Quality Improvement Project: Replacing the Numeric Rating Scale with a Clinically Aligned Pain Assessment (CAPA) Tool

Abstract

CAPA is a multifaceted pain assessment tool that was adopted at a large tertiary Midwest hospital to replace the numeric scale for adult patients who could self-report their pain experience. This article describes the process of implementation and the effect on patient satisfaction scores. Use of the tool is supported by the premise that pain assessment entails more than just pain intensity and that assessment is an exchange of meaning between patients and clinicians dependent on internal and external factors. Implementation of the tool was a transformative process resulting in modest increases in patient satisfaction scores with pain management. Patient reports that “staff did everything to manage pain” had the biggest gains and were sustained for more than 2 years. The CAPA tool meets regulatory requirements for pain assessment.

Introduction

In December 2013, University of Minnesota Medical Center (UMMC) embarked on a quality improvement project that helped transform the culture of pain assessment. The institution struggled with low patient satisfaction scores regarding pain management. Although multiple pain assessment tools were available to staff depending on patient preference or need, the default tool was the numeric (0-10) scale for adult patients who could self-report pain.

Most of the pain assessment tools used by patients to self-report their pain are only one dimensional, indicating only pain intensity. They are quick, easy to use, and economical. Examples include the numeric rating scale, the visual analog scale, Faces Pain Scale–Revised, Iowa Pain Scale, and verbal descriptor scale. Multidimensional scales that include sensory, affective, and evaluative aspects are used in outpatient settings for the management of persistent pain. Examples include the McGill Pain Questionnaire, Chronic Pain Grade Scale, Short Form 36 Bodily Pain subscale, and Brief Pain Inventory. Because these tools require 5-30 minutes of a patient's attention and are more complex to score, they are less useful for assessment of pain in an acute care setting.

Unfortunately, any one-dimensional scale intended to measure intensity of pain does not do justice to the complexity of the patient's pain experience. A numeric value, for example, is but one data point that is meaningless until it is put into context. De C. Williams, Davies, and Chadury (2000) studied the use of simple pain ratings scales. They found patients with chronic pain had difficulty rating pain by a single score. One patient noted difficulty with identifying maximum pain, stating the number is influenced by temporal factors and a sense of desperation.

Franck and Bruce (2009) suggested that “poor compliance with pain assessment guideline may indicate unspoken resistance to use of methods that are overly simplistic, burdensome to patients, often inaccurate and perhaps even disrespectful of clinical expertise and experience” (p. 19). Von Baeyer and Pasero (2017) further asserted that nurses develop “work arounds” to assess pain because “pain scoring is complex, divorced from common sense, does not take account of the social relationship and communication with patients” (p. A-2).

Schiavenato and Craig (2010) claimed pain assessment is a social transaction based on a complex inter-play between the patient and clinician. This concept introduces the clinician as a crucial part of the process. They noted that by including the clinician in assessment, a social interaction is created encompassing the subjective statements from patients with the objective observations of the clinician. Pain assessment becomes an ongoing process that goes beyond a one-dimensional approach.

Pain intensity alone, as measured by the numeric rating scale (NRS), cannot capture the complexity of the pain experience. Additionally, as Schiavenato and Craig (2010) noted, patients modulate pain numbers to express what they believe is in their best interest. For example, patients may inflate number scores to make sure that clinicians appreciate their suffering and elicit an intervention. Other patients may deflate number scores to avoid a painful injection or the side effects of an analgesic. The number given by the patient may not reflect the patient's functional status. For example, a patient may give a low number yet not be able to move because of the pain or give a high number despite oversedation.

As Schiavenato and Craig (2010) explained, patients and clinicians are affected by contributing factors. Table 1 shows how contributing factors mediate pain assessment for both the patient and clinician. The major factors are biologic, sociocultural, developmental, psychologic, experience-empathy, and contextual-situational.

While patients' pain may be affected by their disease process or medications' effects, clinicians may also be dealing with their own pain or reactions to stress. Patients' expression of pain may be modified by gender or cultural origins. Clinicians' interpretations may be colored by biases in response to patients' appearances, social status, and ethnicities. Psychological or developmental factors complicate patients' and clinicians' interpretations of a painful situation. For example, a history of drug addiction may cause fear in a patient destined for surgery, whereas the clinician wrestles with personal views of the patient's comorbidity of substance abuse. Both patient and clinician bring their past experiences with pain to each new interaction; a clinician's level of knowledge and empathy can influence the patient's experience.

In a survey of American Pain Society members, Bakonja and Farrar (2015) found many pain specialists did not use a pain rating scale in their practice because of its inadequacy. They suggested there was a need to develop a new clinically acceptable pain assessment tool that includes “impact of pain on patients and their function” (p. 1249).

In 2012, University of Utah Health Care (2016) developed and piloted a pain assessment tool called CAPA (Clinically Aligned Pain Assessment) after identifying limitations of the NRS. Their tool consisted of five questions regarding comfort, change in pain, pain control, functioning, and sleep (Table 2). Clinician and patient engage in a conversation that addresses the five domains. There is no script. The tool provides standard wording for encoded evaluation of the conversation.

In a pilot study, Donaldson and Chapman (2013) replaced University of Utah's conventional NRS (0-10 scale) with the CAPA tool on selected units. A year after implementation of the CAPA tool, University of Utah's (2016) Press Ganey pain satisfaction scores increased from the 18th percentile to the 95th percentile. This dramatic increase was present on those units that fully embraced use of the tool. Their survey found that 55% of the patients preferred the CAPA tool over the numeric scale. They also found that nurses preferred the CAPA tool by 3 to 1 over the numeric rating scale. Reliability and validity testing were not reported by the University of Utah. Dr. G. Donaldson (personal communication, January 15, 2014), from the University of Utah, stated “the tool was valid for clinical management, but had not been validated for research in clinical studies because it lacked a set of scoring rules for how to convert the responses to numbers.” He recommended using nonparametric tests for any research using the tool.

Before implementation at University of Minnesota Medical Center, providers on the pain and palliative care services were asked to comment on the validity of the tool. Their consensus was that the tool had content or face validity for measuring pain.

The CAPA tool recognizes that the experience of pain is more than just a number. In addition to measuring the intensity of pain, it also examines the effect of pain on functionality and sleep, as well as the efficacy of therapy. There are no scripted questions to ask the patient. The expectation is that the patient and clinician engage in a brief conversation about pain that leads to a coded evaluation. At no point does the patient “rate” anything or check boxes. The health care provider does not supply the answers and ask the patient to choose one. The answers emerge through the conversation. Some sample questions during the assessment of pain may include the following:

• How comfortable are you?

• Are you experiencing any pain?

• Is your discomfort improving or worsening?

• Has the medication/heating pad/ice helped manage your pain?

• Are you able to do what the staff is asking you to do (e.g., walking, coughing, physical therapy)?

• Have you been able to sleep? Is the pain waking you up?

Impressed with the University of Utah's success, UMMC decided to adopt the CAPA tool for pain assessment in areas providing adult health care. The goal was to improve patients' experience with pain management as measured by Press Ganey satisfaction scores.

The implementation of a multidimensional assessment tool in an acute care setting that replaces a one-dimensional intensity tool represents a cultural shift. The numeric scale is ubiquitous; it is found throughout clinical practice and documentation tools. The assumption has been that a numeric rating counts as an assessment of pain, yet the number represents only one piece of data. Without putting the number into a context, the number is meaningless. A simple analogy is that of taking a pulse. A heart rate of 90 is meaningless unless one determines whether the patient has been exercising or was at rest, has an elevated temperature or not, is in severe pain or not.