3 ms·

You're describing Brook's Law from The Mythical Man-Month. It was an observation of collaboration in software engineering specifically, and it cannot be applied

by kedean 4y ago

You're describing Brook's Law from The Mythical Man-Month. It was an observation of collaboration in software engineering specifically, and it cannot be applied universally to every industry. Really, anything that is highly parallel (medicine, teaching, stocking shelves, waitstaff, deliveries) can benefit from hiring more people until you reach saturation, and medicine isn't there or we wouldn't be having the conversation. The problems are caused by a "just in time" approach to staffing, where you have exactly enough people to cover the shifts at bare minimum. What solution would you suggest other than more people? They are not saying to throw more nurses at patients simultaneously, they are saying to hire more nurses so existing ones aren't bound to spent the entire week stretching themselves across the hospital. Also, I think you are being downvoted because you are applying software engineering rules to medicine.

bumby 4y ago
It's interesting because they aren't software engineering rules. If anything, they are industrial engineering rules that pre-date software and certainly older than the mentioned book from 1975. The approach I was using was developed specifically for healthcare and with great effect in some organizations. I know this is HN, but I think it's an error to assume everyone is coming at a problem from a software perspective. >What solution would you suggest other than more people? It obviously depends on the situation but most of the time it comes down to reducing process waste. That may be automation through software where a nurse was hired specifically to only generate reports 40 hours a week, to re-designing a layout that minimizes travel time for nurses when they are delivering to patients. My experience with the staffing situation is that managers did not know how to staff to meet the needs of their patient loads and just revert to simple heuristics that left them understaffed at some times while being overstaffed at others.
- phil21 4y ago
  > re-designing a layout that minimizes travel time for nurses when they are delivering to patients. I think talking about micro-optimizations like this misses the forest through the trees. It's neat and cool. Fun to wring out those last bits of efficiency. But the fact you even need to discuss it shows how hiring adequate amount of bedside staff is the absolute last thing any medical system will do. > revert to simple heuristics that left them understaffed at some times while being overstaffed at others Showing that they were better than modern day automated shift planning. I will submit that if your hospital floor staff is not 50% idle on your average given fully-staffed boring day, you are understaffed. Only extremely exceptional events should cause your staff to be booked 100%. When it happens it should be root cause analyzed and be immediate cause for executive concern. The trope of card playing nurses should be true, because of all industries there are - you want surge capacity in healthcare. Both physically speaking in terms of warm bodies available, as well as mentally speaking in brains not being stressed to their max the entire shift. I understand these ideas don't end up with maximal efficiency. I am likely naive as well - but it's pretty clear that hospital systems are being operated in an intentional manner to keep staffing cut to the bone right up until the point the system or people break. This doesn't just hold true for healthcare, it's endemic in a lot of service industries where the bottom tier employees are expected to live in service of the profession and take up all the slack from above.
  - bumby 4y ago
    >I think talking about micro-optimizations like this misses the forest through the trees. It may be different in practice than you are imagining. It's been years so I don't have the exact numbers, but something on the order of 10-20% increased patient throughput in radiology of a major metropolitan hospital that provides for an underserved community. (Keep in mind, it's not just nurses who have to traverse hospitals but patients. Many of them are old or lack mobility, so the changes are compounded.) I don't consider that "micro" but you may have different expectations. As a comparison, I don't think nurses would think a 20% increase in staffing is trivial, but my preference is to measure at the patient level, because that is the outcome I'm most concerned with. >Showing that they were better than modern day automated shift planning. Can you explain? I'm not following this statement. I was involved because it wasn't working. >you want surge capacity in healthcare. I agree. But like all complicated systems, you rarely get something for nothing. The trope in healthcare is you can choose between quality, cost, and access, but you only get to choose two. Lots of slack in the provider supply side is great for surge capacity, but it generally comes at the expense of cost. That's difficult when many of the same people who complain about the lack of providers also like to point out the cost of healthcare. I would prefer to take system-level view so we aren't essentially just shifting the waste around in the system and instead work to cut it out completely. In this case, excess capacity isn't waste but a desired measure of resilience. But I do think it needs to be measured and managed. My experience has been that nurse managers will just constantly increase the slack in the system rather than address the other issues. And I absolutely agree there are issues, across many industries, where JIT thought-processes are short-sighted. COVID, and prior to that, Fukashima, has demonstrated this error. There are certain critical systems that we do not want operating at maximum input/output efficiency. If the staffing issues I've referenced were consciously framed in that sort of pragmatic, data-driven argument, I would be the first to stand behind it. But often they were really coming from a "we're-overwhelmed-and-don't-know-what-else-to-do-so-just-hire-more-people" perspective.
  - syntheweave 4y ago
    There's a general rule of thumb that I have about any business: if it "looks busy" and people are running around, yelling into phones, unavailable for questions, etc., it's probably either understaffed or ineffectively organized. A workplace where things aren't visibly moving is one that is ready to make plans and respond to events smoothly. The productivity rate is somewhat independent of the "busyness rate" since there are a huge number of ways to stop work from needing to be done in the first place. However, this seems to be unsatisfying to modern management styles. Apparently if you aren't running the employees into the ground like an Amazon warehouse, you're "leaving something on the table".
mcv 4y ago
I'm suddenly reminded of traffic. For a long time, governments tried to fight traffic congestion by adding more lanes. Traffic is highly parallel, so this should work, right? But it doesn't. Traffic grows to fill up those additional lanes and you're stuck with the same congestion. Here, better traffic design is necessary. And room alternative, more efficient forms of traffic (public transport, bikes). I do think nursing (and many similar fields) do need to hire more people, but I also suspect they have to redesign how nurses work. Less overhead, more focus on the core of their work. Streamline the processes, especially the administrative side, and not expect them to fill in for many other kinds of work at the hospital. Get separate specialists for that.

by kedean 4y ago

bumby 4y ago

It's interesting because they aren't software engineering rules. If anything, they are industrial engineering rules that pre-date software and certainly older than the mentioned book from 1975. The approach I was using was developed specifically for healthcare and with great effect in some organizations. I know this is HN, but I think it's an error to assume everyone is coming at a problem from a software perspective. >What solution would you suggest other than more people? It obviously depends on the situation but most of the time it comes down to reducing process waste. That may be automation through software where a nurse was hired specifically to only generate reports 40 hours a week, to re-designing a layout that minimizes travel time for nurses when they are delivering to patients. My experience with the staffing situation is that managers did not know how to staff to meet the needs of their patient loads and just revert to simple heuristics that left them understaffed at some times while being overstaffed at others.

phil21 4y ago
> re-designing a layout that minimizes travel time for nurses when they are delivering to patients. I think talking about micro-optimizations like this misses the forest through the trees. It's neat and cool. Fun to wring out those last bits of efficiency. But the fact you even need to discuss it shows how hiring adequate amount of bedside staff is the absolute last thing any medical system will do. > revert to simple heuristics that left them understaffed at some times while being overstaffed at others Showing that they were better than modern day automated shift planning. I will submit that if your hospital floor staff is not 50% idle on your average given fully-staffed boring day, you are understaffed. Only extremely exceptional events should cause your staff to be booked 100%. When it happens it should be root cause analyzed and be immediate cause for executive concern. The trope of card playing nurses should be true, because of all industries there are - you want surge capacity in healthcare. Both physically speaking in terms of warm bodies available, as well as mentally speaking in brains not being stressed to their max the entire shift. I understand these ideas don't end up with maximal efficiency. I am likely naive as well - but it's pretty clear that hospital systems are being operated in an intentional manner to keep staffing cut to the bone right up until the point the system or people break. This doesn't just hold true for healthcare, it's endemic in a lot of service industries where the bottom tier employees are expected to live in service of the profession and take up all the slack from above.
- bumby 4y ago
  >I think talking about micro-optimizations like this misses the forest through the trees. It may be different in practice than you are imagining. It's been years so I don't have the exact numbers, but something on the order of 10-20% increased patient throughput in radiology of a major metropolitan hospital that provides for an underserved community. (Keep in mind, it's not just nurses who have to traverse hospitals but patients. Many of them are old or lack mobility, so the changes are compounded.) I don't consider that "micro" but you may have different expectations. As a comparison, I don't think nurses would think a 20% increase in staffing is trivial, but my preference is to measure at the patient level, because that is the outcome I'm most concerned with. >Showing that they were better than modern day automated shift planning. Can you explain? I'm not following this statement. I was involved because it wasn't working. >you want surge capacity in healthcare. I agree. But like all complicated systems, you rarely get something for nothing. The trope in healthcare is you can choose between quality, cost, and access, but you only get to choose two. Lots of slack in the provider supply side is great for surge capacity, but it generally comes at the expense of cost. That's difficult when many of the same people who complain about the lack of providers also like to point out the cost of healthcare. I would prefer to take system-level view so we aren't essentially just shifting the waste around in the system and instead work to cut it out completely. In this case, excess capacity isn't waste but a desired measure of resilience. But I do think it needs to be measured and managed. My experience has been that nurse managers will just constantly increase the slack in the system rather than address the other issues. And I absolutely agree there are issues, across many industries, where JIT thought-processes are short-sighted. COVID, and prior to that, Fukashima, has demonstrated this error. There are certain critical systems that we do not want operating at maximum input/output efficiency. If the staffing issues I've referenced were consciously framed in that sort of pragmatic, data-driven argument, I would be the first to stand behind it. But often they were really coming from a "we're-overwhelmed-and-don't-know-what-else-to-do-so-just-hire-more-people" perspective.
- syntheweave 4y ago
  There's a general rule of thumb that I have about any business: if it "looks busy" and people are running around, yelling into phones, unavailable for questions, etc., it's probably either understaffed or ineffectively organized. A workplace where things aren't visibly moving is one that is ready to make plans and respond to events smoothly. The productivity rate is somewhat independent of the "busyness rate" since there are a huge number of ways to stop work from needing to be done in the first place. However, this seems to be unsatisfying to modern management styles. Apparently if you aren't running the employees into the ground like an Amazon warehouse, you're "leaving something on the table".

mcv 4y ago

I'm suddenly reminded of traffic. For a long time, governments tried to fight traffic congestion by adding more lanes. Traffic is highly parallel, so this should work, right? But it doesn't. Traffic grows to fill up those additional lanes and you're stuck with the same congestion. Here, better traffic design is necessary. And room alternative, more efficient forms of traffic (public transport, bikes). I do think nursing (and many similar fields) do need to hire more people, but I also suspect they have to redesign how nurses work. Less overhead, more focus on the core of their work. Streamline the processes, especially the administrative side, and not expect them to fill in for many other kinds of work at the hospital. Get separate specialists for that.