Objective. The purpose of the study was to examine medication use reported by families participating in an urban school-based community intervention program and to relate this use to other social and medical variables.
Design. The design of the study was a cross-sectional questionnaire survey.
Setting. Patients and their families recruited from elementary schools in a community setting were interviewed between December 1991 and January 1992.
Participants. A total of 508 children with asthma were identified by school health records and teacher surveys. Their families confirmed the diagnosis and agreed to enter the study. Questionnaires were completed by 392 families.
Intervention. The 392 families participated in a controlled trial of asthma education after providing the data that are the basis of this report.
Results. More than half of the children took two or more medications for asthma. Thirty-one percent took theophylline alone or in combination with an adrenergic agent; 11% took some form of daily antiinflammatory medication, either cromolyn (8%) or inhaled steroids (3%). The pattern of medication use related to measures of severity and to regular visits to physicians or nurses. In general, however, children were undermedicated. A total of 78 children (20%) reported no medication or over-the-counter medication use, although 37% reported asthma severe enough to be associated with ≥20 days of school missed per month, and 37% had had an emergency room visit for asthma in the past 6 months. More than half of children ≥9 years old supervised their own medication.
Conclusions. We concluded that undermedication is common in poor children with asthma living in urban areas. Antiinflammatory medications are used less commonly than in the general population, and theophylline is used more often. School children may be likely to supervise their own medication.
- ER =
- emergency room •
- OTC =
Medication recommendations for the treatment of chronic asthma have changed a great deal in the past decade. Currently, national and international guidelines for asthma therapy1 2 include daily antiinflammatory medications such as cromolyn, nedocromil, and inhaled steroids for all patients with moderate to severe asthma, as well as for any patient who requires daily treatment with β-adrenergic agonist for asthma symptoms. As a result of these recommendations, the use of antiinflammatory medications has increased sharply, so that by 1993, 22% of asthma medications prescribed by US pediatricians were antiinflammatory agents.3 One of the strongest pressures to institute these recommendations was the epidemiologic evidence that began to accumulate, suggesting that increasing dependence on bronchodilator therapy, especially β-adrenergic agents, was contributing to the worldwide increase in deaths from asthma.4
In the United states, asthma morbidity and mortality has increased disproportionately among poor, minority children living in the inner city.5-7 One proposed explanation for this difference is that asthma is treated as an episodic disease in the inner city; thus, a large proportion of care is given in the emergency room (ER),8 and symptomatic treatment is emphasized in preference to prevention.8 9 Bronchodilator use is higher in Medicaid populations of all ages,10 and data from hospital discharges11 12 suggest that inappropriate medication for asthma might contribute to increased morbidity and mortality among inner-city children. There is little information available regarding asthma medication use in ambulatory children in the inner city. As a part a clinical trial of educational intervention in inner-city school children with asthma, we have collected information on medication use, education, health care use, access to care, and attitudes in a large cohort in Baltimore, MD, and Washington, DC. The present report describes the analysis of this data regarding the patterns of asthma medication use and the factors that affect these patterns.
The information included in this report was derived from questionnaires administered to families enrolled in a study testing the effectiveness of a school-based asthma education and community-based asthma health worker intervention in Baltimore, MD, and Washington, DC. Children were recruited from 42 elementary schools, each with a total enrollment of >350 students, of which >85% were African-American, having school administrators willing to participate in the study.
Initially, 1559 children were identified from school health records as having asthma. A short survey was sent home with the children, and 508 families confirmed that their children had asthma and agreed to participate in the study. After obtaining consent from parents or legal guardians, a telephone interview was conducted from December 1991 to January 1992 by experienced interviewers (Westat Inc, Washington, DC). Families without telephones or those who could not be contacted were mailed a card with a toll-free number for the survey company and offered a $20 incentive to complete the interview. The project was approved by the Joint Council on Clinical Investigation for the Johns Hopkins University.
We developed a 170-item baseline questionnaire that included questions regarding asthma diagnosis, source of regular care for asthma symptoms, school absences, ER use, source of regular medical care, medication type and frequency of use, child's responsibility, activity restrictions, and social, demographic, and behavioral factors.
Questionnaires were completed by 420 families (83%). The final sample included those 392 children who were reported by their caretakers to have active asthma that had been confirmed by a physician.
Classification of Medication
From the questionnaire responses, we created the following four medication categories (see Tables 3 and 4): 1) Patients reporting no medication use were combined with those taking over-the-counter (OTC) medications in a category called none/OTC, because these children's regimen would be the least supervised by the medical system. 2) β-Adrenergic agonist use only, called beta adrenergic, was examined, because this was considered to be symptomatic therapy only. 3) Theophylline alone or in combination with other medications, called theophylline, was separated, both because it was more likely to be used on a regular schedule and has been suggested to have some antiinflammatory properties.3 13 4) Cromolyn and corticosteroid, calledantiinflammatory, usually were used in combination with other medications. Furthermore, we determined that children would be assigned to only one category for each variable that we analyzed; ie, a child using cromolyn was assigned to that category even if adrenergics also were used.
Categoric relationships were examined with χ2analysis. The interaction between potentially related variables was assessed using logistic regression. The analyses were performed using SAS 6.11 (SAS Institute, Cary, NC).
Of the children recruited, 200 were from Washington, DC, and 192 from Baltimore, MD. Mean age was 8.3 ± 2.0 years. Females totaled 230 (59%) and African-Americans 380 (97%). Of the families participating, 78% were headed by single women; 31% of the caretakers had not completed high school, and only 43% were employed full-time. Health care was paid by Medicaid (44%), private medical insurance (32%), and by the family (10%); 14% were enrolled in a health maintenance organization.
As summarized in Table 1, 80 children (21%) were symptomatic ≥20 days each month, and >60% reported symptoms 1 to 2 days a week. Almost half of the children missed ≥6 days of school in the past year. A total of 171 (44%) had visited an ER for asthma, and 7% had been hospitalized in the past 6 months. Fifty-one children (13%) took no medication for asthma, and 107 (27%) took three or more medications.
Patterns of Medication Use
As shown in Table 2, a total of 702 medications were taken by the 392 children in the study; 51 reported taking no medications. The most common medication was a β-adrenergic agonists (54%), followed by theophylline (22%). Antiinflammatory medications such as cromolyn or inhaled steroids were used by 5.3%.
To examine the individual pattern of the children's medication use, the 392 children were grouped into the exclusive categories shown in Fig 1. Of the children, 78 (20%) took either no medication or OTC medications, including 7 who took Primatene Mist. Prescription adrenergic agonists were taken by 221 children (56%), either alone (34%) or with theophylline (22%). A total of 121 children (31%) took theophylline, but only 34 (9%) used theophylline alone. Twenty-nine children took cromolyn, and 13 took inhaled corticosteroids. Of the 17 children who were categorized as taking oral steroids, 6 took oral corticosteroids and β-adrenergic agonist only. These 6 children were not included in further analysis of factors affecting medication use, because it was not possible to distinguish those who took oral steroids regularly from those who took them only intermittently for acute asthma.
Correlates of Medication Use
The pattern of medication use correlated with several measures of asthma severity. As seen in Table 3, children taking a greater number of medications took more antiinflammatory medications (P < .0001). Daily symptoms were not related as strongly to the medication pattern, but interference with usual activity by asthma was related. For example, the greater number of school days a child missed, the less likely he or she was to be untreated or treated only with β-adrenergic agents (P < .0001). ER use was not correlated with medications, but children who had been hospitalized in the past 6 months reported using more theophylline and antiinflammatory medications.
It also should be noted in Table 3 that at every measure of severity, a substantial minority of children were undermedicated. For example, of the 166 children reporting an ER visit in the past 6 months, >50% were untreated or took β-adrenergic agents only, as were 45% of children reporting ≥20 symptomatic days or nights each month.
Difficulties with access to care did not appear to be related to the pattern of medication use. As shown in Table4, 19% of those 276 children whose health plan paid for all medications still took no medications. The proportion of persons taking antiinflammatory medications also was similar among children who received medications through their health plan (12%), compared with those who did not receive free medications (9%). Only 70 persons reported that they had difficulties paying for medications, and their pattern of medication use was nearly identical to the remainder. Persons with private insurance took slightly fewer OTC and antiinflammatory medications, but the overall pattern was not significant (data not shown).
A major factor associated with increased use of antiinflammatory medications was an ongoing relationship with a physician. As seen in Table 4, 287 children reported that they saw a single physician for their asthma care, and this group was significantly more likely to use theophylline and antiinflammatory medications than were patients who received care from a clinic or through the ER (P< .001).
The pattern of medication use also related to parental attitudes (P < .001). Families who answered yes to the question “Do you think medications should be taken regularly?” were more likely to give antiinflammatory medications than those who thought that medications should not be used regularly (20% vs 4%). These families also used OTC or no medications slightly less frequently (19% vs 21%) and β-adrenergic drugs less frequently (30% vs 37%).
From Table 3, it appears that the variables that are related most strongly to the pattern of medications were the number of asthma medications taken by each child and the number of school days missed. Because all three of these variables might be independent measures of asthma severity, the relationship among them was examined by multivariate analysis, using the numbers of school days missed as a dependent variable. Both the pattern of medications (P = .008) and the number of medications (P = .0001) were related to number of school days missed when examined individually. However, the relationship between number of school days missed and the pattern of medications was no longer significant when adjusted for the number of medications. This suggests that the number of medications used was the stronger variable, more closely relating to asthma severity.
A total of 210 children (53%) supervised their own medication. As seen in Fig 2, the proportion of children assuming responsibility for their own medication was strongly age-dependent (P < .001). None of the 4-year-old children were given responsibility for their own medication; 7% and 17% of 5- and 6-year-old children, respectively, were given this responsibility. By age 9 years, >50% of children assumed responsibility for their own medication.
The number of asthma medications was related strongly to self-medication (P < .001). Among those 107 children taking three or more medications, 63 (59%) supervised their own medication, whereas only 28 of 114 children (25%) taking one medication did so. Other indicators of more severe asthma, such as the number of symptomatic days and nights, the number of school days missed, ER use, and hospitalization, were not related.
Being allowed to supervise one's own medication was associated with a higher frequency of antiinflammatory medication use (22% vs 6%) and less OTC, β-adrenergic, or theophylline use (P< .001); however, this relationship disappeared when the data are adjusted for number of medications used. The frequency of self-medication was not affected by number of ER visits, hospitalizations, source or frequency of medical care, or social support. Self-medication was more likely if the caretaker worked outside the home (49% vs 34%; P = .004) and if no one else in the home took asthma medications (47% vs 36%;P = .038).
These data confirm and expand earlier reports11 12 of inappropriate asthma medication use among inner-city children with asthma. The pattern of medication use was surprising in that many children took no medication and 31% took theophylline, either alone or in combination with β-adrenergic agents. To our knowledge, this is the first report that provides data from ambulatory families with asthma; thus, there is no other study with which to compare the frequency of untreated asthma. Recent data regarding prescription patterns among general pediatric populations3 suggest that the use of theophylline reported here is unusually high. We found that children with more severe asthma tended to use more antiinflammatory medications, but that access to medical care or the ability to pay for medication did not influence the pattern of medications used. Finally, we showed that these medications are taken without supervision at a surprisingly young age. Self-medication by the children was increased as more medications were used and as the number of physician visits increased, and may reflect the effects of asthma self-management education.
The use of antiinflammatory medications by these children is less than half of that reported in the general pediatric population in 1993.3 Current recommendations from the National Heart, Lung, and Blood Institute expert panel suggest that antiinflammatory medications should be prescribed for all patients with asthma except those whose disease is mild (either intermittent or persistent).1 2 The pattern of medications that the children in our study used was quite different than this recommendation and confirms earlier reports of children discharged from Boston hospitals.11 12 Of most concern is the substantial number of persons who took as needed (prn) medications or none at all despite nearly daily symptoms, frequent school loss, and ER visits for asthma. In addition, >80% of those who did take regular medications did not use antiinflammatory drugs. It is not clear whether this is because of recommendations by their physician or because of the family's own perception of effective medication. However, the fact that a larger proportion of those who had a regular physician took regular medications and included antiinflammatory medications among them suggests that adherence and self-medication is the more likely explanation.
The children who took no medication or OTC medications are clearly undertreated for their asthma. Among the 78 patients who took no medication for their asthma or took only OTC medication, 33% reported 8 to 14 days of symptoms in the past month, 8% missed >20 days of school in the past year, 37% reported at least one ER visit in the past 6 months, and 7% had been hospitalized in the past year. By any criteria, they were not asymptomatic and would have been better treated with regular antiinflammatory medication. Furthermore, our data suggest that the undermedication was not related to financial barriers, because 19% of patients whose health plan paid for all medications still took OTC medications or none. The most important factor determining whether these children took medication for their asthma, especially antiinflammatory medications, was an ongoing relationship with a single health care provider, either physician or nurse.
We present these data with two caveats. First, the data are derived from questionnaires and as such present a less reliable picture of medication use than do recent reports that have been derived from pharmacy or hospital records.3 11 At the same time, these data provide the first indication that a substantial number of school children with asthma receive no prescription or OTC medications despite reporting frequent symptoms, school loss, ER use, and interference with parent work activity during the day and night. These untreated patients contribute in an important way to the pattern of increased health care use, because they are commonly symptomatic, and they visit the ER for acute asthma as commonly as other children who use prescription medications for their asthma. The second caveat is that we recruited a minority of the asthmatic children in the schools. This was inevitable, because our primary goal was to conduct a clinical trial of behavioral interventions; thus, we chose the most efficient recruitment strategy rather than one that would yield a representative population sample. We can speculate that subjects' willingness to participate relates in some way to asthma severity, family organization, and motivation, but we have no data to estimate these differences.
The pattern of medications among those requiring prescriptions differs markedly from that seen in the general population. The major difference is that theophylline is used more commonly, in 39% compared with 7% of prescriptions prescribed by pediatricians for all patients in 1993 among 5- to 9-year-olds.3 In a paper describing the Michigan Medicaid prescriptions at that time, Gerstman and colleagues10 reported that theophylline use in 1980 to 1986 was similar to national prescribing statistics.13 Our data suggest that this drug may be preferred by physicians in Baltimore, MD, and Washington, DC, who treat inner-city populations.
We found that a surprisingly large proportion of these school children supervised their own medication. Self-medication was more common among children with more severe asthma and among those with more visits to a health care provider. From our data, we cannot speculate whether this related to more frequent symptoms or whether physicians encouraged increased self-management behavior. We also cannot determine how completely this responsibility was assumed by the child; it would seem appropriate if children were given responsibility for taking regular medications but would not be appropriate if all medication use were unsupervised. We found that self-management increased when the parent worked outside of the home, suggesting that working mothers may have to delegate most of the responsibility to the a child at an earlier age. Because there are data showing that children and adolescents with diabetes mellitus who are given greater responsibility for their self-management demonstrated poorer metabolic control,14 future studies should examine this behavior in inner-city families. At present, we would urge all physicians caring for inner-city children with asthma to ask specifically about responsibilities of the child and advise parents when these seem inappropriate.
In summary, these data demonstrate that inner- city children frequently are undermedicated, that antiinflammatory medications generally are underused, and that self-medication is common among elementary school children. We would recommend that physicians caring for inner-city children increase case findings of asthma, and that they inquire about daily symptoms in any child with asthma. In addition to increasing their use of antiinflammatory medication, physicians should be aware that instructions regarding medication use by children <10 years of age needs to be directed not only to the parent but also to the child.
This work was supported by National Institutes of Health Grants HL45312 and AI30773 and by the Hospital for Consumptives of Maryland (Eudowood Fund).
- Received March 3, 1997.
- Accepted August 20, 1997.
Reprint requests to (P.A.E.) Department of Pediatrics, Johns Hopkins Hospital, 600 N Wolfe St, Baltimore, MD 21287.
- ↵National Asthma Education Program Expert Panel Report. Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: US Department of Health and Human Services; 1991. DHHS publication no. 91-3042
- ↵National Asthma Education Program: Expert Panel Report II. Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: US Department of Health and Human Services. 1997. Draft
- Halfon N,
- Newachek PW
- ↵Gerstman BB, Bosco LA, Tomita DK, Gross TP, Shaw MM. Prevalence and treatment of asthma in Michigan Medicaid patient population younger than 45 years, 1980–86. J Allergy Clin Immunol. 1989;1032–1039
- Homer CJ,
- Szilagyi P,
- Rodewald L,
- et al.
- Finkelstein JA,
- Brown RW,
- Schneider LC,
- et al.
- ↵La Greca AM. Children with diabetes and their families: coping and disease management. In: Field TM, McCabe PM, Schneiderman N, eds. Stress and Coping Across Development. Hillsdale, NJ: Erlbaum; 1988:139–159
- Copyright © 1998 American Academy of Pediatrics