Risk of Bias Evaluation of Cross‐Sectional Studies: Adaptation of the Newcastle‐Ottawa Scale

Abstract

Cross‐sectional studies are widely utilized in medical research to estimate prevalence and examine associations. As such, they can serve as a significant source of data for systematic reviews. However, specific considerations are necessary when evaluating the risk of bias (RoB) of cross‐sectional studies, as several potential biases can undermine the validity, reliability, and robustness of their findings. This article introduces a novel, context‐specific tool designed to assess the RoB of cross‐sectional studies for use in systematic reviews. The proposed tool represents an adaptation of the Newcastle‐Ottawa Scale (NOS), originally developed for cohort and case–control studies. Similar to the original NOS, the new tool (named “NOS‐xs”) features a nine‐star rating system to assess six specific items across three main domains: (i) study sample selection, (ii) assessment of exposure(s) and outcome(s), and (iii) confounding factors. Based on the number of awarded stars, studies are categorized as having high (0–3 stars), moderate (4–6 stars), or low (7–9 stars) RoB. The NOS‐xs tool maintains consistency with the original NOS tool, facilitating its integration into systematic reviews that also include cohort and/or case–control studies. While the NOS‐xs is suited to analytical cross‐sectional studies (i.e., association studies), a simplified version (“NOS‐xs2”) is also introduced for descriptive cross‐sectional studies (i.e., prevalence studies). The NOS‐xs2 features a four‐star rating system to assess three of the six specific items included in the NOS‐xs. To streamline their application, spreadsheets for both NOS‐xs and NOS‐x2 are provided.