Stephen W Marshall

University of North Carolina at Chapel Hill, North Carolina, United States

Are you Stephen W Marshall?

Claim your profile

Publications (264)1119.03 Total impact

  • Journal of athletic training 06/2015; DOI:10.4085/1062-6050-50.5.04 · 1.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A report by the Institute of Medicine called for comprehensive nationwide concussion incidence data across the spectrum of athletes aged 5 to 23 years. To describe the incidence of concussion in athletes participating in youth, high school, and collegiate American football. Data were collected by athletic trainers at youth, high school, and collegiate football practices and games to create multiple prospective observational cohorts during the 2012 and 2013 football seasons. Data were collected from July 1, 2012, through January 31, 2013, for the 2012 season and from July 1, 2013, through January 31, 2014, for the 2013 season. The Youth Football Surveillance System included 118 youth football teams, providing 4092 athlete-seasons. The National Athletic Treatment, Injury and Outcomes Network program included 96 secondary school football programs, providing 11 957 athlete-seasons. The National Collegiate Athletic Association Injury Surveillance Program included 24 member institutions, providing 4305 athlete-seasons. All injuries regardless of severity, including concussions, and athlete exposure information were documented by athletic trainers during practices and games. Injury rates, injury rate ratios, risks, risk ratios, and 95% CIs were calculated. Concussions comprised 9.6%, 4.0%, and 8.0% of all injuries reported in the Youth Football Surveillance System; National Athletic Treatment, Injury and Outcomes Network; and National Collegiate Athletic Association Injury Surveillance Program, respectively. The game concussion rate was higher than the practice concussion rate across all 3 competitive levels. The game concussion rate for college athletes (3.74 per 1000 athlete exposures) was higher than those for high school athletes (injury rate ratio, 1.86; 95% CI, 1.50-2.31) and youth athletes (injury rate ratio, 1.57; 95% CI, 1.17-2.10). The practice concussion rate in college (0.53 per 1000 athlete exposures) was lower than that in high school (injury rate ratio, 0.80; 95% CI, 0.67-0.96). Youth football had the lowest 1-season concussion risks in 2012 (3.53%) and 2013 (3.13%). The 1-season concussion risk was highest in high school (9.98%) and college (5.54%) in 2012. Football practices were a major source of concussion at all 3 levels of competition. Concussions during practice might be mitigated and should prompt an evaluation of technique and head impact exposure. Although it is more difficult to change the intensity or conditions of a game, many strategies can be used during practice to limit player-to-player contact and other potentially injurious behaviors.
    05/2015; DOI:10.1001/jamapediatrics.2015.0210
  • [Show abstract] [Hide abstract]
    ABSTRACT: To examine statewide emergency department (ED) visit data for motorcycle crash morbidity and healthcare utilization due to traumatic brain injuries (TBIs) and non-TBIs. North Carolina ED data (2010-2012) and hospital discharge data (2009-2011). Statewide ED visits and hospitalizations due to injuries from traffic-related motorcycle crashes stratified by TBI status. Descriptive study. Descriptive statistics include age, sex, mode of transport, disposition, expected source of payment, hospital length of stay, and hospital charges. Over the study period, there were 18 780 ED visits and 3737 hospitalizations due to motorcycle crashes. Twelve percent of ED visits for motorcycle crashes and 26% of hospitalizations for motorcycle crashes had a diagnosis of TBI. Motorcycle crash-related hospitalizations with a TBI diagnosis had median hospital charges that were nearly $9000 greater than hospitalizations without a TBI diagnosis. Emergency department visits and hospitalizations due to motorcycle crashes with a TBI diagnosis consumed more healthcare resources than motorcycle crash-related ED visits and hospitalizations without a TBI diagnosis. Increased awareness of motorcyclists by other road users and increased use of motorcycle helmets are 2 strategies to mitigate the incidence and severity of motorcycle crash injuries, including TBIs.
    The Journal of head trauma rehabilitation 05/2015; 30(3):175-84. DOI:10.1097/HTR.0000000000000096 · 3.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Overuse injuries result from microtrauma due to repetitive loading combined with insufficient tissue recovery time and can result in both immediate and long-term time loss from sports. Overuse injury rates and patterns differ across college and high school populations, sport, and sex. Descriptive epidemiology study. Surveillance data for 16 sports from the National Collegiate Athletic Association's Injury Surveillance System (NCAA ISS; 2004-2005 through 2008-2009) and 14 sports from High School Reporting Information Online (High School RIO; 2006-2007 through 2012-2013) were analyzed. All reported injuries had an injury mechanism of overuse/gradual onset (college) or overuse/chronic (high school). Overuse injury incidence rates were calculated, and rate ratios with 95% CIs were used to compare subgroups. The rate of overuse injury was 3.28 times higher in college than high school sports (95% CI, 3.12-3.44). The rate of overuse injury among sex-comparable sports was higher in female than male athletes in both populations (college rate ratio, 1.25; 95% CI, 1.16-1.35; high school rate ratio, 1.55; 95% CI, 1.43-1.68). The lower extremity was the most commonly injured body site (college, 69.4%; high school, 70.4%). A larger proportion of overuse injuries among college athletes resulted in time loss of more than 21 days (college, 20.4%; high school, 7.7%) and surgery (college, 5.2%, high school, 2.5%). Overuse injuries can impose a significant burden on college and high school athletes. Interventions addressing prevention of overuse injury are needed. © 2015 The Author(s).
    The American Journal of Sports Medicine 04/2015; DOI:10.1177/0363546515580790 · 4.70 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We evaluated how attention deficit-hyperactivity disorder (ADHD) and learning disability (LD) are associated with concussion history and performance on standard concussion assessment measures. Based on previous reports that developmental disorders are associated with increased injury proneness and poorer cognitive performance, we anticipated that ADHD and LD would be associated with increased history of concussion and poorer baseline performance on assessment measures. Cross-sectional study. Clinical research center. The study sample aggregated data from two separate projects: the National Collegiate Athletic Association Concussion Study and Project Sideline. We analyzed preseason baseline data from 8056 high school and collegiate athletes (predominantly male football players) enrolled in prior studies of sport-related concussion. Measures included demographic/health history, symptoms, and cognitive performance. Attention deficit-hyperactivity disorder and LD were associated with 2.93 and 2.08 times the prevalence, respectively, of 3+ historical concussions (for comorbid ADHD/LD the prevalence ratio was 3.38). In players without histories of concussion, individuals with ADHD reported more baseline symptoms, and ADHD and LD were associated with poorer performance on baseline cognitive tests. Interactive effects were present between ADHD/LD status and concussion history for self-reported symptoms. Neurodevelopmental disorders and concussion history should be jointly considered in evaluating concussed players. Clinical judgments of self-reported symptoms and cognitive performance should be adjusted based on athletes' individual preinjury baselines or comparison with appropriate normative samples.
    Clinical journal of sport medicine: official journal of the Canadian Academy of Sport Medicine 04/2015; DOI:10.1097/JSM.0000000000000207 · 2.01 Impact Factor
  • Osteoarthritis and Cartilage 04/2015; 23:A171-A172. DOI:10.1016/j.joca.2015.02.939 · 4.66 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND North Carolina requires motorcyclists of all ages to wear federally approved safety helmets. The purpose of this article is to estimate the impact of this state law in terms of hospital admissions for traumatic brain injury (TBI) and associated hospital charges. METHODS Hospital admissions of North Carolina motorcyclists with TBIs and associated hospital charges in 2011 were extracted from the North Carolina Hospital Discharge Data system. We estimated hospital admissions and charges for the same year under the counterfactual condition of North Carolina without a universal motorcycle helmet law by using various substitutes (Florida, Pennsylvania, and South Carolina residents treated in North Carolina). RESULTS North Carolina's universal helmet law prevented an estimated 190 to 226 hospital admissions of North Carolina motorcyclists with TBI in 2011. Averted hospital charges to taxpayer-funded sources (ie, government and public charges) were estimated to be between $9.5 million and $11.6 million for 2011, and total averted hospital charges for 2011 were estimated to be between $25.3 million and $31.0 million. LIMITATIONS Cost estimates are limited to inpatients during the initial period of hospital care. This study was unable to capture long-term health care costs and productivity losses incurred by North Carolina's TBI patients and their caregivers. CONCLUSIONS North Carolina's universal motorcycle helmet law generates health and economic benefits for the state and its taxpayers.
    North Carolina medical journal 03/2015; 76(2):70-5.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: The study described initial pharmacological treatment of children in the United Kingdom diagnosed as having ADHD and assessed predictors of medication persistence. Methods: U.K. children ages 3-16 diagnosed as having ADHD between 1994 and 2006 were identified from primary care practice data. Child characteristics, prescription patterns, and initial medication prescribed were described over the study period. The associations of child and clinical factors with medication persistence (defined as initial treatment length greater than six months) were estimated by using binomial regression. Results: Of 2,878 children with an ADHD diagnosis, 46% (N=1,314) received at least one prescription for ADHD medication within two years of diagnosis. The mean initial treatment length was 10.7±.5 months. Only 35% (N=464) of pharmacologically treated children had a treatment length greater than six months after initial medication prescription when the analysis used a 30-day grace period; 57% were persistent in treatment when a less stringent 60-day grace period was used. Children who were initially prescribed long-acting methylphenidate were more likely to persist in treatment than those prescribed standard methylphenidate (risk ratio=1.2, 95% confidence interval=1.1-1.4). Conclusions: A large proportion of children who received medication for ADHD in primary care did not continue in initial treatment for more than six months. Few child or clinical factors were associated with treatment persistence. Epidemiological research about the effects of long-term ADHD medication use should account for the observed limited persistence in medication treatment.
    Psychiatric services (Washington, D.C.) 03/2015; 66(6):appips201300148. DOI:10.1176/appi.ps.201300148 · 1.99 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Over the past decade, there has been significant debate regarding the effect of cumulative subconcussive head impacts on short and long-term neurological impairment. This debate remains unresolved because valid epidemiological estimates of athletes' total contact exposure are lacking. We present a measure to estimate the total hours of contact exposure in football over the majority of an athlete's lifespan. Through a structured oral interview, former football players provided information related to primary position played and participation in games and practice contacts during the pre-season, regular season, and post-season of each year of their high school, college, and professional football careers. Spring football for college was also included. We calculated contact exposure estimates for n=64 former football players (n=32 college football only, n=32 professional and college football). The Head Impact Exposure Estimate (HIEE) discriminated between individuals who stopped after college football, and individuals who played professional football (P<0.001). The HIEE measure was independent of concussion history (P=0.82). Estimating total hours of contact exposure may allow for the detection of differences between individuals with variation in subconcussive impacts, regardless of concussion history. This measure is valuable for the surveillance of subconcussive impacts and their associated potential negative effects.
    Journal of Neurotrauma 01/2015; DOI:10.1089/neu.2014.3666 · 3.97 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Athlete-recalled and clinically documented concussion histories have been used in research on former athletes, but both have limitations. Comparisons of these 2 types of concussion histories are needed to improve the accuracy of estimates of concussion history for future research and clinical care. To estimate the agreement between athlete-recalled and clinically documented concussion histories during college and to explore reasons for differences. Cohort study (diagnosis); Level of evidence, 3. Athlete-recalled concussion histories were provided by a convenience sample of 130 former collegiate athletes using an online questionnaire, and they were individually linked to previously collected clinical data that tracked medically diagnosed concussions at the host institution from 1996 to 2012. The intraclass correlation coefficient (ICC2,1) was used to assess agreement between athlete-recalled and clinically documented concussion histories. Descriptive analyses were performed to assess reasons for disagreement. Agreement between athlete-recalled and clinically documented concussion histories was low (ICC2,1 = 0.21; 95% confidence interval, 0.05-0.37), but it was higher for women (ICC2,1 = 0.65; 95% confidence interval, 0.44-0.79) and for athletes playing more recently (2005-2012; ICC2,1 = 0.39; 95% confidence interval, 0.01-0.67). Of the 53 athletes who self-reported college sports-related concussions, 40% believed that they sustained impacts that should have been diagnosed as concussions but were undetected, and 21% admitted nondisclosure of suspected concussions. Common reasons for nondisclosure included the following: did not think injury was serious enough (91%), did not know it was a concussion (73%), and did not want to leave the game/practice (73%). Given the low agreement between athlete-recalled and clinically documented concussion histories, methodologic research is needed to improve the quality of tools used to assess concussion histories in former athletes. © 2015 The Author(s).
    The American Journal of Sports Medicine 01/2015; 43(3). DOI:10.1177/0363546514562180 · 4.70 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Activity-related patient-reported outcome measures are an important component of assessment after knee ligament injury in young and physically active patients; however, normative data for most activity scales are limited. Objective: To present reference values by sex for the Marx Activity Rating Scale (MARS) within a young and physically active population while accounting for knee ligament injury history and sex. Study Design: Cross-sectional study. Level of Evidence: Level 2. Methods: All incoming freshman entering a US Service Academy in June of 2011 were recruited to participate in this study. MARS was administered to 1169 incoming freshmen (203 women) who consented to participate within the first week of matriculation. All subjects were deemed healthy and medically fit for military service on admission. Subjects also completed a baseline questionnaire that asked for basic demographic information and injury history. We calculated means with standard deviations, medians with interquartile ranges, and percentiles for ordinal and continuous variables, and frequencies and proportions for dichotomous variables. We also compared median scores by sex and history of knee ligament injury using the Kruskal-Wallis test. MARS was the primary outcome of interest. Results: The median MARS score was significantly higher for men when compared with women (χ2 = 13.22, df = 1, P < 0.001) with no prior history of knee ligament injury. In contrast, there was no significant difference in median MARS scores between men and women (χ2 = 0.47, df = 1, P = 0.493) who reported a history of injury. Overall, median MARS scores were significantly higher among those who reported a history of knee ligament injury when compared with those who did not (χ2 = 9.06, df = 1, P = 0.003). Conclusion: Assessing activity as a patient-reported outcome after knee ligament injury is important, and reference values for these instruments need to account for the influence of prior injury and sex.
    Sports Health A Multidisciplinary Approach 01/2015; DOI:10.1177/1941738115576121
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Both high and low indoor relative humidity (RH) directly impact indoor air quality (IAQ), an important school health concern. Prior school studies reported a high prevalence of mold, roaches, and water damage; however, few examined associations between modifiable classroom factors and RH, a quantitative indicator of dampness. We recorded RH longitudinally in 134 North Carolina classrooms (n= 9066 classroom-days) to quantify the relationships between modifiable classroom factors and average daily RH below, within, or above levels recommended to improve school IAQ (30-50% or 30-60% RH). The odds of having high RH (>60%) were 5.8 (95% Confidence Interval (CI): 2.9, 11.3) times higher in classrooms with annual compared to quarterly heating, ventilating, and air conditioning (HVAC) system maintenance, and 2.5 (95% CI: 1.5, 4.2) times higher in classrooms with HVAC economizers compared to those without economizers. Classrooms with direct expansion split systems compared to chilled water systems had 2.7 (95% CI: 1.7, 4.4) times higher odds of low RH (60%) of those without setbacks. This research suggests actionable decision points for school design and maintenance to prevent high or low classroom RH. This article is protected by copyright. All rights reserved.
    Indoor Air 12/2014; DOI:10.1111/ina.12176 · 4.20 Impact Factor
  • Zachary Kerr, Stephen W. Marshall, Johna Register-Mihalik
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: Recurrent concussion may result in neuro-dementia. However, athletes may not disclose sport-related concussion symptoms to others. Research is needed on factors influencing nondisclosure of concussion symptoms by athletes. Methods: Former collegiate athletes who played between 1987-2012 (n=797) completed an online self-administered questionnaire about impacts sustained during sport participation that should have been diagnosed by team medical staff as concussions, but were not. We asked why concussions went undiagnosed and what sources of information influenced respondents’ knowledge of concussions. Results: Of the 214 respondents self-reporting sport-related concussions (26.9% of sample), 44.9% stated that they had sustained impacts that should have been diagnosed by medical staff as concussions, but were not. Common reported factors associated with non-diagnosis were not telling team medical staff (33.2%) and no team medical staff being present (19.2%). Among those with non-disclosed concussions, common self-reported reasons for non-disclosure included: not wanting to leave the game/practice (78.9%); not wanting to let team down (71.8%); and not knowing it was a concussion (70.4%). Sources of knowledge that helped respondents realize impacts may have been undiagnosed concussions were continuing education (20.8%); friends (14.6%); and the media (71.9%), particularly ESPN (32.3%). Conclusion: Among those former collegiate athletes self-reporting sport-related concussions, nearly half of former collegiate athletes reported sustaining undiagnosed sport-related concussions. External influences such as perceived peer expectations and media coverage were influential in former athletes’ concussion reporting. This information may assist in refining public health campaigns that seek to improve disclosure of symptoms (e.g. CDC’s Heads-Up program).
    142nd APHA Annual Meeting and Exposition 2014; 11/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Young athletes suffering from a concussion should be assessed by a trained health care provider. However, many concussions are never diagnosed because children and their parents do not seek care. We examined whether parents’ knowledge and attitudes towards concussion were associated with post-concussion care-seeking for their children. Methods: A cohort of 214 youth sport parents [(63% female; parent age: 44.1(±6.5 yrs); child age: 11.6±(1.9 yrs)] completed a pre-validated concussion knowledge (max possible score=28), attitudes (max possible score=63) and care-seeking survey. Higher knowledge and attitude scores represented more favorable scores. Parents reported number of: diagnosed concussions, times their child displayed signs/symptoms of concussion following a head impact, and care-seeking behaviors following these events. The main outcome was care-seeking for all potential concussions (all-care group) vs. not (lower-care group). Results: Twenty-six percent (n=55) of parents reported that their child had ≥ 1 potential concussion. The majority of these parents (72%, n=40) did not follow-up with a medical provider after each event. Mean attitude score was 51.5(±1.6) in the all-care group and 48.1(±4.2) in the lower-care group. Mean knowledge score was 23.6(±2.5) in the all-care group and 23.6±(2.9) in the lower-care group. More favorable parental attitudes towards concussion were positively associated with care-seeking (OR: 1.7; 95% CI: 1.1, 2.8). Parents’ knowledge was not associated with better care-seeking (OR: 0.9; 95% CI: 0.7, 1.3). Conclusions: While concussion knowledge is undoubtedly important, parental attitudes were more important in predicting care-seeking. Future parental interventions should stress the importance of prompt care for all potential concussions.
    142nd APHA Annual Meeting and Exposition 2014; 11/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To compare the odds of sustaining moderate and severe head impacts, rather than mild, between high school football players with high and low visual performance. Design: Prospective quasi-experimental. Setting: Clinical Research Center/On-field. Participants: Thirty-seven high school varsity football players. Interventions: Athletes completed the Nike SPARQ Sensory Station visual assessment before the season. Head impact biomechanics were captured at all practices and games using the Head Impact Telemetry System. Main Outcome Measures: Each player was classified as either a high or low performer using a median split for each of the following visual performance measures: visual clarity, contrast sensitivity, depth perception, near-far quickness, target capture, perception span, eye-hand coordination, go/no go, and reaction time. We computed the odds of sustaining moderate and severe head impacts against the reference odds of sustaining mild head impacts across groups of high and low performers for each of the visual performance measures. Results: Players with better near-far quickness had increased odds of sustaining moderate [odds ratios (ORs), 1.27; 95% confidence intervals (CIs), 1.04-1.56] and severe head impacts (OR, 1.45; 95% CI, 1.05-2.01) as measured by Head Impact Technology severity profile. High and low performers were at equal odds on all other measures. Conclusions: Better visual performance did not reduce the odds of sustaining higher magnitude head impacts. Visual performance may play less of a role than expected for protecting against higher magnitude head impacts among high school football players. Further research is needed to determine whether visual performance influences concussion risk. Clinical Relevance: Based on our results, we do not recommend using visual training programs at the high school level for the purpose of reducing the odds of sustaining higher magnitude head impacts. Copyright
    Clinical journal of sport medicine: official journal of the Canadian Academy of Sport Medicine 11/2014; DOI:10.1097/JSM.0000000000000143 · 2.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Musculoskeletal injuries during military and sport-related training are common, costly, and potentially debilitating. There is a need to develop and implement evidence-based injury prevention strategies to reduce the burden of musculoskeletal injury. The lack of attention to implementation issues is a major factor limiting the ability to successfully reduce musculoskeletal injury rates using evidence-based injury prevention programs. This article proposes 7 steps that can be used to facilitate successful design and implementation of evidence-based injury prevention programs within the logical constraints of a real-world setting by identifying implementation barriers and associated solutions.
    Clinics in Sports Medicine 10/2014; 33(4):615-632. DOI:10.1016/j.csm.2014.06.012 · 2.58 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Little research has examined concussion outcomes in terms of impact location (ie, the area on the head in which the impact occurred). This study describes the epidemiology of concussions resulting from player-to-player collision in high school football by impact location. METHODS: National High School Sports-Related Injury Surveillance Study data (2008/2009-2012/2013) were analyzed to calculate rates and describe circumstances of football concussion (eg, symptomology, symptom resolution time, return to play) resulting from player-to-player collisions by impact location (ie, front-, back-, side-, and top-of-the-head). RESULTS: Most concussions resulting from player-to-player collisions occurred from front-of-the-head (44.7%) and side-of-the-head (22.3%) impacts. Number of symptoms reported, prevalence of reported symptoms, symptom resolution time, and length of time to return to play were not associated with impact location. However, a larger proportion of football players sustaining concussions from top-of-the-head impacts experienced loss of consciousness (8.0%) than those sustaining concussions from impacts to other areas of the head (3.5%) (injury proportion ratio 2.3; 95% confidence interval 1.2-4.2; P = .008). Players had their head down at the time of impact in a higher proportion of concussions caused by top-of-the-head impacts (86.4%) than concussions from impacts to other areas of the head (24.0%) (injury proportion ratio 3.6; 95% confidence interval 3.2-4.0; P < .001). CONCLUSIONS: Among high school football players who sustained concussions due to player-to-player collisions, concussion outcomes were generally independent of impact location. Recommended strategies for reducing the proportion of top-of-the-head impacts include improved education regarding tackling with proper "head-up" technique.
    Pediatrics 08/2014; 134(3). DOI:10.1542/peds.2014-0770 · 5.30 Impact Factor
  • Z. Y. Kerr, J. D. DeFreese, S. W. Marshall
    08/2014; 2(8). DOI:10.1177/2325967114544107
  • Source
    AOSSM Annual Meeting, Seattle, WA; 07/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: An athlete is thought to reduce head acceleration after impact by contracting the cervical musculature, which increases the effective mass of the head.
    The American Journal of Sports Medicine 06/2014; 42(9). DOI:10.1177/0363546514536685 · 4.70 Impact Factor

Publication Stats

8k Citations
1,119.03 Total Impact Points

Institutions

  • 1996–2015
    • University of North Carolina at Chapel Hill
      • • Department of Epidemiology
      • • Department of Exercise and Sport Science
      • • Injury Prevention Research Center
      North Carolina, United States
  • 2013
    • WakeMed Health & Hospitals
      Raleigh, North Carolina, United States
  • 2012
    • United States Military Academy
      • Department of Mathematical Sciences
      West Point, New York, United States
  • 2011
    • University of Georgia
      • Department of Kinesiology
      Athens, GA, United States
  • 2010
    • Palo Alto Medical Foundation
      Palo Alto, California, United States
  • 2008–2010
    • Duke University Medical Center
      • Department of Community and Family Medicine
      Durham, NC, United States
    • University of North Carolina at Charlotte
      Charlotte, North Carolina, United States
    • Iowa State University
      • Department of Industrial and Manufacturing Systems Engineering
      Ames, Iowa, United States
  • 2009
    • University of Nevada, Reno
      Reno, Nevada, United States
    • Neuroscience & Tms Treatment Center
      Nashville, Tennessee, United States
  • 2003
    • Boston University
      Boston, Massachusetts, United States
  • 1992–1997
    • University of Otago
      • • Injury Prevention Research Unit
      • • Department of Preventive & Social Medicine
      Dunedin, Otago, New Zealand
  • 1994
    • Population Services International
      Washington, Washington, D.C., United States