The 10 Small Steps project demonstrates that a computer tailored, multiple health behaviour intervention can be implemented successfully [22, 23] and improve adherence to healthy behaviours over a 12-month period. The intervention group sustained behaviour changes following a non-contact period of 9 or 12 months. While the dual contact intervention did not result in a statistically significant benefit over the single contact intervention for the former group there was a trend to larger change in some individual behaviours. Importantly, a significant positive change remained in both groups compared to their respective control groups at 12 months.
This intervention trial combined a number of strategies that individually have shown to be important in improving health behaviours in the context of primary care. These include: the use of low-intensity computer-tailored feedback , simultaneous focus on multiple health behaviours , endorsement of GPs [26, 27], and the use of an intervention reinforcement . A direct comparison of the outcomes of the current study with other studies is not possible, as other studies have incorporated some but not all of these elements. For example, a low intensity intervention in a primary care setting for reducing fat and fibre intake was effective after 12 months [29, 30]. Endorsement of the intervention by practitioners was proposed as the possible reason for long term success . A number of primary care interventions applying computer-tailored approaches have demonstrated effectiveness, though they were focused on single risk factors, intensive (including more than one or two sessions with computer-tailored advice) or focused on secondary prevention only [31–33]. Another successful computer-tailored intervention addressed two behaviours simultaneously but was more comprehensive in content .
Many computer-tailored interventions targeting behaviours, show intervention effects declining quickly after intervention completion, despite initial effectiveness [35, 36]. Notably, this was not the case for the '10 Small Steps’ study. At 12 months the participants still showed significant improvements in the health behaviours suggesting that the inclusion of additional strategies alongside computer tailoring, such as the endorsement of GPs, may have added benefits in bringing about and maintaining behaviour change. It is difficult to attribute the positive outcomes observed in the '10 Small Steps’ study to any individual strategy and the synergistic effects of the above proven strategies are likely to be responsible for underlying success in promoting adoption and maintenance of health behaviours in this study.
Most studies evaluating long-term effectiveness of interventions indicate that maintenance of healthy behaviours is difficult. Smoking cessation, for example, has been shown to require a comprehensive and intensive personally tailored approach to help smokers quit in the long term . Similarly, a systematic review of physical activity interventions suggested that additional tailored exercise prescription strategies and booster interventions such as via phone, mail or internet were needed to facilitate long-term (at least 12 months) effectiveness . This lack of intensity in our pragmatic intervention may explain the failure to achieve significant improvements in 'difficult to change’ behaviours such as physical activity, smoking and BMI. However, all intervention and control groups returned positive changes of 3 to 4% in smoking behaviour, which is significant. This indicates that even the 'control group’ received an intervention, simply by completing a health assessment questionnaire sent with approval of their GP, even without additional focused and tailored feedback. Background smoking rates are declining in Australia, but not to this extent in one year. From 1991 to 2004 the prevalence of smoking in Australia fell from 27.1% to 19.0%, a fall of 8.1% in absolute terms, averaging 0.58% per year decline in smoking, which is far less than the observed decline in this study [39, 40].
There were noticeable similarities in dual and single contact groups: increased fish intake, increased use of low fat milk, increased V&F intake and a higher proportion of participants drinking alcohol within guidelines. However, there were some discrepancies between the intervention groups as well: while the single contact group increased the use of spreads other than butter, this was not the case with the dual contact group. The dual contact reduced use of salt which was not the case with the single contact group. It is possible that certain behaviours such as salt intake are habitual and require repeated efforts to change. Habit formation theory posits that habit strength increases as a result of repetition and positive reinforcement and that any type of repetitive behaviour requires decreasing mental effort before eventually becoming habitual [41, 42]. In our study, participants in the dual contact group who reduced salt intake might also have been successful in changing other behaviours during 12 months of intervention. The inconsistencies in the adoption of various behaviours in our study suggest that the role of habit formation in relation to behaviour change needs to be further explored. It is also possible that there is a limit to how much an individual can change and, as a result, individuals may “swap” habits by, for example, giving up one healthy habit to take up another that is perceived as more important. There are studies illustrating the limitations of self-regulatory capacity and the operating of concepts such as decision fatigue, indicating that it might be difficult for an individual to make multiple behavioural changes simultaneously . However, attention to simultaneous or multiple-behaviour change is likely to have a greater impact on public health than sequential or single-behaviour change, possibly by giving individual the autonomy to choose behaviours they perceive as easiest to change. This process can build self-efficacy to change other unhealthy behaviours. Research also suggests simultaneous interventions are more cost effective than sequential interventions which require more resources to repeatedly reach out to participants .
Limitations of our study include the use of a dichotomous scoring system for health behaviours, where relevant sub-threshold change in behaviour remains undetected. Thus our results may have underestimated the real extent of behaviour change. Secondly, the ten component behaviours of the Prudence Score are equally weighted, rather being weighted according to their relative impacts on health. Also, dietary factors are over represented compared to exercise and smoking, which only contributed a single score each. Assessing the impact on morbidity and mortality is beyond the scope of this trial. However, a study employing an equally weighted lifestyle scoring system using all the same items as the Prudence Score (except vegetable and fruit intake and type of spread ) was able to predict mortality in both healthy elderly men and elderly men with established vascular disease [12, 44]. This suggests that the aggregate unweighted score is still a meaningful summary of an individual’s effort to protect their health. As change in health behaviours was examined in two ways (as individual behaviours and as a sum score) it is possible that some significant findings might be attributed to conducting multiple analyses. Thirdly, the use of self-reported data was a potential weakness despite using a previously validated assessment questionnaire . Whilst some behaviour can be monitored objectively, for example physical activity levels, this is problematic for most dietary behaviours, and alternatives are either extremely costly or impractical. However, the same survey instrument was used to assess behaviours before and after interventions. Another limitation of this study is its inability to undertake a cost-effectiveness analysis as it was not one of the planned outcomes of the trial. However, it is important that future studies consider such analysis.
Although the results of this study do not show a major shift in health behaviours, small individual level changes can be meaningful and contribute to reducing the burden of chronic disease on a population level if the intervention is implemented on a large scale. Composite health scores that showed significant improvements are associated with better health [9, 45] and successful aging . The concept of changing health score on the population wide scale is in keeping with Geoffrey Rose’s public health approach, focusing on shifting the distribution of population risk exposure toward a lower mean rather than simply focussing on high risk individuals [47, 48].While the issue of optimal balance between targeted high risk strategies and wider population health strategies is often debated , it is important that these strategies work synergistically. Targeting patients using bio-markers can help identifying high risk population whereas using a simple but comprehensive lifestyle behaviour score such as the Prudence Score can be helpful in addressing determinants of ill health for the entire populations. Such lifestyle scores do not rely on biological tests and are easy to comprehend by the lay public. Moreover, interventions such as 10 Small Steps have the ability to work as a trigger for practitioners to briefly and confidently discuss health risk behaviours with their patients, further enhancing behavioural outcomes.
A key strength of this study was the combination of a number of behaviour change strategies known to be both effective and relevant in the primary care setting. It targeted multiple behaviours in a large sample and assessed for long term change with and without reinforcement. Whilst developing a computer-tailored intervention is initially costly and time consuming, such expert-systems can be cost-effective in the longer term and implementable on a large scale . The simplicity of the intervention itself implies a greater feasibility for translation into practice, a step previously effective interventions have failed due to complexity or effort required to implement. Finally, participants were drawn from general practice and representative of the wider Australian population [3, 50], so the results of this study can be generalised to the primary care setting.