Integrated health services, integrated data sets, what comes first?

Transcription

1 Integrated health services, integrated data sets, what comes first? 23 rd PCSI Conference, Lido, Venice Lisa Fodero & Joe Scuteri

2 Introduction Integrating health services will not only improve patient outcomes but will also result in more cost efficient care by eliminating service duplication and redundancy. Service integration implies a patient centred approach to the collection of data that can be shared by multiple providers in multiple health care settings. Most healthcare providers capture significant data on patients and treatments provided, but few have the ability to share that information with other providers for the benefit of the patient. The data that are collected by providers are largely not standardised and therefore difficult to share.

3 Introduction More than the 60 discretely funded programs form the Australian health care system including: federally funded programs including the Medicare Benefits Scheme (MBS) and Pharmaceutical Benefits Schemes (PBS). programs such as Public Hospitals and the Home and Community Care Program (HACC) that are jointly funded by the Federal and state governments. disease specific programs such as palliative care, mental health, breast cancer screening etc that are funded typically by the Federal Government. Each of these program have different accountability requirements and, as a result, different data sets are collected as part of the process of providing care and to account for the use of funds.

4 Introduction There have been national attempts to standardise data collection through the development of the National Health Data Dictionary (NHDD). Often the data elements defined in the national dictionary do not meet the specific needs of a program, so new program specific data element definitions and associated data domains are developed. Results in an increase in the data collection effort for service providers and inconsistencies in the available data making interpretation and analysis across data sets difficult, if not impossible.

5 Community Health and Outpatient Care Information Project Project undertaken by NSW Health Department. Aims to develop a patient level data collection across all community health and outpatient care settings in the State. Largest project of its kind ever attempted in Australia. Annual collection of approximately 25 million patient level records describing the characteristics of the patients treated and the nature of services provided in the community health and outpatient care settings.

6 Community Health and Outpatient Care Information Project Need for the development of a patient level data collection has built up in recent years for a number of reasons: 1) Community health and hospital outpatient services account for a large and growing proportion of the workload of the NSW health care system, yet little is known about the nature of the services provided and about the patients receiving those services, their ongoing needs and future demands. 2) There has been a considerable shift in the mix of outpatient and sameday patient services largely due to funding incentives which has resulted in information loss on services provided (detailed unit record data are collected on sameday patients but only aggregate data are collected on outpatients).

7 Community Health and Outpatient Care Information Project 3) Although the current Australian Health Care Agreements (between Federal and State Governments) only require the reporting of outpatient data at aggregate level, it is anticipated that the next round of AHCAs will require the collection of patient level data for outpatient care. 4) The availability of enterprise systems such as Cerner and ipm in hospitals and CHIME in community health that have patient scheduling modules has made collection of patient level data for outpatient and community health services more feasible.

8 Scope of CHOCIP Types of NSW public sector services required to report to the CHOCIP include: Public hospitals (including public dental and public psychiatric hospitals), covering: Outpatient medical and nursing services; Outpatient allied health services; Outpatient day procedures; and Outreach services. Public community health services: Centre/campus based services; Home based services; Mobile/Outreach services; Multi-purpose services; Mothercraft services; and Community acute and post acute care services (other than Hospital in the Home service that are reported to the Admitted Patient Data Collection). Justice Health services; and Health One services.

9 Overview of CHOCIP CHOCIP began in 2006 Three phases of CHOCIP: Phase One: Where are we now? Where do we want to be? How do we get there? Phase Two: Infrastructure development (current phase) Phase Three: The roll out (expected start date 1 st September 2008)

10 CHOCIP : Phase Two 13 Project and Data Collection Governance 1.1 Data dictionary 1.2 Business Rules What: Finalise Minimum Data Set (produce Data Dictionary) and associated business rules Why: To standardise and detail data requirements and build system specifications. To standardise interpretation and application of rules in implementing Data Collection When: Start Dec 06, End Mar 08 Who: Project Team, Consultancies 2 Reporting Entity Registration What:Identify data requirements, build a system to collect data, register all clinics/service teams Why: To register and uniquely identify all reporting entities, standardise what is meant by reporting entity When: Start Jan 07, End Dec 07 Who: Project Team, Consultancy 3 Electronic Client Registration What: Mandate electronic Area-level registration of ambulatory clients Why: Uniquely identify each client at the Area level, improve data integrity, minimise data entry burden When: Start Jan 07, End Oct 07 Who: Project Team, Consultancy 4.1 Modifications to Enterprise App. 4.2 Data Extracts from Enterprise App. What: Create State Base Build of ambulatory booking/data collection modules. Implement automated data extracts from these applications Why: Standardise data collection When: Start Jul 07, End Jun 08 Who: SIM, HT, Health Services, Project Team, Vendors/Contractors 5.1 Modifications to Other Source App. 5.2 Data Extracts from Other Source App. What: Implement system modifications to comply with Minimum Data Set. Implement automated data extracts from these applications Why: Standardise data collection When: Start TBD, will extend post Jan 2008 Who: SIM, HT, Health Services, Project Team, Vendors/Contractors 6 Data Extracts for Ancillary Services What: Develop cost effective strategy for ensuring data on ancillary services is available Why: To enable efficient collection of ancillary data When: Start Jan 08, Strategy completed by end Mar 08, Implement strategy post Sep 08 Who: Project Team, SIM, Health Technology 7 Alternative Data Collection Tools What: Select and develop webbased and paper based solutions Why: To enable collection of data across all services When: Start Jan 08, End Jun 08 Who: Project Team, SIM, HT, Consultancy/contractors 8 Proof-of-Concept What: Testing of alternative data collection tools and performance reports in selected sites Why: To fine tune the tools and identify any further issues for implementation of the Collection When: Start Mar 08, End Jun 08 Who: Relevant Health Services, Project Team, Consultancy 9 Data Repository What: Create a repository for incoming data and associated reference tables Why: Securely store data and make it available for analysis When: Start Mar 07, End Aug 08 Who: BIS Program Office, Project Team, SIM, Health Services, DPE, Consultants, Independent Testers 10 Performance Reports What: Develop reports relevant to key stakeholders that can be automatically generated once data are available Why: Data available for clinicians, managers and for assessment of data quality When: Start Aug 07, End Aug 08 Who: Project Team, DPE 11 Change Management, Training, Communication 12 Health Service Implementation Plans

11 Project Methodology The project methodology consisted of five major processes: 1) Review of the proposed MDS to ensure that it would produce the information required to meet the project objectives. 2) Review of a series of 11 data dictionaries for related data collections to extract data element definitions and data domains for data elements that were included in the CHOCIP MDS. 3) Preparation of draft data dictionary for distribution to stakeholders as the basis of a series of workshops to gather input on the most suitable specification of the data elements for the purposes of CHOCIP. 4) Analysis of the consultation findings to prepare the final data dictionary. 5) Preparation of a series of mappings for each data element in the CHOCIP Data Dictionary with the entries in the 11 data dictionaries.

12 Reviewed data dictionaries National Health Data Dictionary Version 12; NSW Health Data Dictionary Version 1.2; NSW Health Drug and Alcohol Data Dictionary Version 5.0; NSW Health Emergency Department Data Dictionary Version 3.2; NSW Health Oral Health Data Dictionary Version 1.4; NSW Health Admitted Patient Data Collection Data Dictionary Version 1.0; NSW Health Sexual Assault Data Dictionary Version 1.0; NSW Health CHIME Data Dictionary of Classifications; NSW Mental Health Data Dictionary Version 3.0; Home and Community Care Data Dictionary Version 2.01; and Hunter Area Health Service Allied Health Data Dictionary (AHMIS).

13 Part One Data Dictionary analysis 28 data elements in Part One of CHOCIP Data Dictionary 17 data elements included in analysis which had specific data domains or codes excluded address, postcode and date data elements

14 Analysis of data element entries in the 11 reference data dictionaries Data Element Number of Data Dictionaries containing data element Number of times most common entry occurs Number of different entries Most common entry used for CHOCIP Sex Yes Aboriginal and Torres Strait Islander origin No Country of birth Yes Disposition status No Preferred language Yes Source of referral No Billing category No DVA card type Yes Estimated date of birth flag No Interpreter required No Service delivery setting Yes Discipline of individual service provider(s) No Service contact mode Yes Outcome of offer No Group or individual service indicator No Anonymous/unidentified client indicator No Initial or subsequent service indicator No

15 Analysis of data element entries in the 11 reference data dictionaries Data Element Number of Data Dictionaries containing data element Number of times most common entry occurs Number of different entries Most common entry used for CHOCIP Sex Yes Aboriginal and Torres Strait Islander origin No Country of birth Yes Disposition status No Preferred language Yes Source of referral No Billing category No DVA card type Yes Estimated date of birth flag No Interpreter required No Service delivery setting Yes Discipline of individual service provider(s) No Service contact mode Yes Outcome of offer No Group or individual service indicator No Anonymous/unidentified client indicator No Initial or subsequent service indicator No

16 Analysis of data element entries in the 11 reference data dictionaries Data Element Number of Data Dictionaries containing data element Number of times most common entry occurs Number of different entries Most common entry used for CHOCIP Sex Yes Aboriginal and Torres Strait Islander origin No Country of birth Yes Disposition status No Preferred language Yes Source of referral No Billing category No DVA card type Yes Estimated date of birth flag No Interpreter required No Service delivery setting Yes Discipline of individual service provider(s) No Service contact mode Yes Outcome of offer No Group or individual service indicator No Anonymous/unidentified client indicator No Initial or subsequent service indicator No

17 Analysis of data element entries in the 11 reference data dictionaries Data Element Number of Data Dictionaries containing data element Number of times most common entry occurs Number of different entries Most common entry used for CHOCIP Sex Yes Aboriginal and Torres Strait Islander origin No Country of birth Yes Disposition status No Preferred language Yes Source of referral No Billing category No DVA card type Yes Estimated date of birth flag No Interpreter required No Service delivery setting Yes Discipline of individual service provider(s) No Service contact mode Yes Outcome of offer No Group or individual service indicator No Anonymous/unidentified client indicator No Initial or subsequent service indicator No

18 Analysis of the use of reference data dictionaries for the purposes of CHOCIP Data Element Number of Data Dictionaries containing data element Defined in National Health Data Dictionary Defined in NSW Health Data Dictionary Entry used Aboriginal and Torres Strait Islander origin 10 Yes Yes NSW Anonymous/unidentified client indicator 1 No No N/A Billing category 7 Yes No Neither Country of birth 10 Yes Yes Both Discipline of individual service provider(s) 4 Yes No Neither Disposition status 9 Yes No Neither DVA card type 7 No Yes NSW Estimated date of birth flag 6 Yes No Neither Group or individual service indicator 2 No No N/A Initial or subsequent service indicator 1 Yes No Neither Interpreter required 6 Yes No Neither Outcome of offer 3 No No N/A Preferred language 9 Yes Yes NSW Service contact mode 4 Yes Yes Neither Service delivery setting 6 Yes No Neither Sex 11 Yes Yes Both Source of referral 9 Yes No Neither

19 Scenario: Impact of multiple data sets Mr Smith, a 70 year old man is hit by a car as he was crossing the road. He sustains a broken left leg and left arm. He is taken by ambulance to the emergency department and treated there, but it is determined that the broken leg needs to be pinned. He is admitted immediately as an inpatient and has surgery the following day. He stays two days in hospital where discharge planners determine that he requires physiotherapy as well as occupational therapy to assess the suitability of his home environment given his restricted mobility. He also requires home nursing to assist with activities of daily living. He is required to return to outpatients two weeks after discharge to consult the orthopaedic surgeon. He finds it difficult to cope with his reduced mobility and inability to look after himself which results in depression requiring the assistance of a psychologist.

20 Analysis of the services received by Mr Smith and the associated data collections Service area Data Set Emergency Department in hospital Emergency Department Data collection Admitted patient unit in hospital Admitted Patient Data collection Hospital based allied health departments Allied Health Data collection Outpatient unit (orthopaedic surgeon) Aggregate outpatient statistics Home nursing service Home and Community Care Data Collection Community mental health team Mental Health Data Collection

21 How CHOCIP addresses these issues? Attempting to standardise the definitions and data domains for all elements in the MDS for community health and outpatient settings. The intention is to rationalise existing reporting requirements across these programs. This will mean that one standarised MDS will be reported and supplementary information for the specific program area will be the only additional data to report. The data collections flagged for integration include: Home and Community Care Data Collection; Drug and Alcohol Data Collection; Mental Health Outcomes and Assessment Tools Data Collection; Allied Health Data Collection; and Chronic Care Performance Indicators.

22 How CHOCIP addresses these issues? In the case of Mr Smith, this development will mean that four of the six collections that collect unit record data on the services he receives will collect it according to standardised and consistent definitions for the common data elements. It also means the common data elements will only be collected once to fulfil the requirements of all these data collections. The Part 1 Data Dictionary project also worked on standardising data definitions with the Emergency and Admitted Patient data collections although there are no current plans to integrate these collections into CHOCIP. Thus the CHOCIP work will contribute to the solution of, but not resolve all the current problems.

23 Conclusion The process of data set definition was found to follow the process of program management in Australia. That is, most programs are funded in silos, they have their own program eligibility criteria, often their own funding models and as a result their own data sets and associated data collections. This phenomenon exists notwithstanding the stated aims of integrating the services for patients of the health system to ensure continuity of care.

24 Conclusion The CHOCIP data dictionary development project has produced a data dictionary that can be applied across the broad range of services covered by the scope of community health and outpatient care. To date, this dictionary defines 28 data elements and their associated data domains. It will be used to support the implementation of the CHOCIP MDS from 1st September It will also be reviewed as part of the Part 2 Data Dictionary Project to include data elements that describe more complicated concepts such as diagnosis and intervention.

25 Lessons learnt The data dictionary exercise has demonstrated that developing integrated health services requires the development of integrated health data sets. It is not a matter of what comes first; the reality is that one is not possible without the other. Without consistent and comparable data across the range of services delivered in community health and outpatient care, it is impossible to identify gaps in service delivery and discontinuity in patient journeys. Data needs to be consistently collected and shared amongst service providers. CHOCIP represents an important step in the process of integrating data sets in support of developing integrated health services. The next challenge will be the analysis of the resultant data to improve methods of service delivery and funding thereby resulting in improved continuity of care for patients of the health system.

26 Thank you NSW Health and NSW Health CHOCIP Project Team Deniza Mazevska Brendan Ludvigsen David Baty Durham Bennett