Journal of Health Psychology: Improving myalgic encephalomyelitis population sampling: Applying an online respondent-driven method to address biases in G93.3 register data - Kielland et al Abstract With widespread late- and under-diagnosing, health register code G93.3 data cannot offer an unbiased sampling frame for myalgic encephalomyelitis, complicating prevalence and demographic distribution assessments. It also remains unclear if all G93.3 cases would meet the Canada Consensus Criteria (CCC). This article describes a novel methodological approach to addressing selection bias when estimating a CCC population’s characteristics, applying an online respondent-driven sampling approach and validated DePaul University algorithms. In a sample of 660 respondents, we assess possible bias in the G93.3 diagnosis by regressing sociodemographic factors on G93.3 status, controlling for medical factors. Results support suggestions that G93.3 register data are biased against those socially deprived. https://journals.sagepub.com/doi/10.1177/13591053251325690
I don’t have the capacity to dig into the results, but I want to highlight the substantial communication with patients that the researches did as a part of running this study: Spoiler: Communication with patients Patient communication Engaging with the target group before and during the survey became essential to reduce non-response rates and increase the target population’s willingness to complete the form. Although time-consuming, such communication paid off professionally in improved participation and rich contextualization. It was reassuring to experience how concern and initial criticism turned to support as we took the time to listen to and explain why our somewhat untraditional research approach had methodological rationale and strengths—and admitted its weaknesses. A YouTube informational video explaining the survey approach, purpose, and content to potential participants was viewed 568 times; a shorter excerpt was viewed an additional 81 times. The project’s Facebook page had approximately 1200 followers. The 14 FAQs with answers posted as we received new queries were each shared by different patient groups to sites unavailable to the research team. We had 32 messenger exchanges with ME-affected persons and their caregivers and additional email and phone exchanges. The community shared four types of concerns. The first was related to the broad eligibility criteria for participation—persons with severe fatigue. They communicated that such broad inclusion criteria in earlier studies had contributed to trivializing the experiences and outcomes of those who met the stricter case-definition criteria for ME. Their questions allowed us to underscore that the collected data would later be reclassified on diagnostic criteria. The second category was concerns, some well-founded, that the criteria would exclude certain groups and individuals. Like most surveys, we had no way to obtain participation from the sickest and loneliest, who had no one to help them complete questionnaires. Further, we would miss those with no social contact, physically or online, with others with fatigue because they had no one to recruit them. The third type was concerns that certain segments of the ME population, notably those within the NMEF, would deliberately exclude non-members from participating for ideological reasons. Here, our rationale was harder to explain: We knew that RDS surveys tended to cross subgroup boundaries and converge toward population representativity and that RDS estimators can adjust for in-group recruitment unless there is a bottleneck. However, we were reassured because we could document this happening also in this survey (Figures 1 and 2). The last group of concerns came from persons who would have liked to participate and felt excluded when not recruited. The NMEF regularly publicizes user surveys in which anyone can take part. It was challenging to explain in plain language that getting representative estimates does not require asking everybody but following a systematic approach to data collection. Reassuringly, concerns about possible exclusions came from presumably both “sides” of the debate.
And look at their data availability! The current article is accompanied by the relevant raw data generated during and/or analysed during the study, including files detailing the analyses and either the complete database or other relevant raw data. These files are available in the Figshare repository and accessible as Supplemental Material via the Sage Journals platform. Ethics approval, participant permissions, and all other relevant approvals were granted for this data sharing.
I'm happy this was published. We need methodology papers Though calling ME/CFS "rare and elusive" is not quite the words I'd have chosen, though it seems to be a phrase used in the field so maybe it's just that I'm not familiar with it.
One glance at the authors (Leonard Jason) and you can predict from past publications that this paper is likely to be based on a conscientious approach and it is no surpise at all to discover it is. In examining biases in ME diagnosis it demonstrates the lack of precision involved in ME diagnosis the world over which is a significant hurdle for individual health management outcomes and the development of consistent research cohorts.
I means you can trust them to consider the impact of their work on ME patients with humane sympathy, Prof Jason being one himself. This is in contrast to some psychologists who practice academic careerist approaches and dont care who it hurts as long as they get the money.
ME Research UK: Research in Norway has found that using population health records may significantly underestimate the number of people with ME/CFS, especially amongst underserved groups. Read more: https://bit.ly/4iZQqtq Read more about underserved groups here: https://tinyurl.com/3svc79r
