A new study on Medicare patients dying soon after emergency department discharges raises questions about staffing and treatment at rural hospitals and other providers who are under pressure to reduce health care costs.

More than 10,000 Medicare patients who do not have life-threatening illnesses die each year in the US within seven days of being released from emergency departments, according to the study, published in the BMJ. Those hospitals with the lowest inpatient admission rates, often hospitals in rural areas, had much higher rates of unexpected deaths.

The study’s lead author said that while the data reflect a fraction of Medicare patient deaths, the finding raises questions about the adequacy of hospital resources in rural and underserved areas and whether the US government’s quest to cut costs — and reduce inpatient admissions from ERs — is also cutting out essential care. 


“There’s no doubt there’s a lot of unnecessary hospital admissions, but this study suggests there’s also avoidable harm from sending people home that shouldn’t go home,” said Dr. Ziad Obermeyer, an emergency medicine physician and professor at Harvard Medical School.

Under the Affordable Care Act, hospitals are under financial pressure to deliver care more efficiently and reduce unnecessary admissions that drive up costs. That has encouraged hospitals to explore alternatives to admitting patients from the ER, such as monitoring them remotely or providing more care at home or in outpatient settings.

Dr. Rade Vukmir, a fellow of the American College of Emergency Physicians, said the study “definitely illustrates that there is a problem.” He said that “admission and discharge strategies” used by Medicare and private insurers influence decision-making, even though most doctors serve as strong advocates for their patients. Manpower and technical resources also vary widely among providers, he said, as do the medical problems that afflict their patients.

“The study brings forth that providing … proper resources and pathways to deliver care will generate the best outcomes,” Vukmir said. “That process should be cooperative with health care providers, insurers, and patients.”

The study examined care delivered to about 16 million Medicare patients between 2007 and 2012. It looked at generally healthy patients, excluding those with life-limiting conditions as well as patients living in nursing homes or undergoing hospice care.

Among discharged patients, about 0.12 percent died within seven days, equating to just over 10,000 deaths annually. Heart attacks and associated cardiovascular problems accounted for the most common cause of death among patients.

The rate of admissions from the ER varied widely among hospitals. The researchers found that hospitals in the lowest quintile of admission rates discharged 85 percent of their patients, compared to 44 percent for those in the highest quintile. 

Hospitals with low admissions had a much higher rate of unexpected deaths — 3.4 times that of hospitals with the highest rate of admissions. Obermeyer noted that the disparity in outcomes was not explained by the level of illness in the hospitals’ patient populations, as the high-admission-rate hospitals tended to see sicker patients.

“It doesn’t seem that the deaths are due to random chance,” he said. “There is something different going on in those low-admission-rate hospitals.”

That difference, he said, is not necessarily attributable to provider error, however. A wide array of factors could be influencing the outcomes, such as limited staffing at poorer hospitals and fewer transportation options in rural areas to allow patients to get follow-up care.

Part of the issue could also be related to protocols related to specific conditions. In cases where patients were diagnosed with chest pain, fewer unexpected deaths occurred, the researchers found. But death rates were higher among patients sent home with confusion, shortness of breath, or those diagnosed with pneumonia who were deemed healthy enough to be discharged.

Obermeyer also said that one of the leading causes of unexpected deaths among the Medicare patients studied was narcotic overdoses. “Those were largely in people who came in with musculoskeletal issues — back pain and other injuries,” he said. “We’re finding the signature of that problem” among Medicare patients as well.

Leave a Comment

Please enter your name.
Please enter a comment.

  • My beloved 88 year old father lost his life when he was misdiagnosed in the ER. Many diagnoscs and Labs were Pedro.

  • This article mentions a push by the ACA limit admissions as a plausible cause, however the study began prior to the ACA. The truth is, there have been stringent admission criteria in place since at least 2004 when I became a physician, and that has nothing to do with Obama. Shame on you for trying to throw a political opinion into medical article without identifying it as such.
    I currently work in a rural ED, and I can tell you it’s actually much easier for me to admit a patient to a rural hospital. Rural hospitals are subsidized differently, and not subjected to DRGs.

  • Having lived in a rural underserved area all my life, this comes as no surprise. An interesting comparison would be to graph this out over the past 20 years or so, to see the pattern that emerges. My money is on significant escalation over the past 4-6 years due to the push from the federal government i.e., Medicare, for pts to have to meet ‘criteria’ for admission or else be sent home or pay for their stay themselves. They will not establish a ‘formal’ set of criteria but expect the provider to use one anyway. At first, it was confusion with a UTI that was no longer covered for inpt stay, then it was hyperglycemia, then it was hypertension, then it was pneumonia/COPD exacerbation, and now it is moving to TIA, renal failure, uncontrolled pain, etc. And what are these syndromes?? the very things that happen to aging elderly who live in poverty in scarcely served areas with inadequate resources to meet their rising healthcare needs. So they get sicker, find it harder to get any assistance, and go home to die ‘unexpectedly’ because Medicare certainly did not want that to happen to them. Nor did Medicare want to pay to take care of them…. Can you spell ‘conundrum’??

  • Rural areas on lower budgets have to hire lesser qualified doctors for the ER, sometimes only a family practitioner. Even with someone that has ER training, they may not have had good training or a really good internship. Then when the ER needs bed space I imagine the less profitable patients are going to be released and moved out faster. Medicare like Medicaid is not going to pay for the better more accurate lab tests and diagnostics always. Most ER’s are only concerned about stabilizing the patient’s vital signs, and then it is “out the door”. If you need further diagnostics and treatment it is up to the patient to follow up with a referred doctor or their own primary care doctor. There are many factors to consider.

    • Hi. I’m “only a family practitioner” that works in a rural ED. The actual reason that family doctors staff rural EDs is because in rural EDs, the ED doctors often admit and care for the patient during their first night. We might deliver babies if the obstetrician is snowed in. And we do a TON of mental health care in our rural ED. So a family doctor is often a more appropriate choice to staff a rural ED, where there is more geriatrics, psych and primary care and less trauma/gun violence/etc. Family physicians go to the same medical schools and do a 3 year residency as well.

  • This is another big data study with more to come that reinforces the importance of clinical interventions – known to be of small importance in overall outcomes. Like the previous male vs female physician article, it is another lazy generalization and a tyranny of the aggregate as illustrated by astute minds starting with those that debunked race vs intelligence to the male female study as indicated by a different Dr. Jha http://www.kevinmd.com/blog/2017/01/deadly-risk-treated-male-physician.html If you want to understand how the US has diverted hundreds of billions for medical error focus with no chance for improvement because people, behaviors, environments, situations, genetics, local resources, and social fabric dominate health outcomes – then you must understand statistical manipulations and limitations. Once again a study is published comparing apples to oranges with too much variation between than within.

  • I live in a rural area. I’m 52 and something happened,my husband found me unresponsive, I couldn’t talk or walk when he finally got me to come to. I waited to go to the er cause they are rude, but after a week and 2 grand Mal seizures I thought, probably better. They did a head CT, said I was fine to go home as it was a subdural hematoma, but, even tho I couldn’t stand, walk, or talk cohesively, hey, it’ll get better. My life was saved when a random nephrologist walked by, asked if they did a UA.no….he wanted one Stat and my kidneys were totally shut down. I spent a month on hemodialysis, but man rural ers suck

  • The ACA has encouraged hospitals to explore alternatives to admitting patients? Hold it right there. There is inpatient care, and there is outpatient care. A patient either requires inpatient care, or they do not. If hospitals are seeking ways to deliver inappropriate levels care in order to save a buck, there should be legal action against them. Shut them down.

    Health insurance plans have been adding restrictions and motivating providers to avoid unnecessary inpatient admissions since the dawn of HMOs and managed care over forty years ago. It is not an onerous new situation created by the ACA. There is nothing new about it.

    Your take-away is that the problem might be the rural location of patients and hospitals with this dilemma? That, and the ACA? Maybe it’s for-profit hospitals that prioritize money over appropriate patient care.

    • Mike you are so spot on…several hospitals have forced my dad out and he is very ill with brain stem stroke, paralysis, (they gave him pneumonia too), A-Fib now, and so much more, he has been to 4 hospitals since Feb 14th the day this happened…my parents have Medicare and pay a lot for Anthem Blue Cross supplement as well – he never gets to stay longer than a week and the day he is admitted they try to force him out saying he is fine – now they took him off antibiotics for pneumonia, Lasix for the removal of fluid in his lungs..and trying to for PT on him daily and say he is fine – his voice is so weak and he says he feels terrible – has lost over 30 lbs and they nearly killed him with the wrong med in a breathing treatment – now they put him on steroids, never take chest x-rays and I hear crackling every time he speaks to me over the phone…yet they all say he is good and always force him to leave…none of this makes any sense except I feel they are trying to euthanize him…I guess nowadays they have a license to kill – he is 75 and was in great health his entire life and now they think he needs to go by by because he has lived long enough…it is evil, greed based and a form of population control.

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy