Technology and Adherence in EHealth: a Systematic Review

Background: Although research has proven that eHealth interventions can be effective, one of the largest pitfalls is the high attrition rates common in research and practice in this field. Research proves that usage of an intervention is crucial to benefit from its potential effectiveness. Furthermore, dose-effect relationships are reported in various studies. Nevertheless, research on attrition has only recently gained the deserved attention whereby important concepts are defined. Drop-out attrition is defined as the percentage of respondents not following the research protocol (e.g. not filling out post-test questionnaires) and non-use attrition is defined as the percentage of respondents who have not used the intervention. Closely related is the concept of adherence, which is defined as the percentage of respondents following the intervention protocol (e.g. completing all modules). Research has been undertaken into characteristics of respondents which might explain attrition and adherence, but there is little research into how technology engages users. One of the main advantages of eHealth is the opportunity technology offers to persuade people to use an intervention. It seems plausible that the same characteristics that make eHealth so volatile might provide solutions to engage and captivate users. Objective: Many studies have been conducted on eHealth interventions. Most of these studies have focused on the effectiveness of an intervention, but include information on the features and functions of the intervention and on attrition rates. Our objective is to systematically review the literature to accumulate and synthesize this information into conclusions on how technology contributes to attrition and adherence. Methods: A comprehensive literature search was conducted using the databases Web of Knowledge, EBSCOHOST, Picarta, Scopus and ScienceDirect. The databases were searched for a combination of the constructs ‘web-based’, ‘intervention’, ‘attrition’ and ‘health’. For each construct, several (6 – 19) keywords were used. Important inclusion criteria were that the intervention was intended to be used repeatedly and was focused on patients and that the research included assessment of effect and information on usage of the intervention. Results: The search yielded 4939 unique titles, of which 347 articles were deemed initially relevant. After title, abstract and full text screening by two researchers, 84 articles were included. Articles studying the same intervention were combined and this yielded a total of 53 unique interventions. First analyses of these interventions show that a variety of features are used of which (psycho)education and tailoring are most common. Adherence is usually low to moderate, with interventions using chronological modules scoring more positively than interventions without a modular structure. Very often, the information on the intervention is insufficient and information on drop-out, non-use and adherence is scarce and unclear. Conclusions: This systematic review provides an opportunity to gain insight into the relationship between the technology used in eHealth interventions and attrition. Thereby it contributes to an emerging understanding and usage of persuasive technology in eHealth. However, research is hampered by unclear reporting of attrition rates and a lack of information on the features and functions of the researched interventions. []