BACKGROUND: A major problem in pain treatment in the newborn is the ability to assess the degree of experienced pain. Several different methods for estimating pain in the newborn have been introduced in recent years. The objectives of this study were to see whether pain scoring was used in Swedish neonatal units, and by which method, and furthermore to observe whether any changes had occurred between 1993 and 1998. METHOD: A questionnaire was distributed to all Swedish neonatal units in 1993 and 1998. RESULTS: In both 1993 and 1998, 86% of all clinics answered the questionnaire. Two-thirds of these clinics used some method for estimating pain in the newborn at both times. Eight clinics used a structured method in the latter period compared to one in the earlier period. A higher proportion of units having full intensive care measured pain compared to other units. CONCLUSION: The study shows that an unchanging low proportion of neonatal units in Sweden attempt to assess pain. There is a minor increase in the number of departments that use a structured method for pain scoring. Documentation of pain intensity is still inadequate and should be improved.