Maternal Depressive Symptoms and Emergency Department Use Among Inner-city Children With Asthma FREE

ENSURING THE appropriate use of emergency department (ED) facilities can play an essential role in the containment of health care resources and costs. It is estimated that 1.7 million visits to the ED are made by children with asthma¹ at a cost exceeding $90 million.² The high use of emergency medical care among low-income, minority children as a primary treatment for asthma is well established.³ Although higher rates of asthma are found among African American and other minority children,⁴ neither disease severity nor age of the child (or the combination) accounts for differences in patterns of health care use.^5- 6 In a review of studies of health care use by children, Riley and colleagues⁷ concluded that the child's health needs explain less than 17% of the variance of health care use. Thus, attention has turned to the role that other factors play in a mother's decision to seek acute medical care for her child.

A limited number of studies have examined the relationship between characteristics of caregivers and ED use for their children. Adolescent mothers who had high rates of social and behavioral problems when they were younger are more likely to bring their children to the ED unnecessarily.⁸ Among preschoolers, Lavigne and colleagues found that parents of children who were younger, of a minority group, and had either a psychiatric diagnosis or history of behavioral problems were more likely to take their children to the ED.⁹ Children of single parents³ and those from poorer families¹⁰ are also seen more frequently in the ED. Parents who perceive their children as being more vulnerable also use urgent care services more frequently.^11- 12

Among young inner-city children with asthma, morbidity is high and asthma is poorly controlled and poorly managed.¹³ Emergency department use is common among this population and often serves as a primary care source for asthma.^13- 14 Inadequate instructions to parents regarding medication use and low adherence were initially identified as risk factors for high ED use.^15- 16 More recently, psychosocial and behavioral correlates were identified in 445 children treated for asthma in an urban ED.¹⁴ Younger children with more symptoms who had previously been hospitalized for asthma were found to be at high risk for receiving ED care. Parental factors such as lack of confidence in the efficacy of asthma medications and lack of use of a criterion for deciding to go to the ED were also associated with increased acute care use. In a multicenter trial in Australia, parental worry and attitudes about the appraisal of asthma severity were also significantly related to repeat ED visits.¹⁷

The psychological status of caregivers may also play a role in the decision to use the ED. The National Cooperative Inner-City Asthma Study investigated psychosocial factors and asthma morbidity among 1528 inner-city children (ages 4-9 years). It was found that 50% of caregivers reported clinically significant levels of psychological distress,¹⁸ as measured by the Brief Symptom Inventory. Caregiver (but not child) mental health was the strongest predictor of hospitalizations for asthma; parents who scored above the clinical cutoff for the Brief Symptom Inventory were nearly twice as likely to have children who were hospitalized.¹⁸ Conversely, parenting style, social support, alcohol abuse, and life stress contributed relatively little to the prediction of asthma morbidity.¹⁹

Inner-city adults,^20- 21 as well as those having children with chronic health problems, including asthma,²² are at increased risk of depression. In neonates, acute care use has been associated with emotional distress in mothers. However, to date, no study has prospectively evaluated whether depressive symptoms of the mother are associated with ED use (without hospital admission) for asthma care for their children. We hypothesized that younger child age, increased asthma morbidity, and maternal depression would be associated with ED use among inner-city school-age children with asthma.

This study combines data from identical surveys administered at baseline and 6 months in 2 community-based interventions designed to evaluate and improve asthma management among inner-city children and their families. In the first survey, 33 inner-city elementary schools from Baltimore, Md, and Washington, DC, were contacted and invited to participate. In the second survey, 32 inner-city Baltimore elementary schools were contacted. Schools were selected if they met the following criteria: (1) more than 85% of students were African American, (2) more than 350 students were enrolled in the school, and (3) administrative consent to participate in a subsequent asthma education study was authorized. The initial surveys were conducted between October 1996 and September 1997 and were approved by the Joint Committee on Clinical Investigation of The Johns Hopkins University School of Medicine, Baltimore.

Children in kindergarten to grade 5 who had an asthma diagnosis listed on their health records were screened for eligibility. Children were eligible for the study if their mothers reported that they had (1) asthma diagnosed by a physician, (2) day or night asthma symptoms, including wheezing, shortness of breath, and/or a cough at least once a week during the past 2 weeks, and/or (3) at least 1 visit for asthma to the ED in the previous 6 months or 1 overnight hospitalization for asthma in the previous year. Ninety-eight percent of the children were African American.

Parents or guardians of eligible children were mailed a consent form. After obtaining informed written consent, telephone interviews were completed. The child's biological mother was the interview respondent 89% of the time. The interview, administered by trained interviewers over the telephone, queried demographic information about the child, as well as asthma symptoms and current management practices. Studies of pediatric populations have generally found good agreement for medically related information when querying information such as hospitalizations and number of visits for asthma.^23- 24

Interviewers were hired through a research contract group and blinded to the study purpose and hypotheses. Within the total sample of 338 respondents, an expanded interview that queried mood was conducted with the last 177 mothers enrolled in the studies; 6-month self-report data on ED use were available for 158 of the children in the baseline sample (89%). Mothers who did not participate in the follow-up interview were more likely to report lower levels of asthma morbidity at baseline (mean ± SD, 6.3 ± 7.2 vs 15.7 ± 11.7; P<.001, days and nights with symptoms per month, respectively), but did not differ in age, level of depressive symptoms, child age, or annual income. All analyses reported were performed on this subset of children and their mothers.

The primary outcome measure was the number of ED visits for asthma (which did not result in hospitalization) reported by the mother between the baseline and 6-month follow-up interview. Participants were asked, "In the last six months, how many times was your child treated for asthma in the emergency room but not hospitalized?" Child measures included an index of asthma morbidity (described below), an estimate of the amount spent for asthma-related medications each month, the number of visits the child had with the primary care provider (PCP) in the past year, and the length of relationship with the PCP.

An index of asthma morbidity was developed at baseline by summing the total number of days and nights during the previous 6 months that the child experienced asthma-related symptoms. This method has been previously shown to have good concurrent validity with other standard measures of asthma morbidity in children.^25- 26 Respondents were asked 2 questions: (1) "For the last six months, on average, how many days per week did cough, wheeze, or shortness of breath limit your child's exercise, ability to play sports or play with friends?" and (2) "For the last six months, on the average, how many nights per week did your child wake up at night with cough, wheeze, shortness of breath or tightness in chest?" Total number of days and nights with symptoms in the previous month were summed. Tertiles were calculated that reflected low (0-8 days and/or nights), moderate (9-16 days and/or nights), and high (≥17 days and/or nights) morbidity.

Maternal measures included demographic information, an assessment of depressive symptoms, and quality of life. Depressive symptoms were assessed using a modified version of the Center for Epidemiological Studies Depression (CES-D) Scale.²⁷ The CES-D has been widely used in clinical trials in both general and psychiatric populations and has sound psychometric properties (ie, reliability, discriminant, and convergent validity). To reduce the burden on participants and, because briefer versions of the test have also been shown to have good predictive accuracy when compared with the full-length version,^28- 29 an 11-item version of the CES-D was used. Scores on each item ranged from 0 to 3 with higher scores reflecting increased dysphoria; scores of 9 or more were judged to reflect clinically significant levels of depressive symptoms, while scores of 8 or less were judged to reflect nonsignificant levels of distress. Quality of life was assessed using an abbreviated 5-item version of the Pediatric Caregiver Quality of Life Questionnaire.³⁰ Mothers were also queried about medication use of the child.

The primary objective of this study was to prospectively determine whether baseline characteristics of inner-city children with asthma and their mothers were associated with reports of ED use in the following 6 months. Because the prevalence of maternal depression was so high in this sample, caregivers were initially divided into 2 categories—low and high levels of depressive symptoms. χ² Analyses and t tests were used to assess differences in demographic, health care utilization patterns, and asthma management practices. To examine characteristics of the child and mother related to ED use, ED visits were dichotomized (no/yes), and Poisson regression was used to estimate crude prevalence ratios (PRs). (We used PRs rather than odds ratios because odds ratios will produce inflated estimates of risk when the prevalence of the outcome [ED use] is high, as in this study.³¹) Poisson regression was also used to derive adjusted PRs, controlling for the effects of child age, parent age, asthma morbidity, depression, and family income. SPSS PC (version 10.0; SPSS Inc, Chicago, Ill) and SAS (version 6.12; SAS Institute Inc, Cary, NC) software were used to analyze the data.

Mothers had a mean ± SD age of 33.3 ± 6.7 years. Of the 140 mothers, most (ie, 70%) of the mothers had completed high school or obtained a General Education Development certificate and reported having state-sponsored medical assistance (56%) or private health care insurance (36%). The children with asthma had a mean ± SD age of 7.9 ± 2.2 years and, 59% of the sample were girls. Mothers reported that the majority (90.5%) of children had been prescribed asthma medications in the past 6 months. More than half (56.5%) had been prescribed a daily asthma medication. Forty-two percent of the children had taken oral steroids for asthma in the past year and 60.1% had a nebulizer at home. The asthma morbidity index was negatively correlated with use of daily asthma medications (r = −0.20, P = .02) and oral steroids (r = −18, P = .03) and positively associated with frequency of home nebulizer use (r = 0.25, P = .02).

Nearly half (ie, 47%) of the mothers reported clinically significant levels of depression. Given this finding, we first examined differences between families in which mothers were categorized as reporting high and low levels of depressive symptoms. As shown in Table 1, mothers with depressive symptoms were significantly more likely to be unemployed (P < .001), be in the lowest income category (ie, <$10 000 per year) (P = .004), and report a lower quality of life (P = .001). Depressive symptoms at baseline and at 6 months were highly correlated (r = 0.53; P<.001).

Few other differences emerged. There were no statistically significant differences in age, sex, or asthma morbidity between the children of mothers with low and high depressive symptoms. Similarly, no differences were observed in mothers' ages or education levels. Thus, family characteristics appeared relatively comparable between groups. There were also no significant differences in reports of lifetime hospitalizations for asthma, the amount spent on asthma medications each month, number of visits for routine asthma care, and length of relationship with the PCP between mothers with low and high levels of depressive symptoms, as shown in Table 2. Overall, the children and their asthma-care patterns, including medication use, appeared relatively comparable between mothers who reported low and high levels of depressive symptoms. Also, in both groups, the estimated time from the child's home to the PCP's office was comparable (ie, 15-30 minutes). However, mothers with high levels of depressive symptoms were 40% (PR, 1.4; 95% confidence interval [CI], 1.0-3.6; P = .04) more likely to report having taken their child to the ED in the following 6 months compared with mothers with low levels of depressive symptoms (Table 3). Mothers who reported the highest tertile of depressive symptoms also reported the most frequent use of the ED (Mantel-Haenszel test, χ²₁ = 6.33, P = .01).

Other factors were also associated with reports of ED use (Table 3). As we had anticipated, mothers of children with the highest asthma morbidity were more likely to report taking their child to the ED (PR, 1.9; 95% CI, 1.4-7.1; P = .006). Mothers between the ages of 30 and 35 years were more than twice as likely to report taking their children to the ED for asthma care (PR, 2.2; 95% CI, 1.9-9.3; P = .001) compared with those who were younger or older. (There was a trend [P = .09] for younger mothers to also use the ED.) Neither child age nor family income was associated with reports of ED use.

Poisson regression was used to evaluate the independent contribution of child age, asthma morbidity, maternal age, depression, and family income in predicting reported ED use. Variables were selected for inclusion based on theoretical importance, significant association in the bivariate analyses, and findings from cross-sectional studies. Variables significantly contributing to reports of ED use included asthma morbidity, maternal age, and depression status (Table 3).

The principal finding of this study is that, in the inner city, mothers with high levels of depressive symptoms are 30% more likely to report taking their children to the ED for asthma care after adjusting for other factors. Among inner-city children with asthma, the age and psychological adjustment of the mother were the strongest predictors of ED use, independent of the child's asthma morbidity. Mothers with depressive symptoms made frequent ED visits, with 12.0% reporting 4 or more ED visits compared with 1.2% of visits by mothers with low levels of depressive symptoms (Table 2).

A secondary finding was that symptoms of depression are common among mothers of inner-city children with asthma. Nearly half (ie, 47%) reported clinically significant levels of depression, which is more than 5 times the national point prevalence rate of 5% to 9%.³² These results are supported by a recent study that found that 39% of inner-city mothers with young children reported high levels of depressive symptoms,³³ and a prospective study that found women with infants who had even 1 ED visit were much more likely to be depressed (odds ratio, 3.2; 95% CI, 1.5-3.0).³⁴ The pervasiveness of depressive symptoms among inner-city mothers is worrisome given the well-documented adverse consequences to both parent and child.³⁵ As expected, mothers with depressive symptoms reported the poorest quality of life. Though widespread, depression remains untreated in up to 90% of those affected.³⁶ In our sample, depressive symptoms were highly correlated at baseline and 6 months later.

Poverty, unemployment, marital discord, child-rearing, urban dwelling, and low socioeconomic status are potent risk factors for maternal depression.³⁷ Depression, in turn, increases health care use. Among adults, depression results in increased health care use, even in otherwise healthy persons.³⁸ A positive relationship has also been found between the use of maternal and child health services. Children who are high health care users are likely to continue this pattern.³⁹ It is likely that these relationships are reciprocal where environmental stressors and having a child with asthma impact psychological adjustment of inner-city mothers.

There are several reasons why depressive symptoms in mothers may be associated with reports of ED use. Maternal psychological distress diminishes the mother's ability to cope with her child's asthma and is a marker for greater negative impact of the child's illness on the family.⁴⁰ Depressed mothers view even mild-to-moderate asthma symptoms as disruptive and difficult to manage.⁴⁰ They may have more difficulty interpreting cues regarding their child's health status (ie, misinterpreting the severity of their child's symptoms), may underestimate their competency to handle the situation, and/or may feel overwhelmed by the child's immediate needs. As anticipated, we found that ED use was highest in children experiencing the most symptoms. However, among mothers with high levels of depressive symptoms seen in the ED, 58% of the children were classified as having mild-to-moderate asthma morbidity, underscoring the importance of nonasthma factors in the mother's decision to seek urgent care. Finally, though there were no differences in rates of prescribed daily asthma medications, oral steroids, or nebulizers, it is possible that asthma management practices differed between mothers with low and high depressive symptoms.

An alternate explanation for greater ED use by mothers with depressive symptoms is that taking their child to the ED may be an attempt to signal their own distress. Mandl et al³⁴ note that some women may find it less threatening or stigmatizing to contact the medical community on behalf of their children than for their own needs. Indeed, when we examined tertiles of CES-D scores at baseline, mothers with the most depressive symptoms also reported the most frequent ED use.

Our finding that depressive symptoms in the caregiver are prospectively associated with reports of ED use is consistent with recent findings from the National Cooperative Inner-City Asthma Study¹⁹ where the caregiver's mental health had the strongest relationship to health care utilization. However, it stands in contrast with 2 previous cross-sectional studies. Watson and Kemper⁴¹ found that after controlling for maternal and child age, maternal depression was not a significant predictor of ED use in their sample of preschool children in an inner city. However, children in their study were significantly younger (mean age, 12.2 ± 15 months) and were seen in an ambulatory care setting an average of 8.5 ± 5.2 visits per year, more than twice as often as children in our sample were seen for regular care (3.5 ± 4.0 months). Because the ambulatory care center routinely screened for and provided additional case management through a special needs clinic to those identified to be at high psychosocial risk, this may have attenuated ED use by increasing social support and providing alternate resources for caregivers who were identified as being at risk. Riley and colleagues⁷ also found that the child's health and maternal patterns of health care use were powerful predictors of utilization while maternal mental health was not. They studied health care utilization of 450 children who were 5 to 11 years old from electronic records of a health maintenance organization over a 2-year period. However, their sample consisted of predominantly white (84%), 2-parent families (91%) with an average household income of more than $60 000. Thus, it is conceivable that as socioeconomic status changes, so too does the role that depressive symptoms play in the caregiver's decision to seek treatment for their child in an acute care setting.

The National Asthma Education Program guidelines were released in 1991⁴² and revised and updated in 1997⁴³ to provide clinical practice and prescribing guidelines for PCPs. An anticipated outcome of improved asthma management was that emergency facilities would be less burdened by children who could be safely treated during an acute asthma attack at home or in the PCP's office. The traditional view has been that caregivers may bring children to the ED because of poorly controlled asthma resulting from deficits in education, skills, or motivation to address the asthma in other settings. Emerging data suggest that the psychosocial status of the caregiver may be an important but often overlooked variable.¹⁸

This is the first study we are aware of to use a prospective design linking reports of depressive symptoms in mothers of children with asthma with subsequent ED use. Other strengths include the use of well-validated psychosocial measures administered by trained clinical interviewers who were blind to the study hypotheses. A limitation is that the caregiver's mental health status was assessed. Children with psychiatric disorders and behavior problems use health care more frequently and have more ED visits.⁶ Child psychiatric factors, either alone or in combination with asthma symptoms, prompted some of the reports of ED use. However, this is less likely given the findings of Weil and coworkers that the strong positive correlation between the mother's and child's mental health eliminated the direct relationship between child's mental health and hospitalizations.¹⁹ We used maternal reports to assess asthma morbidity and ED use; since both reports originate from a single source, they may reflect a shared reporting bias. Future studies would benefit by using independent validated criteria (eg, medical records) when feasible. Finally, we did not have data on the severity of the asthma attack that would help establish the appropriateness of the decision to go to the ED. Some children may have been symptomatic in the evenings or at night when access to their regular health care provider was much more limited.

Emergency services, particularly in inner cities, are increasingly overburdened and unable to cope adequately with the escalating demands that are placed on them. Merely addressing the immediate medical needs of children with asthma is becoming an increasingly difficult task. Clearly, many factors influence asthma morbidity among children living in inner cities, including the psychological adjustment of the caregiver. Frequent ED use may serve as a marker that nonmedical factors contribute significantly to the decision to take a child to the ED. Furthermore, use of the ED for treating asthma in children is unlikely to change significantly if factors not specific to asthma (eg, mother's mental health) remain unidentified and unaddressed.

More research is warranted to confirm these initial findings as well as to assess asthma management practices in depressed mothers and to determine effective intervention methods. Of great concern is the high prevalence of depressive symptoms reported among caregivers in our sample. It is reasonable to assume that, as in the general population, the majority of mothers with depressive symptoms in our study remained untreated. Maternal depression has been linked with profoundly negative short- and long-term outcomes for the child.³⁷ Because effective treatments for depression exist,⁴⁴ the stress for both parent and child associated with depression may be reduced or even potentially avoided.

Recognizing and treating depression among inner-city caregivers must become a high public health priority. The identification of depressive symptoms among caregivers of children with chronic illness is a critical first step. The use of CES-D or other brief screening instruments could be integrated into the routine primary care of inner-city caregivers with chronically ill young children. Identification of mothers at risk may present a unique opportunity to refer and provide opportunities to enhance psychological adjustment. Improved maternal mental health would not only directly impact quality of life but, as this study suggests, may also reduce unnecessary visits to the ED.

Accepted for publication October 25, 2000.

This work was supported by grants HL5201304 and HL5283303 from the National Heart, Lung, and Blood Institute of the National Institutes of Health, Bethesda, Md.

We thank Karen Huss, PhD, and Perla Vargas, PhD, for their helpful comments on earlier drafts of the manuscript.

Corresponding author and reprints: Susan J. Bartlett, PhD, The Johns Hopkins Asthma and Allergy Center, 5501 Hopkins Bayview Cir, Suite 4B.32, Baltimore, MD 21224 (e-mail: bartlett@jhmi.edu).