Admission rates for peptic ulcer in the Trent Region, UK, 1972–2000

Abstract

Background and aim. Peptic ulcer disease is believed to be less common and less severe as a result of modern medical treatment. We therefore examined changes in the admission rates for patients with duodenal ulcer and gastric ulcer, both emergency (for haemorrhage, perforation or severe pain) and for elective surgery, before and since the introduction of the new advances in therapy. These admission indices reflect disease prevalence and severity.

Patients and methods. We identified admission rates during 1972–2000 within the Trent Regional Health Authority, UK (population 4.7 million), from computerised patient information using diagnostic search codes ICD8–10 and expressed as rates per million resident population. Drug expenditure details were obtained from the Department of Health.

Results. Emergency admission rates as a whole changed little, a decline in the young being offset by an increase in the elderly. Haemorrhage was the most common reason (approximately 115 per million for duodenal ulcer and 87 for gastric ulcer) throughout [compared with perforation (80 and 21) and pain (90 and 68)]. In contrast, elective surgery has almost disappeared; this reduction began before the introduction of modern treatment.

Conclusion. Emergency admission rates for duodenal and gastric ulcer for complications or severe pain have fluctuated over the last three decades but with little overall change. In contrast, elective surgery has declined dramatically, as a result of advances in treatment but also from changes in the natural history.

Introduction

Admission for peptic ulcer (PU) was once common: as an emergency for life-threatening complications and electively for planned surgery, the last resort when medical treatment failed.

The last quarter of the 20th century witnessed the development of three new therapies that revolutionised the medical management of PU. This chronic relapsing disease, hitherto medically incurable, was first brought under control in the majority by daily H2 receptor antagonists (H2RA), which resulted in moderate acid suppression. This soon progressed to profound acid inhibition with daily proton pump inhibitors (PPI) giving control to all, almost without exception. The final development was actual cure of the disease by eradication of Helicobacter pylori, most no longer suffering relapses. The remarkable advances in medical therapy would, therefore, be expected to reduce the need for admission.

Reduction in elective surgery is widely recognised and, by the same rationale, admissions for PU complications would also be expected to decline as the treatment became more extensively used. There remains uncertainty, however, whether any such substantial reduction has occurred.

We therefore undertook a descriptive study to assess trends in admission rates for duodenal ulcer (DU) and gastric ulcer (GU), specifically for ulcer complications, severe pain and for elective surgery, to see if the increasing use of modern anti-ulcer treatment has resulted in any reduction. The study was undertaken in our region, the Trent Regional Health Authority (TRHA), the largest health administrative unit in the country until its recent re-structuring. Located in central England, the Region contained industry and occupation representative of the nation as a whole, and had a resident population of 4.5 million in 1972 rising to 4.9 million by 2000.

Our survey starts before the dawn of the new therapeutic era and encompasses the period of major advances in medical treatment.

At discharge or death, the clinical coding offices of individual hospitals examine the case notes of each patient, and code the data in a standardised manner. These computerised data are then collated by the TRHA and were accessed for our study. These records have provided all the data except that for drug use, which was obtained separately. The records specify the following: date of birth, gender, date of admission, diagnosis, whether admission was elective or as an emergency, operation (if any) and its type, and date of discharge or death.

From the TRHA records, we identified patients admitted as an emergency or electively with DU or GU. Based on the database coding, emergency admissions were divided according to the indication: perforation (±associated bleeding), haemorrhage or severe pain but without complications. A proportion of ulcer complications occur in those admitted for other reasons. These may be listed in a secondary position in the series of fields used to record a patient’s diagnoses. To ensure all such patients were included in our survey, the search criteria covered all diagnostic fields.

Elective admissions were narrowed down to those who underwent anti-ulcer surgery, so selection was based on the first diagnostic position only. The diagnostic and operation codes searched for are listed in Table 1.

In-hospital deaths of any of the above patients, including those post-operatively, were used to calculate mortality rates.

The admission rates are expressed as per million of resident population per year, which on average has been 4.7 million during the survey period. The data are presented in graphs as 3-year moving averages specifying the mid-point year, and separately for men and women. Each graph shows crude admission rates subdivided into age-specific rates for 35–64, 65–74 and ≥75 years, and for all ages combined, which includes 0–34 years.

Data on the use of H2RA, PPI and anti-ulcer drugs as a whole were obtained from the Department of Health, Statistics Division 1E, Prescription Cost Analysis System. The amount of H. pylori eradication treatment used could not be quantified as it is not coded as a separate item but by its constituents of a PPI and antibiotics. As non-steroidal anti-inflammatory drugs (NSAID) and anti-platelet drugs (mainly aspirin) may have a bearing on admission patterns, their use was also noted. We have examined prescription items, i.e. the number of times a particular class of drug was prescribed, and the net ingredient cost (NIC), i.e. the basic cost of a drug (this does not take into account any discounts, dispensing costs, fees or income from prescription charges); together, they give a measure of prescribed drug utilisation in the community. These data exclude drugs purchased by patients ‘over the counter’ as we were unable to quantify this.

We have used data for prescriptions dispensed in the community in England as a whole from 1980 (the first available) onwards to reflect the pattern of use in the Trent region as the regional data goes back only to 2000. There is no reason to suppose prescribing patterns in Trent differ from the rest of England.

The 29-year span (1972–2000) of the survey was arbitrarily divided into four periods, each encompassing a particular phase in the evolution of ulcer therapy. Cimetidine, the first H2RA, was introduced in the UK in November 1976. The 5-year period 1972–1976 was therefore just before modern anti-ulcer medical treatment became available. The H2RA then dominated treatment until the launch of the first PPI, omeprazole, in October 1989; these 13 years, 1977–1989, form the second period. The PPI in turn became the predominant therapy from 1990 to 1994, the third period, until superseded by H. pylori eradication. Unlike H2RA and PPI, H. pylori eradication therapy is not a single drug but a combination of already-available therapies. Hence, there was no specific launch date, rather a gradual emergence and then rapid increase in use, which we estimate started around 1995. The final time period was thus 1995–2000.

The change in the rate of decline of elective surgery was calculated by linear regression using Microsoft Office 2000 Excel. Elsewhere, simple descriptive statistics are used.