Medication adherence fundamentals and AI solution to combat non-adherence – Part 3

Yashodeep Sengupta

Combating non-adherence:
Present measures and their pain points

Jim works in the health information technology (healthtech) industry. His latest assignment requires him to do some digging around in the field of medication adherence with an aim to zero in on a proven solution that can benefit his organization.

Here are the links to other blogs of this techies’ primer: part 1, part 2

In the last two parts of our blog series, we saw how Jim explored the non-adherence reality in the medication adherence scenario globally. His notes also included the key reasons for such widespread non-adherence by patients and how they ultimately impact every stakeholder in the cycle. In this blog, Jim attempts to discover the various market offerings to counter medication non-adherence and their limitations.

Since non-adherence impacts all stakeholders in the medication adherence ecosystem, providers, pharma, payers, etc. have already started their efforts to buck the trend of rising non-adherence. However, there is no gold standard approach to ensuring medication adherence. While some measures can be direct in nature involving invasive tests, indirect methods include human interaction and manual monitoring. More advanced ways leverage technology to resist non-adherence.

However, the question arises: Can these steps really make a difference? Let’s explore.

Direct/ indirect steps to counter medication non-adherence

These efforts stand on the pillars of human initiation, implementation and persistence. These are either invasive or non-invasive in nature. Also, this approach involves either patient-driven initiatives (like questionnaires, diaries, etc.) or other stakeholder-driven (like direct observation, monitoring calls, etc.). It falls short of ensuring a considerably foolproof medication adherence. Let’s take a look at some of the key measures in this section and their limitations.

Direct observation

Direct observation includes observing a patient’s medicine-taking behavior. Touted as an impractical way to monitor medication adherence, routine inspection becomes challenging as patients can hide medicines under their tongue and discard them later, and so on. This is used more with patients with single-dose therapies and ones with frequent admissions.

Invasive monitoring

Checking levels of certain drugs in bodily fluids like urine or blood is another way to combat non-adherence. However, there are many limitations to such efforts. For one, patients may find the process intrusive and feel under-confident and nervous about participating in the procedure. Secondly, the test results may not always present the real picture. For example, some drugs can be found in the blood long after stopping medication. On the other hand, certain drugs, like riboflavin, can’t be detected in the body. Additionally, white-coat adherence or “improved patient adherence to treatment around clinic visits” can present a false adherence reality. According to a report, average adherence rates of a group of patients before and after visiting the hospital stood at 88% and 73%, respectively.

Awareness/ monitoring calls

As the name suggests, this step is about calling up/ on patients to check on whether they have been taking medicine doses as prescribed by the medical practitioner or refilled at the pharma centre. The calls can also be about creating awareness about the disadvantages of medication non-adherence. A part of Centers for Medicare & Medicaid Services’ (CMS) Medication Therapy Management (MTM), it ties health plan payments to MTM measures via its Star Program and reimburses MTM activities. However, this measure can be perceived as an intrusive activity, and patients can just hang up or not respond or present false reports. In such circumstances, following up or monitoring becomes a tough task. Any conclusion reached through such calls is flawed more often than not.


Under this low-cost and simple step, patients are encouraged to record their medicine ingestion during the prescribed duration of the course. First, it is completely voluntary where a patient may back out due to disinterest or just be forgetful or careless about it. It also runs the risk of patients manipulating facts or presenting untrue information, with no way to verify it.

Questionnaires and clinical responses

There are a number of popular questionnaires available that ask patients or providers to record their medication adherence efforts to get a sense of how they are doing, through scores. Similarly, patients are made to fill out clinical forms and associated surveys that help providers/ pharmas/ payers gauge how much they adhere to the prescribed medication. Patient interviews with practitioners are yet another way of assessing medication adherence. However, all these processes run the risk of presenting inaccurate results due to bias that may creep in while patients record their responses. Communication gaps between the questioner and the answerer also lead to misinformation. There is no way to ensure that the responses provided by the answerer are true or not. Medication adherence improving aids like calendars, drug cards, medication charts are also not foolproof because of similar behavioral reasons.

Pill count

Pill count is a strategy where patients are required to – either routinely or randomly – visit their medical clinics with their pill strips/ bottles, which inform whether the patient has been taking the prescribed medication properly or not. It is represented by the equation: Number of dosage units dispensed − number of dosage units remained / prescribed number of dosage units per day × number of days between 2 visits. Needless to say, there have been widespread reports of fudging, like switching medicines between bottles, emptying of pills by non-adhering patients before such visits, etc. Patients with chronic ailments can also refill their medicines before running out. This, of course, does not present a real picture of adherence. Moreover, it is an ineffective method with patients with non-discrete dosages.

Prescription refill rates

A part of secondary database analysis, in this method, pharmacies and providers assess medication adherence by comparing prescriptions with refill rates, and also by accessing pharmacy records. It is a very simple calculation method which commits the grave error of mistaking medication possession with medication intake. Gaps in refill are ignored and the need for prolonged therapy with multiple prescriptions is not considered. Mostly, overestimated adherence values lead this method to be inaccurate and misleading.

To counter the limitations of the above methods, a multimeasure approach that takes into account more than one of the measures has been suggested as an alternative. However, the inherent subjectivity, biasness and other behavioral tendencies make this approach weak as well.

View the full infographic here

Digital adherence measures

Like in every aspect of general human life, technology has also found its way into medication adherence measures. With the promise of artificial intelligence (AI), stakeholders have hopped on to the digital bandwagon. Technology-based measures available in the market to combat medication non-adherence are looked upon as new-age, more convenient, user-friendly and insightful, and less invasive. However, given its commercial mass-market nature, lack of customizability and inherent limitations, digital offerings too fail to make the cut.

Let’s look at some key digital offerings available in the market and explore their shortcomings.

Smart pill containers

Nowadays, there are pills that come in bottles and strips with embedded microprocessors that record the time and date whenever a medicine is removed. There are dispensers available too that link up with home voice assistants for reminding and recording functions. Termed as Medication Events Monitoring System (MEMS), these are smart ways to digitally remind patients about medicine intake and record medicine intake. But the principle assumes that whenever a dose is removed, it is consumed by the patient at the particular time. Therefore, the chances of deception are high. There also might be cases when the patient opens the bottle inadvertently, which results in overestimation of adherence. While the monitoring nature of the bottles can deter patients from following the medication, others might transfer contents of a bulky smart bottle to a smaller container for regular use. In these cases too, monitoring correctly becomes a herculean task. Other issues with this measure include high cost of the containers, data integration and software downloads for data retrieval.


These are devices that come embedded on an oral drug, which, when consumed, get dissolved and send alerts to the patient’s and provider’s smartphones through a wearable patch on the patient. So far, there is only one FDA-approved bio ingestible sensor in the market. There are also wristband sensors that capture the motion of a patient’s hand taking medicine out of a container and ingesting it, and send reminders in case of deviations or delays. The deterrent to this measure is its cost viability and various unknowns, including effectiveness on a large population, accuracy of insights, etc. So, in terms of innovation and acceptability, there’s a lot more to be done.

Medical apps

Mobile apps to track and assess medication adherence is seen as a cost-effective and scalable solution. While some apps send text alerts or deploy a chatbot assistant to remind patients about taking prescribed doses of medication, others incentivize medication intake through gamification by offering discount coupons, in-game and encashable reward points, cash discounts, etc. There are also mobile apps that live-capture patients consuming medicine to keep track of their adherence. Although easy and convenient, the measure is not completely foolproof. While there are cases of collected data being resold, most apps are standalone and not integrated to a larger platform. Reminders, financial incentives and gamification are certainly novel, but not a sure-shot way of ensuring medication intake and adherence.

What’s next?

Jim wonders: “It seems most of the existing offerings have drawbacks in containing medication non-adherence effectively.” Tech being the smartest way forward, what stakeholders need is a one-stop digital solution with a high efficacy rate in overturning the non-adherence scenario and its consequences.

To find out the best bet for medication adherence, check out our concluding blog of the series.

Find out before Jim

Do you want to have fast track access to the technology solution Jim eventually finds out to solve medication adherence woes? Learn more.


Your email address will not be published. Required fields are marked *