Sometimes there are different types of patient encounters taking place within the same health facility. In addition to general medical visits for acute or chronic illness, there can be separate clinics for well-child visits, pre-natal and post-natal visits, dental visits, specialist consultations, and so forth. There can also be separate clinics for adult and paediatric cases, and sometimes new patients are separated from re-attendances. Treatment practices for these different types of encounters can be quite different. A study which mixes different types of encounters in an unsystematic way will produce results that are difficult to interpret. Indicators studies should be restricted to a sample of general illness encounters, representing a mix of health problems and ages.

The types of encounters in the sample (e.g. general outpatients only) should be decided beforehand and described clearly to prevent arbitrary decisions by the investigators when they encounter unusual circumstances. This decision should reflect the purposes of the study, and it should also take into account the practical aspects of retrospective or prospective sampling. If samples of adult and paediatric cases are drawn from separate sources, half the sample should be taken from each age group, spread evenly over the chosen period.