Cost and health‐related quality of life for children hospitalized with respiratory syncytial virus in Central China

Abstract Background The economic burden of respiratory syncytial virus (RSV) infection and its impact on health‐related quality of life (HRQoL) are not well‐understood in China. This study assessed total cost and HRQoL for children hospitalized with RSV in Central China. Methods Based on a prospective case series study in Henan Province in 2020–2021, inpatients aged 0–59 months with RSV‐related acute respiratory infections (ARIs) were included into analysis. Total cost included direct medical cost (sum of medical cost before and during hospitalization), direct non‐medical cost, and indirect cost. Direct medical cost during hospitalization data were extracted from the hospital information system. Other costs and HRQoL status were obtained from a telephone survey conducted in the caregivers of the enrolled patients. Results Among 261 RSV‐infected inpatients, caregivers of 170 non‐severe cases (65.1%, 170/261) were successfully interviewed. Direct medical cost per episode was 1055.3 US dollars (US$) (95% CI: 998.2–1112.5 US$). Direct non‐medical cost and indirect cost per episode were 83.6 US$ (95% CI: 77.5–89.7 US$) and 162.4 US$ (95% CI: 127.9–197.0 US$), respectively. Quality adjusted life years (QALY) loss for non‐severe RSV hospitalization was 8.9 × 10−3 (95% CI: 7.9 × 10−3–9.9 × 10−3). The majority of inpatients were <1 year of age comprising significantly higher cost and more QALY loss than older children. Conclusions RSV‐associated hospitalization poses high economic and health burden in Central China particularly for children <1 year old. Our findings are crucial for determining the priority of interventions and allocation of health resources.


| INTRODUCTION
Respiratory syncytial virus (RSV) is a leading cause of hospitalization for young children with acute lower respiratory tract infection, resulting in a considerable disease burden around the world. 1 With 33.0 million RSVassociated acute lower respiratory infection episodes, 3.6 million hospitalizations, and 26,300 in-hospital deaths in children aged 0-59 months in 2019, 2 it was estimated that RSV accounts for about 28% of all acute lower respiratory tract infection episodes and 13%-22% of acute lower respiratory infection related mortalities globally. 3 Additionally, RSV infection creates a substantial economic burden across the globe. 4 The global cost for inpatient management was estimated to be €4.82 billion in 2017 from the healthcare payers' perspective. 5 Several candidates for RSV vaccines, immunoprophylaxis, and immune therapeutics are in clinical trials. 6 Two available preventive measures-palivizumab and polyclonal RSV intravenous immunoglobulin (RSV-IVIG) product-are primarily recommended for children at high risks but not widely used due to higher cost. 5,7 Nirsevimab was recently approved in Europe for the prevention of RSV lower respiratory tract disease in neonates and infants during their first RSV season. 8 Estimations of RSV-related health and economic burden are required for assessing the potential benefits of introduction of these products in countries and regions where no such interventions have been licensed yet, including China. 9 However, there are limited numbers of studies assessing the economic burden and health-related quality of life (HRQoL) associated to RSV infection in China with most of the studies evaluating only the direct medical cost of hospitalization. 10 In the present work, we conducted a prospective study in Central China to estimate the total cost and assess the HRQoL in children hospitalized with RSV-associated acute respiratory infection (ARI).

| Study design and population
This study was based on a prospective case series study conducted at Henan Children's Hospital, the largest tertiary pediatric hospital in Henan province of Central China, between December 29, 2020 and October 20, 2021. Inpatients, aged under 5 years and having an acute onset of symptoms of respiratory infection, with or without fever (depending on whether aged over 3 months or not), were identified as ARIs cases and enrolled. The respiratory specimens from the participants were tested for RSV by real-time reverse transcription polymerase chain reaction (real-time RT-PCR) within 48 h of admission.
Detailed study design and enrollment process were described in the previous study. 11

| Data collection and telephone survey
General characteristics collected for the participants included demographics (age, gender, address, etc.), underlying conditions (preterm birth, congenital diseases, airway dysplasia, etc.), medical history (date of onset, date of hospital admission, date of discharge, etc.), and clinical outcomes (mechanical ventilation, intensive care unit [ICU] admission, complications, etc.). These data were extracted directly from the hospital information system (HIS) and checked for completeness and accuracy by the investigators.
The total cost described in this study included direct medical cost, direct non-medical cost, and indirect cost throughout the whole RSV infection episode ( Figure S1). Direct medical cost was the sum of direct medical cost during and before the hospitalization including cost for examinations, treatment, consultation charges (ward and outpatient consultation), self-medication, and so forth. 12 Direct non-medical cost included cost of transportation, food, and accommodation for the patients and caregivers. Indirect cost accounted for the caregivers' loss of income while taking care of the patients. The direct medical cost during hospitalization was obtained from HIS including both reimbursable charges (covered by medical insurance) and the out-of-pocket expenses. The direct medical cost prior to hospitalization, direct non-medical cost, indirect cost, and the HRQoL status were obtained from the telephone survey.
The telephone survey was conducted between January 20 and November 19, 2021. Caregivers of the enrolled patients were called by a trained nurse, under the supervision of the research investigators, after ≥14 days of discharge. Each telephone number was dialed three times on different days before classifying them as unreachable.
After obtaining verbal consent, the caregivers were requested to provide information related to cost and HRQoL for the corresponding RSV episode. The questionnaire covered the following topics: (i) disease-related information (e.g., how long the symptoms persisted); (ii) direct medical cost prior to hospitalization (e.g., whether visited healthcare facilities before hospitalization, the amount of direct medical expenditure prior to hospitalization, etc.); (iii) direct non-medical cost attributable to the RSV episode; (iv) indirect cost (e.g., the loss of workdays for caregiving); (v) HRQoL of the patients during the episode. If the interviewees were unable to recall quantitative information, for example, medical cost before admission or food cost per day, they were asked to provide a range of cost. The The general characteristics and direct medical cost during hospitalization data were also collected from the HIS for the patients who were unreachable in the telephone survey to examine if there had been any notable differences between the patients who were successfully interviewed and those who were not. In addition, we retrospectively extracted general characteristics and direct medical cost during hospitalization data for 386 RSV-infected inpatients admitted in 2018-2019 using the same procedure to explore if variations existed in direct medical cost during hospitalization between the years due to the circulation of different subgroups or genotypes of RSV. 15,16

| Data analysis
Data analyses were performed in R (version 4.1.2, R Foundation for Statistical Computing, Vienna, Austria). All costs were described as mean with 95% confidence interval (95% CI). 17 Cost spent before 2021 were inflated to the equivalent of 2021 Chinese Yuan (CNY) using the consumer price index for China, [18][19][20] and then were converted to US dollar (US$) according to the average exchange rate of 2021 (1 US$ = 6.45 CNY). General characteristics and costs were also described by two subgroups based on the severity of illness. Severe RSV cases was defined as those with illness that requiring ICU admission, different types of ventilation support during the hospitalization, or in-hospital death. 21 The total cost per RSV episode was the sum of direct medical cost, direct non-medical cost and indirect cost of patients who completed the telephone survey. The direct medical cost was calculated by adding up the direct medical cost during hospitalization and prior to hospitalization of each case. Indirect cost was calculated using the human capital method-by multiplying mean daily income per capita with the number of days the caregivers were absent from the work during the episode. 22 Generalized linear models (GLMs) were performed to explore potential risk factors associated with aforementioned types of cost.
At first, the costs were logarithmically transformed. Then a univariate analysis was done on them to identify potential factors correlated to the cost. Finally, a forward stepwise model selection was performed using Akaike's Information Criterion (AIC) and likelihood ratio test to select the explanatory variables in multivariate analyses. A p-value of <0.05 was considered statistically significant.
PedsQL™ scores could not be used to calculate a quality adjusted life years (QALY) outcome directly as it is a non-preference-based measure (NPBM) instrument. 23 Therefore, a mapping algorithm was adopted to convert the scores of PedsQL™ to the utility score of the EQ-5D-Y (see Supporting Information). 24 Finally, the loss of QALY was estimated by multiplying the difference of health utility during the RSV episode and full health (assuming "full health" utility = 1.0) 25 with the reported duration of RSV illness.

| Sensitivity analysis
As mentioned before, if a range of cost was obtained in the survey (instead of exact cost), the midrange value was taken in the baseline analysis to evaluate the total cost related to RSV hospitalization. Then sensitivity analyses were conducted by using the lower and upper limits of the ranges to test the impact of uncertainty on the results.  Table 1. The majority of patients were at 0-11 months of age (58.2%) and male (58.2%). The vast majority of patients were without underlying conditions (92.7%), that is, otherwise healthy, and for patients <1 year the proportion was 91.4% (139/152). General characteristics of the 170 successfully interviewed and the 91 excluded patients were comparable. However, all the cases that were successfully interviewed was non-severe cases (p = 0.005).

| General characteristics of patients
Only five severe cases were identified, among whom four were under 1 year old (Table S1). Table 2 summarizes direct medical, direct non-medical, and indirect cost for the 170 non-severe RSV patients who were successfully interviewed. The mean direct medical cost was 1055.3 US$ (95% CI: 998.2-1112.5 US$). There was no significant difference in direct medical cost between patients aged 0-5 and 6-11 months (Table S2).

| Total cost of patients hospitalized with RSV
Direct medical cost was significantly higher for patients aged 0-11 months than that for the older ones. Overall, 90.9% of the direct medical cost was spent during hospitalization (959.3 US$, 95% CI: Table S3).
Direct non-medical cost between patients aged 0-5 and 6-11 months also showed no significant difference (Table S2). Significantly higher direct non-medical cost was associated with patients aged 0-11 months than older age groups. Moreover, patients with underlying conditions were likely to pay more direct non-medical expenses than those without ( p = 0.019).

| Factors associated with cost of RSV hospitalization
Univariate and multivariate analyses of combined direct medical cost during hospitalization data for 2018-2019 and 2020-2021 revealed that patients at younger age, development of severe illness, and having at least one complication during the episode were associated with higher direct medical cost during hospitalization. The cost also F I G U R E 1 Flow chart for the enrollment RSV-infected inpatient for total cost and health-related quality of life (HRQoL) analysis.
increased with the extension of hospital length of stay (LOS) (Table S4).
Multivariate analysis of the data from 170 successfully interviewed patients revealed that younger age and longer LOS were associated with higher direct medical cost. Longer LOS was also the only factor associated with higher direct non-medical cost and total cost.
In our study, no particular factor was found to be related to indirect cost (not shown in tables).

| Sensitivity analysis
During the telephone survey, 48.8% (83/170) of caregivers were unable to recall the exact direct medical cost prior to hospitalization, direct non-medical cost, or indirect cost for at least one item and provided ranges instead. The lower and upper limits of the cost ranges were used in the sensitivity analyses (Tables S5-1 Table 3    The reasons for exclusion include unreachable, rejecting to participate or failed to recall cost of the RSV episode. ‡ Mainly including preterm birth, congenital heart diseases, airway dysplasia, asthma and other congenital or chronic comorbidities. § Mainly including pneumonia, bronchitis, respiratory failure, hepatic function abnormality, myocardial injury, and sepsis.

| HRQoL
T A B L E 2 Direct medical, direct non-medical, and indirect cost of non-severe RSV-infected inpatients. F I G U R E 2 Proportion of cost for RSV-infected inpatients with or without underlying conditions (overall 170 patients were successfully interviewed; 159 patients were without underlying conditions and 11 patients had underlying conditions). 0.8). The estimated QALY loss per RSV episode was 8.9 Â 10 À3 (95% CI: 7.9 Â 10 À3 -9.9 Â 10 À3 ). There was no significant difference in the QALY loss between patients aged 0-5 and 6-11 months (Table S2). Patients aged 0-11 months suffered more QALY loss than older age groups ( p = 0.038). There was no significant difference in the QALY loss between patients with and without underlying conditions.

| DISCUSSION
This prospective study was conducted in Central China to estimate the total cost and assess the HRQoL associated with RSV-infected pediatric inpatients by extracting data from HIS and conducting telephone surveys. Our study found that RSV related hospitalization is a key health concern for patients aged 0-59 months, particular for inpatients was also higher than the total health expenditure per capita. 5 In developing countries like Bangladesh, the median outof-pocket expense for RSV-related hospitalization represented 24% of the monthly household income. 29 In Malawi, the cost of each RSV hospitalization was about 15.8% of its GDP per capita. 12  Younger age was found to be associated with increased cost of RSV-related hospitalization in this study, consistent with the findings from Demont et al. showing infants <1 year old represented 80% of the economic burden of RSV among children aged <5 years. 30 Additionally, this study also found that significantly higher direct medical cost during hospitalization was associated with patients having underlying conditions (e.g., preterm birth). This finding is also consistent with previous studies. [31][32][33][34][35][36] However, it should be highlighted that the majority of RSV-infected children did not have any underlying conditions ("otherwise healthy"). Together, such patients consumed 73.6%-92.9% of the total cost spent; this finding is also consistent with previous studies. 35,37,38 For example, Bowser et al. evaluated that full-term infants accounted for 70%-82% of the total RSV hospitalization cost among US children aged 0-59 months. 28 Considering the large population and the high incidence of RSV in Chinese children, 39  Colombian children aged <2 years due to RSV bronchiolitis. 42 In our study, the QALY loss of premature infants did not differ significantly from that of other children and was lower than the QALY loss reported by Glaser et al. However, it was higher than the value reported by Hodgson et al. as our study did not involve outpatients for whom less QALY loss is expected. Other potential factors producing the differences could be selection of different measuring tools, difference in target time period of evaluation, and the discrepancies arising due to the collection of data from parents/caregivers rather than direct measurement of QoL of the young children. 40 Our study has several limitations. Firstly, telephone survey was carried out at least half a month after the discharge making the recall bias inevitable. Secondly, the severe cases could not be reached by the telephone survey preventing us from evaluating direct nonmedical, indirect cost, and HRQoL for severely ill inpatients. Thirdly, some cost might not be taken into account in this analyses. For instance, the interview did not inquire potential medical cost incurred after hospitalization, and the cost of ambulance services was not included in direct medical cost (one case reported being transferred to the hospital by ambulance). The factors above may lead to an underestimation of the overall economic burden during the whole course of RSV infection. Besides, the utility score for fully healthy children was assumed to be 1.0 in the HRQoL study 23 undermining the fact that children might not be "fully healthy" before infection. This could overestimate the QALY loss in our study. Also, the mapping algorithm used in our study was developed based on data from children aged 11-15 years. As a result, the estimated utility of younger children with different clinical characteristics might have a degree of uncertainty. Finally, the data used in the present study were collected for patients of only one hospital in one province. Therefore, the results are unlikely to represent the true scenario of other areas in the diverse healthcare settings of China.

| CONCLUSIONS
Our study revealed that RSV-allied hospitalization causes high economic and health burden, particularly for infants <1 year old. The find-

ACKNOWLEDGMENTS
The authors thank all the patients and caregivers for participating in our study. The authors also give thanks to the staff members at Henan Children's Hospital for their help in field investigation, administration, and data collection.

CONFLICT OF INTEREST STATEMENT
HY has received research funding from Sanofi, Shanghai Roche Pharmaceutical Company, GlaxoSmithKline (China) Investment Co., Ltd., Yichang HEC Changjiang Pharmacertical Co., Ltd., SINOVAC Biotech Ltd. and Shenzhen Sanofi Pasteur Biological Products Co., Ltd.. All other authors declare no competing interests.

PEER REVIEW
The peer review history for this article is available at https://www. webofscience.com/api/gateway/wos/peer-review/10.1111/irv.

DATA AVAILABILITY STATEMENT
The data are not publicly available due to ethical restrictions and privacy. De-identified individual data that supports the findings of this study are available from the corresponding authors on reasonable request. The data requestor will need to sign a data access agreement.