Pharmacy automation is disruptive

Disruptive technology may replace low and medium-quality work (now done by technicians), but will eventually replace the most demanding ones (now done by pharmacists). [Source]

I came across an article outlining the various jobs that are likely to be taken over by machines.

Pharmacists are on the list.

If you think about it, a big chunk of pharmacists’ current job scope can be replaced with automation. With the exception of those requiring clinical judgement, the processes are repetitive and bound by a set of rigid rules and quality measures. Some solutions already exist in the market and we are just waiting for them to get mass adoption. For example, the basic workflow of processing a prescription:

  • Packing (taking the right drugs from the shelf, printing and pasting labels): automation systems to do this already exist. Some hospitals in Singapore are already using them in production.
  • Physical checking (quality control for packing): can be performed using image processing / RFID tagging technologies.
  • Clinical checking (check clinical appropriateness of the drug regimens): using advanced clinical decision support, systems will be able to sieve out the majority of “simple” low-risk prescriptions so human pharmacists do not need to check those.
  • Patient counselling (really talking to the patient): print customized PILs (Patient Information Leaflets) and read it out loud using text-to-speech technology with patient’s preferred language.

These applications do not need further ground-breaking research breakthrough in order to be implemented. The only things keeping these technologies from being adopted are perceived safety risk, slow business cycle of the industry, and bureaucracy.

Is it really going to happen?

There will be proponents of the “old way” citing the importance of human touch – and I agree that it is critical to a certain extent. But how many pharmacists have never ever forgotten advising a patient a relevant counselling point? How many pharmacists can always speak patient’s preferred language? How many pharmacists do not have any lapses due to stress or overwork? How many pharmacists can really digest 100% of patient’s medical history before dispensing the drugs? These are some of the things that machines would eventually perform better than humans do.

Banks have been replacing tellers with ATMs since forever and customers basically have no choice on the matter. (Some banks impose extra charges if you insist on talking to a human for transactions that can be done in ATMs.)

Some will say that the initial cost of these systems would be huge and thus only the biggest establishments would achieve the right economies of scale. But Thomas Watson, the founder of IBM, famously predicted 50 years ago that there would be a world market for maybe five computers. He predicted this because he thought computers would never be smaller than the size of a house and their price would never be within reach of mere mortals. So the person at the forefront of technology 50 years ago was proven to be very, very wrong; I’m sure that some of our prediction could be wrong too in 50 years time!

A few steps at a time

Adoption of automation is gradual. At some point of time, these technologies would mature to the point that we wouldn’t really need pharmacy technicians anymore. The profession would celebrate at this point as it is freed from menial tasks and is now able to focus on where it is perceived to value-add the most (i.e. clinical roles, the hardest to replace by automation).

Then gradually we wouldn’t need pharmacists anymore – at least not in the same capacity as what we have today. Pharmacists might be mainly doing “IT work”, configuring systems and only jump to the floor when system is faulty or in exceptional cases.

It is not a matter of if, but when. If patient is most well-served by machines, it is what is going to be.

3 responses

Leave a Reply