Joint replacement model requires CDI evolution

Under the Comprehensive Care for Joint Replacement (CJR), which began April 1, acute care hospitals in selected geographic areas assume quality and payment accountability for retrospectively calculated bundled payments for lower extremity joint replacement (LEJR) episodes. Episodes begin with admission to an acute care hospital for an LEJR procedure that is paid under the IPPS through MS-DRGs 469 or 470 (Major joint replacement or reattachment of lower extremity with or without MCC, respectively) and end 90 days after the date of discharge from the hospital.

 

The hospitals are permitted to engage physicians and postacute care providers in "collaborator" agreements. Collaborator agreements establish financial risk/reward relationships, and align interests to promote care delivery which will meet or exceed CJR cost and quality outcomes.

 

The cost and quality measures used to assess performance under the CJR are largely derived from reported ICD-10 codes in the inpatient and ambulatory care settings. Clinical documentation improvement (CDI) programs have a tremendous opportunity to get engaged to promote and support the capture of documentation and reported codes impactful to these performance measures. In this article, we will explore a few of the CDI vulnerabilities relevant to this population.

A Medicare fee-for-service beneficiary is included in the CJR model when a claim is submitted for an inpatient encounter assigned MS-DRGs 469 or 470. These surgical MS-DRGs include total hip and knee replacements, ankle arthroplasties, partial hip replacements, lower leg, ankle and thigh reattachments, and hip resurfacing procedures. In the CJR final rule, CMS noted that the majority of the procedures in these MS-DRGs are total and partial hip replacements, and total knee replacements.

 

The key CDI vulnerability associated with CJR patient selection is inaccurate MS-DRG assignment. The included MS-DRGs are replacement?not revision?procedures. Joint revision procedures are more complex, have higher costs, and are therefore assigned to different MS-DRGs (466?468, Revision of hip or knee replacement with or without MCC).

 

If the coder omits assignment of the ICD-10-PCS code for the removal of the original device and only codes the replacement procedure, a patient with a revision?who should be assigned to MS-DRGS 466?468?will instead be misclassified into MS-DRGs 469 or 470, and will skew CJR clinical and cost outcomes.

 

Under the CJR model, reimbursement is determined using a two-sided reimbursement model, which means that if the hospital and "collaborators" deliver patient services below the predetermined target price, they are eligible to share in the savings. However, if the costs exceed the target, they will be required to make a repayment to CMS.

 

CMS noted that patients in MS-DRGs 469 and 470 with fractures, and/or with MCCs, take more resources to manage and thus have higher costs. Therefore, when ICD-10 codes for hip fractures are assigned, or when an MCC is assigned, CMS adjusts the target price for these patients.

 

If we underreport these events either due to documentation and/or code assignment errors, the target price established for our hospital is negatively impacted, and the costs measured for the populations we treat are overstated because we have not accurately reflected patient complexity.

The hip fracture codes which trigger target price adjustment include acute and pathological fractures of the hip. The likelihood of such oversights due to coding and/or documentation challenges is minimal. Nonetheless, due to the significant impact of these omissions under the CJR model, CDI teams are encouraged to build in checks and balances to monitor accurate capture as part of the CDI specialist/coder MS-DRG prebill reconciliation process.

 

MCC capture has been a long-standing challenge for this population due to both the nature of the population and the marginal engagement of some surgeons in complete documentation impactful to MS-DRG assignment.

 

Common MCCs for this population include an acute exacerbation of systolic/diastolic heart failure, encephalopathy, or acute respiratory failure. The use of evidence-based definitions and clear clinical indicators to identify and prompt the capture of such conditions is essential to CJR data quality. Given the aligned incentives encouraged by the CJR model, once the impact of data quality on CJR performance is understood, surgeon engagement barriers should be greatly alleviated.

 

The CJR also stipulates performance for defined quality measures which roll up into a "quality composite." Hospitals must demonstrate minimum quality performance to be eligible to receive any cost savings, and the strength of quality performance impacts the amount of earned savings as well as—in situations where costs exceed established thresholds—the amount of repayment.

 

One of these quality measures, the Risk Standardized Complication Rate (RSCR) for the Total Hip/Total Knee Arthroplasty (THA/TKA) cohort, is a claims-based measure. Performance for this measure contributes 50% of the weight in the CJR quality composite index, in addition to reimbursement under the Hospital Value-Based Purchasing Program. Note that the RSCR cohort is a subset of patients included in the CJR. While the CJR includes inpatient beneficiaries in MS-DRGs 469 and 470, the RSCR includes only those beneficiaries with ICD-10-PCS codes for total hip or total knee arthroplasties. The RSCR further qualifies discharges eligible for the cohort with other requirements.

 

Inpatient CDI programs impact accurate selection of patients in the RSCR cohort, and accurate capture of complications. However, the quality of documentation and coding in both the inpatient and ambulatory settings impacts risk adjustment. The risk adjustment methodology assesses the complexity of the THA/TKA patient population and their likelihood of experiencing complications. The complexity of each patient is then expressed as predicted and expected risk adjusted complication rates, and these rates—in conjunction with the actual (or "observed") number of complications—determine hospital performance for the RSCR measure.

 

Designated comorbidities reported on claims for the 12 months prior to the admission for the THA/TKA procedure are used in the risk adjustment algorithm.

 

The CMS risk adjustment methodology uses a 12-month look-back period, beginning with the admission for the THA/TKA procedure, to identify reported comorbidities. Both Part A and Part B claims are included in the analysis. Some comorbidities only count if they were documented and reported prior to the admission for the THA/TKA procedures (the CMS risk adjustment methodology, at present, does not use present on admission status).

 

CDI vulnerabilities associated with risk adjustment are related to the accurate and complete reporting of chronic comorbidities across the continuum.

 

"Chronic renal insufficiency" must be documented and reported prior to the inpatient admission for the THA/TKA procedure for the comorbidity to contribute to risk adjustment. If the physician's office documents and reports "renal insufficiency" (and does not use the word "chronic" when appropriate), a different ICD-10 code is reported and this comorbidity will not "count."

 

Hospitals required to participate in the CJR model are working diligently to redesign care delivery processes to meet and/or exceed cost and quality outcome thresholds. Many executive teams remain unaware of the impact of documentation and coding on these variables. I hope this article empowers you to step up to the plate and get engaged. CDI programs have an exciting opportunity to expand their scope and impact.