Sunday, February 21, 2010
A few analyses from the Coping With Cancer Study have been published recently, all in the Journal of Clinical Oncology. We've published extensively about the CWCS (see here for more). Briefly, it was a prospective, US multi-institutional study of several hundred advanced cancer patients (& their family caregivers) which measured at baseline, among many things, characteristics of patient coping, communication with clinicians, and care preferences. Patients were followed through death, and caregivers were followed into bereavement.
First is a look at race and predictors of 'intensity' of care at the end of life. This analysis looks at ~70 African American patients and ~230 white patients (essentially all the patients who had died by the time of the analysis who self-reported those racial categories). AA patients were more likely to report wanting (37% vs 22%), and to actually receiving (13% vs 3%) 'intensive' EOL care (receiving intensive EOL care was defined in the study as dying in an ICU after either CPR or mechanical ventilation). Rates of reporting having an EOL discussion with a physician, as well as trust in their physician, were similar between white and AA patients, wheras positive religious coping was more common in AA patients than whites.
In multivariate analysis, religious coping was strongly associated with receiving intensive EOL care in white patients (see also here for another CWC paper on religious coping), but not for AA patients (this is probably because the AA patients were more uniformly positive religious copers than white patients that there just weren't enough to capture measurable differences). In addition, having a preference for intensive EOL care was strongly associated in white patients with actually receiving it; for AA patients there was a not-statistically-significant trend towards this (adjusted OR 4.5, p=.058 compared to an aOR of13 for white patients). That is - white patients were more likely to get what they wanted (at least as measured by self-reported preferences at baseline) than AA patients.
A few thoughts about this. Overall this is very consistent with prior research about racial and cultural differences in EOL care. It's always important to keep in mind however that while these differences are there, and measurable, and reproducible, the groups are more similar than they are different (e.g. vast majority of patients did not receive intensive care at EOL, majorities in both groups reported quality of life/symptoms were more important overall than absolute length of life, etc.). This reality, and the fact that you can never assume what the patient in front of you wants until you ask, should always be kept in the forefront clinically.
What's most interesting about these findings is the disconnect between baseline preferences and actual EOL care in AA patients (at least relative to the white patients). The authors hypothesize that, at least regarding those patients who report at baseline wanting intensive EOL care, AA patients' preferences are more fluid and situation-contingent (e.g. change more readily as death approaches) than white patients who endorse similar goals at baseline. Of course another explanation (which is also discussed) is that AA patients' wishes are not respected to the extent of white patients' at life's end (and in fact having a DNR order at baseline did not seem to be 'protective' from intensive EOL care for AA, as it did with whites). For both groups (the DNR-at-baseline and the preference-for-intensive-EOL-care-at-baseline) the disconnect between patients' wishes and what actually happens is more pronounced in AA patients than white patients. That's disturbing, and worth more investigation.
Second is a look at which patients receive care at the end of life which is consistent with their stated preferences. This analysis looked at baseline patient preferences (for symptom-directed care preferentially over life-extending care or vice versa), actual care received in the last week of life, whether patients reported an EOL conversation with their physicians, and how bereaved family caregivers rated physical and psychological distress in the last week of life.
There are a lot of findings here, and I'll only mention a couple. First is that they note that baseline preferences for life-prolonging care or actual receipt of life-prolonging care in the last week of life were not associated with actually living longer (survival was ~200 days from enrollment in patients preferring symptom relief, and ~160 days for those preferring life prolongation; there were similar findings in patients who actually received life-prolonging care in the last week of life - none of the differences were statistically significant). They defined life-prolonging care in the last week as chemotherapy, tube feeding, or any 'ICU-level' of care. This is reminiscent of the SUPPORT study sub-analysis of cancer patients - preference or receipt of more 'intensive' care at the EOL was not actually associated with living longer.
As far as patients receiving care that was consistent with their baseline wishes, reporting having an EOL conversation with their physician was the only factor consistently associated with this congruence. Regardless of baseline preferences, those patients who received life prolonging care in their last week of life (as defined above) had more physicial and psychological distress in the last week (as reported by family caregivers). So, even for patients who said (at baseline) they preferred life prolonging treatment, receipt of such treatment was associated with more suffering, and lower rated quality of life (compared to patients who didn't receive those treatments, regardless of preference).
I've ruminated at length before about prior CWC findings (showing that having an EOL discussion was associated with improved EOL care/outcomes) - about whether it was the impact of the discussion itself vs. 'there's something different about patients who have these discussions' which leads to the differences in EOL care/outcomes. While this study can't answer that question, to an extent these findings suggest that 'patient protoplasm' is not the major determinant, as even those patients who preferred symptom-control over life-prolongation but received life-prolongation, did poorly. Ie - it's more than a simple phenenomon of those patients who are more comfortable with death/dying/& talking about it 'doing well' through the end. If you ascribe causality to these findings (which should be done cautiously of course) then it seems like we can actively hurt our patients (increase suffering) by the treatments we give them as they die. This is not news of course.....
The final one looks at the impact of spiritual care by medical teams on patient's end of life care. In some ways it is a follow-up to this CWC analysis Tom reported on a year ago. This study looks at patient reports of spiritual support from their medical team, and pastoral care utilization, and finds that those are associated with more hospice utilization, and less aggressive EOL care, even among religious copers.
Loggers, E., Maciejewski, P., Paulk, E., DeSanto-Madeya, S., Nilsson, M., Viswanath, K., Wright, A., Balboni, T., Temel, J., Stieglitz, H., Block, S., Prigerson, H. (2009). Racial Differences in Predictors of Intensive End-of-Life Care in Patients With Advanced Cancer Journal of Clinical Oncology, 27 (33), 5559-5564 DOI: 10.1200/JCO.2009.22.4733
Mack, J., Weeks, J., Wright, A., Block, S., Prigerson, H. (2010). End-of-Life Discussions, Goal Attainment, and Distress at the End of Life: Predictors and Outcomes of Receipt of Care Consistent With Preferences Journal of Clinical Oncology DOI: 10.1200/JCO.2009.25.4672
Balboni, T., Paulk, M., Balboni, M., Phelps, A., Loggers, E., Wright, A., Block, S., Lewis, E., Peteet, J., Prigerson, H. (2009). Provision of Spiritual Care to Patients With Advanced Cancer: Associations With Medical Care and Quality of Life Near Death Journal of Clinical Oncology, 28 (3), 445-452 DOI: 10.1200/JCO.2009.24.8005