Aiming Higher: Results from a State Scorecard on Health System Performance, 2009

October 8, 2009

Authors: Douglas McCarthy, M.B.A., Sabrina K. H. How, M.P.A., and Cathy Schoen, M.S., The Commonwealth Fund

Joel C. Cantor, Sc.D., and Dina Belloff, M.A., Rutgers University Center for State Health Policy

Prepared for the Commonwealth Fund Commission on a High Performance Health System

Feature

State Scorecard Map Use this map to view state-specific rankings and results, and to view the number of lives and dollars each state could save by achieving benchmark levels of performance. Also check out the comparison tool to select years, performance indicators, and states and then generate custom tables and bar charts.

 Key Findings and State Variations by Dimension of Performance

Access

  • For the most part, performance on the State Scorecard's health care access indicators failed to improve from 2003 to 2008. Gaps in health insurance coverage between the top and bottom states remained wide, with uninsured rates for children ranging from 3 percent to 20 percent and rates for adults ranging from 7 percent to over 30 percent.
  • Since the start of the decade—from 1999–2000 to 2007–08—the number of states with high uninsured rates (23% or higher) for nonelderly adults rose from two to nine, while the number with low rates (under 14%) dropped from 22 to 11. In contrast, the number of states with high children's uninsured rates (16% or more) declined from nine to three during this time, reflecting federal support of CHIP.
  • From 2004–05 to 2007–08—the time span represented in the State Scorecard's coverage indicators—trends in coverage were negative in most states for adults and in two of five states for children (Exhibit 3). That this was true even before the severe recession underscores the challenge that states face in ensuring coverage for children and adults in the absence of federal action. 
  • Massachusetts, which had only begun to implement its universal health insurance program during the period covered by the State Scorecard, had the greatest increase in coverage for adults and made gains in coverage for children between 2004–05 and 2007–08, becoming the top-ranked state for the coverage of both adults and children as well as the top-ranked state for access to care overall.
  • Across states, the percentage of adults who reported going without health care because of the cost is  closely associated with insurance coverage and is up to three times greater in states with the highest uninsured adult rates than in states with the lowest uninsured adult rates (19% vs. 7%).

Prevention and Treatment

  • Almost all states improved on process indicators of the quality of hospital treatment (48 states by 5% or better) and nursing home care (38 to 51 states by 5% or better across three indicators). On a set of hospital clinical quality measures, the rate in the five lowest-performing states in 2007 had risen to the level of the five highest-performing states three years earlier. On an expanded set of measures to prevent surgical complications in hospitals, the variation in performance among states narrowed by half.
  • Despite a 30 percent narrowing in state variation on nursing home care, the range has remained wide, with a two-to-five-fold variation between the top-five and bottom-five states.
  • States have failed to match these gains when it comes to the quality of ambulatory care; even in the best states, quality continues to be well below standards. The percentage of adults age 50 and older receiving all recommended cancer screenings and immunizations ranged from a high of just 53 percent in Delaware to a low of 35 percent in Oklahoma. Only about half the states improved by 5 percent or more. The proportion of diabetic patients receiving three basic services to prevent disease complications varied from two-thirds in Minnesota to one-third in Mississippi. The rate worsened or failed to improve significantly in 24 of 42 states for which data were available. Exhibit 5S
  • More than one-quarter of young children in the bottom-five states did not receive timely preventive medical and dental visits and recommended vaccinations, and in the bottom five states more than half of children who needed mental health care did not receive it. Top states, in contrast, achieved vaccination rates of 90 percent and preventive visit and mental health care rates that were 20 and 30 percentage points higher, respectively. Only nine states improved substantially (by 5% or more) on vaccination rates, while 10 lost ground. And only 21 states improved substantially on child mental health care, while 12 declined substantially.
  • In 48 states, there was no appreciable change in the percentage of adults who had a usual source of care—not surprising, given the lack of improvement in health insurance coverage. The proportion of children who received effective, patient-centered care coordination from a primary care medical home ranged from more than two-thirds (69%) in New Hampshire to less than half (45%) in Nevada.
  • Across all states in 2007, there was a divergence in how Medicare patients rated their care, with provider interactions rated more highly and overall care experience rated more poorly than in 2003. (These trends should be interpreted with caution, however, because of changes in survey administration.) More data are needed to judge whether these shifts are an anomaly or represent an enduring change in patients' experiences.

 Potentially Avoidable Use of Hospitals and Costs of Care

  • Hospital admissions among Medicare beneficiaries for ambulatory care sensitive conditions improved (i.e., declined) in a majority of states, although rates fluctuated from year to year—illustrating the importance of looking at long-term trends when assessing improvement. Declining hospital admissions may reflect patients’ improved access to medications for chronic conditions, or incentives provided to manage such conditions better. (The way hospital administrators code diseases for reimbursement purposes also has changed, potentially influencing trends for some conditions.)
  • Hospitalization rates for pediatric asthma declined across most of the 32 states that reported data in both time periods. Yet despite some narrowing in state variation, rates were three times greater in the highest-rate states compared with the lowest-rate
    states, indicating that an opportunity exists for further reductions to benchmark levels. 
  • Hospital admissions and 30-day readmissions among nursing home residents increased by 8 percent and 11 percent, on average, between 2000 and 2006, with negative trends seen in a significant majority of states. Rates went up by 5 percent or more in 29 to 37 out of 48 states for which trend data were available for these two indicators. Rates in the worst-performing states (i.e., those with the highest admission rates) were two to three times higher than in the best-performing states, and the ranges widened.
  • The 30-day hospital readmission rate among all Medicare beneficiaries either failed to improve or increased across most states from 2003–04 to 2006–07, with continued sharp variation across states. Readmission rates in 2006–07 ranged from lows of 13 to 14 percent in the best-performing five states (Oregon, Utah, South Dakota, Nebraska, and Idaho) to highs of 21 to 23 percent in the worst-performing five states (Louisiana, Arkansas, West Virginia, Nevada, and the District of Columbia). Improvements in some states, as well as recent experience in some hospitals, suggest that all states could improve if incentives were better aligned to support care transitions and improve quality of care.
  • Medicare fee-for-service spending per person grew by 6.5 percent per year from 2003 to 2006 for the median state—more than twice the rate of general inflation. The gap in per-beneficiary spending between the highest- and lowest-cost states widened. By 2006, average per-beneficiary spending in the five most costly states was 50 percent higher than average spending in the five least costly states ($9,439 vs. $6,027).
  • Employer premiums (including the employee shares) for a single individual rose an average of 4.5 percent per year in the median state from 2004 to 2008; average annual increases ranged from 8.5 percent in Utah to less than 1 percent in neighboring Nevada. Premiums bought less coverage, as annual deductibles and cost-sharing went up during this time. By 2008, average premiums in the highest-cost states were 30 percent higher than in the lowest-cost states ($5,056 vs. $3,904).

Equity 

  • In most states, there are wide "equity gaps" in performance on access and quality indicators based on income level, health insurance status, and race/ethnicity. Disturbingly, in the majority of states, these equity gaps widened over time. Equity gaps were most likely to worsen for access and coordination of care. (Equity gaps measure the difference between the experiences of vulnerable population groups in each state and the national average for a total of 24 equity comparisons, only 17 of which had data that could be compared over time.)
  • Only eight states—Connecticut, Delaware, New York, Utah, Wisconsin, Oregon, Montana, and Michigan—saw the equity gap narrow, with the vulnerable group improving on more than half of equity indicators and improving relative to the national average. The greatest gains in equity across states were in mortality amenable to health care. Yet even on this indicator, in only half the states was the gap reduced for blacks relative to the national average; moreover, within all states, white–black differences remained large.
  • In those states ranked at the top for equity overall, the gaps between vulnerable groups (low-income, uninsured, and minority) and national averages tended to be smallest. Six of the 13 top-ranked states—Maine, Vermont, Rhode Island, New Hampshire, Delaware, and Iowa—scored in the top quartile on this dimension for all three vulnerable groups. Conversely, five of the 13 states in the bottom quartile of the overall equity rankings score in the bottom quartile for all three groups.
  • In some higher-performing states, traditionally disadvantaged groups reported quality of care that exceeded the national average. For example, the percentage of low-income diabetic patients receiving basic recommended services was higher in 11 states than the national average for all diabetics (44%). In a few instances, the care received by vulnerable groups was on par with that received by the typically advantaged group.
  • The performance patterns for the equity dimension indicate that it is possible to close gaps—and raise the floor on performance—for vulnerable groups in comparison with national averages.

 Healthy Lives

  • Rates of mortality for conditions amenable to health care improved in most states from 2001–02 to 2004–05, but wide regional variation persists. Average death rates were 68.2 per 100,000 persons in the lowest-rate states (Minnesota, Utah, Vermont, Colorado, and Nebraska) compared with 135.4 per 100,000 in states having the highest mortality rates (Mississippi, Louisiana, Arkansas, and Tennessee) and the District of Columbia.
  • Looking just at white mortality rates for conditions amenable to health care, the spread across states is also wide, ranging from a low of 61 deaths per 100,000 in Minnesota to a high of 111 deaths per 100,000 in West Virginia.
  • In all states, potentially preventable deaths among blacks are considerably higher than among whites. Even in the five states with the lowest rates for blacks on this indicator, there is still an average of 92.0 deaths per 100,000 blacks, which exceeds the national average for whites. Preventable deaths among whites have gone down in most states, yet some states have had increases in black mortality, resulting in widening disparities.
  • State variations in breast and colorectal cancer narrowed between 2002 and 2005, as bottom-ranked states improved faster than states with the lowest cancer mortality rates. Notably, rates of colorectal cancer deaths in the bottom states are now at the median state rate observed in 2002.
  • Few states experienced appreciable improvement in their infant mortality rates from 2002 to 2005. Signaling the need for urgent action, several states with already high rates experienced further increases, reaching an average of more than 11.0 deaths per 1,000 births—more than double the rates of states with the lowest infant mortality (4.5 to 5.1 deaths per 1,000 births). 
  • Smoking rates among adults declined by 5 percent or more in the majority of states from 2003–04 to 2006–07. Yet more than one of four adults smoke in high-rate states, compared with just one of 10 in Utah, the lowest-rate state.
  • Obesity is a growing concern across states. As of 2007, at least a quarter of children ages 10 to 17 are overweight or obese in all but three states (although these states are not far behind). And one of three children is overweight or obese in 17 states, with regional patterns closely tracking mortality amenable to health care.