Dynamic treatment regimes (DTRs) are sequences of decision rules that aim to optimize a patient's course of treatment using available information at each stage of follow-up. Much of the DTR methodology has been developed for use within observational datasets under the stable unit treatment value assumption (SUTVA), which states that an individual's outcome is independent of the decision rules followed by others at each stage of treatment. However, this often does not hold in practice due to different forms of interference that occurs within social networks, such as in infectious disease settings where an individual may be less likely to become infected if their neighbours are fully vaccinated. While much of the biostatistical literature has addressed the issue of partial interference, where interference occurs within groups but not across groups, much work has yet to be done for modifying DTR methodology to account for both within- and between-group interference within clustered and hierarchical data. This work explores a modified version of one such DTR methodology, dynamic weighted ordinary least squares regression (dWOLS), using network weights based on propensity score functions modelling both fixed and random effects within groups of patients to account for individual- and group-specific factors affecting the nature of interference present.