J P Tierney, J B Robins*, G S Anthony*, S Bjornsson+, G Orr#
Department of Gynaecology, Southern General Hospital, Glasgow; *Department of Gynaecology, Inverclyde Royal Hospital, Greenock
+Department of Gynaecology, Victoria Infirmary, Glasgow; #Department of Urology, Inverclyde Royal Infirmary, Greenock
Correspondence to: Dr John P Tierney, Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF
E-mail: john.tierney@sgh.scot.nhs.uk
SMJ 2004 49(4): 133-136
Background and aims: To conduct a profile audit of three surgical treatments for urinary stress incontinence through the application of an episode costing process. Methods: Four stage methodology: (1) construction or a profile of care for each of the surgical approaches (setting the standard); 2) calculation of a theoretical profile cost; (3) calculation of the observed costs from real patient episodes; (4) comparison of observed costs with profile costs, (comparing present practice with established standard). Results: Profiles of care were constructed and compared with 39 actual in-patient episodes. Tension free vaginal tape (TVT™ Gynecare) is the cheapest modality of treatment in terms of both the expected profile cost and also observed (real patient) cost. Colposuspension is the most expensive form of treatment with real costs significantly greater than the expected profile. Clinical issues such as length of stay, duration of surgery, patient selection and complication rates were revealed through the exception reporting process. Length of stay is the main determinant of overall cost. Conclusion: It is possible to construct a costed and auditable standard of care for a surgical procedure. This standard can be compared with real patient costs calculated using the same methodology. Exception reporting based on differences between expected and real costs can be used to facilitate the audit of clinical practice. The technique is limited, however, by the need to collect accurate and detailed activity data.
Key words: Audit, incontinence surgery, cost analysis
Urinary incontinence affects 30% of women at some time in their lives. Many surgical techniques exist for the treatment of genuine stress incontinence (GSI), reflecting the varied results of such procedures.1 The ‘gold standard’ has been the Burch Colposuspension with continence rates of 85-90% at one year.1 Although this procedure is extra peritoneal, it is not without morbidity. This has led to the development of less invasive procedures to reduce both surgical and anaesthetic risks.
Injection of bulking agents such as collagen into the bladder neck is a less invasive technique with low morbidity and has been favoured in the elderly, less mobile population.1 Its efficacy reduces with time although the procedure can be repeated. Anaphylaxis, collagen migration and fears over potential prion infection have further restricted the use of these materials.
Tension-free vaginal tape (TVT) has been a recent addition to surgical practice.2 Although this is a relatively new procedure, reported five-year success rates equal that of colposuspension.3,4 Its major advantages are minimal tissue dissection and its suitability as a day surgery procedure, either under regional or local anaesthetic. While it is without doubt that clinical practice should dictate the appropriate mode of treatment for any patient, the relative cost of the available surgical procedures is of increasing importance in assessing ‘value for money’.
Robins and colleagues have previously described a method of determining the expected cost of patient treatment within a theoretical setting through the application of an auditable standard – the profile of care.5, 6 Applying the same episode costing principles to actual patient data allows the real cost of these procedures to be calculated. Not only does this provide a method to compare the costs of different techniques, it also aids audit since variations from the standard can be identified through exception reporting.
The first task was to construct profiles of care for the three modalities of treatment currently used in our unit, namely colposuspension, tension-free vaginal tape (TVT) and bladder neck collagen injection. These profiles were based on contemporary practice, recent evidence, and the experience of the principle surgeons. The intention was that they would represent current good clinical practice for each of the three procedures, (setting the standards). The profiles describe the treatment episode from initial referral to discharge and include out-patient appointments, urodynamics, nurse counselling, and pre-operative assessment as appropriate. Minimum and maximum acceptable deviations, including frequency of minor complications, were also specified within the profile. 5
Each profile of care was costed using information supplied by the relevant hospital finance departments. Detailed information including staff costs, equipment used, investigations performed and theatre time utilised were collated. Overhead allocations for administration, domestic, portering and hotel costs were included.6
Using this information the expected cost for each of the three profiles was generated. Minimum and maximum expected costs were also calculated to provide a cost range that allowed for acceptable deviations from the typical practice described in the profile.
Thereafter, thirty-nine recent cases involving these treatments were reviewed (12 TVT, 13 colposuspensions, 14 collagen injections). Each patient episode was analysed and costed, using the same methodology as applied to the expected profile, to produce the actual episode cost. These actual costs were then compared with the expected costs from the profiles of care, (comparing present practice with standards).
Application of costed data to the profile of care Table 1 shows the typical standard costs for routine preoperative investigations and patient stay in this district general hospital.
This financial information was the basis for the episode costing methodology used in the construction of costed profiles of care. Costs are for 2000/2001. Overhead allocations include administration, capital charges, catering, domestics, laundry, medical records, portering and utilities. Ward and outpatient clinic costs only include medical and nursing overheads. Specific material costs relating to two of the operative procedures are recorded in Table II.
It was assumed for the purpose of profile calculations that only one tape would be used for a T.V.T procedure whereas collagen injection would typically require the use of two ampoules, this being the usual experience of the operating surgeons. The calculated profile costs are reproduced in Table III. A mean, minimum and maximum profile cost is provided for each of the options.
Costed audit and exception reporting
Thirteen consecutive colposuspension procedures were analysed and costed using the principles outlined above. The average patient age was 48.2 years. The mean observed cost was £3100 compared to the expected profile cost of £2693.13. The maximum observed cost was £3802.56 while the minimum observed cost was £2055. Fourteen collagen procedures were reviewed. The average age was 61.8 years. The mean observed cost was £1592.85 compared to the expected profile cost of £1638.27. The maximum observed cost was £2543.27 while the minimum observed cost was £1320.
Twelve TVT procedures were included in the series. The average patient age was 54.3 years. The mean observed cost was £1304.09 compared to the expected profile cost of £1184.57. The maximum observed cost was £1986 and the minimum observed cost was £1050.44. This data is summarised in Table IV.
In addition annotated histograms were produced to demonstrate actual patient costs for each of the three surgical procedures, (Figures 1, 2 & 3). To aid interpretation a vertical dotted line corresponding to the value calculated as the mean expected profile cost has been superimposed onto each chart. Individual patient ‘exceptions’ are highlighted with ‘call-out’ text. A simple comparison of the main drivers of cost for each procedure is made in Table V.
Profiles of care are a practical description of the routine care given to a patient group for a stated diagnosis or operative procedure. Creation of the initial profile requires identification of current best practice. They should be comprehensive, specific and feasible (not idealised), and represent affordable care at a high standard. Profiles of care should focus on the patient, monitor quality and demonstrate the use of resources.5 Costed profiles can also link best clinical practice with treatment cost.6 Profile costs are, of course, theoretical but their value is demonstrated when applied to real patient data. Clinical issues such as length of stay, duration of surgery, patient selection and specific complications are clearly highlighted through the exception reporting process and are the principle drivers of cost. Ideally the expected cost should equal the real cost. Failure to meet the best practice standard is recognized by differences between the expected (profile) and real (episode) patient cost.
Whilst we endeavoured to make allowances for common complications of each procedure within the expected profile costs, atypical or unexpected additions to routine care will clearly inflate the actual patient costs. However, just because the actual episode cost exceeds the expected, this doesn’t necessarily mean that the patient received ‘bad’ care – it is just more likely that there were co-morbidities or complications.
In fact more worrying is the case where the actual episode cost is less than the minimum expected profile cost. In this situation it is more than possible that an investigation or procedure was omitted. Fortunately in this study no patient cost was less than the expected profile cost, although one patient episode came close – a young healthy patient who required minimal pre-operative assessment with straightforward surgery and short in-patient stay. In terms of actual episode cost alone colposuspension proved to be the most expensive mode of treatment for urinary incontinence. This is clearly due to the length of hospital stay involved in this procedure. The actual in-patient cost of collagen injection compares favourably with the expected profile costs, despite the older age group involved. Allowances were made for age group in the construction of the initial profile and this accounts for the wide variation both in acceptable treatment and theoretical cost range, (maximum expected cost minus minimum expected cost), for this specific procedure. Yet in practice the observed episode cost range for collagen injection was much less than we had anticipated. However, TVT was found to be cheapest form of treatment, both in terms of expected cost profile and observed cost. The observed cost variation for TVT was minimal but then again the expected profile cost range was also ‘tighter’, perhaps reflecting more stringent selection criteria for this surgical option.
Clearly the technique described does not constitute a full cost-benefit or cost-effectiveness analysis. Furthermore it must also be realized that the profile cost is the theoretical episode cost and not simply an ideal operation cost. The costing process may be criticised by some accountants as being incomplete but it has been applied in a consistent and clinically meaningful manner to facilitate a ‘bottomup’ clinical process audit. The use of costed clinical audit data can widen discussion in clinical audit and bring in management perspectives. Profiles of care are thus simply a tool to facilitate clinical audit within an episode-costing framework.
Profiles of care are both an audit tool and a practical description of routine care for a given patient group. They document local guidelines and protocols as quality treatment plans.5 Furthermore, the monitoring of variation between the expected cost and real cost leads to the identification of adverse outcomes or changes in clinical practice.6 The methodology described employs the systematic application of a ‘bottom-up’ costing process within an auditable framework and is thus ‘doctor-friendly ’. This process of episode-based profile costing can also be used to generate a sound business case that can be appreciated by management but is grounded on good clinical practice. However, the technique is labour intensive being limited by the need to collect reliable costed information and detailed activity data. In this simple observational audit TVT emerges as the cheapest option and this is largely because the procedure is performed as a day case, thereby minimising hospital admission and its subsequent cost. These results are encouraging since the long-term success of this particular operation appears equal to colposuspension.3,4
REFERENCES
1 Royal College of Obstetricians & Gynaecologists. Surgical Treatment of Urodynamic Stress Incontinence. Guideline No. 35. London: RCOG; ( O c t o b e r 2 0 0 3 ) . [ w w w . r c o g . o r g . u k / r e s o u r c e s / p u b l i c / urodynamic_stress_incontinece_No35.pdf]
2 National Institute for Clinical Excellence. Guidance on the use of tensionfree vaginal tape (Gynaecare TVT) for stress incontinence. Technology Appraisal No. 56. London: NICE; 2003.
3 Ulmsten U, Johnson P, Repazapour M. A 3 year follow up of tension free vaginal tape for surgical treatment of female stress urinary incontinence. Br. J. Obstet Gynaecol 1999; 106: 345 – 50.
4 Ward KI, Hilton P, Browning J. A randomised trial of colposuspension and tension free tape for primary genuine stress incontinence. Neurourol Urodyn 2000; 19: 386 - 88
5 Robins J.B., Anthony G.S., & MacMillan R. Resource Management: Profiles of patient care. Hospital Update Plus 1993; 19(6A): 97-101S.
6 Robins J.B., Wilson K., Anthony G.S., & Williams J. Costing patient care & using costed data in audit. The Clinician in Management 1997; 6(1): 2-8