Home Medication Readiness for Preschool Children With Asthma

This study is a cross-sectional analysis of baseline asthma medication readiness among the caregivers of preschool-aged children at the time of enrollment in the trial Asthma Basic Care Education in Head Start. Participants were recruited from Baltimore City Head Start programs from April 2011 to November 2016. Eligible caregivers had to be the parent or legal guardian of a child aged 2 to 6 years, who reported that a physician diagnosed their child with asthma, and who spoke English. Study staff contacted eligible families who gave permission on screening forms completed at Head Start to confirm eligibility and schedule a home visit. During the home visit, research assistants (RAs) obtained written informed consent from the child’s primary caregiver and conducted the baseline assessment. The Johns Hopkins School of Medicine Institutional Review Board reviewed and approved this study.

At a 2-hour home visit to the primary home of each child enrolled in the study, the RA conducted a structured interview with the child’s caregiver. RAs asked caregivers to report all asthma medications the child had been prescribed and to assemble all of the child’s asthma medications and devices that were currently in the home. RAs visually inspected the medications to assess the 5 readiness criteria for each reported medication. Caregivers also responded to questions on medication beliefs, controller medication adherence, and level of asthma control.¹⁸ 

The medication readiness index includes 5 basic elements of medication availability and knowledge that we considered usual and necessary prerequisites to appropriate use of rescue medications and adherence to controller medications. The 3 availability criteria included whether the caregivers could locate each reported medication, including the nebulizer machine for nebulizer medications (criterion 1), whether each medication was not expired (criterion 2), and whether each medication with a counter had remaining doses (criterion 3). The 2 knowledge criteria included whether the caregiver could correctly identify whether the medication was a rescue or controller medication (criterion 4) and could state any dosing instructions for the medication (criterion 5). The index was developed in consultation with a pediatrician, pediatric pulmonologist, pharmacist, and behavioral psychologist. Survey questions to measure items in the index were reviewed and revised to ensure that they objectively captured the essential components of medication availability and management that are prerequisites for adherence (Supplemental Table 5). We calculated dichotomous measures of “medication readiness,” defined as meeting all 5 criteria in the medication readiness index. Medical records were not available for confirming doctor-prescribed medications and dosing instructions.

Caregiver beliefs about medications were measured by using a 21-item scale, the Pediatric Asthma Medication Beliefs Assessment (PAMBA), which was previously validated in a similar minority urban population.¹⁹ Caregivers rated their agreement with each statement about asthma controller medications on a 5-point Likert scale from “strongly agree” to “strongly disagree” (see Supplemental Table 6 for agreement with individual statements). The scale includes 3 subscales: symptom-based outcome expectancy (8 items), self-efficacy (7 items), and worries and concerns (6 items). Responses to negative statements were scored according to a 1-to-5 Likert scale, with “strongly agree” scored as 1 and “strongly disagree” scored as 5, whereas responses to positive statements were scored in reverse. Item scores were summed to create a total PAMBA score, with a maximum possible score of 105, as well as subscale scores for symptom-based outcome expectancy (maximum of 40), self-efficacy (maximum of 35), and worries and concerns (maximum of 30). Higher PAMBA and subscale scores indicate more positive beliefs about asthma medications.

For each controller medication reported, caregivers answered structured questions about how frequently the child took the controller medication in the past 2 weeks: “not at all,” “a few days (1–3 days),” “several days (4–7 days),” “most days (8–11 days),” “almost every single day (12–13 days),” or “every single day (14 days).” This measure is adapted from other single-item self-reported adherence measures that have been previously validated.^20,21 Caregivers who responded “almost every single day” or “every single day” for any controller were considered to have self-reported adherence in analysis, consistent with the common definition of adherence as taking 80% or more of prescribed doses.²² 

Descriptive statistics for the medication readiness index criteria, self-reported adherence, and health beliefs were calculated for all children whose caregivers completed the medication readiness index assessment. Logistic regression was used to assess the relationship between medication readiness and PAMBA scores. Possible confounders for this relationship were also tested, including caregiver demographics and asthma control, but excluded from the presented models because they were not significant in bivariate analyses. Cross tabulations and the χ² test were used to assess associations between self-reported adherence and medication presence in the home. Two-sample tests of proportions were used to assess for differences between medication type (rescue or controller) and medication readiness index criteria. All analyses were conducted in Stata version 13 (Stata Corp, College Station, TX).

A total of 288 caregivers of Baltimore City Head Start children with asthma completed the baseline surveys including medication readiness index assessments. The caregivers most commonly completing the survey were the birth mother (85%) or birth father of the child (7%; Table 1). Close to 40% of caregivers had not completed high school, 29% reported a high school diploma or general equivalency diploma as their highest educational attainment, and 33% had attended or graduated from college. The mean age of enrolled children at the time of the assessment was 4.2 years (SD = 0.7). The sample is predominantly African American (92%) and covered by state insurance (94%). Sixty percent of the children were boys, and 61% of the children had uncontrolled asthma.

Caregivers reported at least 1 rescue medication or controller medication for 277 (96%) and 161 (56%) children, respectively. The rescue medication most commonly reported was an albuterol metered-dose inhaler (MDI; n = 255, 89%); 58% were prescribed both an albuterol MDI and an albuterol nebulizer (Table 2). The most commonly reported controller medication was fluticasone MDI (n = 120, 42%). Caregivers reported 2 or more controller medications for 39 children (14%) and 2 rescue medications (albuterol MDI and nebulizer) for 168 children (58%).

Among children whose caregivers reported a rescue medication, only 60% had a rescue medication that met all 5 readiness index criteria (Fig 1). Medication readiness was lower for children whose caregivers reported a controller medication, with only 49% meeting all 5 readiness index criteria. Although the proportion of caregivers who were able to locate a rescue medication in the home was nearly identical to that of those who were able to locate a controller medication (78%), caregivers were less likely to meet the knowledge criteria for controller medications, which includes stating any dosing instructions (81% for controller compared with 92% for rescue; P < .01) and identifying the type of medication (controller or rescue) correctly (78% compared with 94%; P < .01).

Among the 2 rescue medications, only 52% of children with albuterol nebulizer prescriptions and 45% of children with albuterol MDIs met all 5 readiness criteria for the respective medication (Table 2). For both albuterol MDI and albuterol nebulizer prescriptions, the criteria that was most likely to be met was caregiver identification of the medication as a rescue medication (93% and 92%, respectively), and the criteria least likely to be met was that the medication had not passed the expiration date (55% and 63%, respectively).

Among children with a controller medication, the proportion whose caregiver met all 5 medication readiness criteria was near or below half for all but 1 medication. For the most commonly reported controller medication, fluticasone MDI, only 44% of children met all 5 readiness criteria. Medication readiness was similar for budesonide nebulizers (48%) and montelukast (51%).

Some medication management patterns are apparent by medication type and mode of delivery. More caregivers correctly identified the purpose of rescue medications than control medications (92% compared with ∼80%; P < .01). For each of the 3 MDI rescue or controller medications, one-third of children actually had an empty canister in the home. Spacers, which are required for correct administration of MDIs, were present for 133 of the 187 children (71%) with any MDI medication in the home. Among the controller medications, the only oral medication, montelukast, was the least likely to be located in the home (68%).

Of the 161 children whose caregivers reported a controller medication, 153 also responded to questions measuring self-reported adherence. Slightly more than half of these caregivers (54%) reported that the child was adherent by taking the medication every day or nearly every day (Table 3). Although caregivers who were able to locate a controller medication in the home were statistically significantly more likely to report that the child was adherent (62% compared with 25%; P < .01), it is worth noting that caregivers for 8 (10%) of the 83 children reported to be adherent could not locate a controller medication in the home. Among the 32 children whose caregivers reported a controller medication but were unable to physically locate one in the home at the time of the interview, 16 (50%) reported the child took a controller medication “most days” or more frequently.

The mean overall PAMBA score was 79.9 (SD: 9.3). Subscale means and Cronbach’s α are included in Supplemental Table 7. The association between caregivers’ medication beliefs as measured according to the PAMBA and meeting all 5 medication readiness criteria was tested for the overall PAMBA score and the 3 subscales (Table 4). Higher scores on the PAMBA, indicating more positive beliefs about asthma medications, were associated with increased odds of meeting all 5 medication readiness criteria for controller medications (crude odds ratio: 1.04; P < .05) but not rescue medications. A 1-point increase in the scale of worries and concerns, indicating the caregiver had fewer worries and concerns about asthma medication, was associated with 1.19 times the odds of controller medication availability in adjusted models (P < .05).

When separate models were compared for the 3 availability criteria and the 2 knowledge criteria, the association between worries and concerns was significant for meeting the knowledge criteria for controller medications (adjusted odds ratio [aOR]: 1.13; P < .05) but not the availability criteria. The subscales for symptom-based outcome expectancy and self-efficacy had no significant association with meeting the medication readiness criteria. High self-reported adherence to controller medications (every single day or almost every single day) significantly increased the odds of meeting all 5 medication readiness criteria for controller medications (aOR: 4.59, P < .01).

Adherence to medication regimens for pediatric asthma control requires caregivers to complete a series of steps that precede actually taking the medication, including filling the prescription, identifying and understanding how to take the medication, organizing storage of the medication, and monitoring medication expiration dates and counter status.²³ Among the population of low-income, predominantly African American households with preschool-aged children with asthma in this study, we observed major gaps in medication readiness, including low knowledge about medications and unavailability of medications. Gaps in the 3 availability criteria were similar for both controller and rescue medications, whereas more caregivers met the 2 knowledge criteria for rescue medications than controller medications. Overall, less than half of children reporting controller medications are ready for daily adherence, and only 60% reporting rescue medications are ready to respond to an asthma exacerbation, by meeting all 5 medication readiness criteria for each class of medication. These findings suggest that poor medication readiness should be considered as an important and potentially modifiable contributor to home-based asthma management. The urban low-income population in this study experiences significantly higher rates of emergency department use and deaths because of asthma exacerbations,¹⁵ and improving home medication readiness may help to improve these outcomes.

In this study, we objectively documented suboptimal home medication management among caregivers of young children with asthma and identified important gaps in medication availability and caregiver knowledge. We extend previous research using self-reported adherence^8,10 and prescription filling patterns^12,–14 to document suboptimal adherence by identifying specific gaps in home medication management that prevent proper adherence. In 1 previous study, asthma medication availability in the home before any intervention was reported, but expiration dates and counter status were not assessed. The researchers reported physical presence of medications among a cohort of Hispanic children between the ages of 5 and 18 in Chicago with uncontrolled asthma.⁸ Two-thirds of children in the Chicago sample (74%) had a rescue medication in the home, whereas only half (49%) had any controller.⁸ Although the composition of our sample is younger, predominantly African American, and 40% of the children had controlled asthma, a similar proportion of households had rescue medications and controller medications physically present in the home (75% and 44%, respectively). Large gaps were also observed in the caregivers’ ability to correctly identify controller medications (32%) but not rescue medications (84%) in the Chicago study.⁸ Another study at the end of a trial to improve asthma management among school-aged children in Rochester, New York, revealed that only 46% of caregivers could accurately recall the name of the asthma medications present in the home but did not verify if the medication was in the home.²⁴ Our results reveal that empty canisters and expired medications also contribute substantially to reducing medication readiness and should be routinely assessed in both research and clinical settings.

The association between a caregiver’s reported worries and concerns about asthma medications and their level of controller medication knowledge and overall medication readiness in our study is similar to associations between beliefs and adherence reported in other studies among different populations of adults^25,–28 and children with asthma.^13,17,29 Researchers for previous studies have reported an association between ICS necessity beliefs, concerns about ICS, and medication refills.²⁷ Negative health beliefs about asthma medications may be a particular concern for minority populations in the United States. Research by Le et al²⁶ among adults with asthma suggests that higher rates of negative health beliefs among minority populations mediates the relationship between minority race and lower adherence. Our findings extend our understanding about the role of medication beliefs and adherence beyond medication taking behavior by demonstrating that worries and concerns about controller medications predict medication readiness.

Our findings about limited medication availability in the home, as well as the association between health beliefs and medication knowledge, indicate an important intervention point in medication adherence.²³ The caregivers in this study reported low educational attainment, which likely contributes to low medication knowledge scores. Another factor that may be contributing to both negative medication beliefs and knowledge gaps is inadequate or ineffective medication counseling from health providers. Although routine asthma care, such as having a nonurgent asthma visit and an asthma action plan, have been associated with improved adherence,³⁰ the majority of surveyed pediatric patients with asthma report that they not are receiving guideline-based care, such as counseling on proper preventive medication use and developing asthma action plans.²⁴ Interventions to improve asthma education can effectively increase adherence and improve outcomes.³¹ Our results suggest that such interventions, particularly when delivered to urban minority populations, should specifically address worries and concerns about medications as well as barriers to maintaining medications in the home. Providing appropriate education will be an important component of efforts to reduce asthma disparities.^9,32 

These results suggest that including home asthma medication readiness in adherence interventions may increase their ability to reduce asthma-related morbidity and mortality. A child whose medication is expired or empty, or whose caregiver cannot locate the medication in the home, is by definition not able to adhere to their controller medication regimen, putting them at greater risk for asthma morbidity and mortality.^6,7 Barriers to filling prescriptions for caregivers of low-income urban children with asthma may include gaps in insurance coverage, difficulties with previous authorization requirements, and competing time demands for meeting other basic household needs.³³ Among children in preschool programs that require a new rescue medication for enrollment, such as Head Start, insurance reimbursement policies that limit refills may require caregivers to go without a medication at home until a new refill is allowed.³⁴ Many families in our sample also divide caregiving responsibilities between multiple caregivers and/or households, and medications may not be present in every home where children spend significant time.³⁴ 

A major strength of this study is the in-home objective measurement of rescue and controller medication availability. The majority of studies in which pediatric medication adherence is assessed rely on self-reported measures or pharmacy fill data, which provide an incomplete picture of the actual medication availability required for children to be ready to adhere to their regimens. However, the cross-sectional nature of this study did not allow us to examine the effects of poor medication readiness on asthma control and health care utilization. Additional research is needed to examine this relationship as well as to understand the factors contributing to poor medication readiness. Furthermore, we did not confirm the parent-reported regimen against medical records.

A striking proportion of low-income urban children with asthma do not have medications readily available in the home. This gap leaves children unable to appropriately adhere to their treatment plan or respond to an exacerbation with a rescue medication, which previous research has revealed increases the risk for asthma morbidity and mortality. Assessment of medication readiness should be incorporated into the care of children with chronic conditions, and interventions are needed to improve medication management.

aOR

adjusted odds ratio

ICS

inhaled corticosteroid

MDI

metered-dose inhaler

PAMBA

Pediatric Asthma Medication Beliefs Assessment

RA

research assistant