Editor—It is now widely accepted that even the most preterm babies experience pain. This is difficult to measure and a number of clinical scales have been developed to make this assessment as objective as possible. Without the use of objective scales, the assessment of pain can become very subjective. We have measured staff perceptions of pain experienced by babies in different clinical situations in a neonatal intensive care unit. Clinical scenarios were presented to nursing and medical staff of the Exeter Neonatal Unit, and they were asked to score on a visual analogue scale the severity of pain they felt a baby experienced in these situations. The scale ranged from no pain to extreme pain on a 10 cm line and staff were asked to mark a point on the line that represented their assessment of the likely level of pain, and they were also asked whether they thought analgesia was necessary for the baby. There were six clinical scenarios:

• a Guthrie test on an awake term baby using a spring loaded Autolet device;

• a ventilated baby of 28 weeks gestation in no obvious distress with normal blood gases;

• a 35 week gestation baby of a diabetic mother who had four attempts at intravenous cannula insertion;

• a term baby with respiratory distress syndrome who developed a pneumothorax needing chest drain insertion;

• a 37 week gestation baby who had grazing of the scalp following a failed ventouse and difficult forceps delivery;

• a 27 week gestation baby who needed a lumbar puncture as part of a septic screen.

The response to the questionnaire was anonymised. Sixty six questionnaires were distributed to 21 doctors and 45 nurses. Fifty six (85%) responded, of whom 18 were doctors (eight men, 10 women) and 38 were nurses (three men, 35 women). The doctors comprised senior house officers, specialist registrars, staff grade, consultants, and nursing staff of sisters, senior nurses, staff, and nursery nurses. The overall scenario score (calculated by totalling the score for each scenario) was significantly higher for nurses (mean (SD) = 28.5 (6.8)) than for doctors (mean (SD) = 35.8 (6.8); p < 001). The scenario score was significantly higher (p < 0.01) for nurses in four of the six clinical scenarios. The two scenarios in which the difference was not significant were the 28 week gestation ventilated baby and the baby with a grazed scalp, although in both situations the mean score was higher for nurses than doctors. In all scenarios, more nurses than doctors thought that analgesia was necessary but this was only statistically significant for the baby needing lumbar puncture (97%v 77%; p = 0.03).

We feel this questionnaire study of our unit highlights important differences in perception of pain between doctors and nurses. Does it reflect a sex difference in the composition of the two groups? Are doctors distancing themselves from the pain that often they inflict when performing practical procedures or are they more aware of potential side effects of the analgesics used? It would be interesting to explore the reasons for these differences.