Thirty-four 6- to 11-year-old children with persistent asthma and one of their parents participated in this cross-sectional study. The 6- to 11-year age range was selected (a) to include children with sufficient literacy levels to participate and (b) to represent the age range of children with the highest prevalence of asthma. The sample was recruited between February and July 2015 from the Arizona Respiratory Center research database of children with asthma who have expressed interest in participating in asthma research. Criteria for inclusion were English-speaking children ages 6 to 11 years at the time of data collection who were prescribed a daily asthma controller medication; the parent/caregiver needed to be able to speak and read English and reside with the child. Exclusion criteria for children included parent report of developmental delay or traumatic brain injury, attention deficit hyperactivity disorder, obstructive sleep apnea, or a recent (within 30 days)/active asthma exacerbation (defined as requiring oral corticosteroids). Children were also excluded if they had participated in an asthma medication adherence intervention study in the past 30 days.
Parents and school-aged children with persistent asthma were recruited from the (blinded for review) Respiratory Center research database. Potential participants were contacted by research personnel using a script to describe the study and review eligibility criteria. Those interested in participating were referred to the principal investigator, who followed up with the families by phone to discuss the study, verify inclusion and exclusion criteria, and schedule a data collection session. Data collection took place during a one-time session at the University of Arizona College of Nursing. After obtaining informed consent (parents) and minor assent (children), parents were given paper-and-pencil study instruments to complete independently in a waiting room. In an adjacent conference room, study personnel read all instruments and questionnaires to the child participants and assisted with completion. Sessions lasted approximately 1.5 to 2.0 hours. After completing all study materials, each parent–child dyad was given one $50 Visa gift card. The University of Arizona Institutional Review Board and Human Subjects Committee approved all procedures.
Parents and children each completed questionnaires with measures described below. The parents also completed a questionnaire on demographic information and child asthma history.
Childhood Asthma Control Test
Childhood Asthma Control Test (C-ACT; Liu et al., 2010) is a validated seven-item parent and child report tool for classifying asthma control in children ages 4 to 11 years. There are four child-reported items and three caregiver-reported items. Scores range from 0 to 27, with higher scores indicating better asthma control. Cut-points have been validated and align with National Heart, Lung, and Blood Institute (NHLBI) asthma guidelines (Liu et al., 2010). Scores of 20 or greater indicate well-controlled, 13 to 19 indicate not well controlled, and 12 or lower indicate very poorly controlled asthma.
Brief Illness Perception Questionnaire
The Brief Illness Perception Questionnaire (BIPQ) is a nine-item self-report questionnaire that assesses cognitive and emotional illness representations (Broadbent, Petrie, Main, & Weinman, 2006). BIPQ Items 1 through 5 were used in this study, those representing symptom/identity, timeline, consequences, personal control, and treatment control. Items are rated on a 10-point response scale, with each item reflecting one illness representation domain. Items are individually scored, with higher scores indicating a more negative effect of the illness representation domain on the individual's life. Several modifications were made to the BIPQ. For clarity, asthma was substituted for your illness. The literacy level for the original BIPQ scale was fourth grade; therefore, simplified language substitutions were also made to the BIPQ to lower the literacy level for child subjects. Modifications include substituting how long your asthma will last instead of how long your asthma will continue, medicine versus treatment, have symptoms versus experience symptoms, and how much does asthma affect your feelings versus your emotions. The modified BIPQ had a second grade literacy level. For parents, the modified BIPQ was used with only language modifications for proxy versus self-reporting (e.g., your child substituted for you). The BIPQ has been validated for use in adults (Broadbent et al., 2006). To our knowledge, this is the first time the BIPQ has been used in children.
Beliefs About Medicines Questionnaire
The Beliefs About Medicines Questionnaire (BMQ)–Specific is a 10-item self-report questionnaire that assesses individual perceived medication necessity and medication concerns (Horne, Weinman, & Hankins, 1999). Using a 5-point response scale (1 = strongly disagree, 5 = strongly agree), five items assess perceived medication necessity and five items assessing medication concerns. Scores for Items 1 through 5 are summed to reflect perceived medication necessity, with higher scores indicating a higher perceived necessity. Scores for Items 6 through 10 are summed to reflect concern about medications, with higher scores indicating higher concerns. The Necessity–Concern differential is calculated by subtracting the concern score from the necessity score; a negative score indicates that concerns outweigh necessity beliefs, whereas a positive score indicates that necessity is higher than concern (Horne et al., 1999).
The BMQ-Specific has been validated in adults (Horne et al., 1999), as well as in parents and children down to age 7 years (Yilmaz, Eroglu, Ozalp, & Yuksel, 2012). Despite prior use of the BMQ in children down to age 7 years, the literacy level of the original scale was at the sixth grade level. Therefore, similar modifications were made to the BMQ to lower the literacy level to ensure child understanding and participation. Substitutions included my life would be hard instead of my life would be impossible; worry about staying on my medicines for a long time instead of worry about the long-term effects of my medicines; worry about always needing my medicines instead of worry about becoming dependent on my medicines. The modified BMQ had a literacy level of second grade. The BMQ-Specific (Parent) is based on the BMQ-Specific (Child), but replaces language to reflect proxy versus self-reporting.
Medication Adherence Reporting Scale for Asthma
The Medication Adherence Report Scale for Asthma (MARS-A) is a 10-item self-report questionnaire for assessing intentional reduction or avoidance of asthma controller medications (Cohen et al., 2009). The MARS-A is not an objective measure of medication adherence. Rather, the MARS-A reflects individual beliefs and behaviors related to controller medications, such as I only use [my controller medication] when I need it and I take it less than instructed. The MARS-A uses a 1-to-5–point response scale, with 1 indicating always to 5 indicating never. Scores are averaged, with high adherence defined as a MARS-A score of 4.5 or higher. The MARS-A is trademarked and modifications are not permitted. The literacy level, however, was at second grade. The MARS-A is validated for use in adults (Cohen et al., 2009), but to our knowledge it had not been validated for parent proxy or child self-reporting at the time of this study.
Data analysis used the Statistical Package for the Social Sciences (SPSS) software version 23 (IBM Corp., 2015). All variables used in parametric testing were examined for assumptions. Variables were transformed if their absolute value of distribution skewness was greater than 1. Variables with positive skew were transformed using square root transformation. Variables with negative skew were reflected, then transformed using square root, and then back-reflected. Transformed variables with an absolute value of skewness still greater than 1 were then transformed using log10 transformation. The resulting variables met assumptions of parametric testing. A p value of less than .05 was used as the criterion for statistical significance.
Descriptive statistics (mean and standard deviation) were used to describe the parent/caregiver and child demographic information, asthma status (C-ACT), asthma illness representations (BIPQ and BMQ), and reported medication adherence (MARS-A). Internal consistency was assessed using Cronbach's α for the C-ACT, BMQ, and MARS-A parent and child scales. Intraclass correlations (ICCs) were used to examine correlations between parent/caregiver and child asthma illness representation scores.
Hierarchical regression analyses were used to assess the relationship between controller medication adherence and asthma illness representations. When parent-reported medication adherence (MARS-A) was used as the dependent variable, the parent illness representation variable was entered on the first step, and the corresponding child illness representation variable was entered on the second step. Conversely, when the child-reported MARS-A was used as the dependent variable, the child illness representation variable was entered on the first step, and the corresponding parent illness representation variable was entered on the second step. This analysis is consistent with the Actor–Partner Interdependence Model used in dyadic analysis, whereby the actor effect is estimated while controlling for the partner effect and then the partner effect is estimated while controlling for the actor effect (Kenny, Kashy, & Cook, 2006).
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
J Pediatr Health Care. 2017;31(6):703-712. © 2017 Mosby, Inc.