Objectives: To examine dimensions of a validated instrument measuring geriatric attitudes of primary care residents and performances on these dimensions between residents and fellows.
Design: Cross-sectional and longitudinal studies.
Setting: An academic medical center.
Participants: Two hundred thirty-eight primary care residents (n=177) and geriatrics fellows (n=61) participated in the study from 1995 to 2000.
Measurements: A 14-item, 5-point Likert scale previously validated for measuring primary care residents' attitudes toward older people and geriatric patient care was used.
Results: Factor analysis showed four dimensions of the scale, labeled Social Value, Medical Care (MC), Compassion (CP), and Resource Distribution, which demonstrated acceptable reliability. Both groups of subjects showed significantly (P<.001) positive (mean>3) attitudes across the dimensions and times, except for residents, who had near-neutral (mean=3) attitudes on MC. Residents' mean attitude scores on the overall scale and the MC and CP subscales were significantly (P<.001) lower than those of fellows over time. Residents and fellows showed different change patterns in attitudes over time. Residents' attitudes generally improved during the first 2 years of training, whereas fellows' attitudes declined slightly. Personal experience was a strong predictor of residents' attitudes toward older patients. Ethnicity, academic specialty, professional experience, and career interest in geriatrics were also associated with residents' attitude scores.
Conclusion: The multidimensional analysis of the scale contributes to better understanding of medical trainees' attitudes and sheds light on educational interventions.