Devising ways to improve medicines adherence

An ingestible sensor made from magnesium and copper is embedded in a pill of the drug aripiprazole. Once the atypical antipsychotic has been ingested, it hits the acidic juices in the stomach and a tiny voltage charge is created that transmits a signal to a patch worn by the patient. The data is then transmitted to an app on the patient’s mobile phone and, with consent, to their doctor or carer. Other patient metrics, including rest, body angle and activity patterns, can also be tracked.

This technology is being designed by Proteus Digital Health and has been added to Otsuka’s Abilify to help patients suffering from severe mental illnesses who may struggle to adhere to their medication regimen and to communicate with their healthcare team about it. Any doubts or confusion about whether a patient has taken their medication are avoided thanks to the record.

This is the first in a new type of ‘digital pill’ and is currently being considered for approval by the US Food and Drug Administration (FDA) although the ingestible sensor part was approved as a medical device by the FDA in 2012. Proteus’s digital medicine pipeline includes six of the most commonly prescribed generic drugs for cardiometabolic conditions.

Digital pills are one approach for a condition where lack of adherence could have adverse consequences, says Carmel Hughes, a pharmacoepidemiologist at Queen’s University Belfast. But she expects that this type of technology will be “a niche area” primarily for psychiatric conditions rather than a panacea for all non-adherent patients. “Affordability will be key for such a technology, and it would probably need to be targeted at those who were non-adherent, and those who would be willing to accept this level of scrutiny,” she adds.

Non-adherence is usually far more complex than a patient simply forgetting to take their medicine. Reasons for it are “highly personalised”, with around 200 individual factors in play, says Hughes. There has been growing interest in psychological approaches to find out why a patient is not taking their medicines, including frameworks such as the capability, opportunity and motivation (COM-B) model of behaviour, so that appropriate interventions can be chosen to address the cause.

Widespread problem

Non-adherence remains a widespread problem. A Cochrane review in 2014 estimated that medicines adherence rates average at around 50%, and have not changed substantially for the past 50 years^[1]
.

Poor adherence has adverse implications for both patients and the healthcare system, says Hughes. Non-adherent patients will not have control over their condition, which can result in avoidable admissions to hospital, she explains. This in turn can lead to further costly medicines being prescribed by a doctor who is unaware that non-adherence to current medication is the problem.

The IMS Institute for Healthcare Informatics estimates that US$269bn could be saved worldwide by improving adherence to medicines.

Reasons for non-adherence

Ascertaining the level of adherence to treatment in a particular patient group is difficult since, digital pills aside, there is no gold standard mechanism for measuring it. “You can ask people, but you are relying on them to tell the truth,” says Hughes. “Studies suggest that people overestimate their own adherence.”

Studies suggest that people overestimate their own adherence

Alternatively, you can check medication records — a proxy for adherence — but this is a measure of whether the patient collects their medicine, not whether they take it. Patients can also be asked to bring their medicines to a healthcare professional so that their pills can be counted but, again, this tactic is open to ‘gaming’.

“Quantifying non-adherence may be of interest to people doing research, but in clinical practice the important question is why someone is non-adherent rather than to what extent,” says Johnson George, a medicines adherence researcher at Monash University, Melbourne, Australia.

While most self-reported measures for assessing adherence cannot distinguish between different types of non-adherence^[2]
, some do give limited information. For example, the Medication Adherence Report Scale was designed to identify barriers and problems to good adherence, and the Tool for Adherence Behaviour Screening (TABS)^[3]
 and the Necessity-Concerns Framework^[4]
 distinguish between intentional and unintentional non-adherence.

However, these frameworks do not indicate how to bring about behaviour change and, to date, many interventions to improve adherence have been employed ineffectively. The COM-B model hypothesises that capability, opportunity and motivation interact to generate behaviour and can therefore explain why a recommended behaviour is not being followed.

“COM-B is one of the frameworks where you can identify issues associated with that particular patient … and then attempt to develop interventions that will hone in on what is the big issue for the individual,” says Hughes.

Capability is defined as a patient’s capacity — both psychological and physical — to understand, remember and plan their treatment. Opportunity relates to factors that are predominantly remote from the patient, including healthcare provision, family support and social influencers. Motivational factors can be conscious and subconscious, and relate to the patient’s beliefs about the potential for side effects, the effectiveness of the treatment and the seriousness of their condition, their anxieties and mood^[5]
. A given intervention needs to change one or more components in the model (see ‘Applying COM-B to factors associated with adherence’).

However, COM-B is labour intensive and time consuming to use so it is geared towards health providers at a strategic level as an aid to improving quality of care, and for research purposes, says Robby Nieuwlaat, author of the Cochrane review on adherence, and a clinical epidemiologist at McMaster University, Hamilton, Canada. Nevertheless, he points out that it is possible for health professionals to ask questions based on the framework to determine the likely cause of non-adherence and identify the most appropriate intervention.

There would not be enough time to use a framework such as COM-B to diagnose the problem, identify appropriate behaviour change interventions and deliver them within one face-to-face consultation, says Hughes. There are around 93 separate behaviour change techniques that could be implemented^[6]
, she points out. Examples include self-monitoring of behaviour, for example, keeping a diary, providing information about the health consequences, implementing prompts or cues and restructuring the social environment.

“Healthcare professionals who are going to be delivering interventions are going to need to be trained in the use of some of these techniques,” she says. “It is quite a different approach to what has been done in the past, where you perhaps have a brief conversation with the patient about how they are getting along with their medicines and whether everything is OK, and that’s the end of it.”

Healthcare professionals who are going to be delivering interventions are going to need to be trained in the use of some of these techniques

What is needed is a practical and broadly generalisable process for assessing adherence that can be easily rolled out to the health service, says Hughes. For example, there has been interest in the use of brief interventions for alcohol and other addictions, and Hughes can see a role for these types of techniques in medicines adherence.

Opportunities to improve adherence

The new medicine service (NMS), provided by community pharmacists in England, is the ideal forum for assessing adherence, says Nick Barber, director of research and evaluation at UK charity The Health Foundation, although he admits that some people will misrepresent their medicine-taking behaviour. “If we pick up 80% of people who are non-adherent but 20% misrepresent that — that is still a hell of a health improvement that we could deliver.”

The NMS, which was introduced in 2011, is performed one to two weeks after a patient is prescribed a new medicine in one of four therapy areas — hypertension, type 2 diabetes, anticoagulation or antiplatelet therapy, and asthma or chronic obstructive pulmonary disease. The pharmacist talks to the patient to find out whether they are taking the medicine, if they consider it to be working, and if they are experiencing side effects. An appropriate intervention can then be recommended.

An evaluation of the NMS published in August 2014 found that it increases the number of patients who are adherent to their treatment by about 10%^[7]
. Furthermore, economic modelling showed that the benefits delivered by the service were below the cost per quality-adjusted life year (QALY) thresholds used by England’s National Institute for Health and Care Excellence in technology appraisals.

Barber is optimistic that the service will be extended to new drug groups. However, he points out that adherence changes over time so it is likely that further interventions will be needed.

A study looking at the impact of pharmacists providing a telephone service modelled on the NMS to patients on long-term treatment found that this approach also improves adherence. A group of 677 patients receiving type 2 diabetes or lipid-lowering medicines from a mail-order pharmacy were randomised to receive the intervention — two telephone consultations, four to six weeks apart — or to a control group. Patients who received the intervention were more likely to be adherent (defined as ≥90% of medication taken in the past seven days) at four week follow up, compared with the control group (adjusted odds ratio 2.20, 95% confidence interval 1.33–3.65, P=0.002)^[8]
.

Once the likely cause of non-adherence has been pinpointed, selecting an appropriate targeted intervention is a logical process. A patient who lacks confidence in the drug or in what the doctor has said, or fears side effects, may need an intervention that involves counselling, education or motivational interviewing. And for those who are not adhering because they are experiencing side effects, a change of dose, regimen or treatment may be the answer.

If the complexity of a patient’s medication regimen is identified as a barrier, simplifying it can help. Research from George and his colleagues at Monash University found that medicines reviews conducted by hospital pharmacists for certain patients reduces the complexity of their regimens. In the study, 205 patients aged over 60 years were given a medicines review intervention and the complexity of their regimen on discharge was compared with that of 186 patients who received usual care. Patients who received the medication review had less complicated regimens — equivalent to ceasing one or two medicines^[9]
.

Researchers at the university are also evaluating the impact of using automated medication dispensers in the homes of elderly people who need to follow complex medicine regimens and cannot cope with a monitored dosage system. The medicines are put in chambers in the dispenser and, when it is time to take a dose, an alarm goes off and the required medication is dispensed in a sachet.

In an increasingly elderly population when treatment is often for several comorbid conditions, multiple medicines use is becoming the standard of care and this presents problems for patients, says Hughes.

“We need to look at how we can best help patients manage multiple drugs and also multiple different types of formulations, all of which could possibly require slightly different behaviours,” she says. A patient could be using a patch, a liquid, tablets and an inhaler, and this can create problems for older people experiencing difficulties with flexibility, eyesight and memory.

Hughes believes there needs to be a much more holistic approach to managing medicines, “right from the formulation of that medicine to the taking of that medicine”. Polypills and digital pills are likely to play a part, she suspects, but will not resolve the problem alone.