COVID-19 Ensuring accuracy of coded data
Helen Nolan a
Introduction
The Hospital Inpatient Enquiry System (HIPE) is managed by the Healthcare Pricing Office (HPO) within the Health Service Executive (HSE). It is responsible for the collection and reporting of hospital inpatient activity data for all public acute hospital inpatient and day case discharges and support and training for the 300 HIPE clinical coding staff. The HPO's coding team manage HIPE data quality and audit activities for approximately 1.7 million discharges per year.
The first case of COVID-19 was reported in Ireland on 28th February 2020. Fast access to accurate COVID-19 data was required to track, monitor and support the health system. Patients with COVID-19 had to be coded within 48 hours of discharge. Due to the pandemic chart based audits carried out by the HPO ceased with desk top reviews performed instead. HIPE data on COVID-19 was widely used across the health service involving more scrutiny of data.
The work carried out by the HPO to ensure that all COVID-19 cases were accurately coded as needed for analysis by Department of health and other agencies is described below.
Methods
New ICD-10-AM codes were released by the Independent Hospital Pricing Authority (IHPA) to enable the collection of COVID-19 data. Several updates of codes and guidelines were issued throughout 2020 and 2021. Each update and change had to be reviewed by the HPO and communicated clearly, efficiently and quickly to the HIPE coders. Edits on the national HIPE data entry system were developed to support coders.
Reviews of COVID-19 cases included the following:- Ventilation:
- Continuous Ventilatory Support (CVS) with 0 CVS hours reported or with 0 ITU days
- CVS hours reported with no CVS procedure and 0 ITU days
- High % of clinically diagnosed COVID-19 versus Lab confirmed cases
- Patients with short LOS with HADx (hospital acquired diagnosis) flag assigned.
- Unspecified pneumonia code versus viral pneumonia code
Queries were issued to each hospital and followed up on promptly. Results highlighted areas where further education was needed and resources such as education sessions on COVID-19 were provided. Frequently asked questions were issued in Coding Notes (issued quarterly).
Results
Data needed to be checked regularly and queries sent to hospitals resulting in some needing correction. As HIPE data was used by a number of agencies there was more awareness and scrutiny of the data with the need to get it right first time.
Conclusions
Desktop reviews of COVID-19 data highlighted areas for review and some inaccuracies of data, a true indication of the coding quality will only be available by chart based audit. We will recommence audits shortly and may identify areas in coding accuracy not noticed by desktop review.
Challenges were identified in the reporting on cases admitted with COVID-19 versus hospital acquired. It is difficult to code hospital acquired COVID without supporting clear documentation and the hospital acquired cases reported on HIPE may not be a true indication of the actual numbers.
a Healthcare Pricing Office, Ireland
Original Version in PDF