Quality of physician care is especially important in the elderly population, as high quality medical care can prevent hospitalization due to chronic conditions [12
]. Insuring high quality of care is therefore an important goal to increase quality of life for those over 65 as well as decrease burden on the health care system. Physicians can enhance patients' perceptions of the quality of care by understanding the differences in perception and assessment of medical care that exist between the general and older population.
Consistent with theory [13
], we found that the caring attitude of the physician is a strong predictor of patient satisfaction. We also confirmed that waiting time and time spent with physician play key roles in the physician rating and satisfaction. Additionally, our study elucidated some of the differences between trends in satisfaction between the elderly and non-elderly. One of the most significant differences between these two groups exists in the time spent with physician and waiting time variables. To illustrate this, Figure is shown to present the relationship between waiting time and patient satisfaction as a function of age. Initially, as waiting time increased, patients over the age of 65 were more forgiving than the younger group. One possible explanation of this difference may be that older patients are simply more likely to give higher satisfaction ratings, as numerous studies have shown [14
]. However it doesn't explain why this trend does not continue as wait time increases; above 30 minute waits, overall patient satisfaction decreased for all age groups. It may be that for non-elderly patients, waiting time is more important when rating their physician. This is reflected in the stronger negative correlation of waiting time to physician satisfaction for the non-elderly as compared to the elderly patients.
Relationship between waiting time and patients' satisfaction as a function of age.
Physician satisfaction also seems to be associated moderately and significantly with patient-reported time spent with the physician in the elderly but not nearly as strongly correlated with satisfaction as in the non-elderly. Other studies [17
] have shown that increasing the visit time may provide physicians with a way to minimize and offset patient dissatisfaction when long waiting times is unavoidable. In following with our results, this may be more effective in the general population than in the older population.
All of these findings raise some important issues which deserve further study, for example: (1) even though elderly patients and non-elderly patients report similar waiting times, why do the elderly patients report overall higher satisfaction scores or why does increased waiting time seem to adversely affect satisfaction less in elderly patients; and (2) why does increased time spent with physicians not positively affect physician satisfaction in this group nearly as strongly as in the non-elderly? One could speculate that physicians are spending enough time with elderly patients, but the issue of waiting in physician offices needs further attention. The ways in which longer physician wait times could differentially affect older adults also may need further study and investigation.
A recent study found that the general practitioner's age was "negatively associated with patients' evaluation of all aspects of age, except accessibility" [19
]. This is an especially interesting finding given that our study focused on the age of the patient rather than the physician. Possible future evaluations are needed to explore this new finding, perhaps taking into account both the age of the physician and patient and evaluating its effects on satisfaction ratings.
Quality of care is clearly a complex, multi-faceted concept. Thus it must be noted that our exploratory analysis had several limitations. First, a cross-sectional study does not permit causal inferences about the results. Second, our findings may be subject to respondent bias, since elderly patients may be less familiar with the use of the internet to rate their quality of medical care. Furthermore, self-reported data is subject to respondent recall bias and may have affected the survey responses we received, especially from the elderly group. Third, we did not measure variables such as race [7
], health status [14
], method of insurance [15
], or patient trust of physician [4
], which have been identified as correlates or possible correlates for patient satisfaction. Additionally many other factors may have influenced patient satisfaction ratings (such as accessibility, level of physician communication clarity, and patient expectations of the visit) but were not measured. Finally, satisfaction ratings in older adults can be heterogeneous; patients 65 to 69 tend to give higher ratings, while those 80 and older tend to give lower ratings [14
]. Therefore, our findings may be affected by potentially greater social desirability bias (patients ranking physician highly because that is the socially desired norm) in elderly respondents.