Cost-effective Alternative Drug Recommendation for Patients


According to an article by CBS News (2019), the prices of medicine are growing by a huge margin every year from 50% to 800%[1]. There are several drugs in the market with identical chemical composition and health benefits, but people are usually prescribed expensive ones that are more widely marketed and patented in the healthcare sector. Fortune magazine reported(2018) that Americans travel to Mexico to buy prescription drugs at a small fraction of the price charged in the US[2][3][4]. So, having a solution that offers accurate alternative drug recommendations will save everyone a huge amount of money. Here, we explore how Caboom can help to build a recommendation system that shows us the cheapest alternatives in just a few clicks.

The Problem 

A recent study published by the Health Affairs journal in 2019 shows that brand drugs increase their price by around 9% on average, which far exceeds the inflation rate. US News pointed out that there are 3 main reasons for drug prices to hike viz brand-medication are protected by the patent right, there is no way patients can understand the actual cost of medication and merge and acquisition of the pharmaceutical industry.

There are cases like, a report from a HuffPost in 2018 shows a drug named Daraprim which was initially priced for $13.50 per pill is hiked to $750 per pill [5]. Data from the Center for disease control and prevention indicates 45.8% of Americans are under at least one prescribed drug. These all pointed out the necessity of an intelligent system that could recommend the alternative cheap drug having the same effect on the branded ones.

What can be the solution?

Fortunately, there are often less expensive generic options available. And they can save you hundreds of dollars. We are talking about a cost-effective approach that encourages the use of more affordable and equally effective drugs. For instance, we can take a look at this example where we have a very essential drug called Coumadin. Coumadin is often prescribed for people with a history of heart attack or stroke. Coumadin is a brand name for the anticoagulant warfarin[6]. Warfarin is also sold as a generic drug, where it is priced 20 times lower than Coumadin on average**.

Not just the recommendation of similar compound drugs, we can leverage the caboom platform analysis to recommend drugs that are made up of different compounds but have the same effect. Most people notice but due to brand value, trending, etc of a certain drug company, they pay way more for the drug effect which they can get in the much cheaper drug.

So, we have two scenarios:

  1. Recommending alternative drugs of identical chemical composition.
  2. Recommending alternative drugs of non-identical composition for the same disease.

Let's take another example of hair loss, Minoxidil (C9H15N5O), cost around $50, and finasteride (C23H36N2O2), costs around $7, both help to reduce hair loss [7]. So there can also be chemically dissimilar drugs but helpful in the same problem.

How can AI enable recommending an alternative cheap drug?

Recommendations for alternative drugs are a tricky thing because the recommendation engine should have to make sure the essential ingredients that are present in branded drugs are present in cheaper alternatives. Moreover, the recommendation model should also be able to recommend the drug that has different ingredients but has the same effect. To learn ingredients, effects, side effects of medicines AI model is trained upon the medication data which then uses those data to learn and recommend alternatives. Because the effects of alternative cheap drugs are the same as the more branded and expensive drug[4]. One could save their money using cheaper alternatives. This work of comparing chemical and effect can now be automated with the help of Artificial intelligence technology.

Caboom: Solution to finding alternative drug

Caboom provides a constraint-based knowledge recommendation engine that helps in recommending alternative drugs when we feed it with the drug description data. Caboom is like a plug and play platform where you just need to upload the data rest of the process from quantifying the quality of your data, pre-processing your data, starting to build a machine learning model all are done in just a few clicks.

People keen to solve problems like these can leverage Caboom as a big enabler and without building any AI workforce and without investing extreme amounts of time and resources, they can just depend on Caboom to enable their AI-driven recommendation platforms. Caboom has abstracted each step while developing a recommendation system in an interactive interface which could demand quite a lot of resources otherwise. 

Traditionally we would need a full-fledged AI team to help discover and put up a recommendation system. However, using Caboom we can guide you to build your system using simple steps that deliver at a smaller fraction of the time and cost.

**Prices are taken from GoodRx, acessed on Sept-17, 2020


See instant AI recommendation results with caboom
Start with your data or a sample for instant results right away.
Request Access