key: cord-1019491-hjgf3ay8
authors: Griffith, Matthew F.; Levy, Cari R.; Parikh, Toral J.; Stevens-Lapsley, Jennifer E.; Eber, Leslie B.; Palat, Sing-I T.; Gozalo, Pedro L.; Teno, Joan M.
title: Nursing Home Residents Face Severe Functional Limitation or Death After Hospitalization for Pneumonia
date: 2020-10-21
journal: J Am Med Dir Assoc
DOI: 10.1016/j.jamda.2020.09.010
sha: 0e99b80f31c8091bff79b915dc81924cf1185597
doc_id: 1019491
cord_uid: hjgf3ay8

OBJECTIVES: Pneumonia is a common cause of hospitalization for nursing home residents and has increased as a cause for hospitalization during the COVID-19 pandemic. Risks of hospitalization, including significant functional decline, are important considerations when deciding whether to treat a resident in the nursing home or transfer to a hospital. Little is known about postdischarge functional status, relative to baseline, of nursing home residents hospitalized for pneumonia. We sought to determine the risk of severe functional limitation or death for nursing home residents following hospitalization for treatment of pneumonia. DESIGN: Retrospective cohort study. SETTING AND PARTICIPANTS: Participants included Medicare enrollees aged ≥65 years, hospitalized from a nursing home in the United States between 2013 and 2014 for pneumonia. METHODS: Activities of daily living (ADL), patient sociodemographics, and comorbidities were obtained from the Minimum Data Set (MDS), an assessment tool completed for all nursing home residents. MDS assessments from prior to and following hospitalization were compared to assess for functional decline. Following hospital discharge, all patients were evaluated for a composite outcome of severe disability (≥4 ADL limitations) following hospitalization or death prior to completion of a postdischarge MDS. RESULTS: In 2013 and 2014, a total of 241,804 nursing home residents were hospitalized for pneumonia, of whom 89.9% (192,736) experienced the composite outcome of severe disability or death following hospitalization for pneumonia. Although we found that prehospitalization functional and cognitive status were associated with developing the composite outcome, 53% of residents with no prehospitalization ADL limitation, and 82% with no cognitive limitation experienced the outcome. CONCLUSIONS AND IMPLICATIONS: Hospitalization for treatment of pneumonia is associated with significant risk of functional decline and death among nursing home residents, even those with minimal deficits prior to hospitalization. Nursing homes need to prepare for these outcomes in both advance care planning and in rehabilitation efforts.

Pneumonia is a common cause of illness among nursing home residents, with approximately 1 case occurring for every 1000 days of nursing home care provided, prior to the current COVID-19 global pandemic. 1 Pneumonia was also the leading cause of hospitalization and death among nursing home residents, accounting for 40% of hospitalizations and 25% to 33% of deaths prior to the current pandemic. 2e7 The current global COVID-19 pandemic has disproportionately affected American nursing home residents, with at least 1 in 11 nursing home residents suffering a confirmed SARS-CoV-2 infection, and has led to an increase in hospitalizations for pneumonia. 8 Approximately 1 in 3 nursing home residents with COVID-19 has been hospitalized, primarily because of pneumonia symptoms (cough, fever, and shortness of breath) and more than 1 in 4 (27%) has died. 9, 10 Advance care planning for episodes of pneumonia was already challenging for residents and facilities; however, the uncertainty surrounding short and long-term effects of hospitalization for pneumonia during the current pandemic has amplified the difficulty of advance care planning. At the same time, Centers for Medicare & Medicaid Services (CMS) considers pneumonia the leading cause of potentially avoidable hospitalizations among nursing home residents and penalizes nursing homes and hospitals for avoidable pneumonia readmissions. 11 The conflicting recommendation by triage tools and health care organizations makes it challenging for nursing home residents, their caregivers, and providers to make informed decisions about the most appropriate treatment decision. To reach informed decisions around hospitalization for pneumonia (including COVID- 19) , residents, families, and caregivers must understand patient preferences, potential benefits, and potential harms. 12 Functional decline is a potential harm of inpatient pneumonia care that must be factored into shared decision making regarding hospitalization for pneumonia. Prior studies have found that functional decline is a potential outcome of hospitalization, particularly among patients with cognitive impairment. 13, 14 Loss of a single activities of daily living (ADL) is associated with a decline in health-related quality of life and an increased likelihood of requiring long-term nursing home care. 15 Among patients with mild to moderate functional limitation at baseline, loss of 1 ADL could have a profound effect on a patient's health-related quality of life and lead to the perception that such a loss would be "worse than death." 16, 17 Although patients and their surrogates may be familiar with the risks of hospitalization for exacerbations of chronic diseases that they have suffered for many years (eg, chronic obstructive pulmonary disease or congestive heart failure), providing them with an estimate of functional outcomes following hospitalization for an unplanned acute illness like pneumonia allows for shared decision making through scenario planning. 18, 19 Every nursing home resident whose care is paid for by Medicare, Medicaid, or the Veterans Health Administration is evaluated using the Minimum Data Set (MDS), a federally mandated, validated instrument that assesses health conditions, disease diagnoses, treatments, and functional and cognitive status. 20 MDS evaluations are completed on admission to the nursing home and quarterly thereafter, as well as at the time of acute changes in clinical status and readmission from a hospital. Information from MDS assessments submitted to CMS. The MDS assessment includes an evaluation of ADL independence available prior to and following hospitalization to quantify functional decline in the perihospitalization period for nursing home residents.

This study was designed to assist nursing home residents, surrogates, providers, and facilities conduct appropriate scenario planning to prepare for unplanned acute care hospitalizations, by describing functional decline among nursing home residents hospitalized for pneumonia and evaluating patient and hospitalization characteristics associated with severe functional decline and death. These findings will aid patients, families, and caregivers in advance care planning and decisions regarding hospitalization during the current pandemic.

This study evaluated CMS data for patients hospitalized with pneumonia identified using Medicare Part A claims submitted between January 1, 2013, and discharge by October 31, 2014.

This was a retrospective cohort analysis of nursing home residents, receiving either post-acute or long-term care, hospitalized for pneumonia during their nursing home stay. After initially identifying all fee-for-service Medicare beneficiaries with a claim submitted for hospitalization during the study period, patients were identified as nursing home residents based on the completion of a MDS assessment during the 120 days prior to the index hospital admission. Hospitalized nursing home residents were included if they had (1) a primary discharge diagnosis of pneumonia or a primary diagnosis of septicemia and secondary diagnosis of pneumonia associated with the hospitalization, (2) age 66 years; and (3) resided in United States (excluding Puerto Rico). We included patients with a primary diagnosis of sepsis and secondary diagnosis of pneumonia as there has been a substantial increase in coding patients with clinical signs and symptoms of pneumonia as having sepsis over the 10 years prior to the study. 21 Nursing home residents were excluded if they were (1) discharged to another hospital following the index admission, as we sought to collect information for only 1 episode of inpatient care per patient, or (2) lacked an MDS assessment submitted within 60 days following the index admission. The first claim submitted during the study period was considered the index hospitalization. This study was reviewed and approved by our institutional review board, who waived the requirement for patient consent.

The primary outcome was a composite of severe functional limitation or death following hospitalization. Severe functional limitation was defined as 4 or more ADL impairments documented in the MDS-ADL Long Form scale completed posthospitalization. ADL assessed bed mobility, transfer, locomotion, dressing, eating, toilet use, and personal hygiene. 22 Most individuals with 4 or more ADL limitations on this scale are functionally bed bound; therefore, we set this cutoff for our definition of severe functional limitation. 23 Failure to perform an ADL without assistance was considered an ADL deficiency. Death was defined as dying during the hospitalization or within 60 days of discharge.

In order to identify whether hypothesized patient characteristics and comorbidities would be associated with functional decline following hospitalization, specific comorbidities and patient characteristics of interest were included in an a priori model. Sociodemographic characteristics including age, race and ethnicity, sex, and state of residence were identified from the Medicare Beneficiary Enrollment file. Patient cognitive status, baseline ADL limitations, active medical comorbidities contributing to the risk of functional limitation (dementia, cancer, aphasia, congestive heart failure, diabetes, stroke, paraplegia, hip fracture, renal failure, schizophrenia, and chronic lung disease) 24e26 and significant patient conditions, including weight loss (loss of 5% or more in last month or 10% or more in last 6 months), difficulty eating, and dependency on a feeding tube, were identified from the MDS completed 120 days or less before the index hospitalization. The model included receipt of invasive mechanical ventilation using procedure codes (Current Procedural Terminology [CPT]).

Bivariate analyses were performed to describe differences in baseline characteristics and characteristics of hospitalization between patients who did and did not experience the outcome. A multivariable logistic regression model was created to evaluate the association between covariates defined in our a priori model and our composite outcome of severe disability and death, adjusted for receipt of mechanical ventilation. Analyses allowed for robust variance estimates to account for clustering of persons within hospitals. All data analysis was conducted in Stata, version 15.0 (Stata Corp, College Station, TX)).

Between January 1, 2013, and October 31, 2014, a total of 256,012 Medicare beneficiaries were hospitalized for pneumonia with an MDS completed within 120 days prior to hospitalization. Among those patients, 41,505 (16.2%) did not have an MDS completed within 60 days following discharge and did not die during that period, so they were excluded. Among excluded patients, 13,822 (33.3%) were discharged to home, an assisted living facility, or an adult family home, 5356 (12.9%) were discharged to hospice care, and 7361 (17.7%) were readmitted to the hospital before their next MDS assessment. The remaining 14,966 (36.1%) patients excluded from the cohort lacked an MDS assessment submitted within 60 days but were discharged from the hospital alive and did not go to any of the previously listed locations. Likely they were discharged to a nursing home but an MDS was not submitted in a timely fashion. The resulting cohort consisted of 214,507 patients. Among patients in this cohort, 56.0% (n ¼ 135,438) were female, 85.1% white (n ¼ 182,298), and 10.2% black (n ¼ 21,774), and had a mean age of 83.1 years (standard deviation 8.1) (Table 1) . Overall, the cohort had a high prevalence of cognitive impairment (Cognitive Performance Score >2, 60.1%, n ¼ 129,005), and 10.1% (n ¼ 19,816) had a feeding tube. Nearly half of patients (47.4%, n ¼ 101,791) had severe disability before hospitalization (Table 2) .

Among the members of the cohort, 79,558 (37.1%) died during or within 60 days of hospitalization and 113,228 (52.8%) had severe disability following admission, for a total of 192,736 (89.9%) with the primary composite outcome. Among patients without severe disability prior to hospitalization (<4 ADL limitations, n ¼ 46,702), 66.3% (n ¼ 30,982) experienced the primary outcome of severe disability or death. The majority of patients with no prehospitalization ADL limitations (52.5%) experienced the composite outcome, as did the majority of patients with all levels of prehospitalization functional limitation. Among patients with severe disability prior to hospitalization, 96.4% (n ¼ 161,754) experienced the primary outcome and 39.3% (n ¼ 66,014) of these patients died during or within 60 days of hospitalization.

Severe functional limitation and dementia prior to hospitalization were associated with severe functional limitation or death following hospitalization for pneumonia (Table 3) . Adjusting for prehospitalization functional and cognitive status, male gender, race (black non-Hispanic, Asian, and Hispanic), certain pre-existing medical comorbidities [congestive heart failure (adjusted odds ratio 1.07, 95% confidence interval 1.03-1. Patients with pre-existing feeding tubes were not at increased risk of the outcome (0.98, 0.87-1.10). Patients who received mechanical ventilation during the hospitalization were at increased risk of the outcome (1.24, 1.19-1.29). Being admitted from a stay in long-term care (admissions not following a prior hospitalization) was associated with increased risk of the primary outcome (1.14, 1.10-1. 19 ).

In the largest study to date of functional decline following hospitalization for nursing home residents, we found that most residents hospitalized for pneumonia developed severe disability or died. Notably, among those with no prehospitalization functional limitation, the majority developed significant disability or experienced death by 60 days following hospitalization. Individuals with severe cognitive, physical or neurologic dysfunction prior to hospitalization were at the greatest risk of severe disability or death. These findings suggest that, during the current pandemic, nursing homes must plan for the influx of a high proportion of nursing home residents returning with severe functional limitation requiring a higher level of care. Additionally, during this time, when admitting new residents or readmitting former residents, providers should discuss likely outcomes following hospitalization for pneumonia to inform advance care planning.

Our findings build on prior studies demonstrating persistent functional decline following hospitalization for infectious and noninfectious acute care admissions among community-dwelling older adults, particularly for those with cognitive and functional impairment assessed on admission. 13,14,27e32 These studies have demonstrated that almost every group of older adults is at risk for a perceptible decline in function following hospitalization that can last for months after discharge or become permanent. However, because community dwelling adults do not undergo detailed functional assessments as performed as part of the MDS, these studies relied on patient or caregiver report of functional status at the time of hospitalization or most recent primary care assessment of functional status. Use of MDS data ensured that we had an accurate assessment of function as close to the time of admission as possible, and a similarly accurate assessment following hospitalization. The few prior studies of functional decline among nursing home patients following pneumonia have relied on assessments performed up to 6 months prior to admission or performed assessments of functional status at the time of discharge, which may have over-or underestimated the severity of function in the days to weeks following discharge. 33, 34 This study is the first to use MDS data to characterize functional decline following hospitalization for pneumonia for the entire spectrum of patients in nursing homes, including patients with few functional and cognitive limitations receiving post-acute care and those with more severe baseline limitations residing in long-term care. Engaging nursing home residents and their families in advance care planning reduces bothersome or burdensome care received at the end of life and improves family member and caregiver satisfaction with care at the end of life. 35e38 However, approximately 4 of 10 nursing home residents fail to have any advance care plan documented. 39, 40 A major barrier to advance care planning is choosing the correct time to initiate it. Discussions occurring too early may not accurately reflect patient preferences over time and discussions that occur too late fail to protect patients from receiving care that was not consistent with their wishes. 41 When informed about the high likelihood of a poor outcome, nursing home residents, their families, and their providers are more likely to engage in advance care planning. 42 Our study demonstrates that nursing home admission, particularly with individuals suffering from cognitive and physical limitations who are admitted for long-term care, is likely the right time to initiate advance care planning around the decision to hospitalize for COVID or other pneumonias, as we found that residents are more likely to experience functional decline or death at 60 days following hospitalization than individuals diagnosed with metastatic nonesmall cell lung cancer. 43 Presently, only 6% to 7% of nursing home residents have documented do-not-hospitalize orders, with the proportion ranging from 4% among those with mild cognitive impairment to 8% to 9% among those with severe impairment. 39,44e46 We anticipate that once aware of how poor prognosis is after hospitalization, nursing home residents, their families, and their providersdwho are already concerned about visitation limitations once hospitalized for COVIDdwill be more likely to engage in advance care planning and the decision not to hospitalize, rather than delay these discussions.

Do-not-hospitalize orders can be effective in preventing hospitalization and burdensome treatment for COVID or other causes of pneumonia, particularly given the stresses of relocation to the hospital and the limited contact family members and caregivers may have with their loved one once hospitalized; however, many patients and families change their minds during episodes of acute illness and agree to hospitalization regardless of documented preferences. 39,47e50 We believe that, in many cases, this is the result of a lack of appropriate scenario planning between patients, their families or caregivers, and nursing home providers. As our findings suggest that the risk of severe functional limitation and death varies among patient populations, there are specific populations of patients that are at greater risk for severe functional limitation and death following hospitalization for pneumonia, suggesting that scenario planning during advance care planning conversations could include discussion of these risk factors when making decisions about COVID hospitalization and do-nothospitalize orders.

There are limitations to this study. First, all patients in our study were hospitalized for pneumonia. Most nursing home residents who develop pneumonia (w65%) are treated in their facility rather than in the hospital. 45, 51, 52 Despite this, septicemia and pneumonia remain the most common reasons for hospitalization of nursing home residents. 53 We were unable to reliably identify individuals treated for pneumonia in the nursing home; however, prior observational studies have demonstrated that individuals hospitalized for pneumonia are no more likely to survive than those treated in facilities and likely experience a reduction in quality of life after hospitalization. 45, 54 Additionally, fewer than 1 in 10 residents with critical illness due to pneumonia are hospitalized, suggesting that our cohort of patients may not have had more severe disease than those treated in facilities. 52 Therefore, it is plausible that we would not find a difference in mortality between the cohort of patients included in this study and those who were treated in the facility. 55 Second, few if any individuals were hospitalized for viral pneumonia due to coronaviruses such as COVID-19, SARS, or MERS. Despite this, we can generalize that the severity of disease and prognosis among those included in our study was similar to those hospitalized with COVID pneumonia, as we found an overall mortality of 37%, higher than the overall mortality for nursing home residents with COVID (27%). At this time, there are no estimates of inpatient mortality for nursing home residents with COVID to provide a direct comparison. Third, we may not have captured all hospitalizations during which pneumonia was treated, as we only looked for individuals with a primary diagnosis of pneumonia or septicemia. Patients with hospital-acquired pneumonia or ventilator-associated pneumonia likely were not included in our study; however, our primary interest was to identify patients who would benefit from scenario planning in the event of pneumonia and not to identify those who acquired pneumonia during a hospital transfer for another indication (eg, stroke, myocardial infarction, or hip fracture).

Because of the high prevalence of comorbid medical conditions, functional limitations, and dementia, nursing home residents are at risk for severe complications resulting from COVID-19. Therefore, it is crucial to discuss possible outcomes of seemingly routine care with patients and their families well in advance so that appropriate decisions can be made in times of acute illness. The results of this study should inform patient and family decisions regarding hospitalization for COVID-19 and other causes of pneumonia at the time of admission to the nursing home as well as inform planning for a high proportion of hospitalized patients returning with new or significantly worsened functional limitations. Further research evaluating severity and duration of decline caused by COVID-19erelated pneumonia from other etiologies is warranted as there are additional comorbidities associated with COVID-19 such as thrombosis and myocarditis that may exacerbate functional loss. 56, 57 Additionally the social isolation resulting from COVID-related closures of nursing homes to outside visitors, leading to functional and cognitive decline among individuals without acute illnesses, warrants further study as well.

Pneumonia in residents of long-term care facilities: Epidemiology, etiology, management, and prevention

Potentially avoidable hospitalizations for elderly long-stay residents in nursing homes

Potentially avoidable hospitalizations of dually eligible Medicare and Medicaid beneficiaries from nursing facility and home-and community-based services waiver programs

Hospital transfers of nursing home residents with advanced dementia

Frequency and pattern of emergency department visits by long-term care residentsda population-based study

Causes of death in elderly nursing home residents

Autopsies and death certificates in the chronic care setting

Division of Nursing Homes/Quality Safety and Oversight Group/Center for Clinical Standards and Quality

Special-Programs-Initiatives-COVID-19-Nursing-Home/COVID-19-Nursing-Home-Dataset/s2uc-8wxp2020

Characteristics associated with hospitalization among patients with COVID-19dMetropolitan Atlanta, Georgia, MarcheApril 2020

Initial and repeated point prevalence surveys to inform SARS-CoV-2 infection prevention in 26 skilled nursing facilitiesdDetroit

Medicare-Medicaid eligible beneficiaries and potentially avoidable hospitalizations

Shared decision makingdPinnacle of patientcentered care

Physical function and disability after acute care and critical illness hospitalizations in a prospective cohort of older adults

Elderly patients with cognitive impairment have a high risk for functional decline during hospitalization: The GIFA Study

The onset of ADL difficulties and changes in healthrelated quality of life

States worse than death among hospitalized patients with serious illnesses

Measuring preferences for health states worse than death

Intervention thresholds: A conceptual frame for advance care planning choices

Exploring the utility of the vignette technique in promoting advance care planning discussions with cancer patients and caregivers

Making the investment count: Revision of the Minimum Data Set for nursing homes, MDS 3.0

Comparison of trends in sepsis incidence and coding using administrative claims versus objective clinical data

Measuring change in activities of daily living in nursing home residents with moderate to severe cognitive impairment

Scaling ADLs within the MDS

Cognition moderates the relationship between facility characteristics, personal impairments, and nursing home residents' activities of daily living

Dementia is the major cause of functional dependence in the elderly: 3-year follow-up data from a populationbased study

Functional limitation and disability associated with congestive heart failure

Long-term cognitive impairment and functional disability among survivors of severe sepsis

The role of intervening illnesses and injuries in prolonging the disabling process

The relationship between intervening hospitalizations and transitions between frailty states

Change in disability after hospitalization or restricted activity in older persons

Recovery of activities of daily living in older adults after hospitalization for acute medical illness

Trajectories of disability among older persons before and after a hospitalization leading to a skilled nursing facility admission

Adverse effects of pneumonia on physical functioning in nursing home residents: Results from the INCUR study

Changes in physical function after hospitalization in patients with nursing and healthcare-associated pneumonia

End-of-life transitions among nursing home residents with cognitive issues

Satisfaction with end-of-life care for nursing home residents with advanced dementia

Improving the end-of-life for people with dementia living in a care home: An intervention study

Systematic implementation of an advance directive program in nursing homes: A randomized controlled trial

Are hospital/ED transfers less likely among nursing home residents with do-not-hospitalize orders?

Advance directives among nursing home residents with mild, moderate, and advanced dementia

Strategic targeting of advance care planning interventions: The Goldilocks phenomenon

Improving end-of-life outcomes in nursing homes by targeting residents at high-risk of mortality for palliative care: Program description and evaluation

Survival by histologic subtype in stage IV nonsmall cell lung cancer based on data from the Surveillance, Epidemiology and End Results Program

Racial differences in hospitalizations of dying Medicare-Medicaid dually eligible nursing home residents

The cost-effectiveness of the decision to hospitalize nursing home residents with advanced dementia

Advance directives in nursing home residents aged > or ¼65 years: United States

Interventions associated with the management of suspected infections in advanced dementia

Changes in care goals and treatment orders around the occurrence of health problems and hospital transfers in dementia: A prospective study

Call the family instead of calling the ambulance

Do-not-hospitalize orders for individuals with advanced dementia: Healthcare proxies' perspectives

Effects of nursing home ownership type and resident payer source on hospitalization for suspected pneumonia

A multifaceted intervention to implement guidelines did not affect hospitalization rates for nursing home-acquired pneumonia

Transitions between nursing homes and hospitals in the elderly population

Does hospitalization impact survival after lower respiratory infection in nursing home residents?

Pneumonia management in nursing homes: Findings from a CMS demonstration project

Thrombosis in hospitalized patients with COVID-19 in a New York City health system

Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China