In this segment we will discuss vertical integration.
Vertical integration is the merging together of
two entities that are at different stages of production.
For example, a merger of a manufacturer and a retailer.
Vertical integration in healthcare typically happens along the care continuum.
That is hospitals integrating with primary care practices,
or with home health agencies.
The former is called backward integration,
as primary care refer patients to hospitals.
The latter is called forward integration,
as hospitals refer patients to home health agencies.
There are many reasons for vertical integration
including synergies in production distribution and marketing,
diversification, securing business, ensuring quality, and so on.
Economic theories for the most part view vertical integration favorably.
The main reason is that transactions along a vertical chain raises the price to
consumers as each entity along this vertical chain seeks to generate a surplus.
In a way, what integration does is cut out the middleman.
Therefore, vertical integration allows firms to better allocate their resources,
offer competitive prices, and increase their sales.
In some cases, vertical integration is deemed anti-competitive especially
if a firm is interested in capturing resources to exclude competitors.
The theoretical literature in economics views
vertical integration as the grouping of assets.
Asset one and asset two are two productive assets owned by separate entities.
Vertical integration places both assets under a single entity.
Unlike economics, the theoretical literature in
management strategy views vertical integration as the grouping of tasks.
Task one, two, and three are part of
a production chain where each task is performed by a separate entity.
Vertical integration places all three tasks under a single entity.
Since assets perform tasks and tasks require assets,
the two streams of literature do not intersect.
Each focuses on a different aspect of vertical integration.
In this example, task one and two are performed by
asset one while task three is performed by asset two.
So, assets and tasks are two sides of the same coin.
However, when we consider the interdependence between assets and
task we need to take into account that some tasks can be performed by multiple assets.
Assets that may or may not be owned by different entities.
In a 2013 paper my colleagues and I called this type of a task a general task.
Task two in our example is a general task,
while task one is dedicated to asset one and task three is dedicated to asset two.
If the two assets are held by separate entities
a conflict is likely to arise as to who gets to perform tasks two.
Now, let's put all these in a more familiar context
where asset one is a hospital and asset two is a home health agency.
Task one represent a surgical intervention which is a hospital dedicated task.
On the other extreme, task three represents help with
daily activities such as bathing which is a home health dedicated task.
Task two, the general task represents
recovery and monitoring which can be performed by hospitals,
as well as, home health agencies.
The question is how much of task two should be performed by the hospital,
and how much by the home health agency?
Consider a patient who arrived at the hospital in bad health
but over the length of her hospitalization her condition improved.
After four days the hospital was ready to discharge the patient.
Recall that under perspective payment systems
the hospital would like to discharge the patient as soon as
possible as it receives a fixed payment per hospitalization regardless of it's length.
On the other hand, the home health agency
agreed to accept the patient after six days in the hospital,
allowing the patient two additional days of recovery time.
Recall that home health agencies are also paid perspectively,
and therefore would like to receive patients who are
relatively healthy and require low intensity of care.
For example, patients who would require fewer visits.
This tension between hospitals and agencies can be resolved by vertical integration.
This is exactly what our study found.
We analyzed nearly 400,000 hospital discharges to home health and found that
vertical integration allows hospitals
to discharge patients to home health agencies earlier,
and in poor health,
and at the same time increase
post hospitalization service intensity without affecting health outcomes.
To conclude our discussion,
let me compare the case of home health to the one of hospice.
We discussed hospice in the previous segment.
Let's go back to our representation of the tension between hospitals and
home health agencies and contrast it with
the relationship between hospitals and hospice agencies.
A patient arrives at the hospital in bad health,
but instead of improving his condition deteriorates.
This line marks a level of health below which the hospital deems the patient incurable.
When the patient reaches this point
the hospital would like to discharge the patient to hospice.
As we discussed in the previous segment,
under a per diem payment schedule and u-shaped cost,
the hospice agency would like to receive the patient as early as possible.
In other words, there is no conflict between
the hospital and the hospice agency over the timing of discharge.
Therefore, unlike the case of home health vertical integration is not needed.
Not surprisingly, while more than a third of
hospitals in the US own a home health agency,
only a handful is integrated into hospice,
put differently under regulated prices,
task alignment is an impetus for vertical integration.
In the next couple of segments,
we will shift our attention to
government regulation and the evidence for it's impact on healthcare markets.
Ezekiel J. Emanuel, MD, PhDDiane v.S. Levy and Robert M. Levy University Professor
Guy David, PhDGilbert and Shelley Harrison Associate Professor of Health Care Management
