Do Combinations of Behavior Change Techniques That Occur Frequently in Interventions Reflect Underlying Theory?

Abstract Background Behavioral interventions typically include multiple behavior change techniques (BCTs). The theory informing the selection of BCTs for an intervention may be stated explicitly or remain unreported, thus impeding the identification of links between theory and behavior change outcomes. Purpose This study aimed to identify groups of BCTs commonly occurring together in behavior change interventions and examine whether behavior change theories underlying these groups could be identified. Methods The study involved three phases: (a) a factor analysis to identify groups of co-occurring BCTs from 277 behavior change intervention reports; (b) examining expert consensus (n = 25) about links between BCT groups and behavioral theories; (c) a comparison of the expert-linked theories with theories explicitly mentioned by authors of the 277 intervention reports. Results Five groups of co-occurring BCTs (range: 3–13 BCTs per group) were identified through factor analysis. Experts agreed on five links (≥80% of experts), comprising three BCT groups and five behavior change theories. Four of the five BCT group–theory links agreed by experts were also stated by study authors in intervention reports using similar groups of BCTs. Conclusions It is possible to identify groups of BCTs frequently used together in interventions. Experts made shared inferences about behavior change theory underlying these BCT groups, suggesting that it may be possible to propose a theoretical basis for interventions where authors do not explicitly put forward a theory. These results advance our understanding of theory use in multicomponent interventions and build the evidence base for further understanding theory-based intervention development and evaluation.


BCT Group
Factor Loading BCTTV1 Number & Label BCT Definition BCT Group 1 823 1.5 Review Behaviour Goals Review behaviour goal(s) jointly with the person and consider modifying goal(s) or behaviour change strategy in light of achievement. This may lead to re-setting the same goal, a small change in that goal or setting a new goal instead of (or in addition to) the first, or no change. .783 1.4 Action Planning Prompt detailed planning of performance of the behaviour (must include at least one of context, frequency, duration and intensity). Context may be environmental (physical or social) or internal (physical, emotional or cognitive). . 729 15.3 Focus on Past Success Advise to think about or list previous successes in performing the behaviour (or parts of it). .717 1.1 Goal Setting (Behaviour) Set or agree a goal defined in terms of the behaviour to be achieved. . 712 15.4 Self-Talk Prompt positive self-talk (aloud or silently) before and during the behaviour. .705 3.3 Social Support (Emotional) Advise on, arrange, or provide emotional social support (e.g. from friends, relatives, colleagues, 'buddies' or staff) for performance of the behaviour. .624 1.2 Problem Solving Analyse, or prompt the person to analyse, factors influencing the behaviour and generate or select strategies that include overcoming barriers and/or increasing facilitators .613 2.3 Self-Monitoring of Behaviour Establish a method for the person to monitor and record their behaviour(s) as part of a behaviour change strategy. .570 8.7 Graded Tasks Set easy-to-perform tasks, making them increasingly difficult, but achievable, until behaviour is performed. .560 1.3 Goal Setting (Outcome) Set or agree a goal defined in terms of a positive outcome of wanted behaviour.
.523 1.6 Discrepancy between Current Behaviour and Goal Draw attention to discrepancies between a person's current behaviour (in terms of the form, frequency, duration, or intensity of that behaviour) and the person's previously set outcome goals, behavioural goals or action plans (goes beyond self-monitoring of behaviour

Overview Background & Aim
The Theories and Techniques of Behaviour Change Project (www.ucl.ac.uk/behaviour-change-techniques) aims to identify hypothesised links between behaviour change techniques (BCTs) and (i) the mechanisms of action through which they influence behaviour and (ii) behavioural theories. The final phase of this research aims to examine whether groups of co-occurring BCTs can be linked to specific theories.

Participants
You are one of 25 experts from 10 countries, selected because you have a high level of expertise in theories and BCTs and provided valuable input in our previous consensus exercise.

Overview of Study
The consensus exercise is a modified Nominal Group Technique, involving 4 rounds. In Round 1, you will be directed to an online questionnaire and asked to list all possible theories that might underlie a particular BCT group. You will be asked to do this for 5 BCT groups (see Appendix A). In Round 2, you will be presented with the same 5 BCT groups, and asked to consider the extent to which each BCT group is linked to a particular theory; the theories you are asked to consider will be those identified by experts in Round 1. In Round 3, you will be invited to participate in an online, asynchronous (i.e. each member can participate at a time of their choosing), anonymous discussion, in which you will be encouraged to exchange views with other experts. Finally, in Round 4, you will be asked to provide final ratings on links between BCT groups and theories (which may or may not be the same as your Round 2 ratings). At this point, you will also have the opportunity to add and/or remove BCTs from each BCT group for each theory.
More detail on Rounds 1 and 2 can be found below; Guidelines for Rounds 3 and 4 will be sent out separately.

Round 1 Guidelines Information Provided
In Appendix A, you will find 5 groups of co-occurring BCTs (i.e. BCTs that were found to appear together frequently in interventions, identified through an exploratory factor analysis). These BCT groups will be presented to you on screen during the consensus exercise. In addition, you will receive a complementary online book containing a description of 83 theories ('ABC of Behaviour Change Theories'), which you may draw on during the task. Log-in details to access this online book will be sent to you by email.

Question
You will be presented with 5 BCT groups, one at a time, and asked to list all theories that you consider to be associated with each group, in the following question:

Response Options
To respond, you will be provided with an open text box in which you can list one or more theories, drawing on the 'ABC of Behaviour Change Theories' and/or your own knowledge and expertise (see Appendix B for a preview of the Round 1 questionnaire).

An intervention was developed that includes the BCTs listed above. Which behavioural theory or theories do you think could have guided the development of this intervention?
Round 2 Guidelines Information Provided You will have access to the same information as above (i.e. the 5 BCT groups and the online book, 'ABC of Behaviour Change Theories').

Question
You will be presented with 5 BCT groups, one at a time, and asked to respond to the following question for all theories listed by experts in Round 1: When providing your ratings, please remember:

Response
(1) In order to ensure consistency across experts, and with the aim of developing a shared understanding of BCT-mechanism of action links, it is very important that you base your answers on the definitions, rather than labels, of the BCTs.
-Definitions for all BCTs will appear on the screen during the consensus exercise, and these can also be found in Appendix A. Please read and re-read these definitions before beginning the task. Definitions can also be accessed via www.bcts.23.co.uk, or through the BCTTv1 smartphone app, which can be downloaded for free via the Apple App and Google Play stores.
(2) In answering the questions, you may draw on your own knowledge and expertise and/or the theories in the 'ABC of Behaviour Change Theories'.
An intervention was developed that includes the BCTs listed above. How confident are you that this group of BCTs is linked to [Theory X]?

Overview:
Thank you for completing the first two rounds of this study. In Round 3, we are asking for your input on links between groups of co-occurring BCTs and theories where there remains uncertainty and/or disagreement. The purpose of this round is to discuss these uncertainties and disagreements to help guide your final round ratings. By providing your input in this round, you will add to our shared understanding of the links between BCT groups and theories, and help to build consensus. This is your opportunity to share your thoughts with other behaviour change experts about the task.

Information Provided:
In Appendix A, you will find confidential data from Round 2 of this consensus exercise, including a summary of all experts' responses, and a reminder of your own responses. You will be able to see the group data on screen as part of the Round 3 discussion. However, we recommend that you read through Appendix A in full before Round 3 begins, in order to give you a chance to reflect on the data to inform your contribution to the discussion.
In Appendix B, you will find step-by-step instructions for how to use 'Loomio', the online discussion forum on which Round 3 will be hosted.

What the task involves:
• We are asking experts to contribute to an online discussion forum, in which you will have the opportunity to share your views about BCT group-theory links rated in Round 2, and to comment on other experts' views. • The online discussion will involve the group of 25 experts who are participating in this study. Your contributions will be anonymous (i.e. you will use an assigned Expert ID and not your name). • We estimate that your participation in Round 3 should take approximately 1.5 hours in all. While you may log in and participate at any time(s) of your choosing, we would recommend that you log in at least three times over the Round 3 period. This is to ensure that you have had a chance to contribute fully to the discussion. • A moderator from the research team will periodically summarise the discussion and raise issues for further consideration.

How to participate:
Step-by-step instructions for participating in the online discussion can be found in Appendix B. In summary: • You will be sent a link to the discussion via an email from the research team. When you click on this link, a website called 'Loomio' will open and you will be prompted to register an account or log in. You can log in using the details you previously registered (i.e. during the last expert consensus study). If, for any reason, you need to register a new account (e.g. if you no longer have a Loomio account), it is very important that you do this using your Expert ID and not your name.
• There will be a number of discussion groups on which you can comment, each specific to a BCT group-theory link for which there was high uncertainty or disagreement in Round 2 (see Appendix A). There will be other discussion groups in which you can contribute thoughts about the task more generally and/or about other BCT group-theory links.
• It may be helpful to focus on the ratings of BCT group-theory links for which (i) you remain very uncertain, and/or (ii) your responses differed from those of the other experts. However, we would also value your thinking on other aspects of the task.

Round 4 Guidelines Overview
Thank you for completing the first three rounds of this study. In Round 4, we are asking you to provide your final ratings for links between groups of behaviour change techniques (BCTs) and behavioural theories. You will be directed to an online questionnaire and asked to consider whether a group of BCTs is linked to a particular theory. In addition, for each BCT group, you will be asked to indicate the theory or theories (up to three) you are most confident are linked. For these theories, you will also have the opportunity to consider whether there are BCTs that should be added to and/or removed from the BCT group.

Information provided:
During the final round, you will see data on screen from Round 2 (i.e. 'In Round 2, X% of experts were very confident that this BCT group is linked to this theory'). For the links that were discussed in Round 3, there will be hyperlinks to the relevant discussion page on Loomio (note that you will need to be signed in to view this content). You will not be asked to discuss links for which there was consensus in Round 2.
Definitions for the BCTs in each BCT group will be presented on screen during the consensus exercise (these are also available in Appendix A). BCT definitions can also be accessed via www.bcts.23.co.uk, and through the BCTTv1 smartphone app, which can be downloaded for free via the Apple App and Google Play stores.
To view theory descriptions, you may draw on the ABC of Behaviour Change Theories book, as in the previous rounds.

What the task involves:
1. You will be presented with 5 BCT groups, one at a time, and asked to respond to the following question (for all theories where consensus was not reached in Round 2): Response Format o In this round, the response options for this question will be: 2. Following completion of the above, you will be asked to identify the theory or theories that you are most confident are linked to each BCT group. o For each BCT group, you will be presented with the theories you rated as 'confident that there is a link' and asked to select those you are most confident about. o You may base your judgements on the frequency and/or strength of links between the BCTs within a group and the constructs of the theory.

Response Format
o The response format for this question will be a click-and-drag format. You will be asked to select a theory and 'drag' it into the 'most confident' box. You will be able to do this for up to three theories. If you select more than one theory, you will be able to re-order theories within the box.

3.
Finally, for each of the theories you identified in the previous step, you will be asked to indicate whether or not there are BCTs you think should be added to, or removed from, the BCT group. You will be provided with an open text box to indicate which (if any) BCTs you would add and/or remove.