April 7, 2017

The Failure of Alaska’s Certificate-of-Need Laws

Testimony before the Alaska Senate Labor and Commerce Committee, Hearing on Repeal of Certificate of Need Program
Key materials

Chairman Costello, Vice Chairman Hughes, and distinguished members of the Labor and Commerce Committee:

Thank you for the opportunity to share my recent work on certificate-of-need (CON) laws as they are applied to health care in the Last Frontier.

My name is Thomas Stratmann. I am a University Professor of Economics and Law at George Mason University in Virginia and a senior research fellow at the Mercatus Center at George Mason University.

INTRODUCTION

In four different academic, data-driven studies, my coauthors and I examined the impact of certificate-of-need laws. We compared economic and health measures between the 35 states that have CON laws to those in states that do not have CON laws.

All four of these peer-reviewed studies are attached as part of my submitted written testimony. These studies use state-of-the-art statistical methods that are well accepted in social sciences, health sciences, and many other areas that analyze data, such that the conclusions are based on apples-to-apples comparisons—that is, I perform the analysis in such a way that states with and without CON laws are comparable. All data I use are publicly available so that my results can be replicated by anyone who chooses to do so.

My findings show that CON laws do not deliver the benefits that were promised to patients. My findings are as consistent as they are unfortunate: across the board, CON laws have failed.

  1. CON harms patients by reducing healthcare quality.
  2. CON harms patients by reducing access to health care.
  3. CON harms patients by reducing medical equipment such as MRI machines and CT scanners that help to diagnose illnesses and thereby prevent premature death.

My findings are consistent with the positions of the Federal Trade Commission and the Department of Justice under both Democratic and Republican administrations, which have argued that CON laws fail to meet their stated goals and that CON laws are harmful to patients. CON laws are not helpful for patients because they reduce the availability of medical care by making it difficult for medical providers to offer their services. Nor are they cost saving.

All of these harmful effects are particularly bad for Alaska because it is geographically distant from the lower 48 states. In the lower 48, patients can travel across state borders to access medical services not provided in their states. For most Alaskans, such travel is cost prohibitive, and they have to live with harmful effects of CON.

BACKGROUND

Certificate-of-need laws require state agency approval before an already-licensed healthcare provider can establish a new healthcare facility and before an already-licensed healthcare provider can expand. In some states, CON laws require permission from state regulators to provide medical services or to purchase medical equipment to which the government otherwise has no objections.

CON LAWS IN ALASKA

CON laws in Alaska require medical providers to obtain government permission to compete for 20 medical services (out of 35 medical services regulated across the US states by CON). Some examples of CON laws are:

  • In Alaska, a hospital needs permission to add a new hospital bed.
  • In Alaska, permission is required for a new provider to open a new hospital.
  • In Alaska, permission is required to purchase an MRI machine, CT scanner, or PET scanner.
  • In Alaska, permission is required to open an ambulatory surgery center.

RATIONALE FOR AND CONCEPTUAL INEFFECTIVENESS OF CON LAWS

While there are numerous claims made about CON laws, the laws’ primary goals include:

  • Ensuring an adequate supply of healthcare resources,
  • Protecting access in rural and underserved communities,
  • Promoting high-quality care,
  • Supporting charity care, and
  • Controlling cost.

Certificate-of-need laws were well intentioned when they were first introduced in states in the mid-1960s. However, the effectiveness of these laws should be measured by their outcomes. Even the best-intended laws might not lead to the best results.

Conceptually, the failure of CON laws might have been expected because CON laws grant a government-protected monopoly to incumbent providers. And basic economics and common sense tell us that government-protected monopolies tend to have negative consequences.

Moreover, CON laws do not have a public health justification. That is, certificate-of-need requirements have nothing to do with public health or safety. Separate state and federal laws govern who is allowed to practice medicine, what type of qualifications are required to do so, and what kind of medical procedures are or are not permitted.

CON laws are designed to restrict competition. And, in a manner unheard of in any other industry I know, in health care, existing hospitals and other medical providers have the opportunity to oppose the CON application of a would-be competitor, simply by claiming that there is no need for that additional medical service. This is akin to McDonalds needing permission from Burger King to open a restaurant in Alaska.

By requiring permission from regulators prior to any change, these state laws limit the ability of healthcare providers to offer cost-effective and innovative healthcare. Yes, CON laws prevent innovation that would otherwise result in less costly medical procedures, less evasive medical procedures, and safer medical procedures. 

One example of a less costly, less evasive, and safer medical procedure is the virtual colonoscopy, as opposed to the traditional optical colonoscopy. When a state requires a certificate of need for MRI machines, as does Alaska, it discourages providers from offering new procedures like virtual colonoscopies. This is because providers first have to get permission from state regulators, which not easy to obtain. The subsequent lack of adequate screening to detect cancer early probably contributes to unnecessary deaths.

EMPIRICAL EVIDENCE OF THE FAILURE OF CON LAWS

My colleagues and I started a project several years ago to analyze data to rigorously test whether each of the stated goals of CON was being achieved.

Specifically, we examined the following claims made by CON proponents:

  • CON laws increase access to medical care facilities.
  • CON improves access to diagnostic services, such as medical imaging services.
  • CON assures that more indigent care is provided.
  • The adoption of CON increases quality of medical care.

We found that CON does not deliver on these promises. There has not been increased patient access to medical care, and the quality of medical care has not been improved. In fact, CON laws have backfired. It turns out that states with CON laws have less patient access to medical care and lower quality of medical care.

CON REDUCES ACCESS TO MEDICAL CARE IN FACILITIES ACROSS THE STATE

One measure of access to medical care is the number of hospitals in a state. To control for the state population served, we measure hospitals per 100,000 population. More hospitals means shorter travel times to hospitals and greater access.

However, the data show that there are fewer hospitals per 100,000 capita in CON states versus in states without CON. In 2011, Alaska had about 25 hospitals. A comparable state without CON has 35 hospitals. So a state without CON has more than thirty percent more hospitals. And this estimate controls for confounding factors—such as age distribution, healthiness of the population, and percentage of the population on Medicaid and Medicare—in order to do an apples-to-apples comparison. This finding suggests that CON reduces access to medical care.

My research also uses another metric to determine the effect of CON on access to medical care—the number of hospital beds available in CON states versus states without CON. And here, we compare states with a CON law that regulates hospital beds versus states that do not regulate beds.

My findings unambiguously show that states without CON have more beds per patient. Why is this important? Well, it means that patients have more choices. They are less likely to be turned away from a hospital. And it might mean that there are hospitals closer to patients.

Alaska also has a CON law for ambulatory surgery centers. Comparing Alaska to statistically similar states without CON laws shows that without a CON, Alaska likely would have 25 centers instead of the 17 it had in 2011, when my data series ends.

CON proponents also say also say that CON laws increase provision of medical care and access to medical care in rural areas. But instead of providing more help for the rural population and better access for the entire state population, as CON proponents claim, CON in fact does the opposite. Alaska has fewer ambulatory surgery centers and fewer hospitals, thus fewer choices. Alaskans in both urban and rural areas have fewer choices because of CON. For example, states comparable to Alaska without CON have eight additional rural hospitals instead of the current roughly 17 hospitals as of 2011.

PATIENTS IN STATES WITH CON HAVE LESS ACCESS TO MEDICAL IMAGING AND OTHER SERVICES

The negative effect of CON on medical supply is not restricted to facilities. Medical inputs such as MRI, CT, and PET scans are also negatively affected. This is because there are CON laws that require permission to purchase such imaging equipment. For example, per year, Alaska residents have about 6,000 MRI scans. My estimates show that residents in states comparable to Alaska but without CON receive almost one-third more MRI scans—that is 8,000 MRI scans. States without CON also have about 30 percent more CT scans than states without CON.

Among the states with no CON laws, North Dakota, with its natural resource boom, might be the most comparable to Alaska. Comparing MRI and CT scan utilization among these states shows that North Dakota residents have more access to medical care than Alaska residents, as measured by utilization. In North Dakota, there are about 190 MRI scans per 1,000 Medicare beneficiaries per year, while in Alaska the number is 170. Similarly, in North Dakota, there are about 440 CT scans per 1,000 Medicare beneficiaries, while there are more than one-third fewer CT scans in Alaska—that is 300 CT scans per 1,000 Medicare beneficiaries.

QUALITY OF HOSPITAL CARE IS LOWER IN STATES WITH CON

In states without CON laws, hospitals have an incentive to compete to attract patients. Hospitals cannot compete as well on prices as most industries do because many of their patients are Medicare and Medicaid patients who can only be charged fixed amounts. But hospitals can compete on different margins, such as quality of service. So there is a strong incentive for hospitals in states without CON to compete for patients by providing better quality of medical services. This incentive does not exist to the same degree in states with CON laws, because in these states, hospitals are shielded by law from competition.

In contrast to this reasoning, some proponents of CON claim that it is good to have fewer hospital providers. They argue that when procedures are concentrated in a few hospitals, physicians have more experience performing operations because they have more volume, and this translates into higher quality of medical services.

To analyze which of these competing views is correct, I used data on the quality of medical services delivered by hospitals. These data come from a publicly available database maintained by the Centers for Medicare and Medicaid Service (CMS). The evidence from the analysis of this data shows CON does not improve quality of medical care.

Unfortunately, however, the numbers are much more alarming than this. The numbers show that CON laws actually reduce hospital quality. Comparing states with CON laws versus those with no CON laws shows that states with CON laws have lower quality of service, as measured by their hospital mortality rates and hospital readmission rates. Specifically, states with CON laws have

  • 0.5 percent more deaths for surgery patients with serious complications,
  • A 0.6 percentage point higher pneumonia mortality rate,
  • A 0.3 percentage point higher heart failure mortality rate, and
  • A 0.4 percentage point higher heart attack mortality rate.

This evidence shows that CON is harmful to patient health and survival in states that have CON laws on the books.

QUALITY OF INDIGENT CARE IS NOT BETTER IN STATES WITH CON

CON proponents sometimes make the claim that CON increases indigent care because successful applicants might commit themselves to increase their medical services to the indigent. However, the data fail to support such optimism. It turns out that hospitals in CON states have only as much indigent care—measured as uncompensated care—as hospitals in states without a CON law. Thus, CON does not lead to additional services for the poor.

CONCLUSION

If all states had CON laws, studying the effect of CON laws would be very difficult because we would not know what the world would look like without CON laws. Fortunately, 15 states do not have CON laws. This allows us to get a glimpse into the world without CON. And when comparing these two worlds, the data show that CON reduces access to medical care overall in the states with CON, in both rural areas and urban areas. CON states have fewer providers, such as hospitals and ambulatory surgery centers. CON results in fewer medical inputs, such as MRI and CT scans and the number of hospital beds. CON does not live up to the claim that it increases indigent care. Moreover, CON reduces quality of medical services.

The takeaway from these findings is that CON laws are bad for Alaska because they reduce the quality of medical care in Alaska, they reduce access for Alaskans, and they reduce opportunities to obtain medical services such as MRI and CT scans. Alaska would be better off if the Last Frontier would join the 15 states that do not have CON laws.