Educational/counseling model health careHow to develop web-based decision support interventions for patients: A process map
Introduction
Over the past decade, considerable progress has been made on the development of ‘decision aids’, which we refer to using the generic term decision support interventions [1]. Trials of these interventions have demonstrated consistent effects on knowledge and participation in decisions [2] although how best to measure the concept of a good decision remains elusive. The International Patient Decision Aids Collaboration (IPDAS) has developed a set of standards for their development and evaluation [3], [4]. There are also proposals to use generic methods to develop patient decision support interventions based on clinical practice guidelines [5]. Since ‘decision aids’ were first defined in 2000 [6], advances in the use of information technology and new ideas about the role of decision support have highlighted the need for further clarification. Further work on the definitions and on the purpose(s) of decision support interventions would allow progress on how to assess their contribution to decision making processes. O’Connor's initial definition emphasised the role of information [7], whilst more recently Elwyn et al. noted the need to provide a basis for accurate affective forecasting and preference construction [1]: helping patients assess how they might feel about potential future events and how to integrate information and construct their preferences.
The role and application of theory to the development of decision support interventions remains problematic [8]. Most decision making theories operate at cognitive levels, and focus on explaining or describing how humans cognitively approach and achieve a decision rather than on how tools could be designed to support decision making. A review of the potential contribution of key decision making theories and notes the lack of a mid level theory which addressing the role and contribution of decision support interventions [9], highlighting the difficulty of integrating theoretical constructs into interventions to influence healthcare processes and encounters. There are also difficulties in (1) ensuring valid and reliable needs assessments that integrate and balance the views of multiple stakeholders; (2) in agreeing the menu of treatment options in the face of shifting evidence base and policy-based appraisals of cost-effectiveness. These challenges all need to be addressed before we can make confident progress in the field.
Furthermore, unprecedented and increasing access to the Internet, including people from lower socioeconomic strata, is game-changing development. The Internet has become the main resource for information and interactive applications, both for patients and for professionals. Developers of decision support interventions will need to adapt and evaluate how to make best use of the Internet. Here are some of the challenges that must be met:
- 1.
Mandatory or optional information navigation: should patients be required to take a pre-determined route through an information source, at least for a prescribed minimum content or should navigation be entirely under their control? If optional navigation is allowed, how should information be organised and layered? Empirical assessments indicate that average users have short attention spans when using websites and do not spend more than 20 min or so using a web-based decision support intervention [10]. Who determines the information that has maximum validity and relevance?
- 2.
Interaction: the Internet provides new challenges to make maximum use of interactive elements: audio, video, gaming technology, avatars etc. There are early attempts to implement interactive tools (e.g., deliberation tools, use of video narratives) but these are pioneering efforts that have not been subjected to extensive evaluations [11], [12]. Their potential to support affective forecasting and preference construction is clear but as yet unrealised [13].
- 3.
Over-engineering: the availability of new interactive methods brings potential pitfalls into the design of decision support interventions. There are trade-offs between what is possible, what is necessary and what is helpful. Decision support interventions need to be both informative, easy to use, absorbing and to add value to the deliberative efforts of users. We do not have sufficient evidence on how to best address these aims and it is possible that the addition of avatars and different interactive tools may not add value.
A research group at Cardiff University has developed a number of web-based decision support interventions. Prosdex (www.prosdex.com) was developed between 2003 and 2006 to support men considering the prostatic specific antigen test. The website is currently recommended in UK guidelines from the NHS Cancer Screening Programme. Amniodex (www.amniodex.com) was developed between 2006 and 2009 to support pregnant women and their partners considering amniocentesis for Down's Syndrome. BresDex (www.bresdex.com) was developed between 2007 and 2010 to help women with breast cancer decide whether to have lumpectomy or mastectomy. Each intervention was designed by following a similar development process. The aim of this article is to share our experiences, to examine the challenges faced by developers and to propose a systematic process map for the development of web-based, perhaps multimedia, decision support interventions.
Section snippets
Process framework
Developing a high quality web-based decision support for patients is a complex, expensive and time-consuming process which requires the co-ordination of many stakeholders and processes. We developed a framework to guide this process. The framework is composed of a set of documents and tasks, coordinated by a Project Management Group, with the support of other groups that provide different forms of expertise (see Box 1, Fig. 1, Fig. 2).
Results—developing web-based decision support interventions: a process map
There are two substantive phases: a content specification phase and a design phase. Both these phases contain a number of steps, and although these can be represented as occurring in sequence, there is a significant amount of overlap that may occur between steps.
Discussion
Based on the experience of developing several web-based interventions, we can outline some general principles: the phases of content specification followed by creative design and tailoring to the target audience. The content specification is fundamental and should go beyond the traditional emphasis on scientific evidence to establish patients’ perspectives on the salience and relevance of the proposed options. The creative phase is less governed by specification. Using an iterative method of
References (21)
- et al.
A generic tool for development of decision aids based on clinical practice guidelines
Patient Educ Couns
(2008) - et al.
Dual equipoise shared decision making: definitions for decision and behaviour support interventions
Implementation Sci
(2009) - et al.
Decision aids for people facing health treatment or screening decisions
Cochrane Database Syst Rev
(2009) - et al.
Developing a quality criteria framework for patient decision aids: online international Delphi consensus process
Brit Med J
(2006) - et al.
Assessing the quality of decision support technologies using the International Patient Decision Aid Standards instrument (IPDASi)
PLoS One
(2009) - et al.
A decision aid for women considering hormone replacement therapy after menopause: decision support framework and evaluation
Patient Educ Couns
(2000) - et al.
Decision aids for people facing health treatment or screening decisions (Cochrane Review)
The Cochrane Library, Issue 1
(2004) - et al.
Where is the theory? Evaluating the theoretical frameworks described in decision support technologies
Patient Educ Couns
(2008) - Elwyn G, Stiel M, Durand MA, Boivin J. Theory-based design of decision support interventions: towards real-world...
- Evans R, Joseph-Williams N, Edwards A, Newcombe RG, Wright P, Kinnersley P et al. Supporting informed decision making...