The evaluation for quality of evidence of cost or economic outcomes through GRADE: A survey of Cochrane reviews

Article type
Authors
Luo X1, Hu Y2, Zhang D3, Li Z2, Chang C2, Wang J2, Wan M4, Wang Z3, Wei D5, Yang K5
1The First Clinical Medical College of Lanzhou University
2School of Basic Medical Sciences, Lanzhou University
3School of Public Health, Lanzhou University
4The First Clinical Medical College of Lanzhou University, Lanzhou 730000, China
5Key Laboratory of Evidence-based Medicine and Knowledge Translation of Gansu Province; Evidence-based Medicine Center, School of Basic Medical Sciences, Lanzhou University; Chinese GRADE Center
Abstract
Background: Evidence of cost or resource use is significant for formulating the recommendation under the GRADE approach when developing guideline. The Cochrane review routinely evaluates the quality of evidence through GRADE.

Objectives: Aim to investigate the evaluation of quality of evidence on cost or economic outcomes in Cochrane reviews.

Methods:We searched Cochrane Database of Systematic Review using the terms 'economic', 'economics', 'socioeconomic', 'insurance', 'cost', without limitation of the date which was from the inception to 5 January 2016. Two reviewers independently identified the reviews rating quality of evidence of cost outcome by GRADE and extracted the data.

Results: A total of 1215 records were retrieved from Cochrane Database, of these 14 were included published from 2012 to 2015. The number of outcomes focused on was ranged from three to seven; however, each review only reported one cost or economic outcome supported by one to six primary studies respectively such as 'Cost of service utilisation', 'Medicine costs', 'Cost-effectiveness', 'Health economic', 'Health costs', 'Economic outcomes', 'Cost', 'Economic data' and 'Cost of care'. The quality of evidence included one moderate, eight low and five very low.

Conclusions: The cost or economic outcomes were rarely reported and rated by GRADE in Cochrane reviews. Most outcomes were considered with low or very low quality of evidence.