The effect of group involvement on post-disaster mental health: A longitudinal multilevel analysis

This manuscript version is made available under the CC-BY-NC-ND 4.0 license: http://creativecommons.org/licenses/by-nc-nd/4.0/ which permits use, distribution and reproduction in any medium, provided the original work is properly cited. This author accepted manuscript is made available following 36 month embargo from date of publication (November 2018) in accordance with the publisher’s archiving policyInvolvement in voluntary associations is a key form of social capital and plays an especially important role following disaster as a venue for coordination and decision-making for the wider community. Yet, relatively little attention has been paid to how group involvement affects mental health, at either the individual or community level. The aim of this study was to assess the impact of involvement in voluntary associations on mental health among residents of bushfire-affected communities. A longitudinal sample of 642 individuals affected by the 2009 Victorian bushfires in south-eastern Australia were surveyed in 2012 and 2014 (3- and 5-years post-disaster). A further subsample (n = 552) of residents residing continuously within 22 bushfire-affected communities were examined for community-level effects using multilevel regression methods. After adjusting for demographics, disaster exposure, and network variables, group involvement at time 1 bore a curvilinear relationship with PTSD at both time points: moderate involvement was most beneficial, with no participation, or high amounts, yielding poorer outcomes. High amounts of group involvement was likewise linked to a greater risk of major depression. Furthermore, communities with higher median levels of group involvement reported lower levels of PTSD symptoms and major depression two years later. With respect to group involvement, more is not always better. For individuals, moderation – if possible – is key. Meanwhile, community-level health benefits come when most people participate to some extent, suggesting that the distribution of involvement across the community is important.This research was supported through an Australian Research Council Linkage Grant (LP100200164), along with the contributions of these partner organisations: Outer East Health and Community Support Alliance, Bendigo Loddon Primary Care Partnership, Lower Hume Primary Care Partnership, Central West Gippsland Primary Care Partnership, Banyule Nillumbik Primary Care Alliance, Central Hume Primary Care Partnership, Australian Red Cross, Australian Rotary Health, Victorian Department of Health, Centrelink