We investigated satisfaction with a Swedish telenursing service and the health-care-seeking behaviour among callers who received a less urgent level of health care than they expected. A postal questionnaire was sent to a random selection of callers (n = 273) to Swedish Healthcare Direct in October 2008. The 'cases' were 18 callers where the telenurse recommended a lower level of health care than the caller expected and who were not in complete agreement with the nurse. The 'controls' were 22 callers who either received a lower recommendation, or were in disagreement with the recommendation. There were no differences between cases, controls and other callers regarding background factors or the telenurse classification of emergency. However, both cases and controls considered their need for health care as more urgent than the other callers. An independent test of the nurses' reception, ability to listen and to take notice of the callers' health problem, showed that nurses who had served cases, had received a significantly lower rating than other nurses. For nurses who had served controls, there was no such difference in rating. Cases and controls had fewer subsequent care visits than other callers, in the three days following the call, although the proportion of emergency visits was higher among cases and controls compared to other callers. If the caller and the nurse disagree about the nurse's recommendations, the consequence can be a dissatisfied caller and more visits to unnecessary high levels of health care. Further training of the nurses may improve the telenurse service.