Social influence on 5-year survival in a longitudinal chemotherapy ward co-presence network

Abstract

Chemotherapy is often administered in openly designed hospital wards, where the possibility of patient-patient social influence on health exists. Previous research found that social relationships influence cancer patient's health; however, we have yet to understand social influence among patients receiving chemotherapy in the hospital. We investigate the influence of co-presence in a chemotherapy ward. We use data on 4,691 cancer patients undergoing chemotherapy in Oxfordshire, United Kingdom who average 59.8 years of age, and 44% are Male. We construct a network of patients where edges exist when patients are co-present in the ward, weighted by both patients' time in the ward. Social influence is based on total weighted co-presence with focal patients' immediate neighbors, considering neighbors' 5-year mortality. Generalized estimating equations evaluated the effect of neighbors' 5-year mortality on focal patient's 5-year mortality. Each 1,000-unit increase in weighted co-presence with a patient who dies within 5 years increases a patient's mortality odds by 42% (β = 0.357, CI:0.204,0.510). Each 1,000-unit increase in co-presence with a patient surviving 5 years reduces a patient's odds of dying by 30% (β = -0.344, CI:-0.538,0.149). Our results suggest that social influence occurs in chemotherapy wards, and thus may need to be considered in chemotherapy delivery.

Keywords: administrative data; chemotherapy; communal coping; generalized estimating equation; jaccard index; longitudinal network; medicine; public health.

Fig. 1

Layout of the chemotherapy ward. Patients begin spells in the waiting room, and are taken to either treatment room 1 or 2 depending on a number of factors.

Fig. 2

Predicted probability of 5-year mortality for patients with varying risk profiles and potential for social influence. Across panels, the first bar represents the predicted probability from model 4 with 0 for all influence terms. The average patient was one who had the median values for all covariates (rounded for dichotomous and categorical variables). This equates to a 69-year-old female whose chemotherapy lasted nine visits over 3 months and spent 30 hours in the ward starting in 2005, with a single diagnosis of a tumor of the ovaries. The low-risk and high-risk patients had values based on the first and third-quartile of the covariates depending on whether the relationship between 5-year mortality and the covariate was negative or positive, respectively. The low-risk patient was a 61-year-old female who visited the ward nine times over the course of a month and spent 30 hours in the ward starting in 2007, with a single tumor of the breast. The high-risk patient was a 79 year-old male whose chemotherapy included two visits to the ward over 4 months and spent 30 hours in the ward starting in 2003, whose primary diagnosis was cancer of the stomach, but had multiple cancer diagnoses. It is important to stress that these patients are not necessarily observed in these exact combinations of covariates; they are chosen in the way they were to demonstrate heterogeneity of the predicted probability of survival. Within each panel, influence terms were given the rounded mean value for the variable in question (refer to Table 2). No influence means the patient was co-present with no-one (never actually observed but gives a baseline probability). “Alters survive” means a patient was only co-present with patients surviving at least 5 years, and “alters die” means a patient was only co-present with patients dying within 5 years. “Both” means a patient was co-present with both types of patients.