*2.2. Study Type*

A cost-minimization analysis was carried out over a three-month period using a societal perspective. Direct costs (healthcare costs corresponding to time spent by professionals and users during visits and travel expenses by users) and indirect costs (patient and caregiver's time) were included. No staff training or equipment costs were included (practitioners used pre-existing devices), since they were not subject to the analyzed interventions. The cost estimate is based on 2019, a year which showed a higher number of telemedicine visits. A sensitivity analysis was carried out increasing the baseline costs. Calculations were performed using a Google Drive spreadsheet. The study was approved by the Ethical Committee for Clinical Research at the Foundation University Institute for Primary Health Care Research Jordi Gol i Gurina (registration number P19/182-P).

### *2.3. Direct Costs*

The Catalan Institute of Health provided anonymized individual data regarding all 52,198 telemedicine consultation services performed during the period November 2011–November 2019. This dataset contains information on a case-by-case basis on the source and destination of every type of telemedicine service and whether it avoided a subsequent face-to-face visit or not. As Table 1 shows, all telemedicine services result in high face-to-face savings, ranging from 72% to 88% of the queries received.


**Table 1.** Number of telemedicine visits and% of face-to-face visits saved, per type.

In order to calculate the derived potential societal savings, di fferential costs attributable to the time spent by practitioners and citizens using telemedicine and usual care were taken into account. From the healthcare system point of view, the savings resulting from this form of intervention are based on the reduction of case managemen<sup>t</sup> time. Whereas in usual care, the time spent on a face-to-face visit with a hospital care professional is 15 minutes, it is calculated that telematic monitoring of the case reduces the time to 5 minutes, redirecting the case back to the primary care professional, who calls the patient for approximately 2 minutes and closes the case, if applicable. If the specialist has any doubts, they can ask the primary care professional to book the patient for a face-to-face consultation (15 minutes). It was taken into account that, although in the teledermatology, teleophthalmology and teleaudiometry services, a primary care doctor is the one who makes the referral, in the case of teleulcers, a (primary and hospital care) nurse reviews the images and sends a reply. Baseline wages are used, according to standard labor agreements, for medical and nursing professionals in primary and hospital care. Travel costs (private car expenses) are calculated using the average travel distance (the methodology is described below) and the baseline price per kilometer.

### *2.4. Indirect Costs*

Productive time (commuting to the hospital) lost by patients and caregivers was considered. The user also benefits from greater agility in the resolution of the case, reducing waiting time, as well as in terms of travel time to a hospital consultation (Hospitals in Manresa and Berga). Employing the methodology used by Vidal-Alaball et al. 2019 [13], through a combination of the R 3.6.1 software (The R Foundation for Statistical Computing, Vienna, Austria), a Google Maps API and the information from each of the user's Primary Care Team (as a proxy for the user's place of residence), together with the referral hospital, a very accurate calculation of the total number of kilometers and time of journeys saved by the intervention was obtained (Figure 2). Therefore, the sample saved 893,820 kilometers (21.58 km per case, for the round trip) and 16,812 hours (25 minutes per case) of travel. The costs to users (patient and caregiver) have been calculated by multiplying travel and consultation time by the average salary/hour.

**Figure 2.** Origin and destination (either Hospital Sant Bernabé in Berga or Althaia Xarxa Assistencial Universitària in Manresa) of telemedicine visits avoided. The thickness of the line corresponds to the number of journeys saved.

> **Table 2.** Direct and indirect costs, for users and for the healthcare system.

Also, according to an aggregate analysis of the users' profiles, it can be observed that the average age of a telemedicine service user is 52, with a standard deviation of 23, suggesting the heterogeneity of the beneficiary profile. If we assume that people aged over 65 (34% of the total sample) and under 16 (8%) require the company of a caregiver during their visits, this means that we have to add the indirect impact in terms of opportunity costs of the time spent by caregivers in 42% of the cases analyzed.

The nature of each type of cost is shown in Table 2.


Finally, Table 3 shows the parameters which were considered when making calculations and their corresponding sources: the hourly wages of professionals, the price per kilometer, the opportunity cost of the user, the total number of visits (saved), consultation time with the specialist with and without telemedicine, the primary care professional's phone call time, and average time and distance for users. The results are shown for both perspectives (i.e., healthcare system and user).


**Table 3.** Calculation parameters.

\* For the comparability between studies, the baseline scenario takes the same parameter as in Vidal-Alaball et al. [12].
