Received date: April 26, 2014; Accepted date: May 23, 2014; Published date: May 27, 2014
Citation: Aagaard-Hansen J, Bonde AH, Jensen BB, Andersen LB, Maindal HT (2014) Inadequate Description of Educational and Behavior Change Theories in Lifestyle Interventions for Type 2 Diabetes Prevention. J Diabetes Metab 5:375. doi: 10.4172/2155-6156.1000375
Copyright: © 2014 Aagaard-Hansen J, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Diabetes & Metabolism
The aim of this article is to explore the extent to which educational and behavior change theories were adequately described in studies aiming at postponing or preventing type 2 diabetes among individuals with prediabetes using lifestyle interventions. This is essential in order to design effective interventions for ‘real life’ settings. A number of databases were searched: Web of Science, PubMed, SCOPUS, Cochrane Library, Cross (MEDLINE, EMBASE, CUCO, BIOSIS), EPPI-Centre: Database of promoting health effectiveness reviews (DoPHER) and EPPI-Centre: The Trials Register of Promoting Health Interventions (TRoPHI). Sixteen randomized controlled trials were selected based on the criteria of prediabetes, defined as impaired glucose tolerance and/or impaired fasting glucose and efficacious lifestyle interventions (diet and/or physical activity). Only four trials made explicit reference to use of educational and behavior change theories in the development of the lifestyle interventions. Another five trials provided partial information, whereas the remaining seven did not refer to theories directly or indirectly. In order to allow replicability and comparison, future studies should explicitly describe and/or consider use of educational and behavior change theories in the development of lifestyle interventions.
Behavior change; Diet; Education; Physical activity; Prevention; Theory; Type 2 diabetes
ADA: American Diabetes Association; BMI: Body Mass Index; CONSORT: Consolidated Standards of Reporting Trials; Credeci: Criteria for Reporting the Development and Evaluation of Complex Interventions in Healthcare; IFG: Impaired Fasting Glucose; IGT: Impaired Glucose Tolerance; PA: Physical Activity; RCT: Randomized, Controlled Trial; T2D: Type 2 Diabetes; TREND: Transparent Reporting of Evaluation with Non-Randomized Designs; WHO: World Health Organization
Assessment of public health interventions should follow methodological rigor and at the same time take into account the complexity of implementing such interventions in real life settings. The Medical Research Council’s evaluation framework  has become if not a golden standard, then a very common reference for development of complex primary health care interventions. In parallel to this development, researchers and editors of scientific journals have made an effort to improve the way scientific studies are reported, e.g. Consolidated Standards of Reporting Trials (CONSORT) , Transparent Reporting of Evaluation with Non-randomized Designs (TREND)  and Criteria for Reporting the Development and Evaluation of Complex Interventions in Healthcare (CreDECI) .
“Adequate application of behavioural science theories is essential for effective behavior change interventions. Planning and evaluation are key elements in developing health promotion programmes. Theories and empirical evidence form the basis for decisions during the planning process ….” . This is in line with attempts to initiate a much needed discussion on the role of theory in health behavior change interventions . This has led to explorations of the usefulness of self-efficacy , intervention mapping [5,8], health behavior theory  and self-determination theory [10,11] in intervention designs.
In the field of diabetes prevention the past two decades’ research has shown the efficacy of intensive lifestyle interventions (diet and physical activity (PA)) in postponing or even preventing the onset of type 2 diabetes among persons with prediabetes. The focus has now shifted to effectiveness and efficiency studies exploring how prevention interventions may be implemented in real life settings. Numerous systematic reviews on lifestyle prevention of type 2 diabetes [12-19] as well as many more scoping reviews have been conducted which mostly focused on the diabetes screening methods and outcomes rather than provision of details about the interventions used. The present article focuses on the extent to which the use of educational and behavior change theories in design of the interventions was explicitly reported.
The following electronic databases were searched in order to identify relevant literature - Web of Science, PubMed, SCOPUS, Cochrane Library, Cross (MEDLINE, EMBASE, CUCO, BIOSIS), EPPI-Centre: Database of promoting health effectiveness reviews (DoPHER), EPPI-Centre: The Trials Register of Promoting Health Interventions (TRoPHI). Variations and combinations of the following search terms were used when searching the databases: (1) pre-diabetes, type 2 diabetes mellitus AND prevention; (2) lifestyle modification, behavior change, health promotion, patient education, intervention, prevention program; (3) physical activity, exercise, diet, weight loss, motivational interviewing, counselling, self-management. A total of 16 trials applying lifestyle interventions (diet and/or PA) with the purpose of delaying or preventing the onset of type 2 diabetes among persons with prediabetes were selected based on the criteria that they should be Randomized, Controlled Trials (RCT) including participants with prediabetes defined as impaired glucose tolerance (IGT) and/or Impaired Fasting Glucose (IFG) (according to relevant American Diabetes Association (ADA) or WHO criteria at the given time). Furthermore, only studies were included which could show significant effect on one or more of the following outcome measurements weight, Body-Mass Index (BMI), waist circumference, IGT, IFG or cumulative incidence of diabetes. No attempts were made to access additional unpublished data or data from project websites. The full texts of the articles were scrutinized with regard to use of terms directly or indirectly referring to educational and behavior change theory. It was not feasible to explore differences in outcome based on whether the trials used educational theories or not.
Lifestyle intervention trials
Sixteen trials were included in the review (Table 1). Most of the 16 trials combined diet and PA intervention. Two studies (DPP [20-22] and IDDP [23,24]) combined lifestyle interventions with pharmaceuticals (Metformin) and some [25-30] addressed smoking and/or alcohol in addition to diet and PA. Most of the trials conducted a relatively intensive initial start-up training program followed by a maintenance program serving to sustain the behavior changes among the study participants. The interventions predominantly comprised health education sessions individually and/or in groups, face-to-face and/or via electronic social media and sometimes combined with clinical and other examinations.
|Trial||Abbreviation||Reference to theory||Reference to educational and behavioral terms|
|Reduced-Fat Diet Intervention in Individuals With Glucose Intolerance ||AUCKLAND||Self-monitoring|
|Ballarat DiabetesPrevention Pilot Initiative ||BDPPI||Motivational interviewing , self-efficacy, self- management principles|
|Diabetes Education & Prevention with a Lifestyle Intervention Offered at YMCA [44,45]||DEPLOY||Empowerment, self esteem|
|Diabetes Prevention Program and Diabetes Prevention Program Outcomes Study [20-22]||DPP||Empowerment strategy, motivational interviewing, self-monitoring|
|Diabetes Prevention Study [25,26]||DPS||Stages of Change Model |
|Da Qing Study [48,49]||DQS|
|Glucose Tolerance Study within the Västerbotten Intervention Programme ||GT-VIP||Problem solving techniques, relapse prevention, self-monitoring|
|Healthy Living Partnership to Prevent Diabetes Study [33,34]||HELP PD||Social Cognitive Theory , the continuing care problem-solving model|
|IndianDiabetesPrevention Programme [23,24]||IDDP|
|JapaneseDiabetesPrevention Project ||JDPP|
|Joetsu Diabetes Prevention Trial ||JDPT|
|The 6-year MalmÖ feasibility study ||MALMÖ|
|New Dietary Education Programme ||NDE||Self management|
|Pre-diabetes Risk Education and Physical Activity Recommendation and Encouragement Programme Study [39,40]||PREPARE||Chaiken’s Dual Process Theory , Gollwitzer’s Implementation Intentions , Leventhal’s Common Sense Model  Social Cognitive Theory |
|Study on Lifestyle and Impaired Glucose Tolerance Maastricht [29,30]||SLIM|
|The Zensharen Study for Prenvention of Lifestyle Diseases ||Zensharen|
Table 1: The 16 trials included in the article and the theories and terms referred to
Reference to educational and behavior change theories
By educational theory we understand a coherent or explanatory model (e.g. 'constructivism') which provides goals, norms and standards for effective education learning processes. The notion of behavior change has connotations to behaviorism including observational processes and external (top-down) attempts to modify behavior . A specific health education activity will always build on a number of pre-assumptions inspired from - or embedded in - one or more models or theories. Although these assumptions might not be explicitly described or defined, they will always guide and influence the practices carried out as part of health education and behavior change interventions.
Four of the studies referred explicitly to educational or behavior change theories. DPS [25,26] indicated that the Stages of Change Model1  was used though few details were provided on the ways in which it was done. HELP PD [33,34] used Social Cognitive Theory2  including the concepts of self-efficacy3 and outcome expectations and incentives. BDPPI  mentioned Social Cognitive Theory , Motivational Interviewing4  and self-management principles . PREPARE [39,40] referred extensively to several theories such as Social Cognitive Theory , Gollwitzer’s Implementation Intentions , Leventhal’s Common Sense Model  and Chaiken’s Dual Process Theory . In addition PREPARE [39,40] measured self-efficacy as part of the study.
In five studies various degrees of reference were made to theoretical terminology. Some educational and psychological concepts were mentioned in passing: empowerment (DEPLOY [44,45], DPP [20-22]), self-esteem (DEPLOY [44,45]), self-management (NDE ), relapse prevention (GT-VIP ) and self-monitoring (AUCKLAND , DPP [20-22], GT-VIP ). DPP [20-22] indicated that motivational interviewing was used as a technique. GT-VIP  mentioned problem solving techniques. Seven of the 16 trials (DQS [48,49], IDDP [23,24], JDPT , JDPP , MALMÖ , SLIM [29,30], Zensharen ) made no explicit reference to educational and behavior change principles or theories. Thus, whereas most of the trials were very specific about the content of the intervention (the what) there were hardly any mentions of how the education was conducted.
The role of social support/influence was mentioned by DEPLOY [44,45] and HELP PD [33,34] as a recognition of the importance of social networks. The only other study alluding to the importance of structural factors (as opposed to individual changes of lifestyle) was HELP PD [33,34], which mentioned physical environment in passing.
Among the 16 trials covered in this article only four (BDPPI , DPS [25,26], HELP PD [33,34] and PREPARE [39,40]) explicitly referred to educational or behavior change theories. As the whole rationale of the selected RCT-based research is to apply lifestyle interventions in order to prevent type 2 diabetes among persons with prediabetes, it is not surprising that the few theories used mainly focus on individual aspects of behavior change (as opposed to structural interventions addressing social and/or built environment). It is, however, remarkable that five of the trials made only sporadic reference to theoretical concepts and seven did not refer explicitly to theories at all. Assuming that that reflects the actual situation, it points to an under-utilization of relevant expertise from other disciplines such as education, psychology and social science.
In their recent review, Baker et al.  made an effort to uncover the ‘behavioral strategies’ utilized in seven trials of which six (DPP [20-22], DPS [25,26], DQS [48,49], IDPP [23,24], JDPP  and GT-VIP ) are also covered here. They found Social Cognitive Theory , the Transtheoretical Model5  and the Theory of Planned Behavior6  to be important. Baker et al.  may be right that the theories mentioned above were indeed specifically considered while planning these lifestyle interventions. But the findings in the present article seem to indicate that this is based on an interpretation of the ways the practical lifestyle interventions were designed in the seven trials, rather than explicit statements in the various articles describing them.
Greaves et al.  found causal evidence that “change in weight was greater when established, well defined behavior change techniques were added to interventions”, but no evidence that “interventions with an explicitly stated theoretical basis … were more effective”. These findings do not rule out the importance of educational or behavior change theory in the design of lifestyle interventions. Partly, the ‘behavior change techniques’ such as specific goal setting, relapse prevention and self-monitoring are implicitly based on theoretical rationales. Partly, the lack of causal evidence may be a consequence of using inappropriate theories or using them inappropriately.
In the present article we have demonstrated inadequate description and/or use of educational and behavior change theories in trials reporting lifestyle interventions for T2D prevention. This critique is not only academically motivated. Use of theories may or may not improve the interventions, but they need to be described in sufficient detail to allow progress of intervention development and replication of the intervention, if shown to be effective. In order to stimulate professional progress, academic exchange of experiences (successes as well as failures) and further scientific development of the area of prevention of type 2 diabetes, it is important to unpack, discuss and challenge the theoretical models that underpin our health education practices.
As the field of T2D prevention research is in the transition from efficacy to effectiveness and efficiency studies, it is essential that theories used are reported in sufficient detail to allow replicability, comparisons and further development. The same conclusion was reached in the recent review of Greaves et al.  The revised CONSORT statement propose as one of the 22 quality items for reporting trials that “precise details of the interventions intended for each group and how and when they were actually administered” be specified . Though focusing on nonrandomized evaluations, the TREND Statement makes the same recommendation and makes in addition the point that “theories used in designing behavioral interventions” should be described . Similarly, the CReDECI guideline highlights the importance of “description of the intervention’s underlying theoretical considerations” as well as “description of all components of the intervention” . The Medical Research Council guidance for evaluation of complex interventions from 2008 also emphasizes the importance of identifying and developing theory and describing the intervention clearly . Abraham and Michie  have suggested a detailed “taxonomy of behavior change techniques used in interventions” that may assist authors in standardizing and thereby clarifying the way lifestyle interventions are reported.
This review has a potential limitation. It restricts itself to information on use of theories in interventions as described in the articles yielded by the published literature. A broader search on homepages, personal contacts, manuals and grey literature might have provided additional knowledge thereby allowing more precise comparisons. However, it could be argued that the same standards of transparency should apply to scientific articles describing health education and behavior change methodologies as pertain to standard laboratory and clinical procedures.
The present article provides an outline of 16 major trials addressing prediabetes by the means of lifestyle interventions - with a focus on the extent to which educational and behavior change theories had been explicitly utilized. Only four of the studies provided sufficient details, five made only sporadic reference to theoretical terms and the remaining seven did not report any information at all.
Researchers studying lifestyle interventions in general and type 2 diabetes prevention in particular should make an effort to abide to existing guidelines for reporting. More specifically, there is a need for future studies to design lifestyle interventions with direct and explicit reference to educational and behavior change theories. This will serve the double purpose of informing future research and assisting practitioners in applying the results in real life settings.
The authors declare that they have no competing interests. The Steno Health Promotion Center at Steno Diabetes Center is funded by Novo Nordisk Ltd. and the Novo Nordisk Foundation.
All authors have participated in the search phase as well as write up. LBA did the main literature search. JAH has drafted the manuscript on which other authors have commented.
We are thankful to colleagues at University of Aarhus and Steno Diabetes Center as well as anonymous reviewers for useful comments on draft manuscripts. Thanks also to the Novo Nordisk library for assistance in the literature search.