Ninth Annual Maternal and Child Health Epidemiology Conference / December 10-12, 2003
RUSSELL S. KIRBY: Actually, I have a disclaimer, I wouldn’t call myself an expert in economic analysis, I am a social scientist, I’ve actually taught economic geography, but wouldn’t call myself an expert. In fact, the presentation that I’m going to give you will show you that I’m definitely not! Because what I’m going to introduce you to are some of the best ways to do this work poorly, and you’ll see for sure that I’m not an expert, but you’ll probably also get some insights that will help you to think about some of the ways to do this well. I’ve put together a number of these top ten lists, and some of you have been at the conference and maybe have seen some of the others, but this one is a brand new list, and when I create these lists I send out emails to likely folks who might have some insights, hopefully somewhat humorous, about the topic…and they contribute things.
This is the first time that I had somebody contribute who didn’t want to be named. All I can tell you is that it is a person who is a health economist, and he thought he was maybe being a little bit…well…touching on some sacred cows a little bit too much, so he didn’t want to be named. But anyway, we’ll look at the ten best ways to do bad economic analyses in Public Health.
The first one is to worship the sacred cows and remember that the federal poverty rate measure is sacrosanct, and it means the same thing in all regions and cultures across the United States, and if you’re interested, you might actually do some internet search and see if you can find out how the federal poverty rate was created, and where it cam from, and you’ll get some sense of why this is one of those things on the top ten list.
Number nine: Deduce! After all, It’s elementary, my dear Watson! First of all, assume all other things being equal. And then, following the completion of your analysis, interpret and apply the results as if you made no heroic assumptions. And if you don’t think that this is a good approach, take a look at what we do with clinical trials…and they’ve been doing the same thing with evidence-based practice for years. So why not do it economic analysis too?
Number eight: What you see is what they got! Charges for healthcare services always reflect the actual cost to the provider or agency of providing those services. So go ahead and use the information on charges in the hospital discharge record or other sources because that’s actually how much they got paid. If any of you have ever received a bill, you’ll know what we’re talking about here.
Number seven: Profits? Or profits? Increasing healthcare costs are solely the result of inflation. That’s the only reason why healthcare costs go up from one year to the next. And since you believe this, you’ll also understand, just for example, that the cost of micro-computers continues to increase and that today’s new Dell costs more than the original TRS or Apple 2E, and of course, it does exactly the same thing. No.
Number six: Can’t you read? Economic jargon should always be used excessively in any Maternal and Child Health report. So, I just gave you an example, of a sense that you might want to put in the summary document of one of your reports. ‘MCH professionals show willingness to pay for cost effectiveness research provided the contingent valuation does not exceed the economist’s indirect costs for the for the standard gamble that quality results will be forthcoming.’ So that’s a nice use of jargon that I think will be really good to put in a report you’re going to send to the legislature…particularly when they’re doing the budgets and then you can join the picket lines, or’ get in line for welfare.
Number five: Cost savings is all that matters! Economic analyses shows that pre-natal care is not directly associated with better birth outcomes. There have actually been a number of studies that show this, so it would be best to conclude, on the basis of these studies, that since prenatal care services are a cost center in the agency budgets, public health programs should not invest in these services for their clients. Because, after all, cost savings is all that matters.
Number four: Try to make it real…compared to what? The economic data in dollars and cents can always be used and analyzed validly with no conversion or adjustment even from differing sources or over time. This point was made clearest to me at a conference I went to in the 1980s where an economic historian from the University of Mississippi said, when he was asked whether his labor data were actually converted into real dollars, he said “How would I know about real wages? I teach at ‘Ole Miss’!” I’m sure other Universities have equitable pay scales.
Number three: (And of course, sate governments do too). From conceptual to operational definition in just one easy step! Race and ethnicity can always be used as proxy measures for income or socio-economic status. And the corollary to this is that any absurd difference in pregnancy outcomes or healthcare utilization across race and ethnic groups, must, therefore, be due to differential income and wealth. Because, after all, there are direct proxies.
Number two: I know a cost when I see one! The following terms are all synonymous and can and should be used interchangeably: ‘cost benefit, cost efficient, cost effective, cost equity, cost utility’…they all mean the same thing! Well, they mean the same thing to ME, they probably don’t mean the same thing to economists…but go ahead and use them interchangeably.
And then finally, number one: It’s all done with smoke and mirrors! There is only one essential difference between health statistics and health economics: in order to do statistics, you actually need data! Although since I’ve put that slide together, I realize it’s not actually true. There are some methods, such as borrowed strength methods in statistics, where you can actually do statistics without, or, well, at least without very much data.
Generally speaking, you need data for statistics.
So that gives you your top ten list of the ten best ways to do bad economic analysis in Public Health. As Scott and Stephen talked about some of the right ways to do things, maybe we see how this all connects up together. Thank you.