A low response rate does not necessarily indicate non-response bias in gastroenterology survey research: a population-based study