乌鸦传媒

In 2009, New York initiated a policy to discourage breast cancer surgery at low-volume centers by denying payment for Medicaid patients. The intent was to improve outcomes across the state for lower-income patients by shifting their surgery to higher-volume centers. But did it work?

To find out, Ann Nattinger, MD, MPH, of Medical College of Wisconsin in Milwaukee, and colleagues, analyzed New York State Cancer Registry data linked with hospital data. The team identified 37,822 women who developed breast cancer in the years before and after the policy was enacted, and used a multivariable "" approach to compare the mortality of Medicaid patients with that for commercially insured patients unaffected by the policy.

"Leveraging the natural experiment afforded by the New York policy, we provide rigorous information about the effectiveness of a centralization policy intervention, for which randomized trials would be difficult to carry out," the researchers explained, reporting their results in the

In the following interview, Nattinger, who is associate provost and senior associate dean for research, discussed the study's results and implications.

Did the New York state policy work? What did you find in terms of mortality outcomes?

Nattinger: Yes, the policy did work. In general, there was a small temporal trend from pre-policy to post-policy, in that breast cancer patients treated after the policy was implemented had slightly reduced 5-year overall mortality compared with those treated before the policy. But the improvement in overall survival was substantially -- and significantly -- greater for the Medicaid patients targeted by the policy than for women with other insurances (P=0.018).

The mortality reduction applied to breast cancer-specific mortality, but not to other causes of death. For example, following policy implementation, the adjusted breast cancer mortality for Medicaid beneficiaries decreased from 6.6% to 4.5%, while that for all other women decreased from 3.9% to 3.8%.

Further supporting the effectiveness of the policy, the differential improvement post-policy in overall and breast cancer-specific mortality observed in New York Medicaid patients was not observed among Medicaid patients treated in nearby New Jersey during the same years.

Although it may seem intuitive, you mentioned that the reasons for better outcomes at higher-volume centers have not been clearly elucidated. What is the current thinking on this?

Nattinger: Efforts to identify the mechanisms by which high-volume hospitals achieve better outcomes of care have shown some association with more current standards of care, but there has been no clearly dominant quality factor in the higher-volume facilities. Breast cancer surgery has a very low postoperative mortality rate, so postop mortality is unlikely to be the key factor.

We believe that surgical case volume is likely a surrogate for enhanced multidisciplinary expertise -- for example, radiation or medical oncology expertise -- and/or other types of hospital resources -- for example, patient navigation or social workers.

It makes sense that hospitals with greater case volumes would be willing to invest in more resources to facilitate higher-quality care.

What about concerns that such a policy might hinder care for some low-income patients, such as those living in rural areas far from a high-volume facility?

Nattinger: That is a valid concern, and one that we considered. We did not find evidence that Medicaid patients experienced difficulty obtaining surgery for their breast cancer, as 86% of patients underwent surgery in New York state after the policy was in place, compared with 86.6% before the policy. In addition, the improvements in 5-year mortality among the Medicaid group speak for themselves.

That said, we are presently conducting analyses to determine whether Medicaid patients traveled further or waited longer to obtain their surgery after the policy was enacted – and especially whether the more rural patients experienced longer travel distances or greater surgical delays.

You mentioned that similar policies have been enacted in Europe. What can you tell us about these policies and their outcomes?

Nattinger: ESMO, the European Society for Medical Oncology, has published regarding early breast cancer. These explicitly recommend that treatment of breast cancer occur in specialized institutions that care for a minimum of 150 new early-stage breast cancer cases per year, and cite improved outcomes in overall and disease-free survival, quality of life, and functional status.

It is further recommended that treatment be carried out in units that provide a multidisciplinary team specialized in breast cancer, consisting of medical oncologists, breast surgeons, radiation oncologists, breast radiologists and pathologists, breast nurses, and a patient navigator. The facility is expected to have access to plastic surgeons, psychologists, physiotherapists, and geneticists.

These guidelines are more stringent than the New York state policy, which limited reimbursement for Medicaid to facilities performing at least 30 breast cancer surgeries annually over a rolling 3-year period. The more stringent policy recommended by ESMO is perhaps more feasible in countries with a greater level of national or regional input into the organization and financing of healthcare services.

It is also possible that a more stringent centralization policy in the U.S. might lead to even better outcomes, but we did not test this hypothesis.

Might this kind of policy be beneficial for patients with other types of cancer?

Nattinger: There has been discussion of centralization for other cancers, including esophageal, pancreatic, colorectal (especially rectal), ovarian, and lung cancers. In some cases, the rationale is the perioperative mortality or complication rate, but for some of these cancers, better outcomes may be associated with the enhanced resources and multidisciplinary care offered by larger facilities.

In all cases, though, the travel time and distance to a high-volume facility must be considered. For less common malignancies in particular, a much lower percentage of hospitals may meet surgical volume thresholds, and the access of patients to timely care could be adversely affected by this type of policy.

Further work should explore which cancer patients can benefit from centralization policies.

Read the study here and expert commentary about it here.