BIOSTATISTICAL ASPECTS OF POPULATION HEALTH

Bio-statistical aspect of Population Health book is important from the perspective of development of newer estimation procedures & analytical techniques that are able to address complex problems public health phenomenon simplistically. The book is an attempt in this direction as it not only suggests new estimation procedures but also demonstrates empirically, the estimates that are superior compared to earlier findings, using existing datasets. The multilevel estimation procedure used by Prof. K K Das, examined the effects of household and village environmental factors on the prevalence of diseases among individuals within and between households in North-Eastern parts of India. The indicators such as availability of doctors in the village, the distances of medical facility from the village, source of drinking water, separate kitchen facility, toilet facility, cooking fuel, type of house, urban-rural set-up have significant impact on disease prevalence and different types of diseases. Dr. Ajay Pandey through his analysis demonstrated the significant gain in neonatal survival among pregnant mothers who adhere to WHO recommended inter-birth birth-interval length of 33 months. The neonatal deaths were also found to be lower in communities that are connected to all weather roads compared to those who are not. This has policy implications for demographically poor performing Empowered Action Group States, as the infrastructure push is needed in these Sate. In recent times especially since 2014 there has been increased allocation by the Central Government for infrastructure development and construction of all weather roads in these states.

Dr. Vivek Verma using Rank Set Sampling (RSS) technique demonstrated the superiority of RSS in situations where the probability of occurrence of an event is not fixed but a random quantity. He showed that, in the estimation of the probability of infant deaths, Bayesian estimators based on ranked set sample not only proved more effective and efficient than any other estimator, but also consistent with the NFHS reported value. Dr. Aditi Baruah derived a statistical model for the distribution of closed birth interval by considering variation in post-partum amenorrhea (PPA) period. Using the model Dr. Aditi demonstrated low risk of conception among Adivasi (Tirbal) female tea gardeners of Assam regardless of their parity. It is found that the risk of conception is low in the population surveyed and is reasonable compared to other methods. Dr. H. Brojeshwor Singh using the data from rural Manipur estimated the average duration of PPA as 6.6 months. PPA is the time interval between the termination of women’s pregnancy and the beginning of the first subsequent menstruation. This finding has immense value from the policy perspective for those designing family planning strategies, especially the PPIUCD/IUCD.

Professor M. Nazrul Islam of Bangladesh suggested new estimation procedure of estimating speed of aging in a population as a function of demographic components i.e. life expectancy and population fertility rates. He tested his estimates with the existing measures of aging velocity using census data of Bangladeshi population for the census years 1981 and 2001. The findings demonstrate that the method suggested are good alternative and consistent over the existing methods. The alternative approach suggests slower aging process than those obtained by the existing measures. Dr. Tandrima Chakraborty of NSSO-India using Weighted Epidemic Chain Binomial Model with one introductory case for four and five member households demonstrated the superiority of model fit to epidemic dataset in studying the pattern of spread of infectious diseases.

Dr. Ramesh K Vishwakarma using Liver Cirrhosis marker data demonstrated the feasibility of computing concordance correlation coefficient (CCC) through an application of prior information using Bayesian approach. The study demonstrated that the Bayesian counterpart of CCC estimates applied between serum bilirubin and albumin among liver cirrhosis patient’s data and its 95% posterior interval for concordance correlation coefficient were found to be very narrow, indicating that the estimates obtained through the suggested method are very precise. Dr. Dharmendra Kumar Dubey determined the predictors of low birth weight among adolescent mothers in Assam-India. Predictors of Low Birth Weight were found to be low levels of education, being poor, fourth & above birth order and mothers being anemic. Prevalence of LBW varied across the districts, with highest reported from Kamrup and Dhubari and lowest reported from Sonitpur and Karbi Anglong districts of Assam in India. Dr. Padum Narayan studied the linkages of son preference over daughters in childbearing process among married couples in India. The findings from the study shows that the parity progression ratios were consistently higher among currently married women who had only daughters at all parities as compared to those who had only sons or both sons and daughters in NFHS-3 (2005-06) as well as NFHS-4 (2015-16) irrespective of the place of residence. The pace of progressions differs substantially between urban and rural areas. Greater parental preference for sons over daughters has been observed in the rural areas as compared to urban areas at all parties in 2015-16. The study suggest that the recent initiate of “BETI Bachao, BETI Padhao” which literally means educate girl child to save girl child is very timely to eliminate the son preference in the society.

Dr. Jagriti Das using Log Normal distribution as an actuarial risk model estimated important actuarial quantities like the probability of ultimate ruin, moments of the time to ruin, the surplus prior to ruin and the deficit at the time of ruin when the underlying claim severity distribution is Log Normal. Dr. Jaishree Prabha developed imputation methods to reduce the impact of non-response sampling error at both the occasions in two-occasion successive (rotation) sampling. Dr. Lipi B Mahanta improvised the Generalized System of Curves that is important in describing frequency distributions for wide a variety of observed distributions.

Scholarly work by authors not only demonstrate their innovative thinking but also is sincere gratitude towards an inspiring mentor, guide & motivator Professor Dilip C Nath who at the age of 70 keeps transforming our dreams into actions. He is currently Professor Emeritus at Royal Global University, Assam and was a Former Vice Chancellor, Assam University Silchar, Assam, India.

DOI:10.5539/9780978430160
URL: https://doi.org/10.5539/9780978430160 

ISBN: 978-0-9784301-6-0 (print)
ISBN: 978-0-9784301-7-7 (ebook)