Residents of long-term care facilities are at increased risk for hospitalization and death due to influenza and pneumonia.^1–4 Vaccination with influenza vaccine and 23-valent pneumococcal polysaccharide vaccine (PPSV23) have been shown to be effective in reducing the incidence of pneumococcal pneumonia and death due to pneumonia and influenza in this population.^5,6 The Advisory Committee on Immunization Practices (ACIP) has recommended annual influenza vaccination for residents of long-term care facilities, regardless of age, since 1988.⁷ While annual influenza vaccination is currently recommended for all persons >6 months of age without contraindications for vaccination, residents of nursing homes and other long-term care facilities are listed as a group at risk for medical complications attributable to severe influenza for whom priority should be given when vaccine supply is limited.⁸ Due to reported outbreaks of pneumococcal disease in nursing homes and other long-term care facilities, in 1997 the ACIP recommended that pneumococcal vaccination status be assessed for residents of nursing homes and other long-term-care facilities.⁹ During 2002 through 2015, published ACIP recommendations indicated vaccination of nursing home residents with 23-valent pneumococcal polysaccharide vaccine (PPSV23).^10,11 In 2016, the recommendation was clarified to indicate that nursing home residence alone was not an indication for PPSV23 vaccination for adults <65 years; however, these residents should be assessed for pneumococcal vaccination status and vaccinated as appropriate based on medical indications.¹²

A 1996 report by the Institute of Medicine on improving the quality of care in nursing homes recommended that all nursing home residents receive periodic standardized assessment of their functional, medical, mental, and psychosocial status.¹³ As a result, legislation requiring uniform resident assessment of all nursing home residents was included in the Omnibus Budget Reconciliation Act of 1987 (OBRA-87).¹⁴ Questions regarding influenza and pneumococcal vaccination status were added to the resident assessment instrument (RAI) in October 2005, when the Centers for Medicare and Medicaid Services (CMS) mandated that all nursing homes offer their residents annual influenza vaccination and at least one lifetime pneumococcal vaccination as a condition of certification.¹⁵

This paper reports influenza vaccination coverage for the 2005–06 through 2014–15 influenza seasons and pneumococcal vaccination coverage from 2006–2014 among residents of CMS-certified nursing homes.

The data for this analysis were obtained from the Minimum Data Set (MDS), which is maintained by CMS and includes the core set of screening, clinical, and functional status elements collected on the RAI, including immunization assessments. Data collected prior to October 1, 2010 were obtained from the MDS version 2.0, and data collected from October 1, 2010 onward were obtained from the MDS version 3.0.^16,17

In accordance with OBRA-87, facilities are required to conduct assessments on all residents at admission, quarterly, annually, whenever there is a significant change in the resident’s status, and at discharge. The RAI is also used for Medicare Prospective Payment System (PPS) assessments to determine payment for Medicare Part A beneficiaries covered under the skilled nursing facilities (SNF) benefit. Assessments are conducted for SNF PPS residents at 5 days, 14 days, 30 days, 60 days, and 90 days post-admission, and when a change in therapy occurs. When the timing of OBRA-87 and SNF PPS assessments coincide, one assessment can be used to satisfy both requirements.^16,17

Influenza and pneumococcal vaccination status is assessed on the RAI by medical record review when possible. If status cannot be determined from the medical record, the resident or the resident’s legal guardian is questioned.^16,17

Influenza vaccination coverage among U.S. nursing home residents increased from 71.4% in the 2005–06 influenza season to 75.7% in the 2014–15 season. This estimate is below the Healthy People 2020 (HP2020) target of 90% influenza vaccination coverage among nursing home residents.¹⁸ Achieving and maintaining high vaccination coverage is important for reducing the excess influenza-related morbidity and mortality in this high risk population.

Pneumococcal vaccination coverage among U.S. nursing home residents from the present analysis of MDS data was 67.4% in 2006, a substantial increase from a prior report using data from the National Nursing Home Survey (NNHS), which reported coverage among nursing home residents age ≥65 years of 23.6%-37.4% from 1995–1999.¹⁹ Coverage continued to increase among residents of all ages from 67.4% in 2006 to 79.9% in 2009. Since 2009, coverage has plateaued at approximately 80%, and remains below the HP2020 target of 90% for pneumococcal vaccination coverage among nursing home residents.¹⁸ The 90% target was met in previous years by five states--Arkansas, Iowa, Minnesota, North Dakota, and South Dakota--and reached 89% in several other states. However, in 2014 coverage had dropped below 90% in all states.

Coverage varied widely by state for influenza and pneumococcal vaccination, with four states having coverage below 70% for both vaccinations in the most recent assessment periods. Differences in vaccination coverage among nursing home residents by state²⁰ and geographic region^19,21–22 have consistently been reported in the literature and are likely due to factors such as differences between states in immunization policies for both residents and health care personnel, differences in state laws regarding nursing home staffing levels,²³ differences in the racial/ethnic distribution of nursing home residents,²⁴ and differences in facility-level characteristics such as bed size, ownership status, and primary payment source.^19,20–22

Although influenza and pneumococcal vaccination coverage remain suboptimal in many states, the shift overall in the reason for non-vaccination for both vaccines from the majority of unvaccinated residents not being offered vaccine to the majority of unvaccinated residents being offered and declining vaccination is an encouraging finding, suggesting that more facilities are now complying with the CMS requirement to offer vaccination to all residents. In the most recent years of data collection, the proportion of unvaccinated residents that had not been offered influenza or pneumococcal vaccination had been reduced to 10.6% and 16.0%, respectively. However, the finding that vaccine refusal now plays the largest role in residents remaining unvaccinated underscores the need for effective patient education in conjunction with access to vaccination. Previous studies have shown that nursing home residents were more likely to receive influenza vaccination if they were advised to do so by a relative or nursing home health care worker,²⁵ and that coverage increases with more frequent physician recommendations.²⁶

Differences in influenza vaccination coverage between white and black nursing home residents have previously been reported in assessments conducted in the 2005–06 through 2008–09 influenza seasons using data from the MDS^20,24,27 and in 2003–04 using data from the NNHS.^28–29 Our analysis revealed that this difference has persisted through the 2014–15 influenza season, and in fact increased from a difference of 7.1 percentage points higher coverage in whites compared with blacks in the 2005–06 season to 9.0 percentage points higher coverage in whites in the 2014–15 season. Similarly, differences in pneumococcal vaccination coverage between white and black nursing home residents have been reported from the NNHS for the years 1995–2004.^{19,21,28–29} While we found that the difference in pneumococcal vaccination coverage between white and black residents was somewhat reduced between 2006 and 2014, in 2014 coverage among black residents remained 8.2 percentage points lower than coverage among white residents.

The vaccination coverage disparity between black and white nursing home residents has been attributed in part to the clustering of black residents in what Mor et al. have termed “lower-tier” nursing homes, characterized by a greater dependence on Medicaid for payment and fewer registered nurses, nurse practitioners and physician assistants, and administrative resources per resident compared with upper-tier facilities.^19,30–32 The notion that lower vaccination coverage among blacks is a result of residence in poorer-quality nursing homes is supported by a study by Bardenheier et al. that found that vaccination coverage was lower for both blacks and whites in nursing homes with higher proportions of black residents.²⁴ However, other researchers have found that black residents were less likely to receive influenza vaccination than white residents in the same facility and that black residents were more likely to refuse vaccination, suggesting that the differences in influenza and pneumococcal vaccination coverage between black and white nursing home residents are not solely attributable to facility-level characteristics.²⁷ In the current study, the differences in vaccination coverage between black and white residents were decreased after controlling for clustering of residents within facilities, but facility-level effects did not completely explain the differences in coverage. Long-standing differences in both influenza and pneumococcal vaccination coverage have been reported between community-dwelling black and white persons ≥65 years, even after adjustment for access to care and other socio-economic factors.^33–37 These differences have been attributed to resistant attitudes and beliefs about vaccination, fewer vaccine-seeking behaviors, poorer provider communication, and less effective provider recommendations among blacks.^34,38–39 In the current study, while vaccine refusal was the most common reason for non-vaccination in white and non-white residents, disparities in access to vaccination likely played a role in lower vaccination coverage among non-white residents, as these residents were more likely than unvaccinated white residents to report not being offered influenza and pneumococcal vaccination.

Little prior research has focused on the disparity between non-Hispanic white and Hispanic nursing home residents; however, in the current study, we found differences between non-Hispanic white and Hispanic residents similar in magnitude to those between non-Hispanic white and non-Hispanic black residents for both influenza and pneumococcal vaccination. We also found that these differences did not persist after controlling for facility-level effects, suggesting the clustering of Hispanics in nursing homes with lower vaccination coverage. Lower influenza and pneumococcal vaccination coverage has been reported among community-dwelling Hispanics aged ≥65 years compared with their non-Hispanic white counterparts.^{33–35,37,40,41} Unlike non-Hispanic black adults aged ≥65 years, studies have found that Hispanics aged ≥65 years were less likely than non-Hispanic whites to report resistant attitudes toward vaccination, but more likely to report that they did not know these vaccines were recommended for them, suggesting that poor communication or language barriers might contribute to lower vaccination coverage in this population.^34,38

Influenza and pneumococcal vaccination coverage increased with increasing age and increasing numbers of comorbidities among nursing home residents in this study in all measurement periods. Increased vaccination coverage with increasing age and increased number of high-risk medical conditions is well-documented, both among nursing home residents^28–29 and community-dwelling persons, even those aged ≥65 years^{35–36,40–42} due to more frequent contacts with medical providers and increased opportunity for vaccination. Medical providers might also have increased awareness of age and condition-based indications for vaccination. Although less pronounced than differences by age and race/ethnicity, we found slightly higher influenza and pneumococcal vaccination coverage among women and among widowed residents compared with those with other marital statuses. However, these differences are likely attributable to higher proportions of older residents in these groups.

The findings in this study are subject to several limitations. Vaccination status is reported to the MDS by individual facilities. Although the RAI users’ manual outlines the preferred procedure for assessing vaccination status, there are no quality checks to ensure that assessments are conducted uniformly across facilities. A recent validation study of the accuracy of influenza vaccination reported to the MDS found that, overall, influenza vaccination coverage measured using RAIs was only 1.8 percentage points lower compared to coverage using medical chart review as the gold standard. However, agreement rates varied widely by facility and state.⁴³ No such validation study has been published for pneumococcal vaccination. Assessment of pneumococcal vaccination status by nursing home staff might be particularly difficult as it is not an annual event like influenza vaccination and requires a longer look-back period. Further work to validate pneumococcal vaccination data reported to the MDS based on Medicare claims data or medical chart review is needed. In the present study, we found disagreement in influenza vaccination status between assessments for approximately 5% of residents each influenza season, and disagreement in pneumococcal vaccination status for approximately 20% of residents each year. Re-vaccination with influenza vaccine is not recommended in the same season. And, while adults with certain immunocompromising conditions or those who were initially vaccinated prior to age 65 were recommended for re-vaccination with pneumococcal vaccine five years after the initial dose, these indications do not explain the high proportion of residents in our data who were reported to be vaccinated on one assessment and then unvaccinated on a subsequent assessment. These inconsistencies were considered to be errors in the subsequent assessments. Higher confidence was placed on earlier assessments, because as more time elapsed between the date of vaccination and the date of assessment, the likelihood increased that residents or their legal guardians could not recall vaccination status or that vaccination in the distant past might not be captured in current medical records. Sensitivity analyses revealed that, depending on the handling of these discrepant assessments in the analysis, actual influenza and pneumococcal vaccination coverage estimates could have been approximately 5% and 10% lower, respectively, than those reported in this study.

An additional limitation of the study is that pneumococcal vaccination coverage estimates for the year 2014 might not reflect changes in the ACIP recommendation for adults aged ≥65 years that were published in September 2014. Prior to September 2014, most adults were recommended to get one lifetime dose of PPSV23, and the question, “Is the resident’s pneumococcal vaccination up to date?” in the MDS 3.0 was most likely interpreted by nursing home staff as referring to one dose of PPSV23. Beginning in September 2014, adults aged ≥65 years were recommended to receive both 13-valent pneumococcal conjugate vaccine (PCV13) and PPSV23, separated by a one-year interval.⁴⁴ While neither the question on the RAI nor the RAI user’s manual were updated to reflect this new recommendation, we cannot discount the possibility that personnel completing the assessment were aware of the change in the recommendation and considered residents aged ≥65 who did not have a dose of both PCV13 and PPSV23 as not up to date. However, as this change would only affect residents who had their first assessment in September 2014 or later, it was not likely to have had a meaningful impact on the coverage estimates in this report.

Influenza and pneumococcal vaccination coverage increased among U.S. nursing home residents from the 2005–06 through 2014–15 influenza seasons and the years 2006–2014, respectively, but remained below national targets of 90% for both vaccines. Non-Hispanic black and Hispanic residents were less likely to be vaccinated compared with non-Hispanic white residents, and these differences persisted over time. Nursing home administrators should employ evidence-based strategies such as standing orders and provider reminders^45–47 to ensure compliance with the CMS mandate to offer influenza and pneumococcal vaccination to all residents along with culturally-appropriate vaccination promotion to increase coverage and protection against disease in this vulnerable population.