HI410 Unit 7 Coding Compliance Audit Example

Discharged Home, Seen by a Home Health Nurse the Next Morning: A Unit 7 Coding Compliance Audit of Forty Short Stays Under the Post-Acute Care Transfer Rule

[Student Name]

Purdue University Global

HI410: Advanced Reimbursement Methodology

Unit 7 Coding Compliance Audit

[Instructor Name]

[Date]

The hospital, the claims, the weights and every dollar figure are composite; the regulations and sources cited are real.

Purpose and Scope

The compliance committee asked health information management to test one field on inpatient claims that coders rarely question: the patient discharge status. For certain groups of diagnoses, Medicare pays a reduced, per diem amount when a stay ends well before the group's typical length and the patient leaves for a skilled nursing facility, or for home with home health care that begins within three days and relates to the hospitalization (42 C.F.R. § 412.4). A status of "discharged home" on such a claim produces a full payment. If the patient actually went to a nursing facility or started home health care, the claim is overpaid. The reverse error also exists, and it costs the hospital money.

Method

The sampling frame was every Medicare inpatient claim in one composite quarter that met three conditions: the group was on the transfer list, the stay, counted with one extra day, still fell short of the group's geometric mean, and the claim was paid. That frame held 212 claims. A random sample of 40 was drawn. For each claim, a credentialed coder who had not coded the original stay compared the claim's discharge status with four sources: the physician's discharge order, the case management discharge note, the nursing discharge summary and the post-acute provider's start-of-care record, where one existed. The audit is a probe, not a statistically valid sample, and its rates should not be projected to all 212 claims without a larger draw. The corrected payment for each changed claim follows the transfer formula: the full group payment divided by the geometric mean length of stay gives a per diem, and the hospital receives twice the per diem for the first day and one per diem for each additional day, up to the full payment. All figures use the hospital's composite adjusted base of $6,313.56.

Findings

Six of the 40 claims, 15.0 percent, carried a discharge status that the record did not support.

1. Claim A, group weight 1.4120, geometric mean 4.6 days, actual stay 2 days. Billed as discharged home. Home health start-of-care visit occurred the morning after discharge for wound care of the surgical site. Full payment $8,914.75; corrected transfer payment $5,813.97; overpaid $3,100.78.

2. Claim B, weight 0.9840, geometric mean 3.4 days, stay 1 day. Billed as home. Case management note: "accepted at skilled facility, transport 15:00." Full payment $6,212.54; corrected $3,654.44; overpaid $2,558.10.

3. Claim C, weight 1.1060, geometric mean 3.9 days, stay 2 days. Billed as home. Home health began two days after discharge for heart failure monitoring. Full $6,982.80; corrected $5,371.38; overpaid $1,611.42.

4. Claim D, weight 1.6380, geometric mean 5.1 days, stay 3 days. Billed as home. Nursing discharge summary: "report called to receiving facility." Full $10,341.61; corrected $8,111.08; overpaid $2,230.53.

5. Claim E, weight 1.2240, geometric mean 4.3 days, stay 2 days. Billed as discharged to a skilled facility. The facility had no bed; the case management note documents a change of plan, and the patient went home without home health. Paid $5,391.48; correct full payment $7,727.80; underpaid $2,336.32.

6. Claim F, weight 0.9120, geometric mean 3.6 days, stay 1 day. Billed as home with home health. The agency's first visit took place on the fifth day after discharge, outside the three-day window. Paid $3,198.88; correct full payment $5,757.97; underpaid $2,559.09.

Every one of the six errors came from a discharge plan that changed after the claim's status had already been chosen, which makes this a process failure, not a coding knowledge failure.

Causes

The errors share one cause. Coders select the discharge status from the physician's discharge order, which records the plan at the moment it is signed. Plans change after that moment. In Claims B and D, the order said home, and case management arranged a facility placement the same afternoon. In Claims A and C, the order said home without mentioning home health, and the referral was made by case management afterward. In Claim E, the plan changed in the opposite direction. In Claim F, a home health referral was made but the agency did not start within three days, which no one at the hospital would have known when the claim was coded. Coding accuracy research consistently finds that errors cluster where the information a coder relies on is incomplete or captured early, not where the rule is unclear (O'Malley et al., 2005).

Financial Summary

1. Overpayments, four claims: $9,500.83.

2. Underpayments, two claims: $4,895.41.

3. Gross dollar error: $14,396.24, or 5.83 percent of the $246,912.30 paid on the 40 sampled claims.

4. Net amount owed to the program: $4,605.42, or 1.87 percent.

The report presents the overpayments and underpayments in separate lines on purpose. The net figure is the amount that changes hands, but it understates the size of the problem by two thirds. A committee that saw only 1.87 percent might conclude that the field is accurate enough; the gross rate shows that one claim in seven is wrong in some direction. The two columns also follow different rules. The four overpaid claims fall under the federal overpayment rule, whose 60-day clock starts on the date this report identifies them (42 C.F.R. § 401.305), and underpayments are corrected through the ordinary claim adjustment process within the timely filing limits.

Recommendations

1. Hold the discharge status field on every claim in a transfer-list group until a case management discharge disposition is filed or 72 hours have passed, whichever comes first. Owner: revenue cycle director. Target: the start of the next quarter.

2. Add the case management disposition note to the documents a coder must read for these groups. Owner: coding manager. Target: 30 days.

3. Build a monthly report matching discharged-home claims in transfer-list groups against home health and nursing facility admission data available to the hospital through its post-acute partners. Owner: health information management analyst. Target: 60 days.

4. Correct the four overpaid claims and return the $9,500.83 through the voluntary refund process within 60 days of this report, and adjust the two underpaid claims. Owner: compliance officer with patient financial services. Target: 45 days.

5. Expand the review to all 212 claims in the frame and to the two preceding quarters, because the cause is systemic. Owner: compliance officer. Target: 90 days.

Conclusion

Discharge status is one field, and it is easy to treat as clerical. For short stays in transfer-list groups, it decides whether the hospital receives a full payment or a per diem, and it is coded at the moment when the patient's destination is least certain. Reimbursement texts describe the transfer rule as a policy that pays for the part of care a hospital actually provided (Casto & White, 2021). This audit shows that applying it correctly depends less on knowing the rule than on waiting for the plan to settle.

References

Casto, A. B., & White, S. (2021). Principles of healthcare reimbursement and revenue cycle management (7th ed.). AHIMA Press.

42 C.F.R. § 401.305 (2024). https://www.ecfr.gov/current/title-42/section-401.305

42 C.F.R. § 412.4 (2024). https://www.ecfr.gov/current/title-42/section-412.4

O'Malley, K. J., Cook, K. F., Price, M. D., Wildes, K. R., Hurdle, J. F., & Ashton, C. M. (2005). Measuring diagnoses: ICD code accuracy. Health Services Research, 40(5p2), 1620-1639. https://doi.org/10.1111/j.1475-6773.2005.00444.x