The Cost of Preventing Readmissions: Why Surgeons Should Lead the Effort

In accordance with the Affordable Care Act, Medicare has instituted financial penalties for hospitals with 30-day readmission rates that exceed a predetermined value. Currently, this value only considers ''excess'' readmissions for myocardial infarction, heart failure, and pneumonia with a maximum fine being one per cent of total Medicare reimbursements. In 2015, this penalty will increase to three per cent and encompass more surgical diagnoses. We retrospectively reviewed a database of adult patients undergoing cardiac surgery treated at our institution in 2012 to establish whether patients with readmissions within 30 days of the index operation could have been managed more cost-effectively without readmission. A calculation of cost efficiency was performed to compare the net hospital profit for two scenarios: admitting patients versus hypothetical preventative measures. Of the 576 patients during the study period, a total of 68 (11.8%) patients with unplanned 30-day readmissions were identified. Outpatient management was determined to have been feasible for 18 (26.5%) patients. Whereas the calculated net profit for readmission was $144,000, inclusion of Medicare's penalty resulted in a loss of $11,950. A readmission reduction program with an annual cost exceeding $11,950 would lead to financial loss. The financial implications of Medicare's readmission penalty alone necessitate the development of cost-effective strategies to reduce rehospitalization.

READMISSIONS ARE ASSOCIATED with increased mor- tality, and rehospitalization rates are used as sur- rogate markers for quality of care.1-5 In accordance with Section 3025 of the Patient Protection and Affordable Care Act, recent Medicare legislation mandates fi- nancial penalties for hospitals with 30-day readmission rates higher than calculated, national threshold values.6-9 At present, this calculated value only considers ''excess'' readmissions for three diagnoses: acute myocardial in- farction, congestive heart failure (CHF), and pneumo- nia with the maximum readmission penalty to a given hospital being one per cent of that institution's regular reimbursements.6, 10 Effective 2015, Medicare's read- mission fines will increase to three per cent and will encompass more diagnostic codes.

Projects to reduce readmissions are highly variable in methodology and efficacy and may paradoxically increase the cost of care.11 Therefore, some institutions have not been eager to champion efforts to reduce readmission as a result of financial concerns. As we have previously reported, reasons for admission of surgical patients are surgery-specific and should be tailored to individual service lines.12 The objective of this pilot study was to use a prospective institutional database to establish whether a population of adult patients undergoing cardiac surgery with readmissions (within 30 days of index operation) to our institution could have been managed cost-effectively without read- mission. We aimed to identify the maximum financial cost of readmission prevention efforts that would still yield a net profit while improving the quality of care.

Methods

A thorough analysis of the current and proposed Centers for Medicare and Medicaid Services (CMS) regulations pertaining to planned versus unplanned readmission and hospital readmission reimbursement penalties was conducted. A retrospective review of our institutional database was performed and all adult patients undergoing cardiac surgery treated during a 12-month period were identified (2012). Detailed preoperative, intraoperative, and postsurgical outcome data were collected in accordance with the Society of Thoracic Surgeons Adult Cardiac Surgery Database Collection Form (version 2.73). Data included the following administrative and financial elements: admission di- agnosis, age, operative procedure(s), length of stay (LOS), insurance status, Medicare reimbursement for original hospitalization, postdischarge readmission date, LOS for readmission, readmission diagnoses (categorized by planned, related/not related to original admission), cost of readmission (including bed, pro- cedure, and pharmacy costs), and reimbursement for rehospitalization.

After review of medical records, two surgeons and a nurse with expertise in cardiac surgery identified a subset of patients within the readmit cohort whose readmissions were considered preventable. An admis- sion was considered preventable if the diagnostic and therapeutic interventions could have been performed in an outpatient setting. This determination was made for each case, taking into consideration each patient's readmission diagnosis, procedures performed on read- mission, and overall readmission course/physiological state. Specifically, if the presenting vital signs were stable and diagnostics and treatments were routinely rendered in the outpatient setting, the rehospitalization was deemed preventable.

Institutional cost data were obtained from the Uni- versity Health Consortium (UHC), and reimbursement information was obtained from Medicare databases. Financial analysis was performed comparing net hos- pital reimbursement in two scenarios: readmitting the cohort as inpatients versus the use of outpatient facilities to render the same treatment. To perform this compar- ison, calculations of cost and reimbursement under both inpatient and outpatient scenarios were carried out.

Using the CMS methodology to determine read- mission adjustment factors and payment adjustment amounts as well as published data from Medicare da- tabases, reimbursement was calculated for each case. The total admission cost was calculated as the sum of direct room costs (i.e., nursing, standard supplies, etc.) multiplied by the LOS and procedural/ancillary ser- vice costs. For each readmitted subject, an estimated net profit calculation was performed as Medicare reimbursement minus total admission cost. This was then compared with the cost of similar diagnostics and procedures as an outpatient. These values were aggregated for the patient population to determine the total net reimbursement under each scenario, inpatient versus outpatient status (see Fig. 1 for an outline of the calculation).

Statistical analyses were used to analyze patient characteristics and to establish whether the difference in net profit for inpatient versus outpatient status was statistically significant. The unpaired, two-tailed, Stu- dent's t test was performed for continuous variables, and the x2 test was performed for categorical data (with a 4 0.05). Statistical analysis was completed using Microsoft Excel (Version 2010; Microsoft Corporation, Redmond, WA). This study was approvedbytheInstitutionalRe- view Board at the University of California, Los Angeles.

Results

Of the 576 patients treated in our facility during the study period, a total of 68 (11.8%) with unplanned 30- day readmissions were identified. It was determined that for 18 (26.5%) of these patients, readmission was ''preventable.'' Comparisons of patient characteristics for the preventable readmission versus nonpreventable readmission groups are depicted in Tables 1 to 3. Both the average time between discharge and readmission and the proportion discharged home were not statisti- cally different between the two groups. The majority of patients in both the preventable (83%) and non- preventable (72%) readmission groups were discharged home after the index operation (Table 2).

The most common original operation for both co- horts was coronary artery bypass grafting and/or valve repair/replacement (Table 3). Patients with preventable 30-day readmissions were more likely: to be older (P 4 0.04), to have a shorter hospital LOS (P 4 0.04), to be in CHF (P 4 0.02), to not have chronic kidney disease (P 4 0.02), and to have been referred to a cardiac rehabilitation facility ( P 4 0.04) (Table 1).

The most common reasons for readmission included CHF and deconditioning (e.g., dehydration, presyncope, malaise), as shown in Figure 2. Other common reasons included respiratory complications, arrhythmia/heart block, anticoagulation issues, gastrointestinal problems, renal failure, and fever/infection (Fig. 2). The calcu- lated reimbursement for these patients' readmissions was $144,000 per patient. However, inclusion of Medi- care's institutional readmission penalty resulted in a net loss of $11,950 per patient (P <0.001).

Discussion

With the institution of the Affordable Care Act and significant changes in healthcare reimbursement, surgeons and other practitioners should be actively involved in the development of strategies to reduce unintended rehospitalization. Our institution has led a number of efforts aimed at reducing avoidable read- missions. Interventions focused on transition of care and postdischarge phases have yielded variable success. To date, a tangible financial value for readmission pre- vention efforts has not been reported. In this study, we evaluated the cost/reimbursement implications of patient readmissions after cardiac surgery. In a hypothetical ex- periment, we examined reimbursement comparisons, within a suitable cardiac surgical patient subgroup, for inpatient admission versus outpatient status. We ini- tially discovered that each readmission was actually yielding profit to the hospital system. However, taking into account Medicare's penalty, readmissions would produce a net loss. Therefore, readmission reduction programs costing more than $11,950/patient/year would result in financial loss.

As seen in our study, the reasons for readmission to surgical services are varied and, in this case, are di- rectly related to the surgical procedure and the medical comorbidities of the patients. We have previously reported that heart failure, insurance status, and race are also significant predictors of readmission after cardiac operations.12 In the present study, one-third of the preventable readmissions were related to symp- toms of heart failure.

The main limitation to this study was the lack of financial data directly from our institution. Most hos- pitals will not release their actual charges and revenue as a result of a variety of reasons. Therefore, we in- stead using previously validated UHC and CMS pub- lished data for our institution. However, because great heterogeneity exists in costs between different medical facilities, our conclusions are limited to ''like'' institutions (i.e., medical centers with similar patient acuity and volume). This limitation may be addressed by future studies aimed at accomplishing similar financial analysis for a wider range of medical facilities. Also, the retrospective nature of this study may have in- troduced inadvertent bias. We tried to limit this by examining the entire medical record and accounting for all of the procedures and treatments performed during rehospitalization.

In summary, the development of cost-effective strat- egies to reduce rehospitalization is both important and advantageous. Given the findings of this study, we be- lieve that readmission prevention programs would be best applied to surgical services with high volumes of low-risk patients. Clearly, there is a subset of read- missions that is not preventable and must be accounted for during CMS penalty calculation. The relatively small operating margin for surgical providers mandates critical review of the care process and reasons for readmission, which are often specific to individual ser- vice lines. As they are most familiar with the entire process of surgical care, surgeons are ideal candidates for leading teams that would devise alternatives to readmission and improve the quality of patient care.

REFERENCES

1. Ashton CM, Del Junco DJ, Souchek J, et al. The association between the quality of inpatient care and early readmission. Med Care 1997;35:1044-59.

2. Ashton CM, Wray NP. A conceptual framework for the study of early readmission as an indicator of quality of care. Soc Sci Med 1996;43:1533-41.

3. Benbassat J, Taragin M. Hospital readmissions as a measure of quality of health care. Arch Intern Med 2000;160:1074-81.

4. Halfon P, Eggli Y, Pretre-Rohrbach I, et al. Validation of the potentially avoidable hospital readmission rate as a routine indicator of the quality of hospital care. Med Care 2006;44: 972-81.

5. Rumsfeld JS, Allen LA. Reducing readmission rates: does coronary artery bypass graft surgery provide clarity? JACC Car- diovasc Interv 2011;4:577-8.

6. Axon RN, Williams MV. Hospital readmission as an ac- countability measure. JAMA 2011;305:504-5.

7. Hansen LO, Young RS, Hinami K, et al. Interventions to reduce 30-day rehospitalization: a systematic review. Ann Intern Med 2011;155:520-8.

8. Joynt KE, Jha AK. Thirty-day readmissions-truth and consequences. N Engl J Med 2012;366:1366-9.

9. Kocher RP, Adashi EY. Hospital readmissions and the Af- fordable Care Act. JAMA 2011;306:1794-5.

10. Krumholz HM, Merrill AR, Schone EM, et al. Patterns of hospital performance in acute myocardial infarction and heart failure 30 day mortality and readmission. Circ Cardiovasc Qual Outcomes 2009;2:407-13.

11. Safran C, Phillips RS. Interventions to prevent readmission: the constraints of cost and efficacy. Med Care 1989;27:204-11.

12. Lancaster E, Postel M, Satou N, et al. Introspection into institutional database allows for focused quality improvement plan in cardiac surgery: example for a new global healthcare system. Am Surg 2013;79:1040-4.

MACKENZIE POSTEL, PAUL N. FRANK, M.D., TOD BARRY, M.B.A., NANCY SATOU, R.N., RICHARD SHEMIN, M.D., PEYMAN BENHARASH, M.D.

From the Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, California

Presented at the 25th Annual Scientific Meeting of the Southern California Chapter of the American College of Surgeons, January 17-19, 2014, in Santa Barbara, California.

Address correspondence and reprint requests to Peyman Benharash, M.D., 10833 Le Conte Avenue, UCLA Center for Health Sciences, Room 62-249, Los Angeles, CA 90095. E-mail: Pbenharash@mednet. ucla.edu.