 |
|
|
Utilization of Care at End of Life: Dartmouth AtlasUtilization of health care services in Vermont at the end of life is markedly different from the rest of the country as a whole. In recent decades, Medicare has invested significantly in making non-hospital sites of care more available, i.e. home health, skilled nursing facilities (SNF), hospice, and long-term care and rehabilitation. The strategic thinking behind this effort was that if these non-hospital care sites were more available, they would be used as alternatives to more expensive acute care hospitalization. At the national level this has not proven to be the case as is highlighted by the authors of the 2008 Dartmouth Atlas. [1]
In contrast, at the end of their life, Vermonters are hospitalized less than in other parts of the country. Vermonters visit their physicians less and experience more continuity of care than elsewhere. They are more apt to avail themselves of home health and skilled nursing facilities, but less likely to be enrolled in hospice programs. Decades of federal strategy to decrease the use of acute care hospitalization by encouraging the use of alternative settings like nursing facilities and home health care seems to have taken root in Vermont, but failed in the country as a whole.
Interestingly, the average amount spent on a Vermonter in the last two years of life is the median amount among all states suggesting that the savings from reduced acute care hospitalization is shifted to alternative sites of care. It is likely that the conservative, non-intensive patient experience in Vermont is preferable to the experience in other parts of the country.[2] These initial revelations about care patterns in Vermont should give pause to local policy makers if they assume that so goes the country, goes Vermont.
General Findings
Low rate of hospitalization; High use of home health and skilled nursing facilities
In contrast to the nation as a whole, the Vermont experience is low use of acute care hospitalization and high use of home health and SNFs. The need for acute care hospitalization has been offset by the utilization of alternative sites of care. Interestingly, the average amount spent on a Vermonter in the last two years of life is the median amount among all states suggesting a cost shift to alternative sites of care.
More Dollars Spent on Outpatient Care; Lowest Spending on Test and X-Rays in the Country
As a consequence of receiving their care in less intensive settings, the distribution of dollars spent in Vermont differs from the overall national pattern. More money is spent on outpatient care and less on inpatient care than in the country as a whole. Also of note, expenditures on diagnostic testing, x-rays and other imaging modalities are the lowest in the country.
Number of Physician Visits is Low; Continuity of Care is High
Finally, Physician utilization is also conservative. Overall number of physician visits is very low; as is the use of specialists. Additionally, the number of physicians seen at the end of life is comparatively low suggesting the continuity of care is higher than in most other parts of the country.
Recommendations
There may be an opportunity to further reduce the use of intensive hospital services at the end of life in the state by increasing enrollment of patients in hospice programs. Policy makers should ensure that state policy decisions are based on Vermont-specific information rather than regional or national data as Vermont utilizations are dramatically different from national norms.
Utilization of Care at End of Life Outline [click on a section to jump ahead]
Data and Background < [back to top]
In 2008 the Vermont Quality Report presented Vermont specific information extracted from the 2008 Dartmouth Atlas of Health Care.[3] The 2008 Report focused on inpatient care at the end of life for elderly Vermonters with chronic illnesses. The following chapter builds on that work by expanding the scope of the Vermont specific information to include care at other sites of care, e.g. home health, outpatient care, hospice care and skilled nursing home care. The cohort of patients included in this report is approximately 12,000 Vermonters who were elderly, had a chronic illness, were Medicare beneficiaries, and died between 2003 and 2005. The researchers at The Dartmouth Institute who author the Atlas have access to the Medicare claims billed on behalf of these individuals.
Facility use [back to top]
According to the authors of the Dartmouth Atlas, nationally there is a positive correlation between the use of acute care hospitalization and the use of home health services and skilled nursing facilities. That is, more acute care hospitalization is associated with more home health and more skilled nursing facility use. Vermont specific data extracted from the Dartmouth Atlas portray a contrary story. The Vermont experience is low use of acute care hospitalization and high use of home health and SNFs.
The Dartmouth authors note that on a national scale, hospice enrollment, in contrast to home health and SNFs, does seem to act as an alternative to hospitalization utilization. Across the nation areas of high hospice enrollment coincide with areas of low hospital use. Again, in Vermont the opposite from the national trend is evident. In Vermont the pattern is low acute care hospitalization in spite of low hospice use.
Figure 1 offers a graphic comparison of Vermont utilization compared to other states. In the figure, all states are ranked against each other. The state with the lowest utilization is ranked as the 1st percentile, and the state with the highest utilization is ranked at the 100th percentile. In Figure 1, Vermont ranks in the 25th percentile in terms of utilization of hospital days, that is, 75% of states see higher hospital utilization. In contrast, Vermont ranks in the 75th percentile for skilled nursing facility days meaning that Vermonters are much more likely to receive care at the end of their lives in SNFs than in other states. Vermont ranks at the 95th percentile for home health visits. Only a few states offer more care through home health care services than in Vermont.
Also in Figure 1, Vermont ranks in the 10th percentile for hospice enrollment; that is, 90% of states have higher enrollment in hospice. These data suggest that, in Vermont, skilled nursing facilities and home health do appear to be used as alternatives to hospitalization. Additionally, there may be an opportunity to reduce acute care hospitalization even further by ensuring appropriate hospice enrollment.
Medicare reimbursement [back to top]
Reimbursement by Sector
Another notable Vermont-specific finding in the 2008 Dartmouth Atlas is that more dollars are spent in outpatient settings and less in inpatient settings when compared to the nation as a whole. In Figure 2a and the other figures in this section, all states are ranked against each other. The state with the lowest amount of dollars reimbursed (spent) by Medicare is ranked in the 1st percentile, and the state with the highest amount of dollars reimbursed is ranked at the 100th percentile. The reader’s attention is called to the comparatively high amount of services being paid for in the outpatient sector as compared to the inpatient sector in Vermont compared to the other 50 states.
Reimbursement by Site
Figures 2b and 2c display the amount of Medicare dollars spent per person during their last two years of life in Vermont compared to the other 50 states. Figure 2b shows the relative amount spent in Vermont compared to the other 50 states broken down into five sectors of care: home health services, outpatient services, SNF services, inpatient services and hospice services. The reader can easily see that per person spending was higher for home health services and outpatient services compared to other states, whereas, comparatively little was spent on hospice and inpatient care. Surprisingly, the amount spent on SNFs is comparatively low in spite of a comparatively high use of SNF days as displayed in Figure 1 above. A plausible explanation for this discrepancy between high utilization and low reimbursement is that Vermont SNFs receive a lower reimbursement rate per day than other states. Further investigation using other sources may be enlightening.
Total Reimbursement
Figure 2c shows the relative amount of total Medicare reimbursements in Vermont compared to other states. The total Medicare reimbursements displayed in Figure 2c is the sum of dollars spent in all the sectors displayed in Figure 2b. Figure 2c displays the total per person spending of Medicare dollars during the last two years of life. Comparing Vermont to other states, Vermont earns the rank of 26th of the 50 states. That is, the total amount spent per individual is greater in 25 states and less in 25 states when compared to Vermont. However, as described above, the source of spending differs in that more is spent in the outpatient sector and in the home than in other states and less is spent on hospice care. And utilization of SNFs is high, though reimbursement is near the median for all states.
The authors of the 2008 Dartmouth Atlas assert that the principal driver of Medicare dollars spent at the end of life is the volume of acute care hospitalizations, not the cost per hospitalization nor the cost of care at other alternative sites. Given the relatively low rate of hospitalization in Vermont, it is surprising that the overall amount of dollars spent on Vermonters at the end of life is not lower.
Apparently in Vermont the relatively high use of alternative care settings and preference for outpatient care offsets the cost savings of avoided inpatient care. Unfortunately, the Atlas does not provide us with any information about overall patient satisfaction with care. There is some suggestive information from Medicare that areas of higher use of intense medical care such as acute care hospitalization, intensive care inpatient settings and high physician utilization actually lead to poorer clinical outcomes, but there is little information about patient satisfaction. It is reasonable to assume that the conservative care patterns observed in Vermont would be preferable to consumers.[4]
Reimbursement on diagnostic testing, imaging, and equipment
The final display in this section on reimbursement focuses on additional aspects of outpatient care. Figure 2d displays the relatively low amounts spent at the end of life in Vermont for durable medical equipment (e.g. wheelchairs, etc, lab tests and imaging e.g. x-rays). Vermont ranks the lowest for all 50 states in terms of dollars spent on testing and imaging and x-rays and in the 17th percentile in terms of expenses for durable medical equipment (DME).
Physician utilization at the end of life [back to top]
Vermont has very conservative utilization of physician services at the end of life relative to other states. In Figure 3a, all states are ranked against each other. The state with the lowest amount of physician visits during the last 2 years of life is ranked as the 1st percentile, and the state with the highest amount of physician visits is ranked at the 100th percentile. Figure 3a illustrates Vermont’s rank for various aspects of physician utilization at the end of life during the last two years. Vermont ranks in the 13th percentile in terms of overall physician visits. There is also relatively low use of specialist physicians.
Figure 3b focuses on utilization during the final six months of life. Again the reader can see continued low utilization of physician resources. Of particular importance are Vermont’s rankings for the number of physicians seen and the percent seeing ten or more physicians at the end of life. These two statistics suggest a higher degree of continuity of care than is typical nationally. That is, during the last six months of life, Vermonters receive the majority of their care from the same physician.
Conclusions
Vermonters are hospitalized less than in other parts of the country. Vermonters visit their physicians less and experience more continuity of care than elsewhere. They are more apt to avail themselves of home health and skilled nursing facilities, but less likely to be enrolled in hospice programs.
There may be an opportunity to further reduce the use of intensive hospital services at the end of life in the state by increasing enrollment of patients in hospice programs as well as patient and physician education about care options at or near end of life. Policy makers should ensure that state policy decisions are based on Vermont-specific information rather than regional or national data as Vermont utilizations are dramatically different from national norms.
[2] Wright AA, Zhang B, Ray A, Mack JW, Trice E, Balboni T, Mitchell SL, Jackson VA, Block SD, Maciejewski PK, Prigerson HG. Associations between end-of-life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment. Journal of the American Medical Association. 2008 Oct 8; 300(14):1665-73.
|
|
|
|
