Understanding and promoting good outcomes PROMs in the Best Practice Tariff for hip and knee replacement Jake Gommon (Pricing Team, NHS England) & Rafael Goriwoda (Patient & Information analytical team, NHS England) 3 rd December 2014
Contents 1 - National Tariff and Best Practice Tariffs (BPTs) 2 - PROMs in the Best Practice Tariff - from patient responses to tariff decision 3 - Beyond payment using PROMs to improve patient outcomes
National tariff and best practice tariffs
National Tariff The National Tariff specifies the rules and rate of financial reimbursement for hospital activity. 2012 Health and Social Care Act transferred responsibility for National Tariff from DH to NHS England and Monitor. 2014/15 first published tariff.
What are BPTs? Introduced in waves since 2010/11 Priced and structured to incentivise best practice BPTs Move away from average cost per episode No one-size fits all approach
Providers will be paid according to their performance. Payment should reflect outcomes, not just activity, and provide an incentive for better quality. 2010 Health White Paper
Best Practice Tariff Introduced in 2014/15 for primary hip and knee replacement Level of reimbursement conditional on patient outcomes HRG code HB12B HB12C HB21B HB21C Description Major Hip Procedures for non Trauma Category 1 with CC Major Hip Procedures for non Trauma Category 1 without CC Major Knee Procedures for non Trauma Category 2 with CC Major Knee Procedures for non Trauma Category 2 without CC
Best Practice Tariff (HB21B) ( ) 6,400 6,300 6,200 6,100 6,000 5,900 5,800 5,700 5,600 5,500 5,400 5,300 Standard tariff BPT 633 Incentive
From patient responses...
The PROMs questionnaires PROMs collections for hip and knee replacements consist of two instruments: EQ-5D, a widely used generic quality of life questionnaire. Oxford Hip Score, Oxford Knee Score: validated, conditionspecific hip and knee questionnaires that have been developed by the University of Oxford. For the Best Practice Tariff we use the condition-specific questionnaire. More sensitive to changes due to the joint replacement procedure, as it includes questions covering specific aspects of quality of life that might be affected by the procedure.
Oxford score: patient feedback on twelve questions...
Patients report big health improvements Patients fill in a pre-op questionnaire at a consultation prior to surgery and a post-op questionnaire about six months after surgery. The difference in Oxford Hip (Knee) scores between the preoperative and post-operative scores can be taken as an indication of the health impact of the surgery - generally very positive.
Average change in Oxford Knee Score post- vs pre-op Outcomes vary (somewhat) across providers 30 Average health gain per provider (2012/13 knee score data) 25 20 15 10 5 0
Case-mix: taking into account differences in patient characteristics There are factors outside of the control of the provider which affect patient-reported health after surgery, e.g. patient characteristics. For each patient we estimate an expected post-operative health taking into account a range of variables (bold = hip, underlined = knee). Average outcomes at provider level are then compared to predicted outcomes to identify providers with outcomes better or worse than expected. Pre-operative health Living alone Age Gender Ethnicity Deprivation Disability Co-morbidites Duration of symptoms prior to surgery Primary diagnosis Assisted in filling in questionnaire
Graphically: funnel plot of providerlevel outcomes (hip)
... to tariff decisions
Using PROMs to determine BPT payment Payment of the BPT applies where providers: Do not have an average health gain significantly below the national average (99.8% sig.); and adhere to the following data submission standards: a minimum PROMs pre-op participation rate of 50%; a minimum National Joint Registry (NJR) compliance rate of 75%; and an NJR unknown consent rate below 25%.
Numbers of trust who have been outliers Procedure Performance outlier Response rate outlier Joint outlier (including NJR) Hips 8 4 29 Knees 6 2 18 2013/14 provisional PROMs data NHS providers
Variations Commissioners may agree to pay BPT where: A provider has a particularly complex case-mix that is not yet appropriately taken into account in the case-mix adjustment; or Recent improvements not yet reflected in the nationally available data; Providers have identified why they are an outlier and have a credible improvement plan in place, the impact of which is not yet known.
Using PROMs to improve patient outcomes Beyond payment decisions
What does it mean to be an outlier PROMs can help trusts and commissioners understand not only their overall outcomes, but also which particular areas are of concern. PROMs data includes information on Overall patient satisfaction with outcomes Patient outcomes on different dimensions, e.g. mobility, pain, usual activities Complications and re-admissions Data can be analysed by patient groups (e.g. age, pre-operative health) It is also worth comparing outcomes across multiple conditions, multiple instruments and over time.
A real-life example of an outlier trust Domain Trust score National Patients saying results were fair at best 18.9% 11.2% Patients saying problems now worse 6.6% 4.2% Patients reaching excellent score (42+), Hip 26% 51% Patients reaching excellent score (42+), Knee 18% 30% Patient with extreme difficulty kneeling down and getting up again afterwards, Knee 63.7% 53.5%.. Dressing themselves, Hip 23.9% 12.2% Patients having had urinary problems 15.9% 11.5% problems with the wound healing. 15.7% 11.3% an allergy or adverse reaction to drug. 13.4% 10.3% problems with their wound bleeding. 8.7% 6.4%
Local level analysis of the data PROMs data is collected for trusts by third party data providers. These providers can give you access to identifiable, patient-level PROMs data. This allows more detailed local analysis though sometimes limited by low numbers- e.g. by consultant by detail of treatment, e.g. prosthesis used etc.
NHS England analysis going forward NHS England (together with the HSCIC) are looking to make the data easier to use and understand. Current pieces of work: Dashboard: summary of trust level PROMs data, highlighting relevant areas of interest. Topics of interest: analysis alongside HSCIC publications to shed light on particular aspects of the data and explore variation in outcomes NHS Choices: looking into ways of incorporating PROMs-based information into NHS Choices PROMs based material to support GPs and patients in shared-decision making
Topics of Interest Make more use of PROMS data and build more interest Gather your views on what areas are most useful to you Prioritise Analysis by NHS England or as part of the quarterly HSCIC topic of interest Broad Areas: Clinical insights and analysis of patient outcomes Methodology/data quality Hospital level outcomes and outliers Please use the listings provided to indicate which areas are most of interest to you. - Thank you for your help.
Please do get in touch! Grateful if you could fill in our questionnaire on topics of interest. More widely, to inform our work, we would like to hear from you, about how you use PROMs data or how you would like to use it in the future, and how we could make it more useful to you. Please do get in touch: rafael.goriwoda@nhs.net england.paymentsystem@nhs.net /resources/pay-syst /statistics/statistical-work-areas/proms www.hscic.gov.uk/prom