There’s a nice class of combinatorial problems that nurses solve every day, to do with the order in which drugs on the pole are fed in to me. As an example, my antifungal medication is in a D5W (dextrose) solution and has to be by itself; the others, in saline, can be in different orders.

My portacath has two lines that go directly into a reservoir that was inserted in my chest, located on the right side above the nipple and level with the armpit. The nurse has discretion, within the rules of drug interaction and delivering only those medications prescribed for me, for when to feed the medication and what other drugs to feed it with. Other constraints include the overall volume of liquid pumped into me, and, in the case of blood, my temperature – blood cannot be fed in you are running a fever since it independently causes your body temperature to rise a few degrees. There is a potential source of error here because of the wide range of solutions to the problem of how to administer the medication. It’s also potentially a time waster since the nurse may have to solve the problem many times over during the shift. In addition, it’s subject to interruption as the doctor may add, modify, or delete a medication order.

Another method of delivering medication is oral. For example, flagyl can be oral or IV. It is preferably given oral to reduce the space required on the pole. However, flagyl tastes extremely nasty and the first time I took it I threw up. For a few times after that I took it in gelcaps, which is where the nurse cuts up a pill and puts it in a gelatin capsule. The merest grain of flagyl on the outside of the capsule, left there by accident as a result of cutting up the pill in house, can still cause me to throw up. And this is what I did. Now I take flagyl by IV, which further reduces the amount of space on the pole.

Yet another method is injection. I get my G-CSF this way via a shot in the belly.

I have not counted the number of different medications and additives (potassium, magnesium, etc.) that I’m taking, but it’s a lot. For the chemo drugs alone, I’m taking 13 different drugs, which in some cases must be given both IV (intravenous) and IT (intrathecal, or direct into the spinal column). So, there are many ways to screw up, as well as very severe consequences in some cases if you overdose or omit the dose. [See James Reason, Beyond the organizational accident: the need for “error wisdom” on the frontline. Quality Safety Health Care 2004:13:28-33]

There is a reverse danger. If all decisions are automated, nurses may perceive that someone is trying to turn them into automatons. While I think there are, and will continue to be better and more sophisticated automated solutions for heath care delivery, we need to make sure that nurses don’t lose opportunities for patient interaction when they can check other less measurable aspects, such as the patient’s mood. [It is commonly held that 80% of all human communication is non-verbal]. Also, nurses take justifiable pride in their ability to solve problems of drug interaction and delivery, and pride in one’s work is a key attribute of a well functioning team