Abstract
Persons living with HIV/AIDS (PLWH) are living long enough to need age- and HIV-related nursing home (NH) care. NH quality of care has been associated with risk for hospitalization; but it is unknown if quality of HIV care in NHs affects hospitalization in this population. We assessed HIV care quality with four national measures adapted for the NH setting. We applied the measures to 2011–2013 Medicare claims linked to Minimum Dataset assessments of resident health, prescription dispensing data, and national reports of NH characteristics. Cox proportional hazards models calculated the risk of all-cause and HIV/AIDS-related hospitalization by HIV care compliance. We identified 1,246 PLWH in 201 NHs with 382 all-cause and 63 HIV/AIDS-related hospitalizations. NH HIV care compliance varied from 24.9% to 64.7%. After regression adjustment, we could detect no difference in all-cause or HIV/AIDS-related hospitalizations by NH HIV care compliance. We postulate that the lack of association may be due to inappropriate HIV care quality measures that do not accurately represent NHs ability to care for PLWH. There is urgent need to create valid NH HIV care quality measures.
Keywords: care quality, HIV and aging, nursing homes, hospitalizations
INTRODUCTION
Demand for HIV care in nursing homes (NH), in addition to care for age-related conditions, will increase due to antiretroviral therapy (ART) prolonging life for persons living with HIV/AIDS (PLWH). A 2015 report noted that NHs will serve approximately 50% of the older adult population meaning NHs may play an important role in the health of up to 300,000 PLWH.1, 2 Overall NH quality is rated and carefully monitored by the Centers for Medicare and Medicaid Services (CMS), yet only one study has reported on NH HIV care quality, finding that quality is variable due to differences between clinical profiles of PLWH within the NH.3 Research also finds that PLWH, 56% of which are insured by Medicare4, are disproportionately admitted to NHs of overall lower quality, which has been linked to higher risk for hospitalizations.5–7
Given the relatively low prevalence of HIV among NH residents, the often high turnover of NH clinical staff, including nurses, physicians, and minimally trained aides, and limited pharmacist oversight, most NHs have little to no experience with HIV care.1, 8 PLWH have unique health profiles as a result of intersecting effects of aging, chronic HIV infection, and ART use; they are physiologically similar to non-HIV infected adults 5–15 years older.8, 9 Furthermore, HIV is sensitive to deviations in care; viral rebound can occur within one month of treatment interruption.10 To that end, a review of the literature showed having less experience treating HIV was associated with increased likelihood to deviate from national HIV care guidelines resulting in, among other outcomes, hospitalizations and emergency room use.11
Though NHs are equipped to provide general clinical care, such as medication administration, pain management, physical therapy, and transportation to off-site healthcare12, it is unclear to what extent they comply with HIV care guidelines. Consequently, PLWH may be at increased risk for HIV care quality-related hospitalizations.
Purpose
This study examined NH HIV care quality as related to risk of first all-cause and HIV/AIDS-related hospitalization for PLWH. We hypothesized that facilities with better quality HIV care would have lower risk of hospitalizations. CMS (DUA# RSCH-2017–51615) and our institution’s Internal Review Board (IRB# 14-02-23) approved this study.
METHODS
Data:
We used a 2011–2013 database of linked files: 1) an all-payer prescription dispensing database from Omnicare; 2) the Minimum Data Set (MDS) version 3.0; 3) Medicare administrative file; 4) Medicare Parts A and B; 5) publicly available Certification and Survey Provider Enhance Reporting (CASPER).
Omnicare data are a subset of claims for all prescriptions dispensed to approximately 3 million individuals across half of the NHs in the US, regardless of the payment source (e.g., Medicaid, Medicare Part D, private insurance, cash). Data elements include the national drug code (NDC), brand and generic prescription names, prescription date, days’ supply, quantity dispensed, and payment source.13
The MDS 3.0 is a nurse-administered physical and mental health assessment. It is collected for all NH patients in Medicare/Medicaid-certified NHs upon admission, quarterly throughout a year, at any change of health status, and upon discharge. Only admission assessments associated with a NH stay of at least 30 days were included in this study.10, 14
The Medicare administrative file provides basic demographics, reason for entitlement, Medicaid eligibility, and mortality. Medicare Parts A and B contain individual-level claims for hospitalizations and outpatient care, respectively.
CASPER is a repository of federally mandated onsite evaluations of all Medicare/Medicaid-certified NHs. Data elements include the NH operational characteristics, health inspections, and aggregate patient characteristics.
Population:
Medicare-eligible individuals with an antiretroviral prescription or 2 diagnoses for HIV and/or AIDS in Medicare Part B or 1 diagnosis in Medicare Part A made up the sampling frame (n=10,445).8, 15 Most (82.6%, n=8,629) were Medicare eligible for the study period, from which we excluded 7,383 that were not newly admitted nor in a NH meeting criteria for at least one measure of HIV care quality — which meant being one of at least five PLWH in a NH facility16 — the final sample size was 1,246 (see flow chart and comparison of included vs excluded PLWH in Supplemental Figure 1 and Supplemental Table 1, respectively).
Nursing Homes:
NHs were identified from the MDS admission assessments for the final sample. NHs had to have at least five PLWH during the study period (n=201).
Measures
Time:
CMS guidelines for using the MDS determined length-of-stay in the NH.17 Time to first hospitalization was the difference between NH admission date (t(0)) and first all-cause, and first HIV/AIDS-related hospitalization date. Censoring included discharge, death, or the study end date (December 31, 2013).
HIV Care Quality:
The National Quality Forum (NQF) is contracted by the US government to develop and validate gold-standard healthcare quality measures.18 There are eight measures of HIV care quality. Four measures, requiring unavailable laboratory data, were excluded. The remaining four measures: 1) CD4 cell count monitoring; 2) prescription of ART; 3) frequency of medical visits; 4) gaps in medical visits, were adapted and applied to the NHs.
Adapting the NQF HIV Care Quality Measures to the NH setting:
Adaptations were based on national HIV care guidelines. The 2011–2013 guidelines recommended that PLWH see a medical provider every three to four months, have CD4 cell count and/or viral load monitoring every three to four months, and have ART prescribed, particularly if CD4 cell count is less than 200 cells/mm3.19, 20
Three measures (CD4 monitoring, medical visit frequency, and gaps in medical visits) were adapted to be based on the number of 90-day intervals per person during which NHs had the opportunity to deliver care; excluding partial 90-day intervals. CD4 monitoring was further adapted to also include viral load monitoring (CPT codes: 86361, 86360, 87536, 87539); viral load monitoring is commonly used concurrently or instead of CD4 cell count.19 Medical visit frequency and gaps in medical visits were identified such that for every 90-day interval per person, there should be at least one medical visit. No medical visit within 90 days of the admission MDS was considered a gap in medical visits.
The prescription of ART measure applied to all PLWH because prescribing of medications is always possible via the NH medical director.
Maximum compliance with all adapted measures was 100%. Higher compliance represented higher quality of HIV care. Dichotomized compliance (low or high) was the main predictor. High compliance NHs were in the upper quartile of compliance relative to other NHs measured.21 High compliance with gaps in medical visits meant there were no gaps in medical visits.
Risk Adjustment:
The 2012nRx DCG/HCC is a validated measure for controlling confounding, and the score generated from the model provides an estimate of patient health acuity.22 It was applied to all models.
Outcomes:
All-cause hospitalizations were identified in Part A as inpatient coded claims. HIV/AIDS-related hospitalizations were a subset having a primary diagnosis code for HIV/AIDS.
Rate and relative risk of hospitalization:
Crude rates for high versus low compliance NHs (calculated as the number of first hospitalizations over the total number of PLWH in the NHs) were compared by T-tests of two proportions (α=.05).
Adjusted Cox proportional hazards with robust estimators, to account for clustering in NHs, predicted the relative risk of first all-cause hospitalization and first HIV/AIDS-related hospitalization by compliance. A test of residuals assessed the proportional hazards assumption.
Covariates:
Grounded in prior research, indicators for having a mental health diagnosis or having a physical health diagnosis were created based on the presence of the following conditions: anxiety, depression, bi-polar disorder, closed brain injury, manic depression, post-traumatic stress disorder, schizophrenia, Alzheimer’s disease, or dementia; cancer, anemia, vascular diseases, cirrhosis, gastro esophageal reflux disease, end stage renal disease, multidrug resistant infections, pneumonia, septicemia, tuberculosis, urinary tract infection, viral hepatitis, wound infections, diabetes, arthritis, and osteoporosis.21 At the NH level, census region, for-profit status, staffing, deficiencies, and facility characteristics were considered.
Model building:
To avoid saturating the models, published methods for variable selection and bootstrapping sensitivity analyses were used.23, 24 The selected covariates, binary high compliance, and the risk adjustment were included in final models. A sub-analysis evaluated only NHs that qualified to be assessed for all four measures (n=34). All tests were run with 95% confidence and using SAS 9.4 (Cary, NC).
RESULTS
Characteristics of PLWH in NHs
We identified 1,246 PLWH in 201 NHs between 2011–2013 (Table 1). On average, they were 59.9±12.02 years, mostly male (73.8%), Black (60.6%), never married (60.4%), and still living by the end of our study period (69.2%). Disability (59.2%) more often than age (33.8%) was the reason for Medicare eligibility. PLWH that had a hospitalization were similar in age to censored PLWH, but were less often male (69.6% v. 75.7%), and more often Black (66.5% v. 58.1%) and deceased by the study end (45.8% v. 24.2%).
Table 1.
Descriptive statistics of PLWH in nursing homes and comparing those with and without hospitalizations
| Person Characteristics | All PLWHa (n=1246) n(%) |
PLWH that had a hospitalization (n=382) n(%) |
PLWH that were censored (n=864) n(%) |
P-valueb |
|---|---|---|---|---|
| All-cause hospitalizations | 382 (30.7) | |||
| HIV/AIDS hospitalizations | 63 (5.1) | |||
| Age (mean ± standard deviation [SD]) | 59.9 ±12.02 | 60.5 ±12.7 | 59.7 ±11.7 | 0.280 |
| Male | 920 (73.8) | 266 (69.6) | 654 (75.7) | 0.025 |
| Race | 0.028 | |||
| White | 334 (26.8) | 83 (21.7) | 251 (29.1) | |
| Black | 755 (60.6) | 254 (66.5) | 501 (58.1) | |
| Hispanic | 143 (11.5) | - | - | |
| Other | 14 (1.1) | - | - | |
| Marital status | 0.386 | |||
| Never married | 752 (60.4) | 217 (56.8) | 535 (61.9) | |
| Married | 141 (11.3) | 44 (11.5) | 97 (11.2) | |
| Widowed | 96 (7.7) | 33 (8.6) | 63 (7.3) | |
| Separated/Divorced | 184 (14.8) | 66 (17.3) | 118 (36.7) | |
| Unknown | 73 (5.9) | 22 (5.8) | 51 (5.9) | |
| Deceased by the end of 2013 | 384 (30.8) | 175 (45.8) | 209 (24.2) | <.001 |
| Study time (mean days ± SD) | 108.7 ±153.7 | 90.7 ±76.8 | 116.7 ±159.7 | 0.006 |
| ADL measure of independenceb | 0.021 | |||
| Independent (ADL 0–5) | 444 (35.6) | 118 (30.9) | 362 (37.7) | |
| Not fully independent (ADL >5) | 802 (64.4) | 264 (69.1) | 538 (62.3) | |
| At least one MH active diagnosis | 675 (54.2) | 190 (49.7) | 485 (56.1) | 0.037 |
| At least one PH active diagnosis | 1134 (91) | 356 (93.2) | 778 (90.1) | 0.073 |
| Risk adjustment scorec | 3.6 ±0.7 | 3.3 ±0.8 | 0.020 | |
| Confusion Assessment Method for delirium | 0.205 | |||
| Delirium Positive | 61 (4.9) | 23 (6) | 38 (4.4) | |
| Eligible for Medicaid | 1122 (90) | 352 (92.1) | 770 (89.1) | 0.100 |
| Medicare entitlement reason (current) | 0.001 | |||
| Age 65+ | 421 (33.8) | 145 (38) | 276 (31.9) | |
| Disability | 738 (59.2) | 199 (52.1) | 539 (62.4) | |
| End Stage Renal Disease | 87 (7) | 38 (9.9) | 49 (5.7) | |
| Medicare entitlement reason (original) | 0.004 | |||
| Age 65+ | 284 (22.8) | 100 (26.2) | 184 (21.3) | |
| Disability | 871 (69.9) | 244 (63.9) | 627 (75.6) | |
| End Stage Renal Disease | 91 (7.3) | 38 (9.9) | 53 (6.1) | |
| Admitted to NH from | 0.424 | |||
| Acute Hospital | 1154 (92.6) | 354 (92.7) | 800 (92.6) | |
| Community (Private home/apt., Board/care, Assisted living, Group home) | 43 (3.5) | 14 (3.7) | 29 (3.4) | |
| Other | 49 (3.9) | 14 (3.7) | 35 (4.0) |
Note: n/a are not applicable to the hospitalization group; - are values omitted because they are smaller than the smallest allowable reportable value for these data;
PLWH: People living with HIV/AIDS;
P-value for the difference between PLWH with and without a hospitalization.
Characteristics of NHs
The 201 NHs had, on average, 9.1±11.2 PLWH (median 7), 151.6±91.2 total patients (median 124), and 173.2±96.9 beds (median 147) (Table 2). Most of the NHs serving PLWH were located in the South (58.7%) or Northeast (20.4%). The majority were for-profit (83.6%). On average, the NHs had 6.1±6.3 substantiated complaints (median 4.0) against them and aides spent the most hours with patients (2.3±0.5; median 2.3).
Table 2.
Descriptive Statistics for n=201 Nursing Homes
|
Nursing Home Characteristics |
Nursing homes (n=201) |
PLWHa served by the NHsb
(n=1246) |
|---|---|---|
| Number of PLWH in the NH | 9.1 ±11.2 | |
| Number of patients in the NH | 151.6 ±91.2 | |
| Number of beds in the NH | 173.2 ±96.9 | |
| Number of substantiated complaints | 6.1 ±6.3 | |
| Hours per patient per day spent with aides | 2.3 ±0.5 | |
| Hours per patient per day spent with LPNs | 0.9 ±0.3 | |
| Hours per patient per day spent with RNs | 0.7 ±0.3 | |
| NH is for profit | 168 (83.6) | 909 (73) |
| NH is dually certified for Medicare and Medicaid | 191 (95) | 1204 (96.6) |
| Type of institution | ||
| Unknown | 7 (3.5) | 38 (3) |
| Dual skilled nursing/nursing facility | 148 (73.6) | 966 (77.5) |
| Separate skilled nursing or Nursing facility | 46 (22.9) | 242 (19.4) |
| NH has a resident council | 124 (61.7) | 694 (55.7) |
| NH Census region | ||
| Midwest | 26 (12.9) | 111 (8.9) |
| Northeast | 41 (20.4) | 400 (32.1) |
| South | 118 (58.7) | 652 (52.3) |
| West | 16 (8) | 83 (6.7) |
Note:
PLWH:People living with HIV/AIDS;
NH: nursing home.
NH Compliance with HIV Care Quality Measures
Thirty four NHs met criteria for CD4/viral load monitoring, frequency of medical visits outside of the NH, and gaps in medical visits (Table 3). All 201 NHs were evaluated for compliance with prescription of ART, but only one quarter of those achieved high compliance (24.9%). Approximately a quarter (26.9%) achieved high compliance on CD4/viral load monitoring, whereas high compliance was achieved by 58.8% for frequency of medical visits outside of the NH and 64.7% for gaps in medical visits outside of the NH.
Table 3.
Comparison of Unadjusted Hospitalization Rates of PLWH by NH Compliance with HIV Quality Care Measures
| Among High compliance NHs | Among Low compliance NHs | |||||||
|---|---|---|---|---|---|---|---|---|
| NHs | PLWH | Rate | NHs | PLWH | Rate | |||
| HIV Care Quality Measures | n (%)a | n (%)a | % | n (%) | n (%) | % | p-value | |
| All-cause Hospitalizations | CD4 cell count and viral load monitoring | 9 (26.5) | 264 (54.4) | 29.5 | 25 (73.5) | 221 (45.6) | 27.1 | 0.89 |
| Prescription of ART | 50 (24.9) | 448 (36) | 28.1 | 151 (75.1) | 798 (64) | 32.1 | 0.60 | |
| Frequency of medical visits outside of the NH | 20 (58.8) | 176 (36.3) | 31.3 | 14 (41.2) | 309 (63.7) | 26.9 | 0.78 | |
| Gaps in medical visits outside of the NH | 22 (64.7) | 188 (38.8) | 30.9 | 12 (35.3) | 297 (61.2) | 26.9 | 0.81 | |
| HIV/ AIDS Hospitalizations | CD4 cell count and viral load monitoring | 9 (26.5) | 264 (54.4) | 5.7 | 25 (73.5) | 221 (45.6) | 4.5 | 0.89 |
| Prescription of ART | 50 (24.9) | 448 (36) | 4.5 | 151 (75.1) | 798 (64) | 5.4 | 0.80 | |
| Frequency of medical visits outside of the NH | 20 (58.8) | 176 (36.3) | 8.0 | 14 (41.2) | 309 (63.7) | 3.6 | 0.60 | |
| Gaps in medical visits outside of the NH | 22 (64.7) | 188 (38.8) | 7.4 | 12 (35.3) | 297 (61.2) | 3.7 | 0.67 | |
n(%) = number and percent of NHs, and number and percent of PLWH. For example, 9 NHs = 26.5% of 34 NHs evaluated and 264 PLWH = 54.4% of 485 PLWH evaluated for CD4 cell count and viral load monitoring.
Note: PLWH=persons living with HIV/AIDS; NH=nursing home
Crude rates of hospitalization
Rates of all-cause hospitalizations ranged from 28.1% to 31.3% in high compliance to 26.9% to 32.1% in low compliance NHs. Rates of HIV/AIDS-related hospitalizations ranged from 4.5% to 8.0% in high compliance to 3.6% to 5.4% in low compliance NHs. There was no statistical difference between crude rates of either hospitalization type across high and low compliance NHs (Table 3).
Risk of hospitalization
High compliance with CD4/viral load monitoring, frequency of medical visits outside of the NH, and gaps in medical visits outside of the NH showed a positive but non-significant risk for all-cause hospitalization risk (Figure 1).
Figure 1.
Adjusted risk of all-cause and HIV/AIDS-related hospitalization by high complice with HIV care quality measures
Similarly, risk for HIV/AIDS-related hospitalizations related to high compliance with each of CD4/viral load monitoring, frequency of medical visits outside of the NH, and gaps in medical visits outside of the NH was positive in direction but not statistically significant, and had wide confidence intervals (Figure 1).
Although not statistically significant, the direction of risk was negative between high compliance with prescription of ART and all-cause hospitalizations, but positive for HIV/AIDS-related hospitalizations (Figure 1).
Limitations
The findings of this study should be interpreted in the context of several limitations. First, the HIV care quality measures were adapted. The adapted measures were not validated. Second, the qualifying sample of NHs is small, limiting generalizability; the study sample excludes private-pay PLWH, managed care enrollees, and is limited to individuals in NHs with records in the Omnicare database. This data has been shown to be comparable to CASPER in representation of NHs25 and reflects only prescriptions filled at the NH pharmacy. Third, the MDS has been validated as a tool for determining the health status of NH resident, but may under-report certain characteristics, particularly if the variables are not linked to claims or quality measures.26 Lastly, insufficient data on HIV/AIDS hospitalizations made confidence intervals wider and reduced power for detecting a difference between high compliance and low compliance NHs.
DISCUSSION
This exploratory study is the first to investigate the association between the quality of HIV care in NHs and risk for all-cause and HIV/AIDS-related hospitalizations of PLWH. In applying adapted national HIV care quality measures to a large Medicare dataset, first we determined that quality of HIV care in NHs varied considerably. Second, we were unable to detect statistically significant differences in risk of all-cause or HIV/AIDS-related hospitalization for PLWH in high versus low compliance NHs.
Caution is warranted, however, in drawing any clear conclusions from this lack of association between quality of HIV care and hospitalization. We encountered significant methodological challenges in this study. It was very difficult to apply the quality measures developed for the community setting to the NH setting. Major issues included low concentration of PLWH newly admitted to NHs, short duration of observation in the NH, and accounting for the fact that NHs are full time clinical settings that could provide most care on-site. Our recommendations based on this work are listed below.
First, there is the low concentration of PLWH in NHs. We identified as many as 877 NHs with PLWH, but many of these facilities had to be excluded from the study sample because they had less than five PLWH total across the three years of data. Although the size of the population of PLWH in NHs is expected to grow, the low relative prevalence of this disease will likely always make a facility-level quality measure of HIV care challenging. For that reason, developing new measures based on methods used for measuring rare diseases might make sense, such that clinical decision making and related quality measures would be based on the health outcomes identified to be most important to PLWH in NHs.27
Second, there is short duration of observation in the NH; we observed our sample for only an average of 108.7±153.7 days over 3 years. The original HIV care quality measures often had time-in-care restrictions that were incompatible with the highly unstable NH population. For example, the original measure for frequency of medical visits required 24 months of observation. New measures could reduce the time-in-care requirement, as we did, requiring only 30+ days in the NH.
Third, NHs are full time clinical settings that can provide most care on-site. The original measure of CD4 cell counts requires two medical visits at least 90 days apart and the measure of gaps in medical visits requires a person be seen by a physician in the first and second half of the measurement year. When revising measures, it might be important to consider that the requirements for medical visits may not be necessary when interactions with providers can take place within the NH facility.
We attempted to address these challenges through our adaptations based on HIV care guidelines. By eliminating time-in-care requirements for a prescription of ART and using 90 day intervals for other measures we identified an additional 247 PLWH and 36 NHs for the study. Still, not all NHs included were evaluated for all quality measures; only 34 NHs qualified for all four measures contributing to high variability between measures.
Despite the challenges, we believe the lack of significant effect between HIV care quality and hospitalizations is notable. This lack of association, which persisted across all models, may bolster 1) recent evidence that the differences between individual PLWH in NHs are more important to care quality and related outcomes than facility level characteristics3, and 2) the idea that new measures of quality could be better if based on outcomes identified as most important by PLWH in NHs.27 To that end, hospitalizations may be the result of deleterious effects of other comorbidities that outweigh the positive effects of high quality disease-specific care, further supporting the need for research examining if care quality measures could be based on outcomes identified as important to PLWH in NHs.28, 29
More work in this area is needed and will depend on the availability of HIV care quality measures appropriate for NHs. Our adaptations may have compromised the construct validity of the measures such that they were assessing extrinsic factors such as NHs’ experience caring for PLWH rather than the quality of HIV care delivered. The original HIV care quality measures were not created in the context of older PLWH, NHs, or related considerations like NH clinical management protocols, despite their creation for application to diverse sources of health data.30
Conclusion
There is urgent need for measures that reliably detect HIV care quality in the NH clinical setting because the aging population of PLWH will increasingly need institutional care. Our best efforts to address the measurement challenges were still insufficient to allow us to detect a consistent relationship between compliance and risk of hospitalization, which suggests missing information on NH setting-specific clinical management and prescribing factors. As a result, we have concluded that future research should develop a new set of measures specific to the population of PLWH in NHs.
Implications
HIV care quality may not be reliably measured in NHs. Future research should explore developing NH HIV care quality measures to better document and understand HIV care quality as it relates to hospitalizations in this increasingly important health care setting for the aging population of PLWH.
Supplementary Material
Supplemental Figure 1. Flow chart to determine sample size.
Supplemental Table 1. Comparing Medicare eligible PLWH included and excluded from the study
Acknowledgements:
This work was assisted by a clinical consultant, Debra Winters, APRN-BC, AACRN and director of the New England AIDS Education and Training Center, Boston, MA
Conflicts of Interest and Source of Funding: Funding was provided by the Agency for Healthcare Research and Quality Dissertation Grant (R36HS025662) and the Harvard Translational Research in Aging Grant (T32 AG023480).
Biographical Sketches
Brianne Olivieri-Mui, PhD, MPH is a post-doctoral fellow at The Hinda and Arthur Marcus Institute for Aging Research at Hebrew SeniorLife and affiliated with Harvard Medical School, Boston, MA. She holds two advisory positions on the Massachusetts Department of Public Health Gay Men’s advisory group and Behavioral Health advisory group and has recently been awarded a T32 grant to continue health services research related to aging and HIV.
Jean McGuire, PhD, MSPH is a Professor of Practice in the Health Sciences Department at Northeastern University, Boston, MA. Her research focuses on population health and disability and long term care service delivery and financing. She was an Assistant Secretary at the MA Executive Office of Health and Human services where she oversaw the disability and long term care portfolio; she has a long history in state, federal, and international HIV/AIDS policy.
John Griffith, PhD is a Clinical Professor of biostatistics in the Department of Health Sciences, Northeastern University, Boston, MA
Sean Cahill, PhD is Director of Health Policy Research at the Fenway Institute, Affiliate Associate Clinical Professor of Health Sciences at Northeastern University, and Adjunct Associate Professor at Boston University School of Public Health. Cahill serves on the Massachusetts Special Legislative Commission on LGBT Aging and the American Heart Association’s Northeast Health Equity Consortium. He is Associate Editor at LGBT Health, and Senior Associate Editor at Annals of LGBTQ Public and Population Health.
Becky Briesacher, PhD is a health services researcher in the Department of Pharmacy and Health Systems Sciences, Northeastern University, Boston, MA. She has expertise in the interrelated relationships among drug policy, prescription drug coverage and ultimately the medication use and health of older adults. She also has over 15 years of academic research experience and served as the Principal Investigator of multiple NIH-sponsored studies.
Footnotes
The authors declare no conflicts of interest.
Contributor Information
Brianne Olivieri-Mui, The Hinda and Arthur Marcus Institute for Aging Research at Hebrew SeniorLife, Harvard Medical School, Boston, MA..
Jean McGuire, Health Sciences Department at Northeastern University, Boston, MA..
John Griffith, Department of Health Sciences, Northeastern University, Boston, MA.
Sean Cahill, Health Policy Research at the Fenway Institute, Affiliate Associate Clinical Professor of Health Sciences at Northeastern University, and Adjunct Associate Professor at Boston University School of Public Health..
Becky Briesacher, Department of Pharmacy and Health Systems Sciences, Northeastern University, Boston, MA..
References
- 1.Centers for Disease Control and Prevention. HIV among people aged 50 and over. 2016. Available at: https://www.cdc.gov/hiv/group/age/olderamericans/index.html. Accessed April 2019
- 2.Nursing Home Data Compendium 2015 Edition. Washington, DC: Department of Health & Human Services: Centers for Disease Control and Prevention; 2015. [Google Scholar]
- 3.Olivieri-Mui B, McGuire J, Griffith J, Cahill S, Briesacher B. Assessing the Quality of Human Immunodeficiency Virus Care in Nursing Homes. J Am Geriatr Soc. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Henry J Kaiser Family Foundation. Medicare and HIV. 2016.
- 5.Meyers DJ, Wilson IB, Lee Y, Cai S, Miller S, Rahman M. The Quality of Nursing Homes That Serve Patients With Human Immunodeficiency Virus. J Am Geriatr Soc. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hellinger FJ. Hospital Use by Persons With HIV in the 21st Century: A 5-State Study. Med Care. 2016; 54:639–44. [DOI] [PubMed] [Google Scholar]
- 7.United States. Department of Health and Human Services. Office of Inspector G. Medicare nursing home resident hospitalization rates merit additional monitoring. Medicare nursing home resident hospitalization rates merit additional monitoring: Washington, D.C. : Department of Health and Human Services, Office of Inspector General; 2013. [Google Scholar]
- 8.Olivieri-Mui B, McGuire J, Cahill S, Griffith J, Briesacher B. People Living With HIV in U.S. Nursing Homes in the Fourth Decade of the Epidemic. J Assoc Nurses AIDS Care. 2019; 30:20–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Guaraldi G, Falutz J, Mussi C, Silva AR. Managing the older adult patient with HIV: Cham: : Adis; 2016. [Google Scholar]
- 10.Hamlyn E, Ewings FM, Porter K, Cooper DA, Tambussi G, Schechter M, et al. Plasma HIV viral rebound following protocol-indicated cessation of ART commenced in primary and chronic HIV infection. PloS one. 2012; 7:e43754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rackal J, Tynan A, Handford CD, Rzeznikiewz D, Agha A, Glazier R. Provider training and experience for people living with HIV/AIDS. Cochrane Database Syst Rev. 2011. [DOI] [PubMed] [Google Scholar]
- 12.Centers For Medicare and Medicaid Services. Skilled nursing facility (SNF) care. 2020. Available from: https://www.medicare.gov/coverage/skilled-nursing-facility-snf-care. Accessed April 2020
- 13.Briesacher BA, Soumerai SB, Field TS, Fouayzi H, Gurwitz JH. Nursing Home Residents and Enrollment in Medicare Part D. J Am Geriatr Soc. 2009; 57:1902–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Davey RT Jr., Bhat N, Yoder C, Chun TW, Metcalf JA, Dewar R, et al. HIV-1 and T cell dynamics after interruption of highly active antiretroviral therapy (HAART) in patients with a history of sustained viral suppression. Proceedings of the National Academy of Sciences of the United States of America. 1999; 96:15109–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Friedman EE, Duffus W. Chronic health conditions in Medicare beneficiaries 65 years old, and older with HIV infection. AIDS. 2016; 30:2529–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McNeish DM, Harring JR. Clustered data with small sample sizes: Comparing the performance of model-based and design-based approaches. Commun Stat - Simul C. 2017; 46:855–69. [Google Scholar]
- 17.MDS 3.0 Quality Measures User’s Manual V9.0 08-15-2015. Centers for Medicare and Medicaid Services; 2015. [Google Scholar]
- 18.The National Quality Forum. About Us. 2019; Available from: http://www.qualityforum.org/About_NQF/. Accessed October 2018
- 19.Aberg JA, Gallant JE, Ghanem KG, Emmanuel P, Zingman BS, Horberg MA. Primary Care Guidelines for the Management of Persons Infected With HIV: 2013 Update by the HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis. 2013; 58:e1–e34. [DOI] [PubMed] [Google Scholar]
- 20.Department of Health and Human Services. Guidelines for the prevention and treatment of opportunistic infections in HIV-infected adults and adolescents: recommendations from the Centes for Disease Control and Prevention, the National Institutes of Health and the HIV Medicine Association of the Infectious Disease Society of America Department of Health and Human Services: Panel on Opportunistic Infections in HIV-Infected Adults and Adolescents; 2013; Available from: http://aidsinfo.nih.gov/contentfiles/lvguidelines/adult_oi.pdf. Accessed March 2016.
- 21.Backus LI, Boothroyd DB, Phillips BR, Belperio PS, Halloran JP, Valdiserri RO, et al. National quality forum performance measures for HIV/AIDS care: the Department of Veterans Affairs’ experience. Arch Intern Med. 2010; 170:1239–46. [DOI] [PubMed] [Google Scholar]
- 22.Li P, Kim Michelle M, Doshi Jalpa A. Comparison of the performance of the CMS Hierarchical Condition Category (CMS-HCC) risk adjuster with the charlson and elixhauser comorbidity measures in predicting mortality. BMC Health Serv Res. 2010; 10:245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Heinze G, Schemper M. A solution to the problem of separation in logistic regression. Stat Med. 2002; 21:2409–19. [DOI] [PubMed] [Google Scholar]
- 24.Heinze G, Wallisch C, Dunkler D. Variable selection - A review and recommendations for the practicing statistician. Biom J. 2018; 60:431–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Group HSRaE. Nursing Facility Total, Average and Median Number of Patients Per Facility and ADL Dependence. CMS OSCAR Data Current Surveys. Washington, DC: American Health Care Association; 2006. [Google Scholar]
- 26.Saliba D, Buchanan J. Development & validation of a revised nursing home assessment tool: MDS 3.0. Rockville, MD: Centers for Medicare and Medicaid Services. 2008. [Google Scholar]
- 27.Slade A, Isa F, Kyte D, Pankhurst T, Kerecuk L, Ferguson J, et al. Patient reported outcome measures in rare diseases: a narrative review. Orphanet J Rare Dis. 2018; 13:61-. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Korthuis PT, McGinnis KA, Kraemer KL, Gordon AJ, Skanderson M, Justice AC, et al. Quality of HIV Care and Mortality Rates in HIV-Infected Patients. Clin Infect Dis. 2015; 62:233–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Harris CM, McKenzie R, Nayak S, Kiyatkin D, Baker D, Kisuule F. Graying of the HIV epidemic: a challenge for inpatient medicine providers. J Community Hosp Intern Med Perspect. 2015; 5:29428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wetherell M eHIVQUAL Indicator Comparison 2014. New York State Department of Health AIDS Institute; Available from: https://www.health.ny.gov/diseases/aids/ending_the_epidemic/docs/key_resources/data_committee_resources/ehivqual_comparison.pdf. Accessed April 2019 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Flow chart to determine sample size.
Supplemental Table 1. Comparing Medicare eligible PLWH included and excluded from the study