Objective: To evaluate the stability of a 6-item smell test, in the context of a prodromal Parkinson’s (PD) cohort.
Background: Short smell tests have been developed for affordable large-scale testing in Parkinson’s (PD) research. Previous work has focused on whether these tests reliably identify individuals with hyposmia or PD, with little focus on test stability and inter-rater agreement over time. This work took place in the context of the UK-based longitudinal PREDICT-PD cohort.
Method: 1519 unique participants completed an abbreviated 6-item version of the University of Pennsylvania Smell Identification Test (UPSIT) between October 2024 and May 2025. Individuals with a diagnosis of PD, pure autonomic failure (PAF) or REM sleep behaviour disorder (RBD) were excluded. Multivariable binomial logistic regression (BLR) models assessed correct identification of each odour over time. To address potential unexpected inter-individual variation, Model 1 adjusted for sex, self-reported or ICD-10-coded hyposmia/anosmia, and the PREDICT-PD risk score (a combination of demographic, clinical, family history and lifestyle factors). Age was not included in Model 1, as it is weighted heavily in the risk score. Model 2 adjusted for age, sex and hyposmia/anosmia only. Additional BLR models examined whether the petrol/gasoline odour identification improved following a manufacturer change in February 2025.
Results: After adjustment, the proportion of individuals correctly identifying four out of the six odours varied significantly over time (Model 1: petrol/gasoline p=0.037, watermelon p<0.001, lemon p<0.001, cinnamon p=0.042). The same odours demonstrated similar change in Model 2 (petrol/gasoline p<0.001, watermelon p<0.001, lemon p<0.001, cinnamon p=0.034). The identification of soap and gas odours remained stable over time. A manufacturing change relating to the petrol/gasoline odour was not associated with improved correct identification.
Conclusion: Several odours in our abbreviated smell test demonstrated longitudinal instability, raising concerns over the reliability of short smell tests and warranting further investigation into odour stability within the full UPSIT.
To cite this abstract in AMA style:
S. Meyer, A. Kuri, J. Bestwick, S. Waters, H. Chohan, A. Shahid, L. Smith, C. Simonet, L. Pérez-Carbonell, E. de Pablo-Fernandez, A. Schrag, A. Noyce. Evaluating the Stability of a Short Smell Test in the PREDICT-PD Cohort [abstract]. Mov Disord. 2026; 41 (suppl 1). https://www.mdsabstracts.org/abstract/evaluating-the-stability-of-a-short-smell-test-in-the-predict-pd-cohort/. Accessed October 1, 2026.« Back to 2026 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/evaluating-the-stability-of-a-short-smell-test-in-the-predict-pd-cohort/