The art of persuasion: designing devices for patients who don’t want to adhere

Published on: 17th December 2015

By some estimates, half of patients with chronic medical conditions fail to take their drugs as prescribed. For many, this is a conscious decision. Intelligent design of drug delivery devices can help to change this.

Within the context of drug delivery devices, the conventional approach to improving therapeutic compliance is often to think in terms of reminders or dose counters. While these devices have a clear role to play, they only tackle one part of the problem – unintentional non-adherence. They do little to explicitly challenge patients who elect not to take their drugs as prescribed. The scale of the problem is hard to quantify, but there is evidence that a significant proportion of patients with chronic medical conditions actively decide not to comply with their prescribed treatment; some estimates attribute intentional non-adherence as high as 70% of the issue 2 .

There is a consensus that the reasons for intentional non-adherence are complex and often idiosyncratic, meaning that it is unlikely that any single intervention will ensure that all patients take their medication as directed. Systemic solutions are needed that help patients to better understand and engage with their therapy, along with drug delivery solutions that cater for the patient’s emotional, as well as physical needs.

This article will explore what device developers can do to confront these issues within realistic commercial constraints that tend to favor ‘standardized’ device solutions.

The Non-Adherence Problem

Medication non-adherence is one of the biggest challenges facing health care providers. It is incredibly difficult to determine exactly how many people are affected, or why patients are not taking their drugs as prescribed, however, current estimates from the World Health Organisation 1 are that 50% of patients around the globe, with long-term illnesses, do not take their medications as prescribed. In the USA alone, it is estimated to cost $290 Billion a year 3 , cause 125,000 deaths annually, and account for 10% to 25% of hospital and nursing home admissions 4 .

Intentional or Non-Intentional?

Unintentional non-adherence is usually related to some form of forgetfulness or confusion. The result is that patients may forget to take a dose, take the wrong amount, repeat a dose, or take the incorrect drug.

Intentional non-adherence, on the other hand, relates to situations where patients are aware of what drugs they should be taking and when. However, they decide that they do not wish to take their drugs as prescribed. This may mean that users fail to take any of the prescribed drugs, end a course prematurely, or take a different dose to that prescribed.

Behaviour change

Ostensibly, the intentional non-adherence challenge is one of behavioral change. One commonly adopted approach is to address this challenge by viewing patients as the problem. Designs then focus on making patients more motivated, preventing them from doing something, or to persuade them to do something else. Fear tactics are one example of this; however, their efficacy is questionable, moreover, they can lead to patient anxiety, often in individuals who have no issues with adherence.

An alternative approach is to view the patient as a rational decision maker that has absorbed the information provided to them, considered it in the context of their particular situation and decided to depart from the prescribed drug regimen. Like the reasons for non-adherence, the way in which individuals make decisions are largely idiosyncratic. However, there have been many attempts in the past to describe decision making activities. One model commonly used is the OODA loop. This describes a feedback loop where decision makers observe the information available to them, they then Orientate this information to their own lives and the specific context, Decide which of the available actions they should adopt, and then act.


Learning from decision making theory, it is evident that the presentation of information alone does not change user behavior, rather it is the process of interpretation, or orientation, that is key. In order to both gain and maintain user engagement, users need to be able to relate the information provided to their own lives and the specific context of use. Accordingly a key opportunity in helping users to adhere to their drug regimen lies in the orientate phase of this decision making cycle.

What is clear is that much of the patient information currently provided is often not being read or not being understood. A Danish study 5 ;found that 40% of elderly patients did not understand the purpose of the drugs they were taking, while only 21% understood the implications of the omission of a drug or dose. This is perhaps unsurprising when viewed in the context of how many drugs some people are taking – with 60% of over 65s in the US taking five or more medications 6 .

A recent study assessing the impact of text messaging on adherence 7 ;also provides some interesting insights. The study involved 303 participants; half of which received SMS alerts prompting them to take their drugs, while the other half, the control group, received no intervention. The first insight is that the reminders were helpful; in fact 60% reported that they were reminded at least once to take a dose that they may have forgotten. This is perhaps unsurprising, but validates the assertion that reminders have a role to play in the adherence challenge.

What gets a bit more interesting is what else the study found. When comparing drop out rates (i.e. those either stopping completely or taking less than 80% of their medication), it is apparent that the drop out rate was considerably lower for those in the test group (25% in the control group compared to 9% of those receiving text messages). Further examination reveals that those in the test group were asked to text back with any concerns that they might have. 15% of the test group reported concerns on at least one occasion, because of uncertainty over the need for treatment, concern over side-effects, or another medical illness. Each case was followed up with a telephone call to address these concerns. This intervention, simply by calling them up and explaining these concerns, resulted in 87% (13% of the test group) resuming treatment.

These findings of this study suggest that actively addressing patient concerns and uncertainty can have a marked impact on adherence rates. For device manufactures the next logical question should therefore be how can medical devices, and the ecosystems that surround them, be designed to support this.

What Can Device Developers Do?

As a systems issue, health care providers are well placed to make changes to adherence levels. In many cases, there are opportunities to simplify drug regimens by viewing them holistically and reducing the number of drugs required. Likewise, the way patients receive their drugs can be simplified, a number of pharmacists are now offering clearly labelled sachets containing all of the drugs a patient should take at a given time.

Many people in the drug development and distribution chain also have their role to play in increasing adherence. Within the marketing team, the way drugs are presented and even named have the potential to impact adoption, packaging design will undoubtedly have an impact on the way drugs are perceived, likewise the way drug devices look, feel, sound, taste, and function will all shape perceptions.

Focusing on medical device development, and more specifically products that patients are using to self-manage their regimens, there remains much that can be done. As highlighted in the text message study, engagement has a key role to play in tackling the adherence challenges. This challenge can be split between gaining and maintaining engagement.

Maintaining Engagement

The concept of seeking to maintain engagement is often more familiar to device developers, involving topics that are well understood by those familiar with good usability engineering practice. This involves minimizing the impact on patients’ lifestyles. Common tactics include focusing on:

  • Convenience and flexibility of use (products should generally be unobtrusive on users’ lifestyles. Extending the time between doses and flexibility in the timing of doses. They should be transportable, allowing users to use them, and safely dispose of them in a wide range of scenarios. Patients should be supported in managing their drug regimen)
  • Time taken (devices should minimize the time required to set up, use, and safely dispose)
  • Complexity of use (devices need to be intuitive to use – matching the cognitive abilities and expectations of the target users)
  • Physical effort and comfort (the forces and postures required to deliver a drug need to be carefully considered and controlled)
  • Capturing and sharing information on adherence (can the device system communicate progress to the patient, providing feedback and reward? Where appropriate, can the system capture the level of adherence for health care providers or carers)

Gaining Engagement

Understanding the patient’s initial decision making process is central to gaining engagement. This is a topic that tends to receive less attention in standards (e.g. IEC 62366) and guidance documents (e.g. HE75).

Playing the decision cycle described in the OODA loop backwards can reveal some interesting insights. Explicitly considering the decisions that users are making, and how their specific view of their condition and context shapes this, is critical. Ultimately, the way we design and present information to users should be driven by a clear understating of the way they orientate themselves with this information.

In this context, the ‘information’ goes well beyond written instructions. It includes all sensorial aspects of the device that patients are interacting with. The form, the color, the material, the way it feels, and the way it responds are all information prompts that shape the way users decide whether to engage.

A second opportunity lies in the decision making section of the model. Carefully controlling the options available to the patient can also assist in the process. Pre-metered doses or treatments in a single pill may help to reduce patients under dosing or overdosing.

There a number of activities that can help device developers to understand the orientate phase.

  • Understanding the context (understanding the condition and the system of therapy, considering how and when patients are engaging with this product and other products used in conjunction)
  • Developing an emotional connection (Getting products out of the medicine cabinet. Designers and developers should ask what makes users love and engage with a product and not want to hide it away. There are many lessons here that can be learnt from how traditionally ‘taboo’ consumer products are now packaged)
  • Information that the users can relate to (Devices should be designed to support simple instruction. Instructions should be limited to major points presented using clear, everyday language and photographs or pictograms, covering why they should follow each instruction, along with how.)


What Does this Mean for ‘Standardized’ Devices?


Based on the guidance captured thus far, it is apparent that devices that resonate with individual needs, or certainly the needs of sub-groups of the population are important. This may be in the form of fun ‘funky’ products for children, or ruggedized products for those who are keen to take part in outdoor pursuits. However, this requirement for multiple variants of the same device presents significant challenges for medical device manufacturers. Even subtle differences in color can mean separate regulatory submissions – resulting in additional cost and potential delays in getting drugs to market.

The result is that, in most cases, a single product must be found that balances the needs of its diverse user base.  This requires detailed consideration to ensure that product can be both standard while still meeting the needs of as wide a patient group as possible.

Arguably, the challenge of maintaining engagement is far better understood and well captured in guidance (e.g. HE75). The FDA focus on safety and efficacy seeks to ensure that the physical forces required and the complexity of use is appropriate for the user population. However, the process prescribed in IEC 62366 does far less to encourage explicit consideration of how devices can be optimized to gain engagement. For that, a different focus is required.

There is a wide range of tools from the fields of human factors and design research that can help to structure this focus. Techniques such as ethnography and semi-structured interviews allow device developers to gain a richer insight into the lives of device users. Furthermore, they can help inform how an emotional connection can be established. In addition, a detailed understanding of decision-making psychology can help structure what information is required, along with where, when and to whom it should be displayed.



1 World Health Organisation report (2003). Adherence to long-term therapies: evidence for action. ISBN 92 4 154599 2
2 Reid. K (2012). The Heart Of The Non-Adherence Epidemic. Available at accessed 06/07/14
3 CVS Caremark (2012). State of the States: Adherence report
4 Smith DL. Compliance packaging: a patient education tool. Am Pharm. 1989;NS29(2):42–45. 49–53.
5 Barat I, Andreasen F, Damsgaard EM (2001). "Drug therapy in the elderly: what doctors believe and patients actually do". British Journal of Clinical Pharmacology 51 (6): 615–622.
6 Belcher VN, et al., (2006). View of older adults on patient participation in medication-related decision making. Journal of general Internal Medicine. 21 (4): 298-303.
7 Wald DS, Bestwick JP, Raiman L, Brendell R, Wald NJ (2014) Randomised Trial of Text Messaging on Adherence to Cardiovascular Preventive Treatment (INTERACT Trial). PLoS ONE 9(12)