Letter to the Editor

Authors


4 January 2009

Dear Editor,

RE: EPIDEMIOLOGY OF ACUTE RHEUMATIC FEVER IN NEW ZEALAND 1996–2005 BY JAINE R ET AL. 2008;44:564

Jaine et al. using hospital discharge data, outline the large ethnic disparities for acute rheumatic fever (ARF) in New Zealand over a recent decade suggesting that ARF epidemiology has not been reviewed recently. Importantly, this subject was reviewed by the New Zealand Child and Youth Epidemiology Service and published online in 2007 reviewing hospital discharge data by age, ethnicity and geographic area of origin but adding interpretation to disparities found by using the New Zealand Deprivation Index (NZ Dep Index) score. Maori and Pacific children are over-represented in the bottom quintile of the NZ Dep Index with a risk ratio for having ARF of 28.65 (95% CI 17.43–47.08) compared with the top quintile as the index.1

The NZ Dep Index is a small area index of deprivation used as a proxy for socio-economic status and made up of variables likely to influence health outcome, for example, income, employment, telephone access, transport access, living space and education.

While genetic/ethnic status cannot be ruled out as a precursor to rheumatic fever, public health measures seem likely to overcome this propensity as witnessed by the very low recurrence rates in a well-delivered secondary prophylaxis programme in New Zealand.2

Household crowding is an important facilitating factor for the spread of infectious disease.3 This was especially well shown in a New Zealand case control study for meningococcal disease, another disease spread via the naso-pharyngeal route,4 where the odds ratio for a case having a crowded house was 10.7 (95% CI 3.9–29.5). Improved housing would be an obvious intervention. However, for cases living in crowded houses, medical interventions can override environmental risk factors at least as shown in inner city Baltimore, USA, in the 1960s, where access to community primary health care clinics was enhanced and rheumatic fever declined by 60% over 10 years.5

Thus, in the absence of direction from the medical literature, we conducted a school-based sore throat clinic programme in a randomised controlled trial designed to improve access to health care known to be a problem in disadvantaged New Zealand Maori and Pacific populations. In spite of the size of this study and more than 86 000 person years, we were able to show only a clinically important effect on ARF and not able to support this statistically.6 This then led to a meta-analysis7 of community and school-based studies of streptococcal throat treatment programmes to prevent rheumatic fever. Of 653 studies screened, only six met the criteria of a randomised controlled trial or trials of before/after design examining treatment of sore throats in a community setting with ARF as an outcome. The data were pooled and a meta-analysis of these trials for ARF control produced a relative risk of 0.41 (95% CI 0.23, 0.70 P= 0.001), which favoured this approach with a suggestion of ∼60% reduction in rheumatic fever. This directs us towards this approach to rheumatic fever control, and we believe new public health initiatives are warranted for ‘at risk’ communities. This supports the approach already being undertaken, with success, in smaller New Zealand communities.8 An economic analysis is being performed to support such an initiative in larger populations.

Ancillary