- Email: [email protected]
Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S
Journal Pre-proof Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S Nicole L. Bolick, MPH, MS, Linglin Huang, MS, Arash Mostaghimi, MD, MPH, Rebecca I. Hartman, MD, MPH PII:
S0190-9622(20)30265-6
DOI:
https://doi.org/10.1016/j.jaad.2020.02.027
Reference:
YMJD 14244
To appear in:
Journal of the American Academy of Dermatology
Received Date: 4 September 2019 Revised Date:
4 February 2020
Accepted Date: 10 February 2020
Please cite this article as: Bolick NL, Huang L, Mostaghimi A, Hartman RI, Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S, Journal of the American Academy of Dermatology (2020), doi: https://doi.org/10.1016/j.jaad.2020.02.027. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
1
Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S.
2 3
Nicole L. Bolick, MPH, MS1,2*; Linglin Huang, MS1*; Arash Mostaghimi, MD, MPH3; Rebecca I.
4
Hartman, MD, MPH3,4
5
1
6
2
7
3
8
Boston, MA
9
4
10
Harvard T.H. Chan School of Public Health, Boston, MA The Brody School of Medicine at East Carolina University, Greenville, NC Department of Dermatology, Brigham and Women’s Hospital and Harvard Medical School,
Department of Dermatology, VA Integrated Service Network (VISN-1), Jamaica Plain, MA
*Denotes co-first authors
11 12
Address Correspondence to:
13
Rebecca Hartman, MD, MPH
14
Department of Dermatology, Brigham and Women’s Hospital
15
221 Longwood Avenue
16
Boston, MA 02115
17
[email protected]
18 19
Conflicts of Interest Disclosures: None
20
Funding Sources: Dr. Hartman is supported by an American Skin Association Research Grant
21
(120795).
22 23
Word Count: 499
24
Table Count: 1
25
Figure Count: 1
26
Reference Count: 4
27 28
To the editor: Skin cancer incidence is rising in the United States (US) despite public health efforts
29
encouraging skin cancer prevention.1 We analyzed temporal trends of skin-cancer related
30
primary (concerned with disease prevention) and secondary (concerned with early disease
31
detection) preventive behaviors to examine for potential areas for improvement.
32
The National Health Information Survey (NHIS) was examined over a ten-year period
33
from 2005 to 2015.2 The NHIS conducts representative population-based annual interviewing of
34
the adult US population with response rates of 70.1% to 80.7% during the examined period.2
35
Our outcomes of interest were: use of sun protective measures (including sun avoidance,
36
protective clothing, and sunscreen use), lifetime history of full body skin examination (FBSE) by
37
a physician, and histories in the past year of indoor tanning and two or more sunburns.
38
Specifically, use of sun protective measures were defined as use always or most of the time
39
when outside for more than one hour on a warm sunny day. Protective clothing included at least
40
one of the following: long pants, hat, or long-sleeved shirt. Of note, a small percentage of
41
individuals answered the sun protection questions by stating that they “don’t go out in the sun.”2
42
These individuals and those with unknown or missing responses were excluded from analysis.2
43
Multivariable logistic regression analyses were used to assess the association between
44
time period and weighted prevalence of sun protective behaviors, adjusting for sex, region,
45
health insurance, alcohol use, smoking status, education, personal and family histories of skin
46
cancer, income, race, and skin reaction to the sun. P-values were adjusted for multiple
47
comparisons. Due to substantial missing data for income, NHIS imputed data was used.2
48
The weighted study included a total of 67,471 individuals. From 2005 to 2015, the
49
unadjusted and adjusted prevalence of most skin cancer preventive behaviors rose, including
50
sun avoidance, sunscreen use, and FBSE, with the exception of wearing protective clothing
51
(Table 1, Figure 1). While the adjusted prevalence of indoor tanning decreased, that of sunburn
52
increased (Table 1, Figure 1).
53
Indoor tanning is substantially decreasing, implying the success of targeted legislative
54
and public health efforts.3 Nevertheless, despite increases in most sun protective behaviors, the
55
rates of primary and secondary skin preventive behaviors remain suboptimal, and the
56
prevalence of multiple sunburns is rising. It is unclear if the rise in sunburns is due to heightened
57
sun awareness causing increased reporting or inadequate use of sun protection. Researchers in
58
Australia have examined temporal changes in sun protective behaviors, and found
59
improvements following the implementation of the SunSmart public health campaign, with
60
decreases in reported sunburns and tan preference, and increases in use of sun protective
61
behaviors, although these benefits may be plateauing more recently.4 Further research is
62
needed in the U.S. population to examine why the prevalence of multiple sunburns is rising and
63
the impact of increased adoption of sun protective behaviors on skin cancer incidence.
64
Additionally, the success of public health efforts against indoor tanning may provide guidance
65
for future public health efforts aimed at skin cancer prevention.
66
67
Acknowledgement
68
Author Contributions: Bolick, Huang, and Hartman had full access to all the data in the study
69
and take responsibility for the integrity of the data and the accuracy of the data analysis. Study
70
concept and design: Bolick, Hartman, and Mostaghimi. Acquisition, analysis, and interpretation
71
of Data: Bolick, Huang, and Hartman. Drafting of Manuscript and revising critically for important
72
intellectual content: Bolick, Huang, Hartman, and Mostaghimi.
73
References
74 75 76 77
1. U.S. Cancer Statistics Working Group. U.S. Cancer Statistics Data Visualizations Tool, based on November 2018 submission data (1999-2016): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; www.cdc.gov/cancer/dataviz, June 2019.
78 79
2. National Center for Health Statistics. Survey Description, National Health Interview Survey, 2015. Hyattsville, MD: National Center for Health Statistics; 2015.
80 81 82
3. Lazovich D. Effect of Parental Permission and Age Restriction Laws on US Adolescent Indoor Tanning Trends. Am J Public Health. 2018;108(7):851-853. doi:10.2105/AJPH.2018.304458
83 84
4. Dobbinson SJ, Wakefield MA, Jamsen KM, et al. Weekend Sun Protection and Sunburn in Australia. Am J Prev Med. 2008;34(2):94-101. doi:10.1016/j.amepre.2007.09.024
85
86 87 88 89
Figure Legend
90
predictive marginal means. Adjusting covariates include sex, census region, health insurance
91
coverage, alcohol use, smoking status, education level, personal history of skin cancer, family
92
history of skin cancer, income, race, and skin reaction to the sun.
Figure 1: Adjusted prevalences of sun protective behaviors over time. The prevalences are the
Table 1 Year, Unadjusted Prevalence, % (95% CI) 2005
2010
2015
Year, Adjusted Prevalence, % (95% CI) Ptrend value*
Adjuste d P2015vs.2
2005
2010
2015
31.7 (30.932.6) 35.9 (35.136.7) 31.5 (30.732.2) 19 (18.419.6) 14.1 (13.514.8) 18.2 (17.718.8)
35.5 (34.736.4) 38.4 (37.539.2) 33.1 (32.234) 22.4 (21.723.1) 6.2 (5.76.7) 21.1 (20.421.8)
36.8 (35.937.6) 37.2 (36.338.1) 34.3 (33.535.1) 22.4 (21.723) 4.1 (3.84.5) 19.9 (19.220.7)
Ptrend value*‡
Adjusted P2015vs.2005 value†
<0.001
<0.001
<0.001
0.098
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.001
005
value† Sun Avoidance Protective Clothing Sunscreen Use Full Body Skin Exam Indoor Tanning Sunburn
31.1 (30.232) 35.5 (34.736.3) 30.6 (29.831.3) 18.1 (17.418.7) 14.8 (14.115.5) 18.7 (18.119.3)
35.4 (34.536.3) 38.5 (37.539.4) 32.7 (31.733.6) 22 (21.222.8) 6.2 (5.76.7) 21.0 (20.221.7)
37.5 (36.538.4) 37.5 (36.638.4) 35.5 (34.736.4) 23.6 (22.924.4) 3.9 (3.54.3) 19.6 (18.820.4)
<0.001
<0.001
<0.001
0.006
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.198
Table 1: Unadjusted and adjusted prevalence of sun protective behaviors by survey year. Adjusting covariates include sex, census region, health insurance coverage, alcohol use, smoking status, education level, personal history of skin cancer, family history of skin cancer, income, race, and skin reaction to the sun.
*Wald tests based on the logistic regression were performed to test for any change in log odds over 2005, 2010 and 2015. †Pairwise Wald tests based on the logistic regression were performed to test for change in log odds in any two time points. P-values were adjusted for multiple comparisons using Benjamini & Hochberg (1995). ‡Median P rule was used to obtain the final adjusted p-values due to multiple imputation for income.
S0190-9622(20)30265-6
DOI:
https://doi.org/10.1016/j.jaad.2020.02.027
Reference:
YMJD 14244
To appear in:
Journal of the American Academy of Dermatology
Received Date: 4 September 2019 Revised Date:
4 February 2020
Accepted Date: 10 February 2020
Please cite this article as: Bolick NL, Huang L, Mostaghimi A, Hartman RI, Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S, Journal of the American Academy of Dermatology (2020), doi: https://doi.org/10.1016/j.jaad.2020.02.027. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
1
Temporal Trends in Primary and Secondary Skin Cancer Prevention in the U.S.
2 3
Nicole L. Bolick, MPH, MS1,2*; Linglin Huang, MS1*; Arash Mostaghimi, MD, MPH3; Rebecca I.
4
Hartman, MD, MPH3,4
5
1
6
2
7
3
8
Boston, MA
9
4
10
Harvard T.H. Chan School of Public Health, Boston, MA The Brody School of Medicine at East Carolina University, Greenville, NC Department of Dermatology, Brigham and Women’s Hospital and Harvard Medical School,
Department of Dermatology, VA Integrated Service Network (VISN-1), Jamaica Plain, MA
*Denotes co-first authors
11 12
Address Correspondence to:
13
Rebecca Hartman, MD, MPH
14
Department of Dermatology, Brigham and Women’s Hospital
15
221 Longwood Avenue
16
Boston, MA 02115
17
[email protected]
18 19
Conflicts of Interest Disclosures: None
20
Funding Sources: Dr. Hartman is supported by an American Skin Association Research Grant
21
(120795).
22 23
Word Count: 499
24
Table Count: 1
25
Figure Count: 1
26
Reference Count: 4
27 28
To the editor: Skin cancer incidence is rising in the United States (US) despite public health efforts
29
encouraging skin cancer prevention.1 We analyzed temporal trends of skin-cancer related
30
primary (concerned with disease prevention) and secondary (concerned with early disease
31
detection) preventive behaviors to examine for potential areas for improvement.
32
The National Health Information Survey (NHIS) was examined over a ten-year period
33
from 2005 to 2015.2 The NHIS conducts representative population-based annual interviewing of
34
the adult US population with response rates of 70.1% to 80.7% during the examined period.2
35
Our outcomes of interest were: use of sun protective measures (including sun avoidance,
36
protective clothing, and sunscreen use), lifetime history of full body skin examination (FBSE) by
37
a physician, and histories in the past year of indoor tanning and two or more sunburns.
38
Specifically, use of sun protective measures were defined as use always or most of the time
39
when outside for more than one hour on a warm sunny day. Protective clothing included at least
40
one of the following: long pants, hat, or long-sleeved shirt. Of note, a small percentage of
41
individuals answered the sun protection questions by stating that they “don’t go out in the sun.”2
42
These individuals and those with unknown or missing responses were excluded from analysis.2
43
Multivariable logistic regression analyses were used to assess the association between
44
time period and weighted prevalence of sun protective behaviors, adjusting for sex, region,
45
health insurance, alcohol use, smoking status, education, personal and family histories of skin
46
cancer, income, race, and skin reaction to the sun. P-values were adjusted for multiple
47
comparisons. Due to substantial missing data for income, NHIS imputed data was used.2
48
The weighted study included a total of 67,471 individuals. From 2005 to 2015, the
49
unadjusted and adjusted prevalence of most skin cancer preventive behaviors rose, including
50
sun avoidance, sunscreen use, and FBSE, with the exception of wearing protective clothing
51
(Table 1, Figure 1). While the adjusted prevalence of indoor tanning decreased, that of sunburn
52
increased (Table 1, Figure 1).
53
Indoor tanning is substantially decreasing, implying the success of targeted legislative
54
and public health efforts.3 Nevertheless, despite increases in most sun protective behaviors, the
55
rates of primary and secondary skin preventive behaviors remain suboptimal, and the
56
prevalence of multiple sunburns is rising. It is unclear if the rise in sunburns is due to heightened
57
sun awareness causing increased reporting or inadequate use of sun protection. Researchers in
58
Australia have examined temporal changes in sun protective behaviors, and found
59
improvements following the implementation of the SunSmart public health campaign, with
60
decreases in reported sunburns and tan preference, and increases in use of sun protective
61
behaviors, although these benefits may be plateauing more recently.4 Further research is
62
needed in the U.S. population to examine why the prevalence of multiple sunburns is rising and
63
the impact of increased adoption of sun protective behaviors on skin cancer incidence.
64
Additionally, the success of public health efforts against indoor tanning may provide guidance
65
for future public health efforts aimed at skin cancer prevention.
66
67
Acknowledgement
68
Author Contributions: Bolick, Huang, and Hartman had full access to all the data in the study
69
and take responsibility for the integrity of the data and the accuracy of the data analysis. Study
70
concept and design: Bolick, Hartman, and Mostaghimi. Acquisition, analysis, and interpretation
71
of Data: Bolick, Huang, and Hartman. Drafting of Manuscript and revising critically for important
72
intellectual content: Bolick, Huang, Hartman, and Mostaghimi.
73
References
74 75 76 77
1. U.S. Cancer Statistics Working Group. U.S. Cancer Statistics Data Visualizations Tool, based on November 2018 submission data (1999-2016): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; www.cdc.gov/cancer/dataviz, June 2019.
78 79
2. National Center for Health Statistics. Survey Description, National Health Interview Survey, 2015. Hyattsville, MD: National Center for Health Statistics; 2015.
80 81 82
3. Lazovich D. Effect of Parental Permission and Age Restriction Laws on US Adolescent Indoor Tanning Trends. Am J Public Health. 2018;108(7):851-853. doi:10.2105/AJPH.2018.304458
83 84
4. Dobbinson SJ, Wakefield MA, Jamsen KM, et al. Weekend Sun Protection and Sunburn in Australia. Am J Prev Med. 2008;34(2):94-101. doi:10.1016/j.amepre.2007.09.024
85
86 87 88 89
Figure Legend
90
predictive marginal means. Adjusting covariates include sex, census region, health insurance
91
coverage, alcohol use, smoking status, education level, personal history of skin cancer, family
92
history of skin cancer, income, race, and skin reaction to the sun.
Figure 1: Adjusted prevalences of sun protective behaviors over time. The prevalences are the
Table 1 Year, Unadjusted Prevalence, % (95% CI) 2005
2010
2015
Year, Adjusted Prevalence, % (95% CI) Ptrend value*
Adjuste d P2015vs.2
2005
2010
2015
31.7 (30.932.6) 35.9 (35.136.7) 31.5 (30.732.2) 19 (18.419.6) 14.1 (13.514.8) 18.2 (17.718.8)
35.5 (34.736.4) 38.4 (37.539.2) 33.1 (32.234) 22.4 (21.723.1) 6.2 (5.76.7) 21.1 (20.421.8)
36.8 (35.937.6) 37.2 (36.338.1) 34.3 (33.535.1) 22.4 (21.723) 4.1 (3.84.5) 19.9 (19.220.7)
Ptrend value*‡
Adjusted P2015vs.2005 value†
<0.001
<0.001
<0.001
0.098
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.001
005
value† Sun Avoidance Protective Clothing Sunscreen Use Full Body Skin Exam Indoor Tanning Sunburn
31.1 (30.232) 35.5 (34.736.3) 30.6 (29.831.3) 18.1 (17.418.7) 14.8 (14.115.5) 18.7 (18.119.3)
35.4 (34.536.3) 38.5 (37.539.4) 32.7 (31.733.6) 22 (21.222.8) 6.2 (5.76.7) 21.0 (20.221.7)
37.5 (36.538.4) 37.5 (36.638.4) 35.5 (34.736.4) 23.6 (22.924.4) 3.9 (3.54.3) 19.6 (18.820.4)
<0.001
<0.001
<0.001
0.006
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.198
Table 1: Unadjusted and adjusted prevalence of sun protective behaviors by survey year. Adjusting covariates include sex, census region, health insurance coverage, alcohol use, smoking status, education level, personal history of skin cancer, family history of skin cancer, income, race, and skin reaction to the sun.
*Wald tests based on the logistic regression were performed to test for any change in log odds over 2005, 2010 and 2015. †Pairwise Wald tests based on the logistic regression were performed to test for change in log odds in any two time points. P-values were adjusted for multiple comparisons using Benjamini & Hochberg (1995). ‡Median P rule was used to obtain the final adjusted p-values due to multiple imputation for income.