Readmission and Mortality in Trauma Patients Utilizing LACE Index

Abstract

Background

Trauma patients are sometimes readmitted unexpectedly soon after discharge from index admission. LACE Index measures four variables (Length of stay, Acuity of admission, Comorbidity, and prior Emergency department visits within 6 months) that are independently associated with death or readmission within 30 days of discharge from hospitalization. Our Trauma Center began utilizing LACE scores to schedule follow-ups for patients discharged home with LACE scores ≥7. We sought to determine factors related to trauma patient readmission and mortality and evaluate the impact of LACE follow-up appointments on 30-day readmission and mortality. We hypothesized that instituting LACE follow-ups would reduce readmissions and mortality.

Methods

We performed a retrospective review (1/1/21-6/30/22 and 7/1/22-12/31/23) evaluating readmission in trauma patients ≥18y before LACE (BL) and during LACE (DL) follow-up appointments. Univariate and multivariable analyses were conducted. A P-value <.05 was considered significant.

Results

1788 patients were included: 872 BL and 914 DL. Readmission was slightly higher DL vs BL and mortality was lower DL vs BL. Multivariable analysis showed patients with cirrhosis had an increased risk of readmission and patients with commercial insurance had a decreased risk of readmission. Multivariable analysis showed attending a follow-up did not impact risk of readmission but decreased risk of mortality.

Conclusion

Our data partially supported our hypothesis; attending an LACE follow-up decreased 30-day mortality but not readmission. Further investigation is warranted to validate the threshold of LACE score that will help prevent readmission and mortality in trauma patients.

Keywords

LACE trauma patient trauma readmission trauma follow-up trauma mortality

Key Takeaways

• 30-day readmission did not decrease with the implementation of LACE follow-up appointments in trauma patients

• 30-day mortality decreased with implementation of LACE follow-up appointments in trauma patients

• Current LACE score thresholds may not hold predictive value for readmission rates with trauma patients.

Introduction

Thousands of patients are admitted for traumatic injuries across the country every day. After receiving life-saving care, these patients are discharged from the hospital to continue their recovery. A portion of these patients are unexpectedly readmitted.¹ Readmission is defined as a nonelective return to an acute care hospital for a variety of reasons. Trauma patients are sometimes readmitted unexpectedly soon after discharge. It is estimated that 3% to 8% of patients will be readmitted within 30 days of discharge and about 10.4% will be readmitted within 90 days.^1-3 Unexpected readmission can result in worse outcomes and is costly to patients with the average cost of a readmission being $16,037.08.⁴ Readmissions are not only costly to patients but also to health systems through utilization of resources as readmitted patients are more likely to have a longer stay in the hospital than their first stay.³ In addition, it is estimated that one in five trauma readmissions are potentially preventable which account for over $300 million in health care costs annually.⁵ It is important to better understand what contributes to unexpected trauma readmission and how to predict it, in order to most effectively prevent readmission and improve patient outcomes.

There are many factors that can contribute to a patient being readmitted unexpectedly. At the provider level, evaluating high- and low-risk patients and their chance of readmission could be used to determine which patients should receive early intervention techniques or appropriate higher-level placement to decrease the chance of readmission. The LACE Index was developed in 2010 to quantify the risk of 30-day readmission and mortality. The LACE Index measures the following four variables: Length of stay, Acuity of admission, Comorbidity, and prior Emergency department visits within 6 months. These variables are independently associated with death or readmission within 30 days of discharge from hospitalization and hold predictive value for medical patients.⁶ The LACE Index has been studied and implemented with medical patients but has not been assessed for trauma surgical patients. As such, there is currently no predictive assessment to ascertain a trauma patient’s relative risk of 30-day readmission or mortality.

In July 2022, our Level 1 Trauma Center began utilizing LACE scores to schedule follow-up appointments for patients discharged home with LACE scores ≥7 in an effort to determine if the LACE Index can be applied to the trauma population. We chose a score of ≥7 as this has been previously validated in the medical population. Our goal was to schedule follow-up appointments within one week of discharge in order to evaluate the progress of recently discharged patients and prevent readmission by providing post-discharge resources as necessary. In this study, we sought to determine factors related to trauma patient readmission and mortality and evaluate the impact of LACE follow-up appointments on 30-day readmission and mortality. We hypothesized that instituting LACE follow-up appointments a week after discharge would decrease the rate of 30-day readmission and mortality.

Methods

After receiving approval from our institutional review board, we conducted a single-center retrospective study of trauma patients ≥18 years old who presented to our Level 1 trauma center between January 1, 2021, and December 31, 2023. Patients were grouped into two categories based on when they were admitted relative to the start of LACE follow-up appointments: Before LACE (BL: 1/1/21-6/30/22) and During Lace (DL: 7/1/22-12/31/23). To determine the attendance rates and efficacy of LACE follow-ups, patients were included if they were discharged to home with a LACE score ≥7. Patients were excluded if they were under 18 years of age, were not admitted during their initial presentation to our hospital, were transferred in or out during their index admission, died during their hospital stay, or were admitted to another service as their disposition was not controlled by our service. We also compared those DL who did not attend a follow-up appointment compared to those who did attend. A sub-group analysis was performed on all patients readmitted to the hospital within 30 days of discharge from their index admission.

We performed our statistical analysis using Stata/SE version 17.0. We conducted a Shapiro-Francia test to evaluate distribution normality. We then ran t-tests for normally distributed variables. Data were reported as mean (standard deviation), P-value. We performed Mann-Whitney U tests for continuous variables that were not normally distributed. Data were reported as median (IQR), P-value. We also conducted Chi-squared tests for categorical variables and reported these data as frequency (percentage of population), P-value. These univariate analyses were run to compare risk factors and outcomes between our BL time period and our DL time period. Further, we compared patients who were readmitted to those who were not readmitted and patients who died to patients who did not die during the DL period.

Multivariable logistic regressions were conducted to assess risk factors predictive of readmission and mortality. A P-value <0.05 was deemed statistically significant for all univariate and multivariable analyses. An analysis was conducted to determine the predictive value of LACE scores at a threshold of 7 for the trauma population. This analysis assessed for sensitivity, specificity, and positive predictive value of readmission.

Results

As displayed in Table 1, a total of 1786 patients met the inclusion criteria for this study with 872 before utilizing LACE scores (Before LACE-BL) and 914 during the implementation of LACE scores and follow-up appointments (During LACE-DL). Interestingly, Table 1 shows no significant association between readmission rates and LACE period: DL (8.52%) vs BL (7.11%, P = 0.269), while mortality was significantly associated with LACE period. Mortality was lower DL (0.66%) compared to BL (1.84%, P = .024).

Table 1.

Univariate Analysis Assessing Association With Implementation of LACE Follow-Up Appointments

	Before LACE (n = 872)	During LACE (n = 914)	P-value
Male^a	497 (57.00)	562 (61.35)	0.061
White^a	775 (88.88)	816 (89.08)	0.695
Hispanic^a	94 (10.84)	89 (9.78)	0.462
Age^b	66 (51-77)	66 (49-77)	0.532
CHF^a	99 (11.35)	98 (10.70)	0.659
MI^a	2 (0.23)	2 (0.22)	0.961
COPD^a	97 (11.12)	88 (9.61)	0.293
CVA^a	26 (2.98)	12 (1.31)	0.014
Hypertension^a	496 (56.88)	520 (56.77)	0.962
Dementia^a	40 (4.59)	36 (3.93)	0.491
Diabetes^a	226 (25.92)	229 (25.00)	0.656
Smoker^a	176 (20.18)	173 (18.89)	0.489
Thyroid disease^a	126 (14.45)	118 (12.88)	0.335
CAD^a	160 (18.35)	165 (18.01)	0.854
Dialysis^a	7 (0.80)	2 (0.22)	0.081
Obesity^a	312 (35.78)	295 (32.21)	0.111
Aspirin use^a	195 (22.36)	207 (22.60)	0.905
Creatinine >2 mg/dL^a	5 (0.57)	4 (0.44)	0.683
PAD^a	20 (2.29)	39 (4.26)	0.020
Cirrhosis^a	19 (2.18)	15 (1.64)	0.402
ISS^b	9 (5-13)	9 (5-12)	0.546
AIS head^c	2.30 (1.22)	2.31 (1.22)	0.945
AIS face^b	2 (1-2)	2 (1-2)	0.660
AIS chest^b	3 (2-3)	3 (2-3)	0.810
AIS abdomen/Pelvis^b	2 (2-3)	2 (2-2)	0.147
AIS extremities^c	2.12 (0.57)	2.09 (0.55)	0.519
AIS external^b	1 (1-1)	1 (1-1)	0.210
ED disposition: Med surg^a	484 (55.50)	520 (56.77)	0.350
ICU LOS^b	0 (0-1)	0 (0-1)	0.082
Hospital LOS^b	3 (2-6)	4 (2-6)	0.144
LACE score^b	9 (8-11)	9 (8-11)	0.401
Readmission^a	62 (7.11)	78 (8.52)	0.269
Mortality^a	16 (1.84)	6 (0.66)	0.024

Abbreviations: CHF, Congestive Heart Failure; MI, Myocardial Infarction; COPD, Chronic Obstructive Pulmonary Disease; CVA, Cerebral Vascular Accident; CAD, Coronary Artery Disease; PAD, Peripheral Artery Disease; ISS, Injury Severity Score; AIS, Abbreviated Injury Score; ED, Emergency Department; Med Surg, Medical Surgical; ICU, Intensive Care Unit; LOS, Length of Stay.

^an(% of population).

^bmedian (IQR).

^cmean (standard deviation).

Of these 914 patients who were discharged home with a LACE score ≥7 DL, we found that 746 (81.44%) had a follow-up appointment scheduled and 491 (65.91%) attended their appointment. Of those 491 patients who attended their follow-up appointment, we found that 7.54% (n = 37) were readmitted within 30 days compared to the 9.65% (n = 41) readmitted of those who did not attend their follow-up appointment or had no follow-up scheduled (P = .881). In those with an appointment made, there was a decrease in 30-day mortality between those who attended their appointment (0.2%) and those who did not (1.96%, P = .011). Table 2 displays patient characteristics and outcomes in those who did not attend their follow-up appointment (n = 423) and those who did attend their follow-up appointment (n = 491). Table 2 shows that patients with coronary artery disease (CAD) were less likely to attend their follow-up appointment (P = .007), and those with commercial insurance were more likely to attend their follow-up appointment (P = .026).

Table 2.

Patient Characteristics and Outcomes Based on LACE Follow-Up Attendance

	LACE follow-up not attended (n = 423)	LACE follow-up attended (n = 491)	P-value
Age^a	67 (51-78)	65 (48-76)	0.051
Race: White^b	375 (88.65)	439 (89.41)	0.900
Sex: Male^b	255 (60.28)	306 (62.32)	0.528
CAD^b	92 (21.75)	73 (14.87)	0.007
PAD^b	19 (4.49)	20 (4.07)	0.755
Diabetes^b	116 (27.42)	113 (23.01)	0.125
CHF^b	48 (11.35)	50 (10.18)	0.571
Dementia^b	17 (4.02)	19 (3.87)	0.908
Current smoker^b	75 (17.73)	98 (19.96)	0.391
Obesity^b	133 (31.44)	162 (32.99)	0.617
Chronic aspirin use^b	100 (23.64)	107 (21.79)	0.506
Arthritis^b	124 (29.31)	142 (28.92)	0.896
COPD^b	48 (11.35)	40 (8.15)	0.102
Dialysis^b	2 (0.47)	0 (0)	0.127
Creatinine >2^b	3 (0.71)	1 (0.20)	0.248
Cirrhosis^b	10 (2.36)	5 (1.02)	0.110
Commercial insurance^b	43 (10.17)	59 (12.02)	0.026
LACE score^a	9 (8-11)	9 (8-11)	0.095
Readmission^b	41 (9.69)	37 (7.54)	0.244
Mortality^b	5 (1.18)	1 (0.20)	0.068

Abbreviations: CAD, Coronary Artery Disease; PAD, Peripheral Arterial Disease; CHF, Congestive Heart Failure; COPD, Chronic Obstructive Pulmonary Disease.

^amedian (IQR).

^bn (% of population).

Multivariable logistic regression analysis as seen in Table 3 revealed that patients with cirrhosis had an increased risk of 30-day readmission (AOR 4.669, 95% CI: 2.049-10.637, P < .001) while patients with commercial insurance had a decreased risk of 30-day readmission (AOR 0.769, 95% CI: 0.636-0.932, P = .007). Multivariable analysis also depicted that attending a follow-up appointment did not impact risk of readmission (AOR 0.671, 95% CI: 0.439-1.028, P = 0.066). When performing a multivariable analysis for mortality (Table 4), attending a follow-up appointment did significantly decrease the risk of mortality (AOR 0.124, 95% CI: 0.016-0.984, P = .048).

Table 3.

Multivariable Logistic Regression Assessing Risk of Readmission for Patients Discharged Home With an LACE Score of ≥7

Readmission
Variable	Unadjusted OR	Unadjusted P-value	Adjusted OR	95% Confidence interval	Adjusted P-value
Age	1.017	0.001	1.011	.998 - 1.024	0.114
Race: White	0.986	0.885	1.106	.910 - 1.344	0.311
Sex: Male	0.793	0.205	0.775	.521 - 1.152	0.207
ISS	1.018	0.177	1.024	.998 - 1.054	0.107
Lace score	1.232	0.001	1.113	1.018 - 1.218	0.019
During LACE follow-ups	1.216	0.27	1.609	1.082 - 2.393	0.019
ED disposition: ICU	1.017	0.057	1.024	1.002 - 1.047	0.035
Hospital LOS	1.023	0.011	1.012	.988 - 1.037	0.338
ICU LOS	1.062	0.019	1.034	.965 - 1.108	0.338
Follow-up attended	0.720	0.088	0.671	.439 - 1.028	0.066
CAD	1.696	0.009	1.333	.823 - 2.159	0.243
CVA	2.261	0.072	1.641	.593 - 4.543	0.340
PAD	1.093	0.851	0.668	.248 - 1.800	0.424
Diabetes	1.430	0.059	1.069	.700 - 1.633	0.757
CHF	2.077	0.001	1.013	.579 - 1.773	0.964
Cirrhosis	6.025	<0.001	4.669	2.050 - 10.637	<0.001
Dementia	1.844	0.082	1.222	.580 - 2.575	0.598
Current smoker	0.936	0.768	0.951	.573 - 1.576	0.844
Obesity	0.948	0.776	0.852	.569 - 1.277	0.439
Aspirin use	1.023	0.912	0.770	.489 - 1.213	0.260
Arthritis	1.461	0.038	1.261	.847 - 1.879	0.253
Dialysis	1.475	0.715	1.010	.116- 8.795	0.993
Thyroid disease	1.128	0.627	0.882	.514 - 1.514	0.650
COPD	1.498	0.113	0.933	.523 - 1.665	0.814
Commercial insurance	0.742	0.001	0.770	.636 - .932	0.007
					AUROC: 0.718

Abbreviations: ISS, Injury Severity Score; ED, Emergency Department; ICU, Intensive Care Unit; LOS, Length of Stay; CAD, Coronary Artery Disease; CVA, Cerebral Vascular Accident; PAD, Peripheral Artery Disease; CHF, Congestive Heart Failure; COPD, Chronic Obstructive Pulmonary Disease.

Table 4.

Multivariable Logistic Regression Assessing Risk of Mortality for Patients Discharged Home With an LACE Score of ≥7

Mortality
Variable	Unadjusted OR	Unadjusted P-value	Adjusted OR	95% Confidence interval	Adjusted P-value
Age	1.036	0.014	1.047	1.008 - 1.088	0.019
Race: White	0.932	0.782	1.101	.666 - 1.821	0.707
Gender: Male	0.542	0.203	0.465	.156 - 1.394	0.171
ISS	1.038	0.213	1.057	.989 - 1.131	0.104
LACE score	1.295	0.001	1.290	1.024 - 1.624	0.031
During lace follow-ups	0.354	0.031	0.498	.178 - 1.394	0.184
ED disposition: ICU	1.005	0.867	1.011	.957 - 1.067	0.707
Hospital LOS	1.008	0.775	0.981	.888 - 1.084	0.704
ICU LOS	1.024	0.744	0.994	.757 - 1.306	0.968
Follow-up attended	0.083	0.015	0.124	.016 - .984	0.048
CAD	3.201	0.008	1.849	.621 - 5.506	0.270
PAD	3.043	0.140	1.174	.200 - 6.892	0.859
Diabetes	1.379	0.486	0.709	.242 - 2.075	0.530
CHF	3.081	0.020	1.115	.326 - 3.818	0.862
Dementia	1.070	0.948	0.338	.036 - 3.175	0.343
Current smoker	1.220	0.699	1.484	.444 - 4.966	0.522
Obesity	0.723	0.501	0.694	.232 - 2.081	0.514
Aspirin use	0.542	0.326	0.343	.092 - 1.278	0.111
Arthritis	0.364	0.105	0.246	.068 - .896	0.034
COPD	2.582	0.065	1.375	.413 - 4.582	0.604
Commercial insurance	1.034	0.859	1.316	.832 - 2.083	0.241
					AUROC 0.865

An analysis of all trauma patients with LACE scores obtained (not just those discharged to home) was completed to determine the predictive value of LACE scores at a threshold of 7 for the trauma population. This showed a positive predictive value (the rate of readmissions for all patients with a score over 7) of 0.08 (86/1134) and a sensitivity (the rate of patients with a score over 7 for all patients with a readmission) of 0.77 (86/111). In comparison, a threshold of 6 had a positive predictive value of 0.07 (100/1412) and a sensitivity of 0.90 (100/111) and a threshold of 8 had a positive predictive value of 0.09 (82/880) and a sensitivity of 0.77 (86/111). Figure 1 illustrates the sensitivity and specificity of every LACE score threshold value. The cutoff of the LACE of 7 is highlighted.

Discussion

We hypothesized that instituting LACE follow-up appointments within a week of discharge to home for trauma patients with an LACE score ≥7 would decrease the rate of 30-day readmission and mortality. Our hypothesis was partially confirmed as we did not find a statistically significant difference in 30-day readmission between DL and BL; however, the rate of 30-day mortality did decrease DL compared to BL. We also found that patients with cirrhosis were more likely to be readmitted within 30 days and that patients with commercial insurance were less likely to be readmitted within 30 days.

Our finding that readmission rates were not significantly different between DL and BL may be explained by patients not having scheduled or not attending their follow-up appointments, since previous research suggests that attendance of post-discharge follow-up appointments does in fact reduce 30-day readmission rates in medical patients.^7,8 There are various factors that can impede a patient’s ability to attend their outpatient appointments. Factors such as language barriers and insufficient communication can result in patients not understanding the importance of the appointment or the details surrounding it.^9,10 Additional socioeconomic factors such as a lack of transportation and a lack of job security can prevent attendance due to not being able to travel safely or not being able to take off work to attend.^9-11 In addition, Coppa et al found that follow-up appointment attendance within one week of discharge was significantly associated with reducing 30-day readmissions.⁸ While our center schedules follow-up appointments within a week post-discharge, it would be worth investigating if scheduling the appointment sooner than that would impact 30-day readmission rates.

In contrast to what we discovered on readmission rates, we found on both univariate and multivariable analysis, that 30-day mortality was lower after implementing LACE follow-up appointments. The fact that while mortality decreased, readmission rate increased during the LACE follow-up implementation, may be explained by a readmission being necessary to prevent other untoward effects (such as mortality) and therefore mortality was prevented due to this. The LACE Index has demonstrated that patients with a score of ≥7 are at a higher risk of post-discharge mortality and previous studies evaluating specific patient populations suggest that early follow-up after hospital discharge reduces mortality rates, which is in line with our findings.^6,12

Our findings that patients with cirrhosis had an increased risk of 30-day readmission is supported by existing literature.^13,14 This is consistent with existing research which found the readmission rate for patients with cirrhosis to be 31.4% compared to a readmission rate of 11-14% for other medical conditions.¹³ Garg et al suggest elevating the post-discharge destination to a short-term rehabilitation or higher-level nursing facility could reduce readmission rates with this patient population; however, a 2023 study found that readmissions for patients with cirrhosis are not likely to be preventable due to the complex and progressive nature of the disease.^13,14 Similar to our findings, previous studies found that patients with cirrhosis on commercial insurance were least likely to be readmitted compared to those on Medicare and Medicaid.¹³

Additionally, previous research supports our finding that patients with commercial insurance have a decreased risk of 30-day readmission.¹⁵ Ferro et al. found that of the patients included in their study who were readmitted, 73.5% had Medicare, 9.1% had Medicaid, and 12.2% had private, or commercial, insurance.¹⁵ Another study found that patients on Medicaid are more likely to be readmitted within 30 days because Medicaid patients were less likely than privately insured patients to receive interventions ultimately leading to increased readmission rates for Medicaid-insured patients.¹⁶ These studies also suggest that because Medicare patients receive more intensive care and less preventative care than privately insured patients, their outcomes, including readmission rates, are worse as well.^15,16

This study is not without limitations. As a single-center, the magnitude of data is limited and this is further coupled with the fact that we only began utilizing LACE score-related follow-up appointments in July 2022, further limiting our sample size. We also did not collect the date and time of readmission and mortality for patients included in this study, and as a result, we cannot further analyze the potential for a survivorship bias in our findings. Another limitation of this study is the COVID-19 status of included patients. Since we did not collect COVID-19 infection status as a part of our data, we cannot account for this as a confounding factor to our findings. Additionally, this is a retrospective study, and information obtained was limited to the medical record. Our center is located in a rural/ suburban area which may mean that our data is not generalizable to other centers in different geographical locations.

As we seek to find a way to accurately predict and subsequently prevent rates of readmission and mortality in trauma patients, it is possible that the LACE score follow-up appointments (at a threshold of 7) on their own are not sufficient or need to be completed sooner. Lowering the threshold may allow for a higher sensitivity and specificity, however, it would increase the follow-up appointments necessary, therefore increasing outpatient resource utilization. While an LACE score of ≥7 may be predictive of readmission in medical patients, a lower score may be necessary to predict readmission in trauma patients. Evaluation of 90-day readmission and morality could give us an additional perspective on outcomes relative to LACE score and appointment attendance. Another factor to consider that may impact 30-day readmission and mortality is that of discharge destination. Further study will help us understand what reasons patients are being readmitted for, what providers can do inpatient and outpatient to mitigate the risk of readmission, and if the LACE index should be adjusted for the trauma population. These additional data points could help inform discharge decisions reveal how to effectively reduce barriers to appointment attendance, and could provide us with impactful means of improving outcomes for our trauma patients after their hospitalization.

Footnotes

ORCID iDs

Julie M. Reichert

Nathan A. Schlessinger

Lindsey L. Perea

Authors Contributions

JHB was involved with data collection and writing. JMR and ILF were involved with data collection, statistical analysis, and writing. MAR, ABT, NAS, KFB, and REW were involved with data collection. LLP was involved with generating the study idea, data collection, statistical analysis, and writing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

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

References

Olufajo

Cooper

Yorkgitis

, et al. The truth about trauma readmissions. Am J Surg. 2016;211(4):649-655.

Petry

Weddle

Richardson

, et al.

Trauma patient readmissions: why do they come back for more?

J Trauma Acute Care Surg. 2015;79(5):717-725.

Morris

Rohrbach

Rogers

, et al. The surgical revolving door: risk factors for hospital readmission. J Surg Res. 2011;170(2):297-301.

Ghabowen

Epane

Shen

Goodman

Ramamonjiarivelo

Zengul

. Systematic review and meta-analysis of the financial impact of 30-Day readmissions for selected medical conditions: a focus on Hospital quality performance. Healthcare (Basel). 2024;12(7):750.

Neiman

Brown

Montgomery

Sangji

Hemmila

Scott

. Targeting zero preventable trauma readmissions. J Trauma Acute Care Surg. 2021;91(4):728-735.

van Walraven

Dhalla

Bell

, et al. Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community. CMAJ. 2010;182(6):551-557.

Bilicki

Reeves

. Outpatient Follow-Up visits to reduce 30-Day all-cause readmissions for heart failure, COPD, myocardial infarction, and stroke: a systematic review and meta-analysis. Prev Chronic Dis. 2024;21:240138.

Coppa

Kim

Oppenheim

Bock

Conigliaro

Hirsch

. Examination of post-discharge Follow-up appointment status and 30-Day readmission. J Gen Intern Med. 2021;36(5):1214-1221.

Chapman

Machado

van der Merwe

Bryson

Smith

. Exploring primary care non-attendance: a study of low-income patients. JPC. 2022;13:1-11.

10.

Sung

O'Driscoll

Gregory

, et al. Identifying barriers to outpatient appointment attendance in patient groups at risk of inequity: a mixed methods study in a London NHS trust. BMC Health Serv Res. 2024;24(1):554.

11.

Alturbag

. Factors and reasons associated with appointment non-attendance in hospitals: a narrative review. Cureus. 2024;16(4):e58594.

12.

Pederson

Gregersen

Barat

Damsgaard

. Early geriatric follow-up after discharge reduces mortality among patients living in their own home. A randomised controlled trial. Eur Geriatr Med. 2017;8(4):330-336.

13.

Garg

Goyal

Obaitan

, et al. Incidence and predictors of 30-day hospital readmissions for liver cirrhosis: insights from the United States national readmissions database. Ann Transl Med. 2021;9(13):1052.

14.

Orman

Desai

Ghabril

, et al. Thirty-day readmissions are largely not preventable in patients with cirrhosis. Am J Gastroenterol. 2024;119(2):287-296.

15.

Ferro

Secemsky

Wadhera

, et al. Patient readmission rates for all insurance types after implementation of the hospital readmissions reduction program. Health Aff (Millwood). 2019;38(4):585-593.

16.

Niedzwiecki

Hsia

Shen

. Not all insurance is equal: differential treatment and health outcomes by insurance coverage among nonelderly adult patients with heart attack. J Am Heart Assoc. 2018;7(11):e008152.