Put healthcare data to work with benchmarking analysis

  • Print
  • Connect
  • Email
  • Facebook
  • Twitter
  • LinkedIn
  • Google+
By Marcella Giorgou | 02 November 2015

Five years after the healthcare reforms of the Patient Protection and Affordable Care Act (ACA), finding solutions to manage healthcare costs beyond cost shifting is more important than ever. Squeezing dollars out of the system can help manage the excise tax in 2018 and makes for a “win-win” situation, financially benefitting both the employee and employer. Where to squeeze is generally specific to each employer. The answer is found in the data, where patterns of trend and utilization can lead to useful benchmarking for employers.


The two major components of healthcare trend are:

  • Changes in utilization of services which may be attributable to:
    • Modifications in medical practices, new technologies or practices, and increases in the supply of services.
    • Changes in overall health and/or attitude of the insured population.
    • Adjustments in benefit designs, utilization review programs, and health insurance industry changes from laws or mandates.
  • Changes in average cost per service which is primarily due to inflation, as well as changes in provider reimbursement agreements, service severity mix, shifting of utilization among service categories (such as increased use of outpatient services versus inpatient services), medical practice patterns, increased use of expensive modern technology, and the expiration of patent protections.

What’s driving trend?

Understanding the major drivers of trend each year is important in the cost management process. In Figure 1, we’ve measured total trend over the past three years. In Figure 2, we’ve broken out the trend by service category. The biggest takeaway from this analysis was the high outpatient trend, which is due to increased utilization of drugs administered in an outpatient facility. An improvement in the office visit benefit resulted in a high professional trend for 2013–2014.

Figure 1: Plan PMPY Allowed Trends

Figure 2: Plan PMPY Allowed Trends by Major Service Category

Utilization and cost benchmarking

A benchmarking analysis involves comparing detailed target performance benchmarks with corresponding claim experience. Target performance benchmarks are based on “Well-Managed” and “Loosely Managed” benchmarks for utilization, allowed amount levels, and plan paid levels, using Milliman’s Health Cost Guidelines (HCGs).1 The HCG benchmarks are calibrated to reflect the demographic profile, geographic profile, and benefit design of the plan.

Loosely Managed utilization levels are representative of plans with some utilization review, preauthorization, and case management. Well-Managed values represent nationwide claim cost and utilization targets in a managed care environment, such as a staff model health maintenance organization (HMO) or a globally capitated provider group, which effectively applies utilization management principles across the entire continuum of medical care, including inpatient care, outpatient facility care, ancillary testing, routine office care, referral physician care, and prescription drugs.

Utilization metrics can be used to measure the effectiveness of utilization management programs or suggest additional opportunities. Utilization is primarily managed by medical management functions. By comparison, unit costs are managed primarily through provider negotiations.

In Figure 3, utilization metrics are used to suggest additional opportunities. Utilization of inpatient medical (in red) has been significantly higher than the benchmarks (in green, which have been decreasing). Inpatient medical will be an issue to address for benefit planning for next year. There are several different initiatives the plan can employ to help reduce unnecessary inpatient utilization. These initiatives include review of facilities that have high avoidable admissions, with possible tightening of participating provider networks and pre-authorization, which can help avoid unnecessary admissions on a case-by-case basis.

Figure 3: Inpatient Medical, Admits/1,000

In the analysis shown in Figure 4, utilization metrics are used to measure the effectiveness of utilization management programs, with a major focus on encouraging proper utilization of the emergency room (ER) and urgent care facilities by reducing the urgent care copay in 2013. Urgent care utilization increased; however, opportunities still exist to change site of service delivery because members continue to over-utilize the ER. Note that in 2015, the plan increased the cost-sharing difference between ER and urgent care facilities, which may result in more appropriate utilization of both ER and urgent care. The impact of this plan design change will be evaluated once credible experience data is available.

Figure 4: Emergency Room and Urgent Care Facility Utilization (per 1,000)

Depending on the size of the plan, there are other Milliman tools available that could help provide additional drill downs of the issues identified above, including:

  • Hospital Inpatient Profiler: This tool addresses high inpatient admits through a review of facilities that have high avoidable admissions and length of stay.
  • Milliman Advanced Risk Adjusters (MARA): Offers population health risk assessment by assigning risk scores that explain expected use by key service components (including ER).


A drill-down of the data for this employer was able to identify some problem areas for the plan, including site of service. Addressing these issues can help the plan operate more efficiently and helps direct benefits strategy away from cost-shifting to the employee. This may result in a mitigation of potential impacts from the excise tax in 2018 as well as maintenance of a plan that is viewed favorably by employees.

1The HCGs are a cooperative effort of Milliman health actuaries and represent a combination of their experience, research, and judgment. An extensive amount of data is used in developing the HCGs and that data is updated annually. The cost models consider utilization and average charge levels for roughly 60 benefit categories and can provide relativities in per capita plan costs between programs of different design, demographics, or geography.