You may not have heard this term, but you have most likely experienced it every January. It makes you anxious to look at your inbox on the electronic health record or your fax machine because there are so many requests to change brands of certain medications. Please note that this is not the usual request to change to a generic, but a request to change between brands. Of course, it does not arrive with a note informing you of the formulary change, but you figure out pretty quickly the reason for all these identical requests.
Non-medical switching occurs when the formularies change medication in a class, such as basal insulin, and the pharmacy does not recognize it as equal, so the pharmacist sends you a request (demand?) for a new prescription so that the pharmacy can dispense what is now preferred on the insurance formulary.
The major frustration for our clinic is the sheer volume of these requests every January. We have tried to think of strategies to manage this sudden workload. If one agrees conceptually to the switch, then the clinic could set up a protocol whereby a nurse reviews each request and approves it. If one does not agree, each request must be reviewed. Either strategy still involves a lot of work for the clinic staff.
Concerns from the medical perspective
From the medical perspective, the important issue is that the insurance company is now deciding that certain medications are “equal,” regardless of the prescriber’s opinion. This means that if the insurance company decides there is no difference between two insulins, they simply refuse to pay for one. The options are limited: go along with the switch or, with a great deal of paperwork (the infamous “prior authorization” or “PA”) and possibly a phone call, you can demand the medication that you had originally chosen. Even if you do get it approved, the cost may be prohibitive since it is no longer “preferred,” thereby requiring a higher copay or no copay.
Although your choice may have been based on specific evaluation of the individual patient, or what is called “personalized medicine,” the insurance company has made a medical practice decision that the two medications are sufficiently similar, rendering your individual assessment irrelevant. In truth, if not for the pharmacy having a different National Drug Code for the newly preferred medication, we would not know about the switch until the next clinic visits, which could be three to 12 months later.
Additional issues with non-medical switching
This then raises the question of whether there is a problem with such switches. There are examples in the literature of alteration in clinical status and increased utilization of health-care resources after such switches. Examples include a 2018 study evaluating the impact of a formulary switch of an SGLT-2 and prior studies with SSRIs or an anti-tumor-necrosis-factor (anti-TNF) biological agent.
The argument in support of such switches contends that on a population level, a 10 percent to 15 percent difference probably would not have much impact on the cost for the insurance carrier. One approach is to argue that if a study comparing the two agents (i.e., insulins) showed non-inferiority, the two could be interchangeable. However, it is rare to have head-to-head studies of medications in the same class. Therefore, the insurance company’s formulary would have had to compare reportedly “similar” studies to conclude that the agents are interchangeable. While this is actually something we do in practice when we review the medical literature to decide the benefits and risks of medications, we do not read the literature to look for interchangeability but for differences that would suggest which patient characteristics identify those more likely to respond to one medication over the other.
Despite our opinions of the literature, the insurance formularies make a decision that exerts a wide impact on patients’ medical care. The result of the switch could include a worsening of his or her medical condition and increased utilization of health resources, as shown in the studies cited. Additionally, nonmedical factors need to be considered, such as patients’ satisfaction with their care, the burden to the clinic and possible abandonment of therapy by the patient. Furthermore, this is one more step in removing the physician’s autonomy. At the most extreme, one could perceive it as an indication that insurance formulary believes they can make medical decisions regarding therapy, so physician input may no longer be needed for implementing the medical care plan.
Awareness is key
While there is no obvious solution to this problem, awareness of it may allow you to plan your clinic time in January a little differently and perhaps create some handouts to explain to patients how to transition between the insulins, what to expect in transition and when to call in (rather than going to the emergency department) so that we can help with the dose adjustments.
Access additional resources and practical information to enhance the care and treatment of your diabetes patients.