Reliability and validity of the pain observation scale for young children and the visual analogue scale in children with burns

Abstract

The aim of this study was to assess if the pain observation scale for young children (POCIS) and the visual analogue scale (VAS) are reliable and valid instruments to measure procedural and background pain in burned children aged 0–4 years. Burn care nurses (n = 73) rated pain from 24 fragments of videotaped children during wound care procedures and during periods of rest using the POCIS and the VAS. Intraclass correlations were used to assess inter-rater and intra-rater reliability for the POCIS and the VAS. Internal consistency for POCIS was assessed by Cronbach's alpha. The POCIS has shown poor to moderate inter-rater reliability, moderate to good intra-rater reliability and an acceptable internal consistency. The VAS turned out to have poor inter-rater reliability and poor to moderate intra-rater reliability. Due to poor results of inter-rater reliability in both scales, construct validation is left undone until more acceptable results are obtained. Factors explaining the results are the large number of raters, the manner they were trained and a lack of variation between pain classes in video fragments. Although not all results were satisfying, an easy to use scale as POCIS has promising qualities and deserves further reliability research.

Introduction

Every year approximately 500 patients with severe burns are admitted to Dutch burn centres [1]. According to Dutch burn centre charts, approximately 30% of these patients are children younger than 4 years old [2]. Burns in young children are mostly caused by hot liquids. Children get, due to their development stage of motor and cognitive skills, hold of cups filled with coffee or tea or pull down hot liquid containers with the tablecloth [3].

Pain caused by severe burns is considered the worst form of pain [4]. Melzack and Wall [5] maintain that pain is such a complex concept with various determining factors that it defies definition. Nevertheless, a definition proposed by the International Association for the Study of Pain, comes up repeatedly in the literature: ‘Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage’ [6]. McCaffery and Beebe [7] recognize two categories of pain, namely chronic and acute pain. Acute pain, like burn pain, has a predictable and limited duration and has a tendency to diminish. Acute pain in burn patients can be divided in background pain and procedural pain.

Adequate management of procedural and background pain is essential to the relationship between the burn patient and the multidisciplinary team. It aids in increasing comfort of the patient and prevents an elevated metabolism, thereby reducing the chance of malnutrition and deterioration of the immune system [8]. Furthermore, adequate pain management correlates highly with reduction of acute stress symptoms during admission in the hospital [9].

Nurses are one group of health-care professionals responsible for the recognition and treatment of pain [10]. Because of their frequent and direct contact with the patient, and through their direct involvement in the phenomenon of pain, burn care nurses have a special responsibility to improve pain management. The ultimate goal is to optimise nursing care in order to achieve good patient results in the burn unit.

Nurses can contribute to pain management by measuring pain. However, there is a need for more clarification regarding instruments to measure background and procedural pain in burned infants, toddlers and preschoolers. This instrument should preferably meet the following criteria. In the first place, the instrument should be reliable and valid for this specific patient population. Secondly, the instrument should be easy and quickly to use and be available in Dutch language. Finally, the instrument should be appropriate for nurses to measure pain, since parents are not present in the burn unit all day.

A literature review was conducted to assess what is known about pain measurement instruments in burned children in the age of 0–4 years. The databases CINAHL and Medline were used. Combinations of the keywords burns and pain measurement, pain assessment or pain scoring and children, non-verbal children, paediatrics, toddlers, preschoolers, infants or preverbal children were used. Consequently, on the base of titles and abstracts a selection was made. Publications without abstracts or names of authors, case studies, studies in other than clinical settings and publications in other than English or Dutch language were excluded.

Relevant literature shows that cognitive, physiological and behavioural components are frequently used to measure pain [11]. Cognitive components can be measured by self-reports, but are only appropriate for children above the age of four [12]. Physiologic parameters such as heart rate, respiratory rate and blood pressure are useful with patients who are not able to interact with their environment, like artificially ventilated children [13], who are however not part of the target group. In addition, measurement of these parameters, when manually assessed, can cause pain and anxiety and can distort data [14]. Behavioural measures are helpful in situations where children are unable to provide self-report. Mills [15] identified three pain behavioural categories in children from 0–3 years old, namely motor movement, communication and facial expression. Children with burns were included in this qualitative study. The following behavioural observation instruments are suggested in burn literature and are examined on the presence of the aspects as described by Mills [15], on age group, easiness of use, validity and reliability and the users group: the observer pain scale [16], [17], the children's hospital of eastern Ontario pain scale (CHEOPS) [8], [13], [16], the objective pain scale [16], the observational scale of behavioural distress, the infant pain behaviour scale, the postoperative pain measure for parents and the face leg activity cry consolability pain assessment tool [13], a discomfort five-point scale [18] and the visual analogue scale (VAS) [19].

However, none of these scales are ready to use because they do not meet all before mentioned criteria. Although the majority of the scales, especially the VAS, is easy to use, based on the necessary time to observe and rate and on the simplicity (the number of scale items and the number of options per item), reliability and validity of the scales is not assessed for our target group. Besides, most of the scales do not cover the age group of our patients. Finally, the scales are not available in Dutch and are not always developed for nurse's use. These findings correspond with the statement of Henry and Foster [20], namely that several behavioural observation instruments have been developed in the past two decades to measure pain in infants and children, but none are specific for burn pain. In addition, Martin-Herz et al. [21] recommend specific investigation of pain assessment for children with burn injuries.

The similarity between the suggested scales is that they are developed to measure acute pain. Given the fact that procedural and background pain are types of acute pain, the literature was screened on Dutch behavioural scales to measure acute pain in this age group. Two instruments were found, namely the pain observation scale for young children (POCIS) [22], measuring postoperative pain after ear, nose and throat surgery and the COMFORT scale [23], measuring postoperative pain on the ICU. The POCIS, derived from the CHEOPS, which includes the three behavioural aspects according to Mills [15], seemed an appropriate scale for research. Firstly, the age group for which the scale was developed corresponds with our target group. Secondly, consisting of seven items with dichotomous answer categories, the POCIS is easy and quickly to use. Trained nurses need less than two minutes to observe in a structured way. Furthermore, an instruction video is available. As the VAS is a frequently used instrument for pain assessment by nurses in children [11], [14], [23], [25], [26], [29], [30], we decided to use the VAS, a scale with ratio properties, in our study as well.

The aim of this study was to assess if the POCIS and the VAS are reliable and valid instruments to measure procedural and background pain in burned children aged 0–4 years. The next research questions were composed:

• What is the reliability (inter-rater reliability, intra-rater reliability and internal consistency) of the POCIS?

• What is the reliability (inter-rater reliability, intra-rater reliability) of the VAS?

• What is the construct validity of the POCIS in relation to the VAS?