BackgroundThe epidemiology of transient synovitis is poorly understood, and the aetiology is unknown, although a suggestion of a viral association predominates.PurposeThis population-based study investigated the epidemiology in order to formulate aetiological theories of pathogenesis.Patient and methodsCases in Merseyside were identified between 2004 and 2009. Incidence rates were determined and analysed by age, sex, season and region of residence. Socioeconomic deprivation scores were generated using the Index of Multiple Deprivation, allocated by postcode. Poisson confidence intervals were calculated and Poisson regression was used to check for trends.ResultsTwo hundred and fifty-nine cases were identified over 5.5 years. The annual incidence was 25.1 (95 % CI 22.1–28.5) per 100,000 0–14 year-olds. Male to female ratio was 3.2:1 (p < 0.001). Mean age at presentation was 5.4 years (95 % CI 5.0–5.8), which demonstrated a near-normal distribution. No relationship was identified between seasonality and incidence (p = 0.64). A correlation was identified with socioeconomic deprivation in Merseyside: incidence rate ratio 1.16 (95 % CI 1.06–1.26, p < 0.001), although further analysis within the subregion of Liverpool did not confirm this finding (p = 0.35).ConclusionsThe normal distribution for age at disease presentation suggests a specific disease entity. The absence of seasonality casts some doubt on the popular theory of a viral aetiology. The absence of a consistent socioeconomic gradient in both Merseyside and Liverpool challenges a previous suggestion of an association with Perthes’ disease. This paper provides ecological evidence that may challenge existing aetiological theories, though transient synovitis remains an enigma.