Duke Research Advantage

By Kevin Schulman, M.D.

Duke Medicine has just announced a $700 million expansion of its physical plant to add 847,000 square feet, including adding 16 operating rooms and renovating 160 beds (that’s over $800 per square foot if you’re doing the math).

This will add a new cancer hospital facility within the Duke campus, and expand the number of intensive care rooms at Duke Hospital. This is a significant investment in infrastructure for the health system, but how will this investment impact the care delivered to patients by Duke Medicine?

Duke Medicine is a fee-for-service hospital system and physician network. This is one of the oldest forms of a provider organizational structure in existence, and pre-dates the development of health insurance in the US. This model is managed to allow patients access to high-end specialty and hospital services. Within this model, hospitalization is not considered a failure, but rather a core service offering. Competition between Duke and other providers is on the basis of technology and specialty service offerings. This type of competition can lead to improvements in the performance of specialist physicians, but can also lead to a phenomena described as a “medical arms race” where hospitals “must” acquire certain technologies to remain attractive in these specific markets.

Hospitals like to point out that their high-tech facilities and services are better than those of competitors (see Robinson and Luft JAMA 1987). This type of competition has been criticized as driving to higher investment in technology, overcapacity, and higher health care costs. There’s some local evidence of this arms race concept with UNC’s investment in a similar cancer facility only nine miles from the Duke campus.

In preparing for the Global Executive MBA residency in London , I reviewed the U.K. National Health Service web site. It is a big contrast to the recent Duke announcement; you will not find any information on ICUs or high-end services, instead you will find a plethora of information on primary care services and access to care.

Why the discrepancy in focus between these two “health systems?” The UK system is an integrated delivery system where people use their primary care physicians to guide them through the system. These physicians are the first point of contact with the system, and primary care services are available in person or electronically from the NHS. In the US, Kaiser Permanente has a similar focus. These organizational designs developed after the advent of health insurance, and combine both insurance and delivery system into a single organizational structure. These systems are focused on managing patient care to reduce the utilization of high-end services. At some level, hospitalization is considered a failure of the system, except when hospital care is the only appropriate/efficient site for the service to be delivered. Further, “competition” is based on access to care when patients need it through primary care networks, and on cost – not on the quality or availability of specialty care networks.

This type of competition may not provide tremendous incentives to improve specialty care service performance, and while this theoretical concern has not been seen empirically, there have been high profile lawsuits about denial of care and inappropriate referrals to low volume providers (see Miller and Luft 2002).

There are clearly at least two approaches to the development of a health system. In many emerging markets, the Duke model is being implemented (especially in for-profit settings). However, as insurance markets develop, it will be interesting to see if we have more interest in an NHS/Kaiser-type system as a result.