Craig Anderson graduated with an Honours degree in Statistics from the University of Glasgow, and then obtained his PhD in Statistics within the same department under the supervision of Dr Duncan Lee and Dr Nema Dean. The title of his thesis was "Identifying Boundaries in Spatial Modelling". After completing his PhD, he moved to Australia to take up a position as a Postdoctoral Research Fellow at the University of Technology Sydney, working with Professor Louise Ryan as part of the ARC Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS). Craig's main interests lie in statistics for health data; specifically spatial and spatio-temporal modelling of disease risk and the modelling of child growth trajectories.
Liu, H, Lindley, R, Alim, M, Felix, C, Gandhi, DBC, Verma, SJ, Tugnawat, DK, Syrigapu, A, Ramamurthy, RK, Pandian, JD, Walker, M, Forster, A, Hackett, ML, Anderson, CS, Langhorne, P, Murthy, GVS, Maulik, PK, Harvey, LA & Jan, S 2019, 'Family-led rehabilitation in India (ATTEND)—Findings from the process evaluation of a randomized controlled trial', International Journal of Stroke, vol. 14, no. 1, pp. 53-60.View/Download from: UTS OPUS or Publisher's site
© 2018 World Stroke Organization. Background: Training family carers to provide evidence-based rehabilitation to stroke patients could address the recognized deficiency of access to stroke rehabilitation in low-resource settings. However, our randomized controlled trial in India (ATTEND) found that this model of care was not superior to usual care alone. Aims: This process evaluation aimed to better understand trial outcomes through assessing trial implementation and exploring patients', carers', and providers' perspectives. Methods: Our mixed methods study included process, healthcare use data and patient demographics from all sites; observations and semi-structured interviews with participants (22 patients, 22 carers, and 28 health providers) from six sampled sites. Results: Intervention fidelity and adherence to the trial protocol was high across the 14 sites; however, early supported discharge (an intervention component) was not implemented. Within both randomized groups, some form of rehabilitation was widely accessed. ATTEND stroke coordinators provided counseling and perceived that sustaining patients' motivation to continue with rehabilitation in the face of significant emotional and financial stress as a key challenge. The intervention was perceived as an acceptable community-based package with education as an important component in raising the poor awareness of stroke. Many participants viewed family-led rehabilitation as a necessary model of care for poor and rural populations who could not access rehabilitation. Conclusion: Difficulty in sustaining patient and carer motivation for rehabilitation without ongoing support, and greater than anticipated access to routine rehabilitation may explain the lack of benefit in the trial. Nonetheless, family-led rehabilitation was seen as a concept worthy of further development.
Anderson, C, Hafen, R, Sofrygin, O, Ryan, L & members of the HBGDki Community 2019, 'Comparing predictive abilities of longitudinal child growth models.', Statistics in Medicine.View/Download from: Publisher's site
The Bill and Melinda Gates Foundation's Healthy Birth, Growth and Development knowledge integration project aims to improve the overall health and well-being of children across the world. The project aims to integrate information from multiple child growth studies to allow health professionals and policy makers to make informed decisions about interventions in lower and middle income countries. To achieve this goal, we must first understand the conditions that impact on the growth and development of children, and this requires sensible models for characterising different growth patterns. The contribution of this paper is to provide a quantitative comparison of the predictive abilities of various statistical growth modelling techniques based on a novel leave-one-out validation approach. The majority of existing studies have used raw growth data for modelling, but we show that fitting models to standardised data provide more accurate estimation and prediction. Our work is illustrated with an example from a study into child development in a middle income country in South America.
In many countries, the monitoring of child growth does not occur in a regular manner, and instead, we may have to rely on sporadic observations that are subject to substantial measurement error. In these countries, it can be difficult to identify patterns of poor growth, and faltering children may miss out on essential health interventions. The contribution of this paper is to provide a framework for pooling together multiple datasets, thus allowing us to overcome the issue of sparse data and provide improved estimates of growth. We use data from multiple longitudinal growth studies to construct a common correlation matrix that can be used in estimation and prediction of child growth. We propose a novel 2-stage approach: In stage 1, we construct a raw matrix via a set of univariate meta-analyses, and in stage 2, we smooth this raw matrix to obtain a more realistic correlation matrix. The methodology is illustrated using data from 16 child growth studies from the Bill and Melinda Gates Foundation's Healthy Birth Growth and Development knowledge integration project and identifies strong correlation for both height and weight between the ages of 4 and 12 years. We use a case study to provide an example of how this matrix can be used to help compute growth measures.
Huque, MH, Anderson, C, Walton, R, Woolford, S & Ryan, L 2018, 'Smooth individual level covariates adjustment in disease mapping', Biometrical Journal, vol. 60, no. 3, pp. 597-615.View/Download from: Publisher's site
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Spatial models for disease mapping should ideally account for covariates measured both at individual and area levels. The newly available 'indiCAR' model fits the popular conditional autoregresssive (CAR) model by accommodating both individual and group level covariates while adjusting for spatial correlation in the disease rates. This algorithm has been shown to be effective but assumes log-linear associations between individual level covariates and outcome. In many studies, the relationship between individual level covariates and the outcome may be non-log-linear, and methods to track such nonlinearity between individual level covariate and outcome in spatial regression modeling are not well developed. In this paper, we propose a new algorithm, smooth-indiCAR, to fit an extension to the popular conditional autoregresssive model that can accommodate both linear and nonlinear individual level covariate effects while adjusting for group level covariates and spatial correlation in the disease rates. In this formulation, the effect of a continuous individual level covariate is accommodated via penalized splines. We describe a two-step estimation procedure to obtain reliable estimates of individual and group level covariate effects where both individual and group level covariate effects are estimated separately. This distributed computing framework enhances its application in the Big Data domain with a large number of individual/group level covariates. We evaluate the performance of smooth-indiCAR through simulation. Our results indicate that the smooth-indiCAR method provides reliable estimates of all regression and random effect parameters. We illustrate our proposed methodology with an analysis of data on neutropenia admissions in New South Wales (NSW), Australia.
Anderson, C, Lee, D & Dean, N 2017, 'Spatial clustering of average risks and risk trends in Bayesian disease mapping.', Biometrical Journal, vol. 59, no. 1, pp. 41-56.View/Download from: UTS OPUS or Publisher's site
Spatiotemporal disease mapping focuses on estimating the spatial pattern in disease risk across a set of nonoverlapping areal units over a fixed period of time. The key aim of such research is to identify areas that have a high average level of disease risk or where disease risk is increasing over time, thus allowing public health interventions to be focused on these areas. Such aims are well suited to the statistical approach of clustering, and while much research has been done in this area in a purely spatial setting, only a handful of approaches have focused on spatiotemporal clustering of disease risk. Therefore, this paper outlines a new modeling approach for clustering spatiotemporal disease risk data, by clustering areas based on both their mean risk levels and the behavior of their temporal trends. The efficacy of the methodology is established by a simulation study, and is illustrated by a study of respiratory disease risk in Glasgow, Scotland.
Anderson, C & Ryan, LM 2017, 'A Comparison of Spatio-Temporal Disease Mapping Approaches Including an Application to Ischaemic Heart Disease in New South Wales, Australia.', International Journal of Environmental Research and Public Health, vol. 14, no. 2.View/Download from: UTS OPUS or Publisher's site
The field of spatio-temporal modelling has witnessed a recent surge as a result of developments in computational power and increased data collection. These developments allow analysts to model the evolution of health outcomes in both space and time simultaneously. This paper models the trends in ischaemic heart disease (IHD) in New South Wales, Australia over an eight-year period between 2006 and 2013. A number of spatio-temporal models are considered, and we propose a novel method for determining the goodness-of-fit for these models by outlining a spatio-temporal extension of the Moran's I statistic. We identify an overall decrease in the rates of IHD, but note that the extent of this health improvement varies across the state. In particular, we identified a number of remote areas in the north and west of the state where the risk stayed constant or even increased slightly.
Huque, MH, Anderson, C, Walton, R & Ryan, L 2016, 'Individual level covariate adjusted conditional autoregressive (indiCAR) model for disease mapping.', International Journal of Health Geographics, vol. 15, no. 1, pp. 1-13.View/Download from: UTS OPUS or Publisher's site
Mapping disease rates over a region provides a visual illustration of underlying geographical variation of the disease and can be useful to generate new hypotheses on the disease aetiology. However, methods to fit the popular and widely used conditional autoregressive (CAR) models for disease mapping are not feasible in many applications due to memory constraints, particularly when the sample size is large. We propose a new algorithm to fit a CAR model that can accommodate both individual and group level covariates while adjusting for spatial correlation in the disease rates, termed indiCAR. Our method scales well and works in very large datasets where other methods fail.We evaluate the performance of the indiCAR method through simulation studies. Our simulation results indicate that the indiCAR provides reliable estimates of all the regression and random effect parameters. We also apply indiCAR to the analysis of data on neutropenia admissions in New South Wales (NSW), Australia. Our analyses reveal that lower rates of neutropenia admissions are significantly associated with individual level predictors including higher age, male gender, residence in an outer regional area and a group level predictor of social disadvantage, the socio-economic index for areas. A large value for the spatial dependence parameter is estimated after adjusting for individual and area level covariates. This suggests the presence of important variation in the management of cancer patients across NSW.Incorporating individual covariate data in disease mapping studies improves the estimation of fixed and random effect parameters by utilizing information from multiple sources. Health registries routinely collect individual and area level information and thus could benefit by using indiCAR for mapping disease rates. Moreover, the natural applicability of indiCAR in a distributed computing framework enhances its application in the Big Data domain with a large number of individual/group level c...
Disease mapping is the field of spatial epidemiology interested in estimating the spatial pattern in disease risk across [Formula: see text] areal units. One aim is to identify units exhibiting elevated disease risks, so that public health interventions can be made. Bayesian hierarchical models with a spatially smooth conditional autoregressive prior are used for this purpose, but they cannot identify the spatial extent of high-risk clusters. Therefore, we propose a two-stage solution to this problem, with the first stage being a spatially adjusted hierarchical agglomerative clustering algorithm. This algorithm is applied to data prior to the study period, and produces [Formula: see text] potential cluster structures for the disease data. The second stage fits a separate Poisson log-linear model to the study data for each cluster structure, which allows for step-changes in risk where two clusters meet. The most appropriate cluster structure is chosen by model comparison techniques, specifically by minimizing the Deviance Information Criterion. The efficacy of the methodology is established by a simulation study, and is illustrated by a study of respiratory disease risk in Glasgow, Scotland.
Anderson, C, Lee, D & Dean, N 2014, 'Bayesian cluster detection via adjacency modelling', Proceedings of COMPSTAT 2014 21st International Conference on Computational Statistics hosting the 5th IASC World Conference, 21st International Conference on Computational Statistics (COMPSTAT 2014), Geneva, pp. 343-350.