Unhealthy habits and the associated risk factors are the most important causes of disease and death in developed countries.1 Evidence shows that 65-80% of cardiovascular disease, 75-90% of type 2 diabetes and 20-30% of all cancers could be prevented if the population were persuaded to adopt a healthy diet, take up physical exercise and stop smoking.2,3 Different studies consistently attribute a 50% decrease in mortality and an eleven year increase in life expectance to the adoption of these habits and moderate alcohol consumption.4–6

We believe that the ideal place to implement this type of interventions that address unhealthy habits is primary healthcare (PHC), for accessibility features and continuity. Despite various interventions have shown their effectiveness,7,8 it is a problem unresolved and healthy lifestyle promotion is far from being integrated in routine primary care practice.9,10

For this reason, in 2008, phase I of the action-research project “Prescribe Vida Saludable” (PVS) was piloted in PHC, in order to set up innovative interventions optimizing the promotion of multiple healthy habits (physical activity, healthy diet, abstinence from smoking) consisting of multiple active ingredients based on main theoretical models of behaviour change and fundamentally, in the 5As strategy (Ask, Advise, Agree, Assist and Arrange follow-up), and modelled by professionals in each intervention centre.11 Professionals in four PHC centres adapted evidence based interventions to their real life context, carrying out organizational changes, including community services (schools, multisport centres, town halls…) and developing new information and communication tools, so that these innovative programmes could be feasible, sustainable and potentially effective. In 2010, phase II was piloted. Its aim was to evaluate the potential feasibility and effectiveness of the PVS in four different reference centres following their usual health promotion practice.

This project is included in the “Strategy to tackle the challenge of chronicity in the Basque Country of the Basque Healthcare Service”: Policy II, prioritize health promotion and disease prevention; Policy III, potentialize the active role of the patient, his responsibility and autonomy; Policy IV, guarantee continued assistance by stimulating multidisciplinary assistance, coordinated and integrated in the different services, levels and sectors of care and Strategic Project 14, clinical professional innovation.

Specifically, the World Health Organization recognizes physical inactivity as one of the principal risk factors for morbidity and mortality.12 The lack of physical activity not only contributes to the increase in prevalence of chronic disease such as cardiovascular disease, obesity, type 2 diabetes, osteoporosis, colon cancer, depression and fall-related injuries, but also contributes to the direct cost of healthcare in developed countries by between 1.5% and 3.0%.13

In the usual clinical practice, general practitioners usually recommend an increase in physical activity, since even a moderate increase has been shown to improve quality of life.14 Nevertheless, the rates and prevalence of physical inactivity remain high.15

The aim of this analysis is to evaluate the efficiency of prescribing physical activity in the PVS programme by means of the estimation of its incremental cost-effectiveness and cost-utility, with respect to the usual clinical practice of promoting physical activity in those who consult for whatever reason.

An economical evaluation of the quasi-experimental piloting of PVS from programme perspective was undertaken. In the Basque Country 8 PHC centres participated: 4 control centres, 4 intervention centres. In the intervention centres, 122 patients between 10 and 65 years old who did not meet with the minimum public health physical activity recommendations (at least 30min of moderate physical activity 5 days per week, or at least 20min of vigorous intensity physical activity 3 days per week16) were selected and subsequently received a first physical activity prescription plan between May 2012 and May 2013. In the control centres a systematic sample by age and sex of 194 patients between 10 and 65 years old who did not meet with the minimum physical activity recommendations during the data collection period were selected. Intervention and control patients were measured at baseline and at 3 months. The census-based deprivation index, used as a proxy for the socioeconomic status, was developed for the MEDEA project,17 where the first quintile includes the areas with higher socioeconomic status and the fifth quintile includes areas with low socioeconomic status.

In the intervention centres, having checked the patients’ physical activity, the doctors and nurses provided summary advice and educational materials to the patients and offered an additional appointment in order to prescribe a personalized physical exercise plan. It was assumed that the intervention centres were carrying out the same procedure as the control centres with the addition of the PVS programme promoting physical activity.

In the control centres were selected 194 patients and 122 in the intervention centres. After 3 months, data was obtained from only 69 patients in the control group (35%) and from 76 patients in the intervention group (62%) (Fig. 1). The patients in the control group had an average age of 45 years whereas the average age in the intervention group was 47 years. The proportion of women was higher in intervention than in control, 68.42% and 50.72%, respectively. Statistically significant differences in gender and deprivation exist between the two groups (Table 2).

Figure 1.

Data collection chronology.

(0.12MB).

The cost data is demonstrated in Table 1. The unitary cost of each action carried out in the PVS programme is 2.88 € for assessment, 9.40 € for advice and 13.91 € for prescription. The average time spent is 3minutes for A1, 7.84minutes for A2 and 13.67minutes for A4. 37% of all professionals engaged in A1 are administrative assistants, 27% are nurses and 36% are doctors. 39% of nurses and 61% of doctors are engaged in A2. And finally, those engaged in A4 are 71% nurses and 29% doctors.

In adjusting the utility of physical activity prescription on utility by age, sex, group, deprivation index and adjusted clinical groups, none of these variables improved significantly the fit of the model to data, only the baseline measurement. Table 3 contains the effectiveness results in the proportion of patients who adopted the healthy habit of physical activity and adjusted utility in QALY of both groups, as well as their costs. The patients in the PVS centres incurred an average cost of 37.04 € (SD: 24.24), while the patients in the control centres didn’t incur any costs because the study was undertaken from the perspective of the PVS programme. The average adjusted utility of the patients PVS is 0.25 QALY (SD: 0.09), while the average adjusted utility of the patients in control centres is 0.22 QALY (SD: 0.04). Using these data, the incremental cost (37.04 €) and the incremental adjusted utility (0.03 QALY) were calculated. The ICUR is 1,234.66 €/QALY and the ICER is 4.12 €. Statistically significant differences in adjusted utility (p-value=0.0310) and in cost (p-value <0.0001) exist between the intervention and control groups.

The cost-utility level appears in Figure 2 and enables analysis of the impact of uncertainty of the RCUI. Thus it can be seen that the uncertainty due to a variability of the parameters which affect the utility is much greater than that produced by those which modify the cost.

Figure 2.

Cost-utility plane. Ordinate axe: incremental cost in euros. Abscissa axe: incremental utility per quality-adjusted life year.

(0.14MB).

The threshold of 30,000 €/QALY is represented on the cost utility level by means of the straight line which allows us to distinguish the cases where an intervention is cost effective (Fig. 2). The simulations represented by points on the right of said straight line are considered cost effective, while those on the left are not. The acceptability curve is shown in Figure 3 and represents the percentage of simulations which can be seen on the right of the straight line. The percentage of simulations with a result below the threshold of 30,000 €/QALY is 98.3%, while 94% of the simulations satisfy the criteria of acceptability (threshold) of 5,000 €/QALY. Furthermore, we can observe that the acceptability curve never reaches 100% of the simulations. This is due to the fact that the ICUR of some simulations is found in the upper left quadrant where the programme is at its most costly and least effective, since its incremental utility growth is negative.

Figure 3.

Acceptability curve. Ordinate axe: percentage incremental net benefit >0. Abscissa axe: criteria of acceptability (threshold) in euros by quality adjusted life year.

(0.1MB).

The prescription of physical activity by means of the PVS innovation thus evaluated presents an efficient cost effective balance in accordance with the results of the pilot scheme in this clinical trial. Both the utility as well as the costs are significantly higher in the intervention group and the ICUR is 1,234.66 €/QALY, far below the 30,000 €/QALY. This estimate is robust in the analysis of additional sensitivity. A large number of the estimates appeared in the quadrant with positive incremental costs and effectiveness.

The economic evaluation manuals show that clinical practice is cost effective when its ICUR is below the threshold.29 In this sense, we have to recognize that in Spain a unanimous criterion on what is an efficient health intervention does not exist, but the threshold of 30,000 €/QALY is considered a frontier between efficient clinical practice and inefficient clinical practice.30 The PVS innovation as method of promoting healthy physical exercise is cost effective with a probability of 98.5% for a threshold of 30,000 €/QALY. The curve of acceptability shows that the uncertainty is low, since more than 94% of the simulations are below the 5,000 €/QALY. These results are better than those of other interventions on healthy lifestyle, with a probability of being cost effective no greater than 93% for thresholds higher than 8,000 €/QALY.31 The differences between groups, principally in gender and deprivation index, mean that the groups are apparently not comparable. We adjusted the analysis by these variables which did not contribute significantly to the model fit, less the baseline measurement.

The professional's dedication to each action is of great importance in the estimation of cost given the perspective of the study. The PVS innovation involves the professionals of the centres in different ways. In spite of the fact that both doctors and nurses carry out A1, it is plausible to consider that A1 could almost entirely be carried out by administrative assistants since they are already the first access point to the centre and at the same time to the PVS programme. Therefore, it is possible that a higher cost is being attributed to this action than necessary.

The rate of replies to the quality of life questionnaire is a limitation of the study. In the intervention centres this rate was 62% compared to 35% in the control centres. It is possible that there is a reply bias in the control centres since those patients who respond to the questionnaire are also the ones with a better quality of life.

In spite of all the previously defined limitations, the results are optimistic when considering the conservative perspective of our analysis and taking into account the results of equivalent studies.32,33 A study with a wider perspective and a longer follow-up period would capture greater improvements in physical activity as well as the prevention of numerous chronic diseases which at the moment involve a heavy care and cost burden. Accordingly, the efficiency would be even greater if we look at it from a larger perspective which includes the QALY attributable to physical activity among the benefits and effectiveness: 3% of discapacity adjusted life years are attributed to physical inactivity per year.34

This study opens the door to a more ambitious study where the effectiveness of the PVS innovation on all the healthy life habits −physical activity, diet, tobacco and alcohol− can be put to the test and this from a healthcare perspective which takes into account the health benefits contained in this kind of preventive intervention. These kinds of interventions have the disadvantage that the huge benefits are not immediately evident and that successful efficacy research not implemented in routine practice of PHC naturally.10 Therefore, the PVS is based on research into implementation.

Miguel Ángel Negrín Hernández.

The corresponding author on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

Substantial contributions to the conception or design of the work: A. Sanz-Guinea and G. Grandes. Acquisition of data: M. Espinosa and C. Martinez. Analysis of data: A. Sanz-Guinea and P. Bully. Interpretation of data: A. Sanchez, H. Pombo, J. Cortada, A. Sanz-Guinea and G. Grandes. Drafting the work: A. Sanz-Guinea and G. Grandes. Revising it critically for important intellectual content: A. Sanchez, C. Martinez, H. Pombo, P. Bully and J. Cortada. All authors reviewed and approved the drafting of the final manuscript.

Project “Evaluación Económica de la Innovación Prescribe Vida Saludable (PVS)”. Expedient: PI13/00573. This study has been funded by Instituto de Salud Carlos III through the project “PI13/00573” (Co-funded by European Regional Development Fund) “Investing in your future”).

None.