I'm pretty sure Bette Davis wasn't talking about health care reform when she warned us in "All About Eve" that it's going to be a bumpy night. But ever since the introduction of the Affordable Care Act 18 months ago or so, this country's been jounced and jostled so ferociously night and day from so many different factions it's a wonder we're not all limping around with black eyes and our arms in slings.

As I mentioned in one of my blogs for H&HN Daily, I don't agree with the blanket criticism of reform. We've known for a long time that fee-for-service is a perverse way to pay for health care. We know that we've got to rein in costs so we don't bankrupt the entire nation. I haven't heard anybody gripe about the emphasis on prevention and wellness; on the need to make care more patient-centered; or on the benefits to quality, safety and, yes, costs, if we could better integrate providers and build a solid continuum of care.

But in the fervor to improve the system, there are bound to be miscalculations, and when what seems like a grand idea turns out to be laden with unexpected, unwelcome consequences, we've got to have the flexibility to rejigger the details.

Take readmissions, for example. A September TrendWatch from the American Hospital Association notes that policymakers are proposing incentives to reduce hospital readmissions by publicly posting data on readmission rates and lowering payments to hospitals with high rates.

The thing is, not all readmissions are the same. There are actually four distinct types. Two types are readmissions planned prior to discharge either as part of a patient's continuing course of care, such as a series of chemotherapy treatments, or to treat an unrelated matter, such as removal of a lung tumor discovered when the patient was hospitalized for a heart attack. Of the two unplanned types of readmissions, one is for those unrelated to the initial hospitalization, such as an injury from a car accident following a stay for surgery. The second type of unplanned readmission is for those related to the initial hospitalization — for instance, to treat a surgical-site infection or adverse reaction to a medication. It's only that last type that hospitals can reasonably be expected to try to limit.

Many readmissions also may be out of hospitals' control because of such factors as patients' chronic conditions, their failure to follow care instructions and their socioeconomic situations. And there is surprising evidence of an inverse relationship between mortality and readmissions that suggests low readmission rates may not always be desirable.

Payment penalties "could exacerbate inequities," the report warns, and "misaligned policies could direct hospitals to reduce readmissions that are appropriate for safe patient care and [that] may actually save lives."

I've been impressed with the willingness of policymakers to respond to concerns from the field about certain aspects of reform. Let's hope they smooth out the bumps when it comes to readmissions.