The development and evaluation of a scale to assess pain in the post-operative neonate.

Debate surrounding the issue of pain management in neonates has mushroomed over the last ten years. Previously held beliefs that neonates do not feel pain because their anatomical make up is different from that of an adult, and that they do not remember pain therefore there is no need to relieve it have been demonstrated as erroneous. Studies such as Volpe (1981), Gilles, Shankle and Dooling (1983) and Beyer and Wells (1989) refuted previously held physiological misconceptions. Anand and Hickeys' 1987 study did much to raise our awareness of the deleterious effects of unrelieved pain in neonates. The impetus for the present study was the wish to improve analgesic techniques in one such group of infants - postoperative neonates. Valid assessment is foundational to improving analgesia and measuring the efficacy of interventions thus broadening our knowledge of safe, effective methods of preventing undue pain in newborns. The research presented here follows four distinct phases. The primary aim of the research was to develop a pain assessment tool. This was initially developed by use of an observational research technique, watching and cataloguing the behaviour of newborns (n=25) over a number of hours in their home environment. Video recordings of normal neonatal behaviour and development were also viewed and empirical evidence from neonatal behaviour experts such as Wolff (1966), Brazelton (1977) and Trevarthan (1977) was drawn upon to provide a detailed overview of neonatal behaviour. Observations were then made on a surgical group of babies (n=34) around normal caregiving episodes. Each observation lasted a number of hours. Some of these episodes were videod for later viewing by 3 clinical psychologists. The qualitative data collected from the observations of these babies (n = 59) was transcribed. The unstructured observations of both real life and video recordings collected by pen and paper provided rich, descriptive information to be analysed qualitatively. Glaser & Strauss (1967) term these "field notes". The field notes were then reduced in order to summarise the information by teasing out themes around which behaviours were clustered (Miles and Huberman 1984). These categories were organised into a detailed scoring system. This was called the Liverpool Infant Distress Score (LIDS). Following initial development the scale was subjected to rigorous reliability and validity tests. After piloting the scale on a further 10 babies undergoing surgery, adjustments were made to the initial scale. The scale was then applied to 31 babies in the peri operative period and a control group of 10 non surgical babies. Validity of LIDS was demonstrated. The value of an assessment tool such as LIDS also lies in its ability to be reproduced consistently and accurately by differing carers. (Melzack 1984). The next part of the study addressed this issue. By teaching the scale to a group of 4 nurses and testing their scores over a number of assessments, inter rater reliability was demonstrated. The final phase of the study compared the subjective scores of two groups of nurses – one experienced neonatal nurses, one paediatric nurses- to the more objective LIDS scores. The results from this final phase of the study suggest that despite an increase generally in nurse awareness regarding pain cues in neonates, pain assessment is still open to subjectivity