Feature:
Healthy Living Move for Health
No.49
January 2007
 
   print / save view 

The Effective Introduction of Physical Activity Programs: The Implementation Challenge.
Marijke Hopman-Rock

 

Abstract
This paper explains why innovation processes, such as the introduction of newly developed physical activity programs, often fail in the implementation phase. Why is it important to carry out research in this area and how may we do this? The implementation challenge is illustrated using examples of the introduction of physical activity programs for older people. Attention is paid to the definition of some widely used concepts and what is known in general about the implementation challenge.

Introduction
Most research on physical activity is in the area of effectiveness: are there favourable results from a program on health variables? Are the results in the experimental group better than in the control group? In the best case, all the available research articles are put together in a scholarly review or a meta-analysis of the effects on personal (health) variables.
However, in most cases the question still remains of what facilitates or impedes the introduction of a new intervention program in practice. This is called the “implementation challenge”. It is not enough to observe health effects in a laboratory situation (as is mostly the case in randomised controlled trials) - programs also have to show their effect in real life situations and to prove their implementation ability in the organisations that will serve the programs to the public. It is known from experience in introduction of innovative programs (1) that, without a systematically guidance of the introduction by program managers, only 7% is used as intended (see figure 1).
Figure 1: Illustration of the “Implementation Challenge”. Use as intended of innovations after one year (Adjusted from Paulussen, 1994)


In this paper it will be explained why innovation processes often fail, why it is important to carry out research in this area and how we may do this, and what we can learn of earlier experiences in the introduction of innovative physical activity programs.
The definition of some widely used concepts will be given. After that, the pathway from basic research to implementation is viewed from the perspective of the researcher and from the perspective of the practitioner, illustrated with examples of my own research. Finally, some basic principles of the diffusion theory of Rogers are given and the RE-AIM framework of Glasgow is introduced to support research on innovation processes.

Concepts and definitions
Margot Fleuren and her colleagues (2) reviewed the determinants of innovations within health care organisations. They carried out a review of the literature and a Delphi procedure among 44 implementation experts. As stated before, only around 7% of the innovative programs were used as intended. Several determinants facilitated or impeded their introduction. Fleuren et al. (2004) identified 50 determinants of the innovation process. They were categorised as:
  • Determinants related to the socio-political context (such as willingness of the patient or participant to comply with the innovation and the persons discomfort);
  • Determinants related to the organisation who delivers the innovation (such as formal reinforcement by the management to integrate the innovation into the organisational policies);
  • Determinants related to the adopting person/user/health professional (such as self-efficacy to perform the behaviour needed; extent to which ownership by the health professionals is perceived);
  • Determinants related to the innovation (such as extent to which the procedures/guidelines are clear; relative advantage; extent to which the innovation is perceived as advantageous);
  • Determinants related to facilities needed to implement the innovation (such as financial resources, existence of equipments and manuals, etc)
To avoid confusion about terminology, the most widely used concepts are defined below, using the definitions by King et al. (3):
Dissemination = the purposive process by which knowledge, including knowledge about an innovation or program, is transferred (especially transfer between researchers and practitioners).
Adoption = the decision-making step when an individual or organisation initially accepts an innovation or program, and commences the process of putting it to use.
Implementation = conducting a program or set of activities as planned. This means that the program manager of the organisation or the individual tries to adhere to the innovation as intended.
Maintenance = the continued use or implementation of a program. How the organisation or individual does its best to continue the use.
Diffusion = the spread or transfer of knowledge, including knowledge about an innovation or program; a passive process. Note the difference between this and dissemination. Dissemination is a purposive process, while diffusion is something that happens spontaneously.
Linkage = the system or patterns of connection between groups (especially researchers and practitioners).
The processes from dissemination to maintenance is mostly referred to as the “implementation process”, but may be better generalised as the “innovation process”. Linkage is important during all the various stages of this process, and researchers, program developers and program managers, as well as practitioners, should continuously put effort into this inter-group communication.
Depending on the perspective of researcher or practitioner, how the innovation process looks like, will vary. We may even distinguish a third perspective: the program manager and policy maker. This third party may be interested in monitoring the whole process.

Innovation from the perspective of the researcher
Nutbeam and his colleagues (4) developed a research model to guide the innovation process (see figure 2).
Figure 2: The researchers’perspective: the model of Nutbeam et al (1990)


(Adapted from Nutbeam et al., J Epi Comm Health 1990; 44: 83-89)

After basic research and theory have been studied, the experimental phase can start (mostly with randomised controlled trials). If this research is successful, then a demonstration study can be planned with relatively more emphasis placed on process evaluation. If this phase is successful, then dissemination studies can be planned. The model ends with the continuing management of implementation (if the innovation was successful of course). These stages in the Nutbeam model will be discussed in more detail, using the TNO Ageing Well and Healthily program (a physical activity program for older adults) as an example (see textbox 1).

Textbox 1: The Ageing Well and Healthily program (Hopman-Rock and Westhoff, 2000)
Ageing Well and Healthy is meant for physically inactive older people (65+) and consists of 6 weekly sessions in which they are informed about health behaviour by a (peer-)educator and learn a low-intensity exercise program which is given by an exercise leader.

Basic Research and Theory
Because nobody wants to re-invent the wheel, one needs to search the literature for information on the type of innovation you want to develop and to implement. You also always need to collect information about determinants of behaviour. A number of theoretical models about determinants of individual behaviour have been developed. An integrative model that is frequently used is the so-called ASS model, where A stands for attitude, S for social norms and S for self-efficacy (5). Attitude, social norms and self-efficacy together determine the individual behavioural intention. It is important to realise that this part of the model has a rather good ability to predict the outcome of individual behavioural intention. It is more difficult, however, to predict actual behaviour. The prediction of actual individual behaviour can be improved by taking information about perceived barriers and about specific skills into account. In the field of exercise behaviour the model of Triandis (6), ‘habits’ are also added as an important predictor of actual behaviour. A problem that is often neglected is the relapse of behaviour (7)[1] . Of course the desired outcome is always the maintenance of a certain “beneficial” individual behaviour. With knowledge from several determinants studies, we were able to design an innovative intervention (Ageing Well and Healthily), with group health education and a low intensity exercise program as the two ingredients to help individual older inactive people to adopt a healthier lifestyle.

Experimental studies
The next step in the Nutbeam research model is to carry out experimental studies.
The intervention can be tested in a small randomised controlled study. At this stage, outcome evaluation is the most important feature. For example, in a study of exercise programs, one could choose to measure blood pressure variables, some physiological measures, and blood variables such as cholesterol and glucose. In the Ageing Well and Healthily study this is exactly what was done (8). Respondents were divided in an experimental group and a control group. Various physiological variables were measured at baseline, before the intervention was started. The study participants were also interviewed at home, using standardised health and quality of life measures.
According to the model of Nutbeam, one should also perform process evaluation at this stage. This can be done while the intervention is being carried out. Process evaluation measures the satisfaction of the individual participants with the intervention, and variables such as the participation level and location and travel problems. Using short questionnaires, completed after each session by both participants as well as the group leaders, can carry out process evaluation. In this way, it was discovered that the peer educators of the Ageing Well and Healthily program needed more time for their part of the sessions. Because this type of research is usually relatively expensive because of the intensive outcome evaluation, it is preferable to do this in small groups only.
Researchers should not forget that a program could be tested perfectly well in a laboratory setting; however, the question remains whether the same results will be obtained when the intervention is carried out in the ‘real’ world.

Demonstration study
In the case of Ageing Well and Healthily, the intervention was found to have beneficial effects in the experimental phase and so we continued with a demonstration study. According to the model of Nutbeam, in this phase, relatively little attention is given to outcome evaluation and more attention is paid to process evaluation.
The researcher still plays an important role. We first had to decide where we wanted to perform our demonstration studies. In our case, we chose a few big towns, some middle-sized towns and some relatively small villages. We then approached people in organisations to see whether they would be willing to collaborate with us. Collaboration with TNO was usually considered attractive because we arranged a lot of things, such as a symposium at the start and the end of the project, protocols and money for the organisation of the local programs. We also provided materials, trained peer-educators and trained exercise leaders.
In this type of study, the randomised-controlled approach is not the most appropriate design. We elected a Community Intervention Trial (8,9). It takes a lot of effort to conduct health interventions in a broad area, in the field, and still keep the study under full control of the researchers. As a researcher, you have to be very creative in finding ways to control the study design. We used a time-delay method where one big town, two middle-size towns and one small village served as controls for four other towns and villages of the same size. At the end of the study we were able to control for the most important outcome variables. For the process evaluation we made use of a specially trained interviewer who functioned independently and could ask questions to participants, educators, exercise leaders and program organisers.

Dissemination and Implementation study
As a researcher, after the demonstration phase you hope that your intervention was successful enough to warrant dissemination and implementation through national introduction. In the case of Ageing Well and Healthily, it was two years after the start of the experimental study before we could start the broader introduction of the program. In this stage, collaboration with national partners in the field is necessary. As a researcher, you no longer decide where programs should be carried out. Your task is to disseminate the knowledge and just wait and see what happens (maybe policy makers and program managers can be assisted to monitor this innovation process, but this requires another kind of researcher).
There are various ways to disseminate information and knowledge about your intervention, for example, by books, and training sessions. In our case, we established a co-ordination centre at TNO in Leiden. We published articles in journals for local professionals. In this context, it was important to present the information in an understandable, practical way, and this was not necessarily the same way we would present the information in an article for a research journal. Once is not enough, you have to be very tenacious. In order to do the right things you may use the model of Fleuren et al. (2) to determine the most suitable determinants of innovation. You should think of making flyers, posters, videos; all possible ways to make people aware of your program. In the case of Ageing Well and Healthily, we collaborated with the Dutch Red Cross organisation, with the association for More Exercise for Seniors, and with the national association of Municipal Health Services. We made posters, videos and flyers to support the collaborating parties in dissemination of Ageing Well and Healthily.
It was possible to monitor dissemination ie where programs were carried out, by whom, with how many participants, were they satisfied, did they change their behaviour, was it easy to disseminate, if not, why not, what can be done to make things better, do organisers, educators, and exercise trainers alter the intervention, if yes, why and in what way?
In the case of Ageing Well and Healthily, dissemination of the program took longer than we had anticipated (9). This was due to problems with the costs of training of the peer-educators and the long time that was needed for local organisations to include Ageing Well and Healthily in their yearly budgets for health interventions in this field. After a few years, we noticed that dissemination and implementation were both slowing down. One of the problems was the ‘not-invented-here syndrome’; organisations were reluctant to implement an innovative intervention that was not developed in their own organisations. This was very disappointing because at the start of the project we also used the so-called “linkage model” of Orlandi et al (10) to ensure that key-organisations for implementation were involved from the beginning.

Innovation from the perspective of the practitioner
The practitioner often works in a consensus-based manner: “if I do this, it will work?”. Why should he or she change his or her habits? What will the advantage be? What does it mean for the daily organisation of the work? It can be very disappointing for researchers if practitioners ignore their beautifully designed innovative interventions. In the next figure the linkage model of Orlandi et al. (1990) is described.
Figure 3: The linkage model for Researchers and Practitioners (Orlandi et al, 1990)

According to the linkage model, researchers and practitioners should exchange knowledge on a regular basis. They should know each other personally and know about each other’s policies, practices and innovations. For example, in the case of Ageing Well and Healthily, we cooperated from the start of the whole process with the director of More Exercise for Seniors, the organisation of exercise leaders. To effectively link the two systems, we used tools such as symposia, articles in non-scientific media, videos, brochures and special training sessions (trained exercise leaders could earn a certificate). Not only should the researcher provide the practitioners with information but the practitioners should also provide the researcher with information and feedback. Researchers should read the journals and leaflets of the practitioner’s organisation and listen to the requests for research to be carried out on a certain topic. In the example of Ageing Well and Healthily, it turned out that exercise leaders of More Exercise for Seniors found it very difficult to cooperate with a senior health educator. Organisation of the intervention was perceived as too difficult and too time-consuming. The exercise leaders were not able to integrate the program into their daily routines. The researchers were perceived as too demanding on them.

Diffusion of innovations
To illustrate the process of adoption of an innovation by practitioners, the model developed by Rogers (11) is introduced. In his (retrospective) diffusion model (see Figure 4) you can consider ‘practitioners’ as the exercise leaders of exercise programs.
Figure 4: Diffusion of innovations: the model of Rogers (2003)

At the start of the implementation, the relevant organisations worked with the innovators and early adopters. These are people who are willing to change. After a while, the early majority will adopt the innovation. The most difficult part, however, is to convince the late majority to adopt the innovative intervention. If you can do this, it will speed up the dissemination and implementation of your intervention. The implementing organisation should be prepared that it can take years, before the full implementation of the innovation is achieved and the so-called ‘laggards’ are reached. And if so, it may be questioned if the interventions are still delivered as intended! In the case of the Ageing Well and Healthily program, we have heard that the interventions have been implemented in more places than we originally knew of. Organisations copied the materials and trained the educators and exercise leaders on their own. The exercise leaders that were reluctant to cooperate with health educators found solutions, such as delivering the health education part by themselves. Of course, the model of Roger is a theoretical model, and nobody want to be labelled as a ‘laggard’. Most laggards have in fact good reasons to be reluctant and will find their solutions to deal with their problems as just described. By now, we have the biggest problem at the tail end: people who have changed the innovative interventions so that these are NOT used as intended!
At TNO we made use of this knowledge and changed the Ageing Well and Healthily program in the desired direction. We are currently busy making all the health education materials easily available using the Internet, to prevent wrong materials being used. We also started training of exercise leaders in delivering health education to older people. Hopefully, this will ease the use of the intervention. As researchers however, we have a problem now: does this adapted intervention still work? To answer this question, researchers may use a framework for monitoring the innovation process called the RE-AIM framework.

Monitoring dissemination and implementation using the RE-AIM framework
A relatively new development in the area of innovation research is the RE-AIM framework of Glasgow et all (12). They identified the following indicators, which should be monitored to get an indication of the innovation process.
These are: Reach, Effectiveness, Adoption, Implementation and Maintenance (see figure 5).
Figure 5: Description of the RE-AIM framework for monitoring of dissemination and implementation of health educational interventions (Glasgow et al, 1999; see website www.re-aim.org)

We are currently using these indicators in our monitoring research regarding the implementation of two innovative physical activity programs for people with osteoarthritis of the hip or knee (13). Both programs are in a re-invention process by the Royal Dutch College of Physiotherapists that will implement these programs among physiotherapists in the Netherlands. We hope to get the answers of this research after a few years of implementation in the field.

Discussion
There is still a long way to go before effective health promotion and innovative physical activity programs are well implemented in the community or health care organisations. It requires substantial financial resources to guide these processes. The whole process starts with the identification of a problem by researchers, practitioners and program managers. Systematic introduction and careful monitoring of the whole innovation process is recommended, in order to identify pitfalls and challenges. Two kinds of researchers are needed: the ones that use the models of Nutbeam and Orlandi to develop and evaluate innovative programs, and researchers who are specialised in guiding and monitoring the whole innovation process after a program is found to be effective in terms of health variables. The model of Fleuren et al. could be a helpful tool to identify determinants of this innovation process and the RE-AIM framework is useful for monitoring, if an effective innovation has be designed and is ready for implementation.
Future research in this area may use the theoretical and practical tools and insights described here, in order to face the “implementation challenge”.

Acknowledgement
Dr. Margot Fleuren is gratefully acknowledged for the critical reading of the first draft of this article.
This paper is an adapted version of:
Hopman-Rock M. The effective introduction of programs for adapted physical activity for older people: what do we know and what can we learn? In: Bergland A, Langhammer B, red. Adapted physical activity and ageing: an interdisciplinary European challenge. Oslo: Oslo University College, Faculty of Health Sciences, 2006:p. 26-41.
[1] mind the difference between individual maintenance of behaviour and maintenance in the context of the innovation process.


References
1. Paulussen, T.G.W. (1994). Adoption and implementation of AIDS education in Dutch secondary schools. Dissertation. Utrecht: Landelijk Centrum GVO.
2. Fleuren, M., Wiefferink, K. & Paulussen, T. (2004). Determinants of innovation within health care organizations. Literature review and Delphi study. Int J Quality in Health Care; 16(2):107-123.
3. King, L., Hawe, P. & Wise, M. (1996). From research into practice in health promotion: a review of the literature on dissemination. Australia: National Centre for Health Promotion;. ISBN: 1 86451 2288.
4. Nutbeam, D., Smith, C. & Catford, J. (1990). Evaluation in health education: a review of progress, possibilities, and problems. J Epid Comm Health;44:83-89.
5. De Vries, H., Dijkstra, M. & Kuhlman, P. (1988). Self-efficacy: the third factor besides attitude and subjective norm as predictor of behavioral intentions. Health Edu Res;3: 273-282.
6. Valois, P., Desharnais, R. & Godin, G. (1988). A comparison of the Fishbein and Ajzen and the Triandis attitudinal models for the prediction of exercise intention and behavior. J Behav Med;11(5):459-72.
7. Laitakari, J., Vuori, I. & Oja, P. (1996). Is long-term maintenance of health-related physical activity possible? An analysis of concepts and evidence. Health Edu Res;11: 463-477.
8. Hopman-Rock, M. & Westhoff, M.H. (2002) Development and Evaluation of “Aging Well And Healthily”: A Health Education and Exercise Program for Community-Living Older Adults. J Aging Phys Act;10: 363-380.
9. Westhoff, M.H. & Hopman-Rock, M. (2002). Dissemination and Implementation of “Aging Well and Healthily”: A Health Education and Exercise Program for Older Adults. J Aging Phys Act;10:381-394.
10. Orlandi, M.A., Landers, C., Weston, R. & Haley, N. (1990) Diffusion of health promotion innovations. In: Glanz, K., Lewis, F.M. & Rimer, B. (eds). Health behavior and health education: theory, research and practice. San Francisco: Jossey-Bas. p. 288-313.
11. Rogers, E.M. (2003). Diffusion of Innovations. New York: The Free Press; (5th ed).
12. Glasgow, R.E., Vogt, T.M. & Boles, S.M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. Am J Public Health; 89:1323-1327.
13. de Jong, O.R.W., Hopman-Rock, M., Tak, E.C.P.M. & Klazinga, N.S. (2004). The Results of an Implementation Study of Two Evidence Based Exercise and Health Education Programmes for Older Adults with Osteoarthritis (OA) of the Knee and Hip. Health Educ Res;19(3):316-25.

Contact
Marijke Hopman-Rock PhD, MSc, MA
TNO Quality of Life, Department of Physical Activity and Health, Leiden, the Netherlands;
BodyatWork Research Center Physical Activity, Work and Health, TNO VU University Medical Center
Amsterdam, the Netherlands
Marijke.Hopman@tno.nl




http://www.icsspe.org/portal/index.php?w=1&z=5