Assessing surgery volumes and medical complication rates

Kennell used logit regression modeling to examine the statistical relationship between annual MTF hospital and surgeon volumes and in-hospital surgical and medical complication rates for bariatric surgery. Independent control variables included surgery type and complexity, patient characteristics, hospital characteristics, and a fiscal year-specific time trend indicator.

Examining the relationship between out-of-pocket costs sharing and preventive benefits

High cost sharing may reduce utilization of key preventive health care benefits. Kennell evaluated the cost-effectiveness and cost saving potential of an array of TRICARE preventive health care benefits for DoD. As part of this study, we prepared a logit regression model which examined the relationship between colorectal cancer (CRC) screening rates and cost sharing, controlling for other factors such as age, gender, risk, sponsor rank, and year of utilization.

Applying propensity score matching models

In conjunction with our subcontractor, RTI, Kennell assisted the Centers for Medicare & Medicaid Services (CMS) in an analysis of the differences in Medicare payments, facility costs, and Medicare margins between patients referred to Medicare’s Long Term Care Hospital (LTCHs) programs and patients with similar conditions and level of acuity who do not use the LTCH program but remain for longer periods in general acute care hospitals. This comparison is difficult because LTCH program patients are generally sicker than others with the same conditions. The study used propensity score matching to control for the selection of sicker patients into LTCHS. The Kennell/RTI team then estimated the impact of referral to LTCHs using the difference between LTCH users and clinically similar non-LTCH users in Medicare payments, costs, margins, and lengths of stay and found that LTCH patients have higher total payments, higher costs, and longer stays.

Forecasting health care utilization after medical facility relocation

For Navy Medicine, Kennell completed studies to evaluate the expected changes in utilization for emergency department and obstetrical use resulting from a relocation of a Navy hospital. We analyzed historical claim and geographic data and used logistic regression to estimate the key characteristics that drive a beneficiary’s decision to use care and where to obtain such care. The results from the analysis were used to inform the physical design of the relevant departments within the new facility.

Preparing analyses of TRICARE population and utilization trends

With the introduction of the TRICARE benefit in the mid-1990s, followed by significant increases in employee cost-sharing over the past decade, many under-65 military retirees have shifted from a reliance on employer-sponsored insurance to reliance on TRICARE as their primary health plan. Kennell has integrated TRICARE population and workload data, in combination with multiple survey sources, to develop a logistic regression model that analyzes the key factors driving historical trends in MHS reliance among under-65 retirees and has used this model to forecast future trends in MHS usage to support TRICARE budgeting and benefit reform analyses.

Evaluating the Patient Centered Medical Home model in Military Treatment Facilities

Kennell evaluated the implementation of the Patient-Centered Medical Home (PCMH) model within military treatment facilities (MTFs). The evaluation was focused on those MTFs that were the earliest PCMH adopters in each of the three military Services. Kennell developed a series of multivariate regression models to compare trends for a range of potential impact measures across these early adopters relative to a comparison group of MTFs that had not adopted PCMH as of the end of FY13. TRICARE beneficiaries enrolled to the regional Managed Care Support (MCS) contractor networks served as a second comparison group. Examples of outcome measures evaluated include PMPM health care costs, hospitalization rates, emergency department usage, HEDIS measures, and survey-based patient satisfaction.

Performing a meta-evaluation of the Centers for Medicare and Medicaid Innovation (CMMI) primary care initiatives

Kennell and its subcontractor, RTI, are conducting a meta-evaluation of six major Medicare primary care initiatives being conducted by CMMI. The review is using mixed methods to address the commonalities and differences in the findings of the six initiatives. Part of the quantitative analysis includes analysis of Medicare claims data to conduct a meta-regression of the impacts of the initiatives.

Completing an econometric evaluation of a chiropractic care benefit

Kennell assisted the Defense Health Agency (DHA) and the Chiropractic Study Working Group (CSWG) in evaluating the Congressionally-required chiropractic care program in the direct care system. The main purpose of the evaluation was to determine whether the Chiropractic Care Program is cost beneficial. To do so we used an econometric model to examine the difference in costs between beneficiaries who used standard treatment approaches and compared them to those who used both standard treatments and chiropractic care in treating a range of musculoskeletal problems. Our approach controlled for diagnoses, age, gender, and other demographic variables. The project evaluation concluded that the use of chiropractic care did not reduce standard treatments costs.