key: cord-0330488-1pji0plq
authors: Price, A.; Contreras-Suarez, D.; Zhu, A.; Schreurs, N.; Measey, M.-A.; Woolfenden, S.; Burley, J.; Bryson, H.; Efron, D.; Rhodes, A.; Goldfeld, S.
title: The impact of ongoing COVID-19 lockdown on family finances and mental health
date: 2021-08-20
journal: nan
DOI: 10.1101/2021.08.15.21262087
sha: 89723b5254a51ddbeb7bed25c31252cb15d7c16c
doc_id: 330488
cord_uid: 1pji0plq

Objectives: Australia's public health restrictions ('lockdown') in 2020 successfully contained the spread of COVID-19. These included a national 'initial' lockdown (March-May), and 'ongoing' lockdown (July-November) for metropolitan Victorian residents only. Australia's experience offers an opportunity to assess impacts of lockdown on families with children, in the relative absence of disease morbidity and mortality. This study (1) described the experience of initial lockdown and (2) evaluated the impact of ongoing lockdown, on family finances and mental health. Methods: Data were drawn from the June and September 2020 Royal Children's Hospital National Child Health Polls. Caregivers of children from the states of Victoria and New South Wales reported on job/income loss; material deprivation (inability to pay for essential items); income-poverty; mental health (Kessler-6); impact on caregiver/child mental health; and caregiver/child coping. Data from N=1207/902 caregivers in June/September were analyzed; Aim (1) with weighted descriptives; Aim (2) with Difference-in-Difference adjusted linear regression models (New South Wales provided the comparator). Results: Following initial lockdown, one-quarter of families reported job/income loss; one-third reported material deprivation. Negative impacts on mental health were reported for half the caregivers and one-third of children. Few caregivers or children had difficulties coping. During Victoria's ongoing lockdown, job/income loss increased by 11% (95%CI: 3-18%); Kessler-6 poor mental health by 6% (95%CI: 0.3-12%) and negative mental health impacts by 12% for caregivers (95%CI: 6-23%) and 14% for children (95%CI: 4-20%). Female (versus male) caregivers, metropolitan (versus regional/rural) families, and families with elementary school-aged children (versus pre-/high-school) were most affected. Conclusions: Ongoing lockdown had negative impacts on mental health, employment, and income, but not deprivation or poverty, likely because of the government income supplements introduced early in the pandemic. Balancing the benefits and harms of lockdown requires planned responses to outbreaks, and evidence-informed financial and mental health supports.

"difficulty paying bills or for necessities", "working longer hours", "filing for unemployment"; "applying for 1 4 0 Government assistance"; "reduced work hours"; "reduced total household income" or "none of the above". A 1 4 1 binary variable describing any job loss (by one or two caregivers) or reduction in income due to COVID-19 1 4 2 (versus not) was created to enable comparison with other Australian studies, e.g. (27) 1 4 3 2. Eight items adapted from the Household, Income and Labour Dynamics in Australia (HILDA) Survey Wave 1 4 4 18 Household Questionnaire Material Deprivation Module (28) asking "In the last month, because of money 1 4 5 pressure did you miss or put off" (response options: "yes" versus "no"): mortgage or rent repayments; 1 4 6

We chose to run linear regression models as interpretation is simpler than logistic regression and most of the 1 9 9 predicted y-values were bounded within 0-1. To check the robustness of these models, we ran marginal effects 2 0 0 probit models for the dichotomous outcomes for the whole sample (not presented), which confirmed the linear 2 0 1 regression output. For all models, the adjusted analyses are presented as the main results (unadjusted analyses are 2 0 2 not included). As 293 caregivers preferred not to disclose their income, these analyses should be interpreted with 2 0 3 caution. Data were analyzed with Stata v13. September, during Victoria's ongoing lockdown, 1769 caregivers were invited and 1434 (81%) completed the 2 1 0 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 20, 2021. ;

Poll. Of these, 902 caregivers (460 Vic, 442 NSW) with 1584 children (786 Vic, 798 NSW) were included in the 2 1 1 analysis. 2 1 2 Table 1 presents the weighted estimates for participant characteristics. Caregivers were an average of 40 years 2 1 3 old, and three in five respondents were female. Each caregiver looked after an average of two children, and two in 2 1 4 five cared for children with additional needs. One in five caregivers had a Health Care Card (an indicator of low 2 1 5 income) and one-quarter were sole caregivers. One-quarter of respondents were born outside of Australia, and a 2 1 6 similar proportion spoke a language other than English at home. One in six identified as living in regional or rural 2 1 7 areas, compared with metropolitan capital cities. Ninety-five percent of respondents had completed high school 2 1 8 (compared with around 85% nationally)(32) and less than 1% identified as Aboriginal or Torres Strait Islander 2 1 9

(compared with 3% nationally).(33) 2 2 0 Experiences after initial lockdown (Aim 1): Table 2 and S1 Table describe families' financial and mental health 2 2 1 experiences in June 2020, and Table 3 presents the adjusted mean differences (MDs) between NSW and Victoria 2 2 2 ('Vic June' column). Twenty-nine percent of NSW caregivers reported job or income loss due to the COVID-19 2 2 3 pandemic, compared with 24% of Victorians (MD=5%; 95%CI: 0.1 to 10%). Tables 2-3 show that financial and 2 2 4 mental health experiences were otherwise similar between states in June. Fifteen percent of families reported low 2 2 5 income, and 30% reported material deprivation (unable to pay for essential items including mortgage or rent; 2 2 6 electricity, gas, water bills; food; healthcare; prescription medicines; home or car insurance; mobile phone bills; 2 2 7 and internet). Caregivers averaged one missed payment out of the eight essential items. One in five caregivers 2 2 8 reported poor mental health according to the K6; half said the pandemic negatively impacted their mental health; 2 2 9 and one-third said their child's mental health was negatively impacted. Thirteen percent of caregivers and 10% of 2 3 0 children were struggling or not coping. 2 3 1 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 1 1 .

It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 20, 2021. The impact of ongoing lockdown (Aim 2): Fig. 1 and Table 3 present the fully-adjusted Difference-in-2 4 8

Difference linear regression analyses testing the impact of ongoing lockdown on families' finances and mental 2 4 9

health. Over the June-September 2020 period, relative to their NSW counterparts, Victorians reported an 11% 2 5 0 increase in job or income loss (95% CI: 3 to 18%). There was no evidence that ongoing lockdown was related to a 2 5 1 change in material deprivation or the proportion of households reporting low income. Ongoing lockdown was 2 5 2 associated with a 0.83 point increase (95% CI: -0.08 to 1.74) in the K6 total score and a 6% increase (95% CI: 0.3 2 5 3 to 12%) in the binary K6 measure of poor mental health. The negative impact of the COVID-19 pandemic on 2 5 4 mental health was also exacerbated by ongoing lockdown, affecting 14% (95% CI: 6 to 23%) more Victorian 2 5 5 caregivers and 12% (95% CI: 4 to 20%) more children relative to their NSW counterparts. Table 3 present the adjusted Difference-in-Difference analyses for the subgroups of caregiver gender 2 5 9 (female/male); child age (0-4 (pre-school), 5-12 (elementary school), 13-17 years (secondary/high-school)); and 2 6 0 location (metropolitan/regional-rural). The evidence for a negative impact of ongoing lockdown on job/income 2 6 1 loss and caregiver mental health was strongest for female caregivers, caregivers of elementary-school aged 2 6 2 children, and caregivers living in metropolitan areas. Notably, Table 3 ('NSW September' column) shows that the 2 6 3 differences between initial and ongoing lockdowns in measures of job/income loss and K6 mental health were in 2 6 4 part due to worsening outcomes for Victorians as well as improving outcomes for NSW families. The proportion 2 6 5 of NSW caregivers reporting job/income loss decreased by 6% (95% CI: 0.2-11%) from June to September. The 2 6 6 same reduction was evident in the poor mental health (K6) of female caregivers (mean difference=6%, 95% CI: 2 6 7 0.3 to 12%). There was evidence that ongoing lockdown exacerbated the negative impact of the pandemic on 2 6 8 children's mental health, for all subgroups except for families living in regional/rural areas. There was no 2 6 9 evidence of a relationship between ongoing lockdown and whether the caregiver or their children were coping or 2 7 0 not, for the cohort overall or by subgroups. 2 7 1 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 20, 2021. ; 1 4 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 20, 2021. Reference group is NSW in June. All differences are mean difference (MD); note, for binary outcomes this is a proportion between 0-1.

Estimated Difference-in-difference interaction adjusted for: child and caregiver age and gender (male/female; noting no other genders were reported), 2 7 6 number of children, child with additional health needs (versus not); one-caregiver family (versus not); owns a Health Care Card (versus not); caregiver 2 7 7 education (<Year 10/Year 10/Year 12/trade or apprenticeship/certificate or diploma/undergraduate/postgraduate); Aboriginal and Torres Strait Islander 2 7 8 (versus not); caregiver born outside of Australia (versus in Australia); Home language other than English (versus speaks English at home); lives regionally 2 7 9

or rurally (versus metropolitan); and SEIFA quintile. Caregiver mental health (total K6 score) and effects of family cluster were accounted for in all child 2 8 0 models using robust estimation. 2 8 1 *p<0.05, ^p<0.1 2 8 2 a Job loss by one or two adults, or reduction in income, due to COVID-19. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 20, 2021. d K6 dichotomized into a binary "poor mental health" (total score 19 or more) versus optimal (total score 6-18). 2 8 7 e Versus "coping/thriving" categories. due to the stress and disruption of home-schooling in July-September. More supervision is required for children in 3 2 2 elementary school than high school, and balancing home schooling with usual paid or unpaid work was a 3 2 3 substantial challenge for families.(42) Finally, metropolitan Victorians experienced stricter and ongoing lockdown 3 2 4 than their regional and rural counterparts. This explains the findings of increased job/income loss and greater 3 2 5 negative impacts on caregiver and child mental health for this subgroup, relative to their NSW counterparts. 19 Unmasked Study which surveyed Australian families with young children (aged 1-5 years) also found that 3 3 0 Victorian caregivers exposed to ongoing lockdown reported increasing mental health symptoms for themselves 3 3 1 and their children.(44) Interestingly, in the Unmasked study, young children living outside of Victoria (exposed to 3 3 2 only the initial lockdown) were still experiencing higher than average levels of anxiety symptoms by the time of 3 3 3

Victoria's ongoing lockdown. We found a similar pattern in our study. For the NSW families exposed to initial 3 3 4 lockdown only, caregivers reported improved mental health for themselves by September but not for their 3 3 5 children. 3 3 6 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) A  n  y  d  e  p  r  i  v  a  t  i  o  n  (  u  n  a  b  l  e  t  o  p  a  y  f  o  r  e  s  s  e  n  t  i  a  l  i  t  e  m . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 20, 2021. ;

Exploring the Impact of COVID-19 on 4

Mental Health Outcomes in Children and Adolescents: A Systematic Review

This study investigated the impacts of COVID-19 lockdown on family finances and mental health in the context 2 9 0 of Australia's minimal disease burden. In June 2020, after an initial national lockdown from March-May, a 2 9 1 quarter of Australian caregivers of children aged 0-17 years reported job or income loss due to the pandemic. One 2 9 2 in three reported material deprivation (being unable to afford essential items such as housing, food, amenities, or 2 9 3 healthcare). One in five caregivers reported poor mental health; half said that the first three months of the 2 9 4 pandemic had negatively impacted their mental health; and a third reported the same negative impact for their 2 9 5 children. By September 2020, Victoria's ongoing lockdown from July onwards was associated with increased job 2 9 6 and income loss, and negative mental health impacts for caregivers and children, compared with neighboring 2 9 7NSW. There was no evidence that ongoing lockdown affected families' experiences of material deprivation or 2 9 8 income poverty. As anticipated, the evidence for the negative impacts of lockdown were strongest for female 2 9 9 caregivers and families living in metropolitan Victoria. While the negative impacts of ongoing lockdown affected 3 0 0 families across children's ages, they were most pronounced for families with children of elementary school 3 0 1 compared with preschool or high school ages. 3 0 2Although the RCH Poll does not have pre-pandemic data for comparison, the financial and mental health impacts 24% and 27% respectively, using the same questions as our study. The third of families reporting material 3 0 8 deprivation is equivalent to pre-pandemic data from the Longitudinal Study of Australian Children.(35) In our 3 0 9 study, one in five families reported poor mental health in June 2020 according to the K6. This aligns with the 3 1 0Australian 'Pulse of the Nation' survey in which 24% of parents reported mental distress across the first months 3 1 1 of lockdown. Despite the global differences between countries' infection rates and governments' approaches to 3 1 2 This study has several strengths. It builds on existing COVID-19 research to show the specific impacts of ongoing 3 3 7 lockdown on families in the relative absence of the virus, i.e., isolating the toll that widespread mortality has on 3 3 8 mental health. The RCH Poll is the only nationally-representative survey of Australian families with children to 3 3 9 assess financial and mental health impacts of COVID-19. It comprised large cross-sectional and nationally 3 4 0 representative surveys, which employed strong methodology (surveys piloted and included the validated K6); 3 4 1 collected data on caregiver and child mental health; surveyed female and male caregivers; and achieved good 3 4 2 response rates. In other Polls, indicators (frequency/prevalence) across a range of topics are almost universally 3 4 3 consistent with more traditionally-obtained estimates, providing support for the sample selection and survey 3 4 4 administration methods. The Difference-in-Difference modeling is well-established method of analyzing policy 3 4 5 change (in our case, lockdown law differences). The natural experiment allowed us to gauge causal relationships 3 4 6 by simulating a counterfactual; that is, what would have happened in the absence of a second lockdown.

There were also limitations. The Poll only samples adults who are 18 years and older, so younger caregivers are 3 4 8 missing. As a subgroup who experience more adversity, this means results may be conservative. Measures were 3 4 9 Children Report 2: Impact of the second wave in Australia on the mental health of young children and parents. 4 9 7

Brisbane. Queensland Centre for Perinatal and Infant Mental Health, Children's Health Queensland Hospital and 4 9 8Health Service. 2021. 4