Association between children's home-schooling and parental psychological distress during the COVID-19 pandemic in Taiwan: Risk and protective factors in a multilevel approach

Abstract

Background

The role home-schooling of children in parental mental health during the COVID-19 pandemic in Taiwan remains unknown. This study aimed to assess the association between parental psychological distress and home-schooling in a socio-ecological context during the peak of the first wave of the COVID-19 pandemic in Taiwan.

Methods

This was a prospective cohort study. In total, 902 parents (father: n = 206, mother: n = 696) who home-schooled children under 18 years of age were recruited by purposive sampling from 17 cities in Taiwan. Data were collected between 19 July and 30 September 2021 through a survey. Multilevel regression models were used to examine the association between parents’ psychological distress and home-schooling considering the characteristics at the person and city levels.

Results

Parental psychological distress was positively associated with difficulty in setting up electronic devices and increased disputes between parents and children, and it was negatively associated with time management and increased time spent bonding with their children during home-schooling (Ps < 0.05). Parents who had a child with health conditions, lived in an extended family, worked from home, lived during the Level 3 alert level, and lived with a median/sporadic level of the COVID-19 community spread by city also reported greater psychological distress (Ps < 0.05). However, parents who had greater household family support reported less psychological distress (P < .05).

Conclusions

Clinicians and policy makers must carefully consider parental mental health while home-schooling during the COVID-19 pandemic in a broader socio-ecological context. A focus is advised on the home-schooling experiences of parents and other risk and protective factors for parental psychological distress at the person and city levels, especially for those with children who require medical interventions and have a medical condition.

1. Introduction

Over the past 3 years, the Coronavirus disease 2019 (COVID-19) pandemic has changed family lives worldwide. In mid-May of 2021, the number of COVID-19 cases in Taiwan jumped abruptly and deaths started to rise steadily. The government put a level-3 emergency alert in place; and parents and children experienced the full shutdown of their country from 15 May to 26 July 2021. This led to a sudden shift from traditional classrooms (in-person instruction) to home-schooling (remote learning) in response to the school shutdowns, which resulted in a completely different learning experience for children and parents.^{1 , 2}

Parents may pass their psychological distress on to their children and practice inappropriate parenting behaviours, which could later contribute to the development of mental problems in these children.3, 4, 5, 6 The association between parental psychological distress and home-schooling during the COVID-19 in Taiwan is unclear. The current findings could help clinicians to assess the mental health of parents who are home-schooling children during additional waves of the COVID-19 pandemic as well as future pandemics.

Home-schooling places considerable demands on parents to, for example, set up electronic devices and ensure adequate internet connectivity. Furthermore, such online set-ups can be accidently disconnected, and parents are responsible for immediate repairs. Additionally, parents have to manage the supervision of their children's learning progress alongside their professional, personal, and parenting roles. Only 14% of primary school students can complete remote school activities without assistance⁷; therefore, parents need to be involved in their children's home-schooling.

By contrast, the impact of home-schooling is not universally negative; current evidence suggests that parents’ experiences of home-schooling are mixed.^{8 , 9} Some parents report positive aspects of home-schooling during school closures, including less time spent commuting and more time spent bonding with their children, which can increase family closeness. Additionally, the home is perceived as a better learning environment for some children.⁸ Therefore, the effectiveness of parent-child interactions or parental time management may increase during home-schooling.

Bronfenbrenner's socio-ecological model¹⁰ provides a comprehensive framework to examine psychological distress during home-schooling by considering the risks and protective factors for parental mental health. These factors are individual (e.g., age, father/mother, and educational level of the parent), interpersonal (social support, child factors, and family factors), organizational (types of work), and community-related (e.g., COVID-19 alert level).^{11 , 12} Furthermore, the recorded level of community spread differed across cities in Taiwan during the COVID-19 epidemic, and previous studies had found that an individual's psychological distress is associated with the level of COVID-19 community spread in their respective city.^{13 , 14} Parental psychological distress during home-schooling is thus associated with multiple factors at the person-as well as the city level.

Given the ongoing impact of COVID-19, there is a clear need to comprehensively quantify the parental psychological distress caused by home-schooling. This study therefore aimed to (1) assess the association between parental distress and negative and positive experiences of home-schooling during the peak of the first wave of the COVID-19 pandemic in Taiwan; and (2) identify other risk and protective factors against parental psychological distress in a broader socio-ecological context.

2. Methods

2.1. Sampling procedures and participants

This was a prospective cohort study with an anonymous questionnaire survey. In total, 992 parents (fathers: n = 206, 22.84%; mothers: n = 696, 77.16%) who were home-schooling children aged ≤18 years were recruited through convenience sampling from 17 cities in Taiwan. Ninety participants (9.07%) were excluded due to failures on three attention check questions. Therefore, 902 parents completed the questionnaire and were included in our analyses; 661 were interviewed online via the survey management software SurveyCake, and 241 filled in the paper-and-pencil version of the questionnaire. The first page of the questionnaire provided information on the purpose of the research and the consent statement, to which the participants had to agree. Completion of the questionnaire took approximately 15–20 mins. Data were collected between 19 July and 30 September 2021.

2.2. Measures

2.2.1. Parental psychological distress

Parental psychological distress was assessed using the Brief Symptom Rating Scale 5 (BSRS-5).¹⁵ This five–item scale has been shown to have excellent validity and reliability in Taiwan and accounts for psychological symptoms such as depressive feelings and low mood.^{16 , 17} Each item was scored on a Likert scale from 0 to 5 (0 = “never,” 1 = “mild,” 2 = “moderate,” 3 = “severe,” and 4 = “very severe”). The sum of these scores ranged from 0 to 20. Cronbach's α for BSRS-5 was 0.88. The BSRS-5 comprises the following five dimensions: (1) Anxiety (feeling tense of “keyed up”), (2) Depression (feeling blue), (3) Hostility (feeling easily annoyed or irritated), (4) Interpersonal Sensitivity (feeling inferior to others), and (5) sleep difficulties. According to Lee's findings,¹⁵ the BSRS-5 is an efficient tool for the screening of suicidal ideation-prone psychiatric inpatients, general medical patients, and community residents. Understanding the differentiated symptom domains for each group and the relationship between them can help health care professionals in their preventative programs and clinical treatment. A total score on the BSRS-5 above 14 may indicate severe mood disorders. Scores between 10 and 14 may indicate moderate mood disorders and scores between 6 and 9 may indicate mild mood disorders.

2.3. General negative and positive experiences of home-schooling

We surveyed the experiences of home-schooling, which have been identified as an important factor for parental mental health during the COVID-19 pandemic.^{8 , 12 , 18 , 19} The negative experiences index was evaluated using the following two items: “The dispute between parents and children increased” and “Dealing with children's home-schooling devices or emergencies took longer and was harder than I expected.” The positive experiences index was evaluated using the following two items: “The time for parents to chat with children increased” and “Time management is more efficient during home-schooling than during classroom learning.” Each item was scored on a Likert-type scale from 1 to 4 (1 = “disagree,” 2 = “slightly disagree,” 3 = “agree,” and 4 = “strongly agree”). The scores of each index ranged from 2 to 8.

2.4. Covariates

2.4.1. City-level variable

2.4.1.1. COVID-19 community level by city

Taiwan was categorized according to three levels of the spread of COVID-19 during the pandemic. The red zone was defined as having high or massive levels of COVID-19 infections, i.e., Taipei and New Taipei City. The yellow zone cities with median or sporadic levels of COVID-19 infections, namely Taoyuan, Miaoli, Taichung, Changhua, Kaohsiung, and Pingtung. The green zone was defined as having low levels of COVID-19 infections, with few suspected or confirmed cases admitted to hospitals; it covered Keelung, Hsinchu, Yunlin, Chiayi, Tainan, Hualien, Yilan, Nantou, and the offshore islands.

2.5. Person-level variables

2.5.1. Individual

(1) Age, reported by parents; (2) Parents, self-reported: “Father” coded as “0” and “mother” as “1”; and (3) Educational level, was classified as “junior high school and below,” “senior high school,” and “university and above.”

2.5.2. Interpersonal

(1) Social support. (a) Perceived household family support was assessed with four items: whether the respondent felt close to their family (emotional support); whether their family helped them when they were in need (instrumental support); whether they provided them with advice when they were in need (informational support); and whether they appreciated their thoughts and behaviors (appraisal support). (b) Perceived support by relatives, (c) perceived friend support, and (d) perceived neighbor support were similarly evaluated with four items. Each item was scored on a Likert-type scale from 1 to 4 (1 = “disagree,” 2 = “slightly disagree,” 3 = “agree,” and 4 = “strongly agree”). Cronbach's α for the household family, the relatives, the friends, and the neighbors support items were 0.86, 0.90, 0.90, and 0.93, respectively. (2) Child factors. (a) Number of children being home-schooled, reported by parents. (b) Children who required medical interventions, coded yes = 1 vs. no = 0. (c) Children with medical diagnosis, including neurodevelopmental disorders (e.g., autism spectrum disorders or developmental delay), neurological conditions (e.g., epilepsy), sensory impairment or other paediatrics diseases, coded yes = 1 vs. no = 0. (3) Family factors. (a) Immigrant families were coded as yes = 1 vs. no = 0. (b) Type of family structure was categorized into nuclear family, extended family, and others.

2.5.3. Organizational

Type of work during home-schooling was categorized into work from home (WFH), alternate working to the office, and regular work in the office (RWO).

2.5.4. Community

The COVID-19 alert levels that were in place during this study were levels 2 and 3. The level 3 COVID-19 alert was in place from 19 to 26 July 2021, and the level 2 alert from 27 July to 30 September 2021.

2.6. Data analysis

The characteristics of the sample are presented in Table 1 . Most parents were university graduates and above (73.95%), with those who only completed junior high school and below representing the smallest group of respondents (1.77%). Few respondents were from immigrant families (3.99%). Most parents lived in a nuclear family (72.28%) and only approximately 20% in an extended family. Furthermore, 32.26% of the children required medical interventions during home-schooling and 10.31% had a diagnosis of neurodevelopmental or neurological disorders. Approximately 30% of parents maintained RWO; however, 41.46% reported WFH, and 24.94% worked alternately. As for the COVID-19 alert level, 43.79% of respondents filled in the questionnaire during a level 3 phase. Most parents lived in a red zone during their reported home-schooling phase (79.38%), while those in a green zone represented the smallest group (7.87%).

Table 1.

Characteristics of the study sample (N = 902).

Characteristics	Mean	SD	n	%
Parental psychological distress	9.81	3.84
Experiences of home-schooling
Negative experiences index	5.06	1.45
Positive experiences index	5.79	1.24
City-level variable
COVID-19 community level by city
Red zone			716	79.38
Yellow zone			115	12.75
Green zone			71	7.87
Person-level variable
Individual
Age	42.48	5.73
Parents
Father			206	22.84
Mother			696	77.16
Education level
Junior and below			16	1.77
Senior			219	24.28
University and above			667	73.95
Interpersonal
Social supports
Household	3.19	0.52
Relatives	2.19	1.25
Friends	2.32	1.28
Neighbors	1.53	1.09
Child factors
Number of children in home-schooling	1.56	0.62
Child having medical intervention			291	32.26
Child with medical diagnosis			93	10.31
Family factors
Immigrant family			36	3.99
Types of family structure
Nuclear family			652	72.28
Extended family			180	19.96
Other family			70	7.76

Characteristics	Mean	SD	n	%
Organizational: Types of work
WFH			374	41.46
Alternately			225	24.94
RWO			303	33.59
Community: During COVID-19 alert level
Level 3			395	43.79
Level 2			507	56.21

WFH, work from home; RWO, regular work in the office.

Multilevel regression models (MLMs) were used to investigate the association between parental psychological distress and home-schooling, and they facilitated the examination of connections at each level (person- and city-level) and the amount of variation taken into account at each level.²⁰ All continuous variables were grand-centered. We used HLM version 7.03 (Scientific Software International Inc.) to apply the MLMs and calculate a population-averaged model with robust standard errors. The data had a multilevel structure, with persons (level 1) nested within cities (level 2), and each multilevel regression model thus consisted of two hierarchical levels. The level 1 model included two experience indices of home-schooling (negative and positive) and individual, interpersonal, organizational, and community-related factors. The level 2 model included COVID-19 community spread by city. Let Y_ij = parental psychological distress, our dependent variable taken on the ith parent associated with the jth city. The level 1 equation of the full model was as follows:

Y_ij = β_0j + β_1jNegative_ij + β_2jPositive_ij + β_3jAge_ij + β_4jMother_ij^(Ref=Father) + β_5jSenior_ij^{(Ref=Junior and below)}+ β_6jUniversity and above_ij^{(Ref=Junior and below)}+ β_7jHousehold_ij + β_8jRelatives_ij + β_9jFriends_ij + β_10jNeighbors_ij + β_11jNumber of children in home-schooling_ij + β_12jChild having medical intervention_ij^(Ref=None) + β_13jChildren with medical diagnosis_ij^(Ref=None) + β_14jImmigrant family_ij^(Ref=None) + β_15jExtended family_ij^{(Ref=Nuclear family)} + β_16jOther family_ij^{(Ref=Nuclear family)} + β_17jWFH_ij^(Ref=RWO) + β_18jAlternately_ij^(Ref=RWO) + β_19jLevel 3_ij^{(Ref=Level 2)} + γ_ij

The level 2 equation for the full model is as follows:

β_0j = γ₀₀ + γ₀₁*Yellow zone_j^{(Ref=Red zone)} + γ₀₂∗Green zone_j^{(Ref=Red zone)} + μ_0j

3. Results

3.1. Parental psychological distress and home-schooling

The intra-class correlation coefficient (ICC), reflecting the variance across the 17 cities (level 2), was 4.12% for parental psychological distress. This indicates that the multilevel modeling approach was appropriate for our analysis.²¹

Table 2 presents the results of Models 1, 2, and 3. Model 1 shows that parental psychological distress was negatively associated with positive experiences index, e.g., effectiveness (P < .001) and positively associated with negative experiences index, e.g., emergency interventions (P < .001). In Model 2, with added variables at the personal level, parental psychological distress was also negatively associated with positive experiences index (P = .037) and positively with negative experiences index (P < .001). The final full model (Model 3) indicated a further significant effect on positive (P = .039) and negative (P < .001) experiences after adding the COVID-19 community spread variable.

Table 2.

Summary of MLM for parental psychological distress (N = 902).

	Model 1			Model 2			Model 3
Fixed effect	Coeff	95% CI	P	Coeff	95% CI	P	Coeff	95% CI	P
Intercept	10.19	9.62–10.77	<0.001	8.12	5.33–10.91	<0.001	7.50	4.77–10.23	<0.001
Experiences ofhome-schooling
Negative experiences index	0.89	0.77–0.01	<0.001	0.74	0.59–0.90	<0.001	0.74	0.58–0.89	<0.001
Positive experiences index	−0.45	−0.57 to −0.33	<0.001	−0.20	−0.39 to −0.01	0.037	−0.20	−0.38 to −0.01	0.039
City-level variable
COVID-19 community level by city
Yellow zone(Ref=Red zone)							1.61	0.95–2.27	<0.001
Green zone(Ref=Red zone)							0.17	−0.67 – 1.00	0.702
Person-level variable
Individual (Parents)
Age				−0.03	−0.07 – 0.01	0.157	−0.03	−0.07 – 0.01	0.113
Parents: other(Ref = Father)				0.29	−0.27 – 0.84	0.309	0.27	−0.28 – 0.82	0.342
Educational level
Senior(Ref = Junior and below)				0.40	−2.31 – 3.12	0.772	0.48	−2.22–0.18	0.730
University and above(Ref = Junior and below)				0.88	−1.82 – 3.58	0.522	0.92	−1.76–3.61	0.501
Interpersonal
Social supports
Household				−1.63	−2.09 to −1.17	<0.001	−1.62	−2.08 to −1.17	<0.001
Relatives				−0.11	−0.31 – 0.09	0.273	−0.13	−0.32 – 0.07	0.209

	Model 1			Model 2			Model 3
Fixed effect	Coeff	95% CI	P	Coeff	95% CI	P	Coeff	95% CI	P
Social supports
Friends				−0.08	−0.28 – 0.09	0.414	−0.06	−0.26 to −0.14	0.566
Neighbors				−0.17	−0.38 – 0.05	0.127	−0.17	−0.38 – 0.04	0.119
Children factors
Number of children in home-schooling				−0.09	−0.46 – 0.28	0.630	−0.10	−0.46 – 0.27	0.605
Child having medical intervention (Ref = None)				0.57	0.10–1.05	0.018	0.59	0.12–1.05	0.015
Child with medical diagnosis (Ref = None)				0.75	0.02–1.49	0.045	0.83	0.10–1.56	0.026
Family factors
Immigrant family(Ref=None)				−0.88	−2.01 – 0.24	0.123	−0.88	−2.00 – 0.23	0.122
Types of family structure
Extended family(Ref=Nuclear family)				0.64	0.09–1.20	0.024	0.62	0.06–1.17	0.029
Other family(Ref=Nuclear family)				0.46	−0.38–1.30	0.282	0.40	−0.43 – 1.24	0.345
Organizational: Types of work
WFH(Ref=RWO)				0.94	0.42–1.47	<0.001	0.96	0.44–1.48	<0.001
Alternately(Ref=RWO)				0.38	−0.20 – 0.96	0.199	0.38	−0.20 – 0.96	0.195
Community: During COVID-19 alert level
Level 3 (Ref=Level2)				0.47	0.02–0.92	0.041	0.45	0.01–0.90	0.047

Random effect	Variance Component	P	Variance Component	P	Variance Component	P
Intercept (μ)	0.53	<0.001	0.44	<0.001	0.00005	>0.500
Individual-level (γ)	12.31		10.85		10.81

WFH, work from home; RWO, regular work in the office.

Additionally, Model 3 indicated other risks and protective factors. On the interpersonal level, parental psychological distress was negatively associated with household social support (P < .001). Moreover, parental psychological distress was higher in those with children who required medical interventions (P = .015) or had neurodevelopmental or neurological disorders (P = .026). Additionally, compared with parents living in a nuclear family, those living in an extended family were more likely to report psychological distress during home-schooling (P = .029). Our organizational-level analysis showed that compared with parents who maintained their RWO, WFH parents were more likely to report psychological distress (P < .001). The community level analysis revealed that parents reported greater psychological distress during level 3 than level 2 alert periods (P = .047). Finally, parents in the yellow zone reported greater psychological distress than those in the red zone (P < .001).

4. Discussion

To the best of our knowledge, this is the first study to examine the association between parental psychological distress and home-schooling experiences during the COVID-19 pandemic in Taiwan. Our main findings are that parents who reported having to deal with urgent emergency interventions or experiencing low effectiveness during home-schooling exhibited more psychological distress. The risk factors for parental mental health during home-schooling due to the COVID-19 pandemic were (1) having a child with health conditions, (2) living in an extended family, (3) working from home, (4) home-schooling during a level 3 alert level, and (5) living in an environment with a median/sporadic level of community spread. In contrast, having a supportive household family was a protective factor for parental mental health.

4.1. Parental psychological distress associated with experience of home-schooling

4.1.1. Negative experiences

Parents who had to manage emergency interventions during remote learning reported greater psychological distress. This pandemic was sudden and unexpected; therefore, most parents were unprepared for this quick shift from traditional learning to home-schooling. Previous research indicates that parental depression and stress due to home-schooling are significantly positively associated with parents' perceived failure to provide at-home education.²² Parents reported that dealing with equipment or devices for home-schooling was difficult. Furthermore, the transition to remote learning required parents to teach their children knowledge, design individualized content, and constantly assess their children's emotional needs. Therefore, they became proxy educators and supervisors of the remote learning of their children.² These added extra responsibilities and stressors to parents' usual tasks during school closures, and they were required to perform multiple, and sometimes conflicting, roles.²³ In addition, some parents had to work longer hours each day to meet home-management and home-schooling obligations, which can affect sleep and reduce time for leisure activities.^{1 , 24}

4.1.2. Positive experiences

The psychosocial impact of home-schooling during the COVID-19 pandemic on parents’ mental health was not only negative. Parents reported that home-schooling increased the time they had to chat with their children, and that eliminating or reducing physical commute times aided their time management, which may have increased closeness and understanding among family members. Some research has reported that the pandemic provided an opportunity to cultivate positive qualities in parent-child interactions, including “appreciation,” “developing tolerance and understanding,” and “learning to cope and develop patience”.⁸ The effectiveness of parent-child interactions and time arrangement increase during home-schooling. However, the precise mechanisms underlying positive effects of home-schooling need further exploration.

4.2. Parental psychological distress associated with other factors

4.2.1. Protective factors

We found that social support from the household family was the strongest protective factor against psychological distress among parents during home-schooling. Calear et al.¹ found that parents who reported higher perceived social support from their child's school tended to exhibit lower levels of psychological distress; and these together with our findings are consistent with the general literature on mental health that links social support to better mental health outcomes.²⁵ The mediating effect of social support on the association of perceived stress and mental health outcomes has long been recognized.^{26 , 27} Social support can increase an individual's resilience and support their adaptive psychological capacities.²⁸

4.2.2. Risk factors

Parents who had a child with health conditions, lived in an extended family, worked from home during the level 3 alert phase, and lived in a city/area with a median/sporadic level of COVID-19 community spread reported greater psychological distress. Previous research has highlighted the potential risk factors for parental psychological distress during the COVID-19 pandemic, which are in line with the factors we identify here, that is, having a child with health conditions,^{29 , 30} living in an extended family,³¹ working from home,²³ and experiencing a level 3 alert phase.³²

It is also worth noting that, compared with parents living in a nuclear family, parents living in an extended family had higher psychological distress. Why do grandparents not reduce parental psychological distress? It may be that many grandparents may fall under the “high-risk” category for COVID-19-related illnesses, or even death, as a result of their age or the presence of underlying health conditions.²⁴ As such, grandparents cannot provide social support to the family, and parents’ duties to care for grandparents would increase significantly during the COVID-19 pandemic. Many parents did not rely on support from grandparents or other family members for childcare or help with parenting-related activities.

Additionally, the psychological distress reported by parents living in cities with only sporadic Covid-19 spread (such as Taoyuan) was higher than those living in cities with massive spread (such as New Taipei City), which was most likely due to the “uncertainty” in cities with only sporadic spread and to the constant fear of infection. Therefore, it was important to determine the associations between parental mental health and home-schooling in a larger socio-ecological context.

Lastly, there were possible associations between parental psychological distress and living with extended family who had Covid-19, immunodeficiency, metabolic syndrome, cardiovascular diseases, or other potential infection risks. While living with extended family may not necessarily be considered bad for mental health, it does increase the chance of elderly having pressure regarding potential infections in the household. While this study did not examine those particular variables, it will certainly be worth exploring further in future research.

4.3. Strengths and limitations

This study has several strengths: (1) the survey was conducted during the peak and slowdown of the first wave of the COVID-19 pandemic in Taiwan. (2) Data were collected during school closures rather than relying on retrospective reports. (3) This was a large and diverse sample of parents who had children ≤18 years of age with and without health conditions. (4) We examined the associations of parental psychological distress and home-schooling with a wide range of socio-ecological factors, including individual, interpersonal, organization, and community-related factors. (5) The collected sample shows strong representativeness of the Taiwanese population regarding families with a wide range of children's ages (under aged 18), as well as the data gathered from 17 of 22 cities in Taiwan. Additionally, we assessed different levels of the spread of COVID-19 across cities, which relates to the perceived risk of infection in the community more than to the official alert level.²² These strengths increase the validity of the findings of this study.

The following limitations should, however, be considered: (1) we relied on the parents' reports of their current psychological distress rather than on longitudinal data. Therefore, we do not know whether parents were psychologically distressed before the COVID-19 pandemic. Additionally, we were unable to assess the parental psychopathology that contributed to the negative outcomes of home-schooling. (2) Many standardized and validated measures were not used to assess parents' and children's experiences of home-schooling in this study, since we chose to not substantially increase the length of the survey; we assumed keeping the questionnaire short would increase the response rate. Furthermore, the items used in this study were designed based on previous COVID-19 research on the effects of home-schooling on parents and family functioning. (3) Data were collected only once rather than in repeated measures. Therefore, the observed association between parental psychological distress and home-schooling does not necessarily indicate a cause-effect relationship. (4) Additionally, the period for data collection was brief; however, we did observe an association between home-schooling and parental distress even within this short time. Further studies are needed to confirm any long-term impact of home-schooling on parental distress. (5) Children with a medical diagnosis were not clearly differentiated in this study. As children with a medical diagnosis account for only 93, we could not differentiate different diagnosis groups in the analysis. However, children with neurological disorders, neurodevelopmental disorders, or sensory impairment may have different effects on parental psychological distress. This is an important issue to be examined in the future study.

5. Conclusions

In Taiwan, school closures have been used to reduce the spread of COVID-19. In response, parents and children had to adapt to new modes of remote learning.³³ The present results clearly show that parental mental health is negatively associated with having to deal with emergencies during home-schooling of a child. However, home-schooling does not only have negative impacts on parental mental health. Parents also reported positive experiences of more parent-child interactions and better time management during home-schooling. Another interesting finding of this study is that social support from the household family is an important protective factor against parental psychological distress and may reduce the negative impact of other risk factors. The risk factors for parental mental health during home-schooling are having a child with health conditions, living in an extended family, working from home, and a median/sporadic level of COVID-19 community spread. Therefore, clinicians and policy makers must carefully consider the mental health condition of parents during home-schooling in a broader socio-ecological context during later phases of the COVID-19 pandemic or future natural disasters (e.g., fires, floods, and earthquakes). When the schools are closed and the needs of home-schooling increase, it would be crucial for the government to aid in the form of remote devices and settings, home-based parent-child activities, and advocacy of informal household social support helping strengthen family functions, especially for those with children who require medical interventions and have a medical condition.

Ethical approval

Ethics committee approval was not required for this anonymous questionnaire study. Furthermore, the study was performed in accordance with the ethical standards as laid down in the Declaration of Helsinki of 1964 and its subsequent amendments or comparable ethical standards.

Availability of data and materials

The datasets used or analyzed during the current study are available from the corresponding author, Hsin-Hui Lu, on reasonable request.

Funding

Editing of this article was funded with a grant from the National Science and Technology Council, Taiwan (MOST 108-2410-H-040-010-MY3; MOST 111-2410-H-040-003-MY4).

Declaration of competing interest

The authors declare that they have no competing interests.