by Robert York, Kenneth Kaufman, and Mark Grube
As a result of unsustainable costs and an inordinate share of gross domestic product, the U.S. health care system has a new business model—one that is transforming the delivery system from hospital-centric sick care to a super outpatient model that will emphasize community-based care.
Fueled by the “power of a million ideas,” the transformation process is under way. Private employers, consumers, public and commercial payers, hospitals and health systems, physicians and allied providers, health care retailers, and other stakeholders are working to reshape care delivery and lower costs while improving quality, outcomes, and access for consumers.
Recent notable initiatives include significant health plan benefit redesign and efforts to reduce corporate health care costs and eliminate uncertainty regarding annual cost exposure. G.E. has a corporate goalto keep health care costs from increasing more than three percent per year and has been meeting that goal. Walgreens has recently announced that it will move 160,000 employees into a private health exchange with a defined contribution to the employee. Organizations such as Advocate and Geisinger have adopted fee-for-value reimbursement strategies and care management strategies that have driven positive results in both quality and costs. These examples are just the tip of the iceberg as new ideas and innovations surface each and every day.
Given the power of disruptive new business models, the question is, “Are these initiatives and innovations making a difference—is the transformation agenda taking hold?”
Design of a Regional Study
To investigate this question, we designed a study to assess whether there is early evidence that progress is being made with the transformation agenda by hospitals and physicians in the greater Chicago Area. To conduct a data-rich study, we were able to obtain comprehensive and current inpatient and outpatient data for the Chicago regional area. Sources included the Illinois inpatient discharge database, proprietary and provider databases, and the U.S. Census Bureau.
To gain a statistically large sample size, the study population covered a 7-county area with 8.5 million residents. These residents collectively comprise 66 percent of the population of Illinois (Figure 1), or 2.7 percent of U.S. population in 2012.
The 7-county study area demonstrated the diversity of demographics, payer sources, types of providers, and geography (city, suburban, rural) common in other major regions nationwide.
In 2012, the 71 hospitals in the 7 counties discharged approximately 970,000 inpatients. Commercial insurers covered 60 percent of the lives and public insurers (Medicare and Medicaid) covered 26 percent. The remaining 14 percent of residents were uninsured (Figure 2).
Figure 1. Illinois County Area
Figure 2. Insurance Coverage
Indicators of Progress: If the transformation agenda is taking hold, we would expect study results to show that:
- Inpatient utilization is declining
- The depth of the decline reflects structural factors at work in changing health care (such as those cited in 3 and 4), not solely recession-linked factors
- Doctors and hospitals are aggressively increasing intensive medical management, and such management is beginning to show positive results
- Accountable and risk-based care is having an attributable statistical impact above and beyond the increased intensive medical management
We carefully compiled and examined the data relevant to these four trends.
What the Data Show: Inpatient Utilization Is Declining and Structural Changes Appear to Be at Work
Confirming the first trend, inpatient utilization in the studied counties declined by approximately 47,000 discharges—dropping from approximately 1,017,000 discharges in 2010 to 970,000 discharges in 2012 (Data include all payers—commercial, Medicare, and Medicaid—and all ages, excepting newborns). Hospital inpatient decline is not a new trend, of course. As documented in previous research, inpatient utilization in the U.S. has been declining during most of the last decade. Relevant to the new study’s results, however, is the fact that inpatient utilization rates per 1,000 declined across allage groups, averaging a 5 percent across-the-board drop. Utilization declines in the adult population ranged from 5 percent for 45-64 year olds to a much larger 8-9 percent for those age 65 and older (Figure 3).
Figure 3. Inpatient Utilization Rates per 1,000: Percentage Change by Age Group
Sources: Kaufman, Hall & Associates, Inc. analysis based on proprietary market and client data; U.S. Census Bureau Population.
Covered by Medicare for most of their health care costs, patients who are 65 years and older would not be expected to reduce their health care use due to economic factors more dramatically than other age groups. Price-sensitive deferral of care due to rising co-payments and deductibles required in Medicare managed care plans may be at work to some degree, but such deferral is unlikely to account for so large a drop in use rates. Something else more structural in nature appears to be at work with this older population.
Additionally, use rates per 1,000 in the greater Chicago region declined in almost every hospital service line. The median drop between 2010 and 2012 across 33 services was also 5 percent (Figure 4). Utilization declines in cardiology (including interventional), medical gastroenterology, general medicine, deliveries, and psychiatry accounted for more than 60 percent of the total decline in volume.
Figure 4. Change by Service Line in Utilization Rate per 1,000 Population: 2010-2012
Note: Not shown are services with discharges of less than 5,500 in the counties, and labels with less than a 2 percent drop. Data exclude MS-DRG 795 normal newborns. Sources: Kaufman, Hall & Associates, Inc. analysis based on proprietary market and client data; U.S. Census Bureau Population.
Table 1 shows the decline in inpatient use rates in surgical areas, including cardiac surgery, spine/back surgery, orthopedics (joint replacements, implants), and urology. Utilization rates declined across all age groups, effectively cancelling out any increase in volume that probably occurred due to population aging and growth. Again, the age groups covered by Medicare showed the highest use-rate drops, suggesting that structural changes, likely including increased use of outpatient settings for care delivery, might be boosting the rate of decline.
Table 1. Inpatient Utilization Rates per 1,000 in Selected Service Lines: Percent Change in Adult Age Groups, 2010-2012
Sources: Kaufman, Hall & Associates, Inc. analysis based on proprietary market and client data; U.S. Census Bureau Population.
Observation stays often are cited as a major contributor to reduced inpatient utilization. The dramatic increase in assignment of patients to observation status by U.S. hospitals, as documented elsewhere, reflects aggressive efforts to proactively move non-acute admissions to lower-cost, outpatient settings.
With increased use of observation status, the number of patients with one-day inpatient stays would be expected to decline dramatically, since one-day patients would be prime candidates for transfer to observation status and outpatient settings rather than inpatient admission.
In fact, the data show that drops between 2010 and 2012 in medical/surgical patients with one-day LOS accounted for only 9 percent of the total drop in medical/surgical volume (Table 2). Of this 9 percent drop, cases with surgical diagnosis-related groups (DRGs) dropped faster than those with medical DRGs, indeed suggesting some continued movement of surgeries into ambulatory surgical settings.
However, most of the volume decline—91 percent—must be attributed to factors other than shift to observation status and movement of short-stay surgical cases to ambulatory settings.
Table 2. Drop in Volume of 1-Day Stays as a Percentage of Total Decline in Medical/Surgical Volume
Sources: Kaufman, Hall & Associates, Inc. analysis based on proprietary market and client data; U.S. Census Bureau Population.
What the Data Show: Providers Are Increasing Intensive Medical Management and Such Management Is Showing Positive Results
To identify whether a more structural change related to improvements in patient care management are starting to contribute to reduced inpatient use rates nationwide, we looked closely at “Ambulatory Care Sensitive Admissions (ACSAs).” The Agency for Healthcare Research and Quality (AHRQ) definesACSAs as patient admissions “for which good outpatient care (related to underlying chronic conditions, such as adult asthma, diabetes, and congestive heart failure) can potentially prevent the need for hospitalization, or for which early intervention can prevent complications or more severe disease.”
AHRQ has identified 16 ambulatory care sensitive admissions, including uncontrolled diabetes admission rate, hypertension admission rate, dehydration admission rate, and asthma in older adults admission rate, as “prevention quality indicators.” A reduction in ambulatory care sensitive admissions to hospitals would suggest that providers are doing a better job of managing patients’ chronic conditions, keeping patients with ambulatory sensitive conditions out of hospitals by meeting their prevention and care needs in outpatient and home settings.
Using our data sources, each patient’s primary diagnosis code at admission was identified as either one of the 16 ambulatory care sensitive admissions, or not one of the 16. The sorting of such codes was completed for patients across service lines.
For example, a patient with “uncontrolled diabetes” as her or her primary diagnosis on admission to a hospital’s endocrinology services would be accounted for in the ”Endocrine” service line as part of the “% ACSAs” population. A patient without an ACSA-related diagnosis would be accounted for as part of the “% All Other Cases in the Service Line” population.
If progress is being made through early care management efforts, the utilization drops in the patients with ambulatory care sensitive admissions (“% ACSAs”) should be greater than those of the patients in the “All Other Cases in the Service Line” population.
The nine services lines where this is indeed the case are outlined with a dotted line in Table 3, which provides the results of the analysis.
For example, discharges of patients with ACSAs fell 12 percent from 2010 to 2012 in the Endocrine service line, while discharges of non-ACSA patients fell only 1.4 percent. Similarly, discharges of patients with ACSAs in the Cardiology, Interventional line fell 23.5 percent while discharges of patients without ACSAs dropped 12.7 percent.
Table 3. Change in ACSA Utilization Drops Versus All Other Cases by Service Line: 2010-2012
Sources: Kaufman, Hall & Associates, Inc. analysis based on proprietary market and client data; U.S. Census Bureau Population.
These data signal that early care management efforts are lowering ambulatory care sensitive admissions in the greater Chicago market, suggesting that doctors and hospitals have started to change the way they care for patients. Improved care coordination and management is successfully keeping patients with chronic conditions out of the hospital, likely moving their inpatient-centric care to a more focused outpatient and home-centric model, as explored later.
What the Data Show: Use of Accountable and Risk-Based Models Is Having an Even Greater Positive Impact
Research next focused on whether care provided under an accountable-style model, which aligns hospital and physician incentives through performance-based arrangements and risk-based reimbursement, is starting to achieve even better results than care under traditional models. Might the new models be accelerating improvements in admissions and lengths of stay, for example?
The results of such an analysis show that ACO-style care outperformed traditional care in reducing avoidable hospital admissions and shortening lengths of stay (Table 4). Discharges of adult patients with ambulatory care sensitive conditions treated under a traditional care model declined 3.8 percent while length of stay declined 2.4 percent. Drops were higher with ACO-style care: discharges declined 6.3 percent and length of stay declined 3.9 percent.
In both cases, the ACO model was achieving better results than the traditional care model, providing early evidence that hospitals and doctors working under accountable care principles are more successful in keeping patients with chronic conditions out of the hospital, and shortening hospital stays when hospitalization is required.
Table 4. Comparative ACSA Utilization Drops, Traditional Care vs. ACO-Style Care: 2010-2012
Source: Kaufman, Hall & Associates, Inc. analysis based on proprietary provider data
What the Data Show: Given the Combined Impact of Trends 1-4, What the Future Might Hold?
We examined what might be ahead for the greater Chicago area given confirmation of the four trends. The regional market still includes a significant portion of inpatient admissions that might be eliminated through proper ambulatory care. These admissions thus represent “vulnerable” admissions for the area’s hospitals.
As noted in Table 5, approximately 139,000 patients with ACSCs and approximately 96,000 medical/surgical patients with one-day stays were admitted to Chicago-region hospitals in 2012. If these 235,000 “vulnerable” inpatient cases were eliminated, the greater Chicago hospital market would experience a 24 percent loss of inpatient discharges and a 15 percent reduction of average daily census for its 71 hospitals.
In other words, the region could lose up to another 1,900 inpatients per day. This provides an early signal that the greater Chicago region, and by analogy many other regions in the nation, are likely not to need as many hospital beds or hospitals going forward. Such dramatic change would require the hospital industry to aggressively “right size,” while moving the focus of care away from the four walls of the hospital. The long-term implications of all of this are equal only to the boundaries of the reader’s imagination.
Table 5. Projected Impact on Greater Chicago Market of Losing “Vulnerable” Admissions
Source: Kaufman, Hall & Associates, Inc.
The results of the greater Chicago area study indicate that the inpatient-centric health care model is declining in the region. The depth of the inpatient utilization drop suggests that structural factors are at work in changing health care provision in the region, not solely recession-linked factors. Examination of admission data for patients with ambulatory care sensitive conditions suggests that doctors and hospitals are changing the way they care for patients with chronic conditions, likely using intensive medical management to keep such patients out of the hospital. Accountable and risk-based care is having a statistical impact above and beyond intensive medical management.
The Chicago regional study indicates that there is indeed early evidence that the transformation agendais taking hold.