Biostatistician with a background in ecology and a history of collaboration with psychology, public health, ecology, environmental health, pharmacology, and seismology. Current research focuses on study design and statistical methods for Ecological Momentary Assessment, a method in the behavioral sciences for collecting data in real time and in the environments of study participants.
Introduction: Ecological momentary assessment (EMA) assesses individuals’ behaviors and moods as ... more Introduction: Ecological momentary assessment (EMA) assesses individuals’ behaviors and moods as they occur in their natural environment. We examined self-reported triggers of temptations/urges to eat and compared percent of times each trigger was reported as the top trigger in the 1st 6 mos. vs. the 2nd 6 mos. of a standard behavioral weight loss intervention. Hypothesis: The distribution of self-reported triggers of urges/temptations changes over time. Methods: We provided participants a smartphone app programmed to permit them to self-initiate completing EMA and report temptations/urges in real time. Questions included: What were the triggers for the temptation/urge? What was the top trigger? Only data from subjects reporting ≥ one urge/temptation were included in the analysis. For each trigger, generalized estimating equations were used to fit logistic regression models to test the hypothesis that proportion of urges/temptations including that trigger did not change between the 1st and 2 nd 6 mos. of EMA surveillance. Results: The sample ( N = 139) was 90.6% female, 82.7% White, 50.9±10.10 years of age with a BMI of 34.2±4.6 kg/m2. There were 3175 reported episodes of temptation/urges. During both 6-mo.periods, sight/smell of food was the most frequently reported trigger occurring in 34.8% of temptations/urges in the first 6 mos., increasing to 35.52% in the second 6-mos. This was followed by stress, hunger/thirst, and habit during both 6-mo. periods. There were no significant differences in the frequency or type of triggers over time, except for “out of nowhere”, which occurred less often in the second period (6.98% vs. 5.42%, p = .02). The triggers are plotted in order of most to least frequent for both periods in the figure below. Conclusions: Fewer temptations/urges were reported in the second 6 mos., which likely is attributed to a decrease in compliance by participants to EMA protocols. The sight/smell of food remained the most frequently reported trigger, which is a concern in an environment that is very food centric.
Background: Ecological momentary assessment (EMA) assesses individuals’ current experiences, loca... more Background: Ecological momentary assessment (EMA) assesses individuals’ current experiences, location and social context in which behaviors occur. During a 12-mo behavioral weight loss intervention and using an EMA app, we asked participants to report any instances of temptations or a lapse from their weight management diet. Objective: Estimate the rate of temptations per unit time and the probability that a lapse will follow based on the location (restaurant, home, work) and the social context ( with others who are eating, completely alone ). Methods: Using a smartphone, participants initiated a report when they experienced a temptation and reported if this led to a lapse. Their location and the social context were recorded during the 5,211 self-reported temptations as well as during 106,960 random assessments over 12 mos. Analyses: A frailty model was used to estimate the rate at which temptations occurred per unit of time and a logistic regression model was used to estimate the probability that a temptation led to a lapse both as a function of location and social context. Results: The sample (N = 150) was predominantly female (90.7%) and White (80.7%), 56.4% were married with 16.4 (2.8) years of education and a mean BMI of 34.0 kg/m 2 (4.6). Temptations occurred most often in a restaurant and the person was with others who were eating or in sight of others who were eating (Fig 1). While temptations occurred infrequently in another person’s home, there was nearly a 65% probability that a lapse would occur there. When with others who are eating , the temptation rate and probability of a lapse were high. Conclusions: Temptations to eat in a manner that is inconsistent with a weight loss plan occur more frequently in certain locations, e.g., restaurants; however, the probability a temptation leads to a lapse is high in a variety of locations and social contexts, e.g., with others who are eating . Interventions need to focus on empowering individuals trying to manage their weight to resist temptations in these at risk locations and social contexts.
Introduction: Research has shown self-efficacy (SE) to be positively associated with adherence to... more Introduction: Research has shown self-efficacy (SE) to be positively associated with adherence to calorie goal when assessment is based on participants’ recall; however, when assessed in real-time, their association is unclear. Our aim was to examine temporal associations between SE for adhering to the daily recommended lifestyle plan and adherence to calorie goal collected in real-time over 12 months. Methods: Ecological momentary assessment (EMA) was used to assess SE for adhering to the daily recommended lifestyle plan in real-time during a behavioral weight loss study. SE was assessed by one question at the beginning of each day in the person’s natural setting “How confident are you that you will be able to stick to your healthy lifestyle plan today?” SE scores ranged from 1 to 10, with 1 reflecting lowest SE. Daily dietary intake was recorded using the Lose It! app. We calculated adherence to the calorie goal by dividing the total number of calories consumed each day by the daily calorie goal, then multiplying by 100 to express the value as a percentage. Adherence for each day was defined as 85%–115%. The mean weekly SE scores and mean weekly number of days adherent to dietary goals were calculated. Random coefficient modeling was used for analysis. Results: The sample (N=150) was 90.7% female, 80.7% white, with a mean age (±SD) of 51.1±10.2 years and a mean BMI of 34.0±4.6 kg/m 2 . Mean weekly SE score (b=-0.02, p
Consider a spatial point pattern realized from an inhomogeneous Poisson process on a bounded Bore... more Consider a spatial point pattern realized from an inhomogeneous Poisson process on a bounded Borel set , with intensity function λ (s; θ), where . In this article, we show that the maximum likelihood estimator and the Bayes estimator are consistent, asymptotically normal, and asymptotically efficient as the sample region . These results extend asymptotic results of Kutoyants (1984), proved for an inhomogeneous Poisson process on [0, T] , where T →∞. They also formalize (and extend to the multiparameter case) results announced by Krickeberg (1982), for the spatial domain . Furthermore, a Cramér–Rao lower bound is found for any estimator of θ. The asymptotic properties of and are considered for modulated (Cox (1972)), and linear Poisson processes.
Rathbun et al. (2007) and Waagepetersen (2008) propose estimating functions for parameters of Poi... more Rathbun et al. (2007) and Waagepetersen (2008) propose estimating functions for parameters of Poisson point process intensity that may be applied when space- and/or time-varying covariates are sampled from a probability-based sampling design. This paper demonstrates that Waageptersen's estimating function is optimal in a class of weighted estimating functions. Copyright 2013, Oxford University Press.
To assess the potential impact of oral rabies vaccination (ORV) on the occurrence of raccoon rabi... more To assess the potential impact of oral rabies vaccination (ORV) on the occurrence of raccoon rabies in the mid-Atlantic region, temporal and seasonal trends of raccoon rabies cases reported in West Virginia from 1990 to 2007 were identified with both descriptive statistical analysis and exploratory time series analysis. Raccoon rabies cases in the non-ORV region maintain an enzootic pattern and increase over time; a bimodal seasonal pattern is observed with a large peak in April and a smaller peak in August. The results of the model indicate that the effect of the ORV intervention to control raccoon rabies was statistically significant. ORV should be attempted in other enzootic raccoon rabies areas.
Advancing the understanding of spatiotemporal aspects of air pollution in the urban environment i... more Advancing the understanding of spatiotemporal aspects of air pollution in the urban environment is an area where improved methods can be of great benefit to exposure assessment and policy support. This paper explores the potential of a technique known as kriging with external drift (KED) to provide high resolution maps of fine particulate matter for a downtown region of Cusco,
Journal of the American Statistical Association, Dec 1, 1994
Page 1. A Space-Time Survival Point Process for a Longleaf Pine Forest in Southern Georgia Stephe... more Page 1. A Space-Time Survival Point Process for a Longleaf Pine Forest in Southern Georgia Stephen L. RATHBUN and Noel CRESSIE* A marked spatial point pattern of trees and their diameters is the result of a dynamic biological process that takes place over time as ...
Spatiotemporal patchiness in the soil environment is thought to be crucial for the maintenance of... more Spatiotemporal patchiness in the soil environment is thought to be crucial for the maintenance of soil biodiversity. It provides diverse microhabitats (allowing resource partitioning), and presents these in a complex mosaic, such that competitors may be spatially and temporally separated (promoting patch dynamics). The objective of our study was to assess the importance of patch dynamics for the maintenance of landscape-level nematode diversity. The spatiotemporal aggregation patterns in populations of five species of the bacterivorous genus Chronogaster were analyzed for a 0.7-ha riparian wetland, during four seasons, using geostatistical methods (spatiotemporal variograms, cross-correlograms, and log-normal kriging). We found that for the three most abundant species, aggregation was significant over distances of 13-39 m, with intraspecific aggregation stronger than interspecific aggregation. In addition, species patterns appeared to be temporally dynamic: local population declines alternated with local blooms at different sites and seasons. One species was consistently present at the same few sites, but the other species showed gradual temporal decreases and increases at different locations in the wetland. Combining nematode patterns with information on soil resource distributions and Chronogaster ecology, indicated that both soil resource patchiness and repeated, but largely unpredictable, disturbances caused by local soil moisture fluctuations could promote landscape-level Chronogaster diversity by providing ample 'probability refuges.' Local species populations, reduced by competition, predation and drought, could recolonize patches either through the awakening of local dormant 'seed populations,' or through limited passive dispersal by wind, water, and phoresis. We suggest that the spatiotemporal Chronogaster data provide preliminary evidence for the importance of patch dynamics in maintaining soil nematode diversity.
Introduction: Ecological momentary assessment (EMA) assesses individuals’ behaviors and moods as ... more Introduction: Ecological momentary assessment (EMA) assesses individuals’ behaviors and moods as they occur in their natural environment. We examined self-reported triggers of temptations/urges to eat and compared percent of times each trigger was reported as the top trigger in the 1st 6 mos. vs. the 2nd 6 mos. of a standard behavioral weight loss intervention. Hypothesis: The distribution of self-reported triggers of urges/temptations changes over time. Methods: We provided participants a smartphone app programmed to permit them to self-initiate completing EMA and report temptations/urges in real time. Questions included: What were the triggers for the temptation/urge? What was the top trigger? Only data from subjects reporting ≥ one urge/temptation were included in the analysis. For each trigger, generalized estimating equations were used to fit logistic regression models to test the hypothesis that proportion of urges/temptations including that trigger did not change between the 1st and 2 nd 6 mos. of EMA surveillance. Results: The sample ( N = 139) was 90.6% female, 82.7% White, 50.9±10.10 years of age with a BMI of 34.2±4.6 kg/m2. There were 3175 reported episodes of temptation/urges. During both 6-mo.periods, sight/smell of food was the most frequently reported trigger occurring in 34.8% of temptations/urges in the first 6 mos., increasing to 35.52% in the second 6-mos. This was followed by stress, hunger/thirst, and habit during both 6-mo. periods. There were no significant differences in the frequency or type of triggers over time, except for “out of nowhere”, which occurred less often in the second period (6.98% vs. 5.42%, p = .02). The triggers are plotted in order of most to least frequent for both periods in the figure below. Conclusions: Fewer temptations/urges were reported in the second 6 mos., which likely is attributed to a decrease in compliance by participants to EMA protocols. The sight/smell of food remained the most frequently reported trigger, which is a concern in an environment that is very food centric.
Background: Ecological momentary assessment (EMA) assesses individuals’ current experiences, loca... more Background: Ecological momentary assessment (EMA) assesses individuals’ current experiences, location and social context in which behaviors occur. During a 12-mo behavioral weight loss intervention and using an EMA app, we asked participants to report any instances of temptations or a lapse from their weight management diet. Objective: Estimate the rate of temptations per unit time and the probability that a lapse will follow based on the location (restaurant, home, work) and the social context ( with others who are eating, completely alone ). Methods: Using a smartphone, participants initiated a report when they experienced a temptation and reported if this led to a lapse. Their location and the social context were recorded during the 5,211 self-reported temptations as well as during 106,960 random assessments over 12 mos. Analyses: A frailty model was used to estimate the rate at which temptations occurred per unit of time and a logistic regression model was used to estimate the probability that a temptation led to a lapse both as a function of location and social context. Results: The sample (N = 150) was predominantly female (90.7%) and White (80.7%), 56.4% were married with 16.4 (2.8) years of education and a mean BMI of 34.0 kg/m 2 (4.6). Temptations occurred most often in a restaurant and the person was with others who were eating or in sight of others who were eating (Fig 1). While temptations occurred infrequently in another person’s home, there was nearly a 65% probability that a lapse would occur there. When with others who are eating , the temptation rate and probability of a lapse were high. Conclusions: Temptations to eat in a manner that is inconsistent with a weight loss plan occur more frequently in certain locations, e.g., restaurants; however, the probability a temptation leads to a lapse is high in a variety of locations and social contexts, e.g., with others who are eating . Interventions need to focus on empowering individuals trying to manage their weight to resist temptations in these at risk locations and social contexts.
Introduction: Research has shown self-efficacy (SE) to be positively associated with adherence to... more Introduction: Research has shown self-efficacy (SE) to be positively associated with adherence to calorie goal when assessment is based on participants’ recall; however, when assessed in real-time, their association is unclear. Our aim was to examine temporal associations between SE for adhering to the daily recommended lifestyle plan and adherence to calorie goal collected in real-time over 12 months. Methods: Ecological momentary assessment (EMA) was used to assess SE for adhering to the daily recommended lifestyle plan in real-time during a behavioral weight loss study. SE was assessed by one question at the beginning of each day in the person’s natural setting “How confident are you that you will be able to stick to your healthy lifestyle plan today?” SE scores ranged from 1 to 10, with 1 reflecting lowest SE. Daily dietary intake was recorded using the Lose It! app. We calculated adherence to the calorie goal by dividing the total number of calories consumed each day by the daily calorie goal, then multiplying by 100 to express the value as a percentage. Adherence for each day was defined as 85%–115%. The mean weekly SE scores and mean weekly number of days adherent to dietary goals were calculated. Random coefficient modeling was used for analysis. Results: The sample (N=150) was 90.7% female, 80.7% white, with a mean age (±SD) of 51.1±10.2 years and a mean BMI of 34.0±4.6 kg/m 2 . Mean weekly SE score (b=-0.02, p
Consider a spatial point pattern realized from an inhomogeneous Poisson process on a bounded Bore... more Consider a spatial point pattern realized from an inhomogeneous Poisson process on a bounded Borel set , with intensity function λ (s; θ), where . In this article, we show that the maximum likelihood estimator and the Bayes estimator are consistent, asymptotically normal, and asymptotically efficient as the sample region . These results extend asymptotic results of Kutoyants (1984), proved for an inhomogeneous Poisson process on [0, T] , where T →∞. They also formalize (and extend to the multiparameter case) results announced by Krickeberg (1982), for the spatial domain . Furthermore, a Cramér–Rao lower bound is found for any estimator of θ. The asymptotic properties of and are considered for modulated (Cox (1972)), and linear Poisson processes.
Rathbun et al. (2007) and Waagepetersen (2008) propose estimating functions for parameters of Poi... more Rathbun et al. (2007) and Waagepetersen (2008) propose estimating functions for parameters of Poisson point process intensity that may be applied when space- and/or time-varying covariates are sampled from a probability-based sampling design. This paper demonstrates that Waageptersen's estimating function is optimal in a class of weighted estimating functions. Copyright 2013, Oxford University Press.
To assess the potential impact of oral rabies vaccination (ORV) on the occurrence of raccoon rabi... more To assess the potential impact of oral rabies vaccination (ORV) on the occurrence of raccoon rabies in the mid-Atlantic region, temporal and seasonal trends of raccoon rabies cases reported in West Virginia from 1990 to 2007 were identified with both descriptive statistical analysis and exploratory time series analysis. Raccoon rabies cases in the non-ORV region maintain an enzootic pattern and increase over time; a bimodal seasonal pattern is observed with a large peak in April and a smaller peak in August. The results of the model indicate that the effect of the ORV intervention to control raccoon rabies was statistically significant. ORV should be attempted in other enzootic raccoon rabies areas.
Advancing the understanding of spatiotemporal aspects of air pollution in the urban environment i... more Advancing the understanding of spatiotemporal aspects of air pollution in the urban environment is an area where improved methods can be of great benefit to exposure assessment and policy support. This paper explores the potential of a technique known as kriging with external drift (KED) to provide high resolution maps of fine particulate matter for a downtown region of Cusco,
Journal of the American Statistical Association, Dec 1, 1994
Page 1. A Space-Time Survival Point Process for a Longleaf Pine Forest in Southern Georgia Stephe... more Page 1. A Space-Time Survival Point Process for a Longleaf Pine Forest in Southern Georgia Stephen L. RATHBUN and Noel CRESSIE* A marked spatial point pattern of trees and their diameters is the result of a dynamic biological process that takes place over time as ...
Spatiotemporal patchiness in the soil environment is thought to be crucial for the maintenance of... more Spatiotemporal patchiness in the soil environment is thought to be crucial for the maintenance of soil biodiversity. It provides diverse microhabitats (allowing resource partitioning), and presents these in a complex mosaic, such that competitors may be spatially and temporally separated (promoting patch dynamics). The objective of our study was to assess the importance of patch dynamics for the maintenance of landscape-level nematode diversity. The spatiotemporal aggregation patterns in populations of five species of the bacterivorous genus Chronogaster were analyzed for a 0.7-ha riparian wetland, during four seasons, using geostatistical methods (spatiotemporal variograms, cross-correlograms, and log-normal kriging). We found that for the three most abundant species, aggregation was significant over distances of 13-39 m, with intraspecific aggregation stronger than interspecific aggregation. In addition, species patterns appeared to be temporally dynamic: local population declines alternated with local blooms at different sites and seasons. One species was consistently present at the same few sites, but the other species showed gradual temporal decreases and increases at different locations in the wetland. Combining nematode patterns with information on soil resource distributions and Chronogaster ecology, indicated that both soil resource patchiness and repeated, but largely unpredictable, disturbances caused by local soil moisture fluctuations could promote landscape-level Chronogaster diversity by providing ample 'probability refuges.' Local species populations, reduced by competition, predation and drought, could recolonize patches either through the awakening of local dormant 'seed populations,' or through limited passive dispersal by wind, water, and phoresis. We suggest that the spatiotemporal Chronogaster data provide preliminary evidence for the importance of patch dynamics in maintaining soil nematode diversity.
Uploads
Papers by Stephen Rathbun