Data Extraction, Quality Assessment, Search Methods
The review team included a doctoral student and two senior professors with consulting by an experienced medical librarian. Variables for QOL and depression/mood were measured by established instruments. Posture was measured by inclinometer, x-ray, chest height, or perceived body postures. As informed by the Shape-Shifter Model,15two levels of search strategy (biologic, behavioral) were employed, guided by the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement.16 Inclusion criteria were: a) posture measured by the standard thoracic kyphosis angle; b) measurement of QOL or depression/mood; c) English language; d) fully published empirical studies with descriptive, cross-sectional, and randomized clinical trial (RCT) designs. Exclusion criteria were: a) surgical posture intervention studies, abstracts, case studies, case reports, dissertations, literature reviews, and qualitative studies. No studies were identified in children with cancer (Figure 2).
First, the existing literature reporting an association between posture and QOL (Biologic Perspective) was searched using PubMed and Ovid Medline with the indexed term for QOL, and the key words postureand thoracic kyphosis (MeSH headings Quality of Life orLife Quality or Health Related Quality of Life orHRQOL ). The initial search included the prior ten years from 2008 to 2018, but due to a low number of results, the time range was increased to the years from 2000 to 2018. The resulting 118 articles were scanned for appropriate titles and abstracts, revealing 8 studies (Table 1). Next, existing research literature investigating the relationship between mood (defined as emotional reactions or depression) and posture (Behavioral Perspective) was searched using Pubmed and Ovid Medline for the prior ten years from 2008 to 2018 using the key wordsposture , thoracic kyphosis , mood ,depression , and anxiety . This resulted in 307 articles. Resulting articles were scanned for appropriate titles and abstracts, revealing 6 studies (Table 2). In summary, 14 total studies met criteria for inclusion in this systematic review.