Abstract
Objective:
A family-tailored education and problem-solving intervention, Supporting Treatment Adherence Regimens (STAR), was developed to address the adherence challenges common in youth with epilepsy and their families. Randomized clinical trial (RCT) results indicated a 21% adherence improvement in the STAR group compared to an education only (EO) group 12-months post-intervention. The current study examined group differences (STAR vs. EO) in epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, and family functioning over time and whether these factors mediated group differences in adherence at 12-months post-intervention.
Methods:
RCT participants included 200 children (ages 2–12) with epilepsy and their caregivers. Children with new-onset epilepsy with adherence <95% were randomized to receive either the STAR (n = 27) or EO (n = 29) intervention. Caregivers completed questionnaires assessing outcomes of interest at baseline, midpoint of the intervention, post-intervention, and 3, 6, and 12 month follow-ups. Regression-based analyses of covariance and longitudinal mixed effect linear models were conducted.
Results:
Results generally revealed no significant group differences across outcomes of interest at post-intervention or over time. However, one significant model did emerge for social problem-solving skills (b = −1.74, p = .04), such that social problem-solving skills scores were initially higher in the STAR group compared to the EO group, then decreased slightly in the STAR group over time while remaining stable in the EO group. None of these factors mediated group differences in adherence at 12-months post-intervention.
Conclusion:
Future research should examine other potential mechanisms of treatment change following adherence interventions, such as STAR. Nonsignificant findings can inform the development of future study designs and intervention efforts.
Keywords: behavioral intervention, epilepsy, adherence, pediatric, compliance, mediators
1. Introduction
Approximately 60% of children with epilepsy and their parents demonstrate suboptimal adherence to anti-seizure medication (ASM) [1]. Nonadherence to ASMs can lead to increased morbidity (e.g., seizures), healthcare utilization, and mortality [2–4]. Across early childhood, adherence barriers for youth with epilepsy include disliking the taste of their ASM, parent forgetfulness, difficulties with pill swallowing, and medication refusal [5]. Despite the importance of adherence for health outcomes (e.g., seizures), few interventions exist for young children with epilepsy [6], with the exception of family-tailored education and problem-solving behavioral interventions developed by Modi and colleagues [7–11].
The Supporting Treatment Adherence Regimens (STAR) intervention was developed to target ASM adherence [8] in young children with epilepsy (2–12 years old) and their families. As part of a randomized clinical trial (RCT), children with adherence less than 95% were randomized to receive either the STAR (e.g., family based problem-solving) or an education only (EO) intervention over eight sessions (six face-to-face and two telephone sessions). Initial RCT results revealed a 21% adherence improvement in the STAR group compared to the EO group at 12-months post-intervention [10]. While these results are promising, examining factors that help explain these treatment effects is an important next step to optimize future iterations of STAR.
The pediatric self-management model [12] highlights the influential nature of individual, family, community, and health care system level factors on pediatric self-management behaviors and adherence. Each factor contains non-modifiable and modifiable contributors to adherence, with the latter serving as important targets for intervention. Multiple modifiable factors from this model have been linked to suboptimal adherence among youth with epilepsy, including less knowledge about epilepsy [13], more barriers to medication adherence [5], more caregiver distress [7], and lower general family functioning [7]. The STAR intervention targeted each of these domains by providing epilepsy-related education and problem-solving strategies to families.
The current study had two key aims. The first aim was to examine group differences (STAR vs. EO) in epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, and family functioning both cross-sectionally (i.e., at post-intervention) and over time (i.e., at post-intervention, 3-, 6-, and 12-month follow-ups). It was hypothesized that families in the STAR group would demonstrate more epilepsy-specific knowledge, fewer barriers to medication adherence, greater problem-solving skills, less caregiver distress, and higher family functioning than the EO group. A secondary aim was to explicitly test whether these factors mediated the relationship between intervention group and adherence at 12-months post-intervention. Specifically, the level of each of these factors at post-intervention was expected to help explain the adherence improvement in the STAR group at 12-months post-intervention.
2. Methods
2.1. Overview of Study Design
Data from this study stem from an enrichment RCT (i.e., only those with nonadherence received the intervention) comparing a family-tailored education and problem-solving adherence intervention (STAR) to an attention control intervention (EO). Detailed methods [8] and primary and secondary outcomes from the larger trial [10] have been published elsewhere.
2.2. Participants
For the larger RCT, eligible children were: (1) between ages 2–12 years, (2) diagnosed with epilepsy during the past 7 months, (3) prescribed only one ASM, and (4) living within 75 miles of the hospital. Children and caregivers were required to read and speak English. Children taking daily medications due to non-epilepsy medical disorders, except allergies and asthma, and/or with a significant developmental delay were excluded from the study. Four hundred and ninety-three families were assessed for eligibility. Two hundred and twenty-eight did not meet inclusion/exclusion criteria and 65 declined to participate. Two hundred families were enrolled in the study, with 56 children ultimately being randomized to either the STAR (n = 27) or EO (n = 29) intervention. Of note, 121 families were not randomized due to high adherence in the larger trial and 23 families withdrew from the study or were lost to follow-up.
2.3. Procedures
Following hospital Institutional Review Board approval, children with epilepsy and their caregivers were recruited through the Comprehensive Epilepsy Center at Cincinnati Children’s Hospital Medical Center from April 2013 to December 2018. Consented families completed baseline questionnaires and were provided with an electronic monitor to assess ASM adherence: Medication Event Monitoring System (MEMS) TrackCap or Vaica SimpleMed+ pillboxes. Participants were then monitored and had the opportunity to be randomized to one of the intervention arms (STAR or EO) due to nonadherence (defined as <95%) at three different time points (1-, 4-, and 7-months following baseline). Participation ended for those with optimal adherence (≥95%) at all three assessment periods.
Only participants randomized to intervention were included in the current study. Randomized participants received eight intervention sessions over a 4-month period. Interventionists, which included masters and Ph.D. level psychologists, were trained by a licensed psychologist and fidelity was maintained through monthly supervision and review of audio sessions with feedback to interventionists. Both the STAR and EO groups discussed epilepsy knowledge deficits and received feedback on the child’s adherence patterns during the first session. The STAR group also received adherence education. The EO group received seven education sessions focused on seizure safety, sleep hygiene, communication and psychosocial comorbidities, and school-based issues. In contrast, the STAR group used a problem-solving approach to address the family’s individualized adherence barriers for the remaining seven sessions.
Randomized participants completed additional questionnaires and adherence electronic monitors were downloaded at the midpoint of the intervention, post-intervention, and three follow-up periods after the intervention (3, 6, 12 months). Medical chart reviews were also conducted at each of these visits.
2.4. Measures
The current study examined epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, and family functioning as possible mechanisms of treatment change. All measures were completed by parents. Parents reported on the child’s age, gender, race/ethnicity, as well as additional family demographic information (e.g., Duncan socioeconomic index [14, 15]), using the Background Information Form. Medical Chart Reviews were conducted to collect information regarding seizure type and ASM prescription. Epilepsy-specific knowledge was assessed using an adapted version of the Epilepsy Knowledge Questionnaire (EKQ) [16], which consisted of 47 true/false items regarding general information about epilepsy and treatment. The percentage of total correct responses was used in this study and psychometric properties of this scale have been shown to be adequate [7]. The Barriers to Medication Adherence subscale from the 27-item Pediatric Epilepsy Medication Self-Management Questionnaire (PEMSQ; α = .68–.85 [17]) was used to assess caregiver’s perceptions of barriers to taking ASMs as prescribed. Items on this scale are reverse scored, such that higher scores represent better self-management. To assess problem-solving skills, the total standardized score from the Social Problem-Solving Inventory-Revised (SPSI-R: Short Form) [18] was used, with higher scores reflecting more adaptive problem-solving. This 25-item measure has been shown to have adequate psychometric properties [18] and assesses an individual’s ability to resolve problems in everyday life [19]. Caregiver emotional functioning was measured using the Global Severity Index T-score from the 53-item Brief Symptom Inventory (BSI; α = .71–.85 [20]) [21]. This index score reflects the overall emotional distress level of caregivers. Finally, the 60-item McMaster Family Assessment Device (FAD; α = .72–.92 [22]) [23] was used to assess family functioning. Both the Problem-Solving and General Family Functioning scales were used, with higher scores representing worse family functioning.
2.5. Data Analysis
Descriptive statistics were calculated for all variables. Participants with incomplete data were retained in the current study to obtain the largest possible sample and included in the analyses using full-information maximum likelihood estimation [19, 24]. All analyses were conducted in Stata version 17 [25] with a robust variance estimator and Mplus version 8.8 [26].
To assess whether there were group differences at post-intervention in epilepsy-specific knowledge (EKQ scores), barriers to medication adherence (PEMSQ scores), social and family problem-solving (SPSI-R and FAD scores, respectively), caregiver distress (BSI scores), and family functioning (FAD scores), we conducted five regression-based analyses of covariance (ANCOVA) models, with intervention group (STAR vs. EO) as our primary predictor, while covarying for baseline levels of each outcome as well as SES (Duncan scores). Standardized effect size differences between the intervention groups on each of the observed outcome scores (i.e., unadjusted) at post-intervention were calculated using Cohen’s d. The following thresholds were used to evaluate effect sizes: d < 0.20 was considered trivial, d = 0.20 was considered small, d = 0.50 was considered a medium effect, and d = 0.80 or larger was considered a large effect [27]. To examine group differences in our outcomes over all time points (baseline, post-intervention, 3-, 6-, and 12-month follow-ups), we estimated longitudinal mixed effect linear models for each outcome, with group as the primary predictor and covarying for baseline Duncan scores.
Finally, we assessed whether epilepsy-specific knowledge (EKQ scores), barriers to medication adherence (PEMSQ scores), social and family problem-solving (SPSI-R and FAD scores, respectively), caregiver distress (BSI scores), and family functioning (FAD scores) levels at post-intervention mediated the relationship between the intervention group and adherence at 12-months post-intervention. We estimated the indirect effect of each of these potential mediators in a multiple mediation model, using 95% bias-corrected bootstrapped confidence to assess the significance of each indirect effect. This mediation model also covaried for Duncan scores and baseline levels of each of the mediators and baseline adherence. We assessed model fit for the multiple mediation model using empirically validated fit indices and levels suggested by Hu and Bentler [28]. Specifically, a root mean square error of approximation (RMSEA) < .05, and a comparative fit index (CFI) and Tucker-Lewis Index (TLI) > 0.95 indicate the hypothesized multiple mediation model fit the observed data well.
3. Results
3.1. Participants
Participant characteristics are presented in Table 1. There were no significant differences in demographic characteristics between the two intervention groups, with the exception of epilepsy type (χ2 = 12.5, p = .03). See Modi et al. for a Consolidated Standards of Reporting Trials (CONSORT) diagram and additional participant information [8]. Descriptive data regarding outcomes of interest are displayed in Table 2. Attrition analyses indicated that missingness for outcomes of interest did not vary by intervention group (STAR vs. EO) or demographic variables at any time point, with the exception of SES. Therefore, SES was included as a covariate in all analyses to ensure valid inferences under the missing at random assumption.
Table 1.
Participant Demographic and Epilepsy-Specific Data (N = 56)
| STAR Intervention (n = 27) | Education Only (n = 29) | |
|---|---|---|
| Characteristics | mean ± SD or n (%) | mean ± SD or n (%) |
| Child Age in years | 7.14 ± 2.80 | 8.15 ± 3.30 |
| Months Since Epilepsy Diagnosis | 2.49 ± 2.04 | 2.64 ± 2.13 |
| Duncan Score | 49.94 ± 19.64 | 48.99 ± 22.58 |
| Child Sex | ||
| Girls | 15 (56) | 14 (48) |
| Boys | 12 (44) | 15 (52) |
| Child Race | ||
| White, non-Hispanic | 18 (67) | 21 (72) |
| White, Hispanic | 0 (0) | 0 (0) |
| Black, non-Hispanic | 5 (19) | 5 (17) |
| Black, Hispanic | 0 (0) | 1 (3) |
| More than one race, non-Hispanic | 4 (15) | 1 (3) |
| More than one race, Hispanic | 0 (0) | 1 (3) |
| Asian | 0 (0) | 0 (0) |
| Seizure Type | ||
| Focal onset | 3 (11) | 8 (28) |
| Generalized onset | 20 (74) | 11 (38) |
| Unknown onset; e.g., unclassified | 4 (15) | 10 (34) |
| Initial Prescribed Anti-Seizure Medication | ||
| Carbamazepine | 0 (0) | 6 (21) |
| Ethosuximide | 12 (52) | 6 (21) |
| Levetiracetam | 5 (19) | 7 (24) |
| Oxcarbazepine | 1 (4) | 2 (7) |
| Lamotrigine | 0 (0) | 0 (0) |
| Topiramate | 1 (4) | 0 (0) |
| Valproic acid | 8 (30) | 8 (28) |
| Primary Caregiver | ||
| Mother/stepmother | 26 (96) | 27 (93) |
| Father/stepfather | 0 (0) | 0 (0) |
| Other; e.g., grandmother, uncle/aunt | 1 (4) | 2 (7) |
| Caregiver Marital Status | ||
| Single | 11 (41) | 6 (21) |
| Married | 11 (41) | 18 (62) |
| Divorced/separated | 4 (15) | 5 (17) |
| Widowed | 1 (4) | 0 (0) |
Note: STAR = Supporting Treatment Adherence Regimens. Higher Duncan scores represent greater occupational attainment.
Table 2.
Descriptive Statistics for Variables of Interest
| Variable | Group | Baseline | Mid-Intervention | Post-Intervention | 3-Month Follow-Up | 6-Month Follow-Up | 3-Month Follow-Up |
|---|---|---|---|---|---|---|---|
| M ± SD | M ± SD | M ± SD | M ± SD | M ± SD | M ± SD | ||
| Epilepsy-Specific Knowledge | EO | 82.75±9.96 | 87.29±6.07 | 86.92±7.55 | 87.32±4.71 | 86.84±6.80 | 85.20±6.98 |
| STAR | 84.41±5.62 | 86.76±6.02 | 88.26±4.53 | 87.09±4.79 | 86.88±5.59 | 85.94±6.26 | |
| Barriers to Medication Adherence | EO | 34.59±4.25 | 33.42±3.82 | 34.90±4.16 | 34.18±4.84 | 34.86±4.94 | 34.95±4.03 |
| STAR | 34.15±6.06 | 34.29±4.30 | 34.67±3.81 | 34.77±3.56 | 34.68±3.96 | 35.76±327 | |
| Social Problem-Solving | EO | 109.41±14.96 | 113.56±15.37 | 107.37±17.97 | 112.33±14.30 | 113.16±13.67 | 110.68±14.79 |
| STAR | 115.85 ±12.16 | 115.05±12.92 | 117.63±12.24 | 107.73±29.50 | 117.35±13.67 | 106.48±27.12 | |
| Family Problem-Solving | EO | 1.83±0.44 | 1.81±0.42 | 1.73±0.42 | 1.76±0.41 | 1.76±0.45 | 1.78±0.38 |
| STAR | 1.76±0.35 | 1.74±0.39 | 1.74±0.43 | 1.83±0.41 | 1.76±0.38 | 1.87±0.37 | |
| Caregiver Emotional Distress | EO | 53.85±12.59 | 51.48±12.78 | 52.00±12.24 | 48.63±13.03 | 52.80±15.21 | 50.14±12.51 |
| STAR | 50.73±9.91 | 53.27±9.84 | 50.21±10.98 | 49.00±10.67 | 50.95±11.23 | 51.15±11.93 | |
| Family Functioning | EO | 1.65±0.41 | 1.63±0.48 | 1.60±0.46 | 1.58±0.45 | 1.64±0.51 | 1.67±0.46 |
| STAR | 1.46±0.36 | 1.51±0.37 | 1.53±0.38 | 1.53±0.41 | 1.57±0.36 | 1.64±0.38 |
Note. Scores for Social Problem-Solving are presented as scaled scores and Caregiver Emotional Distress are presented as T-Scores to aid with interpretation of results.
Regression-based ANCOVA models revealed that there were no significant group differences in epilepsy-specific knowledge (b = 1.17, p = .48), barriers to medication adherence (b = −0.28, p = .81), social problem-solving skills (b = 5.94, p = .06), family problem-solving (b = 0.03, p = .74), caregiver emotional distress (b = 1.78, p = .52) or family functioning (b = 0.04, p = .59) at post-intervention. Effect size differences in the observed outcome scores were trivial to small for epilepsy-specific knowledge (d = 0.22), barriers to medication adherence (d = 0.06), caregiver emotional distress (d = 0.15), family problem-solving (d = 0.01) and family functioning (d = 0.16) at post-intervention. For social problem-solving skills, effect size differences were moderate to large (d = 0.68).
Longitudinal mixed effect models indicated that average intervention group trajectories from baseline to 12-months post-intervention were not significantly different for epilepsy-specific knowledge (b = 0.22, p = .50), barriers to medication adherence (b = 0.13, p = .65), family problem-solving (b = 0.01, p = .41), caregiver emotional distress (b = 0.63, p = .27) or family functioning (b = 0.01, p = .69). However, one significant model did emerge for social problem-solving skills (b = −1.74, p = .04), such that social problem-solving skills scores were initially higher in the STAR group compared to the EO group, then decreased slightly in the STAR group over time while remaining stable in the EO. Group trajectories for each of the outcomes are presented in Figure 1.
Figure 1.
Group Differences in Outcomes of Interest
The multiple mediation model was not a good fit to the data, RMSEA = 0.18, CFI = 0.49, TLI = 0.33. Further, none of the indirect effects were significant (i.e., 95% bias-corrected bootstrapped confidence intervals for each of the estimated indirect effects contained zero). Thus, we have no evidence that the effect of intervention group on adherence at 12-months post-intervention was mediated by epilepsy-specific knowledge, barriers to medication adherence, social or family problem-solving skills, caregiver emotional distress, or family functioning scores at post-intervention.
4. Discussion
Previous research documented a 21% improvement in ASM adherence among children with epilepsy following the STAR intervention relative to the EO group [10]. The current study sought to build upon these initial RCT findings by examining factors that help to explain these treatment effects. Understanding how and why adherence interventions, such as STAR, produce change allows for the creation of more targeted treatments that trigger critical change processes. Contrary to hypotheses, results revealed that there were no significant group differences (STAR vs. EO) in epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, or family functioning following the STAR intervention; however, there was one significant group difference regarding change over time, such that social problem-solving skills decreased for those in the STAR group and remained stable for those in the EO group. This appeared to be driven by higher estimated scores at baseline in the STAR group, with the two groups demonstrating similar social problem-solving skills at 12-months post-intervention. Additionally, none of these factors significantly mediated the effects of treatment on adherence.
These results are quite surprising given the larger adherence and epilepsy literature [5, 7, 12, 13], which suggests that these individual and family factors are important contributors to adherence. Indeed, research indicates that epilepsy knowledge may provide an important foundation for adherence behaviors [29] and that families of children with chronic health conditions must also effectively collaborate, problem-solve, and resolve conflicts to adhere to their child’s medical regimen [30]. The STAR intervention directly targeted these factors by providing comprehensive educational content about epilepsy (e.g., epilepsy restrictions, introduction to ASM adherence) and teaching families a five-step problem-solving framework to support adherence. Yet, the mostly nonsignificant group differences (STAR vs. EO) in this study highlight important methodological considerations for future work in this area.
As depicted in Figure 1, families in this study generally reported few difficulties regarding epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, and family functioning. Such ceiling effects may have limited our ability to detect meaningful changes in these outcomes post-intervention, especially given the smaller sample size in this study. However, these results are also promising and suggest that many families may be doing well from a psychosocial perspective, despite their child’s suboptimal ASM adherence. Additional research is needed to corroborate these findings and should aim to recruit larger samples to increase power and the ability to detect significant effects.
Moreover, it is important to consider that the measures of social problem-solving and generic family functioning used in this study may not have been sensitive to the effects of the STAR intervention [30]. Given that the STAR intervention targets epilepsy-specific family processes (e.g., conflict, communication), measures that capture problem-solving and family functioning as it relates to epilepsy management may be more sensitive to the effects of this intervention. Current measurement options are somewhat limited; thus, there may be a need to create measures that specifically assess family processes related to adherence. With such enhanced measures, we would be able to identify mechanism most likely to impact adherence outcomes and where to focus and target our future interventions. For example, if family conflict were identified as a mechanism of action, we could enhance the STAR intervention to further focus on conflict resolution or ways to better communicate when conflict occurs.
Reporting on nonsignificant results [31], as was done in the current study, is an integral part of the research process. Nonsignificant findings can provide important insights into the validity of current theoretical approaches, illuminate measurement considerations, and guide future research designs. Given that publication bias (i.e., a preference to primarily publish significant results) often occurs in research [32], it is integral that researchers begin grappling with nonsignificant results. This can inform future iterations of adherence interventions in epilepsy and ensure the reliability of treatment effects.
Highlights.
Family-based problem-solving interventions improve adherence to anti-seizure medications.
Epilepsy-specific knowledge, barriers to medication adherence, problem-solving skills, caregiver emotional distress, and family functioning do not mediate these treatment effects.
Families of young children with newly diagnosed epilepsy are generally functioning well from a psychosocial perspective.
The examination of other mediators (e.g., epilepsy-specific family conflict or communication) is warranted.
Acknowledgements
We would like to acknowledge the time and effort of all study participants, as well as our research team members who helped facilitate this research project. A portion of the results included in the current article were presented at the 2023 Society of Pediatric Psychology Annual Conference.
Funding:
This work was supported by the National Institutes of Health-Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD073115).
Declaration of interests
Avani C. Modi and Shanna Guilfoyle report financial support for this research study was provided by National Institute of Health- Eunice Kennedy Shriver National Institute of Child Health and Human Development.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declarations of interest: None
References
- [1].Modi AC, Rausch JR, Glauser TA. Patterns of non-adherence to antiepileptic drug therapy in children with newly diagnosed epilepsy. JAMA 2011;305: 1669–1676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Faught E, Weiner JR, Guerin A, Cunnington MC, Duh MS. Impact of nonadherence to antiepileptic drugs on health care utilization and costs: findings from the RANSOM study. Epilepsia 2009;50: 501–9. [DOI] [PubMed] [Google Scholar]
- [3].Modi AC, Wu YP, Rausch JR, Peugh JL, Glauser TA. Antiepileptic drug nonadherence predicts pediatric epilepsy seizure outcomes. Neurology 2014;83: 2085–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Faught E, Duh MS, Weiner JR, Guerin A, Cunnington MC. Nonadherence to antiepileptic drugs and increased mortality: findings from the RANSOM Study. Neurology 2008;71: 1572–8. [DOI] [PubMed] [Google Scholar]
- [5].Gutierrez-Colina AM, Smith AW, Mara CA, Modi AC. Adherence barriers in pediatric epilepsy: From toddlers to young adults. Epilepsy Behav 2018;80: 229–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Al-Aqeel S, Gershuni O, Al-Sabhan J, Hiligsmann M. Strategies for improving adherence to antiepileptic drug treatment in people with epilepsy. Cochrane Database Syst Rev 2017;2: CD008312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Bakula DM, Junger KW, Guilfoyle SM, Mara CA, Modi AC. Key predictors of the need for a family-focused pediatric epilepsy adherence intervention. Epilepsia 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Modi AC, Glauser TA, Guilfoyle SM. Supporting Treatment Adherence Regimens in young children with epilepsy and their families: Trial design and baseline characteristics. Contemp Clin Trials 2020;90: 105959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Modi AC, Guilfoyle SM, Mann KA, Rausch JR. A pilot randomized controlled clinical trial to improve antiepileptic drug adherence in young children with epilepsy. Epilepsia 2016;57: e69–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Modi AC, Guilfoyle SM, Glauser TA, Mara CA. Supporting treatment adherence regimens in children with epilepsy: A randomized clinical trial. Epilepsia 2021;62: 1643–1655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Williford DN, Guilfoyle SM, Modi AC. Demystifying a family-based epilepsy adherence problem-solving intervention: Exploring adherence barriers and solutions. Clin Pract Pediatr Psychol 2023;11: 66–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Modi AC, Pai AL, Hommel KA, Hood KK, Cortina S, Hilliard ME, Guilfoyle SM, Gray WN, Drotar D. Pediatric Self-management: A Framework for Research, Practice, and Policy. Pediatrics 2012;129: e473–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Carbone L, Zebrack B, Plegue M, Joshi S, Shellhaas R. Treatment adherence among adolescents with epilepsy: what really matters? Epilepsy Behav 2013;27: 59–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Hauser RM. Measuring socioeconomic status in studies of child development. Child Dev 1994;65: 1541–5. [DOI] [PubMed] [Google Scholar]
- [15].Stevens G, Featherman DL. A revised socioeconomic index of occupational status. Soc Sci Res 1981;10: 364–395. [Google Scholar]
- [16].Jarvie S, Espie CA, Brodie MJ. The development of a questionnaire to assess knowledge of epilepsy: 1--General knowledge of epilepsy. Seizure 1993;2: 179–85. [DOI] [PubMed] [Google Scholar]
- [17].Modi AC, Monahan S, Daniels D, Glauser TA. Development and validation of the Pediatric Epilepsy Medication Self-Management Questionnaire. Epilepsy Behav 2010;18: 94–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].D’Zurilla T, Nezu A, Maydeu-Olivares A. Social Problem-Solving Inventory-Revised: Technical Manual. Toronto: Multi-Health Systems, Inc; 2002. [Google Scholar]
- [19].Rubin DB. Inference and missing data. Biometrika 1976;63: 581–592. [Google Scholar]
- [20].Kellett S, Beail N, Newman DW, Frankish P Utility of the Brief Symptom Inventory in the Assessment of Psychological Distress. Journal of Applied Research in Intellectual Disabilities 2003;16: 127–34. [Google Scholar]
- [21].Derogatis LR. Brief Symptom Inventory: Administration, Scoring, and Procedures Manual 4th edition. Minneapolis: NCS Pearson, Inc.; 1993. [Google Scholar]
- [22].Miller IW, Epstein NB, Bishop DS, & Keitner GI The McMaster Family Assessment Device: Reliability and Validity. Journal of Marital and Family Therapy 1985;11: 345–356. [Google Scholar]
- [23].Epstein NB, Baldwin LM, Bishop DS. The Mcmaster Family Assessment Device. Journal of Marital and Family Therapy 1983;9: 171–180. [Google Scholar]
- [24].Schafer JL, Graham JW. Missing data: Our view of the state of the art. Psychological Methods 2002;7: 147–177. [PubMed] [Google Scholar]
- [25].StataCorp. Stata Statistical Software: Release 17. In. College Station, TX: StataCorp LLC; 2021. [Google Scholar]
- [26].Muthén L, Muthén B. Mplus User’s Guide, 8th Ed. Los Angeles: Muthén & Muthén; 1998–2017. [Google Scholar]
- [27].Cohen J Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Erlbaum; 1988. [Google Scholar]
- [28].Hu LT, Bentler PM. Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives. Structural Equation Modeling-a Multidisciplinary Journal 1999;6: 1–55. [Google Scholar]
- [29].Helgeson DC, Mittan R, Tan SY, Chayasirisobhon S. Sepulveda Epilepsy Education: the efficacy of a psychoeducational treatment program in treating medical and psychosocial aspects of epilepsy. Epilepsia 1990;31: 75–82. [DOI] [PubMed] [Google Scholar]
- [30].Psihogios AM, Fellmeth H, Schwartz LA, Barakat LP. Family Functioning and Medical Adherence Across Children and Adolescents With Chronic Health Conditions: A Meta-Analysis. J Pediatr Psychol 2019;44: 84–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Mehler DM, Edelsbrunner PA, Matić K. Appreciating the significance of non-significant findings in psychology. ournal of European Psychology Students 2019;10: 1–7. [Google Scholar]
- [32].Rosenthal R The file drawer problem and tolerance for null results. Psychological Bulletin 1979;86: 638–641. [Google Scholar]