Health care spending and outcomes in US and 10 other countries

​JAMA just published an article trying to ferret out the reasons why the health care spending in the US is so high but our health outcomes so awful (see healthcare costs jama2018 in  dropbox, or doi:10.1001/jama.2018.1150). Given the plethora of numbers and statistics, I will comment on the most important ones (to me, though still lots of numbers).

Details:

--data form 2013-16 from international organizations (including OECD, see comments below), assessing and comparing US data to those of 10 of the highest income countries, which have several different types of health care systems (UK, Canada, Germany, Australia, Japan, Sweden, France, the Netherlands, Switzerland, and Denmark).

Results: (as per the author's chosen domains):

--general demographics:

    --US is by far the largest country with 323 million people, rest of countries with mean of 69 million, but the US has the smallest percentage of people >65 yo (14.5% vs mean of 18.2%)

    --poverty rate (% below poverty line of 60%): US has the highest at 24% vs mean of 18% in the whole group

--spending:

    --spending rate: US has the highest at $9403 per capita vs mean of $5419; US spending at 17.8% of GDP, vs range of 9.6% in Australia to 12.4% in Switzerland​

    --administrative costs (planning, regulating, and managing health care systems and services)​: US highest at 8% of GDP, vs mean of 3% for other countries; 2% for those with national health systems (eg Sweden, UK, Denmark)

        --administrative burden is highest in the US, but is also higher in all countries with insurance-based systems

        --major problems identified by US physicians: 54% complained about time spent on administrative tasks about insurance or claims; 33% about time spent on clinical and data reporting; 16% about time spent on paperwork or disputing medical billing

    --preventive care: US is at the mean of 3% of expenditures, with Canada at 6%, UK at 5%, Netherlands at 4%

    --outpatient care: US is highest at 42% of health care spending, vs mean of 31%

    --health care coverage: US is the lowest at 90%, all others 99.8-100%

    --social spending (spending on old age, incapacity, labor market, education, family, housing) in the US is at 16.7% of GDP, below mean of 19.4%
 --population health:

    --smoking: the US was the lowest at 11.4% of population smoking daily, with mean of other countries being 16.6%

    --obesity/overwt: the US was the highest at 70.1% vs mean of other countries being 55.6%

    --drinking rate and unemployment rates in the US are close to the mean for the whole group

    --health outcomes:

        ​--life expectancy: US had the lowest at 78.8 yrs vs mean of other countries being 81.7 yrs (the US life expectancy varies from 81.3 years in Hawaii to 75 years in Mississippi)

        --infant mortality: US had the highest at 5.8 deaths/1000 live births vs mean of other countries being 3.6/1000 live births

        --neonatal mortality: US had the highest at 4.0 deaths/1000 live births vs mean of 2.6/1000

        --maternal mortality: US had the highest at 26.4/100K live births vs mean of 8.4/100K (!!!!)

        --Low birth wt kids as % of total: US had second highest at 8.1% vs mean of 6.6%

--workforce and structural capacity:

    --physician workforce: in US 2.6 physicians/1000 population vs mean of 3.3

    --% of primary care: US at 43%, same as mean [there are some issues as to how this is defined: internists in several other countries do hospital-based medicine; generalists do primary care. how are the increasing number of hospitalists in the US coded? the % of primary care physicians in the US here is much higher than in several other sources. eg AHRQ found the number to be 33% -- see https://www.ahrq.gov/research/findings/factsheets/primary/pcwork1/index.html ]

    --mean physician salary (adjusted for purchasing power parity): US highest at $218,173 vs range of $86,607 (Sweden) to $202,291 (Australia). mean of $133,723

    --mean nurse salary: US highest at $74,160 vs range from $42,492 (France) to $65,082 (Netherlands). mean of $51,795

    ​--MRI and CT scanners: highest in Japan, 2nd highest in US for MRI (38/1M population) and 3rd highest for CT (41/1M) vs mean of 22 for MRI and 36.5 for CT/1M population

    --hospital beds: US at 2.8/1000 population vs mean of 4.8/1000

--utilization of health care services:

    --US is close to the mean: basically same number of hospital beds and similar utilization rates, same length-of-stay for MI or normal delivery

    --US has same hospital discharge rates for MI, mental/behavioral health/pneumonia/COPD

    ​--US highest in CT scan exams at 245/1000 population (mean 151) and 2nd highest for MRI at 118/1000 (mean 82)

    --US is highest in CABG at 79/100K population (mean 54) and second highest for coronary angioplasty at 248/100K (vs mean 217/100K)

    --US highest for knee replacements at 226/100K population (vs mean of 163), C-sections at 33/100 live births (vs mean of 25), but also well above the mean for cataract surgery and hysterectomies. at the mean for total hip replacements

--pharmaceuticals:

    --​US had the highest pharmaceutical cost per capita at $1443, vs range of $466 (Netherlands) to $939 (Switzerland), with mean of $749

        --for 4 samples (crestor, lantus, advair, humira), US most expensive in general by factor of >2. see commentary below for more info

        --antibiotic prescribing: US 4th at 24 daily doses/1000 population. highest France (29.9), lowest Sweden (12.9), mean 202.

--access and quality:

    --getting appointment within 1-2 days, 2 month or more wait time to see specialist, adequate time spent with primary physician: US at mean

    --perceptions that health system worked well: US 19% (mean 45%); fundamental changes needed: US 53% (mean 45%); need complete rebuild of heath system: US highest at 23% (mean 8%)

    --avoidable hospitalizations: US highest for asthma at 89.7/100K population (mean 42.4), and second highest for diabetes at 191/100K (mean 125.6)

    --quality metrics overall were worse for those on Medicaid

--equity:

    ​--horizontal inequity index (% probability of physician visit in past 12 months by wealth; higher number meaning higher physician access by high-income patient groups): US was the worst (6%), more than 3-fold higher than the second worse(Canada at 1.90). France was 1.20 and Germany 1.00, UK 0.40)

    --unmet need (people not visit physician because of cost, skipped a medical test/treatment/follow-up that was recommended, not fill prescription or skipped doses becuase of cost): US highest at 22.3. range Germany (2.6) to Australia (6.2)

    --unmet need was highest in US both for below-average income (43%, vs mean of 25.5%) and in above-average income (32% with mean of 14.1%)

Commentary:

--so they basically found that, as we know from many many articles/books, the US spends lots on health care but has pretty poor health outcomes, with unequal access for high vs low-income people.  they found pretty similar health resource utilization rates, though they found that the main cost drivers seem to be drugs, admin costs, costs of labor

--this issue of JAMA also had a comparison of drug costs in the US and Canada (from the Medical Letter), noting for 30-day supplies Humira (adalimumab) is $4872 vs $1642, apixaban $419 vs $101, aripiprazole $892 vs $136, dulaglutide $676 vs $209, fluticasone/salmeterol $394 vs $105, lansoprazole $415 vs $63, rosuvastatin $261 vs $55.  (see drug prices US vs canada jama2018 in dropbox, or JAMA 2018; (319(10):1042)

--it was interesting that none of the 4 accompanying editorials pointed out two of the issues i think are most important: inadequate inclusion of the costs for social determinants of health, and lack of insurance/underinsurance/and overall lack of a coherent US health care system

    --i sent out previously a brief review of an important book from 2013, written by 2 public health types (see Bradley EH and Taylor LA. The American Health Care Paradox. BBS Public Affairs. 2013). The health care paradox is the same as in the JAMA article: the US spends much more money on health care than any other country, yet has worse health outcomes than not only the rich countries, but also several of the resource-poorer countries. but these authors come to a very different conclusion: the reason is predominantly that in the US we devote the vast majority of health care dollars to medical care, whereas other high-income countries they have a much larger percentage of their health care dollars devoted to the social items which promote health (the "social determinants of disease"), such as investing in housing, nutrition, education, the environment, and unemployment support. (they quote the WHO: health is defined as "a state of complete physical, mental, and social well-being"). And, as they point out in the book, the US is in the middle of the pack for expenditures as % of GDP if total social service expenditures are included with medical care (and the US is below Italy, Finland, Portugal, and Hungary; countries not even considered in the above JAMA analysis). The US has by far the the highest ratio of health to social expenditures of the 34 OECD countries (Organization for Economic Co-operation and  Development, which includes less resource-rich countries such as Mexico, Turkey, Chile, Czech and Slovak Republics etc). see their figures 1.3 and 1.4. [ie, the US spends a disproportionately large amount of $$ on medical care, a disproportionately low amount on social issues which affect population health, and as a whole, actually does not spend more total money on health care. just a lop-sided allotment. which they argue, reasonably, is a major reason for the "health care paradox"]

    --another major point missed in both the article and the editorials is the issue that in the US there are lots of uninsured and insufficiently insured patients. they quote a 90% insurance rate in the US, though i would add that underinsurance is a really huge issue. This is highlighted by their finding that both low and middle income people did not see clinicians, fill meds, see specialists, etc at an exceedingly disturbing rate. and that their finding of 90% insured is likely to decrease much during the current political climate in the US, leading likely to even worse health outcomes.  A recent blog on patients with chronic pain in a poor, largely minority rural Alabama area noted that 43% of the patients were without insurance (see http://gmodestmedblogs.blogspot.com/2018/03/group-therapy-decreases-chronic-pain.html ). And analyses of different areas in the US show that wealthier urban cities do in fact have much better health outcomes than many poorer areas (eg, health outcomes in rural Mississippi, Louisiana etc are worse than many resource-poor countries). for example, looking at life expectancy, see blog http://gmodestmedblogs.blogspot.com/2016/04/life-expectancy-and-income.html  . 

    ​--i should add that they did comment in the JAMA article that by their analysis the US was not out of line with other countries in terms of investment in social determinants, and that utilization of health services was pretty similar in the different countries. as a result, they felt that there was no evidence that US had poor social conditions leading to more utilization of services. BUT, this seems to be a pretty weak argument: they also note that there was real inequality in terms of accessing care in the US because of cost​, even in the above-average income group, though much worse in low-income (see their data on "equity" above) . likely so many people do not have insurance at all or were really underinsured, which by their own analysis led to decreased utilization of health care (even when they saw a clinician, if the patient did not get the meds or have follow-up because of the costs, this is not good care and does not justify their conclusion that we had similar utilization independent of social conditions). and, i would argue that the definition of social conditions in the JAMA is way too limited (vs in the book on the health care paradox)

    -and they do point out the very higher administrative costs in the US (about 5-6% higher than other countries, with 5-6% being of a very large amount of money: CMS estimates National Health Expenditures in 2006 to $3.3 trillion, so 6%= $200 billion), noting that the costs are much higher in countries overall with a health insurance system vs a national health system (of note, those with national systems did at least as well as the US and the all-inclusive insurance-based systems in terms of all quality and access metrics)

--also, though the salaries of doctors and nurses are higher than in other countries, and this no doubt does add to the overall health care costs, there are real differences in the US than other countries (some commented on in the editorials). we in the US (as opposed to most other countries, and esp the other wealthier nations in this study) have to pay for medical/nursing school (sometimes more than the mortgage on a house), we do have to think about retirement (the US has minimal and often vanishing retirement packages, pensions, etc), and, besides at least in general we do provide some social good. especially in comparison to many business/financial people who make orders of magnitude more money, often have no positive social value (and, in many cases, have negative social value on their way to maximize their own salaries and company profits). so, just leaving the message that a big part of our health care problem is that we pay health care professionals more than in the other 10 countries, creates a wrong target, deflecting from the real one. 

I just have to make a few primary care comments:

    --primary care is markedly underpaid in the US compared to specialties (and i can personally if not enthusiastically corroborate this). this perverse approach (the upside down pyramid) creates several unfortunate sequelae:

         --it is quite onerous for a new residency grad, with likely huge $$ to pay back for college/med school (equivalent to a mortgage) to take a low-paying primary care job in a very expensive city (such as Boston) and expect to be able to buy a condo/house at any point in the near future. quite different from specialists​/those who do more procedures

         --and, related to the poor pay, this situation creates the imperative that we see more patients than we should, in  order for the health center to survive. This means, ironically, that we have less time to take care of all of the patients' issues, which may well lead to ED visits/hospitalizations/specialty visits, which all cost the system hugely more.... and, to further add to the cost of care, specialty visits are more likely to lead to more expensive procedures, many of which have a seemingly pretty constant 30% range of being unnecessary (eg, cardiac caths, or the recent article on knee and hip replacement surgeries -- see hip and knee replacement overutil JAMA2018 in dropbox, or Lam V. JAMA. 2018; 319 (10): 977)

        --yet we get much of the paperwork (the surgeon who wants us to order the MRI prior to surgery). and we do it because we have personal relationships with patients, and we do not want the patient to be hurt because the surgeon does not want to order and get the prior approval from the insurance company (which takes further away from the time we have with patients)

        --as a result, many community health centers are having trouble hiring primary care MDs.  

        --which sets up the ironic situation where there is a deficiency of primary care clinicians (who actually often have the hardest jobs: needing to know the most and broadest array of medicine, taking care of the complexity of patients' medical and psychosocial problems,  and being in the best position to provide the least expensive and most coherent approach to patient care), who are really the buttress of the health care system

        --(that being said, i personally truly love being a primary care physician, developing long-lasting and deep relationships with my patients, and having the broad intellectual stimulation of treating the wide array of patient problems). ​

so, it is useful getting the deluge of statistics in an article as this one in JAMA. and, not surprisingly, it reinforces that we spend way too much money in the US on health care (certainly as compared to other rich countries), but do have pretty miserable health outcomes. but i think this article really misses the point: the US pales so much in comparison to other countries in actually devoting resources to maintaining/improving health, which really is developing and promoting healthy social circumstances which promote healthy lifestyles, and also access to basic health care needs for all. We really lack a coherent system of care which minimizes expenses but covers everyone (as in a national health system), centers the system around primary care/developing long-term relationships with patients, has a clear social preventive approach to care (emphasizing decreasing social/economic disparities and providing for basic needs), and is able to prioritize expenditures (eg, making sure that housing/job training, or access to immunizations is prioritized over spending huge amounts of money on individualized genomic medicine leading to more precise/specific cancer drugs that cost $1M, which might increase the lifespan somewhat of a small number of people...)