TRUST BOARD MEETING JUNE Data Quality Metrics

Transcription

1 a b c Agenda Item: 5 TRUST BOARD MEETING JUNE 2 Data Quality Metrics PURPOSE: Following the recent publication of the Trust s new Information Strategy, it was agreed that the improvement in standards would be measured by a series of metrics reported monthly to the Finance and Performance Committee. At the launch of this process, the new metrics are presented to the Trust Board for information and comment. Previously considered by the Finance & Performance Committee on 6 June 2 Implications: Objectives to which issue relates: Risk Issues: Financial: HR: Healthcare/ National Policy: Legal Issues: Equality Issues: All main objectives Key financial and performance risks associated with below par data and information Incomplete and/or inaccurate coding will have a financial consequence Limited relevance Links to reporting of performance against national targets and the information standards set within the new national contract N/A N/A RECOMMENDATIONS: The Trust Board is asked to: Note the new metrics used to track the progress in improving data quality standards Propose any further changes to the content and format of reporting DIRECTOR: Director of Finance PRESENTED BY: Director of Finance AUTHOR: Head of Information DATE: 22 nd June 2

2 January 28 to March 2 This April Performance Last Year Expected Var % YTD Performance 29/ This Year Expected Var % Month Valid NHS Numbers Coding Accuracy (not yet available) Hospital Deaths (not yet available) A&E Clinical Data CDS Completeness Uncoded Spells Coding Backlog Missing TCI Data (not yet available) Defined Objects (not yet available) Duplicate Case Notes (not yet available) Average Coding Depth Availability of Health Records for Clinic 8-Week Pathway Validation Data Warehouse Load Completion Uncashed Appointment Latency Coded Epsiode Latency Variation in nominal SLA Revenue s Admission Latency Valid Patient Demographics Valid Ethnic Category Codes Valid Postcodes 98.8% 99.% 99.2% -.% Valid NHS Numbers 99.% 99.% 99.2% -.% % Coding Accuracy (not yet available) % % Hospital Deaths (not yet available) % 37.3% 37.7% 39.8% -5.2% A&E Clinical Data 42.% 37.7% 39.8% -5.2% % CDS Completeness 92.% % 4 - Uncoded Spells 37-3,577 3,385 3,54 3.7% Coding Backlog 6,443 3,385 3,54 3.7% Missing TCI Data (not yet available) Defined Objects (not yet available) Duplicate Case Notes (not yet available) % Average Coding Depth % % 98.5% -.2% Availability of Health Records for Clinic 97.8% 98.3% 98.5% -.2% - 8.% 75.4% 7.4% 8-Week Pathway Validation 78.% 8.% 75.4% 7.4% 7.4% 9.% 82.4% 9.2% Data Warehouse Load Completion 84.% 9.% 82.4% 9.2% - 9.6% 9.3%.4% Uncashed Appointment Latency 9.3% 9.6% 9.3%.4% 45.7% 9.8% 95.% -3.3% Coded Epsiode Latency 73.% 9.8% 95.% -3.3% Variation in nominal SLA Revenue s % 64.% 65.2% -.8% Admission Latency 64.% 64.% 65.2% -.8%.% 99.5% 99.9% -.4% Valid Patient Demographics 99.9% 99.5% 99.9% -.4% 63.4% 6.6% 63.9% -5.2% Valid Ethnic Category Codes 6.6% 6.6% 63.9% -5.2% 99.% 99.4% 99.5% -.% Valid Postcodes 99.2% 99.4% 99.5% -.% Publish Date: 26/5/2 Status: Published v. Report Location: S:\2 Information Team\Plans\Information Strategy\Current\ Business Sponsor: Paul Traynor Information Team Lead: Mathew Towers

3 Accuracy Page 2 of 3 Fig. Valid NHS Numbers Fig.2 Coding Accuracy % % 98% 96% 96% 92% 94% 88% 92% 84% 9% 8% 2 A composite indicator for the percentage of outpatient appointments, admitted patients, and A&E attendances with verified NHS Numbers within the monitored period. The target will be set in line to exceed external benchmarks. The data is collected monthly. The percentage of epsiodes without a coding error resulting in a different HRG as identifed through CfH Approved Audit Methodology. The target will be set in line to exceed external benchmarksthe data is collected annually - but will move to quarterly collections from January 2. The presence of a validated NHS Number for a patient record confirms the accuracy of locally held demographic data with the National Spine. This is primarily a Clinical Risk issue. The Audit Commission conduct an annual audit as part of their PbR Data Assurance Framework. Higher accuracy will privode our PCT commissioning colleagues with greater confidence in the accuracy of our SLA invoicing.this is primarily a Compliance issue. Data Services 2/ Workplan item for the Patient Systems Department Data Services 2/ Workplan item for the Clinical Coding Department

4 Accuracy (continued) Completeness Page 3 of 3 Fig.3 Hospital Deaths Fig.4 A&E Clinical Data % % 9% 8% 8% 6% 7% 4% 6% 5% 2% 4% % 3 4 The percentage of attribution errors for Treatment Specialty and Responsible Consultant identified by clinicians on review of hospital deaths report The data is collected quarterly - but will move to monthly collections from April 2. A composite indicator for the percentage completeness of A&E coding for investigations, treatments, diagnoses and outcomes in the monitored period. The target trajectory (currently under review) has currently been set to return to pre-ipm PAS levels for this financial year. The data is collected daily. Ensuring that accurate data is held on the Clinical Outcomes Benchmarking analysis tool provided by Dr Foster. This is primarily a Productivity issue. Completeness is this area is currently very low and provides little clinical information on the interventions occuring in A&E. It is likely that this information will be required in 2/2 for income recovery. This is primarily a Financial issue. Coding Review Group Not yet defined

5 Completeness (continued) Page 4 of 3 Fig.5 CDS Completeness Fig.6 Uncoded Spells.% 2 98.% 8 96.% 6 94.% 4 92.% 2 9.% 5 6 A composite indicator for the percentage completeness of all data items sent to SUS in the various Commissioning Data Sets The data is collected monthly. The percentage of uncoded spells held on SUS at PbR Freeze Date. This will be assessed against the total number of spells held on SUS for the equivalent period in the month-end views. The data is collected monthly. Commissioning Data Sets are sent to SUS on a regular basis and are used for Commissioning reconciliation. The level of completeness of mandatory data items is reviewed by our host PCT. This is primarily a Compliance issue. Uncoded Spells are a direct financial loss to the Trust. This is primarily a Financial issue. Commissioning Information Group (CIG) Data Services 2/ Workplan item for the Clinical Coding Department

6 Completeness (continued) Page 5 of 3 Fig.7 Coding Backlog Fig.8 Missing TCI Data The number of spells still to be coded at the end of the month. A spell is considered to be part of the backlog on the fourth day following discharge. The data is collected daily. The percentage of patients recorded in outpatients as being added to an elective waiting list where there is no waiting list record. The data is collected weekly. A large backlog of uncoded spells creates more uncertainty over the accuracy of monthly financial forecasting and planning assumptions. This is primarily a Financial issue. This is important to track patient pathways and to minimise administrative delays in the system. This is primarily a Productivity issue. Data Services 2/ Workplan item for the Clinical Coding Department Access and Performance Management Group (APMG) Data collection arrangements are currently being finalised

7 Relevance Page 6 of 3 Fig.9 Defined Objects Fig. Duplicate Case Notes 9 The percentage of data items within the Data Warehouse that have a Data Dictionary definition accessible to the end user. The data is collected quarterly. A composite indicator to track the number of temporary and duplicate sets of Health Records in circulation. The data is collected monthly. As the Data Warehouse is redeveloped, a key improvement will be to provide a data dictionary to minimise confusion. This is particularly important for clinical (and any other) staff who have limited experience with the NHS Data Model. This is primarily a Productivity issue. This indicator will address an NHS Information Governance Standard and a recommendation by the Audit Commission to improve the quality of Health Records in use. This is primarily a Compliance issue. Data Services 2/ Workplan item for the Information Department Not yet defined This indicator will be withdrawn until data object lineage is developed within the data warehouse and reporting structures. This indicator will be withdrawn until the information is made available within the data warehouse.

8 Relevance (continued) Reliability Page 7 of 3 Fig. Average Coding Depth Fig.2 Availability of Health Records in Outpatients 4.5 % % % %. 92%.5. 9% 2 The average number of diagnosis codes added for each epsiode of care. The target will be set in line to exceed external benchmarks. The data is collected monthly. A composite indicator for the percentage of outpatient appointments for whom Health Records were available at the point of contact. The target will be set at a locally agreed level. The data is collected monthly. Higher levels of diagnosis coding better reflects the clinical workload within the Trust. This may have an beneficial impact on quality indicators such as HSMR as well as income recovery. This is primarily a Productivity issue. The availability of the patient's health record at the point of contact with the patient is essential to minimise the clinical risk in embarking on a course of treatment. This is primarily a Clinical Risk issue. Coding Review Group Senior Managers' Meeting (SMT)

9 Reliability (continued) Page 8 of 3 Fig.3 8-Week Pathway Validation Fig.4 Data Warehouse Load Completion % % 9% 9% 8% 8% 7% 7% 6% 6% 5% 5% 4% 4% 3 4 The number of records which do not require waiting time adjustments to be made by the 8-Week Validation Team divided by the total number of reviewed patient pathways. The business impact is under review and the target and the trajectory may be subject to revision. The data is collected weekly. The percentage of successful daily loads to the data warehouse available to the information consumer in the monitored period. The data is collected daily. The 8-Week Validation Team make significant retrospective amendments to the patient pathway records. This adds to the administrative overhead yet does not fully address the risk of non-compliance with the NHS Waiting Time Standard. This is primarily a Compliance issue. The Data Warehouse provides a number of operational reports used daily to manage business processes. Delays in the provision of information cause productivity issues and potentially impact patient experience. This is primarily a Productivity issue. Not yet defined Data Services 2/ Workplan item for the Information Department

10 Timeliness Page 9 of 3 Fig.5 Uncashed Appointment Latency Fig.6 Coded Epsiode Latency % % 9% 9% 8% 8% 7% 7% 6% 6% 5% 5% 4% 4% 5 6 The percentage of appointments without administrative and clinical outcomes at the end of the day as a percentage of total appointments in the monintored period. The data is collected daily. The percentage of episodes coded within the PbR Inclusion Date. This is around 5 working days after the month-end The data is collected monthly. Although there is the potential for financial loss if appointments are not 'cashed up', patient pathways are more difficult to manage if the clinic outcomes are not entered in good time. This is primarily a Clinical Risk issue. Our host PCT have set a target at 95%. This is primarily a Compliance issue. Access and Performance Management Group (APMG) Commissioning Information Group (CIG)

11 Timeliness (continued) Page of 3 Fig.7 Variation in nominal SLA Revenue Fig.8 Admission Latency 25 % 2 9% 5 8% 7% 6% 5 5% 4% 7 8 The percentage change in the nominal SLA Revenue total from the PbR Freeze date position - currently calculated for APC activity only. s The data is collected monthly - 2 months in arrears. The proportion of admissions added to PAS on the day of the admission as a pecentage of total admissions in the monitored period. The target will be set at a locally agreed level. The data is collected monthly. This is a indication of the financial impact that the quality of data is having on the income recovery for the Trust. This is primarily a Financial issue. Late data entry on PAS reduced the available time for the Clinical Coding Department to accurate code a patient record. This is primarily a Productivity issue. Not yet defined Not yet defined

12 Validity Page of 3 Fig.9 Valid Patient Demographics Fig.2 Valid Ethnic Category Codes.% % 99.8% 9% 99.6% 8% 7% 99.4% 6% 99.2% 5% 99.% 4% 9 2 A composite indicator for the overall percentage of admitted patients with a valid Date of Birth, valid forename and valid surname in the monitored period. The data is collected monthly. The percentage of admitted patients with valid Ethnic Category codes within the monitored period. Valid codes are those defined by CQC and exclude 'Z' for not stated responses. The data is collected monthly. Without a valid date of birth and name details, the patient will not be matched to the NHS Spine. This is primarily a Clinical Risk issue. Included in the CQC Performance Ratings. Historically, Trust performance has lagged behind the national benchmark. This is primarily a Compliance issue. Data Services 2/ Workplan item for the Patient Systems Department Data Services 2/ Workplan item for the Patient Systems Department

13 Validity (continued) Page 2 of 3 Fig.2 Valid Postcodes.% 99.5% 99.% 98.5% 98.% 2 The percentage of admitted patients (excludes Overseas Visitors) with a postcode that conforms to a standard UK postcode pattern. The data is collected monthly. Invalid postcodes may indicate that the patient's address is incorrect which could lead to correspondance going astray. Higher DNA Rates could be a result of poor hospital communication. This is primarily a Productivity issue. Data Services 2/ Workplan item for the Patient Systems Department

14 Quality Assurance Page 3 of 3 Have the data series date ranges been amended for all graphs? n/a Has the report had management sign-off for distribtion and publication? Have the graph titles been amended to show the correct date range as necessary? n/a Is the report distribution list up to date? n/a Has the commentary been updated for all KPIs? Are there any outlying data points requiring further investigation? Are there any breaches of standards that require management attention? Does the report view on screen as expected? Does the report print out as expected? Number of Pages 3