Contesting misrecognition online: epistemic injustice in contested illness communities
ME/CFS communities online have a history. For decades, patients with ME/CFS were told their illness was functional, psychosomatic, or a matter of deconditioning and catastrophizing. The medical system systematically dismissed their physiological experience. In response, patients built online communities where that experience was recognized and validated — communities that provided what medicine refused to give. These communities also became major information sources: places where illness models were developed, management strategies were exchanged, and a shared understanding of what ME/CFS is and how it behaves was collectively constructed. A 2023 paper by Groenevelt and de Boer in Social Science and Medicine examines what happens to that shared understanding over time — specifically how the dynamics of online community participation, shaped by the experience of systematic disbelief, can cause the community's collective illness model to drift away from what the research actually supports.
Epistemic Injustice: The Starting Point for Distortion
Groenevelt and de Boer use the concept of epistemic injustice to frame their analysis. Epistemic injustice refers to the systematic discrediting of a person's knowledge claims on identity-based grounds — in this context, the widespread clinical dismissal of patient accounts on the grounds that subjective experience is not reliable evidence and that patients lack the expertise to interpret their own physiological state. For ME/CFS patients, this has meant that describing symptoms, reporting worsening after activity, and claiming physiological causality for their condition were routinely met with skepticism, re-attribution to psychological factors, or outright disbelief.
The experience of epistemic injustice creates a specific social environment in online communities. Members have learned that their knowledge claims are vulnerable — that outsiders, particularly those associated with medicine, will challenge, minimize, or dismiss what they report. This shapes how community members present themselves and their illness, and it shapes the community's collective response to information that echoes the same dismissive logic they have encountered clinically.
Performing Credibility Under Scrutiny
The core finding in Groenevelt and de Boer is that the threat of epistemic injustice — the anticipation of being disbelieved — produces specific presentation strategies. Community members who share their illness accounts in public online spaces learn to pre-empt skepticism by performing credibility. This means emphasizing markers that the community has learned are treated as credible: severe and measurable functional impairment, documented test results, visible suffering, and accounts of deterioration that make clear this is a serious physiological condition and not a lifestyle choice or attitude problem.
Simultaneously, aspects of the illness experience that might invite dismissal get downplayed in public presentation. Partial recovery, functional improvement, positive responses to any form of management that has been associated with the delegitimized psychosomatic model — these are either omitted or heavily qualified, because sharing them in a community that has learned to police the illness narrative invites challenge. The result is that what gets shared publicly is not a representative sample of what community members are experiencing. It is the subset of experience that survives the credibility-performance filter — skewed toward severity, persistence, and the absence of improvement.
How Illness Representation Drifts Over Time
As individuals optimize their presentations for community credibility, the aggregate picture of the illness that circulates in the community shifts. The shared illness narrative — the collective model of what ME/CFS is, how it behaves, what it does and does not respond to — is constructed from the accounts that community members share. Those accounts are filtered through the credibility-performance dynamic. Severity is overrepresented. Improvement is underrepresented. Accounts that fit the validated community model circulate widely; accounts that challenge it are questioned or suppressed.
The social reward mechanisms of online platforms amplify this dynamic. Brady and colleagues' analysis of health forum trust dynamics established that the trust signals forum members use are social rather than epistemic — content that resonates with existing community beliefs is trusted regardless of its accuracy, and content that challenges those beliefs is filtered out regardless of its evidential quality. Combined with the credibility-performance dynamic Groenevelt and de Boer describe, the result is a self-reinforcing cycle: the illness model shapes what gets shared, what gets shared shapes the illness model, and the model drifts progressively away from what careful research actually shows toward whatever narrative the social environment has learned to amplify.
What the Research Shows vs. What Communities Sometimes Circulate
The drift documented by Groenevelt and de Boer has specific consequences for how ME/CFS and dysautonomia are understood in patient communities. Research in these conditions has been advancing significantly, producing findings on cerebral blood flow mechanisms, autonomic physiology, and physiological subtypes that change how the conditions should be understood and managed. But the incorporation of that research into community models depends on whether it passes the social filtering and credibility-performance tests. Research findings that challenge established community positions — even when methodologically sound — may circulate less widely than community-resonant narratives that are less evidentially grounded.
This is not a criticism of patients seeking validation and information in spaces that medicine has failed to provide. The epistemic injustice that created the community dynamics Groenevelt and de Boer describe is real. Patients were failed by medicine, and their online communities were a rational adaptive response to that failure. The problem is not the communities themselves. It is the structural dynamics that, when left unexamined, can cause those communities to work against the interests of their members by circulating a picture of the illness that diverges from the research in ways that affect how members understand their condition and make decisions about management.
Reading Online Communities With Appropriate Epistemic Caution
The Groenevelt and de Boer paper provides a framework for engaging with online illness communities more deliberately. The illness picture represented in a community forum is not a random sample of patient experience. It has been filtered by social dynamics that select for severity, persistence, and the absence of improvement — not because those outcomes are necessarily most common, but because the presentation incentives created by years of medical disbelief push in that direction. The research literature shows a range of outcomes in ME/CFS and dysautonomia conditions; communities may not reflect that range.
Saunders and colleagues' meta-analysis established that observing others' negative symptom experiences produces real nocebo effects — the community's severity-skewed illness representation is not merely an informational distortion but an active input to members' symptom experience. Reading accounts of severe and unresponsive illness continuously affects the brain's predictions about one's own illness trajectory. The practical implication is not to avoid online communities but to maintain epistemic awareness of what they are and are not: sources of emotional support and practical management experience, not representative samples of the illness or reliable guides to the research evidence.