Economic burden and secondary complications of influenza-related hospitalization among adults in the US: a retrospective cohort study

Abstract

Objective

This study aims to describe the healthcare resource utilization (HCRU) and direct medical cost of influenza-related hospitalizations to illustrate the persistent economic burden of influenza among adults in the US.

Methods

A retrospective cohort study was conducted using the PINC AI Healthcare Database. Adults hospitalized with a diagnosis of influenza between August 1–May 31 from 2016–2023 were identified and stratified by age (18–49, 50–64 and ≥65 years). The index hospitalization was defined as the individual’s first influenza-related hospitalization during each season. Patient demographics, comorbidities, and hospitalization characteristics were assessed during the index hospitalization. Index hospitalization length of stay (LOS), in-hospital mortality, intensive care unit (ICU) admissions, mechanical ventilation (MV) usage, and costs were evaluated overall and by MV usage, ICU admission, and secondary complication status. Pre-index influenza-related outpatient and emergency department (ED) visits (7 days prior) were also evaluated.

Results

Primarily initiated in the ED, the median LOS for influenza-related hospitalizations was 3–4 days. Inpatient mortality increased with age (2.2–4.4%). Combined mean hospitalization and initial ED visit costs were $12,556–$14,494 (2017/18; high severity season) and $11,384–$12,896 (2022/23; most recent season). Compared to other age groups, adults ≥65 years had higher proportions of hospitalization with no MV or ICU usage. Adults 18–49 years had the highest proportion of ICU admission only, whereas adults 50–64 years had the highest MV usage only and both MV and ICU admission. MV and/or ICU usage was associated with higher hospitalization costs. Increasing proportionally with age, the majority of influenza-related hospitalizations had a secondary complication diagnosis, which were associated with elevated costs.

Limitations

Analysis of this hospital-based administrative database relied on coding accuracy. Only hospital system-associated outpatient/ED visits were captured; the full scope of HCRU was under-ascertained.

Conclusions

The economic burden of influenza-related hospitalizations remains substantial, driven by underlying conditions, MV/ICU usage and secondary complications.

PLAIN LANGUAGE SUMMARY

This study described the healthcare resource utilization (HCRU) and costs for US adults ≥18 years old hospitalized with influenza and associated secondary complications such as pneumonia, asthma exacerbation and malignant hypertension between 2016–2023. The researchers analyzed a hospital admission database and found that, for the healthcare system, average cost per influenza-related hospitalization ranged from $11,384 to $14,494, depending on the influenza season and age of the patient. Over 96% of patients admitted to a hospital initially presented at the emergency department, 20–30% of patients required mechanical ventilation (MV) or intensive care unit (ICU) admission, and the median hospital length of stay was 3–4 days. This study adds to the existing evidence by providing economic burden estimates for the 2022/23 influenza season, the most recent influenza season after the COVID-19 pandemic, and found slightly lower HCRU and cost for influenza hospitalizations relative to prior seasons. Also, the study comprehensively analyzed economic burden by patient age groups and found lower HCRU and costs among patients ≥65 years compared to adults 18–49 years and 50–64 years consistently for all seasons. Additionally, the study found that the proportion of patients with MV usage alone, with MV usage and an ICU admission, and average hospitalization costs were greatest among patients 50–64 years, highlighting the potential benefit of increasing rates of seasonal influenza vaccination among this age group. Finally, the study found higher costs among patients with complications related to their influenza infection compared to patients without complications. Overall, the study found that influenza-related hospitalization can contribute to substantial economic burden in the US in the most recent time period.

Keywords:

• Influenza
• hospitalization
• health care resources utilization
• cost
• secondary complication

JEL CLASSIFICATION CODES:

• I11
• I1
• I
• I12

Introduction

Influenza is an acute respiratory infection caused primarily by influenza viruses type A and B, which circulate during the fall and winter seasons each year. The Centers for Disease Control and Prevention (CDC) estimated that during each year between 2010 and 2020, between 9 million and 41 million people in the United States (US) were infected with influenza, with 140,000 to 710,000 hospitalizations, and 12,000 to 52,000 deaths attributed to the illness.¹ Most influenza cases are relatively mild with symptoms resolving in 1–2 weeks. However, groups including older adults (≥ 65 years of age), children (< 2 years of age), and those with chronic medical conditions are at a higher risk of secondary complications leading to hospitalization, admission to the intensive care unit (ICU), and mechanical ventilation (MV).^2–5 Secondary complications include pneumonia, sepsis, cardiovascular complications, seizures, exacerbation of chronic medical conditions and death.^6–8

In addition to its morbidity and mortality, seasonal influenza has significant economic costs. In 2015, the economic burden of influenza in the US was estimated at $11.2 billion ($6.3–$25.3 billion) with direct medical costs of $3.2 billion ($1.5–$11.7 billion); adults ≥ 18 years of age accounted for 77.8% of the total direct healthcare burden, which was primarily driven by hospitalization costs.⁹ The presence of underlying medical conditions, which are associated with a higher risk of secondary complications, has been shown to result in up to 2.5-fold higher hospitalization costs, as well as increased MV usage and ICU admission.^5,10 Studies have also shown that influenza-related hospitalization costs were greater for those with specific comorbidities.^11,12 Similarly, adults with secondary complications had a 2.3-fold higher rate of hospitalization and 1.5-fold higher mean hospitalization costs compared to adults with no secondary complications during the 12-month period after an influenza infection.¹³

While the clinical and economic burden of influenza illness in the US has been investigated in several published studies, the majority evaluated the burden in influenza seasons including, or prior to, 2015.^9,11–18 There is a limited understanding of the healthcare resource utilization (HCRU) and costs of influenza, particularly in the most recent years where, in addition to the varying seasonal influenza burden attributed to changes in the circulating strains, there was altered social contact and healthcare seeking behavior due to the Coronavirus Disease 2019 (COVID-19) pandemic. In addition, previous studies did not comprehensively evaluate the usage of healthcare resources such as MV, ICU admission, and length of stay among adults of different ages. Furthermore, past literature did not explore how underlying conditions as well as secondary complications associated with influenza differentially contribute to the burden of influenza among adults of different ages in the US. The aim of this study was to provide more comprehensive and contemporary data on the HCRU and costs of influenza-related hospitalization among adults in the US during each of the influenza seasons between 2016 and 2020, as well as the 2022/23 season.

Methods

Study design

This study was a retrospective cohort study using the PINC AI Healthcare Database (PHD; formerly known as the Premier Healthcare Database). The PHD is a hospital-based data source with > 1,190 contributing hospitals representing approximately 25% of annual hospital admissions and 326 million unique patients across the US.¹⁹ The database includes hospital-, patient- and service-level information, such as hospital characteristics, patient demographics, admission and discharge diagnoses, as well as the detailed information and departmental level costs of procedures, medications, and laboratory diagnostic tests performed during inpatient stays. The use of masked identifiers allows for patients to be tracked within hospital systems to capture data from > 9 million inpatient admissions as well as > 89 million visits to emergency departments (ED), ambulatory surgery centers, and alternate sites of care within each PHD hospital system annually. All data are de-identified and fully compliant with the Health Insurance Portability and Accountability Act (HIPAA).

Study population

Adults ≥ 18 years hospitalized with an influenza diagnosis during each of the 2016/17, 2017/18, 2018/19, 2019/20, and 2022/23 influenza seasons were identified from the PHD. The 2020/21 and 2021/22 influenza seasons were excluded from the study due to social distancing measures, limited hospital resources for the treatment of influenza patients, testing patterns, and individual hygiene behavior during the COVID-19 pandemic in the US. The CDC’s Influenza Hospitalization Surveillance (FluServ) Network estimated much lower attack rates, medical visits, hospitalization, and mortality during 2021/22 season and provided no estimates for the 2020/21 season due to minimal influenza activity.^1,20 Consequently, these two seasons were excluded from analyses to avoid potentially biased estimates of the economic burden of influenza. Influenza diagnoses were identified using International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10) diagnosis codes J09.X, J10.X, and J11.X in any position and were required to be documented as “present on admission” (POA). Although the beginning and end of influenza activity in the US varies from year to year, an influenza season was defined in the study as the time period between August 1st and May 31st of each year to capture all influenza-related encounters during the season.^21,22 Patients with a diagnosis or procedure code indicating pregnancy or childbirth during the index hospitalization were excluded from the study. The index hospitalization was defined as the individual’s first influenza-related hospitalization during each influenza season, with the corresponding admission date defined as the index date. Patients from each influenza season cohort were stratified into the following age groups: 18–49 years, 50–64 years and ≥ 65 years. Individuals could contribute to multiple seasons if all eligibility criteria were met within each season.

The findings from the 2017/18 season (high severity season) as well as the 2022/23 season (the most recent season) are reported in detail in the main results. The results for all five influenza seasons are available in the Supplemental Data file.

Covariates

Patient characteristics evaluated during the index hospitalization included age at admission, sex, race/ethnicity, insurance type, and influenza discharge diagnosis position. The presence of chronic conditions associated with a higher risk of complications was assessed during the index hospitalization.ⁱ The presence of these chronic conditions during the index hospitalization were identified by ICD-10 diagnosis codes in primary or secondary positions on hospital discharge records and recorded as POA. Hospital characteristics evaluated during the index hospitalization included hospital location by census region and hospital population served (urban/rural). Patient characteristics, chronic conditions and hospital characteristics were summarized for influenza-related hospitalizations by season and age group.

Outcomes

HCRU and costs

HCRU, costs and in-hospital mortality were summarized for each influenza season and age group. HCRU measured during the index hospitalization included: the proportion of patients that initiated the index hospitalization in the ED, hospital length of stay (LOS), MV usage, ICU admission, and in-hospital mortality. An ED-initiated hospitalization was defined as a hospitalization having an inpatient indicator, an admission type description recorded as “Emergency” or “Urgent”, and ≥ 1 ED charge item. When available, the associated ED visit that initiated the hospitalization was defined as the initial ED visit. MV usage and ICU admission were identified by hospital revenue codes, ICD-10 diagnosis or procedure codes, Healthcare Common Procedure Coding System (HCPCS) procedure codes, and Current Procedural Terminology, 4th Edition (CPT-4) procedure codes (Supplemental Data: Table 1). MV usage included intubation, MV and extracorporeal membrane oxygenation (ECMO). In addition to the index hospitalization, influenza-related hospital-based standalone outpatient encounters within the 7-day period prior to the index hospitalization were identified based on the presence of an influenza diagnosis recorded at any position within an outpatient or ED visit record. Standalone outpatient encounters excluded initial ED visits associated with the index hospitalizations. The number of outpatient visits was measured as the number of distinct outpatient encounters; multiple outpatient encounters on the same day were counted as a single outpatient visit.

Costs were summarized for the initial ED visit, index hospitalization, and combined index hospitalization and initial ED visit, as well as for the pre-index standalone influenza-related outpatient and ED visits. The combined cost of the index hospitalization and initial ED visit was further stratified by MV usage and ICU admission. Patients with missing cost data were excluded from the corresponding cost analysis with patient counts reported for those affected analyses. All costs were adjusted to 2023 US dollars using the medical care component of the Consumer Price Index.²³ In-hospital all-cause mortality was assessed based on a discharge status indicating patient death during the index hospitalization and included death for any reason.

Influenza-related secondary complications

The proportion of patients with influenza-related secondary complications during the index hospitalization was summarized for each influenza season and age group. Secondary complications were defined as diagnoses codes in any position based on prior literature and included pulmonary conditions, metabolic failure, cardiovascular conditions, shock/sepsis, neurologic conditions, musculoskeletal conditions, cerebrovascular conditions and endocrine conditions (Supplemental Data: Table 1).² The proportions of patients with MV usage and/or ICU admission during the index hospitalization, as well as combined index hospitalization and initial ED visit costs, were summarized among patients with and without secondary complications during their index hospitalization for each season and age group. Combined index hospitalization and initial ED visit costs were further stratified by MV usage and ICU admission status.

Statistical analyses

Descriptive statistics were utilized to summarize patient and hospital characteristics, proportion of ED-initiated hospitalizations, number of pre-index standalone outpatient and ED visits, hospital LOS, in-hospital mortality, MV usage and/or ICU admission, proportion of index hospitalizations with secondary complications and costs. Counts and percentages were reported for categorical variables. Continuous variables were summarized using means, standard deviations (SD), medians, and first and third quartiles (Q1, Q3). No formal statistical tests were conducted to compare differences among groups. All results were reported for each season, and stratified into three age groups: 18–49 years, 50–64 years and ≥ 65 years. These age groups were selected to align with the age groups used by the CDC’s FluServ Network²² and the Advisory Committee on Immunization Practices (ACIP) recommendations for influenza vaccination.²⁴ They are also consistent with the age groups used in previously published studies of the economic burden of influenza.^9,11–13 All data analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC) or R (R Foundation, Indianapolis, IN).

Sensitivity analysis

A sensitivity analysis was performed restricting the patient cohorts of each season to those with an influenza diagnosis during their index hospitalization in the primary position. HCRU and cost outcomes associated with the index hospitalization and initial ED visit were summarized by influenza season and age group for this subgroup.

Results

Study population

Patient attrition for the 2017/18 and 2022/23 influenza seasons are presented in Figure 1 (Supplemental Data: Table 2). There were 102,856 and 58,872 adults hospitalized with a diagnosis of influenza between August 1st and May 31st of the 2017/18 and 2022/23 seasons, respectively, who met all selection criteria. Patients ≥ 65 years of age accounted for greater proportions of the influenza-related hospitalizations during both seasons (2017/18: 65.7%; 2022/23: 58.5%), while patients 18–49 years accounted for smaller proportions (2017/18: 11.6%; 2022/23:16.1%).

Figure 1. Patient attrition.

Patient characteristics, chronic conditions and hospitalization characteristics

Patient characteristics during index hospitalizations for each influenza season are presented in Table 1 (Supplemental Data: Table 3). Regardless of age group and season, a slightly higher percentage of hospitalized patients were female (range: 52.9% to 56.5%). Most patients with influenza-related hospitalizations were White or non-Hispanic. Higher proportions of patients 18–49 years had Medicaid and commercial insurance coverage. The proportion of patients with Medicare increased with age and became the primary insurance type in patients ≥ 65 years. The presence of underlying chronic conditions was common; one-third of patients 18–49 years and approximately half of patients 50–64 years had at least one chronic condition, the most frequent of which was diabetes either with or without complications. The presence of chronic conditions increased further to 56.1% among patients ≥ 65 years of age (55.7% in 2022/23); 23.7–25.1% had diabetes without complications, 18.3–21.1% had diabetes with complications, 13.7–17.6% had dementia, and 6.2–7.3% had a malignancy. Regardless of age group or season, more than 85% of influenza-related hospitalizations occurred in urban area hospitals.

Table 1. Descriptive characteristics for patients with an index influenza admission.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
Age
Mean (SD)	37.5 (9.0)	58.0 (4.1)	78.6 (7.7)	36.1 (8.9)	58.3 (4.2)	76.9 (7.4)
Median (Q1, Q3)	39 (31, 46)	58 (55, 61)	79 (72, 86)	37 (29, 44)	59 (55, 62)	76 (71, 83)
Sex, n (%)^a
Male	5,481 (45.9)	10,811 (46.3)	29,403 (43.5)	4,432 (46.7)	7,020 (47.1)	15,239 (44.2)
Female	6,466 (54.1)	12,538 (53.7)	38,143 (56.5)	5,067 (53.3)	7,889 (52.9)	19,224 (55.8)
Ethnicity, n (%)
Hispanic	1,490 (12.5)	1,829 (7.8)	4,116 (6.1)	1,832 (19.3)	1,547 (10.4)	2,978 (8.6)
Non-Hispanic	8,197 (68.6)	16,905 (72.4)	49,177 (72.8)	6,945 (73.1)	12,332 (82.7)	29,107 (84.5)
Unknown	2,263 (18.9)	4,620 (19.8)	14,259 (21.1)	723 (7.6)	1,030 (6.9)	2,378 (6.9)
Race, n (%)
Asian	191 (1.6)	336 (1.4)	1,621 (2.4)	201 (2.1)	214 (1.4)	794 (2.3)
Black	2,778 (23.2)	4,406 (18.9)	6,072 (9.0)	2,541 (26.7)	3,103 (20.8)	3,928 (11.4)
White	7,254 (60.7)	16,151 (69.2)	54,217 (80.3)	5,126 (54.0)	10,038 (67.3)	26,849 (77.9)
Other	1,364 (11.4)	1,995 (8.5)	4,447 (6.6)	1,137 (12.0)	1,084 (7.3)	1,943 (5.6)
Unknown	363 (3.0)	466 (2.0)	1,195 (1.8)	495 (5.2)	470 (3.2)	949 (2.8)
Insurance type, n (%)
Medicare	1,938 (16.2)	7,879 (33.7)	62,587 (92.7)	1,118 (11.8)	4,567 (30.6)	30,823 (89.4)
Medicaid	4,103 (34.3)	5,443 (23.3)	906 (1.3)	4,396 (46.3)	4,557 (30.6)	847 (2.5)
Commercial	3,921 (32.8)	7,613 (32.6)	2,660 (3.9)	2,761 (29.1)	4,400 (29.5)	1,568 (4.5)
Other	392 (3.3)	851 (3.6)	1,129 (1.7)	266 (2.8)	512 (3.4)	1,003 (2.9)
Uninsured	1,596 (13.4)	1,568 (6.7)	270 (0.4)	959 (10.1)	873 (5.9)	222 (0.6)
Influenza diagnosis position, n (%)
Primary diagnosis	4,653 (38.9)	9,812 (42.0)	34,072 (50.4)	2,533 (26.7)	5,394 (36.2)	14,756 (42.8)
Non-primary diagnosis	7,297 (61.1)	13,542 (58.0)	33,480 (49.6)	6,967 (73.3)	9,515 (63.8)	19,707 (57.2)
# of chronic conditions, n (%)
None	7,964 (66.6)	11,899 (51.0)	29,685 (43.9)	6,390 (67.3)	7,542 (50.6)	15,250 (44.3)
≥ 1	3,986 (33.4)	11,455 (49.0)	37,867 (56.1)	3,110 (32.7)	7,367 (49.4)	19,213 (55.7)
Chronic condition^b, n (%)
Diabetes without chronic complications	2,414 (20.2)	6,659 (28.5)	16,032 (23.7)	2,023 (21.3)	4,387 (29.4)	8,645 (25.1)
Diabetes with chronic complications	978 (8.2)	3,808 (16.3)	12,334 (18.3)	744 (7.8)	2,667 (17.9)	7,288 (21.1)
Dementia	17 (0.1)	317 (1.4)	11,870 (17.6)	10 (0.1)	189 (1.3)	4,729 (13.7)
Any malignancy^c	376 (3.1)	1,426 (6.1)	4,196 (6.2)	266 (2.8)	855 (5.7)	2,516 (7.3)
Metastatic solid tumor	110 (0.9)	454 (1.9)	1,029 (1.5)	83 (0.9)	314 (2.1)	673 (2.0)
Rheumatic disease	403 (3.4)	1,028 (4.4)	3,037 (4.5)	241 (2.5)	571 (3.8)	1,538 (4.5)
AIDS/HIV	195 (1.6)	292 (1.3)	64 (0.1)	102 (1.1)	171 (1.1)	65 (0.2)
Hemiplegia or paraplegia	183 (1.5)	327 (1.4)	535 (0.8)	158 (1.7)	217 (1.5)	308 (0.9)
Moderate or severe liver disease	123 (1.0)	326 (1.4)	332 (0.5)	131 (1.4)	208 (1.4)	182 (0.5)
Peptic ulcer disease	79 (0.7)	234 (1.0)	529 (0.8)	64 (0.7)	164 (1.1)	352 (1.0)
Hospital census region, n (%)
East North Central	1,875 (15.7)	4,175 (17.9)	12,380 (18.3)	1,349 (14.2)	2,578 (17.3)	6,239 (18.1)
East South Central	885 (7.4)	1,769 (7.6)	4,408 (6.5)	791 (8.3)	1,491 (10.0)	2,797 (8.1)
Middle Atlantic	1,676 (14.0)	3,344 (14.3)	10,298 (15.2)	1,457 (15.3)	1,984 (13.3)	4,694 (13.6)
Mountain	564 (4.7)	965 (4.1)	3,305 (4.9)	839 (8.8)	1,006 (6.7)	2,569 (7.5)
New England	253 (2.1)	462 (2.0)	1,628 (2.4)	257 (2.7)	430 (2.9)	1,162 (3.4)
Pacific	1,302 (10.9)	2,517 (10.8)	8,452 (12.5)	1,116 (11.7)	1,551 (10.4)	3,804 (11.0)
South Atlantic	3,108 (26.0)	5,718 (24.5)	15,424 (22.8)	2,142 (22.5)	3,438 (23.1)	8,025 (23.3)
West North Central	683 (5.7)	1,521 (6.5)	4,532 (6.7)	355 (3.7)	711 (4.8)	1,622 (4.7)
West South Central	1,604 (13.4)	2,883 (12.3)	7,125 (10.5)	1,194 (12.6)	1,720 (11.5)	3,551 (10.3)
Hospital urbanicity/population served, n (%)
Urban	10,650 (89.1)	20,245 (86.7)	58,588 (86.7)	8,419 (88.6)	12,697 (85.2)	29,251 (84.9)
Rural	1,300 (10.9)	3,109 (13.3)	8,964 (13.3)	1,081 (11.4)	2,212 (14.8)	5,212 (15.1)

Abbreviations. SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile; AIDS, acquired immune deficiency syndrome; HIV: human immunodeficiency virus.

^a A small number of patients had no sex reported in the 2017/18 (18–49: n = 3; 50–64: n = 5; ≥ 65: n = 6) and 2022/23 (18–49: n = 1) influenza seasons.

^b These comorbid conditions were present and listed as a “present on admission” (POA) diagnosis during the index hospitalization.

^c Any malignancy includes solid and hematologic malignancies, but excludes malignant neoplasm of the skin.

HCRU and costs

HCRU and costs during the index hospitalization are presented in Table 2 (Supplemental Data: Table 4). The majority of index hospitalizations were initiated in the ED (2017/18: n = 99,258, 96.5%; 2022/23: n = 57,365, 97.4%). Among index hospitalizations initiated in the ED, the mean cost of the initial ED visit among those who had this cost during the 2017/18 season was $469, $463, and $463 for patients 18–49, 50–64 and ≥ 65 years old, respectively. The mean initial ED visit costs were similar during the 2022/23 season among all three age groups. The median hospital LOS was similar for all patients during both seasons (approximately 3–4 days). Mean total combined costs for the index hospitalization and initial ED visit were lower among patients ≥ 65 years compared to adults in younger age groups for the 2017/18 season (18–49 years: $13,890; 50–64 years: $14,494; and ≥ 65 years: $12,556). Across all three age groups during the 2022/23 season, mean total index hospitalization and initial ED visit costs were lower than those of the corresponding 2017/18 season age groups, with the lowest costs occurring among the youngest age group (18–49 years: $11,384; 50–64 years: $12,896; and ≥ 65 years: $11,755). All-cause inpatient mortality increased with age, ranging from approximately 2.0% to 4.4% during both influenza seasons, with the highest mortality occurring among patients ≥ 65 years.

Table 2. Healthcare Outcomes, resource Utilization and costs during the index hospitalization.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
Emergency department (ED)-initiated index hospitalization^a
n (%)	11,489 (96.1)	22,440 (96.1)	65,329 (96.7)	9,273 (97.6)	14,562 (97.7)	33,530 (97.3)
Initial ED visit costs^b
Count of visits with cost data	10,382 (86.9)	20,397 (87.3)	59,977 (88.8)	8,212 (86.4)	13,060 (87.6)	30,228 (87.7)
Mean (SD)	$469 (411)	$463 (384)	$463 (405)	$465 (510)	$465 (454)	$459 (539)
Median	$390	$392	$391	$427	$434	$430
Q1; Q3	$294; $538	$299; $538	$295; $536	$90; $608	$188; $616	$190; $604
Index hospitalization length of stay (days)
Mean (SD)	4.93 (6.74)	5.51 (7.57)	5.39 (5.03)	4.90 (6.49)	5.64 (7.00)	5.66 (5.87)
Median (Q1, Q3)	3 (2, 5)	4 (2, 6)	4 (3, 7)	3 (2, 5)	4 (2, 6)	4 (2, 7)
Index hospitalization cost
Mean (SD)	$13,136 (31,630)	$13,770 (26,252)	$11,779 (16,613)	$10,650 (23,667)	$12,144 (25,024)	$10,978 (19,806)
Median	$5,831	$6,980	$7,393	$4,837	$5,946	$6,329
Q1; Q3	$3,180; $11,613	$3,846; $13,694	$4,102; $13,503	$0; $10,061	$1,700; $12,806	$1,831; $12,835
Total cost of index hospitalization and initial ED visit
Count of visits with cost data	11,950 (100.0)	23,354 (100.0)	67,552 (100.0)	9,500 (100.0)	14,909 (100.0)	34,463 (100.0)
Mean (SD)	$13,890 (31,703)	$14,494 (26,341)	$12,556 (16,738)	$11,384 (23,846)	$12,896 (25,181)	$11,755 (20,061)
Median	$6,634	$7,727	$8,190	$5,765	$6,842	$7,298
Q1; Q3	$3,896; $12,545	$4,551; $14,584	$4,868; $14,365	$0; $11,090	$2,527; $13,785	$2,756; $13,882
Inpatient mortality
n (%)	238 (2.0)	766 (3.3)	2,873 (4.3)	205 (2.2)	507 (3.4)	1,515 (4.4)

Abbreviations. SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile.

^a ED-initiated hospitalizations were defined as hospitalizations with an admission type recorded as “Emergency” or “Urgent”.

^b The initial ED visit was defined as the ED visit associated with the index hospitalization among patients with ED-initiated hospitalizations. Initial ED visit costs were only determined for patients with non-missing cost data for the ED visit.

The proportion of patients with MV usage and ICU admission during the index hospitalization, as well as the cost of index hospitalizations (including the initial ED visit) stratified by MV usage and ICU admission, are reported in Table 3 (Supplemental Data: Table 5). Most influenza-related hospitalizations were not associated with MV usage or ICU admissions; the proportions of such hospitalizations were slightly higher among patients ≥ 65 years (2017/18: 77.7%; 2022/23: 74.8%) compared to patients 18–49 years (2017/18: 73.4%; 2022/23: 71.0%) and 50–64 years (2017/18: 72.3%; 2022/23: 68.7%). Compared to the other two age groups, patients 18–49 years had the highest proportion of index hospitalizations with ICU admission only (2017/18: 13.2%; 2022/23: 14.9%), while patients 50–64 years had highest proportions of MV usage only (2017/18: 3.9%; 2022/23: 6.4%), as well as both MV usage and ICU admission during the index hospitalization (2017/18: 12.5%; 2022/23: 12.9%). The mean cost of hospitalizations without any MV usage or ICU admission were 24.9% to 44.2% lower across age groups for both seasons when compared to the mean influenza-related hospitalization costs reported in Table 2. MV usage and/or ICU admission was associated with substantially higher hospitalization costs compared to patients with no MV usage or ICU admission during their hospitalization. In particular, patients who had both MV usage and ICU admission during their hospitalization had mean costs that were 3.6 to 7.1 times the costs of patients who required no such medical intervention.

Table 3. Mechanical Ventilation (MV) and Intensive care Unit (ICU) usage and costs during index hospitalization.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
No MV or ICU admission during index hospitalization
n (%)	8,770 (73.4)	16,874 (72.3)	52,507 (77.7)	6,749 (71.0)	10,244 (68.7)	25,789 (74.8)
Total cost of index hospitalization and initial ED visit for patients with no MV or ICU admission
Count of visits with cost data	8,770 (100.0)	16,874 (100.0)	52,507 (100.0)	6,749 (100.0)	10,244 (100.0)	25,789 (100.0)
Mean (SD)	$7,749 (8,793)	$8,612 (8,846)	$9,236 (8,446)	$7,129 (10,872)	$8,040 (10,472)	$8,833 (12,059)
Median	$5,571	$6,405	$7,169	$5,036	$5,794	$6,538
Q1; Q3	$3,507; $8,975	$4,081; $10,397	$4,490; $11,511	$0; $8,863	$2,351; $10,384	$2,746; $11,420
MV only during index hospitalization
n (%)	306 (2.6)	914 (3.9)	2,576 (3.8)	383 (4.0)	958 (6.4)	2,037 (5.9)
Total cost of index hospitalization and initial ED visit for patients with MV only
Count of visits with cost data	306 (100.0)	914 (100.0)	2,576 (100.0)	383 (100.0)	958 (100.0)	2,037 (100.0)
Mean (SD)	$15,684 (22,506)	$17,440 (20,598)	$16,285 (15,031)	$13,655 (32,490)	$12,463 (18,829)	$13,351 (17,624)
Median	$9,024	$11,482	$12,301	$7,809	$8,681	$9,459
Q1; Q3	$5,797; $15,796	$6,760; $19,474	$7,244; $20,460	$1,893; $13,213	$3,012; $15,442	$3,167; $18,089
ICU admission only during index hospitalization
n (%)	1,576 (13.2)	2,656 (11.4)	7,027 (10.4)	1,418 (14.9)	1,787 (12.0)	3,654 (10.6)
Total cost of index hospitalization and initial ED visit for patients with ICU admission only
Count of visits with cost data	1,576 (100.0)	2,656 (100.0)	7,027 (100.0)	1,418 (100.0)	1,787 (100.0)	3,654 (100.0)
Mean (SD)	$13,754 (14,790)	$16,053 (17,879)	$16,459 (16,754)	$11,074 (16,388)	$14,032 (19,040)	$15,030 (18,299)
Median	$9,644	$11,448	$12,685	$7,315	$9,335	$10,486
Q1; Q3	$5,760; $16,971	$6,674; $19,903	$7,094; $21,315	$0; $13,274	$2,633; $17,851	$2,575; $19,834
Both MV AND ICU admission during index hospitalization
n (%)	1,298 (10.9)	2,910 (12.5)	5,442 (8.1)	950 (10.0)	1,920 (12.9)	2,983 (8.7)
Total cost of index hospitalization and initial ED visit for patients with both MV AND ICU admission
Count of visits with cost data	1,298 (100.0)	2,910 (100.0)	5,442 (100.0)	950 (100.0)	1,920 (100.0)	2,983 (100.0)
Mean (SD)	$54,890 (79,879)	$46,036 (58,907)	$37,499 (39,408)	$40,984 (54,765)	$37,811 (55,342)	$31,716 (47,636)
Median	$31,410	$29,687	$27,393	$23,024	$21,586	$19,661
Q1; Q3	$14,389; $67,439	$15,112; $56,355	$15,045; $48,437	$5,355; $56,040	$6,699; $47,798	$3,390; $41,309

Abbreviations. ED, emergency department; MV, mechanical ventilation; ICU, intensive care unit; SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile.

Influenza-related hospital-based standalone outpatient and ED visits prior to the index hospitalization and their associated costs are reported in Table 4 (Supplemental Data: Table 6). Few patients had influenza-related hospital-based outpatient visits during the 7-day period prior to the index hospitalization. The proportion of patients with standalone ED visits during the 7-day period prior to the index hospitalization was lowest among patients ≥ 65 years (2017/18: 1.0%; 2022/23: 1.7%) and highest among patients 18–49 years (2017/18: 3.1%; 2022/23: 3.3%). Mean standalone ED visit costs were highest among patients ≥ 65 years (2017/18: $1,184; 2022/23: $1,346) and lowest among patients 18–49 years (2017/18: $752; 2022/23: $740).

Table 4. Influenza-related standalone outpatient and standalone Emergency department (ED) visits in the 7-day period prior to the index Hospitalization^a.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
Influenza-related outpatient visit
n (%)	78 (0.7)	81 (0.3)	136 (0.2)	37 (0.4)	22 (0.1)	59 (0.2)
Total cost of influenza-related outpatient visits
Count of visits with cost data	78 (100.0)	81 (100.0)	136 (100.0)	37 (100.0)	22 (100.0)	59 (100.0)
Mean (SD)	$784 (814)	$612 (662)	$865 (1,240)	$1,650 (2,583)	$1,147 (1,559)	$1,328 (2,855)
Median	$617	$458	$464	$715	$652	$791
Q1; Q3	$324; $967	$80; $853	$111; $998	$270; $1,419	$93; $1,378	$61; $1,463
Influenza-related ED visit costs
n (%)	374 (3.1)	411 (1.8)	663 (1.0)	310 (3.3)	385 (2.6)	599 (1.7)
Total cost of influenza-related ED visits
Count of visits with cost data	374 (100.0)	411 (100.0)	663 (100.0)	310 (100.0)	385 (100.0)	599 (100.0)
Mean (SD)	$752 (820)	$903 (1,098)	$1,184 (1,534)	$740 (997)	$1,217 (2,189)	$1,346 (3,264)
Median	$573	$661	$739	$530	$755	$721
Q1; Q3	$366; $881	$406; $975	$460; $1,194	$0; $988	$284; $1,248	$240; $1,187

Abbreviations. ED, emergency department; SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile.

^aThese standalone visits are for care received at outpatient and ED facilities that are affiliated with a hospital system in the PINC AI Healthcare Database. Influenza-related visits were identified by an ICD-10 diagnosis code for influenza recorded at any position during the encounter.

Influenza-related secondary complications

Influenza-related secondary complications present during the index hospitalization are reported in Table 5 (Supplemental Data: Table 7). Approximately 78% of hospitalized patients 18–49 years, 89% of patients 50–64 years and 91% of patients ≥ 65 years had at least one secondary complication presenting during their influenza-related index hospitalization. Pulmonary conditions were the most common secondary complication across all three age groups and for both seasons, with pneumonia occurring among ∼40% of patients and asthma exacerbation present in 38.6% of patients 50–64 years, 31.1% of patients ≥ 65 years and 19.6% of patients 18–49 years. Cardiovascular complications were present in approximately 60% of patients ≥ 65 years and more than 40% of patients 50–64 years of age. Metabolic failure was also one of the top secondary complications, presenting in 20.0% to 32.1% of the influenza hospitalizations. Shock syndrome developed in 21.3% to 26.0% of patients during their index hospitalization. Neurologic and endocrine conditions also occurred as secondary complications among influenza hospitalizations with varying proportions.

Table 5. Influenza-related secondary complications during the index hospitalization.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
Influenza-related secondary complication
n (%)	9,403 (78.7)	20,612 (88.3)	60,921 (90.2)	7,272 (76.5)	13,330 (89.4)	31,580 (91.6)
Type of secondary complication, n (%)^a
Pulmonary condition	6,416 (53.7)	15,432 (66.1)	41,007 (60.7)	4,778 (50.3)	9,728 (65.2)	21,123 (61.3)
Pneumonia	4,592 (38.4)	9,451 (40.5)	27,345 (40.5)	3,401 (35.8)	6,135 (41.1)	14,383 (41.7)
Asthma exacerbation	2,413 (20.2)	9,074 (38.9)	20,917 (31.0)	1,785 (18.8)	5,712 (38.3)	10,849 (31.5)
Acute exacerbation of COPD	1,116 (9.3)	7,910 (33.9)	19,120 (28.3)	645 (6.8)	5,027 (33.7)	9,869 (28.6)
Pulmonary collapse	1,064 (8.9)	2,226 (9.5)	6,168 (9.1)	861 (9.1)	1,304 (8.7)	2,882 (8.4)
Metabolic failure	2,393 (20.0)	6,078 (26.0)	19,764 (29.3)	2,030 (21.4)	4,276 (28.7)	11,055 (32.1)
Acute renal failure	2,358 (19.7)	6,035 (25.8)	19,704 (29.2)	2,009 (21.1)	4,251 (28.5)	11,020 (32.0)
Cardiovascular condition	2,444 (20.5)	9,509 (40.7)	39,963 (59.2)	2,009 (21.1)	6,784 (45.5)	21,431 (62.2)
Malignant hypertension	1,639 (13.7)	6,414 (27.5)	28,425 (42.1)	1,325 (13.9)	4,824 (32.4)	15,887 (46.1)
Carditis	655 (5.5)	4,432 (19.0)	21,620 (32.0)	518 (5.5)	2,941 (19.7)	11,012 (32.0)
Acute heart failure	510 (4.3)	1,915 (8.2)	8,662 (12.8)	488 (5.1)	1,681 (11.3)	4,912 (14.3)
Acute myocardial infarction	197 (1.6)	1,049 (4.5)	4,279 (6.3)	322 (3.4)	1,102 (7.4)	3,315 (9.6)
Acute ischemic heart disease	162 (1.4)	540 (2.3)	2,712 (4.0)	181 (1.9)	552 (3.7)	1,770 (5.1)
Shock/sepsis	3,112 (26.0)	5,816 (24.9)	14,389 (21.3)	2,307 (24.3)	3,604 (24.2)	7,515 (21.8)
Any shock syndrome	3,109 (26.0)	5,815 (24.9)	14,376 (21.3)	2,306 (24.3)	3,601 (24.2)	7,512 (21.8)
Neurologic condition	1,104 (9.2)	2,961 (12.7)	12,520 (18.5)	921 (9.7)	1,895 (12.7)	6,415 (18.6)
Encephalopathy	596 (5.0)	1,923 (8.2)	8,627 (12.8)	544 (5.7)	1,460 (9.8)	5,274 (15.3)
Altered mental status	204 (1.7)	612 (2.6)	3,528 (5.2)	187 (2.0)	383 (2.6)	1,637 (4.8)
Endocrine condition	1,676 (14.0)	3,413 (14.6)	6,289 (9.3)	1,573 (16.6)	2,578 (17.3)	4,456 (12.9)
Diabetic ketoacidosis	1,546 (12.9)	3,197 (13.7)	5,880 (8.7)	1,441 (15.2)	2,404 (16.1)	4,142 (12.0)
Diabetic coma	1,044 (8.7)	2,922 (12.5)	5,819 (8.6)	867 (9.1)	2,170 (14.6)	4,049 (11.7)
Diabetic hyperosmolarity	949 (7.9)	2,763 (11.8)	5,453 (8.1)	763 (8.0)	2,043 (13.7)	3,789 (11.0)

^aPercentage of age group with a primary or secondary diagnosis for the secondary complication during the index hospitalization.

Index hospitalization and initial ED visit costs, MV usage and ICU admission among patients with and without a secondary complication during the index hospitalization are reported in Table 6 (Supplemental Data: Table 8). The proportion of influenza hospitalizations with MV usage and/or ICU admission was often several-fold higher when a secondary complication was present. The combined index hospitalization and initial ED visit costs were 1.9 to 2.2 times higher among patients with a secondary complication during the 2017/18 season and 1.6 to 2 times higher during the 2022/23 season.

Table 6. Mechanical Ventilation (MV) and Intensive care Unit (ICU) usage and costs during index hospitalization among patients with and without secondary complications.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 11,950	N = 23,354	N = 67,552	N = 9,500	N = 14,909	N = 34,463
Patients without secondary complications
n (%)	2,547 (21.3)	2,742 (11.7)	6,631 (9.8)	2,228 (23.5)	1,579 (10.6)	2,883 (8.4)
Total cost of index hospitalization and initial ED visit
Mean (SD)	$7,174 (7,760)	$7,533 (9,010)	$7,230 (6,558)	$6,874 (12,168)	$6,879 (8,926)	$7,560 (22,284)
Median	$5,099	$5,418	$5,620	$4,489	$4,871	$5,326
Q1; Q3	$3,268; $8,326	$3,348; $8,729	$3,602; $8,886	$0; $8,301	$1,396; $9,091	$2,098; $9,405
No MV or ICU admission during index hospitalization
n (%)	2,335 (91.7)	2,493 (90.9)	6,192 (93.4)	1,990 (89.3)	1,393 (88.2)	2,622 (91.0)
Total cost of index hospitalization and initial ED visit for patients with no MV or ICU admission
Count of visits with cost data	2,335 (100.0)	2,493 (100.0)	6,192 (100.0)	1,990 (100.0)	1,393 (100.0)	2,622 (100.0)
Mean (SD)	$6,769 (7,290)	$7,004 (7,531)	$6,938 (5,889)	$6,413 (9,779)	$6,434 (7,826)	$6,943 (8,951)
Median	$4,945	$5,305	$5,507	$4,325	$4,769	$5,225
Q1; Q3	$3,198; $7,933	$3,311; $8,315	$3,562; $8,614	$0; $8,020	$1,441; $8,804	$2,140; $9,134
MV only during index hospitalization
n (%)	26 (1.0)	34 (1.2)	54 (0.8)	42 (1.9)	37 (2.3)	44 (1.5)
Total cost of index hospitalization and initial ED visit for patients with MV only
Count of visits with cost data	26 (100.0)	34 (100.0)	54 (100.0)	42 (100.0)	37 (100.0)	44 (100.0)
Mean (SD)	$8,354 (10,246)	$8,483 (6,523)	$10,288 (9,950)	$7,226 (6,723)	$9,838 (10,766)	$8,004 (7,821)
Median	$5,248	$7,336	$7,301	$6,115	$7,210	$6,407
Q1; Q3	$4,133; $8,740	$4,278; $11,624	$4,158; $11,434	$3,300; $11,255	$3,054; $12,993	$935; $11,644
ICU admission only during index hospitalization
n (%)	148 (5.8)	179 (6.5)	358 (5.4)	163 (7.3)	122 (7.7)	184 (6.4)
Total cost of index hospitalization and initial ED visit for patients with ICU admission only
Count of visits with cost data	148 (100.0)	179 (100.0)	358 (100.0)	163 (100.0)	122 (100.0)	184 (100.0)
Mean (SD)	$10,925 (10,291)	$11,020 (17,523)	$10,684 (11,722)	$9,079 (10,642)	$8,165 (11,439)	$8,802 (9,161)
Median	$7,351	$6,313	$7,717	$6,115	$4,943	$6,626
Q1; Q3	$4,044; $14,062	$3,683; $12,804	$4,321; $13,328	$0; $12,840	$0; $10,072	$2,041; $13,101
Both MV AND ICU admission during index hospitalization
n (%)	38 (1.5)	36 (1.3)	27 (0.4)	33 (1.5)	27 (1.7)	33 (1.1)
Total cost of index hospitalization and initial ED visit for patients with both MV AND ICU admission
Count of visits with cost data	38 (100.0)	36 (100.0)	27 (100.0)	33 (100.0)	27 (100.0)	33 (100.0)
Mean (SD)	$15,740 (11,900)	$24,892 (17,371)	$19,091 (15,192)	$22,556 (58,313)	$19,360 (24,074)	$48,008 (188,987)
Median	$12,633	$22,217	$16,929	$7,873	$14,009	$11,450
Q1; Q3	$7,985; $21,036	$12,107; $35,575	$9,258; $26,397	$0; $18,044	$1,818; $26,776	$0; $27,951
Patients with any secondary complication
n (%)	9,403 (78.7)	20,612 (88.3)	60,921 (90.2)	7,272 (76.5)	13,330 (89.4)	31,580 (91.6)
Total cost of index hospitalization and initial ED visit
Mean (SD)	$15,709 (35,292)	$15,420 (27,713)	$13,135 (17,394)	$12,766 (26,257)	$13,609 (26,362)	$12,138 (19,802)
Median	$7,233	$8,186	$8,577	$6,180	$7,176	$7,515
Q1; Q3	$4,171; $14,088	$4,793; $15,636	$5,099; $15,037	$1,773; $12,189	$2,737; $14,444	$2,883; $14,345
No MV or ICU admission during index hospitalization
n (%)	6,435 (68.4)	14,381 (69.8)	46,315 (76.0)	4,759 (65.4)	8,851 (66.4)	23,167 (73.4)
Total cost of index hospitalization and initial ED visit for patients with no MV or ICU admission
Count of visits with cost data	6,435 (100.0)	14,381 (100.0)	46,315 (100.0)	4,759 (100.0)	8,851 (100.0)	23,167 (100.0)
Mean (SD)	$8,104 (9,253)	$8,891 (9,025)	$9,543 (8,686)	$7,428 (11,286)	$8,293 (10,808)	$9,046 (12,344)
Median	$5,804	$6,624	$7,444	$5,232	$5,956	$6,702
Q1; Q3	$3,638; $9,376	$4,238; $10,769	$4,680; $11,879	$1,873; $9,305	$2,547; $10,616	$2,870; $11,688
MV only during index hospitalization
n (%)	280 (3.0)	880 (4.3)	2,522 (4.1)	341 (4.7)	921 (6.9)	1,993 (6.3)
Total cost of index hospitalization and initial ED visit for patients with MV only
Count of visits with cost data	280 (100.0)	880 (100.0)	2,522 (100.0)	341 (100.0)	921 (100.0)	1,993 (100.0)
Mean (SD)	$16,364 (23,213)	$17,786 (20,878)	$16,413 (15,096)	$14,447 (34,276)	$12,569 (19,078)	$13,470 (17,762)
Median	$9,485	$11,665	$12,410	$8,169	$8,748	$9,533
Q1; Q3	$6,138; $16,640	$6,945; $19,857	$7,352; $20,616	$1,893; $15,200	$3,012; $15,523	$3,234; $18,158
ICU admission only during index hospitalization
n (%)	1,428 (15.2)	2,477 (12.0)	6,669 (11.0)	1,255 (17.3)	1,665 (12.5)	3,470 (11.0)
Total cost of index hospitalization and initial ED visit for patients with ICU admission only
Count of visits with cost data	1,428 (100.0)	2,477 (100.0)	6,669 (100.0)	1,255 (100.0)	1,665 (100.0)	3,470 (100.0)
Mean (SD)	$14,047 (15,152)	$16,417 (17,853)	$16,769 (16,927)	$11,333 (16,978)	$14,462 (19,413)	$15,360 (18,602)
Median	$9,950	$11,801	$12,962	$7,443	$9,845	$10,728
Q1; Q3	$5,925; $17,307	$6,964; $20,478	$7,347; $21,717	$0; $13,298	$3,292; $18,350	$2,702; $20,368
Both MV AND ICU admission during index hospitalization
n (%)	1,260 (13.4)	2,874 (13.9)	5,415 (8.9)	917 (12.6)	1,893 (14.2)	2,950 (9.3)
Total cost of index hospitalization and initial ED visit for patients with both MV AND ICU admission
Count of visits with cost data	1,260 (100.0)	2,874 (100.0)	5,415 (100.0)	917 (100.0)	1,893 (100.0)	2,950 (100.0)
Mean (SD)	$56,070 (80,755)	$46,301 (59,196)	$37,591 (39,471)	$41,648 (54,551)	$38,074 (55,620)	$31,534 (43,635)
Median	$32,598	$29,910	$27,457	$24,252	$21,640	$19,720
Q1; Q3	$14,946; $69,047	$15,172; $56,797	$15,079; $48,531	$5,657; $56,869	$6,775; $48,309	$3,744; $41,627

Abbreviations. MV, mechanical ventilation; ICU, intensive care unit; SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile.

Sensitivity analysis

During the 2017/18 influenza season, 38.9% of patients 18–49 years, 42.0% of patients 50–64 years, and 50.4% of patients ≥ 65 years had an index hospitalization with a primary diagnosis of influenza. A lower proportion of patients were hospitalized with a primary diagnosis of influenza during the 2022–2023 influenza season (18–49 years: 26.7%; 50–64 years: 36.2%; and ≥ 65 years: 42.8%). A detailed summary of the sensitivity analysis patient attrition is included in the Supplemental Data: Table 9. HCRU and costs for index hospitalizations with a primary diagnosis of influenza are reported in Table 7 (Supplemental Data: Table 10). Although the initial ED visit costs were similar, hospitalizations with a primary diagnosis of influenza had a slightly shorter LOS and substantially lower mean costs ranging from $6,307 to $8,464 in comparison to hospitalizations with an influenza diagnosis in any position, which had costs ranging from $10,650 to $13,770 depending on age group and influenza season. A primary diagnosis of influenza was associated with 2.3- to 6.6-fold lower inpatient mortality. In comparison to hospitalizations with an influenza diagnosis in any position, primary diagnosis hospitalizations had similar proportions with MV usage only, 1.5- to 1.7-fold lower proportions with ICU admission only and 2.0- to 2.8-fold lower proportions with both MV usage and ICU admission; the associated hospitalization costs, with or without the initial ED visit, were lower among hospitalizations with a primary diagnosis of influenza.

Table 7. Healthcare Outcomes, resource Utilization and costs during index hospitalizations with a primary diagnosis of influenza.

Influenza season	2017/18			2022/23
Age group	18–49	50–64	≥ 65	18–49	50–64	≥ 65
	N = 4,653	N = 9,812	N = 34,072	N = 2,533	N = 5,394	N = 14,756
ED-initiated index hospitalization
n (%)	4,483 (96.3)	9,436 (96.2)	33,086 (97.1)	2,494 (98.5)	5,288 (98.0)	14,432 (97.8)
Initial ED visit costs
Count of visits with cost data	4,077 (87.6)	8,694 (88.6)	30,265 (88.8)	2,226 (87.9)	4,802 (89.0)	12,956 (87.8)
Mean (SD)	$450 (335)	$443 (326)	$445 (356)	$449 (423)	$461 (410)	$460 (533)
Median	$389	$384	$384	$435	$435	$432
Q1; Q3	$295; $522	$297; $520	$292; $518	$134; $598	$217; $609	$211; $603
Index hospitalization length of stay (days)
Mean (SD)	3.51 (3.79)	4.09 (4.24)	4.45 (3.97)	3.53 (3.98)	4.20 (4.45)	4.60 (4.21)
Median (Q1, Q3)	3 (2, 4)	3 (2, 5)	4 (2, 5)	3 (2, 4)	3 (2, 5)	3 (2, 6)
Index hospitalization cost
Mean (SD)	$7,419 (13,354)	$8,372 (13,412)	$8,464 (9,781)	$6,307 (11,124)	$7,997 (13,975)	$8,177 (15,493)
Median	$4,636	$5,500	$6,101	$3,789	$4,918	$5,406
Q1; Q3	$2,699; $7,827	$3,214; $9,341	$3,566; $10,226	$16; $7,406	$1,760; $9,455	$1,811; $10,068
Total cost of index hospitalization and initial ED visit
Count of visits with cost data	4,653 (100.0)	9,812 (100.0)	34,072 (100.0)	2,533 (100.0)	5,394 (100.0)	14,756 (100.0)
Mean (SD)	$8,180 (13,466)	$9,102 (13,495)	$9,253 (9,923)	$7,069 (11,339)	$8,758 (14,131)	$8,988 (15,719)
Median	$5,390	$6,217	$6,904	$4,776	$5,808	$6,379
Q1; Q3	$3,410; $8,624	$3,911; $10,156	$4,309; $11,158	$41; $8,458	$2,575; $10,489	$2,753; $11,101
No MV or ICU admission during index hospitalization
n (%)	4,000 (86.0)	8,319 (84.8)	29,930 (87.8)	2,026 (80.0)	4,225 (78.3)	12,294 (83.3)
Total cost of index hospitalization and initial ED visit for patients with no MV or ICU
Count of visits with cost data	4,000 (100.0)	8,319 (100.0)	29,930 (100.0)	2,026 (100.0)	4,225 (100.0)	12,294 (100.0)
Mean (SD)	$6,413 (6,011)	$7,244 (6,131)	$8,067 (6,759)	$5,392 (5,986)	$6,864 (7,323)	$7,681 (9,815)
Median	$5,014	$5,777	$6,520	$4,399	$5,372	$6,034
Q1; Q3	$3,270; $7,718	$3,769; $9,004	$4,169; $10,130	$862; $7,393	$2,582; $9,184	$2,751; $10,106
MV only during index hospitalization
n (%)	116 (2.5)	345 (3.5)	965 (2.8)	129 (5.1)	365 (6.8)	859 (5.8)
Total cost of index hospitalization and initial ED visit for patients with MV only
Count of visits with cost data	116 (100.0)	345 (100.0)	965 (100.0)	129 (100.0)	365 (100.0)	859 (100.0)
Mean (SD)	$11,899 (10,826)	$14,061 (19,328)	$14,709 (13,580)	$10,152 (11,721)	$10,825 (16,879)	$11,401 (11,895)
Median	$8,838	$9,972	$11,370	$7,620	$8,088	$9,001
Q1; Q3	$5,371; $15,213	$6,252; $15,354	$6,645; $18,295	$3,553; $12,216	$3,379; $13,598	$3,158; $15,689
ICU admission only during index hospitalization
n (%)	356 (7.7)	693 (7.1)	2,145 (6.3)	248 (9.8)	413 (7.7)	992 (6.7)
Total cost of index hospitalization and initial ED visit for patients with ICU admission only
Count of visits with cost data	356 (100.0)	693 (100.0)	2,145 (100.0)	248 (100.0)	413 (100.0)	992 (100.0)
Mean (SD)	$11,367 (10,263)	$11,914 (11,516)	$13,110 (12,292)	$8,901 (11,071)	$8,601 (9,718)	$11,215 (14,207)
Median	$8,292	$8,786	$10,248	$7,032	$6,269	$7,979
Q1; Q3	$4,729; $14,798	$4,959; $15,344	$5,605; $16,794	$0; $11,599	$0; $12,004	$2,118; $14,731
Both MV AND ICU admission during index hospitalization
n (%)	181 (3.9)	455 (4.6)	1,032 (3.0)	130 (5.1)	391 (7.2)	611 (4.1)
Total cost of index hospitalization and initial ED visit for patients with MV AND ICU admissions
Count of visits with cost data	181 (100.0)	455 (100.0)	1,032 (100.0)	130 (100.0)	391 (100.0)	611 (100.0)
Mean (SD)	$38,196 (50,793)	$34,681 (44,693)	$29,973 (30,002)	$26,427 (33,818)	$27,284 (37,692)	$27,938 (55,487)
Median	$22,838	$22,016	$22,079	$13,738	$16,627	$17,615
Q1; Q3	$11,684; $42,885	$11,114; $41,953	$13,518; $37,989	$4,240; $34,166	$6,453; $34,403	$4,480; $32,248
Inpatient mortality
n (%)	14 (0.3)	81 (0.8)	566 (1.7)	11 (0.4)	61 (1.1)	280 (1.9)

Abbreviations. ED, emergency department; MV, mechanical ventilation; ICU, intensive care unit; SD, standard deviation; Q1, 1st quartile; Q3, 3rd quartile.

Discussion

In this retrospective analysis of a large, geographically diverse US hospital-based, all-payer database, the results indicated that substantial economic burden is associated with influenza-related hospitalization for all three age groups. Although there were expected differences between influenza seasons regarding the circulating strains, as well as between age groups regarding underlying chronic conditions and insurance coverage type, patient and hospital characteristics were largely similar across the age groups for the 2017/18 and 2022/23 seasons. HCRU during the index hospitalization was consistently high with over 96% of patients initiating influenza hospitalization in the ED, around 20–30% of patients requiring MV usage or ICU admission while hospitalized, and a median hospital LOS of 3 to 4 days. Influenza-related hospitalization and initial ED visit mean costs ranged from $11,384 to $14,494 depending on the specific season and age group.

The results of the study are consistent with previous retrospective database studies that estimated the HCRU and costs of influenza-related hospitalizations.^9,12,17,18 Using the 2000–2009 MarketScan database, one study estimated that average influenza-related inpatient medical costs during a 21-day follow-up period after the index diagnosis ranged from $8,208 to $9,556 (2010 US dollars [USD]) depending on the influenza season.¹⁸ A study using Veterans Affairs (VA) electronic medical records between 2010 and 2014 suggested influenza hospitalization costs of $13,673 to $14,849 for patients at high-risk for influenza-related complications, and $10,872 to $14,723 for low-risk patients.¹² Using Nationwide Readmission Database data from 2013 to 2014, median influenza-related hospitalization costs were $5,473.20 to $6,220.30 for adults ≥ 20 years.¹⁷ Another study using the 2015 National Inpatient Sample estimated influenza-related hospitalization costs of $7,848.50 to $12,102.27 (2015 USD). The hospital LOS from the current study was also similar to those reported in previous studies.^12,17

National influenza-related hospitalization costs could be estimated using an approach similar to that of previous literature,^9,12 by multiplying the CDC estimated age-specific annual number of influenza hospitalizations^1,25 by the age-specific mean cost of influenza hospitalizations from the current study. For the 2017/18 season, the estimated national expenditure of influenza-related hospitalizations was $8.7 billion (95% uncertainty level (UI): $6.2 billion to $15.0 billion), with $1.1 billion for patients 18–49 years (95% UI: $0.8 billion to $1.8 billion), $1.8 billion for patients 50–64 years (95% UI: $1.4 billion to $2.5 billion), and $5.9 billion for patients ≥ 65 years (95% UI: $4.0 billion to $10.7 billion). Although the influenza burden was lower for the other seasons, the national expenditure was estimated to be $3.8 billion to $6.3 billion for hospitalization among adults over 18 years old. These estimates are higher than what has been reported in the literature, as the current study focused on a more recent time period.⁹ In addition, the methodology for estimating influenza-related hospitalization cases adopted by CDC has changed since 2010.²⁶

The study results indicated variations in the cost of index hospitalizations between age groups, with the lowest mean hospitalization costs typically occurring among patients ≥ 65 years. This finding is consistent with previously published data.^9,11,17,27 A previous study estimated the economic burden of influenza hospitalization in the US and reported costs of $11,908 for patients 18–44 years, $12,102 for patients 45–64 years, $8,330 for patients 65–84 years and $7,849 for patients ≥ 85 years using the National Inpatient Sample (2015 USD).⁹ Another study using the 2013–2014 National Readmission Database estimated lower influenza hospitalization costs among patients 65–84 years compared to patients 20–64 years.¹⁷ Two other studies examined the cost effectiveness of influenza vaccination in the US and estimated hospitalization costs using the MarketScan database. Both studies found similar results, which suggested lower influenza-related inpatient costs in the older age group compared to their younger counterparts.^11,27 The lower hospitalization costs among patients ≥ 65 years may be attributable in part to the high influenza vaccination rates among this age group. Studies evaluating influenza vaccination status among patients hospitalized with influenza suggest that vaccination may attenuate disease severity, reducing the risk of ICU admission and death.^28–31 Vaccination coverage among patients ≥ 65 years increased from 59.6% during the 2017/18 influenza season to 69.7% during the 2022/23 influenza season.³² In comparison, only 26.9% and 35.2% of patients 18–49 years and 39.7% and 50.1% of patients 50–64 years were vaccinated during the 2017/18 and 2022/23 influenza seasons, respectively. Although vaccination status was not available in this study, the results suggest that disease severity during influenza hospitalization may be attenuated among patients ≥ 65 years. Specifically, patients ≥ 65 years experienced lower proportions of influenza-related hospitalization with MV usage and/or ICU admission compared to younger adults, despite having a greater comorbidity burden and presenting with more secondary complications during the index hospitalization. The decreased use of costly healthcare resources such as MV and ICU admission may contribute to lowering hospitalization costs among patients ≥ 65 years.

In the current study, the greatest proportions of hospitalizations with either MV usage alone, or both MV usage and ICU admission were among patients 50–64 years. The cost for index hospitalization was also the greatest among this age group for both influenza seasons. This is also consistent with results from previous literature. One study evaluated the association between influenza virus type and the severity of influenza hospitalization while adjusting for confounding factors such as age.³³ The results indicated that MV/ECMO use and ICU admission were more likely among patients 50–64 years during influenza-related hospitalizations, compared to other age groups. Another study found that MV and ECMO usage during influenza-related hospitalizations was higher among adults 20–64 years compared to those 65–79 years, even after adjusting for confounding factors.¹⁷ From a public health perspective, these findings highlight the potential benefit of influenza vaccination among younger adults to further reduce the disease severity and economic burden of influenza infection among this population.

The results of the study also indicated that hospitalization costs during the 2022/23 influenza season tended to be lower than those of the 2017/18 influenza season. In contrast, both index hospitalization LOS and inpatient mortality were similar between the two influenza seasons. The lower hospitalization costs during the 2022/23 season may be attributed to changes in population composition in terms of age, the presence of chronic conditions and insurance coverage. The release of updated clinical practice guidelines from the Infectious Diseases Society of America and the approval of a new antiviral drug, baloxavir marboxil, at the end of 2018 may have impacted medical treatment practices, resulting in decreased costs over time.^34,35 In addition to promoting the adoption of better prevention and hygiene habits such as hand washing and the use of masks to reduce the risk of infection, the COVID-19 pandemic may have made people more aware of potential infection and the benefit of early testing and treatment of influenza and other viral infections. Early detection of influenza infections and more timely treatment may result in lower HCRU and costs. A combination of these factors may contribute to the lower hospitalization costs during the 2022/23 season. Further monitoring and exploration of the burden of influenza in the post-COVID era is warranted.

Our results indicated that 76.5% to 91.6% of patients had a presenting secondary complication during their index hospitalization, which was slightly higher compared to what has been reported in the past literature. One study reported that 64.4% of adults hospitalized with seasonal influenza had a secondary complication using data from the 2005–2010 Emerging Infections Program (EIP) Influenza Surveillance Network.⁶ Another study found that 64.6% of adult patients were diagnosed with a secondary complication within the 12-month period following an influenza diagnosis based on data from the 1998–2009 US managed care LifeLink database.¹³ The difference in the study results may be due in part to differences between the ICD-9 and ICD-10 diagnostic coding sets as well as the specific conditions included as secondary complications in the studies. Our study also revealed that HCRU and costs were higher among patients presenting with a secondary complication during their index hospitalization. One study reported that the presence of a secondary complication during the 12-month period following an influenza diagnosis resulted in risk-adjusted average hospitalization costs that were 1.4 times higher for patients 18–49 years, 2.0 times higher for patients 50–64 years and 3.9 times higher for patients ≥ 65 years.¹³ Other studies have found that patients with chronic conditions have higher hospitalization costs and increased HCRU in comparison to patients with no at-risk conditions.^5,10–12 As underlying conditions are associated with an increased risk of influenza-related secondary complications, our results are consistent with past literature.

A sensitivity analysis that was performed to evaluate HCRU and costs among patients hospitalized with a primary diagnosis of influenza indicated that there was lower HCRU and inpatient mortality across all age groups, as well as lower index hospitalization costs in comparison to the results of the main analyses. The reduced HCRU and costs that occurred in the sensitivity analysis could potentially be attributable to medical coding practices. Patients who are admitted to the hospital with influenza as well as another medical condition will likely have the other medical condition recorded as the primary diagnosis if the condition is more acute or of greater severity than a viral infection. Accordingly, the difference in cost burden between the main analyses and the sensitivity analysis may be associated with secondary complications of influenza. Thus, it may be important to include the evaluation of secondary complications when estimating the burden of influenza.

Limitations

The findings of this study should be interpreted in the context of several limitations. First, the study only includes data from hospital-based clinical settings. Although this study may capture a portion of outpatient care received by patients, it may not be reflective of the entirety of their outpatient healthcare resource use for influenza and its secondary complications. For example, a portion of patients hospitalized with influenza, particularly those who are older, may be discharged to other care facilities and those discharged to home may need to seek additional care from outpatient practices that are not associated with the PHD hospital system; further studies are warranted to evaluate influenza HCRU and cost in such settings.

Due to the nature of the PHD administrative dataset, the identification of influenza cases, chronic conditions, and influenza-related secondary complications relied on ICD-10 coding accuracy. In addition, due to the nature of the PHD database, a baseline period could not be established for the identification of underlying conditions. As a result, medical conditions that are both chronic conditions and secondary complications had to be classified as either a patient characteristic (chronic condition) or as an outcome (secondary complication). Furthermore, the PHD may not contain the entire patient medical history; thus, the prevalence of medical conditions among the study population may be underestimated in this study. Additionally, influenza vaccination status was not available and influenza treatment was not evaluated. Due to the limited laboratory clinical data available in the PHD, influenza diagnoses were not laboratory confirmed and there is a possibility that they may have included cases with other viral illnesses identified as influenza. However, a validation study of the ICD-10 codes used in the current study reported a sensitivity of 83%, a specificity of 98%, a positive predictive value of 91%, and a negative predictive value of 96%.³⁶ The study also excluded patients with any evidence of pregnancy or childbirth during the index hospitalization to reduce the risk of bias when estimating the economic burden of hospitalization through the inclusion of pregnancy- or delivery-related care and any associated costs. While the PHD contains information from a high proportion of hospitals across the US, it may not be representative of the entire population of hospitalized influenza patients in the US.¹⁹ Finally, although this study focused on describing the economic burden of influenza, comparison to the burden of another respiratory infection would be valuable from a public health policy perspective and is an area of research to be explored in future research projects.

Conclusions

The majority of influenza-related hospitalizations was initiated in the ED and had secondary complications presenting during the hospitalization, which had higher medical costs. Although most influenza-related hospitalizations do not require MV usage or ICU admission, those that required both interventions had exceptionally high HCRU and medical costs. There are variations in the HCRU and costs of influenza-related hospitalization among different age groups and over different seasons due to various factors. Overall, the economic burden of influenza-related hospitalizations remains substantial across all age groups in the most recent years.

Transparency

Author contributions

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article. All authors contributed to study conception and design, data acquisition, analysis, and interpretation, drafting and revising of the manuscript, and have given their approval for this manuscript version to be published.

Reviewer disclosures

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Previous presentations

None.

Acknowledgements

We would like to acknowledge Jingying Zhou and Sachin Heerah for their programming support on this project. Grace Lin, an employee of Novosys Health, provided editorial support in the drafting of this manuscript, which was funded by Pfizer, Inc.

Declaration of funding

The study was sponsored and funded by Pfizer, Inc., New York, NY.

Declaration of financial/other interests

All authors are current employees of Pfizer, Inc.

Data availability statement

Data generated during this study are available upon request.

Notes

i Chronic conditions included dementia, rheumatic disease, peptic ulcer disease, diabetes with and without chronic complications, hemiplegia or paraplegia, any malignancy (solid and hematologic malignancies with the exception of malignant neoplasm of the skin), metastatic solid tumors, moderate or severe liver disease, and acquired immunodeficiency syndrome (AIDS)/human immunodeficiency virus (HIV). Medical conditions that are also considered to be secondary complications of influenza were not evaluated as a chronic condition. 4. Centers for Disease Control and Prevention. People at Higher Risk of Flu Complications: U.S. Department of Health & Human Services; 2023 [updated August 25, 2023; cited 2023 October]. Available from: https://www.cdc.gov/flu/highrisk/index.htm.