Does More Primary Care Increase Healthcare Costs Instead of Lowering Them?

Health care reformers often argue that increasing patients’ access to doctors (especially primary care doctors) can actually lower health care costs in the long run, as these doctors can help diagnose and manage conditions before they lead to more expensive treatments and hospitalizations.  But a new paper by economists Robert Kaestner and Anthony T. Lo Sasso disputes that theory. Here’s the abstract:

By exploiting a unique health insurance benefit design, we provide novel evidence on the causal association between outpatient and inpatient care. Our results indicate that greater outpatient spending was associated with more hospital admissions: a $100 increase in outpatient spending was associated with a 2.7% increase in the probability of having an inpatient event and a 4.6% increase in inpatient spending among enrollees in our sample. Moreover, we present evidence that the increase in hospital admissions associated with greater outpatient spending was for conditions in which it is plausible to argue that the physician and patient could exercise discretion.

The authors further conclude that “the implication of these findings is that expanding health insurance, as recent federal reform (Patient Protection and Affordable Care Act) proposes, will be cost increasing.”


perhaps they should contact the CBO, which has documented how much is saved by expanding coverage as compared to non-coverage- 2 important points here is to: 1- shunt uncovered citizens away from ER visits to MD visits- ER visits are much more expensive; and 2- shunt uncovered citizens to preventative care, which optimally does not include an MD (clinics with nurses or other professionals)- MDs are overpaid and so any mass increase to them preventative or no will be overpriced


Are ER visits really more expensive? This seems more like a situation where it's extremely expensive to treat the first patient, but pretty cheap for subsequent patients. We're going to operate ERs, so we've already committed to the huge fixed expense involved; is the marginal cost of treating an additional patient really that high? It doesn't necessarily require additional staff or facilities.

Joe J

The economics of hospitals is a book in itself, and of ERs is a set of books, though it should be put in the comedy section.
ERs have a high rate of nonpayment. Which is born by those who do pay.

Todd DeYoung

It seems that in order to come to their conclusion you'd also need to look at the costs of inpatient stays driven by early intervention vs those driven by emergency.

Is this addressed beyond the abstract?


Interesting study, however our company has found the opposite to be true for our managed care members. While our model is specific to Medicare, we have found that investing in the primary care practice and building quality initiatives for those physicians has led to direct savings and reduced hospital admissions. In fact, we can't understand why more health plans do not build their delivery model around a strong primary care practice.

Enter your name...

The age of the population might matter. People in their 70s and 80s might make different choices than people in their 30s and 40s when it comes to "discretionary" surgery like most knee surgery.


Healthcare costs still fall in this model, but the "volume" of care rises which makes the spending higher. This still seems like a desirable outcome to me.


I'm left with a so what?

What I gather is the following: Increased outpatient spending leads to increase in inpatient visits.

In fact, that's more or less exactly what you'd expect. If you have more access to say, primary care, you're more likely to also detect things that will require inpatient care.

Now the real questions are:
1. Was the resulting inpatient care (in total) less costly when does this route v. the emergency department route or detected later route?
2. Did this affect actual access to healthcare in a meaningful way? One of the upsides to outpatient care (and primary care, generally) is that you can schedule it. This has it's own value too. It has value from the provider side and from the patient side.

Arguing that one aspect of care increased costs in another is almost entirely the wrong question.

I take my car to the shop every 5k miles. Every once and awhile they find something more expensive to fix than an oil change. I don't get mad that my increased spending on regular service meant that I had to leave my car at the shop once in awhile. Rather, I justify it because having something break down on the express way to work is much more costly even if I end up with my car in the shop in both cases.

Extending the analogy, I'll also note that sometimes they "find" something that may not manifest itself in an expressway breakdown (or perhaps at all during my ownership of the car), and I may end up paying to have it fixed anyway. I guess this means that my spending increased because I knew about it even if I may never have had an issue if it had gone unnoticed.


Billy D

3 interesting things:

1. The study involved a "high deductible" insurance plan, which should generally discourage preventitive visits compared to a low deductible plan. I believe this would actually make their findings understated.

2. It was for a 6-year time period. I assume that most larger health issues that were "prevented" by preventitive visits would have otherwise arisen within that time period, but some issues (e.g. cancer) might have only shown up after the study period. This would seem to work in the opposite direction as #1 above.

3. It was funded by the American Enterprise Institute. They have a certain pro-business/anti-regulation bias that makes you think they knew what they wanted the study to say before it was ever conducted. I'm not making any specific accusations about the study's integrity, but we should keep their bias in mind.


I don't think this result should be surprising. A trip to a primary care physician might be much cheaper than a trip to the ER, but you also have to account for all the people that opted to visit neither until they had an incentive to visit the primary care physician. Plus that expensive ER visit will typically only focus only on relieving the immediate issue and getting the patient out the door, while an outpatient doctor is likely to test & probe for other issues that can lead to further costs.

While all the prevention and early intervention programs sound like they should save money, any savings tends to be very short term. If the prevention or early intervention program actually works to prolong the patients life then those savings are dwarfed by the huge increase in lifetime healthcare costs. They can certainly be worthwhile in terms of quality or quantity of life, but in strictly monetary terms any improvement that increases lifespan is bound to increase the total cost of lifetime care.

The ultimate problem for lowering healthcare costs is that everybody dies from something. Treating the health issues that can be treated affordably just means that more people will live to develop different health problems that are more likely to be horribly expensive. From a purely financial perspective preventing heart disease or catching cancers earlier just means we will eventually have many more people suffering from dementia or having an expensive decade long decline with multiple organs slowly failing.


Jason Ward

An important conclusion that can be drawn from the study which was (in my opinion irresponsibly) left out of the blog post is the following Doctors appear to be the causal factor of this increase rather than users of healthcare services.. In fact, the novel insurance feature that allowed that authors to ferret out this finding was a plan that made inpatient care potentially or actually MORE expensive for the patient. Standard economic theory would clearly predict this would lead to less inpatient care (the plan only allowed money paid into it to be used for outpatient services). This points clearly to the role of physicians in driving overtreatment and the study points this out using rather indirect language. Consumers of healthcare are by definition always under informed about the efficacy of treatment. The need is for more evidence based care decisions and adjustment of physicians' incentives, not less access to healthcare coverage for the marginally insured.

A better headline would have been "Do physician incentives serve to thwart plans designed to lower healthcare costs?"

Jason Ward
Econ student - The University of Illinois at Chicago



I would guess that more primary care = more early detection of problems. The costs of treating these early detections would need to be offset against the cost of the generally more catastrophic events where problems are not picked up early (i.e. cost of inserting a stent into an arterial occlusion vs cost of open heart surgery following an infarction).

The authors state "outpatient care this year may affect inpatient care in the subsequent year." which is a rather short time frame to detect any reduction in relatively rare but expensive and catastrophic outcomes. The conclusion "While the greater costs associated with increased use of primary care may come with improvements in health, a hypothesis in need of more credible evidence, costs will
increase nevertheless." might be subject to scrutiny in terms of a short time horizon.

By not including health outcomes, it is an analysis of costs, but not cost effectiveness.

It may cost 4.6% more, but what if you get 10% more health benefits?



Let's say that primary care finds a tumor that would otherwise kill you a year later if undetected.

Would you care one bit about this study?


In that case, the inpatient care would result in an increase in overall health care costs - exactly the type of sloppy reasoning used to argue against the Act in the first place. Extend the idea: You have no health coverage and do not find the tumor until you're sick enough to go to the Emergency Room, a condition I assume those covered by the unique model mentioned did not have to experience. What happens to the cost of health care then?

Lynn Weathersby

This is a very interesting article, however, the authors did not convince me that a causal relationship exists. It is not a well thought out assumption. There may be a relationship between increased outpatient care services leading to increased inpatient care services but this doesn't really speak to cost as measured by dollars. The cost of medical services are artificially inflated by taking advantage of unsuspecting patients. In example, I took my son to the doctor for an infected cuticle in which a "blister" had developed. The nurse offered to "pop" the blister and put a band-aid on it. I accepted. I subsequently recieved a bill for $624.00 for this "service". I was never informed by the nurse prior to its administration the cost of such a "procedure". In any other industry this kind of business behavior would be considered fraudulent. There are MANY things wrong with the healthcare industry in this country, the most significant is making a profit off illness and disease.



Confounding variables? Distortion of the sample? The sample in question were customers of this insurance which implies first they can afford health insurance and second they chose this policy. I have difficulty extending a "unique health insurance benefit design" to the entire population of consumers of medical care. And I have a much higher level of cognitive dissonance when they claim “the implication of these findings is that expanding health insurance, as recent federal reform (Patient Protection and Affordable Care Act) proposes, will be cost increasing.” To the point I wonder, did American Crossroads fund this research.

Nathan Stockstill

In medicine, outpatient care is separated into primary care (care that prevents illness) and secondary care (care that treats illness). Hospitalization is considered tertiary care. "Primary care providers" provide both primary and secondary care.

Kastner and LoSasso don't take this into account in their study. The arrangement they studied, as well as most insurance arrangements, discourages true primary care. In addition, many people must sacrifice wages to see a doctor when they're not sick so they choose not to do it.

True primary care, virtually by definition, reduces inpatient spending. Secondary care mostly treats conditions that, without care, would never get to the point of requiring hospitalization (although I'd propose that it reduces days of work lost). Just as fear of lawsuits drives unnecessary testing, we also find it results in unnecessary hospitalization.

I've seen studies that use insurance databases to determine purpose of visit. I think Kastner and LoSasso's study would be more valid if they had applied similar data for further analysis.



The subject of the study is access to outpatient treatment, not primary care. These are not the same thing. The headline is misleading.