Can Hospital Reorganization Save Lives?

In the book The Innovator’s Prescription, Harvard’s Clayton Christensen and colleagues have described hospitals as “some of the most managerially intractable institutions in the annals of capitalism”. While there are many factors that make hospitals difficult to manage, they argue that an operational dilemma lies at the heart of this managerial complexity: The tension between standardization and compliance on the one hand and creative autonomous solution finding on the other. These two operational modes require very different management styles and mixing them in the same organizational unit leads to dissonance and confusion. It is no coincidence that R&D and manufacturing are clearly separated in organizations that do both. The trial and error mind-set of R&D is anathema to the Swiss clockwork of a reliable manufacturing operation. While it is rare to find these two operational modes in the same department, it is even rarer to find them intermingled in the daily working lives of individual workers. The exception from the rule are hospitals, where this mix is at the core of much of clinical work. Clinicians serve patients who will benefit from meticulous compliance to protocol and, in the room next door, others whose needs require a creative and iterative trial-and-error process of diagnosis and treatment.

Does it make sense to have these two types of process in the same unit? Or should we separate them out to make general hospitals manageable again? This is the question at the heart of the afore-mentioned book - and it is a big question. Hospitals across the world are organized along specialty boundaries. Historically, the dividing line has been between medicine and surgery but technological advances over the past decades have accelerated this supply-focused organizational structure along specialty boundaries. Replacing a century-old organizational paradigm with a new organizational principle, based on differences in operational requirements of services, will require good evidence before it will be adopted.

While Christensen and colleagues make convincing theoretical arguments for a separation, there is scant empirical evidence to support the proposal. In fact, a critical argument against the separation is the change in patient volume that such a re-organization will necessarily entail. High patient volume is associated with better clinical outcomes, in particular for surgical procedures such as joint replacements or heart bypass surgery.  Therefore, a reduction in volume in a disease segment following the proposed re-organization may erode any gains from an improved operational focus. The recent paper Separate and Concentrate: Accounting for Patient Complexity in General Hospitals by Ludwig Kuntz and colleagues provides evidence that this is not the case.  

The study uses patient-level data from over 250,000 adult patient episodes in 60 German hospitals across 39 disease segments. The outcome of interest was 7-day in-hospital mortality. The authors measured total annualized absolute and relative volumes (as a percentage of total hospital volume) in each disease segment and separated the hospitals into two groups - high-volume and low-volume hospitals - for each disease segment. In addition, patients within a disease segment were separated into three complexity categories: Low complexity patients had a planned hospital stay and no recorded co-morbidities (42% of the sample), while highly complex patients were emergency admissions with at least three recorded co-morbidities (13% of the sample). The remaining 45% of patient episodes in the sample were deemed to be of medium complexity.

The authors estimated the simultaneous effect of relative and absolute volume on 7-day in-hospital mortality, accounting for omitted variable bias through patient-level controls. In addition, the patient’s choice of hospital was modelled explicitly in a simultaneous selection equation, to ameliorate concerns that hospitals with different patient volumes attract patients with different mortality risk.

The study found that the effect of a hospital’s absolute and relative patient volume on 7-day mortality varies by patient complexity. The data provided no support that low complexity patients benefit from high absolute volume in their disease segment. These patients, however, benefit significantly from a high relative volume, suggesting that hospitals whose service portfolio emphasises a specific disease segment provide higher quality care for low complexity patients with the disease. The 7-day mortality decreased from an average of 2.1% in hospitals with a low relative volume to 1.4% in hospitals with a high relative volume of patients in a disease segment. For highly complex patients, the data suggests, perhaps surprisingly, a benefit from low absolute volume in the disease segment, with a significant drop in 7-day mortality from an average rate of 6.1% for hospitals with a high absolute volume to 4.6% for hospitals with a low absolute volume in a disease segment. In contrast to patients with low complexity, relative volume was not a significant predictor of mortality for complex patients. There were no significant absolute or relative volume effects for patients with medium complexity. The results were robust across a wide range of model specifications.

These findings support the proposal to separate the services for more routine patients (planned admission with a low level of clinical complexity) from services for emergency patients and complex planned admissions. They support the delivery of routine services in organizationally separate units that are focused on disease segments. When these routine patients are separated out, the absolute volume of the disease segment in the remaining acute hospital is reduced. The results of the study suggest that this reduction in volume has in fact a positive quality effect for highly complex emergency patients.

The paper also provides evidence that the remaining acute hospitals can further improve their service quality for complex patients by using a disease-based rather than medical specialty-based departmental routing strategy for newly arriving emergency patients. A counterfactual analysis suggests that a reorganization of the sample hospitals based on the two principles of (i) separating routine patients from complex patients and (ii) concentrating complex patients within a disease segment in the same hospital department might have reduced mortality in the sample by 13.4% (95% CI [6.9%; 19.0%]) for low complexity patients and by 11.7% (95% CI [6.1%; 16.9%]) for high complexity patients.

Read the full article at


Kuntz, L, Scholtes, S, and Sülz, S (2019). Separate and concentrate: Accounting for patient complexity in general hospitals. Management Science 65(6):2482-2501.