Expertise and experience in data extractions

Kennell has decades of experience developing, modifying, and using large corporate data sources. With functional expertise in administrative and clinical health data and experience in performing complex data extractions, Kennell has the ability to quickly and accurately respond to information requests.

Information system design and requirements development

Providing full life-cycle support for the Military Health System Data Repository

Kennell has provided full life-cycle support for the Military Health System Data Repository (MDR) including identifying and coordinating data feeds for new sources of military health care data, developing computer programs for integrating data from disparate sources, reviewing data quality issues, identifying solutions to problems identified, and assisting other stakeholders and users of the MDR. The MHS Data Repository serves as the source of data for nearly all MHS systems and Kennell plays a key role in facilitating the transfer and understanding of data outside the MDR. Kennell authored the MDR data dictionaries, provided training to users, consulted with external researchers, and briefed leadership on the status of MHS data.

Developing a user-friendly Management Information System

Kennell staff played a lead role in developing a user-friendly MIS for DoD (the M2 system) which integrates all key MHS data sources for direct care, purchased care, eligibility, and program enrollment. We have continued to support this central M2 system by identifying useful new data sources and specifying the processes necessary for integration into the central system.

Supporting the implementation of an Electronic Health Record

The Military Health System (MHS) is implementing a new Electronic Health Record, called MHS Genesis, which is based off a commercial off the shelf EHR and will be used for data collection for care received in hundreds of military health facilities around the world. Kennell is playing the lead role in identifying functional requirements for data feeds from MHS Genesis. Kennell is also working to redesign data processes to accommodate the transition to the new system.

Data extraction and transformation logic

Creating extractions for research files

Kennell has experience developing specific data extractions for research projects. We consult with researchers to understand the research objectives and methods that will be used to analyze data. We develop methods for selecting cohorts and control groups, including using our knowledge of government health care programs to provide advice about which groups of patients may have censored data. We use the cohort and control groups to then extract detailed research files, specific to each study. Some example projects include the President’s Commission on Wounded Warriors, the Institute of Medicine’s studies on post-traumatic stress disorder, and numerous reports to Congress.

Providing analysis of Veterans Health Administration (VHA) data

Kennell provided data analytics in support of the VA Choice Act Section 201 Assessments under a subcontract to the MITRE Corporation. The data analytics included detailed exploration and prediction of Veteran health care needs and VHA’s ability to effectively and efficiently meet those needs. The twelve individual assessments involved forecasting health care capabilities and resources, physician staffing and productivity measurement, emergency department performance, and appointment scheduling. This required analysis of Veteran demographic data through direct access to the VA’s Corporate Data Warehouse (CDW) within the VA Informatics and Computing Infrastructure (VINCI) platform. For the VA Choice Act Assessments, we developed SAS code to analyze VA appointment, consult, inpatient, emergency department, and outpatient workload at national, VISN, station, and division levels. We provided expert guidance on the proper handling and implementation of business rule logic to reconcile CDW with totals of VHA workload from MedSAS. Our programs efficiently cycled through billions of health care records to quantify diagnosis condition and disease prevalence rates. MITRE and its partners (RAND, McKinsey, and Grant Thornton) also relied heavily on Kennell to implement de-identification algorithms for handling sensitive protected health information (PHI).

Improving the Medicare Chronic Conditions Warehouse (CCW) data

Kennell used the Medicare claims data in the CCW data to examine alternative definitions of chronic conditions, to define standard business rules for defining conditions, to analyze use and cost of care for Medicare beneficiaries, and to prepare extracts for researchers.

Data integration

Integrating VHA and DoD data

Kennell’s analyses routinely require integration of health care population, utilization, and cost data from multiple data systems. In support of the new joint VA DoD Lovell Health Clinic, we integrated VHA and DoD data to calculate budget contributions from TRICARE and we continue to support the Lovell Health Clinic in preparing ongoing reports of linked TRICARE/VHA data. We have also facilitated the merger of DoD and VHA data for projects related to the care of wounded warriors and developed patient quality metrics by properly integrating both VHA and DoD data sources to understand the full picture of care being received for jointly treated patients. Kennell has also developed methods for linking data from the MHS and the VHA for population cohorts. For the National Institute on Drug Abuse (NIDA) and Brandeis University, we looked at Army service members returning from deployment, and followed their health care utilization for the three years following their return from deployment. A main requirement of this study was to determine when the Service member became eligible in the VA, and whether they utilized health care in the MHS or the VHA. In a second study, we looked at Army National Guard and Reserve members returning from deployment who had received referrals on their Post Deployment Health Reassessment (PDHRA) form to find whether the Service member received follow-up care, and in which system.

Integrating Medicare and TRICARE data

We also routinely integrate TRICARE data with information from Medicare. In support of DoD’s annual capitation rate negotiations with a group of HMOs who comprise the TRICARE Designated Providers program, we also have experience integrating TRICARE and Medicare claims data for retirees and their dependents aged 65+.

Integrating different sources of DoD data

Kennell oversaw integration of deployment data from DoD’s DEERS personnel system and the development of TRICARE’s Ill, Injured, and Wounded (IIW) Service Member data file.

Integrating clinical data into a corporate data warehouse

At Kennell, we els in a way that allows analysts to structure their queries efficiently and effectively. Kennell has expertise in this area because we are both developers of large data warehousing systems, as well as analysts and users ourselves. We recently completed the data model design, development, and integration of Clinical Data Repository (CDR)/AHLTA data and the 3M Healthcare Data Dictionary (HDD) into the MHS Data Repository (MDR). Within a few months, Kennell successfully designed, developed, and processed the CDR’s highest value clinical data from fiscal year 2009 forward. The highest value clinical data sources we processed and made available in the MDR include: vital signs, appointments, immunizations, radiology results, laboratory results (including chemistry, pathology, and microbiology results), and medications. These data sources are now some of the most commonly used and requested data types in the MDR for clinical research studies.

De-identification and Privacy Support

Supporting information security and de-identification

Kennell has been the key advisor to the Military Health System (MHS) Office of Privacy and Civil Liberties for more than a decade. In this role, Kennell staff members serve as the subject matter expert for HIPAA determinations about the level of identification of data (protected health information, limited data set, de-identified). We review research related requests for data and advise the Privacy Office on the type of data use agreement that should be in place to protect the information. Kennell is also frequently called upon to prepare de-identified datasets.

Consulting with researchers to provide appropriate research data

When researchers first attempt to study a new health care system, their knowledge of the organization’s data is often lacking. For the MHS Office of Privacy and Civil Liberties, Kennell plays a key role in assisting researchers in obtaining the data they need to do their research. Kennell reviews IRB protocols and meets with the investigators to discuss study goals, methods, cohort and control groups and potential confounding factors. We teach them about the MHS health care systems they are studying, the benefit structure, the beneficiary mix, and limitations to the data, as well as proactively offer advice about additional variables that can be provided to inform the research. We help to design customized research data, and where possible, de-identify the datasets. We also assist researchers in the process of obtaining data sharing agreements.