It’s widely known that it costs a lot more to live in some parts of the country than others. Off the top of your head, you can probably name New York, San Francisco, and Washington, D.C. as examples of cities where it is extremely expensive to live. Of course, that’s just the tip of the iceberg. It’s not like Boston, Chicago, or Los Angeles are particularly cheap. Generally speaking, big cities are more expensive. A large part of that is likely supply and demand.
It’s widely known that it costs a lot more to live in some parts of the country than others. Off the top of your head, you can probably name New York, San Francisco, and Washington, D.C. as examples of cities where it is extremely expensive to live. Of course, that’s just the tip of the iceberg. It’s not like Boston, Chicago, or Los Angeles are particularly cheap. Generally speaking, big cities are more expensive. A large part of that is likely supply and demand. When a bunch of people want to live somewhere, and the amount of habitable space is finite, the cost of that space goes up, and that tends to drive up the costs of everything else in turn. But this also happens on a slightly more macro scale. That is, some regions of the country are more expensive than other regions of the country. For example, the coasts typically have higher costs of living than inland areas, and the northeast and west parts of the country tend to be more expensive than the south or midwest.
In a recent Health Affairs paper, Brendan Saloner and colleagues report the results of an experiment in which a group of actors pretended to be uninsured and made phone calls to attempt to schedule a new patient primary care appointment and find out what it would cost. As the title of their article gives away right from the start, the “average price was $160.” That’s not a particularly shocking number to most of us who have a sense of health care charges, but for the uninsured individual who’s not insulated from those costs, that’s a lot of money to pay to see the doctor for what might amount to about 12 to 15 minutes’ worth of a visit.
The thing I found most interesting, though, was the geographic variation in prices they identified. While they report both the mean and the median costs in their paper, I’m going to focus here on the median costs, because that figure isn’t affected by outliers in the way that the mean can be. The authors investigate the cost of new patient appointments in 10 different states: Arkansas, Georgia, Iowa, Illinois, Massachusetts, Montana, New Jersey, Pennsylvania, Oregon, and Texas. While that’s just a sample of the country, it does a decent job at capturing the various regions of the United States.
I grew curious about how the median cost for a new primary care appointment compared to the cost-of-living in each state. Now, it’s not really fair to look at the overall cost-of-living, because there’s variation within that figure. For example, while city A and city B may cost more than city C, it could be because the cost of housing is much higher in city A than city C, while the cost of transportation and health care are higher in city B than city C. Fortunately, I found a cost-of-living calculator that helped me look specifically at health care costs. The calculator gives each city a score for health care costs that are standardized to a national average value, which equals 100. Thus, a city with a score less than 100 has cheaper health care than the national average, and a city with a score more than 100 has more expensive health care than the national average.
Because I had to pick a city from each of the 10 states that the study included, I went with the state capitals. Now, that poses an issue in that some state capitals are major cities (e.g., Atlanta, Boston) while others are not (e.g., Harrisburg, Salem). So, please, recognize that this is a very back of the envelope approach to looking at this issue that is not without its limitations. Still, doing that, this is how the capitals of the 10 states rank in order from most expensive to least expensive health care (again 100 is the national average):
- Boston, MA 126.6
- Salem, OR 117.7
- Helena, MT 103.6
- Atlanta, GA 102.6
- Springfield, IL 101.8
- Trenton, NJ 101.8
- Austin, TX 100.6
- Harrisburg, PA 91.5
- Little Rock, AR 87.1
- Des Moines, IA 84.7
Okay. So now the question is: How does this translate into the median costs identified in the study? Well, ignoring population weighting and the like, the average score across our 10 capital cities is 101.8, which corresponds to the median primary care visit cost of $125 identified in the study. To scale this down to a national average, we multiply by 98.2% and find that the median national cost of a primary care visit for an uninsured patient is expected to be $123 (I rounded up). We can then figure out the expected cost for each of our cities and compare it to the actual cost identified by the researchers. This is what we find:
- Boston, MA Expected: $156 Actual: $150
- Salem, OR Expected: $145 Actual: $195
- Helena, MT Expected: $127 Actual: $150
- Atlanta, GA Expected: $126 Actual: $126
- Springfield, IL Expected: $125 Actual: $125
- Trenton, NJ Expected: $125 Actual: $125
- Austin, TX Expected: $124 Actual: $125
- Harrisburg, PA Expected: $113 Actual: $125
- Little Rock, AR Expected: $107 Actual: $120
- Des Moines, IA Expected: $104 Actual: $143
So, costs are pretty much what we’d expect in Boston, Atlanta, Springfield, Trenton, and Austin. In fact, Boston is the only place where the actual cost was less than expected. It seems that the actual cost doesn’t go below $120, so places like Harrisburg and Little Rock end up costing a little bit more than expected. Helena also costs about 20% more than expected. But the big shocks are Salem and Des Moines. While Salem is expected to be the second-most expensive, it is actually the most expensive by far–a whopping 30% more costly than the next most expensive locale, Boston. And, at the other end of the spectrum, Des Moines is expected to be the cheapest of all, but it is actually the fourth most expensive on the list. That said, I reiterate that these actual numbers are from a sample within the state and do not necessarily represent the cities as I’m interpreting it here. Nevertheless, it demonstrates that variation in health care costs don’t even necessarily reflect the differences in health care costs we know to exist. That is, there’s variation on top of the variation. Just one of the many reasons why it can be so difficult for people to navigate the health care system, particularly if they’re uninsured.