Medicare Advantage. Payment Year 2020 Advance Notice Part 1. Review of Key Proposals with Expert Insight

Transcription

1 Medicare Advantage Payment Year 2020 Advance Ntice Part 1 Review f Key Prpsals with Expert Insight January 2019 PULSE8 is privileged t bring yu a summary f key Medicare Advantage prgram changes fr Payment This year s changes cntinue the implementatin f imprvements t the risk adjustment prgram as required by the 21 st Century Cures Act f Bullet-Pint Key Changes Prpsed in Part 1 f the 2020 Advance Ntice Payment Cnditin Cunt (PCC) Methdlgy Applicatin f an HCC-cunt-per-member* gruping t risk scre cefficient schedule designed t increase risk scres f members with multiple cnditins. PCCs help accunt fr the added csts arising frm the cmplexity f their treatment. *Cunt f hierarchy-impsed HCCs Only fr use within the EDPS & RAPS inpatient submissins data risk scre calculatin. Alternative CMS-HCC Mdel with PCC Methdlgy Increase frm 83 t 86 Cnditin Cdes CMS internal analysis and research fr PCC implementatin led t the cnsideratin f 10 additinal chrnic HCCs, CMS cnclusin is a prpsal t add 3 new HCCs: HCC 51 Dementia with Cmplicatins HCC 52 Dementia withut Cmplicatins HCC 159 Pressure Ulcer f Skin with Partial Thickness Skin Lss RAPS / EDPS Risk Scre Calculatin Risk scre calculatin: 50% RAPS submissins data, plus 50% EDPS & RAPS inpatient submissins data. RAPS submissins data risk scre calculated using the 2017, Versin 22 CMS-HCC mdel EDPS + RAPS Inpatient data risk scre calculated using the 2020, Versin 23 CMS-HCC Mdel with PCC methdlgy applied. Expanded Review f the Prpsed Changes Payment Cnditin Cunt (PCC) Methdlgy The PCC cncept was intrduced within Part 1 f the PY2019 Advance Ntice released in December The PY2019 Final Ntice (released in April f 2018) prpsed a pstpnement f the PCC mdel t PY2020, with CMS nting their internal vetting analyses wuld cntinue thrughut Finally, in December Page 1 f 5 Curtesy f Pulse8 Yur ONLY Partner fr Risk Adjustment & Quality Analytics Fr mre infrmatin, please visit r call (410)

2 2018, the Part 1 PY2020 Advance Ntice was released, cnfirming the adptin f PCC methdlgy with ne last twist: CMS will adpt either the PY2019 PCC mdel r an Alternative PCC Mdel, detailed in the ntice and bullet-pinted abve. Please nte this critical distinctin regarding cunting HCCs: A member s cunt f HCCs is calculated after the impsitin f the hierarchy rules. Fr example, a member with bth HCC 18 and HCC 19 wuld be gruped as 1 Payment HCCs hierarchy rules state that HCC 18 trumps 19 fr member risk scre determinatin. Pulse8 Insight: It s imprtant t nte that PCC methdlgy is nt a supply f additinal risk scre it s a shifting f the available risk scre frm specific HCCs t an HCC-cunt-by-member risk adjustment functin. In Figure 1 belw, the additive risk scres fr bth prpsed PCC mdels are listed. On the ther hand, Figure 2 reveals the reductin in base risk scre fr three prevalent HCCs: available risk scre is shifting frm Figure 2 (individual HCCs) int Figure 1 (HCC-cunt-by-member). The drp in base HCC risk scres will reduce the average financial impact f individual HCC gap clsures. T maintain and imprve ROI, health plans will need t build gap clsure campaign strategies and analytics t target sub-ppulatins as well as individuals. Figure 1: PCC Risk Scre Cefficients Payment HCC Cunt PY2019 Hld-Over PCC Mdel fr PY2020 PY2020 Alternative PCC Mdel 1 Payment HCCs Payment HCCs Payment HCCs Payment HCCs Payment HCCs Payment HCCs Payment HCCs Payment HCCs Payment HCCs r mre Payment HCCs Cmmunity Nn-Dual Aged Cefficients Figure 2: Example Base HCC Risk Scre Cefficient Reductin HCC Cde and Descriptin Current PY2019 "N Cunt" Mdel Page 2 f 5 PY2019 Hld-Over PCC Mdel fr PY2020 Curtesy f Pulse8 Yur ONLY Partner fr Risk Adjustment & Quality Analytics Fr mre infrmatin, please visit r call (410) PY2020 Alternative PCC Mdel HCC 18 - Diabetes with Cmplicatins HCC 19 - Diabetes withut Cmplicatins HCC 59 - Majr Depressin Cmmunity Nn-Dual Aged Cefficients

3 Pulse8 Insight: Figure 3, belw, prvides estimates t average PMPM risk adjustment payment changes fr bth PY2020 PCC mdels. The estimates in Figure 3 are 50% f the riginal results in accrdance with the PY2020 RAPS 50% plus EDPS 50% risk scre methd. A default average benchmark payment rate f $750/mnth was applied t annualized member mnths t prject the dllar amunts. The actual PY2020 PMPM impact fr a health plan will be the prduct f afre mentined variables (EDPS submissin cmpleteness, cunty benchmark rate, member mnths), a plan s level f disease burden and the six-sided Cmmunity Risk Mdel distributin f the plan s membership. The estimates in Figure 3 are median calculatins between a member ppulatin with high disease burden and higher six-sided mdel distributin and a lw disease burden member ppulatin made up f primarily Cmmunity Nn-Dual Aged beneficiaries. We fund that the additive risk payment impact f the HCC Cunt functin was the nearly equal fr bth PCC mdels abut $3.00 PMPM. The reductin in risk payment due t the lwering f Individual HCC risk scres is rughly twice as impactful (-$4.40 t -$2.10) in the PY2019 Hld-Over mdel than in the PY2020 Alternative mdel. This ffset f $2.30 PMPM is primarily due t an estimated increase f $7.50 PMPM frm the Dementia and Skin Ulcer HCC additins t the Alternative PCC mdel. Lastly, the Demgraphic risk scre variables shw the greatest delta between the tw mdels: PY2019 Hld-Over average f $3.10 t the PY2020 Alternative mdel s -$0.25. Figure 3: PMPM Change Estimates by Risk Payment Cmpnent Risk Payment Cmpnents PY2019 Hld-Over PCC Mdel PY2020 Alternative PCC Mdel HCC Cunt $ 3.00 $ 3.00 Individual HCC $ (4.40) $ (2.10) Net HCC-Derived Risk Payment $ (1.40) $ 0.90 Demgraphic $ 3.10 $ (0.25) Net Ttal Risk Payment $ 1.70 $ 0.65 Payment Year 2020 Alternative HCC Mdel The PY2020 Advance Ntice describes hw CMS research f the PY2019 PCC mdel fund that sme beneficiaries with multiple chrnic cnditins had under-predicted csts. The ntice ges n t say that stakehlder-submitted cmments f the PY2019 PCC mdel pined that it did nt meet the Cures Act requirement t imprve predictive csts fr high needs beneficiaries with multiple chrnic cnditins. The cnfluence f CMS wn research results and stake-hlder cmments was enugh t prmpt the agency t test the impact f adding HCCs fr chrnic cnditins nt included in the current Versin 23 Cmmunity CMS-HCC mdel. Page 3 f 5 Curtesy f Pulse8 Yur ONLY Partner fr Risk Adjustment & Quality Analytics Fr mre infrmatin, please visit r call (410)

4 The Advance Ntice describes, in detail, the prcess used t analyze nn-payment HCCs that wuld merit further analysis fr eventual inclusin. T summarize, the fur basic criteria applied were: 1. Chrnicity f the HCC Determined by applicatin f the empirical data analysis methdlgy detailed in the PY2019 Part 1 Advance Ntice and in the CMS December 2018 publicatin, Reprt t Cngress. 2. Clinical Meaningfulness f the HCC Per the Advance Ntice, the ICDs mapped t an HCC, shuld relate t a reasnably wellspecified disease r medical cnditin 3. Medical Expenditure Predictability f the HCC Again, quting frm the Advance Ntice, The cnditin categry shuld prduce a reasnable and statistically significant estimate f medical expenditures fr Medicare Part A and B benefits. This criterin als states that HCCs whse csts are cvered by related r c-mrbid payment HCCs shuld nt be added. 4. Definitively Diagnsable ICDs f the HCC Only HCCs cmprised f ICDs with minimal clinical discretin shuld be added t the payment mdel. The final filtering metrics applied t nn-payment HCCs fr cnsideratin were 1) sample size greater than 30,000 beneficiaries, 2) annual average cst abve the mean, and 3) predictive rati f less than 0.9. The net result f the fur criteria and three metric filters yielded ten ptential nn-payment HCCs fr inclusin in risk adjustment payment: 1. HCC41 Disrders f the Vertebrae and Spinal Discs 2. HCC42 Ostearthritis f Hip r Knee 3. HCC51 Dementia with Cmplicatins 4. HCC52 Dementia withut Cmplicatin 5. HCC89 Crnary Athersclersis/Other Chrnic Ischemic Heart Disease 6. HCC95 Hypertensin 7. HCC102 Cerebrvascular Athersclersis, Aneurysm, and Other Disease 8. HCC113 Asthma 9. HCC140 Unspecified Renal Failure 10. HCC159 Pressure Ulcer f Skin with Partial Thickness Skin Lss The PY2020 Part 1 Advance Ntice brke dwn the final adjudicatin f the ten cnsidered HCCs as fllws: HCCs 89, 102, 113, and 140 were clinically t vague and/r made-up f ICD cdes with t high a level f diversity. HCCs 41, 42, and 95 are s prevalent within the Medicare ppulatin that their inclusin wuld nt increase the predictive accuracy f the average medical expenditure. Page 4 f 5 Curtesy f Pulse8 Yur ONLY Partner fr Risk Adjustment & Quality Analytics Fr mre infrmatin, please visit r call (410)

5 HCCs 51, 52, and 159 were fund t have met the cnditins fr inclusin, in that they are well-specified, predict medical expenditures, are definitively diagnsed and can indicate significant disease burden. Pulse8 Insight: Pulse8 endrses the prpsed changes, as we d any imprvements t the CMS-HCC Risk Adjustment mdel, and is prud t share with ur custmers the fllwing prjectins fr their impact: Dementia Pulse8 s analyses estimate that HCCs 51 Dementia with Cmplicatins and 52 Dementia withut Cmplicatins cmbine t between 3.5% and 5.5% f plan-established members. The average risk scre fr members with either r bth dementia HCCs (they carry the same base risk scre cefficient) is estimated t be between and based n the six-sided Cmmunity Risk Mdel distributin f the plan. The average estimated, annualized increase t risk adjustment payment fr the cmbined dementia HCCs is $6.75 PMPM as calculated using the parameters listed in Figure 3 abve. Pressure Ulcer / Partial Thickness Skin Lss Pulse8 s analyses estimate that HCC 159 Pressure Ulcer f Skin with Partial Thickness Skin Lss is diagnsed fr abut 0.2% f plan-established members after the CMS-HCC hierarchy is applied. The average risk scre fr members with HCC 159 is estimated t be between and as per the six-sided Cmmunity Risk Mdel distributin f the plan. The average estimated, annualized increase t risk adjustment payment fr the cmbined dementia HCCs is $0.75 PMPM as calculated using the parameters listed in Figure 3 abve. With each passing year, CMS cntinues t advance the predictive accuracy f its mdel. Hwever, these imprvements usually infuse added cmplexity. The latest prpsed changes are n exceptin. Plans lking t ptimize their perfrmance, under such a cmplex and nuanced mdel, need t dedicate themselves t rigrus data analytics, whether internally r with their vendr partner. Page 5 f 5 Curtesy f Pulse8 Yur ONLY Partner fr Risk Adjustment & Quality Analytics Fr mre infrmatin, please visit r call (410)