Publications

We present a case study using the negative binomial regression model for discrete outcome data arising from a clinical trial designed to evaluate the effectiveness of a prehabilitation program in preventing functional decline among physically frail, community-living older persons. The primary outcome was a measure of disability at 7 months that had a range from 0 to 16 with a mean of 2.8 (variance of 16.4) and a median of 1. The data were right skewed with clumping at zero (i.e., 40% of subjects had no disability at 7 months). Because the variance was nearly 6 times greater than the mean, the negative binomial model provided an improved fit to the data and accounted better for overdispersion than the Poisson regression model, which assumes that the mean and variance are the same. Although correcting the variance and corresponding test statistics for overdispersion is a standard procedure in the Poisson model, the estimates of the regression parameters are inefficient because they have more sampling variability than is necessary. The negative binomial model provides an alternative approach for the analysis of discrete data where overdispersion is a problem, provided that the model is correctly specified and adequately fits the data.

OBJECTIVES

To examine the adequacy of caregiver-patient communication in serious illness and its relationship to caregiver burden.

DESIGN

Cross-sectional cohort study.

SETTING

Participants' homes.

PARTICIPANTS

One hundred ninety-three persons aged 60 and older seriously ill with cancer, congestive heart failure, or chronic obstructive pulmonary disease and their caregivers.

MEASUREMENTS

Communication concerns, measured in terms of agreement with statements regarding desire for and difficulty with communication about the patient's illness. Caregiver burden, measured using a 10-item subset of the Zarit Burden Inventory, with scores ranging from 0 to 40 and higher scores indicating greater burden.

RESULTS

Of caregivers, 39.9% desired more communication, and 37.3% reported that communication was difficult. Of patients, 20.2% desired more communication, and 22.3% reported that communication was difficult. Disagreement regarding communication concerns was frequent in caregiver-patient pairs; of caregivers who desired more communication, 83.1% of patients did not, and of patients who desired more communication, 66.7% of caregivers did not. Caregivers who desired more communication had significantly higher caregiver burden scores than did caregivers who did not (9.2 vs 4.7, P<.001), even after adjusting for patient's diagnosis, income, and functional status and caregivers' age, sex, and relationship to the patient.

CONCLUSION

A large proportion of caregivers and seriously ill older persons had an unmet desire for increased communication, although they frequently disagreed with each other about this desire. Caregivers' desire for increased communication may be a modifiable determinant of caregiver burden.

Brief symptom instruments are designed to assess symptoms while maintaining low respondent burden, but they may omit important information. Our objective was to determine whether a representative brief symptom instrument effectively captures the full symptom experience of older adults with advanced diseases. In this cross-sectional study, we interviewed 90 community-dwelling adults with cancer, congestive heart failure, or chronic obstructive pulmonary disease regarding the presence of symptoms in the prior 24 hours. Participants rated the intensity and bothersome nature of 15 symptoms--10 symptoms were included in the Edmonton Symptom Assessment Scale (ESAS) plus 5 supplemental symptoms. Participants reported similar proportions of ESAS and supplemental symptoms. Intensity and "bothersomeness" ratings frequently differed. Brief symptom instruments only provide a limited assessment of the respondent's symptom experience. The benefit obtained from incorporating both the intensity and bothersome nature of a longer list of symptoms may outweigh the potential increase in respondent burden.