A meta-analysis of the original papers on human immunodeficiency virus (HIV) published in Spain was performed to better know the characteristics of HIV infection in Spain. Between 1985 and 1989, 67 investigations fulfilled the inclusion criteria. They were evaluated by means of a quantified questionnaire. The mean quality of the investigations was 44.65 +/- 1.5%. A bias in the selection was highly probable in 80% of the studies, whereas a bias in information and in confusion was less frequent (35.8% and 77.6%, respectively). The investigations of more quality were those carried out recently at the University in collaboration with other Institutions and with participation of sanitary personnel published in microbiologic journals.