Page 59 - ISMCON souvenir 2021

ISMSCON - 2021

Results: All the five states, had a high percentage of women, utilizing institutional delivery services
(95.7%). Institutional delivery was mostly opted in Kerala (almost 100%) and all other states too had a
significantly high percentage (above 90%). Autonomy together with factors such as, state, women’s age,
highest education level, place of residence, wealth index and media exposure had a significant positive
association with institutional delivery. Women having higher level of autonomy had 1.37 times higher
odds for opting institutional delivery, than those having low level of autonomy. Similarly women with
higher educational attainment, greater wealth index and higher media exposure had 22.52, 6.79 and 1.31
times higher odds for availing institutional delivery respectively.
Conclusion: Institutional delivery services is in a good shape in the concerned south Indian states.
However, women autonomy, wealth index, educational attainment and media exposure are some major
factors which positively influence an institutional delivery. Thus, development of strategies for improving
autonomy among women along with the improvement of her education, eradication of poverty and
promotion of topics like importance of institutional delivery via elaborate media exposure are required for
healthy childbirth.

OS22: DOSE RESPONSE META-ANALYSIS FOR CONTINUOUS

OUTCOME- A TWO-STAGE APPROACH

Deepthy M S Kalesh M Karun , Harichandrakumar K.T , N. Sreekumaran Nair 1
1,
2
1
1 Department of Biostatistics, JIPMER, Puducherry, India
2 Division of Biostatistics, MOSC Medical College, Kolenchery, Ernakulam, Kerala, India
Email id: deepthyms27@gmail.com

Introduction

The effectiveness of dose variants of same drug is usually pooled separately for each dose placebo
comparisons or overall differences between the drug and placebo will be looked at without considering
the variations in the doses in the traditional meta-analysis. The proposed work overcomes this limitation
by modelling the relationship between effect sizes and different doses obtained from multiple studies
through a two-stage dose response meta-analysis. Through this method the efficacy of different doses of
the same drug on the outcome and the optimum dose to produce maximum response can be determined
when compared to the traditional model.

Methodology
In the present study, a dose response meta-analysis was performed by updating an existing systematic
review on combination therapy with Zonisamide and anti-Parkinson’s drugs for Parkinson’s disease. The
primary outcome was UPDRS part III scores. Studies reporting at least a combination of two different
doses of drug along with placebo were included. A two stage approach utilizing restricted cubic spline
model and multivariate meta-analysis was performed using R software. The study specific regression
coefficients obtained by fitting appropriate function in the first stage were later combined using multivariate
meta-analysis in the second stage.
Results

The pooled difference in mean UPDRS part III scores were estimated to be -2.11(95%CI:-3.65,-0.58),
-2.82(95%CI: -4.08,-1.57), -3.18(95%CI:-4.40,-1.96),-3.54(95%CI:-5.51,-1.57),-3.90(95%CI:-6.85,-0.95),
-4.26(95%CI:-8.26,-0.26), -4.62(95%CI:-9.69,0.45) and -4.98(95%CI:-11.13,1.17) for different doses
such as 25,50,75,100,125,150,175 and 200 mg/day when compared to placebo. The maximum response
estimated to be -4.98 at 199.96 mg/day using the current method. The dose required to produce 50% and
80% of maximum predicted response were estimated to be 33.61 and 130.71 mg/ day

CONFERENCE SOUVENIR 57