Counting down the days and the remaining chemo infusions until I’m all done. I met with the physiotherapist today and he checked out my neuropathy and strength and balance. He showed my some exercises that I can do to help me get back to normal, or at least accommodate my condition. My balance is a little bit off but it’s good enough that I won’t need a cane. I must admit that the last thing I want to do at the moment is exercise – napping is more where I’m at – but I suppose I must discipline myself.

Coincidentally, the room that I’m in is exactly the same room that I started out in, ward 5B, room 35. It’s very small, but this does give the advantage that I can reach the bathroom while I’m still attached to the IV pole. The ward is in the shape of a three-quarter circle; consequently all the rooms are wedge shaped, since they are on the circumference. The nurse’s station occupies the center of the circle; the next ring is the corridor, often littered with extra chairs and equipment, with terminals and storage closets on the edges; the outer ring has the patient rooms. There are 10 patient rooms in the ward, of which about half are double. For most of my visits here I have had a private room.

I have been reading more of “The Patient from Hell.” The author, Stephen Schneider, is a climatologist working in the field of global warming. He makes distinction between two types of knowledge that I think is important to grasp if we are to understand system safety. The one type of knowledge is the knowledge that is backed up by empirical facts; the other type of knowledge is the knowledge that comes from understanding how a process works. For the first type of knowledge, we can design processes to reduce error. For example, if a doctor submits a chemo order for a patient that is outside of what is considered the generally accepted range, the system can flag it. For the second type of knowledge, even if we don’t have objectively observed data, we can still engage in a hazard analysis and risk assessment, and design controls to try to mitigate the risks. There are difficulties, however with doing this:

• The standard way of doing hazard assessment and risk assessment is to assemble a group of experts and ask them to agree on: 1) How bad would this be if it actually occurred, and 2) What is the likelihood of it happening? You then assign a risk High/Medium/Low based on the two factors. In practice, most risks come out medium, and you only devote resources to those few risks that came out high.
• The hazard analyses and risk assessments are subjective. There may not be sufficient information to come up with good answers and we are reduced to guessing.
• There is a tendency to self-censor and not talk about risks that reflect badly on oneself or one’s colleagues.
• We cannot possibly cover all the possible hazards and risks in this way. After a while you just get satiated.
• Risks developed this way are not defensible. It’s very easy for some third party that outranks you to say ‘No – we don’t accept that that is a risk.’ Sometimes you suspect that that decision is motivated by a hidden agenda, but without facts it is hard to win the battle.

What this means for system safety is that you should never for a moment think that you have all the risks covered. You don’t. Accidents will still happen. Having skilled practitioners in the front line, versed in process knowledge, is needed to catch proto-accidents that did not get caught by the processes. In the medical arena, the folks in the front line are not the specialists or even the attending doctors, but the nurses, who are carrying out the processes. Minor incidents happen all the time and do not get reported. For example, this morning I was given medication meant for the patient in the next room – it was only when I questioned why I was receiving this medication that the nurse realized her mistake. I wonder how many times that patient has received my medication?