Small study bias was appraised by graphical inspection of funnel plots and formally through Begg and Mazumdar rank correlation, Egger's regression intercept, and Duval and Tweedie trim and fill [ 14]. 2189 records were identified through database searching, and 38 were appraised at text level and finally twenty-six seedies (see Appendix) were included including 28.178 patients.
A similar review method was used to appraise the full-text articles; a Full-Text Screening Form was created and applied.
Two authors (ZM and WNW or RA-SS) read the title and abstract of every study identified by the electronic searches and critically appraised the full text of potentially eligible studies.
Our objectives were to appraise the scope of text messaging interventions in health care and to develop a framework for transfer of research findings from HIV to NCDs.
Citations were first screened at the title/abstract level and, if potentially pertinent, (that is, containing any direct or indirect reference to H1N1 infection and ECMO), retrieved in full text and appraised according to the following specifications.
Relevant full text papers were appraised for risk of bias using the Quality in Prognosis Studies (QUIPS) tool, that has been developed for systematic appraisal in studies of prognostic factors 17.
Nine thousand five hundred seventy-eight citations were screened, 63 full text publications reviewed, 38 appraised and 30 articles were included.
First, the readers did not access the full-text article to fully appraise the study results; they only assessed an abstract with or without a limitations section.
In searching PubMed and Web of Science, the titles and abstracts were examined and 346 full text publications were fully appraised as they met the following criteria: English language, anal cancer or its screening in the HIV positive population.
Two reviewers (a lay person and a physician) screened full text articles for eligibility, appraised the articles for their tone (worrisome, neutral, not worrisome), and for the quality of medical data reporting.
After having excluded irrelevant studies, two reviewers (CR and KLW) independently evaluated the full text of the remaining publications, appraised the studies and performed data extraction.
