The ultimate goal of any managed care model (e.g., HMO, PPO, POS, PCMH, ACO, etc.) is to improve the health of the population, enhance the experience and outcomes of the patient, and reduce per capita costs of care. All of these models have at least three things in common:

  1. Members are attributed to a provider.
  2. The provider is accountable for care coordination.
  3. Providers are incentivized through reimbursement or penalties based on their ability to achieve lower costs and improve outcomes/experience.

Undoubtedly, one of the most critical aspects of population health management for a provider is knowing which patients are attributed to you. 

How Attribution Impacts Providers

If I’m a provider in an HMO plan, this is fairly straightforward given that individuals enrolled under these policies are required to select a PCP, or “gatekeeper”. In these models, all services must be rendered or requested by the assigned PCP. As such they tend to have better outcomes in controlling costs, primarily because one person is responsible for the care of that member.

Challenges with Attribution Models

If a provider is in a more flexible model like PPOs, PCMH, or ACOs, it’s not so straightforward; actually, it makes doing the job much more difficult. Why? In these types of models, the member still gets attributed to a PCP, however, its driven based on historical claims data; usually determined based on the plurality of services (i.e., who saw the patient the most). In addition, the PCP does not perform the “gatekeeper” role in these types of policies. In these cases, the member is free to seek services from any other provider in the network without a referral. 

In these more flexible models (e.g., PCMH, PPO, ACO,etc.), attribution is typically determined by the payer on a month to month basis. Providers are informed when a payer submits monthly membership or roster files indicating which patients have been assigned. 

This is problematic for providers for multiple reasons. 

  1. Not all payers determine attribution alike.
  2. Panel churn, where patients are attributed one month and not the next.
  3. Patients are attributed based on urgent care visits.
  4. Patients often see multiple physicians overlapping, especially those that have chronic illnesses, e.g., COPD or CHF.

Careful Considerations

There is no silver bullet attribution method that applies to all contracts. Results under the various methods can lead to a wide range of consequences. Therefore, as a provider of healthcare, you should consider this when signing contracts and request detailed specification on how patients will be attributed to you and your organization under their policy. We highlight some key considerations, but this only scratches the surface.

  1. Is the attribution model patient based or episode-based?
  2. Is it a single attribution or multiple attribution model?
  3. Is it prospective or retrospective?
  4. Is it majority or plurality based?
  5. What is the duration of the measurement period?
  6. How do they handle members with no claims?
  7. Do they attribute at the member or subscriber level?