AIM - To determine the effectiveness of a context-sensitive medicine approach to the rationalisation of the prescribing of proton pump inhibitors (PPIs) in general practice.
DESIGN - Review of all patients taking PPIs and conversion to maintenance therapy where indicated.
SUBJECTS AND SETTINGS - 94 patients taking PPIs at a community health centre.
OUTCOME MEASURES - Success or failure of medication alterations 6 months after the changes had taken place. Reasons for failure and the interval between changes were recorded.
RESULTS - 66 (70%) out of a total of 94 patients were identified as being suitable for dose reduction of their PPI. After 6 months, 63 patients (95%) remained in the study of whom 55 (87%) were taking a lower dose of PPI. Of the 8 patients who left the study, 7 did so because of a deterioration in symptoms.
CONCLUSION - Using a context-sensitive medicine approach and keeping patients fully informed as to why changes were being made to their medication proved to be a highly successful technique in bringing about a rationalisation of the prescribing of PPIs. Substantial savings can be generated for reinvestment in primary care.
Greenhalgh and Worrall1 have argued that the time is now right for evidence-based medicine (EBM) "mark III", for which they have coined the term "context-sensitive medicine". Context-sensitive medicine is the paradigm which they hope will unite, on the one hand, the proponents of the hard science of the "mark I" model of EBM with, on the other hand, the proponents of patient-centred medicine. With the first attempt at this - EBM "mark II" - the pendulum swung too far in the direction of the latter. EBM mark II involved integrating the best available external clinical evidence with individual clinical expertise. It has been argued that EBM mark II appears to allow clinicians to have their cake and eat it, accepting EBM where it fits in with their existing practice but arguing personal experience when it does not.
Context-sensitive medicine retains the high-quality clinical research of the mark I model and also the patient-centredness of the mark II model. However, it introduces an important additional notion in that the clinician's entire decision-making process can be subjected to full scientific scrutiny.
The practice based at the Holsworthy medical centre is staffed by seven doctors (six whole time equivalents) and has approximately 8,500 registered patients. It has a drug budget of £750,000 which, for the year ending March, 1998, was overspent by £25,000 (3 per cent).
The practice decided to top-slice its drug budget to employ a pharmacist based at the drug information centre at the local district general hospital for one day per week for 12 months. The pharmacist attended a rational prescribing seminar organised for the practice by the North & East Devon health authority where it was agreed that a comprehensive review of prescribing should be undertaken using the following approach:
A number of key areas were identified and the initial topic chosen was the prescribing of proton pump inhibitors (PPIs)
Using the resources available to him in the hospital's drug information centre, the pharmacist produced a folder entitled "The evidence base". The folder included relevant articles2–8 to enable the doctors working with the pharmacist to draw up action plans for each topic chosen for review. The folder also contained some material presented in a style that was comprehensible to patients should they wish for further information regarding changes to their medication. Each doctor had a copy of the folder, which was regularly updated by the pharmacist, readily available in his or her consulting room. A copy of the folder was also given to the two local community pharmacists who were notified in advance of any likely changes to the pattern of the doctors' prescribing.
The action plan for the audit of PPIs comprised the following:
Ninety-four patients were reviewed. Sixty-six of these patients (70 per cent) were identified as being suitable for dose reduction of the PPI.
An audit of patients who had had their medication changed was carried out six months later. Of the original 66 patients, three were no longer registered with the health centre, leaving a total of 63 patients (95 per cent) in the study. Overall, 52 (83 per cent) of the remaining 63 patients had been successfully converted on to the recommended medication change. Eight (13 per cent) had had to revert to their original medication because of a deterioration in symptom control (seven patients) or because adverse effects were experienced (one patient). Three patients (5 per cent) either required further modification of treatment or all PPIs were stopped (see Table 1).
Table 1: Results six months after changes had taken place |
|||
| Change to daily dose | Successful | Unsuccessful | Other |
| Omeprazole 20mg to lansoprazole 15mg | 40 | 7 | *3 |
| Lansoprazole 30mg to lansoprazole 15mg | 9 | 0 | 0 |
| Omeprazole 20mg to omeprazole 10mg | 3 | 1 | 0 |
| Total | 52 (83%) | 8 (13%) | 3 (4%) |
| *Of these three patients, one had his proton pump inhibitor stopped and two had their medication changed further to omeprazole 10mg daily | |||
The work carried out in this part of the project took six days of the pharmacist's time. The overall effect of the changes to proton pump prescribing led to a saving of £5,600 over the six-month period audited (based on MIMS prices, July, 1999; see Table 2).
Table 2: Treatment costs of proton pump inhibitors (MIMS, July, 1999) |
||||
| Drug | Treatment dose | Annual cost | Maintenance dose | Annual cost |
| Lansoprazole | 30mg daily | £387 | 15mg daily | £185 |
| Omeprazole | 20mg daily | £393 | 10mg daily | £260 |
| Pantoprazole | 40mg daily | £388 | 20mg daily | £195 |
| Rabeprazole | 20mg daily | £310 | 10mg daily | £198 |
It was recognised at the beginning of the project that the probable key to success was to keep everyone fully informed of potential changes. An article was placed in the local newspaper informing the residents of Holsworthy and surrounding areas that a pharmacist from the local district general hospital would be working with the doctors at the health centre to review prescribing. It mentioned that the pharmacist would be available at the health centre every Tuesday to discuss any concerns patients may have about their medication and in particular that patients would have the opportunity to comment on any drug changes before they took place.
The pharmacists working at local community pharmacies were informed of the rationale behind medication changes so they were in a position to answer inquiries from their customers.
Information was provided in a format that was readily available in the consulting room for access by either the doctor or the patient or could be accessed via the local pharmacy.
For maintenance therapy, lansoprazole 15mg was chosen, partially because research9 has shown that 70 per cent of patients with reflux oesophagitis could be maintained on this dose but mainly because the doctors felt that a change of drug would be more effective than simply lowering the dose of omeprazole, which the majority of patients were currently taking.
Also at current prices it is more cost effective for maintenance therapy (Table 2). Pantoprazole (Protium) and rabeprazole (Pariet) were not licensed for maintenance therapy when this study was carried out.
Although cost savings needed to be made we were adamant that this should not be at the expense of deterioration in patients' symptoms. Indeed, our aim was to improve drug management.
The long-term consequences of powerful inhibition of acid secretion have not yet been fully determined. Lack of gastric acidity may lead to enteric infections and to atrophic gastritis in patients infected with Helicobacter pylori. To date there is no evidence of an increased risk of gastric or colon cancer.10,11
Patients are becoming more aware that drugs have adverse as well as beneficial effects. The patient information leaflet for omeprazole lists 34 potential adverse effects and that for lansoprazole 29 potential adverse effects. Headaches, diarrhoea, stomach pains and skin rashes are among the more commonly reported side effects and these have sometimes been severe enough to lead to discontinuation. Rare side effects include impotence, gynaecomastia and changes in kidney and liver function. Indeed, during this study the PPI was discontinued in one patient because of a deterioration in renal function. Prescribing the minimum effective dose of any drug should lessen the risk of a patient experiencing adverse effects. Also, if the optimum dose is taken, it should prove easier to spot any deterioration in the patient's condition.
By carefully screening patients initially, it was possible to achieve a high success rate of 83 per cent in converting patients on standard treatment doses of PPIs to maintenance therapy.
Partly as a result of this audit, the doctors have adopted a "step up" (lifestyle modifications -> antacids -> H2-receptor antagonist -> treatment dose of PPI) followed by a "step down" (treatment dose of PPI -> maintenance dose of PPI -> H2-receptor antagonist -> antacids) approach for the management of gastro-oesophageal reflux disease. At which point of the ladder a patient enters will depend on the doctor's assessment of the severity of the patient's symptoms. If a PPI is needed then it should be prescribed at a treatment dose initially but, after a suitable period of symptom control, reduced to a maintenance dose.
This study supports previous reviews which suggested that lower doses of PPIs are adequate for continual maintenance in most patients. We feel that the success rate in this project was mainly due to:
Patients in whom the changes were unsuccessful noticed very shortly after the alteration had taken place and were quickly restabilised on their previous medication. A review of PPI prescribing can unlock savings to be used for other community-based health projects.
Mr Wathen is principal pharmacist (drug information and formulary services) at North Devon District Hospital, Raleigh Park, Barnstaple, Devon EX31 4JB. Dr Barker, Dr Edwards, Dr Page, Dr Hillebrandt, Dr Wardle, Dr Shaw and Dr Green-Armytage are general practitioners at Holsworthy health centre, Holsworthy, Devon.
Correspondence to Mr Wathen