Greg Distiller

Continuous-time spatially explicit capture–recapture models, with an application to a jaguar camera-trap survey

Summary Many capture–recapture surveys of wildlife populations operate in continuous time, but detections are typically aggregated into occasions for analysis, even when exact detection times are available. This discards information and introduces subjectivity, in the form of decisions about occasion definition. We develop a spatiotemporal Poisson process model for spatially explicit capture–recapture (SECR) surveys that operate continuously and record exact detection times. We show that, except in some special cases (including the case in which detection probability does not change within occasion), temporally aggregated data do not provide sufficient statistics for density and related parameters, and that when detection probability is constant over time, our continuous-time (CT) model is equivalent to an existing model based on detection frequencies. We use the model to estimate jaguar density from a camera-trap survey and conduct a simulation study to investigate the properties of a CT estimator and discrete-occasion estimators with various levels of temporal aggregation. This includes investigation of the effect on the estimators of spatiotemporal correlation induced by animal movement. The CT estimator is found to be unbiased and more precise than discrete-occasion estimators based on binary capture data (rather than detection frequencies) when there is no spatiotemporal correlation. It is also found to be only slightly biased when there is correlation induced by animal movement, and to be more robust to inadequate detector spacing, while discrete-occasion estimators with binary data can be sensitive to occasion length, particularly in the presence of inadequate detector spacing. Our model includes as a special case a discrete-occasion estimator based on detection frequencies, and at the same time lays a foundation for the development of more sophisticated CT models and estimators. It allows modelling within-occasion changes in detectability, readily accommodates variation in detector effort, removes subjectivity associated with user-defined occasions and fully utilizes CT data. We identify a need for developing CT methods that incorporate spatiotemporal dependence in detections and see potential for CT models being combined with telemetry-based animal movement models to provide a richer inference framework.

HIV/AIDS in the South African construction industry: an empirical study

South Africa has one of the highest HIV/AIDS prevalence rates, with 11% of South Africans currently estimated to be HIV positive. The construction industry has one of the highest prevalence rates of any economic sector. To indicate the nature and extent of HIV/AIDS in the SA construction industry and to establish the degree of association between certain risk factors and the prevalence of HIV/AIDS, the results of HIV tests on 10 243 construction employees drawn from 55 companies nationwide are reported. In addition to the HIV test, a test for tuberculosis (TB) was conducted but proved inconclusive. Permanently employed, semi‐skilled operators and drivers, as well as skilled employees, have the highest prevalence rate of HIV/AIDS; even greater than casual labourers. The group displaying the next highest prevalence is temporarily employed general labourers. With regard to age, middle‐aged employees were shown to be the most at risk. Geographically, KwaZulu‐Natal exhibits the highest HIV prevalence rate, with the Western Cape the lowest. It is recommended that a compulsory, industry‐wide initiative be adopted, funded via levies, involving education, testing and treatment programmes supported via formal certification as part of the completion of work certification process.

Association between psychopathology and substance use among school-going adolescents in Cape Town, South Africa.

Abstract Limited information exists regarding the association between psychopathology and specific substance use in young people both globally and locally. We examined the association between psychopathology and substance use in high school students to determine the nature of the associations and the role of demographic factors in these associations. Grade 8 (N = 480) and Grade II (N = 459) students from 39 high schools in Cape Town, South Africa, completed a self-administered questionnaire. Psychopathology information was obtained from total scores on the Harvard Trauma Questionnaire, Beck Depression Inventory and the Zung Self-Rating Anxiety Scale. Lifetime prevalence rates were calculated for tobacco, alcohol, cannabis and inhalant use. Associations between psychopathology and substance use were determined using regression analyses and structural equation modeling. On adjusting for demographic characteristics, significant associations were found between PTSD and all substance use, between depression, alcohol, cannabis and inhalant use, and between anxiety and cannabis use. The associations of PTSD and depression with alcohol and cannabis use, and between anxiety and cannabis use, were moderated by grade. Although psychopathology and substance use were associated with each other, these associations occurred in accordance with substance and grade. Roles for gender, age and ethnicity emerged in the associations, but further investigation is recommended to examine these.

Rate of pulmonary function decline in South African children with cystic fibrosis

Background . Pulmonary function tests (PFTs) objectively measure the extent and progression of cystic fibrosis (CF) lung disease. The rate of lung function decline in developing countries has not previously been studied. Aim . To investigate the average annual rates of pulmonary function decline in South African children with CF from 1999 to 2006. Methodology : The medical records and best PFT over 3-monthly intervals of children attending the CF clinic at Red Cross War Memorial Children’s Hospital, Cape Town, were retrospectively reviewed and analysed using the mixed model regression method. Results . A total of 1 139 PFT were recorded on 79 patients, with a median (interquartile range) of 14 (6 - 21) PFTs per patient. The mean (standard error) forced expiratory volume in 1 second (FEV1) at age 6 years was estimated at 73.83 (3.34) per cent predicted with an FEV1 decline of 0.23 (0.43)% per annum. FEV1 at age 6 was affected by age at CF diagnosis, genotype, and year of birth. Rate of FEV1 decline was significantly affected by Pseudomonas aeruginosa colonisation and genotype. Conclusions . Although FEV1 at age 6 years was low compared with developed countries, the annual rate of FEV1 decline in South African children with CF was minimal, setting the scene for improved survival in this population.

A spatially explicit capture–recapture estimator for single‐catch traps

Abstract Single‐catch traps are frequently used in live‐trapping studies of small mammals. Thus far, a likelihood for single‐catch traps has proven elusive and usually the likelihood for multicatch traps is used for spatially explicit capture–recapture (SECR) analyses of such data. Previous work found the multicatch likelihood to provide a robust estimator of average density. We build on a recently developed continuous‐time model for SECR to derive a likelihood for single‐catch traps. We use this to develop an estimator based on observed capture times and compare its performance by simulation to that of the multicatch estimator for various scenarios with nonconstant density surfaces. While the multicatch estimator is found to be a surprisingly robust estimator of average density, its performance deteriorates with high trap saturation and increasing density gradients. Moreover, it is found to be a poor estimator of the height of the detection function. By contrast, the single‐catch estimators of density, distribution, and detection function parameters are found to be unbiased or nearly unbiased in all scenarios considered. This gain comes at the cost of higher variance. If there is no interest in interpreting the detection function parameters themselves, and if density is expected to be fairly constant over the survey region, then the multicatch estimator performs well with single‐catch traps. However if accurate estimation of the detection function is of interest, or if density is expected to vary substantially in space, then there is merit in using the single‐catch estimator when trap saturation is above about 60%. The estimators performance is improved if care is taken to place traps so as to span the range of variables that affect animal distribution. As a single‐catch likelihood with unknown capture times remains intractable for now, researchers using single‐catch traps should aim to incorporate timing devices with their traps.

The Effect of Features of the Metabolic Syndrome on Atherosclerotic Risk in Relatively Long-Surviving Patients with Type 1 Diabetes

BACKGROUND Increasing numbers of patients with type 1 diabetes (T1DM) are developing features of the metabolic syndrome. The additional effect of this on the development of atherosclerosis, as inferred by the carotid artery intima media thickness (IMT), has not previously been assessed. The aim of this study was to assess the effect of features of the metabolic syndrome on carotid artery IMT in a cohort of long-surviving patients with T1DM. METHODS Long-surviving patients with T1DM attending the Centre for Diabetes and Endocrinology were assessed regarding their risk factor profile. All underwent measurement of carotid artery IMT. In all, 156 patients who had T1DM for more than 18 years had their carotid artery IMT measured. All had been attending the clinic for over 10 years, and past clinical and laboratory records were available. RESULTS A total of 37 patients had metabolic syndrome according to the International Diabetes Federation (IDF) definition. Those with metabolic syndrome had a significantly increased carotid artery IMT (P = 0.003) compared to those without the syndrome. There was a significant relationship between the number of features of metabolic syndrome and increased atherosclerotic risk according to the carotid artery IMT (P = 0.01). A significant correlation was found between carotid artery IMT and both waist circumference (P < 0.001) and insulin resistance (P = 0.005). CONCLUSIONS In long-surviving patients with T1DM, those that develop metabolic syndrome are more likely to have thicker carotid artery IMT, and, by inference, be at higher risk of atherosclerosis and possibly cardiovascular disease. A linear relationship was present between both waist circumference and insulin resistance and carotid artery IMT.

Nonlinear mixed effects modeling of gametocyte carriage in patients with uncomplicated malaria

BackgroundGametocytes are the sexual form of the malaria parasite and the main agents of transmission. While there are several factors that influence host infectivity, the density of gametocytes appears to be the best single measure that is related to the human hosts infectivity to mosquitoes. Despite the obviously important role that gametocytes play in the transmission of malaria and spread of anti-malarial resistance, it is common to estimate gametocyte carriage indirectly based on asexual parasite measurements. The objective of this research was to directly model observed gametocyte densities over time, during the primary infection.MethodsOf 447 patients enrolled in sulphadoxine-pyrimethamine therapeutic efficacy studies in South Africa and Mozambique, a subset of 103 patients who had no gametocytes pre-treatment and who had at least three non-zero gametocyte densities over the 42-day follow up period were included in this analysis.ResultsA variety of different functions were examined. A modified version of the critical exponential function was selected for the final model given its robustness across different datasets and its flexibility in assuming a variety of different shapes. Age, site, initial asexual parasite density (logged to the base 10), and an empirical patient category were the co-variates that were found to improve the model.ConclusionsA population nonlinear modeling approach seems promising and produced a flexible function whose estimates were stable across various different datasets. Surprisingly, dihydrofolate reductase and dihydropteroate synthetase mutation prevalence did not enter the model. This is probably related to a lack of power (quintuple mutations n = 12), and informative censoring; treatment failures were withdrawn from the study and given rescue treatment, usually prior to completion of follow up.