Unplanned hospital admissions within 24 h after 53,185 surgical procedures at a U.S. ambulatory surgery center

Abstract

Background

Unplanned admission after surgery at an ambulatory surgery center (ASC) is an established measure of the quality of care and can affect the patient’s experience. Previous studies on this topic are generally dated, focused on a single specialty, or studied 30-day admissions after ambulatory surgery. Few studies have reported admission within 24 h after surgery at an ASC which is a different but important measure of the quality of anesthetic and surgical care. Understanding admissions within 24 h of surgery can identify opportunities for improvement immediately after surgery. Therefore, our study was designed to assess the incidence and risk factors for unplanned hospital admissions within 24 h after surgery performed at a hospital ASC.

Methods

After Institutional Review Board approval, a retrospective analysis was performed on all adult patients who underwent surgery at a US ASC between January 1, 2016, and December 31, 2022. Data were obtained from the hospital’s electronic medical record. The study sample was divided into two groups: those with an unplanned hospital admission within 24 h after surgery and those without an unplanned hospital admission. To evaluate risk factors for unplanned hospital admissions, univariate analyses with p value < 0.05 were utilized to identify significant patient variables related to hospital admissions. These variables were further adjusted using a multivariable Firth logistic regression. Descriptive statistics were used to explore the number of patients in different variable categories.

Results

Overall, 53,185 cases were identified for the 7-year period. The incidence of unplanned hospital admission over this period was 0.09% (95% CI 0.07–0.1122%; ranging from 0.05 to 0.12% per year. In the multivariable model, surgery duration (OR 1.010, 95% CI 1.007–1.012, p value < 0.0001), peripheral vascular disease (OR 14.489, 95% CI 4.862–43.174, p value < 0.0001), and deep venous thrombosis (OR 5.527, 95% CI 1.909–16.001, p value = 0.0016) were significantly associated with unplanned hospital admission.

Conclusion

The overall incidence of unplanned hospital admission after surgery at a large tertiary care ambulatory surgery center is very low. This admission rate can also serve as a reference point for future studies and quality improvement initiatives.

Introduction

The future of healthcare will be largely shaped by ambulatory surgery centers (ASCs) (Davis 1987). More than 35 million procedures are performed annually in ambulatory settings in the USA (Suskind et al. 2015; Hall et al. 2017). ASCs have played an important role in enhancing patient satisfaction, improving patient outcomes, and providing high-quality services at a lower cost (Gardner et al. 2005; Yellen and Davis 2001; Bain et al. 1999). Despite advancements in ASC practices, unplanned hospital admission after ambulatory surgery negatively impacts the patient experience and can result in increased costs to an already stressed healthcare system (Imasogie and Chung 2001; Siragusa et al. 2023; Rosen et al. 2024). The incidence of unplanned hospital admission after ambulatory surgery has been a measure of quality in anesthesia care (Shnaider and Chung 2006; Warner et al. 1993; Vila et al. 2003; Martinez Rodenas et al. 2014). The US Center for Medicare and Medicaid Services (CMS) considers unplanned admission directly from an ambulatory surgery facility or to the hospital after initial discharge to home as one of the quality metrics for ambulatory surgery facilities (Available from: https://www.cms.gov/medicare/quality/initiatives/asc-quality-reporting).

Few studies have assessed hospital admissions within 24 h of surgery at ASCs (Melton et al. 2021; Whippey et al. 2013; Brown et al. 2021; Garcea et al. 2008). Most of the previous studies are dated, focused on a single specialty, or studied 30-day admissions after ambulatory surgery (Garcea et al. 2008; Mezei and Chung 1999; Coley et al. 2002; Khan et al. 2005; Caelenberg et al. 2019; Dallas et al. 2017; Teja et al. 2020; Doersch et al. 2019; Bongiovanni et al. 2021; Fancourt-Smith et al. 1990; Gold et al. 1989; Levin et al. 1990; Keller et al. 2023; Paez et al. 2007). Therefore, the primary objective of this study was to study the incidence of unplanned hospital admission within 24 h after surgery at an ASC; a key secondary objective was to identify risk factors associated with these unplanned admissions.

Methods

This retrospective study was approved by the Institutional Review Board (IRB 202300138) at Stony Brook University Hospital. The requirement for written informed consent was waived by the IRB.

Stony Brook Hospital is a tertiary care academic institution. Its ambulatory surgery center has 10 operating rooms and performs approximately 8–9000 procedures annually. Surgical specialties include orthopedics, plastic, general surgery, ophthalmology, ear nose and throat (ENT), urology, gynecology, colorectal, breast, and dental. Unplanned hospital admission was defined as patients who had an unplanned hospital admission within 24 h of surgery at our ASC. Admissions included any patient that was directly transferred from the ASC to the hospital as well as patients who were discharged home from the ASC but then were admitted within 24 h of their anesthetic care.

A query was submitted through our hospital’s Enterprise Reporting Department to obtain data for all adult patients ≥ 18 years. who underwent surgery at this ASC between January 1, 2016, to December 31, 2022. Data were extracted from the anesthesia EMR (Cerner, North Kansas City, Missouri). Missing data were treated as missing, i.e., not imputed. The query included patient information such as demographics including age, gender, Body mass index, duration of surgery (< 1 h, 1–3 h, and > 3 h), American Society of Anesthesiology (ASA 1–4) postoperative nausea and vomiting (PONV) score (0–5), type of anesthesia (general anesthesia, monitored anesthesia care, local, regional), surgical specialty (orthopedics, ophthalmology, urology, gynecology, plastic, breast, ENT, Dental, general and colorectal). We further condensed surgical specialty into General, (Colorectal, Breast, Plastic, Vascular, Surgical Oncology, Trauma), Ortho (Ortho, Pain), Gyn/Urology (Gyn, OB service, Urology), Ophthalmology, and others (Dental, OMFS, ENT, Neuro, Endo, Special procedures). Surgical duration was defined as surgical start time (time of incision) to surgical stop time (dressing applied). To avoid confounding, surgical duration was considered as a surrogate for surgical complexity with less than an hour as low, 1–3 h being moderate, and above 3 h as high. We collected comorbidities (defined as a history of the disorder during the patient’s lifetime) including diabetes mellitus, hypertension, obstructive sleep apnea, hyperlipidemia, coronary artery disease, peripheral vascular disease, atrial fibrillation, supraventricular tachycardia, anemia, asthma, valvular heart disease, congestive heart failure, chronic obstructive pulmonary disease, chronic kidney disease, thyroid disease, deep venous thrombosis, and cancer. Cases with age < 18 years, ASA > 4, and epidural, IV regional, and spinal anesthesia were excluded. Data were divided into those with (cases) and without (controls) admission to the hospital. Cases were studied by the author S. Shah to extract information for reasons of admission.

Statistical analysis

Chi-square tests with exact p values based on Monte Carlo simulation were utilized to examine the marginal association between categorical variables and unplanned readmission status. Wilcoxon rank sum tests were utilized to examine the marginal difference in continuous variables, such as age or surgical duration, between patients who did and did not have an unplanned admission following surgery. Due to the small number of admissions (49 cases), a maximum of five variables could be used in the multivariable logistic regression. Surgical duration time in mins, surgical specialty, anesthesia type, PVD, and DVT were significantly related to admission status based on univariate analyses with p value < 0.05. These variables were further adjusted in a multivariable logistic regression model. Firth’s correction was applied to mitigate the bias caused by rare events and low cell counts for unplanned admissions. Because there was an extremely small number of unplanned readmissions, Firth correction was used to adjust for bias allowing for more reliable estimates in the presence of the low event rate, while also providing more accurate confidence intervals and p values (D Firth 1993).

The odds ratio (OR) > 1 indicated a higher risk of unplanned admission, while an OR < 1 indicated a lower risk of unplanned admission. Statistical analyses were performed by a Biostatistician co-author (S. Stanley) using SAS 9.4.

Results

A total of 61,193 surgeries were performed at the ASC during the 7-year period (2016–2022). We excluded cases ≤ 18 years of age, surgeries with a duration of 0 min reflecting nerve injection procedures, ASA > 4 (incorrect assignments), and anesthesia type (epidural, IV regional, spinal). This resulted in a final sample of 53,185 surgeries. There were 53,136 (99.9079%) uneventful discharges and 49 (0.09213%) unplanned admissions (Fig. 1).

Fig. 1

53,185 surgeries were included in the sample. We excluded cases < 18 years of age, surgeries with a duration of zero minutes reflecting nerve injection procedures, ASA > 4 (incorrect assignments), and anesthesia type (epidural, IV regional, spinal)

The overall incidence of unplanned hospital admissions (2016–2022) was 0.09%; (95% CI 0.07–0.1122%); Because the incidence of unplanned hospital admission was very low, a Poisson Distribution was used to determine the 95% confidence limits, which are labeled by the horizontal bars (Fig. 2).

Fig. 2

Percentage of unplanned hospital admissions after surgery at the ASC from the years 2016–2022. Ninety-five percent confidence limits are shown in the table and also represented by the horizontal bars around the actual/average incidence for that year

The median (interquartile range) age of the entire cohort was 57 (43–69) and the median [interquartile range] BMI for the study was 28.1 (24.7–32.3) kg/m². The majority of subjects were female (60.8%; n = 32,345) and classified as ASA class 2 (56.1%; n = 29,842). General Anesthesia was the most common anesthesia type (49.3%; n = 26,232) followed by MAC (48.3%; n = 25,690) and local (2.4%; n = 1263). Orthopedics was the surgical specialty with the highest number of surgical cases (28.8%; n = 15,302).

Univariate analyses showed significant differences in anesthesia type, surgical duration, and surgical specialty between the two groups. Patients who had an unplanned admission were more likely to have General Anesthesia (cases 75.5%; n = 37 vs. controls 49.3%; n = 26,195, P = 0.0030), a longer surgical duration (cases 64.0 min; 29.0–119.0 vs. controls 29.0; 18.0–58.0, P < 0.0001) and underwent general surgery (cases 36.7%; n = 18 vs. controls 19.1%; n = 10,150). Among comorbidities, PVD (cases 6.1%; n = 3 vs. controls 0.8%; n = 448) and DVT (cases 6.1%; n = 3 vs. controls 1.5%; n = 818) were significantly different (univariate analysis) between the two groups (Table 1).

Table 1 Patient characteristics and univariate regression model for patients with and without unplanned hospital admission

Variables with a p < 0.05 from the univariate analyses were further used in the multivariable logistic regression. Anesthesia type, surgical duration time, surgical specialty, PVD, and DVT were included in this model, which showed increased odds of unplanned admission associated with longer surgical duration time (odds ratio (OR), 1.01; 95% confidence interval (CI), 1.007 to 1.012, P =  < 0.0001 PVD (OR, 14.489; 95% CI, 4.862 to 43.174, P =  < 0.001), and DVT (OR, 5.527; 95% CI 1.909 to 16.001, P = 0.0016 (Table 2)). No significant collinearity was observed.

Table 2 Multivariate logistic regression model for unplanned hospital admission

Surgical causes (55%) were the most common reason for admission followed by medical (22%), anesthesia (16%), and other (6%) (Table 3).

Table 3 Patient reasons for unplanned hospital admissions

Discussion

The overall incidence of unplanned hospital admissions was very low after surgery at a large US academic tertiary care ASC. General Anesthesia, longer surgery duration, and co-morbidities (peripheral vascular disease, and deep venous thrombosis) were associated with a higher risk of unplanned hospital admission. The incidence of hospital admission within 24 h observed in our study (0.09%) was lower than reported in previous studies [0.3–9.5%] (Whippey et al. 2013; Brown et al. 2021; Garcea et al. 2008; Mezei and Chung 1999; Coley et al. 2002; Khan et al. 2005; Caelenberg et al. 2019; Dallas et al. 2017; Teja et al. 2020; Doersch et al. 2019; Bongiovanni et al. 2021; Fancourt-Smith et al. 1990; Gold et al. 1989; Levin et al. 1990). This may be because some of these data are old and modern ASC practices might result in better patient care and thus outcomes. In addition, in our study, there was no statistically significant change noted in incidence during the COVID-19 pandemic.

Comparing our results to previous reports, some previous studies reported admissions within 30 days of discharge, which we agree is important for some surgical complications, e.g., infection (Mezei and Chung 1999; Coley et al. 2002; Teja et al. 2020; Bongiovanni et al. 2021; Keller et al. 2023; Paez et al. 2007). The incidence of unplanned admissions reported in these studies ranges from 1.1 to 2%, which is higher than ours and likely due to the longer follow-up duration in those studies (30 days) vs. ours (24 h). Bongiovanni and colleagues, also presenting data that is older (2009–2011), found that the majority of these unplanned visits were within the first few days of surgery (Bongiovanni et al. 2021). Mezei et al., reporting data from two decades ago, found that surgical complications were the most common reason for admission, followed by medical and anesthesia-related events (Mezei and Chung 1999). Teja et al. proposed an instrument to predict unplanned hospital admissions, however, it did not include disease severity, e.g., ASA physical status, which may be important (Teja et al. 2020). Other studies presenting 30-day admissions after surgery at an ASC include studies focused on a single surgical specialty like urological surgery or a single procedure, e.g., robotic-assisted radical prostatectomy (Keller et al. 2023; Paez et al. 2007) which are important but not helpful to understand outcomes for the larger ASC population.

We observed that patients who had surgery of moderate or high complexity or had general anesthesia were more likely to have an unplanned admission. This is similar to earlier reports (Melton et al. 2021; Whippey et al. 2013; Garcea et al. 2008; Khan et al. 2005; Caelenberg et al. 2019; Levin et al. 1990). It is important to note that ASA was not associated with an increased risk of admissions in our study as has been reported in previous studies (Melton et al. 2021; Whippey et al. 2013; Brown et al. 2021; Caelenberg et al. 2019; Gold et al. 1989; Levin et al. 1990), which may reflect our short-term 24-h definition that might possibly reflect different types of complications than those observed over a longer 30 day period. Consistent with this, we also did not find age to be associated with an increased risk for hospital admissions in contrast to risk.

Melton et al. 2021; Brown et al. 2021; Khan et al. 2005; Fancourt-Smith et al. 1990; Levin et al. 1990). In addition, BMI and OSA were not found to be associated with an increased risk for unplanned admissions in our analysis compared to other studies where a higher BMI was associated with a higher rate of admissions (Whippey et al. 2013). Our findings of a very low admission rate despite many patients with higher age, ASA classification, and BMI, suggest that it may be feasible to expand ASC eligibility to even higher-risk patients.

Surgical causes (55%) were the most common reasons for admission in this cohort. This was followed by admissions due to medical reasons (22%), anesthesia (16%), and other reasons (6%) (Table 3). This is consistent with other studies where surgical causes including bleeding, pain, observation for a surgical complication, and urinary retention were the most common causes (Whippey et al. 2013; Brown et al. 2021; Garcea et al. 2008; Mezei and Chung 1999; Coley et al. 2002; Teja et al. 2020; Paez et al. 2007). Interestingly, only a few patients were admitted due to reasons, e.g., postoperative nausea and vomiting (n = 3), more directly linked to anesthesia.

Some limitations of our study should be considered. The study was a retrospective analysis, so associations (not “cause and effect”) are reported, and potential biases may have been introduced. It was also from a single center so its results might not be generalizable to other ASCs. Our analysis did not include ED visits or hospital admissions to a different hospital outside of our healthcare system since it was not feasible to identify those visits or admissions retrospectively in such a large cohort n = 53,185. However, given the short time frame for our primary endpoint (i.e., 24 h) it seems unlikely that many patients presented to another hospital for care so soon after surgery at our ASC. Our study presents data from a single center and future studies could be a multicenter design to increase external validity. Finally, our study did not include pediatric cases and patients who were ASA > 4 so we cannot generalize our findings to that population.

Our study has numerous strengths including standardized definitions and a large sample of patients [n = 53,185] over a 7-year period with recent data (2016–2022). This is important since many previous studies that reported unplanned admissions within 24 h following ambulatory surgery reflect surgery that occurred 8 to 10 years ago.

Conclusion

The incidence of unplanned hospital admission after surgery at a large US academic center ASC was very low which suggests that modern ambulatory anesthetic care can be safely given to large numbers of patients. This admission rate can also serve as a reference point benchmark for future studies and Quality Improvement initiatives. It is interesting to note that few admissions appeared to be due to classically reported anesthesia-related issues such as PONV.