Hospitals are very focused on avoiding harming patients lately. They have been moving in that direction for a long time, but with health care reform legislation, payments are on the line, which makes something that was a very good idea into an imperative.

In the year 2000, the Institute of Medicine, a non-profit organization that monitors various aspects of medical care, reported that 44,000-98,000 people died each year due to medical errors. This began a nationwide focus on patient safety that has had some, but not enough, impact on outcomes. Hospitals already do not get paid for care of a patient who gets a blood stream infection from their central venous catheter or a urinary tract infection from their bladder catheter, so they have to eat the costs associated with these things. When a hospital is paid a lump sum for a diagnosis (say a patient is admitted with appendicitis) and the patient gets some complication that makes their care longer or more expensive, the amount of money the hospital makes on the whole episode is less. But at some point in the not too distant future all payment will be based on good outcomes and having some event in the hospital that makes things more complicated (and the patient sicker) will hurt the hospital almost as much as it does the patient.

We call the bad things that happen to patients “adverse events” and we try to eliminate all “preventable adverse events.”

Some of these adverse events are really obviously our fault, and others are so preventable that we consider not preventing them to be unconscionable. Our fault would be doing the wrong procedure or the right procedure to the wrong patient or body part, leaving a sponge in a patient’s wound, causing infection of a procedural site by not using sterile technique, giving the wrong medication or the wrong dose, or a medication to which a patient has an allergy. MRI machines have powerful magnets and occasionally make metal objects brought into the room into deadly or injurious projectiles. We have foul evil bacteria in hospitals and if we don’t wash our hands between patients we will transmit bugs such as methicillin resistant staph (MRSA) and Clostridium Difficile from patient to patient. We know that patients who are bedridden or have had orthopedic procedures get blood clots in their legs that can go to their lungs and kill them, so we give them medications that prevent clotting and sometimes contraptions that massage the blood in their legs. We know that patients on ventilators with tubes in their tracheas will develop pneumonia if kept lying flat, so we elevate the heads of their beds. We know that delirious and elderly folks who are weak are liable to fall and break bones so we watch them very carefully. We know that fragile skin on the bottom can break down and cause pressure ulcers if we don’t turn a bedridden patient regularly.

Hospitals are carefully monitored and soundly disciplined if they have too many of these bad things happen, so we really do pay good, and progressively better attention to this sort of thing.

What we don’t necessarily recognize is the huge burden of adverse events that happen in hospitals just because patients are in hospitals, despite or because of the fact that they are being treated by our best and brightest physicians with our best evidence based medicine and fancy technology.

Patients are usually admitted to the hospital because they have something wrong enough that they can’t safely stay home. Sometimes they are admitted because we aren’t sure whether this is true, but want to be on the safe side. When we make the decision to hospitalize a patient, we take on a huge responsibility and expose the patient to very significant risks.

We almost always put an IV in the patient. This is a small sterile tube that goes into a vein and is held in place by something sticky. We then hook the IV up to some sort of fluid with a pump which goes “beep beep beep beep…” when the little tube gets kinked or displaced. We sometimes give the patient various medications through the IV, maybe diuretics to take off some fluids, sedatives to calm them down, antibiotics to kill real or imagined infections, solutions of various salts to increase the blood volume, drugs for nausea, pain, high blood pressure…The beeping wakes them up, but the sedatives make them sleep. They become sleep deprived. The pain medications make them goofy and constipated. The fluids discombobulate their own electrolyte levels or overload them causing swelling and oxygen deficiency. The diuretics, if we went in that direction, cause kidney injury, which is strongly associated with in hospital death. They are not fed because we do tests that require that they not eat, so if they are diabetic their blood sugars drop, and then go too high when they finally get a giant tray of food which is much different than what they eat at home.

Much of what we do to patients is based in our culture of infinite health care resources. We don’t necessarily even need the IV, but put it in anyway, just in case. There is a perverse incentive to do this, since a patient on IV medications of certain types is felt by payers such as medicare to need hospitalization, and one without an IV is not. We are paid for a higher level of care if a patient is getting opiate pain medications by the IV route.We don’t do these things just to make more money, but we are also not immune to perverse incentives. We sometimes do tests without thinking whether they are necessary. We try to avoid fluid overload or dehydration but we don’t necessarily watch people as closely as we should.

Being in a hospital is dangerous. It is also sometimes necessary, and sometimes more dangerous to not be in a hospital. Still. The science of patient safety could link itself more effectively to cost effective care. I would bet that there is actually not one patient admitted to the hospital who does not have a health care associated complication, if we keep in mind that things as seemingly trivial to providers as damage to veins from IVs and blood draws and financial ruin related to hospital costs are truly significant to the patients in our care. We need to be attentive to the fact that every little thing we do, from ordering a medication to ordering a test, carries with it a significant risk, and notice that some portion of our patients’ medical problems stem directly from our best intentions.

Janice Boughton is a physician who blogs at Why is American health care so expensive?