Atul Gawande proves health insurance saves lives in 12 bullets and 430 words

Health care is complex. But Atul Gawande applies his intellect to make it clearer. His email about a research paper on the health effects of insurance is a tutorial on how to summarize effectively.

In the New England Journal of Medicine, Atul Gawande and two colleagues published a 6-page survey paper with 54 footnotes on the quantifiable health impacts of health insurance coverage. But with the debate about health care happening now in the Senate, there was an urgent need to get the word out in a simpler form.

A meaty and data-rich 430-word email

Here’s the email that Dr. Gawande sent to his followers. As you might expect from the author of The Checklist Manifesto, it’s tight and carries a targeted payload. Here’s the whole thing:

Subject: Ariadne Labs | Our paper in the New England Journal today on the effects of health insurance on health and survival

Dear friends, supporters and colleagues:

I wanted to share with you an urgently important analysis we are publishing today in the New England Journal of Medicine reviewing the most rigorous research of the last ten years on what is known about the effects of health insurance on health and survival. (You can find it here.)

The paper comes out as the U.S. Senate leadership prepares to release its proposal to repeal the Affordable Care Act. My co-authors, Harvard Chan School of Public Health economists Ben Sommers and Kate Baicker,
and I reviewed more than forty studies since 2009. The central upshot: we found there is strong evidence that expansions in health coverage have improved people’s access to care and health across multiple domains of well-being and reduced deaths overall.

Here are twelve of our specific findings:

1. People are more likely to have a usual source of medical care and to be able to afford the medical care they need.

2. Fewer medical bills are sent to collections and bankruptcies. Catastrophic out-of-pocket expenses are virtually eliminated.

3. People have greater access to primary and preventive care, chronic illness treatment, and medications.

4. 15-30% more people get screened for high cholesterol and cancer.

5. Nearly twice as many patients take necessary diabetes medication.

6. Depression symptoms (leading cause of disability in US) are reduced by 30%, and more people are diagnosed successfully.

7. More low-income patients get necessary surgery for colon cancer BEFORE it’s too late.

8. Result: 25% more people report being in good or excellent overall health.

9. Longest study: gaining Medicaid cut mortality by 6% over 5 years. The biggest gains came from treatable conditions like heart disease, cancer, and infection.

10. The longer people have coverage, the greater the mortality reduction.

11. Overall, for every ~300 to 800 adults who get coverage, we save one life per year.

12. Increasing coverage through Medicaid has been at least as effective as doing so through private insurance.

Our conclusion: Coverage expansion has made people healthier and helped tens of thousands per year live longer, healthier lives. Weakening coverage will increase medical debts, untreated sickness, and deaths.

As the national debate over the Affordable Care Act reaches this critical point, I hope that this study provides policymakers with a clear understanding of the serious implications of terminating insurance benefits and provides guidance for those whose goals are to prevent wide scale harm to population health and survival.

Yours,
Atul

Why this works (and how it could be better)

Here’s what’s best about this:

• It’s short.
• It’s structured. The numbered format makes it easy to consume.
• It’s relatively jargon free. We all understand words like “low-income patients” and “colon cancer.” Gawande’s ability to explain complex medical and statistical results in lay terms is a gift to all of us.
• It’s quantitative rather than emotional. Most of the political commentary we read is framed in emotional terms, using words like “mean.” Instead, here we learn things like “Nearly twice as many patients take necessary diabetes medication.” That’s much harder to argue with and much more convincing.
• It limits weasel words. Gawande couldn’t resist “urgently important,” “most rigorous,” and “strong evidence” in his opening paragraphs. But the rest of the document uses numbers rather than vague and unconvincing intensifiers.

I still have one quibble: scientific writers can’t seem to escape the passive voice. Here are is the same list with a few of the passive-voice results rewritten as active voice statements. Are they better?

1. People with insurance are more likely to have a usual source of medical care and to be able to afford the medical care they need.

2. The more people who are insured, the fewer medical bills that providers send to collections and bankruptcies. Insurance virtually eliminates catastrophic out-of-pocket expenses.

3. Insured people have greater access to primary and preventive care, chronic illness treatment, and medications.

4. Among people with insurance, 15-30% more people get screened for high cholesterol and cancer.

5. Nearly twice as many insured patients take necessary diabetes medication.

6. Insurance reduces depression symptoms (leading cause of disability in US) by 30%, and enables more people to get a successful diagnosis.

7. More low-income patients with insurance get necessary surgery for colon cancer BEFORE it’s too late.

8. Result: 25% more people report being in good or excellent overall health.

9. Longest study: gaining Medicaid cut mortality by 6% over 5 years. The biggest gains came from treatable conditions like heart disease, cancer, and infection.

10. The longer people have coverage, the greater the mortality reduction.

11. Overall, for every ~300 to 800 adults who get coverage, we save one life per year.

12. Increasing coverage through Medicaid has been at least as effective as doing so through private insurance.

The active voice brings with it the repetition of the word “insurance.” Whether you see this as a feature or a bug depends on what you think the email should emphasize.

I didn’t change items 8-12, because they were already in the active voice. Statements like “for every ~300 to 800 adults who get coverage, we save one life per year” are not only clear, they include all of us in the “we” that are saving lives.

Next time you write something long and complicated — and we must all write things like Gawande’s research paper — think about how you would summarize it in a bulleted email like this. You’re not Atul Gawande, but you can certainly crib his techniques.