INTRODUCTION

There is evidence of the association between a sedentary lifestyle and ischaemic heart disease as well as of the protective effect of exercise, regardless of the total amount of physical activity.1-4 This effect increases with the intensity of the activity carried out. A dose-response association is therefore observed in the reduction of the risk of heart disease, with decreases of 4% for every metabolic equivalent (MET) increase in the intensity of physical activity.5 This connection has also been found with strokes, decreasing their incidence rate and improving prognosis.6 Moreover, physical activity modifies cardiovascular risk factors: systolic and diastolic blood pressure7 drops, high density lipoprotein cholesterol (HDL-C) increases and low density lipoprotein cholesterol (LDL-C) and triglycerides8 decreases.

Based on the proven benefits of aerobic exercise, the American College of Sports Medicine (ACSM) and the American Heart Association (AHA) recommend at least 30 minutes of moderate physical activity, 5 days a week, or 20 minutes of vigorous activity, 3 days a week.9 However, in most developed countries, approximately 70% of the adult population does not follow these recommendations.10,11

In order for sedentary patients to become active, and thus to achieve the benefits that physical exercise produces, various strategies have been employed with different results.12-14 Furthermore, to assess the benefits of physical activity, we used, above all, the increase in physical activity, physical fitness and the modification of certain risk factors.13,14 However, its effect on cardiovascular risk, estimated with scales, was not usually assessed. The use of cardiovascular risk scales helps to assess in a comprehensive manner the improvement in the control of cardiovascular risk factors, and it may be useful to monitor the medium- and long-term effectiveness of physical activity.15,16

This study is part of an experimetal program to promote physical activity (PEPAF) which is carried out by general practitioners in their practice. Its main findings were published recently: the proportion of patients who accomplished the recommendations of physical activity was 4% higher in the intervention group.17

The aim of this paper is to evaluate the effect of the aforementioned program on the control of risk factors and on the estimated cardiovascular risk in a population of sedentary patients between 30 and 74 years of age.

METHOD

Design and Study Population

A detailed description of the clinical trial was published recently.17,18 In brief, it constitutes a cluster-randomized clinical trial, which was carried out from October 2003 to December 2005 in 11 health centers across Spain, considering general practitioners as sampling units. The main reason for the cluster design was to avoid contamination, which would have otherwise occurred among the patients of one doctor, since the study is implemented by each of the doctors but the results correspond to each individual patient.

The 15 groups of the network for preventive activities and health promotion (REDIAPP) were invited to participate, provided that at least 4 doctors in each health center would take part. The study protocol was approved by the research ethics committees in each of the participating centers. In the end, 70 general practitioners from 13 health centers belonging to 8 REDIAPP groups agreed to participate. After signing a collaboration agreement, the doctors were randomly placed into the intervention group (PEPAF group) or in the normal care group (control group) at a 1:1 ratio, using a software program that was centralized and stratified by the health centers. The doctors were assigned anonymously to the comparison groups, using random numbers centrally generated by a computer at the Bizkaia Primary Healthcare Research Unit.

Two centers (12 doctors) left before starting the study because of technical problems, and two doctors refused to participate. Finally, 56 doctors (29 randomly placed into the PEPAF group and 27 to the control group) carried out the study in 11 primary healthcare centers.

The general practitioners recruited patients between the ages of 20 and 80 years who did not meet the aerobic physical activity recommendations of the ACSM (30 minutes of moderate physical activity, 5 days a week, or 20 minutes of vigorous activity, 3 days a week).9 To avoid selection bias, the candidates to be assessed by their general practitioner were selected using systematic sampling carried out by research nurses. They were selected from the list of patients referred for consultation according to the time of their appointment. After attending to the reasons for the consultation, the doctors assessed the patient's physical activity using an electronic algorithm. The computer program guided the doctors in the review of the exclusion criteria, which could efficiently preclude participation for patients who carried out regular physical activity, were under emotional stress, were pregnant on had follow-up difficulties. Patients signed the informed consent documents before the baseline measurement. The study was conducted via a web page, designed to assist doctors in monitoring the research protocol and controlling the recruitment process.

For this study, we selected 3698 patients between 30 and 75 years old to be assessed using the Framingham-D'Agostino cardiovascular risk scale,16 which estimates overall cardiovascular risk, both cardiovascular and cerebrovascular, for this type of population group. The sample size is considered sufficient since—accepting an alpha risk of 0.05 and a beta risk of 0.20 in a bilateral contrast, a standard deviation of 5 points in the difference of cardiovascular risk between baseline and final assessment (since there were 56 clusters), and a 0.01 coefficient of intra-cluster correlation (ICC)—1232 patients were to be allocated in each group, ie, at least 2464. As the final number of participants was 2845 and the average number per cluster was 50, the minimum difference to be detected with a potential of 80% and an alpha risk of 5% was 0.65 units. The cluster sample size calculator (Health Service Research Unit, University of Aberdeen, United Kingdom) was used to estimate the sample size.

Finally, it should be noted that the intervention under evaluation was not designed to modify risk factors or reduce cardiovascular risk, but to increase physical activity in sedentary patients. Therefore, this constitutes an analysis of secondary objectives of the research carried out.

Intervention

The doctors assigned to the PEPAF group gave concise advice and educational materials to all patients and offered them an additional 15-minute consultation to prescribe an individualized exercise plan. The patients who accepted and those did not accept this additional offer formed the advice and prescription of physical activity subgroups, respectively.

The control group doctors carried out their usual care tasks and delayed any systematic intervention on physical activity until the end of the study, unless the motivation or the health problems of patients were directly related with physical inactivity. The training received was similar to that of the intervention group concerning the research procedures and the use of the web page to assess physical activity and develop the recruitment process.

Study Variables

The risk factors used by the D'Agostino scale16 included age, total cholesterol, HDL-C and systolic blood pressure (SBP) as quantitative variables, and sex, use of antihypertensive medicines, smoking, and history of diabetes as dichotomous variables. The risk of cardiovascular morbidity and mortality was assessed using the published equation based on the study by Framingham.16

Blood pressure was measured using the OMRON M7® monitor (Omron Health Care, Ukyo-ku, Kyoto, Japan), according to the recommendations of the European Society of Hypertension.19 The level of lipids and blood glucose was measured at the reference laboratory of each center, following the patient's fasting for at least 8 hours. The data of other risk factors and antihypertensive medicines were recorded on a questionnaire provided by the research nurse. Other variables that could constitute potential confusing factors, including employment status and education level, were collected in the same way.

The body mass index (BMI) was calculated as weight (kg) / height (m).2 Waist circumference was the mean of 3 measurements made at the waist (at the midpoint between the last rib and the iliac crest, using a flexible measuring tape), parallel to the floor and measured after inspiration.

The research nurses were unaware of the group assigned to the participants and carried out their work at a physical activity laboratory, taking measurements of the patient (following his or her recruitment) at baseline and at 12 months.

Data Analysis

To describe the characteristics of the patients studied, we used the mean and standard deviation in quantitative variables and percentages in the qualitative ones. To compare the different subgroups we used the #c2 test for qualitative variables. For quantitative variables, we used the Student t test if the qualitative variable pertained to 2 categories or the ANOVA test with the LSD method for later contrasts if the variable belonged to more than 2 categories. The analysis of changes in the variables analyzed between the PEPAF group and the control group was carried out for treatment purposes. To analyze changes in the variables between baseline assessment and assessment at 12 months we used the Student t-test for paired quantitative data and McNemar's test for qualitative variables. A multilevel analysis was performed, using the generalized mixed effects model, in order to make a multivariate adjustment among different patient groups at the patient level, taking into account the data structure: patients grouped by doctors and doctors grouped by medical centers.

As the study variables were not directly related to the intervention evaluated in the clinical trial and both groups' values were similar reduced at 12 months, they were considered to be random and were withdrawn from the analysis. We re-analyzed the data, drawing the baseline value for those who left the study at 12 months, obtaining almost identical results.

To estimate the effect of the intervention carried out between the baseline assessment and the assessment at 12 months, we used the following formula:

Effectiveness = [Assessment of the PEPAF group at 12 months - Assessment of the PEPAF group at baseline] - [Assessment of the control group at 12 months - Assessment of the control group at baseline].

The statistical significance was set at P<.05. The statistical analyzes were performed using the SPSS / PC+ software, version 15.0 (SPSS Inc., Chicago, Illinois, United States).

RESULTS

Of the 4317 sedentary patients who agreed to participate in the study, 3698 were between 30 and 74 years old. Of these, 1915 were randomly assigned to the PEPAF group (51.8%) and 1783 to the control group (48.2%). 1456 and 1389 subjectes, respectively, completed the study (Figure). The patients who left the project were slightly younger (52.6 vs 49.7 years), with a higher prevalence of women (71% vs 64%) and a lower cardiovascular risk (11.48% vs 12.87%).

Figure 1. Flowchart of the PEPAF study (experimental program for the promotion of physical activity).

Table 1 shows the sample's baseline characteristics, without differences between the PEPAF group and the control group, except for differences in age, education level and the proportion of patients with hypertension and dyslipidemia, which were higher in the control group.

Table 2 shows the baseline values for risk factors and cardiovascular risk, with no difference between the 2 groups except in cardiovascular risk, which was higher in the control group (13.12%) than the PEPAF group (12.02%). Table 2 also displays the changes between the risk factors and cardiovascular risk taken at the baseline assessment and at the assessment at 12 months. A statistically significant decrease in systolic and diastolic blood pressure (control group: 2.93/1.81 mm Hg; PEPAF group: 3.35/1.40 mm Hg) and in pulse pressure (control group: 1.15 mm Hg; PEPAF group: 1.94 mm Hg) was noticeable at 12 months compared to the readings taken at baseline in both groups (P<.05). However, the BMI and waist circumference had not changed. Regarding lipids, there was an increase in HDL-C (control group: 1.73 mg/dL; PEPAF group: 2.67 mg/dL) and a decrease in the atherogenic index (control group 0.12; PEPAF group: 0.16) in both groups (P<.05). Tobacco consumption also dropped in both groups (control group: 13.4%; PEPAF group: 13.3%). Concerning the cardiovascular morbidity and mortality risk, estimated using the Framingham-D'Agostino score, we observed a statistically significant decrease at 12 months in both the PEPAF group (0.79 points; 95% CI, 0.22-1.35) and the control group (0.68; 95% CI, 0.13-1.25). Finally, concerning the effectiveness of the intervention, estimated as the improvement of risk factors and cardiovascular risk in the PEPAF group compared with the improvement obtained in the control group, Table 3 shows that there was no statistical significance in any of the parameters evaluated, except for a slight increase in HDL-C (95% CI, 0.08-1.81mg/dL). There were no relevant changes in the cardiovascular risk score after adjustment for age, sex, education level, hypertension, dyslipidemia, and diabetes.

DISCUSSION

This experimental program aimed at increasing physical activity in sedentary patients and implemented by general practitioners in their practice, was not effective in improving the control of risk factors, except for HDL-C, or the reduction of cardiovascular risk in the intervention group compared with the control group. However, it achieved a decrease in systolic blood pressure (between 2 and 3 mm Hg on average), diastolic blood pressure (between 1 and 1.5 mm Hg on average) and pulse pressure (between 1.4 and 1.7 mm Hg) in both groups. It also improved the lipid profile in both groups, with an increase of HDL-C and a decrease in the atherogenic index. Similarly, there was a decline in smoking with regard to the baseline assessment in both groups. Finally, concerning cardiovascular risk as an overall measure of the program's ability to promote physical activity, which was estimated using the D'Agostino scale, there was a significant decrease in both groups in the assessment at 12 months, but without any differences between them. Although the decline is less than 1% of the risk that was considered to be of clinical relevance when estimating the sample size, it is very close to that figure and we must take into account the trend of the risk increasing over time due to age.

For the correct interpretation of these results, it is important to note that this paper does not assess the effect of physical activity on cardiovascular risk but the widely proven effect of a program promoting physical activity on such a risk. The potential benefit of this program on cardiovascular risk depends on its effectiveness in its original purpose: increasing levels of physical activity significantly. However, although the program was effective, the impact on physical activity was limited, and therefore its effect on cardiovascular risk may not be visible.17We should also note that the inclusion of patients in the study and in the measurements carried out could have been a motivating factor that contributed to improving risk factors and cardiovascular risk, since the change observed was similar in both the PEPAF group and the control group. This fact has already been described by Hardcastle et al20 and Elley et al.21 The main objective of Hardcastle was cardiovascular risk reduction and Elley's purpose was the promotion of physical activity, both obtaining similar results. Neither study showed a decrease in the risk factors or cardiovascular risk in the intervention group compared with the control group; however, improvement was observed in both groups. In the study published by Hardcastle et al20 the intervention group showed a decrease in diastolic blood pressure (0.08 to 4.06mm Hg) and in BMI (0.07 to 0.64), compared with that in the control group. The decrease achieved in the study carried out by Elley CR et al21 in New Zealand in systolic blood pressure (intervention group: 2.58 mm Hg and control group: 1.21 mm Hg) and in diastolic blood pressure (intervention group: 2.62 mm Hg and control group: 0.81 mm Hg), without significant statistical differences, was also similar to the results of our study. Results were also similar to those reported by Whelton et al7 in the meta-analysis on the effect of aerobic exercise on blood pressure (an average decrease of 3.8 PA/2.6 mm Hg). The study by Elley CR et al21 also found no decrease in BMI or total cholesterol, but coronary risk improved, estimated at 4 years in both groups (intervention group: 0.42 and control group: 0.52), without differences between them. Finally, Duna et al22 found an effect similar to ours in the decrease in cardiovascular risk factors (blood pressure, lipids, and weight) in 2 groups with different intervention intensities, aiming to promote physical activity. Nevertheless, there was no difference in the degree of decline between them. In summary, the results obtained by other authors—although the main objective of the intervention carried out was not the same in all of the studies—are in line with those found in our study, where, rather than the actual intervention, the improvement achieved both in risk factors and cardiovascular risk may be related to the inclusion of these patients in the study.

Limitations

We should note some of the study's limitations. First, the intervention assessed was not designed to modify risk factors or to reduce cardiovascular risk, but to increase physical activity in sedentary patients. Therefore, the results presented in this study are an analysis of the secondary objectives. Second, we should always bear in mind that there may be some individual variability in the clinical practice that may limit to some extent the homogeneity of the intervention. Third, although they were examined by different doctors, the fact that the patients of both the PEPAF group and the control group were from the same health center and that the measurement was performed by the same nurses, implies that a certain contamination effect cannot be excluded, which explains why risk factors and cardiovascular risk behaved in the same way in both groups. However, following the experts' recommendations, who believe that concerns regarding contamination are often exaggerated,23 it was decided to randomize doctors by centers, because inter-center correlation may increase the study's strength. Lastly, we should take into account the high number of patients leaving the study (23%), in particular young women, thus participants with a lower cardiovascular risk.

CONCLUSIONS

The improvement achieved both in controlling factors and reducing cardiovascular risk is due to the inclusion of these patients in the project. There were no significant differences between the control group and the group that was encouraged to increase physical activity.

ACKNOWLEDGMENTS

Additional participation: We thank Dr Diego Rada Fernandez de Jauregui of the Unidad de Investigación en Atención Primaria de Bizkaia (Bizkaia Primary Healthcare Research Unit) for his help in the analysis of the data.

PEPAF group members:Equipo director (management team): Unidad de investigación de atención primaria (Primary Healthcare Research Unit). Servicio Vasco de Salud-Osakidetza (Vasque-Osakidetza Health Care Service). (Investigador principal [Principal researcher]: Gonzalo Grandes); co-investigadores (co-researchers): Jesús Torcal, Ricardo Ortega Sánchez-Pinilla, Álvaro Sánchez, Kepa Lizarraga, Javier Serra, and Imanol Montoya. Servicio Vaco de Salud-Osakidetza: CS (Health center) Basauri- Ariz (Ángel Fernandez, Víctor Manuel López, Lourdes Marijuan, Begoña Etxeguren, Víctor Landa, Jesús Martínez, María Mercedes Díez, Juan Ramón Lejarza and Judith González), CS Galdakao (Vidal Salcedo, Idoia Ibáñez, Agurtzane Ortego, Pedro Iraguen, Pilar Echevarria and Virginia Villaverde), CS Algorta (Amaia Ecenarro, María Teresa Uribe, Carmen Moral, Eguskine Iturregui, and Ana Belén Fernandez). Servicio de Salud de Andalucía (Andalusian Health Service): CS Camas-Sevilla (José María Páez, María Ángeles Tarilonte, Concepción Molina, Vicente Rodríguez, Isabel Villafuente and Mercedes Álvarez). Servicio de Salud de las Islas Baleares (Balearic Islands Health Service): CS Dalt Sant Joan-Mahon (Andreu Estela, José María Coll, Ángels Llach, Josep María Masuet, Ana Moll and Mónica Pons). Servicio Catalán de Salud (Catalonyan Health Service): CS Serraparera-Barcelona (Bonaventura Bolibar, Agustí Guiu, Amadeu Díaz, Xavier Martínez, María Dolores Hernández, Jose Ignacio Olivares, Francisco Hernansanz, Rita Ayala and Ana Cascos). Servicio de Salud de Castilla-la Mancha (Castilla-la Mancha Health Service): CS San Fernando-Cuenca III -Cuenca (Vicente Martínez, María del Carmen García, María Ángeles Gabriel, María Luscinda Velázquez, Natividad Ortega, María José Segura, Rodrigo Cerrillo and Patricia López). Servicio de Salud de Castilla y León (Castilla y León Health Service): CS Casa Barco - Valladolid (Carmen Fernandez, Amparo Gómez, Miguel Ángel Díez, Ruperto Sanz, Luis Miguel Quintero and Jose Ignacio Recio), CS La Alamedilla Salamanca (Luis García-Ortiz, José Antonio Iglesias, Manuel Angel Gómez-Marcos, Emilio Ramos, Pilar Moreno, Yolanda Castaño and Nadia Carrillo). Servicio de Salud de Galicia (Galicia Health Service): CS Sardoma - Vigo (Pilar Gayoso, Luciano Casariego, Manuel Domínguez, José Ramón Moliner, Fernando Lago, María Concepción Cruces and Marisa Enríquez). Servicio de Salud de Madrid (Madrid Health Service): CS Guayaba-Madrid (Tomas Gómez Gascón, Javier Martínez, José Antonio Granados, María Ángeles Fernandez, María Isabel Gutiérrez, Carlos San Andrés, Concepción Vargas-Machuca and Cristina Díaz).

ABBREVIATIONS
ACSM: American College of Sports Medicine
AHA: American Heart Association
CI: confidence interval
BMI: body mass index
MET: metabolic equivalent
PEPAF: experimental program for physical activity promotion

VÉASE EDITORIAL EN PÁGS. 1232-4

Funding: The Instituto de Salud Carlos III del Ministerio de Sanidad y Consumo de España funded this project. The project was co-funded with European Regional Development Funds (ERDF) of the European Union (FIS PI02/0015, RETICS G03/170 and RD06/0018).

Correspondence: Dr. L. Garcia Ortiz.
Unidad de Investigación. Centro de Salud La Alamedilla. 37003 Salamanca. Spain
E-mail: Lgarciao@usal.es

Received February 9, 2010.
Accepted for publication June 9, 2010.