Comparing VHA with Fee-for-Service Cancer Care

September 2010, Vol 1, No 4

Boston, MA—The work of 3 Harvard Medical School researchers presented at June’s AcademyHealth Conference provided a snapshot of how cancer care outcomes compare between the Veterans Health Administration (VHA) and the fee-for-service Medicare community. In this case, the notion of “you get what you pay for” may not necessarily be true.

As the largest integrated delivery system in the world, the VHA provided free or low-cost healthcare to 5.5 million veterans in 2007. The papers presented at this session focused on male patients aged >65 years with cancer diagnosed or their first course of treatment received at the VHA between 2001 and 2004. This cohort was compared with a similar population identified from the Surveillance, Epidemiology, and End Results (SEER) Medicare database, whose regions cover about 25% of the United States. The researchers’ analytic strategy and sensitivity analysis accounted for observed differences seen between the 2 populations (eg, the higher percentage of blacks in the veteran population and a lower median income among veterans’ ZIP codes).

After coinvestigator Elizabeth Lamont, MD, MS, established statistically that the VHA’s data were accurate, Nancy Keating, MD, MPH, presented findings on the quality of cancer care in the VHA versus the private sector. Showing a graph depicting a higher rate of stage I colon cancer diagnosis at the VHA compared with SEER (29.6% and 24.5%, respectively), Dr Keating noted, “this is consistent with other studies that demonstrate the VHA has higher rates of early screening than the private sector.”

Dr Keating then compared the VHA and SEER’s propensity scores for standard treatments, including chemotherapy, surgery, radiation, and bisphosphonates for 6 cancer types: colon, small-cell lung, non–small-cell lung, prostate, non-Hodgkin lymphoma (NHL), and myeloma. The results showed the VHA and SEER exchanging the lead position of higher propensity scores for the different procedures. For example, the VHA’s tendency to perform curative surgery for non–small-cell lung cancer was 60.2% versus 65.6% for SEER. Conversely, the propensity score of CHOP (cyclophosphamide, hydroxydaunorubicin [doxorubicin], Oncovin [vincristine], and prednisone/prednisolone) treatments to diffuse large B-cell NHL at the VHA was 71.2% versus 60.7% for SEER. Taking a closer look at instances when the VHA did not appear to perform as well as the private sector, Dr Keating’s team adjusted for unobserved factors, such as the higher smoking rate, lower education level, and more severe comorbidities that veterans are known to have. “If we account for the severity of comorbid illness within the VA population, particularly with chronic obstructive pulmonary disease, we see the differences in treatment erased, or even reversed,” she said.

Is Aggressive End-of-Life Care Better?

Treatment of older patients with cancer at the end of life has become increasingly aggressive over time, despite the absence of evidence that aggressive care at the end of life is associated with better outcomes, according to a second paper presented by Dr Keating, entitled “Cancer Care at the End of Life in the VHA Versus the Private Sector.” Dr Keating’s team looked at 3 measures of end-of-life care for stage IV colorectal and lung cancer: chemotherapy within 14 days of death, admission to the intensive care unit in the last month of life, and >1 emergency department visit in the last month of life. The VHA had a lower propensity score for all 3 measures, but Dr Keating postulated that this may not necessarily be a bad thing. “These are indicators for care that are not potentially useful,” she said. “These lower rates are probably better, although there are some that might argue that.” Dr Keating surmised that the VHA is less aggressive with end-oflife care perhaps “due to the absence of financial incentives for end-of-life measures, or the integrated delivery system may be better structured to limit potentially futile medical care.”

These end-of-life care statistics do not appear to jeopardize survival at the VHA. Mary Beth Landrum, PhD, presented research indicating that survival for older men with colorectal or lung cancer was comparable or better at the VHA when weighed against the private sector.1 However, Dr Landrum acknowledged, “We don’t really understand the relationship between aggressive end-of-life care and patients’ preferences. Ongoing work is looking at that.”

When asked how their studies’ results may impact the larger world of cancer care, Dr Landrum suggested that “the VHA’s less- aggressive care for advanced cancer should be considered for healthcare reform.” Citing the VHA’s quick uptake of expensive medicinal options (eg, rituximab-CHOP chemotherapy) versus its slower implementation of complex equipment that requires highly trained workers (eg, 3-dimensional conformal radiotherapy and intensity-modulated radio therapy), Dr Keating proposed that “large capital outlay may not translate to better survivability. Comparative effectiveness can provide some good data to determine the better route.”

Reference

  1. Keating NL, Landrum MB, Lamont EB, et al. End-of-life care for older cancer patients in the Veterans Health Administration versus the private sector. Cancer. 2010;116(15):3732-3739.