Pain, comorbidities, and clinical decision-making: conceptualization, development, and pilot testing of the Pain in Aging, Educational Assessment of Need instrument

Abstract

Introduction: Pain is highly prevalent in older adults and often contextualized by multiple clinical conditions (pain comorbidities). Pain comorbidities increase with age and this makes clinical decisions more complex. To address gaps in clinical training and geriatric pain management, we established the Pain in Aging-Educational Assessment of Need (PAEAN) project to appraise the impacts of medical and mental health conditions on clinical decision-making regarding older adults with pain. We here report development and pilot testing of the PAEAN survey instrument to assess clinician perspectives.

Methods: Mixed-methods approaches were used. Scoping review methodology was applied to appraise both research literature and selected Medicare-based data. A geographically and professionally diverse interprofessional advisory panel of experts in pain research, medical education, and geriatrics was formed to advise development of the list of pain comorbidities potentially impacting healthcare professional clinical decision-making. A survey instrument was developed, and pilot tested by diverse licensed healthcare practitioners from 2 institutions. Respondents were asked to rate agreement regarding clinical decision-making impact using a 5-point Likert scale. Items were scored for percent agreement.

Results: Scoping reviews indicated that pain conditions and comorbidities are prevalent in older adults but not universally recognized. We found no research literature directly guiding pain educators in designing pain education modules that mirror older adult clinical complexity. The interprofessional advisory panel identified 26 common clinical conditions for inclusion in the pilot PAEAN instrument. Conditions fell into three main categories: "major medical", i.e., cardio-vascular-pulmonary; metabolic; and neuropsychiatric/age-related. The instrument was pilot tested by surveying clinically active healthcare providers, e.g., physicians, nurse practitioners, who all responded completely. Median survey completion time was less than 3 min.

Conclusion: This study, developing and pilot testing our "Pain in Aging-Educational Assessment of Need" (PAEAN) instrument, suggests that 1) many clinical conditions impact pain clinical decision-making, and 2) surveying healthcare practitioners about the impact of pain comorbidities on clinical decision-making for older adults is highly feasible. Given the challenges intrinsic to safe and effective clinical care of older adults with pain, and attendant risks, together with the paucity of existing relevant work, much more education and research are needed.

Keywords: chronic pain; clinical decision-making; geriatric; interdisciplinary; interprofessional; multimorbidity; pain education; survey instrument.

Figure 1

Population flow diagram for scoping data review. Medicare beneficiaries meeting study criteria were selected as illustrated.

Figure 2

Claims data scoping review—identification of potential key elements. Diagnostic rate estimates for common condition ICD-10 codes in CMS Older adults 75–80 years old (yo). (A) Pain condition rate estimates, shown here are the most commonly utilized ICD-10 codes associated with these diagnoses. Pain-associated conditions marked with “*” were included in the assessment instrument. (B) Non-pain conditions selected for this study, shown here are the rate estimates for the most commonly used ICD-10 codes associated with each. Error bars indicate corrected 95% confidence interval, n = 100, net p = 0.05. The gray-scale background in each figure is to alert the viewer to the different x-axis scales.

Figure 3

Pilot PAEAN instrument. Shown is the stem (Prompt) and list of conditions included in the final instrument. The instrument is constructed by placing the stem at the top of the page with the list of conditions along the left margin, each with a Likert scale to the right (one scale associated with each condition). The instruction for the instrument was: “Please rate your agreement with the following statement regarding each of the conditions below:” A 5-point Likert scale (strongly disagree to strongly agree) was used.

Figure 4

Preliminary assessment of respondent agreement that specified conditions impact pain-related clinical decision-making. (A) Cardio-vascular-pulmonary conditions may be viewed a variably impactful. (B) Metabolic conditions may be viewed as relatively less impactful although pilot data suggest that diabetes, renal impairment and hepatic impairment may have a strong impact on decision-making. (C) Pilot data suggest that neuro-psychiatric conditions have a major impact on pain-related clinical decision-making. Error bars represent the 95% confidence interval, corrected for multiple comparisons, n = 26, net p = 0.05. For clarity, error bars are shown in one direction only but pertain bidirectionally, with adjustments for floor (near zero) and ceiling (near 100%) effects.