Change point analysis of normal distribution based on the likelihood ratio principle and its application in the number of outpatients analysis / 中华地方病学杂志

Objective:The authors introduced the change point analysis of normal distribution based on the likelihood ratio principle, analyzed the number of outpatients in a hospital of Luohu District, Shenzhen, to provide scientific basis for rational allocation of health resources.Methods:The authors collected totally 636 number of outpatients' data from 8: 00 to 12: 00 a.m. at 48 time windows at 5 minutes intervals in a hospital of Luohu District, Shenzhen, and analyzed it with single change point analysis of simultaneous change of mean and variance, and discussed when the change point occurred.Results:The average number of outpatient was 13.250 0 for every time window, the change point occurred at 8: 50, the probability was 0.000 025, the average number of outpatient per 5 minutes from 8: 00 to 8: 50 was 7.000 0, the average number of outpatient per 5 minutes from 8: 50 to 12: 00 was 14.897 4, and the ratio of number of outpatients before and after the change point occurred was 1∶2.Conclusion:In the case of no need to determine the base period or compare the data, the change point analysis of normal distribution based on the principle of likelihood ratio provides a new way of statistical analysis and statistical monitoring for the rational allocation of health resources based on the number of patients.

Detection of electroencephalographic, electromyographic, and cardiac variations during wake-sleep transition through change point analysis / Detección de variaciones electroencefalográficas, electromiográficas y cardiacas durante la transición vigilia-sueño por medio del análisis de punto de cambio

Abstract: Introduction: Wake-sleep transition is a continuous, gradual process of change. Most studies evaluating electroencephalogram spectral power during this transition have used variance analysis (ANOVA). However, using this type of analysis does not allow one to detect specific changes in the statistical properties of a time series. Objective: To determine whether change point analysis (CPA) makes it possible to identify and characterize electroencephalographic, electromyographic, and cardiac changes during the wake-sleep transition through a cross-sectional study. Method: The study included 18 healthy volunteers (12 women and six men), from which polysomnography data were obtained during a two-minute transition. Heart rate, respiratory sinus arrhythmia, electroencephalogram spectral power, as well as electromyographic median and mean frequency and electromyographic root mean square were calculated in five-second segments. These segments were analyzed using repeated measures ANOVA, and CPA focused individually and for the group as a whole. Results: Repeated measures ANOVA and CPA by group found decreased levels of alpha and beta power and beta/delta index during wakefulness, and increased theta and delta power levels during sleep. CPA by individual found that only alpha power changed in all participants and failed to identify a specific moment when all the variables studied changed simultaneously. Discussion and conclusion: We consider that CPA provides additional information to statistical analyses such as ANOVA for the specific location of physiological changes during sleep-wake transition.

Resumen: Introducción: La transición vigilia-sueño es un proceso de cambio continuo y gradual. Los estudios que han evaluado el poder espectral del electroencefalograma (EEG) durante esta transición han usado principalmente el análisis de varianza (ANOVA). Sin embargo, con este tipo de análisis no se pueden ubicar con precisión los cambios en las propiedades estadísticas de series de tiempo. Objetivo: Evaluar si el análisis de punto de cambio (APC) permite identificar y caracterizar cambios electroencefalográficos, electromiográficos y cardiacos durante la transición vigilia-sueño mediante un estudio transversal descriptivo. Método: Participaron 18 voluntarios sanos (12 mujeres y seis hombres) a los cuales se les realizó una polisomnografía para determinar un periodo de transición de dos minutos. En segmentos de cinco segundos se calcularon la frecuencia cardiaca, arritmia del sinus respiratorio, frecuencia mediana y media cuadrática del electromiograma y poder espectral del EEG. Estos segmentos se analizaron con ANOVA de medidas repetidas y con el APC que se aplicó de forma grupal e individual. Resultados: Con el ANOVA de medidas repetidas y el APC grupal se encontraron disminución de la potencia alfa, beta e índice beta/delta durante la vigilia e incrementos de theta y delta durante el sueño. Con el APC individual no se identificó un momento específico en el que todas las variables estudiadas cambiaran simultáneamente; además, se encontró que sólo la potencia alfa cambió en todos los participantes. Discusión y conclusión: El APC aportó información adicional al ANOVA ya que permitió conocer la ubicación específica de los cambios en las variables fisiológicas estudiadas durante la transición vigilia-sueño.

Calculation of the ELISA’ s cut-off based on the change-point analysis method for detection of Trypanosoma cruzi infection in Bolivian dogs in the absence of controls

In ELISAs, sera of individuals infected by Trypanosoma cruzi show absorbance values above a cut-off value. The cut-off is generally computed by means of formulas that need absorbance readings of negative (and sometimes positive) controls, which are included in the titer plates amongst the unknown samples. When no controls are available, other techniques should be employed such as change-point analysis. The method was applied to Bolivian dog sera processed by ELISA to diagnose T. cruzi infection. In each titer plate, the change-point analysis estimated a step point which correctly discriminated among known positive and known negative sera, unlike some of the six usual cut-off formulas tested. To analyse the ELISAs results, the change-point method was as good as the usual cut-off formula of the form “mean + 3 standard deviation of negative controls”. Change-point analysis is therefore an efficient alternative method to analyse ELISA absorbance values when no controls are available.

A change point analysis of performance indicators of revised national tuberculosis control program in Gujarat.

Background: In Asia, India is declared as High Tuberculosis/ High HIV/High MDR-TB burden country. The WHO estimates that there are 3100000 cases of TB at a rate of 256/1 lakh population, and the new cases are emerging at the rate of 185/ lakh population every year. To counter TB challenge, the country has implemented Revised National Tuberculosis Programme (RNTCP) since 1997. But, at which critical point of time the programme had significant success or failure in its indicators is important to analyze as this can help programme managers to detect measures that strengthen the impact of program. Aims & Objective: To identify periods when the program indicators had significant change and to corroborate it with activities which may have led to this change. Material and Methods: The RNTCP indicators – NSP case detection rate, NSP case notification rate, Cure rate etc. were analyzed from year 1998-2011. The change point analysis is the method used for this time series analysis. Results: NSP case detection increased significantly from 2002, while cure rate improved from 2005. The sustained case holding is the key to better results. Same way, smear positive retreatment cases, though declining are not reduced to a change point. Conclusion: The analysis suggests that the defaulter rate and failure rate among NSP is maintained in downward direction. It is indicative of better adherence, higher effectiveness and extending reach of the program.

At what age group blood pressure discontinue to increase? An assessment using change-point analysis.

Aim To study at what age group blood pressure ceases to increase for women and men Methods Applying change-point technique, we used our existing database - mega base-line cross-sectional Hulu Langat Health Study that was initiated in 2000 - to locate the most appropriate age limit in planning promotive, preventive and controlling strategies against systolic hypertension. Results Systolic hypertension was found to be constantly increasing for both gender right from the early age until the middle age group. However, women achieved the systolic peak 15 years earlier (at 41-45 years old) than men (at 56-60 years old). Systolic blood pressure was steadily declined after the peak. Conclusions Hypertension intervention, we recommend age before 40 (women) and 55 (men) be the most appropriate period to apply various public health intervention, after that, the action must be exclusively curative.