RWE and Data Analytics
Providing meaningful structure and insights to real world data through cutting-edge analytics
What is RWE?
Real-world data (RWD) are observational data generated in a routine clinical practice, in a natural, uncontrolled setting – outside of conventional clinical trials. Transforming these RWD into meaningful evidence is called Real-World Evidence (RWE).
How do we make a difference?
We carefully analyse RWD to answer key questions related to disease epidemiology and its burden, treatment patterns, healthcare resource utilization, comparative effectiveness and costs in each phase of the drug development (i.e pre-clinical, Phase I-III and post-approval)
To conduct robust analyses, we employ two stage processes – finding and managing the best dataset and data sources to meet the research question; and analysing the data by implementing right algorithms to give it meaning and structure.
Everything you may need
Real-World Data
- Search and storage of dataset
- Integration of multiple dataset
- Data cleaning and management
Real-World Evidence
- Real-world study design
- Protocol and analysis plan development
- Statistical modelling, predictive analytics, machine learning
- Disease and product registries, electronic health records, medical claims and surveys data analytics
- Medical chart review studies
- Social media analysis
- Research on external comparator or standard of care for accelerating drug development
- Evidence gap analysis
- Data visualization
Clinical Trial Data Analysis
- Economic evaluation alongside trial
- Survival analysis
- Patient-level data analysis
- Utility data analysis
Case Study
Tobacco use among pregnant women in low and middle income countries and association between its usage, education and socioeconomic status: Secondary data analysis
The Problem
Tobacco smoking among females of reproductive age group, especially during pregnancy is of particular concern because of the many adverse effects of maternal smoking on pregnancy and infant health outcomes. To help women from low and middle income countries (LMICs) reduce/ quit tobacco use, it is important to to understand the extent and distribution of tobacco use. Therefore, with recent prevalence rate estimates of tobacco use among women of reproductive age group and those who are pregnant from the same data set, will allow a better understanding of whether these rates differ between the two groups of women. Also, current evidence suggests that use of tobacco varies with the type of residence, education and socio-economic class. Smokeless tobacco is more prevalent among those who are less educated and belong to low socioeconomic class. Thus, there is a need to assess whether this distribution varies among pregnant women.
The Solution
- Used Demographic and Health Surveys (DHS) to estimate prevalence rates and possible association of socio-demographic characteristics with tobacco use behavior. These are nationally representative surveys conducted systematically in LMICs.
- Due to complex DHS sample design (two-step stratified cluster design), sampling weights were calculated to account for differential probabilities of selection and participation.
- Analyzed tobacco use prevalence estimates for pregnant and non-pregnant women for exclusive smoking, exclusive smokeless tobacco use, dual use and no tobacco use.
- Estimated pooled prevalence estimate of tobacco use for 42 LMICs to estimate the current tobacco use prevalence among women of reproductive age.
- Performed multinomial logistic regression analysis to see the effect of pregnancy status, type of residence, education, wealth index and age on use of tobacco by women in LMICs.