Unformatted text preview: Chapter 12 References Available at: . (Accessed: February
Bohra-Mishra, P., M. Oppenheimer, R. Cai, S. Feng, and R. Licker. 2017. Climate variability and migration
in the Philippines. Population and Environment 38(3):286-308. doi:10.1007/s11111-016-0263-x.
Available at: . (Accessed:
February 26, 2018).
Boland, C., R. DeKleine, A. Moorthy, G. Keoleian, H.C. Kim, E. Lee, and T.J. Wallington. 2014. A Life Cycle
Assessment of Natural Fiber Reinforced Composites in Automotive Applications. SAE Technical Paper
Boland. S. and Unnasch. S. 2014. Carbon Intensity of Marginal Petroleum and Corn Ethanol Fuels. Life
Cycle Associates Report LCA.6075.83.2014, Prepared for Renewable Fuels Association. Available at:
. (Accessed: February 26, 2018).
Boothe, V.L. and D.G. Shendell. 2008. Potential Health Effects Associated with Residential Proximity to
Freeways and Primary Roads: Review of Scientific Literature, 1999–2006. Journal of Environmental
Boothe, VL., T.K. Boehmer, A.M. Wendel, and F.Y. Yip. 2014. Residential Traffic Exposure and Childhood
Leukemia: A Systematic Review and Meta-analysis. American Journal of Preventive Medicine
Borasin, S., S. Foster, K. Jobarteh, N. Link, J. Miranda, E. Pomeranse, J. Rabke-Verani, D. Reyes, J. Selber,
S. Sodha, and P. Somaia. 2002. Oil: A Life Cycle Analysis of its Health and Environmental Impacts.
[Epstein, P.R. and J. Selber (Eds.)]. Prepared by: Harvard University, Center for Health and the Global
Environment: Cambridge, MA. Available at:
. (Accessed: February 26, 2018).
Bouchama, A., M. Dehbi, G. Mohamed, F. Matthies, M. Shoukri, and B. Menne. 2007. Prognostic Factors
in Heat Wave Related Deaths: A Meta-analysis. Archives of Internal Medicine 167:2170–2176.
Boumans, R.J.M., D.L. Phillips, W. Victery, and T.D. Fontaine. 2014. Developing a Model for Effects of
Climate Change on Human Health and Health-environment Interactions: Heat Stress in Austin,
Texas. Urban Climate 8:78–99. doi:10.1016/j.uclim.2014.03.001.
Boustani, A., S. Sahni, T. Gutowski, and S. Graves. 2010. Tire Remanufacturing and Energy Savings.
MITEI-1-h-2010. Prepared by the Environmentally Benign Manufacturing Laboratory, Sloan School of
Management, Massachusetts Institute of Technology. Available at:
. (Accessed: February 26, 2018).
Bowles, A.E. 1995. Responses of Wildlife to Noise or Wildlife and Recreationists: Coexistence through
Management and Research. Washington, D.C. [Knight, R.L. and K.J. Gutzwiller (Eds.)]. Island Press:
Washington. pp. 109–156. 12-6 Policies to address the impact of coding
differences Impact of coding differences on payment to
MA plans A series of congressional mandates have required CMS to
reduce MA risk scores as a way of addressing the impact
of coding differences. Because of the mandates, CMS
reduced MA risk scores by 3.41 percent in each year
from 2010 through 2013. Starting in 2014, the mandates
specified a minimum reduction of about 4.9 percent,
which increased gradually to about 5.9 percent in 2018,
where it will remain until CMS estimates a risk adjustment
model using MA cost and use data. CMS reduced MA risk
scores by the minimum amount required by law for 2014
through 2018, although larger reductions would have been
allowed. To assess the overall impact of coding differences
on payments to MA plans for a given year, we built
retrospective cohorts of beneficiaries enrolled in either
FFS or MA for all of 2017. We tracked each beneficiary
backward for as long as they were continuously enrolled
in the same program (FFS or MA) or as far back as 2007.
Our analysis calculates differences in risk score growth by
comparing FFS and MA cohorts with the same years of
enrollment (e.g., 2007 through 2017, 2008 through 2017,
etc.), adjusting for differences in age and sex. CMS has taken an additional step to help control the
increased coding intensity in MA by phasing in a new
CMS–HCC model that removes some diagnoses suspected
of being more aggressively coded by MA plans (e.g.,
lower severity kidney disease and polyneuropathy). Our
analysis suggests that the new CMS–HCC model makes
MA risk scores more similar to FFS scores by reducing
them by 2 percent to 2.5 percent relative to the old model.
The new model was phased in during 2014 and 2015, and
MA payments were based entirely on the new model in
Before 2017, the HCC model accounted for dual
enrollment in Medicare and Medicaid with a set of
variables that increased payment for Medicaid enrollees.
This approach treated MA enrollees with partial Medicaid
enrollment and MA enrollees with full Medicaid
enrollment as a single group; however, enrollees with
full Medicaid benefits have Medicare spending that is
significantly higher than enrollees with partial Medicaid
benefits. As a result, risk scores under the old model were
systematically too low for full dual enrollees and too high
for partial dual enrollees.8 In addition to the inaccuracy
in individual risk scores, partial dual enrollees make
up a larger share of dual enrollees in MA than in FFS
Medicare, causing the overall risk scores for MA enrollees
who are enrolled in Medicaid to be inflated under the old
model. For 2017, CMS began differentiating between
MA enrollees with full Medicaid and partial Medicaid
enrollment using separate models that more accurately
determined risk scores for partial benefit and full benefit
Medicaid enrollees.9 We found that the 2017 model
reduced MA risk scores by almost 1 percent by accurately
determining risk scores for subgroups of beneficiaries,
particularly partial dual and full dual enrollees. Figure 13-4 (p. 364) shows the impact of differences
in coding intensity on MA risk scores relative to FFS
for payment years 2013 through 2017 and the amount
by which CMS reduced MA risk scores for the coding
intensity adjustment in each year. The difference between
the lines shows the portion of coding intensity impact that
was not accounted for by payment policies and resulted in
the additional Medicare spending for beneficiaries enrolled
in MA relative to the amount Medicare would have
spent if the same beneficiaries had been enrolled in FFS
Medicare. Three different versions of the CMS–HCC risk
model were used for payment over this period. A blend of
two of these model versions was used for payment in 2014
The impact of coding intensity on MA risk scores changed
over this period, largely because of three factors: changes
to the risk score model used for payment, changes in MA
risk score growth relative to FFS risk score growth, and
the addition of encounter data as a source of diagnostic
Changes in the risk model Our analysis has found that newer versions of the CMS–
HCC model have been less susceptible to diagnostic
coding differences between MA and FFS. Figure 13-4 (p.
364) shows that the version phased in over 2014 to 2016,
removing specific diagnoses with large differences in
MA and FFS coding rates, reduced the impact of coding
differences by 2 percent to 2.5 percent. The version
introduced in 2017, adding separate aged/disabled and
Medicaid enrollment status segments, reduced the impact
of coding differences by almost 1 percent.
Relative risk score growth rates Between 2013 and 2015, our analysis shows that MA risk
score growth outpaced FFS risk score growth by 1 percent Report to the Congress: Medicare Payment Policy | March 2019 363 Chapter 12 References Bradford, M.A., Wieder, W.R., Bonan, G.B, Fierer, N., Raymond, P.A., and Crowther, T.W. 2016.
Managing uncertainty in soil carbon feedbacks to climate change. Nature Climate Change. 6, 751758. doi:10.1038/nclimate3071. (Accessed: Jan 26, 2017).
Brandt, A.R., G.A. Heath, E.A. Kort, F. O’Sullivan, G. Pétron, S.M. Jordaan, P. Tans, J. Wilcox, A.M.
Gopstein, D. Arent, S. Wofsy, N.J. Brown, R. Bradley, G.D. Stucky, D. Eardley, and R. Harriss. 2014.
Methane Leaks from North American Natural Gas Systems. Science 343(6172):733–735.
doi:10.1126/science.1247045. Available at:
. (Accessed: June 17, 2016).
Brandt, A.R., Y. Sun, S. Bharadwaj, D. Livingston, E. Tan, and D. Gordon. 2015. Energy Return on
Investment (EROI) for forty global oilfields using a detailed engineering-based model of oil
production. PloS one 10(12):e0144141. doi:10.1371/journal.pone.0144141. Available at:
. (Accessed: February 26,
Brodrick, C., T.E. Lipman, M. Farshchi, N.P. Lutsey, H.A. Dwyer, D. Sperling, S.W. Gouse III, D.B. Harris,
and F.G. King. 2002. Evaluation of Fuel Cell Auxiliary Power Units for Heavy-Duty Diesel Trucks.
Transportation Research Part D 7:303–315. Available at:
EN=88995268. (Accessed: February 26, 2018).
Brown, C.L., S.E. Reed, M.S. Dietz, and K.M. Fristrup. 2013. Detection and Classification of Motor Vehicle
Noise in a Forested Landscape. Environmental Management 52(5):1262–1270. doi:10.1007/s00267013-0123-8.
Brown, S., S. Hanson, and R.J. Nicholls. 2014. Implications of Sea-level Rise and Extreme Events around
Europe: A Review of Coastal Energy Infrastructure. Climatic Change 122(1-2):81–95.
Brzoska, M. and C. Frohlich. 2015. Climate change, migration and violent conflict: vulnerabilities,
pathways and adaptation strategies. Migration and Development 5(2):190-210.
Brzoska and Frohlich 2015. citing Hsiang, S.M., M. Burke, and E. Michael. 2013. Quantifying the
influence of climate on human conflict. Science 341:6151. doi:10.1126/science.1235367.
Bulka, C., L.J. Nastoupil, W. McClellan, A. Ambinder, A. Phillips, K. Ward, A.R. Bayakly, J.M. Switchenko, L.
Waller, and C.R. Flowers. 2013. Residence Proximity to Benzene Release Sites is Associated with
Increased Incidence of Non‐Hodgkin Lymphoma. Cancer 119(18):3309–3317.
doi:10.1002/cncr.28083. Available at:
D7D76A362.f02t02. (Accessed: February 26, 2018).
Buhaug, H., T.A. Benjaminsen, E. Sjaastad, and O.M. Theisen. 2015. Climate variability, food production
shocks, and violent conflict in Sub-Saharan Africa. Environmental Research Letters 10(12).
doi:10.1088/1748-9326/10/12/125015. Available at:
. (Accessed: February 20, 2018).
12-7 F IFIGURE
GURE Title here....
Impact of coding intensity on MA risk scores was
larger than coding adjustment, 2013–2017 11-XX
11 Impact as a percent above FFS 10
5 Portion of coding intensity
impact not accounted for
by payment policies 4 Impact on MA risk scores used for payment
Coding adjustment applied to MA risk scores 3 CMS–HCC version 2013–2015 2 CMS–HCC version 2014–2016
CMS–HCC version 2017 1
Note: 2014 2015 2016 2017 MA (Medicare Advantage), FFS (fee-for-service), CMS–HCC (CMS–hierarchical condition category). All estimates account for any differences in age and sex
between MA and FFS populations. A blend of two model versions was used for payment in 2014 and 2015. Source: MedPAC analysis of CMS enrollment and risk score files. to 1.5 percent per year, increasing the overall impact of
2016 and 93 percent of MA enrollees in 2017.11 However,
coding intensity on MA risk scores in each year. Between
for enrollees with different encounter-based and RAPSNote: and
scores continued to increase at
based risk scores, the RAPS score tends to be higher.
Overall, encounter-based risk scores were about 2 percent
grew faster than prior years and roughly matched the
lower than RAPS-based risk scores in both 2016 and 2017,
MA risk score growth rate.10 Risk score growth between
despite a decrease in the overall difference by about a half
percent in 2017. The phase-in of encounter-based risk
2015 and 2016 was affected by the transition from ICD–9
scores (see Figure 13-3, p. 362) reduced the overall impact
2017, in the datasheet.
is in the
of coding intensity by about 0.2 percent in 2016 and by
we again found similar growth rates for MA and FFS
• I deleted the years from the x-axis and put in my own.
about 0.4 percent in 2017. For 2018, CMS decreased the
risk scores, with MA risk score growth outpacing FFS
marks andofaxis linesusebecause
they kept risk
to 15 percent,
• The dashed line looked ok here, so I didn’t hand draw it.
period may also have affected the FFS risk score growth
• I can’t delete the legend, so I’ll just have to crop it out in InDesign.
impact of ifMA
• Use direct selection tool to select items for modification.
use the intensity
black selection tool, they w
information We found that MA risk scores for 2017 were about 7
percent higher than for a comparable FFS population. The
• Use paragraph styles (and object styles) to format.
Starting in 2016, CMS blended risk scores based on
decline from our 2016 estimate of 8 percent is the net of
encounter data with risk scores based on RAPS data. We
faster MA risk score growth (0.3 percent), implementing
found that encounter-based and RAPS-based risk scores
a new version of the risk adjustment model (–0.8 percent),
were the same for about 92 percent of MA enrollees in
and increasing the use of encounter data for risk scores 364 The Medicare Advantage program: Status report ...
View Full Document
- Fall '19
- Climate Change, 2010s, Years in the future