Correlation of Patient Activation Measure Level with Patient Characteristics and Type of Vascular Disease

Abstract

Patient activation or level of engagement in one's medical care is linked to hospital readmissions and worse outcomes in a number of diseases. Patients with higher levels of activation are typically guiding their care rather than acting as passive observers of care. This study aims to determine if either patient demographics or type of vascular disease can predict patient activation. All patients presenting over a 4-month period to an outpatient vascular clinic were asked to complete the Patient Activation Measure (PAM) survey. In total, 257 completed surveys were collected. Survey responses were scored on a Likert scale with anchors. Responses are tallied with a score of 1-100 and converted to summary levels 1-4 in accordance with the previously validated scoring system. Level 1 patients are considered disengaged and overwhelmed. Patients in level 2 are becoming aware of their health care, but still struggle. Level 3 patients are taking action, while level 4 represents patients who are maintaining healthy behaviors and pushing further. Chi-squared test and multivariable regression were then performed to determine if patient characteristics or type of disease correlated with activation levels. In total, 257 patients completed the survey. The mean participant age was 67years (±15). Sixteen percent of patients lived alone, 58% were married, and in 39% mean household income was <$50,000. Overall, 21 patients (8.2%) were classified as level 1, 65 (25%) level 2, 94 (37%) level 3, and 77 (30%) level 4. The group comprised 32% peripheral artery disease (PAD), 20% carotid, 18% aortic/aneurysm, 14% venous, and 16% were various other vascular diseases. Over each disease group there was a wide range of activation, but no significant difference between the type of vascular disease and activation level. Chronic limb-threatening ischemia (CLTI) patients comprised 35% (n=29) of the PAD group, and 66% of these patients reported an activation level of 3 (n=10) or 4 (n=9). There was no difference in the levels of activation reported by the CLTI patients compared to the general PAD cohort (P=0.99). Multivariable analysis demonstrated that age, level of education, household income, and type of vascular disease correlated with PAM score, but there was no correlation between length of symptoms, race, or gender. Patient activation is unpredictable using patient characteristics or type of vascular disease, and CLTI patients report high activation levels. Quality databases that collect only patient demographics may not fully capture patient predictors of poor outcomes. Use of the PAM survey should be further explored in vascular patients to correlate activation level with vascular-specific outcomes.