 |
|
|
|
|
|
Quality of Data
There are important improvements in data quality in the 1997/98 edition of the NSW Public Hospital Comparison Data Book which reflect a year long effort at Area Health Service and Health Department levels. NSW Health continues its program to constantly improve the overall quality of data in the various collections with data quality and timeliness measures continuing to be incorporated into the performance contracts between NSW Health and Area Health Service CEOs.
Area Health Services have been closely consulted, with five representatives on the Editorial Committee. This ensured Areas had early input into the overall content and layout of the 1997/98 edition.
Area Health Services provide the data for the Data Book for each of their public hospitals. As in previous years data quality varied between the Areas and between collections. A rigorous process of data verification and checking was undertaken with the Areas to improve the quality and resolve inconsistencies.
The preliminary tables of information drawn from the six source data collections were reviewed by the Casemix Area Network which includes Area Health Service staff responsible for casemix issues. The Network members helped identify incomplete, inaccurate and inconsistent information. Their feedback was then incorporated into a second iteration of tables which were further reviewed by the group.
Individual tables of information were then produced on an Area Health Service basis and circulated to Areas for comment. Feedback was then incorporated into the finalised tables.
The Inpatient Statistics Collection is a key data source. This year output editing was conducted throughout the year, in addition to being done at the end of the collection’s processing cycle, resulting in greatly improved data quality from that collection. As in 1996-97 there was no significant under-reporting of data in the ISC for 1997-98.
As with previous years the most substantial data quality issues arose in the Unaudited Annual Return (UAR) where problems arose in reconciling overall expenditures reported in the Audited Financial Statements with the UAR data. The inclusion of data from the Hospital Costs Data Collection has greatly improved the accuracy and reliability of expenditure information.
Data quality issues should still be taken into consideration, especially by readers examining measures relating to costs. If the measures substantially differ from the peer group figures, then readers should review the components of the data which define the measure (for example the breakdown of expenses by program). Where data are known to be deficient, it is indicated as ‘na’ in the tables.
 |
||
 |
||
 |
||
|
||
 |
||