Back

Identification and management of thyroid dysfunction using at-home sample collection and telehealth services

Gavin, K. M.; Kreitzberg, D.; Gaudreau, Y.; Cruz, M.; Bauer, T. A.

2022-09-29 endocrinology
10.1101/2022.09.29.22280520 medRxiv
Show abstract

IntroductionPrograms aimed at modernizing thyroid care by pairing at-home sample collection methods with telehealth options may serve an important and emerging role in thyroid care. The primary objective of this analysis was to evaluate telehealth utilization, demographics, and clinical characteristics of a cohort of consumer initiated at-home lab thyroid test users who were also offered the option of follow-up telehealth consultations. MethodsThis was a retrospective analysis of real-world data from a de-identified consumer database of home-collected, mail-in Thyroid Tests utilized from March to May 2021 (n=8,152). The mean age was 38.6 (range 18-85) years and 86.6% of individuals identified as female. ResultsSeven percent of test takers fell into a thyroid dysfunction category (0.9% overthypothyroidism, 2.9% subclinical hypothyroidism, 0.1% overt-hypothyroidism, and 3.3% subclinical-hyperthyroidism). Twelve percent of the overall sample opted into a telehealth consultation, with 91.8% receiving a non-treatment telehealth consultation and 8.2% receiving a treatment telemedicine consultation. Sixteen percent of individuals with overt or subclinical thyroid dysfunction engaged in telehealth consultations. Of those opting into a treatment consultation, 59.3% reported a history of thyroid issues, 55.6% indicated wanting to discuss their current thyroid medication, and 48% received a prescription medication. DiscussionThe combination of at-home sample collection and telehealth is an innovative model for screening for thyroid disorders, monitoring thyroid function, and increasing access to care that can be implemented at a large scale and across a wide range of age groups.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.