PURSUANT TO THE PROVISIONS OF THE State Administrative Procedure Act, NOTICE is hereby given of the following action:
Action taken:
Amendment of section 86-1.37 of Title 10 NYCRR.
Statutory authority:
Public Health Law, section 2807-c(35)(b)(v)
Subject:
Potentially Preventable Readmissions.
Purpose:
Implements a revised reimbursement policy related to hospital readmissions that are determined to be potentially preventable.
Text of final rule:
Pursuant to the authority vested in the Commissioner of Health by section 2807-c(35) of the Public Health Law, Subpart 86-1 of Title 10 of the Official Compilation of Codes, Rules and Regulations of the State of New York, is amended by adding a new Section 86-1.37, to be effective upon publication of the Notice of Adoption in the New York State Register, to read as follows:
Part 86-1.37 Readmissions
(a) For discharges occurring on and after July 1, 2010, Medicaid rates of payment to hospitals that have an excess number of readmissions as defined in accordance with the criteria set forth in subdivision (c), as determined by a risk adjusted comparison of the actual and expected number of readmissions in a hospital as described by subdivision (d), shall be reduced in accordance with subdivision (e).
(b) Definitions. For purposes applicable to this section the following terms shall be defined as follows:
(1) Potentially Preventable Readmission (PPR) shall mean a readmission to a hospital that follows a prior discharge from a hospital within 14 days, and that is clinically-related to the prior hospital admission.
(2) Hospital shall mean a general hospital as defined pursuant to section 2801 of the Public Health Law.
(3) Observed Rate of Readmission shall mean the number of admissions in each hospital that were actually followed by at least one PPR divided by the total number of admissions.
(4) Expected Rate of Readmission shall mean a risk adjusted rate for each hospital that accounts for the severity of illness, APR-DRG, and age of patients at the time of discharge preceding the readmission.
(5) Excess Rate of Readmission shall mean the difference between the observed rates of potentially preventable readmissions and the expected rate of potentially preventable readmissions for each hospital.
(6) Behavioral Health shall mean an admission that includes a primary or secondary diagnosis of a major mental health related condition, including, but not limited to, chemical dependency and substance abuse.
(7) Managed Care Encounter Data shall mean claims-like data that describes services provided by managed care plans to their enrollees.
(c) Readmission Criteria.
(1) A readmission is a return hospitalization following a prior discharge that meets all of the following criteria:
(i) The readmission could reasonably have been prevented by the provision of appropriate care consistent with accepted standards in the prior discharge or during the post discharge follow-up period.
(ii) The readmission is for a condition or procedure related to the care during the prior discharge or the care during the period immediately following the prior discharge and including, but not limited to:
(a) The same or closely related condition or procedure as the prior discharge.
(b) An infection or other complication of care.
(c) A condition or procedure indicative of a failed surgical intervention.
(d) An acute decompensation of a coexisting chronic disease.
(iii) The readmission is back to the same or to any other hospital.
(2) Readmissions, for the purposes of determining PPRs, excludes the following circumstances:
(i) The original discharge was a patient initiated discharge and was Against Medical Advice (AMA) and the circumstances of such discharge and readmission are documented in the patient's medical record.
(ii) The original discharge was for the purpose of securing treatment of a major or metastatic malignancy, multiple trauma, burns, neonatal and obstetrical admissions.
(iii) The readmission was a planned readmission or one that occurred on or after 15 days following an initial admission.
(iv) For readmissions occurring during the period up through March 31, 2012, the readmission involves an original discharge determined to be behavioral health related.
(d) Methodology.
(1) Rate adjustments for each hospital shall be based on such hospital's 2007 Medicaid paid claims data and managed care encounter data for discharges that occurred between January 1, 2007 and December 31, 2007.
(2) The expected rate of readmissions shall be reduced by 24% for each hospital for periods prior to September 30, 2010; 38.5% for the period October 1, 2010 through December 31, 2010; and 33.3% on and after January 1, 2011.
(3) Excess readmission rates are calculated based on the difference between the observed rate of PPRs and the expected rate of PPRs for each hospital.
(4) In the event the observed rate of PPRs for a hospital is lower than the expected rate of PPRs, the excess number of readmissions shall be set at zero.
(e) Payment Calculation.
(1) For the excess readmissions identified in paragraph (3) of subdivision (d) of this section, each hospital's projected payment rate for the 2010 rate period, as otherwise computed in accordance with this subpart, will be used to compute the relative aggregate payments, excluding behavioral health, associated with the risk adjusted excess readmissions in each hospital.
(2) For each hospital, a hospital specific readmission adjustment factor shall be computed as one minus the ratio of the hospital's relative aggregate payments associated with the excess readmissions from paragraph (3) of subdivision (d) of this section and the hospital's relative aggregate payments for all non-behavioral health Medicaid discharges as determined pursuant to this subdivision.
(3) Non-behavioral health related payments to hospitals shall be reduced by applying the hospital readmission adjustment factor from paragraph (2) of this subdivision to the applicable case payment or per-diem payment amount for all non-behavioral health related Medicaid discharges to the hospital.
Final rule as compared with last published rule:
Nonsubstantive changes were made in section 86-1.37(d)(2).
Text of rule and any required statements and analyses may be obtained from:
Katherine Ceroalo, DOH, Bureau of House Counsel, Regulatory Affairs Unit, Room 2438, ESP, Tower Building, Albany, NY 12237, (518) 473-7488, email: regsqna@health.state.ny.us
Revised Regulatory Impact Statement, Regulatory Flexibility Analysis, Rural Area Flexibility Analysis and Job Impact Statement
Changes made to the last published rule do not necessitate revision to the previously published RIS, RFA, RAFA and JIS.
Assessment of Public Comment
The Department received a letter with comments and concerns relating to the establishment of Part 86-1.37, regulations implementing the Potentially Preventable Readmissions payment policy. The comments in the letter coordinated by the Healthcare Association of New York State (HANYS) and our response to them are as follows:
1. Comment: HANYS argues that the methodology of adjusting expected experiences to meet targeted rate reductions, which effects over 90% of hospitals, is counterintuitive to promoting appropriate quality related measures.
Response: The Department has established a risk adjusted method that predicts the expected number of PPRs within a 14 day window at each hospital based on historic (2007) claims data. For all hospital admissions that were "at risk" of being followed by a PPR, a statistical model was developed that predicted the likelihood of a subsequent PPR based on the recipient's condition (APR-DRG) and severity of illness, as well as the recipient's age at the time of that initial hospital admission. In this way, the recipient's clinical condition prior to any subsequent readmissions served as the basis for our estimate of whether or not a subsequent readmission was likely to occur.
The expected number of PPRs are derived from the statistical model for all at risk events at each hospital was then compared to the observed number of PPRs at that hospital within 14 days of the at risk event to examine performance standards. As a result of budget negotiations, the statute incorporated a required gross Medicaid savings target. In order to comply with the enabling statute, the expected number of PPRs derived from the statistical model had to be reduced.
Although under the proposed regulations 92% of hospitals would suffer financial penalties, only 3.8% of hospitals will sustain a PPR penalty exceeding 2% of their Medicaid revenue (full annual hospital-specific PPR percentage reduction factors).
2. Comment: HANYS contends that the Department's use of 2007 data does not incentivize hospitals to improve outcomes now and in the future.
Response: 2007 claims were the most recent and reliable records that the Department had available during the development of the PPR methodology. During budget discussions and in the educational webinars the State has indicated that a process to engage stakeholders was underway to begin advancing the base year data for 2011 and forward.
3. Comment: HANYS does not agree with the State's policy to include PPRs where the initial admission was medical/surgical but the readmission has a primary or secondary diagnosis of behavioral health.
Response: The PPR proposal excludes all initial events followed by PPRs and all subsequent readmissions (regardless of whether they were behavioral health related or not) if the initial admission had a primary or secondary diagnosis that was behavioral health related, mental health or substance abuse related (as defined by ICD-9-CM diagnosis codes). However, if the initial admission was not behavioral health related, this admission and all subsequent readmissions (behavioral health or not) were included in all subsequent calculations. The State believes that it is entirely appropriate to include these events since it was the initial medical/surgical admission that was clinically related to subsequent readmissions.
4. Comment: HANYS proposes utilizing other variables in assessing the risk adjusted rates for readmissions, which will recognize the disproportionate impact on safety net providers and other hospitals serving low income communities.
Response: Although we believe clinical risk for readmissions was appropriately assessed using the methodology described, the State is open to further discussions with the Industry on any opportunity to improve the risk adjusted methodology. Regardless of any revisions, unless statutorily changed, $47 million in gross Medicaid savings will have to be achieved.
5. Comment: HANYS contends that non-case payment hospitals should be excluded from the PPR reduction calculation.
Response: A statutory change would be required to exclude non-case payment hospitals and/or low volume Medicaid providers from the PPR reduction calculation.
6. Comment: HANYS is advocating for a reconciliation of savings by hospital to the statewide target reduction due to fluctuations in volume.
Response: A statutory change would be required to reconcile the actual savings achieved to the total penalty calculated for each hospital based on historic data. Although there will most likely be inequities caused by fluctuations in hospital volume and any increase in hospital admissions subsequent to 2007, some of that is captured under DOH's current methodology because the percentage reduction is calculated using projected 2010 revenue (using 2008 Medicaid claims). In addition, the Department plans to use more recent Medicaid claims data (2009) to calculate PPR rates and adjustments in subsequent fiscal years, which should mitigate this issue.
7. Comment: HANYS indicates that the 3M proprietary software is cost prohibitive, thus limiting the ability of hospitals to access the software to monitor and replicate the calculations used by DOH or to inform their ongoing quality improvement efforts.
Response: 3M's APR-DRG grouping software is currently being used by all hospitals and most vendors for Medicaid claiming purposes. In addition, at the request of the State, 3M has worked with the Associations and individual hospitals to negotiate significantly discounted rates for the grouping and PPR software applications.
8. Comment: HANYS believes that Medicaid savings should come from reduced readmission rates, not financial penalties adjusted to meet a budget target. HANYS is concerned that the message this proposal regulation communicates to the provider community is that cost savings, not standards of care, will drive quality of care and patient safety.
Response: The current design of the Medicaid fee-for-service inpatient payment system does not provide incentives to contain avoidable readmissions or promote high quality of care. It presents incentives for providers to increase volume of care rather than to reduce it; reducing readmissions would result in fewer billable discharges. A hospital that has performed poorly, observed higher readmission rates than what was expected, will be penalized.
9. Comment: HANYS encourages an approach targeted toward the dual goals of cost savings and improved care management. HANYS believes a positive first step toward creation of a quality-directed readmissions policy is for DOH to reward hospitals that take action to reduce preventable readmissions.
Response: DOH's original intent was to promote quality improvement through incentive payments to improved/high quality of care performers. The Governor's Executive Budget (2010-11) proposal for PPRs did include financial incentives for those hospitals demonstrating improved rates of readmissions. However, this initiative was not included as part of the final negotiated budget. Incentive payments cannot be made without statutory authority. DOH agrees that there will be associated savings accrued from a decrease in the number of readmissions going forward, and that these savings need to be recognized in some manner (i.e. incentive payment funding).
