Implementation outcomes from delivering ‘acute-on-demand telehealth services’ in primary care: a retrospective observational study

Objectives

The retrospective component of this mixed method study leveraged implementation science to assess the execution of an acute-on-demand telehealth (AODTH) service pioneered in a Singapore public primary care institution using video consultation (VC) targeting the local multiethnic population.

Design

A study was conducted using electronic medical records and AODTH service data from 2023 to 2024. The data were computed to determine the implementation outcomes based on the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework: demographics of users (Reach); conversion to in-patient consultations (Effectiveness); AODTH uptake (Adoption); defaulters, post-VC referral to emergency or specialist outpatient services, medical certification (Implementation); purpose and service utility trend (Maintenance).

Descriptive statistics, Pearson’s χ ² test and regression analyses were executed to report implementation outcomes and associated factors.

Setting

Data were obtained from a primary care clinic (polyclinic) in eastern Singapore.

Participants

AODTH users aged ≥18 years who used the service between 2023 and 2024 were included. No exclusion criteria were applied.

Results

39 811 AODTH VCs were completed by 17 603 unique users, including 44.4% males and Chinese (59%) and Malay (24%) ethnicity. Senior adults aged 65 years and older represented only 2.5% of users and 1.2% of VCs. Monthly VC volumes in 2024 exceeded those in 2023, with some months recording nearly double the consultation numbers and surpassing 2500 VCs in the second half of 2024. Mean default (6.6%) and conversion (2.2%) rates showed a decreasing trend from 2023 to 2024. Referrals to emergency or specialist outpatient services constituted 5.6% of VCs. Acute medical conditions accounted for 82.7% of VCs, chronic disease management 12.4% and other reasons 4.9%, with upper respiratory tract infections comprising almost half of acute cases. Over half (50.9%) of consultations were not prescribed medications, of which 72.3% were issued sick leave. 81.3% of users had two or fewer sick leave issuances, including those with none.

Conclusion

Although AODTH outreach to older adults was low, the service was effective for acute medical management with low in-person consultation conversion despite the lack of physical examination. The rising trends of AODTH users and VCs over 2 years reflect gradual service adoption, while low defaulter and referral rates reflect an appropriate implementation framework. Prevalent medical certification is likely the cornerstone to maintain AODTH service as a means for recovery from acute illnesses.