Abstract
A systematic review of telemedicine projects in Colombia was conducted. We searched electronic databases, and also searched for relevant Internet websites. Each project manager was contacted by telephone to identify projects which had not actually been carried out. They were interviewed to request information about the projects they were managing, and whether they knew of other projects in Colombia. The search process identified 43 different projects, which were classified into two groups: telemedicine research initiatives and projects for providing health-care services via telemedicine. There were 32 projects which provided telemedicine services, of which 14 had been finished, 11 remained active, 4 were being implemented and no data were available about the state of the other 3. Health-care services had been provided using telemedicine to at least 550,000 patients. The projects had connected more than 650 health-care institutions, mainly in deprived areas of the country. Unfortunately, although many projects seem to have had a positive effect, none of them had been rigorously evaluated, and therefore in the absence of scientific evidence no general recommendations can be made. However, the methodology of the present study appears suitable for similar reviews of telemedicine in other developing countries.
Introduction
Telemedicine projects in Colombia were reviewed in studies conducted in 2001 1 and in 2003. 2 However, these previous compilations were not systematic reviews, and projects were listed without further analysis. We have therefore updated the previous work. The aim was to identify all the projects carried out in Colombia which provided services via telemedicine. This information may be helpful in future telemedicine work in developing countries.
Methods
The search strategy of the systematic review comprised four steps. In Step 1, relevant articles were identified from the following databases: MEDLINE, LILACS, Cochrane Library, Current Contents, CINAHL, CUIDEN, Telemedicine Information Exchange (TIE) and IEEExplore. Given the lack of articles from developing countries in indexed publications, 3 other documents were examined, such as references within retrieved articles, and other databases, such as the Centre for Reviews and Dissemination (CRD), the Scientific Electronic Library Online (Scielo), the Royal College of General Practitioners (RCGP), the Association of Telehealth Service Providers (ATSP), the American Telemedicine Association (ATA) and the International Society for Telemedicine (ISFT). The keywords used, in English and Spanish, were Colombia and Colombian and each of the following words: telemedicine, telehealth, ehealth and e-health. When a project was found, it was added to the list, together with the contact information of the project manager, or of an alternate person if the project manager could not be identified.
In Step 2, Internet search engines were used to retrieve more projects. The search engine Google was used with the keyword combination, and its translation into Spanish, shown in Table 1. When using any combination, the following basic rules were followed:
For each combination of keywords, the first 10 hits were examined. When following the links, no more than four levels of navigation were considered; When a project was found, it was added to the list together with its contact information. In addition, two new keywords were added to the list, the first one containing the current combination but filtering the name of the project found, using the option ‘-’ (in Google, to exclude pages where the current project was present in order to broaden the scope of our search) and the second one composed by the name of the project and Colombia (for acquiring specific information of each project prior to the interview); If after one search, less than 10 pages were retrieved and it was impossible to retrieve any new telemedicine project from them, then the next combination was used. Keywords used for the web search
In Step 3, each project manager was contacted by telephone to identify and discard projects which had not actually been carried out. In a second contact, an interview was undertaken to request information about:
The active projects that he or she was managing, including any available evaluations; Other telemedicine projects in the country; Telemedicine experts in Colombia.
If new projects were identified, Step 2(2) was conducted again.
In Step 4, an expert identified by the interviewees was contacted for confirmation about each project. An expert was somebody who was mentioned more times in Step 3(3). If new projects were identified through the expert, then Step 2 was conducted again.
In order to classify the projects, a previously published method 4 was adapted. The information retrieved from projects was classified according to feasibility and impact indicators 5 (see Table 2). The level of evidence for each indicator was rated according to the nine-level categorization 6 shown in Table 3, and recommendations were made by using previously suggested 7 criteria. That is, for levels 1 to 3, high scientific evidence for recommendation existed; for levels 4 to 7, sufficient evidence for recommendation existed; and for levels 8 or 9, the evidence was insufficient.
Evaluated feasibility and impact indicators
Classification of findings
Results
The search process identified 45 different projects. In Step 1, 20 scientific papers were obtained. After reading them, 15 telemedicine projects were identified. Information about Colombian legislation on telemedicine services was also obtained. 8,9
Step 2 provided information about 16 additional projects, and also provided contact information for the ones that had already been identified. Step 3 allowed us to identify 13 more projects, and to discard two (ARCAL 007 10 and Ecopetrol 1,2 ), because they were never carried out. Finally, in Step 4, a list of 42 projects was provided to an expert (an officer from the Communication Office), who confirmed each one and also added a new one, namely the telemedicine activities carried out by ITMS. 11,12
The 43 resulting projects were classified into two groups: telemedicine research initiatives and projects for providing health-care services via telemedicine. The first group included research projects involving the development of hardware and software (such as signal and image processing, telesurgery systems and telemonitoring equipment) for telemedicine projects, which had not yet been validated in patients. The second group consisted of projects in which patients had been managed via telemedicine. A third category was the project from the Colombian Association for Medicine, Informatics and Telehealth, 13 which aimed to create a forum where all those involved in telemedicine development in Colombia could exchange ideas and progress.
Telemedicine research projects
The search algorithm provided information about 10 research projects. Strictly speaking, they were not projects, but research groups devoted to specific telemedicine activities. Therefore, they were not considered in the subsequent analysis. The 10 research groups were:
The Telemedicine Centre from Universidad Nacional de Colombia (UNC). The UNC began its work in 1996 with Teleamazon,
14
a research project to identify those processes within the public health-care system in the Amazonas Department that could be improved using telemedicine; The R & D in New Telecommunication Technologies Group from Universidad del Cauca. In 1999, this group carried out a project that laid the foundations for other telemedicine projects in the university;
15
The Bioengineering Group from UNC. This group mainly conducts research in medical image processing, such as image compression and pattern recognition;
16–18
The Bioengineering Research Group (Grupo de Investigación en Bioingeniería – GIB) from Universidad EAFIT.
19
This group worked on a project for interchanging medical images using the national academic broadband network (Renata); The Colombian Telemedicine Centre (CTC). The CTC conducts research jointly with Universidad Javeriana, mainly related to the development of a virtual platform for educational purposes in surgery;
20–28
The Clinical Engineering Research Group of the Hospital Universitario La Samaritana. This group carried out a project for adapting the PACS and HIS of different Colombian hospitals;
10
The Biomedical Engineering Group (Grupo de Investigación Biomédica-GIB) from the Universidad de los Andes.
29
This group worked with Fundación Santa Fe de Bogotá (FSFB) to propose a methodology for evaluating procedures for digitizing radiology images, and researching on hospital information systems; The FSFB
30
conducted research on cost-effective methods for teleconsulting regarding suicide risk and sexually transmitted diseases in schools of Bogota; The Perception and Intelligent Systems Research Group from the Universidad del Valle. This group developed a system to handle second opinions and digital images through the Internet;
31
The Telehealth Research Group from the Universidad de Caldas. This group compared the efficiency of two methods for teaching the management of childhood illnesses, the conventional one and a virtual course designed by them.
32
Projects for providing health-care services via telemedicine
A total of 32 projects which provided telemedicine services were identified. Out of them, 14 had been finished, 11 remained active, 4 were being implemented and no data were available about the state of the other 3 (see Table 4). Projects were ordered by their current state, and within each group they were ordered chronologically (the most recent first). Table 4 shows whether the indicators of Table 2 had been evaluated in each project, and reflects the level of evidence achieved for each of them (in relation to Table 3). Information about medical speciality and telemedicine services provided, patients attended and health-care centres connected, was also included; for further information see Rey-Moreno. 10
Characteristics of selected studies and evaluated indicators
Institutions: Pu (public); Pr (private), (1) Caprecom (Pu); (2) Esoft Ltd (Pr), (3) ITMS-Colombia (Pr), (4) Fundación Santa Fe de Bogotá (Pr), (5) Telemedicine Centre-Universidad Nacional de Colombia (Pu), (6) Telehealth Research Group(GIT)-Universidad de Caldas (Pu), (7) Universidad del Cauca (Pu), (8) SaludCoop (Pr), (9) Universidad de Cali (Pr), (10) Bioengineering Research Group (GIB)-Universidad EAFIT (Pr), (11) Vision Technology Group (Pr), (12) Hospital Universitario La Samaritana (Pr), (13) Cardiobip Ltd (Pr), (14) Medicine Faculty-Universidad de Antioquía (Pu), (15) Telemedicine Research Group (GITEM)- Universidad Francisco José de Caldas (Pu), (16) Colombian Telemedicine Centre (Pr), (17) Interdisciplinary Telemedicine Team (EIT) - Universidad Pontificia Bolivariana (Pr), (18) Fundación CardioVascular de Colombia (Pr), (19) Dr. Díaz (Pr)
*Status; Act = Active, Imp = Implementation
Medical specialties: C (cardiology); R (radiology); D (dermatology), N (neurology), PS (psychiatry), G (gynaecology); PN (respiratory); U (urology); O (otorhinolaringology); EN (endocrinology); OR (orthopaedics); ON (oncology); GA (gastroenterology); IM (internal medicine); P (paediatrics); GM (general medicine), I (infectious diseases); PA: pathology and H: haematology
Telemedicine services: TC: teleconsultation between medical staff; TM: telemonitoring of vital signs of patients with chronic diseases (at home or in hospital); TEI: tele-education in health for people/patients; TE: tele-education (remote training, support to medical students/health-care staff, continuing education); TA: teleadministration (remote administration process as citation management); O: other telemedicine services (e.g. epidemiology surveillance)
Others: ND: No data; DNA: Did not apply
A common feature of all the projects was the lack of studies assessing their performance and their real impact. Most of the entries in Table 4 show insufficient level of scientific evidence, because they came from descriptive articles (e.g. P8, 33 P15 or P16 34 ) or from anecdotes that were mentioned in the interviews. The only two studies reporting sufficient evidence for recommendation consisted of a comparison of technical effectiveness (I1) between tele-ECG devices (developed by institutions 3 and 18, respectively) and traditional ECG. They were carried out in P4 12 and P30, 13 and both concluded that their devices had the same diagnostic accuracy as the conventional comparators. Some private companies, such as institutions 2, 4 and 18 (see Table 4), mentioned that they had conducted cost-benefit evaluations of their services, but the analyses could not be published for reasons of commercial confidentiality.
The most commonly assessed indicator was I10 (16 projects), showing that many projects in Colombia aimed at improving accessibility to health care for people in deprived areas. Also I8 and I4 had been considered often. Eight projects showed the reduction produced by telemedicine in the time a patient has to wait, and 12 projects showed that telemedicine services are well accepted, although some telemedicine projects (P3, P9) were not accepted initially. It is important to note that despite P18 and P24 reporting initial institutional acceptance, the projects stopped working once they were transferred to the respective health-care institutions.
Very few studies (5 projects) had analysed the improvement in diagnostic capacity (I6) from telemedicine. In P8, an evaluation showed that the availability of second opinions avoided six urgent admissions a year in each health centre. In P15, the manager stated that by using the teleconsultation system, 170 hospital admissions were avoided during 12 months of operation. The isolated conditions of the hospital produced savings of more than US $300,000 per year for the state administration. On the other hand, in P16, thousands of dollars were invested and only 0.7% of the diagnoses were changed.
There was a lack of data assessing the effect of telemedicine on the rest of the indicators, and very few conclusions could be drawn. It is remarkable, for instance, that a poor initial evaluation of the existing technologies had led to P29 remaining in the implementation phase for eight years. In other cases, unforeseen circumstances ruined some projects, such as P22 and P25, which were affected by the economic crisis that struck Colombia at the end of the 1990s.
Institutions providing telemedicine services in Colombia were mostly private. However, public institutions, such as the Colombian government, were also involved in some, for example P1 and P2, which were financed by the Health Office and entrusted to institution 5, 6 and 18 (see Table 4). These projects aimed to improve access to health care.
There was a wide variety of services provided by the projects, ranging from those typical in industrialised countries, such as telecardiology or teleradiology, to those specific to developing regions, such as telemicroscopy for malaria or tuberculosis. Most of the services of telecardiology, teledermatology or teleradiology were provided by teleconsultation using software developed by each institution. In addition, some services, such as those for respiratory medicine in P4, had developed their own hardware.
Discussion
The present systematic review of telemedicine projects in Colombia allowed us to update and extend earlier studies with new information, hence increasing the number of identified projects from 8 to 45. The search strategy was efficient and it served to find almost all the telemedicine projects developed in Colombia up to May 2008. Under Colombian legislation, it is compulsory to report to the government on the start-up of projects which aim at managing patients through telemedicine, and the expert who was selected in the search strategy was the civil servant responsible for the management of those reports. Therefore, we can consider that the results of the search strategy were validated at least for the second group of projects. Thus, the proposed search strategy may be useful for future systematic reviews of telemedicine projects in countries where information is not easily accessible.
The systematic review shows that many institutions in Colombia, both private and public, are involved in the development of telemedicine through several projects. Across these projects, telemedicine services have been provided to at least 550,000 patients, and the projects have connected more than 650 health-care institutions, mainly in deprived areas of the country. Unfortunately, although many projects seem to have had a positive effect, none of them have been rigorously evaluated, and therefore in the absence of scientific evidence no general recommendations can be made. In future, therefore, it would be advisable that financing agencies consider funding only projects which include a well-designed plan for evaluating their feasibility and impact.
Footnotes
Acknowledgements
This work was partially supported by grant TEC2007-68096-C02 from the Spanish Government.
