Gout and Hyperuricaemia in the USA: Prevalence and Trends

Gurkirpal Singh; Bharathi Lingala; Alka Mithal


Rheumatology. 2019;58(12):2177-2180. 

In This Article


While the age-adjusted prevalence of self-reported gout and hyperuricaemia has remained unchanged in the most recent decade from 2007–08 to 2015–16, the estimated total number of persons with self-reported gout has increased from 8.3 million to 9.2 million, reflecting the growth and increased aging of the US population. The diagnosis of gout in the NHANES was self-reported. While it is sometimes believed that self-reported diagnoses of arthritis may not be accurate, a meta-analysis showed a pooled sensitivity of 0.88, 0.75 and 0.71 for self-reported RA, OA and arthritis (unspecified) when compared with a gold standard of physician examination. The specificity of self-reported RA, OA and arthritis (unspecified) were 0.93, 0.89 and 0.79, respectively.[8] Similarly, Sacks et al.[9] showed that a case definition of arthritis based on a self-report of a 'previous diagnosis by a doctor' has a positive predictive value of 74.9% in individuals 45–64 years of age and 91.0% in those ≥65 years of age. The specificity of this definition ranged from 79.6% (45–64 years) to 81.1% (≥65 years). Studies of self-reported gout in physicians and health professionals have found accuracy levels of 94–100% when compared with medical chart review or subsequent detailed questionnaires.[10,11] McAdams et al.[12] studied the reliability of self-reported gout in the community by comparing responses on various questionnaires in the Campaign Against Cancer and Heart Disease (CLUE II) study from 2000 to 2007 and found that 73–75% of persons confirmed the diagnosis of gout in repeat questionnaires. They also found that the sensitivity of a self-report of physician-diagnosed gout was 84%, concluding that 'a self-report of physician-diagnosed gout has good reliability and sensitivity'.[12]

The laboratory findings in the NHANES were objectively ascertained using standardized laboratory methods and, because of the robust survey and sampling design, are likely to be representative of the US population but are not adjusted for use of urate-lowering medications. Our primary definition of hyperuricaemia used a cut-off of 0.40 mmol/dl (6.8 mg/dl), representing a level at which there is supersaturation of the extracellular fluid at 37°C. We used an alternate cut-off of 0.36 mmol/l (6.0 mg/dl), which represents supersaturation at 35°C (believed to be the temperature of the big toe)[4,13] and a target often used in treatment guidelines and clinical trials.[7] As sensitivity analyses, we also analysed a cut-off of 0.48 mmol/l (8.0 mg/dl), a level shown to correspond with a sharp increase in clinical gout[14] as well as renal[15] and cardiovascular complications.[16] The age-adjusted prevalence of hyperuricaemia has declined slightly but the total number of affected individuals is virtually identical [32.5 million in 2015–16 compared with 32.1 million in 2007–08 using our primary definition of 0.40 mmol/dl (6.8 mg/dl)]. While hyperuricaemia is not a 'disease' in itself, high levels of uric acid are associated with several systemic complications and gout.[16]

While the stabilization of self-reported gout and hyperuricaemia prevalence rates is encouraging, our study highlights the still considerable burden of gout and hyperuricaemia in the increasingly aging US population.