Investigating the Role of Social Networks in Antidepressant Medication Initiation Behaviors.

Abstract

Background: Medication adherence in depression continues to challenge patients and providers. Patients’ social networks are theorized to impact medication use decisions, particularly in patients with depression. Pescosolido (1992) proposed the Network Episode Model (NEM) as a dynamic model that integrates social networks in predicting patient behavior. This model had not yet been studied in medication adherence research. This study examined the influence of social networks on medication initiation desires and intentions in patients with depression. In addition to the effect of social networks, the role of illness perceptions and medication beliefs on decisions to initiate medication was evaluated. Methods: A randomized experimental post-test only study design was utilized. A 2 X 2-factor model was used to create patient scenarios where recommendation agreement between community and professional networks (conflicting versus supporting) and physician’s attitude toward community “lay” networks (positive and negative) were manipulated. The scenarios were written based on a hypothetical university student, recently diagnosed with depression. Participants were recruited at a large Midwestern University to participate in a 20-25 minute online survey (Qualtrics). Hypotheses were tested using ANOVA at the 0.05 level of significance. The medication initiation decision model was tested using Structural Equation Modeling (SEM). Results: The overall response rate for the study was 23.5% (n=226). The respondents were primarily Caucasian (76%), Female (58%), Undergraduate (63%) and around 11% self-reported a previous diagnosis or a history of depression. Lay network confirmation or conflict of prescribers’ recommendation had significant impact on both initiation desire and intention. However, professional network attitude toward community “lay” network involvement only affected initiation desire, with the expression of concern toward social network involvement in medication use decisions producing a higher desire to initiate medication therapy. The SEM showed that treatment concerns, beliefs about treatment necessity, and the perceived ability to control the illness were influential on the decision to initiate medication therapy. Conclusion: Results demonstrate the importance of both “lay” network recommendations and professional “treatment” network attitudes toward “lay” networks in medication initiation desires and intentions. Health care providers must be cognizant of the potential role social networks play in patient medication use decisions.