
Data about your health and that of your community can empower you to make — or demand — changes. When there are gaps in the record or the data don’t exist, participatory data collection empowers people to contribute to the public conversation. Access to data is access to power.
On November 17-19, 2017, Data for Black Lives will host their inaugural conference at MIT, centering data as an instrument of accountability, protest, and collective action.
My friend Max Clermont will moderate a session entitled, “Code Black: Closing the Health Equity Gap,” and invited me to join a distinguished panel that will include Selwyn Rogers, Cheryl Dorsey, and Paulah Wheeler.
The panel description:
The distribution of health and disease is not random. Health disparities between Blacks and whites run deep, no matter the age or ailment. Here are the facts: Blacks have a higher mortality rate than any other racial or ethnic group for eight of the top 10 causes of death. African-American adults are nearly 50% more likely to be obese than their white counterparts. And Blacks are six times more likely than whites to be victims of intentional violence. This breakout session will explore the representation, access, and care delivery gaps in our health system. It will also examine the role that racism, poverty, and the lack of opportunity play in perpetuating disparities.
In preparation for the discussion, I have been thinking about how data has played a role in empowering the individuals and communities I’ve observed.
As a researcher, I tracked how people documented their health and the health of loved ones, participating in the collection of personal data. I also tracked disparities in technology adoption and use. As CTO at HHS, I oversaw initiatives to open up access to data on an individual level, such as through the Blue Button on FHIR project, and at the population level, such as the data sets posted on HealthData.gov.
To dig into these issues further, I looked for recent examples of how open data empowers people:
- Researchers analyzed vital statistics data for Michigan for 2008-15 (before and during the water crisis in Flint) and found that Flint’s fertility rates decreased by 12%, fetal death rates increased by 58%, and overall health at birth decreased compared to other cities in Michigan. Read the PDF report: http://www2.ku.edu/~kuwpaper/2017Papers/201703.pdf
- Researchers analyzed government data and found that in New Jersey neighborhoods with a significant share (27% or higher) of black residents, all children—regardless of race—are at higher risk for asthma.
- ProPublica leverages publicly available (and FOIA’d) data to create tools that people can use to look up doctors, hospitals, nursing homes, and other health care facilities. For example, their Dollars for Doctors analysis of Medicare data shows that where a hospital is located and who owns it make a big difference in what share of its doctors take industry payments.
National surveys also help to map the landscape:
- The 2016 National Survey of Children’s Health found that 38% of of children in every state have had at least one Adverse Childhood Experience or ACE, which puts them at increased risk for health problems later in life. Two-thirds of black children (64%) have had a traumatic experience, compared with about 40% of white children and 51% of Latino children.
- A 2015 Pew Research Center study found few differences among whites, blacks, and Latinos when it comes to tech device use.
When the data doesn’t exist yet, participatory data collection empowers people:
- AIR Louisville is a community program that uses “smart” connected inhalers to help map – and improve – the asthma problem in Louisville.
- Streetwyze is a mobile mapping platform that enables real-time, community-generated data to be integrated with predictive analytics so that health care providers, hospitals, and cities can track health equity indicators, improve service delivery, and predict future trajectories for vulnerable populations. See, for example, their collaboration with Aclima on air quality.
If you have other examples to share — or questions you would like to see addressed — please comment below.
Featured image: Black Lives Matter by Johnny Silvercloud on Flickr.
Seeing this post on the heels of reading the interview with Nikhil Wagle of Dana-Farber is quite stunning. I met Nikhil during Susan G. Komen’s Big Data 4 Breast Cancer event in Menlo Park where I spoke about Maureen, her passing, breast cancer and its connection to my work on CLOUD.
His interview with the Breast Cancer Research Foundation is excellent. The link is here:
https://www.bcrf.org/blog/investigating-breast-cancer-dr-nikhil-wagle
This excerpt shows the connection to the theme of your post here:
“Dr. Nikhil Wagle: Sure, it’s true that black women get breast cancer at earlier ages. They tend to get more advanced stages of the disease, and they tend to die more frequently from breast cancer. There are a number of disparities that some of which can be explained by health disparities, and others of which may be differences in biology. There’s been a lot of work looking into that. One of the things we’re really interested in is to understand the differences in tumor biology and try to see if we can try and address some of those disparities.”
Thanks for making the jump from Twitter, Gary! I’m grateful to learn about this interview with Dr. Wagle.
First thanks for the link to the article about children and asthma in NJ. Before I looked at the charts in the article I guessed the counties were I thought those rates were higher. Except for the zip codes outside AC I was pretty close. My choices based on lower income and more heavy industry. I had not thought about the impact of living by large highways.
To add to your list: Check the Resources provided on the #gyncsm website for our chat on Gyn Cancer Disparities which lists a number of journal and media articles related to various gyn cancer disparities. http://gyncsm.blogspot.com/2017/03/gyn-cancer-health-disparities-chat.html
For example this study published in 2015 which showed Black women are less likely to survive endometrial cancer. There are links to the actual study in this AACR article. https://www.aacrfoundation.org/Science/Pages/black-women-less-likely-to-survive-endometrial-cancer.aspx
I look forward to hearing more about your panel experience.
Dee
Thanks, Dee! The chat summary is a fantastic way to capture insights and ideas that are tweeted and might otherwise disappear. Much appreciated!
Knowledge is power. Data therefore is power. Removing access to data is removing power. The longer I live (especially in the medical system), the more I see how true this is.
For those who haven’t seen the definition of empowerment I use, it’s at the start of this 5 minute “slidecast” video: dave.pt/empoweredengaged. In short, empowerment is increasing someone’s ability to take effective action … and keeping us apart from our data DECREASES that ability, and is thus bluntly and clearly disempowering.
See the BMJ cover in 2013 “Let the patient revolution begin.” https://i0.wp.com/participatorymedicine.org/epatients/wp-content/uploads/sites/3/2015/11/BMJ-cover-let-the-patient-revolution-begin.jpeg?ssl=1
Susannah,
One of my favorite open data sources is SafeCity — a crowdsourcing platform for sexual assault, which was started in India, but anyone can post.
http://safecity.in/about/
http://safecity.in/
Dave, Ranit, thanks so much for adding those resources!
Another community colleague, @ALSAdvocacy shared this research article, published in April 2017, with the comment that “CDC ALS Registry has published repts depicting ALS as white guys’ disease. Are they just the ones in “the system?”:
Evaluating the completeness of the national ALS registry, United States
Quoting from the abstract:
“Our objective was to evaluate the completeness of the United States National ALS Registry (Registry). We compared persons with ALS who were passively identified by the Registry with those actively identified in the State and Metropolitan Area ALS Surveillance project. Cases in the two projects were matched using a combination of identifiers, including, partial social security number, name, date of birth, and sex…
…The cases identified by the surveillance project that did not match cases in the Registry were more likely to be non-white, Hispanic, less than 65 years of age, and from western states. (emphasis added) The methods used by the Registry to identify ALS cases, i.e. national administrative data and self-registration, worked well but missed cases. These findings suggest that developing strategies to identify and promote the Registry to those who were more likely to be missing, e.g. non-white and Hispanic, could be beneficial to improving the completeness of the Registry.”
Please keep adding resources, tools, and insights to the comments here and on Twitter. I’m grateful for the help as we prepare for the Data for Black Lives event in November!
Thanks, Susannah.
My further few cents…
http://als-advocacy.blogspot.com/2017/10/which-people-with-als-count.html
This is more than an “oops” moment. This is serious.
Our data represent us. With a poorly understood and poorly diagnosed disease, the Registry’s failure to count some demographic groups has real consequences.
— We don’t know how many cases of ALS are not diagnosed before people die. Saying that ALS is white guys’ disease can be a self-fulfilling assertion. How many doctors miss the ALS diagnosis in people of color? After all, the CDC said it was mostly white guys’ disease.
— Good epidemiological data should inform science and priorities. Misleading data are perhaps steering precious research funds down some wrong paths.
— Finally, the CDC brags of its Registry’s old-fashioned emailing system to inform some people with ALS of some clinical trials. Now we know it actually contributes to the problem of lack of diversity in trial participation.
Every person with ALS counts. We must stop the madness of publishing data that suggest that all people with ALS look like the incomplete subset of people that the CDC’s ALS Registry found.
This should be the end of the CDC’s ALS Registry. An algorithm, process, and reporting that systematically discriminate are just wrong.
Thanks, Cathy, for the follow up!
Thanks for bringing up an important topic, Susannah.
Last night on the news there was coverage of my local police department’s new implicit-bias training program. Perhaps we need something similar for all who are engaged in information sciences related to healthcare.
Cathy, got a link to video of that news item??
I’ve been particularly moved by the work done by groups that create valuable data when institutions in power refuse to collect or release it. A good example of this is the work done by Fatal Encounters (http://www.fatalencounters.org/) to create “an impartial, comprehensive, and searchable national database of people killed during interactions with law enforcement.”
The Washington Post also does similar work: https://www.washingtonpost.com/graphics/national/police-shootings-2017/
A blog post from the Black Health Matters site caught my eye, in which pharmacist Megan Nichole asks the following question:
What can Black millennial healthcare providers do to increase minority clinical trial enrollment?
Click through to get more detail, but I like the 3 actions she highlights:
1. Enhance the quality of online information about clinical trials.
2. Connect potential participants to satisfied participants and trustworthy providers.
3. Start in our own communities.
I’ll be listening for more information & action items along these lines when I attend the PCORI annual meeting. It kicks off on Oct. 31 with a keynote by Freddie White-Johnson, Founder and President of the Fannie Lou Hamer Cancer Foundation and Program Director for the Mississippi Network for Cancer Control and Prevention at The University of Southern Mississippi. The panel that will follow includes Bishop Simon Gordon, whose ministry includes outreach and partnership in health care research. I’ll be blogging about this panel, too, so stay tuned!
Thank you Susannah for anchoring this discussion. Practical lessons from ongoing projects remain hard to share; this is important.
I’ll add this bit, a very interesting project analysis called “Lessons Learned from an Experiment in Infrastructuring” by Gwen Ottinger about the “Meaning from Monitoring” project. I appreciate this hands-on report about challenge of engaging communities in the design and use of activist tools – the complexity is much higher than it looks from a distance. Worth reading!
Meaning from monitoring: https://www.fairtechcollective.org/experiments
Lessons Learned from an Experiment in Infrastructuring: https://toxicnewsdotorg.files.wordpress.com/2016/03/infrastructure-may-2017.pdf
Everyone,
Thank you for your comments and insights leading up to the Data for Black Lives conference. The videos for all the panels and keynotes are
nowno longer available on the livestream link below:https://livestream.com/D4BL
Here’s the agenda so you can see who is speaking and on what topics.(Unfortunately the 2018 agenda is no longer online.) The opening keynote by Ruha Benjamin was one of the best speeches I’ve seen in the past few years – a must-watch.If you have a chance to watch the health panel video, you’ll see that Max Clermont, the moderator, did not shy away from bringing up the health effects of racism. As Zinzi Bailey captured in a tweet:
“Health departments & providers recommend lifestyle changes for hypertension. What are the ‘lifestyle changes’ I can make to reduce my exposure to racism?”
This part of the conversation reminded me of Alexandra Drane’s work on the “Unmentionables” of health & health care — all the stuff that deeply affects our health (but nobody wants to talk about in public). Stress, caregiving, sleep, money problems, sexual health, substance abuse, adverse childhood events…and the list goes on.
The questions we got from the audience sparked more ideas than we had time to express. The one that sticks with me was, essentially: How might data scientists contribute to the conversation about universal health care and make access more equitable? And does access to health insurance make a difference in life expectancy?
Selwyn Rogers said, essentially, no, access to health insurance does not, by itself, extend people’s lives. He went on to talk about the social sphere, underlining some points he’d made earlier about how your zip code affects your life span.
My contribution was to ask data scientists to look at the natural experiments we are experiencing in our country and in our communities. Like what we observed when Medicaid was expanded in Oregon, which you can read more about here:
The Oregon Health Insurance Experiment
Researchers found that people who got access to health insurance for one to two years had higher rates of health care utilization; lower rates of financial strain; lower rates of depression; and no change in physical health. I pointed out that we’d been talking about the “smog” of stress and racism that hangs over black communities — a phrase introduced by Ruha Benjamin — and health insurance is one way to remove some of the smog from people’s lives.
Update: registration for the 2019 Data for Black Lives conference is now open.
If you’re interested in learning more about the inaugural conference, check out the D4BL press page.