After missing warning signs of spiking childhood lead poisoning that occurred few months after switching to a corrosive river water source in 2014, outside pressure forced officials at the Michigan Department of Health and Human Services (DHHS) to closely scrutinize their data in July 2015. They discovered scientifically conclusive evidence of an anomalous increase in childhood lead poisoning in summer 2014 immediately after the switch in water sources, but stood by silently as Michigan Department of Environmental Quality (MDEQ) officials repeatedly and falsely stated that no spike in blood lead levels (BLL) of children had occurred. DHHS employees later stonewalled efforts by outside researchers who questioned MDEQ statements and withheld damning documents but released incomplete data suggesting that there were no problems — DHHS officials ultimately fessed up only after incidence of childhood lead poisoning skyrocketed above 10% in the two Flint zip codes with highest water lead risk during summer 2015!

DHHS Pressured to Examine Data in mid-July 2015

DHHS’ engagement in the Flint water crisis, started in early 2015 with two phone calls from Lee-Anne Walters to a state lead poisoning nurse in Lansing identified only as “Karen.” When asked about the phone call last night, Walters said she was at first expressing concern over high lead in water measurements in Flint, but on the second call she was in tears because her son’s blood lead had increased to 6.5 ug/dL.

According to Walters, “Karen” stated that “He is barely lead poisoned.  If CDC had not changed their lead poisoning standard from 10 down to 5, we would not be having this conversation.”  Angry, Walters protested that Karen was minimizing the problem, at which point “Karen” cut her off and said “I am working with kids in their 40’s and 50’s. It is just a few IQ points…it is not the end of the world.” FOIA e-mails do reveal a “Karen” at DHHS who acknowledged receiving phone calls on the dates Walters called (the caller’s name is redacted in the FOIA), but elsewhere this person reveals to colleagues that the last time she actually worked with a child who had blood lead above 45 ug/dL was in 2009.

The next time the Flint water issue was apparently raised at DHHS, was on July 23^rd, when Linda Dkykema (DCH) sent out an e-mail titled “R.E.Director’s Office Assignment- Flint- need update ASAP” on the heels of the memo written by EPA’s Miguel Del Toral.

Misinformed from the Outset: MDEQ Garbage-in All but Ensures DHHS Garbage-out

The Director’s Office update was tainted and probably doomed from the start, because Dykema logically reached out to MDEQ’s Steve Busch for background who provided her the following insights:

“The city of Flint recently conducted drinking water testing throughout the city with special attention to those areas known to have old service lines. The city water supply is in compliance with the lead rule….DEQ has not seen a change in the city’s compliance with the lead rule since switching to the Flint River source….Regarding the home with high drinking water lead levels: some years ago the supply line that serves the neighborhood was replaced, but somehow this house was not connected to the new line, such that the family’s drinking water supply was coming from the old corroded lead pipe. None of the neighbors water had elevated lead level… Regarding the EPA drinking water official quoted in the press articles: the report that he issued was a result of his own research and was not reviewed or approved by EPA management. He has essentially acted outside his authority.”

In other words, Dykema and DHHS started out their investigation with demonstrably false MDEQ talking points as assumptions. Nonetheless, a team of researchers proceeded to examine their data for possible increased incidence of lead poisoning in Flint’s children after the water switch.

Anomalous Increase in Lead Poisoning Noted—But False Conclusion Provided Publicly by MDEQ

The DHHS team immediately found evidence of a problem. Specifically, in the summer months of July, August and September of 2014 immediately after the switch, blood lead levels in Flint had been much higher than normal. On July 27^th of 2015, an e-mail asked “can you quickly run any tests to see if the difference in the first graph is statistically significant,” and the result quickly came back that it was. The team tried to re-run the data adding extra years to see if it changed the conclusions, but it did not because “there does appear to be a higher proportion of EBLL <lead poisoning> last summer than usual.”

A July 28^th e-mail summarized the exercise by noting “This doesn’t say anything about causality, but it does warrant further investigation ” and a two page memo was created for the Director’s Office (see excerpt Figure 1). The “quick detection of abnormal variation” in July, warranting further investigation, does not seem to have gone any further, and the study results did not see the light of day until this FOIA request.

However, MDEQ’s Brad Wurfel did publicly begin stating that a study had been done, and using those results for public relations purposes as early as September 6^th, 2015. Ironically, it was used to discredit high lead in water data collected by Virginia Tech:

“….state officials are questioning the Virginia Tech findings. “The samples don’t match the testing that we’ve been doing in the same kind of neighborhoods all over the city for the past year,” says Brad Wurfel, a state Department of Environmental Quality spokesman. Wurfel says DEQ has conducted two rounds of testing in the past year. He adds that the Department of Community Health conducts its own blood level lead testing in Flint. “With these kind of numbers,” Wurfel says, “we would have expected to be seeing a spike somewhere else in the other lead monitoring that goes on in the community.”

In other words, Wurfel was completely misrepresenting the conclusions of the DHHS July 2015 blood lead study—to claim there was no spike in children’s blood lead when there was one. The FOIA does not reveal precisely where, or how, Wurfel was so misinformed. But the record is also perfectly clear that DHHS sat by silently as Wurfel repeated these false statements throughout September.

Outside Researchers Asking to Analyze Flint’s Blood Lead Data Were Stonewalled: “Yikes!”

On September 2^nd 2015, Professor Marc Edwards of Virginia Tech (primary author of this article and source of the Freedom of Information Act documents cited herein) made a simple data request to DHHS. Edwards had previously made an identical request in November 2006 for a blood lead study in Lansing, and that data was provided without any problems. Edwards did not expect problems this time.

His request sat without response until a reminder e-mail was sent, at which point DHHS employee Robert Scott apologized for the delay, and provided the information needed to process the request, acknowledging:

Yes, I think this <study> sounds great.  There has been some concern about the water source change in Flint, and in fact we had a call about it today. 

The request then sat another 3 days, at which point Edwards inquired again and DHHS said that they were busy and if Edwards wanted the data in “a week rather than a month or so-then please send me a paragraph explaining why.” Edwards wrote:

Yes, I think there is clearly some urgency to the situation. MDEQ has publicly stated that your blood lead records, are showing that there is no public health concern for residents in Flint. The levels of lead in Flint water, that we are finding in our water sampling, are certainly in a range that can cause childhood lead poisoning. Indeed, one child has already, likely been lead poisoned from exposure to high lead in water. I think the fact that you already have other teams working on these records, indicates a high level of interest, and urgency. Congressional interest in the safety of the water is also very high, and this will be an important issue in deciding options for treating the water, in the weeks and months ahead.

If red flags about the data request had not already gone up, the e-mails clearly show alarms were raised immediately throughout all of DHHS after Scott wrote to colleges at 1:09 pm on September 11, 2015 that:

“The attached was submitted to me along with a request for de-identified data, which should be no problem. When you have a few minutes you might want to take a look at it. Sounds like there might be more to this than what we learned previously. Yikes!.”

A flurry of e-mails followed, and within the hour, Edwards request that Scott had just said “should be no problem,” was in fact encountering all kinds of problems. Specifically, a privacy specialist immediately asked for (among other things):

“I want to know more about the study..Saying that the researcher wants to verify the claim of MDEQ is not sufficient…. It does not have to be a great deal longer, but it should explain what the intended use of the blood lead data is. It is also important to know exactly how it will be used for research in order to justify the release….”

The next few weeks were an exercise in frustration, during which it became perfectly clear to Edwards that DHHS had no interest in fulfilling the routine data request.

In the meantime, emails during the same time period indicate that the state spent plenty of time scrutinizing the Virginia Tech data (e.g., “The issue has been ongoing on for awhile now, but picked up steam in July when the Virginia Tech folks got involved.”).   Moreover, Dr. Mona Hanna-Attisha of Hurley Medical Center had made her own data request in mid-September, and she was having trouble getting it processed as well. Edwards wrote to DHHS on September 24th:

I just spoke to this young researcher at Hurley, and apparently, she has been unable to get access to the state blood lead records. I have to say, it is very disturbing that the state keeps issuing these blood lead reports and statements in their press releases, and refuses to share the data backing them up with outside researchers. Even worse, state reps are running around claiming that these reports are proof that Flint water is safe to drink. Can you tell me why it is so difficult to get this data, and why your agency is raising so many obstacles to sharing it with everyone who asks?  I note that I have been asking to see your data since MDEQ first sent it to reporters back in August, and I count 10 e-mails that I sent responding to all your questions. As of yet, you have given me nothing in response. Yet you have been sending reporters one report after another. It seems your agency is more interested in public relations than sound science.

DHHS Scott immediately wrote to colleagues claiming “I’m not very happy with him <Edwards> right now,” and drafted a response that apologized for delays, his vacation, and acknowledging “my time has been largely taken up with my Department’s response to the current situation in Flint.” Hence, the FOIA reveals that DHHS had created their own catch-22: because they stood by and said nothing while MDEQ made false claims about what the blood lead data showed, they had no time to provide data to researchers that wanted to test the truthfulness of the claim.

Scott’s DHHS colleague Nancy Peeler asserted that Edwards e-mail had “intent to escalate and spin things,” and that he (Scott) should not get “caught up in that.” After reading the response, Edwards withdrew his request with a plea to let Hanna-Attisha have access to the state’s data, and within an hour the two DHHS employees were congratulating one another on their success:

DHHS Peeler to DHHS Robert Scott: “Really nice email, thanks.”

DHHS Scott to DHHS Peeler: “Thanks-he sent a couple of responses that I’ll share on Monday. 🙂”

An Agency “more interested in public relations than sound science.”

Three sets of e-mails involving Peeler and Scott, sent within a day of the above exchange, illustrate the unethical behavior occurring within DHHS at the time—all in the name of public relations and at the expense of sound science.

1) “I’m Sure This One Is Not for the Public.” Unbeknownst to Edwards, the day before he asserted in writing that DHHS was behaving unethically by hiding results to help cover up the problem and discredit the findings of Hurley’s Mona Hanna-Attisha–they were doing just that. At 3:45 pm on September 24^th, Scott wrote an e-mail entitled “One more attempt to recreate Hurley.” The e-mail confirmed that Mona-Hanna Attisha’s analysis showing that blood lead had spiked after the switch to Flint River was verified with the DHHS data. But he also noted that “I’m sure this one is not for the public.” When informed that DHHS had knowledge supporting her analysis late last night, Hanna-Attisha confirmed that the state never did share those results with her–rather, the record shows DHHS was working full time to attack the idea there was any problem with childhood lead poisoning in Flint.

2) DHHS Sends Scrubbed Data to Edwards, Hanna-Attisha and Dr. Reynolds (Mott Children’s Health Center): But Withholds Damning July Report. In response to Edwards’ request for the final data supporting Brad Wurfel’s months of assertions that blood lead had not increased, Scott provided both Edwards and Hanna-Attisha with documentation. Unfortunately, it was not the final DHHS memo from late July—the report that demonstrated DHHS knew of a serious lead spike that occurred in summer of 2014. Instead, Scott sent a revised version of the data that had been created just two days before, which had no mention of the statistically significant spike in childhood lead poisoning. Likewise, responding to an earlier request from Dr. Reynolds, DHHS considered sending the July report to the Governor and their Director’s Office, but instead sent a revised version of the data with no hint of problems with blood lead in Flint’s children.

3) Peeler’s “Secret Hope” is Granted to Handle Free Press Reporter. Reporter Kristi Tanner from the Detroit Free Press was doing her own analysis of the Flint blood lead trends. Repeating the approach of Hanna-Attisha using DHSS data, comparing incidence of lead poisoning before to after the water switch, Tanner demonstrated a statistically significant increase in childhood lead poisoning. A flurry of DHHS e-mails to handle her analysis, went back and forth in DHSS–none of it considered acknowledging that she was correct. After all, Scott’s analysis showing the same thing, was surely “not for the public.” Instead, the following exchange occurred between Peeler, Scott and DHHS public relations.

DHSS Peeler to DHSS Scott (September 25, 2015 1:20 pm):

“I think Bob is the best person to speak to them about the lead data, if you are comfortable with that Bob. My secret hope is that we can work in the fact that this pattern is similar to recent past. 🙂 “  

DHHS Scott promptly delivered a quote (9/25/2015 1:19 pm):

“While the trend for Michigan as a whole has shown a steady decrease in lead poisoning year by year, smaller areas such as the city of Flint have their bumps from year to year while still trending downwards overall. Does that sound reasonable?”

Public Relations to Robert Scott (DHSS) (9/25/2015 1:36 pm):

“I like what you had to say. That’s basically what I told her, but she wants to hear it from someone other than a spokesperson. 🙂 Are you available for a call with her.”

The smiley and happy faces inserted in the above e-mails, are indeed part of the original documents. To her credit, Tanner did not buy the states arguments, but did note the DHHS conclusion that the increase was “seasonal and not related to the water supply.”

Also “Not for the Public:” Data That Caused the State to Admit They Were Wrong.

By October 2^nd, the DHHS finally acknowledged that there was a serious problem in Flint. It confirmed that children living in the two zip codes (48503 and 48504) identified by Virginia Tech has having highest lead in water risks , also had increased incidence of childhood lead poisoning after the switch to Flint River. Just as Hurely’s Mona—Hanna Attisha had asserted. But they have never, to date, publicly released their in-house data demonstrating the extent of the problem.

Our FOIA reveals a shocking DHHS graph created in October 2015. It shows the statistically significant spike in blood lead that occurred in summer 2014– the scientific result that DHHS has never publicly acknowledged. Even more horrifying, it also shows blood lead skyrocketing in summer of 2015, to the point where 9.5-12.5 percent of children in Flint’s two high risk zip codes were lead poisoned. This is in the range of Hurley’s more recent estimate that incidence of childhood lead poisoning in some neighborhoods was as high as 15%. It goes without saying, surely, that this data was also “not for the public.”

FOIA Response Delayed by False Claims of “Litigation Hold” and “Attorney Client Privilege.”

DHHS fought hard to delay release of these documents. After the legal time allowed for FOIA elapsed, and Edwards had paid for the request, DHSS fabricated claims of a “Litigation Hold” and “Attorney Client Privilege” as cause for further delays. We want to publicly thank Nancy Kaffer for pointing out the absurdity of these claims. We are also deeply indebted to Senator Ananich and his staff, who fought for release of these documents to us late Friday night.

 

MDHHS emails obtained via FOIA requests regarding blood lead levels of kids in Michigan, specifically in Flint

Download (PDF, 8.94MB)

This is the first batch of emails. We will possibly be releasing a second set sometime in January.

Primary Author: Dr. Marc Edwards

Acknowledgements: Siddhartha Roy