I can or can not get no satisfaction.

February 15, 2012 2:02 PM

How to Feed and Grow Your Health Care System. A landmark study has established that patient satisfaction is correlated with mortality - in the wrong way. The more satisfied, the greater mortality. What accounts for this dynamic? And what are the implications for healthcare costs and available political options? 'One of the primary findings itself raises concern—a 26% mortality excess among the most satisfied patients, an effect size that far exceeds that for all other, more immediate, study outcomes (eg, a 12% excess in hospitalizations).' Brenda E. Sirovich, MD, MS wrote a response to the study.

The study: The Cost of Satisfaction A National Study of Patient Satisfaction, Health Care Utilization, Expenditures, and Mortality by Joshua J. Fenton, MD, MPH; Anthony F. Jerant, MD; Klea D. Bertakis, MD, MPH; Peter Franks, MD was published online in Arch Intern Med. February 13, 2012. doi:10.1001/archinternmed.2011.1662

Conclusion from the study:

"Conclusion: In a nationally representative sample, higher patient satisfaction was associated with less emergency department use but with greater inpatient use, higher overall health care and prescription drug expenditures, and increased mortality."

Brenda E. Sirovich, points out that the dynamics of healthcare are responsible for escalating costs and worse health outcomes:

"Positive feedback systems abound in health care, for both physicians and patients. Diagnostically, almost any unnecessary, or discretionary, test (particularly imaging) has a good chance of detecting an abnormality; acting on that abnormality has an excellent chance of producing a favorable outcome (because a good outcome was already highly likely). Having obtained an excellent outcome, ostensibly owing to a test that was seemingly unnecessary, a natural reaction would be thereafter to perform (or undergo) even more discretionary testing on (or by) patients, with an increasingly negligible likelihood of benefit and greater risk of net harm.

Consider thyroid cancer: incidence of papillary carcinoma (by far the most common type) has tripled over a 30-year period, with an abundance of very small cancers that appear nonlethal.8 The excess cases almost certainly represent pseudodisease (destined never to cause symptoms during a patient's lifetime)—patients who cannot possibly benefit from having had their cancer detected, but can be, and likely are, harmed. However, in the eyes of the patient, her loved ones (and casual acquaintances), and her physicians, she was snatched from the jaws of a premature death by a vigilant physician who thought he felt something on examination or who inexplicably ordered thyroid ultrasound examination. The lesson learned, for all, will surely be to be increasingly vigilant in the future.

Ransohoff et al7(p665) explains, "The point is that . . . decisions for aggressive intervention—screening or treatment—may be positively reinforced when patients and physicians view the decisions from the perspective of an individual person."

Even if patients and physicians were to see through this illusion, overcoming the "more is always better" fallacy of health care remains an enormous challenge. In a recent survey published in the Archives,9 we found that nearly half of US primary care physicians believed that their own patients were receiving too much medical care, and they identified potent systemic incentives encouraging aggressive practice. Practicing physicians have learned—from reimbursement systems, the medical liability environment, and clinical performance scorekeepers—that they will be rewarded for excess and penalized if they risk not doing enough.

More aggressive practice, therefore, improves not only patients' perceived outcomes, but also those of physicians (reimbursement, performance ratings, protection against lawsuits), and the positive feedback loop of health care utilization is fueled at two ends."

'A positive feedback system is not in fact positive (ie, favorable)—it represents an unstable system, one that cannot control its own growth or demise. It is time that we, as a profession and as a society, take responsibility for controlling this unrestrained system, by working to overcome the widespread misconception that more care is necessarily better care and to realign the incentives that help nurture this belief.'

posted by VikingSword (20 comments total)

Wonder why they didn't do a comparison with similar metrics from other countries health systems?

Because, without that, it's just "Broken system is broken - and here's a different way it's broken".
posted by Pinback at 2:20 PM on February 15, 2012 [4 favorites]

This is a very timely study, given the increasing professional resistance to under-50 mammogram and prostate screening.
posted by mek at 2:54 PM on February 15, 2012 [2 favorites]

"[R]espondents in the highest patient satisfaction quartile (relative to the lowest patient satisfaction quartile) had lower odds of any emergency department visit (adjusted odds ratio [aOR], 0.92; 95% CI, 0.84-1.00), higher odds of any inpatient admission (aOR, 1.12; 95% CI, 1.02-1.23), 8.8% (95% CI, 1.6%-16.6%) greater total expenditures, 9.1% (95% CI, 2.3%-16.4%) greater prescription drug expenditures, and higher mortality (adjusted hazard ratio, 1.26; 95% CI, 1.05-1.53)."

This means, "patients who had higher satisfaction also had higher mortality". The summary makes it sound like, "systems with higher mortality rates produce higher patient satisfaction." It really matters which of these is true. If the first is true, a possible interpretation is that patients in high-mortality groups were satisfied because their life-threatening complaints were taken seriously, unlike someone who "only" has, say chronic pain. If the second is true, it implies that patients like to be "overtreated" and that the people who know best need to start rationing care.
posted by Ralston McTodd at 2:55 PM on February 15, 2012 [4 favorites]

Yeah... It's also possible that doctors are just more sympathetic with patients with terrible, life threatening conditions, while being dismissive of patients with more minor problems.
posted by delmoi at 3:01 PM on February 15, 2012 [3 favorites]

And patients with life-threatening conditions also are more likely to believe, often correctly that their lives were saved by their care.
posted by Ralston McTodd at 3:03 PM on February 15, 2012 [3 favorites]

I should have included "Adjusting for sociodemographics, insurance status, availability of a usual source of care, chronic disease burden, health status, and year 1 utilization and expenditures". I do think that "health status" can be subjective, and I'm not sure if they mean status at the time of the study or during the period treatment.
posted by Ralston McTodd at 3:06 PM on February 15, 2012 [1 favorite]

Yeah... It's also possible that doctors are just more sympathetic with patients with terrible, life threatening conditions, while being dismissive of patients with more minor problems.

&

And patients with life-threatening conditions also are more likely to believe, often correctly that their lives were saved by their care.

Please note:

"Adjusting for sociodemographics, insurance status, availability of a usual source of care, chronic disease burden, health status, and year 1 utilization and expenditures,[...]"
posted by VikingSword at 3:08 PM on February 15, 2012

I do think that "health status" can be subjective

Even if that is true, why would the "subjective" be biased in only one direction?
posted by VikingSword at 3:10 PM on February 15, 2012

Patient Satisfaction? Press Ganey Statistics.
posted by unliteral at 3:16 PM on February 15, 2012

They used the 12-Item Short Form Health Survey to determine health status. From what I can tell, and the information I can find online is sketchy, this asks patients questions mainly about if their physical or mental health problems are interfering with various activities in their lives (housework, climbing stairs, socializing). It sounds like non-life-threatening conditions, like chronic pain, could affect the scores as much as or more than many life-threatening ones.
posted by Ralston McTodd at 3:18 PM on February 15, 2012

chronic disease burden, health status

I'm sympathetic, but adequate measurement of these to capture the endogeneity problem is virtually impossible.
posted by a robot made out of meat at 3:23 PM on February 15, 2012

They used the 12-Item Short Form Health Survey to determine health status.

But also:

"To address otherwise unmeasured morbidity and propensity to use care, we included the following year 1 utilization measures: total health care expenditures, number of office visits, indicators of any emergency department visits and any inpatient admissions, and the number of drug prescriptions."

However, in addition they asked an overall question:

"We also included a single-item self-rated health measure in which patients rate their health as excellent, very good, good, fair, or poor. This single-item predicts mortality and inpatient and outpatient utilization independent of the 12-Item Short Form Health Survey.28"

So what might such a single self report be worth in accuracy and more importantly morbidity prediction?

Actually, a great deal, and appears highly accurate:

Tell Me How You Are, and I Know How Long You Will Live

"The way people rate their health determines their probability of survival in the following decades. Researchers from the Institute of Social and Preventive Medicine at the University of Zurich demonstrate that for ratings ranging from "excellent," "good," "fair" and "poor" to "very poor," the risk of mortality increases steadily -- independently of such known risk factors as smoking, low education levels or pre-existing diseases."

"Now, researchers from the Institute of Social and Preventive Medicine at the University of Zurich demonstrate that self-rated health is also linked to the probability of survival or death over a long period of more than thirty years."
posted by VikingSword at 3:25 PM on February 15, 2012 [4 favorites]

They specifically addressed these objections in:

"We repeated each model with the exclusion of patients with poor self-rated health and 3 or more chronic diseases. This was done because of the possibility that these patients may be more dependent on (and satisfied with) their physicians but more likely to use hospital care and to die."

The findings held.
posted by VikingSword at 3:35 PM on February 15, 2012 [1 favorite]

Perhaps patients are less satisfied with doctors who tell them things they need to hear rather than things they want to hear. For example, "You are fat and need to lose weight. Now." Or "You need to quit drinking. Now. " Or "You need to stop sitting around all the time and exercise. Now." That would likely lead to less patient satisfaction but better health compared to doctors who do not urge their patients strongly to do difficult but important things.
posted by Justinian at 7:20 PM on February 15, 2012 [1 favorite]

higher patient satisfaction was associated with less emergency department use
Going to the emergency room means that something that probably sucks happened to you. Even if you went to the ER for something relatively routine, for some reason, you would still be likely to experience a multi-hour wait.

but with greater inpatient use,
To me inpatient use means visiting the doctor at a scheduled appointment at a convenient time, about something predictable, and being seen reasonably quickly; the whole experience being much less traumatic than the emergency room.

higher overall health care and prescription drug expenditures,
Partially the cost=value illusion, and partially from the simple fact of having experienced more pain-relieving and problem-reducing treatments and drugs.

and increased mortality
The dead cease to suffer. I suspect that on the scale they have used, one year of misery and then death will be less misery than two years of misery and then death. I may be misreading that and they may only be counting patient reports of happiness/misery prior to death and the ones who report greater happiness tend to die 26% more often; in that case, it could be that they suffered greater pain before they were made comfortable. A patient who suffers a suppurating abscess will benefit more from anaesthetic than one with a mild toothache, simply because the anaesthetic sets the pain level to near-zero in either case and the abscess hurt more to start with.
posted by aeschenkarnos at 10:36 PM on February 15, 2012 [1 favorite]

This post and the one below make an excellent pair.
posted by TedW at 4:54 AM on February 16, 2012

Inpatient means that someone is admitted to the hospital. Visiting your doctor in his office is an outpatient visit (an ED visit is also considered outpatient, and in fact some hospital stays are considered outpatient as well). But inpatient indicates that you are having an acute, potentially life-threatening episode of disease. It can be quite a bit more traumatic than visiting the ED.

They are using satisfaction with physician communication as a proxy for overall satisfaction with care. So pain and suffering isn't necessarily a factor here.

This is particularly interesting: However, in our study, more satisfied patients were more likely to rate their health as excellent and had better physical and mental health status than less satisfied patients. This might be pointing to a correlation between general health literacy and satisfaction. If I understand heath better, I will feel better about my interactions with my physician. Also, if I'm health literate and in good health and I go to my doctor with a concern, he or she is more likely to take me seriously.

The real take-away is this: Advocates of patient experience metrics argue that systematic routine measurement of patient satisfaction is a powerful quality improvement tool for physicians and health plans.1 While we do not believe that patient satisfaction should be disregarded, our data suggest that we do not fully understand what drives patient satisfaction as now measured or how these factors affect health care use and outcomes.

A portion of hospital reimbursement from Medicare is now dependent on patient satisfaction scores (HCAHPS). This raises questions about whether this is wise.
posted by jeoc at 6:42 AM on February 16, 2012 [1 favorite]

"Doc, I'm gonna need 120 Xanax and an increase in my OxyContin dose. Oh, and some broad spectrum antibiotics to take in case I get sick on my fishing trip next week."

"Did you ever get that colonoscopy I referred you for?"

"Hell no! I ain't lettin' no one stick nothin' up my ass!"

It pissed me off when Angie's List started doing reviews of health care because it quickly turned into the highest rated doctors consistently had comments like "I went to four different doctors before Dr. Satisfaction gave me what I wanted."

I'm jaded because my practice is mostly uninsured, homeless, substance abusing, and mentally ill, but there are a large number of patients who make very stupid decisions about their health care. This is why market-based approaches to health care mostly fail.

At my clinic, we often say to each other, if you aren't pissing off some of your patients, there's something seriously wrong the way you practice medicine.
posted by Slarty Bartfast at 7:27 AM on February 16, 2012 [4 favorites]

I will definitely be sharing these articles with my colleagues; every hospital administrator I have encountered is obsessed with patient satisfaction and believes that it is synonymous with quality care (and it will also let me expalin that browsing MeFi at work really is work-related use of the computers).

This quote really sums it up:
"The point is that . . . decisions for aggressive intervention—screening or treatment—may be positively reinforced when patients and physicians view the decisions from the perspective of an individual person."

In reality, as The Fat Man said:
The delivery of good medical care is to do as much nothing as possible.
(I am skeptical of anyone who expresses an opinion on healthcare without having read The House of God)
posted by TedW at 8:06 AM on February 16, 2012

Using a self assessment of health status as a control when explaining a self assessment of satisfaction might have some unintended effects.
posted by ~ at 4:17 AM on February 17, 2012