 |
 |
|
Rapid Responses to:
|
|
Electronic letters published:
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif){kind=icon}
In Response:
Quality of care and vulnerability
|
|
|||
|
Takahiro Higashi, MD, PhD Kyoto University Department of Epidemiology and Healthcare Research, Neil Wenger, Paul Shekelle
Send rapid response to journal:
thigashi{at}pbh.med.kyoto-u.ac.jp Takahiro Higashi, et al.
|
We interpret the findings very differently than Dr. Garrett and colleagues, although we agree with their conclusion that there is a strong argument for palliative medicine for many of the patients in the study sample. Many of the quality indicators in the ACOVE measurement set focused on pain and end of life care. Dr. Garrett and colleagues state that there is a lack of association between our quality measurement and survival, yet the analyses in the article demonstrate a strong relationship.(1) Their concern that only 1 of 9 of the most prevalent quality indicators (in Table 2) demonstrates a statistically significant relationship between receiving the care process and survival misses the point that these ancillary analyses, which are statistically underpowered and unadjusted, were presented only to provide insight into potential mechanisms of the relationship between process and outcome. We focused on the direction of the point estimates, showing that the relative risk of death for those who passed the quality indicators was below 1 in 8 of the 9 quality indicators. Dr. Garret and colleagues expressed concerns about the selection of community-dwelling vulnerable older persons using the VES-13 and were interested in the age distribution within our sample, especially the proportion of persons 85 years or older among those with a VES-13 score of 5 or less. There were 135 persons 85 years or older, among whom 79 had the VES score of 5 or less. Furthermore, they were concerned about the relationship between age and survival in the studied cohort. The VES-13, which contains age, was a strong predictor of survival, as expected. We share Dr Garrett and colleagues’ excitement about our findings of association between quality and survival and their commitment to quality care for older patients. 1 Higashi T, Shekelle PG, Adams JL, et al. Quality of care is associated with survival in vulnerable older patients. Ann Intern Med. 2005;143:274-81. Conflict of Interest:None declared |
|||
|
|
|||
|
Stephanie L Garrett, MD University of Louisville, James G. O'Brien, MD, and Toni P. Miles, MD, PhD
Send rapid response to journal:
s0garr04{at}louisville.edu Stephanie L Garrett, et al.
|
LETTER TO THE EDITOR: In a recent meeting of our fellows and departmental faculty, we discussed Higashi et al (1). There was considerable excitement because we are committed to quality care for our frail older patients. To our surprise the analyses as presented did not support the conclusions. The authors state that ‘…8 of these 9 quality indicators… patient(s) … receiv(ing) recommended care was less likely to die …’ (Table 2). Only one of these 9 quality indicators were significant – the exact opposite of the text. Pneumococcal vaccine lowered the risk of death by 54% while all the others did not alter risk. Figure 1 comparing high and low quality care groups showed a clear separation only after 800 days. During the first 400 days, the lines are indistinguishable. The authors used the Vulnerable Elders Survey – 13 to create the sample. Patients who score a 3 or higher with the VES-13 are at 4 times the risk for death over the next two years (2). During the 3 year follow-up period 23% of patients in the overall sample died. The paper does not report proportion dead within categories of quality score. The authors use age as a surrogate for severity of illness and see no association with VES-13 (Figure 3). The implication, though not explicitly stated, is that the lack of relationship between age and VES-13 means that no statistical confounding is present. However, in every other report that has ever been published, age is significantly associated with risk of death. The lack of a direct test for association between age and survival is a major oversight. VES-13 has a selection bias that would systematically label robust persons who are 85 years and older as vulnerable. In the scoring system, anyone who is in this age group gets an automatic 3 points. It would be informative to look at the age distribution within VES-13 groups. We suspect that the persons who are 5 or less are in this subgroup. The VES-13 appears to do a very good job of identifying persons who are within 1 – 2 years of death. Comfort care, goals of care, advance directives are quality indicators for end of life. The lack of association between the ACOVE quality indicators (3) and survival in this sample makes a strong argument for palliative care medicine. Sorting out robust older adults from frail ones will result in quality of care for all. References 1. Higashi T, Shekelle PG, Adams JL, Kamberg CJ, Roth CP, Solomon DH et al. Quality of care is associated with survival in vulnerable older patients. Ann Intern Med 2005, 143(4): 274 – 281. 2. Saliba D, Elliott M, Rubenstein LZ, Solomon DH, Young RT, Kamberg CJ et al. The Vulnerable Elders Survey: A Tool for Identifying Vulnerable Older People in the Community. JAGS 2001 49: 1691 – 1699. 3. ACOVE Quality Indicators. Ann Intern Med 2001, 135(8): Part 2 653 – 667. Conflict of Interest:None declared |
|||
