Gastroenterological treatment is often comprised of a series of multiple interventions, each one associated with a potential benefit, as well as harm. The analysis addresses the question of when to stop a series of consecutive interventions and maximize medical benefit. The benefit and harm associated with gastroenterological interventions are modeled as 'continuous' influences on patient health or as a 'discrete' sequence of random events. The analysis suggests that the benefit of a sequence of gastroenterological interventions is likely to accumulate following few interventions at the beginning of therapy and that any prolonged sequence is likely to inflict as much harm as benefit. It is impossible to reach a state of perfect health with inherently imperfect interventions. Health cannot be raised above a level that equals the ratio of expected benefit/(expected benefit+expected harm). As the achievement of perfect health constitutes an unattainable goal, a sequence of gastroenterological interventions should not be continued much further beyond its initial success.