Design of a Health Information System Enhancing the Performance of Obesity Expert and children Teams

The globally increasing prevalence of childhood obesity is one of the most serious public health challenges of the twenty-first century. Due to the need for multi-professional therapies that require a high amount of personnel and financial resources, IT-supported interventions promise help. So far, meta-studies show their limited impact on health outcomes. This work presents therefore a design theory that helps constructing health information systems (HIS) that positively impact the performance of obesity expert and children teams. Team performance is measured through self-reports, patients’ adherence to therapy and positive health outcomes. In order to assess the utility of the proposed design theory, its underlying design process was adopted by an interdisciplinary team of therapists, patients, their parents, IS researcher and computer scientists. This team developed and evaluated several HIS services collaboratively over the course of two years. Results of this design process show first evidence of the utility of the HIS design theory. However, challenges with regard to the design process still exist and are discussed. Keywords: Health information system, design theory, design process, childhood obesity, therapy, prevention, evaluation. Kowatsch et al. / Design of a HIS enhancing the performance of obesity expert and children teams Twenty Second European Conference on Information Systems, Tel Aviv 2014 2 1 Introduction The prevalence of overweight and obesity in childhood has dramatically increased over the last decade (Aeple et al., 2013; Sassi, 2010). According to the 2013 fact sheet of the World Health Organization, more than 40 million children under the age of five were overweight in 2011. The prevalence of this disease pattern has tripled in Europe since the 1980s, whereas this effect can be primarily observed in childhood and adolescence (Branca et al., 2007, p. 9). In Switzerland and in most Central and Northern European countries, the overweight rates of children and adolescents have doubled between 1997 and 2007 (Aeberli et al., 2010; Knöpfli et al., 2007; Zimmermann et al., 2004). Although the prevalence of pediatric adiposity did level off until 2012 (Murer et al., 2013; Rokholm et al., 2013), it is expected that almost 20% of the Swiss children will suffer from overweight or obesity in 2022 (Schneider et al., 2009b). This epidemic dissemination has not only psychological and physiological drawbacks for those being affected, but it has also serious implications for the public and private healthcare sector: First, already during childhood severe cardiovascular and metabolic co-morbidities are present (L'Allemand-Jander, 2010). Second, pediatric obesity tracks into adulthood in more than 75% of 7–11 year old children (Andersen et al., 2010; Reilly and Kelly, 2011; Toschke et al., 2008). Third, public health costs are drastically increasing due to overweightand obesity-related comorbidities (Hänggli et al., 2008; Wabitsch and Mooss, 2009; Wang et al., 2011). In Switzerland, for example, obesity-linked disease costs have more than doubled from 2.600 Mio USD in 2001 to 5.800 Mio USD in 2006 (Schneider et al., 2009a) which equals to almost 10% of all health expenses in Switzerland (Swiss Statistical Office). In order to address these serious issues, multi-professional programs with physical activity, nutritional and behavioral components have been proposed (Sempach et al., 2007) and shown to have positive effects on therapy outcomes (Ho et al., 2012; Oude et al., 2009; USPSTF, 2010), for example, on body composition and metabolic parameters (Savoye et al., 2007). Similar beneficial effects could be demonstrated in low-threshold primary care interventions, if performed by specialized healthcare providers (Sargent et al., 2011). However, implementation of conventional familyand home-based childhood obesity interventions is challenging for parents, often requiring them to attend multiple educational sessions. Attrition rates for traditional interventions are frequently high, namely 30% to 50%, due to competing demands for parents' time (Knowlden and Sharma, 2012). For example, in Switzerland, less than one percent of overweight children can participate in those programs due to limited personal and financial resources (Hänggli et al., 2008). In addition, obesity interventions are also required in rural regions but they are not affordable today as well (ibid.). In summary, limited personnel resources and costs have a negative impact on health care supply while, at the same time, multi-professional programs are strongly recommended. Health information systems (HIS) have not only the potential to improve outcomes of health interventions but also to reduce their costs significantly (Agarwal et al., 2010; Spring et al., 2013; Spring et al., 2012). However, up till now, effects on obesity-related health outcomes were rather small and un-sustained (e.g. Illner et al., 2012; Ngo et al., 2009; Reed et al., 2012). It is therefore still an open issue how to design evidence-based HIS tailored to overweight and obese children that significantly improve health outcomes in the long term. In order to address these challenges, we propose a design theory that aims at building novel HIS that improve the performance of obesity expert and children teams. The applicability of the design theory is demonstrated by the design and evaluation of one HIS artifact, which is embedded into a health care concept with these overall goals: (1) to increase the data quality of patients’ self-monitoring activities, e.g. measuring steps per day, (2) to intensify the feedback from patients to therapists, (3) to visualize the implementation of lifestyle modifications in the home environment, (4) to document therapyrelevant everyday situations by high-resolution data, e.g. measuring mood or diet-related situations, (5) to reduce the families’ time demands and number of on-site consultations while allowing for Kowatsch et al. / Design of a HIS enhancing the performance of obesity expert and children teams Twenty Second European Conference on Information Systems, Tel Aviv 2014 3 punctual interventions from distance, and (6) to appeal to the patient and to support her to reach therapeutic goals. Additionally, potential side effects of IT use have to be addressed in order to avoid excessive gaming or screen time. For that purpose, a HIS design theory for IT-supported childhood obesity programs is presented. In line with this theory, several HIS services are collaboratively developed and evaluated by obesity experts, overweight and obese children, their parents, IS researchers and computer scientists. These HIS services are then evaluated by longitudinal field studies with therapists, children and their parents as part of existing obesity therapy and prevention programs. The remainder of this paper is structured as follows. We next motivate the current work by outlining lack of evidence-based IT interventions for childhood obesity. Against this background, we then describe a very first version of a HIS design theory for IT-supported childhood obesity interventions followed by a detailed description of its underlying design process. An overview of the HIS services is presented afterwards before results from two particular evaluations are described. The results are then discussed with regard to the design theory and its design process. This paper concludes with a summary and outlook on future research. 2 Shortcomings of IT Interventions for Childhood Obesity Today, it is state-of-the-art that obesity programs address three, i.e. physical, nutrition and behavioral components (Hauner et al., 2007; Sempach et al., 2007) as multi-component programs have been demonstrated to be more efficacious than mono-component interventions (USPSTF, 2010). Moreover, goals of obesity therapy in the growing child do not exclusively focus on weight loss, but on improvement of lifestyle and mental health, since these are preconditions for long-term beneficial outcomes (Epstein et al., 2010; Sempach et al., 2007). The lack of patients’ resources and health budgets (Hänggli et al., 2008) initiated several attempts to increase the efficiency and quality of obesity interventions by HIS, but current reviews indicate that effects of HIS on obesity-related health outcomes were rather small and un-sustained (e.g. Connelly et al., 2013; Illner et al., 2012; Lieffers and Hanning, 2012; Ngo et al., 2009; Reed et al., 2012; Wieland et al., 2012; Williamson et al., 2006). For example, effects of low-threshold internet based interventions on improvement of lifestyle or obesity were shown, but limited, i.e. not sustained after two years, due to reduced usage of the web application (Williamson et al., 2006). In addition, the high attrition rate from self-administered IT-based systems is related to self-motivation and literacy, characteristics known to be reduced in a large group of overweight children, e.g. those originating from migration families or with lower socioeconomic and / or educational background (l'Allemand et al 2012;Rokholm et al 2010). In line with these results, Table 1 shows a selection of IT-supported obesity interventions and commercial applications of which six shortcomings can be observed. First, the majority of interventions do not consider a multi-professional approach as recommended, i.e. a concurrent focus on physical activity, nutrition intake and behavioral support is not present. Second, only a few IT-supported health interventions are tailored to children and adolescents with their individual needs and requirements. Third, evidence-based effects on therapy outcomes are rare if at all present. Fourth, none of these applications has been evaluated as part of an existing multi-professional obesity intervention for children and adolescents. Fifth, none of these interventions or applications was either adapted to or evaluated with multi-culture obesity programs, which is crucial in the sense that cultural differences must be taken into account during their development. Finally and to the best of our knowledge, none of these applications was co-designed by therapists, children, their parents, IS researcher and computer scientists according to an design process for constructing evidence-based HIS and only few have incorporated the target group as co-designers (indicated by an asterisk * in Table 1). Kowatsch et al. / Design of a HIS enhancing the performance of obesity expert and children teams Twenty Second European Conference on Information Systems, Tel Aviv 2014 4 Due to these facts, we next describe a first version of a HIS design theory for childhood obesity interventions. Name Reference Component of program Tailored to Evidence-­‐based effects on health