August 8th, 2008

a word about scheduling

As a young 'un, my sister (whose early education and career path I emulated assiduously) said that because she was able to estimate accurately (or at least, more accurately than her colleagues), she was highly desirable (as in Bell Labs highly desirable) to employers. This struck my young and impressionable brain as An Important Datum and I never forgot it.

On a slight tangent, I have a sister-in-law who made partner at a NYC law firm -- part time (she almost made it at a different one earlier but wound up moving out West for a while. She says she is Highly Desirable to clients because she's actually polite to them, even on the phone, which is apparently enough to really differentiate a lawyer in NYC. But that I learned recently, and other than passing it along as An Important Datum, I'm not sure what use it will ever serve me. It might help you or someone you know tho so don't hesitate to pass it along.

Returning to the ability-to-estimate question, despite knowing in principle the importance of this, it took me a few years to get a handle on Actually Doing It, so I've committed my own share of Hideously Aggressive Estimates. By the time I got to Amazon, however, I had a system worked out, and it basically amounted to this. Break the work down into recognizable bits. Estimate each bit, assuming you get to work on it 4 days a week, 2-3 hours a day (because with the level of interruptions I suffered, that was optimistic). Request time estimates from anyone else involved, and add comparable amounts of padding. Then build into the entire schedule the Sick and Vacation estimate (this is a seasonally adjusted percentage of days lost to sick days and vacation days, that can incorporate actual plans if known, but is otherwise produced as a Wild Assed Guess). This is different from putting together a full-time-on-project estimate and then padding by n percent -- the numbers can vary wildly, and if you have a real handle on how much time you have during the week to do Real Work On This Project and your Sick/Vacation guess is good, you can really nail the due date pretty consistently. It's really fun to do this to other people (take their estimate and hand back what you really think it will take -- and be right).

I'm sure that scheduling appointments like in a health care providers office is different. There's a longish thread over at the Well on the NYT where a lot of bitching about doctor's are always late is going on, and the doctors say well there are emergencies and people who've been around the block a few times point out that those emergencies happen with _stunning_ regularity. So I'm going to take a little stab at How to Schedule Appointments to Minimize Being Late for Clients (because I don't like the term patient. It implies we should be. And as Mr. A. said, patience is one of the Lesser Virtues).

(1) Based on how late you are running each day (or how early you have to stack the pre-start-of-day-appointments), schedule enough open appointments to return you to a reasonable length day (you define). If by some miracle, there is no one who needs that same-day slot, or who spills over into that slot, or whatever, you can use that time for coffee, a meal, charting, etc. If you were already building in some number of slots and still spilling over, build in more.

Just to indicate that I have a clue how this works, there were days at Amazon that I brought a magazine. I couldn't get enough time to read e-mail and respond, so I'd sit and read a paragraph or two of an article in a magazine until the next person showed up at my door who Had To Talk to Me Right Now Because It Was SOOOOOO Important. Maybe you shouldn't schedule _any_ appointments in advance. Especially for the first week after school starts (if you see children). Or during flu season. Or whenever the really-busy-time is for you.

(2) Refer patients to another provider. Aggressively. If you're that busy, you aren't helping them. Help them find someone else who can. Elsewhere, we call this delegation. If you're doing a bang up job as a health care provider, you should think of yourself as requiring a team of people to replace what you have been doing; start finding that team and handing work off to them. Now.

(3) _In addition to_ scheduling open appointments, stagger less-than-appointment length blank periods between _every single appointment_. So if you have 15 minute appointments, don't schedule more often than 15 minutes + n. That way, if someone runs over, you've at least got n. If you genuinely cannot help people in the time slot that you've been assigning, expand the slot and then pad. If you cannot do this and stay in business, that's a strong clue to find another business. If you think you can help your client in that time slot, but your clients differ with you, see #2.

(4) Once you have this system up and running, you should see the following pattern: you are waiting for the beginning of the next set appointment as often as you are late to the next set appointment. The distribution and details will vary, but the goal is to find the middle. Once you've got that nailed down, start training your patients to arrive 10-30 minutes early, so they can _fill in the blank spots_. Good practices do this. I've shown up early and gotten someone else's slot because they weren't early enough for it -- or they were late. Either way, I benefited, efficiency benefited, etc.

(5) Stop taking new clients when you hit a point where serving your existing population is taking up almost all of your time. Be really serious about this. See #2.

There are almost certainly practices and specialties and locations where this set of rules is doomed. (ETA: Docs who take free patients who qualify, for example, are probably going to burn right through all these rules for economic reasons.) Rural areas with inadequate service is an obvious one. There's one really common scenario, however, that does _not_ excuse failure to adhere to this set of rules: your clients want to come in almost exclusively in the mornings before school/work, in the afternoon or evenings after school/work, on Fridays, or on the weekend. A lot of practices as a result are zombieville during the daytime (I _love_ this and I have for years) and madhouses during those hours on the edge. Further, they run late a lot, due to traffic, child care snafus, unexpected meetings at work, blah, blah, blah. Probably a lot of no shows, too (altho with the number of people claiming 20-25%+ no shows, I gotta wonder what the hell is going on). It seems to me that you probably should just close up shop and treat the whole thing as shift work. Clearly, you wouldn't lack for clients.

Obviously, this stuff can only be implemented at the business level (and solo practices are a bygone operation, so it's probably not the health care provider's decision). But it _can_ be implemented at the business level -- I've been to practices that consistently worked this way. Of course, they also are heavy on the nurse-practitioner/physician's assistant/non-MD staff. Maybe that's the real key.

ETA: And what _is it_ with people and eye contact? There's this whole other subthread about people demanding eye-contact or being upset when eye-contact is demanded. I thought this was a well-understood difference -- some cultures do, some don't, it's usually a power thing, and of course there's the whole ASD issue.

one of the most fun things about that excitement in Seattle

Was that you got to walk down the middle of the street without having to worry about cars.

Well, NYC is going to try closing a street in Manhattan -- almost 7 miles of it -- for 3 Saturdays. Like a big ole block party.

I particularly like the way Bloomberg is quoted as describing it:

"This has been done in Bogotá for 30 years. They love it. It’s phenomenally popular and it probably will work here. If it doesn’t, at least we’ll have tried."

Between running the city for several years now in a remarkably scandal-free fashion, and this kind of communication style, it's no wonder to me why this man's popularity only increases.

_Clinical Decision Making_, David M. Eddy

Subtitled: From Theory to Practice: A Collection of Essays from the Journal of the American Medical Association

Back on June 23rd, I posted about several things, including what a really good book about decision theory should explain.

This does not meet my criteria for what a really good book about decision theory should explain. However, it's still worth putting a substantial amount of time and energy into, and has a lot to reward a reader, whether they are in the medical industry or not.

Each chapter is a previously published essay (footnotes explain what, if anything, was changed for this publication; usually it's pretty minor things like the exact title and possibly a term-of-art changed between original publication and this book) on the subject of cost, benefits, harms, medical practice, what is being done, why we can't continue to do things this way indefinitely, what doctors should do instead, and what non-doctors will do to doctors if doctors don't take effective action for themselves. Pronto.

There are a number of basic assumptions built into Eddy's View of the World, and they are at odds with a lot of my View of the World. Like, he thinks people should be rational when deciding how to Live Their Life (or Die Their Death). He never/only rarely and indirectly allows for the possibility that common medical practices do far more harm than good -- so his worst case scenario is no-benefit, whereas mine is way-more-harm-than-if-you-dealt-with-your-thing-yourself. This is why Eddy, at least for a while, practiced medicine, whereas I contemplate the appeal of Christian Science at irregular intervals. He also seems to really believe that medicine in the past had a good grasp of what was worth doing and what wasn't; anyone with a decent grasp of any given time-and-place in history knows that this hasn't ever been true in the past. A reasonable corollary is that it is not true now. And _that_ is part of the appeal of Eddy. He basically thinks we don't know what the hell we're doing in medicine right now AND we're throwing crazy amounts of money at it and THAT is not sustainable. He tackles how to carve away at the (growth of) costs while improving quality.

The essays are from the 1980s and 1990s, so they span Oregon's proposal (but not its actual implementation after the Feds said, um, you _do_ know about the ADA, right? And this whole priority scheme you've devised which aims first and foremost at a return to full health/ability is a violation of the ADA. So fix it.) and the Clinton administrations proposal. Eddy consulted for Kaiser Permanente (so we can guess right there I'm biased in Eddy's favor) and helped them devise practice guidelines. During the 1980s, Eddy's perspective was, dude, there's a lot of unjustifiable variation (cf Wennberg) so we need Research and we need Guidelines and computers and math are probably quite necessary. During the 1990s, Eddy morphed a bit, got sucked into the preventive screening thing and got considerably more frantic on the, hey, this is _so_ not sustainable! issue. He also seems to have gotten progressively more suspicious of virtually all specialties, at least when it came to explaining how effective their standard practices were.

Eddy includes a lot of himself in this book. Several essays are imagined conversations with his (deceased) dad -- and you can imagine this could be a big deal for a 4th generation doctor who is working his way slowly around to questioning the entire enterprise of medicine. The concluding chapter is touching (and you know me -- that's not a word I normally go anywhere near) and inspiring, a description of his mother's decline and how the (remaining) family worked together to help her get where she wanted to go. Eddy is a numbers guy, and Eddy has no mortal clue how to translate what he is advocating into language understandable to someone who does things because they feel right or because everyone else is doing them or whatever. He recognizes this, but shies away from coming up with any alternative explanation. If you can't meet Eddy in RationalWorld (with some numbers), you won't meet Eddy anywhere.

Here's the problem with preventive screening of all sorts. You take hundreds of thousands if not millions of _healthy people_ and, based on an imperfect -- often very imperfect -- test, subject them to a series of medical risks and ultimately diagnose some fraction of them as suffering from a disease which if it had caused them to come in with symptoms would be widely recognized as a pretty bad thing (breast or prostate or cervical cancer, say). You treat them (again, risky). The hope is that you caught it before they got sick, and if you do this with enough people, over the population as a whole, the rate of people getting sick and having nasty unpleasant outcomes (death) from that disease will go down. But if it turns out that in the end, the total number of people dead from treatment and disease is equal to the unscreened population due to disease alone unscreened, you spent a lot of money with no gain. And if it turns out that you killed a bunch of asymptomatic people by "treating" them for a "disease" AND you still have the same number of deaths due to disease, you just spent a whole helluva lot of money to kill healthy people. There's a lot of reason to believe that's _exactly_ what we're doing with PSA tests, and we might well be doing it with mammograms for 40-50 year old women, and heck, we might be doing in too many women with pap smears, but that's actualy probably not the case.

And that's not counting the people who survived "treatment" for an asymptomatic disease -- but, say, don't have a dick or a boob or a womb any more. And might really want one. As a for instance.

If we had really amazingly perfect research, maybe we could avoid doing this. Or at least spot it after the fact. But Eddy even does a pretty good job of explaining why this is kinda hard -- not to mention expensive -- to do. He thinks we should do it (so do I); at the time he was writing these articles, he was a bit more optimistic about how that would turn out than I am.

If Eddy ran the world, it would be, on balance, a better place. I think. If a lot of people listened to Eddy, remained skeptical, but adopted heuristics like, when in doubt, don't, the world would be, on balance, a _much_ better place. I don't know how to get us from here to there.

But that does not mean I'm giving up. Somewhere around here, there's a book on public policy. Maybe I'll track it down and read it. After all, I'm almost to the last month of pregnancy, when I focus on how I haven't had the baby yet, am not ruler of the universe yet, haven't ascended yet, etc.

doctors who don't like crises, and doctors whose ancestors include doctors

I've commented in previous reviews (in previous pregnancy) on what happens when the offspring of docs write about doctoring, specifically in regards to ante/prenatal care (Thomas Strong is the American example; Annie Oakley, IIRC, is the Brit).

David Eddy, in addition to providing a bang-up example of someone born to a doctor (4th gen) becoming really insightful and critical of a lot of aspects of medicine, demonstrates another salient characteristic of doctors I mentioned previously: he doesn't like medical crises. Odent and Grantly-Read were both army medics (different armies) who went into obstetrics in part because they were looking for something non-heroic. Eddy didn't go into obstetrics (altho he apparently is known in part for sleuthing down the origin of the once-a-cesarean-always-a-cesarean guideline), of course, but I can't help but feel that all of the above mentioned doctors, whether OB or not, gravitated towards wonkitude/policy theory -- both to cut down on the crises and because they've lived and breathed this world so long they can't really help themselves.

The more I read, the more I feel compelled to conclude that we're all doomed to dissatisfaction with medical care, especially here in the US, but generally as well.

Well, it's not like dissatisfaction is an unusual state of mind for me.