'Low-value Care' More Prevalent in Hospital-based Primary Care Practices

The use of low-value care was common among patients who saw someone other than their primary care physician at a hospital-based primary care practice. Practice location, rather than practice ownership, is the driving factor behind the disparity.

A study in this week's issue of  JAMA Internal Medicine finds that hospital-based primary care practices are more likely to make referrals to specialists and order expensive imaging and other unneeded tests for patients with common conditions than do their colleagues in community-based practices.

Practice location, rather than practice ownership, is the driving factor behind the disparity.

Low-value Care Misunderstood By Patients

The use of low-value care was even more common among patients who saw someone other than their primary care physician at a hospital-based primary care practice.

Study senior author Bruce Landon, MD, a general internist and Harvard Medical School professor of healthcare policy and of medicine at Beth Israel Deaconess Medical Center, spoke with HealthLeaders about the study. The following is a lightly edited transcript.

HLM: Do you believe your findings represent an accurate reflection of what is occurring nationally in the healthcare delivery system?

Landon: These findings might not apply to a particular physician or visit. But we can say that across a very large nationally representative sample of visits, on average, primary care visits to hospital-based vs. community-based visits for these three common conditions resulted in more over use of these lower-value services at hospital-located clinics than in community-based practices.

Unwise Medical Choices Stubbornly Defy Eradication

There are some relatively straightforward reasons for those findings. For example, there is much more readily available imaging equipment in hospitals. In the hospital-based practice where I work there is a CT Scan and an MRI one floor down. If I want to order just a plain film, it's a click of a button.

More commonly, in community-based practices, the patient will have to get in his car and go to an imaging center or a hospital and that is one more hurdle. I'm in a hospital-based practice that has seven floors of specialists. It's probably just easier and more readily available for me to involve specialist colleagues.

Of course, a lesson from this is that clinicians and managers and those who are practicing in hospital-based practices need to be at the very least aware of these tendencies, whether it's through data, feedback, measurements, education, etc., to try to make it so that physicians actually think twice before using these lower-value services that are easily accessible to them.

HLM: Is there any evidence, anecdotal or otherwise, that hospitals are pressuring physicians to do these low-value tests?

Landon: We compared those practices that were owned by a health system or hospital to those that were physician-owned or community-based practices. For the most part the community based practices looked alike.

The only subtle difference was the hospital-owned community-based practices used referrals more than physician-owned practices.

So there was a slight difference in the use in referrals and one could postulate that once you are owned by a larger system, they make it easier for you to get patients in to specialists and make that referral process happen. But, they did not use more low-value CTs, MRIs, plain X-rays and the like.

HLM: Can steps can be taken to curb the use of these low-value services.

Landon: One of the issues with all low-value services is that what's low value in some situations is high value in other situations. Everything has to be taken into account through a clinical lens. If I were trying to reduce low-value services at hospital-based practices, there are a few things I would do.

The first would be measurement. You can't manage what you can't measure. Identifying the rate of use, and which physicians have a higher tendency of using low-value services and feedback and educational meetings, those sorts of things are good steps.

HLM: What other trends did your data identify?

Landon: Continuity of care seemed to be an important issue. When a hospital-based clinic patient was seeing his own primary care physician, there did not tend to be overuse.

When patients were seeing a covering physician, someone's colleague who presumably was less familiar with the patient, there tended to be more overuse, more referrals, more CTs and more MRIs.

I'll speculate on why that might be the case with hospital-based practices but not in community-based practices. Community practice locations tend to be smaller and have more fulltime physicians.  And those physicians, because they are a smaller, tighter-knit group of people, also tend to know each other's patients better.

Hospital-based practices tend to be larger. Academic hospitals in particular tend to have a lot of clinicians relative to the number of patients because there are also researchers and teachers.

The familiarity with your colleagues' patients might not be as high at a hospital-based practice. Therefore, one of the other things we should try to focus on would be to take steps to improve the visit-based continuity within those practices.<!--pagebreak-->

HLM: Can you put a dollar figure on the global cost of delivering this low-value care?

Landon: No. We didn't include an estimate in the paper. Frankly, the specialist referrals you could even postulate would cost more because that could lead to downstream procedures and imaging. That is one of the prime reasons against trying to not refer patients with routine back pain.

HLM: Could this trend worsen as hospitals and health systems acquire more physician practices?

Landon: You should not interpret our findings as being completely against the concept of hospitals buying practices, even though there are other reasons why that might not be a good thing.

But, this evidence does not suggest that there are going to be huge spikes in utilization of low-value services when hospitals buy physician practices.

HLM: What would you like to see done with your findings?

Landon: This is a study that supports changing the site-of-care differentials for hospital-based practices. The people whose attention I'd like to get are the practice managers and clinicians in hospital-based practices to make them more aware of this tendency so they can take internal management steps to bring data to bear and educate and use other methods to impact their own physicians' use of low-value services.