Article type
Year
Abstract
Background: prevalence estimates are important in public health since they support priority-setting-definition, public health policies, and health technologies assessment. Despite their importance, there is still a lack of standards and uncertainty related to the critical appraisal of this type of study.
Objectives: to describe and classify the components of instruments used to critically assess prevalence studies.
Methods: we searched Medline, Embase, Web of Science and grey literature using terms related to 'prevalence' and 'critical appraisal' in order to identify tools applicable for risk of bias assessment of prevalence studies. After identification of eligible tools, we extracted all items from each of them and classified them into at least one of four predefined domains: 'population and setting', 'outcome measurement', 'statistics' and 'other'. Two review authors independently carried out study selection, data extraction and classification of the components.
Results: we included 30 tools in our review, eight (26.7%) specifically designed to assess prevalence studies and 22 (73.3%) applicable for this purpose. We identified 119 unique items, from which 12 (10.1%) were classified as 'population and setting', 16 (13.4%) as 'outcome measurement', 14 (11.8%) as 'statistics' and 77 (64.7%) as 'other'. There was great variability among items assessed in each tool; not all tools assessed all domains and there was overlap among items in some tools.
Conclusions: we identified a comprehensive set of items useful to assess the risk of bias of prevalence studies. This set can be used to guide the conduct of prevalence studies, the assessment of prevalence studies and the development of tools to critically assess prevalence studies.
Patient or healthcare consumer involvement: prevalence estimates are very important for patients and healthcare consumers because prevalence impacts on health system policies and priority-setting definition. It is important to consider the risk of bias of these estimates during the process of decision making, in order to base health decision making on better judgment.
Objectives: to describe and classify the components of instruments used to critically assess prevalence studies.
Methods: we searched Medline, Embase, Web of Science and grey literature using terms related to 'prevalence' and 'critical appraisal' in order to identify tools applicable for risk of bias assessment of prevalence studies. After identification of eligible tools, we extracted all items from each of them and classified them into at least one of four predefined domains: 'population and setting', 'outcome measurement', 'statistics' and 'other'. Two review authors independently carried out study selection, data extraction and classification of the components.
Results: we included 30 tools in our review, eight (26.7%) specifically designed to assess prevalence studies and 22 (73.3%) applicable for this purpose. We identified 119 unique items, from which 12 (10.1%) were classified as 'population and setting', 16 (13.4%) as 'outcome measurement', 14 (11.8%) as 'statistics' and 77 (64.7%) as 'other'. There was great variability among items assessed in each tool; not all tools assessed all domains and there was overlap among items in some tools.
Conclusions: we identified a comprehensive set of items useful to assess the risk of bias of prevalence studies. This set can be used to guide the conduct of prevalence studies, the assessment of prevalence studies and the development of tools to critically assess prevalence studies.
Patient or healthcare consumer involvement: prevalence estimates are very important for patients and healthcare consumers because prevalence impacts on health system policies and priority-setting definition. It is important to consider the risk of bias of these estimates during the process of decision making, in order to base health decision making on better judgment.