Abstract
Dear Editors:
We read with great interest the article “User Satisfaction with Asynchronous Telemedicine: A Study of Users of Santa Catarina's System of Telemedicine and Telehealth,” which was published in this journal by von Wangenheim et al. 1 This article has some advantages and disadvantages from our perspective, as noted in the following statements.
The article's methods, results, and discussion have very illustrative tables and diagrams, and the metric associated with patient satisfaction and health professionals is well described, but the statistical analysis has not been described as well for other parts of the article. 1
The title of this article, “User Satisfaction with Asynchronous Telemedicine and Telehealth,” is a general subject, and, as you know, telehealth studies focus on the fields of health education of sophisticated people, education of users, supervision, sharing of local public health information, assessment of research, and telemedicine. In fact, the subjects that are tackled in this article (the “Satisfaction Study with the Tele-electrocardiography Service” from the point of view of the patients in basic healthcare units and the “Satisfaction Study with the Service of Electronic Delivery of the Result of the Clinical Analysis” from the point of view of the epidemiological surveillance staff) raise merely a small part of the range of the wider concepts. Therefore, it is impossible to extend these results to the total system. 2
Another matter that drew our attention was the time difference between the two surveys. One of them was performed in the second half of 2007, and the other was done in the first half of 2010. From our point of view, telemedicine and telehealth are definitely time-dependent concepts, and ignorance of the effect of promotion of technology can cause unreliable results.
This study has a brief and useful questionnaire in its appendix section that would be helpful in performing the same studies in other countries with circumstances to similar to those in Santa Catarina.
Finally, it is possible to extend this study in the economics groundwork and examine the cost-effectiveness in telemedicine and telehealth. 3,4
References
Response to Fayaz‐Bakhsh and Goodarzi
Aldo von Wangenheim, PhD
Dear Editors:
My colleagues and I are delighted to provide this response to the comments of Fayaz‐Bakhsh and Goodarzi regarding our article. 1
The instruments (questionnaires) we applied were very simple, and the research questions were straightforward. Thus we did not see the need to perform any more elaborate statistical analyses.
Of course, it is always possible to generalize, taking the degree of satisfaction of a few key user groups and consider them as valid for all groups. In our article, we decided not to do this because we understood that we were investigating the reaction to the introduction of telemedicine of very different and specific populations and that the reactions these populations have shown and the expectations and requirements they have are not necessarily able to be generalized or considered equivalent (e.g., considering them applicable to medical doctors, nurses, or dentists using the system). The first group we interviewed was composed of patients, mainly from small and medium‐sized upstate cities. The second group was composed of public servants with a wide range of educational backgrounds who performed mainly bureaucratic work in small and medium‐sized towns. We understand that the perception of quality and, consequently, parameters that generated satisfaction depended on different aspects for each of those groups. For the patients, the most important factor was agility and ease of access to the service because historically the major complaint against public health services in Brazil is frustration. For this population, ease of use, for example, was not a factor to be considered because these individuals usually do not have direct contact with the system. For public servants working with laboratory examination data, on the other side, ease of use is a key factor because in small upstate cities their background on computer usage skills is expected to be low. Because of these large demographic differences, we had to develop different analysis instruments, in the form of different questionnaires, applied using different media (paper forms for the patients, online surveys for the public servants). If we decided to apply this study to another population, such as physicians or nurses, there would have to be developed yet another analysis instrument in the form of another questionnaire (which we are doing now—for heavy users such as physicians, standard user satisfaction and usability measurement instruments, such as Brazilian Unitary Health System (SUS) and Nielsen criteria, can be adapted and used). The intent of our article 1 was to present the first results we obtained.
The Santa Catarina State Integrated Telemedicine and Telehealth System is a very large public initiative, and various services and modules have been developed and deployed during the 7 years that the project has been running. For both surveys, we decided to perform the data collection about 2 years after each service (in this specific case, “Telecardiology” and “Electronic Delivery of Lab Analyses”) was started and after initial problems with each service were solved. This was performed in order to guarantee that the applications were mature and to avoid the complication that problems related to software development and local infrastructure introduced an effect on the outcome.
The objective of our article was to study specifically user satisfaction. Theoretically it is possible to perform this extension, both in the area of epidemiology, trying to answer the question “Does telemedicine affect positively the morbidity profile of a population through the offer of faster and better diagnoses?,” and in the area of healthcare economics, trying to answer the question “Does telemedicine affect positively the costs of healthcare through the reduction of state‐paid patient locomotion and reduction of severe illnesses through faster diagnoses?” Epidemiological data can be collected very effectively and in almost real‐time and over large areas through telemedicine. In this point we have been working on data collection since 2005, and our objective is to provide clear epidemiological evidence that shows if telemedicine is or not correlated with a measurable impact on morbidity. First data, on more than 230,000 telecardiology examinations performed between 2005 and 2010, have already been published. 2 We are now working on the 2010–2013 data, which were acquired using DICOM Structured Reporting and a specially developed controlled vocabulary. Data on other kinds of examinations are also being processed. Concrete healthcare economics data are more difficult to obtain: for the travel costs for the situation “after telemedicine” we have very concrete data, but for the situation before 2005 data collection was poor. On the other hand, answers for the evolution of morbidity‐related healthcare costs can be provided through epidemiological data already being collected through telemedicine.