Document Sample
F1 Powered By Docstoc
					                                                 Am J Public Health 1999 89:1861-3

          Effectiveness of a comprehensive multisectoral campaign
           to increase seat belt use in Greater Athens area, Greece

Running Title: Increasing seat belt use: impact of a campaign

         Eleni Petridou,1,2 MD, MPH, Dimitrios Trichopoulos,1,2 MD, PhD,
         Matina Stappa,3 DDS, Yannis Tsoufis,4 MEC, Alkistis Skalkidou1,5
                  and the Hellenic Road Traffic Police Department

1. Center for Research and Prevention of Injuries, Dept of Hygiene and
   Epidemiology, Athens University Medical School, Athens, Greece
2. Dept of Epidemiology, Harvard School of Public Health, Boston, USA
3. Dept of Health Education, Ministry of Education, Greece
4. Ministry of Transportation, Greece
5. TROHOPEDIA-Greek Youth for Road Safety

Address for Correspondence:
Eleni Petridou, MD, MPH
Assoc. Professor of Epidemiology
Dept of Hygiene and Epidemiology
Athens University Medical School
75 M. Asias Str., T.K. 115 27
Athens, Greece
Tel. 301 7773 840
Fax. 301 7773 840 or 9324 300

Effectiveness of a comprehensive multisectoral campaign
to increase seat belt use in Greater Athens area, Greece


Objectives. To assess the effectiveness of a comprehensive campaign aiming to
increase seat belt use in Athens.
Methods. In 1996 a survey focusing on seat belt use was undertaken amongst
occupants of 1 400 passenger cars. From October 1997 to June 1998 the campaign
was implemented, while seat belt law enforcement was not intensified. In 1998
another inspection survey of 2 250 cars was undertaken.
Results. The program brought only a 6% increase in compliance, but there was an
estimated gain of about 50 averted deaths and 1 500 averted injuries.
Conclusions. An intensive campaign to increase seat belt use, not complemented with
law enforcement, resulted in moderate gains.

Key words: effectiveness; education information campaign; seat belt; safety coalition;
             inspection survey


        Among the European Union member states, Greece has one of the highest
death rates from motor-vehicle crashes.1,2 Several reasons may contribute to this
discrepancy, and variability in seat belt use is likely to be an important one.3-6 Indeed,
the prevalence of seat belt use among car occupants in Greece is reported as being one
of the lowest in European Union.6,7 The rising mortality from motor-vehicle
accidents in Greece and a series of sensational fatal car crashes, have created a fertile
soil among opinion leaders for the launching of a comprehensive program-project to
increase seat belt use in Greece.8
        Most efforts to increase seat belt use are based on awareness raising
complemented with law enforcement.9-12 This program has eventually focused on an
education and information campaign, because the Road Traffic Police Department did
not intensify enforcement of existing legislation for mandatory seat belt use. Indeed,
the number of citations for seat belt law violations slightly declined in the Greater
Athens area during the nine-month implementation period, in comparison to the
respective period of the previous year. Therefore, the only component of the program
that could have had an impact on changes in seat belt use in mid 1998, was the
education and information campaign. After a long preparatory period, the campaign
was launched in October 1997 and lasted until June 1998, with residual activities

Materials and Methods

        The program was coordinated by the Center for Research and Prevention of
Injuries, University of Athens Medical School and was contributed to by six
Ministries and over 50 other governmental and non governmental organizations.
Each institutional member of this coalition had each own independent budget and the
total cost of the campaign amounted to about three million US dollars.
        The activities of this campaign were heavily concentrated, although not
limited, to the Greater Athens area, where 40% of the 10 million Greek population
reside. Specific actions comprised production and distribution of televised and radio
messages, internet posting, a multipurpose video and a variety of eye-catching
advertising materials. Miscellaneous events, such as painting and poster competitions

among school students and training sessions for traffic policemen, schoolteachers, day
care workers and journalists were also organized.
       A pre-intervention survey in summer 1996, followed by a post-intervention
one in summer 1998 were provisioned. During the pre-intervention survey some
1 400 passenger cars and during the post-intervention survey some 2 250 passenger
cars were randomly stopped by road traffic policemen in 10 sites of the same
randomly selected secondary roads, five systematically selected main road arteries,
and five sites in the principal highways linking the capital of Athens with the rest of
the country. Thereafter, a trained interviewer inspected the car restraint availability
and use and recorded demographic variables of the occupants, basic car
characteristics, date and time of the interview. No refusals were noted.
       As a criterion, to assess the effectiveness of the campaign, the increasing
frequency in seat belt use between the two inspection surveys was used. The data
were modeled through multiple logistic regression.13,14 Because many cars in Greece
are not provided with rear seat belts and, even if available, rear seat passengers rarely
use them, modeling was restricted to front seat occupants older than 11 years old, for
whom a seat belt was available.

       Table 1 shows seat belt availability in occupied car seats and seat belt use,
when available, during the two inspection surveys by a series of potential predictors.
It appears that the campaign has been moderately effective. More specifically, seat
belt use increased from slightly less than 20% to slightly more than 25%. The
increase was somewhat larger among men than among women and more evident
among persons 25-64 years old, concerned virtually front seat occupants, and was
mostly found in those travelling highway journeys.
       Table 2 shows multiple logistic regression-derived, mutually adjusted odds
ratio for seat belt use by specified categories of a series of variables. The odds for
seat belt use were substantially higher in 1998 than in 1996 (odds ratio for seat belt
use in 1998 versus 1996, 1.8). Female and older (65+ years) car occupants were also
substantially and significantly more likely to wear a seat belt. Drivers were more
likely than front seat passengers to be seat belted and seat belt use was particularly
high in highways and particularly low in low speed suburb roads. Frequency of seat
belt use was lower during the weekends and during night hours. Interaction of year of

inspection with the other co-variates was not striking with one exception: the odds
ratio for seat belt use in highways was 3.2 in 1998 but only 2.0 in 1996.

       The objective of this study was to evaluate the effectiveness of a nine-month
campaign aiming to increase seat belt use. After adjustment for factors that affect seat
belt use, the increase in the prevalence of seat belt use could be confidently attributed
to the effect of the campaign (last 4 columns in Table 1 and first entry in Table 2).
Although the paired surveys approach does not have the advantages of randomization,
likely confounders were accurately measured and properly adjusted for.15
       The results indicate that the campaign was effective, although not strikingly
so. Moreover, the increased prevalence in seat belt use was apparently accomplished
through persuasion alone, because enforcement of legislation was not intensified
during the period of the campaign. Nevertheless, the documented increase of seat belt
use is expected to have reduced car passenger fatalities in Greater Athens by about
8%, assuming that non use of seat belt increases fatality by at least 50%.8 The meager
figures, however, can be translated into about 50 human lives that were saved and
1 500 road traffic injuries that were averted, simply due to increase in seat belt use.
The increase in seat belt use was particularly evident during highway journeys and
was also somewhat more pronounced among drivers in the 25-64 age bracket. This
appears to reflect the widely held misconception that seat belt use should be reserved
for highway driving. Although improvement is needed across the board, particular
emphasis should be given to the need of seat belt use in low speed journeys and
among rear seat passengers. There is also a need to concentrate future efforts on
persons less than 25 years old as well as on night and weekend car users, who appear
to represent population subgroups with higher risk taking attitudes.16,17
       In conclusion, a massive educational campaign, that was launched by a
coalition as broad as any in the Greek public health history but was not practically
supported by seat belt law enforcement, turned out to be only moderately successful.
In retrospect, this should not be surprising. Other populations, with stronger and
longer-lasting traditions in public health activism, have faced similar difficulties in
their efforts to increase seat belt use and it took them considerable time to reach the
currently enviable levels of car restraint use. After completion of the program,
however, new regulations regarding the standards of child restraints have been passed,

whereas legislation on mandatory seat belt use of rear seat car occupants is pending in
the Greek parliament. Provision of supportive services has also currently been
developed and mainly focuses on loan schemes for child car restraints, quality checks
for existing seat belts and consultative services for installment of missing seat belt in
old cars.


This study has been supported in part by a grant from Directorate General VII of the
European Union. We would like to acknowledge the valuable contribution at
different stages of the planning, organization and implementation of the campaign of:
   J. Breen (European Transport Safety Council)
   B. Utterstrom (Swedish National Society for Road Safety)
   M. Safwenberg (Swedish National Society for Road Safety)
   TROHOPEDIA (Greek Youth for Road Safety – a volunteers initiative)
   S. Kiose (Campaign Coordinator, CEREPRI)
   N. Dessypris (Statistician, CEREPRI)
   All other contributors to the campaign


1. World Health Organization. World Health Statistics Annual (1994). Geneva:
   WHO, 1995.

2. European Transport Safety Council. A strategic road safety plan for the
   European Union. Brussels: European Transport Safety Council, 1997.

3. Norin H. The seat belt wearing law in Sweden and its effect on occupant injuries
   in Volvo cars. Goteborg: Volvo Car Corporation, 1977.

4. Christian MS, and Bullimore DW. Reduction in accident injury severity in rear
   seat passengers using restraints. Injury. 1989; 20:262-264.

5. National Highway Traffic Safety Administration. Second report to Congress-
   Effectiveness of occupant protection systems and their use. Washington, DC:
   NHTSA, 1996.

6. European Transport Safety Council. Seat belts and child restraints: increasing
   use and optimizing performance. Brussels: European Transport Safety Council,

7. Petridou E, Hellenic Road Traffic Police Department. Person, time and place
   predictors of seat belt use in Athens, Greece. J Epidemiol Community Health.
   1998; 52:534-535.

8. Petridou E, Skalkidou A, Ioannou N, Trichopoulos D, and the Hellenic Road
   Traffic Police. Fatalities from non-use of seat belts and helmets in Greece: A
   nationwide appraisal. Accid Anal Prev 1998; 30:87-91.

9. Makinen T, Wittink RD, Hagenzieker MP. The use of seat belts and contributing
   factors-an international comparison. Report R-91-30. Leidschendam: SWOV;

10. Dussault C. Effectiveness of a selective traffic enforcement program combined
   with incentives for seat belt use in Quebec. Health Education Research, Theory
   Practice. 1990; 5:217-223.

11. Johnston JJ, Hendricks SA, and Fike JM. Effectiveness of behavioral safety belt
   interventions. Accid Anal Prev. 1994; 26:315-323.

12. Koch D, Medgyesi M, and Landry P. Saskatchewan’s occupant restraint program
   (1988-94):    performance   to   date.   Regina,   Saskatchewan:   Saskatchewan
   Government Insurance (SGI), 1995.

13. Prentice RL, and Breslow NE. Retrospective studies and failure time models.
   Biometrika. 1978; 65: 153-158.

14. Breslow NE, and Day NE. Statistical methods in cancer research. Vol I. The
   analysis of case-control studies.   IARC Scientific Publications No 32. Lyon:
   IARC, 1980.

15. MacMahon B, & Trichopoulos D. Epidemiology: Principles and Methods. 2nd
   Edition. Little Brown and Co. Boston, 1996.

16. Assum T. Attitudes on road accident risk. Accid Anal Prev. 1997; 29:153-9.

17. Petridou E, Zavitsanos X, Dessypris N, Frangakis C, Mandyla M, Doxiadis S,
   Trichopoulos D. Adolescents in high-risk trajectory: clustering of risky behavior
   and the origins of socioeconomic health differentials. Prev Med. 1997; 26:215-

Table 1:      Seat belt availability in occupied car seats and seat belt use, when available, by time, place, and personal characteristics.
             Greater Athens area, 1996aand 1998: individuals 12 years or more.
Characteristic                   Seat belt used          Seat belt         Seat belt not    Percent used when        Percent of seats
                                       (a)              available,         available (c)     Seat belt available      with seat belt
                                                                                                     a                        a
                                                       not used (b)                                a 100               (a + b) 100
                                                                                                       a +b                    a + b+ c
                               1996       1998       1996        1998      1996      1998       1996          1998      1996          1998
 Male                           165        359         771       1031       172       137        18            26         84              91
Female                          149        298         602        877       252       227        20            25         75              84

Age (years)
 12-24                           52        117         285        540       170       168        15            18         67              80
 25-34                           79        170         367        502        87        62        18            25         84              92
 35-64                          154        335         648        786       140       109        19            30         85              91
 65+                             29         35          73         80        27        25        28            30         79              82

Occupied seat
 front, driver                  179        338         556        685         6         2        24            33         99          100
 front, non driver              116        287         368        574         7         3        24            33         99          100
 Rear                            19         32         449        649       411       359         4             5         53           66

Time of inspection
 day hours                      238        471         883       1211       317       237        21            28         78              88
 night hours                     76        186         490        697       107       127        13            21         84              87

Place of inspection
 highways (entries –exits)      109        224         352        332        97        66        24            40         83              89
 main roads                      84        210         327        698       108       137        20            23         79              87
 suburb roads                   121        223         694        878       219       161        15            20         79              87

Day of inspection
 Weekdays                       282        534       1178        1423       386       269        19            27         79              88
 Weekend                         32        123        195         485        38        95        14            20         86              87
Data from: Petridou et al.: Person, time and place predictors of seat belt use in Athens. J Epidemiol Commun Health 1998;52:534-535

Table 2: Multiple logistic regression derived, mutually adjusted odds
         ratios (OR) for seat belt use and associated 95% confidence
         intervals (CI) by specified categories of a series of variables.
        Analysis restricted to occupants of front seats with belt
        availability, from both surveys (n=3103)

Characteristic                      OR                 (95% CI)

Year of inspection
   1996                            reference
   1998                            1.8                 (1.5, 2.1)

      Male                         reference
      Female                       1.8                 (1.5, 2.1)

Age (years)
   12-24                           1.0                 (0.8, 1.2)
   25-34                           1.0                 (0.8, 1.2)
   35-64                           reference
   65+                             1.9                 (1.3, 2.7)

Occupied seat
   front, driver                   1.3                 (1.1, 1.7)
   front, non driver               reference

Time of inspection
   day hours                       reference
   night hours                     0.7                 (0.6, 0.8)

Place of inspection
   highways (entries-exits)        2.6                 (2.2, 3.2)
   main roads                      1.3                 (1.1, 1.6)
   suburb roads                    reference

Day of inspection
   weekdays                        reference
   weekend                         0.8                 (0.6, 0.9)


Shared By: