Value-Based Purchasing Comes to Home Care

On Monday, July 6, CMS issued the pre-publication copy of the proposed FY 2016 Home Health PPS Rule (the “Proposed Rule”).  The Proposed Rule will impose further cuts to home health reimbursement.  This is done in part through a continued application of the Affordable Care Act rebasing, which CMS has determined is appropriate, after concluding last year’s rebasing cut had no impact on home health care access.  The Proposed Rule also includes a nominal case mix or “coding creep” adjustment and another reweighting of case mix.  CMS projects that the impact of the changes to reimbursement contained in the Proposed Rule will result in a 1.8 percent reduction in home health payments in CY 2016.  This translates into a $350 million cut to home health reimbursement.  These cuts will continue to place heavy burdens on providers, as the already narrow margins in home health continue to shrink at a time when other proposals will result in increased costs to providers.

Although the negative rate adjustment is of great importance to providers, there is a bigger proposed change unveiled in the Proposed Rule.  This change is the Home Health Value-Based Purchasing Program (“HHVBPP”).  HHVBPP will bring value-based purchasing directly to home health.  Home health has been indirectly impacted by VBP because VBP initiatives for hospitals and physicians has required providers to focus on showing hospitals and physicians how the home health providers can assist them with achieving the quality goals.  This “upstream” impact has made home health providers more aware of their quality and outcomes data and the need to achieve certain quality goals as part of obtaining referrals from hospitals, SNFs and ACOs.

With the HHVBPP, the need to achieve specified quality metrics will no longer be driven by the indirect pressure of referral sources but will directly impact home health providers’ reimbursement.  This proposal does not come as a complete surprise because in last year’s HHPPS rule, CMS sought input regarding the “elements of the model, size of the payment incentives and percentage of payments that would need to be placed at risk in order to spur home health agencies (“HHAs”) to make the necessary investments to improve the quality of care for Medicare beneficiaries; the timing of the payment adjustments; and how performance payments should be distributed.”  CMS received a number of comments but notes that it did not receive any strong counter arguments or alternative proposals to HHVBPP.  It will, therefore, move forward with the VBP model in 2016.

CMS proposes to launch a five-year pilot program in nine states.  The program will begin on January 1, 2016 and continue through December 31, 2022.  Providers that are not part of this program initially should not simply ignore it.  Providers that are not in the pilot should use this as an opportunity to learn and prepare because it is extremely likely, given recent statements by CMS, that HHVBPP will become a permanent part of HHPPS.  And, as we will see below, this proposed VBP program is somewhat more onerous to poor performing providers than other VBP programs CMS has implemented.

Which states will be in the demonstration?

The first question most providers will ask is, “am I going to be in this demonstration?”  The proposed rule identifies nine regions, and CMS has randomly selected a single state out of each grouping.  This result is nine participating states: Massachusetts, Maryland, North Carolina, Florida, Washington, Arizona, Iowa, Nebraska and Tennessee.

CMS’s proposed methodology for selecting the states that will be designated as the participating states results in state groupings that “reflect regional variations and enhance the generalizability of the model.”  CMS notes that if they change the selection methodology based upon the comments, these nine states could change or states could be replaced by a selection of sub-state regions.

Do I have to participate?

As proposed, all providers in each of the nine states must participate.  If you are in one of the nine states, you are in the HHVBPP – no exceptions.  Although the program is statewide, participation is based upon state-specific provider numbers.  This means that if you operate in a state that is required to participate and a state that is not required to participate, Nebraska and Oklahoma for example, only your provider number(s) that are in Nebraska will participate.  The Oklahoma numbers will not be involved in the program.  However, all of your provider numbers in Nebraska will participate.

The requirement that all providers in a state participate is both a result of comments from the last value-based home health pilot project and to eliminate the data being skewed.  CMS is concerned that if they did not require all providers in a state to participate, the data would be skewed because only the quality providers, that thought they would qualify for an increase in payments, would apply to participate. This would result in the data not accurately reflecting quality of care.


To ensure providers are competing against comparable providers, CMS proposes two limitations on the model.  Provider performance will be measured against only other providers in the same state.  Furthermore, providers in each state will be categorized into one of two “cohorts.”  These two cohorts are the larger volume cohort and the smaller volume cohort.  CMS defines these cohorts based upon HHCAHPS participation.  Providers that are large enough to participate in HHCAHPS will be in the larger volume cohort and providers that do not participate in HHCAHPS will be in the smaller volume cohort.  In states where the number of providers in the smaller volume cohort is small enough to prevent fair competition, these providers will be placed into the larger volume cohort.

Data gathering and reporting

Performance scores and peer rankings will be determined by using CY 2015 as a baseline year.  Subsequent years will be used to measure performance for each HHA.  The program will provide each HHA with quarterly and annual reports the agency can use to “assess and track their performance.”  The quarterly reports will be issued beginning in July 2016.  The report will be based upon data from January 1, 2016 – March 31, 2016.  Subsequent quarterly reports would be available in  October (April 1 – June 30 quarter), January (July 1 – September 30) and April (October 1 – December 31).  These reports will specify both the HHAs specific performance results (absolute and improvement) and a comparison to other providers within its cohort throughout the entire state.

The annual report will be the annual performance and adjustment report.  This report will inform the agency of its payment adjustment and provide a rationale for the adjustment.  It will then inform the agency when the adjustment would be applied.  The payment adjustment report would be agency-specific and only seen by the agency.  This report will be issued in August of the year following the year for which the report is generated.  This means that in August 2017, a provider will receive their annual performance/adjustment report for CY 2016.  The adjustment proposed in this report will be implemented beginning January 1 of the following year.  Using August 2017 as an example, the provider will see the payment adjustment stated in the August 2017 report implemented in January 2018.

Public Quality Report

The program would include a third and final report.  This report would be an annual, publicly available, quality report.  This report would be designed to provide industry stakeholders information on each agency’s performance.  CMS intends this report to be available to referral sources, and others, to allow them to verify they are referring their patients to quality providers.  This would strongly imply that the publicly available report would include scoring and/or ranking information.  This could be especially damaging to providers, as it may put providers into a “quality hierarchy” within the state.  Providers whose scores are lower than their competitors and, therefore, farther down the hierarchy, may stop receiving referrals as the hospitals and SNFs move to refer patients to providers that are perceived to be “of higher quality.”

Providers may want this report to include not just rankings but scoring information.  As noted below, rankings are relative.  This makes it possible for providers with good scores to have a relative rank that may appear worse than their absolute score would indicate.  This would lead to the conclusion that reporting the actual performance score could be beneficial.  This public report may have a more negative impact than any future payment adjustments.

Payment Adjustments

Payment adjustments based upon performance are the main incentive in HHVBPP.  The Proposed Rule allows providers to receive an increase to their reimbursement for good performance or see a further reduction for poor performance.  Although the program would begin in 2016, the first adjustment will not occur until 2018.  This reflects a process of setting the baseline for performance based upon 2015 data, collecting 2016 data and then measuring performance.  Performance for 2016 would be announced in 2017, and the adjustment would be implemented in 2018.

The payment adjustments will start at +/- 5 percent for 2018 and 2019.  They will then move to +/- 6 percent for 2020 and end at a maximum of +/- 8 percent for 2021 and 2022.  This range of adjustments is much more aggressive than other Value-Based Purchasing programs Medicare has implemented.  In explaining the phase-in of adjustments, CMS did not explain why it proposed to use an adjustment scale that was significantly greater than that used for other providers.  This is a concern because providers that find themselves facing reductions in reimbursement ranging as high as 5 to 8 percent may have a hard time staying in operation, given the other rate reductions they are facing.  Moreover, the Proposed Rule will determine adjustments based upon provider performance relative to others in the cohort, not based upon absolute performance.

Quality Measures

Quality will be measured in the HHVBPP using 25 currently existing measures (10 process measures and 15 clinical measures), plus 4 new measures.  The 4 new measures are not currently reported by agencies to CMS.  All agencies in the selected states will be required to report these new measures.  The new measures are “Advanced Care Planning,” “Adverse Event for Improper Medication Administration and/or Side Effects,” “Influenza Vaccination Coverage for Home Health Care Personnel” and “Herpes Zoster (Shingles) Vaccination received by HHA Patients.”  Because these are new measure, agencies will receive points simply for reporting these measures.

Measuring Performance

Agency performance will be measured using a Total Performance Score.  The score will be developed using all measures, broken into categories and each measure will have equal weight.  The 25 existing measures will account for 90 percent of the TPS.  The other 10 percent of the TPS will come from the 4 new measures, which will each count equally towards this 10 percent.  CMS will award points for each of the measures in each category.  Points will be awarded in each category based upon the higher of the providers quality achievement score or quality improvement score in each category.  This allows agencies that have made significant improvements in quality but whose overall quality score is still low to be recognized as having good performance.

An HHA is awarded points only for measures for which the HHA has provided twenty home health episodes of care per year.  Measures for which points are awarded are called applicable measures.  If an agency has a measure for which it has provided fewer than twenty episodes of care during the year, the measure would not be an applicable measure.  If, due to low volume, an agency has fewer than five applicable measures in a year, that agency would not be subjected to any payment adjustment.  Points for applicable measures are aggregated to generate a TPS.  For all measures, other than the “new measures” an agency will be awarded from 0 to 10 points.  For New Measures, an agency receives 10 points for reporting the measure and 0 points for failing to report the measure.  An agency’s maximum number of points for measures other than the “New Measures” is 250.  If an agency, due to low volume reported fewer than 25 applicable measures, the total possible points would be reduced by ten points for each measure that was not reported.  This allows for low volume agencies that report at least 5 measures to be awarded a score that can be compared to other providers.

Performance Scoring

For purposes of scoring agency performance, CMS will establish benchmarks for each of the measures, other than the four “New Measures.”  Agencies will be awarded points for these measures as follows.  If the agency’s score equals or exceeds the calculated benchmark, which is defined as the mean of the top 10 percent of all HHAs’ performance on the particular measure, the agency will receive 10 points.  If the agency’s score on a measure equals or exceeds the calculated Achievement Threshold, which is defined as the median of all HHAs’ performance on the specified quality measure, during the period, the agency will receive between 1 and 9 points as calculated using a specified formula.  If an agency’s score on a measure is less than that of the Achievement Threshold, the agency will receive 0 points.  The benchmarks and achievement thresholds will be calculated separately for each cohort in each state to ensure similar providers are scored against each other.

Improvement Scoring

Agencies will also have an opportunity to earn points based upon improvement.  For each measure, CMS proposes to use either the performance score or the improvement score, depending upon which one would result in more points being awarded to the agency.  The same benchmark would be used for improvement scoring on each measure.  An agency would receive ten points if its performance score exceeded the benchmark.  If the agency’s score did not exceed the benchmark, but did exceed the agency’s own baseline period score, it would receive between 0 and 10 points, determined by a proposed formula.  The agency’s baseline period score is simply it performance on the particular measure during the CY 2015 period.  If its performance score is less than its baseline period score, the agency would receive 0 points.

The total points for the measures are divided by the total possible points (10 / the number of applicable measures) and multiplied by 90.  Then the total points earned for reporting New Measures is divided by 40 and multiplied by 10.  These two numbers are totaled to reach a TPS between 0 and 100.  This score is then used to determine the adjustment.

Transforming the TPS Scoring into an Adjustment

CMS proposes to use a linear exchange function to translate the performance scores into payment adjustments.  In this process, the average TPS score for agencies in a state and cohort will be equal to a 0 percent adjustment.  Agencies whose TPS are below the average will receive a payment reduction.  Agencies whose TPS is above average will receive a payment increase.  The amount of any increase or decrease will depend upon where the agency falls along the curve and the slope of the curve.

This means that an agency’s adjustment is based not only upon their TPS but upon their TPS relative to other agencies in the cohort.  Because the average becomes 0, it is possible, depending upon the slope of the function, for a provider that performs well but not as well as other providers in its cohort to have its payments reduced.  This same issue arises in other VBP models but is more concerning in the HHVBPP because the adjustments are much higher.

Review and Recalculation Request

Under CMS’s proposal, agencies would be provided two opportunities to review scoring information.  They will have the quarterly reports as well as the annual report that announces the adjustment.  Agencies will be able to challenge any adjustment if they identify a discrepancy or calculation error.  They can challenge the scores reported in the quarterly reports within ten days of receiving them.  CMS will only adjust the scores if they determine they made a mistake in calculating the numbers.  Similarly, the annual reports, which will be received in August before the adjustments can be challenged, but an agency will only be able to request a scoring recalculation and will be required to include a specific basis for the requested recalculation.


Based upon the Proposed Rule, value-based purchasing is as much the future of home health care as it is the future for other Medicare providers.  Unfortunately, CMS proposes to make the HHVBPP more onerous in terms of the size of adjustments.  The comments made on this program will be extremely important.  Providers that are not in one of the nine program states need to not only watch how the program develops in those states but begin preparing now so that you are ready when HHVBP comes to the rest of the industry.  With the numerous reimbursement cuts the industry has received over the past few years, the risk of ongoing negative adjustments for performance should drive agencies to improve the quality assurance efforts with the goal at being at or near the top of performance in their state and cohort when this program goes national.