Earlier this month, CVS announced plans to buy Aetna— one of the nation’s largest health insurers—in a $69 billion deal. Aetna and CVS pitched the deal to the public largely on the promise of controlling costs and improving efficiency in their operations, which they say will inhere to the benefit of consumers. The media coverage since the announcement has largely focused on these claims, and in particular, on the question of whether this vertical integration will ultimately lower health care costs for consumers—or increase them. There are both skeptics and optimists. A lot will turn on the effects of integrating Aetna’s insurance with CVS’s pharmacy benefit manager services.
But CVS and Aetna also flag another potential benefit that has garnered less media attention—the promise in combining their data. CVS CEO Larry Merlo says that “[b]y integrating data across [their] enterprise assets and through the use of predictive analytics,” consumers (and patients) will be better off. This claim merits more attention. There are three key ways that Merlo might be right.
First, lack of price transparency is a huge problem under the current health care model. Patients are supposed to be price sensitive and turn down low-value and unnecessary care, but most of the time they don’t know what care will cost at decision-time. Part of the problem is that providers don’t have real-time access to insurer information. Has this patient met his or her deductible? What is the patient’s coinsurance obligation? And so forth. Merging the provider’s (CVS’s) systems with the insurer’s (Aetna’s) would solve this problem and should, at least in theory, make it easy for providers to quote patients real-time, accurate cost information before patients consent to pay for care.
Price transparency legislation has proliferated in recent years, but it usually just requires disclosure of chargemaster pricing (basically the “rack rates” of health care). While this information might be better than nothing, it doesn’t reflect the prices that insurers negotiate with providers, and it doesn’t provide particularly useful information to individual patients who want to know what they specifically will be asked to pay for a procedure. CVS should now be able to do better. This should be an important part of its business plan as it commits to turning its retail clinics into “community-based sites of care” that give patients alternatives to expensive and often inconvenient hospital-based care.
Second, unnecessary care is a major unsolved industry problem. According to the Institute of Medicine, it is the predominant factor in U.S. health care expenses, accounting for about a third of healthcare spending in the US ($750 billion out of $2.6 trillion). Data and incentives are currently impediments to addressing the problem. It is hard to determine what care is unnecessary and in what circumstances. And then it’s hard to give providers incentives to counsel against unnecessary treatments and procedures, particularly when they are paid for delivering more care in a fee-for-service reimbursement model. Vertical integration, here between a payor and a provider, might prove very useful. It may give the payor the incentive to mine its treasure trove of data to better identify sources of unnecessary care. And it might incentivize providers to heed the results.
Third and relatedly, CVS itself has noted the potential for mining and providing better information to patients to help them make decisions and manage their care. Making health care decisions is hard. Patients are variably medically literate and even those who understand the jargon are prone to biases in decision-making. With access to claims data, CVS should be able to provide more personalized information to patients about the likely outcomes of various treatment choices.
None of these reasons for optimism are a slam dunk. There will be tremendous cost involved in integration and while the promise of using analytics in positive ways is certainly worth exploring, it is not without hurdles, including concerns about patient privacy and confidentiality. But in a world where consumer costs have dominated the discussion, it is at least worth thinking more about the impact that the integration of payor and provider data might have.
