INTEGRATIVE REVIEW: BEHAVIORAL INTERVENTIONS FOR PHYSICAL ACTIVITY PRACTICE

This study aimed to carry out an integrative literature review on the effectiveness of interventions in physical activity (PA) practice in the general population. The search was carried out in articles indexed in online databases: Scopus, CINAHL and Medline. Studies in English or Brazilian Portuguese were included, with evidence levels 2 or 3, published between 2004 and 2008. The final sample consisted of 14 studies. In 57.1% of the studies, interventions were effective for behavior change to practice PA. The diversity of target populations, assessment instruments and intervention designs makes it difficult to compare results and build evidence on the effectiveness of interventions for PA promotion.


INTRODUCTION
Regular physical activity (PA) results in systemic benefits, such as a lower heart frequency, increased cardiac debit and decreased blood pressure.
The effects of physical exercise are not only related to the biological sphere, but also to its positive effects on psychological and psychosocial health (1) .Adherence to physical exercise programs aiming for health promotion is low in the general population, representing a major public health concern.Countless personal and environmental factors are implied in how demographic, biological, psychological, social and physical factors are related to the program.Strategies need to be developed that make people adopt a more active lifestyle, knowing, preventing and/or controlling the risk factors present in their lifestyle (2)   .Changes in health behavior result from reciprocal relations between the environment, personal factors and behavioral attributes (3) .Considering the importance of adopting a more active lifestyle, nurses, as one of the categories responsible for educative activities with the healthy or ill population, need to develop, put in practice and assess the effectiveness of interventions aimed at optimizing regular PA practice and, consequently, a healthier lifestyle.

OBJECTIVES
This study aimed to carry out an integrative review of Brazilian and international literature about the design and effectiveness of interventions used to stimulate PA practice in different population groups.

METHODS
The steps recommended by literature were used to elaborate the integrative review (4-5)   .

Study categorization, analysis and interpretation
A specific instrument was used to assess articles for inclusion in reviews, which Ursi constructed and validated in an earlier study (21) .The analysis of the articles was based on the concepts of quantitative research with experimental and quasi-experimental designs (22) and on literature about the research theme (5)   .
In half of the studies, the intervention was based on a theoretical framework: 28.6% was based on Motivational Theories and 21.4% on Multiple-Stage Models (Table 3).A majority (71.4%) used the isolated increase in PA as the dependent variable, with walking as the most explored target behavior, followed by the association between PA and healthy eating (28.6%), and between PA, healthy eating and giving up smoking (7.1%).The mean duration of interventions was 37.9(±23.3)weeks, with a minimum duration of 8 and a maximum of 72 weeks.The methodology was considered appropriate in 42.8% of the studies (10- 11,15,19-20) , considering criteria as viability, reproducibility, method clarity and application of instruments and objective measures; in 14.3%, the design of the steps was not clear, impeding reproducibility (13,16)   ; and, in 28.6%, no criteria were mentioned for including/excluding subjects.In data analysis, parametric (42.8%) and variance and covariance (42.8%) models predominated.

DISCUSSION
The findings evidence that most of the publications on PA interventions has been disseminated in journals from the Health Science Area, classified in different subjects.These findings are consistent with other reviews (23)(24) .
The mean follow-up period coincided with the duration of the intervention, indicating the need for longitudinal designs, with a view to evidence on the long-term effectiveness of the intervention.
In half of the studies analyzed, strategies were theory-based, especially motivational theories.
However, among the 57.1% whose results evidenced a significant increase in PA, strategies had been theory-based in only 28.6%, corresponding to a mere 14.3% of the sample.Nowadays, there is a largescale debate among experts about the utility and perceived barriers in the application of behavioral theories (23)(24)(25) .These theories point towards a generalized and carefully interpreted systematic summary of empirical evidence on behavior, and its application is expected to improve the effectiveness of interventions for behavioral modification.It has been argued, however, that most theories offer important support as to which needs should be changed, and not how these changes can be induced (26)   .Hence, it would be important for theories, besides explaining the subject's motivation to adopt the behavior, to advance in knowledge about how behavioral determinants of change can be modified and how these determinants can be translated into methods, strategies and effective instruments for behavioral change (26)   .The diversity of the target population, the intervention methods and assessment measures used, as well as the lack of standardization in self-reports on PA in the studies under analysis, limit the identification of which of strategies' potential attributes are associated with the effectiveness of an intervention.1063 Among the 57.1% of studies with a significant increase in PA, intervention strategies comprised counseling, written information/orientations and didactical material, either separately or jointly, with substantially different application/follow-up periods.These disparities do not permit comparisons and generalization of findings and, consequently, make it more difficult to construct evidence on the effectiveness of the interventions for PA promotion.Although this review involved a limited number of studies, its findings indicate the short-term effectiveness of interventions for PA promotion, in line with results from earlier reviews (24,26) .The small number of interventions incorporated into the (health and education) service routine is highlighted, which reveals a continuing gap between theoretical conception and practical application.Interventions with a multidisciplinary approach were also rare, directed at the family's involvement in the adoption of PA practice.

Study limitations
Limitations were related to the exclusive use of DeCS and MeSH descriptors to locate the articles in the electronic databases, which can explain that a limited number of studies was recovered during the period, and also that studies interventions to promote physical activity in the general population?Sample selection Articles in English or Brazilian Portuguese were considered eligible if published in journals indexed in Medical Literature Analysis and Retrieval System on-line (MEDLINE, PubMed version), Cumulative Index to Nursing and Allied Health Literature (CINAHL) and SCOPUS, between August 2004 and August 2008, departing from the descriptors Intervention Studies, Motor Activity and Behavior, according to the Medical Subject Headings (MeSH), and their equivalents in Brazilian Portuguese according to the Health Sciences Descriptors (DeCS).The search was carried out in the three databases at the same time, in September 2008, using the descriptors combined by the Boolean connector "AND".As a result, 102 articles were found: 64 in SCOPUS, 26 in MEDLINE and 12 in CINAHL.The titles and abstracts of these 102 articles were carefully read, 40 were excluded because they were inadequate for the guiding questions, 3 were duplicates (found in more than one database), 15 were not available online in their full version in the collection of the University's Library System and 16 were meta-analyses or reviews.Twenty-eight articles were selected for analysis of their full version, including articles with evidence level 2 (individual research with experimental design) or 3

a c c e
s s e d e l e c t r o n i c a l l y w e r e i n c l u d e d , w h i c h restricted access to all studies elected for the review.CONCLUSIONS This integrative review about the use of behavioral interventions in the period from August 2004 to August 2008 evidenced that, in 57.1% of the studies, interventions effectively promoted PA; of these, however, intervention strategies had been based on theory in few studies.The range of assessment instruments and interventions limits comparisons and result generalizations and make it more difficult to construct evidence about the effectiveness of interventions to promote PA in the short and long terms.

Table 1 -
List of PA measurement instruments, tests and variables measured in the studies included in the integrative review.Campinas, 2009 Continue...

Table 2 -
Continuation( * Wellness, Academics and You Program; † Promoting Activity and Changes in Eating)