This is the first community-based study to provide close follow-up of all ARI that children suffer in the first 5 years of life. There was a good retention rate of study subjects and accurate data collection on respiratory infections has been achieved through regular telephone calls and validation with daily diaries. This cohort at high genetic risk of atopy, however, represents infants from small families of higher socio-economic group. More than half the cohort was first born, 74% were breastfed for more than 6 months, only 32.2% attended day care in early infancy, only 4% were exposed to maternal antenatal smoking and 11.4% to postnatal environmental tobacco smoke. Thus the findings may not reflect the findings in a general population. However, to counterbalance the shortcomings of this study, early childhood is a time of particular susceptibility for ARI and asthmatics have recently been demonstrated to have a profound deficiency in innate antiviral immunity.7 Thus infants at high-risk of developing allergic diseases, such as the children in this study, are the population most likely to become ill with respiratory viral infections and are therefore the best group to study aspects of ARI.
Frequency, type of ARI and viruses associated with ARI
The frequency of four episodes/year in the first 2 years and 2–3 episodes/year between 2 and 5 years is lower than that reported in some other studies that have described rates of 6–9 episodes/year.2,6,8 Data on ARI incidence in community studies are scarce and the majority of studies have involved older children or been observational studies in day care settings.7,9,10 Environmental factors associated with ARI include young age, environmental tobacco smoke (ETS) exposure, poverty, lack of breastfeeding, day care attendance and overcrowding.2,11–15 The lower incidence rate in this study could have arisen because the cohort were from small families of higher socio-economic status, the majority were ‘only’ children, most of them were breastfed for more than 6 months, and only 11% exposed to ETS. While these factors may limit the generalisability of our findings, they do not compromise the validity of the results.
More than two-thirds of the ARI involved only the upper respiratory tract. The majority of the infants had two episodes of URI in the first year and 53% reported at least one episode of LRI. Schwartz et al. found infants in their study had nine ARI/year with only 13% involving the lower respiratory tract.10 Fergusson and Horwood,6 in a prospective study in Christchurch, New Zealand, found 64.1% of children under the age of 2 years experienced at least one LRI. The definition of LRI was ambiguous and based on parental recall. Other studies2,4 found higher incidences of 250 episodes per 100 child-years. These rates were drawn from studies of close populations such as private paediatric clinic and day care facilities. Studies relying on medical attendance to draw conclusions on LRI incidence would be unreliable as many episodes of LRI resolve spontaneously without requiring medical attention.
In this study, as expected, most of the ARI occurred during the colder winter months of June to August. Clear seasonal trends in ARI rates have been demonstrated with relatively fewer infections in the summer months.16
The majority of ARI in the first year of life, including wheezy LRI, were found to have viral aetiology, with rhinoviruses and RSV the most common pathogens identified, supporting reports by other studies that viruses are the primary respiratory pathogens in ARI in early childhood.17,18
Duration and management of symptoms
Runny/blocked nose was the most common presenting symptom, followed by cough. In the majority of episodes, both these symptoms resolved within 2 weeks of onset. A total of 27.0% of the ARI episodes were associated with a fever of >38°C, which resolved in the majority of cases within the first 3 days.
Ray et al. reported on URI respondents at a walk-in clinic, and found 84% had phlegm, 95% cough, and 38% reported associated fever.19 With that pattern of symptoms, paracetamol was used in half of the ARI, while cough mixtures, antitussives and other cold medications were used in 53% of ARI, particularly if the child had a ‘moist, wet’ cough. Jhaj et al. found 88% of episodes of URI in children were treated with cough mixtures and other cold medications.20
Forty-five per cent of all episodes of ARI in this study presented to their family physicians, with a slightly higher percentage presenting in the first 2 years, or if the child had symptoms of a LRI. As the majority of children from this cohort are ‘only’ children, their parents may have felt less experienced in managing ARI symptoms and may thus be more likely to seek medical attention, especially when the child is younger, or when their child had a rattly or wheezy chest, which they may have considered a more serious illness.
Antibiotics were used in 23.6% of all episodes of ARI, with more LRI being treated with antibiotics than URI. This rate is markedly lower than that found in comparable studies in other countries. Antibiotic prescriptions for ARI in the primary care setting show markedly differing rates from 74% to 79% in the UK, Canada and Mexico,21–23 56% to 67% in the USA, Italy and Sweden,24–26 with the lowest rate of 26% in Malaysia.27 These studies also show higher rate of antibiotic prescription for LRI compared with URI (75% and 46%),28 (90% and 76%),29 (91% and 58%),22 (81% and 44%).25 The rate of antibiotic prescription for ARI in Australia has dropped from 42.1% in 1998–1999 to 35.1% in 2002–2003 reflecting the effectiveness of ongoing education strategies to reduce the use of antibiotics in the primary care setting.21,30
An association was found between antibiotic use and the duration of symptoms, perhaps reflecting parental concern over the length of their child’s illness and physicians’ perception of parental expectations of the need for antibiotic treatment.
There is little doubt that antibiotic resistance is a direct consequence of antibiotic use. People who had recently received antibiotics have been demonstrated to be more likely to be infected by antibiotic-resistant bacteria,29,31,32 while bacterial resistance to one drug or drug class increases the likelihood of developing resistance to unrelated antibiotics.
In Australia, Streptococcus pneumoniae resistance to antibiotics typically used to treat URTI has increased dramatically in the preceding few years.33 An Australian study that followed 484 preschool children for 25 months in the community setting found that rates of pneumococcal resistance are higher among children who had used more antibiotics and that the likelihood of carrying penicillin-resistant pneumococci was doubled in children who had received a beta-lactam (i.e. a penicillin or a cephalosporin) antibiotic in the preceding 2 months.34
Imprudent prescribing of antibiotics for ARI, particularly URI, which are, in the main viral in origin, has no doubt contributed to the rising trend of antibiotic resistance world-wide. Just under a quarter (23.6%) of the ARI in the present study were treated with antibiotics and although no trends were found between age groups, children with LRI were more likely to be treated with antibiotics. The majority of wheezy LRI episodes in hospitalized preschool children are caused by viruses, particularly RSV.35 In this study, more than two-thirds of the episodes of ARI, including wheezy LRI in the first year of life had viral aetiology5 and education of parents and physicians on the viral nature of the majority of ARI could thus contribute an effective strategy for changing therapeutic behaviour.
Risk factors to ARI
The finding that male children have more ARI in the first 3 years is similar to other studies.36,37
Children who were exposed to other children either at home or at day care had significantly more ARI, LRI and wheezy LRI in the first year. These findings are consistent with the literature,2,14,38 and can be explained by the increased exposure through contact with other children and the spread of respiratory pathogens.
The low prevalence of ETS exposure, prolonged breastfeeding and low rate of day care attendance were felt to have contributed to the lower incidence of ARI in this study. A systematic public health approach in the management of ARI to maximise preventive measures such as increased public education to raise awareness about the detrimental effects of ETS, protective effects of breastfeeding, reduced exposure to other children through delaying child care attendance would help to minimise transmission rates to reduce prevalence rates and economic impact. Clinicians can promote and reinforce these public health messages through parental education during consultations. Governments should ban all tobacco advertisements, legislate to stop smoking in public places as well as private vehicles, impose higher levies on cigarettes and tobacco products; and use revenue raised from these sources for further public education.