Ours is the first large population-based cohort study of low-income women to examine the determinants of patient satisfaction with breast cancer treatment. Nearly three quarters of women in our sample reported they were extremely satisfied with the breast cancer care they had received. This high rate of satisfaction has been noted in the literature.10,16,18–20,24–26
Explanations for such elevated satisfaction rates include fear of losing services, social desirability bias, or ingratiating response bias, which could occur when patients attempt to ingratiate themselves with the researchers/providers, especially when confidentiality cannot be assured in the survey.27
Other authors postulate that respondents may be reluctant to criticize for fear of unfavorable treatment in the future28
or that dissatisfaction is expressed only when an extreme negative event occurs.29
Our sample consisted of women who had been enabled by the BCCTP in terms of providing them access to breast cancer care that they possibly would not have received otherwise or would have accessed with much greater difficulty. In such a scenario, it is plausible that the high reported satisfaction rate is an accurate representation of the true rate of satisfaction in our sample.
In terms of the determinants of satisfaction, older women were more likely to be extremely satisfied compared with younger women. Evidence suggests that older people tend to be more satisfied with their healthcare provision in both the primary care and in-hospital settings.30–32
Possible reasons for this are postulated to be more modest expectations in the elderly and their reluctance to criticize.33
Less acculturated Latinas had five times the odds of non-Hispanic white women of being extremely satisfied. This effect size is more than twice the effect size noted in the unadjusted analysis. After controlling for education, marital/partner status, and self-efficacy, we are left with the pure effect of ethnicity and acculturation in the model. It is conceivable that less acculturated Latinas are more satisfied because of the greater gratitude they feel on having received care for breast cancer than others, who might have better access to care. Their expectations from the system may be lower than those of the more acculturated Latinas, who by virtue of being more familiar with their environment may have higher expectations in line with those of native-born women. This reasoning is in contrast to that of Fitzpatrick's theory of “the need for the familiar” determining patient satisfaction, which states that patients from cultures/ethnicities different from the West are unlikely to be familiar with the medical canons and, thus, less likely to be satisfied.34
An important finding of our analysis was the lower odds of satisfaction among women who believed they could have been diagnosed sooner. Arriving at a diagnosis of breast cancer is a process that begins with the woman (or her physician) finding an abnormality, deciding to get it investigated further, and interaction with the healthcare system for a definitive diagnosis, which could involve visits to multiple providers and coordination among them. This process may be delayed in women who do not have health insurance and, thus, face greater barriers to accessing care. In our sample, slightly more than half the women (51%) had a >60-day delay between the time they first became aware that something was wrong to their biopsy or surgery. This diagnostic delay could be the result of individual level factors (e.g., financial or transport barriers, regret at not getting regular mammograms), system level factors (e.g., long waiting times or repeat biopsies); it could also represent the overall dissatisfaction at the late stage of diagnosis. Nonetheless, it is important to realize that this perceived delay can have a negative impact on self-reported satisfaction with care.
Women who received or were continuing to receive chemotherapy or radiotherapy were more likely to report being extremely satisfied. This could be a reflection of their perception that all that was possible was being done in terms of treatment to help them overcome their cancer, as exemplified by Wensing's theory of goal seeking, which avers that major concerns for patients are resolution of their health problems.35
Noh et al.10
also found receipt of radiotherapy to be a significant determinant of patient satisfaction after breast cancer surgery. It should be noted, however, that not all women would benefit from these treatment options, and those not receiving chemotherapy or radiotherapy may include those who did not follow treatment plans, those who did not need it, and women who did not have the time to travel to the clinic to complete the chemotherapy or radiotherapy.
Additionally, increased information giving was strongly associated with greater satisfaction, which is in consonance with the literature.36,37
For example, Liang et al.37
found that greater patient-physician communication was associated with a sense of choice, treatment, and satisfaction with care in older women with breast cancer. Furthermore, physician emotional support was also strongly associated with satisfaction and supports findings in the literature.36,38,39
in primary care found that satisfaction with treatment by general practitioners was related to perceptions of interest and devotion by the physician rather than to technical skills. A similar finding was reported by Silliman et al.,41
who found that older women with breast cancer relied heavily on their physicians for support.
Finally, greater physician solicitation of patient input in the decision-making process was associated with greater satisfaction, similar to what has been reported in the literature.15,42–46
This finding in our population is noteworthy because it indicates that even in a relatively disempowered population, participatory decision making independently affects satisfaction. Modifying such physician behavior could be a valuable target in this population for improving satisfaction, in addition to modifying training programs for physicians so as not to assume that lower socioeconomic patients should be communicated with in a paternalistic fashion. This finding must be modulated by the fact that not all women may want or desire a highly participatory role in decision making.47,48
One study found that nearly 69% of patients preferred a passive role in decision making,49
and a meta-analysis of patient preference studies reported that most studies found higher proportions of patients desiring information rather than a greater role in decision making.50
Patients who have emigrated from societies with healthcare systems characterized by paternalistic decision making may be uncomfortable with a participatory approach. It is plausible that for many of the Latina women in our sample, the discussion with the surgeon about the choice between lumpectomy and mastectomy could have been their first experience of being involved in medical decision making.
In terms of healthcare policy, our results indicate that interventions, such as Continuing Medical Education for physicians focusing on improving communication skills, which have been shown to improve patient satisfaction,51
could also be beneficial for this provider/patient population. Women report high levels of need for information on a variety of issues pertaining to breast cancer,52
and such information giving can be augmented by the provision of specialized media, such as the Interactive Breast Cancer CDROM,53
patient-focused psychoeducational interventions,54
or a decision board.55
Interventions to address informational needs must ensure that they address a woman's needs throughout her treatment and are not a one-time event. A formal approach, along the lines of the Breast Buddy Breast Care Program at Kaiser Permanente, can be used to combine these patient level interventions with a patient advocate and mentoring from breast cancer survivors.18
Modifying physician behavior when interacting with such disempowered populations by actively soliciting their input in decision making and by targeting outreach activities in these communities (i.e., with access barriers that lead to perceptions of diagnostic delay) are other actionable policy recommendations.
Although our results were obtained from a large population-based cohort of women, a few caveats should be kept in mind. Because the study was conducted in a sample of low-income, medically underserved women in a specific Medicaid breast cancer treatment program in California, external generalizability of the findings to other low-income populations may be limited. Further, the cross-sectional study design prevents an assumption of causality between predictor and dependent variables. We measured satisfaction with a single question, which precluded an examination of the different constructs of satisfaction. In addition, we did not have data on stage of disease, patient coping style, or personality, which may affect patient satisfaction,17,56
although a recent study reported that stage of breast cancer was not associated with surgery satisfaction.20
Finally, the quality of our data depends on the accuracy of patient self-report on physician communication; thus, recall bias may be an issue in our study. However, it has been noted that people who have undergone a sudden and life-threatening health crisis manifest very clear recall of the details surrounding the event57
; breast cancer patients, for example, can recall the precise time they first noticed their symptoms.58
In conclusion, our population-based analysis shows that satisfaction is an important outcome in this population and is reported at high levels similar to those found in other, more enabled populations, but its determinants are unique. Age, ethnicity/acculturation, receipt of chemotherapy and radiotherapy, physician emotional support, and collaborative decision making impact on satisfaction in low-income breast cancer patients. In addition, the perception of diagnostic delay is a strong predictor of low satisfaction in this group, an aspect that has not been highlighted in the literature.