How can you know your child's in a good PICU? — assessing quality of care

There are about 400 PICUs in the United States. It is hard to know the exact number because, although several organizations have proposed standards, these facilities are self-identified. So a hospital is free to designate a place as a PICU and paint the letters on the door to it. How can you tell if your child is in a good one? Of course you can use the general reputation of a hospital to decide — a recognized children’s hospital will most likely have a good one — but are there any reliable measures of quality?

The question of quality of care (and the related one of getting good value for our healthcare dollars) should revolve around outcomes — how did the patients fare? After all, that’s the bottom line. However, that question can be difficult to answer, and the reason for that difficulty relates to our essential humanity — we are all unique. Thus the same treatment applied to different persons, even if they have the same disease, can yield different results, because no two individuals truly have the “same” disease — it expresses itself differently in each of us. Still, we can’t use this uniqueness to dodge the issue; clearly, some things work and some things don’t, and some facilities do them better than others. How can we sort them out?

A huge issue for outcomes research is case-mix. That is, the only way to compare how one PICU does with another is if the patient populations are nearly exactly the same. If they aren’t, it’s not a fair comparison. We generally can compare very large PICUs with each other because they care for so many children that small case-mix differences wash out. But that may not be not true for smaller ones, and even among the large ones there are differences in the sort of children they serve.

Since outcome research is so hard, what is often done is to use a proxy for it, something called a process marker. The notion is that one can monitor how a hospital is following a recommended process, such as standard procedures for immunizations. If the hospital is doing well at that, by assumption it probably is doing well at other things. But that is an assumption, although there is research to support it. You can find examples of this here and here, at the National Quality Measures Clearinghouse.

One of the most rigorous outcome measurement programs is in the area of organ transplantation. This is because the organization that regulates how precious organs are used demands that all hospitals submit reports of how well their patients are doing. You can even read and compare these for yourself here. There are still problems of case-mix, since hospitals vary in choosing precisely what patients they will transplant, but overall the system gives great transparency to what is happening to patients.

So back to the original question: how can you know a PICU is a good one? If you dig (and have the time), you can find out how well a hospital does with complicated, planned procedures like organ transplants or heart surgery. But most children land in a PICU from some acute, unplanned condition. In that case all a parent can realistically do is check the credentials of the facility (such as here) and of the doctors (such as here and here). One day we will have more, since both the public and those who pay the medical bills — insurance companies and governmental agencies — are rightly demanding it.