Background: The needs of information for the primary health care centers planning and management, and the poor situation we started from, have generated a large amount of information systems, which, as a general rule, have not been sufficiently evaluated. Since 1986, in the Area de Gestión, 7, Centro, of the ICS here exists an information system of the general medicine services with a sampling method (ANAC-2). The validation of some aspects of the process and content is shown in order to evaluate the quality of information.
Methods: The problems arisen during the process of collecting data from nine centers are analyzed during six months and its information content is compared with the one of each system with a standard respect their value. To evaluate the concordance, we have used a graphic representation of the differences of each system with a standard respect their mean value, and the "limits of agreement".
Results: On relation with the problems of collecting data, two centers show a nonfulfillment of the observation calendar higher than 20% and the logical divergences are not important. The kind of visits distribution is quite correct, even if the estimate of the whole number of visits is higher than 20% in two centers. In the activity indicators, the system of reference has a tendency to give average values lower than the ANAC-2, with the exception of prescription/visit. In referrals and prescriptions, the use of different sources of information between systems produces an average difference of 3.3 interconsults/100 visits and 0.8 prescriptions/visit respectively. Generally, the limits of agreement are wide and become unacceptable in laboratory.
Conclusions: The study carried out is evaluated positively, for it detects the problematical areas which can be modified or require further studies. The importance of validating the information systems is emphasized, in spite of difficulties.