Racial/Ethnic Disparities in Preventive Care Practice Among U.S. Nursing Home Residents

Abstract

Objective: To assess racial/ethnic disparities in preventive care practices among U.S. nursing home residents. Method: To implement the Institute of Medicine definition of health care disparity, we used the rank-and-replace adjustment method to assess the disparity in receipt of eight preventive care services among residents and evaluate trends in disparities. The sampling design (stratification and clustering) was accounted for using Stata 11. Results: The 2004 National Nursing Home Surveys data show White residents were more likely to have pain management, scheduled toilet plan/bladder retraining, influenza vaccination, and pneumococcal vaccination than Black residents. White residents were also more likely to have scheduled toilet plan/bladder retraining than residents of Other race/ethnicity. Significant Black–White disparities in receipt of influenza vaccination and pneumococcal vaccination were found. Time trend analysis showed that disparities were neither exacerbated nor reduced. Conclusion: Persistent racial/ethnic disparities in preventive care among nursing home residents exist. We urge the development and implementation of targeted interventions to improve the quality of preventive care in nursing homes.

Keywords

disparity preventive care nursing homes

Introduction

Nursing home quality has been a serious concern for several decades (Mor, Zinn, Angelelli, Teno, & Miller, 2004). To improve quality of care, market-based reforms such as publicly rating the care provided in nursing homes were initiated in the late 1990s and the Nursing Home Compare online went live in 2002 (Centers for Medicare and Medicaid Services[CMS], 2002). Yet some have expressed concern that these initiatives and other market-based reforms may have had the unintended consequence of exacerbating disparities in care among nursing home residents (Casalino et al., 2007; Chien, Chin, Davis, & Casalino, 2007; Mor et al., 2004), including segregation of nursing homes into a two-tier system (Mor et al., 2004). Limited data exist on whether disparities in preventive care between Whites and Blacks and Whites and residents of Other race/ethnicity widened or narrowed in the past decades. To our knowledge, only one study provided descriptive data on the trend of influenza vaccination in U.S. nursing homes, and it only estimated the crude prevalence of influenza vaccination among White and Black residents without adjusting for other factors (B. H. Bardenheier, Wortley, & Shefer, 2009). The trend over time of pneumococcal vaccination by race/ethnicity has not been examined. Furthermore, most prior nursing home disparity research focused on preventive care for specific diseases. Such studies found that Black and Hispanic diabetic residents were less likely to receive antidiabetic medications than non-Hispanic Whites (Allsworth, Toppa, Palin, & Lapane, 2005; Spooner, Lapane, Hume, Mor, & Gambassi, 2001); that Asian/Pacific Islanders, Blacks, and Hispanics at risk for secondary stroke received anticoagulants less often than non-Hispanic White residents (Christian, Lapane, & Toppa, 2003; Quilliam & Lapane, 2001); and that Black nursing home residents on antipsychotic drugs in Arkansas were less likely than their White counterparts to take the medication (Hudson, Cody, Armitage, Curtis, & Sullivan, 2005). Our study extends prior research by examining racial disparities in general preventive care services in nursing homes.

Most prior research on racial/ethnic disparity of care in nursing homes generally did not adequately consider the role of mediating factors such as underlying health conditions and health care system–related variables (Grabowski & McGuire, 2009). In the Institute of Medicine (IOM) Unequal Treatment report (IOM, 2003), a disparity is defined as a difference in treatment provided to members of different racial or ethnic groups that is not justified by the underlying health conditions or treatment preferences of patients. Our study assessed racial disparity in preventive care practice in U.S. nursing homes using the 2003 IOM definition of disparity. The objectives of this study are to assess the disparities in receipt of preventive care among nursing home residents of different racial groups and to evaluate the trend of disparity in receipt of influenza and pneumococcal vaccinations over the past decade.

Method

Conceptual Framework

According to the 2003 IOM report, a difference in health care could be due to three factors:

1) clinical appropriateness of the intervention, patient needs, and patient preferences; 2) the operation of healthcare systems and the legal and regulatory climate in which healthcare systems function; and 3) discrimination at the individual, patient-provider level, including biases, prejudices, stereotyping, and uncertainty in clinical communication and decision-making (IOM, 2003, p. 49).

Furthermore, the report defines disparities in health care as racial or ethnic differences in the quality of health care that are due to Factors 2 and 3. Differences in care due to Factor 1—appropriateness of intervention, patient need, and patient preference—are not considered to be part of the disparity.

Prior research (Cook, McGuire, & Miranda, 2007; Cook, McGuire, & Zaslavsky, 2012; McGuire, Alegria, Cook, Wells, & Zaslavsky, 2006) has operationalized clinical appropriateness and need as individual health status, health conditions, or characteristics predictive of health status, such as age and sex, and are referred to as need variables. In survey data, differences due to measures of health care system characteristics (health insurance, location of nursing home [urban/rural], and other nursing home characteristics) can be considered to fit into the operation of health care systems category. For example, if Medicaid-enrolled residents receive poorer nursing home services, and racial/ethnic minority residents are disproportionately enrolled in Medicaid, then the operation of the health care system may be affecting disparities through this system-level variable. The independent effect of race and ethnicity (operationalized by the race/ethnicity indicator coefficient) is commonly used as a proxy of discrimination (Blank, Dabady, & Citro, 2004). Patient preference data are scarce because this variable has seldom been surveyed.

To implement the IOM definition of disparity, we estimated the disparity of preventive care among three racial/ethnic groups (non-Hispanic White, non-Hispanic Black, and Other race/ethnicity), allowing the health care system–related factors, but not the need factors, to contribute to the disparity calculation.

Data Source

Data for this analysis were from the four National Nursing Home Surveys (NNHS) in 1995, 1997, 1999, and 2004. The NNHS is a two-stage, cross-sectional probability sampling survey of U.S. nursing homes (Gabrel & Jones, 2000a, 2000b; Jones, Dwyer, Bercovitz, & Strahan, 2009). The series of NNHS have collected facility information and data on current residents. All nursing homes included in this survey had at least three beds and were certified by Medicare or Medicaid or had a state license to operate as a nursing home.

In each NNHS, a sample of nursing homes was selected from a data set (universe file) listing all U.S. nursing homes. Each universe file was constructed primarily from provider data for the period from the CMS and state licensing lists. In each participating nursing home, NNHS interviewers drew a random sample of current residents at the time of the interview. Up to 6 current residents were sampled at each nursing home in the 1995, 1997, and 1999 surveys and 12 current residents in the 2004 survey. Detailed information on each resident was collected from staff members at the nursing home, typically registered nurses, from medical records, and other documentation. Residents were not directly interviewed. Data collected included residents’ demographic characteristics, date of admission, current functional status, and admission and current diagnoses. There were 8,056, 8,138, 8,215, and 13,507 residents included in the 1995, 1997, 1999, and 2004 NNHS respectively.

Measures

Outcome variables

We examined residents’ receipt of eight preventive care services (yes = 1, no = 0) using 2004 NNHS data including pain management, behavioral problem management, skin/wound care, continence management, restorative care, scheduled toilet plan/bladder retraining to manage incontinence, influenza vaccination, and pneumococcal vaccination. We further evaluated the trends of influenza vaccination and pneumococcal vaccination using 1995, 1997, 1999, and 2004 NNHS data.

Covariates

We classified the residents into three racial/ethnic groups: non-Hispanic White, non-Hispanic Black, and Other race/ethnicity (“Other race” for short). Following prior research (Cook et al., 2007; Cook et al., 2012; McGuire et al., 2006), the need variables we included were age (<65, 65-74, 75-84, and 85+ years), sex, marital status (married/not married), and length of stay (<1, 1-<2, 2-<4, 4 or more years). The intensity of residents’ needs in five activities of daily living (ADLs, transferring, eating, toileting, dressing, and bathing) was aggregated ≤2, 3, and 4-5). We only included four ADLs (eating, toileting, dressing, and bathing) in the trend analysis of influenza vaccination and pneumococcal vaccination from 1995 to 2004 because of a wording change for the transferring measure in the 2004 NNHS (Zhang et al., 2010). Residents’ comorbidities were identified from the recorded current or admission diagnoses using ICD 9 (the International Classification of Diseases, Ninth Revision, Clinical Modification) codes (nervous system, infectious/parasitic, cancer, endocrine/metabolic, blood, circulatory, respiratory, digestive, genitourinary, skin, musculoskeletal, injury and poisoning, other mental disorder, symptoms and ill-defined conditions, and supplementary classification). The total number of comorbidities for each resident was totaled and classified into three categories (≤2, 3-4, and ≥5 comorbidities).

We included the health care system–related variables that indicate a resident’s socioeconomic status in all of the four NNHS surveys. These variables include payment source (Medicaid, Medicare, private insurance, or other payments), characteristics of the nursing home (bed size ≤99, 100-199, 200+ beds; for-profit vs. non-profit status), and location (urban area vs. others; see Table 1).

Table 1.

Characteristics of White, Black, and Other Race/Ethnicity Nursing Home Residents: 2004 NNHS.

	White (n = 11,448)	Black (n = 1,384)	Other (n = 675)

	% (95% CI)	% (95% CI)	% (95% CI)
Need variables
Age
<65	9.46 [8.59, 10.33]***	21.78 [18.84, 24.73]	22.98 [18.37, 27.59]
65-74	10.14 [9.46, 10.81]	18.72 [16.44, 21.00]	18.73 [15.29, 22.16]
75-84	31.73 [30.64, 32.82]	29.03 [26.21, 31.86]	32.16 [28.00, 36.31]
≥85	48.68 [47.27, 50.08]	30.46 [27.68, 33.25]	26.13 [21.47, 30.80]
Female	73.16 [71.96, 74.36]***	63.29 [60.30, 66.29]	58.85 [54.62, 63.09]
Married	20.86 [19.88, 21.84]	16.76 [14.26, 19.25]	21.99 [17.84, 26.14]
Number of comorbidities
≤2	9.90 [8.92, 10.88]***	11.44 [9.00, 13.89]	12.66 [9.26, 16.07]
3-4	37.38 [36.01, 38.75]	43.23 [40.26, 46.20]	39.83 [35.33, 44.33]
≥5	52.72 [50.96, 54.48]	45.33 [41.71, 48.94]	47.51 [42.12, 52.90]
Number of ADL needs
≤2	14.20 [13.20, 15.19]	13.34 [10.85, 15.83]	15.54 [11.69, 19.38]
3	6.50 [5.96, 7.04]	5.26 [4.01, 6.51]	5.98 [3.68, 8.28]
4-5	79.30 [78.17, 80.44]	81.40 [78.62, 84.17]	78.48 [74.18, 82.79]
Length of stay (years)
<1	43.86 [42.52, 45.21]	43.26 [40.32, 46.19]	49.37 [44.69, 54.06]
1-<2	18.60 [17.70, 19.49]	17.17 [15.04, 19.30]	17.76 [14.43, 21.10]
2-<4	19.98 [19.04, 20.92]	20.80 [18.51, 23.10]	16.57 [13.51, 19.62]
≥4	17.56 [16.53, 18.58]	18.77 [16.29, 21.25]	16.30 [12.97, 19.62]
Health care system–related variables
Health insurance/primary payment
Medicaid only	26.49 [24.88, 28.10]	44.52 [39.82, 49.23]	46.79 [40.45, 53.14]
Medicare only	33.37 [31.80, 34.95]	33.09 [29.07, 37.11]	29.09 [23.39, 34.78]
Private insurance	12.62 [11.46, 13.79]	5.96 [4.17, 7.75]	5.96 [3.35, 8.58]
Other insurance/payment	27.51 [25.69, 29.33]	16.43 [12.61, 20.24]	18.16 [12.10, 24.22]
Bed size
≤99	34.35 [32.80, 35.90]***	21.77 [17.39, 26.16]	27.73 [21.1, 34.32]
100-199	52.38 [50.39, 54.37]	56.16 [49.85, 62.47]	53.25 [44.85, 61.64]
≥200	13.27 [11.56, 14.99]	22.07 [15.80, 28.34]	19.02 [11.68, 26.36]
Urban location	73.62 [72.69, 74.56]***	84.52 [81.40, 87.64]	86.47 [82.08, 90.86]
For profit ownership	59.22 [55.75, 62.69]***	75.20 [69.42, 80.98]	69.00 [60.16, 77.83]

Note. NNHS = National Nursing Home Surveys; CI = confidence interval; ADL = activity of daily living.

***

p < .001.

Statistical analysis

To implement the IOM definition of disparity of care, we conducted a three-step process, first on 2004 data to estimate the disparity in receipt of eight preventive care services, and then over multiple years of data to assess trends of disparities of receipt of influenza and pneumococcal vaccinations.

Using the 2004 NNHS survey data, we first estimated a multivariate logistic regression model to assess the probability of receiving each of the eight preventive care services as a function of all covariates (all variables in Table 1, including race). We included interaction terms (race × major payer source) and evaluated model fit statistics using the modified Hosmer–Lemeshow test (Archer & Lemeshow, 2006).

Second, we used the rank-and-replace adjustment method to create two counterfactual minority resident populations: Black race, Black distribution of health care system–related covariates, but White distribution of need covariates and Other race, Other race distribution of health care system–related covariates, but White distribution of need covariates (Cook et al., 2012; McGuire et al., 2006). This was implemented in three steps. In Step 1, a need-based score of each resident was defined as the sum of the terms (the coefficient times the covariate) of the fitted logistic regression model corresponding to need variables. In Step 2, individual residents were then assigned survey-weighted ranks within their race based on this need score. In Step 3, the need-based scores of each non-White resident (Black and Other race) were replaced by those of the equivalently ranked White resident. Thus, a resident of Black or Other race with a need-based score at the p-th percentile for Blacks or Other race would be reassigned the need-based score of the White resident at the p-th percentile for Whites.

Third, we estimated the predicted probability of having the eight preventive care services for the White group and two counterfactual populations, with the difference between the White and each non-White counterfactual population constituting the IOM-concordant disparity. To calculate standard errors, we drew simulated samples with replacement, using the bootstrap procedure (Efron, 1979), replicating the sampling design (stratification and clustering) of the NNHS to create 100 samples, and the variance of these 100 estimates was determined.

We re-estimated disparities using the same model specification as described above and executed the rank-and-replace adjustment to obtain disparity estimates of receipt of influenza and pneumococcal vaccinations in nursing homes using 1995, 1997, and 1999 NNHS surveys data. Using estimates and standard errors for each racial/ethnic group and each time point, we were able to evaluate disparities between years, within years (White residents vs. Black residents, and White residents vs. Other race residents), and trends in disparities (1995-2004). Data analyses were implemented using SVY routines in Stata 11. Results were considered significant if p value was less than .05.

Results

Table 1 presents nursing home resident population characteristics by the three racial groups in 2004. Non-Hispanic White residents were more likely to be in the age categories of 85 years and older (p < .001), female (p < .001), and have more comorbidities (p < .001) than non-Hispanic Black and Other race residents. They were more likely to reside in a smaller nursing home (≤99 beds; p < .001), but less likely to be in a nursing home located in an urban area (p < .001), and less likely to be in a nursing home that is for-profit (p < .001), than non-Hispanic Black and Other race residents.

Table 2 presents multiple logistic regression results for 2004 data. After controlling for need and health care system–related variables, non-Hispanic Black residents were less likely to have pain management (adjusted odds ratio [AOR] = 0.40, p < .01), scheduled toilet plan/bladder retraining (AOR = 0.66, p < .05), and influenza vaccination (AOR = 0.70, p < .05) than non-Hispanic White residents; Other race residents were less likely to have scheduled toilet plan/bladder retraining (AOR = 0.45, p < .001) than non-Hispanic White residents.

Table 2.

Multiple Logistic Regression Results of Association of Race, Need, and Health System–Related Variables With Selected Preventive Care Practices, 2004 NNHS.

	Pain management	Behavioral problem management	Skin/wound management	Continence management	Restorative care	Scheduled toilet plan/bladder retraining	Influenza vaccination	Pneumococcal vaccination
Race/ethnicity (vs. White)
Black	0.40**	0.62	0.76	0.77	0.89	0.66*	0.70*	0.70
Other	0.98	0.89	1.01	1.25	1.01	0.45***	0.83	0.94
Need variables
Age (vs. <65)
65-74	1.18	1.84**	1.38*	0.59**	1.07	0.46***	0.76**	0.75**
75-84	1.11	1.37	1.46**	0.78	1.09	0.70***	0.87	0.89
≥85	1.17	1.12	1.17	0.81*	1.11	0.78***	1.05	0.95
Number of ADL needs (vs. ≤2)
3	0.71*	0.81	0.36***	0.12***	0.36***	0.09***	0.95	1.03
4-5	0.89	1.18	0.530**	0.49**	0.68**	0.55***	0.94	1.01
Female	1.11	0.91	0.88	1.05	1.07	1.14*	0.98	1.02
Married	1.01	0.97	1.16	1.03	1.12	1.11	0.96	1.02
Number of comorbidities (vs.≤2)
3-4	0.71	0.58**	0.81	0.83	0.76*	0.88	0.92	0.83
≥5	0.86	0.97	0.81	0.98	0.87*	1.01	0.97	0.82***
Length of stay (vs. <1 year)
1-<2	1.21	0.78	1.64**	1.02	0.98	0.74***	0.37***	0.38***
2-<4	1.15	1.05	1.07	1.34*	0.98	1.10	0.91	0.74**
≥4	1.22	1.12	0.83	1.10	1.07	1.04	1.30*	0.87
Health system–related variables
Payment source (vs. Medicaid)
Medicare only	0.97	0.72*	1.18	0.70*	1.16	0.89	0.74**	0.96
Private insurance	0.95	0.69	1.54*	0.98	1.23	0.97	0.79	0.81*
Other insurance/payment	0.87	0.82	0.95	0.91	1.08	0.97	1.11	1.00
Bed size (vs. ≤99)
100-199	1.50	0.95	0.77	1.38	1.33	1.12	1.11	0.94
≥200	1.27	0.72	0.90	1.24	1.16	1.01	0.99	0.89
Urban location	0.84	1.13	0.94	0.79	0.80*	0.95	0.68***	0.92
For profit ownership	0.59**	0.79	0.95	0.67*	0.78*	0.79*	0.83	0.53***
Race by payment source interaction terms
Black × Medicare	1.42	1.08	1.22	1.54	1.13	1.05	0.93	0.86
Hispanic × Medicare	0.84	1.43	1.26	1.69	0.86	1.42	1.08	1.08
Black × Private	2.32	1.91	1.27	0.80	1.21	1.82	0.54	1.00
Hispanic × Private	0.54	NA	0.79	0.79	0.64	1.66	1.38	1.84
Black × Other	2.76*	1.14	1.26	1.16	0.97	0.72	0.98	0.72
Hispanic × Other	0.03**	0.40	NA	0.83	0.55	1.11	1.07	0.96

Note. NNHS = National Nursing Home Surveys; ADL = activity of daily living.

p < .05. **p < .01. ***p < .001.

Estimates of disparity in receipt of the eight preventive care services according to the IOM definition are summarized in Table 3. Non-Hispanic Black residents were 3.18 percentage points (p < .001) less likely to have pain management care, 9.23 percentage points (p < .001) less likely to have a scheduled toilet plan/bladder retraining, 9.16 percentage points (p < .001) less likely to have received influenza vaccination, and 13.02 percentage points (p < .001) less likely to have received pneumococcal vaccination than non-Hispanic White residents. In addition, Other race residents were 12.32 percentage points (p < .001) less likely to have a scheduled toilet plan/bladder retraining than non-Hispanic White residents.

Table 3.

Disparity in Receipt of Selected Preventive Care by Race: 2004 NNHS.

	Pain management	Behavioral problem management	Skin/wound management	Continence management	Restorative care	Scheduled toilet plan/bladder retraining	Influenza vaccination	Pneumococcal vaccination

	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)	% (SE)
White (n = 11,448)	6.71 (0.28)	5.64 (0.26)	6.40 (0.27)	6.08 (0.25)	18.84 (0.49)	34.26 (0.68)	76.96 (0.59)	56.20 (0.53)
Black (n = 1,384)	3.53 (0.53)	3.88 (0.53)	5.34 (0.71)	4.83 (0.58)	16.16 (1.22)	25.03 (1.36)	67.80 (1.74)	43.18 (1.64)
Other race (n = 675)	4.81 (0.98)	4.95 (1.04)	5.63 (1.11)	7.92 (1.33)	15.61 (1.57)	21.94 (1.91)	73.91 (2.09)	54.85 (2.65)
B–W difference	−3.18 (0.68)***	−1.75 (0.56)	−1.06 (0.77)	−1.25 (0.63)	−2.67 (1.24)	−9.23 (1.41)***	−9.16 (1.77)***	−13.02 (1.74)***
O–W difference	−1.90 (1.02)	−0.68 (1.10)	−0.77 (1.13)	1.85 (1.34)	−3.23 (1.69)	−12.32 (2.04)***	−3.05 (2.24)	−1.35 (2.67)

Note. NNHS = National Nursing Home Surveys; B = Black; W = White; O = Other.

p < .05. **p < .01. ***p < .001.

Figure 1 displays trends of receipt of influenza and pneumococcal vaccinations among non-Hispanic White, non-Hispanic Black, and Other race residents from 1995 to 2004 after adjustment for need variables. No significant trend was observed from 1995 to 2004 for these three racial groups. Among non-Hispanic White residents, the rates of influenza vaccination were 77.01%, 80.28%, 80.70%, and 76.96%, in 1995, 1997, 1999, and 2004, respectively; 76.53%, 74.58%, 79.40%, and 67.80% among non-Hispanic Black residents; and 79.97%, 75.36%, 78.56%, and 73.91% among Other race/ethnicity residents. Rates of pneumococcal vaccination, among non-Hispanic White residents, increased from 41.56% in 1995 and 50.00% in 1997 to 63.58% in 1999, but decreased to 56.20% in 2004. Among non-Hispanic Black residents, rates increased from 36.72% in 1995 and 41.14% in 1997 to 51.09% in 1999, but decreased to 43.18% in 2004. Among Other race residents, rates increased from 39.76% in 1995 and 44.84% in 1997 to 59.41% in 1999, but decreased to 54.85% in 2004. The 2004 influenza vaccination shortage could account for the reduction in 2004 (McQuillan et al., 2009).

Figure 1.

Trend of influenza and pneumococcal vaccinations coverage by race: 1995, 1997, 1999, and 2004 NNHS.

Figure 2 displays the disparity trend in receipt of influenza and pneumococcal vaccinations from 1995 to 2004 after adjustment for need variables. The overall Black–White disparity trend from 1995 to 2004 was not significant, though an upward trend was shown in disparity in receipt of vaccinations between 2004 and 1995. Specifically, the Black–White disparity in receipt of influenza vaccination increased from 0.48 percentage points in 1995 to 9.16 percentage points in 2004 (p < .001); for pneumococcal vaccination, the disparity increased from 4.84 percentage points in 1995 to 13.02 percentage points in 2004 (p < .001). The 2004 influenza vaccination shortage may have exacerbated the Black–White disparity in a time of scarcity.

Figure 2.

Trend of disparity in receipt of influenza and pneumococcal vaccination, 1995, 1997, 1999, and 2004 NNHS.

In addition, significant disparities were identified in the NNHS surveys in some, but not all, survey years. Specifically, non-Hispanic Black residents were less likely to have influenza vaccination than non-Hispanic White residents by 5.71 percentage points (p < .01) in 1997 and 9.16 percentage points (p < .001) in 2004. In addition, non-Hispanic Black residents were less likely to have pneumococcal vaccination than non-Hispanic White residents by 8.86 percentage points (p < .01) in 1997, 12.50 percentage points (p < .001) in 1999, and 13.02 percentage points (p < .001) in 2004 (Figure 2).

Discussion

Using the IOM disparity framework, this is the first study to examine disparity trends in receipt of influenza and pneumococcal vaccinations among White, Black, and Other race residents in the past decade. Moreover, this study found that White–Black disparities existed in receipt of four of the eight preventive care services examined. That is, fewer non-Hispanic Black residents had pain management, scheduled toilet/bladder retraining, and influenza and pneumococcal vaccinations than non-Hispanic White residents. In addition, residents of Other race/ethnicity were less likely to receive scheduled toilet/bladder retraining than White residents. These findings reinforce earlier evidence that there are persistent disparities in care between White and Black residents in nursing homes (Bernabei et al., 1999; Christian et al., 2003; Quilliam & Lapane, 2001).

Compared with the non-institutionalized population, nursing home residents are present at the facility and there are few non-medical costs for receiving preventive care (e.g., no need to drive to a clinic to obtain the vaccination). Therefore, access to care is not a major reason for the racial disparities in preventive care in nursing homes. While health care disparity in nursing homes has been well documented, the causes need more investigation. A better understanding of the sources of disparity would help in the design of interventions (Konetzka & Werner, 2009). Implementing the IOM definition of disparity, which recognizes the role of socioeconomic differences associated with race/ethnicity as mediators of disparities (Cook et al., 2012; McGuire et al., 2006), we found that health care system–related factors, including payment source and facility size and location, were significant contributors to racial/ethnic disparities in preventive care practice because (a) the racial/ethnic groups significantly differed on these characteristics and (b) because they were significant predictors of preventive care. For example, Black and Other race/ethnicity residents were significantly more likely to reside in for-profit nursing homes, which was a significant negative predictor of receiving five of the eight preventive services.

Our trends analysis of 1995, 1997, 1999, and 2004 NNHS survey data showed no reductions in disparities in influenza and pneumococcal vaccinations, and there is suggestive, though not significant, data that show Black–White influenza vaccination disparities were exacerbated between 1995 and 2004. Significant Black–White disparity in receipt of influenza vaccination was found in the 1997 and 2004 NNHS surveys, and significant Black–White disparity in receipt of pneumococcal vaccination was found in the 1997, 1999, and 2004 NNHS surveys. These findings indicate inequities in vaccination for Black nursing home residents over the last decade. For example, the Black–White disparity in receipt of influenza increased from 0.48 percentage points in 1995 to 9.16 percentage points in 2004, and the Black–White disparity in receipt of pneumococcal vaccination increased from 4.84 percentage points in 1995 to 13.02 percentage points in 2004. Our results are similar to a study using the 2005-2006 Minimum Data Set which also reported a crude disparity estimate of 8.3 percentage points (B. H. Bardenheier, Wortley, Ahmed, Hales, & Shefer, 2010). Our findings are also similar to two earlier studies using NNHS data (B. H. Bardenheier et al., 2009; Li & Mukamel, 2010). However, these two studies excluded residents of Other races, and only crude estimates were reported without adjusting for other covariates.

Pneumococcal vaccination rates for Black, White, and Other race residents all decreased from 1999 to 2004: from 63.58% to 56.20%, from 51.09% to 43.18%, and from 59.41% to 54.85%, respectively. In general, the 1995-2004 NNHS surveys data show that vaccination coverage in nursing homes remains well below the Healthy People 2010 and 2020 goals of 90% (U.S. Department of Health and Human Services, 2010). Nursing home residents are vulnerable to influenza and pneumococcal complications because of their frail health and because the virus can spread easily in a nursing home environment (Beck-Sague, Banerjee, & Jarvis, 1993; Kingston & Wright, 2002; Menec, MacWilliam, & Aoki, 2002; Muder, 2000). The vaccination is the single most effective way to prevent influenza and related consequences, including hospitalization and death (Jefferson et al., 2005; Muder, 2000). Of note, various strategies such as standing order protocols have been proposed to increase vaccination in nursing homes (McKibben, Stange, Sneller, Strikas, & Rodewald, 2000).

Moreover, our findings are consistent with previous studies among non-institutionalized individuals (Center for Disease Control and Prevention, 2003; Lu, Singleton, Rangel, Wortley, & Bridges, 2005). In the three flu seasons from 2006 to 2009, the National Minimum Data Set showed a statistically significant increase in flu vaccination rate for both White and Black residents, but disparities between White and Black residents still existed (Cai, Feng, Fennell, & Mor, 2011). The National Health Interview Surveys (NHIS) also indicated an increasing trend of disparity in vaccine coverage for those 65 years and older from 1989 to 2001 (Center for Disease Control and Prevention, 2003).

The market-based reform initiatives implemented since 2002 could have brought about unintended consequences (Mor et al., 2004). Consumers from different racial, ethnic, and socioeconomic groups may not have equal access to and derive equal benefit from quality information; health care providers may not be equally able to respond to quality improvement incentives because of lack of resources. If traditionally underserved groups are less likely to use quality information and/or are more likely to access providers with greater restraints on providing high-quality care, disparities may worsen even as average access and quality increase (Konetzka & Werner, 2009).

There is evidence that the implementation of Nursing Home Compare had modest or no effect on quality measures in nursing homes (Grabowski & Town, 2011; Zinn, Spector, Hsieh, & Mukamel, 2005). For instance, data from the Nursing Home Compare and the Online Survey Certification and Recording (OSCAR) system showed only small improvements in quality measures from 2003 to 2004 (Castle, Engberg, & Liu, 2007). Over a longer time frame, our results indicate that the Black–White disparity in receipt of vaccinations increased significantly between year 1995 and year 2004, though the overall trend is not significant.

Other research indicates that Black residents tend to be clustered in “lower-tier” nursing homes characterized by a greater dependence on Medicaid for payment, understaffing, and care deficiencies (Grabowski, 2004; Mor et al., 2004). These nursing home facilities are likely to have insufficient resources and manpower, and/or lack the ability to maintain a centralized IT system to track residents’ vaccination history or to implement evidence-based immunization programs, such as standing orders (B. Bardenheier, Shefer, Tiggle, Marsteller, & Remsburg, 2005). Studies also found that the differences and Black–White inequities in influenza vaccination coverage rate in nursing homes were correlated with the percentage of Black residents in the facility (B. Bardenheier, Wortley, Ahmed, Gravenstein, & Hogue, 2011; B. Bardenheier, Wortley, Shefer, McCauley, & Gravenstein, 2012). Yet, of special note, evidence shows that the racial vaccination gaps becomes smaller in nursing homes with the implementation of procedures like standing orders for influenza vaccinations, allowance of verbal consents, and routine review of facility-wide vaccination rates (B. Bardenheier, Shefer, et al., 2011). These and other continuous and comprehensive efforts are needed to address health care disparities in U.S. nursing homes.

Limitations

First, our analyses were conducted on repeated cross-sectional data and therefore could only examine associations, but not causal effects. Second, although we controlled for residents’ payment source and facility characteristic as health care system–related variables in our model analyses, other socioeconomic and health care system variables, such as residents’ education level, nursing home staff ratio, and standing order for influenza vaccinations, were not included because these variables were not available in the surveys. Third, the data used are reported from chart reviews by facility staff. Thus, the receipt of preventive care could be underreported or the status, like immunization, could be misclassified. Fourth, no variable was included to account for residents’ preference of preventive care. Yet, given the general positive effect associated with the preventive care for residents’ health, we believe that the potential cofounding effect from this type of variable is minimal. Lastly, we are not able to provide meaningful interpretation about “Other race” since respondents in “Other race” were heterogeneous.

Conclusion

Elimination of racial and ethnic disparities in health is an important national agenda (U.S. Department of Health and Human Services, 2010). Our findings of disparities in preventive care among racial/ethnic minority nursing home residents warrant special concern given the increasing diversity of the U.S. population and nursing home residents. For example, between 1999 and 2008, the number of elderly Hispanics and Asians living in U.S. nursing homes grew by 54.9% and 54.1%, respectively, while the number of elderly Black residents increased 10.8%. During the same period, the number of White nursing home residents declined 10.2% (Feng, Fennell, Tyler, Clark, & Mor, 2011). Despite the expectation that nursing homes should be able to provide appropriate immunizations and other preventive care services to all residents, racial/ethnic disparities in preventive care practice have been persistent over decades. Our study findings indicate that there is differential preventive care for Blacks and Other race/ethnicity residents, after adjustment for need and health care system variables, and that differential treatment in for-profit nursing homes and among Medicaid enrollees contribute to disparities in a number of preventive care measures. It is critical to develop targeted interventions to improve preventive care in nursing homes, especially for Black and Other race/ethnicity residents.

Footnotes

Acknowledgements

We thank the National Nursing Home Surveys participants without whom this study would not be possible.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Cook was supported by the National Institute of Mental Health (R01 MH091042: PI Cook).

References

Allsworth

J. E.

Toppa

Palin

N. C.

Lapane

K. L.

(2005). Racial and ethnic disparities in the pharmacologic management of diabetes mellitus among long-term care facility residents. Ethnicity & Disease, 15(2), 205-212.

Archer

Lemeshow

(2006). Goodness-of-fit test for a logistic regression model fitted using survey sample data. The Stata Journal, 6, 97-105.

Bardenheier

Shefer

Ahmed

Remsburg

Rowland Hogue

C. J.

Gravenstein

(2011). Do vaccination strategies implemented by nursing homes narrow the racial gap in receipt of influenza vaccination in the United States? Journal of the American Geriatrics Society, 59, 687-693.

Bardenheier

Shefer

Tiggle

Marsteller

Remsburg

R. E.

(2005). Nursing home resident and facility characteristics associated with pneumococcal vaccination: National nursing home survey, 1995-1999. Journal of the American Geriatrics Society, 53, 1543-1551.

Bardenheier

Wortley

Ahmed

Gravenstein

Hogue

C. J.

(2011). Racial inequities in receipt of influenza vaccination among long-term care residents within and between facilities in Michigan. Medical Care, 49, 371-377.

Bardenheier

Wortley

Shefer

McCauley

M. M.

Gravenstein

(2012). Racial inequities in receipt of influenza vaccination among nursing home residents in the United States, 2008-2009: A pattern of low overall coverage in facilities in which most residents are black. Journal of the American Medical Directors Association, 13, 470-476.

Bardenheier

B. H.

Wortley

Ahmed

Hales

Shefer

(2010). Influenza immunization coverage among residents of long-term care facilities certified by CMS, 2005-2006: The newest MDs quality indicator. Journal of the American Medical Directors Association, 11, 59-69.

Bardenheier

B. H.

Wortley

P. M.

Shefer

(2009). Influenza vaccine in African-American and White nursing home residents: Is there a gap? Journal of the American Geriatrics Society, 57, 2164-2165.

Beck-Sague

Banerjee

Jarvis

W. R.

(1993). Infectious diseases and mortality among US nursing home residents. American Journal of Public Health, 83, 1739-1742.

10.

Bernabei

Gambassi

Lapane

Sgadari

Landi

Gatsonis

. . . Mor

(1999). Characteristics of the SAGE database: A new resource for research on outcomes in long-term care. SAGE (Systematic Assessment of Geriatric drug use via Epidemiology) Study group. Journals of Gerontology: Series A Biological Sciences and Medical Sciences, 54, M25-M33.

11.

Blank

R. M.

Dabady

Citro

C. F.

(2004). Measuring racial discrimination. Washington, DC: National Academies Press.

12.

Cai

Feng

Fennell

M. L.

Mor

(2011). Despite small improvement, black nursing home residents remain less likely than whites to receive flu vaccine. Health Affairs, 30, 1939-1946.

13.

Casalino

L. P.

Elster

Eisenberg

Lewis

Montgomery

Ramos

(2007). Will pay-for-performance and quality reporting affect health care disparities? Health Affairs, 26, w405-414.

14.

Castle

N. G.

Engberg

Liu

(2007). Have nursing home compare quality measure scores changed over time in response to competition? Quality & Safety in Health Care, 16, 185-191.

15.

Center for Disease Control and Prevention. (2003). Racial/ethnic disparities in influenza and pneumococcal vaccination levels among persons aged > or =65 years–United States, 1989-2001. MMWR Morbidity and Mortality Weekly Report, 52(40), 958-962.

16.

Centers for Medicare and Medicaid Services. (2002). Nursing home quality initiatives overview. Baltimore, MD: Author. Retrieved from http://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/NursingHomeQualityInits/index.html

17.

Chien

A. T.

Chin

M. H.

Davis

A. M.

Casalino

L. P.

(2007). Pay for performance, public reporting, and racial disparities in health care: How are programs being designed? Medical Care Research and Review, 64(5 Suppl.), 283S-304S.

18.

Christian

J. B.

Lapane

K. L.

Toppa

R. S.

(2003). Racial disparities in receipt of secondary stroke prevention agents among US nursing home residents. Stroke, 34, 2693-2697.

19.

Cook

B. L.

McGuire

T. G.

Zaslavsky

A. M.

(2012). Measuring racial/ethnic disparities in health care: Methods and practical issues. Health Services Research, 47(3, Pt. 2), 1232-1254.

20.

Cook

B. L.

McGuire

Miranda

(2007). Measuring trends in mental health care disparities, 2000 2004. Psychiatric Services, 58, 1533-1540.

21.

Efron

(1979). Bootstrap methods: Another look at the jackknife. Annals of Statistics, 7, 1-26.

22.

Feng

Fennell

M. L.

Tyler

D. A.

Clark

Mor

(2011). The Care Span: Growth of racial and ethnic minorities in US nursing homes driven by demographics and possible disparities in options. Health Affairs, 30, 1358-1365.

23.

Gabrel

Jones

(2000a). The National Nursing Home Survey: 1995 summary. Vital and Health Statistics 13(146), 1-83.

24.

Gabrel

Jones

(2000b). The National Nursing Home Survey: 1997 summary. Vital and Health Statistics 13(147), 1-121.

25.

Grabowski

D. C.

(2004). The admission of blacks to high-deficiency nursing homes. Medical Care, 42, 456-464.

26.

Grabowski

D. C.

McGuire

T. G.

(2009). Black-White disparities in care in nursing homes. Atlantic Economic Journal, 37, 299-314.

27.

Grabowski

D. C.

Town

R. J.

(2011). Does information matter? Competition, quality, and the impact of nursing home report cards. Health Services Research, 46(6, pt. 1), 1698-1719.

28.

Hudson

T. J.

Cody

Armitage

T. L.

Curtis

M. A.

Sullivan

(2005). Disparities in use of antipsychotic medications among nursing home residents in Arkansas. Psychiatric Services, 56, 749-751.

29.

Institute of Medicine. (2003). Unequal treatment: Confronting racial and ethnic disparities in health care. Washington, DC: National Academy Press.

30.

Jefferson

Rivetti

Rudin

Di Pietrantonj

Demicheli

(2005). Efficacy and effectiveness of influenza vaccines in elderly people: A systematic review. Lancet, 366, 1165-1174.

31.

Jones

A. L.

Dwyer

L. L.

Bercovitz

A. R.

Strahan

G. W.

(2009). The National Nursing Home Survey: 2004 overview. Vital and Health Statistics 13(167), 1-155.

32.

Kingston

B. J.

Wright

C. V.

Jr. (2002). Influenza in the nursing home. American Family Physician, 65. 75-78.

33.

Konetzka

R. T.

Werner

R. M.

(2009). Disparities in long-term care: Building equity into market-based reforms. Medical Care Research and Review, 66, 491-521.

34.

Mukamel

D. B.

(2010). Racial disparities in receipt of influenza and pneumococcus vaccinations among US nursing-home residents. American Journal of Public Health, 100(Suppl. 1), S256-S262.

35.

P. J.

Singleton

J. A.

Rangel

M. C.

Wortley

P. M.

Bridges

C. B.

(2005). Influenza vaccination trends among adults 65 years or older in the United States, 1989-2002. Archives Internal Medicine, 165, 1849-1856.

36.

McGuire

T. G.

Alegria

Cook

B. L.

Wells

K. B.

Zaslavsky

A. M.

(2006). Implementing the Institute of Medicine definition of disparities: An application to mental health care. Health Services Research, 41, 1979-2005.

37.

McKibben

L. J.

Stange

P. V.

Sneller

V. P.

Strikas

R. A.

Rodewald

L. E.

(2000). Use of standing orders programs to increase adult vaccination rates. MMWR Recommendations and Reports, 49(RR-1), 15-16.

38.

McQuillan

Daley

M. F.

Stokley

Crane

L. A.

Beaty

B. L.

Barrow

. . . Kempe

(2009). Impact of the 2004-2005 influenza vaccine shortage on pediatric practice: A national survey. Pediatrics, 123, e186-192.

39.

Menec

V. H.

MacWilliam

Aoki

F. Y.

(2002). Hospitalizations and deaths due to respiratory illnesses during influenza seasons: A comparison of community residents, senior housing residents, and nursing home residents. Journals of Gerontology: Series A Biological Sciences and Medical Sciences, 57, M629-M635.

40.

Mor

Zinn

Angelelli

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.

41.

Muder

R. R.

(2000). Approach to the problem of pneumonia in long-term care facilities. Comprehensive Therapy, 26, 255-262.

42.

Quilliam

B. J.

Lapane

K. L.

(2001). Clinical correlates and drug treatment of residents with stroke in long-term care. Stroke, 32, 1385-1393.

43.

Spooner

J. J.

Lapane

K. L.

Hume

A. L.

Mor

Gambassi

(2001). Pharmacologic treatment of diabetes in long-term care. Journal of Clinical Epidemiology, 54, 525-530.

44.

U.S. Department of Health and Human Services. (2010). Healthy people 2020. Retrieved from http://www.healthypeople.gov/2010/hp2020/Objectives/TopicAreas.aspx

45.

Zhang

Decker

F. H.

Luo

Geiss

L. S.

Pearson

W. S.

Saaddine

J. B.

. . . Albright

(2010). Trends in the prevalence and comorbidities of diabetes mellitus in nursing home residents in the United States: 1995-2004. Journal of the American Geriatrics Society, 58, 724-730.

46.

Zinn

Spector

Hsieh

Mukamel

D. B.

(2005). Do trends in the reporting of quality measures on the nursing home compare web site differ by nursing home characteristics? Gerontologist, 45, 720-730.