33 Risk factors: a systematic review 2 METHODS The protocol of this study was based on the PRISMA statement.32 The study was registered in PROSPERO international prospective register of systematic reviews during the data extraction phase (registration number CRD42020167173). Search strategy and information sources A comprehensive search was performed using the bibliographic databases PubMed, Embase.com, Scopus, the Cochrane Library, Cinahl (via Ebsco) and PsycINFO (via Ebsco) from inception to 15 August 2019 in collaboration with a medical librarian (Linda J. Schoonmade, Annelienke M. van Hulst and Shosha H.M. Peersmann). Search terms included controlled terms (MeSH in PubMed, Emtree in Embase, Thesaurus terms in Cinahl and PsycINFO) as well as free text terms. The following search terms were used (including synonyms and closely related words) as index terms or free-text words: “ALL” and “children” and ‘steroids” and “adverse effects” or “APR” or “sleep problems.” The search was performed without date or language restrictions. Duplicate articles were excluded. The full search strategy for all databases can be found in Appendix 1. In addition, reference lists of all included studies and relevant reviews were manually searched (cross-reference check) for potential additional studies by two authors (Annelienke M. van Hulst and Shosha H.M. Peersmann). Eligibility criteria and study selection All studies were independently screened by two researchers (Annelienke M. van Hulst and Shosha H.M. Peersmann). First, studies were screened on title and abstract using reference program Rayyan.33 Studies that met the following predefined inclusion criteria were included: (a) study population of children aged 0–18 years old, (b) diagnosed with ALL, (c) receiving steroids (e.g., dexamethasone, prednisone) as part of their leukemia treatment, (d) including an APR or sleep outcome. All types of outcome measurements (questionnaires, observational, chart review, and actigraphy) were deemed eligible. Studies were excluded if they only entailed adults or animals, were nonpeer reviewed (congress abstract/poster), only reported neurocognitive measures or nonacute behavioral or sleep outcomes (late effects). Second, full-texts were screened and included if any of the risk factors of behavior and sleep mentioned above were evaluated. As stated before, risk factors that were not predefined could also be included. Studies were excluded if no original data was reported (reviews), it entailed a duplicate, a case report (series) or if full-text was unavailable. Case reports and relevant reviews were set aside to check references. In addition, articles that reported on outcomes of ALL trials were kept apart, as these articles were not designed to meet
RkJQdWJsaXNoZXIy MTk4NDMw