Preusser Research Group, Inc.



Task Order 11


 Breath Test Refusals in DWI Enforcement


August 2005


T.J. Zwicker, J. Hedlund, and V.S. Northrup


This is an interim report for the study titled “Determine Whether There Is an Increase in Breath Test Refusals and Develop and Evaluate a Promising Program to Deter Refusals.” The project is being conducted by Preusser Research Group under contract number DTNH22-98-D-45079 Task Order 11.


The number of alcohol-related fatalities decreased 37 percent from 25,165 in 1982 to 15,935 in 1998. However, recent numbers of alcohol-related fatalities for 2000 (16,653), 2001 (17,400), 2002 (17,419), and 2003 (17,013) indicate that fatalities continue to exceed the numbers seen in the mid-1990's. It appears that new initiatives are required to achieve additional reductions.

New initiatives to achieve additional reductions in alcohol-related fatalities require an understanding of the problem and efforts to affect it so far. In two recent studies, researchers have identified States that achieved the largest reductions in alcohol-related fatalities from 1982 to 1996 (Ulmer, Hedlund, and Preusser, under review) and examined the reductions as a function of driver age, with a particular emphasis on youth (Hedlund, Ulmer and Preusser, 2000). It appears that stronger laws have been effective in reducing the number of people who choose to drink and drive, but some believe that these same laws have produced an unwanted consequence of higher breath test refusal rates in some States. Offenders receive implied consent penalties for refusal in most States, but refusals may help offenders avoid a DWI conviction, which carries more severe penalties.

A reduction in the number of test refusals will increase the effectiveness of the administrative and criminal systems so offenders can no longer avoid penalties, may help to identify more problem drinkers, and may help identified problem drinkers get some help.


The three goals of this study are (1) to document the extent of the breath test refusal problem, (2) to investigate the reasons for breath test refusals or lack of a significant percentage of refusals in selected States, and (3) to develop, implement, and evaluate effective and efficient countermeasures to deal with the problem. The first and second goals are covered in this report. A later report will detail the results of the program implementation and evaluation.

Method A review of the administrative sanctions and criminal penalties for breath test refusal in each State, DC, and Puerto Rico was conducted. After the review, each State, the District of Columbia, and Puerto Rico were contacted to obtain breath test refusal data for the period from 1996 to 2001. Five case-study States were selected to learn more about the causes for refusals. Connecticut, Maryland, Florida, Louisiana, and Oklahoma were selected because they all had rates above the national average and provided a mix in terms of the magnitude of refusal rates and variations in impaired driving laws and practices. Refusal rates ranged from slightly above average to far above the national average. Laws pertaining to test refusals ranged from moderately weak to good. The case studies consisted of interviews with prosecutors, judges, defense attorneys, police officers, police supervisors, and administrative unit officers to (1) evaluate the arrest, breath test, administrative, and judicial processes, and (2) identify refusal problems, barriers, and potential solutions.


State laws vary widely with regard to administrative and criminal penalties for refusal. All States but one have administrative sanctions for refusal. Depending upon the State, the administrative sanctions are sometimes more stringent than those for failing a breath test. Most States do not criminalize refusal, but many admit refusals in criminal cases. Most of those that admit refusals in criminal cases do not permit refusals to be used as evidence of guilt. Some States have provisions to force a blood alcohol concentration (BAC) test after a refusal.

The distribution of refusal rates amongst States is depicted below in figure 1. The rates shown for 2001 are somewhat higher than rates reported for 1987 in an earlier study (Jones, Joksch, and Wiliszowski, 1991).

Figure 1.
2001* Breath Test Refusal Percents by State**

figure 1 - click for long description

*2000 data were used for Massachusetts and New Jersey
**Complete data was not available from AZ, CO, MO, NV, NY, SD, VA, VT, or WY

Refusal rates for some individual States differ markedly from 1987 to 2001, which may be due to real differences or due to real differences combined with differences produced by different data collection methods used by these States for assessing refusal rates in 1987 and in 2001. Refusal rates nationwide remained stable at about one-quarter of all drivers arrested for DWI from 1996 to 2001. States with statistically significant changes in refusal rates during this period were split evenly between those with increases and those with decreases. In general, the States where refusal rates decreased already had low rates, and States where rates increased already had high refusal rates. No State with a significant increase in refusals criminalizes refusal. Six of the eight States with statistically significant decreases in refusals have hard license suspensions in which no hardship or work permits are available during the suspension period.

Case Study Results

In Connecticut, 75 percent of those refusing the test are first-time offenders who would receive much less severe administrative penalties and the same criminal penalties for taking and failing a breath test. They become eligible immediately for a work permit during the entire administrative suspension period if they fail a BAC test and almost always receive the Pretrial Alcohol Education program, which results in dismissal of their criminal cases after one year. First-time offenders reportedly often refused based on a lack of understanding of these consequences.

In Maryland, the majority of the refusals are reportedly from first-time offenders. First-time offenders who fail breath tests can receive permission to drive during their entire administrative suspensions, while those who refuse and want to continue driving must have an interlock device installed for one year. First-time offenders almost always have their cases pled down to a lesser impaired driving charge and receive Probation Before Judgment (PBJ), which results in the dismissal of their criminal cases and no record of a DWI, regardless of whether they take or refuse the test. The consensus advice for first-time offenders was to take the breath test because of the reduced administrative penalties and the same PBJ outcome for their criminal cases. As in Connecticut, many first-time offenders in Maryland do not understand these consequences.

In Florida, refusal benefits all offenders arrested for DUI. The increased severity of the administrative suspension for refusal is mitigated by the availability of hardship permits that can be obtained if the person does not have a prior BAC test refusal. Both defense attorneys and prosecutors agreed that the consequences for refusal are less severe than the consequences of conviction, even for a first offense. A new law in Florida criminalizing the second refusal may lead to a reduction in refusals by repeat offenders, but the law has not been in effect long enough for State officials to determine its impact.

In Louisiana, first-time offenders have a high rate of breath test refusal. Refusing the breath test benefits the criminal cases of all offenders arrested for DWI. Without a test result, district attorneys have a much more difficult time getting a DWI conviction and usually reduce the charges to obtain a guilty plea. Refusal has the added benefit of avoiding the sanctions contained in Louisiana's high-BAC law if the offender's test result would have been .15 or higher. In at least one jurisdiction, a judge has issued warrants to order blood tests for those who refuse. The strategy may be the best solution to reduce refusals without new legislation.

In Oklahoma, repeat offenders usually refuse the breath test, but most first-time offenders reportedly take it. All those interviewed agreed that it is to the advantage of any offender to refuse the test. The administrative penalties are essentially the same for those who fail and those who refuse a breath test, which means that the criminal case outcome affects refusal more than the administrative sanctions. First-time offenders almost always have their charges pled down regardless of whether they refuse or fail the test.


In many States across the country officers are instructed to read verbatim to suspects from an Administrative License Revocation (ALR) card - the information on this card is read to the suspect when the officer requests a breath sample, and this provides the suspect with information regarding the consequences of refusing to provide the sample. Officers are often instructed to read this information verbatim to ensure that each suspect receives the same information in a uniform manner that limits any possible coercion on the part of the officer. 

In Connecticut and Maryland first-time offenders typically receive less severe penalties for failing the test, even with a high BAC, than for refusing the test. It is believed that many first-time offenders refuse the test because they do not understand these consequences. For States such as these, State officials may want to review their process for notifying suspects of both the administrative and criminal consequences of refusing to provide a breath sample. However, States such as Connecticut and Maryland must then consider the consequences of these offenders receiving less severe sanctions. In addition, if these offenders are not convicted of an alcohol-related offense and are later stopped for DWI, they will not be considered repeat offenders.

Louisiana, and 11 other States with similar laws or case law (e.g., Arizona, California, and Wisconsin), could reduce breath test refusals by encouraging officers to obtain a warrant, when needed, to draw blood for a chemical test. Warrants could be obtained for as many types of DWI arrests as judges would be willing to issue warrants, such as cases in which the arrested driver had a minor in the vehicle, cases where the driver is suspected of having a BAC above .15, or in cases of driver involvement in an injury crash. Judges who may not be willing to issue warrants to draw blood for a chemical test for the standard first-time DWI offenders may be more willing to issue a warrant for more egregious offenders. Warrants are already used in at least one jurisdiction in Louisiana. The extent of their use is not known, but they have reportedly eliminated the problem of refusals in cases where they are issued.

It is believed that Oklahoma and Florida are unlikely to reduce refusal rates substantially without new legislation. In each State, the administrative and criminal penalties for refusal are less severe than those for taking and failing the breath test.