I want to create a rich description of the experiences of adolescents with Type 1 Diabetes, and then use this to create designs that address adolescents' adherence to their treatment regimens.
Ideally, no tool can perfectly substitute any kind of observations including shadowing, in which you can observe the distinction between what people do and what people say they do. However, diary can provide you some information from the actor's point of view.
Alongside info from the patient, you may consider asking for permission to see their clinical records - you can then compare them to gain a broader understanding of the issues - I'm submitting a paper on this to QHR in the next few weeks based on my experiences of an ethnographic review of an occupational therapy led back to work programme for disabled people - happy to pass this on once it's been accepted for publication.
I agree that the diary method will be most appropriate if you are not present. if you are able to make an initial visit to the setting to familiarize yourself and make the observations it would give some credibility to the diary recordings.