Household food insecurity, income loss, and symptoms of psychological distress among adults following the Cyclone Amphan in coastal Bangladesh

Introduction Cyclone Amphan swept into Bangladesh’s southwestern coast at the end of May 2020, wreaking havoc on food security and economic stability, as well as possibly worsening mental health. We studied the prevalence of post-cyclone stressors in adults following the cyclone and its association with symptoms of psychological distress. Methods We conducted a cross-sectional study in a coastal district of Bangladesh. A five-item brief symptom rating scale was used to measure the symptoms of psychological distress. Household food insecurity was measured using the USAID Household Food Insecurity Access Scale questionnaire. We estimated adjusted prevalence ratios (aPRs) using robust log-linear models adjusted for potential confounders. Results A total of 478 adults (mean [SD] age, 37.0[12.6] years; 169[35.4%] women) participated in the study. The prevalence of moderate-to-severe psychological symptoms and suicidal ideation was 55.7% and 10.9%, respectively. Following the cyclone, 40.8% of the adults reported severe food insecurity, and 66% of them reported moderate-to-severe mental health symptoms. Also, 54.4% of women and 33.7% of men reported severe food insecurity in the households. Moreover, 25.5% of respondents reported no income or a significant income loss after the cyclone, and 65.5% of them had moderate-to-severe psychological symptoms. Also, 13.8% of respondents reported housing displacement because of severely damaged houses, and 68.2% of them reported moderate-to-severe psychological symptoms. The high prevalence of mental health symptoms was found in women (aPR = 1.41, 95% CI = 1.06–1.82), people with severe food insecurity (aPR = 1.63, 95% CI = 1.01–2.64), and people who lost jobs or lost a major income source (aPR = 1.25, 95% CI = 1.02–1.54). Conclusion Following cyclone Amphan, many low-income individuals saw their income drop drastically while others were unemployed and living with severe food insecurity. The result suggests gender inequalities in food-security after the cyclone. Immediate action is needed to ensure household food-security for reducing the burden of mental illness. Rising opportunities of paid-jobs and decreasing income-loss, especially for the poor people, can have a protective impact on psychological distress. However, due to the high prevalence of severe psychological symptoms, long-term mental health services are required among the population of coastal Bangladesh.


S2 Appendix. Calculation of response rate
The response rate (RR) is the number of complete interviews with reporting units divided by the number of eligible reporting units in the sample. The following formula was used to calculate the response rate [1].
Where, RR1 = Response rate I = Complete interview P = Partial interview R = Refusal and break-off NC = Non-contact O = Other UH = Unknown if household/occupied HU UO = Unknown, other The details of these quantities for our study are given in the following: I and P=The in-person household survey was conducted in which housing units are sampled from an address-based sampling frame of eight selected camps using systematic sampling technique. We consider less than 50% of all applicable questions answered (with other than a refusal or no answer) equals break-off, 50%-80% equals partial, and more that 80% equals complete. We found complete answered from 478 participants (i.e., I=478. We found 8 of the participants did not complete the questionnaire for BRS-5 (i.e., P=8).
R= Refusals and break-offs consist of cases in which some contact has been made with the housing unit and a responsible household member has declined to do the interview, or an initiated interview results in a terminal break-off. Thirteen of the participants refused to take part in the interview, but no one took a terminal break-off after initiating an interview (i.e., R=13).
NC= Non-contacts in in-person household surveys consist of three types: a) unable to gain access to the building, b) no one reached at housing unit, and c) respondent away or unavailable. 6 housings were not accessible due to security condition, and we were not able to reach at housing unit during the interview period (i.e., NC=6).
O= Other cases represent instances in which the respondent is/was eligible and did not refuse the interview, but no interview is obtainable because of: a) death; b) the respondent is physically and/or mentally unable to do an interview; c) language problems; and d) miscellaneous other reasons. We did not face any language problem to exclude participants. We found 2 physically ill (bed-bound or palliative care) patients from household during the interview (i.e., O=2). UH= Cases of unknown eligibility and no interview include situations in which it is not known if an eligible housing unit exists and those in which a housing unit exists, but it is unknown whether an eligible respondent is present in the housing unit. It was not a scenario in our situation (i.e., UH=0).
UO= Not eligible cases for in-person household surveys include: a) out-of-sample housing units; b) not-a-housing unit; c) vacant housing units; d) housing units with no eligible respondent; and e) situations in which quotas have been filled. In a systematic sampling technique, we found a total of 33 households without any adult respondents during the interview (UO=33).

Reference:
[1] The American Association for Public Opinion Research. 2016. Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys. 9 th edition. AAPOR.

Income loss by gender
Income loss due to cyclone

Household food insequrity
Income loss due to cyclone