Page 66 - ISMCON souvenir 2021

ISMSCON - 2021

that article is in the process of being published elsewhere. Objective prior information was obtained from
published literature sensitivity and specificity of similar tests. Non-informative Beta (1,1) prior was used for
the prevalence of sepsis. Posterior distributions of prevalence and other measures of diagnostic accuracy
were obtained by applying Bayes theorem and employing the Markov Chain Montecarlo approach. We
compared conditional independence and conditional dependence model, where conditional dependence
means there is a possibility that false positive and false negative errors simultaneously occur.
Results:
The conditional dependence model had a higher estimate of prevalence (20% point higher) when
compared with the conditional independence model. However, the sensitivity and specificity of the
methods were comparable between the models.
Conclusions :
LCMs combine the prior knowledge with the information from the data on imperfect diagnostic tests
to estimate test parameters like sensitivities, specificities, and PPV along with the prevalence. The
conditional dependence model accounts for any dependence between the tests.

OS31: Application of Bio statistical Techniques for Summarizing

Evidence; Systematic Review and Meta-Analysis

JITENDRA RAMAKANT GAWDE
Advanced Centre for treatment, research and education in cancer, kharghar
jgawade12@gmail.com

Abstract
In today’s world of uncertainties, you can’t say what will happen in the next second. World is changing
at every second as many fields are taking birth. Some fields like medical are growing at faster rate. The
race of development is getting more interesting and fields like research and development are running
very ahead in the race. Many researches are being published and many are still in queue and it is also the
need of the hour. Many bio statistical techniques are being used for summarizing evidence of similar kind
of research. Systematic review and Meta-analysis are one of those techniques which are used commonly
and produces results with great accuracy. The chances of producing bias and inaccurate results are
very less however, you may not always be able to find the highest level of evidence to answer your
research question. Both the techniques are called together but have uniqueness in itself. In this paper
we will see how we can use this two techniques for summarizing evidence of our research question. The
only difference is in systematic review, you try to analyze your evidence in a systematic way whereas
meta-analysis is a statistical process which combines data of multiple separate studies for a specified
research question. In recent years, systematic reviews and meta-analyses have been actively performed
in various fields. Therefore this paper throw light on few techniques which can be used for summarizing
evidence to your research question.

Keywords: systematic review, meta-analysis, evidence, summarizing, research question

OS32: BIOSTATISTICS – A PERQUISITE TO MEDICAL RESEARCH

J. KEZIA ANGELINE , DR.V. SRIVIDHYA 2
1
1Consultant Biostatistician for the NIH-IPV Project, St. John’s Medical College and Research Institute, Bangalore;
E-mail: johnkezangetsy@gmail.com
2Assistant Professor (SG), Avinashilingam Institute for Home Science and Higher Education for Women,
Coimbatore – 641 043

64 CONFERENCE SOUVENIR