This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Environmental Contexts of Ultimate Decisions: Why White Nursing Home Residents Are Twice as Likely as African American Residents to Have an Advance Directive

¹ Departments of Economics and Health Behavior and Administration, University of North Carolina at Charlotte.
² Departments of Sociology and Anthropology and Communication, George Mason University, Fairfax, Virginia.

Address correspondence to Jennifer L. Troyer, PhD, Department of Economics, University of North Carolina at Charlotte, Charlotte, NC 28223. E-Mail: jtroyer{at}uncc.edu

	Abstract

TOP Abstract Methods Results Discussion References

 
Objectives. The purpose of this study was to determine the extent to which observed differences between White and African American nursing home residents in having an advance directive are attributable to differences between the groups in personal characteristics, the organizational environment of the nursing home, and the geographical environment of the counties in which the nursing homes are located.

Methods. By using the Medical Expenditure Panel Survey Nursing Home Component matched with county-level measures from the Area Resource File, we modeled the probability of having an advance directive as a function of nursing home resident, facility, and county characteristics for African American and White residents.

Results. The probability of having an advance directive was 27.0% for African American residents and 63.6% for White residents. Nearly half of this 36.6 percentage point gap could be explained by group differences in personal, facility, and county characteristics.

Discussion. County characteristics play a more prominent role than do personal or facility measures in explaining the observed ethnic gap in the prevalence of advance directives. Additional studies should focus further on geographic, health status, and attitudinal variations among nursing home residents that may account for the remaining ethnic difference in the prevalence of advance directives among nursing home residents.

IDEALLY, well before someone in long-term care begins entering a trajectory toward death, the individual, family members, and health care providers should engage in ongoing discussions about treatment options (Travis et al., 2002). A part of this advance care planning process may be the establishment of one or more advance directives—documents that specify someone's wishes regarding treatment decisions or identify a proxy for health care decisions if the individual should become incapable of making decisions. Proponents of advance care planning and advance directives have argued that they extend autonomy (Davidson, Hackler, Caradine, & McCord, 1989; High, 1987), an argument that contributed to the passage of the Patient Self-Determination Act of 1991. The Act requires that persons in health care and long-term care institutions reimbursed by Medicare or Medicaid, including nursing homes, be informed of their rights to engage in advance care planning and to establish advance directives. Recent studies have found that efforts to encourage advance care planning in nursing homes increased markedly after passage of the Act (Molloy et al., 2000).

Since the promulgation of the legislation, several authors have documented ethnic differences in the use of advance directives (Castle & Mor, 1998; Degenholtz, Arnold, Meisel, & Lave, 2002; Kellogg & Ramos, 1995; Kiely, Mitchell, Marlow, Murphy, & Morris, 2001; McAuley & Travis, 2003; O'Brien et al., 1997; Suri, Egleston, Brody, & Rudberg, 1999). Compared to White residents, African American nursing home residents, in particular, tend to be much less likely to have an advance directive and are more likely to desire aggressive interventions near the end of life. This finding is robust, even when controlling for other pertinent factors. Clearly, there are many reasons why African American individuals may be less likely than White individuals to establish an advance directive, including their historical mistreatment in health care and their resulting mistrust of health professionals (Caralis, Davis, & Wright, 1993; Dula, 1994), as well as their religious beliefs (McAuley, Pecchioni, & Grant, 2000). Such factors should be understood at the individual level and respected. However, knowing more about the basis for group differences in the use of advance directives between White and African American individuals is also important. Therefore, the major objective of this research was to determine the extent to which ethnic differences in advance care planning are attributable to differences in the personal characteristics of African American and White nursing home residents, the facilities in which they reside, and the counties in which the facilities are located. To our knowledge, this is the first study to consider the factors that may underlie White–African American differences in advance care planning with a nationally representative sample of nursing home residents while controlling for resident, facility, and county characteristics.

We founded the conceptual framework for this investigation on the assumption that the process of advance care planning in long-term care is both multifaceted and complex, and, as a result, investigations of the probability of having an advance directive must take into account several categories of variables. Specifically, decisions about advance directives in nursing homes are potentially influenced by (a) numerous personal factors, (b) the micro-environmental characteristics of the facility, and (c) the broader social and economic environment represented by the county in which the facility is located. These three levels of influencing factors, discussed in more detail in the following paragraphs, have a long theoretical tradition in health services and health research, including the writings of René Dubos (1961, 1987). Dubos emphasized the interplay between proximate and distal social and physical external environments and the person's internal environment in shaping health, health care, and health-related decisions.

Much of the prior work on ethnic differences in the use of advance directives has focused on personal characteristics of the individual. Several studies (Bradley, Wetle, & Horwitz, 1998; Castle & Mor, 1998; McAuley & Travis, 2003) have identified positive associations between age and having an advance directive. Prior work (Bradley et al.; McAuley & Travis) has also shown that the following individuals are more likely than others to have an advance directive: (a) more educated individuals (McAuley & Travis), (b) men (O'Brien et al., 1997), (c) residents whose stay is reimbursed by Medicaid (Bradley et al.; McAuley & Travis; Suri et al., 1999), and (d) persons with a living child (Eleazer et al., 1996; McAuley & Travis). Furthermore, longer nursing home stays may be correlated with age, having more time to implement an advance directive, or having enhanced relationships with staff—all factors that could influence whether a patient establishes an advance directive. Studies have also shown that having a diagnosis or condition that is correlated with terminal decline, such as cancer, is related to having an advance directive (Castle & Mor; McAuley & Travis).

Regarding facility characteristics, there is evidence of ethnic segregation of residents into different types of nursing homes, with African American individuals being far more likely White individuals to reside in facilities with higher percentages of residents funded by Medicaid and with fewer nurses (Mor, Zinn, Angelelli, Teno, & Miller, 2004). Lower staffing levels—especially of professional staff—may result in staff spending less time with residents in general and providing limited supportive communication about advance care planning, thereby limiting the likelihood of patients having an advance directive. Castle and Mor (1998) studied a large sample of nursing home residents in 10 states in order to consider how facility characteristics influenced the probability that residents had an advance directive. They found that ownership type, chain affiliation, facility size, occupancy rate, nurse staffing levels, and the proportion of residents funded by Medicaid all impacted the probability that a resident would adopt an advance directive. Therefore, notwithstanding the probability that some individuals enter the facility with previously established advance directives, the nursing home environment may impact the overall prevalence of advance directives among residents.

Geographic characteristics clearly play a role in whether a nursing home resident establishes an advance directive. Castle and Mor (1998) found that the level of concentration in the local nursing home market and whether the state in which the facility was located had a prospective or retrospective Medicaid reimbursement scheme were significant for some types of advance directives. Others have identified significant state (Kiely et al., 2001) and regional (Levin et al., 1999) differences regarding advance directives. Buchanan, Bolin, Wang, Zhu, and Kim (2004) found that nursing home residents in rural facilities were, on average, much more likely to have an advance directive than urban dwellers and that African American residents were more likely than White residents to reside in urban facilities. Given both the strong evidence of geographic variation in the prevalence of advance directives among nursing facility residents, and the fact that both staff and residents are likely to originate relatively near facilities, it is reasonable to assume that there are geographic differences in such factors as income, poverty, education, or ethnic and age composition that could at least partially explain ethnic differences in the prevalence of advance directives. Some studies (Fisher & Wennberg, 2003; Fisher et al., 2000) demonstrate the power of location variables in the utilization and quality of acute care. In addition, other authors have considered the effect of geographically based measures on racial differences on home health care use (White-Means & Rubin, 2004), on medical care usage by disabled individuals (White-Means, 2000), and on self-rated health status (Cagney, Browning, & Wen, 2005). Therefore, although prior research has not specifically investigated this line of reasoning, we view geographic characteristics as potentially important factors in explaining African American–White differences in the prevalence of advance directives.

	METHODS

TOP Abstract Methods Results Discussion References

 
Data
Our analysis is based upon Round 1 of the Medical Expenditure Panel Survey Nursing Home Component (MEPS-NHC) matched with county-level measures obtained from the Area Resource File. The MEPS-NHC is a nationally representative sample of individuals who were residents of nursing homes on January 1, 1996. Its sample size (3,209 residents) is sufficiently large to allow for an in-depth study of our two target ethnic groups (White and African American individuals), and the MEPS-NHC has a battery of items regarding both residents and facilities. Furthermore, we are able to merge the MEPS-NHC with information regarding the county in which the facility is located. For detailed information on the design and methods of the MEPS-NHC, see Potter (2001). Of the 3,209 residents, we excluded 93 (3%) from our sample because they were of an ethnicity other than African American or White. In addition, we excluded 320 records because of missing data for some of the resident characteristics. Of the remaining 2,796 records, we eliminated 131 observations with missing facility data. The research data set for our current study included 2,665 residents in 730 facilities.

Measures
All results presented in this article take into account the complex sample survey design of the MEPS-NHC; they use the appropriate survey weights (in order to give unbiased estimators for all nursing home residents in the United States) and account for cluster sampling and stratification (in order to eliminate bias in standard error estimates). We used the Stata statistical software package (StataCorp, 2003) for all analyses. The presence of an advance directive was indicated if at least one of the following four types of advance directives was in the resident's record or chart on January 1, 1996: (a) a living will, (b) a do-not-resuscitate order, (c) a do-not-hospitalize order, or (d) limits on feeding, medication, or other treatments.

Resident Characteristics
Table 1 presents the variables we used in our analysis. We included two broad types of resident characteristics: demographic measures and measures of health status and functional ability. Due to the large number of missing values for education, we measured resident education in this study by using two binary variables: one for whether education was recorded, and one for whether graduation from high school was recorded.

Facility Characteristics and County Characteristics
We used measures similar to those employed by Castle and Mor (1998), discussed earlier, in order to address group differences in facilities. As we previously noted, although there is evidence of geographic variation in the use of advance directives, virtually no research has examined whether county characteristics are useful in explaining African American–White differences in the prevalence of advance directives. Counties are useful geographic divisions to employ in this type of analysis because of the widely available county-level measures through the Area Resource File. In addition to considering whether the nursing home was in a metropolitan versus nonmetropolitan county, we considered per capita income, education, poverty levels, racial composition, and the proportion of the population aged 65 and older in the facility's county. Most, but not all, nursing home residents originate in the county in which the facility is located (McAuley, Pecchioni, & Grant 1999; McAuley & Usita, 1998). Therefore these measures were acceptable controls for the socioeconomic differences at the county level that may shape the geographic environment in which communication and decision making about advance directives occurs.

Probit Estimates: Full Sample
In order to more closely examine the effect of ethnicity on the gap in the proportion of White and African American individuals with advance directives, we first modeled the probability of having an advance directive as a function of resident, facility, and geographic characteristics for all African American and White residents. The estimated marginal effect associated with the African American binary variable can, for a resident with average characteristics, reveal the degree to which being African American decreases that resident's probability of having an advance directive when controlling for other factors.

Probit Estimates Using Subsamples: African American and White Individuals
Following the single-equation probit estimation, we subdivided the data into two samples: White residents and African American residents. By using these subsamples, we estimated probit models of the probability of having an advance directive. We performed these calculations separately for two reasons. First, separate estimates of marginal effects can reveal whether personal, facility, and county characteristics have the same impact for each group on the likelihood of having an advance directive. Second, the separate estimates may be used in conjunction with subsample means in order to determine how much of the White–African American difference in the probability of having an advance directive may be attributable to differences in measured average group characteristics.

More formally, the outcome of interest is whether the resident has any advance directive. The probability of having an advance directive is specified

where X_iR represents individual, facility, and county characteristics attributable to individual i with ethnicity type R, Y_iR = 1 implies that the individual has an advance directive, ß is the vector of parameters, and represents the cumulative density function for the standard normal. We used the estimated coefficients, , and sample means in order to compute estimated marginal effects. For continuous variables in X_iR, such as county-level per capita income, the marginal change in the k^th continuous variable, x_k, on the probability that Y_iR = 1 is

where is the normal probability density function and is the mean value of X over the sample used to estimate the model. For binary variables in X_iR, the effect of switching the k^th binary variable from 0 to 1 is

where x_ik1 indicates that the binary variable takes a value of 1, x_ik0 indicates that the variable takes a value of 0, and the remaining elements of X_iR are set at the sample mean. For example, the estimated marginal effect associated with the African American binary variable reveals the degree to which being African American increases (or decreases, if negative) a resident's probability of having an advance directive for a resident with otherwise average characteristics.

Using the estimates of ß, which for the African American sample is _B and for the White sample is _W, the vectors of African American (X_iB) and White (X_iW) characteristics, and the size of the African American (n_B) and White (n_W) samples, the degree to which the gap in the probability of having an advance directive is explained by differences in measured characteristics of African American and White residents as follows:

where

In words, Explained indicates the degree to which the gap in the probability of having an advance directive is explained by differences in the measured characteristics of African American and White residents and the facilities and counties in which they are situated. Similar to methods presented by Cotton (1988) and Neumark (1988) for continuous dependent variables, we used a technique for use with binary dependent variables found in Troyer (2002) in order to further decompose Equation 4. Doing so allowed us to determine the amount of the Explained portion of the gap that was attributable to each explanatory variable, where the contribution of each explanatory variable depended on the ethnic group differences in the means for that variable and the effect of that variable on the probability of having an advance directive.

	RESULTS

TOP Abstract Methods Results Discussion References

 
Table 1 presents sample means for African American and White residents for each of the analytical variables. As shown, 63.4% of all White residents had at least one advance directive, but only 27.0% of African American residents had any advance directive. This implies that without controls, White residents are 2.35 times more likely than African American residents to have an advance directive. Other differences in the average characteristics of African American and White residents displayed in Table 1 suggest that it would be useful to explore empirically how these differences affect the probability of having an advance directive.

The second and third columns of Table 2 contains the marginal effects and the p values from the probit model for the full sample. Compared with White residents, African American residents were approximately 23.0% less likely to have an advance directive, when controlling for other factors that may affect having an advance directive and assuming that the characteristics of White and African American residents have the same impact on advance directive implementation.

Given the relatively large number of explanatory variables and the relatively small number of African American residents in the sample (218), many of the coefficients for the explanatory variables were not significant at the conventional.05 level. As with the problem of multicollinearity, when one has a micronumerosity problem, the estimated coefficients on the explanatory variables remain unbiased, but the confidence intervals tend to be wide. However, the decomposition of the gap in the probability of having an advance directive does not require statistically significant coefficient estimates.

Table 3 contains information on the size of the ethnic gap in the probability of having an advance directive, the extent to which the gap is explained by differences in personal, facility, and county characteristics, and the degree to which each explanatory variable helps to explain group differences in the probability of having an advance directive. Of the 36.6% gap between African American and White residents in the probability of having an advance directive, 44.0% (or 16.0 of the 36.6 percentage points) could be explained by differences in the average characteristics of African American and White residents.

Among facility characteristics, the most important variable in explaining the probability gap was the proportion of Medicaid recipients in the nursing home. A higher Medicaid census was associated with a lower probability of having an advance directive, and African American residents tended to be in homes with higher proportions of Medicaid residents.

In general, demographic differences were more important than resident health status variables in explaining the probability gap between groups. The three most important demographic measures were possession of a high school diploma, the presence of a living child, and being aged 85 or older. In all three cases, these factors increased the probability of having an advance directive, and White residents were more likely than African American residents to possess the characteristics.

It is important to note that some of the characteristics actually served to increase the difference in use of advance directives (negative signs on the contribution value in Table 3). African American residents were on average more likely than White residents to be diagnosed with Alzheimer's disease, which had a small positive effect on the probability of having an advance directive. Thus, ethnic group differences in the prevalence of Alzheimer's disease did not help to explain the lower average rates of implementation of advance directives for African American residents.

	DISCUSSION

TOP Abstract Methods Results Discussion References

 
Building upon previous documentation of African American–White differences in the prevalence of advance directives among nursing home residents, we based this research on a conceptual framework following the tradition of René Dubos (1961, 1987) in emphasizing the importance of examining the personal, institutional, and geographic factors that might explain these differences. Our work goes several steps further than prior research by using a nationally representative sample, incorporating all three levels of potential explanatory factors, and determining the relative impact of personal, institutional, and geographic factors on the African American–White difference in the use of advance directives. Although group differences in the prevalence of advance directives persist after controlling for these factors, 44.0% of the gap in the probability of having an advance directive is explained by differences in the measured group characteristics.

A large portion of the explained difference comes from group differences in facility and county characteristics. Nearly all prior work on ethnic differences in advance directives has failed to include facility-level measures, and, to our knowledge, this is the first study to attempt to quantify and assess the impact of a battery of county-level characteristics. We recognize that all work that uses environment-level measures has a potential causality problem, whereby conclusions about individual resident choices about advance directives may be correlated with aggregate county-level measures of ethnicity, poverty, income, and so on, without being caused by these measures. As such, readers must interpret with care the effects of these measures.

The most important facility characteristic in explaining the probability gap is the proportion of Medicaid recipients in the nursing home. A higher Medicaid census was associated with a lower probability of having an advance directive, and African Americans tended to be in homes with higher proportions of Medicaid residents. In the face of high Medicaid populations (and related lower average revenue), facilities may be unable to commit the resources necessary to support residents in advance care planning.

Given the impact of institutional and geographic characteristics, future research should focus on continuing to identify the underlying location- and facility-specific factors that influence differences use of in advance directives. Among the facility factors that may be important to consider are: (a) the type and amount of staff training and whether there are special internal policies or programs regarding when or how to provide information about advance care planning, (b) administrative and staff attitudes about advance directives, and (c) whether staff deal differently with White and African American individuals in the presentation of advance directive options.

Among the underlying location-specific factors that may influence African American–White differences in the utilization of advance directives are social norms, differential availability of and access to health care and long-term care, and other factors that generate incentives or disincentives by ethnic group to engage in advance care planning. Along similar lines, future research should continue to pursue the development of richer measures that help to explain geographic differences in advance directives that are correlated with ethnicity in the United States. Geographic variables such as physician supply, hospital characteristics, relationships of hospitals and nursing homes, and availability of home health agencies may influence local nursing home populations and their use of advance directives. One avenue for such research is the inclusion of information about the county of prior residence, in addition to characteristics of the county in which the facility is located, because the facility county is not always the nursing home resident's county of origin. Inclusion of appropriate state-level variables might also improve researchers' understanding of African American–White differences in the use of advance directives. Still another avenue for research into geographic factors is the inclusion of finer geographic categories, such as census tracts or zip codes, which should permit refinements in the definition of geographic location (Morrill, Cromartie, & Hart, 1999). Also, researchers have developed methods of determining nursing home markets based upon geographic units that are smaller than counties (Zwanziger, Mukamel, & Indridason, 2002). Inclusion of this finer geographic detail should improve the operationalization of both market areas of facilities and communities of residence.

One of the limitations of this study is the incomplete set of resident characteristics in the MEPS-NHC. For example, if more detailed health status measures had been available, we might have found that differences in diagnoses would have played a more prominent role in explaining ethnic group differences in the use of advance directives. In addition, the fact that a considerable percentage of the difference in White–African American implementation of advance directives remains in spite of the introduction of numerous individual, facility, and geographic variables points to the importance of ethnically relevant, health-related personal factors in the differential utilization of advance directives. These factors might include attitudes about death and dying (Waters, 2001); religious values (Gamble, 1993; Roberson, 1985; Waters); historically based concerns on the part of African Americans about the true intentions of those who provide information about advance directives and related apprehensions about the implications of signing advance directives (Dula, 1994; Waters); and expectations that such decisions should be made by a trusted friend or family member (Waters). Further research should consider the extent to which these issues affect the differences in the utilization of advance directives. This information could ultimately be used to generate more ethnically sensitive approaches to providing information about advance directives in long-term care facilities. One other potential limitation of this study is that the MEPS-NHC data were collected in January 1996. It would be very useful to consider more recent information on advance directives, though there is evidence that the percentage of nursing home residents with advance directives has not increased since this period (McAuley & Travis, 2003). Some imprecision in the analyses may be introduced by the fact that we did not statistically account for the multilevel data, although this problem is most likely to affect significance levels, which are not critical to our work. In a similar vein, it would be valuable to examine rates of advance directive use across similar counties and facilities, which would require a larger sample than is available through the MEPS-NHC. Finally, we should also note that having an advance directive on file does not necessarily mean that it will be acted upon. Advance directives are, at best, limited indicators of the current outcomes of end-of-life discussions that should accompany health care and long-term-care decision making in old age.

	Acknowledgments

 
We would like to thank D. E. B. Potter from the Agency for Healthcare Research and Quality for assisting us in gaining access to, and for providing information about, the Medical Expenditure Panel Survey Nursing Home Component data. All conclusions and analyses are those of the authors.

	Footnotes

 
Decision Editor: Charles F. Longino Jr., PhD

Received for publication March 30, 2005. Accepted for publication December 1, 2005.

	References

TOP Abstract Methods Results Discussion References

 

• Bradley, E., Wetle, T., Horwitz, S. (1998). The Patient Self-Determination Act and advance directive completion in nursing homes. Archives of Family Medicine, 7,417-423.[Abstract/Free Full Text]
• Buchanan, R., Bolin, J., Wang, S., Zhu, L., Kim, M. (2004). Urban/rural differences in decision making and the use of advance directives among nursing home residents at admission. Journal of Rural Health, 20,131-135.[Medline]
• Cagney, K. A., Browning, C. R., Wen, M. (2005). Racial disparities in self-related health at older ages: What difference does the neighborhood make? Journal of Gerontology: Social Sciences, 60B,S181-S190.[Abstract/Free Full Text]
• Caralis, P. V., Davis, B., Wright, K. (1993). The influence of ethnicity and race on attitudes toward advance directives, life-prolonging treatments, and euthanasia. Journal of Clinical Ethics, 4,155-165.[Medline]
• Castle, N., Mor, V. (1998). Advance care planning in nursing homes: Pre- and post-Patient Self-Determination Act. Health Services Research, 33,101-124.[Medline]
• Cotton, J. (1988). On the decomposition of wage differentials. Review of Economics and Statistics, 70,236-243.
• Davidson, K., Hackler, C., Caradine, D., McCord, R. (1989). Physicians' attitudes on advance directives. Journal of the American Medical Association, 262,2415-2419.[Abstract/Free Full Text]
• Degenholtz, H., Arnold, R., Meisel, A., Lave, J. (2002). Persistence of racial disparities in advance care plan documents among nursing home residents. Journal of the American Geriatrics Society, 50,378-381.[Medline]
• Dubos, R. (1961). Mirage of health: Utopias, progress, and biological change. Garden City, NY: Doubleday.
• Dubos, R. (1987). Man, medicine, and environment. New York: New American Library.
• Dula, A. (1994). African American suspicion of the health care system is justified: What do we do about it? Cambridge Quarterly of Health care Ethics, 3,347-357.[Medline]
• Eleazer, G. P., Hornung, C. A., Egbert, C. B., Egbert, J. R., Eng, C., Hedgepeth, J., et al. (1996). The relationship between ethnicity and advance directives in a frail older population. Journal of the American Geriatrics Society, 44,938-943.[Medline]
• Fisher, E. S., Wennberg, J. E. (2003). Health care quality, geographic variations, and the challenge of supply-sensitive care. Perspectives in Biology and Medicine, 46,69-79.[Medline]
• Fisher, E. S., Wennberg, J. E., Stukel, T. A., Skinner, J. S., Sharp, S. M., Freeman, J. L., et al. (2000). Associations among hospital capacity, utilization, and mortality of U.S. Medicare beneficiaries, controlling for sociodemographic factors. Health Services Research, 34,1351-1362.[Medline]
• Gamble, V. N. (1993). A legacy of distrust: African Americans and medical research. American Journal of Preventive Medicine, 9,35-38.[Medline]
• High, D. (1987). Planning for decisional incapacity: A neglected area in ethics and aging. Journal of the American Geriatrics Society, 35,814-820.[Medline]
• Kellogg, F., Ramos, A. (1995). Code status decision-making in a nursing home population: Processes and outcomes. Journal of the American Geriatrics Society, 43,113-121.[Medline]
• Kiely, D., Mitchell, S., Marlow, A., Murphy, K., Morris, J. (2001). Racial and state differences in the designation of advance directives in nursing home residents. Journal of the American Geriatrics Society, 49,1346-1352.[Medline]
• Levin, J. R., Wenger, N. S., Ouslander, J. G., Zellman, G., Schnelle, J. F., Buchanan, J. L., et al. (1999). Life-sustaining treatment decisions for nursing home residents: Who discusses, who decides, and what is decided? Journal of the American Geriatrics Society, 47,82-87.[Medline]
• McAuley, W. J., Pecchioni, L., Grant, J. (1999). Admission-related migration by older nursing home residents. Journal of Gerontology: Social Sciences, 54B,S125-S135.[Abstract]
• McAuley, W. J., Pecchioni, L., Grant, J. A. (2000). Personal accounts of the role of God in health and illness among older rural African American and White residents. Journal of Cross-Cultural Gerontology, 15,13-35.[Medline]
• McAuley, W. J., Travis, S. S. (2003). Advance care planning among residents in long-term care. American Journal of Hospice and Palliative Care, 20,353-359.
• McAuley, W. J., Usita, P. (1998). A conceptual model for the mobility patterns of nursing home admissions. The Gerontologist, 38,726-734.[Abstract]
• Molloy, D. W., Guyatt, G. H., Russo, R., Goeree, R., O'Brien, B. J., Bedard, M., et al. (2000). Systematic implementation of an advance directive program in nursing homes: A randomized controlled trial. Journal of the American Medical Association, 283,1437-1444.[Abstract/Free Full Text]
• Mor, V., Zinn, J., Angelelli, J., Teno, J. M., Miller, S. C. (2004). Driven to tiers: Socioeconomic and racial disparities in the quality of nursing home care. Milbank Quarterly, 82,227-256.[Medline]
• Morrill, R., Cromartie, J., Hart, L. G. (1999). Metropolitan, urban, and rural commuting areas: Toward a better depiction of the U.S. settlement system. Urban Geography, 20,727-748.
• Neumark, D. (1988). Employers' discriminatory behavior and the estimation of wage discrimination. Journal of Human Resources, 23,279-295.
• O'Brien, L. A., Siegert, E. A., Grisso, J. A., Maislin, G. M., LaPann, K., Evans, L. K., et al. (1997). Tube feeding preferences among nursing home residents. Journal of General Internal Medicine, 12,364-371.[Medline]
• Potter, D. E. B. (2001). Methodology report no. 3: Design and methods of the 1996 MEPS nursing home component. Rockville, MD: Agency for Healthcare Research and Quality. Retrieved October 26, 2004, from http://www.meps.ahrq.gov/Papers/MR3_98-0041/MR3.htm.
• Roberson, M. H. B. (1985). The influence of religious beliefs on health choices of Afro-Americans. Topics of Clinical Nursing, 7,57-63.
• StataCorp. (2003). Stata statistical software: Release 8.0. College Station, TX: Author.
• Suri, D., Egleston, B., Brody, J., Rudberg, M. (1999). Nursing home resident use of care directives. Journal of Gerontology: Medical Sciences, 54A,M225-M229.[Abstract]
• Travis, S. T., Bernard, M., Dixon, S., McAuley, W. J., Loving, G., McClanahan, L. (2002). Obstacles to palliation and end-of-life care in a long-term care facility. The Gerontologist, 42,342-349.[Abstract/Free Full Text]
• Troyer, J. (2002). Decomposing the effect of marital status on migration. Applied Economics Letters, 9,641-644.
• Waters, C. M. (2001). Understanding and supporting African Americans' perspectives of end-of-life care planning and decision making. Qualitative Health Research, 11,385-398.[Abstract/Free Full Text]
• White-Means, S. I. (2000). Racial patterns in disabled elderly persons' use of medical services. Journal of Gerontology: Social Sciences, 55,S76-S89.[Abstract/Free Full Text]
• White-Means, S. I., Rubin, R. M. (2004). Is there equity in the home health care market? Understanding racial patterns in the use of formal home health care. Journal of Gerontology: Social Sciences, 59B,S220-S229.[Abstract/Free Full Text]
• Zwanziger, J., Mukamel, D. B., Indridason, I. (2002). Use of resident-origin data to define nursing home market boundaries. Inquiry, 39,169-172.

This article has been cited by other articles:

				H. Temkin-Greener, N. Zheng, S. A. Norton, T. Quill, S. Ladwig, and P. Veazie Measuring End-of-Life Care Processes in Nursing Homes Gerontologist, July 2, 2009; (2009) gnp092v1. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation