G N Rajkumar, D R Small, I G Conn
Department of Urology, Southern General Hospital, Glasgow
Correspondence to: Dr D Small, Principal Physicist, Departments of Urology & Clinical Physics, Southern General Hospital, Glasgow G51 4TF Tel: 0140 201 1528 Email: Doug.Small@sgh.scot.nhs.uk
SMJ 2005 50(2): 61-63
Introduction
Computerised audit systems have been used in various medical departments. These have been shown to be a valuable tool in resource modelling1 and in maintaining standards of work among medical2 and secretarial staff.3,4 The computerised audit system in our department was set up in August 1995 to collect information on all clinical activity in the department. It provides database recall of all clinical data, and also prompt generation of discharge summaries. The diagnosis and procedure codes (second edition; 1996), issued by the British Association of Urological Surgeons were used. These incorporate the ICD 10 codes for diagnosis and the OPCS 4 for procedures. These codes form an integral part of the audit system but are also added to, making the system more versatile. The system is based on one that was already in use in the surgical and otolaryngology departments of another Glasgow Trust but has been extensively modified to serve the urological specialty. Since August 1995, data on 22,671 patients undergoing in-patient or day-case treatment have been entered. Many of these patients have multiple records due to the nature of our workload.
Materials and methods
Data entry form Data are collected on a standard form.(Fig1) The top part of the form is for standard demographic details of the patient, dates of admission and discharge and method of admission. There is provision to indicate when the form was actually passed to the respective consultant following discharge of the patient and when the form was completed. These fields were added as part of a management audit. The next part of the form is for the diagnosis codes and has provision for co-morbidity. This is followed by the procedure codes along with date and surgeon codes. Fields for complications – either general or procedure specific - are completed as they occur, or on discharge. There are fields to indicate pending reports from pathology, radiology, biochemistry and urodynamics. The discharge outcome field has a numbered menu. The amount of information given to patients or their relatives can also be indicated. The follow-up fields indicate whether the patient is to be seen in clinic or readmitted either as an in-patient or in the day case unit. On the reverse of the form, there is space for additional comments in free text, where results of investigations and management plans are entered.
To further assist in diagnostic and procedure coding a smaller booklet with commonly used codes has been developed. The British Association of Urological Surgeons (BAUS) handbook is however available for reference. These data then generate a discharge letter in a standard format on a sheet of A4 for the patient record and mailing to the general practitioner. More complex or complicated cases may require a formal dictated summary. Data on lithotripsy for urological stone disease and urodynamic investigations are entered on separate forms.
Lithotripsy data entry form
Anecdotally, we felt that the contracted-out lithotripsy service was not achieving the stone clearance success rates that either the provider or the literature claimed. In order to assess this we created a lithotripsy module in the audit system and after data collection for a year, were able to present the provider with objective data to back up our claim.
In addition to demographic details, this form has entry fields for side, site, size and position of the calculus and also type of calculus whether single, staghorn or in a cluster. The field for site has options indicating upper, middle or lower calyx, renal pelvis, upper, middle or lower third of ureter or the vesico-ureteric junction. The size field classifies stones as <5mm, 5–10mm, 10–20mm, 20–30mm and >30mm and also stores the absolute value of the size. The treatment fields include data on the power and number of shocks given to each stone. The field for outcome of treatment can indicate whether the treatment was successful or not and whether the case needs further review by the consultant. It can also indicate whether the stone has moved to a different location and might be amenable to a different modality of treatment. The field for further treatment has options for further lithotripsy, PCNL, ureteroscopic intervention, stent insertion or removal, nephrostomy or a combination of any of the above. There is also space for comments in free text.
Urodynamic data entry form
Our department has a specialist interest in urodynamics and receives referrals from all over Scotland. A separate module was created for urodynamics for the following purposes:
• Instant recall of investigation result when an outside consultant phones.
• Selection of cases for regular urodynamic training meetings.
• Selection of cases for research projects.
• Summarising referral pattern for management purposes.
This form has 3 sections. The first section has demographic details and the type of urodynamic investigation along with the result of the investigation. The next section has fields for previous surgery (24 options), neurological history (14 options) and presenting symptoms (19 options). The fields on neurological conditions have options of indicating the level of lesion and whether it is a complete or an incomplete lesion. The last section has the findings of the urodynamic study. Filling and voiding detrusor function are characterised by choosing from drop-down menus offering the relevant ICS terminology.5 Numerical data can also be input for bladder parameters such as capacity, flow rates, urethral pressure.
The lithotripsy and urodynamic data entry modules are understandably more elaborate than the routine clinical ones, but the former are usually completed by the specialist staff involved in these procedures at the time they are performed.
Network and software
All data were initially stored on a dedicated personal computer within the urology department and regularly backed up to disk. Within a year, the system was migrated to an IT department server and networked to seven users. All the secretaries are connected via the hospital intranet and their personal computers are networked using Novell software. The file server is backed up daily by the IT department. The system application is DATAEASE version 5 which is a relational database with a query language. This version of DATAEASE is DOS based and although there are windows versions available we have yet to feel the need to upgrade. In general DOS based systems are faster and easier to maintain.
At present there are records of 22,671 patients with 46,410 admissions. There are 71 forms, 183 relationships between forms and to date over 430 queries have been written. The DATAEASE query language allows for reasonably subtle queries of the database. When even more specific data are required or a complex format is required, MS visual basic is used, after the DATAEASE query language has run a filter first. This is the case with monthly ureteric stent surveillance, where the DATAEASE query extracts details of all patients who have ever had a ureteric stent inserted and the visual basic routine establishes whether it is still in-situ and if so for how long. The three-monthly activity data have a particular format to enable instant summary of the log of trainees or consultants and this is possible only with a high level programming language such as visual basic. On the other hand, the monthly mortality and morbidity data are all produced by the DATAEASE queries alone.
Codes for diagnoses and procedures were established by a sub-committee of BAUS in 1996. They are based on the ICD10 and OPCS4 systems. These were electronically imported to the audit system. In order to allow more complex recording of sub specialist activity, one extra alphanumeric character has been added, for example Ambulatory Urodynamics is M478C, Stretching of tension free vaginal tape is M528A. Neither of these codes was in the original data which were: “urethral catheterisation for urodynamics”-M478 and “placing of tension free vaginal tape”-M528
Results
Simple and routine collection of data is the philosophy of this computerised audit system. The ready access to data and the ability to build sophisticated queries have enabled numerous departmental audits, many of which have been done instantaneously. This system has greatly facilitated many clinical, managerial and training audits with implications for resource planning and management. To date over 430 queries have been processed. The following are a few examples:
Clinical audit
The insertion of an indwelling ureteric stent is an increasingly common procedure and is used in diverse urological conditions. This has led to an increased incidence of stent related morbidity and complications .6, 7 However one serious situation, inevitably associated with delayed management, is complication associated with a retained stent or a forgotten stent.8 We have a system whereby any patient who has a stent in-situ for more than 150 days is automatically flagged, along with details of date of insertion and the respective consultant, to enable us to review the situation and take action accordingly.
In managing patients with urological malignancy, the audit system enables us to identify patients with more than one malignancy and arrange appropriate follow up. Cancer registries are becoming more dependent on hospital information systems for collection of their data.9 Our data on newly diagnosed cancer patients are retrieved every month and passed on to the Information and Statistics department of the NHS in Scotland via the Scottish Cancer Registry. An annual review of the number of transurethral resections of prostate (TURP) performed over a seven-year period has shown a steady downward trend. This has implications in resource planning and in predicting training patterns for trainees. (Fig2)
One early example of the system allowing the audit loop to be closed involved a simple comparison of all the procedures being done by four consultants in our department. It was apparent that one consultant was performing many more cystoscopies than the three others for the management of benign prostatic hyperplasia. After discussion and consultation of the literature 10 the practice of the one surgeon was changed. Clearly counting the cystoscopies did not necessarily require a sophisticated audit program but the point is that we did not know a priori that cystoscopy was the variant procedure; the system was used to compare all procedures and cystoscopy was the one that fell out as the outlier.
Managerial audit
A well designed computerised audit system has been shown to be a powerful tool in resource planning and management.1 Information gained from our system has considerable advantage for the hospital management. For example the audit of our lithotripsy service showed that the actual rate of stone clearance was lower than predicted by the lithotripsy service providers and the literature11 and resulted in the company undertaking further sessions free of charge to obtain stone clearance. This was a short audit lasting only a year comprising 65 patients with 85 stones. Table 1 shows the data presented to the company supplying the mobile lithotripsy service.
There is great pressure for in-patient beds and there have been suggestions that adequate provision of day care surgery facilities would lead to a more efficient use of in-patient beds.12,13 As a management exercise we recently postulated that all minor and intermediate cases could be managed in a “23 hour" ward. We analysed around 17,000 minor and 466 intermediate procedures, and found that 38% of minor cases stayed over one day and 32% of intermediate cases stayed three days or fewer. We felt that with adequate resource provision for a formal preassessment clinic we could potentially save over 4500 bed days per year running a "23 hour" ward.
Urodynamic discharge letter
The urodynamics module has been extremely useful for research and teaching. Probably its most useful routine application is in the production of an automated GP letter. Fields from the audit system populate a report in MS ACCESS. A final diagnosis and follow up from the consultant are entered and a discharge letter generated without the need for dictation. A screen shot from the letter is shown in figure 3.
Training audit
One of the features of the computerised system is the ability to produce summaries of clinical activity. At present we produce a summary every three months and also annually. This can be compared with any previous year or period. The procedures have been summarised into 50 categories and further grouped into major, intermediate and minor. Two of our regular training meetings are in urodynamics and lithotripsy. The audit system greatly simplifies the task of selecting the cases for these. For example it is a very simple query to select all recent cases of neurogenic detrusor overactivity. During our monthly departmental meetings, morbidity and mortality data are readily produced and relevant discharge summaries are generated for further discussion. A computerised personal audit for surgical trainees has been shown to be very beneficial.14 For the benefit of our trainees, a list of their procedures can be exported either to Microsoft Excel or Access and produce an accurate logbook for training purposes.
Conclusion
This computerised audit system is a cheap, labour saving and effective tool, which in today ’s climate of clinical governance is a must for any department. A wide variety of clinical, managerial and training audits has been performed with relative ease. The dual approach to querying (DATAEASE and Visual Basic) allows a very sophisticated level of querying and output style. The cost of an eightuser networked version of DATAEASE is approximately £300. Note: Figures 1 and 3 are available from the authors.
ACKNOWLEDGEMENT: We would like to acknowledge the work of Chris Sheldon, Roger Quinn and Howard Smith of North Glasgow University Hospitals NHS Trust in the design of the initial database.
REFERENCES
1 Ellis BW, Rivett RC, Dudley HA. Extending the use of clinical audit data: a resource planning model. Br Med J 1990 Jul 21; 301 (6711): 159–62.
2 Earnshaw JJ, Hayward JK, Horrocks M, Baird RN. The importance of vascular surgical audit to surgeons, patients and purchasers. European Journal of Vascular Surgery 1992 Sep; 6(5): 540–4.
3 Dunn DC, Dale RF. Combined computer generated discharge documents and surgical audit. Br Med J Clinical Research Ed. 1986 Mar.22; 292(6523): 816–8.
4 Llewelyn DE, Ewins DL, Horn J, Evans TG, McGregor AM. Computerised updating of clinical summaries: new opportunities for clinical practice and research? Br Med J 1988 Dec 10; 297(6662): 1504–6.
5 Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourology and urodynamics 2002;21(2):167-178.
6 S. Richter; A. Ringel; M. Shalev; I. Nissenkorn: The indwelling ureteric stent: a “friendly” procedure with unfriendly high morbidity. BJU International, Vol 85, Issue 4, March 2000, pages 408 – 411.
7 Tolley, David: Ureteric stents far from ideal. The Lancet, Volume 356 (9233). 9 September 2000, 872–873.
8 Monga, Manoj; Klein, Eric; Casteneda-Zuniga, WilfidoR; Thomas, Raju: the forgotten indwelling ureteral stent: a urological dilemma. J Urol, Vol 153(6). June 1995. 1817–1819.
9 Moss SM, Smith J, Nicholas DS. The quality of histopathology data in a computerised cancer registration system: implications for future audit care. Public health 1997 Mar; 111(2): 101–6.
10 Madsen FA, Bruskewitz RC. Cystoscopy in the evaluation of benign prostatic hyperplasia. World Journal of Urology 1995; 13:14–16.
11 Lingeman JE. Extracorporeal shockwave lithotripsy. Development, instrumentation, and current status. Urologic Clinics of North America 1997 Feb 1; 24(1): 185–211.
12 Glass RE, Thomas PA. Surgical audit in a district general hospital: a stimulus for improving patient care. Annals Of The Royal College Of Surgeons Of England 1987 May; 69(3): 135–9.
13 Wyllie JH, Kidson IG, Wyllie DH. Pursuing efficiency in surgical practice. Br Med J 1988 Nov 26; 297(6660): 1368–71.
14 Beard JD, Brennan JA, Budd JS. A computerised personal audit for surgical trainees. Annals Of The Royal College Of Surgeons Of England 1990 Jul; 72(4): 46–8