The misuse of prescription drugs is a pressing public health crisis in the U.S. It is fuelled by high-risk prescribing. Building on the sociology of deviance (Becker, 1963) we argue that high-risk prescribing consists of two distinct practices. The first – routinely prescribing to patients whose prescription-fill patterns are consistent with misuse or abuse – conforms with the sociological definition of deviance. The second – routinely overprescribing to patients within possible bounds of medical use – does not, and we refer to it as “liminal.” To explore the social structures in which liminal and deviant prescribers are embedded, we constructed patient-sharing networks of 500,472 physicians from 213.9 million prescriptions over a four-year period. We find that deviant and liminal prescribers have starkly different networks: an increase in the size or cohesion of the network is associated with a reduction in liminal prescribing but an increase in deviance. We also find no cross-practice social influence, suggesting a lack of “slippery slope” in which liminal prescribing escalates over time and becomes deviance. Liminal prescribing is far more prevalent than deviance (16.14% compared to 2.38%) and accounts for most of the oversupplied benzodiazepines (55.8% compared to 8.7%). Our study highlights the distinctiveness of liminal practices and the importance of studying them.
We utilize longitudinal social network data collected pre–COVID-19 in June 2019 and compare them with data collected in the midst of COVID in June 2020. We find significant decreases in network density and global network size following a period of profound social isolation. While there is an overall increase in loneliness during this era, certain social network characteristics of individuals are associated with smaller increases in loneliness. Specifically, we find that people with fewer than five “very close” relationships report increases in loneliness. We further find that face-to-face interactions, as well as the duration and frequency of interactions with very close ties, are associated with smaller increases in loneliness during the pandemic. We also report on factors that do not moderate the effect of social isolation on perceived loneliness, such as gender, age, or overall social network size.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
'/%3e%3cpath d='M200 752.362h200v300H200z' style='fill:%23fff;fill-opacity:1;stroke:none' transform='translate(0 -752.362)'/%3e%3cpath d='M400 752.362h200v300H400z' style='fill:%23ff883e;fill-opacity:1;stroke:none' transform='translate(0 -752.362)'/%3e%3c/svg%3e)