AJPH EDITORIALS
community violence, however, is not. Community violence is a cause of mental health problems, but it is also a consequence of the same structural inequalities (e.g., inequitable education funding, food and housing insecurity, discrimination) that generate stressors that also the increase risk of poor mental health. A problem of such complexity is unlikely to be meaningfully addressed through individual-level interventions. Promoting mental health and preventing violence requires population-based approaches, which LHDs are potentially
well positioned to lead. Much progress remains, however, to advance the role of LHDs in mental health promotion and violence prevention and transform Winslow’s1 vision into a reality. Jonathan Purtle, DrPH, MSc
ACKNOWLEDGMENTS The author is supported by the National Institute of Mental Health (R21MH111806; R25MH08091). I thank Rosie Mae Henson, MPH, for her thoughtful feedback on this article and Note. The contents of this article are solely the responsibility of the author and
do not necessarily represent the official views of the National Institute of Mental Health.
1. Winslow C-E. Public health at the crossroads. Am J Public Health. 1926; 16(11):1075–1085.
5. Purtle J, Klassen AC, Kolker J, Buehler JW. Prevalence and correlates of local health department activities to address mental health in the United States. Prev Med. 2016;82:20–27.
2. Cohen N, Galea S., eds. Population Mental Health: Evidence, Policy, and Public Health Practice. New York: Routledge; 2011.
6. Purtle J, Peters R, Kolker J, Klassen AC. Factors perceived as influencing local health department involvement in mental health. Am J Prev Med. 2017;52(1):64–73.
3. Krueger J, Counts N, Riley B. Promoting mental health and well-being in public health law and practice. J Law Med Ethics 2017;45(1, suppl):37–40.
7. Walker ER, Kwon J, Lang DL, Levinson RM, Druss BG. Mental health training in schools of public health: history, current status, and future opportunities. Public Health Rep. 2016;131(1): 208–217.
REFERENCES
4. Fowler PJ, Tompsett CJ, Braciszewski JM, Jacques-Tiura AJ, Baltes BB. Community violence: a meta-analysis on the effect of exposure and mental health outcomes of children and
Do Our Cells Pay the Price When We Sit Too Much? See also Xue et al., p. 1425.
Excessive sedentary behaviors, typically characterized as sitting for prolonged periods, are an increasingly important health risk in our current age of televisions and computers. The link between television watching and obesity was first reported more than 30 years ago.1 Since then, extensive data have confirmed this connection in children and adults. In addition, emerging evidence suggests that, independent of exercise levels, sedentary behaviors are associated with higher risk of weightrelated chronic diseases— including cardiometabolic conditions and certain cancers— as well as overall mortality (reviewed in de Rezende et al.2). What remains unclear, however, is the influence of such behaviors on the function of critical physiological systems in the body. In this issue of AJPH, the report by Xue et al. (p. 1425) helps shed light on this important question.
1360
Editorial
Du
MAGNITUDE OF THE PROBLEM To place such knowledge in context, it is important to highlight the prevalence of sedentary behaviors in the United States. According to nationally representative data from the most recently available cycle of the National Health and Nutrition Examination Survey (NHANES), almost one third of US adults aged 20 years or older reported sitting nine or more hours a day in 2013 to 2014 (Figure 1). Television viewing in particular is the most commonly reported regular activity separate from work or sleep, and accounts for a large proportion of total sitting time. Recent Nielsen data indicate that US adults watch an average of more than five hours of television a day.3 These figures are troublingly high when compared with the Centers for Disease Control and Prevention
adolescents. Dev Psychopathol. 2009; 21(1):227–259.
guidelines to avoid all inactivity,4 and underscore the critical need to better understand the longterm biological consequences of televison watching and other types of sedentary behaviors. In recognition, this question was deemed a priority area for future research by a recent Sedentary Behavior Workshop sponsored by the National Institutes of Health.5
RELATIONSHIP WITH TELOMERE LENGTH To this end, the report by Xue et al. marks an important contribution to the literature. The authors investigated the relationships between sedentary
behaviors, physical activities, and leukocyte telomere length in a cross-sectional study of 518 Chinese adults aged 20 to 70 years. Telomeres are DNA– protein structures that cap and protect the ends of chromosomes.6 Because telomeres shorten with cell division or damage, telomere length can act as a quantifiable proxy of biological aging and damage accumulated across a lifespan. Telomere length has been linked to a number of health outcomes in observational studies, and may facilitate chronic disease development.6 Thus, understanding the associations between sedentary behaviors, physical activities, and telomere length not only helps assess the degree to which these modifiable behaviors may affect cellular damage, but can also help uncover if altered cellular aging is a potential mechanism driving previously observed
ABOUT THE AUTHOR Mengmeng Du is with the Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, and the Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA. Correspondence should be sent to Mengmeng Du, Memorial Sloan Kettering Cancer Center, Department of Epidemiology and Biostatistics, 485 Lexington Ave, Floor 2, New York, NY 10017 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link. This editorial was accepted June 14, 2017. doi: 10.2105/AJPH.2017.303981
AJPH
September 2017, Vol 107, No. 9
AJPH EDITORIALS
REMAINING CONSIDERATIONS
Percentage
50 36.6
40 26.2
30
18.3
20 10
9.0
7.8
2.0
0 0 to < 3
3 to < 6
6 to < 9 9 to < 12 12 to < 15
≥ 15
Hours per Day Sitting Note. Data were weighted to be nationally representative. Error bars indicate 95% confidence intervals that incorporate these sample weights.
FIGURE 1—Percentage of US Adults Aged 20 Years or Older Reporting Sitting or Reclining, by Hours Each Day, in 2013 to 2014: National Health and Nutrition Examination Survey
associations between these behaviors and health outcomes. In the study by Xue et al. conducted in 518 adults from Southwest China, independent of physical activity and energy intake, each one-hour increase in television watching was associated with a 72-base-pair decrease in mean telomere length. This translated to a difference of approximately 1.2 to 1.8 years in biological age. Although this association is modest, the difference in telomere length for a three-hour increase in television watching would be roughly comparable to the difference in telomere lengths comparing smokers with nonsmokers.7 The authors note no associations with any of the other sedentary behavior or physical activity variables.
STRENGTHS OF THE STUDY The study by Xue et al. has several important strengths, including the use of Southern Blot, long considered the gold standard of telomere length measurement, as well as the comprehensive assessment of sedentary behaviors, physical activities, and potential
September 2017, Vol 107, No. 9
AJPH
predictors of telomere length unavailable in similar studies, such as socioeconomic status and stress levels. As the authors note, levels and types of sedentary behaviors and physical activity vary widely across populations (which they suggest may account in part for the reported null associations with exercise) and it is important to study associations in different geographical regions and racial/ ethnic groups. Importantly, reported estimates were both adjusted and unadjusted for percent body fat, physical activities (or sedentary behaviors for analyses of activity), and total energy intake. Sedentary behavior may increase risk of chronic disease by altering energy expenditure or energy intake.2 Here, the observed association with television watching remained after they adjusted for physical activities and energy intake, suggesting that other factors may drive the association between television watching and telomere length. For instance, television watching may be associated with unhealthy eating habits or greater exposure to unhealthy lifestyle choices (via ads), or television may displace time spent on other activities.2
Despite the strengths, the findings of Xue et al. should be interpreted with caution for several reasons. First, as the authors note, the body of literature examining sedentary behaviors and telomere length is scarce— previously comprising two cross-sectional studies with inconsistent results. Thus, large follow-up studies are needed to confirm the observed association with television watching. In particular, beyond the scope of the current report, these studies are needed to carefully tease apart the role of diet—whether as a confounder or mediator—in the relationship between television watching and telomere length. Second, telomere length was assessed with a single measure, precluding the evaluation of telomere shortening. Although resource-intensive, a prospective study with at least two measures of telomere length would be required to test if sedentary behaviors or physical activities affect telomere shortening. Third, in addition to television watching, the authors explored associations with many types of sedentary behaviors (e.g., computer or phone, screen-based, total) and physical activities (e.g., walking, household activities, moderate-to-vigorous activities, calisthenics or aerobics). It is possible given this exploratory design that the association with television watching was attributable to chance, although studies evaluating studies of other health conditions support the observed association with television watching.2 Moreover, the authors evaluated non–television-watching sedentary behaviors and physical activities as continuous variables; associations may have been missed
if they were U-shaped, as has been suggested for physical activity.8 Finally, the authors report a difference by age in the association between television watching and telomere length— with the association present in younger participants (statistically significant in those aged 20–40 years, and suggestive in those aged 41–55 years), but not in participants aged 56 to 70 years. Similar findings have been reported, but data examining age differences are limited and exploratory.2 Future large studies should directly examine this hypothesis. Taken together, Xue et al. contribute compelling data to the emerging body of evidence for the long-term biological consequences of television watching. Many questions remain, however, and these findings warrant follow-up. Confirming the connection between television watching and accelerated aging on a cellular level would provide further proof of principle for telomere length to serve as a proxy that could help identify and compare modifiable behaviors most likely to cause similar types of cellular damage. In addition, these data would provide the impetus for mediation studies to confirm if short telomeres are a possible mechanism that may, in part, underlie published relationships between television watching and health. This could, in turn, help identify individuals most at risk for the detrimental health effects of such behaviors. Although everyone could benefit from minimizing sedentary behaviors, such knowledge can inform more direct targeting of public health interventions aimed at minimizing such behaviors and promoting physical activity. Mengmeng Du, ScD
Du
Editorial
1361
AJPH EDITORIALS
television viewing in children and adolescents. Pediatrics. 1985;75(5):807–812.
ACKNOWLEDGMENTS The author is supported by P30 CA008748 from the National Cancer Institute, National Institutes of Health. For the NHANES statistical analysis, the author thanks Kelli O’Connell for providing analytic support and Elizabeth D. Kantor for providing guidance.
REFERENCES 1. Dietz WH Jr, Gortmaker SL. Do we fatten our children at the television set? Obesity and
2. de Rezende LF, Rodrigues Lopes M, Rey-Lopez JP, Matsudo VK, Luiz Odo C. Sedentary behavior and health outcomes: an overview of systematic reviews. PLoS One. 2014;9(8):e105620. 3. Nielsen Co. Total audience report Q1 2016. Available at: http://www.nielsen. com/us/en/insights/reports/2016/thetotal-audience-report-q1-2016.html. Accessed June 13, 2017.
4. Centers for Disease Control and Prevention. Current physical activity guidelines. Available at: https://www.cdc. gov/cancer/dcpc/prevention/policies_ practices/physical_activity/guidelines.htm. Accessed June 13, 2017. 5. Thyfault JP, Du M, Kraus WE, Levine JA, Booth FW. Physiology of sedentary behavior and its relationship to health outcomes. Med Sci Sports Exerc. 2015;47(6):1301–1305.
Transgender people face frequent experiences of discrimination, violence, social and economic marginalization, and abuse across the lifespan. International efforts to track the murder of transgender people suggest that a transgender person is murdered at least once every three days.1 However, in the United States there is no formal data collection effort that can be used to describe the nature, frequency, or extent of transgender homicides. In an effort to address this gap, Dinno (p. 1441) used nongovernmental organization (NGO) data on the murder of transgender people and various estimates of the transgender population in the United States as well as federal data on cisgender people to create a range of estimates of the transgender homicide rate compared with the cisgender homicide rate from 2010 to 2014. Findings suggest that transgender people overall may not face a higher risk of being murdered than do cisgender people but that young transgender women of color almost certainly face a higher chance of being murdered.
1362
Editorial
Stotzer
DATA CONCERNS Scholars focusing on the experiences of transgender people have historically hesitated to specify the relative risk of transgender homicide because of substantial limitations of available data. The definition of transgender itself varies and can represent a very broad or very narrow category of people who defy traditional expectations of gender. Although this definitional issue may seem academic at first, it has significant consequences for how to categorize both murder victims and the estimated transgender population. Of those two numbers, the size of the transgender population is better researched, although a fairly narrow definition of transgender is used. Current estimates have been derived from studies that use random sampling and include measures of gender identity that go beyond the simple binary construct of “male” or “female”2–4 (although frequently only providing a third, “transgender” or “other” category). Some states have chosen to include questions about gender
8. Savela S, Saijonmaa O, Strandberg TE, et al. Physical activity in midlife and telomere length measured in old age. Exp Gerontol. 2013;48(1):81–84.
6. Calado RT, Young NS. Telomere diseases. N Engl J Med. 2009;361(24):2353–2365.
Data Sources Hinder Our Understanding of Transgender Murders See also Dinno, p. 1441.
7. Du M, Prescott J, Kraft P, et al. Physical activity, sedentary behavior, and leukocyte telomere length in women. Am J Epidemiol. 2012;175(5):414–422.
identity in national surveys, such as the Behavioral Risk Factor Surveillance System,2,3 that have then been extrapolated to create national estimates, but there are no systematic federal surveys that include questions to identify transgender people. These preliminary estimates will likely need frequent revision, because younger cohorts identify as transgender in greater numbers than do older cohorts.4 Younger people are also choosing a wider variety of labels for their gender identity (e.g., genderqueer, nonbinary) that may not be captured in questions that ask only whether the respondent is male, female, or transgender. However, gender identity questions will not be included in the 2020 Census, one of the best tools for outlining the demographics of the US population, delaying an official count of transgender people by at least another decade. Thus, although we have better
estimates about transgender population size in the United States than ever before, those estimates should be used cautiously. Although we are getting closer to an accurate idea of transgender population size, the issue of defining transgender has also plagued the tracking of transgender homicide, among other data issues. The primary method of collecting crime data at the federal level is through the Federal Bureau of Investigation’s Uniform Crime Reports, a system in which local law enforcement agencies voluntarily submit crime statistics from their jurisdiction to the Federal Bureau of Investigation each year in aggregate form. Notoriously problematic because of the voluntary nature of the program and the various ways that laws vary or are interpreted across jurisdictions, the Uniform Crime Reports are also often assumed to be a significant undercount of crime. In addition, there are few opportunities for local law enforcement agencies to indicate that a victim (or perpetrator) is transgender outside the gender identity– motivated bias crimes category. Another federal data source,
ABOUT THE AUTHOR Rebecca L. Stotzer is with the Myron B. Thompson School of Social Work, University of Hawai‘i, M anoa. Correspondence should be sent to Rebecca L. Stotzer, University of Hawai‘i at Manoa, Myron B. Thompson School of Social Work, 2430 Campus Rd., Honolulu HI, 96813 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link. This editorial was accepted June 8, 2017. doi: 10.2105/AJPH.2017.303973
AJPH
September 2017, Vol 107, No. 9
community violence, however, is not. Community violence is a cause of mental health problems, but it is also a consequence of the same structural inequalities (e.g., inequitable education funding, food and housing insecurity, discrimination) that generate stressors that also the increase risk of poor mental health. A problem of such complexity is unlikely to be meaningfully addressed through individual-level interventions. Promoting mental health and preventing violence requires population-based approaches, which LHDs are potentially
well positioned to lead. Much progress remains, however, to advance the role of LHDs in mental health promotion and violence prevention and transform Winslow’s1 vision into a reality. Jonathan Purtle, DrPH, MSc
ACKNOWLEDGMENTS The author is supported by the National Institute of Mental Health (R21MH111806; R25MH08091). I thank Rosie Mae Henson, MPH, for her thoughtful feedback on this article and Note. The contents of this article are solely the responsibility of the author and
do not necessarily represent the official views of the National Institute of Mental Health.
1. Winslow C-E. Public health at the crossroads. Am J Public Health. 1926; 16(11):1075–1085.
5. Purtle J, Klassen AC, Kolker J, Buehler JW. Prevalence and correlates of local health department activities to address mental health in the United States. Prev Med. 2016;82:20–27.
2. Cohen N, Galea S., eds. Population Mental Health: Evidence, Policy, and Public Health Practice. New York: Routledge; 2011.
6. Purtle J, Peters R, Kolker J, Klassen AC. Factors perceived as influencing local health department involvement in mental health. Am J Prev Med. 2017;52(1):64–73.
3. Krueger J, Counts N, Riley B. Promoting mental health and well-being in public health law and practice. J Law Med Ethics 2017;45(1, suppl):37–40.
7. Walker ER, Kwon J, Lang DL, Levinson RM, Druss BG. Mental health training in schools of public health: history, current status, and future opportunities. Public Health Rep. 2016;131(1): 208–217.
REFERENCES
4. Fowler PJ, Tompsett CJ, Braciszewski JM, Jacques-Tiura AJ, Baltes BB. Community violence: a meta-analysis on the effect of exposure and mental health outcomes of children and
Do Our Cells Pay the Price When We Sit Too Much? See also Xue et al., p. 1425.
Excessive sedentary behaviors, typically characterized as sitting for prolonged periods, are an increasingly important health risk in our current age of televisions and computers. The link between television watching and obesity was first reported more than 30 years ago.1 Since then, extensive data have confirmed this connection in children and adults. In addition, emerging evidence suggests that, independent of exercise levels, sedentary behaviors are associated with higher risk of weightrelated chronic diseases— including cardiometabolic conditions and certain cancers— as well as overall mortality (reviewed in de Rezende et al.2). What remains unclear, however, is the influence of such behaviors on the function of critical physiological systems in the body. In this issue of AJPH, the report by Xue et al. (p. 1425) helps shed light on this important question.
1360
Editorial
Du
MAGNITUDE OF THE PROBLEM To place such knowledge in context, it is important to highlight the prevalence of sedentary behaviors in the United States. According to nationally representative data from the most recently available cycle of the National Health and Nutrition Examination Survey (NHANES), almost one third of US adults aged 20 years or older reported sitting nine or more hours a day in 2013 to 2014 (Figure 1). Television viewing in particular is the most commonly reported regular activity separate from work or sleep, and accounts for a large proportion of total sitting time. Recent Nielsen data indicate that US adults watch an average of more than five hours of television a day.3 These figures are troublingly high when compared with the Centers for Disease Control and Prevention
adolescents. Dev Psychopathol. 2009; 21(1):227–259.
guidelines to avoid all inactivity,4 and underscore the critical need to better understand the longterm biological consequences of televison watching and other types of sedentary behaviors. In recognition, this question was deemed a priority area for future research by a recent Sedentary Behavior Workshop sponsored by the National Institutes of Health.5
RELATIONSHIP WITH TELOMERE LENGTH To this end, the report by Xue et al. marks an important contribution to the literature. The authors investigated the relationships between sedentary
behaviors, physical activities, and leukocyte telomere length in a cross-sectional study of 518 Chinese adults aged 20 to 70 years. Telomeres are DNA– protein structures that cap and protect the ends of chromosomes.6 Because telomeres shorten with cell division or damage, telomere length can act as a quantifiable proxy of biological aging and damage accumulated across a lifespan. Telomere length has been linked to a number of health outcomes in observational studies, and may facilitate chronic disease development.6 Thus, understanding the associations between sedentary behaviors, physical activities, and telomere length not only helps assess the degree to which these modifiable behaviors may affect cellular damage, but can also help uncover if altered cellular aging is a potential mechanism driving previously observed
ABOUT THE AUTHOR Mengmeng Du is with the Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, and the Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA. Correspondence should be sent to Mengmeng Du, Memorial Sloan Kettering Cancer Center, Department of Epidemiology and Biostatistics, 485 Lexington Ave, Floor 2, New York, NY 10017 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link. This editorial was accepted June 14, 2017. doi: 10.2105/AJPH.2017.303981
AJPH
September 2017, Vol 107, No. 9
AJPH EDITORIALS
REMAINING CONSIDERATIONS
Percentage
50 36.6
40 26.2
30
18.3
20 10
9.0
7.8
2.0
0 0 to < 3
3 to < 6
6 to < 9 9 to < 12 12 to < 15
≥ 15
Hours per Day Sitting Note. Data were weighted to be nationally representative. Error bars indicate 95% confidence intervals that incorporate these sample weights.
FIGURE 1—Percentage of US Adults Aged 20 Years or Older Reporting Sitting or Reclining, by Hours Each Day, in 2013 to 2014: National Health and Nutrition Examination Survey
associations between these behaviors and health outcomes. In the study by Xue et al. conducted in 518 adults from Southwest China, independent of physical activity and energy intake, each one-hour increase in television watching was associated with a 72-base-pair decrease in mean telomere length. This translated to a difference of approximately 1.2 to 1.8 years in biological age. Although this association is modest, the difference in telomere length for a three-hour increase in television watching would be roughly comparable to the difference in telomere lengths comparing smokers with nonsmokers.7 The authors note no associations with any of the other sedentary behavior or physical activity variables.
STRENGTHS OF THE STUDY The study by Xue et al. has several important strengths, including the use of Southern Blot, long considered the gold standard of telomere length measurement, as well as the comprehensive assessment of sedentary behaviors, physical activities, and potential
September 2017, Vol 107, No. 9
AJPH
predictors of telomere length unavailable in similar studies, such as socioeconomic status and stress levels. As the authors note, levels and types of sedentary behaviors and physical activity vary widely across populations (which they suggest may account in part for the reported null associations with exercise) and it is important to study associations in different geographical regions and racial/ ethnic groups. Importantly, reported estimates were both adjusted and unadjusted for percent body fat, physical activities (or sedentary behaviors for analyses of activity), and total energy intake. Sedentary behavior may increase risk of chronic disease by altering energy expenditure or energy intake.2 Here, the observed association with television watching remained after they adjusted for physical activities and energy intake, suggesting that other factors may drive the association between television watching and telomere length. For instance, television watching may be associated with unhealthy eating habits or greater exposure to unhealthy lifestyle choices (via ads), or television may displace time spent on other activities.2
Despite the strengths, the findings of Xue et al. should be interpreted with caution for several reasons. First, as the authors note, the body of literature examining sedentary behaviors and telomere length is scarce— previously comprising two cross-sectional studies with inconsistent results. Thus, large follow-up studies are needed to confirm the observed association with television watching. In particular, beyond the scope of the current report, these studies are needed to carefully tease apart the role of diet—whether as a confounder or mediator—in the relationship between television watching and telomere length. Second, telomere length was assessed with a single measure, precluding the evaluation of telomere shortening. Although resource-intensive, a prospective study with at least two measures of telomere length would be required to test if sedentary behaviors or physical activities affect telomere shortening. Third, in addition to television watching, the authors explored associations with many types of sedentary behaviors (e.g., computer or phone, screen-based, total) and physical activities (e.g., walking, household activities, moderate-to-vigorous activities, calisthenics or aerobics). It is possible given this exploratory design that the association with television watching was attributable to chance, although studies evaluating studies of other health conditions support the observed association with television watching.2 Moreover, the authors evaluated non–television-watching sedentary behaviors and physical activities as continuous variables; associations may have been missed
if they were U-shaped, as has been suggested for physical activity.8 Finally, the authors report a difference by age in the association between television watching and telomere length— with the association present in younger participants (statistically significant in those aged 20–40 years, and suggestive in those aged 41–55 years), but not in participants aged 56 to 70 years. Similar findings have been reported, but data examining age differences are limited and exploratory.2 Future large studies should directly examine this hypothesis. Taken together, Xue et al. contribute compelling data to the emerging body of evidence for the long-term biological consequences of television watching. Many questions remain, however, and these findings warrant follow-up. Confirming the connection between television watching and accelerated aging on a cellular level would provide further proof of principle for telomere length to serve as a proxy that could help identify and compare modifiable behaviors most likely to cause similar types of cellular damage. In addition, these data would provide the impetus for mediation studies to confirm if short telomeres are a possible mechanism that may, in part, underlie published relationships between television watching and health. This could, in turn, help identify individuals most at risk for the detrimental health effects of such behaviors. Although everyone could benefit from minimizing sedentary behaviors, such knowledge can inform more direct targeting of public health interventions aimed at minimizing such behaviors and promoting physical activity. Mengmeng Du, ScD
Du
Editorial
1361
AJPH EDITORIALS
television viewing in children and adolescents. Pediatrics. 1985;75(5):807–812.
ACKNOWLEDGMENTS The author is supported by P30 CA008748 from the National Cancer Institute, National Institutes of Health. For the NHANES statistical analysis, the author thanks Kelli O’Connell for providing analytic support and Elizabeth D. Kantor for providing guidance.
REFERENCES 1. Dietz WH Jr, Gortmaker SL. Do we fatten our children at the television set? Obesity and
2. de Rezende LF, Rodrigues Lopes M, Rey-Lopez JP, Matsudo VK, Luiz Odo C. Sedentary behavior and health outcomes: an overview of systematic reviews. PLoS One. 2014;9(8):e105620. 3. Nielsen Co. Total audience report Q1 2016. Available at: http://www.nielsen. com/us/en/insights/reports/2016/thetotal-audience-report-q1-2016.html. Accessed June 13, 2017.
4. Centers for Disease Control and Prevention. Current physical activity guidelines. Available at: https://www.cdc. gov/cancer/dcpc/prevention/policies_ practices/physical_activity/guidelines.htm. Accessed June 13, 2017. 5. Thyfault JP, Du M, Kraus WE, Levine JA, Booth FW. Physiology of sedentary behavior and its relationship to health outcomes. Med Sci Sports Exerc. 2015;47(6):1301–1305.
Transgender people face frequent experiences of discrimination, violence, social and economic marginalization, and abuse across the lifespan. International efforts to track the murder of transgender people suggest that a transgender person is murdered at least once every three days.1 However, in the United States there is no formal data collection effort that can be used to describe the nature, frequency, or extent of transgender homicides. In an effort to address this gap, Dinno (p. 1441) used nongovernmental organization (NGO) data on the murder of transgender people and various estimates of the transgender population in the United States as well as federal data on cisgender people to create a range of estimates of the transgender homicide rate compared with the cisgender homicide rate from 2010 to 2014. Findings suggest that transgender people overall may not face a higher risk of being murdered than do cisgender people but that young transgender women of color almost certainly face a higher chance of being murdered.
1362
Editorial
Stotzer
DATA CONCERNS Scholars focusing on the experiences of transgender people have historically hesitated to specify the relative risk of transgender homicide because of substantial limitations of available data. The definition of transgender itself varies and can represent a very broad or very narrow category of people who defy traditional expectations of gender. Although this definitional issue may seem academic at first, it has significant consequences for how to categorize both murder victims and the estimated transgender population. Of those two numbers, the size of the transgender population is better researched, although a fairly narrow definition of transgender is used. Current estimates have been derived from studies that use random sampling and include measures of gender identity that go beyond the simple binary construct of “male” or “female”2–4 (although frequently only providing a third, “transgender” or “other” category). Some states have chosen to include questions about gender
8. Savela S, Saijonmaa O, Strandberg TE, et al. Physical activity in midlife and telomere length measured in old age. Exp Gerontol. 2013;48(1):81–84.
6. Calado RT, Young NS. Telomere diseases. N Engl J Med. 2009;361(24):2353–2365.
Data Sources Hinder Our Understanding of Transgender Murders See also Dinno, p. 1441.
7. Du M, Prescott J, Kraft P, et al. Physical activity, sedentary behavior, and leukocyte telomere length in women. Am J Epidemiol. 2012;175(5):414–422.
identity in national surveys, such as the Behavioral Risk Factor Surveillance System,2,3 that have then been extrapolated to create national estimates, but there are no systematic federal surveys that include questions to identify transgender people. These preliminary estimates will likely need frequent revision, because younger cohorts identify as transgender in greater numbers than do older cohorts.4 Younger people are also choosing a wider variety of labels for their gender identity (e.g., genderqueer, nonbinary) that may not be captured in questions that ask only whether the respondent is male, female, or transgender. However, gender identity questions will not be included in the 2020 Census, one of the best tools for outlining the demographics of the US population, delaying an official count of transgender people by at least another decade. Thus, although we have better
estimates about transgender population size in the United States than ever before, those estimates should be used cautiously. Although we are getting closer to an accurate idea of transgender population size, the issue of defining transgender has also plagued the tracking of transgender homicide, among other data issues. The primary method of collecting crime data at the federal level is through the Federal Bureau of Investigation’s Uniform Crime Reports, a system in which local law enforcement agencies voluntarily submit crime statistics from their jurisdiction to the Federal Bureau of Investigation each year in aggregate form. Notoriously problematic because of the voluntary nature of the program and the various ways that laws vary or are interpreted across jurisdictions, the Uniform Crime Reports are also often assumed to be a significant undercount of crime. In addition, there are few opportunities for local law enforcement agencies to indicate that a victim (or perpetrator) is transgender outside the gender identity– motivated bias crimes category. Another federal data source,
ABOUT THE AUTHOR Rebecca L. Stotzer is with the Myron B. Thompson School of Social Work, University of Hawai‘i, M anoa. Correspondence should be sent to Rebecca L. Stotzer, University of Hawai‘i at Manoa, Myron B. Thompson School of Social Work, 2430 Campus Rd., Honolulu HI, 96813 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link. This editorial was accepted June 8, 2017. doi: 10.2105/AJPH.2017.303973
AJPH
September 2017, Vol 107, No. 9
