[Evaluation of the quality of the treatment indicator of the State Information System on Drug Addiction].

Background: In Spain, the problems caused by cocaine and opiates consumption are monitored through the State Information System on Drug Abuse (SEIT); one of its indicators is the number of admissions for ambulatory treatment for these drugs abuse/addiction (Treatment Indicator).

Methods: The information quality of this indicator was evaluated retrospectively during 1988 by comparing the data notified to SEIT with those registered in the treatment centres. For that, a random sample of cases among the clinical records from the notifying centres, and another among the cases registered in the SEIT Central Unit, were selected and crossed, calculating the precision and concordance indexes and observing the possible differences between both samples. In addition, mechanisms of notification and information organization were examined in the selected centres.

Results: In 1988, at state level, the sensibility was 81% and the predictive positive value 96%. As a whole, the SEIT underestimated the number of cases by 16%. Although the sample was not representative by Autonomous Communities, important between-regions differences in precision were observed. No significant differences of the information quality were found in relation to the centre size (although quality was lower in small centres) nor between the "true cases" notified and those not notified. Disparities, among centres, in relation to mechanisms of notification and information organization were found.

Conclusions: The SEIT is of a great value to know the characteristics of the persons treated for opiates or cocaine in Spain; but the between-regions variations on its precision might limit seriously geographic comparisons in the number of cases. It is necessary to evaluate periodically the indicator quality and to improve the consistency of data registration mechanisms.