 19 studies were quantitative and non-randomized, five studies were qualitative interviews, three studies were quantitative descriptive studies, one study was mixed methods. Of these 61 studies, three described interventions specifically for older adults and were quantitative and randomized studies [39][40][41]. The studies were performed in the following countries: the United States of America (25), the United Kingdom (7), Germany (4), Australia (5), Denmark (4), Canada (4), the Netherlands (4), Spain (2), France (2), Ireland (1), Italy (1), Estonia (1) and New Zealand (1).
The aforementioned three studies [39][40][41] that specifically targeted older adults focused on interventions with personalized feedback and information provision. Interventions for the general populations included therapy sessions, frequently including motivational interviews or motivational enhancement with other educational material. Some interventions offered a stepped care process with personalized feedback on alcohol. The way in which the interventions were delivered differed widely. Personal treatment and the Internet were the most mentioned ways.
The quality assessment results (MMAT score) are shown in Table 4. Overall, the quality of the studies was generally high (4 or 5) or moderate (3). Only three studies were rated low (2) [42,43] or poor (1) [44].                    Intervention: more aware of how much they routinely drink, and to make healthier choices.
(1) Providing information about health consequences (E); providing information about emotional consequences (E); encouraging self-monitoring of behavior (E); encouraging self-monitoring of outcomes of behavior (E) and encouraging behavioral experiments (E).
Appreciated the friendly, non-threatening tone and that the message was straightforward (M), meaningful, achievable (M), and was gainframed-i.e., emphasised the benefits of drinking less rather than the harms of drinking too much (M) → reduction of alcohol consumption (O)

Quantitative nonrandomized
Intervention: Two-week implementation intention interventions that linked high-risk situations with alternative responses Control: Two-week implementation intention interventions for selected situations and responses but did not link these together Intervention: (1) Cut back on drinking over the following two weeks (E) (2) and fill in an "if-then" worksheet format. Response (linked high-risk situations with alternative responses) (E) → with a significant reduction in alcohol consumption when drinking was reported (O) → more abstinent days (O) Control: To try to cut back on drinking over the following two weeks (E) and (3)

Qualitative interviews
Intervention: The participants were asked to abstain from drinking alcohol for 6 weeks, during which period they were to maintain their "normal" social behavior and obligated to keep a diary of their experiences with abstinence Intervention: (1) Abstain from drinking alcohol for 6 weeks (E), during which period participants were to maintain their "normal" social behavior (E) → the participants reporting the largest decrease in consumption were the persons reporting the highest initial consumption level (O) Intervention: (1) Abstain from drinking alcohol for 6 weeks (E), during which period the participants were to maintain their "normal" social behavior (E), producing increased awareness of the role of alcohol in their lives (M).
(2) Participants expressed more insights into their expectations of social gatherings and how to fulfil them (M) → the participants reporting the largest decrease in consumption were the persons reporting the highest initial consumption level (O).
(3) More participants reported that they now made conscious decisions about their alcohol consumption prior to participating in a social gathering and that they would feel more comfortable complying with those decisions (M) → some started to drink at a slower pace, and others started bringing their own water bottles (O) Intervention 1: Drinktest (online personalized feedback intervention) plus prototype alteration (feedback regarding prototype alteration tailored to gender, drinking behavior (also including normative feedback), intentions, and prototypical self-characterization) Intervention 2: Drinktest (online personalized feedback intervention) plus cue reminder Intervention 3: Drinktest (online personalized feedback intervention) plus prototype alteration and cue reminder Control: Original Drinktest (1) received feedback tailored to demographic background (gender drinking behavior (also including normative feedback), intentions, and prototypical self-characterization), alcohol consumption and intentions to reduce drinking. These messages reflected on personal drinking levels in comparison with the Dutch norm and peers' drinking behavior Intervention 1: (1) Received feedback tailored to gender, drinking behavior (also including normative feedback) (E), intentions and prototypical self-characterization; (2) the prototype message reflected on characteristics that the participants evaluated as personally desirable or undesirable by evaluating themselves on 11 characteristics (E); (3) participants were encouraged to reduce their drinking to achieve their desired characteristics and, in turn, to be positively valued by peers (E); (4) then, participants were guided in their goal setting by selecting an action plan to achieve the desired characteristics (E) → reduction of alcohol consumption (O) Intervention 2: (1) Received feedback tailored to demographic background (gender), alcohol consumption and intentions to reduce drinking. These messages reflected on personal drinking levels in comparison with the Dutch norm and peers' drinking behavior (E). Participants were guided in their goal setting by selecting an action plan to achieve the desired characteristics (E). (2) Feedback was provided that reflected on their action plans, explaining that a cue reminder may help them to remember their plans (E) (if made) and they received a free silicone bracelet by mail. If participants did not want to receive the bracelet, they were encouraged to select a piece of their own jewellery or another object of frequent use (E) → reduction of alcohol consumption (O) Not studied

Study Quality (MMAT)
Intervention 3: (1) Drinktest plus prototype alteration, cue reminder and feedback tailored to gender, drinking behavior (also including normative feedback) (E), intentions and prototypical self-characterization. (2) The prototype message reflected on characteristics that the participants evaluated as personally desirable or undesirable by evaluating themselves on 11 characteristics (E); (3) participants were encouraged to reduce their drinking to achieve their desired characteristics and, in turn, to be positively valued by their peers (E). (4) Participants were guided in their goal setting by selecting an action plan to achieve the desired characteristics (E). (5) Feedback was provided that reflected on their action plans, explaining that a cue reminder may help them to remember their plans (E) (if made) and they received a free silicone bracelet by mail. If participants did not want to receive the bracelet, they were encouraged to select a piece of their own jewellery or another object of frequent use (E) → reduction of alcohol consumption (O) Control group: Original Drinktest: (1) received feedback tailored to demographic background (gender), alcohol consumption and intentions to reduce drinking. These messages reflected on personal drinking levels in comparison with the Dutch norm and peers' drinking behavior → reduction in alcohol consumption (O)

Qualitative interviews
Aim: to inform recruitment and retention strategies by exploring users' motivations and experiences in using a novel, Internet intervention, the Hello Sunday Morning (HSM) program.

Quantitative nonrandomized
Intervention Hello Sunday Morning (HSM): An Australian social media health promotion "movement" that asks participants to set a personal goal publicly to stop drinking or reduce their consumption, for a set period of time, and to record their reflections and progress on blogs and social networks

A. Practitioner-inperson-individual
Paying attention to drinking behavior (1) motivational exercises to change behavior (2) pointing out the health disadvantages of drinking behavior (3) helping to develop networks Interventions make people think and act differently about alcohol consumption and seek help from family and friends

Less or no alcohol consumption
The relationship between the patient and practitioner (1) Empathic behavior of therapist Patient and practitioner collaborate in the identification of additional sessions, judged best to meet the patient's clinical needs and the relationship between the patient and the therapist improves Less or no alcohol consumption

C. Practitioner-inperson-relatives
The status of the relationship -Less or no alcohol consumption Teaching the partner to deal with drinking behavior When the non-drinking partner is taught to deal with the behavior of the drinking partner, this can lead to more understanding and support from the non-drinking partner for the drinking partner.

D. Practitioner-inperson-group component
Motivating to change lifestyle (1) regarding personal relationships, nutrition and exercise (2) and coping with desires for alcohol -Less or no alcohol consumption Motivating to change lifestyle delivered in a workplace setting (1) discussion of alcohol use and its consequences (2) training element to change behavior and reduce alcohol use (3) personal advice is given on alcohol use

E. No practitioner-not in-person-individual.
Web based interventions (1) personal feedback (2) comparing own results with others (same phase, age group, gender or country) (3) compared with the previous data of the participant. Mobile phone interventions could provide insight into how much someone drinks through the information provided and this leads to realization of their own drinking behavior

F. No practitioner-not in-person-group component
Intervention to abstinent people (1) Not drinking alcohol for a certain period or to drink less (2) share this experience with peers Intervention to abstinent people (with or without problematic drinking behavior (1) from drinking alcohol for a certain period or to drink less and (2) to share this experience with peers makes people aware of their alcohol consumption and reduces alcohol consumption.
Less or no alcohol consumption

Themes
We were interested in how (which elements of interventions), in which context and why (which mechanisms) interventions prevent or reduce (problematic) alcohol consumption among older adults. The results were first categorized according to their mode of delivery (i.e., the context): (1) practitioner or no practitioner involvement; (2) in-person or not; and (3) individual treatment, group treatment or treatment with relatives' involvement. Consequently, six different modes of delivery were found: (A) practitionerin-person-individual; (B) practitioner-not in-person-individual; (C) practitioner-inperson-relatives; (D) practitioner-in-person-group component; (E) no practitioner-not in-person-individual; and (F) no practitioner-not in-person-group component. Then, for every mode of delivery, one or more findings were provided about how (which elements of interventions) and, when found, why (by which mechanism) these elements contributed to the prevention and reduction of (problematic) alcohol consumption for (older) adults. Table 4 provides a summary of the studies' characteristics. Table 5 provides a summary of the results.

Paying attention to drinking behavior
From the treatments that were delivered by a practitioner, in-person and individually, four effective elements were present: (1) motivational exercises [53,56,60]; (2) pointing out the health disadvantages of drinking behavior [51,52,79]; (3) helping to develop networks [54,57,58,60]; or a combination of these approaches [46]. Paying attention to drinking behavior yields results. Interventions make people think and act differently about alcohol consumption [57] and seek help from family and friends [57,58]. In many studies, the drinking behavior of the control group also changes, although they receive a much smaller intervention [45,53] or no intervention at all [79].

The relationship between the patient and the therapist
The relationship between the patient and the practitioner is of great importance for a successful outcome of the treatment [42,47,49,50,59]. More treatments can improve the relationship between patient and therapist [59]. If the practitioner shows certain behavior [42,47,50], such as reflective listening to the patient, the relationship also improves. There are also indications [47,50] that, if the patient and the practitioner collaborate in the identification of additional sessions judged best to meet the patient's clinical needs, the relationship improves and alcohol consumption is reduced.

Personal contact and feedback
Of the treatments that were delivered by a practitioner, via telephone or online and individually, five effective elements were present. If a counselling session is given over the phone by a practitioner and a (1) workbook is sent out afterwards on how to reduce alcohol consumption [39,61] or if (2) personalized feedback is given before or after the telephone sessions [39,62,64], then drinking behavior is reduced, also among older adults.
(3) If an in-person session is followed by a phone call [65], this also helps to reduce drinking behavior.

Online communication and feedback
If treatment is given via online communication by means of (4) assignments or modules undertaken by the participant about his or her drinking behavior followed by a chat session with the practitioner about the assignments [63] or is (5) followed by feedback from the practitioner [66], then the drinking behavior is reduced. For none of these elements were the reasons why they were effective and which mechanisms they triggered found.

C. Practitioner-in-person-relatives
Regarding the treatments that were delivered by a practitioner, in person and included the involvement of relatives, two effective elements were found.

The status of the relationship
The (1) status of the relationship with (marriage) partners/family members influences the outcome of the intervention [61,68,[70][71][72][73]. By influencing this status, the treatment can also lead to a successful outcome [69,71].
Teaching the partner to deal with drinking behavior The partner can be (2) taught to deal with the drinking behavior of the partner through therapy [61,69,70,72] or through (video) information [68], which can lead to lead to alcohol reduction of the drinking partner. If the non-drinking partner is taught to deal with the behavior of the drinking partner, this can lead to more understanding and support from the non-drinking partner for the drinking partner [69]. The drinking partner is then better advised not to use alcohol or to moderate alcohol consumption.

D. Practitioner-in-person-group component
Of the treatments that were delivered by a practitioner, in-person and in a group setting or in a group setting at work, two effective elements were present.

Motivating to change lifestyle
Brief group interventions focusing on (1) motivating participants to change their lifestyles regarding personal relationships, nutrition and exercise [79] and coping with desires for alcohol [65,74] lead to alcohol reduction.

Motivating to change lifestyle delivered in a workplace setting
If an intervention is given in a work setting in which (1) alcohol use and its consequences are discussed [76][77][78] and/or in which a (2) training element is offered that intends to change behavior and reduce alcohol use [76][77][78] and/or (3) personal advice is given on alcohol use [76], this leads to lower (risky) alcohol use. For none of these elements were the reasons why they were effective given.

E. No practitioner-not in-person-individual
In relation to treatments that were not delivered by a practitioner, were not in-person and were individual, five effective elements were present.

Web-based interventions
Web-based interventions that give (1) personal feedback [43,64,81,85,86,89,90,95,97] and of which the respondents' result is also (2) compared with the results of people who are in the same phase [43,81] or have the same age group, gender or country of origin [85,89,95] or is compared with the previous data of the participant [81] ensure lower alcohol consumption. Web-based interventions based on cognitive behavioral therapy (CBT) that (3) gradually teach the participant skills for refusing drinks, dealing with cravings, etc., result in lower alcohol consumption [53]. Web-based interviews for older adults that also contain elements of personalized feedback and complement this with information on each person's own specific risks of alcohol consumption as well as information on the effects of alcohol on health, medication use and functional status and recommendations for safe drinking [40,41] lead to lower alcohol consumption.

Telephone based interventions
When a (mobile) phone intervention consist of a (1) self-guided program or modules or steps in which coping strategies and control functions for many alcohol-related issues are taught [83,[87][88][89]96], this could lead to less alcohol consumption and less binge drinking. Mobile phone interventions provide insight into how much someone drinks and leads to realization of their own drinking behavior [87,88]. The provision of (2) self-help material on the consequences of alcohol use and motivating behavioral change in relation to alcohol use [84] during a telephone-based intervention leads to less alcohol use.

F. No practitioner-not in-person-group component
In the treatments that were not delivered by a practitioner, not in-person and in an online group setting, two effective elements were present: Intervention of abstinent people (with or without problematic drinking behavior) (1) from drinking alcohol for a certain period or to drink less [44,98,99] and (2) to share this experience with peers [44,98] makes people aware of their alcohol consumption and reduces alcohol consumption.

Discussion
We were interested in how (which elements of interventions), in which context and why (which mechanisms) interventions prevent or reduce (problematic) alcohol consumption among older adults. We found information on the functioning of alcohol interventions for the general population (which often were designed for an 18+ population and therefore also included older adults). Three effective elements of interventions were identified in several types of contexts for the general population. Two of these three effective elements were also found in the interventions especially designed for older adults.
The second effective element was being in contact with others and communicating with them about (alcohol) problems. Sometimes practitioners help participants to develop (new) social networks [46,57,58,60]. Sometimes the family members or partners of the participants are taught to understand the drinking habits of their loved ones and how to support them in drinking less or abstaining from drinking [70,72]. Contact with peers and colleagues is also an important factor. Participants have to share their experience of abstinence for a period with their peers [44,94] or discuss with their colleagues, in a work setting, alcohol use and its consequences [76][77][78]. The importance of the role of contact with others or social networks on alcohol consumption has been acknowledged previously [100,101]. The A study by Robinson et al. [102] showed strong negative associations between empathic processing (the thoughts or feelings of others and responding accordingly) and social support and both the consequences of drinking and the percentage of drinking days.
Providing participants with personalized feedback about their drinking behavior is the third commonly found effective element across the context settings. This element leads to results in interventions that are given by a practitioner in-person [67,76] or by a practitioner via telephone [39,63,65] but also when the feedback is provided through computer-generated communication [40,41,43,64,81,85,86,89,95,97]. The effect of personalized feedback on alcohol consumption was described as important in an earlier review of online alcohol interventions [103]. The study by Riper et al. [104] showed that singlesession, individually personalized feedback without professional guidance can be effective in reducing risky alcohol consumption in young and adult problem drinkers.
The element of the provision of information on several-alcohol related issues was also found among one of the three interventions especially designed for older adults [39]. In addition, the element of providing personalized feedback was found in two of the three interventions for older adults [40,41]. The element of contact with others was not found in the three interventions especially designed for older adults. This is striking because contact with others is especially important for older adults since loneliness is a problem for that age group [105] and there is a relationship between the use of alcohol and loneliness [12,106,107].
We only found three studies on the prevention or reduction of alcohol consumption that were specifically designed for older adults. The reason for this low number of studies could be that the results of the aging of the population (people in general are becoming older and the absolute number of older adults is rising) have only become clear in the last few years and will increase in the years to come. The importance of research into the reduction and prevention of alcohol use in older adults has only recently become more apparent.

Limitations
We did not include grey literature in our review because our aim was to give an overview of the scientific peer-reviewed literature on interventions for older adults to reduce or prevent (problematic) alcohol use first. If we had included grey literature, we might have found more interventions designed specifically for older adults. Although we included many randomized controlled trials, we could not perform a meta-analysis because of the heterogeneity among the interventions, the study populations and the results. We chose to limit the operationalization of the context to the mode of delivery to make it easier to compare the contexts of the studies. For many studies, other information about the context was scarce or incomplete. If this information had been provided, a better comparison of contexts would have been possible. We only included Western high-income countries since problematic drinking behavior is highest among the population in these countries. Non-western countries were excluded because drinking culture, and thus also offered interventions to older adults, differs from western countries. This may limit the generalization of this study to other countries. Results can be generalized to the general (older) population, but not to specific groups (e.g., pregnant people, veterans) since drinking culture is different among these subgroups. Future research might investigate other vulnerable subgroups. Another limitation is that not in all articles was the 'why' mechanism addressed, indicating that a complete overview of why some interventions were effective is lacking in current research reports. Future research about why interventions were effective and especially why interventions are effective for older adults is necessary. Despite the limitations, this study provides a broad overview of which elements of interventions are effective in preventing or reducing alcohol use as well as indicating why these elements are effective.

Practical Implications
This literature review identified three major effective elements of interventions: (1) providing information on the consequences of alcohol consumption; (2) being in contact with others and communicating with them about (alcohol) problems; and (3) personalized feedback about drinking behavior. Two of these elements, information provision and personalized feedback, are related to creating awareness. This is also a common answer to why an intervention works. People became aware of their alcohol consumption and what it means for their bodies. For developers of new interventions concerning the reduction or prevention of alcohol consumption of (older) adults, but also for policy makers, it could be a good start to look at what creates awareness regarding alcohol consumption for that specific target group. The third effective element, contact with others and communicating about (alcohol) problems, is also an element that is important for developers of interventions and policy makers. People explain that sharing their experiences of (reducing) alcohol consumption helps them. In doing so, it is important that friends and family are supportive of the choice of the person to reduce or stop drinking and respond empathically about this choice. This could be difficult for some friends or family members as drinkers tend to seek each other out and then influence each other's use [100]. Developers of interventions and policy makers could therefore facilitate the process of helping (older) adults to develop contacts with people that are supportive of their choice to reduce or prevent their alcohol consumption.

Scientific Recommendations
We only found three studies on the prevention or reduction of alcohol consumption that were specifically designed for older adults. In order to provide adequate interventions to help reduce or prevent alcohol consumption for older adults, more research is necessary on what creates awareness regarding alcohol consumption for this target group. Moreover, research on how to help older adults develop contacts with people that are supportive of their choice to reduce or prevent their alcohol consumption is necessary, because these contacts are helpful in reducing or preventing alcohol consumption.

Conclusions
This study provides answers to the questions of how (which elements of interventions), in which context and why (by which mechanisms), interventions prevent or reduce (problematic) alcohol consumption among older adults. Most of the studies were not especially designed for older adults but also included older adults. The findings of this study highlight three major effective elements of interventions: (1) providing information on the consequences of alcohol consumption; (2) being in contact with others and communicating with them about (alcohol) problems; and (3) personalized feedback about drinking behavior. Two of these elements were also used in the interventions especially designed for older adults. In order to provide adequate interventions to help reduce or prevent alcohol consumption for older adults, more research is necessary on what creates awareness regarding alcohol consumption for this target group.