It’s gone, it’s back: A prospective study on the COVID-19 pandemic-related shortages and mental health of Australian families

ABSTRACT

Our aim was to explore the association between COVID-19 pandemic-related product shortages and symptoms of stress, anxiety, and depression in Australian families, concurrently and longitudinally, while controlling for demographic, health, and psychological characteristics. This prospective study used two waves of data (baseline, Time 0 = April 2020; Time 1 = May 2020) from a longitudinal cohort study of Australian parents of a child aged 0‐18 years. Parents were surveyed at baseline about whether they had experienced product shortages related to COVID-19. DASS21 was used to measure symptoms of depression, anxiety, and stress at both waves. The sample included 2,110 participants (N = 1,701, 80.6% mothers). About 68.6% of the respondents reported being impacted by one or more shortages. Product shortages correlated significantly with higher combined and individual scores for anxiety, depression, and stress (r = 0.007 to 0.18, all p < 0.001) at baseline. At Time 1, parental emotion regulation explained 4.0% of the variance (p < .001). Our findings suggest a role for improving parental emotion regulation in coping with stressors, such as shortages and lockdowns.

KEYWORDS:

• COVID-19
• families
• longitudinal
• mental health
• parents
• product shortages

Introduction

Product shortages were a signature of the beginning of the COVID-19 pandemic around the world. Scientific papers have reported shortages of ventilators (Iyengar et al., 2020), blood for transfusion (Gniadek et al., 2020), medication (Ann Pulk et al., 2020), acute beds (Vincent & Creteur, 2020), face masks (Rowan & Laffey, 2020) and shields (Gomes et al., 2020) and hand sanitiser (Berardi et al., 2020). However, other product shortages, such as toilet paper, food and other daily-use products, reported by daily newspapers (Davis et al., 2020; Wright, 2020), have not been well documented in academic papers.

An Australian study (n = 600) (O’connor et al., 2020) showed that 17% of shoppers admitted to panic buying in April 2020 (during country-wide lockdowns) versus 6% in June 2020 (after country-wide lockdowns). Australians have experienced one of the highest rates of COVID-19 related panic buying worldwide (Norman, 2020). The peak of this behaviour happened when the Australian federal government tightened social distancing rules and extended the lockdown in late March 2020.

Despite the prevalence of panic-buying, psychological responses to panic buying and shortages in daily-use products have received little attention. What research there is has instead sought to identify personality and demographic traits predictive of stockpiling behaviour (Garbe et al., 2020). An Australian study (Anglim & Horwood, 2021) showed that, in terms of personality traits, those less agreeable, more anxious, and less able to cope with uncertainty were more likely to stockpile products. The study suggested that stockpiling of products may be used as a coping mechanism to reduce anxiety about uncertainty. Indeed, humans do not cope well with perceived threats and unpredictability, with intolerance of uncertainty a common feature of many mental disorders (Rosser, 2019). In a broader context, the link between disasters and mental health has been well documented in the literature (Makwana, 2019). However, even within this extensively researched field, very little attention has been given to the impact of product/resource shortages, and anxiety, depression or stress, except for studies showing that disasters do occur more commonly in socioeconomically disadvantaged countries with a pre-existing problem of resource unavailability (Math et al., 2015). Therefore, the potential relationship between shortages themselves and the mental health of parents has not been examined to date, particularly in the context of the COVID-19 pandemic. Furthermore, given this lack of relevant studies, in the present project, we included independent variables identified as pertinent to disasters and mental health (e.g. resilience) rather than specifically to product shortages and mental health at the time of disasters. We have previously provided an overview of these potential predictors elsewhere (Mikocka-Walus, Stokes, et al., 2021).

The present study

Families with young children have been particularly affected by the COVID-19 pandemic (Westrupp et al., 2021) and are likely to be significantly affected by product shortages. Firstly, many necessary child-related products, such as nappies, baby wipes, and medication, were affected by panic-buying related shortages, and it would have been difficult to find appropriate substitutes (Grose, 2020). Second, most parents in Australia juggle paid work with raising their family and may have struggled to shop around to locate required products (Evans et al., 2020; Living in Australia: Household, Income and Labour Dynamics in Australia (HILDA), 2022). Third, anxiety, stress, and distress are known to ‘crossover’ to affect other family members in a household, including children. There is considerable evidence showing that parental stress and anxiety have long-term adverse effects on children’s mental health and developmental outcomes (A. Morris et al., 2017; A. S. Morris et al., 2010; Berking & Wupperman, 2012; Jeong et al., 2021; Sanders & Mazzucchelli, 2013; Trentacosta & Shaw, 2009; World Health Organization, 2018). Therefore, it is critical that we understand the full impact of product shortages on Australian families to best identify those families most in need of support and to prevent the possible impact of product shortages on mental health during future pandemics/disasters. The present study aims to explore the association between COVID-19 pandemic-related product shortages and symptoms of stress, anxiety and depression in Australian families, concurrently and longitudinally over a period of 2–4 weeks when the shortages first appeared (April/May 2020) while controlling for demographic, health and psychological characteristics.

Material and methods

Design

This prospective study is part of an ongoing longitudinal cohort study of Australian parents of a child aged 0 - 18 years (Westrupp et al., 2020). The study was conducted via online surveys during the COVID-19 period.

Selection criteria

Participants were eligible to participate in the current study if they were currently a resident in Australia and spoke English, were 18 years or over, and were the current parent of a child aged 0-18 years. Survey information and advertisements were written in English, so it was expected that people with adequate English fluency completed the survey.

Recruitment

Families were recruited via social media advertisements and paid online recruitment platforms, such as Facebook, Reddit, and Prolific. Both paid and unpaid recruitment strategies were used. One parent per family completed the survey.

Procedure

The advertisements included a web link that directed participants to the Qualtrics survey. The landing page for the survey contained a brief description of the project. Participants were then asked two eligibility questions. Eligible participants were then presented with a Plain Language Statement and Online Consent form. A follow-up survey was later administered, with reminders sent as necessary.

Measures

The study used a mixture of validated measures, researcher-developed questions (e.g. demographics) and items/subscales adapted from pre-existing measures (e.g. Longitudinal Study of Australian Children (LSAC)). Overall, 10 main measures/adapted items were used plus several questions on demographics, health characteristics, shortages and COVID-19.

Dependent variables

Symptoms of stress, anxiety and depression were measured by the Depression and Anxiety Scale (DASS-21) (Lovibond & Lovibond, 1995). The DASS total score is the sum of the three subscales: Depression, Stress and Anxiety. Each subscale has seven items rated on a 4-point scale from 0 ‘did not apply to me at all’ to 3 ‘applied to me very much, or most of the time’, thus the DASS total score ranges from 0 to 63. DASS-21 was collected at Time 0 (April 2020) and Time 1 (May 2020).

Independent variable

Product shortages were measured by the question: Have you been impacted by any shortages related to COVID-19? Participants were asked whether they had experienced (1) toilet paper shortages, (2) food shortages, (3) medicine shortages, and (4) other shortages (please list). Shortages were measured at Time 0 and were coded as binary (0 = no impact, 1 = impact).

Other variables

All collected at Time 0.

Demographics: age and gender; age of oldest child; number of children; parental education; income; resided in a disadvantaged area; resided in Victoria (the state with the heaviest lockdown).

COVID-19 factors: the following items were adapted from the CoRonavIruS Health Impact Survey (CRISIS) V0.1. (Merikangas et al., 2020) – Participant or family members affected by COVID-19, financial problems or housing and food insecurity, feelings and attitudes about COVID-19, appraisals of COVID-19 as a serious health risk, and whether they are likely to catch COVID-19 (all rated on a 7-point scale from ‘strongly disagree’ to ‘strongly agree’).

Health characteristics: presence of a chronic physical condition and presence of a child’s diagnosis (e.g. ADHD; Autism, Asperger’s).

Positive affect in the past 4 weeks was measured using the Positive Affect scale of the PANAS-Short Form (Thompson, 2007), which is a 5-item, 5-point scale ranging from ‘very slightly or not at all’ to ‘extremely’ (α = 0.80).

Emotion regulation was measured by the Difficulties in Emotion Regulation Scale-16 (DERS) (Bjureberg et al., 2016). This is a 16-item, 5-point scale ranging from ‘almost never’ to ‘almost always’ (α = 0.95).

Introversion/Extraversion was measured by a 1-item, 7-point scale ranging from ‘introvert’ to ‘extrovert’ developed by the authors.

Resilience was measured by the Brief Resilience Scale (BRS) (Smith et al., 2008), a 6-item, 5-point scale ranging from ‘strongly disagree’ to ‘strongly agree’ (α = 0.88).

Experiences on the Close Relationships Scale were measured using Relationship Structures (ECR-RS) (Fraley et al., 2011). This is a 9-item, 7-point scale ranging from ‘strongly disagree’ to ‘strongly agree’ and divided into two subscales: Attachment anxiety (α = 0.90) and attachment avoidance (α = 0.86).

Stressful life events over the past 12 months were measured by an 8-item, 2-point scale rated as yes/no (Brugha & Cragg, 1990). Items were summed up to form a total score.

Immediate and prior stressors consisted of a count of COVID-19 related risks (0–5) including illness, employment issues, financial issues; a count of prior stressful life events (0–7); and a count of financial deprivations over the last 12-months (0–7).

Social support was measured by 1 item from the LSAC (Soloff et al., 2005), rated on a 4-point scale.

Global child health was measured by the LSAC (Soloff et al., 2005). It is a 1-item recoded to binary (0 = poor/fair/good versus 1 = very good/excellent).

Analysis and data preparation

The missing data were assessed for randomness using Little’s MCAR test and were found to be missing at random. Missing values were addressed through multiple imputation (Mackinnon, 2010). Variables imputed include all dependent and independent variables. We used a fully conditional Markov Chain Monte Carlo method to impute missing data with all included variables serving as predictors, using 100 iterations. We utilised a maximum of 250 draws per case and a maximum of 250 draws per parameter. We ran multiple regressions of the changes in symptoms of stress, anxiety and depression over the period, with scores at Time 0 subtracted from Time 1, giving a positive value if scores increased, and a negative value if they decreased. This was undertaken for the symptoms of stress, anxiety and depression overall (all three subscales summed), and for each of anxiety, depression, and stress measures separately. All linear regression models were conducted with five sets of variables included in the following order: (1) demographic variables, (2) health variables, (3) immediate and prior stressors, (4) psychological variables, and (5) measures of product shortages. Effect sizes (R²) were interpreted as per the recommendations for psychological research: effect-size r < .20 small; >/= .20 medium; >/= .30 large and >/= .40 very large (Funder & Ozer, 2019). Given the number of tests undertaken, it was decided to set the type I error rate to .01 as suggested by a Bonferroni correction for five tests at .05.

Results

Demographic, health, and psychological characteristics

The final sample contained 2,110 participants, of whom 1,701 (80.6%) were mothers (Table 1). Participants’ age ranged from 19 to 69 (M = 38; SD = 7). The average age of the children was 8.66 years (SD = 5.14).

Table 1. Demographic details of sample and mean responses to items are included in the analyses.

Variable	Male	Female	Other			Miss.%
Parent gender	442	1895	9			0.8%
	Mean	SD	Min	Max
Age	38.30	7.07	18	69		0.2%
Child’s age	8.66	5.14	1	19		0.0%
SEIFA disadvantage	1013.68	59.14	719	1139		1.1%
	Yes	No
Residing in Victoria	50.7%	49.0%				0.3%
Did not complete high school	9.4%	90.6%				0.0%
Household income <$52K	13.8%	84.4%				1.8%
Single Parent Family	11.0%	89.0%				0.0%
English not spoken at home …	4.6%	95.4%				0.0%
Child global health poor/fair/good …	75.6%	8.9%				15.5%
	0	1	2	3	4+
Number of children	0	671	1089	431	173	<0.1%
Count of financial deprivations …	1914	233	101	64	53	0.0%
Count of COVID-19 stressors …	1076	664	384	215	26	0.0%
Count of stress events …	834	720	323	132	65	12.3%
	Yes	No
Impacted by toilet paper shortage	48.6%	51.4%
Impacted by food shortage	48.1%	51.9%
Impacted by medicine shortage	22.9%	77.1%
Impacted by other shortage	10.2%	89.8%
Impacted by one or more shortages	68.6%	31.4%
	Mean	SD	Min	Max	<Mean
Introversion	4.42	1.60	1	7	44.9%	8.3%
Positive affect (PANAS)	14.70	3.70	5	25	42.0%	10.8%
DERS total score	31.01	12.27	16	80	55.3%	10.6%
Brief Resilience Scale score	3.46	0.76	1	5	38.3%	11.2%
Attachment Anxiety score	3.30	1.78	1	7	50.0%	13.1%
Attachment Avoidance score	3.32	1.32	1	7	48.3%	13.0%
COVID-related psych. risk factors	9.95	2.98	3	18	39.5%	7.8%
T0 DASS depression	9.83	8.46	0	42	58.9%	9.6%
T1 DASS depression	9.11	8.19	0	42	34.0%	44.74%
T0 DASS anxiety	5.79	6.78	0	42	54.6%	9.56%
T1 DASS anxiety	4.56	5.87	0	42	37.2%	44.78%
T0 DASS stress	15.49	8.87	0	42	49.1%	9.51%
T1 DASS stress	14.60	8.42	0	42	59.6%	44.74%

Baseline (Time 0)

Overall, 68.6% of the respondents reported being impacted by one or more shortages. Approximately half the sample reported being impacted by toilet paper (48.6%) and food (48.1%) shortages. One-fourth of the sample were impacted by medicine shortages (22.9%) and 10% were affected by other product shortages (e.g. hand sanitiser, tissues and nappies).

DASS Total scores (i.e. scores for Anxiety, Depression and Stress subscales combined) were significantly correlated with toilet paper (r = .16), food (r = .18) and medicine (r = .19) product shortages (p < .001) but uncorrelated with other product shortages (r = .02). A similar pattern of correlation between product shortages and the component scales of stress, anxiety and depression was observed, albeit the correlation between stress and toilet paper shortages was slightly lower (r = .11) (Supplementary Tables 1 and 2). While correlations between product shortages at Time 0 and DASS scores at Time 1 were somewhat lower, they were still statistically significant.

Table 2. Results of the analysis of DASS at time 0. The upper part of the table gives results for each level of the model overall. The lower part of the table gives results obtained in the final step for each variable.

Level		R^2	SE	-95%CI	+95%CI	ΔR^2	p
1	Demographic variables	0.054	0.001	0.053	0.056	0.054	<.001
2	Health variables	0.144	0.001	0.143	0.146	0.090	<.001
3	Immediate & prior stressors	0.204	0.001	0.203	0.206	0.060	<.001
4	Psychological variables	0.626	0.001	0.625	0.627	0.422	<.001
5	Effect of shortages	0.628	0.001	0.626	0.629	0.001	.393
Level	Final model pooled results	B	SE	−95%CI	+95%CI	sr	p
	Constant	2.895	2.159	−1.344	7.134		.180
1	Age	−0.062	0.020	−0.102	−0.022	−0.073	.002
1	Child’s age	0.060	0.030	0.002	0.118	0.048	.042
1	Number of children in the family	0.029	0.122	−0.210	0.268	0.005	.810
1	Household income <$52k per year	−0.010	0.301	−0.601	0.581	−0.001	.974
1	SEIFA disadvantage	0.001	0.002	−0.002	0.005	0.017	.445
1	Participant state	−0.053	0.188	−0.422	0.315	−0.006	.776
1	Parent gender	−0.065	0.238	−0.531	0.401	−0.006	.784
1	Highest level of schooling	−0.138	0.334	−0.793	0.516	−0.009	.679
2	Treated for a chronic phys. condition?	0.096	0.204	−0.304	0.497	0.010	.636
2	COVID-related psych. risk factors …	0.265	0.033	0.199	0.330	0.173	<.001
2	Child global health poor/fair/good …	0.746	0.332	0.095	1.398	0.051	.025
2	Does the child have a diagnosis?	0.330	0.216	−0.095	0.754	0.034	.128
3	COVID-related environ. stressors …	0.383	0.093	0.200	0.567	0.088	<.001
3	Stressful life events …	0.036	0.095	−0.150	0.222	0.008	.704
3	Count of financial deprivations …	0.533	0.114	0.310	0.757	0.102	<.001
4	Attachment Avoidance	−0.011	0.090	−0.187	0.165	−0.003	.904
4	Introversion	−0.085	0.061	−0.205	0.035	−0.030	.165
4	Brief Resilience Scale	−0.211	0.153	−0.512	0.090	−0.030	.169
4	Difficulties with Emotional Regulation	0.321	0.010	0.301	0.342	0.567	<.001
4	Positive Affect	−0.308	0.029	−0.364	−0.251	−0.234	<.001
4	Do you get enough help …	−0.843	0.230	−1.295	−0.392	−0.084	<.001
4	Treated for a mental health condition?	0.340	0.207	−0.066	0.746	0.035	.100
4	Attachment Anxiety	0.148	0.067	0.015	0.280	0.049	.029
5	Impacted by toilet paper shortage	0.326	0.210	−0.086	0.737	0.034	.121
5	Impacted by food shortage	0.213	0.205	−0.189	0.616	0.022	.299
5	Impacted by medicine shortage	0.414	0.241	−0.058	0.886	0.037	.086
5	Impacted by other shortage	−0.032	0.310	−0.641	0.576	−0.002	.917

Note: Level indicates the step that variables entered the model.

The effect of demographic variables, health variables, immediate and prior stressors, psychological variables, and measures of shortages was first evaluated against the symptoms of stress, anxiety and depression overall score at Time 0. As shown in Table 2, the full model explained 62.8% of the variance of the symptoms of stress, anxiety and depression at Time 0. Demographic variables explained 5.4% (p < .001), health variables explained a further 9% (p < .001), immediate and prior stressors explained an additional 6% (p < .001), psychological variables explained 42.2% (p < .001), and the effect of shortages did not add significantly to the model but added a further 1%.

The final model results show that six variables had the greatest individual effects on DASS Total. In order of magnitude, these were difficulties with emotional regulation (sr = .567, p < .001), positive affect (sr = −.234, p < .001), COVID-related psychological risk factors (sr = .173, p < .001), financial deprivation prior to the pandemic (sr = .102, p < .001), COVID-19-related environmental stressors, such as illness, employment, financial, or housing issues (sr = .088, p < .001), and whether the respondent believed they had adequate support from their social network (sr = −.084, p < .001).

Change over time (Time 0 to Time 1)

The full model explained 7.8% of the variance of the change in symptoms of stress, anxiety, and depression (Table 3). Demographic variables explained 1.5% (p < .001), health variables explained a further 1.8% (p < .001), immediate and prior stressors explained an additional 0.5% (p < .001), psychological variables explained 4% (p < .001), and the effect of shortages did not add significantly to the model but added a further 0.1%. Therefore, shortages had limited impact on mental health over time.

Table 3. Results for the analysis of change in DASS. The upper part of the table gives results for each level of the model overall. The lower part of the table gives results obtained in the final step for each variable.

Level		R^2	SE	-95%CI	+95%CI	Δ R^2	p
1	Demographic variables	0.015	0.001	0.013	0.017	0.015	<.001
2	Health variables	0.033	0.001	0.031	0.035	0.018	<.001
3	Immediate & prior stressors	0.038	0.001	0.036	0.040	0.005	<.001
4	Psychological variables	0.078	0.002	0.075	0.081	0.040	<.001
5	Effect of shortages	0.078	0.002	0.075	0.082	0.001	.719
Level	Final model pooled results	B	SE	−95%CI	+95%CI	sr	p
	Constant	−3.311	2.381	−7.994	1.373		.165
1	Age	0.032	0.025	−0.018	0.082	0.036	.209
1	Child’s age	−0.005	0.038	−0.079	0.070	−0.003	.906
1	Number of children in the family	0.119	0.164	−0.205	0.444	0.020	.468
1	Household income <$52k per year	0.383	0.418	−0.445	1.211	0.027	.361
1	SEIFA disadvantage	0.002	0.002	−0.002	0.005	0.019	.431
1	Participant state	0.627	0.279	0.075	1.180	0.069	.027
1	Parent gender	0.195	0.348	−0.496	0.885	0.017	.577
1	Highest level of schooling	0.477	0.487	−0.489	1.443	0.030	.330
2	Treated for a chronic phys. condition?	−0.010	0.254	−0.511	0.491	−0.001	.968
2	COVID-related psych. risk factors …	−0.135	0.041	−0.215	−0.054	−0.085	.001
2	Child global health poor/fair/good …	−0.804	0.423	−1.639	0.030	−0.052	.059
2	Does the child have a diagnosis?	0.166	0.265	−0.357	0.689	0.016	.533
3	COVID-related environ. stressors …	0.267	0.133	0.004	0.530	0.058	.046
3	Stressful life events …	0.085	0.134	−0.180	0.349	0.018	.528
3	Count of financial deprivations …	−0.277	0.150	−0.574	0.020	−0.051	.067
4	Attachment avoidance	−0.107	0.109	−0.322	0.108	−0.026	.329
4	Introversion	0.016	0.078	−0.137	0.169	0.005	.838
4	Brief Resilience Scale	0.163	0.197	−0.226	0.553	0.022	.409
4	Difficulties with emotional regulation	−0.070	0.013	−0.097	−0.044	−0.142	<.001
4	Positive affect	0.090	0.036	0.019	0.160	0.067	.013
4	Do you get enough help …	−0.169	0.304	−0.771	0.433	−0.016	.580
4	Treated for a mental health condition?	0.387	0.259	−0.123	0.897	0.038	.137
4	Attachment Anxiety	0.143	0.090	−0.035	0.320	0.045	.115
5	Impacted by toilet paper shortage	−0.039	0.259	−0.549	0.470	−0.004	.879
5	Impacted by food shortage	0.104	0.265	−0.420	0.627	0.010	.697
5	Impacted by medicine shortage	−0.325	0.300	−0.915	0.266	−0.028	.280
5	Impacted by other shortage	0.357	0.415	−0.463	1.177	0.024	.391

Table 3 reveals that only COVID-related psychological risk factors (sr = −.085, p = .001), specifically difficulties with emotional regulation (sr = −.142, p < .001), contributed to the change significantly. Models were also examined for the change in depression, with a total R² = .055, of which the change in R² due to psychological variables was R² = .035. For change in anxiety, the total R² was .073, with change in R² due to health variables being R² = .025 and psychological variables being R² = .030. For changes in stress, the total R² was .045, with the change in R² due to psychological variables being R² = .021. Therefore, product shortages had only a very small effect on mental health at Time 1.

Discussion

In this study COVID-19 related shortages correlated significantly with symptoms of stress, anxiety, and depression at baseline; however, they contributed only a very small proportion of variance to changes in these symptoms from baseline to 2–4 weeks later. Instead, the strongest contributor to the combined measure of stress, anxiety and depression for Australian families were parents’ difficulties with emotional regulation.

The results may be explained by the acute nature of the shortages. Had the shortages been greater in their duration and severity, their impact on mental health over time might have been more pronounced. In addition, the Time 0 assessment occurred during a lockdown, while the Time 1 assessment was in mid-May, when restrictions were easing, with the initial panic subsiding and the outlook for Australia optimistic at that time. It may suggest that parents still exhibiting mental health symptoms at Time 1 may have been those with more enduring emotion regulation difficulties, who struggled to cope after the crisis, while the majority were no longer distressed by the shortages. Furthermore, the present sample of parents was of relatively good health, with just 30% reporting a chronic health condition. In a study with a chronically ill sample (n = 831), a significant relationship between product shortages and psychological distress, which was associated with COVID-19 prevalence and fear of COVID-19, was found (Mikocka-Walus, Skvarc, et al., 2021). Since the chronically ill sample relied heavily on products, such as medications and specialty foods even short-term shortages affected their lives significantly.

The psychological characteristics of the participants clearly overshadowed the temporary inconvenience of the shortages. This may be a byproduct of an overlap between psychological variables and the DASS scores, for example, scores of resilience tend to be lower with higher anxiety/depression scores. However, it may also indicate that to prevent deterioration of mental health at the time of major stresses, interventions supporting emotion regulation in parents should be implemented. Parents’ capacity to manage their emotions during stress is important for parents’ own mental health as well as their children’s ongoing functioning (S. Havighurst & Kehoe, 2017). Established programs exist for teaching emotion regulation skills in parents (e.g. Tuning into Kids) (S. S. Havighurst & Harley, 2007/2010), which may be adapted to non-normative stressors like COVID-19. Given research indicating that mindfulness practice leads to healthy emotion regulation (Roemer et al., 2015), parents may be supported to try mindfulness during crises. Even brief mindfulness exercises have been shown to improve psychological outcomes (Howarth et al., 2019). Public health efforts to raise awareness about evidence-based strategies for supporting emotion regulation (specifically, brief skills that parents can realistically adopt) are likely to reap rewards.

Limitations

While shortages in the current study were common, our measure is unable to differentiate between low-level shortages causing little disruption and urgent shortages. Further, we focused on three specific types of product shortages and did not explore shortages of ventilators, face masks and acute beds. In fact, our free-text analysis showed that many participants were affected by the shortages of hand sanitiser, gloves, baby wipes, and other everyday use products. Further, a reverse direction of the relationship between product shortages and mental health could exist, namely that feeling more psychological distress is associated with more attention to product shortages. However, we could not test this hypothesis as we only asked about product shortages once.

Conclusion

While the experience of product shortages did contribute to stress, anxiety, and depression at baseline, little effect of the shortages on mental health over time was identified. A focus on emotional regulation rather than shortages might be useful to improve stress, anxiety, and depression in parents during times of pandemics.

Authors’ contributions

AMW, MS and EW conceived the study. EW collected data. MS and EW conducted analysis. AMW, MS and EW interpreted data. AMW drafted the manuscript. All authors contributed to and approved the final draft of the manuscript. AMW is accepting full responsibility for the conduct of the study. She has access to the data and has control of the decision to publish.

Consent to participate

Participants provided informed written consent.

Data accessibility statement

The data that support the findings of this study are available on request from the senior author. The data are not publicly available due to restrictions.

Ethics approval

The study received approval from the Deakin University Human Ethics Advisory Group.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.