Examining intra- and inter-device reliability of pressure-mediated reflection spectroscopy in a multi-state sample of healthy adults

Authors: SISSON, SUSAN - University Of Oklahoma Health Sciences Center; HELMS, EMILY - University Of Oklahoma Health Sciences Center; Casperson, Shanon; HASNIN, SAIMA - University Of Illinois; JILCOTT PITTS, STEPHANIE - East Carolina University; STAGE, VIRGINIA - North Carolina State University; LONG, CHRISTOPHER - Center For Nutrition And Health Impact; MASSEY-SWINDLE, TAREN - University Of Arkansas; DEV, DIPTI - University Of Nebraska; BRAUN, ASHLEA - Oklahoma State University; STOOKEY, JODI - San Francisco Department Of Public Health; CAPE, ROWENA - San Francisco Department Of Public Health; BALDWIN, JONATHAN - University Of Oklahoma Health Sciences Center

Submitted to: Public Health Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/17/2025
Publication Date: 9/25/2025
Citation: Sisson, S.B., Helms, E., Casperson, S.L., Hasnin, S., Jilcott Pitts, S., Stage, V.C., Long, C.R., Massey-Swindle, T., Dev, D.A., Braun, A., Stookey, J.D., Cape, R., Baldwin, J. 2025. Examining intra- and inter-device reliability of pressure-mediated reflection spectroscopy in a multi-state sample of healthy adults. Public Health Nutrition. 28. Article e169. 1-11. https://doi.org/10.1017/S136898002510116X.
DOI: https://doi.org/10.1017/S136898002510116X

Interpretive Summary: Because circulating carotenoids are deposited in the skin, technologies that measure skin carotenoids have emerged as a valid non-invasive method of assessing fruit and vegetable intake. The Veggie Meter® is a low cost, portable device that allows for a rapid (less than 1 minute) assessment of skin carotenoids making it a valuable tool to evaluate fruit and vegetable intake in a variety of research and community-based public health nutrition settings. However, it is not known how these measurements compare across devices and geographically diverse populations. This multicenter study was conducted at 8 research sites across the United States, each with at least two Veggie Meter® devices, and 282 participants completed the study. The results from this study established that the observed margin of difference within a device can be as high as 14% and between devices can be as high as 22%. As such, device variability necessitates practical consideration for practitioners and researchers using the Veggie Meter® to evaluate intervention effectiveness and examine epidemiological associations with health outcomes. Implications of this research include ensuring that the same equipment is used for a participant across an intervention timeframe and that the actual changes observed may need to be larger than 14% to be truly attributed to the intervention.

Technical Abstract: Objective: To examine the intra- and inter-device reliability of devices using pressure-mediated reflection spectroscopy (the Veggie Meter®). Design: A cross-sectional research study was conducted across 8 sites in the United States. Using two Veggie Meters® at each site, participants completed five, counter-balanced pairs of finger scans. Intra-device comparisons included intra-class correlation coefficients (ICC) and calculation of the Coefficient of Variation (CV) and 95% confidence intervals (CI) of each device/site; hypothesized to be = 6%. Inter-device comparisons included ICC, Absolute Relative Differences (ARD) and 95% CI, and equivalence; both hypothesized to be = 10%. Setting: Eight sites across the United States Participants: Across sites, participants’ (n=282) average age ranged 24.7-39.0 years; sex ranged 60.0%-85.7% women, and Non-Hispanic White ranged 20.0%-94.3%. Results: Intra-device ICC ranged from 0.77-0.99. The CV ranged from 6.2%-14.2%, with an average of 8.8%. A majority (63%; n=10) of the Veggie Meter® devices had significantly higher CV from the hypothesized 6%. Inter-device ICC ranged from 0.58-0.94. The Absolute Relative Difference (ARD) ranged from 7.5%-22.0%, with an average of 13.9%. ARD in a majority (n=5) of sites was significantly higher than the hypothesized 10%. Five sites (63%) demonstrated equivalence below the hypothesized 10%. Conclusions: Our study demonstrates the intra-device and inter-device reliability to be moderate to high, as per ICC. We offer guidance that the observed margin of difference within a device was up to 14%, with an average of 9%. We offer guidance that the observed margin of difference between devices was up to 22% with an average of 14% between devices.