Results
The search yielded a total of 884 publications (Fig. 1). Of these,
534 (60.4%) were excluded (references that did not include
both an abstract and full text (331); those that were duplicated
between databases (14); those with no mention of any kind
of health IT, beyond opinion articles, editorials and essays on
generic issues related to the subject (139); and publications
dealing solely with technical specifications for the writing of
software, not necessarily for the health sector (50)).
According to the PAHO classification (Table 1), most of
the publications identified (27.7%) described the development
and/or implementation of information systems at health facilities,
including hospitals, clinics, physician’s offices and diagnosis
and treatment support centres. These information systems were
Table 1. Distribution of health information technology
references, according to area of applicability
Area of applicabilitya No. of references %
Systems for health-care facilities 97 27.7
Decision-support systems 81 23.1
Electronic data exchange 66 18.9
Educational activity support 49 14.0
Electronic patient registration 37 10.6
Telemedicine 11 3.1
Medical imaging 9 2.6
Total 350 100
a According to Pan American Health Organization classification.
used for data storage, processing, recovery or diffusion purposes.
This group also included articles about systems designed for the
management of clinical and administrative information within
a specific facility or between different facilities.
The next group (23.1% of the publications identified)
was related to decision-support systems that facilitate clinical
and administrative decision-making by means of interactive
dialogues. These include clinical diagnosis, individual monitoring
applications, facility and institution management applications
and “virtual health libraries”. The third group (18.9%),
were publications dealing with electronic data exchange,
including general infrastructure designed to allow interaction
and information exchange between the users and services and
between systems themselves, by means of the Internet and
electronic mail. The next group dealt with support systems
for educational activity, directed towards distance-learning and improvement of teaching ability in the education and training
of health-care professionals (14.0%). Electronic patient registries
were the subject of 10.6% of the publications. These are
systems that integrate and promote access (from a single site to
multiple locations) to collections of clinical and administrative
data concerning the patient, based on a distributed database
and including different means of support, such as intelligent
optical card technology.
The final group comprised publications related to telemedicine
such as support systems for diagnosis teleconferences,
transmission of high-resolution images and vital signs for longdistance
diagnosis and robotic telesurgery. These subjects accounted
for 3.1% of the publications, and medical imaging
systems designed to store, process, recover and transmit medical
images for 2.6%.
Following a detailed analysis of the 350 abstracts included
in Table 1, 135 were excluded. These were publications
concerned with data exchange, educational activities, and telemedicine
or medical imaging, or because of language problems
(three). Of the 215 remaining references, 163 were not concerned
with the evaluation of technology (Fig. 1).
We attempted to obtain the full texts of the remaining 52
publications: 45 of them were successfully located, corresponding
to 13% of the total. We then proceeded to the identification
of advantages, problems and potential solutions related to the
use of computerized systems in primary health care. From the
reference lists included in the articles, and further searches using
other sources, seven more relevant publications were located,
giving a total of 52.