METHODS: The patients studied consisted of 58 people with an established diagnosis of ulcerative colitis (UC) and Crohn's disease (CD) as determined by a combination of clinical, radiological, endoscopic and histological criteria. The patients were residents of a well-defined geographical area in the northern part of Singapore and had been referred to the single regional hospital. Epidemiological data including sex, age, ethnicity, family history and disease type and extent were collected from case records and patient interviews.

RESULTS: There were 37 UC and 21 CD patients. Of the patients with UC, 67.5% were Chinese, 13.5% were Malay and 19% were Indian. The CD group consisted of 81% Chinese, 9.5% Malay and 9.5% Indian patients. The study population from which the patients were drawn was approximately 0.5 million in size.

METHODS: Recurrent abdominal pain was defined as at least three episodes of abdominal pain, severe enough to affect a child's activities over a period longer than 3 months. A health-care consulter was defined as a child who had been brought to see a doctor regarding recurrent abdominal pain at least once in the past year. Children aged between 9 and 15 years were randomly chosen from schools in the city of Petaling Jaya, given questionnaires to fill in and interviewed to determine whether they fulfilled the above criteria for having symptoms of recurrent abdominal pain and for being a consulter. Bivariate analysis and multiple logistic regression analysis were performed on the data obtained.

RESULTS: One hundred and forty-three (9.61%) children fulfilled the criteria for recurrent abdominal pain out of a total of 1488 schoolchildren interviewed. There were 65 (45.5%) consulters and 78 (54.5%) non-consulters. Among the consulters, the male to female ratio was 1:1.4, while among the non-consulters, the ratio was 1:1.1. On bivariate analysis, the Chinese had a significantly lower likelihood to consult a doctor (P = 0.02), while the other two races did not show any increase in consultation (Malays, P = 0.08; Indians, P = 0.21). Among those with severe pain, there was a significantly higher prevalence of consulters (P < 0.01). Furthermore, those whose sleep was interrupted by abdominal pain were more likely to consult (P < 0.01). Children who had consulted a doctor were more likely to be missing school because of abdominal pain (P < 0.01). Following multiple logistic regression analysis, ethnicity was no longer a significant predictor.

METHODS: A prospective study of 485 consecutive patients who underwent colonoscopy during a 22-month period was performed. All patients answered a detailed questionnaire. Indications for colonoscopy and the findings were recorded.

RESULTS: The mean age of the study population was 55.7 +/- 14.7 years. There were 221 (45.6%) males and 264 (54.4%) females. Sixty-five (13.4%) were Malays, 298 (61.4%) were Chinese and 112 (23.1%) were Indians. Multiple backward stepwise regression analysis revealed that independent predictors for colorectal cancer (odds ratio [95% CI]) were the presence of rectal bleeding (4.3 [4.0-8.0]) and iron deficiency anemia (4.0 [3.6-10.2]). In those aged 50 and over, male gender (4.5 [2.2-9.3]) and abdominal pain (3.1 [1.4-6.7]) were also significant positive predictors of cancer.