In 2012, the New Zealand (NZ) Government announced that one of the top ten goals of the public sector was to reduce the incidence of first episode acute rheumatic fever (ARF) by two-thirds, to 1.4 per 100,000 by mid-2017. To achieve this goal, the Government expanded the Rheumatic Fever Prevention Programme (RFPP), which was established in 2011.

However, there are important gaps in knowledge about the aetiology, pathogenesis and risk factors for ARF that currently limit the ability to develop and implement effective interventions. This paper presents the protocol for a study to identify risk factors for ARF in NZ, including a description of relevant aspects of ARF epidemiology in NZ, a literature review of the current knowledge of ARF risk and protective factors, and a description of the study’s aims, hypotheses, and methods.


Epidemiology and Impact of ARF

The epidemiology of ARF in NZ had been described in several sources up to 2014 when this study commenced. ARF epidemiology has also been described for some specific regions of NZ. These analyses showed several key features of the epidemiology that affect the design of research studies investigating this disease. Performed an additional descriptive analysis to inform the development of the protocol for this study. For the purpose of this brief analysis, we largely used hospitalisation data for the 5-year period 2010 to 2014 leading up to the start of this study, although some analyses covered the 20-year period of 1995 to 2014. These data were filtered and analysed using a standard method adopted by the NZ Ministry of Health for estimating the incidence of initial ARF hospitalisations. This method inevitably has some error in case classification as hospitalisations tend to over-count the number of diagnosed cases by around 25–33%. Additionally, at least 10% of true cases may go undetected.


Study Aims and Research Questions

The primary aim of this study was to identify potentially modifiable risk factors for ARF with the ultimate goal of producing robust evidence to support policies and programmes to decrease rates of ARF in high-risk NZ populations.



The study design is a prospective population-based case-control study. Considerable effort was put into considering potential risks to the study effectiveness and minimising these risks.

Once recruited into the study, cases and matched controls (or their parent/caregiver if aged less than 16 years old) will be interviewed using a study questionnaire, as described in full article. Blood, throat and nasal swabs, and hair specimens will also be obtained on a sub-sample of study participants, as described in full article.

The goal of the data analysis is to support effective investigation of the aims and research questions. There is a particular focus on identifying important modifiable risk factors. The analysis attempts to produce ‘…the most accurate (valid and precise) effect estimates obtainable from the data…rather than simply improve the fit…’. The approach to the analysis is described in full article.


Results and Conclusions

This study will quantify the association between ARF and a range of potentially modifiable risk factors including: adverse environmental exposures (notably household crowding, bed sharing, poor indoor environments, tobacco smoke exposure); limited resources for personal care (notably washing, teeth cleaning); poor nutrition (notably sugar sweetened beverages); and poor oral health (decayed, missing, and filled teeth).

It will also assess the potential protective effect of good access to primary health care services, including general practitioners, school-based sore throat management services, and oral health care. The study will investigate the association of ARF with preceding treatable infections (notably sore throats, skin infections and scabies) as well as host factors that might assist in targeting ARF prevention services (notably family history of ARF/RHD). All of this information can be used in the short to medium term to guide improved prevention measures and refine current programmes.

In addition, this study is seeking to provide information to better understand the role of circulating GAS types, immunological markers, inherited factors and early life exposures. Such knowledge will assist medium to longer term strategies, including the development of improved diagnostic markers for ARF and an effective GAS vaccine. The results will also contribute to international understanding about the pathophysiology of ARF.

A major strength of this study is that it is taking a comprehensive approach covering organism, host and environmental factors that may be associated with ARF. It has also started with a well-developed model of how these factors may influence the risk of ARF, based on an extensive review of published literature on ARF aetiology. Having closely matched controls as well as community controls will enable the study to examine a wide range of specific environmental risk factors.

There are very few high-quality studies that have investigated the aetiology of ARF. The current study aims to fill some of the considerable knowledge gaps that are currently preventing an evidence-informed approach to the prevention of this disease. We are using the methods outlined in this protocol paper to achieve this aim and look forward to reporting our findings in due course.