Physician perception and recommendation of insulin pens for patients with type 2 diabetes mellitus

Objective: To determine the way in which physicians perceive insulin pens and the factors that affect their recommendations and their patients' pen initiation/use. Research design and methods: Primary care physicians (N = 300; 108 family practice, 192 internal medicine) and endocrinologists (N = 225) were recruited from national databases (IMS and AMA) to participate in an Internet survey. All data are self-reports. Measures include five dependent variables (the extent of suitability for patient segments, two pen-related behaviors - their frequency of presenting patients with the pen as a treatment option and strength of pen recommendation - and extent of their patients' pen initiation and use) and several potential correlates: practice characteristics; therapeutic philosophy and practices; perceptions of insulin pens (convenience, facilitation of self-care, blood glucose control efficacy, cost/coverage). Correlates of dependent variables were examined using multiple linear regression. Results: Significantly (p < 0.05) more patient pen use and/or successful pen initiation was reported by physicians who: (1) were more involved in clinical practice, were early adopters of clinical innovations or instructed their patients regarding insulin use; (2) reported less insulin mixing or therapeutic inertia; (3) perceived pens as efficacious and facilitating self-care; and (4) presented and recommended pens to their patients. More frequent presentation of pens to patients and/or stronger recommendation of pen use was reported by physicians who were early innovation adopters, reported less insulin mixing or therapeutic inertia, and perceived pens as convenient and efficacious (p < 0.05). Limitations: The study is a cross-sectional survey using a self-administered questionnaire yielding unverified self-reports. Physicians treating few patients with diabetes mellitus or prescribing no insulin were excluded from the study. Conclusions: Physicians' pen-related actions are a function of their personal and practice characteristics as well as their perceptions of pens. Physicians' presentation of pens as an option is strongly associated with perceived pen convenience, but physicians' pen recommendations and the estimated pen use/initiation of their patients are most strongly associated with the perception of clinically relevant pen attributes - glucose control efficacy and/or self-care facilitation.