For healthcare professionals only


Abnormality rates on barium meal examination in different racial groups

BMJ 1995; 310 doi: (Published 08 April 1995) Cite this as: BMJ 1995;310:909
  1. R Orme, senior registrara,
  2. A Novosadek, registrarb,
  3. J R Lee, consultanta
  1. a Department of Radiology, General Hospital, Birmingham B4 6NH
  2. b Department of Radiology, Dudley Road Hospital, Birmingham B18 7QH
  1. Correspondence to: Dr R Orme, Department of Radiology, Toronto Hospital, General Division, 200 Elizabeth Street, Toronto, Ontario, Canada M5G 2C4.
  • Accepted 12 December 1994

Barium meal examination of the upper gastrointestinal tract is a widely performed investigation readily available to general practitioners and hospital outpatient departments. In central Birmingham a high proportion of the population is of Asian (from the Indian subcontinent) descent. We investigated differences in rates of abnormality on barum meal examination in Asian and white patients attending our departments.

Patients, methods, and results

The reports of barium meal examinations of patients attending two radiology departments in central Birmingham during 1992 were retrospectively analysed. Racial origin was indicated on the request form at the time of examination and was recorded in all patients. All retrievable radiological reports of Asian patients and a comparable number of white patients were studied. Reports were classified as showing (a) normal findings, including small amounts of gastrooesophageal reflux; (b) minor abnormalities, including conditions such as hiatus hernia, gastric erosions, and duodenitis; and (c) major abnormalities including malignancy or suspected malignancy and gastric or duodenal ulceration.

The data were analysed using {chi}2 tests, and subgroups were also compared according to age, sex, and source of referral.

Of 1362 patients attending for barium meal examination, 960 (70.5%) were white and 361 (26.5%) were Asian, with 41 others (3.0%). Radiological reports of 286 Asian and 331 white patients were retrieved and analysed. The mean age of Asian patients was lower than that of white patients (44.5 v 51.0 years). In addition, the proportion of young patients (those under 45) was significantly higher in Asians (59% (165/282) v 38% (125/331) in white patients, difference 21% (95% confidence interval 13.2% to 28.8%); {chi}2=32.7, df=2, P<0.001). No significant difference was found between men and women in either racial group. Over three quarters of Asian patients (79% (225/286)) had been referred directly by their general practitioner compared with just over half of white patients (53% (176/331), difference 26% (18.7% to 33.3%); {chi}2=42.7, df=1, P<0.0005). There was, however, no significant difference in the rate of abnormality between patients referred by general practitioners and those from hospital outpatient departments in either racial group, and there was no excess of referrals from any particular source.

Abnormality rates were higher in white patients, the overall rate being 47% compared with 25% in Asians (difference 22% (14.7% to 29.4%); {chi}2 32.0, df=1, P<0.001) (table). The age standardised rate for Asians was 28%, giving a relative excess of abnormalities in white patients of 89% using crude rates and 65% after standardising for age. Rates of both major and minor abnormalities were significantly increased in the white patients (table).

Abnormality rates on barium meal examination in Asian and white patients by age, sex, means of referral, and type of abnormality

View this table:


Over a quarter (27%) of patients attending for barium meal examination were Asian, although Asians constituted 16% of the hospital population in the 1991 census1; this gives a relative excess of attendances among Asians of 68%. Asians may consult a general practitioner more often, rates of up to three times those of white patients being seen among Pakistani men, and Asians are also more likely to have repeated consultations.2 3 4 In our study 79% of Asian and 53% of white patients were referred directly by their general practitioner, implying that patient management may vary in different racial groups. Overall rates of abnormality were, however, similar regardless of the means of referral.

Young patients (under 45) were over-represented in the Asian group, an age group in which abnormality rates are low,5 but significant differences were found in all age groups and after standardising for age. Communication difficulties in some Asian patients may contribute to inaccuracies in clinical diagnosis, but such difficulties are less likely in younger patients, among whom the differences remain striking.

Variations in prevalence of gastrointestinal symptoms and differing patterns of patient presentation and patient management are all factors that could be examined in more detail in the community and in hospitals, and they might provide opportunities to optimise investigation and management of patients in different racial groups.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract