Poor reporting of physical activity and exercise interventions in youth mental health trials: A brief report

Abstract Aim To describe the quality and completeness of the description and reporting of physical activity and exercise interventions delivered to young people to promote mental health or treat mental illness. Methods We conducted a series of scoping reviews identifying 64 controlled trials of physical activity and exercise interventions delivered to young people. We extracted: intervention characteristics, personnel and delivery format, the intensity, duration, frequency and type of physical activity or exercise. Results There was limited reporting of intervention details across studies; 52% did not provide information to confidently assess intervention intensity, 29% did not state who delivered the intervention, and 44% did not specify the intervention delivery format. Conclusions We recommend that authors adhere to the CONSORT reporting requirements and its intervention reporting extensions, (a) the Template for Intervention Description and Replication, (b) Consensus for Exercise Reporting Template and (c) as part of this, detail the frequency, intensity, time and type of physical activity recommendations and prescriptions. Without this, future trials are unable to replicate and extend previous work to support or disconfirm existing knowledge, leading to research waste and diminishing translation and implementation potential.

recently conducted a series of scoping reviews (M. Pascoe et al., 2020;M. C. Pascoe et al., 2020a; M. C. Pascoe et al., Under Review) to examine the evidence for physical activity/exercise as a mental health promotion strategy and treatment approach for mental health disorders in young people (mean age 12-25.9 years). As part of this work, we extracted data from each included study for physical activity/exercise intensity and delivery format. Through this, we found a general paucity of clear specification and reporting of physical activity interventions for mental health promotion and treatment. The characteristics of descriptions and reporting are presented and discussed in the current brief report.

| METHODS
We conducted a series of three scoping reviews (M. Pascoe et al., 2020;M. C. Pascoe et al., 2020a; M. C. Pascoe et al., Under Review) in line with PRISMA-ScR guidelines (Tricco et al., 2018) and following the five stage framework outlined by Arksey and O'Malley (2005). To determine study eligibility, the following criteria were applied to the studies identified in the initial search: population mean age between 12 and 25.9 years; published from 1980 to 2017, included a physical activity or exercise intervention and a comparison or control condition; reported at least one mental health symptom outcome, such as depression symptoms (American College of Sports Medicine, 1998;Norton, Norton, & Sadgrove, 2010); study designs were randomized or non-randomized controlled trials and studies were published in English. Excluded studies were unpublished studies and non-intervention studies.

| Information sources
We conducted searches using 'Evidence Finder' (www.orygen.org.au), which is a comprehensive database of all available published controlled trials and systematic reviews of interventions in the youth mental health field (De Silva, Bailey, Parker, Montague, & Hetrick, 2018;Hetrick, Parker, Callahan, & Purcell, 2010). The searchable database is populated annually using comprehensive and systematic searches of the Embase, MEDLINE, PsycINFO and Cochrane Library databases, coupled with strict and reproducible inclusion criteria to identify studies. It includes research published from 1980 to 2017 and contains all available prevention, treatment and relapse-prevention studies in young people (mean age 6-25 years), across the following mental illnesses: anxiety, depression, bipolar, eating disorders, psychosis, substance use and suicide-self harm. It contains controlled trials (including randomized controlled trials and quasi-randomized studies), systematic reviews and meta-analyses, published in English. Unpublished trials are not included within the Evidence Finder. The following criteria were applied to the search engine (https://www.orygen.org.au/Training/Evidence-Finder): (a) mental health or substance use problem: 'all'; (b) stage of illness: 'all'; (c) treatment/intervention: 'complementary and alternative interventions', followed by 'Physical activity/exercise'; (d) publication date: 'all'.
Title/abstract and then full text screening were independently undertaken by at least two authors. Data charting (Arksey & O'Malley, 2005) was undertaken by a single author using a specifically designed extraction form and was checked by a second author. Reference lists of identified literature were searched for suitable primary research based on titles in the first instance, and if relevant abstracts and full text review.
Title/abstract screening was undertaken by two authors (Michaela C. Pascoe and Alan P. Bailey). Full texts were independently reviewed by two authors (Michaela C. Pascoe, Alan P. Bailey and Melinda Craike). There were no conflicts. Data charting (Arksey & O'Malley, 2005) was undertaken by a single author (Michaela C. Pascoe) using a specifically designed extraction form (Table 1) and extraction was checked by a second author (Tim Carter and Alan P. Bailey).
Data were obtained directly and only from the published articles.
Two authors reviewed each study for objective (heart rates [HR], %maximal HR, %HR reserve, %1-repetition maximum, percent of maximal-oxygen-uptake [%VO 2max ]) and subjective (ratings of perceived exertion) measures of exercise intensity, and classified intervention intensity for aerobic (Norton et al., 2010) and resistance exercise (Garber et al., 2011). Where interventions were poorly described, we attempted to estimate an exercise intensity based on the compendium of exercise energy expenditure (Ainsworth et al., 2000). A critical appraisal of individual sources of evidence was not conducted in the current scoping review.

| RESULTS
We identified a total of 64 controlled trials that investigated physical activity as a treatment for mental illness or mental health promotion strategy among young people. As shown in Table 1, we found a general paucity of reporting of the characteristics of the physical activity interventions. We found that only 31 (48%) provided sufficient information to accurately assess the intensity of the intervention with regards to objective (HR, %maximal HR, %HR reserve, %1-repetition maximum, %VO 2max ) and subjective (ratings of perceived exertion) measures of exercise intensity (Garber et al., 2011;Norton et al., 2010). We were unable to determine the intervention intensity in 33 (52%) of the studies. Only 51 (80%) of studies stated the physical activity or exercise type (ie, aerobic/resistance and/or form, that is, yoga, tai chi).
The delivery format was also poorly described; 19 (29%) did not state who delivered the intervention; 28 (44%) did not specify if the intervention was delivered in a group or individual format; 9 (14%) did not report the intervention dose as defined by both duration and frequency.

| DISCUSSION
This report highlights an important limitation in the literature and a need for better specification and clearer reporting of physical activity interventions for mental health promotion and treatment for youth. Whether the type and intensity of the physical activity engaged in is prescribed or self-selected also should be specified. Autonomy is proposed as one of three basic psychological needs fundamental to positive mental health (Craft, Perna, Freund, & Culpepper, 2008;Ryan & Deci, 2000) and participants experience a greater psychological tolerance to higher intensity physical activity when intensity is self-selected, rather than imposed (Ekkekakis, Parfitt, & Petruzzello, 2011). The environment in which the intervention is delivered should also be clearly stated, given that there is evidence indicating that physical activity undertaken in natural environments may have a more positive mental health benefit compared to physical activity undertaken indoors (Coon et al., 2011). Finally, it is important to report who delivered the intervention, as our scoping reviews

| Reporting recommendations
We recommend that in addition to adhering to the minimum reporting We also acknowledge that self-selected and therefore variable physical activity interventions, as well as behaviour change interventions aimed to increase physical activity and exercise, may be unable to meet the recommended reporting requirements. Most of the current recommendations apply to prescribed exercise/physical activity interventions, however, self-selected variable physical activity interventions, as well as behaviour change interventions, could still measure changes in engagement in physical activity levels and report on this as an outcome (objective and subjective measures of physical activity).
Poor reporting prevents replication and delays progress in the research field as replication of previous work is important to confirm or reject existing knowledge and is the building block to extending current knowledge by adding, subtracting and integrating new components in subsequent trials (T. C. Hoffmann et al., 2013). Poor reporting creates research waste either through inadequate trial replication and extension, or through lost opportunity to implement findings into practice (Glasziou et al., 2014;T. C. Hoffmann et al., 2013). Finally, poor reporting ultimately creates an implementation gap as practitioners and the young people they serve are unable to take timely advantage of potentially effective interventions (Deenik et al., 2019; T. C. Hoffmann et al., 2013).
The method of delivery also needs to better reported, including detailed information regarding who delivered the intervention and their level of training, the environment in which the intervention was delivered, if the interventions were self-selected based on preference or prescribed, and delivered individually or in groups. Improved reporting of physical activity/exercise interventions could shed light on whether minimum dosages are required to improve mental health outcomes in young people, or whether interventions promoting messaging such as 'move more, sit less' and 'more is probably better' are sufficient (Teychenne et al., 2020). This will directly benefit both clinical prescriptions of physical activity/exercise and public health guidelines for mental health.

ACKNOWLEDGEMENTS
We would like to acknowledge our recently deceased co-author, Professor Nigel Stepto, who made a significant contribution to this work and to the field of physical activity, exercise and wellbeing, more generally. He will be remembered warmly.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.