The wafer test: a semi-quantitative test to screen for xerostomia

OBJECTIVE: To develop a screening test for xerostomia. METHODS: A cross-sectional study was conducted among 152 healthy subjects aged <20-60 yr, 30 patients with primary Sj?gren's syndrome and 60 patients with other connective tissue diseases, sampled randomly. A validated screening questionnaire for sicca syndrome and the Schirmer-1 and wafer tests were carried out in all subjects. In addition, non-stimulated whole salivary flow was measured in a random sample of 113 participants. The main outcome was the time of dissolution of the wafer. RESULTS: Time of dissolution of the wafer was 2.8+/-2.1 min in the healthy group, 3.3+/-1.5 min in the connective tissue diseases group, and 9.2+/-3.9 min in the primary Sj?gren's syndrome group (P<0.001). The correlation coefficient between the wafer test and non-stimulated whole salivary flow was -0.60 [95% confidence interval (CI) -0.47, -0.71]. A cut-off value of 4 min ('wafer 4') showed sensitivity of 92.9%, specificity of 71.7%, a positive predictive value of 31.7%, a negative predictive value of 98.6%, accuracy of 74.3%, an ROC (receiver operating curve) value of 82.3 and a likelihood ratio of 3.3 (95% CI 2.3, 4.6) for xerostomia. The proportion of patients with wafer 4 was 8% in the healthy group, 23% in the connective tissue diseases group and 93% in the primary Sj?gren's syndrome group (P<0.001). Wafer 4 was a significant predictor of xerostomia after controlling for age, gender, temperature and relative humidity. CONCLUSION: The wafer test is valid and reliable for identifying subjects with xerostomia.