primary care physicians/providers actually do matter!!!

primary care physicians/providers actually do matter!!!

A recent analysis found a pretty clear relationship between primary care physician supply and important health outcomes (see primary care dec mortality jamaintmed2019 in dropbox, or doi:10.1001/jamainternmed.2018.7624).

Details:

-- this study compared information on changes in primary care and specialist physician supply with mortality in the US over the period of 2005 to 2015

-- primary care physician numbers were obtained from the American Medical Association Physician Masterfile; population counts from the US Census Bureau; health outcomes from the National Center for Health Statistics and the Human Mortality Database.

-- 3142 counties were assessed, 7144 primary care service areas, and 306 hospital referral regions

-- The 5 major categories of health-related deaths were: cardiovascular, cancer, infectious diseases, respiratory tract diseases, and substance use or injury

-- comparing 2005 to 2015:

    -- percent of non-metro areas increased from 67 to 73%, % in poverty increased from 15.3 to 16.3%, mean household income decreased from $54,038 to $48,600, population with a less than high school education decreased from 18% to 13%, population greater than 65 yo increased from 15% to 18%, Hispanic population increased from 7% to 9%, uninsured population under 65 yo decreased from 25% to 12%, hospital beds per 100,000 decreased from 359 to 295, tobacco use increased from 18% to 21%, obesity rates increased from 28% to 32%, high-pollution days increased from 4.7 to 6.2%;  there was no significant change in percent female, percent black, unemployment rate

-- primary outcome was life expectancy, and secondary outcomes were cause-specific mortality and restricted mean survival time (see below)

Results:

-- primary care physician supply increased from 196,014 physicians in 2005 to 204,419 in 2015, but:

    --since there were disproportionate losses in primary care while the population increased, the mean density of primary care physicians decreased from 46.6 to 41.4 per 100,000 population. Greater losses in rural areas.

-- For every 10 additional primary-care physicians per 100,000 population,there was an associated 51.5 days increase in life expectancy (29.5-73.5 days; 0.2% increase) 

-- for every 10 additional specialist physicians per 100,000 population, there was a 19.2 day increase (7.0-31.3 days)

-- 10 additional primary-care physicians per 100,000 population was associated with reduced cardiovascular, cancer, and respiratory mortality by between 0.9%-1.4%

    -- cardiovascular mortality: decreased 30 deaths per million

    -- cancer mortality: decreased 24 deaths per million

    -- respiratory mortality: decreased 8.8 deaths per million

-- for 10 additional specialist physician supply:

    -- cardiologists: decreased 49 cardiovascular deaths per million

    -- pulmonologists: decreased 11 deaths per million

--ancillary and sensitivity analyses:

    -- they comment on the high level of completeness of the mortality surveillance at the county-level and primary care physician supply at an area-level

    -- analysis of alternative geographic levels (even though they are controlling for lots of potential confounding factors, people in different counties might access health care in different ways, so they repeated the analyses using alternative geographic levels, including the primary-care service area and hospital referral region): similar results, 51.5 day life expectancy per 10 additional positions at the county level, 117 days at the primary-care service area level, and 158 at the hospital referral region level

    -- instrumental variable analysis (assessing variables (primary care physician supply) that influences the outcome (mortality) but is not subject to reverse causality (i.e. the directionality is lower primary-care physician supply leading to increased mortality). For example the federal Public Service Loan Forgiveness program, a major financial policy inducement for physicians to select primary-care, but does not affect specialist density: an increase of 10 primary-care physicians per 100,000 was associated with a 90 day increase in life expectancy

    -- individual-level analysis (to decrease the likelihood of the “ecological fallacy” that on a population basis there was a correlation between decreased primary-care physicians and mortality, but that on an individual level perhaps the increased mortality was in those individuals who actually got more primary care physician care). For this they used the mortality data from the Social Social Security Death Master File of the Optum Clinformatics Data Mart, 2003 through 2016, of 1,505,554 individuals, to assess the association between area-level primary care physician supply and individual-level life expectancy, and assessed the “restricted mean survival time”, the actual exposure to area-level primary care physician density): survival time increased by 114 days per decade of exposure to 10 more primary care physicians per 100,000 population

    -- falsification testing (e.g. examining whether the primary care physician supply might affect variables not expected to have an association with that physician supply: increased primary-care physician supply was not significantly associated with deaths from interpersonal violence (e.g. murder), suggesting that the primary care physician supply changes in mortality were more likely for effects that those physicians might have influenced

    -- the role of other clinicians (e.g. including primary care nurse practitioners and physician assistants along with physicians): an increase of 36.4 days conferred by an increase of 10 of the combo of these primary care clinicians per 100,000 population

Commentary:

-- one advantage of the study is that they did lots of sensitivity analyses to assess the rigors of their conclusions; this was not just a quick and dirty study. Though one concern with mortality data is the accuracy of death certificates: the signers (myself included) are often guessing as to what factors led to the death of the patient. And, of course, this was not a randomized control trial, so there could well have been important unaccounted for confounders, and therefore their results cannot assert causality

-- from their analysis, the association of primary care physician density and life expectancy was approximately 1/5 the magnitude of the association between poverty and life expectancy (33 days vs 149 days, for a 2-SD increase)

-- in looking at the change in some of the demographic variables, it is pretty noticeable that from 2005-2015 there was a 50% decrease in uninsured patients under 65yo and there was a 30% increase in high-pollution days.  And these 2 really important issues (health insurance and pollution/climate change) are galloping backwards in the current political climate in the US (maybe that's the type of climate change we need???)

    -- see http://gmodestmedblogs.blogspot.com/2016/06/air-pollution-and-heart-disease.html  for a study and comments on air pollution and heart disease

    -- see http://gmodestmedblogs.blogspot.com/2019/01/health-effects-of-climate-change.html for a review of data on climate change and health

so, 

some significant validation for what we do in primary care. but, alas, the US  health care system does not value/validate primary care, even though we provide the most clinically effective (and cost-effective) care from an overall perspective. But fewer medical residents are entering primary care in part because of the current disincentives, including relatively poor pay vs other clinical fields (a huge issue for recent grads, with large mortgage-levels of debt), typically highly truncated visits to deal with the often huge medical/psychosocial problems of patients (also related to poor reimbursement, and the need to see lots of patients in order to get the relatively low salary we do), and the avalanche of paperwork that overwhelms us (pharmacy requests for prior approvals, insurance company forms to deal with, other patient forms where "just have your primary care provider fill this out for you"). And, of course, the remarkable time-sink also known as an electronic medical record. 

Even most of our local Accountable Care Organizations, which presumably need to rely on the efficiency and lower costs of primary care providers, are largely run by hospital networks (which are really run by and for specialists, and most of the saved money goes back to the hospitals) and not by primarycare organizations. 

but, we primary care physicians do have the ability to develop quite profound long-term relationships with patients, which can be as rewarding to us as to them....

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to blogs since 8/15/17:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just type in a name in the search box and get all the blogs with that name in them

to access older blogs from the BMJ website, from October 2013 until 8/15/17: go to http://blogs.bmj.com/bmjebmspotlight/category/archive/ 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list