- Email: [email protected]
Indirect, out-of-pocket and medical costs from influenza-related illness in young children
Vaccine 30 (2012) 4175–4181
Contents lists available at SciVerse ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Indirect, out-of-pocket and medical costs from influenza-related illness in young children夽 Ismael R. Ortega-Sanchez a,∗,1 , Noelle-Angelique M. Molinari a,1 , Gerry Fairbrother b,1 , Peter G. Szilagyi c,1 , Kathryn M. Edwards d,1 , Marie R. Griffin d,1 , Amy Cassedy b,1 , Katherine A. Poehling d,1 , Carolyn Bridges a,1 , Mary Allen Staat b,1 a
Centers for Disease Control and Prevention, United States Cincinnati Children’s Hospital Medical Center, United States c University of Rochester, School of Medicine and Dentistry, United States d Vanderbilt University, School of Medicine, United States b
a r t i c l e
i n f o
Article history: Received 2 January 2012 Received in revised form 9 April 2012 Accepted 16 April 2012 Available online 27 April 2012 Keywords: Influenza Children Out-of-pocket cost Indirect cost
a b s t r a c t Background: Studies have documented direct medical costs of influenza-related illness in young children, however little is known about the out-of-pocket and indirect costs (e.g., missed work time) incurred by caregivers of children with medically attended influenza. Objective: To determine the indirect, out-of-pocket (OOP), and direct medical costs of laboratoryconfirmed medically attended influenza illness among young children. Methods: Using a population-based surveillance network, we evaluated a representative group of children aged <5 years with laboratory-confirmed, medically attended influenza during the 2003–2004 season. Children hospitalized or seen in emergency department (ED) or outpatient settings in surveillance counties with laboratory-confirmed influenza were identified and data were collected from medical records, accounting databases, and follow-up interviews with caregivers. Outcome measures included work time missed, OOP expenses (e.g., over-the-counter medicines, travel expenses), and direct medical costs. Costs were estimated (in 2009 US Dollars) and comparisons were made among children with and without high risk conditions for influenza-related complications. Results: Data were obtained from 67 inpatients, 121 ED patients and 92 outpatients with laboratoryconfirmed influenza. Caregivers of hospitalized children missed an average of 73 work hours (estimated cost $1456); caregivers of children seen in the ED and outpatient clinics missed 19 ($383) and 11 work hours ($222), respectively. Average OOP expenses were $178, $125 and $52 for inpatients, ED-patients and outpatients, respectively. OOP and indirect costs were similar between those with and without high risk conditions (p > 0.10). Medical costs totaled $3990 for inpatients and $730 for ED-patients. Conclusions: Out-of-pocket and indirect costs of laboratory-confirmed and medically attended influenza in young children are substantial and support the benefits of vaccination. Published by Elsevier Ltd.
1. Introduction Seasonal influenza, estimated to result in over 200,000 hospitalizations and 31 million outpatient visits annually in the United States (US) [1,2], disproportionally affects young children and the
Abbreviations: OOP, out-of-pocket costs; ED, emergency department; NVSN, New Vaccine Surveillance Network; IQR, inter-quartile range. 夽 The findings and conclusions expressed are those of the authors and do not necessarily represent the official views of the CDC or DHHS. ∗ Corresponding author at: Centers for Disease Control and Prevention, 1600 Clifton Rd., NE MS-A-34, Atlanta, GA 30333, United States. Tel.: +1 404 639 2834; fax: +1 404 639 4960. E-mail address: [email protected] (I.R. Ortega-Sanchez). 1 For the New Vaccine Surveillance Network (NVSN). See Appendix A. 0264-410X/$ – see front matter. Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.vaccine.2012.04.057
elderly [2–6]. Approximately 20% of all influenza-attributable hospitalizations and 12% of all outpatient visits are among children aged ≤5 years [1], although they comprise only 7% of population [7]. Direct medical costs of influenza illness in US children have been estimated to be >$1.2 billion per year [1]. Although these costs are high, they nevertheless underestimate the overall economic burden of influenza disease. Influenza disease also has an economic impact on family caregivers: family caregivers miss work time (i.e., indirect costs) and incur illness-related out-of-pocket (OOP) expenses such as co-payments, travel and lodging costs, over-the-counter medications and other expenses. Although data are available regarding medical costs for laboratory-confirmed influenza [8,9], limited economic data are available regarding work time lost by caregivers [10–12], costs incurred in medical settings [13], or OOP costs [11]
4176
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
incurred by families of children with influenza. When assessing the economic impact of the benefits of influenza vaccination on families, most societal cost analyses and cost-effective analyses have relied primarily on ad hoc assumptions or extrapolations from studies of other diseases [1,11–15]. The objective of our study was to estimate the indirect and OOP costs to caregivers of children <5 years with laboratory-confirmed medically attended influenza. We also examined the direct medical costs attributable to these influenza cases. We compared the medical and non-medical resource utilization rates and associated costs by type of medical care for children seen in the hospital, ED, and outpatient settings.
2. Methods 2.1. Overview The CDC-funded New Vaccine Surveillance Network (NVSN) was used to evaluate a representative group of children aged <5 years with laboratory-confirmed medically attended influenza during the 2003–2004 influenza season previously described [16]. Briefly, county children with acute respiratory infection and/or fever who were hospitalized or seen in emergency department (ED) and outpatient settings were enrolled. Enrolled children had nasal and throat swabs collected and tested for influenza by viral culture and polymerase chain reaction. For this study, enrolled county children with laboratory-confirmed influenza were identified and data were collected from medical chart reviews, medical billing records, and follow-up interviews with parents or caregivers. For the 2003–2004 study year, additional questions and telephone follow-up calls with parents or guardians were conducted to collect additional information on costs of illness. Data collection and analyses included rates of healthcare resource utilization (e.g., days of hospitalization, number of doctor visits), time missed by caregivers, OOP expenses (e.g., co-pays, travel expenses) and medical costs. Rates and costs were estimated by the highest level of medical care received (see description below) and compared across high-risk conditions for influenza-related complications as designated by the Advisory Committee on Immunization Practices.
2.2. Data 2.2.1. Surveillance and enrollment NVSN conducted influenza surveillance in three US counties, Monroe County, New York (Rochester), Davidson County, Tennessee (Nashville), and Hamilton County, Ohio (Cincinnati). After informed consent, 85% of eligible county children who were admitted with a diagnosis of acute respiratory tract infection and/or fever were enrolled [16]. Children were excluded if they had respiratory symptoms >14 days, fever and neutropenia associated with chemotherapy, were hospitalized in the prior 4 days or transferred from another surveillance hospital, or were newborns never discharged from the hospital. Surveillance hospitals captured ≥95% of all county children hospitalized with acute respiratory illness or fever [8,16]. County children presenting to selected clinics and emergency departments with symptoms of acute respiratory tract infection or fever were enrolled [8,16]. Children were enrolled 1–2 days per week in the clinic and 3–4 days per week in the only ED associated with a children’s hospital in each county with systematic rotation of days and shifts to obtain a representative sample. The percentage of all visits attributable to acute respiratory tract infection or fever for county children seen at these emergency departments was 30% in Nashville, 70% in Rochester, and 95% in Cincinnati [16].
All children had nasal and throat swabs obtained for influenza testing. Children were determined to have laboratory-confirmed influenza if these nasal and throat swabs were positive for influenza by duplicate reverse-transcriptase-polymerase-chain reaction (RTPCR) test or viral culture [16]. 2.3. Caregivers’ interview Parents or guardians of children with laboratory-confirmed influenza were interviewed by telephone three to four weeks after hospitalization discharge or ED or clinic visit. Multiple attempts to contact caregivers were made with an overall response rate of 84.6%. Reasons for refusal to participate in the study were not collected. Data encompassed caregivers’ time lost from work and expenses related to the influenza illness, including illness-related expenses that may have occurred two weeks prior to and three weeks after hospitalization, or one week before through one week after an ED or clinic visit. These data were collected using a standardized questionnaire along with caregivers’ employment status, income and work time shifts (see Fig. 1, details are also included below). 2.3.1. Medical charts and billing records Information on demographics, clinical procedures and medical resources used were collected by reviewing medical charts of enrolled patients. Detailed medical cost data for inpatients and ED patients were also provided by each surveillance site. Sources of medical cost data included summary billing records from unified billing forms (known as UB-92s) and similar accounting systems containing both costs and charges. In all cases, these medical data contained detailed line-item charges for all diagnostic tests, therapeutics, supplies, and room fees. Although medical billing data for outpatients were not gathered, data on medical resource utilization were collected during chart reviews at outpatient clinics. 2.4. Analysis Rates of healthcare utilization and associated costs per influenza patient were estimated following standard procedures [17,18] and using patient-specific data collected from medical billing records, medical chart reviews and follow-up interviews with patient caregivers. Using the highest level of medical care received by influenza patients, the data were then categorized into three mutually exclusive levels of care: inpatients, ED-patients and outpatients [1,11,16]. Estimates for each level of care included the resources and costs of the lower level of care, if applicable. For example, data for inpatients included not only hospitalization but also ED and outpatient care that might have occurred before or after hospitalization; and data for ED patients also include outpatient care, if any. Fig. 1 describes the categories of resource utilization and costs by medical care setting and source of data. We also collected information on the number of clinic or ED visits, the length of stay when hospitalized and the number of workdays lost by caregivers (Fig. 1). The number of medical procedures, including radiological procedures and laboratory tests were also assessed. 2.4.1. Cost categories Costs were grouped into three categories: (1) OOP expenses (medical and non-medical payments or co-payments, over-thecounter drugs, travel and phone call expenses related to the child’s illness), (2) indirect costs (wages and income associated with time missed from work during the child’s illness) and (3) medical costs. Medical costs were grouped into four subcategories: diagnostics (e.g., radiology, laboratory), therapeutics (e.g., pharmaceuticals, blood products, intensive care therapies), room fees and supplies (e.g., operating, recovery and ward rooms, ICU, sedation, ED units,
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
4177
Laboratory-confirmed influenza patients enrolled from active surveillance
ED-patients
Outpatients
Inpatients
*
Resource utilization
Costs
*
DIRECT From medical chart review and billing records -Outpatient and ED care (visits) -Hospitalization (days) -Medical procedures (number) -Laboratory (tests) -Radiology (x-rays) From caregivers’ From medical interview billing records Out-of-pocket costs Medical costs -Medical copayments -Room & supplies -Physician fees -Non medical payments -Pharmacy -Over-the-counter drugs -Telephone calls -Laboratory -Travel and logging -Radiology
INDIRECT From caregivers’ interview -Caregivers time off from work, school or other activities -Caregivers time off for visits to or stays in healthcare institutions
From caregivers’ interview -Value of work time off -Value of time off for Dr Visits -Value of time off for Hospital or ED
*Although specific to the level of medical care, descriptions of the types and sources of data inED-patients and Outpatients are similar to those described in the table for inpatients. No medical billing data were collected for outpatients. Fig. 1. Categories of resource utilization and direct and indirect costs by type of cost, medical care setting and source of data.
and transportation), and physician services (e.g., fees from visits, consultations, and procedures) (see Fig. 1). Since a specific physician cost-to-charge ratio was not available from hospitals, fees from physician services as reported in medical billing records were included after adjusting them with a generic cost-to-charge ratio [8]. Unlike physician fees, all other medical costs were adjusted using hospital- and department-specific cost-to-charge ratios. All costs were later adjusted for inflation to December 2009 US dollars using the Medical Component of the Consumer Price Index [20] for medical cost, and the Gross Domestic Product deflator index [21] for OOP and indirect costs. 2.4.2. Valuing time missed Caregivers were also asked about the uses and sources of time lost at the household level: the number of hours allocated to doctor visits, hospitalization stays or home care during the child’s illness was used to account for the uses of caregivers’ time. For caregivers, efforts were made to identify whether time lost was from work, school or college, or other activities.
Although efforts were made to collect data on household income and then calculate the prorated hourly earnings for time lost to influenza, due to the limited response to income questions we relied on an indirect approach. To estimate the value of caregivers’ time off work (i.e., indirect costs) we used non-occupation-specific hourly earnings data for individuals belonging to the surveillance sites (i.e., $17–$22 per hour) as reported in the National Compensation Survey from the by the US Bureau of Labor Statistics [19]. 2.4.3. Statistical analysis Means, medians, and inter-quartile range (IQR) values of resource utilization rates and costs for each influenza final outcome were calculated. Estimates within each outcome were differentiated by patient risk and age group. Because previous and preliminary univariate analyses showed skewed costs by a few cases with extremely high medical costs [8], standard non-parametric analyses of variance (Mann–Whitney–Wilcoxon [22,23] and Kolmogorov–Smirnov score tests [23]) were used for comparisons among categories and subcategories of resource utilization
4178
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
Table 1 Characteristics of enrolled children with laboratory-confirmed influenza. Influenza cases N Age 0–5 months 6–23 months 24–59 months Gender Female Male Risk of complicationb High Non high Race/ethnic group White Black Other or unknown Health insurance Uninsured Public Insurance Private Insurance Household income <$10,000 $10,000 to <$30,000 $30,000 to <$50,000 $50,000 to <$100,000 ≥$100,000 Refused, unknown, missing
Inpatients (%)
ED-patients (%)
Outpatients (%)
67
122
92
30 (48)a 18 (29) 14 (23)
16 (13)a 48 (39) 58 (48)
9 (10)a 38 (41) 45 (49)
41 (61) 26 (39)
50 (41) 72 (59)
50 (54) 42 (46)
14 (21) 53 (79)
27 (22) 94 (78)
21 (23) 71 (77)
32 (51) 27 (43) 4 (6)
41 (34) 53 (43) 28 (23)
26 (32) 35 (43) 19 (24)
3 (5) 30 (45) 34 (51)
6 (5) 81 (68) 33 (28)
6 (7) 55 (63) 27 (31)
25 (37) 13 (19) 7 (10) 9 (13) 4 (6) 11 (16)
29 (24) 37 (30) 9 (8) 8 (7) 8 (7) 35 (29)
29 (32) 18 (20) 8 (9) 11 (12) 5 (5) 27 (29)
a
Percent figures do not add to 100 due to rounding. High risk (HR) for influenza related complications is based on the CDC Advisory Committee on Immunization Practices definition [32]. Conditions included asthma, cancer, diabetes, heart disease, immune deficiency, kidney disorders, sickle cell disease, cystic fibrosis, bronchopulmonary dysplasia, blood disorders, liver disorders, and seizure disorders. b
rates and costs. The exact form of non-parametric tests was performed to obtain a higher accuracy on statistical differences among subgroups. 3. Results Data were collected from 67 hospitalized patients (i.e., inpatients), 122 ED-patients, and 92 clinic patients (i.e., outpatients) with laboratory-confirmed influenza cases during the 2003–2004 influenza winter season (Table 1). No deaths were detected among the surveillance population. Nearly half (48%) of inpatients were less than 6 months old, while almost half of ED-patients (48%) and outpatients (49%) were 24–59 months old. A high proportion had public health insurance (i.e., Medicaid) and had household incomes of less than $30,000 (Table 1). 3.1. Indirect costs About 75% of families with an inpatient stay reported work time lost during the child’s influenza illness (Table 2). Fewer families reported time off from school or college (6–10%) or from other unpaid activities (10% for inpatients, 18% for ED-patients, and 14% for outpatients) (Table 2). In all cases, caregivers cared for patients at home, escorted them to medical visits, spent time in hospitals (if child was hospitalized) and provided help in various other forms (e.g., taking care of other non-sick kids, making runs to pharmacies, etc.). Caregivers of hospitalized children with influenza reported, on average, 73 h of work time lost, mostly for hospital stays (Table 3). ED patients averaged 19 h of caregivers’ work time lost, mostly due to ED visits while outpatients also averaged at least 7 h of caregivers’ work time lost (Table 3). The average indirect cost (e.g., caregiver time cost) for inpatients was estimated to be $1456 (median $1089; IQR $1319), mainly due to work time missed during hospital stays (82%) (Table 3). Indirect cost incurred by caregivers of ED-patients was $383 (median
$125; IQR 268) and by caregivers of outpatients was $222 (median $85; IQR 280). No statistical difference was found among indirect cost between patients with identified high risk conditions and those without (p > 0.427 for inpatients, p > 0.486 for ED-patients, p > 0.137 for outpatients) (Table 4). Similarly, among children aged ≤23 months versus 24–59 months no statistical difference was found in inpatients ($1321 versus $1525, p > 0.323) and outpatients ($257 versus $198, p > 0.150); and only ED-patients reported a marginal statistical difference by age group ($503 versus $217, p > 0.022).
3.2. Out-of-pocket costs Approximately 55% of families of inpatients reported medical payments, over-the-counter medication and travel costs related to the child’s influenza illness. Expenses were also reported by 71% and 63% of families of ED-patients and outpatients, respectively (Table 2). The average OOP cost among all inpatients was approximately $178 (median $50; IQR 84) with medical payments (>43% of total OOP), travel expenditures (>24%) and phone calls costs (15%) as principal components (Table 3). For ED-patients, average OOP costs were $125 (median $35, IQR 57) mainly due to travel (53%) and prepaid phone card, internet fee or payphone call expenses (26%). The average OOP cost for outpatients were approximately $52 (median $29, IQR 56) again due to travel (32%) and phone call expenses (20%). The estimated medians for OOP payments in patients were not statistically different (p > 0.10) between high-risk and non-high risk groups (not shown). Although variations in components of OOP costs (i.e., copays and travel costs) were observed across age groups, analyses showed no statistical differences in the total OOP cost among inpatients aged ≤23 months versus 24–59 months ($182 versus $162, p > 0.133), and outpatients ($61 versus $39, p > 0.438). Only ED-patients showed a statistical difference ($181 versus $63, p > 0.052).
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
4179
Table 2 Types of costs incurred among enrolled influenza cases by level of medical care. ED-patients n (%)a
Outpatients n (%)a
122
92
50 (75) 4 (6) 7 (10)
57 (47) 8 (7) 22 (18)
39 (42) 9 (10) 13 (14)
37 (55) 17 (25) 37 (55) 17 (25) 34 (51)
89 (71) 17 (14) 89 (71) 35 (29) 88 (62)
58 (63) 18 (20) 58 (63) 29 (32) 58 (63)
Inpatients n (%)a N Caregivers reporting time off From work From school From unpaid activity Out-of-pocket costs Medical Nonmedical Over the counter Telephone costs Travel costs Healthcare utilizationb Outpatient care Hospitalization Medical proceduresc Laboratoryd Radiology
67
67 (100) 67 (100) 60 (90) 50 (75) 34 (51)
122 (100) n/a 47 (39) 3 (3) 31 (25)
92 (100) n/a 4 (4) 1 (1) 4 (4)
Medical costs
67 (100)
122 (100)
n/a
a
Percent figures represent proportion of caregivers reporting specific economic data. Data include patients utilizing at least one unit by specific healthcare resource during influenza illness. c Medical procedures include reports of mechanical ventilation, oxygen, antiviral drug treatment, treatment of bacterial pneumonia, ear infection, or sinusitis. Medical procedures may also include treatments for influenza patients with asthma, bronchopulmonary dysplasia, cystic fibrosis, heart disease, immune deficiency, kidney disorders, and seizure disorders. d Laboratory tests could include a single or multiple blood draws with multiple measures. Note: Healthcare utilization data were collected from medical records while medical costs data were from billing records. No medical costs were collected for outpatients. b
Table 3 Out-of-pocket and caregivers time costs per child with influenza by level of final medical care.
Out-of-pocket costs ($) Total out-of-pocket costs Medical payments* , a Non-medical payments* , b Over-the-counter drugs** Telephone calls* Travel and logging * , c Caregivers time off (h) Total time off from work* Time off for Dr visits** Time off for hospital or ED visits* Caregivers time costs ($)* Total value of work time off Value of time for Dr visits Value of time for hospital or ED
Inpatients
ED-patients
Outpatients
n = 67
n = 122
n = 92
Mean
(SD)
178 78 18 9 29 44
(535) (418) (31) (13) (98) (137)
50 5 0 5 0 10
(68) (13) (62)
73 8 61 1456 164 1201
(1395) (277) (1267)
Median
IQR
Mean
(SD)
84 15 25 15 6 19
125 11 4 11 33 67
(348) (13) (16) (13) (151) (200)
55 4 48
66 9 65
19 7 6
1089 80 965
1320 174 1202
383 146 127
(42) (10) (31) (903) (211) (670)
Median
IQR
Mean
(SD)
35 7 0 7 0 10
57 15 0 15 4 19
52 10 4 10 17 11
(144) (15) (14) (15) (102) (15)
6 5 0
13 7 0
11 7 –
(17) (14) –
125 98 0
268 135 0
222 131 –
(366) (290) –
Median
IQR
29 6 0 6 0 4
56 15 0 15 4 13
0 3 –
15 5 –
85 50 –
280 104 –
a Medical payments may include copayments for medical visits or prescriptions, cost of additional laboratory test or X-rays, cost of traditional or homeopathic medicines or treatments. b Non-medical payments may include hired childcare fees for sick child or for other healthy children while taking care of sick child, home conditioning and cleaning for sick child. c Travel and logging may include expenses for gas, tolls, parking bus fares, taxi fares and hotel room fees. * Non significant (p > 0.10), all Kruskal–Wallis tests: one-way analysis of variance by ranks to compare median values by medical care setting (Pr > Chi square). ** p < 0.05, all Kruskal–Wallis tests: one-way analysis of variance by ranks to compare median values by medical care setting (Pr > Chi square).
3.3. Medical costs Inpatient medical costs were approximately $3990 (median $2929; IQR $1470) with room fees and supplies (55%) and physician fees (23%) as principal components (Table 4). No statistically significant difference was found in medical costs of high risk versus non-HR inpatients. Average medical costs of ED-patients were approximately $730 (median $402; IQR 274) with room and supplies (>50%) and other physician fees (>35%) as principal cost components (Table 4). Overall, median medical costs of ED-patients were not statistically different across risk groups ($274 for high risk versus $260 for non-high risk; p > 0.10). Analyses showed no statistical differences (p > 0.585) in medical costs among inpatients aged <23 months ($4239) and 24–59 months ($3294). On other hand,
marginally significant statistical differences in medical cost among ED-patients aged ≤23 months ($935) and 24–59 months ($502) were observed (p > 0.069).
4. Discussion This is the first study to estimate the OOP and indirect costs of laboratory-confirmed medically attended influenza illnesses among young children. We found a substantial economic burden on families as a result of medically attended influenza illnesses in children aged <5 years in addition to already substantial medical costs [1]. Caregivers lost on average 73 h of work for inpatients, 19 h for ED-patients and 6 h for outpatients with influenza. As a result,
4180
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
Table 4 Medical resource utilization and costs per child with influenza by medical care setting and risk group. All patients, n = 280
Inpatient Medical resources Outpatient care (visits) Hospitalization (days) Medical proceduresb Laboratory (tests) Radiology (X-rays) Medical facility stay related costs ($)c Room and supplies Physician fees ED patient Medical resources Outpatient care (visits) Medical proceduresb Laboratory (tests) Radiology (X-rays) Medical facility visit related costs ($)c Room and supplies Physician fees Outpatientd Healthcare utilization Outpatient care (visits) Medical proceduresb Laboratory (tests) Radiology (X-rays)
Non high risk, n = 218
p valuea
High risk, n = 62
Mean
Median
IQR
Mean
Median
IQR
Mean
Median
IQR
1.50 3.10 10.70 5.00 0.80 3990 2186 899
1 3 2 4 1 2929 1646 585
1 1 4 8 1 1470 1603 555
1.50 3.20 10.80 6.10 0.80 4296 2308 966
1 3 3 6 1 2927 1530 606
1 1 4 7 1 1573 1803 576
1.80 2.60 10.40 0.90 0.70 2830 1725 644
1 2 1 0 1 3115 1790 572
1 1 2 1 1 1056 399 382
0.279 0.063 0.044 <0.001 0.321 0.436 0.309 0.439
1.40 2.70 0.10 0.30 730 370 262
1 0 0 0 402 209 166
1 1 0 1 307 302 128
1.40 2.00 0.10 0.30 796 414 260
1 0 0 0 398 209 166
0 1 0 1 359 315 128
1.40 5.30 0.10 0.30 517 223 274
1 0 0 0 406 209 175
1 1 0 1 154 302 128
0.451 0.259 0.335 0.446 0.480 0.055 0.083
1.27 0.09 0.02 0.04
1 0 0 0
1 0 0 0
1.33 0.06 0.01 0.03
1 0 0 0
1 0 0 0
1.10 0.20 0.00 0.10
1 0 0 0
1 0 0 0
0.194 0.254 0.501 0.438
a
Mann–Whitney test: two-sample one-side nonparametric test of ranks to compare median values by influenza risk condition. The number of medical procedures reported includes mechanical ventilation, oxygen, antiviral drug treatment, treatment of bacterial pneumonia, ear infection, or sinusitis. Medical procedures in inpatients or Ed-patients may also include treatments for asthma, bronchopulmonary dysplasia, cystic fibrosis, heart disease, immune deficiency, kidney disorders, and seizure disorders. c Total costs and two main component costs for hospitalized and ED-patients are shown. The two main component costs are part of the total medical facility costs. These costs data were obtained through review of billing records. d No medical cost from billing data were collected for outpatients. b
wages and self-employed income lost for households amounted to $1456, $383 and $222 for inpatients, ED-patients and outpatients, respectively. Out-of-pocket costs averaged $178 for inpatients, $125 for ED-patients and $52 for outpatients. While these costs are considerable, they do not include the economic value of time lost from leisure or other activities such as attending school or college. Our estimates for indirect and OOP costs were substantially higher than previously estimates from other studies for influenza or other respiratory infections [11–15,24–26]. For example, two recent studies included assumptions that families lose 0.89 workdays (or 7 h) per ill school-aged child [24] and 1.9 workdays (15 h) per ill child attending day-care [25]. While these estimates are similar to our findings for patients seen in an outpatient and ED setting, they substantially underestimate the burden to parents for hospitalized children. Thus, prior studies that did not include such costs substantially underestimated the economic impact of medically attended cases of influenza among young children. Due to data limitations, most cost-effectiveness studies on vaccinating young children against influenza have imputed lower family expenses than those estimated in this study. For example, for each day of hospitalization, two previous studies imputed less than one day of work missed by caregivers [11,12] compared with 2.5 days (i.e., 61 h or 7.6 work days missed divided by 3.1 days in hospital) estimated in our study (Tables 3 and 4). Consequently, the value of time lost per case calculated out of assumed time lost for caregivers of inpatients ($540) [11] were about half of those estimated in our study for hospitalized children (Table 3). In addition, OOP costs in previous studies was limited to medical co-pays for outpatient visits, $8, and over- the-counter drugs between $5 and $7 [11,26] while our study’s estimated OOP costs are substantially higher and include additional costs not captured in other studies (Table 3).
In line with our estimates of medical resource utilization, other studies of young children hospitalized with laboratory confirmed influenza have found a similar length of hospital stay of generally 2–3 days [9,27,28]. However, a wide variation exists in the literature in regards to hospitalization costs. Costs among non-high risk patients in our study were comparable with previous studies [11,29]. Nonetheless, for hospitalized children with more complex conditions, our study reported lower medical costs compared to those estimated for the same risk group in other studies [29,30]. The following strengths and limitations should be considered when interpreting the data we report. We used data collected prospectively from three geographic sites to identify children with laboratory confirmed influenza. Although participation in the surveillance system was high (over 70% of families consented to participate), costs may have been different among children whose families participated compared to those that did not. We obtained item-based medical and non-medical resource and cost data assembled with information from multiple sources, including medical records, billing data and interviews with caregivers. Recall bias may have occurred about OOP expenses and other indirect costs for families as caregivers were interviewed 3–4 weeks after the child’s medically attended illness. Cost data on daycare absenteeism beyond hospitalization or home care were not collected as these may have resulted in duplication of costs. In addition, sample size limited our ability to conduct detailed sub-analyses of influenza illness-related costs and resource utilization among smaller age and medical risk groups. This could be particularly important since studies have shown that a small number of children often account for a disproportionate share of medical care expenditures [8,29,30]. Moreover, the top 10% of medical costs observed in children with complex conditions requiring high levels of care, may be much higher (approximately $5402) [8], than the average cost estimated in our study ($3990) and yet lower than those reported
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
in other influenza cost studies [29,30]. In addition, data for this study was collected during the 2003–2004 influenza season, a season characterized by predominant circulation of influenza A (H3N2) virus and high pediatric morbidity and mortality [31]. It is possible that illness severity was greater in this year compared to less severe influenza seasons or seasons when seasonal influenza A (H1N1) or influenza B predominated [16]. In conclusion, this study found the economic impact of medically attended influenza in children and their families to be substantial and much higher than previously estimated. These data emphasize the benefits of preventive interventions and should contribute to broadening the acceptance of influenza vaccination. Acknowledgements We acknowledge the collaboration of Michol Holloway, Meredith E. Tabangin, Vanessa Florian, Linda Jamison and Diana Henderson from Cincinnati Children’s Hospital Medical Center; Christina Albertin, Geraldine Lofthus and Ken Schnabel from the University of Rochester School of Medicine and Dentistry; Carol Ann Clay, Diane Kent, Yuwei Zhu, Erin Keckley and Nayleen Whitehead from the Vanderbilt University School of Medicine. Clinical, cost and epidemiologic data were provided by the New Vaccine Surveillance Network (NVSN). Conflict of interest: MRG reports receiving grant funding from MedImmune. KME receives funding from NIH to conduct influenza vaccines studies. All other authors, no conflict of interest. Funding: This research was completed while the primary authors were employees of the Centers for Disease Control and Prevention (CDC). Initial support was provided by the National Vaccine Program Office (NVPO) of the DHHS. Appendix A. New Vaccine Surveillance Network (NVSN) Team Nashville: Kathryn M. Edwards, Marie R. Griffin, Katherine A. Poehling, Diane Kent, Ann Clay, Mariah Daley, Carolyn Cooper, Stephanie Mayer, Yuwei Zhu, John Williams, Jody Peters, Amy Herrygers. Rochester: Peter Szilagyi, Caroline B. Hall, Geoffrey A. Weinberg, Laura Shone, Stanley Schaffer, Kenneth Schnabel, Gerry Lofthus, Sandra Ambrose, Christina Albertin, Linda Anderson, Andrea Marino, Rebecca Martinez, Gladys Lathan, Charlene Freundlich, Doreen Francis. Cincinnati: Mary Allen Staat, Gerry Fairbrother, Ardythe Morrow, David Bernstein, Rick Hornung, Linda Jamison, Emilie Grube, Vanessa Wimmer, David Witte, Pam Groen, Joel Mortensen, Cindi Ventrola, Michol Holloway and Diana Henderson. CDC: Marika Iwane, Jim Alexander, Jennifer Reuer, Iddrisu Sulemana, Ranee Seither, France Walker, Benjamin Schwartz, Carolyn Bridges. References [1] Molinari NA, Ortega-Sanchez IR, Messonnier ML, Thompson WW, Wortley PM, Weintraub E, et al. The annual impact of seasonal influenza in the US: measuring disease burden and costs. Vaccine 2007;25(June (27)):5086–96. [2] Thompson WW, Shay DK, Weintraub E, Brammer L, Bridges CB, Cox NJ, et al. Influenza-associated hospitalizations in the United States. JAMA 2004;292(11):1333–40. [3] Thompson WW, Comanor L, Shay DK. Epidemiology of seasonal influenza: use of surveillance data and statistical models to estimate the burden of disease. Journal of Infectious Diseases 2006;194(Suppl. 2 (November)):S82–91 [Review]. [4] O’Brien MA, Uyeki TM, Shay DK, Thompson WW, Kleinman K, McAdam A, et al. Incidence of outpatient visits and hospitalizations related to influenza in infants and young children. Pediatrics 2004;113:585–93.
4181
[5] Neuzil KM, Mellen BG, Wright PF, Mitchel Jr EF, Griffin MR. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. New England Journal of Medicine 2000;342:225–31. [6] Neuzil KM, Zhu Y, Griffin MR, Edwards KM, Thompson JM, Tollefson SJ, et al. Burden of interpandemic influenza in children younger than 5 years: a 25-year prospective study. Journal of Infectious Diseases 2002;185:147–52. [7] U.S. Census Bureau Annual Estimates of the Resident Population by Sex and Five-Year Age Groups for the United States: April 1, 2000 to July 1, 2008 (NC-EST2008-01). http://www.census.gov/popest/national/asrh/NC-EST2008sa.html [accessed October 2009]. [8] Fairbrother G, Cassedy A, Ortega-Sanchez IR, Szilagyi PG, Edwards KM, Molinari NA, et al. High costs of influenza: direct medical costs of influenza disease in young children. Vaccine 2010;28(July (31)):4913–9 [Epub 2010 May 31]. [9] Ampofo K, Gesteland PH, Bender J, Mills M, Daly J, Samore M, Byington C, Pavia AT, Srivastava R. Epidemiology, complications, and cost of hospitalization in children with laboratory-confirmed influenza infection. Pediatrics 2006;118(December (6)):2409–17. [10] Iskander M, Booy R, Lambert S. The burden of influenza in children. Current Opinion in Infectious Diseases 2007;20(June (3)):259–63 [Review]. [11] Meltzer MI, Neuzil KM, Griffin MR, Fukuda K. An economic analysis of annual influenza vaccination of children. Vaccine 2005;23(January (8)):1004–14. [12] Prosser LA, Bridges CB, Uyeki TM, Hinrichsen VL, Meltzer MI, Molinari NA, Schwartz B, Thompson WW, Fukuda K, Lieu TA. Health benefits, risks, and costeffectiveness of influenza vaccination of children. Emerging Infectious Diseases 2006;12(October (10)):1548–58. [13] Pisu M, Meltzer MI, Hurwitz ES, Haber M. Household-based costs and benefits of vaccinating healthy children in daycare against influenza virus: results from a pilot study. Pharmacoeconomics 2005;23(1):55–67. [14] Luce B, Nichol KL, Belshe RB, Frick KD, Li SX, Boscoe A, et al. Cost-effectiveness of live attenuated influenza vaccine versus inactivated influenza vaccine among children aged 24–59 months in the United States. Vaccine 2008;26(June (23)):2841–8. [15] Luce BR, Zangwill KM, Palmer CS, Mendelman PM, Yan L, Wolff MC, Cho I, Marcy SM, Iacuzio D, Belshe RB. Cost-effectiveness analysis of an intranasal influenza vaccine for the prevention of influenza in healthy children. Pediatrics 2001;108(August (2)):E24. [16] Poehling KA, Edwards KM, Weinberg GA, Szilagyi P, Staat MA, Iwane MK, et al. The underrecognized burden of influenza in young children. New England Journal of Medicine 2006;355(July (1)):31–40. [17] Gold MR, Siegel JE, Russell LB, Weinstein MC, editors. A practical guide to costeffectiveness in health and medicine. New York: Oxford University Press; 1996. [18] Haddix AE, Teutsch SM, Corso PS, editors. Prevention effectiveness: a guide to decision analysis and economic evaluation. New York: Oxford University Press; 2003. [19] U.S. Bureau of Labor Statistics. National Compensatory Survey; December 2007–January 2009. Available from: http://stats.bls.gov/ncs/ [accessed October 2009]. [20] US Bureau of Labor Statistics. ConsumerPrice Index. Available from: http://stats.bls.gov/cpi/ [accessed October 2009]. [21] US Bureau of Economic Analysis. Price Indexes for Gross Domestic Product. National Income and Product Accounts. US Department of Commerce. Available from: http://www.bea.gov/index.htm [accessed October 2009]. [22] Xie J, Priebe CE. Generalizing the Mann–Whitney–Wilcoxon statistic. Journal of Nonparametric Statistics 2000;12:5. [23] Sheskin DJ. Handbook of parametric and nonparametric statistical procedures. Florida: CRC Press; 2004. [24] Li S, Leader S. Economic burden and absenteeism from influenza-like illness in healthy households with children (5–17 years) in the US. Respiratory Medicine 2007;101(June (6)):1244–50 [Epub 2006 December 6]. [25] Heikkinen T, Silvennoinen H, Peltola V, Ziegler T, Vainionpaa R, Vuorinen T, et al. Burden of influenza in children in the community. Journal of Infectious Diseases 2004;190:1369–73. [26] Hibbert CL, Piedra PA, McLaurin KK, Vesikari T, Mauskopf J, Mahadevia PJ. Cost-effectiveness of live-attenuated influenza vaccine, trivalent in preventing influenza in young children attending day-care centres. Vaccine 2007;25(November (47)):8010–20. [27] Cohen GM, Nettleman MD. Economic impact of influenza vaccination in preschool children. Pediatrics 2000;106(November (5)):973–6. [28] Iskander M, Kesson A, Dwyer D, Rost L, Pym M, Wang H, et al. The burden of influenza in children under 5 years admitted to the Children’s Hospital at Westmead in the winter of 2006. Journal of Paediatrics and Child Health 2009;(November) [Epub ahead of print]. [29] Hall JL, Katz BZ. Cost of influenza hospitalization at a tertiary care children’s hospital and its impact on the cost-benefit analysis of the recommendation for universal influenza immunization in children age 6 to 23 months. Journal of Pediatrics 2005;147(December (6)):807–11. [30] Keren R. Flu-related hospitalization in children. LDI Issue Brief. 2006 December–2007;12(January (3)):1–4. [31] Bhat N, Wright JG, Broder KR, Murray EL, Greenberg ME, Glover MJ, et al. Influenza-associated deaths among children in the United States, 2003–2004. New England Journal of Medicine 2005;353(December (24)):2559–67. [32] Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices. MMWR 2006;55(RR10):1–42.
Contents lists available at SciVerse ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Indirect, out-of-pocket and medical costs from influenza-related illness in young children夽 Ismael R. Ortega-Sanchez a,∗,1 , Noelle-Angelique M. Molinari a,1 , Gerry Fairbrother b,1 , Peter G. Szilagyi c,1 , Kathryn M. Edwards d,1 , Marie R. Griffin d,1 , Amy Cassedy b,1 , Katherine A. Poehling d,1 , Carolyn Bridges a,1 , Mary Allen Staat b,1 a
Centers for Disease Control and Prevention, United States Cincinnati Children’s Hospital Medical Center, United States c University of Rochester, School of Medicine and Dentistry, United States d Vanderbilt University, School of Medicine, United States b
a r t i c l e
i n f o
Article history: Received 2 January 2012 Received in revised form 9 April 2012 Accepted 16 April 2012 Available online 27 April 2012 Keywords: Influenza Children Out-of-pocket cost Indirect cost
a b s t r a c t Background: Studies have documented direct medical costs of influenza-related illness in young children, however little is known about the out-of-pocket and indirect costs (e.g., missed work time) incurred by caregivers of children with medically attended influenza. Objective: To determine the indirect, out-of-pocket (OOP), and direct medical costs of laboratoryconfirmed medically attended influenza illness among young children. Methods: Using a population-based surveillance network, we evaluated a representative group of children aged <5 years with laboratory-confirmed, medically attended influenza during the 2003–2004 season. Children hospitalized or seen in emergency department (ED) or outpatient settings in surveillance counties with laboratory-confirmed influenza were identified and data were collected from medical records, accounting databases, and follow-up interviews with caregivers. Outcome measures included work time missed, OOP expenses (e.g., over-the-counter medicines, travel expenses), and direct medical costs. Costs were estimated (in 2009 US Dollars) and comparisons were made among children with and without high risk conditions for influenza-related complications. Results: Data were obtained from 67 inpatients, 121 ED patients and 92 outpatients with laboratoryconfirmed influenza. Caregivers of hospitalized children missed an average of 73 work hours (estimated cost $1456); caregivers of children seen in the ED and outpatient clinics missed 19 ($383) and 11 work hours ($222), respectively. Average OOP expenses were $178, $125 and $52 for inpatients, ED-patients and outpatients, respectively. OOP and indirect costs were similar between those with and without high risk conditions (p > 0.10). Medical costs totaled $3990 for inpatients and $730 for ED-patients. Conclusions: Out-of-pocket and indirect costs of laboratory-confirmed and medically attended influenza in young children are substantial and support the benefits of vaccination. Published by Elsevier Ltd.
1. Introduction Seasonal influenza, estimated to result in over 200,000 hospitalizations and 31 million outpatient visits annually in the United States (US) [1,2], disproportionally affects young children and the
Abbreviations: OOP, out-of-pocket costs; ED, emergency department; NVSN, New Vaccine Surveillance Network; IQR, inter-quartile range. 夽 The findings and conclusions expressed are those of the authors and do not necessarily represent the official views of the CDC or DHHS. ∗ Corresponding author at: Centers for Disease Control and Prevention, 1600 Clifton Rd., NE MS-A-34, Atlanta, GA 30333, United States. Tel.: +1 404 639 2834; fax: +1 404 639 4960. E-mail address: [email protected] (I.R. Ortega-Sanchez). 1 For the New Vaccine Surveillance Network (NVSN). See Appendix A. 0264-410X/$ – see front matter. Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.vaccine.2012.04.057
elderly [2–6]. Approximately 20% of all influenza-attributable hospitalizations and 12% of all outpatient visits are among children aged ≤5 years [1], although they comprise only 7% of population [7]. Direct medical costs of influenza illness in US children have been estimated to be >$1.2 billion per year [1]. Although these costs are high, they nevertheless underestimate the overall economic burden of influenza disease. Influenza disease also has an economic impact on family caregivers: family caregivers miss work time (i.e., indirect costs) and incur illness-related out-of-pocket (OOP) expenses such as co-payments, travel and lodging costs, over-the-counter medications and other expenses. Although data are available regarding medical costs for laboratory-confirmed influenza [8,9], limited economic data are available regarding work time lost by caregivers [10–12], costs incurred in medical settings [13], or OOP costs [11]
4176
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
incurred by families of children with influenza. When assessing the economic impact of the benefits of influenza vaccination on families, most societal cost analyses and cost-effective analyses have relied primarily on ad hoc assumptions or extrapolations from studies of other diseases [1,11–15]. The objective of our study was to estimate the indirect and OOP costs to caregivers of children <5 years with laboratory-confirmed medically attended influenza. We also examined the direct medical costs attributable to these influenza cases. We compared the medical and non-medical resource utilization rates and associated costs by type of medical care for children seen in the hospital, ED, and outpatient settings.
2. Methods 2.1. Overview The CDC-funded New Vaccine Surveillance Network (NVSN) was used to evaluate a representative group of children aged <5 years with laboratory-confirmed medically attended influenza during the 2003–2004 influenza season previously described [16]. Briefly, county children with acute respiratory infection and/or fever who were hospitalized or seen in emergency department (ED) and outpatient settings were enrolled. Enrolled children had nasal and throat swabs collected and tested for influenza by viral culture and polymerase chain reaction. For this study, enrolled county children with laboratory-confirmed influenza were identified and data were collected from medical chart reviews, medical billing records, and follow-up interviews with parents or caregivers. For the 2003–2004 study year, additional questions and telephone follow-up calls with parents or guardians were conducted to collect additional information on costs of illness. Data collection and analyses included rates of healthcare resource utilization (e.g., days of hospitalization, number of doctor visits), time missed by caregivers, OOP expenses (e.g., co-pays, travel expenses) and medical costs. Rates and costs were estimated by the highest level of medical care received (see description below) and compared across high-risk conditions for influenza-related complications as designated by the Advisory Committee on Immunization Practices.
2.2. Data 2.2.1. Surveillance and enrollment NVSN conducted influenza surveillance in three US counties, Monroe County, New York (Rochester), Davidson County, Tennessee (Nashville), and Hamilton County, Ohio (Cincinnati). After informed consent, 85% of eligible county children who were admitted with a diagnosis of acute respiratory tract infection and/or fever were enrolled [16]. Children were excluded if they had respiratory symptoms >14 days, fever and neutropenia associated with chemotherapy, were hospitalized in the prior 4 days or transferred from another surveillance hospital, or were newborns never discharged from the hospital. Surveillance hospitals captured ≥95% of all county children hospitalized with acute respiratory illness or fever [8,16]. County children presenting to selected clinics and emergency departments with symptoms of acute respiratory tract infection or fever were enrolled [8,16]. Children were enrolled 1–2 days per week in the clinic and 3–4 days per week in the only ED associated with a children’s hospital in each county with systematic rotation of days and shifts to obtain a representative sample. The percentage of all visits attributable to acute respiratory tract infection or fever for county children seen at these emergency departments was 30% in Nashville, 70% in Rochester, and 95% in Cincinnati [16].
All children had nasal and throat swabs obtained for influenza testing. Children were determined to have laboratory-confirmed influenza if these nasal and throat swabs were positive for influenza by duplicate reverse-transcriptase-polymerase-chain reaction (RTPCR) test or viral culture [16]. 2.3. Caregivers’ interview Parents or guardians of children with laboratory-confirmed influenza were interviewed by telephone three to four weeks after hospitalization discharge or ED or clinic visit. Multiple attempts to contact caregivers were made with an overall response rate of 84.6%. Reasons for refusal to participate in the study were not collected. Data encompassed caregivers’ time lost from work and expenses related to the influenza illness, including illness-related expenses that may have occurred two weeks prior to and three weeks after hospitalization, or one week before through one week after an ED or clinic visit. These data were collected using a standardized questionnaire along with caregivers’ employment status, income and work time shifts (see Fig. 1, details are also included below). 2.3.1. Medical charts and billing records Information on demographics, clinical procedures and medical resources used were collected by reviewing medical charts of enrolled patients. Detailed medical cost data for inpatients and ED patients were also provided by each surveillance site. Sources of medical cost data included summary billing records from unified billing forms (known as UB-92s) and similar accounting systems containing both costs and charges. In all cases, these medical data contained detailed line-item charges for all diagnostic tests, therapeutics, supplies, and room fees. Although medical billing data for outpatients were not gathered, data on medical resource utilization were collected during chart reviews at outpatient clinics. 2.4. Analysis Rates of healthcare utilization and associated costs per influenza patient were estimated following standard procedures [17,18] and using patient-specific data collected from medical billing records, medical chart reviews and follow-up interviews with patient caregivers. Using the highest level of medical care received by influenza patients, the data were then categorized into three mutually exclusive levels of care: inpatients, ED-patients and outpatients [1,11,16]. Estimates for each level of care included the resources and costs of the lower level of care, if applicable. For example, data for inpatients included not only hospitalization but also ED and outpatient care that might have occurred before or after hospitalization; and data for ED patients also include outpatient care, if any. Fig. 1 describes the categories of resource utilization and costs by medical care setting and source of data. We also collected information on the number of clinic or ED visits, the length of stay when hospitalized and the number of workdays lost by caregivers (Fig. 1). The number of medical procedures, including radiological procedures and laboratory tests were also assessed. 2.4.1. Cost categories Costs were grouped into three categories: (1) OOP expenses (medical and non-medical payments or co-payments, over-thecounter drugs, travel and phone call expenses related to the child’s illness), (2) indirect costs (wages and income associated with time missed from work during the child’s illness) and (3) medical costs. Medical costs were grouped into four subcategories: diagnostics (e.g., radiology, laboratory), therapeutics (e.g., pharmaceuticals, blood products, intensive care therapies), room fees and supplies (e.g., operating, recovery and ward rooms, ICU, sedation, ED units,
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
4177
Laboratory-confirmed influenza patients enrolled from active surveillance
ED-patients
Outpatients
Inpatients
*
Resource utilization
Costs
*
DIRECT From medical chart review and billing records -Outpatient and ED care (visits) -Hospitalization (days) -Medical procedures (number) -Laboratory (tests) -Radiology (x-rays) From caregivers’ From medical interview billing records Out-of-pocket costs Medical costs -Medical copayments -Room & supplies -Physician fees -Non medical payments -Pharmacy -Over-the-counter drugs -Telephone calls -Laboratory -Travel and logging -Radiology
INDIRECT From caregivers’ interview -Caregivers time off from work, school or other activities -Caregivers time off for visits to or stays in healthcare institutions
From caregivers’ interview -Value of work time off -Value of time off for Dr Visits -Value of time off for Hospital or ED
*Although specific to the level of medical care, descriptions of the types and sources of data inED-patients and Outpatients are similar to those described in the table for inpatients. No medical billing data were collected for outpatients. Fig. 1. Categories of resource utilization and direct and indirect costs by type of cost, medical care setting and source of data.
and transportation), and physician services (e.g., fees from visits, consultations, and procedures) (see Fig. 1). Since a specific physician cost-to-charge ratio was not available from hospitals, fees from physician services as reported in medical billing records were included after adjusting them with a generic cost-to-charge ratio [8]. Unlike physician fees, all other medical costs were adjusted using hospital- and department-specific cost-to-charge ratios. All costs were later adjusted for inflation to December 2009 US dollars using the Medical Component of the Consumer Price Index [20] for medical cost, and the Gross Domestic Product deflator index [21] for OOP and indirect costs. 2.4.2. Valuing time missed Caregivers were also asked about the uses and sources of time lost at the household level: the number of hours allocated to doctor visits, hospitalization stays or home care during the child’s illness was used to account for the uses of caregivers’ time. For caregivers, efforts were made to identify whether time lost was from work, school or college, or other activities.
Although efforts were made to collect data on household income and then calculate the prorated hourly earnings for time lost to influenza, due to the limited response to income questions we relied on an indirect approach. To estimate the value of caregivers’ time off work (i.e., indirect costs) we used non-occupation-specific hourly earnings data for individuals belonging to the surveillance sites (i.e., $17–$22 per hour) as reported in the National Compensation Survey from the by the US Bureau of Labor Statistics [19]. 2.4.3. Statistical analysis Means, medians, and inter-quartile range (IQR) values of resource utilization rates and costs for each influenza final outcome were calculated. Estimates within each outcome were differentiated by patient risk and age group. Because previous and preliminary univariate analyses showed skewed costs by a few cases with extremely high medical costs [8], standard non-parametric analyses of variance (Mann–Whitney–Wilcoxon [22,23] and Kolmogorov–Smirnov score tests [23]) were used for comparisons among categories and subcategories of resource utilization
4178
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
Table 1 Characteristics of enrolled children with laboratory-confirmed influenza. Influenza cases N Age 0–5 months 6–23 months 24–59 months Gender Female Male Risk of complicationb High Non high Race/ethnic group White Black Other or unknown Health insurance Uninsured Public Insurance Private Insurance Household income <$10,000 $10,000 to <$30,000 $30,000 to <$50,000 $50,000 to <$100,000 ≥$100,000 Refused, unknown, missing
Inpatients (%)
ED-patients (%)
Outpatients (%)
67
122
92
30 (48)a 18 (29) 14 (23)
16 (13)a 48 (39) 58 (48)
9 (10)a 38 (41) 45 (49)
41 (61) 26 (39)
50 (41) 72 (59)
50 (54) 42 (46)
14 (21) 53 (79)
27 (22) 94 (78)
21 (23) 71 (77)
32 (51) 27 (43) 4 (6)
41 (34) 53 (43) 28 (23)
26 (32) 35 (43) 19 (24)
3 (5) 30 (45) 34 (51)
6 (5) 81 (68) 33 (28)
6 (7) 55 (63) 27 (31)
25 (37) 13 (19) 7 (10) 9 (13) 4 (6) 11 (16)
29 (24) 37 (30) 9 (8) 8 (7) 8 (7) 35 (29)
29 (32) 18 (20) 8 (9) 11 (12) 5 (5) 27 (29)
a
Percent figures do not add to 100 due to rounding. High risk (HR) for influenza related complications is based on the CDC Advisory Committee on Immunization Practices definition [32]. Conditions included asthma, cancer, diabetes, heart disease, immune deficiency, kidney disorders, sickle cell disease, cystic fibrosis, bronchopulmonary dysplasia, blood disorders, liver disorders, and seizure disorders. b
rates and costs. The exact form of non-parametric tests was performed to obtain a higher accuracy on statistical differences among subgroups. 3. Results Data were collected from 67 hospitalized patients (i.e., inpatients), 122 ED-patients, and 92 clinic patients (i.e., outpatients) with laboratory-confirmed influenza cases during the 2003–2004 influenza winter season (Table 1). No deaths were detected among the surveillance population. Nearly half (48%) of inpatients were less than 6 months old, while almost half of ED-patients (48%) and outpatients (49%) were 24–59 months old. A high proportion had public health insurance (i.e., Medicaid) and had household incomes of less than $30,000 (Table 1). 3.1. Indirect costs About 75% of families with an inpatient stay reported work time lost during the child’s influenza illness (Table 2). Fewer families reported time off from school or college (6–10%) or from other unpaid activities (10% for inpatients, 18% for ED-patients, and 14% for outpatients) (Table 2). In all cases, caregivers cared for patients at home, escorted them to medical visits, spent time in hospitals (if child was hospitalized) and provided help in various other forms (e.g., taking care of other non-sick kids, making runs to pharmacies, etc.). Caregivers of hospitalized children with influenza reported, on average, 73 h of work time lost, mostly for hospital stays (Table 3). ED patients averaged 19 h of caregivers’ work time lost, mostly due to ED visits while outpatients also averaged at least 7 h of caregivers’ work time lost (Table 3). The average indirect cost (e.g., caregiver time cost) for inpatients was estimated to be $1456 (median $1089; IQR $1319), mainly due to work time missed during hospital stays (82%) (Table 3). Indirect cost incurred by caregivers of ED-patients was $383 (median
$125; IQR 268) and by caregivers of outpatients was $222 (median $85; IQR 280). No statistical difference was found among indirect cost between patients with identified high risk conditions and those without (p > 0.427 for inpatients, p > 0.486 for ED-patients, p > 0.137 for outpatients) (Table 4). Similarly, among children aged ≤23 months versus 24–59 months no statistical difference was found in inpatients ($1321 versus $1525, p > 0.323) and outpatients ($257 versus $198, p > 0.150); and only ED-patients reported a marginal statistical difference by age group ($503 versus $217, p > 0.022).
3.2. Out-of-pocket costs Approximately 55% of families of inpatients reported medical payments, over-the-counter medication and travel costs related to the child’s influenza illness. Expenses were also reported by 71% and 63% of families of ED-patients and outpatients, respectively (Table 2). The average OOP cost among all inpatients was approximately $178 (median $50; IQR 84) with medical payments (>43% of total OOP), travel expenditures (>24%) and phone calls costs (15%) as principal components (Table 3). For ED-patients, average OOP costs were $125 (median $35, IQR 57) mainly due to travel (53%) and prepaid phone card, internet fee or payphone call expenses (26%). The average OOP cost for outpatients were approximately $52 (median $29, IQR 56) again due to travel (32%) and phone call expenses (20%). The estimated medians for OOP payments in patients were not statistically different (p > 0.10) between high-risk and non-high risk groups (not shown). Although variations in components of OOP costs (i.e., copays and travel costs) were observed across age groups, analyses showed no statistical differences in the total OOP cost among inpatients aged ≤23 months versus 24–59 months ($182 versus $162, p > 0.133), and outpatients ($61 versus $39, p > 0.438). Only ED-patients showed a statistical difference ($181 versus $63, p > 0.052).
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
4179
Table 2 Types of costs incurred among enrolled influenza cases by level of medical care. ED-patients n (%)a
Outpatients n (%)a
122
92
50 (75) 4 (6) 7 (10)
57 (47) 8 (7) 22 (18)
39 (42) 9 (10) 13 (14)
37 (55) 17 (25) 37 (55) 17 (25) 34 (51)
89 (71) 17 (14) 89 (71) 35 (29) 88 (62)
58 (63) 18 (20) 58 (63) 29 (32) 58 (63)
Inpatients n (%)a N Caregivers reporting time off From work From school From unpaid activity Out-of-pocket costs Medical Nonmedical Over the counter Telephone costs Travel costs Healthcare utilizationb Outpatient care Hospitalization Medical proceduresc Laboratoryd Radiology
67
67 (100) 67 (100) 60 (90) 50 (75) 34 (51)
122 (100) n/a 47 (39) 3 (3) 31 (25)
92 (100) n/a 4 (4) 1 (1) 4 (4)
Medical costs
67 (100)
122 (100)
n/a
a
Percent figures represent proportion of caregivers reporting specific economic data. Data include patients utilizing at least one unit by specific healthcare resource during influenza illness. c Medical procedures include reports of mechanical ventilation, oxygen, antiviral drug treatment, treatment of bacterial pneumonia, ear infection, or sinusitis. Medical procedures may also include treatments for influenza patients with asthma, bronchopulmonary dysplasia, cystic fibrosis, heart disease, immune deficiency, kidney disorders, and seizure disorders. d Laboratory tests could include a single or multiple blood draws with multiple measures. Note: Healthcare utilization data were collected from medical records while medical costs data were from billing records. No medical costs were collected for outpatients. b
Table 3 Out-of-pocket and caregivers time costs per child with influenza by level of final medical care.
Out-of-pocket costs ($) Total out-of-pocket costs Medical payments* , a Non-medical payments* , b Over-the-counter drugs** Telephone calls* Travel and logging * , c Caregivers time off (h) Total time off from work* Time off for Dr visits** Time off for hospital or ED visits* Caregivers time costs ($)* Total value of work time off Value of time for Dr visits Value of time for hospital or ED
Inpatients
ED-patients
Outpatients
n = 67
n = 122
n = 92
Mean
(SD)
178 78 18 9 29 44
(535) (418) (31) (13) (98) (137)
50 5 0 5 0 10
(68) (13) (62)
73 8 61 1456 164 1201
(1395) (277) (1267)
Median
IQR
Mean
(SD)
84 15 25 15 6 19
125 11 4 11 33 67
(348) (13) (16) (13) (151) (200)
55 4 48
66 9 65
19 7 6
1089 80 965
1320 174 1202
383 146 127
(42) (10) (31) (903) (211) (670)
Median
IQR
Mean
(SD)
35 7 0 7 0 10
57 15 0 15 4 19
52 10 4 10 17 11
(144) (15) (14) (15) (102) (15)
6 5 0
13 7 0
11 7 –
(17) (14) –
125 98 0
268 135 0
222 131 –
(366) (290) –
Median
IQR
29 6 0 6 0 4
56 15 0 15 4 13
0 3 –
15 5 –
85 50 –
280 104 –
a Medical payments may include copayments for medical visits or prescriptions, cost of additional laboratory test or X-rays, cost of traditional or homeopathic medicines or treatments. b Non-medical payments may include hired childcare fees for sick child or for other healthy children while taking care of sick child, home conditioning and cleaning for sick child. c Travel and logging may include expenses for gas, tolls, parking bus fares, taxi fares and hotel room fees. * Non significant (p > 0.10), all Kruskal–Wallis tests: one-way analysis of variance by ranks to compare median values by medical care setting (Pr > Chi square). ** p < 0.05, all Kruskal–Wallis tests: one-way analysis of variance by ranks to compare median values by medical care setting (Pr > Chi square).
3.3. Medical costs Inpatient medical costs were approximately $3990 (median $2929; IQR $1470) with room fees and supplies (55%) and physician fees (23%) as principal components (Table 4). No statistically significant difference was found in medical costs of high risk versus non-HR inpatients. Average medical costs of ED-patients were approximately $730 (median $402; IQR 274) with room and supplies (>50%) and other physician fees (>35%) as principal cost components (Table 4). Overall, median medical costs of ED-patients were not statistically different across risk groups ($274 for high risk versus $260 for non-high risk; p > 0.10). Analyses showed no statistical differences (p > 0.585) in medical costs among inpatients aged <23 months ($4239) and 24–59 months ($3294). On other hand,
marginally significant statistical differences in medical cost among ED-patients aged ≤23 months ($935) and 24–59 months ($502) were observed (p > 0.069).
4. Discussion This is the first study to estimate the OOP and indirect costs of laboratory-confirmed medically attended influenza illnesses among young children. We found a substantial economic burden on families as a result of medically attended influenza illnesses in children aged <5 years in addition to already substantial medical costs [1]. Caregivers lost on average 73 h of work for inpatients, 19 h for ED-patients and 6 h for outpatients with influenza. As a result,
4180
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
Table 4 Medical resource utilization and costs per child with influenza by medical care setting and risk group. All patients, n = 280
Inpatient Medical resources Outpatient care (visits) Hospitalization (days) Medical proceduresb Laboratory (tests) Radiology (X-rays) Medical facility stay related costs ($)c Room and supplies Physician fees ED patient Medical resources Outpatient care (visits) Medical proceduresb Laboratory (tests) Radiology (X-rays) Medical facility visit related costs ($)c Room and supplies Physician fees Outpatientd Healthcare utilization Outpatient care (visits) Medical proceduresb Laboratory (tests) Radiology (X-rays)
Non high risk, n = 218
p valuea
High risk, n = 62
Mean
Median
IQR
Mean
Median
IQR
Mean
Median
IQR
1.50 3.10 10.70 5.00 0.80 3990 2186 899
1 3 2 4 1 2929 1646 585
1 1 4 8 1 1470 1603 555
1.50 3.20 10.80 6.10 0.80 4296 2308 966
1 3 3 6 1 2927 1530 606
1 1 4 7 1 1573 1803 576
1.80 2.60 10.40 0.90 0.70 2830 1725 644
1 2 1 0 1 3115 1790 572
1 1 2 1 1 1056 399 382
0.279 0.063 0.044 <0.001 0.321 0.436 0.309 0.439
1.40 2.70 0.10 0.30 730 370 262
1 0 0 0 402 209 166
1 1 0 1 307 302 128
1.40 2.00 0.10 0.30 796 414 260
1 0 0 0 398 209 166
0 1 0 1 359 315 128
1.40 5.30 0.10 0.30 517 223 274
1 0 0 0 406 209 175
1 1 0 1 154 302 128
0.451 0.259 0.335 0.446 0.480 0.055 0.083
1.27 0.09 0.02 0.04
1 0 0 0
1 0 0 0
1.33 0.06 0.01 0.03
1 0 0 0
1 0 0 0
1.10 0.20 0.00 0.10
1 0 0 0
1 0 0 0
0.194 0.254 0.501 0.438
a
Mann–Whitney test: two-sample one-side nonparametric test of ranks to compare median values by influenza risk condition. The number of medical procedures reported includes mechanical ventilation, oxygen, antiviral drug treatment, treatment of bacterial pneumonia, ear infection, or sinusitis. Medical procedures in inpatients or Ed-patients may also include treatments for asthma, bronchopulmonary dysplasia, cystic fibrosis, heart disease, immune deficiency, kidney disorders, and seizure disorders. c Total costs and two main component costs for hospitalized and ED-patients are shown. The two main component costs are part of the total medical facility costs. These costs data were obtained through review of billing records. d No medical cost from billing data were collected for outpatients. b
wages and self-employed income lost for households amounted to $1456, $383 and $222 for inpatients, ED-patients and outpatients, respectively. Out-of-pocket costs averaged $178 for inpatients, $125 for ED-patients and $52 for outpatients. While these costs are considerable, they do not include the economic value of time lost from leisure or other activities such as attending school or college. Our estimates for indirect and OOP costs were substantially higher than previously estimates from other studies for influenza or other respiratory infections [11–15,24–26]. For example, two recent studies included assumptions that families lose 0.89 workdays (or 7 h) per ill school-aged child [24] and 1.9 workdays (15 h) per ill child attending day-care [25]. While these estimates are similar to our findings for patients seen in an outpatient and ED setting, they substantially underestimate the burden to parents for hospitalized children. Thus, prior studies that did not include such costs substantially underestimated the economic impact of medically attended cases of influenza among young children. Due to data limitations, most cost-effectiveness studies on vaccinating young children against influenza have imputed lower family expenses than those estimated in this study. For example, for each day of hospitalization, two previous studies imputed less than one day of work missed by caregivers [11,12] compared with 2.5 days (i.e., 61 h or 7.6 work days missed divided by 3.1 days in hospital) estimated in our study (Tables 3 and 4). Consequently, the value of time lost per case calculated out of assumed time lost for caregivers of inpatients ($540) [11] were about half of those estimated in our study for hospitalized children (Table 3). In addition, OOP costs in previous studies was limited to medical co-pays for outpatient visits, $8, and over- the-counter drugs between $5 and $7 [11,26] while our study’s estimated OOP costs are substantially higher and include additional costs not captured in other studies (Table 3).
In line with our estimates of medical resource utilization, other studies of young children hospitalized with laboratory confirmed influenza have found a similar length of hospital stay of generally 2–3 days [9,27,28]. However, a wide variation exists in the literature in regards to hospitalization costs. Costs among non-high risk patients in our study were comparable with previous studies [11,29]. Nonetheless, for hospitalized children with more complex conditions, our study reported lower medical costs compared to those estimated for the same risk group in other studies [29,30]. The following strengths and limitations should be considered when interpreting the data we report. We used data collected prospectively from three geographic sites to identify children with laboratory confirmed influenza. Although participation in the surveillance system was high (over 70% of families consented to participate), costs may have been different among children whose families participated compared to those that did not. We obtained item-based medical and non-medical resource and cost data assembled with information from multiple sources, including medical records, billing data and interviews with caregivers. Recall bias may have occurred about OOP expenses and other indirect costs for families as caregivers were interviewed 3–4 weeks after the child’s medically attended illness. Cost data on daycare absenteeism beyond hospitalization or home care were not collected as these may have resulted in duplication of costs. In addition, sample size limited our ability to conduct detailed sub-analyses of influenza illness-related costs and resource utilization among smaller age and medical risk groups. This could be particularly important since studies have shown that a small number of children often account for a disproportionate share of medical care expenditures [8,29,30]. Moreover, the top 10% of medical costs observed in children with complex conditions requiring high levels of care, may be much higher (approximately $5402) [8], than the average cost estimated in our study ($3990) and yet lower than those reported
I.R. Ortega-Sanchez et al. / Vaccine 30 (2012) 4175–4181
in other influenza cost studies [29,30]. In addition, data for this study was collected during the 2003–2004 influenza season, a season characterized by predominant circulation of influenza A (H3N2) virus and high pediatric morbidity and mortality [31]. It is possible that illness severity was greater in this year compared to less severe influenza seasons or seasons when seasonal influenza A (H1N1) or influenza B predominated [16]. In conclusion, this study found the economic impact of medically attended influenza in children and their families to be substantial and much higher than previously estimated. These data emphasize the benefits of preventive interventions and should contribute to broadening the acceptance of influenza vaccination. Acknowledgements We acknowledge the collaboration of Michol Holloway, Meredith E. Tabangin, Vanessa Florian, Linda Jamison and Diana Henderson from Cincinnati Children’s Hospital Medical Center; Christina Albertin, Geraldine Lofthus and Ken Schnabel from the University of Rochester School of Medicine and Dentistry; Carol Ann Clay, Diane Kent, Yuwei Zhu, Erin Keckley and Nayleen Whitehead from the Vanderbilt University School of Medicine. Clinical, cost and epidemiologic data were provided by the New Vaccine Surveillance Network (NVSN). Conflict of interest: MRG reports receiving grant funding from MedImmune. KME receives funding from NIH to conduct influenza vaccines studies. All other authors, no conflict of interest. Funding: This research was completed while the primary authors were employees of the Centers for Disease Control and Prevention (CDC). Initial support was provided by the National Vaccine Program Office (NVPO) of the DHHS. Appendix A. New Vaccine Surveillance Network (NVSN) Team Nashville: Kathryn M. Edwards, Marie R. Griffin, Katherine A. Poehling, Diane Kent, Ann Clay, Mariah Daley, Carolyn Cooper, Stephanie Mayer, Yuwei Zhu, John Williams, Jody Peters, Amy Herrygers. Rochester: Peter Szilagyi, Caroline B. Hall, Geoffrey A. Weinberg, Laura Shone, Stanley Schaffer, Kenneth Schnabel, Gerry Lofthus, Sandra Ambrose, Christina Albertin, Linda Anderson, Andrea Marino, Rebecca Martinez, Gladys Lathan, Charlene Freundlich, Doreen Francis. Cincinnati: Mary Allen Staat, Gerry Fairbrother, Ardythe Morrow, David Bernstein, Rick Hornung, Linda Jamison, Emilie Grube, Vanessa Wimmer, David Witte, Pam Groen, Joel Mortensen, Cindi Ventrola, Michol Holloway and Diana Henderson. CDC: Marika Iwane, Jim Alexander, Jennifer Reuer, Iddrisu Sulemana, Ranee Seither, France Walker, Benjamin Schwartz, Carolyn Bridges. References [1] Molinari NA, Ortega-Sanchez IR, Messonnier ML, Thompson WW, Wortley PM, Weintraub E, et al. The annual impact of seasonal influenza in the US: measuring disease burden and costs. Vaccine 2007;25(June (27)):5086–96. [2] Thompson WW, Shay DK, Weintraub E, Brammer L, Bridges CB, Cox NJ, et al. Influenza-associated hospitalizations in the United States. JAMA 2004;292(11):1333–40. [3] Thompson WW, Comanor L, Shay DK. Epidemiology of seasonal influenza: use of surveillance data and statistical models to estimate the burden of disease. Journal of Infectious Diseases 2006;194(Suppl. 2 (November)):S82–91 [Review]. [4] O’Brien MA, Uyeki TM, Shay DK, Thompson WW, Kleinman K, McAdam A, et al. Incidence of outpatient visits and hospitalizations related to influenza in infants and young children. Pediatrics 2004;113:585–93.
4181
[5] Neuzil KM, Mellen BG, Wright PF, Mitchel Jr EF, Griffin MR. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. New England Journal of Medicine 2000;342:225–31. [6] Neuzil KM, Zhu Y, Griffin MR, Edwards KM, Thompson JM, Tollefson SJ, et al. Burden of interpandemic influenza in children younger than 5 years: a 25-year prospective study. Journal of Infectious Diseases 2002;185:147–52. [7] U.S. Census Bureau Annual Estimates of the Resident Population by Sex and Five-Year Age Groups for the United States: April 1, 2000 to July 1, 2008 (NC-EST2008-01). http://www.census.gov/popest/national/asrh/NC-EST2008sa.html [accessed October 2009]. [8] Fairbrother G, Cassedy A, Ortega-Sanchez IR, Szilagyi PG, Edwards KM, Molinari NA, et al. High costs of influenza: direct medical costs of influenza disease in young children. Vaccine 2010;28(July (31)):4913–9 [Epub 2010 May 31]. [9] Ampofo K, Gesteland PH, Bender J, Mills M, Daly J, Samore M, Byington C, Pavia AT, Srivastava R. Epidemiology, complications, and cost of hospitalization in children with laboratory-confirmed influenza infection. Pediatrics 2006;118(December (6)):2409–17. [10] Iskander M, Booy R, Lambert S. The burden of influenza in children. Current Opinion in Infectious Diseases 2007;20(June (3)):259–63 [Review]. [11] Meltzer MI, Neuzil KM, Griffin MR, Fukuda K. An economic analysis of annual influenza vaccination of children. Vaccine 2005;23(January (8)):1004–14. [12] Prosser LA, Bridges CB, Uyeki TM, Hinrichsen VL, Meltzer MI, Molinari NA, Schwartz B, Thompson WW, Fukuda K, Lieu TA. Health benefits, risks, and costeffectiveness of influenza vaccination of children. Emerging Infectious Diseases 2006;12(October (10)):1548–58. [13] Pisu M, Meltzer MI, Hurwitz ES, Haber M. Household-based costs and benefits of vaccinating healthy children in daycare against influenza virus: results from a pilot study. Pharmacoeconomics 2005;23(1):55–67. [14] Luce B, Nichol KL, Belshe RB, Frick KD, Li SX, Boscoe A, et al. Cost-effectiveness of live attenuated influenza vaccine versus inactivated influenza vaccine among children aged 24–59 months in the United States. Vaccine 2008;26(June (23)):2841–8. [15] Luce BR, Zangwill KM, Palmer CS, Mendelman PM, Yan L, Wolff MC, Cho I, Marcy SM, Iacuzio D, Belshe RB. Cost-effectiveness analysis of an intranasal influenza vaccine for the prevention of influenza in healthy children. Pediatrics 2001;108(August (2)):E24. [16] Poehling KA, Edwards KM, Weinberg GA, Szilagyi P, Staat MA, Iwane MK, et al. The underrecognized burden of influenza in young children. New England Journal of Medicine 2006;355(July (1)):31–40. [17] Gold MR, Siegel JE, Russell LB, Weinstein MC, editors. A practical guide to costeffectiveness in health and medicine. New York: Oxford University Press; 1996. [18] Haddix AE, Teutsch SM, Corso PS, editors. Prevention effectiveness: a guide to decision analysis and economic evaluation. New York: Oxford University Press; 2003. [19] U.S. Bureau of Labor Statistics. National Compensatory Survey; December 2007–January 2009. Available from: http://stats.bls.gov/ncs/ [accessed October 2009]. [20] US Bureau of Labor Statistics. ConsumerPrice Index. Available from: http://stats.bls.gov/cpi/ [accessed October 2009]. [21] US Bureau of Economic Analysis. Price Indexes for Gross Domestic Product. National Income and Product Accounts. US Department of Commerce. Available from: http://www.bea.gov/index.htm [accessed October 2009]. [22] Xie J, Priebe CE. Generalizing the Mann–Whitney–Wilcoxon statistic. Journal of Nonparametric Statistics 2000;12:5. [23] Sheskin DJ. Handbook of parametric and nonparametric statistical procedures. Florida: CRC Press; 2004. [24] Li S, Leader S. Economic burden and absenteeism from influenza-like illness in healthy households with children (5–17 years) in the US. Respiratory Medicine 2007;101(June (6)):1244–50 [Epub 2006 December 6]. [25] Heikkinen T, Silvennoinen H, Peltola V, Ziegler T, Vainionpaa R, Vuorinen T, et al. Burden of influenza in children in the community. Journal of Infectious Diseases 2004;190:1369–73. [26] Hibbert CL, Piedra PA, McLaurin KK, Vesikari T, Mauskopf J, Mahadevia PJ. Cost-effectiveness of live-attenuated influenza vaccine, trivalent in preventing influenza in young children attending day-care centres. Vaccine 2007;25(November (47)):8010–20. [27] Cohen GM, Nettleman MD. Economic impact of influenza vaccination in preschool children. Pediatrics 2000;106(November (5)):973–6. [28] Iskander M, Kesson A, Dwyer D, Rost L, Pym M, Wang H, et al. The burden of influenza in children under 5 years admitted to the Children’s Hospital at Westmead in the winter of 2006. Journal of Paediatrics and Child Health 2009;(November) [Epub ahead of print]. [29] Hall JL, Katz BZ. Cost of influenza hospitalization at a tertiary care children’s hospital and its impact on the cost-benefit analysis of the recommendation for universal influenza immunization in children age 6 to 23 months. Journal of Pediatrics 2005;147(December (6)):807–11. [30] Keren R. Flu-related hospitalization in children. LDI Issue Brief. 2006 December–2007;12(January (3)):1–4. [31] Bhat N, Wright JG, Broder KR, Murray EL, Greenberg ME, Glover MJ, et al. Influenza-associated deaths among children in the United States, 2003–2004. New England Journal of Medicine 2005;353(December (24)):2559–67. [32] Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices. MMWR 2006;55(RR10):1–42.