A few months ago, I heard a presentation on organizational structure that seemed to hold a valuable insight for physician organizations — and for health care organizations of all sizes. The presenter, André L. Delbecq, DBA, is the J. Thomas and Kathleen L. McCarthy Professor of Management in the Leavey School of Business and Administration at Santa Clara University, Santa Clara, Calif. In his presentation, Delbecq contrasted two organizational structures and their underlying philosophies. You've probably run into the same contrast in other venues.

The petrified forest

He started with the older, highly centralized, bureaucratic model that shaped the development of large industrial organizations in the early 20th century. In that model, power is concentrated in the top leadership, and that leadership governs the whole enterprise by command. Relationships are highly formalized and hierarchical. In some sense, customers are at the very bottom of the hierarchy: They're told what they want to buy and — in the old paradigm — they're happy with it.

Pause a minute and ask yourself whether this description reminds you of any health care organization you know, or contract with, or work for.

Delbecq argued that the centralized, hierarchical model is obsolete. It grew from management theory of the early 20th century and had a place in that era, but it became less and less appropriate to our society. With the GI Bill sending veterans to college, the educational level of the work force outside the hierarchy grew by leaps and bounds. The civil rights movement and women's rights movement challenged the corporate bastions of white male power. Foreign competition and competition from domestic upstarts began to eat into the foundations of the corporate monoliths because the competitors were both faster and more innovative.

New growth

The power of innovation was an important theme in Delbecq's presentation. Delbecq argues that, for an organization to be innovative, it must be small and large at the same time: small, because it takes a small, flexible organization to remain action oriented and empowered; large, because innovation requires a broad base of resources.

The “large but small” paradox is resolved in the organizational model that Delbecq held up next to the traditional hierarchy: a “federalist” structure. An organization built on this model is not run from the top down. Each business unit is allowed to operate fairly autonomously and to maintain its own style.

The top leadership focuses on industry-level strategies; it provides a vision for the organization as a whole and ways of measuring progress toward that vision. It advises and supports the business units, but it does not manage them. The individual strategic business units are the most important nodes in the network; the rest is infrastructure. Each business unit does what it does best and works to please its customers, not the board of directors. Each business unit functions as a small action-oriented enterprise while benefiting from the resources of a much larger organization.

Now ask yourself whether this description reminds you of any health care organization you know, or contract with, or work for.

Health care

Certainly the conventional, hierarchical structure has been tried again and again in health care, far beyond the point where it was declared obsolete in the rest of the economy. The article by Rod Aymond and Theodore Hariton, MD, in this issue chronicles some of the results. But the “federalist” structure is rare, at least, if not unheard of. Remember, we're not talking here just of the difference between a staff-model organization and a physician network. Even if the former is hierarchical, the latter is not likely to be the network of “energy centers” that Delbecq spoke of. In such a network, to paraphrase Delbecq, each practice would have to be highly responsive to patients' needs; accountable for its own costs, quality and revenue; innovative and strategically minded. More, it would have to resonate to the vision of the larger organization and meet its high standards of performance.

As we've seen again and again, a contract does not make a network. In fact, I suspect that the kind of federation of business units that Delbecq spoke of probably requires incorporation of high-performing, like-minded groups into one entity. Can it be done in health care? I think it is being done here and there in a faltering, hesitant way. It can't be easy or fast, since it involves finding or growing those high-performing, like-minded groups. The thought that it might be happening at all is comforting, though, since it suggests that there might, after all, be a future for health care.