Abstract
Background:
Despite decreases in readmissions among Medicare beneficiaries after the implementation of the Hospital Readmissions Reduction Program, older adults living with multiple chronic conditions (MCCs) continue to experience higher readmission rates. Few strategies leverage nursing to identify patients at risk for readmission.
Objectives:
Examine the effect of nurse assessments of discharge readiness on 30-day readmissions.
Research Design:
Cross-sectional study linking 3 secondary data sources (i.e., nurse survey, hospital survey, and Medicare claims data) representing 424 hospitals.
Subjects:
188,806 Medicare surgical patients with MCCs.
Measures:
Discharge readiness was derived from the 2016 RN4CAST-US survey. Medicare claims data was used to determine MCC count. The outcome was 30-day readmissions across MCC count.
Results:
The average discharge readiness score was 0.45 (range = 0–0.86) indicating that, in the average hospital, less than 50% of nurses were confident their patient or caregiver could manage their care after discharge. Nearly 8% of patients were readmitted within 30 days of discharge; the highest rates of readmissions were among individuals with ≥5 MCCs (4,293, 13.50%). For each 10% increase in the proportion of nurses in a hospital who were confident in their patients’ discharge readiness, the odds of 30-day readmission decreased by 2% (95% CI [0.96, 1.00]; p = 0.028) for patients with 2–4 MCCs and 3% (95% CI [0.94, .99]; p = 0.015) for patients with ≥5 MCCs, relative to patients with 0–1 MCCs.
Conclusions:
Nurse assessments of discharge readiness may be a useful signal for hospitals to reduce readmissions and examine factors interfering with discharge processes.
Keywords: Multimorbidity, multiple chronic conditions, older adults, readmissions, discharge readiness, nursing
INTRODUCTION
Older adults living with multiple chronic conditions (MCCs) experience worse outcomes during and after hospitalization, including adverse events, functional decline, mortality, and readmissions.1–3 Inadequate pre-discharge preparation, lack of patient and family readiness for discharge, and poor discharge care coordination are all contributors to unplanned readmissions.4 Optimal discharge planning is crucial for older adults living with MCCs who often experience difficulty post-discharge with performing activities of daily living, frequent changes in health status, and multiple transitions in care settings and providers.5,6 A core responsibility of registered nurses is ensuring that patients are prepared for discharge and thus are a key resource to preventing unplanned readmissions.
Although the timing of discharge is a team decision,7 nurses are ideally positioned to assess patient’s readiness for discharge due to their bedside presence and involvement in virtually all aspects of patient care throughout the hospital stay.8 For example, in addition to monitoring patient status and providing direct care, nurses are able to assess social risk factors which can contribute to readmission likelihood such as patients’ access to outpatient care and reliable transportation.9 Nurses are also able to assess patients’ health literacy, housing, and determine if patients have a strong support network (i.e., formal or informal caregiving support) —critical information which can be integrated into the discharge planning process.9,10 Nurses take on much of the responsibilities of the discharge process, including education and care coordination.8 As such, nurses’ assessments of discharge readiness show promise in identifying patients at high risk for readmission; furthermore, aggregating nurses’ discharge readiness assessments may provide a novel means of distinguishing differences in the overall capacity of a hospital to provide consistent and high-quality care that minimizes readmission risk.
Nurse reported discharge readiness is a promising avenue to reduce readmissions as research suggests that nurses are reliable informants of healthcare quality and patient safety.11,12 Nurses’ assessments of care quality have been shown to be associated with objective patient outcomes, including mortality and failure to rescue, or patient death after an adverse event, suggesting that nurses may have important knowledge regarding discharge readiness.11 However, limited research has evaluated nurses’ assessments of discharge readiness and their link to patient outcomes. The research to date has assessed the linkages between nurses’ assessments of patient discharge readiness and post-discharge utilization, including emergency department visits and hospital readmissions, and the large majority of this work has not included sufficient numbers of hospitals to understand how nurses’ assessments of readiness are associated with readmissions across hundreds of hospitals.4,7,13 For example, Weiss and colleagues found that nurses’ assessments of discharge readiness were associated with fewer readmissions in a sample of 162 medical-surgical patients discharged from 1 of 4 Midwestern hospitals4 and in a sample of 254 adult medical-surgical patients discharged from one tertiary medical center in Eastern US.14 Utilizing this evidence to inform practice, Weiss and colleagues developed a randomized control trial that included a structured assessment of discharge readiness implemented across 33 Magnet hospitals, but found that neither nurse or patient assessments of discharge readiness successfully reduced readmissions.13
Our study builds on this previous research as there has been no formal evaluation of whether variation of nurses’ assessments of discharge readiness across hundreds of hospitals are associated with readmissions among older adults living with MCCs, a population at high risk for readmission. By using a large and representative dataset, spanning 424 hospitals, and using the assessments of thousands of nurses, we seek to understand whether nurses’ assessments of discharge readiness are useful for identifying patients at high risk for readmission.
METHODS
Design and Data
This was a secondary analysis of three cross-sectional data sources that were linked using a common hospital identifier. The 2016 RN4CAST-US survey provided information on nurse assessments of discharge readiness from a 30% random sample of licensed registered nurses who were surveyed in California, Florida, New Jersey, and Pennsylvania.15 There was a 26% initial response rate, and an intensive nonresponse survey was completed, yielding an 87% response rate.15 Statistically significant differences were not observed between initial respondents and nonrespondents on most hospital measures, minimizing the concern for nonresponse bias.15 These four states were selected for their diversity of urban and rural regions and because they include approximately 25% of the annual discharges from Medicare beneficiary hospitals.15 The 2016 American Hospital Association (AHA) Annual Survey provided information on hospital characteristics, including the number of hospital beds, teaching status, and technology status. Patient data was obtained from the Center for Medicare and Medicaid Services (CMS) Medicare Provider Analysis and Review (MedPAR) 2016 annual data set.
Sample
Hospitals.
The analytic sample included, on average, 26 nurses (range: 10–127) from each of the 424 non-federal, acute care hospitals located in California, Florida, New Jersey, and Pennsylvania.
Nurses.
We limited the nurse sample to include only direct care, hospital nurses working on adult units, including adult medical or surgical, oncology, ICU, the emergency department, operating room/recovery room, psychiatric unit, hospice/palliative care, outpatient/same day/procedures, and rehabilitation/long-term care. Nurses working on pediatric, maternity/newborn, PICU, NICU, or “other hospital setting” were excluded from the analysis.
Patients.
The total patient sample included 188,806 Medicare fee-for-service beneficiaries ages 65+ who were discharged alive following hospitalization for a general, orthopedic, or vascular surgical procedure as identified by Medicare Severity Diagnosis Related Groups (MS-DRG) (e.g., Appendix 1). Patients were excluded if they left against medical advice, were discharged to a location other than home/self-care and had more than one surgical procedure during the index admission. Surgical patients were used because there is well-validated risk adjustment and because these populations have common procedures across acute care hospitals.16–18 In the future, other patient populations and settings should be considered.
Measures
Discharge Readiness.
Nurses’ assessments of discharge readiness were derived from the 2016 RN4CAST-US survey, which included a question that asked nurses to rate their level of confidence in their patients or their caregivers ability to manage care after discharge on a 4-point Likert-type scale ranging from ‘not at all confident’ to ‘very confident’. Individual nurse responses of discharge readiness were dichotomized into ‘not at all confident’/’somewhat confident’ and ‘confident’/’very confident’ and averaged at the hospital level to produce the proportion of nurses who were confident in their patient being ready for discharge. This measure has been previously used to describe variation in discharge readiness across hospitals.19 For statistical modelling, we scaled the hospital-level discharge readiness variable so that any effect on readmissions corresponds with an increase of 10% in the proportion of nurses who are confident their patients or caregivers can manage care after discharge (i.e., discharge readiness).
Multiple Chronic Conditions.
A list of 20 MCCs reported and defined by the U.S. Department of Health and Human Services (HHS) Office of the Assistant Secretary of Health (OASH)20 was used to identify patients with MCCs. ICD-10 codes were used to classify each condition. Only 15 of the 20 conditions were included in this study because CMS does not measure all 20 OASH conditions (i.e., autism, hepatitis, HIV, schizophrenia, and substance abuse disorders are excluded). For our analysis, the number of chronic conditions were categorized into levels of 0–1 MCCs, 2–4 MCCs, and ≥5 MCCs.
30-day Readmissions.
CMS’ validated Risk-Standardized Readmission Measures were used to determine all-cause 30-day readmissions to any adult, non-federal acute care hospital following discharge to home (i.e., to self-care) from the index hospitalization for a general, orthopedic, or vascular surgical procedure.21–24
Covariates
We controlled for patient and hospital characteristics to minimize potential confounding and better determine the relationship between nurse assessments of discharge readiness and 30-day readmissions. For patient characteristics, we controlled for age, sex, number of MCCs, and MS-DRG. One hundred and seventy MS-DRG codes were used for the identification of surgical procedure type (i.e., general, orthopedic, and vascular surgery).
Structural hospital characteristics were obtained from the AHA data set. We controlled for the number of hospital beds, teaching status, and technology status as these characteristics have been shown to be associated with patient outcomes and account for other potentially confounding structural hospital characteristics.25 Number of hospital beds was categorized into ≤100 beds (small), 101–250 beds (medium), and >250 beds (large). Teaching status was grouped into nonteaching with no residents/fellows, minor teaching with a ratio of 1:4 residents/fellows to bed, or major teaching with a ratio of >1:4 residents/fellows to bed.26,27 Technology status was dichotomized so that high technology included hospitals with the capacity to perform open heart surgery and/or organ transplant.26,27
Data Analysis
Frequencies and percentages were used to describe discharge readiness across hospitals. Chi-squared tests of significance for categorical variables and one-way analysis of variance (ANOVA) for continuous variables were used. The same descriptive analyses were used to describe the patient sample overall and across the levels of MCCs. For statistical modelling, we used multivariable regression to estimate the relationship between nurses’ assessments of discharge readiness and readmissions among older adults living with MCCs. To account for the clustering of patients within hospitals, we used Huber-White sandwich estimators.28 We sequentially tested our models, starting with the unadjusted model to estimate the relationship between discharge readiness and MCCs (separately) and 30-day readmission. Model 2 controlled for patient and hospital characteristics. Model 3 tested the interaction effects of discharge readiness and MCCs on 30-day readmission. We then conducted a post-hoc stratified analysis to further determine the interaction effects of discharge readiness on 30-day readmissions across the levels of MCCs. Lastly, using expected frequencies derived from the interaction model, we estimated the counterfactual for readmissions, or what the expected reduction in readmissions would be if patients received care in hospitals with nurses who reported greater confidence in discharge readiness. We then used cost data to determine an estimate of the cost savings that would result in greater discharge readiness. All statistical analyses were completed in STATA Version 17.0.29 with the significance level set at <0.05.
RESULTS
Descriptives
Hospitals.
Table 1 presents the characteristics of the 424 hospitals overall and relative to discharge readiness. Hospitals were divided into quartiles based on the average discharge readiness score, with the 1st quartile including hospitals with the lowest average discharge readiness score (1st quartile mean=0.23, range=0–0.33) and the 4th quartile including hospitals with the highest average discharge readiness score (4th quartile mean=0.65, range=0.56–0.86). The average discharge readiness score for all hospitals was 0.45 (range=0–0.86) with higher scores indicating higher proportions of nurses who are confident in patients’ readiness for discharge. Although there was variation in discharge readiness across hospitals, the average discharge readiness score did not vary significantly as a function of any of the hospital characteristics.
Table 1.
Characteristics of Hospital Sample by Discharge Readiness
| Overall (n = 424) |
Q1 (n = 107) |
Q2 (n = 109) |
Q3 (n = 102) |
Q4 (n = 106) |
p-value | |
|---|---|---|---|---|---|---|
| Discharge Readiness, m (SD) | .45 (.16) | .23 (.08) | .41 (.03) | .51 (.03) | .65 (.07) | -- |
| Number of Hospital Beds | ||||||
| Small (≤ 100 beds) | 14 (3.30) | 4 (28.57) | 2 (14.29) | 4 (28.57) | 4 (28.57) | 0.416 |
| Medium (101–250 beds) | 159 (37.50) | 48 (30.19) | 39 (24.53) | 31 (19.50) | 41 (25.79) | |
| Large (> 250 beds) | 251 (59.20) | 55 (21.91) | 68 (27.09) | 67 (26.69) | 61 (24.30) | |
| Technology Status | ||||||
| High | 261 (61.56) | 59 (22.61) | 68 (26.05) | 71 (27.20) | 63 (24.14) | 0.180 |
| Teaching Status | ||||||
| None | 178 (41.98) | 36 (20.22) | 48 (26.97) | 43 (24.16) | 51 (28.65) | 0.377 |
| Minor | 202 (47.64) | 60 (29.70) | 51 (25.25) | 49 (24.26) | 42 (20.79) | |
| Major | 44 (10.38) | 11 (25.00) | 10 (22.73) | 10 (22.73) | 13 (29.55) |
Note. Discharge readiness reflects the proportion of nurses who are confident their patient or caregiver can manage their care after discharge; range (0–1).
Patients.
Characteristics of the surgical patient sample overall and across the levels of MCCs are presented in Table 2 (n=188,806). As age increased, the number of MCCs a patient had also increased, and females had fewer MCCs compared to males. Patients who underwent vascular surgery had the greatest number of chronic conditions with nearly 50% having ≥5 MCCs, followed by general (30%) and then orthopedic (20%). Consistent with reports, the most frequently occurring comorbidities were hypertension (72%) and hyperlipidemia (92,804, 49%)30. The least common comorbidities were dementia (2%) and stroke (0.15%). A total of 14,704 (7.79%) individuals were readmitted within 30 days of discharge from the hospitals with the highest rates of readmissions among individuals with ≥5 MCCs (4,293, 13.50%).
Table 2.
Characteristics of Surgical Patient Sample by MCCs
| Overall (n = 188,806) |
0–1 CCs (n = 49,878) |
2–4 CCs (n = 107,132) |
5+ CCs (n = 31,796) |
p-value | |
|---|---|---|---|---|---|
| Age, m (SD) | 73.65 (6.56) | 72.33 (5.85) | 73.91 (6.58) | 74.86 (7.12) | <.0001 |
| Gender | |||||
| Male | 97,359 (51.57) | 22,079 (44.27) | 56,687 (52.91) | 18,593 (58.48) | <.0001 |
| Female | 91,447 (48.43) | 27,799 (55.73) | 50,445 (47.09) | 13,203 (41.52) | |
| Surgical Group | |||||
| General | 65,501 (34.69) | 19,611 (39.32) | 36,307 (33.89) | 9,583 (30.14) | <.0001 |
| Orthopedic | 75,386 (39.93) | 27,110 (54.35) | 41,808 (39.02) | 6,468 (20.34) | |
| Vascular | 47,919 (25.38) | 3,157 (6.33) | 29,017 (27.09) | 15,745 (49.52) | |
| OASH Comorbidity | |||||
| Hypertension | 136,250 (72.16) | 15,933 (31.94) | 89,934 (83.95) | 30,383 (95.56) | <.0001 |
| Hyperlipidemia | 92,804 (49.15) | 4,140 (8.30) | 62,602 (58.43) | 26,062 (81.97) | <.0001 |
| CAD | 61,191 (32.41) | 1,633 (3.27) | 36,410 (33.99) | 23,148 (72.80) | <.0001 |
| Diabetes | 43,967 (23.29) | 1,097 (2.20) | 25,964 (24.24) | 16,906 (53.17) | <.0001 |
| CKD | 33,163 (17.56) | 603 (1.21) | 16,354 (15.27) | 16,206 (50.97) | <.0001 |
| Asthma | 29,781 (15.77) | 832 (1.67) | 13,627 (12.72) | 15,322 (48.19) | <.0001 |
| Cancer | 24,424 (12.94) | 2,773 (5.56) | 14,662 (13.69) | 6,989 (21.98) | <.0001 |
| Arthritis | 23,809 (12.61) | 1,870 (3.75) | 14,130 (13.19) | 7,809 (24.56) | <.0001 |
| COPD | 21,539 (11.41) | 96 (.19) | 7,912 (7.39) | 13,531 (42.56) | <.0001 |
| Depression | 17,453 (9.24) | 1,102 (2.21) | 10,219 (9.54) | 6,132 (19.29) | <.0001 |
| CHF | 16,842 (8.92) | 94 (.19) | 5,514 (5.15) | 11,234 (35.33) | <.0001 |
| Cardiac Arrhythmias | 10,625 (5.63) | 435 (.87) | 5,756 (5.27) | 4,434 (13.95) | <.0001 |
| Osteoporosis | 7,743 (4.10) | 710 (1.42) | 4,729 (4.41) | 2,304 (7.25) | <.0001 |
| Dementia | 4,025 (2.13) | 160 (.32) | 1,999 (1.87) | 1,866 (5.87) | <.0001 |
| Stroke | 274 (.15) | 8 (.02) | 109 (.10) | 157 (.49) | <.0001 |
| 30-day Readmission | 14,704 (7.79) | 2,535 (5.08) | 7,876 (7.35) | 4,293 (13.50) | <.0001 |
Note. DR reflects the proportion of nurses in the hospital who are confident their patient can manage care after discharge; range (0–1).
Multivariable Regression Modelling
Effects of Discharge Readiness on 30-day Readmissions.
Table 3 displays the results of the multilevel, logistic regression models assessing the association between hospital discharge readiness (operationalized as the percentage of nurses in each hospital who were confident in discharge readiness) and readmissions, while accounting for the number of MCCs a patient had. In the unadjusted Model 1, when compared to patients with 0–1 MCCs, the odds of 30-day readmissions were higher for patients with either 2–4 MCCs by 1.48 (OR=1.48, 95% CI [1.40, 1.56]) and 2.92 (OR=2.92, 95% CI [2.73, 3.12]) for patients with ≥5 MCCs. In the unadjusted Model 1, every 10% increase in the percentage of nurses confident in discharge readiness at the hospital level was associated with a 3% decrease in the odds of a patient being readmitted (OR = 0.97, 95% CI [0.95, 1.00]; p=0.028).
Table 3.
Sequential Modeling Indicating the Effects of Discharge Readiness and MCC on 30-day Readmissions, Older adult surgical patients
| Patient Outcomes | Unadjusted (Model 1) |
Adjusted for patient and hospital characteristics (Model 2) |
Fully adjusted with interaction of DR and MCC (Model 3) |
|||
|---|---|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| 30-day Readmission | ||||||
| Discharge Readiness | .97 (.95, 1.00) | .028 | .98 (.96, 1.00) | .053 | 1.02 (.99, 1.05) | .214 |
| MCC | ||||||
| 2–4 CCs | 1.48 (1.40, 1.56) | <.0001 | 1.07 (1.01, 1.13) | .015 | 1.32 (1.12, 1.57) | .001 |
| ≥5 CCs | 2.92 (2.73, 3.12) | <.0001 | 1.21 (1.12, 1.32) | <.0001 | 1.55 (1.27, 1.89) | <.0001 |
| Interaction | .015 | |||||
| Discharge Readiness*2–4 CCs | .96 (.93, .99) | .008 | ||||
| Discharge Readiness*≥5 CCs | .95 (.92, .99) | .007 | ||||
Results from Model 2 are fully adjusted, accounting for patient and hospital characteristics. The addition of these covariates made little impact on the model as we continued to find consistent associations between MCCs and readmissions, and modest, but significant associations between discharge readiness and post-discharge readmissions. We found that for every additional 10% increase in discharge readiness at the hospital level, the odds of a patient being readmitted decreased by 2% (OR= .98, 95% CI [0.96, 1.00]; p =0.046). When compared to patients with 0–1 MCCs, the odds of 30-day readmissions remained higher for patients with either 2–4 MCCs by 1.21 (OR=1.21, 95% CI [1.15, 1.28]; p=<0.0001) and 1.77 (OR=1.77, 95% CI [1.67, 1.89]; p=<0.0001) for patients with ≥5 MCCs.
Even when we adjust for all comorbidities, the results are similar. For example, for every 10% increase in discharge readiness, the odds of a patient being readmitted decreased by 2% (OR=0.98, 95% CI [0.96, 1.00]; p=0.053). When compared to patients with 0–1 MCCs, the odds of 30-day readmissions remained higher for patients with either 2–4 MCCs by 1.07 (OR=1.07, 95% CI [1.01, 1.13]; p=0.015) and 1.21 (OR=1.21, 95% CI [1.12, 1.32]; p=<0.0001) for patients with ≥5 MCCs.
Model 3 is similar to Model 2 but differs in that we include an interaction of the MCC and discharge readiness variables. The significant interaction term (p-value=0.012; chi-square=8.78) suggests that the effect of discharge readiness differed for patients at various levels of MCCs. When compared to patients with 0–1 MCCs, for every 10% increase in discharge readiness, the odds of 30-day readmissions decreased for patients with either 2–4 MCCs by 4% (OR=0.96, 95% CI [0.93, 0.99]; p=0.010) and 5% (OR=0.95, 95% CI [0.92, 0.98]; p=0.005) for patients with ≥5 MCCs.
The post-hoc, stratified analysis (Table 4) used to further determine the interaction effects of discharge readiness on 30-day readmissions across the levels of MCCs showed that for each 10% increase in discharge readiness, the odds of 30-day readmission decreased modestly by 2% (OR=0.98, 95% CI [0.96, 1.00]; p=0.028) for patients with 2–4 MCCs and 3% (OR=0.97, 95% CI [0.94, 0.99]; p=0.015) for patients with ≥5 MCCs, while discharge readiness had no statistically significant effect on readmissions for patients with 0–1 MCCs (OR=1.02; 95% CI [0.99, 1.05]; p-value=0.231). Figure 1 shows the predicted probability of 30-day readmissions across levels of discharge readiness and stratified by MCCs. In line with our results detailed in Table 4, the greatest, statistically significant decrease in predicted probability of readmission is seen among patients with ≥5 MCCs as discharge readiness increases, followed by patients with 2–4 MCCs. However, there is a small but statistically insignificant increase in predicted probability of readmission among patients with 0–1 MCCs as discharge readiness increases.
Table 4.
Adjusted Effects of Discharge Readiness Stratified by MCCs
| 0–1 MCCs | 2–4 MCCs | 5+ MCCs | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| 30-day Readmissions | ||||||
| Discharge Readiness | 1.02 (.99, 1.05) | .291 | .98 (.96, 1.00) | .029 | .97 (.94, 1.00) | .021 |
Notes: Model is adjusted for patient (i.e., age, sex, MCC count, and MS-DRG) and hospital (i.e., number of hospital beds, teaching status, and technology status) characteristics.
Figure 1.
Predicted Probability of 30-Day Readmissions Across Discharge Readiness, Stratified by MCCs
DISCUSSION
Using a sample of 424 US hospitals, we found that nurse assessments of discharge readiness corresponded with 30-day readmission rates for older adults living with MCCs. In alignment with current evidence,2,31 we also found that 30-day readmissions rose as the number of chronic conditions increased among older adults, and when compared to older adults with 0–1 MCCs, nurses assessments of discharge readiness became increasingly important in decreasing readmissions as the number of chronic conditions increased. In light of value-based payment models, evidenced-based strategies to decrease readmissions are paramount, particularly among at-risk patient populations like older adults with MCCs.32,33
Yet, few strategies leverage nursing or recognize nursing as a resource for identifying patients at risk for readmission despite nurses being the only clinician at the hospital bedside 24/7. Robust evidence suggests that nurses’ assessments of care quality and safety translate into objective patient outcomes like mortality and failure to rescue.11,12 Consistent with these findings, we show that nurses’ assessments of discharge readiness are associated with 30-day readmissions, highlighting a critical intervention point for hospital administrators, payers, and policy makers looking to address post-acute care service utilization for a patient population that accounts for a large majority of readmissions.1–3
We also show that many hospitals have room for improvement in terms of discharge readiness. For example, in the average hospital, less than half of nurses (45%) were confident that their patients or caregiver could manage their care after discharge. Further, there was substantial variation in discharge readiness across hospitals with some hospitals having only 23% of nurses expressing confidence in discharge readiness while other hospitals had as much as 65% of nurses expressing confidence, indicating that this is an area that hospitals can target for future organizational intervention. Given that the largest effect for discharge readiness was observed among older adults with 5 or more MCCs, it may be that many of the reasons for readmission are outside nurses’ control and more related to disease severity or social health needs that nurses cannot sufficiently address in their work setting. This could be why discharges occur despite nurses’ lack of confidence in their patients’ readiness – they feel there is nothing more they can do.
Our findings are consistent with a larger literature linking nursing resources (nurse education level, nurse staffing levels, nursing work environments) with patient outcomes, showing that investments in nursing are a tangible way to improve patient care.34–36 Of importance to note, these nursing resources, along with nurse assessments of discharge readiness are modifiable and subject to organizational intervention, unlike many structural hospital characteristics that cannot be changed. For example, we show that although discharge readiness varied across hospitals, it did not systematically vary across hospital characteristics (i.e., number of hospital beds, teaching status, technology status). Thus, improving discharge readiness may represent an obtainable goal for hospitals regardless of size, teaching, or technology status.
To improve discharge readiness and reduce readmissions, hospitals incorporate varying strategies (e.g., case managers, transitional care programs, standardized discharge teaching) that are customizable to the structure of their organization,13 but few interventions are designed to deliberately address multimorbidity.37 Calls for such interventions and improvements in care have been added to research priorities of several organizations including the Veterans Affairs, the National Institute of Health, and the AHRQ.38–40 It is understood that any improvements to hospital discharge processes need to incorporate person/caregiver-centric models and holistic, evidenced-based interventions (e.g., AHRQ’s IDEAL Discharge Planning strategy).
To optimize patient outcomes post-discharge, researchers have called for the addition of structured discharge readiness assessments, incorporating both patient and nurse perspectives, to assist the medical team to more effectively tailor discharge plans to meet the needs of patients prior to discharge.13 Although patient perspectives are a valuable indicator of quality, the findings regarding their assessments of discharge readiness and how it relates to actual readmissions is mixed.4,13,14 Not only are nurses a convenient and accessible source of information, their holistic training and perspective coupled with their around-the-clock presence at the bedside can help identify patients at high risk for readmissions. Using nurses as informants of discharge readiness may be especially helpful when caring for older adults living with MCCs.
Strengths and Limitations
This study used detailed data from a large sample of nurses working across 424 hospitals to build upon current evidence and evaluate whether variation across hospitals in nurses’ assessments of discharge readiness are associated with readmissions among older adults living with MCCs. To do this, we relied on a single item discharge readiness measure, which offers high utility as it is not burdensome to the busy clinician but can also be easily deployed in the electronic health record to flag patients for another evaluation before discharging home. However, there are some limitations to our study as the one-item measure does not allow us to determine what factors influenced nurses’ assessments of discharge readiness (e.g., patient-caregiver dynamics, diagnosed chronic conditions, caregiver ability, social determinants of health, clinical needs, care coordination, access to care, etc.), which would be helpful in improving the hospital discharge process. Future work determining patient discharge readiness would benefit from inclusion of these factors, which nurses are ideally positioned to report on in addition to their overall assessment of discharge readiness. We also recognize the limitations of examining MCCs by count rather than cluster (i.e., specific combinations of conditions). Incorporating MCC clusters would enhance our understanding of the impact of MCCs, and the effect that discharge readiness has on reducing readmissions across various MCC clusters as improving the discharge process will require targeted interventions (e.g., discharge teaching) specific to MCC clusters, not count. Lastly, the cross-sectional study design prevents causal inferences about the relationship between nurses’ assessments of discharge readiness and readmissions.
Conclusion
Our findings suggest that nurses are valuable informants of discharge readiness among older adults living with MCCs — a population that is increasingly complex and interdependent. Nurses are intricately involved in all aspects of a patient’s acute care stay, which uniquely positions them to the needs — clinical and social — of patients. Considering the constant pursuit to decrease readmissions, hospitals would be remiss to not leverage the informed assessments of nurses to identify older, multimorbid adults that are at high risk for readmission.
Supplementary Material
Acknowledgments:
We thank the Penn T32 staff for their assistance in preparation for and during the submission process of this manuscript.
Funding:
This study was supported by the National Institute of Nursing Research (NINR) (R01NR014855, Aiken and T32NR007104, Aiken). Schlak was a postdoctoral research fellow supported by NINR Comparative and Cost-Effectiveness Research Training for Nurse Scientists (T32NR014205, CO-PIs Poghosyan, Stone) at Columbia University School of Nursing during the time of writing.
Footnotes
Conflicts of Interest: The authors declare no conflict of interest. This study does not represent the views of the federal government or the Department of Veterans Affairs.
Contributor Information
Colleen A. Pogue, University of Pennsylvania School of Nursing, Center for Health Outcomes and Policy Research, Philadelphia, PA, USA;.
Amelia E. Schlak, Department of Veteran Affairs. Address: 810 Vermont Avenue, NW, Washington, DC, 20420.
Matthew D. McHugh, University of Pennsylvania School of Nursing, Philadelphia, PA, USA, Address: 418 Curie Blvd., Philadelphia, PA 19104.
References
- 1.Buttorff C, Ruder T, Bauman M. Multiple Chronic Conditions in the United States. RAND Corporation; 2017. doi: 10.7249/TL221 [DOI] [Google Scholar]
- 2.Aubert CE, Schnipper JL, Fankhauser N, et al. Patterns of multimorbidity associated with 30-day readmission: a multinational study. BMC Public Health. 2019;19(1):738. doi: 10.1186/s12889-019-7066-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Newman D, Tong M, Levine E, Kishore S. Prevalence of multiple chronic conditions by U.S. state and territory, 2017. PLoS One. 2020;15(5):e0232346. doi: 10.1371/journal.pone.0232346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Weiss M, Yakusheva O, Bobay K. Nurse and patient perceptions of discharge readiness in relation to postdischarge utilization. Med Care. 2010;48(5):482–486. doi: 10.1097/MLR.0b013e3181d5feae [DOI] [PubMed] [Google Scholar]
- 5.Provencher V, Clemson L, Wales K, et al. Supporting at-risk older adults transitioning from hospital to home: who benefits from an evidence-based patient-centered discharge planning intervention? Post-hoc analysis from a randomized trial. BMC Geriatrics. 2020;20(1):84. doi: 10.1186/s12877-020-1494-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McMartin K Discharge Planning in Chronic Conditions. Ont Health Technol Assess Ser. 2013;13(4):1–72. [PMC free article] [PubMed] [Google Scholar]
- 7.Weiss ME, Piacentine LB, Lokken L, et al. Perceived Readiness for Hospital Discharge in Adult Medical-Surgical Patients. Clinical Nurse Specialist. 2007;21(1):31–42. [DOI] [PubMed] [Google Scholar]
- 8.Luther B, Wilson RD, Kranz C, Krahulec M. Discharge Processes: What Evidence Tells Us Is Most Effective. Orthop Nurs. 2019;38(5):328–333. doi: 10.1097/NOR.0000000000000601 [DOI] [PubMed] [Google Scholar]
- 9.Brooks Carthon JM, Hedgeland T, Brom H, Hounshell D, Cacchione PZ. “You only have time for so much in 12 hours” unmet social needs of hospitalised patients: A qualitative study of acute care nurses. J Clin Nurs. 2019;28(19–20):3529–3537. doi: 10.1111/jocn.14944 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Phillips J, Richard A, Mayer KM, Shilkaitis M, Fogg LF, Vondracek H. Integrating the Social Determinants of Health into Nursing Practice: Nurses’ Perspectives. Journal of Nursing Scholarship. 2020;52(5):497–505. doi: 10.1111/jnu.12584 [DOI] [PubMed] [Google Scholar]
- 11.McHugh MD, Stimpfel AW. Nurse reported quality of care: A measure of hospital quality. Research in Nursing & Health. 2012;35(6):566–575. doi: 10.1002/nur.21503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Smeds-Alenius L, Tishelman C, Lindqvist R, Runesdotter S, McHugh MD. RN assessments of excellent quality of care and patient safety are associated with significantly lower odds of 30-day inpatient mortality: A national cross-sectional study of acute-care hospitals. International Journal of Nursing Studies. 2016;61:117–124. doi: 10.1016/j.ijnurstu.2016.06.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Weiss ME, Yakusheva O, Bobay KL, et al. Effect of implementing discharge readiness assessment in adult medical-surgical units on 30-day return to hospital: the READI randomized clinical trial. JAMA network open. 2019;2(1):e187387–e187387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Weiss ME, Costa LL, Yakusheva O, Bobay KL. Validation of Patient and Nurse Short Forms of the Readiness for Hospital Discharge Scale and Their Relationship to Return to the Hospital. Health Serv Res. 2014;49(1):304–317. doi: 10.1111/1475-6773.12092 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lasater KB, Jarrín OF, Aiken LH, McHugh MD, Sloane DM, Smith HL. A Methodology For Studying Organizational Performance. Med Care. 2019;57(9):742–749. doi: 10.1097/MLR.0000000000001167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Aiken LH, Sloane DM, Bruyneel L, et al. Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study. The Lancet. 2014;383(9931):1824–1830. doi: 10.1016/S0140-6736(13)62631-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ball JE, Bruyneel L, Aiken LH, et al. Post-operative mortality, missed care and nurse staffing in nine countries: A cross-sectional study. Int J Nurs Stud. 2018;78:10–15. doi: 10.1016/j.ijnurstu.2017.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schlak AE, Aiken LH, Chittams J, Poghosyan L, McHugh M. Leveraging the Work Environment to Minimize the Negative Impact of Nurse Burnout on Patient Outcomes. International Journal of Environmental Research and Public Health. 2021;18(2):610. doi: 10.3390/ijerph18020610 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McHugh M, Ma C. Hospital nursing and 30-day readmissions among Medicare patients with heart failure, acute myocardial infarction, and pneumonia. J Nurs Adm. 2013;43(10 Suppl):S11–18. doi: 10.1097/01.NNA.0000435146.46961.d1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Goodman RA, Posner SF, Huang ES, Parekh AK, Koh HK. Defining and measuring chronic conditions: imperatives for research, policy, program, and practice. Prev Chronic Dis. 2013;10:E66. doi: 10.5888/pcd10.120239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lindenauer PK, Bernheim SM, Grady JN, et al. The performance of US hospitals as reflected in risk-standardized 30-day mortality and readmission rates for medicare beneficiaries with pneumonia. J Hosp Med. 2010;5(6):E12–18. doi: 10.1002/jhm.822 [DOI] [PubMed] [Google Scholar]
- 22.Keenan PS, Normand SLT, Lin Z, et al. An administrative claims measure suitable for profiling hospital performance on the basis of 30-day all-cause readmission rates among patients with heart failure. Circ Cardiovasc Qual Outcomes. 2008;1(1):29–37. doi: 10.1161/CIRCOUTCOMES.108.802686 [DOI] [PubMed] [Google Scholar]
- 23.Krumholz HM, Merrill AR, Schone EM, et al. Patterns of hospital performance in acute myocardial infarction and heart failure 30-day mortality and readmission. Circ Cardiovasc Qual Outcomes. 2009;2(5):407–413. doi: 10.1161/CIRCOUTCOMES.109.883256 [DOI] [PubMed] [Google Scholar]
- 24.Ross JS, Mulvey GK, Stauffer B, et al. Statistical models and patient predictors of readmission for heart failure: a systematic review. Arch Intern Med. 2008;168(13):1371–1386. doi: 10.1001/archinte.168.13.1371 [DOI] [PubMed] [Google Scholar]
- 25.Schlak AE, Aiken LH, Chittams J, Poghosyan L, McHugh M. Leveraging the Work Environment to Minimize the Negative Impact of Nurse Burnout on Patient Outcomes. International Journal of Environmental Research and Public Health. 2021;18(2):610. doi: 10.3390/ijerph18020610 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Aiken LH, Clarke SP, Cheung RB, Sloane DM, Silber JH. Educational levels of hospital nurses and surgical patient mortality. JAMA. 2003;290(12):1617–1623. doi: 10.1001/jama.290.12.1617 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Silber JH, Williams SV, Krakauer H, Schwartz JS. Hospital and patient characteristics associated with death after surgery. A study of adverse occurrence and failure to rescue. Med Care. 1992;30(7):615–629. doi: 10.1097/00005650-199207000-00004 [DOI] [PubMed] [Google Scholar]
- 28.Huber PJ. The behavior of maximum likelihood estimates under nonstandard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics. 1967;5.1:221–234. [Google Scholar]
- 29.Stata Statistical Software: Release 17. StataCorp [Google Scholar]
- 30.Center for Disease Control and Prevention. Chronic Diseases in America | CDC. Published 2022. Accessed April 21, 2022. https://www.cdc.gov/chronicdisease/resources/infographic/chronic-diseases.htm
- 31.Marengoni A, Angleman S, Melis R, et al. Aging with multimorbidity: A systematic review of the literature. Ageing Research Reviews. 2011;10(4):430–439. doi: 10.1016/j.arr.2011.03.003 [DOI] [PubMed] [Google Scholar]
- 32.Lowe G, Dohrmann J, Chapman J, Stamford L, Driscoll A. Older Persons Complex Care Nurse Practitioner model to reduce hospital readmissions: A pilot study. Collegian. Published online July 12, 2022. doi: 10.1016/j.colegn.2022.06.013 [DOI] [Google Scholar]
- 33.Gerteis J, Izrael D, Deitz D, et al. Multiple Chronic Conditions Chartbook: 2010 Medical Expenditure Panel Survey Data. AHRQ Publications No, Q14–0038; 2010:52. [Google Scholar]
- 34.Lasater KB, Sloane DM, McHugh MD, Porat-Dahlerbruch J, Aiken LH. Changes in proportion of bachelor’s nurses associated with improvements in patient outcomes. Research in Nursing & Health. 2021;44(5):787–795. doi: 10.1002/nur.22163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Sloane DM, Smith HL, McHugh MD, Aiken LH. Effect of changes in hospital nursing resources on improvements in patient safety and quality of care: A panel study. Med Care. 2018;56(12):1001–1008. doi: 10.1097/MLR.0000000000001002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lasater KB, McHugh M, Rosenbaum PR, et al. Valuing hospital investments in nursing: multistate matched-cohort study of surgical patients. BMJ Qual Saf. 2021;30(1):46–55. doi: 10.1136/bmjqs-2019-010534 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kastner M, Hayden L, Wong G, et al. Underlying mechanisms of complex interventions addressing the care of older adults with multimorbidity: a realist review. BMJ Open. 2019;9(4):e025009. doi: 10.1136/bmjopen-2018-025009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bierman AS, Wang J, O’Malley PG, Moss DK. Transforming care for people with multiple chronic conditions: Agency for Healthcare Research and Quality’s research agenda. Health Services Research. 2021;56(S1):973–979. doi: 10.1111/1475-6773.13863 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.National Institute of Nursing Research. National Institute of Nursing Research Strategic Plan Working Group Draft Framework for 2022–2026. Published 2022. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/ https://www.ninr.nih.gov/sites/files/docs/strategic_planning_working_group_framework_report_051021_v1_508c.pdf
- 40.VA Health Services Research & Development Science. Health Services Research and Development Research Priorities. Published 2021. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/ https://www.hsrd.research.va.gov/funding/PriorityDomains2019.pdf
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.