Descriptive analyses of the trustworthiness of healthcare trials and reviews abstracts: a systematic review protocol

• Grant number: 18/17932-7
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate (Direct)
• Start date: January 14, 2019
• End date: March 13, 2019
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Leonardo Oliveira Pena Costa
• Grantee: Dafne Port Nascimento
• Supervisor: Maurits W. Van Tulder
• Host Institution: Pró-Reitoria de Pós-Graduação, Pesquisa e Extensão. Universidade Cidade de São Paulo (UNICID). São Paulo , SP, Brazil
• Institution abroad: University Amsterdam (VU), Netherlands
• Associated to the scholarship: 16/17853-4 - Analyses of reporting quality and accuracy of abstracts of randomized controlled trials and systematic reviews in Physiotherapy and other health care areas, BP.DD

Abstract

Background: Abstracts of randomized controlled trials and systematic reviews should be clear, concise and should reflect exactly what was accomplished in the study. Studies in different healthcare areas assessed their abstracts and found discrepancies such as omission of information, inconsistent/inaccurate information with the full-text, low reporting quality of abstracts and spin of results/conclusions. This PhD proposal comprises three studies related to reporting quality, spin of information and accuracy of abstracts of randomized controlled trials and systematic reviews in physiotherapy and other healthcare areas, which are: 1) Description and analyses of abstracts of randomized controlled trials in physiotherapy for low back pain; 2) Description and analyses of abstracts of systematic reviews in physiotherapy for low back pain; 3) Systematic review comprising all studies that summarized the accuracy, reporting quality and spin of information of healthcare abstracts of randomized controlled trials and systematic reviews.Method: The first two studies were already completed, with methodology as follows: 1) A random sample of 200 randomized controlled trials in physiotherapy for low back pain, published from 2010 to 2015, indexed in PEDro database was analyzed. 2) The full sample of 65 systematic reviews of physiotherapy for low back pain in PEDro, published from 2015 to 2017, was also analyzed. Information regarding study characteristics, reporting quality, spin of information and accuracy of abstract and full-text was extracted. Primary outcomes were based on items fully reported from CONSORT and PRISMA recommendations for abstracts, besides items that indicate the presence of spin in the results and conclusion sections. Abstract information was compared with information in the full-text. Article and journal's descriptive data were collected to analyze whether factors such as, journal's impact factor, open access and endorsement of CONSORT/PRISMA recommendations, were associated with the reporting quality of abstracts. Statistical analyses were descriptive using frequency in percentage, besides linear regression analyses. The third study will be included in the fellowship program, which is: 3) A systematic review comprising all studies that summarized the accuracy, reporting quality and spin of information of healthcare abstracts of randomized controlled trials and systematic reviews.Fellowship program proposal: We would like to run the third study in collaboration with the VU University, with methodology as follows: The systematic review (please see 'Protocol 1' attached) of studies analyzing healthcare abstracts (randomized controlled trials and systematic reviews) will be conducted through PubMed, MEDLINE, CENTRAL, Cochrane Library and EMBASE databases, in July 2018. Summarized data about abstracts reporting quality, spin of information and accuracy with the full-text will be extracted and analyzed descriptively.Expected results: Results of this PhD proposal may encourage authors in better reporting and interpreting of abstracts of randomized controlled trials and systematic reviews. Furthermore, journal editors and reviewers must take actions to improve the trustworthiness of abstracts upon the review process, before publication. (AU)