Professionals’ perspectives on interventions to reduce problematic alcohol use in older adults: a realist evaluation of working elements

Objectives This study set out to understand how (which elements), in what context and why (which mechanisms) interventions are successful in reducing (problematic) alcohol use among older adults, from the perspective of professionals providing these interventions. Design Guided by a realist evaluation approach, an existing initial programme theory (IPT) on working elements in alcohol interventions was evaluated by conducting semistructured interviews with professionals. Setting and participants These professionals (N=20) provide interventions across several contexts: with or without practitioner involvement; in-person or not and in an individual or group setting. Data were coded and links between contexts, elements, mechanisms and outcomes were sought for to confirm, refute or refine the IPT. Results From the perspective of professionals, there are several general working elements in interventions for older adults: (1) pointing out risks and consequences of drinking behaviour; (2) paying attention to abstinence; (3) promoting contact with peers; (4) providing personalised content and (5) providing support. We also found context-specific working elements: (1) providing personalised conversations and motivational interviewing with practitioners; (2) ensuring safety, trust and a sense of connection and a location nearby home or a location that people are familiar with in person and (3) sharing experiences and tips in group interventions. Furthermore, the mechanisms awareness and accessible and low threshold participation were important contributors to positive intervention outcomes. Conclusion In addition to the IPT, our findings emphasise the need for social contact and support, personalised content, and strong relationships (both between client and practitioner, and client and peers) in interventions for older adults.

Whilst the manuscript makes some references to motivational interviewing, many of the interventions and related elements seem very relevant to an extensive 'brief interventions' literature (albeit existing in many forms/ names e.g, predominantly 'Screening and Brief Interventions and Referral to Treatment' in the US).Some further reference to the relevance of these findings to the brief intervention literature I think should be incorporated within the introduction and implications.
Under 'Interpretation of findings', the authors state there are two important mechanisms that contribute to positive outcomes.The first is identified as 'awareness' but is not clear what the second is (perhaps knowledge?).
Concerning the mechanism of awareness, some recent conceptual and empirical work has focused on this (termed 'problem recognition') in non-help seeking drinking groups [1,2] which it may be worth highlighting.Notably, one recent study on problem recognition [3] also empirically supports the important point made about the threat of 'alcoholic' labeling undermining 'awareness'/problem recognition.
2) Please describe the Initial Program Theory of Boumans et al in greater detail that you refer to.Readers will likely be unfamiliar with the work and it is problematic not describing it in greater detail on line 55.
3) Please fix the formatting in tables 2 and 3 so that they have greater readability.Table 2 had "professional" across three lines, this was the same for "alcohol use" and "relatives".This also goes for Table 3 such as "target group intervention".4) In the results section, please label each italicized component as either a context, element, mechanism, or outcome.I believe that most of these are elements, but it would assist the reader in better understanding what each component is without confusing elements for mechanisms for example.This could be easily changed by writing for example on Line 51 "Professionals confirmed that the element (1) paying attention to drinking behavior..." 5) Please define context, element, mechanism and outcome better on lines 44-49 for readers unfamiliar with this methodology.You put them together nicely but they still lack robust definitions.3 is great, I just wanted to commend you all on how you put that information together.

Reviewer #1
Thank you for the opportunity to review 'Professionals' perspectives on interventions to reduce and prevent problematic alcohol use in older adults: A realist evaluation of working elements.'I found this to be an interesting study addressing an important issue of alcohol use amongst older adults.
Comment 4: My main comment is that overall I found the manuscript somewhat hard to follow, even after several readings.In particular, the narrative organization/categorization of the interventions took me some time to comprehend (the tables were easier to follow).As such I think the most significant improvement to this manuscript would be to try and reconfigure the presentation of the 'six combinations of contexts' such that these are easier to comprehend.Reply: We thank the reviewer for the comments and suggestion.We try to present the contexts more easier.First, we made a summarized table of our program theory to provide a better overview of our findings to ensure that the categorization of contexts is easier to comprehend.Second, we added some sentences in the introduction with extra explanations and clarifications of the contexts.Third, we mentioned in the result section that the contexts we describe are the same contexts that the IPT conformed to.

Revised text:
"Table 4 provides an overview of the summarized findings of the program theory."(See page 14 and 15 for Table 4).Comment 5: Whilst the manuscript makes some references to motivational interviewing, many of the interventions and related elements seem very relevant to an extensive 'brief interventions' literature (albeit existing in many forms/ names e.g, predominantly 'Screening and Brief Interventions and Referral to Treatment' in the US).Some further reference to the relevance of these findings to the brief intervention literature I think should be incorporated within the introduction and implications.Reply: We thank the reviewer for the suggestion.We acknowledge the relevance of SBIRT for older adults in reducing problematic alcohol use.We agree with the suggestion that we should include the relevance of SBIRT in our introduction, and therefore we mentioned SBIRT as one of the existing approaches that are appropriate for older adults.Additionally, we addressed the relevance in our discussion section, explaining how our findings can be applied to existing brief interventions.

Revised text:
"There are several existing approaches and specific interventions to reduce alcohol use, such as Screening, Brief Intervention, and Referral to Treatment (SBIRT), cognitive behavioral therapy (CBT) or motivational enhancement therapy, that are considered to be appropriate for older adults"(See page 3) "First, practitioners should have attention for the elements that are related to the role of the practitioner in in-person interventions, more specifically in existing brief interventions which are already often used by practitioners."(See page 18) Comment 6: Under 'Interpretation of findings', the authors state there are two important mechanisms that contribute to positive outcomes.The first is identified as 'awareness' but is not clear what the second is (perhaps knowledge?).Reply: Thanks for this comment.Indeed, we didn't include the word mechanism.Moreover, we were not consistent in how we mentioned the second mechanism throughout the manuscript and therefore did some extra minor changes throughout the manuscript.

Revised text:
"Second, accessible and low-threshold participation is an important mechanism in interventions for older adults."(See page 17) We made textual changes when referring to the second mechanism accessible and low-threshold participation throughout the manuscript.(See page 2,15,17,18) "This study is set out to confirm, refuse, or refine the IPT, consisting of the working elements, mechanisms and outcomes across six different contexts, as identified by Boumans et al (22)."(See page 4) Comment 10: Please fix the formatting in tables 2 and 3 so that they have greater readability.Table 2 had "professional" across three lines, this was the same for "alcohol use" and "relatives".This also goes for Table 3 such as "target group intervention".Reply: We thank you for this pointing this out.We have revised the tables such that they are now easier to read.Table 2 is now on a separate page with landscape orientation and the cell sizes in Table 3 have been adjusted such that words are on two lines.
Comment 11: In the results section, please label each italicized component as either a context, element, mechanism, or outcome.I believe that most of these are elements, but it would assist the reader in better understanding what each component is without confusing elements for mechanisms for example.This could be easily changed by writing for example on Line 51 "Professionals confirmed that the element (1) paying attention to drinking behavior..." Reply: Thank you for this suggestion.It indeed makes the content easier to follow.We have not labelled each component with an (e), (m) or (o).

Revised text:
"Below, we describe the working elements (e), mechanisms (m) that contribute to reductions in alcohol use (o)." ( See page 9) We made changes throughout the result section to label each element (e), mechanism (m) and outcome (o).(See page 9-12) Comment 12: Please define context, element, mechanism and outcome better on lines 44-49 for readers unfamiliar with this methodology.You put them together nicely but they still lack robust definitions.

Reply:
We understand that providing more information about the Realist Evaluation can be helpful for readers who are not familiar with this approach.Therefore, we decided to provided more detailed information about this approach.In two incidents mechanisms are followed with (m) and in two incidents with (e).. Is this a mistake?
In several places the manuscript refers to 'responsible' alcohol use.Ideally I suggest replacing this with 'lower risk' as there is arguably a greater degree of subjectivity and possibility of judgement (i.e., drinking above guidelines is 'irresponsible') associated with 'responsible'.
As a possibly pedantic suggestion, it may be worth adding the letter identifiers (e,m,o) to the table headings for extra clarity.
In the discussion, I suggest replacing the instances of the word 'minimally' with 'at least' for clarity.E.g., "were found in minimal in three contexts" would be clearer as "were found in at least three contexts"

GENERAL COMMENTS
This has already been reviewed, so there is little for me to comment upon.However, I wanted to state that I appreciate this research given the subject matter and I believe it to be important.
> On page 4 there was a typo of M for mechanism.An E was in place of the M.
> On page 16 the word "providing" should be italicized next to "4)".
> I was interested, though, as to what the average interview length was and the range of interview times.Especially because it was mentioned that interviewees may have wanted to leave early by agreeing with the IPT.Please mention this somewhere in the paper.
Other than those minor revisions it was a great piece on an important and often overlooked topic.

VERSION 2 -AUTHOR RESPONSE
Comment 1: Please complete a thorough proofread of the text and correct any spelling and grammar errors that you identify.
Reply: We completed a proofread of the text, conducted by native English speakers.
Revised text: "Realist evaluation (RE) works from the assumption that interventions and their elements (E) work differently in different contexts (C).The context refers to the circumstances in which interventions operate, e.g. an online context in which an intervention is provided.Interventions may be successful in some contexts and not in others because the underlying processes through which the interventions bring change, called the mechanisms (M), are triggered to a different extent and lead to different intervention outcomes (O).Therefore, the interaction between the intervention elements (E), contexts (C), and mechanisms (M) play an important role in shaping the intervention outcomes (O)." (See page 4) Comment 13:

Comment 2 :
Page 18, last paragraph -please italicise 'awareness' for consistency when referring to mechanisms, i.e.First, awareness is an important mechanism in preventing or reducing problematic alcohol use.Reply: We thank you for pointing this out and we changed this in the manuscript.Revised texts: First, awareness is an important mechanism in preventing or reducing problematic alcohol use.(See page 17) Comment 3: Please check the citation on page 4, last paragraph, line 63 for Boumans et al. (22).In the reference list, Boumans et al. is number 21 and Ettner et al. is number 22.

Table 4 .
Summary of the program theoryThe mechanisms that are presented are the summarized mechanisms from Table3, along with the numbering of the original mechanisms as indicated in Table 3 "First, in interventions that were delivered in the context […] Second, in interventions that were delivered in the context […] Third, in interventions that were delivered in the context […] Fourth, in interventions that were delivered in the context […] Fifth, in interventions that were delivered in the context […] Sixth, in interventions that were delivered in the context"(See page 3 and 4) "This study is set out to confirm, refuse, or refine the IPT, consisting of the working elements, mechanisms and outcomes across six different contexts, as identified by Boumans et al. (22).More specifically, the objective of this study is to understand from the perspective of professionals providing these interventions how (i.e., which elements), in which context, and why (which mechanisms) interventions are successful in reducing (problematic) alcohol use among older adults."(See page 4) "We found data across six combinations of contexts that the IPT conformed to." (See page 9) tailoring (1.1.,1.2., 1.5.a.), improving communication with practitioner (1.3.a., 1.3.c.), engagement (1.5.b), and emotional support (1.5.c) Table 3 is great, I just wanted to commend you all on how you put that information together.Reply: We would like to thank the reviewer for this compliment.