Child health

Environmental strategies must be complemented by behavioural approaches to help children learn to use roads safely In Australia, pedestrian injury is the leading cause of death among 1–14-year-olds.1 In 2000, 38 child pedestrians in this age group died2 and about 1140 (29 per 100 000) were hospitalised, often with lengthy stays, because of injuries sustained when hit by a vehicle.1 These rates decrease with age and are lowest for 10–14-year-olds.1 The most recent comparison with other OECD countries shows that Australia has the 13th lowest pedestrian fatality rate for 0–14-year-olds,3 with slightly more pedestrian deaths among 0–5-year-olds than the median rate for all OECD nations (1.03 per 100 000 versus 0.89 per 100 000).4 The primary predictors of this child pedestrian trauma relate to the interaction between the characteristics of the child and the design and nature of the road environment to which the child is exposed. A Western Australian case–control study of child pedestrians aged 1–14 years identified four key environmental and behavioural factors that independently predicted the likelihood of child pedestrian injury.5 These comprised the volume of traffic encountered by the child, presence of visual obstructions, availability of footpaths on the child’s street of residence, and the child’s behaviour. Predictors also vary according to the age of the child. Whereas 1–2-year-olds are more likely to be hit by a reversing vehicle, the most common cause of pedestrian trauma in 3–9-year-olds is mid-block “dart-out” (entering the road between intersections and not seeing, or misjudging, a gap in traffic).6,7 Pedestrians aged under 10 years are particularly vulnerable because of their small physical size and underdeveloped abilities for dealing with traffic situations, both cognitive (attention focus, interpreting traffic signs) and perceptual (locating sounds, judging speed, peripheral vision).6 Given these limitations, children under the age of 10 do not have the ability to cross roads without adult help. Ten to 14-year-olds are also vulnerable, but more because of their failure to apply safe pedestrian skills than because of their lack of skills. Further, road trauma in this age group may also be associated with general delinquency and problem behaviour.8 Many of these predictors of pedestrian trauma can be prevented or modified and are therefore amenable to intervention.7 While debate continues about the merits of environmental (passive) versus behavioural (active) intervention strategies to reduce this road trauma, evidence suggests that all are necessary, and no single strategy is sufficient.6 A multifaceted approach that combines strategies targeting the behaviour of all road users (including education, training and publicity), the road environment and vehicle design have been found to be the most effective.3 Strategic approaches involving public health, education, health promotion, urban planning, engineering and motor vehicle design are required. Consequently, while efforts are needed to make the road environment safer for pedestrians by reducing the speed and volume of traffic to which they are exposed, it is also necessary for pedestrians (particularly children) to learn how to use these road environments safely. Yet research into behavioural approaches to pedestrian safety has lagged behind environmental research.9 Behavioural programs for children need to be developmentally appropriate and include modelling and training by an adult in a social context and road environments relevant to the child. Programs also need to be interactive and involve problem-solving with consistent and prompt feedback from a caring adult, rather than merely following rules.10 The use of didactic knowledge-only strategies (such as rote learning of rules) is inappropriate, as younger children are not able to generalise this learning to real roads.11 Roadside training and, to a lesser extent, realistic simulations appear to improve visual timing and gap selection, to increase the ability to identify safe and dangerous crossing locations, and to enhance learning of appropriate strategies for crossing at parked cars. Such training has produced positive results with children as young as 5 years.12,13 Several new approaches to children’s pedestrian safety education are being tested. These include programs targeting younger children and adults who care for children, and use of new technologies. Two Australian reviews recommend targeting 0–5-year-olds, arguing that, with good quality pedestrian safety training, young children could demonstrate a rudimentary conception of danger, which improves with age.14,15 Some Australian jurisdictions have developed curricula and materials based on these approaches, such as the Victorian “Starting Out Safely” program. Engaging parents and helping them recognise their important role in their children’s pedestrian safety has the potential to significantly enhance children’s safety on and near roads.14 Parents provide the best role models and one of the only means for children to receive the necessary personalised one-on-one training and to practise crossing real roads. Technologies that use interactive simulations (eg, visual reality computer) coupled with real road experience and “pretend” road practice (the “pretend” road is set up parallel to a real one) can enable children to practise their skills, receive consistent and instant feedback and repetition, and be introduced with careful control to the complexity of traffic.16 While reviews of the impact of behavioural and environmental programs on preventing road trauma in children have demonstrated mixed success, it is apparent that programs need to take a comprehensive preventive approach with modifications to the road environment, enforcement, engineering and education. While much still needs to be done to determine the optimal mix and “dose” of these approaches to reduce child pedestrian trauma, every effort must be made to keep children safe near traffic and roadways that are becoming increasingly busy and complex.

Donna S Cross EdD · Margaret R Hall PhD

Early diagnosis of language disorders and autism is important, and early intervention for autism and some language disorders makes a difference. Developmental surveillance of children to detect these disorders should be a routine part of medical practice. The persistence and pervasiveness of communication and socialising deficits differentiate children with autism from those with specific developmental language disorders. Hearing and vision assessment is essential in any communication disorder. Interventions, targeted to identified areas of need, should encompass communication enhancement, behavioural therapy, educational modification, parent education and family support. Pharmacological interventions have an important but discrete role in autism, but there are no magic bullets. It is important to remember that the normal childhood illnesses occur in children with developmental disorders. Parents should be directed to reliable websites on the Internet, and given information and books to read as well as phone numbers of relevant services (eg, autism associations). There is a need for increased government financial support for early intervention programs.

John Wray FRACP · Helen Knott BApplSc(S · Natalie Silove FRACP, MMed, FCP(SA)

A child with atopy produces IgE antibodies after exposure to common environmental allergens. The atopic diseases (eczema, asthma and rhinoconjunctivitis) are clinical syndromes each defined by a group of symptoms and signs. Not all children with atopy will have atopic disease or develop symptoms after exposure to an allergen. Both genetic and environmental factors determine the development of atopic disease. The presence of specific IgE antibodies to environmental allergens is determined with skin prick or radioallergosorbent testing in children with atopy. Test results should be interpreted in the context of the clinical history and further investigations (eg, allergen avoidance or challenge). Management of atopic disease is frequently symptomatic, but it is important to avoid identified allergen triggers. Immunotherapy may be considered in selected school-age children with severe rhinoconjunctivitis. Preventing atopic disease in high-risk infants and hindering progression of disease in children with established disease are the areas of active research.

Michael S Gold MD, FCP, FRACP · Andrew S Kemp PhD, FRACP

Objective: To evaluate the effect of a behaviour modification program, taught to parents in a single visit to a trained nurse, in improving sleep performance in newborn infants.Design: Randomised controlled trial.Setting and participants: 268 families with normal newborn infants in the community, recruited between October 1996 and March 1997 from birth notices published in a South Australian daily newspaper.Intervention: A 45-minute consultation with a nurse 2–3 weeks after the birth, including a tutorial discussion on normal sleep patterns in newborn infants, supported by retained written material and, for infants with weight gain < 30 g daily, referral to their usual postnatal care provider.Main outcome measures: Hours of daytime sleep (0600–1800), night sleep (1800–0600) and total sleep per 24 h; and number of daily records with total sleep ≥ 15 h per 24 h, assessed by 7-day sleep diary at ages 6 and 12 weeks.Results: 268 families returned at least one sleep diary (137/171 intervention, 131/175 control), recording 3273 days. Two intervention infants were referred for low weight gain. Total sleep time was 15 h or more per 24 h on 62% of recorded days in the intervention group, compared with 36% in the control group (P < 0.001). At 6 weeks of age, intervention infants slept a mean 1.3 h per day more than control infants (95% CI, 0.95–1.65), comprising a mean 0.5 h more night sleep (95% CI, 0.32–0.69) and 0.8 h more daytime sleep (95% CI, 0.56–1.07). At 12 weeks, intervention infants slept a mean 1.2 h per day more (95% CI, 0.94–2.14), comprising 0.64 h more night sleep (95% CI, 0.19–0.89) and 0.58 h more daytime sleep (95% CI, 0.39–1.03). There was no significant difference in crying time between the groups.Conclusions: A single consultation supported by written material in the first 3 weeks of a child’s life improves sleep performance at 6 weeks of age. This improvement is maintained at 3 months.

Brian G Symon MD · John E Marley MD · A James Martin MD · Emily R Norman MB BS

Constipation is common in children, with prevalences ranging from 0.3% up to 28%. In most children, constipation is functional (ie, without objective evidence of a pathological condition). Painful defecation has been proposed as the primary precipitant of functional faecal retention in early childhood. Faecal soiling is often secondary to constipation, and may occur during spontaneous relaxation of the sphincters precipitated by rectal distension. Management in general practice combines behavioural modification techniques with prolonged courses of laxatives. Treatment is usually successful, but may take up to 6–12 months. Significant numbers of children with initially good responses to therapy for constipation relapse in the long term. Long-term relapse is more frequent in children under 4 years at onset of symptoms and in whom there is a history of faecal soiling associated with constipation.

Anthony G Catto-Smith MD, FRACP

Bedwetting (nocturnal enuresis) is common. It occurs in up to 20% of 5 year olds and 10% of 10 year olds, with a spontaneous remission rate of 14% per year. Weekly daytime wetting occurs in 5% of children, most of whom (80%) also wet the bed. Bedwetting can have a considerable impact on children and families, affecting a child’s self-esteem and interpersonal relationships, and his or her performance at school. Primary nocturnal enuresis (never consistently dry at night) should be distinguished from secondary nocturnal enuresis (previously dry for at least 6 months). Important risk factors for primary nocturnal enuresis include family history, nocturnal polyuria, impaired sleep arousal and bladder dysfunction. Secondary nocturnal enuresis is more likely to be caused by factors such as urinary tract infections, diabetes mellitus and emotional stress. The treatment for monosymptomatic nocturnal enuresis (bedwetting with no daytime symptoms) is an alarm device, with desmopressin as second-line therapy. Treatment for non-monosymptomatic nocturnal enuresis (bedwetting with daytime symptoms — urgency and frequency, with or without incontinence) should initially focus on the daytime symptoms.

Patrina H Y Caldwell FRACP, PhD · Elisabeth Hodson MRCP, FRACP · Jonathan C Craig FRACP, PhD · Denise Edgar RGN, BN

Ursula Bayliss,* Christopher Cowell,† James Hong,‡ Veronica Wiley,§ Bridget Wicken¶ * Clinical Nurse Consultant, §Principal Scientist, ¶ Clinical Director, NSW Newborn Screening Programme, † Head, Institute of Endocrinology & Diabetes, The Children's Hospital at Westmead, Westmead, NSW 2145; ‡ Paediatrician, North Gosford Medical Centre, North Gosford, NSW. bridgetwATchw.edu.au To the Editor: We report an acute presentation of congenital hypothyroidism in a child almost 6 years old. The condition was not detected by newborn screening. Screening of all neonates started in New South Wales in July 1977, with thyroid stimulating hormone (TSH) being measured in dried blood spots taken from a heel-prick blood sample (currently at 2–3 days of age). A whole-blood TSH level of 40 mIU/L or above triggers a request for full thyroid function testing, whereas with a level of 20–39 mIU/L a second sample is requested. We have screened over 2.3 million babies and detected 690 babies with congenital hypothyroidism. Ten babies with dyshormonogenesis or ectopic thyroid tissue had normal results and were missed by the screening test. Since screening started, “juvenile hypothyroidism” not associated with thyroid antibodies has all but disappeared. A healthy girl aged 5 years 11 months presented with acute dysphagia and drooling. There were no previous dysphagic symptoms. Initially, epiglottitis was suspected; however, at endoscopy a lingual thyroid was visualised at the base of her tongue, and this was confirmed by a technetium scan. She had normal growth and development, with both height and weight at the 50th centiles, a pulse rate of 90 beats/min, and normal deep tendon reflexes. The whole-blood TSH level at newborn screening on Day 3 was 40 mIU/L (reference range [RR], < 20 mIU/L). Thyroid function testing at another hospital on Day 10 showed a serum TSH level of 16.6 mIU/L and a serum free thyroxine (FT4) level within the normal range (12 pmol/L; RR, 11–30 pmol/L). These results were interpreted as normal, whereas, in fact, the TSH level was above the reference range for 10 days of age (< 10 mIU/L), although within the reference range for 2–7 days. On the patient’s admission for treatment of acute dysphagia, the TSH level was 10.9 mIU/L and the FT4 level was 18 pmol/L. A diagnosis was made of compensated hypothyroidism secondary to the ectopically placed lingual thyroid. Thyroxine treatment was commenced on diagnosis, and regular follow-up arranged. Three months after the start of treatment, the results of thyroid function tests (FT4, 17 pmol/L; TSH, 2.7 mIU/L) were within the normal range. Acute presentation of a lingual thyroid is most unusual.1 This case emphasises that further investigations must be performed when thyroid function test results are equivocal. Unfortunately, the thyroid status was considered normal because the FT4 value was within the normal range. All babies whose TSH results remain elevated while the FT4 levels are normal should have a thyroid scan, as we recommend when reporting results.

Ursula Bayliss · Christopher Cowell · James Hong · Veronica Wiley · Bridget Wicken

Aim: To ascertain the incidence of autism spectrum disorders in Australian children.Setting: New South Wales (NSW) and Western Australia (WA), July 1999 to December 2000.Design: Data were obtained for WA from a prospective register and for NSW by active surveillance.Main outcome measures: Newly recognised cases of autism spectrum disorders (defined as autistic disorder, Asperger disorder and pervasive developmental disorder not otherwise specified [PDD-NOS]) in children aged 0–14 years; incidence was estimated in 5-year age bands (0–4 years, 5–9 years, 10–14 years).Results: In WA, 252 children aged 0–14 years were identified with autism spectrum disorder (169 with autistic disorder and 83 with Asperger disorder or PDD-NOS). Comparable figures in NSW were 532, 400 and 132, respectively. Most children were recognised with autistic disorder before school age (median age, 4 years in WA and 3 years in NSW). Incidence of autistic disorder in the 0–4-years age group was 5.5 per 10 000 in WA (95% CI, 4.5–6.7) and 4.3 per 10 000 in NSW (95% CI, 3.8–4.8). Incidence was lower in older age groups. The ratio of all autism spectrum disorders to autistic disorder alone was 1.5:1 in WA and 1.3:1 in NSW, and rose with age (1.8:1 and 2.9:1 in 10–14-year-olds in WA and NSW, respectively).Conclusions: These are the first reported incidence rates for autism for a large Australian population and are similar to rates reported from the United Kingdom. Ongoing information gathering in WA and repeat active surveillance in NSW will help to monitor any future changes.

Katrina Williams PhD, FRACP, FAFPHM · Megan Helmer MHlthSc(CDM) · Craig M Mellis MPH, MD, FRACP · Marshall Tuck MPH · Emma J Glasson PhD · Carol I Bower MSc, PhD, FAFPHM · John Wray FRACP

Obesity in children and adolescents has reached alarming levels — 20%–25% of children and adolescents are overweight or obese, and 4.9% of boys and 5.4% of girls are obese. Rates of obesity have increased significantly in Australia from 1985 to 1995, with the prevalence of overweight doubling and obesity trebling. Body mass index (related to reference standards for age and sex) is recommended as a practical measure of overweight and obesity in children, and is used in monitoring individual progress in clinical practice. Obesity in childhood and adolescence may be associated with a range of medical and psychological complications, and can predispose individuals to serious health problems in adult life, including type 2 diabetes, hypertension, dyslipidaemia and non-alcoholic steatohepatitis. Obesity interventions for which there is some evidence include family support, a developmentally appropriate approach, long-term behaviour modification, dietary change, and increased physical activity and decreased sedentary behaviour. Prevention of obesity in children and adolescents requires a range of strategies involving changes in both the microenvironment (eg, housing, neighbourhoods, recreational opportunities) and the macroenvironment (eg, food marketing, transport systems, urban planning).

Jennifer A Batch MB BS, MD, FRACP · Louise A Baur MB BS, PhD, FRACP

Emma J Best,* Monica M Lahra,† Pam Palasanthiran‡ * Paediatric Infectious Diseases Fellow, † Microbiology Registrar, Neonatal Medicine, Royal Prince Alfred Hospital, Sydney, NSW. ‡ Infectious Diseases Physician, Sydney Children’s Hospital, Level 4, High Street, Randwick, NSW 2031; PalasanthiranpATsesahs.nsw.gov.au To the Editor: Pasteurella multocida is an oral commensal of domestic pets known to be an opportunistic human pathogen after traumatic animal contact. The most common infections in humans are skin and pulmonary infections. This report outlines a case of P. multocida meningitis, which has not previously been reported in Australia. A 19-day-old girl presented with a 12-hour history of fever and poor feeding. Her temperature was 39.5°C, and she was irritable, with no localising signs or skin lesions. A full septic screen was performed. Cerebrospinal fluid (CSF) showed a neutrophilic pleocytosis and gram-negative coccobacilli. She was treated with intravenous cefotaxime and gentamicin. Within 24 hours both CSF and blood cultures showed growth of gram-negative bacilli. The initial Gram stain, growth on chocolate agar and positive oxidase and catalase tests were suggestive of a Haemophilus species. However, further biochemical tests revealed the organism to be P. multocida. The infant made an excellent clinical recovery, with normal neurological and growth assessments at 6 and 12 months. The family owned two cats but reported no contact between their baby and the pets. A single tonsillar swab performed on each cat by a veterinarian 10 days after the baby’s presentation failed to isolate Pasteurella species. The family elected to keep the pets. Pasteurella meningitis occurs at extremes of age, in the immunocompromised (associated with liver cirrhosis, renal disease and haematological malignancies) and after traumatic head injury.1 Infants aged under 1 year account for almost half the cases of P. multocida meningitis. On review of the literature, we found 37 reported cases of P. multocida infection in infants (Box).1-5 In more than three-quarters of these cases, there was known contact with household animals — in more than half of these contact was non-traumatic (licking or presumed handling of the pet). Molecular studies in one of the cases with no history of traumatic contact confirmed that P. multocida isolates from pet and infected child were indistinguishable.2 This infection is unusual, and, given the popularity of household pets, the risk appears low. However, this case highlights the relative immunocompromise of newborn infants, and is a reminder of the importance of hand hygiene and preventing contact between newborn infants and pets. Details of 38 case reports of invasive Pasteurella multocida infection in infants (including current case)1-5 Mean age (range) 2.6 months (1 day– 11 months) Type of infection Meningitis 30 (79%) Puerperal sepsis, chorioamnionitis 7 (18%) Bacteraemia (postnatal) 1 (3%) Nature of animal contact Traumatic (scratch, bite) 9 (24%) Non-traumatic 22 (58%) Unknown 7 (18%) Type of animal (n = 31) Cat 16 (52%) Dog 11 (35%) Both 4 (13%)

Emma J Best · Monica M Lahra · Pam Palasanthiran

Megan L Counahan,* Ross M Andrews,† Rick Speare‡ * Surveillance Manager, Communicable Diseases Section, Department of Human Services, Level 17/120 Spencer Street, Melbourne, VIC 3000; † Senior Research Fellow, Centre for International Child Health and Clinical Epidemiology and Biostatistics Unit, Murdoch Children’s Research Institute, Melbourne, VIC; ‡ Professor, School of Public Health and Tropical Medicine, James Cook University, Townsville, QLD. megan.counahanATdhs.vic.gov.au To the Editor: For parents to treat head lice (pediculosis) effectively in their children, it is necessary for them first to recognise it is present. We conducted a school-based screening program involving 1838 children from 16 randomly selected primary schools in Victoria between May and October 2001 (participation rate, 55.2%).1 As part of this program, we compared a written report from parents on their child’s pediculosis status against results of our examination (7–10 days later). We examined the scalp and hair of each child for lice (“crawlers”) or viable louse eggs (“active infestation”) and dead or hatched louse eggs (“inactive infestation”) using white hair conditioner, which makes lice and eggs easier to see with the naked eye, and a fine-toothed head lice comb. This is a validated, accurate and sensitive diagnostic technique.2 Parents were unaware of the proposed screening date, and the study team was unaware of the parents’ reports. We compared parental report about pediculosis against results of our screening for 1179 children who could be matched with completed questionnaires. Overall, 149 children (12.6%) had active pediculosis, but parents reported head lice in only 36 children (3.0%) (Box 1 and Box 2). These comprised 24 of the 149 children with confirmed pediculosis (16%), and another 12 children who did not have pediculosis when examined. The positive predictive value (PPV) of parental report was 66.6%, indicating that parental reporting was not a reliable indicator of pediculosis. An implication of the low PPV is that some children may have been unnecessarily treated with insecticide for an infestation they did not have. On the other hand, a substantial proportion of children with head lice had not been identified by their parents and could contribute to ongoing transmission within schools. While it was possible they were infected subsequent to completion of the questionnaire, this seemed unlikely, as 72% were found to have multiple louse eggs, indicating a longer duration of infestation than the 7–10 days since the questionnaire was completed. Our study clearly demonstrates that parental reporting of head lice in their children is unreliable. We suggest several possible reasons: parents did not see the head lice, did not recognise them, or used a diagnostic technique with a lower sensitivity than the method we chose, such as examining dry hair. It is also possible that parents were inhibited from reporting pediculosis by the possible repercussions, such as exclusion of the child from school. Indeed, children whose parents failed to answer the question about pediculosis had a higher prevalence of head lice than those whose parents answered. Nevertheless, parents’ management of pediculosis is likely to improve if a sensitive detection method is used. To improve the sensitivity of parental diagnosis and control of head lice we recommend that parents be instructed to screen their children weekly using hair conditioner and combing. 1 Screening results compared with parental report Pediculosis by parent report Pediculosis on examination Yes No Total Yes 24 12 36 No 99 969 1068 Unsure 26 49 75 Total 149 1030 1179 2 Sensitivity and specificity of parental report versus screening Pediculosis prevalence By parental report 3.0% (36/1179) By screening 12.6% (149/1179) Sensitivity 16.1% (24/149) Specificity* 98.8% (1018/1030) Positive predictive value 66.6% (24/36) Negative predictive value* 89.0% (1018/1143) * Specificity and negative predictive value were calculated after grouping “unsure” and “no” responses.

Megan L Counahan · Ross M Andrews · Rick Speare

In most cases, childhood leukaemia has a fetal origin, but multiple molecular events are required after birth for pre-leukaemic cells to progress to leukaemia. Cure rates for acute lymphoblastic leukaemia (ALL) now approach 80%. A high level of minimal residual disease detected by polymerase chain reaction in patients with ALL in remission has profound prognostic importance and is the focus of a major Australian study attempting to prevent relapse in these children. Greater awareness of the late effects of chemotherapy has led to changes in the treatment protocols for ALL, with improvement in neurocognitive outcomes and reduced rates of second malignancies. Pharmacogenetics is a new field of research that aims to enhance treatment efficacy by assessing the individual’s metabolism of and response to chemotherapeutic agents. Targeted therapies currently being developed show some promise of being able to further improve cure rates. Adolescents with ALL have a better prognosis if treated with paediatric rather than adult protocols.

David S Ziegler MB BS · Glenn M Marshall MB BS, FRACP · Luciano Dalla Pozza MB BS, FRACP · Keith D Waters MB BS, FRACP

CorrectionRe: “Prescribing of amino acid formula”, by Andrew S Kemp in the 15 November 2004 issue of the Journal (Med J Aust 2004; 181: 574-575). The author’s position and address were omitted. Dr Kemp is Professor of Paediatric Allergy, The Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145. andrewk5ATchw.edu.au The html and pdf versions of his letter published in the eMJA were corrected on 21 December 2004.

Andrew S Kemp

Gwendolyn L Gilbert,* Heather F Gidding,† Josephine Backhouse,‡ Peter B McIntyre§ * Director, ‡ Serology Project Officer, Centre for Infectious Diseases and Microbiology, Institute of Clinical Pathology and Medical Research, PO Box 533, Wentworthville, NSW 2145. † Epidemiologist, § Director, National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases, University of Sydney, Westmead, NSW. LyngATicpmr.wsahs.nsw.gov.au To the Editor: Varicella vaccine was licensed in Australia in 2000. It is safe and efficacious and can prevent significant acute morbidity, significant out-of-pocket expenses for parents of affected children and, in Australia, prevents an estimated 450 admissions to hospital and one death per year.1 In September 2003, the National Health and Medical Research Council (NHMRC) recommended giving varicella vaccine to all children at 18 months of age.2 The net effect of childhood immunisation on varicella morbidity will depend on vaccination coverage. Modelling of Australian (unpublished National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases data) and UK3 seroprevalence data suggests that, for a range of vaccine efficacy estimates, 80% coverage is required before morbidity is reduced in adults. Accurate Australian coverage data are unavailable, as varicella vaccine is not funded under the National Immunisation Program.2 National serosurveillance can provide a practical alternative estimate of vaccine uptake. The first Australian national serosurvey of vaccine-preventable diseases, for which sera were collected between July 1996 and February 1999, established baseline seroprevalence for future assessment of the effects of changes in the vaccination schedule. It showed that the incidence of varicella was highest in 5–9-year-old children,4 indicating that infant vaccination would provide optimal protection. The second national serosurvey (with sera collected in 2002) is under way, with methods identical to those of the first.4 We compared varicella IgG levels in children aged 1–5 years, testing 459 sera in the first serosurvey and 380 in the second (see Box). The proportion with protective or equivocal antibody levels increased between the two serosurveys. The difference was statistically significant only in 3–4-year-olds, which is consistent with the greatest uptake when children enter childcare. Although there is some variation in varicella incidence over time, the changes are consistent with only modest vaccine uptake in the time that varicella vaccine has been available in Australia. The only other available estimates of varicella vaccine uptake are from GP consultations5 and reports to the Australian Childhood Immunisation Register (ACIR), which also indicate it is modest. For example, only 6.2% of children aged 4 years were reported to the ACIR to have received varicella vaccine (Brynley Hull, Epidemiologist, National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases, personal communication). This is probably an underestimate, as there is no incentive for notifying varicella vaccination. However, a low vaccination uptake is consistent with use only in the private sector, and could lead to an increase in adult morbidity (despite an overall reduction in infection rates) because of a higher average age of infection (unpublished National Centre for Immunisation Research data). Unless vaccine is provided in the routine immunisation program at no cost to parents, uptake is unlikely to meet the 80% target required to reduce disease burden in all ages.3 Varicella IgG seroprevalence in preschool children in the 1st and 2nd national serosurveys 1st serosurvey, July 1996 to February 1999 2nd serosurvey, 2002 Age group No. Positive Negative Equivocal* No. Positive Negative Equivocal* Increase in positive and equivocal (95% CI) P 1 to < 3 years 138 27 (19.6%) 111 (80.4%) 0 152 34 (22.4%) 118 (77.6%) 0 2.8% (−6.6 to 12.2) 0.6 3 to < 5 years 214 72 (33.6%) 140 (65.4%) 2 (0.9%) 152 73 (48.0%) 79 (52.0%) 0 13.4% (3.3 to 23.6) 0.01 5 years 107 61 (57.0%) 46 (43.0%) 0 76 43 (56.6%) 33 (43.4%) 0 −0.4% (−15.0 to 14.1) 1.0 Total 459 160 (34.9%) 297 (64.7%) 2 (0.4%) 380 150 (39.5%) 230 (60.5%) 0 4.2% (-2.4 to 10.8) 0.2 *Sera giving equivocal results by enzyme immunoassay were retested and most were resolved by immunofluorescence (IF). These results represent the few that were still equivocal by IF and are probably low-titre positive results.

Gwendolyn L Gilbert · Heather F Gidding · Josephine Backhouse · Peter B McIntyre

Ailsa Goulding,* Andrea M Grant,† Peter L Davidson‡ * Professorial Research Fellow, † Assistant Research Fellow, Department of Medical and Surgical Sciences, ‡ Research Fellow, Injury Prevention Research Unit, University of Otago, Dunedin, New Zealand. Ailsa. GouldingATstonebow.otago.ac.nz To the Editor: The interesting article on playground safety and arm fractures by Sherker and Ozanne-Smith1 documents a steady increase in hospitalisation rates for arm fractures among Victorian children between 1987 and 2002. It would appear that the increase over this 15-year period was about 45%. These figures may represent the tip of the iceberg, as few children with arm fractures are admitted to hospital and most are treated on an outpatient basis. Increases of similar magnitude in forearm fractures in adolescence have also been reported recently in the United States.2 It would seem that children are becoming more vulnerable to such fractures. This is a concern, as fractures of the distal forearm are extremely common during growth. Indeed, about a quarter of all fractures during childhood and adolescence occur at this site. One factor that might be contributing to this rising incidence of arm fractures during growth is increasing childhood adiposity. Childhood obesity has increased sharply in Australian children over recent years.3 Obese children fall with more force, and, although they may have more bone for chronological age than children of healthy bodyweight, adaptive increases in bone mass are not enough to accommodate their high bodyweight gain, placing them at a biomechanical disadvantage during falls on the outstretched arm.4 Our studies of consecutive series of girls and boys with distal forearm fractures indicate that a high proportion of these children are overweight. Moreover, in a 4-year prospective study of 170 girls, we found that high bodyweight at baseline increased the risk of new fractures.5 In their study, Sherker and Ozanne-Smith measured the heights and weights of 402 children under 13 years of age who had broken their arm falling from playground equipment between 2000 and 2002. We wonder how many of these children were overweight or obese for their age. Examination of the body mass index values would provide this information, and we ask that the authors report these data for both girls and boys using international cut-off points. We would like to know whether or not overweight is contributing to rising rates of arm fracture in Australian playgrounds.

Ailsa Goulding · Andrea M Grant · Peter L Davidson

Shauna Sherker,* Joan Ozanne-Smith† * Postdoctoral Research Fellow, NSW Injury Risk Management Research Centre, University of New South Wales, Sydney, NSW; † Chair of Injury Prevention, Accident Research Centre, Monash University, Melbourne, VIC. Shauna. SherkerATunsw.edu.au In reply: The prevalence of obesity has increased dramatically among Australian children, particularly over the past 2 decades.1 However, the role of obesity as a risk factor for fall-related arm fracture remains unclear. Using standard definitions for child overweight and obesity,2 our study population3 (Box) did not demonstrate quite as high a prevalence of obesity as had been previously reported for Victorian schoolchildren.1 This case series was part of a larger, yet to be published case–control study, the results of which indicate no significant difference in body mass index between children who fell from playground equipment and fractured their arm (cases) and those who fell and landed on their arm with no significant injury (controls). The negative public health effects of increasing prevalence of obesity among Australian children highlight the need to promote safe and enjoyable physical activity. Playground equipment provides a very popular means of physical activity for children. Improving playground safety standards to minimise the risk of arm fracture — a traumatic, costly and preventable childhood injury — has never been more urgent. Children with arm fracture caused by falls from playground equipment, showing proportion of children who were normal weight, overweight or obese.* * Based on unpublished data from Sherker and Ozanne-Smith.3

Shauna Sherker · Joan Ozanne-Smith

Child health will only improve when local structures are further strengthened, enabled and supported Most countries in the western Asia-Pacific region have made consistent gains in child survival over the past 25 years (Box 1).1,2 Notable exceptions to this positive trend are Papua New Guinea (PNG) and East Timor. Sadly, a static child mortality rate such as has occurred in PNG, where the population has doubled over the past 25 years, means that, in this new century, about twice as many children are dying per year as in the mid-1970s. The health inequities between Australia and its nearest neighbours are many, and breathtaking in magnitude. In this issue of the Journal, McGain et al (page 687) document 87 deaths from snakebite at Port Moresby General Hospital (PMGH) (the largest hospital in PNG) over a 10-year period.4 This hospital serves a population of about 500 000 people. By comparison, throughout the whole of Australia since 1981 there have been an average of 2.6 snakebite deaths per year.5 McGain et al point out that lack of antivenom is a major reason for the high mortality, and that the cost of antivenom in real terms is 40 times greater in PNG than in Australia. The reasons for this include a per-capita gross national income that is 2.9% of that in Australia,2 price mark-up with privatisation of overseas distribution from Australian suppliers, and an Australian government subsidy for antivenoms sold in Australian hospitals that does not apply elsewhere. It is not difficult to suggest some feasible solutions to this problem: an Australian government subsidy for overseas developing-country purchasers, and/or direct supply from the manufacturer to the PNG Department of Health. However, snakebite is just one small piece of a very large puzzle of poor child health outcomes in PNG and the Asia-Pacific, and lack of antivenom is only one factor in high death rates from snakebite. Other factors are more systemic: limited access to health services, limitations in the quality of health systems, inadequate manpower, and poor management and financing. These problems are greater in rural areas, where 85% of the population lives; they have a major impact on all causes of avoidable child mortality, and are more difficult to solve than the lack of snake antivenom. PerspectiveAt PMGH, for every child who dies from snakebite, more than 50 die from other conditions that have been eradicated or controlled in Australia. In a 12-month period in 2001–2002 there were 238 child deaths at PMGH, of which four were from snakebite. Of 195 deaths in which the cause could be certified, 29 were from measles, 35 from meningitis (about a third of which were caused by Haemophilus influenzae type b [Hib]), 14 from HIV, 7 from tuberculosis, 66 from pneumonia and 11 from acute gastroenteritis.6 Throughout PNG, about a third of Hib isolates are resistant to available antibiotics (principally chloramphenicol),7,8 and rates of HIV are rising rapidly. The persistence of these infections in PNG despite the existence of effective methods of prevention or control should be a cause for concern and action in Australia as well as in PNG. Two-thirds of all child deaths are associated with moderate to severe malnutrition. On the role of aid and economic developmentAustralia currently allocates $435 million in aid to PNG (representing 20% of Australia’s official development assistance and 0.26% of its gross national income9 — far short of the benchmark of 0.7% agreed to by rich nations at the Earth Summit in 1992, and only achieved by Scandinavian countries and The Netherlands10). Much of Australia’s aid is now tied to strengthening law-making and law-enforcement facilities and financial management, but a proportion is allocated to social services (principally health and education). Aid programs in PNG often find themselves “between a rock and a hard place”. Sustainable development cannot occur in an environment of poor governance. When existing systems are not functioning well, one outcome, sometimes occurring by default and sometimes by design, has been the development or evolution of “parallel projects”, which circumvent existing government structures to achieve a flow of services or information to the periphery. There is a tension between this project approach and the building of genuine long-term capacity (ie, the resources and structures that enable self-sustainability) within government programs. However, if inequity within PNG and between PNG and Australia is to be reduced, aid allocated to social services must be spent in ways that will strengthen local systems so that services reach the most marginalised communities. One example of this dilemma is the Women’s and Children’s Health Project, funded by the Australian government and launched in PNG in 1997 (funding will cease at the end of 2004). The project has allocated $10 million a year to improve child and family health services. Credit must be given for its achievements, such as improvements to the vaccine “cold chain” (previously a major limitation on the quality of vaccines distributed in remote areas) and training and capacity support in some rural areas. However, results have generally been disappointing. Only a small proportion of the aid money has filtered down to the villages and settlements where child mortality is highest. Much has been consumed by large infrastructure costs in Port Moresby. In an attempt to tick off activities as completed achievements, weak and sometimes frustratingly inefficient government systems have often been circumvented using a “parallel project” mentality, rather than taking the much slower approach of working with and strengthening existing local structures. In a thought-provoking but pessimistic review, Professor Helen Hughes, Senior Fellow at the Centre for Independent Studies, has argued that aid has failed the Pacific nations.11 She believes that aid has created an ambiguity of independence, an environment in which government funds are spent on consumption rather than economic development, elevated exchange rates, and provided fodder for political corruption. These factors, plus high tariffs, have hindered manufacturing for domestic markets and export of agricultural products; reduced employment opportunities, skill development and entrepreneurship outside the government sector; and encouraged dependent welfare states. A partial solution suggested by Hughes11 is to make receipt of aid conditional on achieving certain goals, under an agreement of mutual obligation. This would require removing aid from government budgets, with mutual agreement between recipient and donor countries on its use, mutual monitoring, and disbursement subject to regular account auditing. Successful examplesPapua New Guinea (PNG) needs human capacity to provide a quality health service. In some areas of endeavour, this has been achieved. The Paediatric Society of PNG is one example of the slow and successful development of indigenous technical and professional capacity. This is a story of committed engagement by many paediatricians over four decades, building on the foundation laid by the late Professor John Biddulph. Progress has been based on the principles of quiet example and mentorship, working together at the front line of healthcare and grappling with everyday problems. In the past decade there has been increasing development of a few subspecialty areas and extraclinical skills, such as public health, research, evidence-based understanding, policy development, advocacy and child health nursing capacity. Australian public hospitals and individual paediatricians have played key enabling roles in this development, and, in turn, their support has been greatly assisted by AusAID through the PNG Medical Officer, Nursing and Allied Health Professional program and its predecessors. The outcomes are impressive. Locally trained PNG paediatricians now provide services in most of the 20 provinces, and contribute substantially to all areas of public child health, policy and service delivery.12 The PNG standard treatment manual,13 along with the National Government Health Plan, is a blueprint for a quality child health service in a resource-poor setting, and has been reproduced in many other developing countries and in internationally adopted strategies. These advances have only been achieved through the work of vital national child health institutions, the PNG Paediatric Society and the Department of Child Health at the University of PNG, with aid projects providing background support at various stages. Limiting factors to progressDespite some successful programs, many activities in PNG have not resulted in health gains where they are needed. Support areas of the health service remain weak: health and human resources management at all levels, drug and vaccine procurement, distribution and stock management, and health financing. Primary care, the most essential form of healthcare in rural areas but the least robust and most vulnerable level of the health service, has suffered the most because of these deficiencies.14 There needs to be a similar concentration on building capacity and commitment in these areas, improving efficiency, and minimising waste of resources and squandering of funds.15 The beginnings of progress in some of these areas have occurred. Health management is stronger in some provincial health services and hospitals now than it was 10 years ago, partly as a result of structural reforms and support and mentoring for management capacity provided by the AusAID-funded Health Sector Support Program. More needs to be done, but sustainable change will only occur slowly, tailoring strategies to individual situations — an approach that is at odds with some aid projects, whose designers often propose a “one size fits all” formula for rolling out the latest Big Idea, with little critical evaluation of outcomes. Equity and conditionality as principles of aidConditional aid, as Hughes suggests,11 might be a useful strategy, providing direct funding to carefully selected high-priority areas, with an agreement that certain process milestones will be reached. Potential examples might be Australian government funding of Hib vaccine, conditional upon completion of the national supplemental immunisation activities16 and achieving coverage of over 80%; or subsidising the purchase of snake antivenom, nevirapine and ceftriaxone, conditional upon improvements in drug procurement and national distribution systems. A further condition to ensure commitment and sustainability would be the understanding that the PNG government would take over responsibility for funding after a mutually agreed period of time. These targeted interventions would have broad benefits to the health service, would enable the implementation of new (to PNG) and highly effective interventions, and would enhance equity within PNG and between our two countries. However, there are some risks with conditional agreements. The withholding of interventions if conditions are not met would continue to hurt the people who are innocent of any waste or corruption — nurses and doctors who struggle every day to provide good healthcare, and the patients who suffer from the effects of a lack of quality services. Support should be given to local training institutions rather than aid projects running unsustainable training programs. A portion of the aid budget could be provided to build key areas of capacity by financially supporting individuals or groups committed to collaboration in ways that are appropriate to Melanesian society. AusAID and the PNG Health Department are currently proposing a step in this direction by establishing a Capacity Building Service Centre, which will place more emphasis on engaging locally successful individuals to act as mentors, build capacity, and result in significant changes to external contracting. The approaches outlined above are complementary and would reduce inequity between Australia and its Pacific neighbours. In some ways, they would be a departure from some current large aid projects, whose economic benefits often spin back to the donor country, and whose resources are consumed by project infrastructure that duplicates government institutions. I can see little place in a country like PNG for health projects that are managed by overseas private consortia. International tendering for health projects in a country that has no structured health management organisations can scarcely improve equity. This model has worked better for some aid development projects, such as road and water supply contracts, in which local engineering companies have won contracts, thus contributing to local development, employment and economic growth. However, the idea that health aid should be corporatised in a country that desperately needs an effective public health system is fundamentally flawed. No easy answersThere are no easy answers to how Australia can best assist regional countries. Ongoing engagement remains necessary at many levels — between governments, professional societies, institutions, and individuals. Without this there can be no mutual understanding, which provides the basis for progress and is crucial for regional peace. Some of the best examples of success suggest that sustained, quiet and modest-budget collaboration by committed groups or individuals who treat each other as equals will be the most effective strategy. As yet I have not addressed the problems of child mortality in the worst-affected country in the region — East Timor. A sustained collaborative approach would be of great benefit to this small country (Box 2). The view of PNG as a “failed state” is wrong. Progress is being made in many areas. Now is not the time for Australia to abandon PNG or our closest Pacific neighbours, but to learn from institutions and areas that have achieved much, to support them to do more, and to tailor approaches to specific situations. Australia also has much to learn from Pacific countries — quiet persistence, patience and a sense of community are qualities that might help us have a more realistic view of what progress really means. 1 Trends in mortality in children under 5 years (per 1000 live births) in the Asia-Pacific region over the past 50 years1-3 * In a Demographic Health Survey (DHS) in East Timor in 2003, the mortality rate in 2003 among children under 5 years of age was estimated to be 107 per 1000 live births. Mortality rate estimates from the previous eras, represented on the graph, are also based on retrospective data from the 2003 DHS, so the accuracy of these trends is uncertain. Reliable data from previous years are not available. 2 Increasing East Timor’s capacity to meet its child health needs The problems of capacity in East Timor are even deeper than in Papua New Guinea (PNG), and the child health system is in a much more embryonic stage of development. Currently, there are no East Timorese paediatricians, which is a major impediment to sustainable progress, local leadership, autonomy and direction. However, collaboration between the East Timorese Ministry of Health, the University of PNG and the Royal Australasian College of Physicians (RACP) will hopefully see East Timorese doctors trained in child health, largely in PNG, with some additional clinical experience in rural hospitals in Australia, over the next 5–10 years. This will provide training in settings that are similar to those of East Timor, foster personal and institutional connections between two developing countries, promote a developing-country university as a regional centre for high-quality specialist training, and minimise the risk of “brain drain” that would exist if specialist RACP Fellowship training were done in Australia.

Trevor Duke MD, FRACP, FJFICM

To the Editor: We report an unusual case of Ross River virus (RRV) disease in a 7-year-old child. The patient presented to her general practitioner with fever, rash and pain in the lower limbs. Swelling of the joints of the hands and left knee was found, with a widespread rash that covered the trunk, limbs and face. The rash comprised lesions of varying types, including maculopapular, vesicular and petechial lesions (Box). Rash in a child with Ross River virus disease The general practitioner transferred the patient to the state tertiary paediatric service. No antibiotics were given before transfer. On arrival at Princess Margaret Hospital for Children, Perth, the patient was unwell, with a fever of 38°C. A provisional diagnosis was made of septicaemia (probably meningococcal), and treatment was begun with intravenous ceftriaxone. Extensive investigations were performed, but results of all initial serological, polymerase chain reaction and culture investigations were negative. Rheumatology review was requested because of the prominent arthritic component of the illness. This revealed widespread polyarthritis, and the illness was felt to be a reactive or post-infectious process. The child’s family raised the possibility of RRV disease, as her grandmother had had this disease several years previously, and the child had stayed overnight at her grandmother’s home in a coastal lake area 2 weeks before disease onset. The area had abundant mosquitoes, as well as kangaroos, which are vertebrate amplifiers for RRV.1 Serological tests for RRV were performed 3 days after admission, and were negative for IgG and positive for IgM. Repeat serological testing during convalescence showed a fourfold rise in IgG titre (from 80 to 320), confirming the diagnosis of RRV disease. The patient’s rash decreased over several days. She had persistent synovitis in the left knee at review 3 weeks after admission. At review at 8 weeks all symptoms and signs had resolved, and she had full function. This case highlights the fact that, while RRV disease with severe symptoms and arthritic manifestations is uncommon in children, it nevertheless should still be considered in the differential diagnosis of children with a febrile and arthritic disease.1,2 This child’s illness appears to have been a reactive vasculitis and polyarthritis, which, while well recognised with other infections, is not well described in association with RRV disease in children. RRV arthritis is caused by joint infection, and treatment is currently based on empirical anti-inflammatory regimens. During the recent RRV disease epidemic in Western Australia, 1174 notifications for RRV disease were received between 1 October 2003 and 31 March 2004. Of these, 21 patients were aged 15 years or younger. Thus, while RRV disease is an infrequent illness in children, it does occur, and should be considered in the differential diagnosis of a child who presents with a febrile illness, rash and joint symptoms from an area with known autochthonous transmission of RRV.

Kynan T Feeney · Kevin J Murray · Amanda J Whittle · Gary K Dowse

Trial: Hatakka K, Savilahti E, Pönkä A, et al. Effect of long term consumption of probiotic milk on infections in children attending day care centres: double blind, randomised trial. BMJ 2001; 322: 1327-1329. QuestionIn children attending daycare centres (population), does long term consumption of probiotic milk (intervention) prevent infections (outcome)? Trial detailsObjective: To examine whether long term consumption of a probiotic milk could reduce gastrointestinal and respiratory infections in children in daycare centres. Design: Randomised, double-blind, placebo-controlled study over 7 months. Setting: 18 daycare centres in Helsinki, Finland. Participants: 571 healthy children aged 1–6 years; 282 in the intervention group (mean age, 4.6 years; SD, 1.5 years) and 289 in the control group (mean age, 4.4 years; SD, 1.5 years). Intervention: Milk with or without Lactobacillus GG. Average daily consumption of milk in both groups was 260 mL. Main outcome measures: Number of days with respiratory and gastrointestinal symptoms, absences from daycare because of illness, respiratory tract infections diagnosed by a doctor, and courses of antibiotics. Results: Children in the Lactobacillus group had fewer days of absence from daycare because of illness (4.9 days [95% CI, 4.4–5.5] versus 5.8 days [95% CI, 5.3–6.4]; absolute difference, 16% [P = 0.03]). Corresponding age-adjusted findings were 5.1 days (95% CI, 4.6–5.6) for the Lactobacillus group versus 5.7 days (95% CI, 5.2–6.3) for the control group; age-adjusted difference, 11% (P = 0.09). There was also a relative reduction of 17% in the number of children having respiratory infections with complications and lower respiratory tract infections (unadjusted absolute reduction, 8.6%; 95% CI, -17.2% to -0.1%; P = 0.05; age-adjusted odds ratio, 0.75; 95% CI, 0.52–1.09; P = 0.13) and a 19% relative reduction in courses of antibiotic for respiratory infection in the Lactobacillus group (unadjusted absolute reduction, -9.6%; 95% CI, -18.2 to -1.0; P = 0.03; adjusted odds ratio, 0.72; 95% CI, 0.50–1.03; P = 0.08). Conclusions: Lactobacillus GG may reduce respiratory infections and their severity among children in daycare. The effects of the probiotic Lactobacillus GG were modest but consistently in the same direction. CommentaryRationale for the trialChildren attending daycare centres are more likely to suffer gastrointestinal and respiratory infections than children cared for at home or in small family groups.1 There are public health and economic consequences, including direct medical costs and the indirect cost of parents taking time off work to care for sick children.2 Studies support the use of probiotics to reduce the incidence of antibiotic-associated diarrhoea3,4 and to hasten recovery from rotavirus diarrhoea.5 Probiotic bacteria may have a beneficial effect on the host immune response by altering intestinal microbial balance. Trial methodsThis was a randomised, double-blind, placebo-controlled trial carried out in 18 daycare centres in children aged 1–6 years over 7 months, which included winter. Daycare staff, parents, children and investigators were blinded or unaware of treatment allocation. Randomisation was effectively concealed, as children were allocated to intervention or control groups by a computer-generated block-randomisation procedure, with stratification on the basis of age and daycare centre. The randomisation procedure was successful in equalising baseline characteristics between placebo and active groups, apart from the control group having slightly younger children and more children with more than five recent infections. There was excellent follow-up, with nearly 90% of children completing the study, although reasons for withdrawal were not stated. Groups were analysed on an intention-to-treat basis. A random selection of 100 faecal samples was assessed to confirm compliance. Outcome measures included days of respiratory or gastrointestinal symptoms, days of absence from daycare because of illness, and number of upper respiratory tract infections complicated by lower respiratory tract infections. In summary, this trial was well designed and conducted. The average compliance in both groups was 60%. Unfortunately, there was a difference in age distribution between the two groups after randomisation. The investigators decided before the study that a minimum clinically relevant beneficial effect would be a 20% difference between the groups. Thus, based on previously reported episodes of illness in children attending daycare, they used a power calculation to estimate that, to detect a 20% difference with a power (or sensitivity) of 90% with 95% confidence, they needed to enrol 250 children per group. They were able to achieve this goal. New informationThe authors claim that milk containing Lactobacillus GG slightly reduced the incidence of respiratory infections and antibiotic treatment in children. The effect of probiotics was modest when adjustments were made for age. Of the measured variables — days of any illness, days of respiratory or gastrointestinal symptoms, and absence because of illness — only for absence because of illness was there a difference between the intervention and control groups that approached statistical significance, although there was a trend towards less illness for the other variables in the Lactobacillus group. It is useful to look at the tables of results in the article in question. Table 2 presents the raw data, and it is clear from scanning both the unadjusted and age-adjusted results that the effect is modest at best. The unadjusted days of absence because of illness was 4.9 days in the probiotic group and 5.8 days in the control group (ie, an improvement of 0.9 days); when adjusted for age, the difference between the probiotic group and the control group was 0.6 days. This raises the question of whether a reduction of 0.6 days in duration of absence because of illness is actually clinically useful. When the episodes of illness were diagnosed by a doctor (Table 3 of the article), the overall episodes of illness and courses of antibiotics prescribed for infections were significantly reduced for the probiotic group. Of all antibiotic courses prescribed, there were 119 in the probiotic group versus 144 in the control group, with an absolute percentage reduction of 8.0% (95%CI, -16.6 to 1.0). The P value was 0.07, thus approaching, but not reaching, statistical significance. However, the age-adjusted results (reported as odds ratios) are less impressive and the statistical significance of these findings is further reduced, with a 95% confidence interval crossing unity, and a P value of 0.17 — not significant. The results are thus entirely consistent with a chance difference. Implications for clinical practice This is the first large, high quality study to examine this interesting research question.6 There were no reported harmful effects of the Lactobacillus GG; costs were not fully explored in the report. The author of the accompanying editorial stated that probiotics “show promise but bigger studies are needed”.7 This is not entirely correct, as the study was adequately powered to detect a 20% difference and it failed to report a convincing clinical effect. However, it is likely that further clinical research will be undertaken in this and related areas, such as the use of probiotics to hasten recovery from acute gastroenteritis in children and the prevention of antibiotic-associated diarrhoea.

Mark G Coulthard MB BS, FRACP · Craig M Mellis MD, MPH, FRACP

Gastroenteritis in children is still a common reason for consulting a general practitioner and for hospital admission. Rotavirus is the most common cause of gastroenteritis in children and accounts for half of all hospital admissions for severe acute infectious diarrhoea. Most children with gastroenteritis do not develop dehydration and can be treated at home. Children with mild to moderate dehydration should be treated with low osmolarity oral rehydration solutions, and those with severe dehydration or shock need to be admitted for administration of intravenous fluids. Lactose-free feeds should not be routinely used after acute gastroenteritis, but there is some evidence that a lactose-free diet may reduce the duration of diarrhoea. Antimotility drugs are rarely indicated in children with gastroenteritis, as the potential risks outweigh the benefits. The development of a rotavirus vaccine would provide huge public health benefits and cost savings. Other preventive strategies include educating people about personal and food hygiene and encouraging breastfeeding.

Elizabeth J Elliott MD, FRACP, FRCPCH · Jacqueline R Dalby-Payne MB BS, PhD, FRACP

To the Editor: Recent media reports have claimed numerous health benefits for A2 milk1,2 (eg, “new wave milk”, “wonder milk”). It is becoming more widely available, particularly in health food shops, and is advertised on Queensland television. We believe it is important to offer clear information about this product and cows’ milk allergy. A2 milk is produced by cows homozygous for the A2 polymorphic variant (his→pro) at amino acid 67 of the b-casein gene. A difference in degradation patterns of the A1 and A2 variants is purported to lead to differences in immunological or pharmacological effects,3-5 which we will not comment on here. Regarding cow’s milk allergy, β-casein is one of at least seven proteins in cows’ milk with allergenic significance (α-, β- and κ-casein, α- and β-lactoglobulin, lactoferrin and transferrin). One would not expect a single amino-acid difference in one protein to have a significant effect on milk allergenicity. We have found in discussion with parents of milk-allergic children, as well as from inquiries from the community to AllergySA, that there is a perception that A2 milk may be less allergenic than “normal” milk (which contains A1 and A2 b-casein). Although most proponents of A2 milk have made no explicit claims about allergenicity — and indeed some have cautioned against the use of A2 in milk-allergic individuals — there have been media reports that may have led to this perception.6 However, these reports are misleading. For example, it is quite likely that children with a previous history of cow’s milk allergy who have been found to tolerate A2 milk have in fact “grown out” of the allergy, which is the usual natural history. Others may never have had true milk allergy. We obtained a sample of pure A2 milk from A2 Dairy Marketers (Acacia Ridge, QLD) and used it for skin-prick testing of 11 consecutive milk-allergic children (Box). The tests compared A2 milk with “normal” (A1/A2) milk and cow’s milk protein extract. The mean diameter of the wheal raised by normal milk was not significantly different to that raised by A2 milk (8.2 mm for normal milk v 10.7 mm for A2 milk; P = 0.09, paired t test). No patient had a negative reaction to A2 milk when the reaction to normal milk was positive. We did not perform an oral challenge with A2 milk in these children, as many had experienced severe allergic reactions, and the predictive value of a positive skin-prick test in the presence of a clear recent history of clinical allergy is high. We therefore caution that A2 milk should not be used by those with IgE-mediated cow’s milk allergy, particularly those who have had recent severe reactions to milk. Mean wheal diameter* (mm) on skin-prick testing Patient Normal milk† A2 milk† Cow’s milk extract‡ Histamine positive control 1 12 10 8 4.5 2 11.5 12 11 5.5 3 4 8 6 15 4 8 11 10.5 3 5 12 8 6 9 6 3 5 2 9 7 7 15 7 10 8 7 7.5 5 7.5 9 6 7.5 4 3.5 10 13 25 4.5 3 11 7 9 3 5 Mean 8.2 10.7 6.1 6.8 * As wheals produced are not necessarily circular, it is standard to report diameter as the mean of two measurements taken perpendicular to each other. Results for all negative controls were 0 mm. † Normal and A2 milk were stored frozen, and aliquots thawed for testing. They do not produce wheal reactions in non-allergic individuals. ‡ Cows’ milk extract is manufactured for skin-prick allergy testing by Hollister-Stier, Wash, USA, and purchased from Richard Thomson, Sydney, NSW.

William B Smith · Deryn Thompson · Margaret Kummerow · Patrick Quinn · Michael S Gold

To the Editor: There appear to be regional differences in the prescribing of amino acid infant formula in Australia. This is possibly due to differing practices in use of this formula as a first-line treatment for cow’s milk allergy or as a strategy for preventing allergy. This has financial implications, as the cost to the Pharmaceutical Benefits Scheme (PBS) of amino acid formula is $371 per prescription, compared with $106 for hydrolysed protein formula.1 In infants at high risk of allergic disease who are unable to be completely breastfed, there is evidence that prolonged feeding with a formula based on hydrolysed cow’s milk protein rather than conventional cow’s milk formula reduces infant and childhood allergy.2,3 There is no clear evidence that amino acid formula should be substituted for extensively hydrolysed protein formula as a primary preventive strategy.3 The current PBS indication for hydrolysed protein formula is treatment of intolerance to both cow’s milk and soy protein, but not primary allergy prevention. Similarly, current PBS guidelines restrict the use of amino acid formulas to proven intolerance to cow’s milk, soy protein and protein hydrolysate. Among children who are allergic to cow’s milk, 10% or less are also sensitive to protein hydrolysate formula.4 Thus, if current guidelines were followed, one might expect nine times the use of hydrolysed protein formula compared with amino acid formula. I obtained statistics on PBS items supplied for the period January 2003 to January 2004 from the Health Insurance Commission (www.hic.gov.au/statistics/dyn_pbs/forms/pbs_tab1.shtml) for hydrolysed protein formula (item numbers 2676W and 8259Q) and synthetic amino acid formula (item numbers 3066J, 8443J, 8574G and 8575H). These showed that 8374 hydrolysed protein formula items were supplied, half the number of amino acid formula items (16 886). Numbers of amino acid formula items supplied per 1000 children aged 4 years and younger were calculated using population statistics from the Australian Bureau of Statistics census figures 2001. These are compared in the Box with numbers of paediatric physicians per 1000 children (obtained from the Royal Australasian College of Physicians 2004) and paediatric allergists (derived from the Australasian Society of Clinical Immunology and Allergy membership handbook 2003). Prescribing practice varied markedly between states and territories. The Australian Capital Territory, New South Wales and Victoria had six to seven times more amino acid formula items per 1000 children than Western Australia. This did not appear related to numbers of paediatricians or paediatric allergists, as Western Australia had a similar number of paediatricians and more paediatric allergists per 1000 children than NSW and Victoria. The differences found were unlikely to be related to variation in numbers of adult immunology/allergy specialists, who are unlikely to treat many infants aged under 2 years. Nor were they likely to be due to differing prevalence of combined milk, soy and protein hydrolysate intolerance, as the prevalence of allergic disease does not differ markedly between Australian states. For example, the prevalence of atopic eczema at age 6 years in four cities (Adelaide, Melbourne, Sydney and Perth) was very similar, ranging from 10.1% to 11.4%.5 It seems unlikely that 80% of cases of combined intolerance are being missed in Western Australia. The estimated cost to the PBS for amino acid formula for 2003–2004 of $7 107 627 was 10 times that of hydrolysed formula ($757 570). Amino acid formula prescription rates, January 2003 to January 2004, compared with numbers of paediatric physicians and allergists per 1000 children aged 4 years or younger Amino acid formula items per 1000 children Paediatric physicians per 1000 children Paediatric allergists per 1000 children Australian Capital Territory 22.3 0.79 0 New South Wales 18.8 1.02 0.033 Victoria 17.8 1.00 0.030 Tasmania 12.3 0.53 0.033 South Australia 9.3 1.01 0.067 Northern Territory 9.1 0.92 0 Queensland 5.9 0.72 0.008 Western Australia 3.3 0.99 0.049

Andrew S Kemp

Objective: To examine predictors of death in young offenders who have received a custodial sentence using data routinely collected by juvenile justice services.Design: A retrospective cohort of 2849 (2625 male) 11–20-year-olds receiving their first custodial sentence between 1 January 1988 and 31 December 1999 was identified.Main outcome measures: Deaths, date and primary cause of death ascertained from study commencement to 1 March 2003 by data-matching with the National Death Index; measures comprising year of and age at admission, sex, offence profile, any drug offence, multiple admissions and ethnic and Indigenous status, obtained from departmental records.Results: The overall mortality rate was 7.2 deaths per 1000 person-years of observation. Younger admission age (hazard ratio [HR], 1.4; 95% CI, 1.0–1.9), repeat admissions (HR, 1.8; 95% CI, 1.1–2.9) and drug offences (HR, 1.5; 95% CI, 1.0–2.1) predicted early death. The role of ethnicity/Aboriginality could only be assessed in cohort entrants from 1996 to 1999. The Asian subcohort showed higher risk of death from drug-related causes (HR, 2.5; 95% CI, 1.1–5.5), more drug offences (relative risk ratio [RRR], 13; 95% CI, 8.5–20.0) and older admission age (oldest group v youngest: RRR, 9.3; 95% CI, 1.3–68.0) than non-Indigenous Australians. Although higher mortality was not identified in Indigenous Australians, this group was more likely to be admitted younger (oldest v youngest: RRR, 0.31; 95% CI, 0.15–0.63) and experience repeat admissions (RRR, 1.6; 95% CI, 1.0–2.4).Conclusions: Young offenders have a much higher death rate than other young Victorians. Early detention, multiple detentions and drug-related offences are indicators of high mortality risk. For these offenders, targeted healthcare while in custody and further mental healthcare and social support after release appear essential if we are to reduce the mortality rate in this group.

Carolyn Coffey BSc, GradDipEpi · Andrew W Lovett FRACP · Eileen Cini BSc(Hons) · George C Patton MD, FRANZCP · Rory Wolfe PhD · Paul Moran MD, MRCPsych

Objective: To describe the impact of a hepatitis A vaccination program for Indigenous children in north Queensland.Design: Enhanced surveillance of all notified cases of hepatitis A in north Queensland from 1996 to 2003.Setting: North Queensland; population, 596 500 people, including about 6900 Indigenous children aged under five years.Interventions: Hepatitis A vaccine was provided to Indigenous children in north Queensland from February 1999; two doses were recommended (at 18 months and 2 years of age), as was catch-up vaccination up to the sixth birthday.Results: In the 4 years 1996–1999, 787 cases of hepatitis A were notified in north Queensland, 237 (30%) of which were in Indigenous people. The average annual notification rates in Indigenous and non-Indigenous people during this period were 110 and 25 cases per 100 000 persons, respectively. In the first 4 years after introduction of the vaccination program (2000–2003), 66 cases of hepatitis A were notified. Only nine of the 66 (14%) were in Indigenous people. The average annual notification rates in Indigenous and non-Indigenous people in 2000–2003 were 4 and 2.5 cases per 100 000 persons, respectively.Conclusion: Hepatitis A seems to have been eradicated from Indigenous communities in north Queensland very soon after the vaccination program began. The rapid decline in notifications in non-Indigenous as well as Indigenous people suggests the program quickly interrupted chains of transmission from Indigenous children to the broader community. To our knowledge this is the first evidence that a hepatitis A vaccination program targeting a high-risk population within a community can reduce disease in the broader community. Hepatitis A vaccine should be provided to other high-risk Indigenous children elsewhere in Australia.

Jeffrey N Hanna MPH, FAFPHM · Susan L Hills MTH, FAFPHM · Jan L Humphreys

A series looking at everyday practice in the light of the evidence The health of the first two of Shakespeare’s seven ages of man — “the infant, mewling and puking” and “the whining school-boy”1 — is the domain of paediatricians, paediatric surgeons and general practitioners. Paediatrics is a relatively new discipline. The first Professor of Paediatrics in Australia, Sir Lorimer Dods, was appointed to the University of Sydney in 1949, but it was not until the late 1960s that a common written examination for paediatrics and adult medicine was abandoned, and the medical world formally recognised that children are not just little adults when it comes to health and disease. Today, paediatrics is a vibrant discipline encompassing the whole gamut of medical subspecialties, plus some exclusive to childhood, such as neonatology and child development. 1 Topics covered by the Paediatrics series Problem crying in infancy Acute infectious diarrhoea and dehydration Prevention and treatment of obesity Bedwetting, constipation and toileting issues Autism and language disorders Atopic disease Obstructive and other sleep disorders Developmental, learning and behavioural problems Minor trauma Care of the child in Australian society The differences between the disciplines of paediatrics and internal medicine are far greater than those embodied in the observation that children are not just scaled-down adults. Most of the children admitted for acute care to Australian paediatric hospitals are still cared for by general paediatricians, while it is a long time since generalists fulfilled this role for adults in tertiary hospitals. Subspecialty practice in paediatrics is also different: children’s health problems differ from those of adults, and subspecialty practice is almost exclusively hospital or university based. In this issue of the Journal, we begin a Practice Essentials series on paediatrics that will focus on the common problems confronting paediatricians and, by extension, general practitioners. We could have chosen to cover the recent technological and pharmacological advances in paediatrics (of which there are many), but most, although not all, of these have their origin in adult medicine, partly because of a correct emphasis on the ethical considerations of experimentation in children, and partly because of the commercial considerations of the pharmaceutical and biotechnology industries. The problems we have chosen to feature are low technology and predominantly concern development and neurocognition. In 2002, 62 general paediatricians in Victoria were asked which of the clinical conditions they dealt with were the most difficult.2 The responses were illuminating: 26% listed conduct disorder, 24% family dysfunction, 18% eating disorder, 15% autism spectrum disorder, 13% children at risk, and 13% attention deficit disorder. A previous study looked at how six paediatricians spent their time during 12 months of community paediatric consultation:3 3875 of 14 711 (26.3%) consultations were for attention deficit hyperactivity disorder and learning problems, and 1917 (13.0%) were for intellectual disability. The most common medical condition was asthma (1470 consultations; 10%), followed by constipation and/or encopresis (966; 6.6%), and urinary tract infection and enuresis combined (815; 5.6%). In planning this series, potential topics were focused through the “camera obscura” of a group of general practitioners. The topics chosen reflect very closely the experience of the Victorian general paediatricians, and include developmental and learning problems, autism and language disorders, constipation, bedwetting, and the most common eating disorder — obesity. A complete list of the topics is given in Box 1. Problematic childhood behaviours often make a major contribution to family dysfunction, and problem crying in infancy and sleep disturbance can test the resilience of families and marriages. These disorders often do not have a significant underlying physical problem, but may be symptomatic of difficult family relationships. Paediatricians have to keep in mind that their patients grow up and stewardship will pass into other hands. Equally, we cannot practise in isolation — we need to be aware of the childhood lifestyle disorders that are determinants of adult disease. The child is indeed “father of the man”.4 The pandemic of obesity facing Australia and other developed countries has its genesis in childhood. Realistic management of this problem must involve strategies to decrease sedentary behaviour, particularly television and computer viewing in childhood, and also to promote physical activity and appropriate dietary intake. Another determinant of lifestyle disease in adulthood is upper-airway obstruction. Children who snore have been shown to have a neurocognitive disadvantage compared with their peers of the same age and socioeconomic group. The series concludes with an article on the care of children in Australian society. This reminds us about the social determinants of health, which are fundamental to our understanding of child growth and development.5 Academic paediatricians teach medical students to be a conduit for entitlements for their patients. Paediatricians need to become more involved in policy-making and advocacy. Parents of children with autism, as well as those with children with chronic disability, are beginning to demand this and they should be heeded. Finally, in addressing each of these important topics, the authors have sought to apply the best evidence available. Each article includes some evidence-based practice tips, with the level of evidence graded according to the National Health and Medical Research Council’s system for assessing evidence (Box 2). We hope you will enjoy this series and that it will provoke debate and discussion. 2 Designation of levels of evidence of the National Health and Medical Research Council6 Level I: Evidence obtained from a systematic review of all relevant randomised controlled trials. Level II: Evidence obtained from at least one properly designed randomised controlled trial. Level III-1: Evidence obtained from well-designed pseudorandomised controlled trials (alternate allocation or some other method). Level III-2: Evidence obtained from comparative studies (including systematic reviews of such studies) with concurrent controls and allocation not randomised, cohort studies, case-control studies, or interrupted time series with a control group. Level III-3: Evidence obtained from comparative studies with historical control, two or more single-arm studies, or interrupted time series without a parallel control group. Level IV: Evidence obtained from case series, either post-test or pretest/post-test.

Richard T L Couper MB ChB, FRACP · Richard L Henry MD, FRACP, DipClinEpid · Michael South DCH, FRACP, MD