How well described are the interventions with physical activity in randomized clinical trials?

Article type
Authors
Aleman A1, Castellano G1, Tomasso G2, Bello R3, Trujillo J1, Colomar M2, Andrade Miranda A4
1Departament of Preventive Medicine, School of Medicine. University of the Republic
2Clinical and Epidemiological Research Unit Montevideo
3 Hospital General de Agudos “Dr. Ignacio Pirovano”
4Cochrane Argentina, Departamento de investigación del Instituto Italiano de Buenos Aires
Abstract
Background: physical activity is defined by the World Health Organization (WHO) as any bodily movement produced by skeletal muscles that require energy expenditure. Lack of physical activity has been identified as one of the main risk factors for global mortality (6% of deaths globally). An increasing number of studies to assess interventions including physical activity have been published in the last years. In Cochrane Central Register of Controlled Trials (CENTRAL), the number of trials including physical activity in their title has increased from 545 in 2014 to 865 in 2017, more than 50%. However, physical activity interventions require a detailed description in order to be reliably implemented.

Objectives: the objective of this study is to assess the accuracy of the descriptions of physical activity interventions in randomized clinical trials studies published in 2018 and included in CENTRAL.

Methods: we performed a literature search with the term 'physical activity' from inception up to 31 March 2019 in CENTRAL. Included studies met the following criteria:
1) complete primary randomized clinical trials with final results published;
2) a 'physical activity' component in the intervention; and
3) published in English or Spanish in 2018.

All included studies, after assessment of title and abstract for inclusion criteria, were double-blind, full-text screened with TIDieR (Template for Intervention Description and Replication) tool to assess if they reached the required description of the intervention in order to warranted replicability. We report the proportion and 95% confidence interval (CI) of TIDieR items included in the description of the intervention globally and item by item.

Results: the CENTRAL search retrieved 699 studies, 142 studies met the inclusion criteria. Reasons for exclusion, in order of frequency, were: non-complete studies (protocols or conference presentations), no physical activity intervention, duplications or publications in other languages than Spanish or English. Only 3% (95% CI 1 to 8) of the studies included description of all 12 TIDieR items. Items 1 (90%, 95% CI 94 to 83), 2 (85%, 95 CI 91 to 77) and 8 (86, 95% CI 91 to 78) were the most reported and items 9 (30%, 95% CI 39 to 22), 10 (12%, 95% CI 19 to 7) and 12 (26%, 95% CI 35 to 19) were the least reported. These data include results from one reviewer assessment, we will present double-blind assessment in the Colloquium.

Conclusions: even though studies on physical activity interventions have increased in the last years, the description of its characteristics in order to facilitate implementation, is not accurate in a high proportion of studies. TIDieR should make it easier for authors to describe their interventions, and for readers to use the information.

Patient or healthcare consumer involvement: no