Assessing explanatory variables of perceived stress to disaster: implications for risk research

Abstract

The COVID-19 pandemic continues to increase social, economic, and psychological risks, including increased perceived stress – or the degree to which a person perceives a stressor and their ability to cope with it. The current study uses novel data collected during the COVID-19 pandemic to assess the role of a range of demographic and disaster-related experiential variables on perceived stress (n = 744). Hierarchical linear regression indicates that women experience greater perceived stress than men and as age and educational attainment increase, perceived stress decreases. Respondents experiencing rent/mortgage stress, job loss due to the COVID-19 pandemic, anticipated reliance on others, and worry about ongoing impacts of COVID-19 on their physical health, the economy, and personal relationships also experience increased perceived stress. Results provide empirical evidence of risks stemming from the multiple concerns (i.e., financial, psychological, and physical health) of U.S. residents regarding the COVID-19 disaster. Findings indicate the need for policy and legislative actions, such as the U.S.-wide eviction moratorium, to support individuals suffering from multiple impacts from the pandemic and to reduce perceived stress and its attendant risks including increased incidents of posttraumatic stress and depression.

Keywords:

• Disaster risk
• perceived stress
• economic stress
• mental health

Introduction

The COVID-19 pandemic continues to be a profound and disrupting force and, as such, provides lessons to be learned to better understand the often-compounding risks of disaster for individuals. To this end, the current study identifies socio-demographic and experiential factors that influence perceived stress to the COVID-19 disaster. Building on prior risk research into COVID-19’s impacts (e.g., Brown, Coventry, and Pepper 2021; Bryce et al. 2020; Schneider et al. 2021; M-Amen et al. 2021) and following the call for new research efforts into its psychological impacts (Obschonka, Gewirtz, and Zhu 2021), we aim to identify risks from the pandemic by investigating relationships between socio-demographic characteristics and perceived stress to help inform social systems that are better able to respond to the next global pandemic.

Studying disasters – defined by the UN as a severe disruption of the functioning of a community due to hazardous events (UN Office for Disaster Risk Reduction 2022) — provides important insights needed to develop and inform better disaster preparation and response (Osofsky and Osofsky 2021). Experiences of ongoing disasters likely increase adverse psychological (e.g., increased PTSD), physical (e.g., higher musculoskeletal pain) and social (e.g., increased community fracturing) risks (Osofsky and Osofsky 2021; Cannon et al. 2021).

Psychological consequences of disaster

Research suggests that psychological impacts from disaster, such as post-traumatic stress disorder (PTSD), depression and perceived stress, increase post-disaster (Osofsky et al. 2015; Ferreira, Buttell, and Cannon 2020; Schwartz et al. 2015). Perceived stress is understood as the degree to which a person perceives a stressor and how well they feel they can adapt to it (Caplan 1981). Increased perceived stress likely increases the risk of depression and posttraumatic stress (Rodríguez et al. 2010). Such psychological impacts have been linked with an increased risk of adverse physical health effects, such as increased musculoskeletal pain and cardio-respiratory complaints (e.g., Spitzer et al. 2009; Pacella, Hruska, and Delahanty 2013).

Recent research into the COVID-19 pandemic has investigated changes in perceived stress in multiple populations including, college-age students, finding that increased anxiety has been linked with increased perceived stress (Aslan, Ochnik, and Çınar 2020), and increased perceived stress linked to increased alcohol use (Charles et al. 2021); parents, finding that they exhibited greater parental stress and high depressive symptoms during the COVID-19 pandemic (Brown et al. 2020); and nursing students, finding they exhibited moderate perceived stress (Aslan and Pekince 2021). In sum, recent research suggests mental health and substance use risks from stressors brought on by the COVID-19 pandemic.

Psychological impacts from disaster vary by Socio-Demographic characteristics

Increases of adverse mental health outcomes vary by socio-demographic characteristics (Vezzali et al. 2018; Leon et al. 2007). Some disaster research indicates that women, older individuals, and those with less education experience greater psychological problems post disaster (Barbosa-Leiker et al. 2013; Xu and Wu 2011). For example, Leon et al. (2007) found that after disaster exposure, women were more likely to indicate greater levels of perceived stress when compared with men, and that increased perceived stress was positively associated with increased PTSD symptomology. Patel et al. (2018), using the well-validated Perceived Stress Scale (Cohen 1994), found that ethnicity, previous experience with depression and anxiety were each positively associated with increased perceived stress after disaster. Similarly, Schwartz et al. (2016), in their study of Hurricane Sandy survivors, found that a third of the study sample exhibited high stress, and that mental health issues and Hispanic ethnicity were positively associated with increased perceived stress.

Such relationships between mental health impacts and socio-demographic characteristics appear to hold true for the COVID-19 pandemic, with recent research suggesting a dramatic increase in reported experiences of adverse mental health conditions, including anxiety disorder and depression related symptoms (Czeisler et al. 2020; Riehm et al. 2021; Twenge and Joiner 2020). Such mental health impacts have been found to disproportionately impact women (Riehm et al. 2021; Torales et al. 2020), and specific subpopulations, in particular, Hispanic and Black communities (Czeisler et al. 2020).

Social and economic concerns stemming from experiences of disaster

In addition to socio-demographic factors that may influence perceived stress during and after disaster, the extant research suggests that there are disaster related worries and experiences that also increase the risk of adverse psychological issues (Esterwood and Saeed 2020; Shigemura et al. 2020). For example, research has indicated disasters can fracture communities and the support they provide their members (Zhou et al. 2018; Lai et al. 2015), leading to increased rates of posttraumatic stress and depression (Finelli and Zeanah 2019; Kronenberg et al. 2010). Further compounding such risks, disasters have been shown to strain personal relationships producing relationship violence risk factors and eroding key social and emotional supports (Picou, Marshall, and Gill 2004; Schokkenbroek et al. 2021). Disasters, including the COVID-19 pandemic, have generated their own economic concerns. For instance, a recent study of German residents during the pandemic found that although virus-related concerns decreased over time, economic worries arising from the pandemic remained (Rosman et al. 2021).

Data and methods

The purpose of this study was to investigate explanatory, or independent, variables of perceived stress during the COVID-19 pandemic among a sample of respondents living in the U.S. to identify psychological and social risks to disaster. The study sample includes 744 adult participants who completed an online survey during the COVD-19 pandemic. Data were collected over a 12-month period, starting in the first week of April 2020 and closing the last week of March 2021. A convenience sampling strategy was used to recruit a ‘river’ sample, an important type of sample for exploratory studies with sub-populations (Lehdonvirta et al. 2021). The study was approved by X University Social/Behavioral Institutional Review Board (IRB; reference: 2020-556). Informed consent was conducted through the survey link as approved by the IRB. Distribution of the online survey was through one of the research team members’ personal social media accounts (e.g., Facebook, Instagram and LinkedIn) and advertised on the X School of Social Work social media outlets and website for a period of 12 months. Criteria for inclusion required participants be 18 years or older. This study advances methodological developments in risk research by considering psychological and social risks to disaster together. Specifically, the following measurements were included a) perceived stress measured by the Perceived Stress Scale, b) COVID-19 related stressors, and c) personal and household demographics. The Qualtrics survey took an estimated 10 minutes to complete.

Measures

Outcome variable

Perceived stress scale (PSS)

The study includes a recoded variable for total score on the Perceived Stress Scale (PSS). The PSS has documented correlations with a variety of health-related measures including stress, self-reported health, health services, health behavior, smoking status and health seeking behaviors (Cohen 1994). Scores on the PSS range from 0 to 40, with higher values on the PSS indicating greater perceived stress. The PSS categorizes perceived stress as low, with scores ranging from 0 to 13, moderate (14 to 26), and high (27 to 40). The PSS has excellent psychometric properties (Lee 2012; Roberti, Harrington, and Storch 2006) and disaster risk research using the PSS has found ethnicity, previous disaster experiences, and mental health problems, such as depression, likely increased perceived stress (Jiang et al. 2020; Schwartz et al. 2015; Leon et al. 2007). The PSS showed excellent internal consistency with a Cronbach’s α of 0.88 (n = 728).

Independent variables

Demographic variables

Similar to other risk studies investigating social vulnerability (e.g., Cannon, Twigg, and Rowell 2003), we included the following demographic variables and response categories: (1) age; (2) gender (1 = men, 2 = women); (3) race (1 = white, 2 = not white); (4) relationship status (1 = not in a relationship, 2 = in a relationship); (5) employment status (0 = not employed, 1 = employed); (6) education (1 = less than high school, 2 = high school, 3 = some college, 4 = associate degree, 5 = bachelor’s degree, 6 = graduate degree); and (7) residential status (0 = does not own house, 1 = own house).

COVID-19 experiential variables

The purpose of the following set of independent variables is to determine which experiences associated with COVID-19 influenced perceived stress. The following independent variables and response categories were included: (1) rent/mortgage stress (1= stressed worried, 2= not stressed worried); (2) nutritional stress (1= stressed worried, 2 = not stressed worried); (3) ‘Has the COVID-19/Coronavirus led to you losing your job?’ (1 = yes; 0 = no); (4); ‘Has the COVID-19/Coronavirus led to you losing income?’ (1 = yes; 0 = no); (5) ‘Do you think you will need help and cooperation from others (e.g., family, friends, or neighbors) to recover from the impact of COVID-19/Coronavirus?’ (1 = very little help, 2 = little help, 3 = neither very little help nor very much help, 4 = much help, 5 = very much help); (6) ‘Do you think you will need help and cooperation from others (e.g., government or non-governmental organizations) to recover from the impact of COVID-19/Coronavirus?’ (1 = very little help, 2 = little help, 3 = neither very little help nor very much help, 4 = much help, 5 = very much help); (7) ‘How worried are you now about any ongoing impacts of the COVID-19/Coronavirus pandemic on the physical health of you or any member of your immediate family?’ (1= not at all worried, 2= a little worried, 3 = moderately worried, 4 = very worried); (8) ‘How worried are you now about any ongoing impacts of the COVID-19/Coronavirus on the economy in your community (1= not at all worried, 2= a little worried, 3 = moderately worried, 4 = very worried)?’; and (9) ‘How worried are you now about any ongoing impacts of the COVID-19/Coronavirus Pandemic on the relationships with family and friends for you or any member of your immediate family?’ (1= not at all worried, 2= a little worried, 3 = moderately worried, 4 = very worried).

Analytic strategy

Multiple hierarchical linear regression with two models were performed to identify how much of the variance in perceived stress, measured by the PSS, among survey respondents was explained by demographic variables and experiences related to COVID-19.$$Y=b_1{X}_1+b_2{X}_2+b_3{X}_3+\dots +a$$

Model 1 included demographic variables (i.e., age; sex; race; relationship; employment; education; and residential status) to investigate whether they related to an increased level of self-reported perceived stress for the study sample. Model 2 added to the set of demographic variables, a set of COVID-19 experiential variables (i.e., rent/mortgage stress, nutritional stress, job loss, lost income, assistance from family/friends and neighbors, assistance from government/non-governmental organizations, worry about ongoing impacts of COVID-19 on physical health, worry about ongoing impacts of COVID-19 on the economy, worry about ongoing impacts of COVID-19 on relationships) to investigate key linkages to an increased level of perceived stress for the sample. SPSS 27 was utilized to conduct the final data analysis.

Results

Descriptive statistics of demographic characteristics

The sample of 744 participants had a mean age of 47.68 years (SD = 15.35). Of the sample, 80.9% identified as women (n = 602), 18.7% as men (n = 139) and 0.3% (n = 3) chose the other category. The majority of the sample identified as White (83.1%; n = 602) with 10.3% identifying as Black (n = 77), 1.9% as mixed or bi-racial (n = 14), 1.6% as other (n = 12), 1.3% as Middle Eastern (n = 10), 1.1% as Asian (n = 8), and 0.7% as Native American or Alaska Native (n = 5). Regarding relationship status, most participants were in a relationship (71.2%; n = 530). The majority of the sample were employed (73.2%; n = 544). With respect to educational attainment, only 1.7% had less than a high school diploma (n = 13), followed by 4.4% with a high school diploma or GED (n = 33), 9.3% with some college (n = 69), 3.9% with an associate degree (n = 29), 19.1% with a bachelor’s degree (n = 142), and 61.6% with a graduate degree (n = 458). More than two-thirds of the sample owned their home with 67.9% reporting homeownership (n = 505).

Regarding the outcome variable, the PSS, respondents had a mean score of 20.42 (SD = 6.63) for the 10-item scale. The moderate perceived stress group had the highest representation with 66.7% (n = 496), followed by the high perceived stress group with 17.7% (n = 132), and finally the low perceived stress group was 15.6% of the sample (n = 116). Detailed descriptions of demographic variables are presented in Table 1.

Table 1. Demographic characteristics of the sample.

	Participants (n = 744)
		Range
Characteristic	Mean/%	n	SD
Age (in years)	47.68	744	15.35
		18-91
Gender
Male	18.7	139
Female	80.9	602
Other	.3	3
Race
White	83.1	618
Black	10.3	77
Native American/Alaska Native	.7	5
Asian	1.1	8
Middle Eastern	1.3	10
Mixed or Bi-Racial	1.9	14
Other	1.6	12
Relationship
In a relationship	71.2	530
Single	28.8	214
Employment
Employed	73.2	544
Unemployed	27.8	200
Education
Less than 12 years/No HS Diploma	1.7	13
HS Diploma/GED	4.4	33
Some College	9.3	69
Associate Degree	3.9	29
Bachelor Degree	19.1	142
Graduate Degree	61.6	458
Residential
Own House	67.9	505
Rent	32.1	239
Perceived Stress Scale Mean Score	20.42		6.63
Perceived Stress Category
Low Perceived Stress (0-13)	15.6	116
Moderate Perceived Stress (14-26)	66.7	496
High Perceived Stress (27-40)	17.7	132

Descriptive statistics of COVID-19 experiential variables

Questions regarding COVID-19 experiences were also included in the survey to better understand the role such experiences play in measuring perceived stress. Rent/mortgage stress was reported by nearly one third of the sample (31.5%; n = 235) and nutritional stress was observed among a quarter of the sample (22.7%; n = 269). COVID-19 resulted in job loss for 13.3% of the sample (n = 99) and 36.2% (n = 260) of respondents reported a loss of income because of the pandemic. Of the sample, 19.4% (n = 154) reported they would need assistance during the pandemic from family, friends, or neighbors. With respect to governmental and non-governmental support, the need was greater with 26.3% (n = 197) of respondents indicating they would require help from such entities. More than 2/3^rd of respondents were worried about the ongoing impacts of the COVID-19 pandemic on their physical health and that of their families (67.5%; n = 501) and about a fourth of respondents reported being very worried (25.4%; n = 190). Approximately one third of respondents reported being moderately worried about the impacts of COVID-19 on the economy (30.6%; n = 228) and more than half indicated they were very worried (54.1%; n = 402). Respondents also reported they worried about pandemic impacts on their relationships with friends and family with 70.3% of the sample reporting some degree of worry (n = 526). Detailed descriptions of the COVID-19 experiential variables for the sample are presented in Table 2.

Table 2. Descriptive statistics of COVID-19 experiences of the sample.

	Participants (n = 744)
Characteristic	Mean/%	n
Rent/Mortgage Stress
Stressed/Worried	31.5	235
Not Stressed/Worried	69.5	509
Nutritional Stress
Stressed/Worried	22.7	169
Not Stressed/Worried	77.3	575
COVID-19 Lead to Losing Job
Lost Job	13.3	99
Did not lose Job	86.7	645
COVID-19 Lead to Losing Income
Losing Income	36.2	260
Did not Lose Income	63.8	474
Assistance from Family/Friends & Neighbors
Very little help	36.9	275
Little help	23.7	176
Neither very little help nor very much help	20.1	149
Much help	10.9	81
Very much help	8.5	63
Assistance from Government & Non-Governmental Organizations
Very little help	33.1	246
Little help	24.4	181
Neither very little help nor very much help	16.2	120
Much help	13.2	98
Very much help	13.1	99
Worried about ongoing impacts of COVID-19 on physical health of self or immediate family
Not at all worried	7.1	53
A little worried	27.9	207
Moderately worried	39.6	294
Very worried	25.4	190
Worried about ongoing impacts of COVID-19 on the economy in your community
Not at all worried	1.9	14
A little worried	13.4	100
Moderately worried	30.6	228
Very worried	54.1	402
Worried about ongoing impacts of COVID-19 on the relationships with family and friends for you or any member of immediate family
Not at all worried	29.3	218
A little worried	29.4	219
Moderately worried	24.5	182
Very worried	16.8	125

Perceived stress model testing

Bivariate correlations of all variables included in the hierarchical linear regression models are presented in Tables 3 and 4, respectively. The correlation matrices indicate no pattern of multicollinearity among the variables. Model 1: Demographic Variables Explain Perceived Stress

Table 3. Bivariate correlations of all variables included in Model 1.

	1.	2.	3.	4.	5.	6.	7.	8.
1. Total PSS Score	1.00
2. Age	−0.297**	1.00
3. Gender	0.196**	−0.154**	1.00
4. Race	0.25	−0.121**	−0.013	1.00
5. Relationship status	0.146**	−0.369**	0.1**	0.068	1.00
6. Employment	0.033	0.198**	0.053	−0.04	0.01	1.00
7. Education	−0.176**	0.073*	0.063	0.029	−0.142**	−0.244**	1.00
8. Residential status	0.2**	−0.414**	0.043	−0.05**	0.175**	0.008	0.532**	1.00

Table 4. Bivariate correlations of new variables included in Model 2.

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.
1. Total PSS Score	1.00
2. Rent/Mortgage stress	−0.375**	1.00
3. Nutrition stress	−0.336*	0.741**	1.00
4. COVID-19 led to job loss	−0.128**	0.083**	0.054	1.00
5. COVID-19 led to income loss	−0.188**	0.302**	0.217**	0.016	1.00
6. Cooperation from family	0.287**	−0.271**	−0.276**	0.114	−0.188**	1.00
7. Cooperation from government	0.226**	−0.338**	−0.301**	−0.041	−0.16**	0.584**	1.00
8. Worry over ongoing COVID-19 impacts on physical health	0.324**	−0.134**	−0.189**	−0.042	−0.034	0.274**	0.249**	1.00
9. Worry over ongoing impacts on the economy	0.258**	−0.129**	−0.131**	0.025	−0.149	0.161**	0.24**	0.287**	1.00
10. Worry over ongoing impacts on relationships	0.339**	−0.23**	−0.259**	−0.071	−0.071	0.326**	0.275**	0.414**	0.236**	1.00

A summary of the regression coefficients for the model is presented in Table 5. The R² statistic was statistically significant (F (7,724) = 38.077, p < 0.001, R² adjusted =0.131), indicating that 13.1% of the variance in perceived stress was explained by demographic variables. Results show that age (small to medium effect size, f²= .07), gender, (small effect size, f²=.03) and educational attainment (small effect size, f²=.02) were each statistically significant. Surprisingly, as age decreases, there is a corresponding increase in perceived stress. According to the model, women reported higher perceived stress levels than men. Moreover, decreases in educational attainment led to an increase in perceived stress. Age was the most robust explanatory variable in the model (β = .24). A post-hoc power analysis was conducted using the sample of 735 respondents for which there were data, a 7-independent variable equation, and an alpha level of p < 0.05. The post-hoc power analysis showed greater than 0.9 power, indicating there was more than adequate power to conduct this statistical analysis (Rosner 2011).

Table 5. Model 1 of hierarchical linear regression for demographic factors of perceived stress (n = 735; df = 7).

					95% CI	95% CI
	B	β	t	p	Lower Bound	Upper Bound
Age***	−0.104	−0.241	−5.744	.001	−0.139	−0.068
Gender***	2.635	.605	4.358	.001	1.448	3.822
Race	.340	.019	0.539	.59	−0.898	1.578
Relationship	−0.077	−0.006	−0.151	.88	−1.083	0.928
Employment	−0.071	−0.005	−0.137	.891	−1.095	0.952
Education***	−0.754	−0.147	−3.985	.001	−1.126	−0.383
Residential	−0.816	−0.058	−1.41	.159	−1.952	0.32
Constant	24.799		14.41	.001	21.421	28.178

Notes: n = 735; df = 7.

*** p < .001

Model 2: COVID-19 experiences explain perceived stress

The full model of both sets of demographic and COVID-19 experiential variables (Model 2) was statistically significant, R² = 0.345 (F(16, 708) = 23.281, p < 0.001), indicating that 34.5% of the variance in perceived stress was explained by both demographic and COVID-19 experiential variables. A summary of the regression coefficients for this model are presented in Table 6. Statistically significant explanatory variables of perceived stress were, in addition to age, gender, and education (that were discussed above in Model 1), rent/mortgage stress (small to medium effect size, f²=.06), job loss resulting from COVID-19 (small effect size, f²=.01), an anticipated need for greater help and cooperation from family, friends, or neighbors to recover from the impact of COVID-19 pandemic (small effect size, f²=.01) and worry over ongoing impacts of COVID-19 on physical health (small effect size f²=.03), the economy (small effect size, f²=.02), and personal relationships (small effect size, f²=.02).

Table 6. Model 2 of hierarchical linear regression for demographic and COVID-19 factors of perceived stress (n = 727; df = 9).

					95% CI	95% CI
	B	β	t	p	Lower Bound	Upper Bound
Age***	−0.091	−0.212	−5.742	.001	−0.122	−0.06
Gender***	2.167	0.129	4.074	.001	1.123	3.211
Race	0.981	0.056	1.728	0.084	−0.133	2.095
Relationship	−0.007	−0.001	−0.015	.988	−0.895	0.882
Employment	0.24	0.018	0.519	.604	−0.667	1.147
Education*	−0.358	−0.07	−2.054	.04	−0.701	−0.016
Residential	0.319	0.023	0.62	0.535	−0.691	1.329
Rent/Mortgage Stress***	−1.123	-.204	−4.231	0.001	−1.644	−0.602
Nutrition Stress	−0.089	−0.015	−0.317	0.751	−0.643	0.464
COVID-19/Coronavirus lead to you losing your job*	−1.273	−0.071	−2.288	0.022	−2.364	−0.181
COVID-19/Coronavirus lead to you losing income	−0.845	−0.061	−1.883	0.06	−1.726	0.036
Cooperation from others (e.g. family, friends, neighbors)*	0.433	0.09	2.32	0.021	0.067	0.799
Cooperation from others (e.g. government or non-governmental organizations)	−0.18	−0.041	−1.026	0.305	−0.524	0.164
Worried about ongoing impacts of COVID-19 on physical health***	1.111	0.151	4.288	.001	0.602	1.619
Worried about ongoing impacts of COVID-19 on the economy***	1.153	0.281	4.1	.001	0.601	1.705
Worried about ongoing impacts of COVID-19 on relationships***	0.989	0.159	4.501	.001	0.558	1.42
Constant	20.373		8.888	.001	15.872	24.873

* p < .05; ***p < .001.

According to the model, rent/mortgage stress led to an increase in perceived stress. Similarly, job loss due to the pandemic led to an increase in perceived stress. The more respondents thought they would require cooperation from friends, family, or neighbors to recover from the pandemic, the more their perceived stress increased. Greater worry about the impacts of the pandemic on their physical health, the economy, and personal relationships also contributed to increased perceived stress. Rent/mortgage stress was the most robust explanatory variable in the model (β = .204). A post-hoc power analysis was conducted using the sample of 727 respondents for which there were complete data, a 9-independent variable equation, and an alpha level of p < 0.05. The post-hoc power analysis showed greater than 0.90 for the detection of significant relationships, indicating there was more than adequate power (i.e., power * 0.90) to conduct this statistical analysis.

Discussion

Findings from this study are significant in that they advance insights into contributing factors to perceived stress during an ongoing, long-term disaster with implications for risk research. Consistent with extant research into the relationship among gender and disasters, broadly (Xu and Wu 2011; Leon et al. 2007), and COVID-19 specifically (Schokkenbroek et al. 2021), this research similarly demonstrates that women experience greater perceived stress during COVID-19 than men meaning women may be more at-risk for adverse psychological impacts from the disaster. Interestingly, our results indicate that younger respondents showed greater levels of perceived stress, which could be tied to not just the concerns over physical health but also the impacts of the pandemic on the economy and personal relationships with friends and family, consistent with prior research (Rosman et al. 2021). This point is further supported by the second regression model, which showed that immediate impacts of the pandemic, such as rent/mortgage stress and job loss due to the pandemic increased perceived stress for the sample and explained almost 30% of the variance in perceived stress.

Compounding these immediate impacts, were ongoing stressors in the form of worrying about the pandemic’s long-term impacts on physical health, the economy, and personal relationships. These ongoing concerns can pose risks for individuals for years to come as they may translate into ongoing mental health issues as prior research suggests experiences of disasters and their myriad consequences often lead to increased rates of posttraumatic stress and depression (Finelli and Zeanah 2019; Kronenberg et al. 2010) and trigger mental disorder episodes or precipitate first-time episodes (Esterwood and Saeed 2020; Shigemura et al. 2020). This, in turn, may increase the risks of turning to drug and alcohol use to cope (Osofsky et al. 2014).

Decision-makers, particularly in local, state, and federal legislatures and executive branches, should consider these contributors to perceived stress, and the mental and physical health risks of increased perceived stress when confronting ongoing disasters, generally, and the COVID-19 pandemic, specifically. For instance, the U.S. nation-wide rent moratorium has prevented the displacement of millions of people during a public health and economic crisis (Shroyer 2021). However, with the rent moratorium expiring in many places across the country, residents will need policies that help them to keep their housing and alleviate rent/mortgage stress, which was a significant contributor to increased perceived stress in our sample. Similarly, economic planning to secure a quick and long-sustaining recovery particularly as the Omicron variant continues to surge in much of the U.S. is needed to alleviate the ongoing stress related to worries over economic prospects.

Given the established relationship between disasters and negative mental health impacts described above (e.g., PTSD), there are clear implications for targeting vulnerable populations going forward. Risk management practices, including those of providers of mental health treatment, should incorporate questions about perceived stress and coping into their practice with their existing clients. Further, efforts should be made to make mental health services available to a wider audience during the initial stages of a disaster to reduce risks from mental health problems. One of the ways this goal can be advanced is through telemedicine, the use of which has increased dramatically during the COVID-19 pandemic (Schokkenbroek et al. 2021). Efforts should be made to increase the accessibility of this service to poor and chronically underserved communities. One way to do this is to establish publicly funded and accessible sites with tablets or computers in distressed communities where people can come and see a doctor or therapist remotely. This strategy bypasses the digital divide often present in poor communities and is also likely to increase service utilization. This type of crisis-oriented work has a long history in disaster responses and should be modified to incorporate the lessons learned about risks associated with health care access and social isolation during COVID-19.

Limitations

There are two important limitations to this research that are necessary to consider. First, the sample is not representative, which limits the generalizability of the findings. More specifically, women, higher educational attainment, and white as a racial category were overrepresented in the sample due to the convenience sampling strategy utilized in this study. Although useful for understanding contributors to perceived stress during an ongoing and unfolding crisis, future research should endeavor to collect data from representative samples to increase generalizability of results. Recent research does suggest such river samples are important for exploratory studies as they include non-demographic subpopulations (Lehdonvirta et al. 2021). Second, this research uses cross-sectional data which cannot be used to assess causality. Although difficult to attain, future research should consider collecting longitudinal data to identify changes in COVID-19 related experiences and perceived stress over time to shed further light on risks to ongoing disasters.

Conclusions

The results of this study suggest that the COVID-19 pandemic has created stress and uncertainty for many. Although most of this sample was well educated, owned their own homes (a historical sign of financial success and stability) and were overwhelmingly employed during the pandemic, they still reported a moderate level of perceived stress. It seems likely that Americans that were less financially and employment secure would have reported even more stress. Taken as a whole, these findings suggest the need for pro-active and preventative government policies and programs that build on what was learned during the Covid-19 pandemic and target those groups most likely to need help to reduce or prevent risks associated with severe psychological problems from developing during the next variant wave and future pandemics.

Ethical compliance section

The authors have no funding to disclose. All procedures performed in studies involving human participants were in accordance with the ethical standards of the Tulane University Socio/Behavioral Institutional Review board and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The authors declare they have no conflict of interest. Informed consent was obtained by participants agreeing to participate in the anonymous survey in accordance with the approved study plan by the Tulane University Socio/Behavioral Institutional Review Board.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.