Multiple indicators multiple causes modeling of socio-demographic disparity effects on word-of-mouth consumer behavior towards healthcare purchasing

Word-of-mouth (WOM) expressed as basic face-to-face WOM (bWOM) and electronic or internet-based WOM (eWOM) is the most common form of communication among consumers. The impact of socio-demographic characteristics on WOM preferences among consumers of pharmaceutical products is a largely unexplored area. The study aimed to examine the role of socio-demographic attributes or covariates (gender, age, educational status, and employment status) on preference for WOM behavior, and also, to determine the comparative impact of bWOM and eWOM on purchase intentions among consumers. A cross-sectional study of 1,118 randomly selected customers of community pharmacies in Nigeria with online questionnaires for data collection. Based on latent variable modeling and cognitive dissonance theory, hypotheses were developed and tested using the Multiple-Indicators Multiple-Causes (MIMIC) co-variance-based structural equation modeling technique. Results revealed that the developed measurement model for the study was adequately fitted and valid. The two-factor structure of WOM was confirmed. Positive correlations were found among exogenous covariates such as age, gender, employment, and educational status. Gender and educational status covariates did not influence respondents' preferences. Younger customers had a higher preference for bPOM. Unemployed respondents had higher tendency to use bWOM and eWOM compared to employed respondents. Also, eWOM had a stronger influence on purchase intentions compared to bWOM. Policymakers and regulators in the pharmaceutical industry should use knowledge of consumers' demographic disparities and communication preferences in formulating consumer awareness and education programs about proper drug use on social media platforms. The study adds to the literature by developing an integrated model to examine the effects of demographic disparities on consumer preferences using the MIMIC framework.