Biomedical subjects

Andy H Lee

Publications and source records attributed to Andy H Lee.

At least 19 recordsLinked to original sources

PubMed2007

A score test for overdispersion in zero-inflated poisson mixed regression model.

Count data with extra zeros are common in many medical applications. The zero-inflated Poisson (ZIP) regression model is useful to analyse such data. For hierarchical or correlated count data where the observations are either clustered or represent repeated outcomes from individual subjects, a class of ZIP mixed regression models may be appropriate. However, the ZIP parameter estimates can be severely biased if the non-zero counts are overdispersed in relation to the Poisson distribution. In this paper, a score test is proposed for testing the ZIP mixed regression model against the zero-inflated negative binomial alternative. Sampling distribution and power of the test statistic are evaluated by simulation studies. The results show that the test statistic performs satisfactorily under a wide range of conditions. The test procedure is applied to pancreas disorder length of stay that comprised mainly same-day separations and simultaneous prolonged hospitalizations.

Adult↗

PubMed2006

Multilevel logistic regression modelling with correlated random effects: application to the Smoking Cessation for Youth study.

A multilevel logistic regression model is presented for the analysis of clustered and repeated binary response data. At the subject level, serial dependence is expected between repeated measures recorded on the same individual. At the cluster level, correlations of observations within the same subgroup are present due to the inherent hierarchical setting. Two random components are therefore incorporated explicitly within the linear predictor to account for the simultaneous heterogeneity and autoregressive structure. Application to analyse a set of longitudinal data from an adolescent smoking cessation intervention that motivated this study is illustrated.

Adolescent↗

PubMed2006

A decade of change in breastfeeding in China's far north-west.

BACKGROUND: There have been considerable changes in breastfeeding practices in China over the past forty years. However China is a very large country, and breastfeeding rates in different parts of China vary considerably. The objective of this paper is to identify and compare breastfeeding types and rates between 1994-1996 and 2003-2004 in Shihezi, Xinjiang Uygur Autonomous Region, PR China. METHODS: In 1994-1996, a study of breastfeeding (n = 2197) was undertaken in Shihezi, Xinjiang, PR China. A decade later in 2003-2004, a longitudinal study (n = 545) of infant feeding practices was undertaken in the same area. RESULTS: The 'any breastfeeding' rates at 1, 4 and 6 months were 94%, 82% and 78% respectively in the early 1990s. A decade later, breastfeeding at 1 month was lower, but rates at 4 and 6 months remained the same. In 2004 the 'full breastfeeding' rate at one month was significantly higher (57%) than a decade earlier (38%), but after 3 months there was a rapid decline. This reflected a shift in the way complementary foods are introduced: the initial introduction was later, but by a higher proportion of mothers. CONCLUSION: The rate of breastfeeding at one month is significantly lower in 2003-2004 when compared to 1994-1996. The 'full breastfeeding' rates were initially higher, but after 3 months were then lower. The Chinese national breastfeeding targets were not reached in either period of the study. These studies show the need to further promote full or exclusive breastfeeding and further longitudinal studies are necessary to provide the detailed knowledge about risk factors required for health promotion programs.

Journal Article↗

PubMed2006

Retrospective analysis of catheter-related infections in a hemodialysis unit.

OBJECTIVE: To compare the incidence rates of catheter-related bloodstream infection associated with different vascular access methods in patients receiving hemodialysis. SETTING: Tertiary care public hospital in Western Australia. DESIGN: Retrospective analysis of surveillance data collected by the hospital's infection control department. METHODS: The number of confirmed bloodstream infections for each type of vascular access was identified for the period from July 2002 through June 2003. The corresponding number of patient-days was determined to calculate the infection incidence rates. The serially correlated data were then analyzed using Poisson generalized estimating equations. RESULTS: A total of 32 confirmed bloodstream infections were identified. Infection rates, in number of infections per 1,000 patient-days, were as follows: 0.4 for native arteriovenous fistulae; 2.86 for synthetic arteriovenous grafts; 4.02 for permanent, tunneled, cuffed central venous catheters; and 20.2 for temporary, nontunneled, noncuffed central venous catheters. Compared with permanent catheters, the monthly infection rate associated with the temporary catheters was significantly higher (incident rate ratio [IRR], 5.025 [95% confidence interval {CI}, 1.532-16.484]; P=.008) and that of arteriovenous fistulae was significantly lower (IRR, 0.099 [95% CI, 0.030-0.324]; P=.001). The monthly infection rate for arteriovenous grafts was not significantly different from that for permanent central venous catheters (IRR, 0.702 [95% CI, 0.246-2.008]; P=.510). CONCLUSIONS: A hierarchy of infection risk associated with vascular access type is evident. Native arteriovenous fistulae should be recommended for all patients receiving chronic hemodialysis, to minimize infection.

Bacterial Infections↗

PubMed2006

A score test for zero-inflation in correlated count data.

To account for the preponderance of zero counts and simultaneous correlation of observations, a class of zero-inflated Poisson mixed regression models is applicable for accommodating the within-cluster dependence. In this paper, a score test for zero-inflation is developed for assessing correlated count data with excess zeros. The sampling distribution and the power of the test statistic are evaluated by simulation studies. The results show that the test statistic performs satisfactorily under a wide range of conditions. The test procedure is further illustrated using a data set on recurrent urinary tract infections.

Aged↗

PubMed2006

Fragility and crash over-representation among older drivers in Western Australia.

This study assessed age-related and gender differences in the relative contribution of fragility and crash over-representation to serious injuries per crash-involved driver in Western Australia. Police-reported crashes for the period 1998-2003 were extracted from the Western Australian Road Injury Database. For each passenger vehicle driver age and gender group, serious injuries per crash-involved driver and driver involvements in crashes per 100 million vehicle-kilometre travelled (VKT) were calculated as the respective measure of fragility and crash over-representation. Results from the decomposition method of analysis showed that older drivers over the age of 70 sustained serious injury rates more than twice as high as those of the 30-59-year-old drivers. Fragility increased with age, contributing between 47% and 95% for drivers above 65 years, but crash over-representation was the dominant factor for male drivers above 80 years. In contrast, fragility contributed little to the excess injury risk of younger drivers under the age of 30. The importance of fragility as a contributing factor to the inflated serious injury risk per vehicle-kilometre travelled for older drivers suggested that road safety initiatives should be directed towards the protection of vehicle occupants as well as screening for their driving ability.

Accidents, Traffic↗

PubMed2006

Validity of a food-frequency questionnaire for elderly men in southeast China.

OBJECTIVE: To investigate the internal validity of a food-frequency questionnaire (FFQ) developed for a case-control study of prostate cancer in southeast China. DESIGN: A comprehensive questionnaire comprising a quantitative FFQ and a short food habit questionnaire (SFHQ) was developed and modified from previous cancer and nutritional studies. The Goldberg formula (ratio of energy intake (EI) to basal metabolic rate (BMR), EI/BMR) was used to assess the validity of the FFQ by making comparisons with physical activity levels. Physical activity levels were measured by the estimated total metabolic equivalents (MET) and the ratio of energy expenditure (EE) to BMR (EE/BMR). Correlation analyses were undertaken to compare the SFHQ variables with those of the quantitative FFQ. SETTING: Hangzhou, Zhejiang Province, China. SUBJECTS: A total of 404 men over 45 years old with or without prostate cancer were recruited from eight hospitals. RESULTS: The partial correlation coefficients, controlling for age and family history of prostate cancer, were moderate to high (P<0.05) for preserved foods intake, fat consumption and tea drinking variables between the SFHQ and the quantitative FFQ. The average EI/BMR was 1.72, with 76% of subjects exceeding the Goldberg cut-off value of 1.35. Apart from weight, BMI, EE/BMR and MET, there were no significant differences in characteristics between low (<1.35) and normal EI/BMR groups. CONCLUSIONS: The FFQ is demonstrated to be a valid instrument to measure energy and food intake for elderly men in southeast China.

Basal Metabolism↗

PubMed2006

Estimating population-based injury prevention practices.

This study piloted a survey suitable to generate statewide prevalence estimates for injury prevention practices and/or behaviours. A number of prevalence estimates of injury prevention practices and/or behaviours were generated in the pilot. For example, the prevalence of a household smoke alarm was high (732/1000 households) and yet only 44% of households had tested the functionality of the alarm in the 12 months preceding the survey. Findings of this kind point to an increasing need for primary care practitioners to play a leading role in targeting preventive strategies for injury. In addition, the survey methods and approach highlight the value of such information to health departments and non-government agencies in terms of targeting priority injury risk behaviours and the evaluation of injury-related intervention programmes. The survey has been designed so the methods can be applied nationally and internationally.

Adult↗

PubMed2006

Multi-level zero-inflated poisson regression modelling of correlated count data with excess zeros.

Count data with excess zeros relative to a Poisson distribution are common in many biomedical applications. A popular approach to the analysis of such data is to use a zero-inflated Poisson (ZIP) regression model. Often, because of the hierarchical study design or the data collection procedure, zero-inflation and lack of independence may occur simultaneously, which render the standard ZIP model inadequate. To account for the preponderance of zero counts and the inherent correlation of observations, a class of multi-level ZIP regression model with random effects is presented. Model fitting is facilitated using an expectation-maximization algorithm, whereas variance components are estimated via residual maximum likelihood estimating equations. A score test for zero-inflation is also presented. The multi-level ZIP model is then generalized to cope with a more complex correlation structure. Application to the analysis of correlated count data from a longitudinal infant feeding study illustrates the usefulness of the approach.

Adult↗

PubMed2006

Protective effects of green tea against prostate cancer.

Prostate cancer has the third highest incidence of all cancers in men worldwide with incidence and mortality being particularly high in affluent, developed countries. Tea, especially green tea, has demonstrated promise in the prevention of several cancers. Green tea contains several components including catechins, a category of polyphenols that have chemopreventive properties. Although evidence from epidemiological studies is not comprehensive, it is strengthened by animal and in vitro evidence suggesting that consumption of tea is associated with decreased risk or progression of prostate cancer. Emerging evidence and potential biological mechanisms for the role of green tea in prostate cancer prevention are presented in this review.

Animals↗

PubMed2006

Under-nutrition affects time to recurrence of gastroenteritis among Aboriginal and non-Aboriginal children.

This study investigated whether under-nutrition affected time to hospitalization for recurrence of gastroenteritis in Australian children. Linked hospitalization records of all infants, born in 1995 and 1996 in Western Australia, who were admitted for gastroenteritis during their first year of life (n=1001), were retrieved. A survival frailty model was used for determining the factors influencing the recurrent times over the subsequent seven years. Aboriginality and under-nutrition were significantly associated with an increased risk of recurrence (hazard ratios of 2.59 and 1.28). Hospitalizations due to gastroenteritis were common among Aboriginal children who had a higher mean re-admission rate and much shorter intervals between re-admissions than other patients. The proportion of patients with recurrence was also significantly higher for Aboriginals (38.5%) than for other patients (14.2%). Gastroenteritis remains a serious problem in Aboriginal children. This presents a complex challenge to be addressed with public-health principles, political determination and commitment, and adequate resources.

Cohort Studies↗

PubMed2005

Influence diagnostics for two-component Poisson mixture regression models: applications in public health.

In many medical and health applications, Poisson mixture regression models are commonly used to analyse heterogeneous count data. Motivated by two data sets drawn from public health studies, influence diagnostics are proposed for assessing the sensitivity of the fitted two-component Poisson mixture regression models. Under various perturbations of the observed data or model assumptions, influence assessments based on the local influence approach are developed for detecting clusters and/or individual observations that impact on the estimation of model parameters. Results from studies on recurrent urinary tract infections and maternity length of stay illustrate the usefulness of the influence diagnostics.

Cluster Analysis↗

PubMed2005

Estimating crashes involving heavy vehicles in Western Australia, 1999-2000: a capture-recapture method.

A two-sample exploratory study of police and hospital records was undertaken to estimate the number of fatalities and serious injuries for heavy vehicle drivers involved in a crash in Western Australia. The capture-recapture method was used to assess differences and similarities in characteristics of heavy vehicle drivers from both sources. Each heavy vehicle driver involved in a crash from the police report was matched against the heavy vehicle driver's hospitalisation record from the Hospital Morbidity Data System, with surname, initials, date of birth, gender, date of crash, road user type and vehicle type as matching fields. The estimated number of fatalities and serious injuries to heavy vehicle drivers from 1st July 1999 to 31st December 2000 was 5 and 59, respectively, which was 25 and 31% higher based on the capture-recapture methodology than the aggregated (non-overlapping) total officially reported to the police and hospitals. No significant age difference (p>0.05) was found for drivers involved in a heavy vehicle crash between the two sources (37 years versus 40 year of age). However, female heavy vehicle drivers were over-represented in the hospital records (11%) compared to the police records (1%). The capture-recapture approach is useful for evaluating the completeness of data sources and identifying biases within datasets. The underestimation of heavy vehicle drivers seriously injured and killed has important implications for heavy vehicle safety management and resource allocation in Western Australia.

Accidents, Traffic↗

PubMed2005

An incremental EM-based learning approach for on-line prediction of hospital resource utilization.

OBJECTIVE: Inpatient length of stay (LOS) is an important measure of hospital activity, health care resource consumption, and patient acuity. This research work aims at developing an incremental expectation maximization (EM) based learning approach on mixture of experts (ME) system for on-line prediction of LOS. The use of a batch-mode learning process in most existing artificial neural networks to predict LOS is unrealistic, as the data become available over time and their pattern change dynamically. In contrast, an on-line process is capable of providing an output whenever a new datum becomes available. This on-the-spot information is therefore more useful and practical for making decisions, especially when one deals with a tremendous amount of data. METHODS AND MATERIAL: The proposed approach is illustrated using a real example of gastroenteritis LOS data. The data set was extracted from a retrospective cohort study on all infants born in 1995-1997 and their subsequent admissions for gastroenteritis. The total number of admissions in this data set was n = 692. Linked hospitalization records of the cohort were retrieved retrospectively to derive the outcome measure, patient demographics, and associated co-morbidities information. A comparative study of the incremental learning and the batch-mode learning algorithms is considered. The performances of the learning algorithms are compared based on the mean absolute difference (MAD) between the predictions and the actual LOS, and the proportion of predictions with MAD < or = 1 day (Prop(MAD < or = 1)). The significance of the comparison is assessed through a regression analysis. RESULTS: The incremental learning algorithm provides better on-line prediction of LOS when the system has gained sufficient training from more examples (MAD = 1.77 days and Prop(MAD < or = 1) = 54.3%), compared to that using the batch-mode learning. The regression analysis indicates a significant decrease of MAD (p-value = 0.063) and a significant (p-value = 0.044) increase of Prop(MAD < or = 1) with the incremental learning algorithm. CONCLUSIONS: The incremental learning feature and the self-adaptive model-selection ability of the ME network enhance its effective adaptation to non-stationary LOS data. It is demonstrated that the incremental learning algorithm outperforms the batch-mode algorithm in the on-line prediction of LOS.

Algorithms↗

PubMed2005

Evaluating the effectiveness of a participatory ergonomics approach in reducing the risk and severity of injuries from manual handling.

Manual handling is the greatest contributor to non-fatal injury and disease in the workplace, commonly accounting for one-third of national injury counts. Interventional strategies that have focused on selecting or modifying the worker have been ineffective in reducing injury risk. In recent times, participatory ergonomics has been widely adopted as a process to reduce the risk of injury from manual handling but it is not well validated as an intervention. This study evaluated the effectiveness of a participatory ergonomics risk assessment approach in reducing the rate and severity of injuries from manual and non-manual handling sustained by a cohort of 137 cleaners within a hospital setting. The date of injury and the workers' compensation claim cost and hours lost from work were obtained for each injury incurred during the 4-year pre-intervention and 3-year intervention period. The age, gender and hours worked were ascertained for every cleaner whether injured or not. Using generalized linear mixed modelling analysis, reductions of rate of injury by two-thirds, workers' compensation claim costs by 62% and hours lost by 35% for manual handling injuries were found to be associated with the intervention period. Although the cleaners experienced a significant intervention period reduction in non-manual handling injury rate, the corresponding changes in severity of injury were not significant. The success of the intervention supports the adoption of a participatory ergonomics approach in reducing the rate and consequence of injuries in the workplace.

Adult↗

PubMed2005

Do dietary lycopene and other carotenoids protect against prostate cancer?

To determine whether dietary intake of lycopene and other carotenoids has an etiological association with prostate cancer, a case-control study was conducted in Hangzhou, southeast China during 2001-2002. The cases were 130 incident patients with histologically confirmed adenocarcinoma of the prostate. The controls were 274 hospital inpatients without prostate cancer or any other malignant diseases. Information on usual food consumption, including vegetables and fruits, was collected by face-to-face interviews using a structured food frequency questionnaire. The risks of prostate cancer for the intake of carotenoids and selected vegetables and fruits rich in carotenoids were assessed using multivariate logistic regression, adjusting for age, locality, education, income, body mass index, marital status, number of children, family history of prostate cancer, tea drinking, total fat and caloric intake. The prostate cancer risk declined with increasing consumption of lycopene, alpha-carotene, beta-carotene, beta-cryptoxanthin, lutein and zeaxanthin. Intake of tomatoes, pumpkin, spinach, watermelon and citrus fruits were also inversely associated with the prostate cancer risk. The adjusted odds ratios for the highest versus the lowest quartiles of intake were 0.18 (95% CI: 0.08-0.41) for lycopene, 0.43 (95% CI: 0.21-0.85) for alpha-carotene, 0.34 (95% CI: 0.17-0.69) for beta-carotene, 0.15 (95% CI: 0.06-0.34) for beta-cryptoxanthin and 0.02 (95% CI: 0.01-0.10) for lutein and zeaxanthin. The corresponding dose-response relationships were also significant, suggesting that vegetables and fruits rich in lycopene and other carotenoids may be protective against prostate cancer.

Carotenoids↗

PubMed2005

Sensitivity of test for overdispersion in Poisson regression.

Overdispersion or extra-Poisson variation is very common for count data. This phenomenon arises when the variability of the counts greatly exceeds the mean under the Poisson assumption, resulting in substantial bias for the parameter estimates. To detect whether count data are overdispersed in the Poisson regression setting, various tests have been proposed and among them, the score tests derived by Dean (1992) are popular and easy to implement. However, such tests can be sensitive to anomalous or extreme observations. In this paper, diagnostic measures are proposed for assessing the sensitivity of Dean's score test for overdispersion in Poisson regression. Applications to the well-known fabric faults and Ames salmonella assay data sets illustrate the usefulness of the diagnostics in analyzing overdispersed count data.

Bias↗

PubMed2005

Moderate physical activity and prostate cancer risk: a case-control study in China.

A case-control study was conducted to investigate the association between physical activity and prostate cancer risk in China. A total of 130 recent histopathologically confirmed cancer cases and 274 controls were interviewed. Information collected included type, intensity and duration of various physical activities. The adjusted odds ratios of prostate cancer risk for physical activity variables were obtained from multivariate logistic regression models. Results showed that moderate physical activity was inversely related to the prostate cancer risk, with adjusted odds ratio being 0.20 (95% confidence interval 0.07-0.62) for the upper vs. the lower quartiles of weekly metabolic equivalent task-hours. The dose-response relationship was also significant (p = 0.015). The findings suggest that moderate physical activity may be protective against prostate cancer for Chinese men.

Aged↗