Gasping for breath, Michael (name changed) pushed back in his chair. His legs felt like lead weights. He could not delay the rescue call any longer. The oxygen in his blood had dipped into the 80s.

An X-ray showed a shadow over the lower part of Michael’s right lung, and the hospital admitted him to treat a pneumonia. A smoker for decades, he was relieved a few days later to be discharged home. His cough had dried up. His legs still felt weak, but he had not been up walking like he would have normally. A little trouble swallowing he attributed to a diet of institution food. And if his eyelids drooped a little, there had not been a good night’s sleep since before he came in.

The prospective payment structure that pays a hospital the same rate for an admission whether it lasts two days or ten encourages facilities to push patients through each phase of care. This is called throughput, and from triage to emergency room to admission to discharge, every unit of a hospital is so focused on throughput that occasionally the whole purpose of the enterprise (to identify and treat illness) can be overshadowed.

Financial interests so often overtake patient outcomes in order of priority that we do not always recognize this as the pathologic behavior of a business-first, everything-else-second system. Health care institutions enjoy some incredible privileges, and that label of nonprofit did at one time convey something important about how a business was meant to function. But today, monetary incentives are just as often obstacles to care that patients deserve as the system itself is a facilitator of that care.

Throughput is nothing more than the available rationale for a profit-first approach. It sounds like it might be reasonable, and that is part of its danger. Naturally, there are exceptions: When a smaller hospital sends a patient to a larger one for a specialty procedure, it makes sense that the larger hospital would send the patient back afterwards, preserving space to provide the same service to another patient right away. But when a community’s only hospital functions more as a staging area guiding patients onto the next steps while finding something to treat (and, importantly, to bill for) without looking for the root cause, fundamental needs may go unmet.

After discharge, Michael lost some weight without trying. A dry cough nagged here and there. He found himself choosing softer foods. One evening Michael could not stand from his chair, and he knew where he would be headed. Another X-ray showed a consolidation in the same spot as before. A different scan found scattered nodules. Images of the rest of his body revealed lesions everywhere, and after a biopsy he was started on palliative chemotherapy, too far progressed now to cure.

What should have happened differently? Clinical people involved in Michael’s care were trained to connect these dots (not only to find and treat the pneumonia, but to contextualize the specific deficits not explained by that alone) during his first stay. But with pressure to get him out the door within a predetermined timeframe, the significance of this constellation of findings was not fully appreciated when that might have changed the trajectory.

By quality metrics the hospital did fine for Michael. This proves that our systems are built in a way which makes failure inevitable. But slow is smooth, and smooth is fast: No matter what we are doing, getting things right the first time is the best, fastest, and ultimately cheapest approach. Where we lose this is at the intersection of financial incentive and time horizon. Hospital payment encourages strategies which appear fast, those which may be great for quarterly metrics but devastating to long-term outcomes and long-term spending.

As long as superficial metrics are tied to reward, there will be managers eager to produce those metrics on paper, with little regard for, and perhaps limited understanding of, the patient outcomes set aside. As long as a hospital can be run as a profit generator first and care center second, there will be institutions which pursue those managers for leadership roles.

These are big problems, but they are begging to be solved. Perhaps the lever of change must be pulled from outside of the system: Varying rules between states provide the experimental conditions which prove financial pressures can stimulate hospital systems to make investments that produce great care outcomes (things like reasonable staffing and census ratios). The long-term savings of these measures far exceed their expense. In the long run, they are great business, too. Unfortunately, that only matters if the people occupying leadership roles are invested in the long run, and that is not what they are paid to care about today. Penalties for egregious care failures like this one might provide some motivation, but a counterbalance to non-clinical performance measures is sorely needed.

John Corsino is a physical therapist.