Assessing the validity of self-reported history of rash caused by metal or jewellery.

BACKGROUND: The use of metal-containing bio-devices is becoming increasingly common. Self-reported history of dermatitis with metal exposure is not established as being predictive for metal allergy.

OBJECTIVES: To assess the validity of two screening questions addressing metal allergy.

METHODS: At Massachusetts General Hospital Contact Dermatitis Clinic, 2132 consecutive patients were asked either 'Do you get rashes when jewellery touches your skin' (Q1; N = 1816) or 'Do you get rashes when metal touches your skin?' (Q2; N = 316) before being patch tested.

RESULTS: Testing showed that 20% of subjects had positive reactions to nickel, 7.4% had positive reactions to cobalt, and 5.8% had positive reactions to chromium. Q1 was 40% sensitive (95%CI: 35-45%). The positive predictive value (PPV) was 51%, and the negative predictive value (NPV) was 82%. Q2 was 77% sensitive (95%CI: 68-84%). The PPV was 71%, and the NPV was 84%. Q2 was 37% more sensitive than Q1 (p < 0.0001), with a higher relative risk (RR) (4.75, p < 0.001) than Q1 (RR = 3.01, p < 0.001).

CONCLUSIONS: Patient-reported metal allergy, although not perfect, is a reasonable method for metal allergy screening to help identify those needing further objective evaluation by patch testing.