To assess whether repeat olfactory testing in initially hyposmic subjects improves prediction of conversion to Parkinson disease (PD).

Olfactory testing has been proposed as a screening tool for the diagnosis of PD due to its low cost and ease of completion. Hyposmia is sensitive for the development of PD, but lacks specificity. We assess whether repeat smell testing improves its diagnostic characteristics for predicting conversion to PD in a community cohort.

Participants completed up to 12 years of clinical and imaging evaluations as part of PARS. Olfaction was assessed with the University of Pennsylvania Smell Identification Test (UPSIT), where accuracy less than the 15^th percentile of age and gender norms was considered hyposmic.  Development of PD was defined as clinical diagnosis on annual assessments or DAT deficit (less than 65% of age-expected lowest putamen specific binding ratio) on [¹²³I] β-CIT SPECT scans completed approximately every two years. Kaplan Meier survival analysis and Cox proportional hazards regression were used to estimate the hazard ratio for developing PD based on olfactory testing.

Of 155 participants who were hyposmic at baseline, 46 (30%) reverted to normosmia. None of 88 initially normosmic subjects and only 1 of 46 (2%) of reverters developed evidence of PD, compared to 23 of 109 (21%) subjects with consistent hyposmia. Consistent hyposmia had sensitivity of 96% and specificity of 61% for the development of clinical or imaging evidence of PD, compared to 100% and 40%, respectively, for hyposmia on a single, initial test.  In Cox models accounting for potential confounding factors, the hazard ratio for developing PD between consistent hyposmia and reversion was 10.2 (95% CI: 1.3-77.2, p=0.03). 

Serial smell testing improves detection of prodromal PD compared to a single test. Repeat testing may be a cost-effective strategy to identify at-risk patients for early diagnosis and disease modification studies.