Category Archives: Health Services Research

Fact Check: the Colorado Case for Repeal of the Affordable Care Act (and replacement with the Affordable Health Care Act) is a False Narrative

The Affordable Health Care Act (AHCA or “Trumpcare”) will cut 24 million from insurance by 2026 including 410,600 from Colorado (use this interactive feature to see how many will be impacted in your state).  Colorado’s own Senator Cory Gardner has expressed his reservations about the 14 million projected to lose Medicaid, describing Medicaid as providing “access to life-saving health care services” for the 1.3 million Coloradans—and 72 million Americans—who get their insurance through Medicaid, including the Medicaid expansion part of the Affordable Care Act (ACA). Given the ominous projections about the AHCA’s impact compared to staying with the ACA, how did Republican legislators end up with such a destructive bill after 8 years of decrying the disastrous results of the ACA? The unfortunate reality of what the ACA-to-AHCA transition would mean highlights the accomplishments of the ACA as it is on the verge of elimination: it has improved healthcare access and outcomes for millions of Americans.

The mythology on which the Republican members of congress have based their anti-ACA rhetoric is exemplified by the talking points used by our Colorado legislators. Senator Gardner has reiterated his anti-ACA talking points in his two recent tele-town halls.   As he has before, Senator Gardner gave three arguments for why the ACA must be repealed:

  • Health insurance premiums in Colorado are rising because of the ACA
  • Coloradans lost their health insurance because of the ACA
  • Coloradans can’t access their doctor because of the ACA

Each argument bent the truth, casting the ACA as causing the problem rather than what the data shows it is: part of the solution. If the ACA did not cause the problem, repeal is not the solution.

  • Premiums are rising.  True, but we don’t know how much is due to the ACA and the rise is exaggerated. Premiums were rising before the ACA. This year’s rise brought lower-than-predicted premiums into line with the 2017 rates predicted by the CBO in 2009. The 400% rise cited by Senator Gardner in both tele town halls– of “people whose premium rose from $300 to $1500 a month”—grossly overstates the average rise of 20.4% in Colorado–an increase that primarily impacts those who buy individual policies—about 3% of the population–and is often offset by subsidies.
  • Hundreds of thousands of Coloradans lost insurance: Not true.   By “lost insurance”, the Senator is referring to “churn”—transitioning between plans, sometimes a result of canceled plans—and not about a net loss in insured people in Colorado. The rate of churn was not increased by the ACA. Most cancelations—about 250,000—involved limited benefit plans that were not ACA-compliant, and thus had to be either made ACA-compliant or dropped when the ACA was enacted.  Despite the churn, Colorado has nevertheless seen a net increase of over 530,000 in the number insured. In fact, Coloradans continue to enroll in plans in record-breaking numbers, with a 12% increase seen over the same time last year in the 2017 open enrollment period.
  • People can’t see their doctors because of high deductibles. Not true.  The 2015 Colorado Health Access Survey shows that, since the ACA was passed, increased coverage has meant increased access. Coloradans have steadily increased their rate of preventive health visits by 7% (from 61.9% in 2013 to 66.1% in 2015) and have lowered their rate of skipping doctors’ visits due to costs by 15% (from 12.3% in 2013 to 10.4% in 2015).

Although rising premiums, churn and barriers to accessing doctors were not eradicated by the ACA, this review of Colorado data shows that the ACA is part of the solution, not part of the problem. To do what is best for Coloradans, Senator Gardner needs to step away from his false narrative about the ACA and address real issues reflected in Colorado data.  He has taken a first step in raising concerns about threats to Medicaid. There are ways to strengthen the ACA to address some of the real issues in healthcare, but to repeal it based on a false narrative disseminated by Senator Gardner and his Republican colleagues would be a tragedy.  Let’s look at the facts, and work together to make the system better, for Colorado and the rest of America.

 

 

Advertisements

The Affordable Care Act is Good for Colorado (and Repeal is Not)

The effects of the Affordable Care Act of 2010 in each state depend on various factors, such as the number of uninsured individuals in the state and the governor’s receptiveness to the law’s provisions. In this post, I focus on the benefits of the Affordable Care Act (ACA) in Colorado, primarily from the perspective of my job as an emergency department (ED) physician, in support of my argument that the ACA must not be repealed–instead, I recommend that it be strengthened. In a prior post, I summarized estimates of the impact of the GOP’s proposed ACA repeal/replace plans on coverage and consumer costs.

Effects of the ACA on Uninsurance Rate in Colorado

In Colorado, the number of uninsured individuals declined by 25.5%, from 729,000 individuals in 2013, or 14.1% of the population, to 543,000 individuals in 2014, or 10.3% of the population. colorado-uninsurance-rate

Colorado was one of 32 states (including DC) that had expanded Medicaid as of 1/1/2017. In fiscal year 2014, annual Medicaid enrollment in Colorado increased by 26.1%. Average monthly enrollment increased 59% from 783,420 to 1,244,031 pre-ACA to post-ACA.

Drivers of Healthcare Cost and Utilization from the Emergency Department Perspective

Healthcare access and coverage–so improved under the ACA–drive a lot of what happens in the emergency department (ED). We can think of this in 3 categories:

  1. How the ACA has impacted who comes into the ED (and the hospital).
  2. How the ACA has decreased the amount of services we ED physicians provide to patients while they are in the ED (and hence the cost).
  3. How the ACA has impacted what happens at the end of the ED visit (should I stay or should I go?)

drivers-of-ed-utilization

How the ACA has impacted who comes into the ED (and the hospital)

In Colorado, the number of ED visits has declined with ACA implementation. This varies state, by state, but in Colorado, we have seen a clear decline in the rate of ED visits per insured person–a trend driven by the expansion of Medicaid insurance.

decline-ed-visits-all-ages-co

The trend is also apparent when we restrict the analysis to children only:decline-ed-visits-children-co

(This and other data are available from the Colorado All Payer Claims Data summary tools, at Colorado Medical Price Compare, available at https://www.comedprice.org/#/reports )

The decline in population rate of ED visits among Medicaid patients did not result in a net decline in the number of ED visits–in 2014, the overall number of Medicaid ED visits went up, largely because more patients were on Medicaid. But, among those on Medicaid, during the ACA rollout, population rates of ED visits went down. As expected, the number of un-insured ED visits declined in Colorado as the number of Medicaid visits rose.

medicaid_ed_100814-01

Increasingly, patients are more likely to visit their medical home (primary care provider) because they are more likely to have one, thanks to the ACA.  When we do see patients in the ED, patients are more likely to have had started treatment before coming in (saw or called their medical home). Diverting more ED visits to the medical home is good for many reasons, not the least of which is cost. cost-ed-vs-pcp

The decline in ED visits is also a Colorado Medicaid performance metric for Colorado’s seven Regional Care Collaborative Organizations (RCCOs)–RCCOs received payment incentives if they achieved a 3% decline in ED visits per patient in their RCCO. Although most RCCOs did not achieve this decline, all achieved some decline in 2016:rcco-level-ed-visit-reduction-medicaid

How the ACA has decreased the amount of services we ED physicians provide to patients while they are in the ED (and hence the cost)

If we know a patient in the ED has no coverage (no financial access to primary care), the ED visit costs more for many reasons, including:

  • The patient is more likely to have tests done in the ED because there is no primary care provider to order tests and follow test results
  • The patient is more likely to have subspecialty consultations instead of office follow-up visits with primary care providers
  • The patient is more likely to get more conservative (more aggressive) treatment in the ED, rather than waiting a day or two to see if the patient really needs more aggressive testing/treatment, because there is no medical home to provide that testing/treatment in a day or two.

 

How the ACA has impacted what happens at the end of the ED visit

If we, as ED physicians, know a patient has financial access to primary care, we are more likely to discharge them home rather than admit them to the hospital. This is reflected in the decrease in hospital admissions relative to the rate of ED visits among all ages:

decline-ed-admit-rate-all-ages-co

The same trend is seen among children only:decline-ed-admit-rate-children-co

As an example of how this works, one of the most common reasons young children are admitted to the hospital in Colorado (and nationwide) is for a common lower respiratory viral syndrome called bronchiolitis. Several of my colleagues at Children’s Hospital of Colorado have performed the research (here and here) to show the circumstances under which a child is safely discharged home on home-oxygen therapy. This is cost-saving in that hospitalizations are expensive, and also result in many missed days of work for parents. The criteria for discharge to home (rather than hospital admission) in our hospital’s evidence based guideline include access (within 24 hours) to a medical home for a follow up check:

bronchiolitis-home-oxygen-ccg

Before the ACA permitted so many more access to a medical home, patients with common conditions like bronchiolitis were more likely to be admitted. Discharging more to home means fewer inpatient stays:

decline-ip-visits-all-ages-codecline-ip-visits-children-co

Thus, for Colorado consumers, the ACA has helped improve coverage, thereby reducing the per-insured rate of ED visits, the per-ED-visit rate of inpatient admissions and the overall rate of inpatient admissions, both for adults and children. It has primarily done this by allowing more Coloradans to access care where they should be able to: in a medical home. By reducing these costly forms of healthcare utilization, EDs are less crowded for those who truly need ED care. By reducing the proportion of uninsured visitors to the ED and hospital, the ACA has also decreased the amount of uncompensated care, permitting more hospitals to remain in business.

What we miss when we look only at the cost of healthcare coverage

With the Republicans on the verge of dismantling the Affordable Care Act and Medicaid, much of their focus has been on how much healthcare insurance “costs”. I thought it would be timely to revisit a 2015 analysis in the American Journal of Public Health, “Considering Whether Medicaid is Worth the Cost: Revisiting the Oregon Health Study.” (full text) The investigators performed a cost-effectiveness analysis using data from the Oregon Health Study (OHS), the experiment that began in 2008 when the state randomly selected uninsured participants to apply for Medicaid coverage creating a randomized controlled trial of a social policy. They concluded that Medicaid is in fact a cost-effective program.

The authors’ cost-effectiveness analysis found that the observed benefit of providing Medicaid was $62,000 per quality-adjusted life-year (QALY) gained. As a reference point, placing smoke detectors in homes provides a benefit of $210,000 per QALY.

The study reminds us that health benefits are best understood broadly and over the long term, rather than through a narrow focus on one year’s “cost” in the cost-effectiveness balance.

Retail Urgent Care Clinics Do Not Decrease Emergency Department Visits

A study published online today in Annals of Emergency Medicine (“Association Between the Opening of Retail Clinics and Low-Acuity Emergency Department Visits”) demonstrated that, contrary to expectations, retail clinics had little effect on rates of low-acuity visits to nearby emergency departments (EDs). This contradicts the popular theory that retail clinics would reduce ED visits.  A 2015 report “Building a Culture of Health: The Value Proposition of Retail Clinics“, found that consumer use of retail clinics was rising, with consumers citing convenient hours/location and perceived lower cost as the main reasons for choosing retail clinics for care. Another study projected that 13% to 27% of US ED visits could be treated in retail clinic settings, with an estimated cost savings of $4.4 billion.

However, this theory did not align with the new study’s findings. During the study period (2007-2012), the number of retail clinics grew from 130 to nearly 1,400 in the 23 states studied. During the same period, the rate of retail clinic penetration – in other words, the proportion of each ED’s catchment area that overlaps with a 10-minute drive radius of a retail clinic – more than doubled (8.1 to 16.4).

In the 2,053 emergency departments in these states, more retail clinics did not correlate with fewer low-acuity ED visits; however, increased retail clinic density did correlate with a slight reduction in the subset of low-acuity visits, albeit only among the privately insured. As an example, for an ED with 40,000 annual visits and 50% private insurance, there would be approximately 8 fewer visits per year for each 10% increase in retail clinic density.

 

The accompanying editorial offers three theories as to why retail clinics increase health care use:

  1. They meet unmet demands for care. The demand for episodic acute care often exceeds the supply of physicians or facilities in many communities, and retail clinics attempt to meet this need.
  2. Motivations for seeking care differ in EDs and retail clinics. A patient who might otherwise not seek care for a cough would stop by the neighborhood retail clinic but would not invest the time and travel to schedule an appointment with their medical home or risk a long ED wait.
  3. Groups of people who are more likely to use EDs for low-acuity conditions–particularly patients withMedicaid and ED super-users–do so because they have little access to other types of care, including retail clinics. Retail clinics follow the money–they locate in affluent areas in which few Medicaid patients live. Also, Medicaid plans often do not cover retail clinic use.  ED superusers commonly have Medicaid insurance and infrequently live in affluent neighborhoods, and often have complex medical/mental health conditions that cannot be cared for in retail clinics.

As an ED physician, my primary concern is that these clinics introduce added fragmentation into an already fragmented health care system. I treat many patients who have sought care for an illness from 1-2 different retail clinics in the past week, without seeing their regular doctor, and then end up in my ED with no record (other than the patient’s recollection) of what happened at those clinics and why. I do my best to reconstruct the story–the antibiotics given for the sore throat on Monday at one retail clinic where they did not check a strep test, followed by a visit to a second retail clinic on Wednesday for the ongoing sore throat where they sold them different antibiotics from the co-located pharmacy and where they did check a strep test and some type of blood test that were both negative–and make a coherent decision based on guessing why each retail clinic did what it did and how the patient is today.

A second concern is illustrated by my patient anecdote. Retail clinics tend to compete with primary care providers for patient business, and, in so doing, they prescribe more antibiotics to meet patients’ expectations.

If patients continue to vote with their feet, using convenient retail clinics in lieu of their primary care providers, my hope is that we find a way to better integrate them into the healthcare system so that we are all working collaboratively to best serve the patient.

 

Paid Sick Leave Lowers Costs

An Upshot column “The High Costs of Not Offering Paid Sick Leave” argues that employees and their co-workers may be better off with an incentive to take time off when sick. About 45 percent of the American work force does not have paid sick leave; that’s about 50 million workers. Families with less ability to afford unpaid time off are more likely to lack paid sick leave. According to a study in Health Affairs, 65 percent of families with incomes below $35,000 had no paid sick leave, while the same was true of only 25 percent of families with annual incomes above $100,000. Those without sick leave were farm ore likely to go to work sick (“presenteeism”), as well as to forgo seeking medical care for themselves or for an ill family member. Although expanding health insurance helps people pay for health care, it does nothing to help them afford to take time off to get it.

Another study in Health Services Research by a Cornell economist supports the theory that paid sick leave could reduce the spread of contagion. This study found that each week, up to 3 million U.S. employees go to work sick. Females, low-income earners, and those aged 25 to 34 years have a significantly elevated risk of presenteeism behavior.

The Upshot column presents the benefits of paid sick leave for one illness, flu, although morbidity and mortality from other contagious conditions would also be reduced:

Paid sick leave slows the spread of disease. Cities and states that require employers to offer paid sick leave — Washington, D.C.; Seattle; New York City; and Philadelphia, as well as Connecticut, California, Massachusetts and Oregon — have fewer cases of seasonal flu than other comparable cities and states. Flu rates would fall 5 percent if paid sick leave were universal. According to one estimate, an additional seven million people contracted the H1N1 flu virus in 2009 because employees came to work while infected. The illnesses led to 1,500 additional deaths.

Beyond reducing flu deaths, paid sick leave is associated with

Though a few cities and states mentioned above mandate employers provide paid sick leave, the Family and Medical Leave Act requires only unpaid sick leave be provided by employers with more than 50 workers. A new Obama administration rule is extending paid sick leave to ~300,000 private-sector employees working on government contracts starting on Jan. 1.  Policy makers should consider the potential public health implications of their decisions when contemplating guaranteed sick leave benefits.

Obamacare Succeeds in Expanding Insurance Coverage

Amidst news of the increasing premiums and decreasing choice of plans in some healthcare markets, the New York TimesUpshot column brings us this visual display of the impressive and varied impact of Obamacare on insurance coverage rates from 2013-2016. States that decided to expand their Medicaid programs saw much larger declines in their uninsured rates compared with those that didn’t. The proposed mechanisms for this effect were that Medicaid expansion

  • provided a new coverage option for childless adults below or near the poverty line
  • helped spur many people who were already eligible for the program to sign up
  • boosted enrollment in Obamacare’s marketplace plans.

The column analyzed the difference in insurance coverage expansion between states that did and did not take the Medicaid expansion:

screen-shot-2016-10-31-at-5-01-03-pm

AHRQ and CMS Public Reporting Measures Fail to Describe the True Safety of Hospitals

A new study from the Johns Hopkins Armstrong Institute for Patient Safety and Quality, published in the journal Medical Care, performed a systematic review and meta-analysis of two sets of safety measures used for pay-for-performance and public reporting The measures evaluated in the study are used by several public rating systems, including U.S. News and World Report’s Best Hospitals, Leapfrog’s Hospital Safety Score, and the Center for Medicare and Medicaid Services’ (CMS’) Star Ratings.

The two sets of measures evaluated are:

The investigators first performed a systematic review of all published medical research since 1990, looking for studies that addressed the validity of the HAC  and PSI measures.  They identified only 5 of these 40 safety measures with enough data in these prior studies to permit a pooled meta-analysis:

  • A. Iatrogenic Pneumothorax (PSI 6/HAC 17)
  • B. Central Line-associated Bloodstream Infections (PSI 7)
  • C. Postoperative hemorrhage/hematoma (PSI 9)
  • D. Postoperative deep vein thrombosis/pulmonary embolus (PSI 12)
  • E. Accidental Puncture/Laceration (PSI 15)

The investigators then performed a meta-analysis, pooling the results of all studies about the validity of each of these measures.  Their findings in the figure below show that in pooled studies (the diamond at the bottom of each lettered rectangle) only one measure–Measure E, PSI 15 (Accidental Puncture and Laceration)–met the investigators’ criteria for validity: a positive predictive value of at least 80% (indicating that at least 80% of the patients determined by the measure to have an accidental puncture or laceration truly had an accidental puncture or laceration.) Actual occurrence (reference standard) of each reported safety event was determined, in each individual study, by medical chart review.

Screen Shot 2016-06-02 at 6.12.26 AM

Measure D, PSI 9 (Postoperative hemorrhage or hematoma) came close to the 80% PPV threshold, with a pooled PPV of 78.6%.

Based on these findings, the investigators conclude that these measures, widely used for public reporting and pay-for-performance, should not be used for either purpose:

 CMS and others have created payment incentives based on hospitals’ performance for a variety of hospital-acquired complications, which are measured with the respective PSIs and HAC measures. Policy makers and payers have argued that the PSIs and HAC measures are good enough for reporting and pay-for-performance, whereas many providers believe they are not. Our results suggest that the PSIs and HAC measures may not be valid enough and/or have insufficient data to support their use for these purposes. This is especially true given the potential financial impact these pay-for-performance approaches may have on the narrow financial margins on which most hospitals function.