Dear Editor

It is timely that Chamberlain and Solomon [1] are proposing an extended zero blood alcohol limit for young drivers and it will definitely save more young lives; however, we believe that drinking and driving is an important part of this complex problem and we have to streamline all other Graduated Driver Licensing (GDL) components in order to realize the significant impact of GDL in saving our children in the future.

For more than a century alcohol has been recognized as one of the principal risk factors for motor vehicle crashes.[2] Alcohol-related motor vehicle crashes represent a leading cause of morbidity and mortality, particularly in young people, carrying an immeasurable human cost, as well as an enormous burden to society. Eighty-one percent of Canadians have rated drinking and driving as one of the most significant social issues they must face today, placing it ahead of other prominent issues, including health care, pollution, and the state of the economy.[3]

Young drivers continue to be a major traffic safety concern. Several factors contribute to the increased risk of traffic crash-related fatalities among adolescents and young adults including less experience driving, higher rates of drinking and driving, excessive speeding and lower rates of seat belt use. Given the significance of alcohol impaired driving to youth mortality, a key issue is to enhance the effectiveness of prevention policy and programming. Mothers Against Drunk Driving Canada has recommended that the zero BAC limits for young drivers be extended to the age of 21.[1]

Previous studies in other countries have also shown that an increased BAC limit is a difficult transition for young drivers (after GDL from a zero tolerance of alcohol to a dangerously higher BAC limit) and it was a major factor which contributed to their increased crash involvement in Australia. [4] Another recent study showed that new drivers experienced a difficult transition from the zero tolerance policy to the adult driver BAC limit (0.08 percent).This has been compromising overall traffic safety in New Zealand. [5]

In addition, New South Wales introduced extended zero tolerance laws to their new drivers.[6] Young drivers are inexperienced not only in driving but also in drinking and when they combine the two activities, this could be deadly. It is not surprising given that the risks of a fatal crash while driving at the current Canadian legal BAC limit (and in many other motorized countries including New Zealand, UK, USA) are alarmingly high even for experienced mature drivers.

The legal BAC limit in most motorized countries is too high; people often mistakenly believe that they may drive up to a BAC of 0.08 percent, overlooking the fact that driving is still impaired at lower concentrations. To set a blood alcohol limit so high that a 72 kg man can drink four bottles of beer and still be under the legal limit has consequences for drunk drivers, passengers of the vehicle and all other innocent vulnerable road users. Importantly, it may adversely influence a person’s estimates of their relative risk of injury or death while driving. Therefore drinking and driving legislation policies and decisions about enforcement need to be hinged on the scientific evidence. [2]

As Rivara and colleagues (2001) [7] pointed out, one of the most effective enabling factors for injury control is legislation. Importantly, laws must be capable of communicating the public health and traffic safety hazards of drinking and driving. As veteran pediatric injury prevention experts like Simons-Morton and Winston (2006) [8] have shown, laws have the power to change drivers’ perceptions of the value of safety practices.

However, we must understand that the youth traffic crash problem is interwoven with several other factors. [9] Therefore, it is important to have an overall approach that is built on the foundation of GDL laws to prevent young driver crashes. We should push for more comprehensive legislation and better enforcement based on the currently available best evidence.

To achieve the Canadian national road safety strategy target 2010, various road safety strategies for new drivers must receive priority. [10] It is evident that our current GDL laws have been steering young drivers in the right direction. However, in addition to extended zero BAC limits, until a complete cell phone ban, maximum speed limits, and compulsory seat belt laws are incorporated into the graduated licensing system throughout Canada, it is unlikely that the national road safety strategy target for 2010 will be achieved. One of the major reasons teens are killed or seriously injured when involved in traffic crashes is lack of seat belt use. It is well known that when drivers drink and drive, they tend to lack seat belt use. If we have effective policies that separate drinking and driving we could reduce non-belted youth drivers in our communities. In addition, GDL laws that explicitly include requirements for seat belt use in all phases, and sanctions that prohibit “graduation” to the next licensing phase if there is a seat belt citation, could increase teen seat belt use substantially. A zero-tolerance program for nonuse of safety belts, use of cell phone while driving and exceeding safe posted speed limits could be implemented, with immediate loss of license or other administrative penalties resulting for non-compliance. One major step we need to take irrespective of anything urgently is that we need to promote responsible driving among our children.

No policy can be effective unless it is adequately implemented and enforced, and there is awareness of both the policy and the enforcement efforts on the part of the intended targets. As with all above zero-tolerance programs, enforcement and strategic media campaigns to increase youth’s awareness of the law and of its enforcement efforts could significantly increase the effectiveness of these laws.

References

[1]. Chamberlain E, Solomon R. Zero blood alcohol concentration limits for drivers under 21: lessons from Canada. Inj Prev. 2008; 14(2):123-128.

[2].Desapriya EB. Alcohol limit for drink driving should be much lower. BMJ 2004; 328(7444):855-6.

[3]. Beirness, D.J., Simpson, H.M., Mayhew, D.R., et al; The Road Safety Monitor 2005- Drinking and Driving- Traffic Injury Research Foundation 2005

[4]. Senserrick TM. Graduation from a zero to .05 BAC restriction in an Australian graduated licensing system: a difficult transition for young drivers? Annu Proc Assoc Adv Automot Med. 2003;47:215-31.

[5]. Senserrick, T., Harworth, N., Review of literature regarding National International young driver training licensing and regulatory systems. Melbourne: Monash University Accident Research Centre, 2005.

[6]. Keall M.D., Frith W.J., Patterson T.L. The influence of alcohol, age and number of passengers on the night-time risk of driver fatal injury in New Zealand. Accid Anal Prev. 2004;36(1):49-61.

[7]. Rivara FP, Bennett E, Crispin B, Kruger K, Ebel E, Sarewitz A. Booster seats for child passengers: lessons learned for increasing their use. Inj Prev. 2001;7(3):210-3.

[8]. Simons-Morton BG, Winston FK. Translational research in child and adolescent transportation safety. Eval Health Prof. 2006;29:33 DOI: 10.1177/0163278705284442.

[9].Desapriya E, Joshi P, Pike I. Effects of graduated driver licensing on fatalities in 16-year-olds. Pediatrics 2006;118(5):2252-3.

[10]. Canada’s Road Safety Targets to 2010 http://www.tc.gc.ca/roadsafety/tp/tp13736/pdf/CRS_Target.pdf (accessed 15th March 2008)

Dear Editor,

We are grateful for the letter of Desapriya et al. regarding our recent Special Feature, and agree with the sentiments it expresses. We, too, view our proposal to extend BAC limits to the age of 21 as part of a much larger initiative to reduce traffic crashes among youth. Indeed, we outlined a more comprehensive approach to the issue in our 2006 report, Youth and Impaired Driving in Canada: Opportunities for Progress, which was published by MADD Canada and Allstate Insurance. The report includes proposals not only for graduated licensing and extended BAC restrictions, but also for more effective police enforcement powers and the regulation of alcohol sales. The full report can be downloaded at www.madd.ca.

Regards,

Erika Chamberlain and Robert Solomon

Dear Editor

First I want to thank the authors for an excellent study. As the authors discuss, the benefit of using photoelectric technology to reduce smoke alarm disablement and thereby fire deaths has never been adequately communicated to the public. (Note: This author has made repeated requests, accompanied by extensive research, to the National Fire Protection Association (NFPA), Underwriters Laboratories (UL) and the Consumer Product Safety Commission (CPSC) to inform the public but no action has been taken.)

The authors may be interested to know that the following language has been contained in the Massachusetts State Building Code since 1998. "Section 919.3 - Where required: single and multiple station smoke detectors or household fire warning systems shall be installed and maintained in full operating condition in the locations described in 780 CMR 919.3.1 through 919.3.3. Any smoke detector located within 20 feet of a kitchen or within 20 feet of a bathroom containing a tub or shower shall be a photoelectric type smoke detector."

This decision was based on research submitted by this author that is not available in the public health literature.

From a study published in NFPA’s Fire Journal, “ ... We favor photoelectric detectors to reduce rates of nuisance alarms from cooking and to provide optimal protection from cigarette related fires. Electrical detectors with battery back-up are the detectors of choice, except in communities such as remote villages in Alaska, where alternating current is non-existent or unreliable. If ionization detectors are installed, they should be located at least 20 feet, and preferably 25 feet, from stoves and at least 10 feet from bathroom doors if possible." (KuKlinski, Diana, Berger, Lawrence, and Weaver, John, "Smoke Detector Nuisance Alarms - A Field Study in a Native American Community", Fire Journal (Sept/Oct 1996) pp. 65-72.)

In a study in Woodlands, Texas, 90% (115/126) of the total number of false alarms were recorded by the ionization detectors, 86% (83/95) of the non-malfunction alarms were caused by cooking. This study clearly shows that the most common source of false alarms in a residential setting is cooking and that ionization detectors are clearly more susceptible to these types of false alarms. (Moore, D.A., "Remote Detection and Alarm for Residences: The Woodlands System", U.S. Fire Administration, Emmitsburg, Md., May 1980.)

In a manual published by one manufacturer titled "A Method For Improving Smoke Detector Codes In The United States, the manufacturer recommends using photoelectric detectors in, "Existing small apartments where kitchens or open flame heaters are adjacent to sleeping area." This manufacturer also recommends using a photoelectric detector if you have to place a detector within 20 ft of a furnace or heater. ("A Method for Improving Smoke Detector Codes In the United States", prepared by BRK Electronics, (1987).)

The authors have contributed valuable research to the previous work. I have already forwarded it to the various boards and committees that I am dealing with which type of smoke alarm to require.

If I may add some additional comments on the following topics discussed by the authors.

1. The authors cite 2 studies showing that, "smoke alarms are effective interventions for injuries from residential fires."

Comment

I agree that smoke alarms are better than nothing but this information has to be put in context. The authors conclude that having a smoke alarm is better than not having a smoke alarm. This is almost self evident. But neither investigated whether or not one type was better than another or if there were scenarios were it should have made a difference but didn’t. Neither author was probably aware, because it was never publicized, that in the late 80's UL and the smoke alarm manufacturers decided to de-sensitize smoke alarms, i.e. ionization smoke alarms, to help reduce the nuisance alarm problem. Since this de-sensitization has taken place the % of fire fatalities with working alarms has doubled. (US Fire Administration Data)

A 2004 Report issued by the NFPA, after this de-sensitization, found that smoke alarms reduced the risk of dying in a fire by only 7% for apartment dwellers. This report also estimated that the reduction risk for all residential occupancies was only 21%. (Ahrens, M., “U. S. Experience with Smoke Alarms and Other Fire Detection Alarm Equipment,” National Fire Protection Association, Quincy, MA November 2004.) In any case, researchers cannot identify how much of the reduction in risk is not due to the effectiveness of the detector but rather due to occupant characteristic that go along with owning a smoke detector: higher income, newer construction, better evacuation plans etc? All of these factors would contribute to a reduction in fire risk. As a consequence, the actual reduction in risk due to the effectiveness of the detector is probably much less that 21%. Smoke alarms do reduce risk but do they reduce it as much as we think they do?

Since, according to the CPSC, approximately 90% of the smoke alarms in use are ionization, then these statistics refer to the effectiveness of ionization alarms. This relatively small amount of risk reduction is due to the failure of the ionization alarm to adequately sense smoke created by smoldering fires, such as those started by smoking. "NIST officials told the Boston City Council's Public Safety Committee in August that "ionization alarms may not always alarm, even when a room is filled with smoke from a smoldering fire." Scientists have reached similar conclusions in Norway, Australia and England." ("Smoke Detector Alarming Limits," Roylance, F., Baltimore Sun, March 4, 2008.)

2. The authors discuss the cost of different technologies and list the following prices: ionization - $10, photoelectric $15, and dual sensors for $25.

Comment

The Boston Fire Department has provided photoelectric smoke alarms in our free give-away program for approximately ten years. The last pricing information that we received was $5.75 for ionization and $8.00 for photoelectric. It appears that much of the price differential at the retail level is due to marketing as opposed to manufacturing cost differences. It is not unreasonable to assume that if photoelectric alarms captured 90% of the market that benefits of "economies of scale" would change the price differential. In addition, in new construction the total price included wiring and labor so the total price differential is negligible.

Let me conclude with the following information:

Due to this author’s research the Massachusetts Board of Fire Prevention Regulations approved the following language. (It will become effective on October 1, 2008.)

32.02: Definitions

Approved monitored battery power smoke detector, shall consist of a working device, as defined in M.GL.c148, s. 26D. However such device shall not include a device that employs ionization technology as the sole means of smoke detection.

Approved primary power smoke detector, shall consist of a working device, as defined in M.GL.c148, s. 26D. However such device shall not include a device that employs ionization technology as the sole means of smoke detection.

32.03 Installation of Smoke Detectors In the Vicinity of a Kitchen, Bathroom or Other Areas.

No smoke detector employing ionization technology shall be installed within 20 feet of an entryway to a kitchen, or bathroom containing a bathtub or shower or any other room or area which features any item which produces steam or moisture. The 20 foot measurement shall include and extend into common areas of a multi-family dwelling, if applicable.

Due to this author’s research the Governor of Vermont will sign the following legislation. (It will become effective on January 1, 2009.)

(a) A person who constructs a single-family dwelling shall install one or more photoelectric-only-type smoke detectors in the vicinity of any bedrooms and on each level of the dwelling, …

(b) Any single-family dwelling when transferred by sale or exchange shall contain one or more photoelectric-only-type smoke detectors in the vicinity of any bedrooms and on each level of the dwelling installed in accordance with the manufacturer's instructions …

This research paper is additional evidence supporting my efforts to educate the public regarding the differences in smoke alarm technology and I appreciate the authors’ thoroughness. Hundreds of lives can be saved each year if the public is provided with this type of information.

Jay Fleming Deputy Chief Boston Fire Dept.