*3.2. Sensitivity Analysis.*

\*

We looked at di fferences in responses to the eight questions in the questionnaire according to professional categories, identifying significant statistical di fferences in three cases. First, with respect to question 3, nursing sta ff rated the quality of care delivered by the telemedicine services as significantly better compared with medical sta ff (*p* < 0.001). Second, regarding question 6, medical sta ff reported having experienced more technical di fficulties than nursing sta ff (*p* < 0.05). Finally, with regard to question 7, medical sta ff stated having experienced more organizational and other di fficulties that might have a ffected the quality of care delivered by the telemedicine services than nursing sta ff (*p* < 0.001) (Table 3). The di fferences in responses by gender and age have no statistical di fferences between groups.

Looking at di fferences in responses to the eight survey questions, in relation to the number of times respondents have used any of the various telemedicine services in the previous year, significant statistical di fferences were found in three questions. Respondents who used telemedicine services more often rated the quality of care of these services as significantly worse, compared with respondents who use them less often (Figure 3) (*p* < 0.05). They also stated having experienced more technical di fficulties and more organizational and other di fficulties compared with respondents that used telemedicine services less often (*p* < 0.001).


**Table 3.** Differences in Responses by Professional Categories.

**Figure 3.** Quality of Care of the Telemedicine Consultation, by Number of Times Used.

#### *3.3. Multivariate Linear Regression Model*

A positive correlation (PCC = 0.728) was found between the overall quality of the telemedicine consultation and the rating of the technical quality of said consultation, between the overall quality of the telemedicine consultation and the future use of telemedicine (PCC = 0.583), and between the technical quality of the telemedicine consultation and the future use of telemedicine (PCC = 0.505). A negative correlation (PCC = 0.531) was found between organizational and other difficulties that might affect the quality of care delivered by the telemedicine service and the future use of telemedicine. We found no correlation between the perceived quality of telemedicine rendered and the perceived barriers of use (Table 4).

The multivariate linear regression showed two variables to be good predictors for the future use of telemedicine services: the overall quality of the telemedicine consultation (*p* < 0.005) which positively affects the future use, and organizational (or other) difficulties that might affect the quality of care delivered by the telemedicine service (*p* < 0.001), which negatively impacts the future use (Table 5).


**Table 4.** Linear Correlations between Q1–Q8.

#### *Int. J. Environ. Res. Public Health* **2020**, *17*, 4092


**Table 5.** Multivariate Linear Regression.

Dependent variable: Q8 Would you continue to use the telemedicine service?
