In reply: We appreciate the opportunity to respond to Curtis's critique1 of our study2 and would like to address the concerns raised.
Cross‐sectional surveys are routinely used to estimate prevalence in a population. Moreover, we have already acknowledged that cross‐sectional surveys have inherent limitations, including potential response biases, and we have transparently reported these limitations in the discussion section of our article (in keeping with journal guidelines) so that our results could be interpreted in context.
Curtis suggests that our findings may overestimate the true prevalence of post‐coronavirus disease 2019 (COVID‐19) condition (long COVID), stating it “could be as low as 3%.”1 This estimate appears to assume that all non‐respondents were free of long COVID, which is highly unlikely. While acknowledging that the prevalence of long COVID reported across studies can vary by different methods and case definitions used,3 the prevalence in our study population aligns well with other recent investigations, including cohort4 and meta‐analysis5 studies.
Although the absolute (weighted) proportion with long COVID could be an overestimate of long COVID in the population, as stated in our study, the relative risk estimates for factors associated with long COVID, which are based on internal comparisons, remain valid.
Finally, regarding “abstract bias” and “media release bias”, we have made concerted efforts to communicate our findings responsibly. For example, in the abstract, we did not report the “estimated prevalence of long COVID in Western Australia” but rather that “18.2% of survey respondents reported symptoms consistent with long COVID.”2 Our public communications have consistently highlighted the study's methodology to minimise misinterpretation. We agree that maintaining public trust in science is paramount and are committed to transparency in research.
We believe our study contributes valuable insights into the burden of long COVID in a highly vaccinated population and we thank Curtis for providing an opportunity to reiterate some of the strengths and limitations of this work.
No relevant disclosures.