Childhood Asthma and Parental Antidepressant Use in a Nationwide Danish Cohort

Introduction

Asthma is the most common chronic illness among children.¹ Caregivers are responsible for symptom monitoring and medication administration,² which places a significant physical and emotional burden, and adversely affects caregivers’ mental health.³ Caregivers of children with asthma experience higher rates of depression and anxiety, and the prevalence correlates with offspring’s asthma severity.⁴ However, the association between parental mental health and childhood asthma burden, including severity, exacerbation burden, comorbidities, or multiple offspring with asthma, remains poorly understood.

Conversely, families’ mental health impacts family dynamics and child well-being.⁵ Maternal depressive symptoms are inversely correlated with asthma control, school attendance,⁶ asthma knowledge, and medication adherence⁷ among under-resourced populations. Furthermore, caregiver depression and inadequate emotional environment contribute to offspring depression,⁸ which has been implicated in paediatric asthma morbidity⁹ and mortality.¹⁰

Most studies on parental depression and childhood asthma have focused on the mothers of the affected children.^11,12 Treatment of maternal depression leads to improved asthma management and reduced asthma costs.¹³ Fathers of children with asthma usually comprise less than 10% of the caregiver sample in studies that investigated caregiver mental illness,¹⁴ and the effect of paternal depression management on offspring asthma control is not known.

Mental health problems among parents of children with asthma have been associated with lower family socioeconomic status (SES) family. Low SES is a social determinant of health and is independently associated with a higher disease burden.¹⁵ However, the extent to which lower SES and parental mental health independently contribute to children’s asthma burden, particularly in universal healthcare systems, has not yet been well elucidated. In the present study, we investigated the association between caregiver depression and childhood asthma, and the contribution of SES in a Danish cohort comprising of children with actively treated asthma.

Methods

Study Cohort

Children aged 2–17 years with actively treated asthma, defined as the redemption of ≥2 canisters of ICS during 2015, were enrolled in the REASSESS Youth cohort, a nationwide Danish paediatric asthma cohort with data provided by Statistics Denmark, the Danish National Prescription Database and the National Patient Registry as previously described.¹⁶ Families with children with actively treated asthma were matched 1:1 by Statistics Denmark to control families based on the number of siblings, size of the city of residence, families’ disposable income, families’ highest education level, and the mother’s age. All matches were unique. Data was collected from the date of the first ICS redemption (“index date”) in 2015 for a prospective period of 730 days.

Study approval was granted by the Capital Region of Copenhagen’s Data Safety Board (ref. P-2021-602). As per Danish law, access to patient journals does not require informed consent if granted by the Capital Region of Copenhagen’s Data Safety Board. According to Danish law, access to patient journals does not require informed consent, if granted by the Capital Region of Copenhagen’s Data Safety Board. This study was conducted in accordance with the principles of the Declaration of Helsinki. All data and supporting materials are available through Statistics Denmark upon application; however, approval from data sources may be required as per Danish law.

Caregivers were defined as the adult(s) residing with a case child on the index date. A family was defined as co-residing adults, biological and non-biological siblings, and the case child. Biological parents were defined as the individuals registered in birth records as biological mothers or fathers. Children without known caregivers, those with less than 365 days of ICS treatment data, and those belonging to families without a suitable match were excluded (Figure 1).

Figure 1 Flowchart of study inclusion and exclusion criteria in a nationwide cohort of childhood asthma.

Asthma Definitions and Classifications

Based on medication redemption in 730 prospective days from the index date, the GINA 2020 Treatment Steps were used to classify the asthma severity for each child.¹⁷ ICS treatment was calculated as the average daily dose (calculated as beclomethasone standard particle dose equivalents) during the observation period as previously described¹⁸ and stratified according to the age-appropriate GINA 2020 dosing chart.¹⁷

Possible severe asthma was defined based on GINA 2020 Step 3 or 4 with either two moderate or one severe or near-fatal exacerbation for individuals aged 2–11 years, and GINA Step 4 with exacerbation frequency as detailed above, or GINA Step 5 regardless of exacerbations for individuals aged 12+ years.

Disease control was defined based on short-acting β2-agonist (SABA) use quantified by annualised redeemed doses during the observation period. Excessive SABA use was defined as twice the acceptable control criteria for children aged 2–11 years (2 daily puffs at least two days a week, totalling 400 annual doses) or 600 (ages 12+)) annual redeemed doses of SABA.^17,19

Exacerbations were defined as moderate when they were associated with the redemption of at least 37.5 mg of prednisolone for 5 days, severe when associated with hospitalisation (ICD-10 codes DJ45, DJ46, DJ96, DJ960 or DJ969), and near-fatal exacerbations when hospitalisation was to an intensive care unit with or without supplemental ventilation. Re-exacerbations within ± 14 days were considered treatment failures and were not counted as an additional exacerbation.

Socioeconomic Definitions

The variables used to define SES include family workforce attachment defined as the highest SES at the index date, ranked in the following order: Employed (or self-employed), Currently undergoing education, Transfer income recipient (Main source of income stemming from welfare transfers) and Retiree or Unclassifiable. Additional socioeconomic variables included family residence classified based on the number of inhabitants in the municipality of residence at the index date and family disposable income adjusted for the number of cohabiting individuals to correct for the economics of scale,²⁰ family highest attained education level ranked as higher education (Bachelor’s degree or higher), vocational education (at least two years of vocational studies in addition to mandatory schooling) or basic education (ca. 12 years of mandatory schooling).

Antidepressant Use and Comorbidity Definitions

Parental antidepressant use was defined based on the redemption of >1 prescription belonging to the N06A ATC-code group after the birth of the child.

Comorbidities were included as covariates and were defined as the non-respiratory Charlson score, with updated weights by Quan et al,²¹ used either as the total household score, or individual (maternal/paternal) scores depending on the outcome.

Statistical Analyses

Descriptive statistics on demographic variables are presented as median (interquartile range, IQR). Wilcoxon rank-sum test or χ² -test was used depending on continuous or categorical data for groupwise comparisons.

Logistic multivariate regression was used to investigate the impact of childhood asthma (yes/no; none/mild-to-moderate/possible severe), childhood asthma control (controlled/uncontrolled; exacerbating/non-exacerbating) and total family childhood asthma burden (one/multiple children with asthma) on the odds of caregiver antidepressant use. Analyses were adjusted for child age, family disposable income quartile, highest attained family education level, city of residence size, family workforce attachment and comorbidity (Charlson score).

R 4.1.3 (The R Foundation, AU) was used for statistical analyses and Ggplot2 was used for creating figures.

Results

We identified 38,608 families for the 45,820 children with actively treated asthma in 2015. After excluding children below 2 years of age, those without known caregivers, those with episodic ICS treatment, and those without found matches, the final cohort comprised 28,595 children and their families (Figure 1). The median age of the children was 8 years, and 59% of the children were boys. The 28,595 children and their families were 1:1 matched with control families with children without asthma, such that the case and control families did not differ in median child age, maternal and paternal age, number of siblings, or socioeconomic variables (Table 1 and Table 2). Within the cases, the prevalence of uncontrolled asthma was 16% and 4.8% of children had possible severe asthma. In terms of exacerbations, 7.8% of children experienced at least one asthma exacerbation, either moderate or severe (Table 1).

Table 1 Demographics Characteristics of the Children with Actively Treated Asthma, of Their Families, and of the Control Families in a Nationwide Danish Cohort

Table 2 Socioeconomic Characteristics of Families with Children with Actively Treated Asthma and of Control Families

Prevalence of Caregiver Depression

The overall prevalence of caregiver antidepressant use was higher among case families as compared to controls (17% versus 14%, respectively). Maternal antidepressant use was seen in 12% of case families as compared to 9.3% of the control families (p<0.001). Paternal antidepressant use was also higher in case (6.2%) compared to control families (5.3%, p<0.001), although the difference was smaller than that observed in the mothers (Table 1).

Determinants of Caregiver Depression

Odds of caregiver antidepressant use were higher in the case families compared to controls (OR 1.29 (1.23–1.35)), with slightly higher odds among mothers than fathers. There were higher odds of caregiver antidepressant use among cases with uncontrolled asthma compared to controlled asthma (OR 1.43 (1.31–1.56) vs 1.26 (1.20–1.33)).

After adjusting for family asthma burden, defined as either parental asthma or additional children with asthma, the odds of caregiver depression in case families remained increased compared to controls, with trends towards lower odds among those with parental asthma and higher odds of antidepressant use among those with more than one child with asthma. Intriguingly, the odds of caregiver antidepressant use were higher among families with children who had had no exacerbations compared to children who experienced at least one exacerbation during the study period. There was no significant association between caregiver antidepressant use and asthma severity (Figure 2).

Figure 2 Odds for antidepressant use in families of children with actively treated asthma and 1:1 matched control families. Analyses adjusted for matching variables, children’s age, and family non-respiratory Charlson score.

Positive predictors of caregiver antidepressant use among families with children with actively treated asthma included welfare recipiency, residence outside of the capital region and higher Charlson scores. Conversely, higher parental education level and higher disposable income were protective factors against the use of antidepressants among caregivers. (Table 3 and Figure 2).

Table 3 Predictors of Caregiver Antidepressant Use in Families of Children with Actively Treated Asthma Compared to Control Families

Association Between Caregiver Depression and Childhood Asthma Outcomes

The odds for adverse childhood asthma outcomes dependent on the caregiver's use of antidepressants was investigated after adjusting for socioeconomic covariates (Figure 3). The odds of poor asthma control were higher among families with caregiver antidepressant use. The odds of possible severe asthma were lower in families with caregiver antidepressant use. No significant association between asthma exacerbations and antidepressant use was observed. Additionally, we found that caregiver antidepressant use was not associated with their child’s ongoing ICS doses (Table 4).

Table 4 The Distribution of Parental Antidepressant Use Based on Inhaled Corticosteroid Dosing of Children with Actively Treated Asthma

Figure 3 Odds for adverse asthma outcomes in children with actively treated asthma with parental antidepressant use. Analyses adjusted for size of city of residence, families’ disposable income, families’ highest education, children’s age, and family non-respiratory Charlson score.

Discussion

In the present study, we demonstrated that caregiver antidepressant use was higher among families with children actively treated for asthma compared to families with children without known and treated asthma, even after adjusting for socioeconomic factors. The association between caregiver antidepressant use and childhood asthma was stronger in mothers than in fathers. Higher parental education and family income were protective against antidepressant use among caregivers with children with asthma. Although poor asthma control in children was more prevalent among families with caregiver antidepressant use than among non-antidepressant users, there was no association with asthma severity.

Although the link between caregiver depression and childhood asthma has been examined²² to the best of our knowledge, this is the first study to show a positive association between caregiver depression and childhood actively treated asthma, using medication retrieval rather than symptom questionnaires, independent of socioeconomic status in a large nationwide cohort. Our findings are consistent with those of a meta-analysis¹⁴ and support a recent nationwide Danish study²³ which reported an increased risk of asthma in children whose parents had mental health conditions. In this study, we built on the work of Heuckendorff et al,²³ and report not only higher odds of childhood asthma, but also higher odds of poorly controlled asthma among children with caregivers on antidepressant use, independent of SES.

The prevalence of antidepressant use by caregivers in families with children actively treated for asthma was lower than the 25% prevalence reported in a Spanish study.²⁴ A Danish nationwide study estimated a 39% prevalence of caregiver mental health conditions among parents of children, irrespective of their health status.²⁵ These differences are explained by our stringent inclusion of children with continuous ICS use and of active antidepressant use in a system in which antidepressants are not the first-in-line treatment, and lower symptom reporting has been observed among adults in Denmark.²⁶ Irrespective of these differences, our study and prior published data highlight the burden of caregiver depression among children with asthma and identify the necessity of addressing parental needs in the clinical management of childhood asthma, which is likely not addressed during caregivers’ medical visits.

The effects of childhood asthma on caregiver depression are linked to psychosocial influences of life stressors. Caregivers of children with chronic illnesses are exposed to greater stress than caregivers of healthy children.²⁷ Asthma knowledge,²⁸ medication adherence,²⁸ and mitigation of dust mite exposure²⁹ are associated with less parental stress, while Everhart et al³⁰ found that paediatric asthma-related stressors such as asthma severity and SES, had a cumulative negative impact on caregiver quality of life. The age of disease onset is also important since parents of younger children assume more responsibility for asthma management than parents of older children,² whereas asthma control and knowledge tend to improve with age.^31,32 In light of these associations, addressing barriers to parental asthma knowledge could facilitate better paediatric asthma control and improvement in parental mental health.

There is an additive interaction between maternal mental illness and disadvantaged SES on childhood asthma.²³ Indeed, Freeman et al³³ demonstrated that the risk of depression was inversely correlated with SES in the European adult population. In our study, we found that antidepressant use among caregivers of children with persistent asthma was positively correlated with lower SES.

Our observed association between poor asthma control and caregiver antidepressant use is consistent with the known negative effect of caregiver depression on childhood asthma outcomes³⁴ and on school attendance, with higher absenteeism.⁶ One possible explanation is the link between depression severity and altered cognitive function including compromised memory, processing speed and executive functioning,³⁵ which could contribute to ineffective asthma management and a higher likelihood of exposing children to potential triggers such as tobacco exposure.³⁶ In fact, depressed caregivers of children with asthma endorsed lower confidence during their interactions with the child’s medical providers and described more unmet needs.³⁷ Notably, a recent study found that caregiver depression remission improved childhood asthma outcomes, which was partially mediated by improvements in the severity of children’s depressive symptoms.³⁸

Limitations

The present study was strengthened by data obtained from nationwide registries providing a powerful statistical analysis, a strict definition of parental depression and a matched control group. However, we acknowledge that there are several limitations to our observational study related to study design and availability of data. First, the inclusion criteria are based on the redemption of ICS and thus exclude SABA-only treated children. Second, using the outcome of antidepressant use, we exclude many individuals with depression who are receiving active antidepressant therapy. Third, the study is limited to two years of annualised data, yet asthma is a variable disease with a burden changing over time. Despite these limitations, we would like to highlight the strengths of the REASSESS Youth cohort, which include a high degree of data completeness, universal linkage between children and parents, and a robust non-biased data capture due to legal requirements to submit data to Statistics Denmark from the various sources used. Our family-based match allowed robust comparisons and is also a strength of the study.

Conclusion

In this nationwide study of 28,595 children with actively treated asthma, we found that caregiver antidepressant use was more prevalent in families with children with asthma than in control families, independent of socioeconomic factors. Parental antidepressant use was associated with poor asthma control in children, even after adjusting for socioeconomic status, signalling that caregiver mental health can have an impact on childhood asthma.