Our Capabilities
Feasibility Investigation
Wondering if a research idea can be implemented? Feasibility Investigation is the first and foremost step. We will take care of the sample size estimation, power analysis, as well as preliminary descriptive results. Our preliminary analysis often provides valuable information to overcome biases inherent to observational studies and allows stakeholders to make a informed decision at the very beginning. Based on our initial findings, we will also come up with a most feasible approach to tackle the research questions.
Study Design
We can never over-emphasize the importance of study design. Conclusions from a flawed study design are meaningless. On the contrary, well-designed observational studies have been shown to provide similar results to randomized clinical trials. With right research questions, we will develop well rounded study designs suitable for each specific question and available data. For pharmacoepidemiology studies, we are highly experienced in cohort study design and case-control study design. In addition, when necessary, we will develop novel study designs including base-study-cohort approach and the emulated trial design, which are increasingly adopted in observational studies currently.
Data Cleaning
Even for seasoned programmers and statisticians, cleaning and handling huge amount of data can be a daunting task. With our innovative and carefully designed standard operating procedure (SOP), we make sure data are cleaned efficiently and avoid issues that are prone to human error. We specialize in missing value imputation using different methods ranging from imputation using mean/median values to more advanced methods, such as multivariate imputation by chained equation, K nearest neighbors (KNN) and deep learning. After the data pipeline, there will be a final data set created that is cleaned, checked, and most importantly, analyzable.
Advanced Analytics
Estimating proper models will ensure research questions are answered appropriately. We make every effort to estimate, validate statistical models and check assumptions. We deploy a broad range of statistical solutions that harness the power of statistical inference including:
-Generalized Linear Models (Logistic Regression, Modified Poisson Regression)
-Survival Analysis and Competing Risks Models
-Longitudinal and Event History Analysis
-Quasi-experimental Design and Causal Inference (Propensity Score Matching, Inverse Probability Treatment Weighting)
-Prediction Models
-Bayesian Statistical Models
Data Visualization
Through our visualization tools and techniques, we help our clients communicate information and deliver the take-aways clearly and effectively. We strive for providing insights into complex data by communicating its key-aspects in a more intuitive way. Especially we are proficient in visualizing geographic information and complex data with spatial attributes. Our in-house Geographic Information System (GIS) team will bring a unique expertise in creating informative graphics for disease incidence, prevalence, medication utilization, to name a few.
Medical Writing
With our statistical expertise and domain knowledge, we aim to transform dry clinical facts into professional scientific articles and reports. We hope our writings will enhance the reading experience in general, resulting in increased knowledge translation. Corresponding to our clients' dissemination needs, we will produce various formatted reports including internal communications, academic journal publications, national and local conference presentations, or a structured summary in a brief.
We Are Most Knowledgable in
CMS compiles claims data for Medicare and Medicaid patients across a variety of categories and years. This includes Inpatient and Outpatient claims, Master Beneficiary Summary Files, and many other files. Indicators from this data source have been computed by personnel in CDC’s Division for Heart Disease and Stroke Prevention (DHDSP). This is one of the datasets provided by the National Cardiovascular Disease Surveillance System. The system is designed to integrate multiple indicators from many data sources to provide a comprehensive picture of the public health burden of CVDs and associated risk factors in the United States. The data are organized by location (national and state) and indicator.
More Info: https://healthdata.gov/dataset/center-medicare-medicaid-services-cms-medicare-claims-data
IBM® MarketScan® Research Databases are a family of data sets that fully integrate many types of data for healthcare research, including:
• De-identified records of more than 250 million patients (medical, drug and dental)
• Productivity (workplace absence, short- and long-term disability, and workers’ compensation)
• Laboratory results
• Health risk assessments (HRAs)
• Hospital discharges
• Electronic medical records (EMRs)
More Info: https://www.ibm.com/products/marketscan-research-databases
Clinical Practice Research Datalink (CPRD) is a real-world research service supporting retrospective and prospective public health and clinical studies. CPRD is jointly sponsored by the Medicines and Healthcare products Regulatory Agency and the National Institute for Health Research (NIHR), as part of the Department of Health and Social Care.
CPRD collects de-identified patient data from a network of GP practices across the UK. Primary care data are linked to a range of other health related data to provide a longitudinal, representative UK population health dataset. The data encompass 45 million patients, including 13 million currently registered patients.
For more than 30 years, research using CPRD data and services has informed clinical guidance and best practice, resulting in over 2,400 peer-reviewed publications investigating drug safety, use of medicines, effectiveness of health policy, health care delivery and disease risk factors.
More Info: https://www.cprd.com/