JMIR Publications

ABSTRACT

Background:

In recent years, people with mental health problems are increasingly using online social networks to exchange social support. For example, in online depression communities, patients could share their experiences, provide valuable information, and obtain emotional support to fight against the diseases. However, it is still a critical issue that how they develop online social support networks to exchange informational support and emotional support.

Objective:

We aim to investigate which user attributes have significant effects on the formation of the informational support network and emotional support network in online depression communities, and further examine whether there is an association between the two social networks.

Methods:

This study attempted to use social network theory and construct exponential random graph models to help understand the informational support network and the emotional support network in online depression communities. Then, we retrieved available data composed of 74,986 topic posts from 1,077 members in an online depression community in China. An informational support network of 1,077 participant nodes and 6,557 arcs and an emotional support network of 1,077 participant nodes and 6,430 arcs were constructed respectively to examine the endogenous effects (purely structural effects) and exogenous effects (actor-relation effects) for both support networks separately and cross-network effects between the two networks.

Results:

The results found some important structural features on the formation of two support networks, such as reciprocity (r=3.6247, p<.001; r=4.4111, p<.001) and transitivity (r=1.6232, p<.001; r=0.0177, p<.001). The results also provide support for the effects of some individual factors on the formation of the two networks respectively. There are no significant homophily effects for gender (r=0.0783, p=0.2043; r=0.1122, p=0.2462) in the two support networks. There is no tendency for the users who have high influence (r=0.3253, p=0.0529) and write more posts(r=0.3896, p=0.0676) or newcomers (r=-0.0452, p=0.6627) to form informational support ties more easily. But long-term users (r=0.6680, p<.001) or those who provide more replies to other posts (r=0.5026, p<.001) are more likely to form informational support ties. Those users who have high influence (r=0.8325, p<.001), spend much time online (r=0.5839, p<.001), write more posts (r=2.4025, p<.001), and provide more replies to other posts (r=0.2259, p<.001), are more likely to form emotional support ties. But newcomers (r=-0.4224, p<.001) are less likely to receive emotional support as compared to experienced old-timers. Besides, we found that there are a significant entrainment effect (r=0.7834, p<.001) and a non-significant exchange effect (r=-0.2757, p=0.3219) between the two networks.

Conclusions:

These findings have important theoretical and practical implications for online depression communities. The managers of online depression communities should employ different policies for different types of patients to develop their social support networks and according to their individual characteristics.