We operationalized this with the following question: “We asked you about your expectations when it was decided that your would have a gastrostomy feeding tube. We asked surrogates at 3 and 6 months whether their expectations had been met. We dichotomized the 4-level response, which ranged from strong agreement to strong disagreement for each question into agree versus disagree analyses using the numerical 4-point scale gave similar results. Higher scores indicated higher expectations. The final index was composed of the sum of 14 items reflecting expectations for: improved nutrition being more comfortable less pain fewer problems eating greater independence not feeling hungry or thirsty quality of life having a decreased risk of pneumonia living longer more freedom better health fewer coughing episodes and greater enjoyment of food (included as a distracter to assess global optimism). The content of the interviews included an index of items reflecting possible expectations for improvement from the gastrostomy tube that was generated from review of the outcomes literature and from open-ended interviews with experienced physicians. We initially contacted the family member who signed the procedure consent for the feeding tube, confirming that they were the family member “most familiar with the patient’s care.” Most interviews were conducted several days to 2 weeks after feeding tube insertion.

To ensure a uniform source of data, we attempted to interview a surrogate in all cases. We excluded trauma patients because they often have a very good prognosis, and the feeding tube is transient. Early deaths and metastatic cancer were excluded because our aims were to examine the medium-term (3- to 6-month) functional outcomes and fulfillment of expectations with tube feeding. Exclusion criteria were: non-ENT (ear, nose and throat) metastatic malignancy, major trauma, noncompetent patient with no surrogate decision-maker, ward of the state or patient death before completion of consent. Inclusion criteria were: age >17 years and ability of patient or surrogate to speak English. Patients were identified by daily checks of inpatient gastrointestinal procedures, interventional radiology schedules and chart review.

Patients who received an initial gastric feeding tube were recruited at two North Carolina Hospitals: a university-based 608-bed tertiary care hospital and a 500-bed community hospital about 45 miles away.

1433 task3