Mental Illness Prevalence and Disparities Among Hurricane Sandy Survivors: A Two-Year Retrospective

Background/Purpose:  Hurricane Sandy struck New Jersey (NJ) and New York (NY) on October 29, 2012, killing more than 100 persons and affecting millions of residents. Literature suggests that natural disasters have significant potential impacts on the mental health of survivors. Racial disparity in mental health was documented after Hurricane Katrina in New Orleans, but this aspect has not been reported after Sandy. Two of our study questions aim to redress this: 1) Did the racial/ethnic subgroups found at risk for or reporting mental illness after Katrina experience similar vulnerability after Sandy? 2) Were sociodemographic characteristics associated with mental illness in other disaster literature, such as employment, income, and education, also associated with mental illness after Sandy?

Methods: The study utilized weighted survey data from Associated Press NORC (AP-NORC) interviews in 2014 with 1,009 Sandy survivors from twelve affected neighborhoods in NY and NJ. We used K6 prevalence measures of “probable mental illness” to classify the respondents into “non-mental illness” (NMI), “mild/moderate mental illness” (MMI), and “serious mental illness” (SMI).  We used ordered logistic regression to analyze mental health status among Sandy survivors with respect to the characteristics listed in the research questions plus the level of direct impact from Sandy that respondents reported (not at all, a little, moderate, very, extremely affected). Given the sample weights of this survey data, we used Taylor linearization method to calculate standard errors and variance estimators.

Results: Sandy survivors exhibited levels of mental illness that were about twice the national average (11% versus 6% for MMI; 6% versus 3% for SMI respectively). Although representation of blacks is much smaller in our study (11% versus 56%) compared to post-Katrina studies, we also found that blacks were more susceptible than whites to MMI and SMI (Odds Ratio-OR= 3.03, p=0.038). Additionally, Hispanics were more likely to have MMI and SMI compared to whites (OR= 2.11, p=0.019). Further, the higher probable SMI (12%) and MMI (16%) among those with higher unemployment (as well as less income and education) are largely attributable to the disproportionately higher number of blacks and Hispanics than whites with these characteristics. Finally, higher levels of perceived impact from Sandy were strongly associated with higher rates of mental illness and accounted for much of the disparity between both blacks and Hispanics compared to whites.

Program/Practice Implications: The severity of mental illness among Sandy survivors differed significantly among racial and ethnic groups, and was strongly associated with their employment status and the perceived impact of Sandy on their lives. These differences suggest that psychosocial interventions after such disasters, such as psychiatric services and re-employment services, should focus more heavily on socially vulnerable subgroups. While the high unemployment rate (40%) among Sandy victims may have been influenced by the Great Recession, the strong relationship between unemployment and mental health in this study suggests that mental health interventions after natural disasters should also have an employment component.  Future research will further elaborate the intersection of disasters, unemployment, and racial/ethnic subgroups after US hurricanes in the 2000s.