The pace of medical innovation outstrips our ability to evaluate those innovations against savings and/or quality goals. There is always a new technology or a new drug that promises to transform the management of a particular disease. Device and pharmaceutical companies will continue to push for payment of experimental therapies, especially when existing treatments do not work for specific populations or individuals. In addition to negotiating lower healthcare service delivery rates for its members, one might argue that health insurance companies exist to push back on unproven technologies or treatment plans when lower-cost alternatives exist.
The current approach to managing medical demand through prior authorization (First, provider and patient agree to move forward with a specific treatment. Then, the payer tells the provider and the patient that the service is not covered.) pits payers and providers against each other with the patient feeling they are not receiving the care they need. No randomized trial or meta-analysis will ever completely reassure a provider or a patient that a treatment will definitely not work for their specific situation. Most coverage determination guidelines are based on consensus statements from specialty societies. Even without the inclusion of leading scientists who may receive funding from pharmaceutical or technology companies, specialty societies usually prefer to be seen at the forefront of their fields, pushing for newer, if not necessarily better, treatments.
From a payer’s perspective, the indicators of a treatment’s effectiveness may be limited to changes in claims (e.g., medications, office visits, emergency room visits, hospitalizations). A more holistic measurement process would include patient-reported health outcomes (e.g., functional status, quality-of-life). Once a payer identifies a population who has received a novel intervention, the payer would then want to compare that population’s experience with the intervention against a population that did not receive the intervention. Ideally, the payer would also have access to those variables that might influence the outcome(s) of interest. In the absence of randomized, controlled data, this approach may be the fastest way to obtain effectiveness information about a novel intervention across a variety of patient subgroups.
Authorizing individual procedures, technologies or medications seems to be most cost-effective when targeted toward interventions that
- Incur enough costs to consider applying the overheard required to implement authorization,
- Are only beneficial to specific subgroups, and
- Are likely to be rejected more often than not (depending on the intervention’s cost, the review process may be more expensive than approving the intervention for all qualifying individuals)
There are other financial levers (e.g., deductibles, co-payments) that affect a payer’s decision to subject an intervention to the prior authorization process. Reversible interventions (taking an oral medication daily) may be viewed differently than irreversible procedures (obtaining a hip replacement). Among these criteria, only one is focused on the patient’s health. Obtaining more meaningful information about a treatment’s effectiveness in the wild could help payers make more appropriate coverage decisions or alter their stepped-therapy protocols to account for conditions that might simultaneously manage healthcare costs while providing needed new services to qualifying patients.
Most healthcare payers (the federal and state governments included) have not traditionally been in the business of tracking patient improvements to determine an intervention’s effectiveness. Collecting additional data from members is time-consuming and subject to gaps in reporting and data entry errors. But specialty societies and the research community have not been able to respond to market needs to measure the effectiveness of the ever-increasing number of interventions to address different health conditions. Diverting a relatively small amount of funds (e.g.,, reducing or waiving copays in exchange for members completing surveys) towards measuring effectiveness during an evaluation period may help tailor future prior authorization policies and reduce costs overall.
The process to measure patient-oriented outcomes would need to be broad enough to encompass a meaningful population, but discrete enough to provide concrete information to decision-makers. Some quality-of-life measures are rated on a 5- or 7-point scale, a scale that may be difficult to find meaningful improvements in health. Other measures may be too tedious to expect individuals to do repeatedly. In addition to patient-reported information like functional status and quality-of-life, payers and patients facing a decision to acquire a device or apply for a medication would probably want additional information like:
- Additional medications needed,
- Symptoms from the condition or other elements of the regimen,
- Likelihood of future complications, and
- Other services that might be required
Without any clear guidance from specialty society and patient advocacy groups, payers may need to start with some data request among members receiving a service and then iterate based on feedback from patients and others.
Applying measurement rigor to services that have traditionally required prior authorization might simultaneously improve the perception of payers who use the tool while providing information that might not be found elsewhere. Although this type of evidence would not be as robust as a randomized, controlled trial, the observational data would provide new information about patient groups excluded from traditional clinical trials and support longer treatment periods. Patients who provided information about their improvement (or lack thereof) would know that their responses would be used to tailor prior authorization efforts moving forward as well as helping inform other patients about the intervention’s effectiveness.