Administration

Nathan Kaufman: How silos undermine U.S. healthcare

Healthcare players of all stripes optimize the parts and fail the whole.

Published September 25, 2025 5:24 pm | Updated September 30, 2025 3:31 pm

In healthcare, the players are many: Physicians, hospitals, insurers, pharmaceutical companies and middlemen are all acting rationally in their interest or according to narrow incentives. But the results of focusing on optimizing these siloed components are skyrocketing costs, frequent shortages, inefficient healthcare systems and inconsistent patient outcomes.

The disconnect violates a fundamental principle of systems thinking, as articulated by operations research expert Russell Ackoff. To paraphrase, Ackoff asserts that a system is not the sum of its parts, but the product of their interactions, and that optimizing each part in isolation can perversely diminish the effectiveness of the whole.

The pharmaceutical supply chain offers a vivid example. This chain involves drug manufacturers, distributors, pharmacy benefit managers (PBMs), pharmacies, health plans and hospitals, each focused on their own efficiency and profit maximization. A manufacturer might reduce production to minimize costs, and a PBM might limit drugs in a formulary to maximize its rebates. Individually, these choices may seem sensible; collectively, they can translate into suboptimal outcomes. One part of the network’s drive to cut costs or boost margins often shifts burdens downstream, and patients ultimately bear the cost of higher prices and shortages.

A similar pattern emerges in healthcare delivery, where patients frequently navigate a fragmented maze of providers. In today’s fend-for-yourself, highly specialized system, a single patient — say, an elderly person with diabetes and heart disease — may see an endocrinologist, a cardiologist and a primary care physician, while also having to regularly visit labs and imaging centers, all in separate silos. Each provider strives to deliver high-quality care within their respective area of expertise. Still, without strong coordination, the patient can end up with duplicate tests, conflicting treatment plans or gaps in care, resulting in higher costs and suboptimal outcomes.

How hospitals participate

Even within single organizations, such as a health system, silos prevail. Individual departments — such as emergency departments, operating rooms (ORs), intensive care units and back-office units like IT, human resources (HR) and finance — often operate as semi-independent entities with their own siloed objectives and budgets. These internal silos, which focus narrowly on their assigned objectives, can result in suboptimal quality and increased costs.

For example, hospital administration might assign cost-reduction targets to each department. If the radiology and laboratory departments each cut unnecessary capacity to save money in their silo, the clinical floors might experience delays in getting imaging or lab results, slowing down patient care and prolonging hospital stays — ironically raising total costs.

Suppose the supply chain department is tasked with reducing implant costs. They accomplish their siloed task by limiting vendors for implants; the cost per implant may be reduced. All well and good, but unhappy surgeons may take their patients and associated revenue to another hospital, resulting in a negative overall financial impact from this siloed strategy.

Finally, within a health system, I have rarely seen a person who has been given the authority and accountability to manage the overall clinical and financial performance of a service line, such as spine care, which depends on multiple clinical and administrative silos, including OR, hospitalists, HR, finance and legal. As a result, critical decisions are slow and marginalized by the vertical objectives of all the silos involved, and the spine program is suboptimized.

Everyone is acting rationally in self-interest, but the outcome is costly and suboptimal. This fragmentation is exactly what Ackoff warned against — diminishing overall effectiveness by focusing on parts in isolation rather than the interactions that truly drive outcomes.

Solving the disconnect

If siloed thinking is the culprit, how do we fix it? For health systems, the solution starts with acknowledging that the problem is fixable, instead of yielding to the idea that the situation is intractable.

Hospital leaders must rethink and reallocate system-wide accountability and visibility. Health systems need to return to basics, focusing on optimizing quality, service and cost of the horizontal care delivery experience rather than achieving individual vertical, siloed objectives within the organization.

This requires flattening the organization and clarifying objectives, authority and accountability.

Appoint a directly responsible individual

Steve Jobs relied on a concept of assigning a directly responsible individual (DRI) to manage horizontally across silos, and that is a key solution for healthcare. At Apple, major tasks or outcomes had a single DRI who managed across silos, ensuring each performed for the overall benefit of the company. In health systems, this framework can clarify performance, accountability and authority across the myriad silos, such as for managing patient stays and optimizing the performance of key service lines. Time frames for implementation need to be dictated by market opportunities and needs, rather than by the IT or HR departments’ needs.

The DRI is not a specific job title; it clearly designates authority and responsibility for achieving a particular organizational goal. The DRI could be a regional executive, the local system leader or even a consultant. Support services such as HR, IT and legal would be told that because the DRI is held accountable for performance outcomes, their role is to support this performance. In organizations lacking clear accountability or authority, different silos often blame each other for the poor overall performance.

Current data is key

But assigning a DRI is only part of the solution. The other critical piece is giving those individuals real-time, cross-silo data using integrated dashboards with actionable benchmarks. By aggregating metrics from multiple systems into one shared view, dashboards provide instant transparency.

Real-time dashboards enable a health system to monitor performance and address issues promptly, rather than waiting 30-60 days for retrospective reports. For example, live tracking of bed availability is essential to ensure patients are discharged on time and beds are ready for the next patient. Executives should have high-level dashboards that alert them to overall performance in key areas, while the department leaders should have more detailed data.

Dashboards and DRIs work together; one provides visibility, while the other encourages accountability. The saying often attributed to Peter Drucker is, “If you can’t measure it, you can’t manage it.”

Meanwhile, a chief medical officer should have a dashboard that tracks the length of stay (LOS) in hours for each patient floor. The nurse in charge of each floor should have a dashboard monitoring the LOS for every patient on that floor, and with one click, be able to drill down to the medical record to determine why a patient is staying longer than expected. These dashboards identify unnecessary delays by floor and bed in real time, allowing the appropriate party to address and resolve discharge delays immediately.

Quality metrics and trends should also be tracked in real time to identify problems early and prevent a trend in patient harm. Delays and denials should also be monitored daily to quickly detect changes in payer behavior.

Bringing the pieces together

The combination of dashboards and DRIs can help shift healthcare from a system of isolated excellence to one of integrated performance.

We have little control over our nation’s dysfunctional healthcare system. However, what we do know is that physicians and health systems must make significant improvements in the cost and quality of care episodes.

The focus should be on optimizing the overall performance of the health system, which may mean allowing the suboptimization of some of its parts. But that is how real systems work. 

Advertisements

googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text1' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text2' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text3' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text4' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text5' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text6' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text7' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-leaderboard' ); } );

{{ loadingHeading }}

{{ loadingSubHeading }}

We’re having trouble logging you in.

For assistance, contact our Member Services Team.

Your session has expired.

Please reload the page and try again.