The Problem of Suspended and Revoked Drivers Who Avoid Detection at Checkpoints

Objective: Although driver license suspension and revocation have been shown to improve traffic safety, suspended or revoked (SR) drivers who continue to drive—which appears to be the majority—are about 3 times more likely to be involved in crashes and to cause a fatal crash. In California and many other U.S. states, drivers are typically mailed notices requesting that they surrender their licenses when they are SR for reasons other than driving under the influence of alcohol or drugs (DUI), yet they frequently do not comply. Typical procedures at DUI checkpoints in California and other U.S. states include inspecting driver licenses and checking for signs of intoxication during brief contacts with law enforcement officers. Hence, these checkpoints are in fact DUI/license checkpoints in California and many other states. The purpose of this study was to estimate the extent to which SR drivers avoid being detected at DUI/license checkpoints for SR driving, because they illegally retained possession of their license cards. Method: Law enforcement officers used electronic license card readers at DUI/license checkpoints in Sacramento, California, to record data for 13,705 drivers. The SR status of all contacted drivers was determined after the checkpoints and compared to law enforcement citation records from the checkpoints. Results: Although only 3% of the drivers contacted at the checkpoints were SR, about 41% of SR drivers were able to pass through undetected because they presented license cards that they illegally retained. Drivers SR for DUI-related reasons were more likely to be detected, whereas those SR for failure to provide proof of financial responsibility (insurance) were less likely to be detected. Conclusion: The fact that many SR drivers are able to pass through DUI/license checkpoints undetected weakens both the specific and general impacts of checkpoints for deterring SR driving and may diminish the effectiveness of suspension and revocation actions for reducing the crash risk posed by problem drivers. Using license card readers that can quickly identify SR drivers in real time during routine traffic stops and at DUI/license checkpoints warrants further consideration.


Introduction
Driver license suspension and revocation are countermeasures intended to reduce the driving risk posed by problem drivers. In California and other U.S. states, licensees can have their driving privileges suspended or revoked (SR) for various reasons related directly (e.g., driving under the influence of alcohol or drugs [DUI]) or indirectly (e.g., failure to carry vehicle liability insurance) to traffic safety and also for nondriving-related behaviors (e.g., failure to pay child support). Compared to validly licensed drivers, those under suspension or revocation for any reason have elevated crash and traffic conviction rates (Gebers and DeYoung 2002) and are about 3 times more likely to be involved in fatal crashes and to cause a fatal crash (Brar 2012). Approximately 1.3 million California licensees are SR at any given time, representing about 4% of all licensed California drivers. Suspension and revocation have been shown to be effective specific and general deterrents for reducing crashes and traffic violations (Masten and Peck 2004;Wagenaar and Maldonado-Molina 2007) and for decreasing DUI recidivism (Rogers 1997;Tashima and Marelich 1989). Driving while under suspension or revocation in California is a misdemeanor (California Vehicle Code [CVC] §14601), as is possessing an SR license card (CVC §14610). Any vehicle operated by a driver under suspension or revocation is subject to impoundment (CVC §14602.6) and, if the driver has prior convictions for driving while SR and owns the vehicle, forfeiture (CVC §14607.6). Vehicle impoundment is an effective specific deterrent for reducing subsequent traffic convictions and crashes among SR drivers who are caught driving (DeYoung 1999;Voas and DeYoung 2002). Despite these potential consequences, it is estimated that as many as 75% of SR drivers continue driving during their suspension or revocation period (Coppin and Van Oldenbeek 1965;Hagen et al. 1980;Lenton et al. 2010; Ross and Gonzales 1988), although they report driving less often and more carefully to avoid detection (Clark and Bobveski 2008;Ross and Gonzales 1988). Drivers SR for any reason are estimated to represent about 6% of all driving exposure on California roadways (Brar 2012).
Many SR drivers in the United States do not surrender their license cards as required by law when they are notified of their suspension or revocation actions. Drivers in California and other U.S. states are typically ordered to surrender their licenses when they receive a written order of suspension or revocation from their driver licensing agency. For example, in California the orders are mailed to SR drivers and (in most cases) direct them to return any licenses in their possession via mail or by bringing them to a Department of Motor Vehicles (DMV) office or, if they do not have possession of their licenses, to indicate the reason they no longer have them. For DUI offenders in California and most other states, licenses are usually confiscated by law enforcement during arrest if an administrative per se (APS) license suspension action is initiated, or license surrender may be ordered as part of postconviction suspension or revocation actions, which are again typically implemented by mailed orders. Judges will also occasionally confiscate driver licenses during court proceedings.
In all cases where license suspension or revocation orders are served, there exist pathways by which drivers can circumvent the license surrender processes. For example, drivers given an APS suspension-for whom law enforcement officers immediately collect any license in their possession-could apply for a duplicate license prior to when the licensing agency receives and processes the APS notice of suspension, which in California can take up to 10 days after the arrest or detention. Drivers ordered to surrender their licenses by mail could simply ignore the orders or acknowledge and return the orders but falsely claim that their license cards were lost. The extent to which SR drivers illegally retain possession of their license cards is unknown, but it is thought to be a frequent occurrence in California given the large volume of suspension or revocation orders that are returned as undeliverable (estimated to be about 25%) and the fact that many successfully delivered orders are subsequently returned without surrendered licenses attached. Retaining license cards may protect SR drivers from detection by law enforcement. For example, if these SR drivers are stopped by law enforcement for traffic violations and the officers do not electronically check the status of their licenses against licensing agency databases, the drivers may not be detected as being SR.
In addition to license checks during routine traffic stops, another enforcement measure that is used to deter suspended and revoked driving is law enforcement checkpoints/roadblocks. For example, although the primary purpose of DUI checkpoints is to provide a general deterrent against DUI (NHTSA 2008), another important function is to detect persons driving without valid licenses and remove them from the road. This results from the fact that procedures at DUI checkpoints in California and other U.S. states typically include inspecting driver license cards (Brown 1991). Hence, these checkpoints are in fact DUI/license checkpoints in California and many other states.
The interaction between law enforcement and drivers who enter DUI/license checkpoints in California and other U.S. states is usually brief (under 30 s) and typically involves hav-ing officers look for signs of alcohol or drug impairment and visually check whether the license cards appear valid (Brown 1991). There is no way to know that drivers are under suspension or revocation just by looking at their license cards, yet for most drivers who enter checkpoints there are no electronic checks of license validity using licensing agency records. Under the current license screening method used during DUI/license checkpoints in most U.S. jurisdictions, SR drivers who have not properly surrendered their licenses will only be identified if they are flagged for further DUI screening, present an expired license, or do not present a license for inspection. Therefore, many SR drivers in California and elsewhere may pass through these checkpoints undetected, which would weaken both the specific and general impact of the checkpoints for deterring suspended and revoked driving, as well as potentially diminish the effectiveness of suspension and revocation actions for reducing the driving risk posed by problem drivers.
Given the prevalence and high-risk nature of suspended and revoked drivers in general, improved means of identifying and removing them from the road is a desirable traffic safety objective. The purpose of the present study was to estimate the percentage of drivers who avoid being detected at DUI/license checkpoints for driving while suspended or revoked because they illegally retained possession their license cards. We also investigated whether detection rates differed as a function of the reason for the suspension or revocation.

Checkpoint Data Collection Procedures
The Sacramento Police Department Metro-DUI Enforcement Team electronically captured the driver license (DL) numbers of all drivers who presented a DL at 17 DUI/license checkpoints between December 2012 and August 2013, using Veriscan M-310 Handheld license card readers. The Veriscan M-310 Handheld license card readers were purchased by the Sacramento Police Department from: IDScan.net New Orleans, Louisiana. These card readers scan bar codes or magnetic stripes on licenses and identification cards and are capable of reading data for cards issued in all U.S. states. The card readers were used to record DL data only; they did not check or alert officers about license validity, and functionality that would normally alert users of license expiration was disabled. For licenses that could not be read by the card readers, or in cases when officers did not have access to scanners, the officers recorded the DL numbers on paper logs created for this purpose. About 2% of licenses could not be read by the card readers, typically because they did not have 2-dimensional bar codes (i.e., older licenses) or the magnetic stripes were damaged. In cases where drivers did not have a DL in their possession, officers searched for the drivers in the DMV database and obtained a DL number if one existed. Except for the added step of scanning or hand recording the DL numbers of contacted drivers, officers did not alter their typical checkpoint procedures, which are described below.

Typical Law Enforcement Checkpoint Procedures
The typical procedures followed by the Sacramento Police Department Metro-DUI Enforcement Team during the 17 DUI/license checkpoints are detailed in this section.
Traffic at the checkpoint locations is funneled down to one lane, with an adjacent lane coned off to provide standing space for law enforcement officers and volunteers who distribute DUI-related literature to drivers at the end of each contact. Eight to 10 uniformed law enforcement officers are spaced at approximately one-car-length intervals in this adjacent lane. The lead vehicle entering the checkpoint is stopped, creating a group of vehicles. Officers approach each vehicle, ask the drivers to present their DLs, and quickly check for indications of alcohol or drug impairment. The contacts are brief, generally less than 30 s. Drivers with license cards that appear to be valid and who show no signs of impairment are allowed to exit the checkpoint once the lane clears in front of their vehicle. Drivers who present an expired DL or a suspect DL (e.g., it does not appear to belong to the driver) or who do not present a DL are directed into a triage area to have their license statuses verified against the DMV licensing database. Drivers showing signs of impairment are taken into the triage area for further DUI screening. When all contacted vehicles have exited the checkpoint, the next group of vehicles is stopped and the procedure is repeated. Drivers detained for license issues or DUI testing are either cited, arrested, or released depending on their status after further screening. There is a zero-tolerance policy in regard to suspended and revoked drivers at the checkpoints. If they are detected, regardless of the reason for the suspension or revocation, they are always cited and their vehicles are towed and impounded for up to 30 days.

Data Processing and Analysis
Data were downloaded from the card readers after each checkpoint, and copies of citations and arrest logs were collected. License numbers and other identifying information from the card readers, paper logs, citations, and arrest logs were compiled into a database and compared to police volunteer staff's independent counts of vehicles that passed through the checkpoints. Typically the license counts were within 2% of the independent counts of vehicles, indicating that the officers indeed recorded the license numbers for most contacted drivers.
The license status of the contacted drivers during the night of each checkpoint was subsequently checked against the licensing database of the California DMV and compared with arrest and citation data from the checkpoints. The percentages of SR drivers who passed through the checkpoints undetected were calculated overall and separately by the particular reasons the drivers were SR (e.g., DUI, failure to appear for a court date, or a non-driving-related reason). Because a driver's privilege can be simultaneously SR for multiple reasons, the earliest suspension or revocation action still in effect that should have required surrender of the license card was used to classify the reason for the suspension or revocation. Note that for drivers with out-of-state licenses, it was not possible to determine license status because their records are not present in the California DMV database.
The drivers' reasons for suspension or revocation were grouped into 8 categories for presentation purposes: 1. APS-related reasons, which are typically due to arrests or detainments of drivers with blood alcohol concentrations in excess of the legal limits for their age and license class; 2. DUI-related reasons, which are related to convictions for DUI; 3. physical and mental (P&M) conditions/lack of skill reasons, which are due to P&M impairments that affect safe driving ability (e.g., dementia), or poor driving skills not otherwise linked to a known condition; 4. negligent operator/serious offender reasons, which are related to the accumulation of excessive demerit points for traffic violation convictions and at-fault crashes, or due to convictions for egregious driving behaviors (e.g., hit-andrun crashes or reckless driving); 5. failure to appear/failure to pay reasons, which are due to failures to appear for court hearings or pay court-levied fines; 6. financial responsibility-related reasons, which are associated with failures to provide proof of vehicle liability insurance under required conditions, such as for crash involvements reported to the DMV; 7. non-driving-related reasons, which are for issues such as failure to pay child support and graffiti convictions; and 8. out-of-state suspension/revocation reasons, which are due to being SR in another state as indicated on the National Driver Register Problem Driver Pointer System.

Results
A total of 13,705 drivers were contacted at the 17 DUI/license checkpoints. Overall, 89.2% had a valid California license, 3.3% were SR, 2.4% were unlicensed or had an expired license, 2.8% were licensed out-of-state, and for 2.2% the licensing status could not be determined; see Figure 1 and when DL numbers were not read by the card readers due to damaged bar codes and when the officers made errors while entering DL numbers on the paper logs. Overall at least 5.7% of contacted drivers had a license problem that would have been of interest to law enforcement, had they all been detected during the checkpoints. Among the 453 SR drivers contacted, the most common reason for suspension or revocation was failure to appear/failure to pay (60.3%, Table 1 and Figure 2). The next most common reasons were for a DUI conviction (12.4%), an APS arrest (8.4%), a non-driving-related reason (7.1%), failure to provide proof of financial responsibility (5.7%), a negligent operator or serious driving offense (3.3%), a P&M condition or lack of driving skill (2.2%), and, finally, an out-of-state suspension or revocation (0.7%). Therefore, the majority of SR drivers contacted at the checkpoints (92.9%) had their licenses SR for a reason that was directly or indirectly related to a traffic safety issue.
Though the majority (59.2%) of the SR drivers contacted at the checkpoints were cited by law enforcement, 40.8% were not cited for driving with a SR license (Table 1). The percentages Fig. 2. Distribution of reasons that drivers contacted at checkpoints were suspended or revoked. detected (Figure 3) differed as a function of the reasons for the suspensions or revocations, χ 2 (7, N = 453) = 29.06, P < .001. Specifically, the majority of drivers suspended for failure to provide evidence of financial responsibility were able to pass through the checkpoints undetected (73.1%), whereas fewer drivers SR because of an APS arrest (21.1%) or DUI conviction (25.0%) were undetected. Furthermore, all of the drivers SR due to an out-of-state suspension or revocation were undetected at the checkpoints, although there were few such drivers (n = 3).

General Discussion of Findings
The purpose of this study was to estimate the extent to which suspended and revoked drivers who continue to drive avoid detection at checkpoints in California because they illegally retained possession of their license cards. Although the majority of SR drivers were successfully detected by law enforcement at the checkpoints, usually because they did not present Fig. 3. Differences in law enforcement detection of drivers suspended or revoked for various reasons. a driver license for visual inspection, about 41% of them were able to pass through undetected because they presented license cards that should have been surrendered. Though the overall number of drivers contacted at the checkpoints who were SR was small-representing only 3.3% of all drivers contacted-the high percentage who were not detected by law enforcement is concerning given that identifying such drivers is one of the purposes of DUI/license checkpoints.
We also investigated differences in law enforcement detection of SR drivers as a function of the reasons for the suspensions or revocations. Drivers SR for DUI-related reasons were underrepresented among those who were able to pass through the checkpoints undetected. This would be expected because the APS procedure, which generally is part of DUI arrests in most U.S. states, includes the confiscation of the license cards by law enforcement. The fact that detection rates were lower among drivers SR for other reasons suggests that the practice used in California and other U.S. states of mailing notices ordering SR drivers to surrender their licenses is less successful for effecting license surrender.

Study Limitations
The drivers contacted at DUI/license checkpoints presented in this study are from a single geographic region of the state. Therefore, the results presented here do not necessarily generalize to other cities, counties, or regions of California or to jurisdictions outside of the state. However, to the extent that DUI/license checkpoint procedures are similar across California-as might be expected if all California agencies follow the functional and legal guidelines for checkpoints recommended by NHTSA and the courts (Ingersoll v. Palmer, 1987;NHTSA 2006)-there is no reason to suspect that SR drivers possessing valid-appearing licenses would have any less difficulty passing undetected through checkpoints in other California jurisdictions. In addition, given that DMV orderof-suspension procedures are uniform throughout California, it seems likely that the rates of illegal retention of license cards among suspended and revoked drivers are similar statewide. Though driver's licenses are typically also only visually checked at DUI checkpoints conducted in other U.S. states (Brown 1991), the procedures presented in this study should be replicated in other states to determine whether large percentages of SR drivers are similarly undetected at checkpoints conducted outside of California.
Recent California DMV estimates indicate that approximately 4% of licensees statewide and 5% of licensees in Sacramento County are SR at any given time. Using quasi-induced exposure methods, Brar (2012) estimated that SR drivers represent about 6% of the driving exposure on California roadways. The percentage of SR drivers who were contacted at the checkpoints (3.3%) underestimates the state and county percentages for several reasons. First, some drivers who are SR actually cease driving altogether, and others choose to drive less often to avoid detection (Clark and Bobveski 2008). Second, the license status of 2.2% of the contacted drivers was unknown because some license cards failed to scan and were either recorded incorrectly or not recorded at all. It seems likely that some of these drivers would have been found to be SR if it had been possible to check their license statuses. Using license card readers that are capable of reading one-dimensional bar codes-which are more universal and less prone to corruption-would likely reduce the rate of unrecorded driver license numbers in the future. Third, drivers are alerted by signage that they are approaching checkpoints and are allowed to avoid them by making legal turns before entering the coned-off areas. It seems reasonable to surmise that SR drivers may be more inclined to avoid entering checkpoints, although the extent to which this occurs is unknown. Furthermore, drivers may avoid checkpoints altogether through advance notice from websites, social media, or other communications. For example, the Sacramento Police Department gives 24-h notice on their website and typically issues press releases about upcoming checkpoints. Given all of these reasons, the estimates of suspended and revoked drivers from the present study should be considered at best a conservative lower-bound estimate of the prevalence of SR drivers who continue driving.
The number of SR drivers contacted at the DUI/license checkpoints was not large. However, the fact that so many of them were able to pass through without being detected and cited for driving while SR is troubling because it weakens both the specific and general impacts of checkpoints for deterring suspended and revoked driving and may diminish the effectiveness of suspension and revocation actions for reducing the crash risk posed by problem drivers. Though license suspensions and revocations reduce subsequent crashes and traffic violations among SR drivers (Masten and Peck 2004), the majority of SR drivers continue to drive anyway (Coppin and Van Oldenbeek 1965;Hagen et al. 1980;Lenton et al. 2010;Ross and Gonzales 1988). The effectiveness of driving-related penalties is greatly dependent upon drivers' perceptions of the likelihood of being caught (NHTSA 2008), as reflected by the fact that SR drivers who continue to drive have reported driving less often and more carefully to avoid detection (Clark and Bobveski 2008;Ross and Gonzales 1988). Nonetheless, the evidence suggests that drivers under suspension or revocation perceive their risk of detection for continued driving to be very low (Knoebel and Ross 1997). To the extent that they are indeed able to avoid detection while they drive, they will likely continue to do so, posing an elevated traffic safety risk (Brar 2012;Gebers and DeYoung 2002). Furthermore, those who are able to drive undetected avoid other consequences that are effective for controlling the risk of problem drivers, such as vehicle impoundment (DeYoung 1999;Voas and DeYoung 2002). The potential effectiveness of checkpoints-which are one of the few strategies available for deterring SR driving and reducing the associated elevated traffic safety risk-is undermined if SR drivers are able to avoid detection because they present illegally retained license cards to law enforcement and continue driving without penalty. It therefore behooves licensing and law enforcement agencies to maximize license surrender upon suspension or revocation and to improve the detection of SR drivers during checkpoints and routine traffic stops, to reduce this mechanism of nondetection and increase SR drivers' perceptions of the likelihood of being caught.
Given the findings of this and other studies that have shown that drivers under suspension and revocation for various reasons continue to drive (Coppin and Van Oldenbeek 1965;Hagen et al. 1980;McCartt et al. 2003), future efforts should focus on improving the detection of these drivers. License plate readers have been suggested as an improved detection tool (DeYoung 2013). However, license plate readers are limited to alerting law enforcement to issues related to the registered owner of the vehicle, and the driver of a vehicle is not always the owner. Methods specific to the licensee-such as using card readers that can quickly identify SR drivers in real time during routine traffic stops or at DUI/license checkpoints-may be more promising for aiding identification and enforcement and therefore warrant further investigation. Although the findings of this study indicate that a high percentage of SR drivers avoid detection at DUI/license checkpoints, it is important to note that they also show that checkpoints indeed result in the detection and at least temporary removal of a large number of SR drivers from the road. However, using card readers that can quickly identify SR drivers in real time has the potential to make these checkpoints even more effective.
DUI/license checkpoints are general deterrents for drinking and driving that are associated with reductions in alcoholrelated crashes (DeYoung 2013;NHTSA 2008). Ignition interlock devices (IIDs)-an increasingly popular countermeasure used to try to control DUI recidivism-have been shown to reduce recidivism and alcohol-related crashes among offenders while they are installed in their vehicles (Elder et al. 2011). However, this sanction is limited by the fact that many DUI offenders choose to remain SR rather than install IIDs (Elder et al. 2011). For example, in an ongoing pilot study in which all DUI offenders in 4 California counties are required to install IIDs, less than half actually do so. Like other SR drivers, those who do not install IIDs likely continue to drive. Hence, the effectiveness of IID programs like the one in California is also undermined to the extent that SR drivers are able to avoid detection because they are able to present illegally retained license cards to law enforcement.
One factor that influences whether California courts will convict drivers cited for driving on SR licenses (CVC §14601) is whether the DMV records show proof that the drivers were aware of their suspension or revocation at the time of the citation, which is called "proof-of-service." DeYoung et al. (1997) evaluated the effects of various suspension notice mailing strategies and found that using certified mail instead of firstclass mail greatly increased proof-of-service rates on DMV records, which, in turn, increased subsequent conviction rates for SR driving. The DMV began using certified mail to deliver orders of suspension and revocation based on the results of that study, but the practice was eventually discontinued due to costs, and the orders are currently delivered by regular first-class mail. Although law enforcement officers in California have access to this proof-of-service information when they verify license validity against DMV records, it does not influence whether or not they issue citations for SR driving. However, it does help determine the authority section that they use to impound the vehicles of SR drivers. Furthermore, if proof-of-service is deemed to be inadequate, the officers provide proof-of-service when they issue citations. Given that prevention of SR driving would likely be attenuated by low conviction rates for this offense that result from cases where service was deemed to be inadequate by the courts, investigating ways to improve proof-of-service in California and other U.S. states seems worthy of further investigation.