As part of an evaluation of the training of hospice nurses in communic
ation skills, the selectivity of patients in disclosing their concerns
and the ability of nurses to register ail the concerns disclosed were
studied. Forty-two nurses were recruited from two hospices in the nor
th of England. They were asked to determine and write down patients' c
urrent concerns before and after training, and nine months later. Thei
r interviews were tape recorded to permit rating of the concerns discl
osed. After each interview a research nurse used a semistructured inte
rview and the Concerns Checklist to elicit patients' concerns. The Spi
elberger State Anxiety Scale and Hospital Anxiety and Depression Scale
were then administered to assess patients' mood. In total, 87 patient
s were thus assessed. Patients were highly selective in what they disc
losed and showed a strong bias towards disclosing physical symptoms. O
verall, 60% of concerns remained hidden and concerns about the future,
appearance and loss of independence were withheld more than 80% of th
e time. Patients who were more anxious or depressed were less likely t
o disclose concerns. The nurses registered only 40% of the concerns di
sclosed to them at interview, and less than 20% of patients' concerns
were identified approprately. The nurses were selective in the categor
ies of concerns that they registered. Pain, family worries, appetite a
nd weight loss, nausea and vomiting were noted most frequently, while
concerns about cancer, bowel function, treatment and emotional worries
were not registered. The patients' main concern was identified and re
corded in only 45% of cases. Overall, it was found that hospice patien
ts selectively disclosed physical symptoms while nurses did not elicit
or register patients' concerns accurately. Nurses therefore need to i
mprove their ability to elicit and register all of their patients' con
cerns and to pay particular attention to those who are anxious and dep
ressed.