This summer, I worked for the company everyone hates. Doctors, nurses, patients, pharmacists- everyone hates dealing with us and with the paperwork we create. I worked for a pharmacy benefits management company.

Those of you not in the pharmacy world might ask: What is a pharmacy benefits management company? The standard answer, which I always found extremely circular and frustrating, is that we manage pharmacy benefits for various insurance companies. In practice, it means that we are contracted by an insurance company to handle the everyday affairs of prescription claims. The patient takes a prescription to the pharmacy, they electronically bill the insurance company, and the insurance company's computer software automatically (according to a standard protocol) accepts or rejects the claim. When the pharmacy gets a rejection, either they or the doctor's office can call us, the PBM, to fight it. When the doctor's office sends in a prior authorization, we get it and we either make the decision, call the doctor's office for more information, or call and tell them it's been rejected.

We didn't have the final say in most cases, but we knew the general rules. No Nexium unless you've tried (and failed) Prilosec OTC. No Allegra unless you've failed Claritin. The list goes on. And so the doctors hated us because we were denying their patients meds that they felt were superior, and because we made them go search the patient's chart to find out whether they'd ever been on Claritin. And the pharmacies hated us because they had to tell the patients that their drug wasn't covered.

But with healthcare costs (especially prescription drug costs) spiralling higher and higher, insurance premiums rising, and the possibility of universal healthcare that would cost taxpayers however-many trillion dollars, there is clearly a role in healthcare for cost containment.

Let's take proton pump inhibitors (PPIs). We got a lot of prior authorizations for PPIs. Most of the insurance companies we worked for would approve Prilosec (omeprazole) with no problems. If you wanted to use a different PPI, you would have to show that the patient had used Prilosec and still had symptoms. To me, this was a reasonable rule. The efficacy rate of omeprazole is very similar to that of the other PPIs (such as rabeprazole, lansoprazole, pantoprazole, or esomeprazole). In fact, esomeprazole is just an isomer of omeprazole. They are essentially the same drug, except that esomeprazole is brand-name only (Nexium) and costs probably ten times as much. Doctors know this, or they should if they routinely prescribe PPIs. Yet we still got tons of requests for Nexium, or Protonix, or Aciphex for brand-new patients. Maybe a drug rep had given some samples to a doctor. Maybe the patient wasn't satisfied with something that was "just" an OTC drug and wanted something "better." There could be any number of reasons. I would call the doctor's office and ask if the patient had ever tried Prilosec. The nurse would say no, and I would say okay, they have to try Prilosec before we'll pay for anything else. And they would say okay, and a few hours later I would be able to check and see that the patient had filled a prescription for Prilosec.

So in that case, it makes sense to have someone regulating the transaction whose primary interest is economic. The insurance company saves money (sometimes over $100/month), the patient pays a lower copay, and insurance premiums nationwide go down (or at least don't go up).

Another benefit to PBMs is in catching medication errors (which, again, saves money). The best example of this is in Advair inhalers. There are two types of Advair inhalers- a traditional aerosol inhaler and a diskus. Unfortunately, the directions are different. The inhaler is taken as two puffs twice daily, and the diskus is taken as one inhalation twice daily. So when we saw a prescription for Advair diskus, two puffs twice daily, we knew that some doctor had confused the directions. We would call to get it fixed because the insurance company wouldn't want to pay for two inhalers in one month- but in the process, we saved a patient from overdosing on salmeterol and getting tachycardia or even a heart attack.

Of course, this isn't all-inclusive and won't catch everything, but it's always good for the system to have more checks.

The major problem with an insurance company or a PBM is the inherent conflict of interest. It was in our best interests to choose the lower-priced Prilosec over the equally-effective and higher-priced Protonix...but it was also in our best interests to choose a less effective but lower-priced drug over one that was more effective but higher-priced. And if there was ever a judgement call on a case, it almost always resulted in us rejecting the claim. Transdermal patches, for example, are really expensive in general, and we were very reluctant to cover them if there was an oral version available. A request for transdermal patches might be rejected solely because the patient had other oral meds in their profile and clearly was physically able to take drugs orally. Even a liquid version of a drug for a child might be questionable if the child was older than 6 or 7. There were times when I dreaded calling the doctor's office back, because I disagreed with the judgment that had been made and I knew that the doctor's office would too.

How can we get rid of this conflict of interest while still being cost-conscious? There is, obviously, no easy answer. Perhaps one of the best systems is what's used in hospitals, where a P&T committee, made up of the doctors who will be prescribing these drugs to their patients, is able to create the formulary for the entire hospital. This isn't very practical in the real world, unless we want to create groups of doctors who would also have a financial interest in an insurance company. (Which, of course, would create other conflicts of interest.) So I really don't know what the answer is. Educating patients and doctors helps, of course, and so does choosing conscientious pharmacists to work at PBMs. But that won't solve the problem entirely.

It's just one of the problems that will hopefully be addressed when our country's healthcare system is overhauled, over the next few years.