Appendix C: the evidence
This appendix sets out a summary of the key behaviour change theories (empirical, theoretical and methodological) and other, additional evidence used to inform the recommendations. It also sets out a brief summary of findings from the economic appraisal and the fieldwork.
The reviews, economic appraisal and fieldwork report are available.
The reviews were unable to capture all material related to behaviour change. This is because the evidence is broad, the methods used are diverse and the assumptions made about science, knowledge and explanation vary considerably. Some evidence focuses on particular components of human actions, much is theoretical, and some consists of models of human behaviour (see also appendix B). The Programme Development Group (PDG) has also, therefore, drawn on a range of theoretical and methodological evidence. This evidence is briefly outlined below.
Antonovsky argued that there are 'health-giving' or 'health-generating' factors in many situations. These 'salutogenic' factors can help people withstand or respond positively to stressors, crises or difficulties. They help to protect against vulnerability and disease and may help maintain good mental and physical health. Lazarus argued that people develop habitual ways of coping with life. However, although they may be highly effective from the individual's point of view, some coping mechanisms (like smoking or excessive alcohol consumption) may damage their health and the health of others. Behaviour change and readiness to change behaviour takes place in this context.
Bourdieu argued that many of the things that people do and believe are so familiar and habitual that they go largely unnoticed (because they are part of their 'habitus'). This makes changing them very difficult.
Giddens argued that society was the product of interaction between individual human behaviour and the social structure. He argued that the human actions or agency produce societal patterns. The patterns repeat themselves to such a degree that structures emerge. Although those structures change, sometimes gradually, sometimes rapidly, individuals are aware of them and orient their actions in line with them (and are constrained by them).
The Theory of Planned Behaviour (TPB) is the most widely applied model of beliefs, attitudes and intentions that precede action (Ajzen 2001; Conner and Sparks 2005). TPB proposes that intention is the main determinant of action and is predicted by attitude, subjective norms and perceived behavioural control (PBC). PBC is a person's perception of whether or not they can control their actions and is closely related to Bandura's construct of self-efficacy (1997). Both PBC and self-efficacy are likely to bolster intentions and sustain action because people are more likely to attempt actions that are controllable and easy to perform.
The PDG drew on other sources for a general understanding of wider public health issues. These included:
The former Health Development Agency's evidence base.
Conner M, Norman P (2005) editors. Predicting health behaviour: research and practice with social cognition models. Maidenhead: Open University Press.
The health economic analysis compared and contrasted the cost-effectiveness of behaviour change interventions aimed at reducing coronary heart disease (CHD) and delivered across the life course. Two phases were completed. The first involved a review of the cost-effectiveness of interventions designed to promote healthier lifestyles and to reduce the risk of developing CHD. In the second phase, a model was developed to determine the cost effectiveness of a population-based behaviour change intervention.
Many interventions aimed at tackling multiple risk factors fell into the 'likely to be very cost effective' category (£0 to £20,000/per cost per quality adjusted life year [QALY]). These included a mix of population-level and individual interventions for adults over the age of 30.
Interventions aimed at changing the behaviour of adults with specific CHD risk factors (such as smoking, poor diet and low levels of physical activity) fell into the 'likely to be very cost effective' category. Two non-advisory interventions (labelling of foods containing trans-fatty acid and a population-based programme promoting a healthier diet) also fell into the 'likely to be very cost effective' group.
Significant gaps in the evidence were noted. There was little evidence on the cost-effectiveness of using behaviour change interventions with specified sub-groups (for example, 19- to 30-year-olds, low-income groups, pregnant women, and particular ethnic or disadvantaged groups). The quality of evidence was also a cause for concern. For example, there was a lack of reliable data from which to extrapolate the long-term health outcomes. In addition, only a limited number of economic evaluations had been conducted alongside RCTs of behaviour change interventions to reduce CHD.
In the second phase, a deterministic Markov chain simulation model was developed of a population-wide intervention to lower cholesterol and prevent CHD. The intervention was carried out in Norway in 1990. It included a mass media campaign and information delivered to a range of sectors including academia, the agricultural sector and schools. The model extrapolated the results to England and Wales in the first decade of 2000.
In the base case, an incremental cost-effectiveness ratio (ICER) of £87 per QALY (£116 per life year) was estimated. However, it was noted that the health benefits were underestimated, as this model only reported those related to CHD. Sensitivity analysis estimated that the intervention would be highly cost effective in a wide range of situations.
Fieldwork aimed to test the relevance, usefulness and the feasibility of implementing the recommendations, and the findings were considered by the PDG in developing the final recommendations. The fieldwork was conducted with commissioners, service providers and practitioners involved in a wide range of services and activities relevant to health-related behaviour change. For details, see the fieldwork report on generic and specific interventions to support attitude and behaviour change at population and community levels.
Fieldwork participants were fairly positive about the recommendations and their potential to support attitude and behaviour change at the individual, community and population levels.
The recommendations were seen to reinforce aspects of a range of government policies and initiatives, including providing support to achieve certain public service agreement (PSA) targets (for example, to reduce teenage pregnancies and to reduce health inequalities).
While participants did not view the recommendations as offering a new approach, the principles they are based on have not been implemented universally. They indicated that wider and more systematic implementation would be achieved if there was:
clarity about how the recommendations apply to people in different roles
more information about how to implement some of the recommendations
further information on how compliance with the recommendations will be determined.