Improving myalgic encephalomyelitis population sampling: Applying an online respondent-driven method to address...

Area of Science:

• Medical informatics
• Epidemiology
• Public health

Background:

• Health register data, specifically code G93.3, is widely used for myalgic encephalomyelitis (ME) research.
• However, issues with late and under-diagnosis mean G93.3 data may not accurately represent the true prevalence or demographics of ME.
• It is also uncertain if all individuals with a G93.3 code meet established diagnostic criteria like the Canada Consensus Criteria (CCC).

Purpose of the Study:

• To develop and apply a novel methodological approach to address selection bias in estimating the characteristics of the CCC-defined ME population.
• To assess the potential bias in G93.3 diagnosis by examining sociodemographic factors in relation to G93.3 status.

Main Methods:

• An online respondent-driven sampling approach was employed to recruit participants.
• Validated DePaul University algorithms were used in the analysis.
• Sociodemographic and medical factors were regressed against G93.3 status in a sample of 660 respondents.

Main Results:

• The study's findings suggest that G93.3 health register data are biased.
• Specifically, the data appear to be biased against individuals from socially deprived backgrounds.
• This indicates that G93.3 codes may underrepresent certain demographic groups within the ME population.

Conclusions:

• Health register data coded as G93.3 are not an unbiased sampling frame for myalgic encephalomyelitis (ME) prevalence studies.
• The findings highlight significant selection bias, particularly against socially deprived populations.
• A respondent-driven sampling approach combined with validated algorithms offers a method to better estimate characteristics of the CCC-defined ME population.