| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
TRACK to:
|
|
Electronic letters published:
![[Read Comment]](/icons/misc/sectionDOWN.gif){kind=icon}
Re: Disadvantaged populations are also complex
Disadvantaged populations are also complex
Primary Care & the Pursuit of Health Equity
|
|
|||
|
Robert L. Ferrer, San Antonio, TX, United States Physician; University of Texas Health Science Center at San Antonio
Send response to journal:
|
George Rust and Elizabeth Bayliss make important comments about the multivariate nature of social disadvantage. George Rust reminds us of a sometimes overlooked truth, that racial and ethnic disparities in both health and health care are often driven by factors other than race and ethnicity, and that many of those factors are amenable to policy interventions. Elizabeth Bayliss makes a related argument, with which I also agree, that it is important to acccount for and understand the interdependence of risk factors for poor access to health care. In this regard, Figures 3 and 4 [in the paper] presented multivariate results that simultaneously adjust for the other markers of potential disadvantage. I have also performed several of the stratified analyses she recommends, finding that the results reported in the paper seem robust when stratified by geographic category or insurance status. In explanation, I would offer that although disadvantage is complex, family physicians' practice characteristics mitigate many of the paths to poor access: they charge less than specialists, they work where the population lives, and they make up a significant fraction of community health center staff. So, one could argue that workforce reform is perhaps more necessary, not less, in the absence of reform for universal access. Nevertheless, it seems very likely, as Dr. Bayliss suggests, that enactment of universal access would outweigh any access improvement created through workforce interventions. Whether or not increasing the supply of family physcians (or nurse practitioners) would reduce disparities is an open question; to the extent that the added workforce replicated the patterns of the old, it is plausible. In the paper however, I was making the opposite argument, that further declines in the primary care workforce are likely to erode access for disadvantaged Americans. To me at least, this is less controversial. Who else will step in? Competing interests: None declared |
|||
|
|
|||
|
Elizabeth A. Bayliss, Denver, CO, USA Clinician Researcher, Kaiser Permanente and UCHSC
Send response to journal:
|
After an analysis of demographic factors associated with access to different medical specialties, Dr. Ferrer concludes that primary care clinicians provide a disproportionate share of ambulatory care to disadvantaged populations and to those with chronic conditions (though it appears that this applies primarily to FPs and mid-levels, not pediatricians and general internists). Therefore from a policy perspective, primary care in general, and family medicine in particular, is essential to the continued care of all demographic groups. We have to be a little careful not to leap to the assumption that just because FPs provide more care to disadvantaged populations than do the other provider groups examined, that subsequent policies to increase numbers of FPs could reduce health disparities by increasing visit access for these at-risk populations. This approach assumes that a major problem behind health disparities is that disadvantaged populations are getting insufficient access to care due to insufficient numbers of primary care clinicians. But disadvantaged populations are complex and are ‘disadvantaged’ for multiple reasons. For example, in this analysis the really low odds of access to all providers were for uninsured patients; thus encouraging me to leap to another assumption that it may be ineffective to advocate for improvements in health disparities via changes in the primary care workforce in the absence of substantial health- insurance insurance reform. In order to explore these issues, I am curious about possible interactions between the independent variables that describe demographic characteristics. Although looking at interactions runs the risk of creating a complicated picture, that is precisely the point—many of these characteristics that describe ‘disadvantaged’ populations are likely to be associated with each other as well as with the ‘outcome’ of provider type. For example, what does relative access to these providers look like if the data are stratified by MSA/non-MSA, or by insured/uninsured? Will that further inform policy decisions on the overall role for primary care? Or the need for broader insurance coverage? Or primary and specialty reimbursement from public insurance programs? This article is an intriguing contribution to characterizing ‘disadvantaged’ populations and determining their care needs. Studies such as this can serve as starting points to further explorations in which we combine population-level data with patient-level data to help us define more accurately the care needs of various complex patient populations. Competing interests: None declared |
|||
|
|
|||
|
George S. Rust, Atlanta, GA USA Professor of Family Medicine, National Center for Primary Care, Morehouse School of Medicine
Send response to journal:
|
In his recent article (“Pursuing Equity”), Dr. Ferrer makes the very important point that primary care clinicians (especially family physicians) are a critical source of care for those who are disadvantaged. As such, primary care clinicians are an essential buffer that improves homeostasis in a fragmented and dysfunctional health care system by reducing extreme inequities. The converse is also compelling – our nation’s failure to support primary care as a national priority may exacerbate health care inequities and worsen health outcomes for our diverse U.S. population.
One caution about this type of study -- there is a tendency to think of “poor, black, and uninsured” all in one breath, without recognizing the heterogeneity within each racial or ethnic group. Instead, it is important to make a distinction in this type of analysis between modifiable predictors of access and equity (poverty, uninsurance, residential segregation, access to affordable and culturally-relevant care, etc.), vs. risk factors (race, ethnicity, gender, etc.) that are not modifiable either by patients or by society. We published just such an analysis(1) of MEPS data in 2004, looking at advantaged vs. disadvantaged individuals (uninsured, poor, and no usual source of care) solely within the African-American population, to demonstrate the powerful impact of these modifiable risks on utilization even within a minority population characterized by a high level of adverse outcomes attributable to racial health disparities(2). We found up to six-fold variation in utilization of office visits and prescription medicines when comparing well-insured, middle & upper-income African American respondents vs. African Americans who were poor and uninsured, as well as dramatic differences in the ratio of ED visit rates to hospital admission rates.
Bottom line? Racial disparities in patterns of health utilization are often closely related to factors we as a society could fix, if we so chose. From Dr. Ferrer’s analysis, we can add primary care access to the list, along with previously known factors such as uninsurance, individual and neighborhood poverty, residential segregation, and education(3).
1. Rust G, Fryer GE, Strothers H, Daniels E, McCann J, Satcher D. Modifiable Determinants of Health Care Utilization within the African-American population. J National Medical Assoc, Sept. 2004; 96(9):1169-1177.
2. Satcher D, Fryer GE, McCann J, Troutman A, Woolf S, Rust G. What If We Were Equal? A Comparison of the Black-White Mortality Gap in 1960 and 2000. Health Affairs, March 2005.
3. Williams DR, Jackson PB. Social sources of racial disparities in health. Health Aff (Millwood). 2005 Mar-Apr;24(2):325-34.
Competing interests: None declared |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
TRACK: