Development of an instrument to measure barriers to arthritis pain management in older adults
Hiemenz, Melinda Lee
MetadataShow full item record
The purpose of this methodological study was to develop and estimate the psychometric properties of the Barriers to Arthritis Pain Management Instrument (BAPMI), an instrument intended to measure older adults' perceptions of barriers that prevent them from managing their arthritis pain. The BAPMI was developed based on findings from a focus group study with older adults with arthritis in which pain management barriers were identified. The sample consisted of 174 independent living older adults who attended senior centers. Inclusionary criteria included: (a) being 65 years of age and older, (b) able to speak and read English, and (c) having a diagnosis of arthritis. Psychometric properties of the resulting 16-item BAPMI, the final version of the initial 28 items tested, included content validity (CVI = .84), internal consistency (alpha = .87), and stability ( r = .82, p = .01). Item reduction using a combination of Cronbach's alpha and principal components analysis (PCA) with varimax rotation, resulted in the final scale with high estimates of content validity, internal consistency, and stability. Concurrent validity was estimated using the Barriers Questionnaire, an instrument developed to measure barriers to cancer pain management. While a moderate correlation was expected, a low correlation (r = .25, p. = .01) demonstrated that the barriers interfering with cancer and arthritis pain management are different. Construct validity was estimated using: (a) PCA for exploration of the BAPMI's possible underlying dimensions, and (b) staged path analysis for testing a model that included Barriers as a construct. PCA revealed four underlying dimensions: (a) Emotional Distress, (b) Pain-Movement Cycle, (c) Relationship with Health Care Provider, and (d) Age-Related Expectations, all of which were identified as themes in the focus group study. The model tested, using staged path analysis, included Chronic Pain Experience (CPE) as the dependent variable and five independent variables: Age, Functional Disability, Pain Intensity, Self-Efficacy, Barriers, and Perceived Pain Management Success (PPMS). An R2 of .46 was obtained when CPE was regressed on the IVs of PPMS, Barriers, and Self-Efficacy. All the relationships with Barriers were in the predicted direction except age, which had a low and inverse relationship with Barriers.