Abstract
Background and aims
The aim of this study was to investigate the nationwide trend of feeding tube placement during the COVID-19 pandemic and to explore its potential implications for the healthcare system.
Methods
This was a nationwide retrospective trend analysis of feeding tube placement in adult hospitalizations from 2016 to 2020 using the National Inpatient Sample database. We analyzed the total number of feeding tubes placed endoscopically PEG and percutaneous endoscopic jejunostomy [PEJ]), surgically, and by Interventional Radiology (IR).
Results
The average monthly case number of feeding tube placements increased from 20,902 before the pandemic to 22,358 during the pandemic, which corresponds to .8% and 1.0% of hospitalizations, respectively. The odds of undergoing a feeding tube placement during the pandemic were significantly higher than before the pandemic (odds ratio [OR], 1.20; 95% confidence interval [CI], 1.15-1.24; P < .001). Quantitatively, there were 23,904 (95% CI, 17,294-30,515) excess feeding tube placements, or 12,954 (95% CI, 7,116-18,793) excess feeding tube placements in patients surviving hospitalizations during the pandemic. The majority of the increase in feeding tube placements were done endoscopically, as shown by a higher number of excess cases (25,052; 95% CI, 21,543-28,560) compared with the surgical group (4548; 95% CI, 3336-5761) and the IR group (–4106; 95% CI, –9488 to 1276).
Conclusions
Our study highlights the significant increase in feeding tube placements during the COVID-19 pandemic. Future research is warranted to accurately estimate the long-term care needs and evaluate the potential impact of increased feeding tube placements during the pandemic on the healthcare system.
The emergence of SARS-CoV-2 in December 2019 resulted in a global pandemic of COVID-19, with significant morbidity and mortality in the United States. Patients with COVID-19 and ventilator-dependent chronic respiratory failure frequently require enteral nutrition through nasogastric or orogastric tubes in the intensive care unit (ICU). The prolonged use of these tubes can lead to adverse events, including an increased risk of aspiration pneumonitis due to impaired swallowing and incomplete closure of the distal esophageal sphincter. Feeding tubes for enteral nutrition may be considered as an alternative, although they are also associated with adverse events such as pneumonia, ileus, peristomal infection, buried bumper syndrome, and stomal leakage. Despite these potential adverse events, feeding tubes may provide relief for both patients and healthcare practitioners, while providing reliable means of nutrition upon discharge. Therefore, feeding tube placement in conjunction with tracheostomy in patients with ventilator-dependent chronic respiratory failure secondary to COVID-19 may decrease ICU length of stay and free up ICU beds for other patients during the pandemic.
Despite the limited research on feeding tube placement in the United States during the COVID-19 pandemic, it remains a topic of significant importance due to the nationwide challenges of staffing shortages and infection control in the new pandemic-era hospital environment. Using the National Inpatient Sample (NIS), the largest nationwide inpatient database, we aimed to examine the volumes and outcomes of feeding tube placement procedures from 2016 to 2020 to gain a more comprehensive understanding of the feeding tube utilization during the pandemic. We hope that this study better informs the allocation of healthcare resources in the future.
Methods
Study design and database description
This study was a retrospective, cross-sectional examination of hospitalizations involving feeding tube placement at acute care hospitals in the United States in 2020, with data from 2016 to 2019 analyzed for contextual comparison. Throughout this article, the term "feeding tube placement" is used to describe all procedures involving the placement of gastrostomy and jejunostomy tubes, whether they were inserted by Surgery, Interventional Radiology (IR), or Gastroenterology. Specific techniques are mentioned only when analyzing or discussing such procedures. The inclusion criteria for this study required adult hospitalizations, defined as patients aged ≥18 years on hospital admission. The data used in this study were extracted from the NIS from 2016 to 2020. No institutional review board evaluation was required for the use of limited data sets in this study under the Health Insurance Portability and Accountability Act of 1996.1
Identifying diagnosis and procedures
The International Classification of Diseases, Tenth Revision (ICD-10), Procedure Coding System (ICD-10-PCS), was used to identify procedures, including feeding tube placement. The ICD-10-PCS is a standardized system developed by the Centers for Medicare & Medicaid Services (Bethesda, Md, USA) in partnership with 3M Health Information Management (Cedar Valley, Utah, USA) that uses alphanumeric codes to represent medical procedures. These codes are composed of 3 to 7 digits, with the first digit indicating the specific section of medical practice, and subsequent digits specifying the body system, root operation, body part, approach, and device used. In this study, the codes "0DH6" and "0DHA" were used to identify gastrostomy and jejunostomy, respectively. Subsequent digits were used to specify the procedural techniques used, such as PEG and percutaneous endoscopic jejunostomy (PEJ) placement by Gastroenterology, and feeding tube placement by Surgery or IR. In addition, ICD-10-Clinical Modification codes were used to identify various disorders and symptoms, as presented in Supplementary Table 1 (available online at www.igiejournal.org).
Study outcomes and variable definition
To characterize the patient population, we extracted the following variables from the NIS: age, sex, race, median household income, primary expected payer (ie, insurance), hospital size (based on the number of beds) (Supplementary Table 2, available online at www.igiejournal.org), census region, urban or rural location, and academic or nonacademic status of the inpatient institution. Comorbidity burden was quantified by using the Elixhauser comorbidity index.2 In-hospital all-cause mortality rates were determined by using discharge status in the NIS, as the NIS does not record cause of death data. Hospitalization costs and charges for the various years were adjusted for inflation compared with the inflation rate in July 2020.
Statistical analysis
Statistical analyses were conducted by using STATA version 17.0 (StataCorp, College Station, Tex, USA). Survey commands were used in all analyses to account for the stratification, clustering, and weighting to produce nationally representative unbiased results with Healthcare Cost and Utilization Project–provided methodology.3 Continuous variables were analyzed by using the Student t test if they were normally distributed and the Wilcoxon rank sum test if they were not continuous. Proportions were compared by using the χ2 test.
To predict pandemic monthly feeding tube placement volumes based on pre–COVID-19 pandemic data, trigonometric regression analysis was used, which accounts for both time trends and yearly seasonality. Sine and cosine functions reflecting seasonal patterns were used to capture the variations in feeding tube placement volumes throughout the year. By incorporating these functions into a linear regression model and fitting it to the pre-pandemic data, a model was developed that predicts expected feeding tube placement volumes during the pandemic in the absence of the pandemic, considering both time trends and seasonal patterns. Excess feeding tube placement procedure volume was calculated as the difference between the actual and predicted feeding tube placement volumes during the pandemic period. Using the trigonometric regression model derived from pre-pandemic data, we generated predictions and their corresponding 95% confidence intervals (CIs) for feeding tube placements during the pandemic months. The excess feeding tube placement volumes and their 95% CIs were then obtained by subtracting the actual volumes from the predicted volumes and lower/upper bounds of the CIs. Logistic regression was used to compare binary variables, and linear regression was used to compare continuous variables. Multivariate regression was used to compare outcomes with adjustment for confounders. Confounders were identified as variables with a univariate regression P value <.2 or deemed clinically relevant. All P values were 2-sided, with a significance level of .05.
Results
Patient and hospital characteristics
Table 1 compares the characteristics of hospitalizations with feeding tube placements that occurred before and during the COVID-19 pandemic. The median age of patients included in this study was 68 years before the pandemic and 67 years during the pandemic. The proportion of female patients was slightly lower during the pandemic period. Scores on the Elixhauser comorbidity index, a measure of the burden of pre-existing medical conditions, were similar between the 2 periods. However, hospitalizations with feeding tube placements performed during the pandemic were characterized by a higher prevalence of acute end-organ damage, sepsis, tracheostomy, intubation, and palliative care, all measures indicating a higher severity of illness. There were also slight differences in the racial distribution of patients included in the study before and during the pandemic, with a lower percentage of white patients and a higher percentage of black, Hispanic, and Asian or Pacific Islander patients during the pandemic period. Median household income and insurance status were largely similar between the 2 periods, as were the characteristics of the hospitals, including region, bed size, and urban location. However, the percentage of teaching hospitals was higher during the pandemic.
Table 1.
Patient and hospital characteristics of hospitalizations with feeding tube placement before and during the COVID-19 pandemic
| Characteristic | Pre-pandemic∗ |
Pandemic∗ |
P value |
|---|---|---|---|
| (n = 1,045,100) | (n = 223,575) | ||
| Age, median (IQR), y | 68 (57-78) | 67 (57-77) | <.001 |
| Sex, female, % | 44.1 | 43.3 | .003 |
| Elixhauser comorbidity index score, median (IQR) | 5 (4-7) | 5 (4-7) | <.001 |
| Race, % | <.001 | ||
| White | 63.6 | 59.0 | |
| Black | 19.1 | 20.8 | |
| Hispanic | 10.0 | 12.0 | |
| Asian or Pacific Islander | 3.5 | 3.8 | |
| Native American | .5 | .7 | |
| Other | 3.4 | 3.8 | |
| Median household income in the patient's ZIP code, % | .527 | ||
| $1-$42,999 | 32.6 | 32.4 | |
| $43,000-$53,999 | 25.5 | 26.1 | |
| $54,000-$70,999 | 23.0 | 22.3 | |
| $71,000 or higher | 19.0 | 19.3 | |
| Insurance, % | <.001 | ||
| Medicare | 63.8 | 61.2 | |
| Medicaid | 14.6 | 15.9 | |
| Private | 19.1 | 20.2 | |
| Self-pay | 2.5 | 2.6 | |
| Critical conditions, % | .0 | .0 | |
| Acute end-organ injury | 64.2 | 71.0 | <.001 |
| Sepsis | 32.2 | 38.7 | <.001 |
| Tracheostomy | 19.8 | 21.2 | <.001 |
| Intubation | 26.3 | 29.3 | <.001 |
| Palliative care | 12.5 | 15.5 | <.001 |
| COVID-19, % | .0 | 13.9 | <.001 |
| Hospital region, % | .500 | ||
| Northwest | 17.5 | 19.2 | |
| Midwest | 21.8 | 22.0 | |
| South | 41.8 | 39.7 | |
| West | 18.8 | 19.1 | |
| Hospital bed size, % | .094 | ||
| Small | 13.6 | 15.6 | |
| Medium | 27.0 | 25.5 | |
| Large | 59.4 | 59.0 | |
| Hospital urban location, % | 95.6 | 95.6 | .850 |
| Teaching hospital, % | 78.4 | 82.1 | <.001 |
IQR, Interquartile range.
The pre-pandemic period is defined as January 2016 to February 2020, and the pandemic period is defined as March 2020 to December 2020.
Trend of feeding tube placement
The average monthly frequency of feeding tube placement increased from 20,902 before the COVID-19 pandemic to 22,358 during the pandemic, corresponding to .8% and 1.0% of hospitalizations, respectively. The odds of undergoing a feeding tube placement during the pandemic was significantly higher than before the pandemic (odds ratio [OR], 1.20; 95% CI, 1.15-1.24; P < .001). However, when controlling for COVID-19 and adjusting for age, sex, race, household income, and the Elixhauser comorbidity index, the difference in the odds of undergoing a feeding tube placement during the pandemic compared with before the pandemic was less statistically significant (adjusted odds ratio [aOR], 1.04; 95% CI, 1.01-1.09; P = .011). Figure 1A illustrates a trend of increased feeding tube placement during the pandemic. The trend falls within predicted feeding tube placement rates when COVID-19–related hospitalizations were excluded (Fig. 1B), consistent with the aforementioned findings. Quantitatively, there were 23,904 (95% CI, 17,294-30,515) excess feeding tube placements in any patients (discharged alive or died) or 12,954 (95% CI, 7116-18,793) excess feeding tube placements in patients who survived hospitalization during the pandemic. In addition, an increase in length of stay for 1 day was associated with a 7% increase in the odds of feeding tube placement (OR, 1.07; 95% CI, 1.07-1.07; P < .001). Moreover, mechanical ventilation was associated with a 15.7 times increased odds of feeding tube placement (OR, 15.7; 95% CI, 15.4-15.9; P < .001).
Figure 1.
The trend of feeding tube placement performed from 2016 to 2020. A, Trend of all feeding tube placements performed. B, Trend of feeding tube placements performed excluding COVID-19 as a diagnosis.
Feeding tube placement and comorbidities
During the 2020 pandemic year, 13.9% of patients who had a feeding tube placement during hospitalization were diagnosed with COVID-19, with 6.3% having COVID-19 as the primary diagnosis and 7.6% as a secondary diagnosis. Compared with patients who had feeding tube placements during a hospitalization before the COVID-19 pandemic, those who had the procedure during the pandemic were significantly more likely to require intubation (27.5% vs 24.7%, P < .001) and tracheostomy (21.2% vs 19.8%, P < .001). Patients who had feeding tube placements during the pandemic were also more likely to have acute organ decompensation (including acute kidney injury, coagulopathy, altered mental status, shock, acute respiratory failure, or liver failure; 71.0% vs 64.2%, P < .001), sepsis (38.7% vs 32.2%, P < .001) or required palliative care (15.5% vs 12.5%, P < .001).
Feeding tube placement subtypes according to technique
During the COVID-19 pandemic, the odds of hospitalized patients receiving feeding tube placements by gastroenterologists (ie, via endoscopic approach) were significantly higher (OR, 1.37; 95% CI, 1.28-1.47; P < .001) compared with patients admitted before the pandemic. Similarly, the odds of patients receiving feeding tube placements performed by IR (OR, 1.10; 95% CI, 1.06-1.14, P < .001) and surgeons (OR, 1.32; 95% CI, 1.24-1.41, P < .001) were also significantly higher during the pandemic. Figure 2 illustrates the trend of feeding tube placement techniques over time. The greatest increase in feeding tube placements was through the endoscopic approach, as indicated by a higher excess number of cases (25,052; 95% CI, 21,543-28,560) compared with surgical approaches (4548; 95% CI, 3336-5761). There was no significant increase in the volume of feeding tube placements by IR (excess cases, –4106; 95% CI, –9488 to 1276).
Figure 2.
The trend of feeding tube placement according to procedural approach. IR, Interventional Radiology; GE, Gastroenterology.
Feeding tube placement outcomes
During the COVID-19 pandemic, adverse events related to feeding tube placement occurred more frequently (6.8% vs 6.1%; OR, 1.13; 95% CI, 1.07-1.19; P < .001). This remained true after adjusting for various covariates such as patient age, sex, elective procedure status, Elixhauser comorbidity index score, race, insurance, hospital size, teaching status, COVID-19 infection, acute end-organ decompensation, sepsis, intubation, tracheostomy, and palliative care (adjusted OR [aOR], 1.26; 95% CI, 1.20-1.33; P < .001). Specifically, hemorrhage (aOR, 1.21; 95% CI, 1.03-1.43; P = .024) and tube malfunction (aOR, 1.41; 95% CI, 1.33-1.50, P < .001) were more commonly observed during the pandemic. Although the mortality rate seemed higher in patients who received feeding tube placement during the pandemic (13.3% vs 9.6%), the adjusted odds of death were slightly lower (aOR, .94; 95% CI, .89-1.00; P = .040). There was no significant difference in length of hospital stay between pre-pandemic and pandemic hospitalizations with feeding tube placement, although the cost of hospitalization was higher ($73,698 vs $59,970, P < .001) (Table 2).
Table 2.
Feeding tube placement outcomes before and during the COVID-19 pandemic
| Variable | Pre-pandemic∗ | Pandemic∗ | Multivariate regression† |
|---|---|---|---|
| Any adverse event | 6.1% | 6.8% | 1.24 (1.17 to 1.32), <.001 |
| Hemorrhage | .4% | .5% | 1.21 (1.03 to 1.43), .024 |
| Infection | 1.0% | .9% | 1.01 (.90 to 1.14), .847 |
| Malfunction | 4.3% | 5.3% | 1.41 (1.33 to 1.50), <.001 |
| Other | 1.0% | .7% | .88 (.77 to 1.00), .051 |
| Mortality | 9.6% | 13.3% | .94 (.89 to 1.00), .040 |
| LOS (d) | 18.9 | 21.0 | .11 (–.23 to .45), .533 |
| Cost of hospitalization‡ | $59,970 | $73,698 | 5498.41 (3379.92 to 7616.90), <.001 |
LOS, Length of hospital stay.
The pre-pandemic period refers to January 2016 to February 2020, and the pandemic period refers to March 2020 to December 2020.
In the multivariate regression, the listed results are as follows: for binary outcomes, adjusted odds ratio with 95% confidence interval and P value; for continuous outcomes, beta-coefficient (mean difference) with 95% confidence interval and P value.
Hospitalization costs and charges for the various years were adjusted for inflation to July 2020.
Discussion
The current study showed that there were 23,904 (95% CI, 17,294-30,515) excess feeding tube placements beyond predicted in all patients and 12,954 (95% CI, 7,116-18,793) excess feeding tube placements in patients surviving hospitalizations during the COVID-19 pandemic in 2019 in the United States. At the national level, the odds of undergoing a feeding tube placement during the pandemic were significantly higher than before the pandemic. This substantial increase in feeding tube placements is likely attributed to the significant number of patients requiring prolonged hospitalization and mechanical ventilation. Our findings support the association between these factors and feeding tube placement, with each day of increased length of stay being associated with a 7% increased odds of feeding tube placement and the odds of feeding tube placement being 15.7 times higher in patients requiring mechanical ventilation. Previous research has shown that the administration of early enteral nutrition, delivered via continuous infusion through a nasogastric tube before switching to a feeding tube in patients requiring prolonged ventilation, significantly reduces the risk of mortality among critically ill patients with COVID-19.4, 5, 6, 7, 8 As an added benefit, continuous enteral nutrition may decrease the risk of exposure to the virus for healthcare providers.6
Our investigation found that there was a significant increase in the number of patients who received feeding tubes and survived hospitalization during the COVID-19 pandemic in 2020. This surge in feeding tube use may potentially lead to a public health burden related to the care and management of these devices, warranting further investigation. Patients who required new feeding tubes were more likely to be discharged to a facility or inpatient rehabilitation/nursing services.9 In addition, patients experiencing feeding tube malfunction may have been directed to the emergency department for evaluation and imaging.10 In response to this increased need, the adoption of telemedicine and advanced home care technology has surged, offering support for basic enteral nutrition feeding administration and feeding tube care. These new technologies may offer secondary benefits such as increased access to care and reduced risk of exposure to the virus. However, with the higher volume of feeding tube placements and the subsequently higher rate of hemorrhage and tube malfunction observed during the pandemic, we anticipate a significant increase in the need for feeding tube care in the post-pandemic era. Our quantitative results may assist healthcare policy makers in estimating the upcoming increased burden on the healthcare system and in planning accordingly.
Despite the observed increase in feeding tube placements during the COVID-19 pandemic, it is probable that their use continues to be insufficient for meeting patient needs. The use of nasogastric tubes for enteral nutrition may have been prolonged due to concerns regarding the high risk of COVID-19 transmission in the context of gastrostomy tube placement.10 In addition, patients requiring replacement of existing gastrostomy or jejunostomy tubes were likely constrained by scheduling delays of nonemergent procedures early in the pandemic. Taking these factors into consideration, the need for feeding tube placement during the remainder of the pandemic may be even higher, highlighting the potential impact on the healthcare system.
Our study showed that adverse events, particularly hemorrhage (aOR, 1.21; 95% CI, 1.03-1.43; P = .024) and tube malfunction (aOR, 1.41; 95% CI, 1.33-1.50; P < .001), were more frequently observed during the COVID-19 pandemic, even after adjusting for confounding variables. There are several potential explanations for this phenomenon. First, although previous retrospective studies have suggested that the overall post-procedural adverse event rate for feeding tube placement in patients with COVID-19 is comparable to patients without COVID-19,11,12 our study specifically compared adverse events before and during the pandemic. Therefore, the increased risk for adverse events may have affected patients both with and without COVID-19 during the pandemic, which may be related to the higher volume of procedures. Second, certain confounders were not available in our database, most notably the use of anticoagulation. The utilization of therapeutic doses of anticoagulation in all hospitalized patients, which was proposed early on during the pandemic, is still a topic of controversy.13,14 However, given the high rate of pulmonary embolism and deep venous thrombosis associated with COVID-19, therapeutic anticoagulation may be responsible for the higher risk of bleeding adverse events in patients receiving gastrostomy/jejunostomy noted in our study.
Given the observed higher mortality rate among patients who underwent feeding tube placement during the COVID-19 pandemic, it is crucial to consider the potential for reverse causation. Patients with more severe illness might have required feeding tube placements more frequently, contributing to the increased utilization rather than the feeding tube placements directly causing higher mortality. This suggests that the increase in feeding tube placements observed during the pandemic could be predominantly influenced by the heightened severity of illness among hospitalized patients. Nonetheless, feeding tube placements were associated with adverse events as denoted by procedure-specific ICD-10 codes. Further research is warranted to elucidate the causal relationship between disease severity, feeding tube placement utilization, adverse events, and their subsequent consequences in patients who received feeding tube placements during the COVID-19 pandemic.
To the best of our knowledge, the current study is the first to identify that the increased volume of feeding tube placements was due to endoscopic approach procedures (ie, PEG/PEJ), which were mostly performed by gastroenterologists, even though all subspecialties faced the same risk of staff contamination, equipment contamination, and the use of personal protective equipment.10 The etiology of this discrepancy in clinical practice is unclear and merits further exploration.
The current study has several limitations. First, because the NIS is an administrative database, diseases, procedures, and comorbidities were identified by using ICD-10 codes. The use of ICD-10 codes may potentially result in diagnostic misclassification. The primary outcome measure of this study, namely feeding tube placement volumes, was identified by using ICD-10-PCS codes, which have been shown to possess high validity due to their use in hospital billing. Although the use of ICD-10 codes in previous studies adds some support to their reliability,15 the lack of alternative methods for conducting trend analyses of procedures at the national level using sources other than administrative databases must also be acknowledged. We adhered to the proposed methodologic standards of research using the NIS to ensure strict quality of our analyses.16 Second, the NIS does not differentiate between comorbidities/adverse events that occur during a hospitalization and those that were already included in a patient's history. This discrepancy may affect the interpretation of the outcomes related to feeding tube–related adverse events. Of note, the availability of ICD-10 codes that specifically indicate adverse events related to feeding tube placement (eg, gastrostomy hemorrhage), rather than general adverse events (eg, hemorrhage due to a procedure), increases confidence in the validity of these codes. Lastly, due to the limitation of the NIS database, as a cross-sectional database, our study was not able to investigate how long the tube feeding was needed for each patient. Therefore, the exact number of patients requiring long-term care after feeding tube placement is unclear. This is a crucial aspect to be explored in the future.
The current study has many strengths. To the best of our knowledge, this study is the first to systematically assess the volume of feeding tube placement in the United States during the COVID-19 pandemic at a national level. Furthermore, the NIS is a nationally representative database that includes patients from a diverse array of hospitals, including those of varying sizes, teaching and nonteaching status, urban and rural locations, and private or public ownership across 49 states. This approach allows the current study to accurately reflect the real-world inpatient experience in the United States during the pandemic. These strengths make this study unique, and they provide valuable insight into the impact of the COVID-19 pandemic on feeding tube placement in patients in the United States.
In conclusion, the current study found that inpatient utilization of feeding tube placements, particularly those performed endoscopically by gastroenterologists, increased in the United States during the COVID-19 pandemic. This escalation seems related to the pandemic, as adjusting factors related to COVID-19 and patients reduced the likelihood of feeding tube placement, whereas monthly volumes remained within expectations when excluding COVID-19 hospitalizations. There was a significantly higher population of patients who received feeding tubes and survived hospitalization during the COVID-19 pandemic in 2020. Therefore, we anticipate a potential public health burden related to the care and treatment of feeding tube–related adverse events in the coming years, emphasizing the need for further research to accurately estimate long-term care needs and evaluate the impact on the healthcare system. Our quantitative analysis assists healthcare workers, administrators, and policy makers in anticipating the burden’s magnitude and planning necessary adjustments.
Disclosure
The authors declare no conflict of interest.
Appendix
Supplementary Table 1.
ICD-10-CM codes used to identify GI diagnoses
| Diagnosis | ICD-10-CM codes |
|---|---|
| Gastrostomy/jejunostomy | |
| Surgical | 0DH60UZ, 0DHA0UZ |
| IR | 0DH63UZ, 0DHA3UZ |
| PEG/PEJ | 0DH64UZ, 0DH68UZ, 0DH67UZ, 0DHA4UZ, 0DHA8UZ, 0DHA7UZ |
| Gastrostomy/jejunostomy adverse events | |
| Hemorrhage | K94.21, K94.11 |
| Infection | K94.22, K94.12 |
| Malfunction | K94.23, K94.13 |
| Unspecified | K94.20, K94.29, K94.10, K94.19 |
| COVID-19 | U071 |
| Intubation and mechanical ventilation | 0BH17EZ and (5A1935Z OR 5A1945Z OR 5A1955Z) |
| Palliative care | R00.0∗ |
| Hypotension | I95∗ |
| Acute respiratory failure | J96.0∗ |
| Shock | R57∗ |
| Acute kidney injury | N17∗ |
| Coagulopathy | D65∗, D696∗ |
| Sepsis | A267∗, A021∗, A227∗, A241∗, A312∗, A327∗, A39∗, A40∗, A41∗, A427∗, A5486∗, B007∗, B377∗, O020∗, O0337∗, O0387, O0487, O0737∗, O0882∗, O8604∗, R652∗, R7881∗, T80211, T8144∗, T826∗, T827∗, I2601∗, I2690∗, I76∗, O883∗, M00∗, I33∗ |
| Mental status change | R40∗, R4182∗, G934∗ |
| Liver failure | K720∗ |
Asterisk (∗) stands for any length of character string (including 0).
ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification; IR, Interventional Radiology; PEJ, percutaneous endoscopic jejunostomy.
Supplementary Table 2.
Bed size categories∗
| Region | Small | Medium | Large |
| Northeast region | |||
| Rural | 1-49 | 50-99 | ≥100 |
| Urban, nonteaching | 1-124 | 125-199 | ≥200 |
| Urban, teaching | 1-249 | 250-424 | ≥425 |
| Midwest region | |||
| Rural | 1-29 | 30-49 | ≥50 |
| Urban, nonteaching | 1-74 | 75-174 | ≥175 |
| Urban, teaching | 1-249 | 250-374 | ≥375 |
| Southern region | |||
| Rural | 1-39 | 40-74 | ≥75 |
| Urban, nonteaching | 1-99 | 100-199 | ≥200 |
| Urban, teaching | 1-249 | 250-449 | ≥450 |
| Western region | |||
| Rural | 1-24 | 25-44 | ≥45 |
| Urban, nonteaching | 1-99 | 100-174 | ≥175 |
| Urban, teaching | 1-199 | 200-324 | ≥325 |
National Inpatient Sample description of data elements. Available at: https://www.hcup-us.ahrq.gov/db/vars/hosp_bedsize/nisnote.jsp. Accessed October 5, 2022.
References
- 1.DUA Training—Accessible Version Healthcare Cost and Utilization Project data use agreement course. https://www.hcup-us.ahrq.gov/DUA/dua_508/DUA508version.jsp Available at:
- 2.Elixhauser A., Steiner C., Harris D.R., Coffey R.M. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27. doi: 10.1097/00005650-199801000-00004. https://www.jstor.org/stable/3766985 Available at: [DOI] [PubMed] [Google Scholar]
- 3.HCUP methods series calculating National Inpatient Sample (NIS) variances for data years 2012 and later. https://www.hcup-us.ahrq.gov/reports/methods/2015_09.jsp Available at:
- 4.Detopoulou P., Tsouma C., Papamikos V. COVID-19 and nutrition: summary of official recommendations. Top Clin Nutr. 2022;37:187–202. doi: 10.1097/TIN.0000000000000286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Martindale R., Patel J.J., Taylor B., Arabi Y.M., Warren M., McClave S.A. Nutrition therapy in critically ill patients with coronavirus disease 2019. JPEN J Parenter Enteral Nutr. 2020;44:1174–1184. doi: 10.1002/jpen.1930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Barazzoni R., Bischoff S.C., Breda J., et al. ESPEN expert statements and practical guidance for nutritional management of individuals with SARS-CoV-2 infection. Clin Nutr. 2020;39:1631–1638. doi: 10.1016/j.clnu.2020.03.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ojo O., Ojo O.O., Feng Q., et al. The effects of enteral nutrition in critically ill patients with COVID-19: a systematic review and meta-analysis. Nutrients. 2022;14:1120. doi: 10.3390/nu14051120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.World Health Organization Clinical management of severe acute respiratory infection (SARI) when COVID-19 disease is suspected. https://apps.who.int/iris/handle/10665/331446 Available at:
- 9.Kapur S., Cantor Z., Bishop P., Benjaram S., Lohia P. Need for inpatient rehabilitation and skilled nursing services in COVID-19 patients post hospitalization: a retrospective cohort study. Ann Palliat Med. 2023;12:90–102. doi: 10.21037/apm-22-858. [DOI] [PubMed] [Google Scholar]
- 10.Wells Mulherin D., Walker R., Holcombe B., Guenter P. ASPEN report on nutrition support practice processes with COVID-19: the first response. Nutr Clin Pract. 2020;35:783–791. doi: 10.1002/ncp.10553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shah A., Ahmed Z., Zeineddine F., Quigley E.M.M. Safety of percutaneous endoscopic gastrostomy placement in patients with SARS-CoV-2 infection. Gastroenterology Res. 2022;15:263–267. doi: 10.14740/gr1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lipcsey M., Stein D.J., Yu R.L., et al. High rate of percutaneous gastrostomy placement in COVID-19 patients with low overall complications. Tech Innov Gastrointest Endosc. 2021;23:385–387. doi: 10.1016/j.tige.2021.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tritschler T., le Gal G., Brosnahan S., Carrier M. Point: should therapeutic heparin be administered to acutely ill hospitalized patients with COVID-19? Yes. Chest. 2022;161:1446–1448. doi: 10.1016/j.chest.2022.01.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jimenez D., Rali P., Doerschug K. Counterpoint: should therapeutic heparin be administered to acutely ill hospitalized patients with COVID-19? No. Chest. 2022;161:1448–1451. doi: 10.1016/j.chest.2022.01.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Peery A.F., Crockett S.D., Murphy C.C., et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the United States: update 2021. Gastroenterology. 2022;162:621–644. doi: 10.1053/j.gastro.2021.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Khera R., Angraal S., Couch T., et al. Adherence to methodological standards in research using the National Inpatient Sample. JAMA. 2017;318:2011–2018. doi: 10.1001/jama.2017.17653. [DOI] [PMC free article] [PubMed] [Google Scholar]