The Law of Attrition
By Gunther Eysenbach. Journal of Medical Internet Research, 2005
2,543 citations recorded by the source corpus.
In an ongoing effort of this Journal to develop and further the theories, models, and best practices around eHealth research, this paper argues for the need for a “science of attrition”, that is, a need to develop models for discontinuation of eHealth applications and the related phenomenon of participants dropping out of eHealth trials. What I call “law of attrition” here is the observation that in any eHealth trial a substantial proportion of users drop out before completion or stop using the appplication. This feature of eHealth trials is a distinct characteristic compared to, for example, drug trials. The traditional clinical trial and evidence-based medicine paradigm stipulates that high dropout rates make trials less believable. Consequently eHealth researchers tend to gloss over high dropout rates, or not to publish their study results at all, as they see their studies as failures. However, for many eHealth trials, in particular those conducted on the Internet and in particular with self-help applications, high dropout rates may be a natural and typical feature. Usage metrics and determinants of attrition should be highlighted, measured, analyzed, and discussed. This also includes analyzing and reporting the characteristics of the subpopulation for which the application eventually “works”, ie, those who stay in the trial and use it. For the question of what works and what does not, such attrition measures are as important to report as pure efficacy measures from intention-to-treat (ITT) analyses. In cases of high dropout rates efficacy measures underestimate the impact of an application on a population which continues to use it. Methods of analyzing attrition curves can be drawn from survival analysis methods, eg, the Kaplan-Meier analysis and proportional hazards regression analysis (Cox model). Measures to be reported include the relative risk of dropping out or of stopping the use of an application, as well as a “usage half-life”, and models reporting demographic and other factors predicting usage discontinuation in a population. Differential dropout or usage rates between two interventions could be a standard metric for the “usability efficacy” of a system. A “run-in and withdrawal” trial design is suggested as a methodological innovation for Internet-based trials with a high number of initial dropouts/nonusers and a stable group of hardcore users.
Subfields: Pharmacy
View Paper: https://www.semanticscholar.org/paper/6a6b1436ff134f6452e85f5498f7bba7815685e4
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Pharmacy
- Thomas Richardson
- Derek Richards
- David C. Mohr
- Stephen M. Schueller
- J. V. van Gemert-Pijnen
- Hans C. Ossebaard
- Saskia M. Kelders
- Linda Gonder-Frederick
Citations in the corpus
- Conceptualising engagement with digital behaviour change interventions: a systematic review using principles from critical interpretive synthesis (2016, background): “This kind of conceptualisation has emerged due to the observation that while many download and try DBCIs, sustained usage is typically low [21 – 24].”
- A Systematic Review of the Impact of Adherence on the Effectiveness of e-Therapies (2011, neutral): “It is thought that the variation in adherence and persistence may be due to the participant’s discretion in using e-therapy [26], rather than therapy that is prescribed, as in a drug trial [13].”
- Computer-based psychological treatments for depression: a systematic review and meta-analysis. (2012, methodology): “Dropout from treatment is a continued cause of concern, especially with interventions that offer no support (Eysenbach, 2005).”
- Realizing the Potential of Behavioral Intervention Technologies (2013, background): “Although empowering users to be more actively involved in the process is likely to increase agency (Moriarty, 2010), adherence is a limitation encountered by many BITs (Eysenbach, 2005), and we discuss it in more detail in the Adherence section.”
- Gratitude and well-being: a review and theoretical integration. (2010, background): “However, as these interventions are givenwithout human contact, attrition is commonly very high (Eysenbach, 2005), creating problems as the therapies are most effective when followed to conclusion.”
- Gamification of Cognitive Assessment and Cognitive Training: A Systematic Review of Applications and Efficacy (2016, background): “This is a particular problem for Web-based studies where participants can simply close the browser window and drop out of the study the moment they decide it is not worth their time [5].”
- Artificially intelligent chatbots in digital mental health interventions: a review (2021, background): “DMHIs are effective when used as recommended; however, these interventions have been criticized for low usage rates and high dropout [34–37].”
- Persuasive System Design Does Matter: A Systematic Review of Adherence to Web-Based Interventions (2012, background): “One of the issues that is frequently addressed is the problem of non-adherence [11, 13-17], which refers to the fact that not all participants use or keep using the intervention in the desired way.”
- Objective User Engagement With Mental Health Apps: Systematic Search and Panel-Based Usage Analysis (2019, background): “However, the impact of digital interventions is limited by their ability to engage users in therapeutic activities and to support user adherence to the therapeutic process [9,10].”
- Behavioral Intervention Technologies: Evidence review and recommendations for future research (2010, positive): “Adherence to Web-based interventions has been widely acknowledged to be a problem [25,28] and should constitute a standard outcomemeasure and a focus of new research.”
- A Test of Two Positive Psychology Interventions to Increase Employee Well-Being (2014, background): “Another potential cause for attrition was that all contact with participants was online; past research suggests that the lack of face-to-face contact can increase attrition (Eysenbach 2005).”
- Self-Compassion and Body Dissatisfaction in Women: A Randomized Controlled Trial of a Brief Meditation Intervention (2014, background): “According to Eysenbach (2005), attrition rates for studies conducted over the Internet tend to be much higher than for studies conducted through more traditional means, particularly with self-help applications.”
- The influence of social networking sites on health behavior change: a systematic review and meta-analysis (2015, background): “Indeed, retention rates of general SNS studies included in this review are very promising at around 80%, and they shed new light on the ‘law of attrition’ of online interventions.(46,63)”
- Development of a New Approach to Guided Self-Help via the Internet: The Swedish Experience (2008, background): “From a research point of view we found early that the risk of dropping out of treatment was high (Eysenbach, 2005).”
- Harnessing Social Media for Health Promotion and Behavior Change (2013, background): “Attrition, when participants stop usage or are lost to follow-up, has also been identified as a fundamental issue for evaluation of online interventions (Eysenbach, 2005).”
- Just-in-Time Adaptive Interventions (JITAIs) in Mobile Health: Key Components and Design Principles for Ongoing Health Behavior Support (2017, background): “” However, given the rapid, unexpected, and ecological nature of these states, as well as the law of attrition [54] in mHealth interventions, JITAIs in mobile health are also motivated by the need to address relatively rapid changes in mechanisms underlying adherence and retention.”
- CONSORT-EHEALTH: Improving and Standardizing Evaluation Reports of Web-based and Mobile Health Interventions (2011, methodology): “In addition to the traditional CONSORT flow diagram we also highly encourage the provision of an attrition diagram (CONSORT-EHEALTH item 13b-i) in the results sections, illustrating the login behavior of participants in all groups over time as a survival curve [19].”