Respiratory viral infections are one of the next group of diseases likely to be targeted for prevention in childhood by the use of vaccines. To begin
collecting necessary epidemiology and cost information about the illnesses
caused by these viruses, we conducted a prospective cohort study in 118 Melbourne
children between 12 and 71 months of age during winter and spring 2001. We
were interested in calculating an average cost per episode of community-managed
acute respiratory disease, in identifying the key cost drivers of such illness,
and to identify the proportion of costs borne by the patient and family.
There were 202 community-managed influenza-like illnesses identified between
July and December 2001, generating 89 general practitioner visits, and 42
antibiotic prescriptions. The average cost of community-managed episodes
(without hospitalisation) was $241 (95% CI $191 to $291), with the key cost
drivers being carer time away from usual activities caring for the ill child
(70% of costs), use of non-prescription medications (5.4%), and general practice
visits (5.0%). The patient and family met 87 per cent of total costs. The
lowest average cost occurred in households from the highest income bracket.
Acute respiratory illness managed in the community is common, with the responsibility
for meeting the cost of episodes predominantly borne by the patient and family
in the form of lost productivity. These findings have implications for preventive
strategies in children, such as the individual use of, or implementation
of public programs using, currently available vaccines against influenza
and vaccines under development against other viral respiratory pathogens.
Commun Dis Intell 2004;28:509–516.
Top of page
For some time now there has been a divergence between what vaccines the
National Health and Medical Research Council (NHMRC), more recently on the
advice of the Australian Technical Advisory Group on Immunisation (ATAGI),
recommends Australians should receive, and what is paid for by the National
Immunisation Program (NIP). The NIP is a Commonwealth, State and Territory
Governments’ initiative that provides certain vaccines free of charge to
This divergence previously only applied to recommendations for older
Australians, in particular, influenza and 23-valent polysaccharide pneumococcal
vaccines. Influenza vaccine was first recommended for older Australians in
the third edition of, what is now called, the Australian Immunisation
Handbook in 19861 but was only funded
nationally in 1999.2 A general recommendation
for use of polysaccharide pneumococcal vaccine in older Australians was first
made in the fifth edition of the immunisation handbook (1994),3
and the Commonwealth Government has announced funding for a national program
to commence in 2005.4
But as Burgess and McIntyre reported recently,5
the release of the eighth edition of the handbook6
has seen this divergence between recommended and funded vaccines extend to
children. Varicella vaccine is on the Australian Standard Vaccination Schedule
at 18 months of age; there is a universal recommendation for a primary course
of the relatively expensive seven-valent conjugate pneumococcal vaccine;
and inactivated poliomyelitis vaccine is recommended when appropriate combination
vaccines become available. An infant program and a catch-up program for children
under the age of two years for pneumococcal vaccine commenced at the beginning
of 2005,4 but there is currently no provision
to fund either universal childhood varicella vaccination or a transition
to inactivated poliomyelitis vaccine.
This emerging discrepancy between recommended and funded vaccines is
only likely to widen. One of the next major groups of diseases preventable
by use of vaccines is likely to be the respiratory viral infections of childhood.
Injectable influenza vaccines are currently licensed for Australian children
down to the age of six months, but are currently recommended only for children
in high-risk groups.6 A trivalent, cold-adapted,
influenza vaccine (CAIV-T), containing live-attenuated virus and delivered
intranasally, was licensed in the United States of America (USA) in 2003
for healthy five to 49-year-olds,7 with the likelihood
of younger and older age indications in the future. Other vaccines against
respiratory viral infections, including respiratory syncytial virus and parainfluenza
viruses, are currently under development. Information about the epidemiology
and costs of acute community-based respiratory illness in children, particularly
those costs borne by the patient and family, are required to guide future
vaccine use and other control measures. Given the circulation patterns of
these viruses, particularly respiratory syncytial virus and influenza virus,
we collected information about respiratory illness in winter and spring.
We report here burden information for community-based respiratory illnesses
in urban Australian children, and use these to calculate an average cost
for these episodes.
Top of page
We conducted a prospective cohort study of healthy children in metropolitan
Melbourne, Victoria, between 1 July and 1 December 2001. The Royal Children’s
Hospital Ethics in Human Research Committee approved the study and written
informed consent was obtained from parents/guardians. Methods for this study
have been described elsewhere.8 Eligible children
were between 12 and 71 months of age at enrolment without pre-existing chronic
respiratory or other medical problems. Children aged between 12 and 23 months
were recruited largely via maternal and child health nurses (MCHNs) and immunisation
providers in 23 local council areas across greater Melbourne. We invited
participation from families with older children who had previously participated
in a (non-respiratory pathogen) vaccine study conducted by our group, and
also distributed flyers and posters through childcare centres. More than
one child per family could be enrolled. We collected household demographic
features at enrolment. Gross household income was collected in four brackets:
bracket 1, ≤ $21,000; bracket
2, $21,101 to $33,000; bracket 3, $33,001 to $56,000; and bracket 4, > $56,000.
Parents/guardians completed a symptom diary card for each day the child
was on the study. We designated an important respiratory illness to be an
influenza-like illness (ILI) using the criteria described by Belshe et
al in the CAIV-T efficacy study conducted in the USA during
1996 and 1997.9 An ILI was defined as having occurred
if a child had at least one category A symptom or at least two category B
symptoms (Table 1). All information about symptoms was from parental report
only. Individual episodes began on the first day on which there were sufficient
symptoms to meet the definition of an ILI, and finished on the final day
there were any documented symptoms associated with the ILI. A new episode
was deemed to have commenced if there were three or more symptom-free days
since the last day with any symptoms of the previous episode. Number and
duration of ILIs was ascertained; incidence rates were calculated using child-months
(person-time) as the denominator, and 95 per cent confidence intervals (CI)
were produced using the standard method for incidence rate data.10
Table 1. Defining symptoms of an influenza-like illness
| Category A symptoms
|| Category B symptoms
- fever (either identified without measurement or a measured temperature
of 37.6° C or higher by axillary thermometer)
- shortness of breath
- pulmonary congestion (moist cough)
- pneumonia (diagnosed by a healthcare provider)
- ear infection (suspected by parent/guardian or diagnosed by healthcare
- runny nose/nasal congestion
- sore throat
- muscle aches
- decreased activity (lethargy/weakness)
We used incident-based costing to derive an average cost of community-managed
episodes (not including illnesses in which there was a hospitalisation).
Once a child developed an ILI we asked parents to complete a burden diary
on healthcare use, travel costs seeking healthcare (including car used and
kilometres travelled), medication usage, investigations performed, time spent
seeking healthcare during the episode, and excess time spent caring for a
sick child—that is, time over and above that normally required for the care
of the child when well.
Cost data were calculated from a societal perspective using 2001–2002
financial year Australian dollar values (Table 2). Discounting is not relevant
as costs were collected in a single year. Direct and indirect costs were
included, and we allocated costs as being borne either by the patient and
family, the healthcare sector, or by another sector.11
Details of sources for all costs are provided (Tables 2 and 3). An average
cost per episode was calculated using the total number of illnesses, not
just those where burden information was available, as the denominator.
Table 2. Summary of resources consumed during 202 influenza-like illnesses in 118 Melbourne
children during winter and spring 2001
||Patient and family sector
||% ILI cost
|General practice visits*
|Other healthcare provider visits†
|Hospital emergency department visit (no admission)‡
|Other prescription medication||
|Over-the-counter and other medication¶
|Paid childcare for other children**
|Travel costs seeking healthcare††
|Time seeking healthcare‡‡
Time away from work, pay
Time away from work, no pay lost
Time away from usual activities
|Excess time caring for ill child‡‡
Time away from work,
Time away from work, no pay lost
Time away from usual activities
| Sector total§§
| Sector cost per ILI
| Sector per cent
Total cost per ILI
Carer time spent seeking healthcare and excess time spent caring for
an ill child were collected in three categories: time away from work with
pay lost; time away from work with no pay lost; and time away from usual
activities. We applied a sex-weighted hourly rate derived from the Australian
Bureau of Statistics average weekly earnings (females: $19.69 per hour; males:
$22.44 per hour) for reported times.12 For time
away from work with pay lost and time away from usual activities we allocated
the cost to the patient and family sector; and for time away from work with
no pay lost, we allocated the cost to the employer (other sector), who was
paying for working hours not performed.
We identified the key cost drivers for illness, and calculated an average
resource unit used per episode and 95 per cent confidence intervals (95%
CI) using standard methods.13 Where information
was not available for an illness, we applied a zero value for missing data
when calculating means and CIs. One-way sensitivity analyses were undertaken
by using the 95 per cent confidence limits for these key cost drivers, and
we calculated an average cost for all episodes, by including those illnesses
where there was a hospitalisation. We also used the confidence limits to
perform multi-way sensitivity analyses, with a least expensive and most expensive
scenario for community-managed episodes.
Calculations were performed using Microsoft Excel.
Top of page
One hundred and twenty-one children from 80 households were enrolled;
complete individual and household demographic data about 118 children (98%)—52
females and 66 males—from 78 households (97.5%) were available and these
are included in this analysis. These 118 children provided 14,430 child-days
(477.3 child-months) of follow-up between 1 July and 1 December 2001. Most
study households came from the highest annual income bracket: 73 per cent
from bracket 4 (income > $56,000); 15 per cent from bracket 3; 6 per cent
from bracket 2; and 5 per cent from bracket 1. There were 15 households with
a couple and one child, 41 with a couple and two children, 17 with a couple
and three children, three with a couple and four children, one household
with a single parent and two children, and one household with a single parent
and three children. Eight children (7%) were one year of age at enrolment,
62 (53%) were two years of age, 19 (16%) were three years of age, 18 (15%)
were four years of age, and 11 (9%) were five years of age.
There were 202 ILI community-managed episodes identified, giving an incidence
rate of 0.42 ILIs per child-month (95% CI 0.36 to 0.48). There were three
episodes that resulted in hospitalisation,8 and
these were not included in general calculations. During the period, 21 children
had no episodes of ILI, 35 children had one episode, 30 children had two
episodes, 24 children had three episodes, five children had four episodes,
and three children had five episodes. We received costing information for
180 (89%) of these illnesses (Table 2). The illnesses where we did not receive
burden data were shorter (median duration: 2.5 days versus 5 days) and less
likely to have parent-reported fever or ear infection8
(proportion with uncomplicated illness: 77% versus 48%), compared to those
illnesses where burden data were available. Parents may have been less likely
to report burden information for illnesses they felt were trivial, or resulted
in no excess resource consumption.
Using the costs from these 180 for all 202 illnesses gave an average
cost per ILI episode of $241 (95% CI $191 to $291). The average cost using
only those illnesses we had information on was $270. The key cost driver
for ILI in children was carer time spent caring for the ill child away from
usual activities, making up 70 per cent of total costs. Females spent an
average of 6.38 hours per episode (95% CI 4.61 to 8.15) caring for the ill
child away from their usual activities, and males an average of 1.95 hours
per episode (95% CI 1.05 to 2.84). The next most important non-carer time
related drivers were use of non-prescription medication (5.4% of total costs,
244 episodes of use, 95% CI 215 to 273), and general practitioner visits
(5.0% of total costs, 89 visits, 95% CI 68 to 110 visits).
The average cost per episode was lowest for those illnesses occurring
in households from the highest income bracket: bracket 4, $208; bracket 2,
$290; bracket 1, $377; and bracket 3, $449. These rankings remained the same
when illnesses where there was no information available were removed from
average calculations (bracket 4, $235; bracket 2, $327; bracket 1, $431;
and bracket 3, $474).
Funding the resource use during illness was predominantly the responsibility
of the patient and family, with this sector being responsible for meeting
87 per cent of total costs. The healthcare sector met five per cent of costs,
and other sectors met eight per cent of costs.
As key costs drivers, carer time away from usual activities, non-prescription
medication, and general practice visits were individually varied in one-way
sensitivity analyses, according to the upper and lower 95 per cent confidence
limits. The average cost per episode varied little for the sensitivity analyses
involving non-prescription medication and general practice visits (Table
3), but ranged from $186 to $296 when carer time away from usual activities
was varied. The one-way sensitivity analysis which included the three illnesses
with hospitalisations increased average cost per episode to $287 (Table 3).
Two scenarios were tested producing a least expensive average cost per episode
of $177, and a most expensive average cost per episode of $304 (Table 3).
Unsurprisingly, these values varied little from those generated in the one-way
analyses of carer time away from usual activities.
Table 3. One-way and multi-way sensitivity analyses for average cost of episodes
||Average cost per episode
|General practice visits
||Number of general practice visits and dependent variables*
||Lower value: 68 visits
Upper value: 110 visits
|Over-the-counter and other medication
||Number of episodes of over-the-counter and other medication
||Lower value: 215 episodes
Upper value: 273 episodes
|Carer time away from usual activities
||Time spent caring from ill child away from usual activities
||Lower value: 5.67 hours
(4.61 female, 1.05 male)
Upper value: 10.99 hours
(8.15 female, 2.84 male)
||Addition of three ILIs with a hospitalisation
||All costs for these ILIs added to total costs†
|Least expensive scenario
||General practice visits* Over-the-counter and other medication
Carer time from usual activities
215 episodes of use
Female carers: 4.61 hours per episode
Male carers: 1.05 hours per episode
|Most expensive scenario
||General practice visits*
Over-the-counter and other medication
Carer time from usual activities
273 episodes of use
Female carers: 8.15 hours per episode
Male carers: 2.84 hours per episode
Top of page
As demonstrated by our findings, acute respiratory illness in healthy,
urban children during winter and spring is common, with the costs borne largely
by the patient and family. These findings have implications for preventive
strategies in Australian children, particularly vaccine use. The impact carer
time away from usual activities has on the average cost per episode can be
seen in a number of ways: the proportion of total costs made up by this single
variable (70%); and in the multi-way sensitivity analyses producing least
and most expensive cost per episode scenarios varying little from the one-way
sensitivity analysis of this variable alone. Not including carer time away
from usual activities, as recommended for submissions to have drugs listed
on the Pharmaceutical Benefits Scheme,14 would
substantially under-estimate the true impact of community-managed disease
of this nature. In this regard, these illnesses may be similar to chickenpox,
being common and usually community-managed, with the direct costs of a proposed
infant vaccination program in Australia outweighing the direct costs associated
with not implementing such a program.15
A significant proportion of the illnesses identified in this study are
likely to have been caused by respiratory viral infections, including respiratory
syncytial virus, influenza virus, parainfluenza viruses, human metapneumovirus,
coronaviruses, adenoviruses, and rhinoviruses. The Victorian Infectious Diseases
Reference Laboratory identifies 2001 as a year of normal seasonal activity
for influenza from the collaborative sentinel influenza surveillance scheme.16
Injectable influenza vaccine is licensed in Australia for children down to
six months of age. The recent licensing in the USA of the intranasal CAIV-T
vaccine provides the possibility of better access, acceptability, and delivery
of public influenza vaccination programs, especially if the license for use
extends to a lower age-group. The current price of the vaccine, though set
to fall to USA$23.50 for the 2004/2005 influenza season in the United States
of America,17 will remain an impediment to its
wider use. Vaccines against other respiratory viruses are under development,
but still likely to be some way off; the possibility for preventing such
illnesses at present is limited to influenza. Beginning in 2004 the Advisory
Committee on Immunisation Practices (ACIP) in the USA have made injectable
influenza vaccine part of the routine childhood immunisation schedule—for
children from six months up to two years of age. This recommendation extends
to household contacts (including older children) and out-of-home caregivers
of all children less than two years of age.7 Interest
in this recommendation was driven by the USA Centers for Disease Control
and Prevention initiating national surveillance for paediatric influenza-associated
deaths.18 It is also possible that use of vaccine
in this age group may lead to reduced incidence of disease in other age-groups
due to herd protection, similar to effects seen from vaccinating school-aged
children against influenza,19,20 and more recently,
seen in vaccinating USA infants with conjugate pneumococcal vaccine.21
Vaccination programs against illnesses that are largely managed in the
community may not appear cost-effective if the impact of lost productivity
is ignored. Vaccines against influenza and other viral respiratory pathogens
may be recommended for young children in the near future, but may not pass
the cost-effectiveness hurdle for public funding. There are few published
studies looking at the cost-effectiveness of influenza vaccine specifically
in children. Given the findings of our study, it will not be surprising that
a childhood influenza vaccine program could be potentially cost-saving if
indirect costs are included.22 The reduction in
indirect costs is central to the economic benefits of vaccination,23–25
with previous studies showing that these benefits are greatest when parents
are prevented from missing work to care for an ill child.24
The majority of households (73%) in our study came from the highest income
bracket. This compares with approximately 40 per cent of Victorian family
households being in this income range, according to 2001 Census data.26
This may have had a number of impacts: members of lower income households
have been shown to have a higher incidence of respiratory viral infections,
thought to be due to the impact of crowding;27
but in this study we did not find a lower rate of illness in children from
high income families.8 It could be argued that
parents from relatively higher income households might be more likely to
expend more resources in caring for a sick child, as compared with those
from a lower income household; but we found that the average cost per episode
was lowest in households from the highest income bracket. In our study, high-income
households were more likely to have both parents spending some time working
outside the home. Parents in these households might have different thresholds
for seeking medical attention or using medication for illnesses that are
perceived to be mild or of minor significance. If anything, due to the lower
average cost per episode in higher income households, the over-representation
of such households in our study may have made our cost estimate conservative.
Costing studies such as this, together with studies that measure the
relative role of specific pathogens, will not only inform local cost-effectiveness
studies, but failing public funding of programs, will provide important information
for vaccine providers and parents about the likely benefits of paying for
available vaccines themselves.
Top of page
We would like to thank the research staff who assisted with this study—Jacinta
O’Sullivan, Samantha Colquhoun, Ethna Macken, and Sally Mizrahi. Recruitment
of younger children for this study was only possible through the kind assistance
of local government Maternal and Child Health Nurses in the greater Melbourne
area. We extend our appreciation to the children and families who participated
in the study. Stephen Lambert is a National Health and Medical Research Council
Public Health Postgraduate Scholar. Support for this study was provided in
part by a grant to the Murdoch Childrens Research Institute from CSL Ltd.
Top of page
1. National Health and Medical Research Council. Immunisation Procedures. 3rd edition. Canberra:
Australian Government Publishing Service, 1986.
2. Taylor A, Wilson D,
Dal Grande E, Gill T. National Influenza Survey - a population survey of
vaccination uptake in Australia, October 2000. Adelaide: South Australian
Department of Human Services, 2001.
3. National Health and
Medical Research Council. The Australian Immunisation Procedures
Handbook. 5th edition. Canberra: Australian Government Publishing
4. The Hon Tony Abbott MHR. Media Release: Commonwealth Government to provide pneumococcal vaccine for both young and old. Commonwealth Department of Health and Ageing, 2004. Available from: http://www1.health.gov.au/internet/ministers/publishing.nsf/Content/health-mediarel-yr2004-ta-abb078.htm?OpenDocument&yr=2004&mth=6 Accessed on 22 October 2004.
5. Burgess MA, McIntyre PB. Vaccines: the new Australian best-practice schedule. Med J Aust 2004;180:494–496.
6. National Health and Medical Research Council. The Australian Immunisation Handbook.
8th edition. Canberra: Australian Government Publishing Service, 2003.
7. Centers for Disease Control and Prevention. Prevention and control of influenza - recommendations
of the Advisory Committee on Immunisation Practices (ACIP). MMWR
Morb Mortal Wkly Rep 2004;53 (no. RR–6):1–40.
8. Lambert SB, O'Grady KF, Gabriel SH, Nolan TM. Respiratory illness during winter: a cohort study
of urban children from temperate Australia. J Paediatr Child Health 2004 (in press).
9. Belshe RB, Mendelman
PM, Treanor J, et al. The efficacy of live attenuated,
cold-adapted, trivalent, intranasal influenzavirus vaccine in children. N
Engl J Med 1998;338:1405–1412.
10. Rothman KJ. Epidemiology: an
introduction. New York: Oxford University Press, Inc, 2002.
11. Drummond MF. Methods for the
economic evaluation of health care programmes. 2nd edition. Oxford: Oxford
University Press, 1997.
12. Australian Bureau of Statistics.
Average weekly earnings Australia: November 2001. Volume 6302.0. Canberra:
Australian Bureau of Statistics, 2002.
13. Kirkwood BR, Sterne JAC. Essential
medical statistics. Second edition. Oxford: Blackwell Science, 2003.
14. Commonwealth Department of Health
and Ageing. Manual of Resource Items and their associated costs for use in
major submissions to the Pharmaceutical Benefits Advisory Committee involving
economic analyses. Canberra: Australian Government, 2002.
15. Scuffham PA, Lowin AV, Burgess
MA. The cost-effectiveness of varicella vaccine programs for Australia. Vaccine
16. Victorian Infectious Diseases
Reference Laboratory. Sentinel Influenza Surveillance. Victorian Infectious
Diseases Reference Laboratory, 2004. Available from: http://www.vidrl.org.au/publications/pubs2004.htm Accessed 22 October 2004.
17. Medimmune Inc. MedImmune Selects Henry Schein as Exclusive Distributor of FluMist™, MedImmune Substantially Reduces the Vaccine's Price. 2004. Available from: http://phx.corporate-ir.net/phoenix.zhtml?c=83037&p=irol-investornewsArticle&ID=581615&highlight=; Accessed 22 October 2004.
18. Centers for Disease Control and Prevention. Update: influenza activity — United States, 2003–04 season. MMWR
Morb Mortal Wkly Rep 2004;53:284, 286–287.
19. Reichert TA, Sugaya N, Fedson
DS, Glezen WP, Simonsen L, Tashiro M. The Japanese experience with vaccinating
schoolchildren against influenza. N Engl J Med 2001;344:889–896.
20. Monto AS, Davenport FM, Napier JA, Francis Jr T. Modification of an outbreak of influenza in Tecumseh,
Michigan by vaccination of schoolchildren. J Infect Dis
21. Whitney CG, Farley MM, Hadler
J, et al. Decline in invasive pneumococcal disease
after the introduction of protein-polysaccharide conjugate vaccine. N
Engl J Med 2003;348:1737–1746.
22. Nichol KL. The efficacy, effectiveness
and cost-effectiveness of inactivated influenza virus vaccines. Vaccine
23. White T, Lavoie S, Nettleman
MD. Potential cost savings attributable to influenza vaccination of school-aged
children. Pediatrics 1999;103:e73.
24. Cohen GM, Nettleman MD. Economic
impact of influenza vaccination in preschool children. Pediatrics
25. Luce BR, Zangwill KM, Palmer
CS, et al. Cost-effectiveness analysis of an intranasal influenza vaccine
for the prevention of influenza in healthy children. Pediatrics
26. Australian Bureau of Statistics. Household and Family Estimates, Australia, Electronic Delivery, June 2001. Catalogue 3236.0.55.001, 2004. Available from:
http://www.abs.gov.au/Ausstats/abs@.nsf/Lookup/490EBCFADEE3238DCA256EA8007DF182 Accessed 22 October 2004.
27. Monto AS. Epidemiology of viral respiratory infections. Am J Med 2002;112 Suppl 6A:4S–12S.
28. Commonwealth Department of Health and Ageing. Medicare Benefits Schedule, November 2001. Commonwealth Department
of Health and Ageing, 2004. Available from: http://www7.health.gov.au/pubs/mbs/archive.htm Accessed on 22 October 2004.
29. Commonwealth Department of Health and Ageing. Medicare statistics: average patient contribution per service (patient and bulk billed services out-of-hospital only; table B6) Australian Government, 2001. Available from:
http://www1.health.gov.au/internet/main/publishing.nsf/Content/Medicare+Statistics-1 Accessed on 2 September 2004.
30. Commonwealth Department of Health
and Ageing. Schedule of Pharmaceutical Benefits for Approved Pharmacists
and Medical Practitioners. Canberra: Australian Government, 2001.
31. MIMS Australia. MIMS annual.
25th edition. Sydney: MediMedia Australia Pty Ltd., 2001.
32. Royal Automobile Club of Victoria.
Vehicle operating costs. Melbourne: RACV, 2002.
33. Commonwealth Department of Health
and Ageing. National Hospital Cost Data Collection. Hospital Reference Manual,
Round 6, for the 2001–202 financial year. Canberra: Australian Government,
Top of page
1. Murdoch Childrens Research Institute, Royal Children’s Hospital, Parkville, Victoria
2. School of Population Health, University of Melbourne, Melbourne, Victoria
3. Program Evaluation Unit, School of Population Health, University of Melbourne, Melbourne, Victoria
Corresponding author: Dr Stephen Lambert, School of Population Health, Level 5/207 Bouverie Street, University of Melbourne Victoria 3010. Telephone: +61 3 8344 9330. Facsimile: +61 3 9348 1827. Email: firstname.lastname@example.org
This article was published in Communicable Diseases Intelligence Vol 28 No 4, December 2004.