Keep Calm and Take Your Meds!
Why non-adherence to your prescription is bad and how 1mg can be your trusted saviour.
Abstract. Patient non-adherence is a major barrier in effectively meeting the objectives of an established treatment plan. Non-adherence not only jeopardises the opportunity for the patient to recover completely from the disease, the consequences of non-adherence and their subsequent remediation add costs and pressure to the already stretched healthcare system, thereby having an amplified negative effect. The objective of this article is two-fold: Firstly, to highlight the consequences of non-adherence, especially from a pharmacokinetic perspective and secondly, to highlight a technological solution that can address this problem in a holistic and multidimensional way.
1. Introduction
All things are poison, and nothing is without poison. … It is the right dose that differentiates a poison from remedy”
Patient non-adherence is a significant medical challenge with rates of non-adherence being as much as 30% for short-term and as much as 60% for some long-term therapies, [1]. The economic cost of non-adherence to the public healthcare system is massive and is a global problem. For example, poor adherence costs the United States of America in excess of $100 billion annually in preventable healthcare spending, [2]. In rural parts of India, rates of adherence are believed to be considerably worse than those in the urban areas. A variety of reasons are behind non-adherence, such as:
- forgetfulness,
- cost of the medicine,
- inability to visit a pharmacy,
- experiencing adverse events,
- and complexity of the prescribed treatment plan.
Patient non-adherence typically manifests through two actions: not following the prescribed time interval between two successive doses and omission of one or more than one of the recommended number of doses. Some drugs are ‘forgiving’, that is, the duration of their therapeutic action is more than double of the time interval between two successive doses, which gives them a certain degree of robustness against non-adherence of the patients towards the prescribed dose regimen, [3]. But even for such drugs, a large degree of non-adherence may lead to therapeutic failure.
The adverse effects of non-adherence from a patient’s perspective are not difficult to imagine — opportunity loss to recover from the disease and harsher interventions that might become necessary at a later stage because the original treatment plan can no longer suffice. For instance, failure to follow the prescribed dose regimen for antibiotics can not only result in treatment failure, it can also lead to the development of antibiotic resistance which is a major health risk for the entire community. For Type 2 diabetes, poor adherence to the prescribed medication has been proven to be associated with inadequate glycemic control, increased morbidity and mortality, and increased costs, [4].
When your doctor prescribes a dose regimen for a medicine, a lot of science goes behind in designing of that dose regimen. Dose finding is a difficult problem and the onus of determining the optimal dose regimen — one which balances the twin objectives of safety and efficacy — lies with the pharmaceutical industry. In order to win marketing approval from the regulator, the company must demonstrate through clinical data that the recommended dose regimen is optimal, that is, a sufficiently strong therapeutic action exists and the prevalence of the adverse effects is within an acceptable limit.
An optimal dose regimen has the optimal number of doses, the correct dose sizes and the right dosing intervals. Furthermore, the pharmacokinetics (PK) — the science which studies the speed at which a drug gets absorbed, distributed, metabolised and eliminated from the body — varies from patient to patient. Even for the same patient, the optimal dose is a function of the bodyweight and the surface area. Which explains why children are generally prescribed smaller doses as compared to the adult doses of the same drug.
Pharmacokinetics (PK) is the science which studies the speed at which a drug gets absorbed, distributed, metabolised and eliminated from the body. In other words, PK describes what the body does to the drug.
Non-adherence is, thus, an antithesis to the science of optimisation of the dose regimen. In this article, we show through an illustration how non-adherence hampers the optimality of a dose regimen and how 1mg app can address this problem in a multidimensional way.
2 Importance of Adherence
2.1 Optimisation of a Dose Regimen
As mentioned in Section 1, a pharmaceutical company has to determine the dose regimen that is both: safe and effective for the test drug in order to get marketing approval. The optimal concentration of the drug that is required to be maintained in the blood is often known, the challenge is to find the dose regimen which enables that.
[5] present an algorithm which enables computation of an optimal dose regimen while estimating the parameters of the PK model which describes the concentration-time relationship of the drug, say, C(t). An optimal dose regimen is one which enables the concentration of the drug in the blood to be as close as possible to the ideal target concentration, say, Ctgt, for a time period, T, as shown in Figure 1. The areas where C(t) is greater than Ctgt are the areas of over-exposure to the drug (marked with a ‘+’ sign) and the areas where C(t) is smaller than Ctgt are the areas of under-exposure to the drug (marked with a ‘-’ sign). Large areas of over-exposure would be associated with toxicity and adverse effects while large areas of under-exposure would be associated to treatment inadequacy and potential disease escalation. The objective function that is optimised by [5] is the sum total of these areas of over- and under-exposure.
Let n doses of the drug are to be administered at times (0, t_1, …, t_n) with the corresponding dosing time intervals as (t_1, t_2 — t_1, …, t_n — t_(n-1)) = (v_1, …, v_(n-1)) (say). In [6], the author describes a simulation study with two instances of non-adherence as discussed below.
2.2 Dose Intervals not Maintained
This is the case when the patient deviates from the prescribed time point of ingesting the dose — taking it earlier or later. Let n = 5 and let the prescribed dosing time intervals be (v_1, …, v_4) = (8, 8, 8, 8) hours, that is, the 5 doses are prescribed at time points t = 0, 8, 16, 24, 32 h. Let the actual vector of time intervals for a patient be (u_1, …, u_4) where u_i’s, i = 1, …, 4 are drawn independently from a N(8,1) distribution (that is, a normal distribution with mean 8 and standard deviation of 1) for every patient. In this scenario, by definition of the normal distribution, 68% of the dosing intervals lie inside the interval [7, 9] h since for a N(µ, σ²) distribution, 68% of the area lies between [µ − σ, µ + σ]. Also, since the mean is 8 h, the simulated time points will be centred at the prescribed dosing time points. Therefore, the dose schedules followed by most patients will only mildly deviate from the prescribed time point, thus making this scenario a case of minor non-adherence. Figure 2 presents the distribution of the simulated dosing time points. As expected, the spread in the distributions increases in the latter time points because their variabilities get supplemented with the variabilities in the preceding time points. Figure 3 presents the simulated concentration profiles of a few patients.
As to the effect of this mild non-adherence, the dose regimen administered to the 10,000 patients was, on average, 6% less optimal (in terms of safety and efficacy) with a 30% increase in the variability as compared to the baseline case when all doses are ingested at the exact prescribed dosing time points. As mentioned before, since we considered a scenario of mild non-adherence, this shows that even minor departures can make the treatments sub-optimal for the patients.
2.3 Omission of Doses
In this scenario, we consider the case of dose omission, i.e., skipping of one or more doses by the patients. Let p, the probability of a patient skipping a single dose, be 0.5%. Thus, for a single patient, the probability of not ingesting at least one of the prescribed doses is 1 − (1 − p)^5 = 2.5%. Again, this is a case of mild non-adherence. In this scenario, the dose regimen administered to the 10,000 patients was, on average, 17% less optimal with a 167% increase in the variability as compared to the baseline case. In this example, it can be observed that skipping a dose has a much larger impact on the efficacy and safety of the treatment as compared to mild departures from the prescribed dosing time points.
3. A Technological Solution for Promoting Adherence
As discussed in the previous section, even minor departures from the prescribed dose regimen can have a large effect on the success of the treatment therapy. The prescribed dose regimen is chosen by the pharmaceutical company after satisfying the regulator with extensive clinical data and any departures by the patient to follow it is a loss to the patient as the treatment becomes sub-optimal. Not following the prescribed dose regimen can result in therapeutic failure or unwanted side effects.
1mg Technologies is India’s leading digital consumer healthcare platform with the aspiration of becoming a one-stop solution for all healthcare needs. As mentioned in Section 1, there are a number of reasons why patients exhibit non-adherence. We discuss below how 1mg is addressing those.
3.1 Non-adherence because of Forgetfulness
1mg app, [7], enables setting up reminders so that the patients get alerted by their smartphones about the time of their medicine. Given the ubiquitousness of smartphones and the penetration of mobile internet in both rural and urban India, this feature is expected to mitigate the problem of missing doses or not taking them at the designated time. Patients can record the dosing schedules for all the prescribed medicines and can help themselves be compliant to the prescribed dose regimen. Figure 4 gives a snapshot of this functionality.
3.2 Adverse Effects Driven Non-Adherence
Experiencing adverse events is another reason many patients stop following or deviate from the prescribed dose regimen. Experiencing adverse events can be harrowing for the patient and may spur the patient to lose faith in the entire treatment plan. Furthermore, in India, not only there is a shortfall of approximately 700,000 doctors but the distribution of doctors is highly uneven, e.g., the density of doctors in Chandigarh is ten times that in the worst state, Meghalaya, [8]. In such scenarios, in order to remain adherent or have their prescribed regimen modified, patients can use the 1 mg platform to get expert medical advice (e-consultations) from the comfort of their home. Enabling consultation by specialist doctors is expected to especially help patients residing in areas of poor doctor density to maintain adherence to their treatment regimens. Figure 5a shows a snapshot of this functionality.
Furthermore, the platform enables patients to order diagnostic tests from their smartphones which can help the doctor in understanding the current status of the disease and can provide swift advice to the patient, as shown in Figure 5b.
3.3 Non-Adherence because of Inadequate Access
Another reason behind non-adherence is the patient’s inability to visit a pharmacy. With 1mg, medicines can be ordered from the comfort of the patient’s home by uploading a valid prescription thus removing this as an obstacle in being compliant. A local pharmacy’s inventory can be unreliable — a patient may not get all the prescribed medicines which, in the case of 1mg, is alleviated by the presence of multiple vendors. Moreover, 1mg provides comprehensive information about cheaper substitutes of the drug (if available), thus potentially reducing the cost of the treatment plan significantly. Figure 6 gives a snapshot of this feature.
3.4 Non-Adherence because of complexity of the regimen
As mentioned in the beginning of this article, long and complex treatment plans are generally associated with higher rates of non-adherence. While the pharmaceutical industry strives to design simpler dose regimens (for example taking only 1 slightly large dose instead of 6 smaller doses), often patients have some unresolved questions and fears about the drugs that they are prescribed. 1mg provides comprehensive information about the all the drugs sold on the platform and allows patients to ask any specific questions that they might have to a qualified doctor and these interactions are visible to all visitors on the platform. Furthermore, aggregated responses from a medicine specific survey are presented at the product page which give information about the general characteristics of the drug, as reported by the actual users, as shown in Figure 7.
These initiatives create a virtual community for all users of that medicine and can reassure patients about their individual prescriptions, thus warding off any potential non-adherence.
4. CONCLUSION
Patient non-adherence is a ubiquitous problem having serious ramifications for the patients and the entire healthcare system. We discussed in this article that even a low degree of non-adherence can result in high impact on the efficacy and safety of the therapy. As the reasons for non-adherence are diverse in nature, a patient-centric approach and a one-stop solution which intertwines with patients’ lives is required. While a plethora of technology enabled solutions are available which solve specific reasons behind non-adherence in a piecemeal fashion, the services offered by 1mg app, [7], are expected to address the problem of non-adherence in a holistic way and should be a boon to the patient community and the healthcare system at large by being an integrated healthcare platform which enables patients to take complete charge of their healthcare requirements and maintain adherence through their smartphones.
For any questions or comments about this blog, please reach out to the author at: https://in.linkedin.com/in/kabir-soeny-phd-10ab2033
REFERENCES
[1] Jin, J., Sklar, G.E., Min, Sen Oh V., Chuen, Li S. (2008) Factors affecting therapeutic adherence: A review from the patient’s perspective. Ther Clin Risk Manag. 2008. doi:10.2147/tcrm.s1458
[2] Osterberg L, Blaschke T. Adherence to medication. The New England Journal of Medicine. 2005. doi: 10.1056/NEJMra050100
[3] Urquhart, J. (1996) Patient non-adherence with drug regimens: Measurement, clinical correlates, economic impact, European Heart Journal. doi: 10.1093/eurheartj/17.suppl_A.8
[4] Polonsky, William H, and Robert R Henry. Poor medication adherence in type 2 diabetes: recognizing the scope of the problem and its key contributors. Patient preference and adherence vol. 10 1299–307. 22 Jul. 2016, doi:10.2147/PPA.S106821
[5] Soeny, K., Bogacka, B. and Jones, B. (2019) Dose optimisation with simultaneous pharmacokinetic estimation in adaptive clinical trials, Statistical Methods in Medical Research. doi: 10.1177/0962280219852582.
[6] Soeny, K. (2017) Pharmacometrically Driven Optimisation of Dose Regimens in Clinical Trials, PhD thesis, Queen Mary University of London, London.
[7] 1mg Technologies Pvt. Ltd. (2020). 1mg [Mobile application software]. Retrieved from https://play.google.com/store/apps/details?id=com.aranoah.healthkart.plus&hl=en.
[8] Anand S, Fan V. (2016) The health workforce in India. Geneva: World Health Organization; Human Resources for Health Observer Series №16; https://www.who.int/hrh/resources/16058health_workforce_India.pdf
[9] 1mg Technologies Pvt Ltd. (2020). 1mg [Web URL]. www.1mg.com
