The short version

Claim we investigated

People in LEAD were 58% less likely than people in the control group to be arrested.

Misleading. LEAD participants were arrested less often, but 58% measures odds; their actual chance of arrest was about a quarter lower. After entry, 58% of LEAD participants and 80% of a comparison group were arrested again, about 27% fewer. The 58% figure is a drop in odds, which looks far bigger than the drop in chance when arrests are common.

Quote or cite this fact-check
FactHub rates the claim “People in LEAD were 58% less likely than people in the control group to be arrested” Misleading (high confidence): LEAD participants were arrested less often, but 58% measures odds; their actual chance of arrest was about a quarter lower. Reviewed November 3, 2026. https://facthub.org/n/article/lead-seattle-58-percent-arrests/4

Key points

What the evidence establishes

  • Established. After joining, 58% of LEAD participants were arrested again, compared with 80% of similar people booked and prosecuted as usual.
  • Context. The study's 58% figure is lower odds of arrest. By our approximation, with 80% of the comparison group arrested, that is about a 22% lower chance.
  • Established. The evaluators accepted WPC's point on the wording. LEAD's co-creator said WPC "is not wrong" but still used "likelihood."
  • Not shown. The study wasn't randomized. Without arrests on old warrants, the effect shrank to 34% lower odds and was only marginally significant.
  • Context. The Washington Policy Center is right about the wording but miscounted jobs, housing and deaths, and overstated federal funding.
  • Established. Congress set aside at least $72.5 million for LEAD sites from 2018 to 2026, inside an existing opioid grant program.
Important context: LEAD, short for Law Enforcement Assisted Diversion, began in Seattle in 2011. Officers send people suspected of low-level drug or sex-work offenses to case managers and legal help instead of booking them. A University of Washington team compared 203 participants with 115 similar people booked and prosecuted as usual, from 2009 to 2014. In August 2026 the Washington Policy Center, a free-market think tank, said the program's success claims were wrong. This article rates the most repeated version, from the 2015 announcement.

Over the entire course of the evaluation to date, people in LEAD were 58% less likely than people in the control group to be arrested

Susan E. Collins, Seema L. Clifasefi and Heather S. Lonczak, University of Washington LEAD evaluation team One-page evaluation summary attached to the April 8, 2015 LEAD press release · · Assertion

Context: The summary's second bullet, under 'This evaluation found that LEAD was associated with significantly reduced recidivism.' The full report it summarized said '58% lower odds.' In August 2026 the evaluators said they had used 'likelihood' where they should have used 'odds' and were correcting it.

Source: LEAD press release and UW evaluation summary (April 8, 2015) · View in the dataset

The evidence

The conclusion rests on the combined record. Expand each item to see its role and limitations.

Independence note: This check uses 7 methods across 9 institutions and 1 country or international grouping.
Evidence map
Method Institution / country Year(s) What it shows Key limitation Primary links
Reading the 2015 evaluation report, the 2017 peer-reviewed paper and the 2015 announcement summary, and converting the reported odds into a chance of arrest. University of Washington HaRRT Center, FactHub Research Team United States 2015, 2017, 2026 The 2017 paper reports 58% lower odds of arrest over the whole evaluation and 60% over six months. The 2015 report wrote "60% lower odds (likelihood)." The one-page summary released with the April 2015 announcement turned this into "58% less likely." Odds compare arrested to not arrested, so they move much more than the chance when most people are arrested. With 80% of the comparison group arrested, 58% lower odds means about 63% of LEAD participants arrested: roughly 22% less likely. In 2026 the evaluators accepted WPC's point on wording and corrected the odds wording in the 2017 housing paper. Our conversion applies the adjusted odds to raw after-entry shares, so it is an approximation. The papers don't show how the stated percentages were computed from the printed odds ratios, and we could not reproduce them. The evaluators' model also adjusted for each group's starting arrest rate and time in the study, so the true equivalent is uncertain. The raw after-entry gap, about 27%, gives a similar answer. Collins, Lonczak and Clifasefi, University of Washington UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi) LEAD coalition and UW LEAD Evaluation Team University of Washington HaRRT Center Clifasefi, Lonczak and Collins, University of Washington FactHub Research Team FactHub Research Team
Comparing the published shares of each group arrested at least once before and after entering the evaluation. University of Washington HaRRT Center, FactHub Research Team United States 2009, 2014, 2015, 2017 Before entry, 83% of LEAD participants and 77% of the comparison group had an arrest. After entry, LEAD fell to 58% while the comparison group rose to 80%. The after-entry gap is 22 percentage points, so LEAD participants were about 27% less likely to be arrested in raw terms. The evaluators cite the same 22-point difference. The 58% after-entry share is a coincidence; it is not the 58% in the claim. These are unadjusted shares. The groups differed in age and sex, and the periods before and after entry varied in length by person, which the evaluators' models adjusted for. LEAD participants also started with more arrests (83% vs 77%), so part of their drop could be a return toward normal after a peak, which raw shares cannot separate. Measured instead from each group's own start, LEAD fell 25 points while the comparison group rose 3, which puts LEAD about a third below where it might otherwise have been. That is larger than a quarter but still far from 58%. University of Washington HaRRT Center UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi) UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi) Collins, Lonczak and Clifasefi, University of Washington FactHub Research Team
Reading the evaluators' analysis with arrests to serve warrants removed and their discussion of felony charges, and the critique of both. University of Washington HaRRT Center, Washington Policy Center United States 2015, 2017, 2026 Many arrests were to serve warrants, often for missed court dates on old charges. With those removed, LEAD participants had 34% lower odds of arrest, and the result was only marginally significant (p = .09). Over six months it was not significant. The evaluators say fewer warrants is partly how LEAD works, because legal help resolves old cases. The prosecutor's office knew who was in LEAD and could weigh it in felony charging, so the 39% lower odds of a felony charge may partly reflect its decisions. The evaluators call felony charging a less clean indicator. Whether fewer warrant arrests reflect less crime or only fewer missed court dates can't be told from this data. The 2017 paper does not show the analysis without warrant arrests. UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi) University of Washington HaRRT Center Washington Policy Center (Eric Zimmerman) University of Washington HaRRT Center Collins, Lonczak and Clifasefi, University of Washington
Reading the U.S. Justice Department's CrimeSolutions profile and the study's own statement of its design. U.S. Department of Justice, Office of Justice Programs, University of Washington HaRRT Center United States 2017, 2025 CrimeSolutions, the Justice Department's program-rating service, rates LEAD "Promising" for arrests and felony charges. Its outcome table describes the result correctly, as 56 percent lower odds of arrest, though its rating line says "reducing the likelihood" of arrest. It lists the 2017 study as its evidence. Police shifts in that study were randomly assigned, but comparison neighborhoods were added later, so its authors call it nonrandomized. The review repeats the same study rather than adding new data. Its profile says it is no longer updated. U.S. Department of Justice, Office of Justice Programs, CrimeSolutions Collins, Lonczak and Clifasefi, University of Washington Collins, Lonczak and Clifasefi, University of Washington
Searching official, program and congressional records for the claim, and comparing archived and current versions of pages. Seattle Police Department, LEAD Seattle-King County, Seattle City Council, U.S. House of Representatives United States 2015, 2016, 2026 The Seattle Police Department posted the line in April 2015, and the post is still up. LEAD's own website said participants were "58% less likely to be arrested" as late as April 2026, then changed it to "58% lower odds." In April 2026 LEAD's operator told a City Council committee that LEAD "reduced recidivism for all crimes by 58%." In 2016 Rep. Suzan DelBene told Congress LEAD cut recidivism "by as much as 60 percent." Defending LEAD in August 2026, its co-creator still called it a "58% reduced likelihood." We checked the sources critics named and a sample of others, not every use of the figure. Seattle Police Department LEAD Seattle-King County (Internet Archive copy, April 29, 2026) LEAD Seattle-King County Purpose Dignity Action (LEAD), Seattle City Council Public Safety Committee The Center Square (Carleen Johnson), republished by News-Shield Congressional Record, U.S. House
Reading each enacted Commerce-Justice-Science explanatory statement from 2017 to 2026 and the committee reports it made binding. U.S. Congress, Washington Policy Center United States 2018, 2019, 2020, 2021, 2022, 2023, 2024, 2025, 2026 From 2018 to 2026, enacted budgets reserved part of the Justice Department's opioid grant program for LEAD sites. The minimums were $2.5 million in 2018, $5 million in 2019, $10 million in 2020 and $11 million a year after that, except 2025. That totals $72.5 million. From 2022 on, the amounts sit in committee reports that the enacted statements give the force of their own text. The 2025 budget continued the 2024 law, and we found no new LEAD directive for it. WPC's headline says "nearly $100 million"; its year-by-year list, by our sum, comes to about $80 million. These are minimums within a larger opioid grant program, not new money, and not amounts actually awarded. Award records that name LEAD are too incomplete to total. Congressional Record, U.S. House U.S. Senate Committee on Appropriations Congressional Record, U.S. House Congressional Record, U.S. House Congressional Record, U.S. House U.S. House Committee on Appropriations U.S. House Committee on Appropriations U.S. Senate Committee on Appropriations U.S. Senate Committee on Appropriations U.S. Congress Washington Policy Center (David Boze) FactHub Research Team Congressional Record Congressional Record Congressional Record Congressional Record
Comparing the Washington Policy Center's figures with the housing and employment paper's table and the evaluators' person-level counts. Washington Policy Center, University of Washington HaRRT Center United States 2017, 2026 The 2017 housing and employment paper reports, for each month of follow-up, the share of participants housed or employed, then averages those shares. WPC turned the rise in the average share employed into "three" people who gained jobs, and the average share housed into a 13% success rate. Counting people instead, the evaluators report 11 of 162 newly employed and 45 of 145 (31%) newly housed, with 81% sheltered or housed at some point. WPC then set nine deaths against its job figure. The evaluators confirm a plain employed-or-not measure did not change significantly. The person-level counts come from the evaluators' 2026 response, not the peer-reviewed paper, and the study had no comparison group for jobs or housing. MyNorthwest / KIRO Newsradio Washington Policy Center (Eric Zimmerman) Clifasefi, Lonczak and Collins, University of Washington University of Washington HaRRT Center University of Washington HaRRT Center Collins, Lonczak and Clifasefi, University of Washington
1. The study measured lower odds, not a lower chance Reading the 2015 evaluation report, the 2017 peer-reviewed paper and the 2015 announcement summary, and converting the reported odds into a chance of arrest.

The 2017 paper reports 58% lower odds of arrest over the whole evaluation and 60% over six months. The 2015 report wrote "60% lower odds (likelihood)." The one-page summary released with the April 2015 announcement turned this into "58% less likely." Odds compare arrested to not arrested, so they move much more than the chance when most people are arrested. With 80% of the comparison group arrested, 58% lower odds means about 63% of LEAD participants arrested: roughly 22% less likely. In 2026 the evaluators accepted WPC's point on wording and corrected the odds wording in the 2017 housing paper.

Limit: Our conversion applies the adjusted odds to raw after-entry shares, so it is an approximation. The papers don't show how the stated percentages were computed from the printed odds ratios, and we could not reproduce them. The evaluators' model also adjusted for each group's starting arrest rate and time in the study, so the true equivalent is uncertain. The raw after-entry gap, about 27%, gives a similar answer.

Verify: Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56 LEAD Program Evaluation: Recidivism Report (March 27, 2015) Innovative Law Enforcement Assisted Diversion (LEAD) Program Is Showing Success (press release with evaluation summary, April 8, 2015) HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): findings and concession Seattle's Law Enforcement Assisted Diversion (LEAD) Program: Within-Subjects Changes on Housing, Employment, and Income/Benefits Outcomes and Associations With Recidivism. Crime & Delinquency 63(4) (2017): 429-445, corrected September 2026 Saved analysis: what the reported lower odds mean for the chance of arrest Saved analysis: odds estimates as the evaluators reported them

2. After joining, 58% of LEAD participants were rearrested, against 80% of the comparison group Comparing the published shares of each group arrested at least once before and after entering the evaluation.

Before entry, 83% of LEAD participants and 77% of the comparison group had an arrest. After entry, LEAD fell to 58% while the comparison group rose to 80%. The after-entry gap is 22 percentage points, so LEAD participants were about 27% less likely to be arrested in raw terms. The evaluators cite the same 22-point difference. The 58% after-entry share is a coincidence; it is not the 58% in the claim.

Limit: These are unadjusted shares. The groups differed in age and sex, and the periods before and after entry varied in length by person, which the evaluators' models adjusted for. LEAD participants also started with more arrests (83% vs 77%), so part of their drop could be a return toward normal after a peak, which raw shares cannot separate. Measured instead from each group's own start, LEAD fell 25 points while the comparison group rose 3, which puts LEAD about a third below where it might otherwise have been. That is larger than a quarter but still far from 58%.

Verify: HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): findings and concession LEAD Program Evaluation: Recidivism Report (March 27, 2015) LEAD Program Evaluation: Recidivism Report (March 27, 2015): significance and six-month change Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56 Saved analysis: how much less often LEAD participants were arrested after entry

Share of Seattle LEAD and control participants arrested, before and after entryPercent of people in Seattle's LEAD diversion program and in a system-as-usual control group with at least one arrest before and after entering the University of Washington evaluation, October 2009 to July 2014.{"width":960,"height":540,"plot":{"left":90,"top":70,"right":926,"bottom":482},"yDomain":[0,100],"series":[{"name":"LEAD","color":"#176b87"},{"name":"Control","color":"#b6542c"}],"points":[{"series":0,"x":194.5,"y":140.04,"xLabel":"LEAD, before","yLabel":"83"},{"series":0,"x":403.5,"y":243.04,"xLabel":"LEAD, after","yLabel":"58"},{"series":1,"x":612.5,"y":164.76,"xLabel":"Control, before","yLabel":"77"},{"series":1,"x":821.5,"y":152.4,"xLabel":"Control, after","yLabel":"80"}]}Share of Seattle LEAD and control participants arrested, before and after entry020406080100LEAD, beforeLEAD, afterControl, beforeControl, afterGroup and periodLEADControlFactHub · Share arrested at least once (%)
After entering the study, 58% of LEAD participants were arrested at least once, against 80% of the comparison group: about 27% fewer, not 58%. Shares are unadjusted. View source chart
3. The effect shrinks without warrant arrests, and felony charges are a weaker test Reading the evaluators' analysis with arrests to serve warrants removed and their discussion of felony charges, and the critique of both.

Many arrests were to serve warrants, often for missed court dates on old charges. With those removed, LEAD participants had 34% lower odds of arrest, and the result was only marginally significant (p = .09). Over six months it was not significant. The evaluators say fewer warrants is partly how LEAD works, because legal help resolves old cases. The prosecutor's office knew who was in LEAD and could weigh it in felony charging, so the 39% lower odds of a felony charge may partly reflect its decisions. The evaluators call felony charging a less clean indicator.

Limit: Whether fewer warrant arrests reflect less crime or only fewer missed court dates can't be told from this data. The 2017 paper does not show the analysis without warrant arrests.

Verify: LEAD Program Evaluation: Recidivism Report (March 27, 2015) HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): employment, housing, deaths and warrants National Retraction Brief: Widespread and systematic factual errors in Law Enforcement Assisted Diversion (LEAD), Part 1, version 1.3 (August 16, 2026) HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): findings and concession Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56

4. A federal review rates LEAD promising, on one nonrandomized study Reading the U.S. Justice Department's CrimeSolutions profile and the study's own statement of its design.

CrimeSolutions, the Justice Department's program-rating service, rates LEAD "Promising" for arrests and felony charges. Its outcome table describes the result correctly, as 56 percent lower odds of arrest, though its rating line says "reducing the likelihood" of arrest. It lists the 2017 study as its evidence. Police shifts in that study were randomly assigned, but comparison neighborhoods were added later, so its authors call it nonrandomized.

Limit: The review repeats the same study rather than adding new data. Its profile says it is no longer updated.

Verify: Program Profile: Law Enforcement Assisted Diversion (LEAD) Program (Seattle, Washington) Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56 Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes (2017): design and officer discretion

5. Who repeated "58% less likely" Searching official, program and congressional records for the claim, and comparing archived and current versions of pages.

The Seattle Police Department posted the line in April 2015, and the post is still up. LEAD's own website said participants were "58% less likely to be arrested" as late as April 2026, then changed it to "58% lower odds." In April 2026 LEAD's operator told a City Council committee that LEAD "reduced recidivism for all crimes by 58%." In 2016 Rep. Suzan DelBene told Congress LEAD cut recidivism "by as much as 60 percent." Defending LEAD in August 2026, its co-creator still called it a "58% reduced likelihood."

Limit: We checked the sources critics named and a sample of others, not every use of the figure.

Verify: Study: Innovative Seattle/King County Diversion Program Shows Success (SPD Blotter, April 8, 2015) Evaluation page for the 2017 housing and employment paper, as archived Evaluation page for the 2017 housing and employment paper (current) Diversion & LEAD utilization since Seattle drug ordinance adoption (April 28, 2026) WATCH: Creator defends drug diversion program amid conflicting success claims (August 12, 2026) Comprehensive Opioid Abuse Reduction Act of 2016, Amendment No. 3 offered by Ms. DelBene (May 12, 2016), p. H2307

Independent UW researchers funded by Laura & John Arnold Foundation found LEAD reduced recidivism for all crimes by 58%, reduced felony charges by 39%

Purpose Dignity Action (LEAD), LEAD's operator, presenting to the Seattle City Council Public Safety Committee Slide 'Evaluation & research' in the committee's agenda materials · · Assertion

Context: Shown in an update on LEAD use since Seattle's 2023 drug ordinance. The same slide continues that LEAD 'resulted in 41 fewer days in jail per person annually, and lowered the odds of entering prison by 88%.'

Source: LEAD presentation, Seattle City Council Public Safety Committee (April 28, 2026) · View in the dataset

It is a 58% reduced likelihood of being arrested compared to a comparison group.

Lisa Daugaard, LEAD co-creator, Purpose Dignity Action Interview with The Center Square · · Assertion

Context: Responding to WPC, she first said 'It's not a 58% reduction in the absolute number of arrests for the lead participants.' She added, 'But WPC is not wrong,' and said the figure 'should not be sort of flattened or further simplified as a 58% reduction in the number of arrests.'

Source: The Center Square, 'Creator defends drug diversion program amid conflicting success claims' (August 12, 2026), as republished by News-Shield · View in the dataset

6. Federal budgets set aside at least $72.5 million for LEAD Reading each enacted Commerce-Justice-Science explanatory statement from 2017 to 2026 and the committee reports it made binding.

From 2018 to 2026, enacted budgets reserved part of the Justice Department's opioid grant program for LEAD sites. The minimums were $2.5 million in 2018, $5 million in 2019, $10 million in 2020 and $11 million a year after that, except 2025. That totals $72.5 million. From 2022 on, the amounts sit in committee reports that the enacted statements give the force of their own text. The 2025 budget continued the 2024 law, and we found no new LEAD directive for it. WPC's headline says "nearly $100 million"; its year-by-year list, by our sum, comes to about $80 million.

Limit: These are minimums within a larger opioid grant program, not new money, and not amounts actually awarded. Award records that name LEAD are too incomplete to total.

Verify: Explanatory statement, Consolidated Appropriations Act, 2018 (March 22, 2018), p. H2093 Senate Report 115-139, Commerce, Justice, Science appropriations bill, 2018 Explanatory statement, Consolidated Appropriations Act, 2019 (February 13, 2019), p. H1811 Explanatory statement, Further Consolidated Appropriations Act, 2020 (December 17, 2019), p. H10966 Explanatory statement, Consolidated Appropriations Act, 2021 (December 21, 2020), p. H7942 House Report 117-97, Commerce, Justice, Science appropriations bill, 2022 House Report 117-395, Commerce, Justice, Science appropriations bill, 2023 Senate Report 118-62, Commerce, Justice, Science appropriations bill, 2024 Senate Report 119-44, Commerce, Justice, Science appropriations bill, 2026 Full-Year Continuing Appropriations and Extensions Act, 2025 (Public Law 119-4) U.S. Congress misled by false LEAD crime-reduction claims, leading to nearly $100 million in federal funding (September 9, 2026) Saved analysis: LEAD set-asides in enacted federal budgets, 2018-2026 Explanatory statement, Consolidated Appropriations Act, 2022 (March 9, 2022) Explanatory statement, Consolidated Appropriations Act, 2023 (December 20, 2022) Explanatory statement, Consolidated Appropriations Act, 2024 (March 5, 2024) Explanatory statement, Commerce, Justice, Science appropriations act, 2026 (January 8, 2026), p. H255

7. WPC's job, housing and death figures read monthly averages as head counts Comparing the Washington Policy Center's figures with the housing and employment paper's table and the evaluators' person-level counts.

The 2017 housing and employment paper reports, for each month of follow-up, the share of participants housed or employed, then averages those shares. WPC turned the rise in the average share employed into "three" people who gained jobs, and the average share housed into a 13% success rate. Counting people instead, the evaluators report 11 of 162 newly employed and 45 of 145 (31%) newly housed, with 81% sheltered or housed at some point. WPC then set nine deaths against its job figure. The evaluators confirm a plain employed-or-not measure did not change significantly.

Limit: The person-level counts come from the evaluators' 2026 response, not the peer-reviewed paper, and the study had no comparison group for jobs or housing.

Verify: John Curley Show article on LEAD (August 12, 2026) National Retraction Brief: Widespread and systematic factual errors in Law Enforcement Assisted Diversion (LEAD), Part 1, version 1.3 (August 16, 2026) Seattle's Law Enforcement Assisted Diversion (LEAD) Program: Within-Subjects Changes on Housing, Employment, and Income/Benefits Outcomes and Associations With Recidivism. Crime & Delinquency 63(4) (2017): 429-445, corrected September 2026 HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): findings and concession HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): employment, housing, deaths and warrants Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56

Common questions 9

“The 58% claim has been corrected now that the error is known.”

Short answer: Only in places

Some pages have been fixed, but not all. LEAD's website now says "58% lower odds," and the 2017 housing paper carries a September 2026 correction. The police department's 2015 post still says "58% less likely," and in August 2026 LEAD's co-creator still called it a "58% reduced likelihood."

See the evidence: Who repeated "58% less likely"

“LEAD caused participants to be arrested less.”

Short answer: Likely in part; not proven

Probably in part, but this study can't prove it. Police shifts were randomly assigned, but comparison neighborhoods were added later, so the design is nonrandomized. Officer favoritism is an unlikely cause, since only about 40 of 1,300 officers took part. But much of the gap was in warrant arrests, and without them the result was only marginally significant.

See the evidence: A federal review rates LEAD promising, on one nonrandomized study

“LEAD reduced felony charges by 39%.”

Short answer: 39% lower odds

The study found 39% lower odds of a felony charge, not 39% fewer charges. Prosecutors knew who was in LEAD and could weigh it, so the evaluators call this a less clean measure than arrests. After entry, 20% of LEAD participants and 31% of the comparison group were charged with a felony.

See the evidence: The effect shrinks without warrant arrests, and felony charges are a weaker test

“No clear difference in arrests rates can be reliably claimed, but also cannot be altogether ruled-out.”

Our rating: Misleading

Arrest rates clearly differed: 58% of LEAD participants versus 80% of the comparison group, a highly significant gap. What is uncertain is how much LEAD caused, especially for non-warrant arrests.

See the evidence: After joining, 58% of LEAD participants were rearrested, against 80% of the comparison group

“Of 163 or 164 LEAD participants without jobs, only three gained jobs.”

Our rating: False

Three is the rise in the average number employed in a given month, not a count of people. The evaluators count 11 of 162 who got jobs during follow-up; a plain job measure showed no significant change.

See the evidence: WPC's job, housing and death figures read monthly averages as head counts

“LEAD had a 13% success rate in securing permanent housing for participants who lacked it.”

Our rating: False

13% is the share housed in an average month. Counting people, 45 of 145 who lacked housing (31%) got it, and 81% had shelter or housing at some point. No comparison group was studied.

See the evidence: WPC's job, housing and death figures read monthly averages as head counts

“Three times as many LEAD participants died as gained jobs during the study.”

Our rating: False

Nine LEAD participants died during the evaluation; 11 were newly employed during follow-up. The claim set a death count against a misread monthly job average.

See the evidence: WPC's job, housing and death figures read monthly averages as head counts

“According to an initial study, LEAD successfully reduces recidivism by as much as 60 percent.”

Our rating: Misleading

The 60% was lower odds of arrest over six months, not 60% less reoffending. Over those months, the share of LEAD participants arrested held about steady while the comparison group's rose by half.

See the evidence: The study measured lower odds, not a lower chance

“Congress was misled by false LEAD crime-reduction claims, leading to nearly $100 million in federal funding.”

Our rating: Mixed

Congress did hear inflated figures, and budgets then reserved LEAD money. But enacted set-asides total $72.5 million from 2018 to 2026, inside an existing opioid grant, not nearly $100 million.

See the evidence: Federal budgets set aside at least $72.5 million for LEAD

Conclusion

The evaluation found a real difference, and the way it was announced inflated it. After entering the study, 58% of LEAD participants were arrested at least once, against 80% of the comparison group. That is 22 points, or about 27% fewer. The evaluators' adjusted model put this as 58% lower odds of arrest. Odds and chance differ most when an outcome is common, and here most people in both groups were arrested. By our approximation, 58% lower odds means roughly 22% less likely. Either way, the plain-English claim of "58% less likely" more than doubles the effect.

The error started in the evaluators' own one-page summary in April 2015 and spread through police, program and government materials. The full report and the 2017 paper reported odds, though the report also wrote "odds (likelihood)." The evaluators have accepted WPC's point on the wording, and some pages have been fixed. Others, including the police department's 2015 post, still carry it, and LEAD's co-creator still used "likelihood" in August 2026.

How much of the difference LEAD caused is less certain. Police shifts were randomly assigned to LEAD or comparison, but comparison neighborhoods were added later, so the evaluators call the design nonrandomized. The evaluators balanced the groups statistically, and the arrest result was significant (p = .001). But with arrests on old warrants removed, the gap shrank to 34% lower odds and was only marginally significant (p = .09). The 39% lower odds of a felony charge may partly reflect prosecutors, who knew who was in LEAD.

The Washington Policy Center is right about the 58% wording, and it gets credit for forcing corrections. Several of its own numbers go wrong in a similar way: it read monthly averages as head counts for jobs and housing, set deaths against a misread job figure, and overstated federal money. Confidence is high because the claim, the study's numbers and the corrections are all on the record.

How we checked
  1. Defined the claim as the 2015 announcement's wording, "People in LEAD were 58% less likely than people in the control group to be arrested," read the way most people would: as a 58% lower chance of arrest.
  2. Read the University of Washington evaluators' March 2015 report, the 2017 peer-reviewed recidivism paper, the 2017 housing and employment paper, and their August 2026 response to the Washington Policy Center.
  3. Recorded the published arrest shares for each group before and after entry, checked figure labels against the PDF text, and converted the reported odds into a chance of arrest. The arithmetic is in the FactHub dataset.
  4. Traced who repeated the 58% claim: police, program, city council and congressional records, comparing archived and current versions of the program's web page.
  5. Checked the Washington Policy Center's counter-claims against the evaluators' tables and their person-level counts, and its federal funding figure against each enacted budget statement from 2017 to 2026.
  6. Recorded limits: the study was not randomized, person-level job and housing counts are not peer reviewed, and set-asides are minimums rather than amounts awarded.

Editorial note: Scope: the claim that LEAD participants were 58% less likely to be arrested, as stated in the 2015 announcement, and the main counter-claims in the Washington Policy Center's 2026 brief. The article does not judge whether LEAD should be funded or expanded, and it does not evaluate LEAD programs outside Seattle or after 2014, which this study did not cover. The odds-to-chance conversion is our own arithmetic, shown in the dataset. Person-level job and housing counts come from the evaluators' 2026 response and are not peer reviewed.

Expert-review status

No external specialist review has been recorded.

Funding and conflicts

Funding: No external funding for this piece.

Ownership: Published by the FactHub Research Team.

Conflicts: None known.

Sources 36

Sources are labeled by kind, with what this fact check takes from each, so readers can check the evidence independently.

  1. Peer-reviewed study Collins, Lonczak and Clifasefi, University of Washington. Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning 64 (2017): 49-56 Free copy DOI: 10.1016/j.evalprogplan.2017.05.008

    The peer-reviewed paper comparing 203 LEAD participants with 115 people processed as usual.

    Reliability: Has a stake in this finding. Written by the evaluators of the program, funded by the Laura and John Arnold Foundation; LEAD staff sat on the evaluation advisory committee. The authors declared no conflicts.

    • The paper reports the results as lower odds: 60% over six months, and 58% for arrest and 39% for a felony charge over the longer term.

      What the source says
      Compared to controls, LEAD participants had 60% lower odds of arrest during the six months subsequent to evaluation entry; and both a 58% lower odds of arrest and 39% lower odds of being charged with a felony over the longer term.
      Abstract

    • The evaluators themselves put the raw difference at about 22 percentage points.

      What the source says
      reduced recidivism by about 22 percentage points compared to the system as usual
      Section 4.3

    • Qualifies: The planned randomization was dropped, so the study is nonrandomized.

      What the source says
      the originally planned randomization schema was relaxed, and a nonrandomized controlled design was employed in its place
      Section 4.4, Limitations

    • Qualifies: The prosecutor's office knew who was in LEAD and could weigh it in felony charging, so lower felony odds may partly reflect its decisions.

      What the source says
      As an unblinded operational partner, the prosecutor's office could take into account LEAD participation and progress in the program when deciding whether and when to file felony charges.
      Section 4.2

    • Qualifies: Nine LEAD participants and two comparison participants died during the five-year evaluation; comparison deaths were not systematically tracked.

      What the source says
      including 9 LEAD participants (4.43%) and 2 (1.74%) control participants
      Section 3.2.2
  2. Peer-reviewed study Collins, Lonczak and Clifasefi, University of Washington. Seattle's Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes (2017): design and officer discretion Free copy DOI: 10.1016/j.evalprogplan.2017.05.008

    Further points from the peer-reviewed recidivism paper.

    Reliability: Has a stake in this finding. Written by the program's evaluators, funded by the Laura and John Arnold Foundation.

    • Qualifies: People were assigned by whether they were arrested on a LEAD or a comparison police shift.

      What the source says
      police officer shifts were randomized to be either LEAD or control shifts, and eligible individuals were allocated to those conditions if they were arrested during the respective shifts
      Section 2.2

    • Qualifies: Comparison neighborhoods were added later because too few eligible people were left in the original area.

      What the source says
      control areas (in addition to control shifts) were added to the evaluation
      Section 2.2

    • Few officers knew who was in LEAD, so officer favoritism is an unlikely explanation for fewer arrests.

      What the source says
      Only approximately 40 of 1300 Seattle Police Department officers were involved in the LEAD program.
      Section 4.1
  3. Academic UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi). LEAD Program Evaluation: Recidivism Report (March 27, 2015)

    The evaluators' full report, including analyses with and without arrests to serve warrants.

    Reliability: Has a stake in this finding. The same evaluators' report to LEAD's policy group before peer review; copy hosted by IssueLab.

    • The report described the six-month result as lower odds and added 'likelihood' in parentheses.

      What the source says
      the LEAD group had 60% lower odds (likelihood) of arrest during the six months subsequent to evaluation entry
      Summary of findings, p. 2

    • The share of LEAD participants arrested fell by 30% (83% to 58%) while the comparison group's rose by 4%.

      What the source says
      the proportion of control participants who were arrested at least once subsequent to evaluation entry increased by 4%, whereas the proportion of LEAD participants who were arrested subsequent to evaluation entry decreased by 30%
      Summary of findings, p. 3

    • Qualifies: Without warrant arrests, the long-term result was 34% lower odds and only marginally significant.

      What the source says
      LEAD participants had 34% lower odds of being arrested at least once when warrant-related arrests were removed. This effect was marginally significant (p = .09).
      Summary of findings, p. 3

    • Qualifies: Over six months, the analysis without warrant arrests found no significant effect.

      What the source says
      Analyses of exclusively nonwarrant-related arrests indicated no significant LEAD effects.
      Summary of findings, p. 2
  4. Academic UW LEAD Evaluation Team (Collins, Lonczak, Clifasefi). LEAD Program Evaluation: Recidivism Report (March 27, 2015): significance and six-month change

    Further points from the evaluators' full report.

    Reliability: Has a stake in this finding. The same evaluators' report to LEAD's policy group before peer review; copy hosted by IssueLab.

    • The long-term arrest result was statistically significant at p = .001.

      What the source says
      The LEAD effect on arrests over time was statistically significant (p = .001).
      Summary of findings, p. 3

    • Over six months, the share of comparison participants arrested rose by half while LEAD's held about steady.

      What the source says
      the proportion of control participants who were arrested increased by 51%, whereas the proportion of LEAD participants who were arrested plateaued (+6%)
      Summary of findings, p. 2
  5. Party to the dispute LEAD coalition and UW LEAD Evaluation Team. Innovative Law Enforcement Assisted Diversion (LEAD) Program Is Showing Success (press release with evaluation summary, April 8, 2015)

    The announcement of the evaluation, with the evaluators' one-page summary attached.

    Reliability: Has a stake in this finding. Issued by LEAD's backers (contacts at the ACLU of Washington and Public Defender Association); the summary is signed by the evaluators.

    • The evaluators' summary said LEAD participants were 58% less likely to be arrested, the wording rated here.

      What the source says
      Over the entire course of the evaluation to date, people in LEAD were 58% less likely than people in the control group to be arrested
      Summary of LEAD Program Evaluation: Recidivism Report, second bullet · Susan E. Collins, Seema L. Clifasefi and Heather S. Lonczak, UW evaluators
  6. Party to the dispute University of Washington HaRRT Center. HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): findings and concession

    The evaluators' 11-page reply to the WPC brief, hosted by LEAD's operator.

    Reliability: Has a stake in this finding. The evaluators defending their own work; they say they did not receive or administer LEAD program funding. Person-level counts in it are not in the published papers.

    • The evaluators accept WPC's point about the wording and say they are correcting it.

      What the source says
      One point in the Brief is well-taken and is being corrected.
      p. 1

    • The share of LEAD participants arrested fell from 83% to 58%, while the comparison group's rose from 77% to 80%.

      What the source says
      Pre-to-post arrest prevalence fell from 83% to 58% among LEAD participants and rose from 77% to 80% among controls
      p. 2, Aim 1

    • Contradicts: Counting people, 11 of 162 participants without jobs at the start were employed at some point during follow-up.

      What the source says
      Of the 162 participants not employed at baseline, 11 became newly employed at some point during the follow-up, bringing the total employed during follow-up to 23.
      p. 5

    • Contradicts: Counting people, 45 of 145 who lacked permanent housing at referral obtained it, 31%, not 13%.

      What the source says
      of the 145 participants who were documented as lacking permanent housing at referral, 45 obtained it during the follow-up, which is 31% not 13%.
      p. 9

    • Qualifies: The evaluators concede felony charges are a weaker measure of behavior than arrests.

      What the source says
      felony charging is a less clean indicator of behavior change than arrest
      p. 7
  7. Party to the dispute University of Washington HaRRT Center. HaRRT Center's Response to the Washington Policy Center's "National Retraction Brief" (August 16, 2026): employment, housing, deaths and warrants

    Further points from the evaluators' reply.

    Reliability: Has a stake in this finding. The evaluators defending their own work; the housing and death figures are their own calculations, not peer reviewed.

    • Contradicts: 81% of participants unhoused at baseline had shelter or permanent housing at some point during follow-up.

      What the source says
      Of the 145 participants who were documented as unhoused at baseline, 81% had shelter or permanent housing at some point during follow-up
      p. 9

    • Contradicts: Deaths were a cumulative count over the project, while the job figure was a monthly average, so the two can't be compared directly.

      What the source says
      The deaths are a cumulative person count across the 3-yr active project period
      p. 4

    • Qualifies: The evaluators confirm a plain employed-or-not measure did not change significantly.

      What the source says
      Binary employment status (employed versus unemployed) did not change significantly.
      p. 2, Aim 3

    • Qualifies: The evaluators argue fewer warrant arrests are partly how LEAD works, not a flaw in the comparison.

      What the source says
      Reductions in warrant arrests are partly a mechanism of the intervention, not a confound of the outcome.
      p. 6

    • After entry, 20% of LEAD participants and 31% of the comparison group were charged with a felony, unadjusted.

      What the source says
      consistent with our published raw prevalences (20% versus 31% post-entry)
      p. 7
  8. Peer-reviewed study Clifasefi, Lonczak and Collins, University of Washington. Seattle's Law Enforcement Assisted Diversion (LEAD) Program: Within-Subjects Changes on Housing, Employment, and Income/Benefits Outcomes and Associations With Recidivism. Crime & Delinquency 63(4) (2017): 429-445, corrected September 2026 Free copy DOI: 10.1177/0011128716687550

    The evaluators' study of housing, jobs and income for 176 LEAD participants, with no comparison group.

    Reliability: Has a stake in this finding. Written by the program's evaluators; single-group design, so it cannot show cause.

    • The paper was corrected in September 2026 to replace likelihood wording with odds wording.

      What the source says
      This article has been updated to correct several descriptions of odds ratios in the Abstract, Results, and Discussion sections
      Correction notice, p. 1

    • Contradicts: The published job and housing percentages after referral are averages over each month of follow-up, not counts of people.

      What the source says
      Postreferral values are comprised of the percentage of individuals fitting that category averaged over each month of the 18-month follow-up period.
      Table 1, note

    • Qualifies: With no comparison group, the housing and job changes can't be credited to LEAD.

      What the source says
      Causal conclusions cannot therefore be drawn based on these findings.
      Discussion, limitations
  9. Government U.S. Department of Justice, Office of Justice Programs, CrimeSolutions. Program Profile: Law Enforcement Assisted Diversion (LEAD) Program (Seattle, Washington)

    The Justice Department's rating of LEAD's evidence.

    Reliability: Federal program-rating service; reviewers scored the single 2017 study. The page says it is no longer updated.

    • The federal review describes the result as lower odds of arrest, rated Promising, based on one study.

      What the source says
      Individuals who participated in the program had 56 percent lower odds of being arrested at least once
      Outcomes: Arrests

    • Qualifies: The federal profile rates LEAD Promising, but its rating line itself says likelihood rather than odds.

      What the source says
      The program is rated Promising for reducing the likelihood of being arrested and being charged with a felony offense.
      Program summary

    • Qualifies: The profile lists the 2017 Collins, Lonczak and Clifasefi paper as the study behind its ratings.

      What the source says
      Collins, Susan E., Heather S. Lonczak, and Seema L. Clifasefi. 2017.
      Evidence base
  10. Government Seattle Police Department. Study: Innovative Seattle/King County Diversion Program Shows Success (SPD Blotter, April 8, 2015)

    The police department's post on the evaluation, still online.

    Reliability: Has a stake in this finding. SPD is a LEAD partner on its policy coordinating group.

    • The police department repeated the 58% less likely claim, and the post is unchanged.

      What the source says
      People in LEAD were 58% less likely than people in the control group to be arrested.
      Third paragraph
  11. Party to the dispute LEAD Seattle-King County (Internet Archive copy, April 29, 2026). Evaluation page for the 2017 housing and employment paper, as archived

    The program's website before the 2026 dispute.

    Reliability: Has a stake in this finding. Website of LEAD, run by Purpose Dignity Action.

    • In April 2026 the program's own site still said participants were 58% less likely to be arrested.

      What the source says
      LEAD participants 58% less likely to be arrested for any crime in any jurisdiction compared to a similarly situated control group.
      Page summary
  12. Party to the dispute LEAD Seattle-King County. Evaluation page for the 2017 housing and employment paper (current)

    The same page after the program changed its wording.

    Reliability: Has a stake in this finding. Website of LEAD, run by Purpose Dignity Action.

    • The program's page now says 58% lower odds of re-arrest.

      What the source says
      LEAD participants had 58% lower odds of re-arrest for any crime in any jurisdiction, compared to a similarly-situated control group.
      Page summary
  13. Party to the dispute Purpose Dignity Action (LEAD), Seattle City Council Public Safety Committee. Diversion & LEAD utilization since Seattle drug ordinance adoption (April 28, 2026)

    LEAD's presentation in the committee's agenda materials.

    Reliability: Has a stake in this finding. Presented by LEAD's operator while the council weighed LEAD funding and expansion.

    • LEAD told the City Council in 2026 that it reduced recidivism by 58% and felony charges by 39%.

      What the source says
      Independent UW researchers funded by Laura & John Arnold Foundation found LEAD reduced recidivism for all crimes by 58%, reduced felony charges by 39%
      Slide: Evaluation & research
  14. News report The Center Square (Carleen Johnson), republished by News-Shield. WATCH: Creator defends drug diversion program amid conflicting success claims (August 12, 2026)

    Interview with LEAD co-creator Lisa Daugaard.

    Reliability: Wire story from The Center Square, a nonprofit news service, republished by a member paper.

    • LEAD's co-creator agreed WPC was right that the figure was oversimplified, yet still called it a reduced likelihood.

      What the source says
      It is a 58% reduced likelihood of being arrested compared to a comparison group. But WPC is not wrong.
      Daugaard quote · Lisa Daugaard, LEAD co-creator
  15. Advocacy group Washington Policy Center (Eric Zimmerman). National Retraction Brief: Widespread and systematic factual errors in Law Enforcement Assisted Diversion (LEAD), Part 1, version 1.3 (August 16, 2026)

    The critique of LEAD's evidence claims.

    Reliability: Has a stake in this finding. Free-market think tank; the brief recommends pausing LEAD expansion and restoring arrest and jail for drug crimes.

    • Qualifies: WPC says no clear difference in arrest rates can be claimed.

      What the source says
      No clear difference in arrests rates can be reliably claimed, but also cannot be altogether ruled-out.
      Section III.B, p. 4 · Eric Zimmerman, Washington Policy Center

    • Qualifies: WPC argues LEAD's advantage comes partly from fewer warrant arrests after dropped or reduced charges.

      What the source says
      The LEAD group appears to benefit from reduction in warrant arrests due to a higher rate of dropped and reduced charges, while the control group does not.
      Section III.B, p. 4 · Eric Zimmerman, Washington Policy Center

    • Qualifies: WPC says three times as many LEAD participants died as gained jobs.

      What the source says
      compared to the small number estimated as gaining jobs during the study, three times as many LEAD participants died.
      Section I, p. 2 · Eric Zimmerman, Washington Policy Center

    • Qualifies: WPC put LEAD's permanent-housing success rate at 13%.

      What the source says
      This is equivalent to a 13% success rate in securing permanent housing for the LEAD participants who lacked such housing upon enrollment.
      Section III.E, pp. 6-7 · Eric Zimmerman, Washington Policy Center

    • Qualifies: WPC built its job figure from the average number employed in a given month, then compared it with the number employed at entry.

      What the source says
      an estimated average of 16 (9.03%) had employment in a given month of the 18-month follow-up period
      Section III.D, p. 5 · Eric Zimmerman, Washington Policy Center
  16. News report MyNorthwest / KIRO Newsradio. John Curley Show article on LEAD (August 12, 2026)

    Article on Eric Zimmerman's radio interview.

    Reliability: Talk-radio station article quoting Zimmerman's on-air remarks.

    • Qualifies: Zimmerman said only three of about 163 participants without jobs gained jobs.

      What the source says
      how many additional people in that group gained jobs?…The answer is three
      Zimmerman quote on employment · Eric Zimmerman, Washington Policy Center

    • Qualifies: Zimmerman put the number of participants without jobs at the start at 163 or 164.

      What the source says
      163 or 164 people didn’t have jobs when they came into LEAD
      Zimmerman quote on employment · Eric Zimmerman, Washington Policy Center
  17. Government Congressional Record, U.S. House. Comprehensive Opioid Abuse Reduction Act of 2016, Amendment No. 3 offered by Ms. DelBene (May 12, 2016), p. H2307

    House floor debate on adding law-enforcement diversion programs to federal opioid grants.

    Reliability: Official record of House proceedings.

    • Rep. DelBene told the House LEAD reduced recidivism by as much as 60%, citing the initial study.

      What the source says
      According to an initial study, it successfully reduces recidivism by as much as 60 percent.
      p. H2307 · Rep. Suzan DelBene (D-Washington)
  18. Government Congressional Record, U.S. House. Explanatory statement, Further Consolidated Appropriations Act, 2020 (December 17, 2019), p. H10966

    The enacted 2020 budget statement for Justice Department grants.

    Reliability: Joint explanatory statement accompanying an enacted appropriations law.

    • The 2020 budget reserved at least $10 million of the opioid grant program for LEAD sites.

      What the source says
      of which no less than $10,000,000 shall be made available for additional replication sites employing the Law Enforcement Assisted Diversion (LEAD) model
      Comprehensive Opioid, Stimulant, and Substance Abuse Program, p. H10966
  19. Government U.S. Senate Committee on Appropriations. Senate Report 119-44, Commerce, Justice, Science appropriations bill, 2026

    Committee report made binding by the enacted January 2026 explanatory statement.

    Reliability: Congressional committee report; the enacted statement gives its language the same weight as its own.

    • The 2026 budget reserves at least $11 million of the $185 million opioid grant program for LEAD sites.

      What the source says
      Of the $185,000,000 for COSSUP, not less than $11,000,000 shall be made available for additional replication sites employing the Law Enforcement Assisted Diversion [LEAD] model
      Comprehensive Opioid, Stimulant, and Substance Use Program
  20. Government U.S. Congress. Full-Year Continuing Appropriations and Extensions Act, 2025 (Public Law 119-4)

    The full-year stopgap law that funded the 2025 budget year at 2024 terms.

    Reliability: Enacted public law.

    • Qualifies: The 2025 Justice Department budget was funded by continuing the 2024 appropriations law.

      What the source says
      division C of Public Law 118-42
      Sec. 1101(a)
  21. Advocacy group Washington Policy Center (David Boze). U.S. Congress misled by false LEAD crime-reduction claims, leading to nearly $100 million in federal funding (September 9, 2026)

    Summary of Part 2 of the WPC brief, on federal funding.

    Reliability: Has a stake in this finding. Free-market think tank; the linked PDF lists set-asides by year but gives no total.

    • Qualifies: WPC says false claims led Congress to nearly $100 million in LEAD funding.

      What the source says
      U.S. Congress misled by false LEAD crime-reduction claims, leading to nearly $100 million in federal funding
      Headline
  22. FactHub record FactHub Research Team. Saved analysis: how much less often LEAD participants were arrested after entry

    Arithmetic on the published after-entry arrest shares.

    • After entry, LEAD participants were 22 points, or 27.5%, less likely to be arrested in raw terms; 21.2% for non-warrant arrests.

  23. FactHub record FactHub Research Team. Saved analysis: what the reported lower odds mean for the chance of arrest

    Conversion of the reported odds into a chance of arrest.

    • With 80% of the comparison group arrested, 58% lower odds equals about a 22% lower chance of arrest.

  24. FactHub record FactHub Research Team. Saved analysis: odds estimates as the evaluators reported them

    The printed odds ratios, percentages and p-values.

    • Qualifies: The printed odds ratios don't match the stated percentages by simple subtraction, and the papers don't explain the conversion.

  25. FactHub record FactHub Research Team. Saved analysis: LEAD set-asides in enacted federal budgets, 2018-2026

    Year-by-year LEAD set-asides from enacted budget statements, beside WPC's list.

    • Enacted LEAD set-asides total $72.5 million for 2018-2026; WPC's own list sums to about $80 million.

  26. Government Congressional Record, U.S. House. Explanatory statement, Consolidated Appropriations Act, 2018 (March 22, 2018), p. H2093

    The enacted 2018 budget statement for Justice Department grants.

    Reliability: Joint explanatory statement accompanying an enacted appropriations law.

    • The enacted 2018 statement adopted the Senate report language on LEAD.

      What the source says
      Senate report language on the Law Enforcement Assisted Diversion (LEAD) model is adopted.
      p. H2093
  27. Government U.S. Senate Committee on Appropriations. Senate Report 115-139, Commerce, Justice, Science appropriations bill, 2018

    The Senate report whose LEAD language the 2018 budget adopted.

    Reliability: Congressional committee report, adopted on this point by the enacted statement.

    • The adopted 2018 language reserved at least $2.5 million for LEAD sites.

      What the source says
      Of the $21,000,000 for COAP no less than $2,500,000 shall be made available for additional replication sites employing the Law Enforcement Assisted Diversion [LEAD] model
      Comprehensive Opioid Abuse Program
  28. Government Congressional Record, U.S. House. Explanatory statement, Consolidated Appropriations Act, 2019 (February 13, 2019), p. H1811

    The enacted 2019 budget statement for Justice Department grants.

    Reliability: Joint explanatory statement accompanying an enacted appropriations law.

    • The 2019 budget reserved at least $5 million for LEAD programs.

      What the source says
      including no less than: $5,000,000 for Law Enforcement Assisted Diversion (LEAD) programs
      Comprehensive Opioid Abuse Program, p. H1811
  29. Government Congressional Record, U.S. House. Explanatory statement, Consolidated Appropriations Act, 2021 (December 21, 2020), p. H7942

    The enacted 2021 budget statement for Justice Department grants.

    Reliability: Joint explanatory statement accompanying an enacted appropriations law.

    • The 2021 budget reserved at least $11 million for LEAD sites.

      What the source says
      Of the $185,000,000 for COSSAP, no less than $11,000,000 shall be made available for additional replication sites employing the Law Enforcement Assisted Diversion (LEAD) model
      COSSAP, p. H7942
  30. Government U.S. House Committee on Appropriations. House Report 117-97, Commerce, Justice, Science appropriations bill, 2022

    House report the enacted 2022 statement made binding.

    Reliability: Congressional committee report; the enacted statement gave its language the same weight as its own.

    • The 2022 House report, made binding by the enacted statement, kept at least $11 million for LEAD-model sites.

      What the source says
      no less than $11,000,000 shall continue to be made available for sites employing models consistent with LEAD
      COSSAP
  31. Government U.S. House Committee on Appropriations. House Report 117-395, Commerce, Justice, Science appropriations bill, 2023

    House report the enacted 2023 statement made binding.

    Reliability: Congressional committee report; the enacted statement gave its language the same weight as its own.

    • The 2023 House report, made binding by the enacted statement, kept at least $11 million for LEAD-model sites.

      What the source says
      no less than $11,000,000 shall continue to be made available for sites employing models consistent with LEAD
      COSSUP
  32. Government U.S. Senate Committee on Appropriations. Senate Report 118-62, Commerce, Justice, Science appropriations bill, 2024

    Senate report the enacted 2024 statement made binding.

    Reliability: Congressional committee report; the enacted statement gave its language the same weight as its own.

    • The 2024 Senate report, made binding by the enacted statement, reserved at least $11 million for LEAD sites.

      What the source says
      no less than $11,000,000 shall be made available for additional replication sites employing the Law Enforcement Assisted Diversion [LEAD] model
      COSSUP
  33. Government Congressional Record. Explanatory statement, Consolidated Appropriations Act, 2022 (March 9, 2022)

    Joint explanatory statement accompanying the enacted appropriations law; it gives the named committee report the force of its own text.

    Reliability: Official explanatory statement for an enacted appropriations law, printed in the Congressional Record.

    • The enacted 2022 statement gives House Report 117-97, which holds the $11 million LEAD floor, the same weight as its own text.

      What the source says
      97 (``the House report'') carries the same weight as language included in this joint explanatory statement
      Division B, Commerce, Justice, Science
  34. Government Congressional Record. Explanatory statement, Consolidated Appropriations Act, 2023 (December 20, 2022)

    Joint explanatory statement accompanying the enacted appropriations law; it gives the named committee report the force of its own text.

    Reliability: Official explanatory statement for an enacted appropriations law, printed in the Congressional Record.

    • The enacted 2023 statement gives House Report 117-395, which holds the $11 million LEAD floor, the same weight as its own text.

      What the source says
      395 (``the House report'') carries the same weight as language included in this joint explanatory statement
      Division B, Commerce, Justice, Science
  35. Government Congressional Record. Explanatory statement, Consolidated Appropriations Act, 2024 (March 5, 2024)

    Joint explanatory statement accompanying the enacted appropriations law; it gives the named committee report the force of its own text.

    Reliability: Official explanatory statement for an enacted appropriations law, printed in the Congressional Record.

    • The enacted 2024 statement gives Senate Report 118-62, which holds the $11 million LEAD floor, the same weight as its own text.

      What the source says
      62 (``the Senate report'') carries the same weight as language included in this joint explanatory statement
      Division C, Commerce, Justice, Science
  36. Government Congressional Record. Explanatory statement, Commerce, Justice, Science appropriations act, 2026 (January 8, 2026), p. H255

    Joint explanatory statement accompanying the enacted appropriations law; it gives the named committee report the force of its own text.

    Reliability: Official explanatory statement for an enacted appropriations law, printed in the Congressional Record.

    • The enacted 2026 statement gives Senate Report 119-44, which holds the $11 million LEAD floor, the same weight as its own text.

      What the source says
      Senate Report 119-44 (``the Senate report'') carries the same weight as language included in this joint explanatory statement
      Division A, p. H255
  37. FactHub dataset Seattle LEAD evaluation: arrest rates, reported odds and federal set-asides

    • Tables: arrest-rates

      Analysis arrest-rates-whole

    • Tables: arrest-rates

      Analysis post-entry-gap

    • Tables: arrest-rates, reported-estimates

      Analysis odds-to-probability

    • Tables: reported-estimates

      Analysis reported-estimates

    • Scope: Entire dataset release

      Analysis federal-set-asides

    • Over the entire course of the evaluation to date, people in LEAD were 58% less likely than people in the control group to be arrested
      Susan E. Collins, Seema L. Clifasefi and Heather S. Lonczak, University of Washington LEAD evaluation team, · In the dataset
    • Independent UW researchers funded by Laura & John Arnold Foundation found LEAD reduced recidivism for all crimes by 58%, reduced felony charges by 39%
      Purpose Dignity Action (LEAD), LEAD's operator, presenting to the Seattle City Council Public Safety Committee, · In the dataset
    • It is a 58% reduced likelihood of being arrested compared to a comparison group.
      Lisa Daugaard, LEAD co-creator, Purpose Dignity Action, · In the dataset
    • No clear difference in arrests rates can be reliably claimed, but also cannot be altogether ruled-out.
      Eric Zimmerman, Policy Director, Public Safety Solutions, Washington Policy Center, · In the dataset
    • Let's say 163 or 164 people didn't have jobs when they came into LEAD, and how many additional people in that group gained jobs? [...] The answer is three
      Eric Zimmerman, Washington Policy Center, · In the dataset
    • This is equivalent to a 13% success rate in securing permanent housing for the LEAD participants who lacked such housing upon enrollment.
      Eric Zimmerman, Policy Director, Public Safety Solutions, Washington Policy Center, · In the dataset
    • For example, compared to the small number estimated as gaining jobs during the study, three times as many LEAD participants died.
      Eric Zimmerman, Policy Director, Public Safety Solutions, Washington Policy Center, · In the dataset
    • According to an initial study, it successfully reduces recidivism by as much as 60 percent.
      Suzan DelBene, U.S. Representative (D-Washington), · In the dataset
    • U.S. Congress misled by false LEAD crime-reduction claims, leading to nearly $100 million in federal funding
      Washington Policy Center (David Boze), Free-market think tank, · In the dataset
  38. FactHub chart lead-seattle-arrest-share

Update history 1 change, last

  1. Initial version.

Feedback these versions applied

For developers: ClaimReview data

This page includes standardized ClaimReview metadata describing the exact claim, FactHub's finding, publisher, and review dates. You can inspect the record below and confirm that it matches the article you just read.

View ClaimReview JSON 
{"@context":"https://schema.org","@type":"ClaimReview","url":"https://facthub.org/n/article/lead-seattle-58-percent-arrests/4","datePublished":"2026-11-03","dateModified":"2026-11-03","author":{"@type":"Organization","name":"FactHub","url":"https://facthub.org/"},"claimReviewed":"People in LEAD were 58% less likely than people in the control group to be arrested.","reviewRating":{"@type":"Rating","alternateName":"Misleading"},"image":"https://facthub.org/node-assets-v2/article/lead-seattle-58-percent-arrests/4/spd-car/social-1200x630.jpg"}

This metadata helps machines discover the fact-check. It does not prove the conclusion or guarantee special treatment by a search engine.

Challenge this fact-check

Every source above is linked so you can check it yourself. Identify the specific conclusion you dispute and provide verifiable evidence that could change it. Credible challenges deserve review and a visible response.

Submit evidence Send general feedback instead

Technical details: node handle article:lead-seattle-58-percent-arrests