Spatial and temporal variations in tuberculosis incidence, Nepal.

Tuberculosis (TB) is an important public health problem in Nepal. The aim of this study was to investigate the spatial and temporal variations in TB incidence in Nepal. Data regarding TB cases were obtained from the Nepal National Tuberculosis Center (NTC) for 2003-2010 and analyzed. Models were developed for TB incidence by gender, year and location using linear regression of log-transformed incidence rates. Apart from a relatively small number of outliers, these models provided a good fit, as indicated by residual plots and the r-squared statistic (0.94). The overall incidence of TB was 1.31 cases per 1,000 population with a male to female incidence rate ratio of 1.83. There were trends of increasing incidence in TB for recent years among both sexes. There were marked variations by location with higher rates occurring in the Terai region and relatively moderate and low rates of TB in the Hill and Mountain regions, respectively. TB incidence was also higher in the capital city Kathmandu and other metropolitan cities. A log-linear regression model can be used as a simple method to model TB incidence rates that vary by location and year. These findings provide information for health authorities to help establish effective prevention programs in specific areas where the disease burden is relatively high.

Estimating liver cancer deaths in Thailand based on verbal autopsy study.

BACKGROUND: Liver cancer mortality is high in Thailand but utility of related vital statistics is limited due to national vital registration (VR) data being under reported for specific causes of deaths. Accurate methodologies and reliable supplementary data are needed to provide worthy national vital statistics. This study aimed to model liver cancer deaths based on verbal autopsy (VA) study in 2005 to provide more accurate estimates of liver cancer deaths than those reported. The results were used to estimate number of liver cancer deaths during 2000-2009. METHODS: A verbal autopsy (VA) was carried out in 2005 based on a sample of 9,644 deaths from nine provinces and it provided reliable information on causes of deaths by gender, age group, location of deaths in or outside hospital, and causes of deaths of the VR database. Logistic regression was used to model liver cancer deaths and other variables. The estimated probabilities from the model were applied to liver cancer deaths in the VR database, 2000-2009. Thus, the more accurately VA-estimated numbers of liver cancer deaths were obtained. RESULTS: The model fits the data quite well with sensitivity 0.64. The confidence intervals from statistical model provide the estimates and their precisions. The VA-estimated numbers of liver cancer deaths were higher than the corresponding VR database with inflation factors 1.56 for males and 1.64 for females. CONCLUSIONS: The statistical methods used in this study can be applied to available mortality data in developing countries where their national vital registration data are of low quality and supplementary reliable data are available.

Treatment outcome of multidrug-resistant Mycobacterium tuberculosis in Nepal.

The purpose of this study was to examine factors associated with treatment outcome of multidrug-resistant tuberculosis (MDR-TB) cases in Nepal. A retrospective analysis of MDR-TB cases by demographic determinants and treatment was conducted. A total of 494 MDR-TB cases were registered from 2005 to 2008, with data obtained from the National Tuberculosis Center. Chi-squared tests were used to assess statistically the association between smear and culture conversion and treatment outcome. Determinants were analyzed with the use of Kaplan-Meier curves and Cox proportional hazards models to generate estimates of the associations with the time to treatment outcome. Sputum conversion status and culture conversion status were positively associated with treatment outcome for MDR-TB. In a multiple Cox proportional hazards regression model, no determinants were found to be associated with time to cure.

Geographical variations in all-cause mortality in Thailand.

In this study, we examined age-specific death rates among men and women from various districts in Thailand using mortality data from 1999 to 2001. A Poisson generalized linear model was used for analysis. To adjust for large variations in resident populations among districts, the 926 districts in Thailand were reduced to 235 "superdistricts" based on a minimum population of 200,000. The Poisson model incorporating additive factors for age-group and superdistrict generally provided a good fit for these data. The fitted mortality rates among the 235 superdistricts were compared with the overall means for each gender (637 per 100,000 for males and 415 per 100,000 for females). Thematic maps were created with three different colors signifying each superdistrict's mortality rate compared to the mean. Northeastern Thailand had higher than average mortality for both males and females. Lower than average death rates were found in southern Thailand with the exception of Phuket and Narathiwat, and in Bangkok, except for females in the superdistrict containing Nong Chok and Lat Krabang Districts. This modeling and mapping approach is a useful preliminary tool enabling public health planners to determine statistically valid geographical variations in mortality and to develop effective interventions.

A statistical method for estimating under-reported incidence rates with application to child diarrhea in Thai provinces bordering Cambodia.

Diarrhea is a major health problem in Thailand, but reported data of disease incidence are known or suspected to be under-reported. This study aimed to develop a statistical model for estimating the annual incidence of hospital diarrhea cases among children under five years. Data regarding diarrhea patients 0-4 years old were collected for the National Notifiable Disease Surveillance (Report 506) about Thai provinces bordering Cambodia during 1999-2004 by the Ministry of Public Health. A log-linear regression model based on the prevailing seasonal-trend pattern was used for diarrhea incidence as a function of quarter, year and district, after imputing rates where under-reporting was evident, using populations obtained from the 2000 population census. The model also takes any spatial correlation between districts into account, using the generalized estimating equation (GEE) method. Diarrhea incidence had seasonal peaks in the first quarter (January to March) and the trend steadily increased from 1999 to 2004. Results from such studies can help health authorities develop prevention policies.

Modeling the incidence of tuberculosis in southern Thailand.

The aim of this study was to examine the trend, seasonal and geographic effects on tuberculosis (TB) incidence in the fourteen southern provinces of Thailand from 1999 to 2004. Data were obtained from the National Notifiable Disease Surveillance Report (506), Ministry of Public Health. The joint effects of gender, age, quarterly season and location on the TB incidence rates were modeled using both negative binomial distribution for the number of cases and log-linear distribution for the incidence rate; then these models were compared. The linear regression models provided a good fit, as indicated by residual plots and the R2 (0.64). The model showed that males and females aged less than 25 years had similar risks for TB in the study area. Both sexes had their risk increased with age but to a much greater extent for men than women, with the highest rate noted in males aged 65 years and over. There was no evidence of a trend in the annual incidence of TB during 1999-2004, but the incidence has a significant season variation with peaks in the first quarter over the six year period. There were also differences in the incidence rate of TB both within and between provinces. The high risk areas were in upper western and lower southern parts of the region. The log-linear regression model could be used as a simple method for modeling TB incidence rates. These findings highlight the importance of selectively monitoring geographic location when studying TB incidence patterns.

A statistical method for forecasting demographic time series counts, with application to HIV/AIDS and other infectious disease mortality in southern Thailand.

This study investigated regional and temporal patterns of death reported from infectious diseases (including HIV/AIDS) in 14 provinces of southern Thailand over the period 1999-2004, using data obtained from the Thailand Bureau of Policy and Strategy, Ministry of Public Health. Causes of deaths were identified using the International Classification of Diseases 10th revision (ICD-10), and mortality incidence rates were then calculated using populations obtained from the 2000 population census. Poisson and negative binomial lagged observation-driven regression models for mortality incidence were fitted to the data separately for HIV/ AIDS and other infectious diseases. Overall, the hospital mortality rates started to increase sharply in 2003 - 2004. The in-hospital mortality for HIV/AIDS showed peaks in urban districts and decreased from north to south with mortality for males approximately double that of females. For other infectious diseases, an upward trend in hospital mortality age 40 and over started in 2003-2004, particularly among persons reported as dying from septicemia, while showing a slightly increasing trend for other infectious diseases. Identifying the real cause of hospital deaths recorded as septicemia would substantially improve hospital mortality data quality.