Correcting the fixed-effect estimator for endogenous switching

In this paper, we propose a two-step estimator for panel data models in which a binary covariate is endogenous. In the first stage, a random-effects probit model is estimated, having the endogenous variable as the left-hand side variable. Correction terms are then constructed and included in the main regression.

Effect of nonnormal random components on level and power in group-randomized trials

The use of group-randomized trials is particularly widespread in the evaluation of health care, educational, and screening strategies. Group-randomized trials represent a subset of a larger class of designs often labeled nested, hierarchical, or multilevel and are characterized by the randomization of intact social units or groups, rather than individuals. The application of random effects models to group-randomized trials requires the specification of fixed and random components of the model. The underlying assumption is usually that these random components are normally distributed. This research is intended to determine if the Type I error rate and power are affected when the assumption of normality for the random component representing the group effect is violated. ^ In this study, simulated data are used to examine the Type I error rate, power, bias and mean squared error of the estimates of the fixed effect and the observed intraclass correlation coefficient (ICC) when the random component representing the group effect possess distributions with non-normal characteristics, such as heavy tails or severe skewness. The simulated data are generated with various characteristics (e.g. number of schools per condition, number of students per school, and several within school ICCs) observed in most small, school-based, group-randomized trials. The analysis is carried out using SAS PROC MIXED, Version 6.12, with random effects specified in a random statement and restricted maximum likelihood (REML) estimation specified. The results from the non-normally distributed data are compared to the results obtained from the analysis of data with similar design characteristics but normally distributed random effects. ^ The results suggest that the violation of the normality assumption for the group component by a skewed or heavy-tailed distribution does not appear to influence the estimation of the fixed effect, Type I error, and power. Negative biases were detected when estimating the sample ICC and dramatically increased in magnitude as the true ICC increased. These biases were not as pronounced when the true ICC was within the range observed in most group-randomized trials (i.e. 0.00 to 0.05). The normally distributed group effect also resulted in bias ICC estimates when the true ICC was greater than 0.05. However, this may be a result of higher correlation within the data. ^

The Effect of Slum Redevelopment on Child Health Outcomes: Evidence from Mumbai

As the population of urban poor living in slums increases, governments are trying to relocate people into government-provided free housing. Slum redevelopment affects every part of a household’s livelihood, but most importantly the health and wellbeing of younger generations. This paper investigates the effect of slum redevelopment schemes on child stunting levels. Data was collected in forty-one buildings under the slum-redevelopment program in Mumbai. The study demonstrates through a fixed effect regression analysis that an additional year of living in the building is associated with an increase in the height-for-age Z-score by 0.124 standard deviations. Possible explanations include an improvement in the overall hygienic environment, sanitation conditions, indoor air pollution, and access to health and water facilities. However, anecdotal evidence suggests that water contamination, loss of livelihood and increased expenses could worsen health outcomes for residents. This study prompts more research on the health effects of slum redevelopment projects, which are becoming increasingly common in the rapidly urbanizing developing world.

Investigating the patterns and determinants of life satisfaction in Germany following reunification

This paper investigates the patterns and determinants of life satisfaction in Germany following reunification. We implement a new fixed-effect estimator for ordinal life satisfaction in the German Socio-Economic Panel and find negative effects on life satisfaction from being recently fired, losing a spouse through either death or separation, and time spent in hospital, while we find strong positive effects from income and marriage. Using a new causal decomposition technique, we find that East Germans experienced a continued improvement in life satisfaction to which increased household incomes contributed around 12 percent. Most of the improvement is explained by better average circumstances, such as greater political freedom. For West Germans, we find little change in average life satisfaction over this period.

Quantifying the costs of drought : new evidence from life satisfaction data

We estimate the cost of droughts by matching rainfall data with individual life satisfaction. Our context is Australia over the period 2001 to 2004, which included a particularly severe drought. Using fixed-effect models, we find that a drought in spring has a detrimental effect on life satisfaction equivalent to an annual reduction in income of A$18,000. This effect, however, is only found for individuals living in rural areas. Using our estimates, we calculate that the predicted doubling of the frequency of spring droughts will lead to the equivalent loss in life satisfaction of just over 1% of GDP annually.

Determinants of China’s regional innovation capacity

This paper investigates the determinants of China’s regional innovation capacity (RIC) and variations in these determinants between different types of regions. Based on the framework of national innovation capacity (NIC) and research on innovation system, this paper develops a framework of RIC in the Chinese context. Using panel data from 1991 to 2009, clustering analysis is first employed to classify regions according to their innovation development path. Panel data regressions with fixed effect model are conducted to explore the determinants of RIC and how these vary across the different regional clusters. We find that the 30 regions can be clustered into three groups, and there are considerable differences in the drivers of RIC between these different regional groups.

The economics of happiness : a lifetime perspective

The three studies in this thesis focus on happiness and age and seek to contribute to our understanding of happiness change over the lifetime. The first study contributes by offering an explanation for what was evolving to a ‘stylised fact’ in the economics literature, the U-shape of happiness in age. No U-shape is evident if one makes a visual inspection of the age happiness relationship in the German socio-economic panel data, and, it seems counter-intuitive that we just have to wait until we get old to be happy. Eliminating the very young, the very old, and the first timers from the analysis did not explain away regression results supporting the U-shape of happiness in age, but fixed effect analysis did. Analysis revealed found that reverse causality arising from time-invariant individual traits explained the U-shape of happiness in age in the German population, and the results were robust across six econometric methods. Robustness was added to the German fixed effect finding by replicating it with the Australian and the British socio-economic panel data sets. During analysis of the German data an unexpected finding emerged, an exceedingly large negative linear effect of age on happiness in fixed-effect regressions. There is a large self-reported happiness decline by those who remain in the German panel. A similar decline over time was not evident in the Australian or the British data. After testing away age, time and cohort effects, a time-in-panel effect was found. Germans who remain in the panel for longer progressively report lower levels of happiness. Because time-in-panel effects have not been included in happiness regression specifications, our estimates may be biased; perhaps some economics of the happiness studies, that used German panel data, need revisiting. The second study builds upon the fixed-effect finding of the first study and extends our view of lifetime happiness to a cohort little visited by economists, children. Initial analysis extends our view of lifetime happiness beyond adulthood and revealed a happiness decline in adolescent (15 to 23 year-old) Australians that is twice the size of the happiness decline we see in older Australians (75 to 86 yearolds), who we expect to be unhappy due to declining income, failing health and the onset of death. To resolve a difference of opinion in the literature as to whether childhood happiness decreases, increases, or remains flat in age; survey instruments and an Internet-based survey were developed and used to collect data from four hundred 9 to 14 year-old Australian children. Applying the data to a Model of Childhood Happiness revealed that the natural environment life-satisfaction domain factor did not have a significant effect on childhood happiness. However, the children’s school environment and interactions with friends life-satisfaction domain factors explained over half a steep decline in childhood happiness that is three times larger than what we see in older Australians. Adding personality to the model revealed what we expect to see with adults, extraverted children are happier, but unexpectedly, so are conscientious children. With the steep decline in the happiness of young Australians revealed and explanations offered, the third study builds on the time-invariant individual trait finding from the first study by applying the Australian panel data to an Aggregate Model of Average Happiness over the lifetime. The model’s independent variable is the stress that arises from the interaction between personality and the life event shocks that affect individuals and peers throughout their lives. Interestingly, a graphic depiction of the stress in age relationship reveals an inverse U-shape; an inverse U-shape that looks like the opposite of the U-shape of happiness in age we saw in the first study. The stress arising from life event shocks is found to explain much of the change in average happiness over a lifetime. With the policy recommendations of economists potentially invoking unexpected changes in our lives, the ensuing stress and resulting (un)happiness warrant consideration before economists make policy recommendations.

Gender bias in schooling : the case for Bhutan

Differing parental considerations for girls and boys in households are a primary cause of the gender gap in school enrolment and educational attainment in developing countries, particularly in Sub-Saharan Africa and South Asia. While a number of studies have focused on the inequality of educational opportunities in South Asia, little is known about Bhutan. This study uses recent household expenditure data from the Bhutan Living Standard Survey to evaluate the gender gap in the allocation of resources for schooling. The findings, based on cross-sectional as well as household fixed-effect approaches, suggest that girls are less likely to enrol in school but are not allocated fewer resources once they are enrolled.

Postoperative chemotherapy for non-small cell lung cancer : a systematic review and meta-analysis

Background Postoperative chemotherapy is currently not recommended for resected non-small cell lung cancer in many countries and centers. Recently, results of several large randomized clinical trials were reported with conflicting evidence. Accordingly, we sought to determine whether postoperative chemotherapy is associated with improved survival compared with that after surgical intervention alone. Methods Randomized clinical trials with cisplatin- or uracil plus ftorafur-containing regimens were included and evaluated separately. A systematic review that included randomized clinical trials performed before 1995 was identified and found to be of adequate quality. Further randomized controlled trials were identified by searching MEDLINE, EMBASE, and the Cochrane Controlled Trials Register from 1995 through 2004. In addition, the reference lists of articles and conference abstracts were searched. The logarithm of the hazard ratio and its standard error were calculated, and a fixed-effect model was used to combine the estimates. Results There were 7200 patients enrolled in 19 trials included in the analyses. An overall estimate of 13% relative reduction in mortality (95% confidence interval, 7%-19%) was found. There was 11% relative reduction in mortality associated with postoperative cisplatin (95% confidence interval, 4%-18%; P = .004) and 17% associated with uracil plus ftorafur (95% confidence interval, 5%-27%; P = .006) compared with that after surgical intervention alone. This means that there would be an additional survivor at 5 years for 25 patients treated with cisplatin or for 30 patients treated with uracil plus ftorafur. Conclusions Postoperative chemotherapy is associated with improved survival compared with that after surgical intervention alone. Selected patients with completely resected non-small cell lung cancer should be offered chemotherapy. Copyright © 2004 by The American Association for Thoracic Surgery.

Institutional determinants of growth-aspiration entrepreneurship

The thesis is a country-level study on the institutional and human capital determinants of growth-aspiration entrepreneurial activity. By using country-level panel-data analysis, the study is to our knowledge the first to test to what extent country-level human capital accumulation is associated with the prevalence of growth-aspiration entrepreneurship. Overall findings of the study suggest that there are different effects of the institutional determinants on the prevalence of growth-aspiration entrepreneurship in developing countries and developed countries. The study also found that country-level human capital moderates the effects of the institutional environment.

Understanding causal relationships between the built and social environment, health and associated behaviours and risk factors

Introduction: Built environment interventions designed to reduce non-communicable diseases and health inequity, complement urban planning agendas focused on creating more ‘liveable’, compact, pedestrian-friendly, less automobile dependent and more socially inclusive cities.However, what constitutes a ‘liveable’ community is not well defined. Moreover, there appears to be a gap between the concept and delivery of ‘liveable’ communities. The recently funded NHMRC Centre of Research Excellence (CRE) in Healthy Liveable Communities established in early 2014, has defined ‘liveability’ from a social determinants of health perspective. Using purpose-designed multilevel longitudinal data sets, it addresses five themes that address key evidence-base gaps for building healthy and liveable communities. The CRE in Healthy Liveable Communities seeks to generate and exchange new knowledge about: 1) measurement of policy-relevant built environment features associated with leading non-communicable disease risk factors (physical activity, obesity) and health outcomes (cardiovascular disease, diabetes) and mental health; 2) causal relationships and thresholds for built environment interventions using data from longitudinal studies and natural experiments; 3) thresholds for built environment interventions; 4) economic benefits of built environment interventions designed to influence health and wellbeing outcomes; and 5) factors, tools, and interventions that facilitate the translation of research into policy and practice. This evidence is critical to inform future policy and practice in health, land use, and transport planning. Moreover, to ensure policy-relevance and facilitate research translation, the CRE in Healthy Liveable Communities builds upon ongoing, and has established new, multi-sector collaborations with national and state policy-makers and practitioners. The symposium will commence with a brief introduction to embed the research within an Australian health and urban planning context, as well as providing an overall outline of the CRE in Healthy Liveable Communities, its structure and team. Next, an overview of the five research themes will be presented. Following these presentations, the Discussant will consider the implications of the research and opportunities for translation and knowledge exchange. Theme 2 will establish whether and to what extent the neighbourhood environment (built and social) is causally related to physical and mental health and associated behaviours and risk factors. In particular, research conducted as part of this theme will use data from large-scale, longitudinal-multilevel studies (HABITAT, RESIDE, AusDiab) to examine relationships that meet causality criteria via statistical methods such as longitudinal mixed-effect and fixed-effect models, multilevel and structural equation models; analyse data on residential preferences to investigate confounding due to neighbourhood self-selection and to use measurement and analysis tools such as propensity score matching and ‘within-person’ change modelling to address confounding; analyse data about individual-level factors that might confound, mediate or modify relationships between the neighbourhood environment and health and well-being (e.g., psychosocial factors, knowledge, perceptions, attitudes, functional status), and; analyse data on both objective neighbourhood characteristics and residents’ perceptions of these objective features to more accurately assess the relative contribution of objective and perceptual factors to outcomes such as health and well-being, physical activity, active transport, obesity, and sedentary behaviour. At the completion of the Theme 2, we will have demonstrated and applied statistical methods appropriate for determining causality and generated evidence about causal relationships between the neighbourhood environment, health, and related outcomes. This will provide planners and policy makers with a more robust (valid and reliable) basis on which to design healthy communities.

Family-based programmes for preventing smoking by children and adolescents (Review)

Background There is evidence that family and friends influence children's decisions to smoke. Objectives To assess the effectiveness of interventions to help families stop children starting smoking. Search methods We searched 14 electronic bibliographic databases, including the Cochrane Tobacco Addiction Group specialized register, MEDLINE, EMBASE, PsycINFO, CINAHL unpublished material, and key articles' reference lists. We performed free-text internet searches and targeted searches of appropriate websites, and hand-searched key journals not available electronically. We consulted authors and experts in the field. The most recent search was 3 April 2014. There were no date or language limitations. Selection criteria Randomised controlled trials (RCTs) of interventions with children (aged 5-12) or adolescents (aged 13-18) and families to deter tobacco use. The primary outcome was the effect of the intervention on the smoking status of children who reported no use of tobacco at baseline. Included trials had to report outcomes measured at least six months from the start of the intervention. Data collection and analysis We reviewed all potentially relevant citations and retrieved the full text to determine whether the study was an RCT and matched our inclusion criteria. Two authors independently extracted study data for each RCT and assessed them for risk of bias. We pooled risk ratios using a Mantel-Haenszel fixed effect model. Main results Twenty-seven RCTs were included. The interventions were very heterogeneous in the components of the family intervention, the other risk behaviours targeted alongside tobacco, the age of children at baseline and the length of follow-up. Two interventions were tested by two RCTs, one was tested by three RCTs and the remaining 20 distinct interventions were tested only by one RCT. Twenty-three interventions were tested in the USA, two in Europe, one in Australia and one in India. The control conditions fell into two main groups: no intervention or usual care; or school-based interventions provided to all participants. These two groups of studies were considered separately. Most studies had a judgement of 'unclear' for at least one risk of bias criteria, so the quality of evidence was downgraded to moderate. Although there was heterogeneity between studies there was little evidence of statistical heterogeneity in the results. We were unable to extract data from all studies in a format that allowed inclusion in a meta-analysis. There was moderate quality evidence family-based interventions had a positive impact on preventing smoking when compared to a no intervention control. Nine studies (4810 participants) reporting smoking uptake amongst baseline non-smokers could be pooled, but eight studies with about 5000 participants could not be pooled because of insufficient data. The pooled estimate detected a significant reduction in smoking behaviour in the intervention arms (risk ratio [RR] 0.76, 95% confidence interval [CI] 0.68 to 0.84). Most of these studies used intensive interventions. Estimates for the medium and low intensity subgroups were similar but confidence intervals were wide. Two studies in which some of the 4487 participants already had smoking experience at baseline did not detect evidence of effect (RR 1.04, 95% CI 0.93 to 1.17). Eight RCTs compared a combined family plus school intervention to a school intervention only. Of the three studies with data, two RCTS with outcomes for 2301 baseline never smokers detected evidence of an effect (RR 0.85, 95% CI 0.75 to 0.96) and one study with data for 1096 participants not restricted to never users at baseline also detected a benefit (RR 0.60, 95% CI 0.38 to 0.94). The other five studies with about 18,500 participants did not report data in a format allowing meta-analysis. One RCT also compared a family intervention to a school 'good behaviour' intervention and did not detect a difference between the two types of programme (RR 1.05, 95% CI 0.80 to 1.38, n = 388). No studies identified any adverse effects of intervention. Authors' conclusions There is moderate quality evidence to suggest that family-based interventions can have a positive effect on preventing children and adolescents from starting to smoke. There were more studies of high intensity programmes compared to a control group receiving no intervention, than there were for other compairsons. The evidence is therefore strongest for high intensity programmes used independently of school interventions. Programmes typically addressed family functioning, and were introduced when children were between 11 and 14 years old. Based on this moderate quality evidence a family intervention might reduce uptake or experimentation with smoking by between 16 and 32%. However, these findings should be interpreted cautiously because effect estimates could not include data from all studies. Our interpretation is that the common feature of the effective high intensity interventions was encouraging authoritative parenting (which is usually defined as showing strong interest in and care for the adolescent, often with rule setting). This is different from authoritarian parenting (do as I say) or neglectful or unsupervised parenting.

Combined pharmacotherapy and psychological therapies for post traumatic stress disorder (PTSD)

BACKGROUND PTSD is an anxiety disorder related to exposure to a severe psychological trauma. Symptoms include re-experiencing the event, avoidance and arousal as well as distress and impairment resulting from these symptoms.Guidelines suggest a combination of both psychological therapy and pharmacotherapy may enhance treatment response, especially in those with more severe PTSD or in those who have not responded to either intervention alone. OBJECTIVES To assess whether the combination of psychological therapy and pharmacotherapy provides a more efficacious treatment for PTSD than either of these interventions delivered separately. SEARCH STRATEGY Searches were conducted on the trial registers kept by the CCDAN group (CCDANCTR-Studies and CCDANCTR-References) to June 2010. The reference sections of included studies and several conference abstracts were also scanned. SELECTION CRITERIA Patients of any age or gender, with chronic or recent onset PTSD arising from any type of event relevant to the diagnostic criteria were included. A combination of any psychological therapy and pharmacotherapy was included and compared to wait list, placebo, standard treatment or either intervention alone. The primary outcome was change in total PTSD symptom severity. Other outcomes included changes in functioning, depression and anxiety symptoms, suicide attempts, substance use, withdrawal and cost. DATA COLLECTION AND ANALYSIS Two or three review authors independently selected trials, assessed their 'risk of bias' and extracted trial and outcome data. We used a fixed-effect model for meta-analysis. The relative risk was used to summarise dichotomous outcomes and the mean difference and standardised mean difference were used to summarise continuous measures. MAIN RESULTS Four trials were eligible for inclusion, one of these trials (n =24) was on children and adolescents. All used an SSRI and prolonged exposure or a cognitive behavioural intervention. Two trials compared combination treatment with pharmacological treatment and two compared combination treatment with psychological treatment. Only two trials reported a total PTSD symptom score and these data could not be combined. There was no strong evidence to show if there were differences between the group receiving combined interventions compared to the group receiving psychological therapy (mean difference 2.44, 95% CI -2.87, 7.35 one study, n=65) or pharmacotherapy (mean difference -4.70, 95% CI -10.84 to 1.44; one study, n = 25). Trialists reported no significant differences between combination and single intervention groups in the other two studies. There were very little data reported for other outcomes, and in no case were significant differences reported. AUTHORS' CONCLUSIONS There is not enough evidence available to support or refute the effectiveness of combined psychological therapy and pharmacotherapy compared to either of these interventions alone. Further large randomised controlled trials are urgently required.

WNT16 influences bone mineral density, cortical bone thickness, bone strength, and osteoporotic fracture risk

We aimed to identify genetic variants associated with cortical bone thickness (CBT) and bone mineral density (BMD) by performing two separate genome-wide association study (GWAS) meta-analyses for CBT in 3 cohorts comprising 5,878 European subjects and for BMD in 5 cohorts comprising 5,672 individuals. We then assessed selected single-nucleotide polymorphisms (SNPs) for osteoporotic fracture in 2,023 cases and 3,740 controls. Association with CBT and forearm BMD was tested for ~2.5 million SNPs in each cohort separately, and results were meta-analyzed using fixed effect meta-analysis. We identified a missense SNP (Thr>Ile; rs2707466) located in the WNT16 gene (7q31), associated with CBT (effect size of -0.11 standard deviations [SD] per C allele, P = 6.2×10-9). This SNP, as well as another nonsynonymous SNP rs2908004 (Gly>Arg), also had genome-wide significant association with forearm BMD (-0.14 SD per C allele, P = 2.3×10-12, and -0.16 SD per G allele, P = 1.2×10-15, respectively). Four genome-wide significant SNPs arising from BMD meta-analysis were tested for association with forearm fracture. SNP rs7776725 in FAM3C, a gene adjacent to WNT16, was associated with a genome-wide significant increased risk of forearm fracture (OR = 1.33, P = 7.3×10-9), with genome-wide suggestive signals from the two missense variants in WNT16 (rs2908004: OR = 1.22, P = 4.9×10-6 and rs2707466: OR = 1.22, P = 7.2×10-6). We next generated a homozygous mouse with targeted disruption of Wnt16. Female Wnt16-/- mice had 27% (P<0.001) thinner cortical bones at the femur midshaft, and bone strength measures were reduced between 43%-61% (6.5×10-13<P<5.9×10-4) at both femur and tibia, compared with their wild-type littermates. Natural variation in humans and targeted disruption in mice demonstrate that WNT16 is an important determinant of CBT, BMD, bone strength, and risk of fracture. © 2012 Zheng et al.

Association between endometriosis and the interleukin 1A (IL1A) locus

STUDY QUESTION Are single-nucleotide polymorphisms (SNPs) at the interleukin 1A (IL1A) gene locus associated with endometriosis risk? SUMMARY ANSWER We found evidence for strong association between IL1A SNPs and endometriosis risk. WHAT IS KNOWN ALREADY Genetic factors contribute substantially to the complex aetiology of endometriosis and the disease has an estimated heritability of ∼51%. We, and others, have conducted genome-wide association (GWA) studies for endometriosis, which identified a total of nine independent risk loci. Recently, two small Japanese studies reported eight SNPs (rs6542095, rs11677416, rs3783550, rs3783525, rs3783553, rs2856836, rs1304037 and rs17561) at the IL1A gene locus as suggestively associated with endometriosis risk. There is also evidence of a link between inflammation and endometriosis. STUDY DESIGN, SIZE, DURATION We sought to further investigate the eight IL1A SNPs for association with endometriosis using an independent sample of 3908 endometriosis cases and 8568 controls of European and Japanese ancestry. The study was conducted between October 2013 and July 2014. PARTICIPANTS/MATERIALS, SETTING, METHODS By leveraging GWA data from our previous multi-ethnic GWA meta-analysis for endometriosis, we imputed variants in the IL1A region, using a recent 1000 Genomes reference panel. After combining summary statistics for the eight SNPs from our European and Japanese imputed data with the published results, a fixed-effect meta-analysis was performed. An additional meta-analysis restricted to endometriosis cases with moderate-to-severe (revised American Fertility Society stage 3 or 4) disease versus controls was also performed. MAIN RESULTS AND THE ROLE OF CHANCE All eight IL1A SNPs successfully replicated at P < 0.014 in the European imputed data with concordant direction and similar size to the effects reported in the original Japanese studies. Of these, three SNPs (rs6542095, rs3783550 and rs3783525) also showed association with endometriosis at a nominal P < 0.05 in our independent Japanese sample. Fixed-effect meta-analysis of the eight SNPs for moderate-to-severe endometriosis produced a genome-wide significant association for rs6542095 (odds ratio = 1.21; 95% confidence interval = 1.13–1.29; P = 3.43 × 10−8). LIMITATIONS, REASONS FOR CAUTION The meta-analysis for moderate-to-severe endometriosis included results of moderate-to-severe endometriosis cases from our European data sets and all endometriosis cases from the Japanese data sets, as disease stage information was not available for endometriosis cases in the Japanese data sets. WIDER IMPLICATIONS OF THE FINDINGS SNP rs6542095 is located ∼2.3 kb downstream of the IL1A gene and ∼6.9 kb upstream of cytoskeleton-associated protein 2-like (CKAP2L) gene. The IL1A gene encodes the IL1a protein, a member of the interleukin 1 cytokine family which is involved in various immune responses and inflammatory processes. These results provide important replication in an independent Japanese sample and, for the first time, association of the IL1A locus in endometriosis patients of European ancestry. SNPs within the IL1A locus may regulate other genes, but if IL1A is the target, our results provide supporting evidence for a link between inflammatory responses and the pathogenesis of endometriosis. STUDY FUNDING/COMPETING INTEREST(S) The research was funded by grants from the Australian National Health and Medical Research Council and Wellcome Trust. None of the authors has competing interests for the study.