Evaluation of the diagnostic accuracy of nonverbal signs used by medical staff to assess postoperative pain: A prospective study
Postoperative pain relief is an essential component of postoperative patient care. However, assessing pain is difficult, because pain is subjective. An evaluation based on self-reported pain, such as the numeric rating scale (NRS) and visual analogue scale, is important for accurately assessing pain.1
However, medical staff often do not use any scale for self-reported pain for a variety of reasons, including a lack of trust in the ability of such scores to accurately measure patients’ level of pain.2 Instead, medical staff rely on nonverbal signs of pain (NVSP), for example, facial expressions. Although these signs are commonly used in the clinical setting, reliance on only the NVSP can underestimate pain and result in inadequate postoperative pain control.2 Additionally, to evaluate pain, medical staff often ask patients questions such as ‘Do you have pain?’ and/or ‘Do you need analgesics?’ Careful consideration of the patients’ complaints of pain has been suggested to improve the accuracy of pain evaluation.1
Although it is generally accepted that the diagnostic accuracy of the NVSP is considerably lower than that of patients’ complaints of pain, no study has supported this. We objectively evaluated the diagnostic accuracies of these two methods of pain assessment.
Ethical approval was provided by the institutional ethics committee of Nagaoka Red Cross Hospital, Nagaoka, Japan (Chairperson Dr Hideo Morishita) on 3 September 2012. Written informed consent was obtained from all patients. Patients aged at least 20 years who underwent surgery under general anaesthesia and were subsequently admitted to the surgical ICU (SICU) were enrolled. Patients with disturbed consciousness and communication disorders were excluded.
SICU nurses were trained to observe NVSP and use the NRS. Pain assessment started on the morning of the next day following surgery with the evaluation of the NVSP. They assessed NVSP before any care or treatment was provided. The following nonverbal sings listed in the acute pain subscale of the North American Nursing Diagnosis Association International (NANDA-I) were assessed: facial mask, position to avoid pain, protective gestures, diaphoresis, irritability, sighing, restlessness and moaning.3 Then the nurses asked whether patients had pain, and if they complained of pain, the nurses asked whether the pain was difficult to ignore. Next, the nurses asked whether the patients requested analgesics. Finally, the nurses assessed the NRS score, which comprised a scale from zero (no pain) to 10 (worst imaginable pain).
In total, 289 patients admitted to the SICU after undergoing surgery under general anaesthesia were screened. Of these, 14 patients were excluded (eight were less than 20 years old and six had communication disorders); therefore, 275 were enrolled. During the initial screening of the collected data, we found that the data for 60 patients could not be used because of inadequate records (n = 35), a sudden change in condition (n = 2), inability to understand the NRS (n = 2) or because the patient was discharged from the SICU on the same day of surgery (n = 21). Ultimately, 215 patients were analysed. Table 1 shows the patients’ demographic characteristics.
The sensitivity and specificity of the NVSP and patients’ complaints of pain and requests for analgesics were evaluated as follows. A positive finding was defined as the observation of a sign, whereas a negative finding was defined as the nonobservation of a sign. Significant pain was defined as pain that was difficult to ignore.
Table 2 shows the sensitivity and specificity (95% confidence interval) of the NVSP, and patients’ complaints of pain and requests for analgesics. All sensitivity values for the NVSP were low, ranging from 0.10 to 0.27, whereas the specificity values were high, ranging from 0.94 to 0.99.