Background

The estimated number of children with chronic health conditions in the United States is 15 to 18 million [1]. These large numbers of children rely on their caregivers for the majority of their care and health management by virtue of them being dependents. The parent/caregiver roles include learning about their child’s condition, giving medications, ensuring that the child performs procedures, and providing transportation to appointments with health care providers [2].

The number of adults, who have grown up with technology, in particular interactive media such as computer-delivered education and video games, has been rising with each successive generation. Although the use of computer-based technology and video games has been described as a means to teach children to self-manage conditions such as cancer, asthma, and diabetes, the state of the literature on the use of interactive technology and/or video games to teach parents about their children’s conditions has not been well characterized [3–9]. Such interventions have been described in the literature as “serious games” [10]. The goal of this systematic review of the literature was to identify and evaluate research that had used interactive media approaches to educate parents of pediatric patients with chronic conditions and their relative effectiveness.

Methods

We conducted a review of publications using the following databases: MEDLINE, PsychINFO, CINAHL, and EMBASE, and we reported the findings based on the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) criteria [11]. The search terms used are presented in Table 1. Additionally, the Cochrane database of systematic reviews was accessed, and an review focused on interactive health communication applications acted as a source for additional articles that the reviews examined [12]. To be included in the review, an article had to report data evaluating an educational intervention using either interactive media or interactive games for parents. Any original research design was eligible. The titles of the articles were scanned to identify pertinent studies for abstract evaluation based on the presence of an educational intervention in the title. Two authors independently examined abstracts from the most relevant articles; any disagreement between the authors was settled by a senior advisor. They also performed reference chaining using the discovered studies as a starting point. They focused on the following research question (see Table 2). The primary research question was: Does the use of “serious games” (games intended to educate) and/or interactive media for parents improve health outcomes in children with chronic conditions (e.g. asthma, diabetes, chronic kidney disease, cystic fibrosis, and cardiac abnormalities)?

Table 1 Search terms: search strategies and key words
Full size table
Table 2 Population, intervention, comparator, outcome, time allowed for outcomes, time of search of the literature, study designs allowed (PICOTTS)
Full size table

They determined internal validity, potential for biases, accuracy and appropriateness of the analysis and applicability used in each study. The overall quality of the study was determined through an application of the GRADE guidelines [13–19]. They also determined the literacy level of the tools used in these the studies using the Flesh-Kincaid methodology.

Inclusion/exclusion criteria

Only observational studies, cohort studies, and randomized control trials were included in the final review. They included studies that went as far as 1986, the earliest year that this topic was introduced in the literature [20]. For cohort and randomized trials, they did not differentiate studies on the basis of time between intervention and post-testing, when applicable. The participants for the studies needed to include parents in the exposure group.

Data extraction

The studies were examined and basic information was extracted from each study. For quantitative studies data was collected on: (1) the sample size in the study; (2) the composition of the study population; (3) the measurement tool that used in the study; and (4) outcomes. Further information on the potential for biases and analysis performed in each of the quantitative studies is presented in Additional file 1 as an evidence table.

Critical appraisal

Two co-authors critically appraised each manuscript based on previously established criteria to assess the size of each study as well as potential biases, confounders, measurement precision, generalizability, and the meaning of the findings from the study. The studies were evaluated using the GRADE methodology, using such things as risk of bias and inconsistency, to grade the studies as very low, low, moderate, and high quality [19]. If there was a disagreement existed on the quality assessment of a study, the study was discussed with a senior advisor until a consensus was reached on the grading of the study.

Data synthesis

Because of heterogeneity in interventions and study designs they did not attempt meta-analysis. The overall trend in the results and the degree of certainty in the results were considered when assessing the body of literature pertaining to our focused questions.

Results

Our initial search identified 4367 papers, but only 12 fulfilled the criterion established for final analysis as noted in Fig. 1. Most of the studies that explored outcomes of knowledge and/or skill showed an improvement in these outcomes after implementation of their particular intervention (Table 3). The quality of the studies found varied greatly; the majority being very poor quality studies with only one high quality study (Table 4).

Fig. 1
figure 1

PRISMA flow diagram

Full size image
Table 3 Synthesis table
Full size table
Table 4 GRADE quality evaluation
Full size table

One study examined the level of knowledge that parents had regarding asthma and found an improvement in scores using the “serious game” that was presented in the study [5]. This study did not control for child’s percentage of life with the condition, so parents experience with the disease or knowledge base overtime was not evaluated (P = 0.06, [95 % Confidence Interval [22.71, 24.29]). The questions that were used to assess the knowledge of the participants both before and following the intervention had a standard grade level that varied greatly.

One randomized study used human simulators for improving the knowledge and management of diabetes by parents [21]. Although this was a small cohort study, improvement in self-efficacy and knowledge were promising, no statistical significance was found. This study did not include information on the child’s disease duration.

The largest study we found was an observational study that looked at the use of educational kiosks at various public locations [22]. This study’s participants found this form of interactive media to be educational; 49 % of the first time users who completed the exit survey, planned to discuss the topics presented through the kiosks with health providers. Although the study did not specify the questions used within the kiosk, there was no association between the ease of kiosk use and participant education level. Almost half of the kiosk events (47 %) came from one location (a fast food restaurant), which could have acted as a source of selection bias since participants could have different exposure level than those recruited in a library.

The next randomized study assessed the effect of intense therapy with dieticians, psychologists and providers on the self-management skills of parents whose children were diagnosed with atopic dermatitis [23]. The intervention showed improvement in medical treatment confidence by the caregivers, resulting in a greater reduction (although not statistically significant) in disease severity among the intervention participants compared to controls based on the SCORAD (a survey for scoring atopic dermatitis). This translated into a decrease of 20.5 points in the intervention arm and 16.2 points in the control arm (p = 0.21, t = 1.27). To measure rumination, the authors relied on the Trier scale for measuring coping with disease, but there are not actual values provided for the intervention and control arms of the study.

One randomized looked at the effects on a computer-based asthma education program on quality of life, peak flow measures, and parental knowledge [24]. There was no statistical difference between the intervention and control groups concerning asthma knowledge or asthma symptoms. There was a small improvement 0.4 in the intervention versus 0.3 in the control group in terms of their correct responses to a questionnaire related to air quality control, although this was not statistically significant (p > 0.05, 95 % CI [−0.3, 1.1]). This study was concerning for the lack of control for asthma severity among the intervention and control groups.

Another randomized control trial examined the effects of an interactive health learning device on children with asthma, with particular emphasis on limitations on physical activity, peak flow readings, and use of health services [25]. The study did show a difference between the groups in terms of peak flow readings (OR of 0.43, 95 % CI [0.23–0.82], p = 0.01) and limitation in physical activity (OR of 0.52, 95 % CI [0.29–0.94], p = 0.03). There was not a statistical difference between the groups in terms of hospitalizations (OR of 0.99, 95 % C [0.25, 3.88], p = 0.96) or emergency department (ED) visits (OR of 0.59, 95 % CI [0.26, 1.35], p = 0.21). The study was underpowered by the author’s own calculations, so the certainty of the effects detected is in question.

One study took a particular interest in urban, minority children with asthma in a randomization control trial of the ability computer-based education program to affect knowledge and self-efficacy [26]. The trial did demonstrate an improvement in knowledge scores of participants (knowledge scores of 21.1, 95 % CI [19.38 to 22.82], p < 0.01) and self-efficacy (mean 56.5, 95 % CI [53.38, 59.62], p = 0.04). There are concerns about the lack of controlling for disease severity or performance of children in school, which were both measured, in the data analysis.

Another diabetes focused study tested the effect of a computer based education program on blood glucose readings and disease knowledge in a match case–control trial [27]. The study did show improvement in blood glucose readings prior to lunch (F =10.922, p < 0.02) and prior to dinner (F = 7.221, p < 0.025). The lack of controlling for the factors used to match the cases and controls, as well as the small sample size, raised concerns of the internal validity of the study.

One study focused on children with leukemia assessed a CD-ROM based education program through a randomized control trial that emphasized sense of control [28]. The showed improvement in sense of control (F = 6.38, p = 0.004). Data analysis that did not try to control for parental education, which the authors measured, was a concern for the internal validity of the study since this factor represented a potential confounder in the development of a sense of control in the participants.

Another large study included 228 children in a randomized trial examining an Internet-based interactive media education program for asthma as it related to disease knowledge and caregiver quality of life [29]. Although there was an improvement in disease knowledge (mean change in score (2.52, 95 CI [−0.38, 5.42], p = 0.029) in the intervention group. The study’s internal validity was hindered by a limited recall interval for the quality of life questionnaire.

Another study of high-risk urban children with asthma assessed the ability of a multimedia education software to affect ED visits as well as parental and youth knowledge of disease [30]. The study did not observe a statistical difference in emergency department, although there was notable decrease in both groups utilization of the ED. Parental knowledge did not show a significant difference between the groups, but the knowledge of the children showed greater improvement in the intervention group (F =18.78, p < 0.001).

A notable study examined the effects of an online parent information support program on parental management of chronic kidney disease (CKD), parental empowerment, and the level of support from fathers concerning CKD care [31]. Parents did have an improvement in their perceived competence in disease management, although not statistically significant (2.6, 95 % CI [−1.6, 6.7], p = 0.213). Parental empowerment (−0.2, 95 % CI [−0.5, 0.2], p = 0.404) and the level of support from fathers (−4.3, 95 % CI [−24.7, 16.2], p = 0.667) were not significantly affected by the intervention.

The overall trend in the effects was positive in most of the studies. Only one study fit the criteria to be considered a high quality study by the GRADE guidelines [29]. Despite this lack of many high quality studies, there did appear to be a trend of positive results that showed an effect of the “serious games” on various intermediate, as well as some clinically relevant, outcomes.

Discussion

We found limited evidence supporting the effectiveness of interactive media to educate parents of children with chronic conditions. However, a consistency of positive results supports the idea that parent education is possible through interactive media or games. The magnitude of this effect cannot be accurately determined from the studies reviewed above due to a lack of certainty in many studies, as well as varying measures that were targeted by the study designs. The effects that were shown on knowledge add to the causal linkage since parent knowledge has been associated with health outcomes in pediatric patients with chronic conditions.

It is important to understand the ability of parents to receive and process health information for the continued management of their child’s chronic condition [2]. To optimize this capacity, parental education should be tailored to their needs, culture and literacy level. It has been shown that parents go through a sense of disorientation when their child is diagnosed with a chronic condition [32]. Afterwards it may be expected that parents or caregivers will be responsible for the management of their child’s chronic condition until the patient is ready to self-manage their health.

Interactive and game platforms for parent education may be appealing because they can draw on skills that parents may have developed in their youth. These parent education platforms could also account for parental health literacy as the information could be presented with visual and auditory signals, promoting disease knowledge and management, impacting patient health outcomes. Among the studies examined in this review, only a handful demonstrated interventions that were tailored for parents with varying degrees of health literacy.

The majority of the studies showed positive and/or informative findings, which may be a reflection of publication bias in that only these findings appear to be reported. Although this is a legitimate concern, the scarcity of studies shows a gap in the literature regarding the use of interactive media for parents as learners of their child’s chronic condition. This gap represents an opportunity for improving the knowledge base of the primary care giver of children with chronic health and mental conditions. Interactive technology and/or serious games are tools that would allow parents, as well as children, to develop their self-efficacy skills in appealing formats/platforms.

From our systematic search of the literature, the studies were underpowered for the most part. This lack of large cohorts is another area that can be improved upon in order to strengthen the body of evidence. It may be argued that the lack of large cohorts makes the findings in the studies described statistically insignificant. While these small studies are not conclusive, they show trends that would be well served by being tested in larger, powered trials. Additionally, the clinical significance of findings in each of the studies, were taken into consideration. The increased knowledge that caregivers can gain from the use of interactive media has been shown to translate to decreased use of health care [33].

Another implication may be, the improved self-efficacy of the caregiver may translate to the child. One plausible sequence of events is that the child sees their caregiver gaining a handle on the information relating to their chronic illness. From there the child is inspired to learn about their illness so that they may emulate their caregiver.

Conclusion

The literature has few studies that show improvement in intermediate measures of disease management for parents of children with chronic conditions. There were some studies that examined health outcomes, such as morbidity from disease and or health care utilization, but this area is a potential source for future research. Few large interventional studies also points to a gap in the current literature. While there does not appear to be a large amount of existing data, the evidence that is present has encouraging results.

Our findings show that interactive media could potentially serve as a tool to educate parents/caregivers about their child’s physical/mental health condition. The use of games to educate children has been explored in children with some chronic conditions [20]. It would stand to reason that parents could learn from similar games that are tailored for them. The literature at this time shows a lack of a large number of high quality studies that support this idea, and this area is one possible avenue for further research.