In clinical and epidemiological researches, continuous predictors are often discretized into categorical variables for classification of patients. When the relationship between a continuous predictor and log relative hazards is U-shaped in survival data, there is a lack of a satisfying solution to find optimal cut-points to discretize the continuous predictor. In this study, we propose a novel approach named optimal equal-HR method to discretize a continuous variable that has a U-shaped relationship with log relative hazards in survival data. The main idea of the optimal equal-HR method is to find two optimal cut-points that have equal log relative hazard values and result in Cox models with minimum AIC value. An R package ‘CutpointsOEHR’ has been developed for easy implementation of the optimal equal-HR method. A Monte Carlo simulation study was carried out to investigate the performance of the optimal equal-HR method. In the simulation process, different censoring proportions, baseline hazard functions and asymmetry levels of U-shaped relationships were chosen. To compare the optimal equal-HR method with other common approaches, the predictive performance of Cox models with variables discretized by different cut-points was assessed. Simulation results showed that in asymmetric U-shape scenarios the optimal equal-HR method had better performance than the median split method, the upper and lower quantiles method, and the minimum p-value method regarding discrimination ability and overall performance of Cox models. The optimal equal-HR method was applied to a real dataset of small cell lung cancer. The real data example demonstrated that the optimal equal-HR method could provide clinical meaningful cut-points and had good predictive performance in Cox models. In general, the optimal equal-HR method is recommended to discretize a continuous predictor with right-censored outcomes if the predictor has an asymmetric U-shaped relationship with log relative hazards based on Cox regression models.

Objective: To probe the spatiotemporal patterns of the incidence of scarlet fever in Beijing, China, from 2005 to 2014. Methods: A spatiotemporal analysis was conducted at the district/county level in the Beijing region based on the reported cases of scarlet fever during the study period. Moran’s autocorrelation coefficient was used to examine the spatial autocorrelation of scarlet fever, whereas the Getis-Ord Gi* statistic was used to determine the hotspot incidence of scarlet fever. Likewise, the space-time scan statistic was used to detect the space-time clusters, including the relative risk of scarlet fever incidence across all settings. Results: A total of 26,860 scarlet fever cases were reported in Beijing during the study period (2005–2014). The average annual incidence of scarlet fever was 14.25 per 100,000 population (range, 6.76 to 32.03 per 100,000). The incidence among males was higher than that among females, and more than two-thirds of scarlet fever cases (83.8%) were among children 3–8 years old. The seasonal incidence peaks occurred from March to July. A higher relative risk area was mainly in the city and urban districts of Beijing. The most likely space-time clusters and secondary clusters were detected to be diversely distributed in every study year. Conclusions: The spatiotemporal patterns of scarlet fever were relatively unsteady in Beijing from 2005 to 2014. The at-risk population was mainly scattered in urban settings and dense districts with high population, indicating a positive relationship between population density and increased risk of scarlet fever exposure. Children under 15 years of age were the most susceptible to scarlet fever.

Evidence shows that multiple factors, such as socio-economic status and access to health care facilities, affect tuberculosis (TB) incidence. However, there is limited literature available with respect to the correlation between socio-economic/health facility factors and tuberculosis incidence. This study aimed to explore the relationship between TB incidence and socio-economic/health service predictors in the study settings. A retrospective spatial regression analysis was carried out based on new sputum smear-positive pulmonary TB cases in Beijing districts. Global Moran’s I analysis was adopted to detect the spatial dependency followed by spatial regression models (spatial lag model, and spatial error model) along with the ordinary least square model were applied to examine the correlation between TB incidence and predictors. A high incidence of TB was seen in densely populated districts in Beijing, e.g., Haidian, Mentougou, and Xicheng. After comparing the R², log-likelihood, and Akaike information criterion (AIC) values among three models, the spatial error model (R² = 0.413; Log Likelihood = −591; AIC = 1199.76) identified the best model fit for the spatial regression model. The study showed that the number of beds in health institutes (p < 0.001) and per capita gross domestic product (GDP) (p = 0.025) had a positive effect on TB incidence, whereas population density (p < 0.001) and migrated population (p < 0.001) had an adverse impact on TB incidence in the study settings. High TB incidence districts were detected in urban and densely populated districts in Beijing. Our findings suggested that socio-economic predictors influence TB incidence. These findings may help to guide TB control programs and promote targeted intervention.

Background: Maternal Mortality is a public health problem in Nepal, which was highest in 1990 among the South Asian countries. Associated factors of maternal mortality are various; among them maternal health services such as antenatal (ANC) and postnatal care (PNC) services are the main. Methods: A multivariate secondary data analysis out based on Nepal Demographic Health Surveys 2001 and 2006. Logistic regression models was performed to compare the utilization of the ANC and PNC services, with background characteristics of women aged between 15 to 49 years old. Results: A total of 8913 reproductive aged groups (15-49) women were taken for analysis and the mean age was 28.59±7.040 years. Logistic regression analysis revealed that health facility delivery (AOR=1.297, 95% CI=1.135-1.481), PNC check-up at health facility (AOD=4.442, 95% CI=2.815-7.011) and PNC service with a skilled health worker (AOD=4.533, 95% CI=2.753-7.465) utilized more in 2006 compared to 2001. This study also found that highly educated women had (AOD, 95% CI=10.823-22.968) more utilized the heath facility during pregnancy and (AOD, 95% CI=2.194-16.950) more likely during a PNC check-up, whereas, educated women were less (AOR=0.043, 95% CI=0.007-0.254) likely consult with a skilled professional. Similarly, antenatal care (ANC) visits (4 or more than four) and ANC visit in the first trimester were increased (95% CI=1.137-1.518) and (AOD=1.041, 95% CI=0.924-1.173) respectively. This study found that educated women, those who were living in urban areas, were more likely to use maternal health services compared to other regions. Conclusion: Increased in utilization of the ANC and PNC services through skilled health workers in a health facility among cohorts of educated women. However, the improvements were not equally distributed across the all regions in the country. This utilization of maternal health services is not sufficient to achieve the MDG goal. Because, health facility delivery is poor and counterpart home delivery is still high in Nepal.