238 by xiangpeng


									Predictors of Injury Among Adult
Recreational In-Line Skaters:
A Multicity Study
Richard M. Seldes, MD, Jeane A. Grisso, MD, MSc, Jeff R. Pavell, DO,
Jesse A. Berlin, ScD, Virak Tan, MD, Brock Bowman, MD, Judith L. Kinman, MA,
and Robert H. Fitzgerald, Jr, MD

      In-line skating is currently one of the     Each data collection site was an area where a
fastest growing recreational sports in the        large number of skaters were present. Per-
United States and is gaining much popularity      sons wearing in-line skates were potentially
worldwide. Since its introduction to the gen-     eligible, regardless of whether they were
eral United States population in the 1980s,       actively skating. The interviewers were
participation had grown to an estimated 31        instructed to focus on one area ofthe site and
million people nationally in 1996, most of the    choose a potential skater. The interviewer
growth occurring within the previous 5 to 7       bypassed that particular skater and, on the
years.' dramatic increase in injuries due to      basis of a predetermined list of randomly
in-line skating has resulted from this recent     generated numbers, counted the nth skater in
rise in popularity. According to the Consumer     the immediate area, approached that skater,
Product Safety Commission, an estimated           completed the interview, and used the next
 105 000 in-line skating injuries occurred in     random number on the list and the same for-
 1996, representing a 191% increase from          mat to determine the next subject. As a
 1993.2 This has led to an increase in large-     means of preventing repeated interviews,
scale safety awareness programs:-7                potential subjects were asked whether they
      Several studies have described fre-         had participated in the interview at a prior
quency rates and types of injuries from in-line   time.
skating using injury databases, emergency               An adult recreational in-line skater was
departnent case series, or small surveys. 814     defined as a person 18 years of age or older
However, to date there have been no commu-        who participated in in-line skating for at least
nity-based studies describing risk factors for    1 hour a week for at least 1 month out of the
in-line skating injuries based on skating prac-   year. The main outcome variable was a self-
tices or exposure. In addition, most studies      reported injury within the past year. An
have failed to differentiate among the differ-    injury was defined as an impairment caused
ent types of in-line skaters. The styles of in-   by in-line skating that required the skater to
line skating can be classified as recreational,   change his or her usual skating patter, take
aggressive, racing, and roller hockey.6           medication, or seek medical attention. When
      We conducted a cross-sectional survey       a skater reported an injury, detailed questions
of adult recreational skaters in 6 major cities   were asked about the body part(s) injured,
in the United States to identify risk factors     safety gear use, and skating experience at the
for injury from in-line skating, estimate the     time of injury. The events that caused the
prevalence of such injuries, and describe the
use of safety gear and the skating habits of
this population.                                  Richard M. Seldes, Virak Tan, and Robert H.
                                                  Fitzgerald, Jr, are with the Department of Ortho-
                                                  paedic Surgery, Hospital of the University of Penn-
Materials and Methods                             sylvania, Philadelphia. Jeane A. Grisso, Jesse A.
                                                  Berlin, and Judith L. Kinman are with the Center
Study Design and Data Acquisition                 for Clinical Epidemiology and Biostatistics,
                                                  University of Pennsylvania School of Medicine.
                                                  Jeff R. Pavell is with the Department of Physical
      Subjects were recruited from 6 US           Medicine and Rehabilitation, New York University
cities (New York, Philadelphia, Houston,          Medical Center, New York City. Brock Bowman is
Chicago, Los Angeles, San Francisco) where        with the Department of Physical Medicine and
recreational in-line skating is popular.' The     Rehabilitation, Baylor University Medical Center,
study period was August 1996 to February          Houston, Tex.
1997.                                                   Requests for reprints should be sent to
      We developed a detailed interview-style     Richard M. Seldes, MD, Hospital of the University
                                                  of Pennsylvania, Department of Orthopaedic
questionnaire based on pilot interviews with      Surgery, 2nd Floor Silverstein Pavilion, 34th and
in-line skaters of different ages and skating     Spruce Sts, Philadelphia, PA 19104 (e-mail:
abilities. At each of the chosen data collec-     rseldes@mail.med.upenn.edu).
tion sites, subjects were randomly selected.            This paper was accepted June 18, 1998.

                                                                    February 1999, Vol. 89, No. 2
                                                                                                                            Public Health Briefs

injury and the medical treatments received        skaters, respectively. Only 6% of skaters
were recorded.                                    consistently wore all 4 types of safety gear.        TABLE 1-Selected Characteristics
                                                                                                                of Adult Recreational In-
      Several variables were selected as pos-     Agreement rates comparing observed safety                     Line Skaters in 6 Major
sible risk factors. Skating exposure variables    gear use and self-reported use ranged from                    US Cities, 1996/97
included self-reported total months of skat-      90% to 95%, depending on the type of gear.                    (n = 938)
ing, hours of skating per week, and miles               Ninety-nine skaters (11%) reported 118                                      Sample
                                                  injuries. The most common body part                  Characteristic
skated per week. Behavioral variables
included a history of perfonning tricks, self-    injured was the wrist (15% of all reported           Geographic location, %
reported skating skill, self-reported safety      injuries), followed by the knee (12%), and             New York                     48
gear use, and a history of receiving skating      the most common injury types were frac-                 Houston                     25
instruction. Observed skating gear use was        tures (19% of all reported injuries) and con-           San Francisco                9
                                                                                                          Philadelphia                 7
recorded for 200 skaters to validate self-        tusions (19%). Sixty-five percent of injuries           Chicago                      7
reports. Environmental variables included         required medical attention. The most com-              Los Angeles                   4
skating location most often used and history      mon circumstances that caused an injury              Sex, %
of collisions. Demographic variables              were collisions or trying to avoid collisions          Male                          61
included age, sex, education level, marital       (26%), followed by falls while doing tricks            Female                        39
status, and body mass index (weight in            (21%). Only 7% of injured skaters were               Marital status, %
kilograms divided by height in meters             wearing all 4 types of safety gear at the time         Single                        71
squared).'5                                       of injury. Sixty-two percent of skaters had            Married                       23
                                                                                                         Divorced                        6
                                                  more than 1 year of skating experience
Statistical Analysis                              before their injury.                                 Education level, %
                                                        Risk factors for injury, as determined           Graduate degree               21
                                                                                                         College degree                54
      Continuous variables were compared          by the bivariate analysis, included skating in         Less than college degree      25
via independent-sample t tests or Wilcoxon        a street or area with railings and ledges            Reason for participation, %
tests, depending on the distribution of the       (P = .001), a history of performing tricks or           Exercise                      33
variables. Categorical variable distributions     stunts (P = .001), greater number of hours of          Fun                            57
were compared by means of X2 statistics. For      skating per week (P = .001), greater number            Transportation                  6
ordinal variables, we also computed the X2        of total months skating (P= .001), and a               Other                           4
test for trend. Unadjusted odds ratios (ORs)      self-reported expert skill level (P=.001)            Confidence level, %
and 95% confidence intervals (CIs) were cal-      (Table 2).                                              Beginner                      36
                                                                                                          Intermediate                  48
culated for all variables of interest. Logistic         Further analysis of those who reported            Expert                        16
regression models included all variables that     an expert skill level revealed that this group       Instruction, %
were statistically significant in unadjusted      included a significantly higher proportion of           Lessons                       12
analyses, that were considered clinically         men (P = .001), skated more hours per week             Videos/books                    4
important based on a priori hypotheses, or        (P = .0001), skated more miles per week                 None                          84
whose addition to the models substantially        (P = .0001), and had more months of skating          Usual skating location, %
changed the estimates of the effect of other      experience (P=.0001) than the rest of the               Street                        32
factors. SAS for Windows, version 6.11            sample. Also, this group was more likely to             Park path                     61
                                                                                                          Rink                           5
(SAS Institute Inc, Cary, NC), and EGRET,         skate in areas with railings, ramps, and                Ramps, railings, or ledges     2
version 0.25.1 (Statistics and Epidemiology       ledges (P=.001); perform tricks or stunts            Perform tricks or stunts, %
Research Corp, Seattle, Wash), were used to       (P=.001); wear less safety gear (P=.001);              Yes                            25
perform the analysis.                             and sustain an injury (P=.001) than begin-              No                            75
                                                  ners and those at intermediate skill levels.         Collisions, %
                                                        Independent risk factors for injury, as           Yes                           36
Results                                           determined from the logistic regression                 No                            64
                                                  models (Table 3), included a history of per-         Acute injury, %
     We interviewed 964 of 1014 skaters           forming tricks or stunts (P= .03), number of            Yes                           11
randomly selected for participation (response     hours skating per week (P = .006), and skat-            No                            89
rate: 95%). Of the 964 subjects, 26 were          ing location most frequently used (P < .001).        Age, y, mean ± SD              30 ± 8
excluded because they were younger than 18        The skating location that was most associ-           Total months
years of age, leaving 938 adult recreational      ated with an increased prevalence of injury             skating, mean ± SD         25 ± 21
skaters for the analysis. The typical skater      was an area with railings, ramps, and ledges         Hours skating per week,
was a well-educated, single, healthy adult        (vs skating on streets) (adjusted OR = 8.5,             mean ± SD                    6±6
who skated primarily for exercise or fun, did     95% CI = 2.5, 29.1). The hours of skating
                                                  per week variable was associated with the             Note. The survey cities were New York,
not take lessons, refrained from doing tricks                                                             Philadelphia, Houston, Chicago, Los
or stunts, had been in-line skating for           prevalence of injury in a linear dose-                  Angeles, and San Francisco.
approximately 2 years prior to the interview,     response relation, with skating more than 10
and skated an average of 6 hours per week         hours per week having the highest risk
(Table 1).                                        (adjusted OR = 4.0, 95% CI = 1.8, 8.9). It is
      Safety gear use was generally low. Hel-      important to note that performing tricks or       Discussion
mets, elbow guards, knee guards, and wrist         stunts was a predictor of injury regardless of
guards were reportedly used sometimes or           the skater's self-reported skill level or level       To our knowledge, this is the first com-
never by 76%, 86%, 72%, and 41% of                 of skating experience.                            munity-based study of in-line skating assess-

February 1999, Vol. 89, No. 2                                                                              American Journal of Public Health 239
Public Health Briefs

                                                                                                    recreational skaters were most likely to per-
  TABLE 2-Selected Characteristics of Injured and Uninjured In-Line Skaters in                      form tricks while skating, wear less safety
           6 Major US Cities, 1996/97 (n = 938)
                                                                                                    gear, and sustain an injury. Another impor-
                                                Uninjured          Injured                          tant finding was the overall low rate of safety
              Characteristic                    (n = 839)         (n = 99)              P           gear use reported among all adult in-line
  Sex, %                                                                                                  Previous studies of in-line skating
     Male                                          59               71
     Female                                        41               29                              injuries have been based on injury databases,
  Total months skating, %                                                             .001 b        emergency department series, or small sur-
     1-12                                          29               15                              Veys.8,9,11-14We believe that the present study
    13-24                                          36               33                              design had several advantages over these
    >24                                            35               52                              types of studies. Through this survey, we
      Total months skating, mean ± SD           25 ± 21           32 ±25                            obtained data from a large sample of both
  Hours skating per week, %                                                           .001 b        uninjured skaters and injured skaters, includ-
    1-2                                            35               13                              ing those whose injuries did not necessarily
    3-5                                            31               29
    6-10                                           23               30                              result in emergency department care. In
    >10                                            11               28                              addition, through an in-depth standardized
      Hours skating per week, mean ± SD           6±6              9±7                              interview, we were able to obtain detailed
  Miles skated per week, %                                                            .001 b        information regarding skating exposure,
    <5                                             28               12                              skating habits, and specific events surround-
     5-9                                           24               17                              ing reported injuries. To enhance the general-
     10-19                                         27               40                              izability of our study, we used a large sample
     .20                                           21               31
      Miles skated per week, mean ±SD            12 ± 14          18 ±19                            of randomly selected skaters, achieved a
  Perform tricks or stunts, %                                                         .001a         high response rate, and recruited subjects
    Yes                                            22               45                              from 6 cities in various regions of the United
    No                                             78               55                              States.
  Confidence level, %                                                                  .001 a             However, our study design did involve
    Beginner                                       37                19                             some limitations. Information bias may have
    Intermediate                                   48                54                             occurred because data were based on self-
    Expert                                         15                27                             report and, with the exception of safety gear
  Note. The survey cities were New York, Philadelphia, Houston, Chicago, Los Angeles, and           use, could not be validated. Selection bias
    San Francisco.                                                                                  may have been present because skaters who
  aDerived from x2 test comparing the given attribute between the injured and uninjured             had sustained a severe injury previously may
    skaters.                                                                                        have been excluded from the current general
  bTest for trend.                                                                                  skating population. Thus, the prevalence of
                                                                                                    severe injuries may have been underesti-
                                                                                                    mated. The results and conclusions of the
  TABLE 3-Risk Factors for Injury From In-Line Skating                                              present study pertain only to adult recre-
                                                                                                    ational in-line skaters.
                                         Unadjusted                       Adjusted                        Most of the concern regarding in-line
                                     Odds Ratio (95% Cl)     Odds Ratio (95% Cl)         P          skating injuries has been directed toward
                                                                                                    beginners.37 Most studies to date have
   Hours skating per week                                                              .006         reported that beginners are at the greatest
     1-2                                 Reference                Reference                         risk for injury and that the events causing
     3-5                                2.5 (1.3, 5.0)           2.1 (1.1, 4.3)
     6-10                               3.5 (1.8, 6.8)           2.6 (1.2, 5.4)                     injury are most often due to a loss of balance
     >10                                6.3 (3.1, 12.7)          4.0 (1.8, 8.9)                     or a change in skating surface.'1'2,14'16 In the
   Usual skating location                                                              <.001        present study, the most commonly injured
     Street                              Reference                Reference                         skater was a self-reported expert with more
     Ramps, rails, ledges               9.5 (3.2, 28.0)          8.5 (2.5, 29.1)                    than 1 year of skating experience, and the
     Park path                          0.5 (0.3, 0.8)           0.7 (0.4,1.2)                      most common circumstances that caused an
     Rink                               0.6 (0.2,1.9)            0.5 (0.2,1.6)
                                                                                                    injury were collisions or performing tricks
   Perform tricks or stunts                                                             .03         while skating. Based on our findings, we
     No                                  Reference                Reference
     Yes                                2.8 (1.8, 4.3)           1.8 (1.1, 3.0)                     believe that more experienced adult recre-
                                                                                        .01         ational in-line skaters are at an increased risk
   Education level                                                                                   for injury. To date, this at-risk population has
     Less than college                   Reference                Reference
     College degree                     0.9 (0.6,1.4)            1.5 (0.9, 2.7)                     not been recognized.
     Graduate degree                    0.4 (0.2, 0.8)           0.6 (0.3,1.3)                             In conclusion, we recommend that safe
                                                                                                     skating education programs recognize this
   Note. Cl = confidence interval.                                                                   at-risk population and consider specifically
                                                                                                     targeting more advanced skaters in their
ing risk factors for injury. The results from       skating more hours per week are significant      campaign messages. Advanced skaters
this large survey indicate that skating loca-       predictors of injury from in-line skating. In    should be made aware that, even after they
tion, performing tricks while skating, and          this adult population, more experienced          have learned the basic skating techniques

240 American Journal of Public Health                                                                                 February 1999, Vol. 89, No. 2
                                                                                                                                                                    Public Health Briefs

and gained confidence in their skating abil-                        writing of the paper. All authors are guarantors for                  9. Orenstein J. Injuries and small-wheel skates.
ity, they still have a substantial risk for injury                  the integrity of the research.                                           Ann Emerg Med. 1996;27:204-209.
and should continue to adhere to fundamen-                                                                                               10. Schieber RA, Branche-Dorsey CM. In-line
                                                                                                                                             skating injuries: epidemiology and recommen-
tal safety principles. More important, the                          References                                                               dations for prevention. Sports Med. 1995;
performance of tricks while skating places a                          1. American Sports Analysis Summary Report.                            19:427-432.
skater at significantly higher risk for injury,                          Hartsdale, NY: American Sports Data Inc; 1995.                  11. Calle SC, Eaton RG. Wheels-in-line rollerskat-
regardless of the skater's experience level.                          2. National Electronic Injury Surveillance                             ing injuries. J Trauma. 1993;3 5:946-95 1.
                                                                         System. Washington, DC: US Consumer Prod-                       12. Banas M, Dalldorf P, Marquardt J. Skateboard
Targeting and educating more experienced                                 uct Safety Commission; 1993-1996.
skaters in safe skating campaign messages                                                                                                    and in-line skate fractures: a report of one sum-
                                                                      3. Advocacy and Communications. In-Line Skat-                          mer's experience. J Orthop Trauma. 1992;6:
may assist in lowering rates of injury due to                            ing Is on a Roll. Chicago, Ill: American Med-                       301-305.
in-line skating. D                                                       ical Association; 1996.
                                                                      4. Play-It-Safe Campaign. Chicago, Ill: American                   13. Schieber RA, Branche-Dorsey CM, Ryan GW.
                                                                         Academy of Orthopaedic Surgeons; 1996.                              Comparison of in-line skating injuries with
                                                                                                                                             rollerskating and skateboarding injuries.
Contributors                                                          5. Asphalt Bites Campaign. Minnetonka, Minn:
                                                                                                                                             JAMA. 1994;271:1856-1858.
                                                                         Rollerblade Inc; 1996.
Richard Seldes designed and planned the study, ana-                   6. Gear Up Campaign. Kensington, Md: Intema-                       14. Schieber RA, Branche-Dorsey CM, Ryan GW,
lyzed the data, oversaw and conducted the standard-                      tional In-Line Skating Association; 1996.                           Rutherford GW, Stevens JA, O'Neil J. Risk
ized interviews, and wrote the paper. Jeff Pavell,                    7. Consumer Product Safety Alert. Skate but                            factors for injuries from in-line skating and the
Virak Tan, and Brock Bowman contributed to the                           Skate Safely-Always Wear Safety Gear. Wash-                         effectiveness of safety gear. N Engl J Med.
design and planning of the study, analyzed the data,                     ington, DC: US Consumer Product Safety                              1996;335:1630-1635.
conducted the standardized interviews, and con-                          Commission; 1995.                                               15. Garrow J, Webster J. Quetelet's index (W/H2) as
tributed to the writing of the paper. Jeane Grisso,                   8. Adams S, Wyte C, Paradise M, Castillo J. A                          a measure of fatness. Int J Obes. 1985;9:
Jesse Berlin, and Judith Kinman designed and                             prospective study of in-line skating: observa-                      147-153.
planned the statistical analysis and contributed to the                  tional series and survey of active in-line                      16. Heller DR, Routley V, Chambers S. Rollerblad-
writing of the paper. Robert Fitzgerald contributed to                   skaters-injuries, protective equipment, and                         ing injuries in young people. J Paediatr Child
the planning of the study, analysis of the data, and                     training. Acad Emerg Med. 1996;3:304-311.                           Health. 1996;32:35-38.

     ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ .i,. i ,
                       t.~~~~~~~ f"
                                                                     Wife Abuse Among Women of
                                                                     Childbearing Age in Nicaragua
                                                                      Mary Carroll Ellsberg, MPH, Rodolfo Pefia, MD, MPH, Andre's Herrera, MD,
                                                                      Jerker Liljestrand, MD, PhD, and Anna Winkvist, PhD

  #S       Ri                                 .. .....i''
                                                                                            Wife abuse is increasingly recognized
                                                                                      as a global public health concern.1 Although
                                                                                      reliable prevalence data are scarce, it is esti-
                                                                                      mated that between 20% and 50% of
                                                                                      women in most countries have experienced
                                                                                      physical violence from an intimate part-
                                                                                                                                         population-based data, whether this reflects
                                                                                                                                         an actual increase or improved reporting.

                                                                                      ner.2-6 Wife abuse has been associated with              This article presents the results of the
    W iSiig...Z;~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.             a variety of adverse health outcomes for          first population-based survey carried out in
                                                                                      women and children, including trauma,7'8           Nicaragua on wife abuse. The study aimed to

                                                                                       low birthweight,9"0 gynecological dis-
                                                                                       orders,'1 depression, 13 suicide, 4 and sexu-     Mary Carroll Ellsberg is with the Department of
                                                                                       ally transmitted diseases.'5 Few studies have     Epidemiology and Public Health, Umea University,
                                                                                       found significant risk factors among women        Umea, Sweden, and the Department of Preventive
                                                                                       for wife abuse,'6"17 although some risk fac-      Medicine, Autonomous University of Nicaragua,
                                                                                       tors have been consistently associated with       Le6n. She is also with the Center for Health and
                                                                                       violent men, such as witnessing violence as       Gender Equity, Takoma Park, Md. Rodolfo Pefia and
                                                                                       a child, poverty, stress, alcohol use, and cul-   Anna Winkvist are with the Department of
                                                                                                                                         Epidemiology and Public Health, UmeA University.
                                                                                       tural norms that discriminate against             Rodolfo Penia is also with the Department of
                                                                                       women. 16"18-20 Awareness regarding wife          Preventive Medicine, Autonomous University of
       ,,, ide wi0 . W1 j;b* '
                         ik iu  30M                                                    abuse has increased greatly in Nicaragua, in      Nicaragua. Andres Herrera is with the Departnent of
                                                                                       part as a result of the growing number of         Preventive Medicine, Autonomous University of
                                                                                       nongovernmental organizations providing           Nicaragua. Jerker Liljestrand is with the Departnent
                                                                                       health, legal, and psychological services for     of Reproductive Health and Research, World Health
                                                                                       battered women, as well as advocacy to            Organization, Geneva, Switzerland.
                                                                                                                                               Requests for reprints should be sent to Mary
                                                                                       improve laws and public policy with regard        Carroll Ellsberg, MPH, Center for Health and Gender
       &   244)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Tmi
              I FZiXkX, ;fiut ;9 S,                                                    to domestic violence.2"22 Although the            Equity, 6930 Carroll Ave, Suite 910, Takoma Park,
                                                                                       reported incidence of wife abuse has              MD 20912 (e-mail: mellsbergggenderhealth.org).
                              V t , . .,~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . . .. increased,23 it is unknown, in the absence of
                                      .~ ~                                                                                                     This paper was accepted July 6, 1998.
                        ~ ~~T'
                             ~ ~
                10~~~~~~~~~~~~ ~
                                                                                                                                                 American Journal of Public Health 241

To top