Michael Hallstone, Ph.D., Principal Investigator Department of Justice Administration

1. Driving Under the Influencein the City & County of HonoluluA Profile of Arrestees, Case Outcomes, Collisions,Injuries, and Stakeholder Recommendations Michael Hallstone, Ph.D., Principal Investigator Department of Justice Administration University of Hawaii–West Oahu and Paul Perrone, Chief of Research and Statistics Crime Prevention & Justice Assistance Division Department of the Attorney General

2. Acknowledgments The City & County of Honolulu Police Department (HPD) Sgt. Robert Lung of HPD for coordinating the collection of arrest report data The Department of the Attorney General’s Hawaii Criminal Justice Data Center (HCJDC) Liane Moriyama and John Maruyama of the HCJDC for their assistance in providing criminal history data. Stakeholders Sergeants Robert Lung and Clyde Yamashiro of HPD; William C. Bagasol of the Office of the Public Defender; Jim Fulton and Charles P. Izumoto of the City & County of Honolulu’s, Department of the Prosecuting Attorney; Connie Abram, Carol McNamee, and Annelise Rossi of Mothers Against Drunk Driving–Hawaii, and R. Patrick McPherson, a private attorney who specializes in handling DUI cases.

3. Study Goals • To provide a more comprehensive statistical description of DUI in C&C Honolulu • To document the concerns and policy recommendations of stakeholders associated with DUI issues in Hawaii

4. Data Sources • 503 DUI arrest reports randomly selected from calendar year 2001 (approx. ¼ of all arrests) • Used 2001 as wanted a majority of cases to go through courts • Criminal history data for each DUI arrestee from Criminal Justice Information System (CJIS) • Stakeholder meeting, comments, and recommendations • No crime data are perfect but these are unique to Hawaii and are “case-level” and rare for DUI

5. “Typical DUI Arrestee” • Please see report for complete demographics • A male in his twenties to early thirties • A county resident • Only somewhat more likely to have a lower or middle income-level occupation than a higher paid profession • A Blood Alcohol Concentration (BAC) in the .129 to .141 range at the time of his arrest (the legal limit in Hawaii is .08) • And is a first-time DUI arrestee without an extensive criminal history.

6. Average BAC • Average BAC= .135 • Varies tremendously from .000 to .323 • Especially top 25% - ranges from .174 to .323 • BAC is affected by complexities of “real world” policing • Must transport to a station with “legally admissible” testing device • Making “PAS” device legally admissible evidence could possibly increase number of DUI convictions (see recommendations)

7. Criminal Histories • As a group DUI arrestees do not have extensive DUI or overall criminal histories • Future research should compare DUI arrestees to both the general population and criminal subpopulation • “Are DUI arrestees more criminally involved than the general population or criminals in general?”

8. DUI Criminal History • DUI History • Over three-quarters of the sample had no prior DUI convictions • Almost 90 percent had no more than one. • Only 5 percent of the sample had three or more prior DUI convictions at the time of their arrest. Three prior convictions mark the threshold for being charged under Hawaii’s 2001“habitual drunk driver” law (HRS §291-4.4) as well as the current statute (HRS § 291E-61.5). • Thus, small group of “persistent offenders”– (see recommendations) • Most likely for most of this group their DUI offenses are a part of general criminal involvement and social disorganization --very highly correlated (.697, p<.001) NOT IN REPORT. Plain English: people who have many DUI convictions also have many other types of criminal convictions • Errata sheet “DUI misdemeanors”

9. General Criminal History • Approximately two-thirds of the sample had no arrests or convictions of any type at the time of their arrest and, of those who did, three-quarters of the offenses fell in the misdemeanor and petty misdemeanor/violation categories. • Small group with extensive criminal histories • Again, future research should compare the criminal histories of DUI arrestees to those of both the general population and the criminal subpopulation.

10. Relating Criminal History and BAC • DUI arrestees with prior DUI arrests and/or convictions, and/or prior criminal convictions of any type (but not prior arrests) have significantly higher average BACs than do DUI arrestees without such priors • Plain English: DUI arrestees with “priors” have more alcohol in their system • However, the average difference is in the range of .02 BAC. As explained in the report, these results should be interpreted with caution.

11. Case Outcomes • High Conviction Rates • ~83% of all arrestees • ~90% of arrestees with BACs above legal limit of .08 • Perhaps gives support to the idea going through the trouble of making the “PAS” device “legally admissible evidence” • “public/private partnership” – would advocacy groups would be willing to help fund this? • alcohol industry could be asked to contribute

12. DUI-Related Motor Vehicle Collisions (“MVCs”) • DUI-related fatalities not included (or known), but topic for further research • About 30% of DUI arrests involved MVC • Half single and multiple MVCs • Slightly less than one-third of the people who were involved in DUI-related MVCs were injured. Of those who were injured, just over two-thirds sustained injuries that required emergency transportation to a hospital. • Additional research should be conducted to determine if DUI-related MVCs are more dangerous than non-DUI-related collisions in terms of the likelihood and/or severity of injuries.

13. Average BAC and MVCs • Surprising result: Average BAC actually lower in MVCs which result in an injury than in accidents that did not result in an injury • Small difference of about .01 to .02 BAC – about one drink?

14. BAC Test “Refusers” and “Consenters”Criminal Histories • 28 (5.6%) arrestees refused to take BAC test • On average, BAC test “refusers” have greater number of prior DUI arrests (1.25 vs. 0.59) than “consenters” • Also more misdemeanor, petty misdemeanor, and violation convictions of any type • Plain English: BAC test refusers have had more DUI arrests and overall contact with the criminal justice system than do BAC test consenters • This makes sense: more “savvy” defendant

15. Conviction Rates of “Refusers” and “Consenters” • Although not statistically significant: Conviction Rate for Refusers = 69% Conviction Rate for Consenters = 83% MEDIA PLEASE DO NOT REPORT!! • Too few “refusers” in sample for statistical tests -- additional research need to determine whether difference is “real”

16. Recommendations Overview • Simplify BAC test consent form • Approve PAS device BAC as admissible evidence • Focus on deterring “average citizen” • Conduct additional research

17. Simplify BAC Test Consent Form • Strong consensus amongst stakeholders that current form is too long and can be dramatically shortened • Arrestees are intoxicated while a long form is being read to them • Neighbor islands use much shorter forms successfully • Proposed revision in Appendix

18. Make PAS BAC Legally Admissible • PAS results likely higher than “station” BAC • To make PAS legally admissible • Urge the State Department of Health to approve the PAS as a blood testing device in Hawaii. • Set up a program to field test individual PAS devices on a monthly basis in order to assess and ensure accuracy. • Opportunity for “public/private” partnership? • Set up a training program to ensure that police officers use the PAS properly in the field. • Opportunity for “public/private” partnership?

19. Focus on Deterring “Average Citizen” • Huge reductions in DUI over past 25 years probably due to deterring average citizen rather than hardcore alcohol/drug addict • Average citizen is largest pool of potential DUI offenders • Threat of punishment (deterrence) most effective for people with “a stake in conventional life” • Government agencies and advocacy groups should work cooperatively to obtain both public and private funding, as appropriate, for these efforts.

20. What Works According to Research • DUI deterrence results from combined effects of a number of variables that result in: more DUI enforcement and news coverage • These result in a greater perceived risk of arrest and thus a reduction in DUI • Government and advocacy groups should work cooperatively to find “private/public partnerships” create funding to create more DUI enforcement and news coverage • Taxpayers unlikely to support “government doing it all.”

21. Download a Copy of the Report cpja.ag.state.hi.us Mahalo!