The AIHW collects data against a set of national Key Performance Indicators (nKPIs) from primary health care organisations that provide health care to Aboriginal and Torres Strait Islander Australians. The nKPI data are collected every 6 months, reported back to organisations at the individual service level and compiled for national reporting purposes. The raw data received from health organisations are carefully checked to identify any data quality issues and are corrected, in consultation with services, through an ‘exception reporting’ process before the data are used in any type of reporting. This working paper identifies the most common data quality issues that lead to the issuing of exception reports, and suggests a number of options to improve the data collection process that will reduce the number of exception reports issued.

Overview

The aim of this paper was to:

• identify the most common data quality issues in the nKPI data collection
• identify the reasons for these issues
• provide options for consideration to reduce the number of exception reports.

Exception reports are issued when data quality problems are identified in data submitted to the AIHW from health services. Exception reports are a mechanism through which health services can manually correct their nKPI data by enabling services to resubmit data. The issuing of exception reports and the resubmission process adds to the time taken to finalise data submissions and increases the workload for health services because of complex data checking, investigation and resubmission.

This paper examines the exception reporting process through:

• the extent of data quality issues and the number of exception reports issued over time
• the characteristics of health services that have data quality issues and exception reports
• the main reasons for exception reports
• the indicators that cause the most issues for health services.

While the AIHW has collated and analysed data on the issues outlined above, this paper does not cover all the possible factors that can contribute to data quality problems, for example, issues caused by inexperienced staff, errors in data entry or software faults. This analysis is based on a data validation process (which ensures quality data is collected) and comments provided by health services.