Clinical Coding for BM and PBSC Transplant Procedures
Summary
Background
The Wessex Blood and Marrow Transplantation Unit in Southampton submit transplant activity data each (calendar) year to the European Group for Blood and Marrow Transplantation (EBMT) Registry. Prior to submission this data is validated against the hospital patient administration system (PAS) to ensure transplant activity is accurate and consistent. A number of discrepancies between the data prepared for EBMT and PAS were highlighted. The hospital PAS system identified inaccurate and inconsistent clinical coding for transplant procedures. This resulted in a significantly lower number of transplant episodes being recorded than in reality and the complexity of the procedures was not reflected. In addition, our Commissioning Consortium used PAS data to generate their own report on transplant activity to review current and future funding for the Wessex Blood and Marrow Transplantation service and this again showed reduced transplant activity.
We started to work closely with the Clinical coding department and the Commissioners to produce consistent codes that should be used for all transplant procedures. The International procedure coding system (OPCS) does not currently provide specific codes for peripheral stem cell transplant procedures so the level, consistency and accuracy of the coding was dependent on individual interpretation. We decided to raise this with the BSBMT which then led to us undertaking an audit across all the UK transplant centres.
Plan
Two questionnaires were designed, one for completion by the clinician and one for completion by the clinical coding department. The aim was to determine the codes used by all UK transplant centres, the level of contact between clinicians and coding departments, how coding is undertaken, the level of coding and whether centres routinely validate their EBMT data submission against their hospital administration system.
Process
Both questionnaires were sent 27th April 2006 to 56 teams in 53 transplant centres and replies collected by Keiren Towlson on behalf of BSBMT. All responses were received by November 2006 and passed back to us for analysis. A total of 16 responses were received from the clinicians and a total of 18 responses from the clinical coding departments. In some centres the same person had completed both forms.
Results
These indicated that clinician and clinical coding department responses to the question regarding their contact with each other did not correlate. Less than half of the centres who responded validate their EBMT activity data against their hospital PAS. 16 transplant centres report they code transplant complications but only 14 centres have access to case notes to code these procedures / complications. 2 centres reported codes for harvest and donor harvest that do not reflect these procedures. 1 centre reported coding BM and PBSC autologous transplant as a blood transfusion. 1 centre reported coding BM and PBSC allogeneic transplant as donation of other tissue.
Conclusions
Allocating specific personnel in coding would improve the consistency of coding for transplant procedures. Regular contact between transplant teams and clinical coders is likely to improve the quality and consistency of transplant procedure coding. BM autologous and allogeneic transplant coding is the most accurate. This is reflected by the fact that there are specific OPCS codes available. Accuracy of clinical coding for bone marrow and peripheral stem cell transplant procedures is vital for both costing and quality purposes especially in a payment by results system. There is a need for specific national guidance and the new OPCS codes may more accurately reflect PBSC transplant procedures in the future.