VELMOOR COLLEGE OF HEALTH SCIENCES Case Study Source Document — HIM Program Title: The Data Behind the Discharge Student Scenario: Maya Tsai, HIM 125 — Health Information Systems --- BACKGROUND Maya Tsai is a second-semester HIM BS student completing a data quality assignment in HIM 125. She has been given a simulated discharge dataset from a small community hospital (Lakeview General — simulated) containing 20 patient discharge records. Her task is to identify discrepancies between two length-of-stay (LOS) measures: LOS_calendar (the number of calendar days from admission to discharge date) and LOS_billing (the LOS value submitted on the billing claim). In a correctly maintained dataset, LOS_calendar and LOS_billing should match for most records. Any discrepancy may indicate a data entry error, a system calculation difference, or a potential billing compliance issue. Maya's assignment asks her to flag all discrepancies, classify each by probable cause, and summarize her findings in a brief quality audit memo. --- THE DISCHARGE DATASET (20 RECORDS) Columns: patient_id | admit_date | discharge_date | LOS_calendar | LOS_billing | discrepancy_flag | notes Row 01: PT-1001 | 2027-01-02 | 2027-01-05 | 3 | 3 | NO | Match Row 02: PT-1002 | 2027-01-04 | 2027-01-06 | 2 | 2 | NO | Match Row 03: PT-1003 | 2027-01-05 | 2027-01-08 | 3 | 4 | YES | LOS_billing overstated by 1 day Row 04: PT-1004 | 2027-01-06 | 2027-01-10 | 4 | 4 | NO | Match Row 05: PT-1005 | 2027-01-07 | 2027-01-11 | 4 | 3 | YES | LOS_billing understated by 1 day Row 06: PT-1006 | 2027-01-09 | 2027-01-12 | 3 | 3 | NO | Match Row 07: PT-1007 | 2027-01-10 | 2027-01-15 | 5 | 5 | NO | Match Row 08: PT-1008 | 2027-01-11 | 2027-01-13 | 2 | 4 | YES | LOS_billing overstated by 2 days — high risk Row 09: PT-1009 | 2027-01-12 | 2027-01-16 | 4 | 4 | NO | Match Row 10: PT-1010 | 2027-01-14 | 2027-01-17 | 3 | 3 | NO | Match Row 11: PT-1011 | 2027-01-15 | 2027-01-20 | 5 | 5 | NO | Match Row 12: PT-1012 | 2027-01-16 | 2027-01-18 | 2 | 2 | NO | Match Row 13: PT-1013 | 2027-01-17 | 2027-01-22 | 5 | 6 | YES | LOS_billing overstated by 1 day Row 14: PT-1014 | 2027-01-18 | 2027-01-21 | 3 | 3 | NO | Match Row 15: PT-1015 | 2027-01-19 | 2027-01-24 | 5 | 5 | NO | Match Row 16: PT-1016 | 2027-01-20 | 2027-01-22 | 2 | 2 | NO | Match Row 17: PT-1017 | 2027-01-21 | 2027-01-25 | 4 | 4 | NO | Match Row 18: PT-1018 | 2027-01-22 | 2027-01-24 | 2 | 1 | YES | LOS_billing understated by 1 day Row 19: PT-1019 | 2027-01-23 | 2027-01-28 | 5 | 5 | NO | Match Row 20: PT-1020 | 2027-01-24 | 2027-01-27 | 3 | 3 | NO | Match --- DISCREPANCY SUMMARY Total records: 20 Records with match: 14 (70%) Records with discrepancy: 6 (30%) — IDs: PT-1003, PT-1005, PT-1008, PT-1013, PT-1018 [correction: also PT-1018] Discrepancy breakdown: LOS_billing overstated (potential overbilling concern): 3 records (PT-1003, PT-1008, PT-1013) LOS_billing understated (potential underbilling / revenue loss): 2 records (PT-1005, PT-1018) High-risk discrepancy (>1 day difference): 1 record (PT-1008 — overstated by 2 days) Overall discrepancy rate: 30% High-risk discrepancy rate: 5% (1/20) --- ANALYSIS NOTES FOR MAYA'S AUDIT MEMO A 30% discrepancy rate in LOS fields exceeds the typical acceptable threshold for well-maintained hospital data (usually <5%). This dataset should be treated as a data quality alert requiring investigation at the source system level. The HIM team should: 1. Pull the original admission and discharge records for PT-1003, PT-1005, PT-1008, PT-1013, and PT-1018. 2. Determine whether the LOS_billing field is populated by an automated system or manual entry — manual entry is more likely to cause single-day overstatements. 3. Escalate PT-1008 (2-day overbilling) to the compliance officer for review under the simulated Lakeview General billing compliance policy. 4. Recommend a system audit to identify whether a date calculation logic error is producing consistent +1 or -1 discrepancies. 5. Document findings in the Data Quality Incident Log. --- LEARNING OBJECTIVES (HIM 125 Alignment) 1. Understand the components of a discharge dataset and the relationship between clinical and billing data fields. 2. Apply a systematic discrepancy identification process to tabular health data. 3. Distinguish between overstatement and understatement discrepancies and their different risk profiles. 4. Understand when a discrepancy requires escalation to compliance vs. routine correction. 5. Practice writing a concise data quality audit memo. --- DISCUSSION QUESTIONS 1. Why is a 30% discrepancy rate significant, and how does it compare to typical data quality benchmarks? 2. What is the difference between a data entry error and a system calculation error — and why does that distinction matter for fixing the problem? 3. PT-1008 has a 2-day overbilling discrepancy. What is the potential financial and regulatory impact if this record were submitted to Medicare without correction? 4. Should Maya report PT-1008 to the compliance officer before or after completing the full dataset review? Justify your answer. 5. How might this dataset look different if LOS_billing were populated by an automated system tied directly to the EHR's admission/discharge module? --- INSTRUCTOR NOTES (DO NOT REPRODUCE ON STUDENT-FACING PAGE) This case maps to HIM 125 Week 9: Data Integrity and Quality Auditing. The simulated Lakeview General Hospital does not exist. The 30% discrepancy rate is intentionally high to make the exercise clearly meaningful; in a more advanced course, students could be given a dataset with a subtler error pattern. PT-1008 is the "high-stakes" record and is typically the focal point of class discussion. The dataset can be extended to 40 or 50 records by instructors who want a larger analysis exercise. Expected discussion time: 30–40 minutes. --- SOURCE DOCUMENT NOTES This document is a curricular clinical simulation exercise created for the Velmoor College of Health Sciences website. All patient IDs, dates, hospital names, and data values are simulated. Created for clinical instructional simulation purposes.