University of Southern California Dissertations and Theses

Personal hair dye use and risk of B-cell non-Hodgkin’s lymphomas among adult women in Los Angeles County

(USC Thesis Other)

Personal hair dye use and risk of B-cell non-Hodgkin’s lymphomas among adult women in Los Angeles County

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

PERSONAL HAIR DYE USE AND RISK OF B-CELL NON-HODGKIN’S
LYMPHOMAS AMONG ADULT WOMEN IN LOS ANGELES COUNTY

by

Xinxin Xu

———————————————————————————

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)

December 2010

Copyright 2010 Xinxin Xu

ii

ACKNOWLEDGEMENTS

This work was done under the direction and supervision of my guidance
committee chair, Dr. Leslie Bernstein. I would like to give my deepest gratitude and
appreciation to Dr. Leslie Bernstein for her invaluable guidance and persistent help.
I would like to extend my special thanks and gratitude to my committee
members: Dr. Wendy Mack and Dr. Stanley Azen for their thoughtful comments on
my thesis.
I would also like to thank my colleagues and friends at City of Hope: Jane
Sullivan-Halley, Huiyan Ma, and Yani Lu for their valuable advice and
encouragement. Finally, I would like to thank my family-my husband Dalin and my
daughter Michelle. Without their patient love and silent support, this dissertation
would not have been possible.
iii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ii
LIST OF TABLES iv
ABSTRACT v
CHAPTER I: INTRODUCTION 1
CHAPTER II: METHODS 4
Study population and non-Hodgkin’s lymphoma subtype classification 4
Data collection 6
Exposure assessments 7
Statistical analysis 8
CHAPTER III: RESULTS 10
Participant characteristics 10
Personal hair dye use and risk of B-cell NHL 12
Type, color, initial calendar year of personal hair dye use and risk of B-cell NHL 13
Personal hair dye use and risk of B-cell NHL by race/ethnicity 14
Personal hair dye use and subtypes of B-cell NHL 16
CHAPTER IV: DISCUSSION 18
REFERENCES 29

iv

LIST OF TABLES

Table 1: Participant characteristics 11
Table 2: Odds ratio (OR) (95% confidence interval, CI) of B-cell non-
Hodgkin lymphoma associated with hair dye use in Los Angeles
women 13

Table 3: Odds ratio (OR) (95% confidence interval, CI) of B-cell non-
Hodgkin lymphoma associated with different type/color/ initial
calendar year of hair dye use in Los Angeles women 14

Table 4: Adjusted odds ratio (OR) (95% confidence interval, CI) of B-
cell non-Hodgkin lymphoma associated with different
type/color/initial calendar year of hair dye use by race/ethnicity
in Los Angeles women 15

Table 5: Adjusted odds ratio (OR) (95% confidence interval, CI) of B-
cell non-Hodgkin lymphoma associated with different
type/color/ initial calendar year of hair dye use by three major
histologic subtypes in Los Angeles women 17

v

ABSTRACT

Study Aim: There have been inconsistent reports regarding personal hair dye use and B-
cell non-Hodgkin’s lymphomas (NHL) risk. This study aims to evaluate the relationship
between women’s personal hair dye use and the risk of B-cell NHL. Methods: A 1:1
matched population-based case-control study of women conducted in Los Angeles
County recruited 986 case-control pairs. Case participants were diagnosed with
histologic confirmed primary B-cell NHL between 2004 and 2008. Control participants
were individually matched to the index cases by age, gender, race, and neighborhood of
residence. Exposure information was obtained by in-person interviews. Multivariable
conditional logistic regression was performed to estimate odds ratios (ORs) and 95%
confidence intervals (CIs) of B-cell NHL associated with personal hair dye use overall,
by race, and by NHL subtypes, adjusting for potential confounders (alcohol status,
smoking status, and education levels). Results: Compared to women who had never
used any hair dye products, those who had ever used personal hair dyes had a 38%
greater risk for B-cell NHL (OR=1.38, 95%CI=1.05-1.31). A younger age at first hair
dye use, a longer duration of hair dye use, and a greater number of hair dye applications
were all associated with an elevated risk of B-cell NHL (p-values for trend=0.007, 0.006,
0.025, respectively). Relative risk estimates were higher for those who had ever used
permanent dyes (OR=1.46, 95%CI=1.11-1.93), black hair dyes (OR=1.76, 95%CI=1.18-
2.63), and also higher for those who initiated hair dye use prior to 1980 (OR=1.45,
95%CI=1.08-1.96). The influence patterns were similar across different racial/ethnic
groups. Increased risk of follicular lymphoma and chronic lymphocytic leukemia/small
vi

lymphocytic lymphoma (CLL/SLL) were observed in this study. Conclusion: The results
from this study support the hypothesis that hair dye use, and particularly using permanent
hair dyes, using black hair dyes, and year of initiation of use prior to 1980 increase the
risk of B-cell NHL, especially the follicular lymphoma and CLL/SLL subtypes. Further
research is warranted to address the effects of specific types, colors, or components of
hair dye products, as well as the underlying biological mechanisms which mediate the
increase in B-cell NHL risk.

1

CHAPTER I
INTRODUCTION

Non-Hodgkin’s lymphoma (NHL) is the sixth most diagnosed cancers in the U.S.
According to the American Cancer Society (ACS), an estimated 65,540 new cases will
occur (male: 35,380; female:30,160) in 2010 (Jemal, Siegel, Xu, & Ward, 2010). The
NHL incidence rate is higher in males than in females, Whites have the highest incidence
while the lowest incidence is observed in American-Indians/Alaska Natives (Altekruse et
al., 2009). Since 1990, a slight but steady increase in the incidence of NHL (rising by
about 1.1% each year) has been observed among women in the U.S., but not men (ACS,
2010). HIV infection, immunosuppression and certain viral infection-related cases of
NHL account for only a small fraction of this increase in incidence and other risk factors
are still largely unknown (La Vecchia & Tavani, 1995; Baris & Zahm, 2000; Fisher &
Fisher, 2004).
NHL is comprised of a number of lymphoid malignancies including B-cell and T-
cell lymphomas(Fisher & Fisher, 2004). Among B-cell lymphomas, which account for
the majority of NHL, diffuse large B-cell lymphomas (DLBCL) (28%), chronic
lymphocytic leukemia/small lymphocytic leukemia (CLL/SLL) (20%), and follicular
lymphoma (13%) are the three major subtypes(Morton et al., 2006). Given the different
incidence patterns of NHL histologic subtypes, their etiologies may vary (Morton, et al.,
2006).
Personal hair dyes are widely used among women. In the U.S. more than 70% of
the female population uses hair dye products (Zhang et al., 2008). Hair dyes contain
2

chemicals or ingredients some of which may be carcinogenic and/or mutagenic in
animals (Corbett, 1999; Nohynek, Antignac, Re, & Toutain, 2010). Epidemiological
studies conducted to investigate the association between personal hair dye use and NHL
risk have shown inconsistent results (Zahm et al., 1992; Grodstein, Hennekens, Colditz,
Hunter, & Stampfer, 1994; Holly, Lele, & Bracci, 1998; Miligi et al., 2005; Tavani et al.,
2005; de Sanjose et al., 2006; Mendelsohn et al., 2009; Wong, Harris, Wang, & Fu, 2010).
In general, depending on how long they will last on hair, hair coloring products
are classified as permanent, semi-permanent, temporary and gradual dyes (Bolduc &
Shapiro, 2001). Permanent or oxidation dyes, which account for the majority of the hair-
coloring market, remain in the hair shaft despite shampooing. Other types of hair dyes
remain for a more limited time or do not penetrate the hair shaft. It has been suggested
black hair dyes contain higher concentrations of chemicals than other color dyes (Zahm
et al., 1992; Bailey & Skare, 2008; Zahm et al., 2008). This has raised concerns that
permanent and black hair dyes may increase the risk of NHL, yet evidence is
inconclusive (Zahm et al., 1992; Grodstein, Hennekens, Colditz, Hunter, & Stampfer,
1994; Altekruse, Henley, & Thun, 1999; Zhang et al., 2004; Miligi et al., 2005; de
Sanjose et al., 2006; Morton et al., 2007). Because of “clear evidence” or “some evidence”
for carcinogenicity in animals, some ingredients of coloring products have been
discontinued in dyes that are sold since the early 1980’s (Nohynek, et al., 2010). Whether
new dyes are safe and whether higher risk associated with discontinued dyes may persist
in affecting NHL risk are important to determine. Therefore it is necessary to examine
further the influence of permanent, black hair dyes, and initial calendar year of use on the
association between hair dyes and NHL risk.
3

Since the etiology of NHL can differ across histologic subtypes, the association of
NHL with personal hair dye use may differ by subtype. Unfortunately, few studies have
investigated the association by NHL subtype (Miligi et al., 2005; Morton et al., 2007;
Zhang, et al., 2008). In addition, no previous studies have explored the association
between hair dyes and NHL risk in different race/ethnic groups, although NHL incidence
differs substantially by race (Morton, et al., 2006).
Therefore, to comprehensively understand the risk of NHL associated with
different characteristics of personal hair dye use, we examined the relationship between
personal hair dye use and B-cell NHL risk overall, by race, and by common histologic
subtypes using data collected in a population-based case-control study of women
conducted in Los Angeles County .

4

CHAPTER II
METHODS

Study population and non-Hodgkin ’s lymphoma subtype classification
Patients with newly diagnosed primary B-cell NHL between May 1, 2004 and
March 31, 2008 in Los Angeles County were identified through a rapid case-
ascertainment (RCA) system by the Cancer Surveillance Program (CSP), a population-
based cancer registry for Los Angeles County(USC, 2002). Case participants were
restricted to English speaking women aged between 20 and 79 years old and resided in
Los Angeles County on their dates of diagnosis. Women with a previous diagnosis of
NHL, Hodgkin’s disease, leukemia, multiple myeloma, acquired immune deficiency
syndrome (AIDS), or human immunodeficiency virus (HIV) infection were excluded
from this study. Of 1848 eligible patients with B-cell NHL, 1013 patients (54.8%)
completed in-person interviews, 284 patients refused to participant, 251 patients were
deceased, 137 patients were not interviewed within 18 months, 93 patients moved from
Los Angeles County or were not able to be located, 59 patients were too ill or mentally
disabled for interview, 10 patients were not contacted without their physicians’
permission, and 1 patient partly completed the interview.
Case participants and population-based female control participants were
individually matched on age (within five year age groups), race/ethnicity (Hispanic white,
non-Hispanic white, African American, Asian/other), and socioeconomic status of the
neighborhood of residence. Women who were previously diagnosed with NHL, who
were non-English-speaking or HIV positive were excluded as control participants. We
5

used the neighborhood walk procedures to screen control participants by walking the
neighborhood using a specific algorithm based on the residence of a case participant in
that neighborhood. All interviewed cases of a specific socioeconomic status were placed
in a pool and their neighborhoods were walked in the specific walking sequence until a
potentially eligible control participant was identified who matched one of the cases on all
of the other criteria. Our representatives came to the house to ascertain whether this
eligible control participant lived in the residence. If no-one responded, we left an
explanatory letter with postage-paid return envelope. We then followed up this woman by
additional letters and telephone. If a potential eligible control refused to participate, we
identified the next eligible woman in the walk sequence until she was interviewed. Using
this procedure, of the 1250 potentially eligible control participants, 1038 completed the
interview (83.0%), 166 women refused to participate, 35 women moved from Los
Angeles County or were not able to be located, 9 women were too ill for interview, 1
women were deceased, 1 women were not interviewed within 18 months.
We excluded 8 unmatched case participants from this analysis as well as 33 extra
controls (either we found another eligible control earlier in the walk sequence or deemed
the matched case ineligible based on later diagnosis information), and 15 case-control
pairs (who were mismatched on race). We also excluded 4 case-control pairs from our
analysis because either the case or the corresponding control had unknown history of hair
dye use, or all of the information regarding hair coloring was missing, leaving 986 case
participants and 986 matched control participants in the final analysis.
Subtypes of B-cell NHL were classified based on the World Health Organization
(WHO) histological classification (2008 update version). In our study, NHL case
6

participants were classified into four subtypes: Diffuse large B-cell lymphoma (DLBCL)
(International Classification of Diseases for Oncology codes M-9678, M-9679, M-9680,
M-9684), follicular lymphomas (M-9690,M-9691,M-9695,M-9698), chronic lymphocytic
leukemia/small lymphocytic lymphoma (CLL/SLL)( M-9670,M-9823), and a combined
group of the remaining subtypes (M-9590, M-9591, M-9671, M-9673, M-9833, M-9835
and M-9836). Pathology reports were reviewed by an expert to confirm or reclassify
subtypes and to exclude any ineligible case participants.
Data collection
Overall, case participants were recruited within 30-35 days of diagnosis via the
RCA procedures. We first sent a letter to the primary physician of the patient to
determine if there were contraindications to contacting the patient. We then sent a letter
(including a form to be returned) to those who could be contacted for our study. We
contacted patients by telephone if they did not respond to our letter within two weeks. We
conducted an in-person interview with eligible case participants who agreed to participate
at their homes. Control participants were interviewed using similar procedure as case
participants.
All participants completed a structured, standardized questionnaire to obtain
information on demographic information, history of hair dye use and other life style risk
factors (smoking, alcohol) before the reference date (diagnosis date for case participants,
and the same year and month for individual matched control participants). In particular,
the participants were asked whether they had ever used any type of hair dye products. If
the answer was yes, for each type of hair dye, we obtained the color of hair dyes, age at
first and last use, the frequency (times of hair dye use per week, month or year) and
7

duration of use (total years used). During the interview, we provided all participants with
a list of hair dye products to help them recall lifetime use of hair dye products.
Exposure assessments
We considered a woman as a hair dye user (i.e., “ever user”) if she had ever used
any hair dye product, either at home or in a beauty salon on her hair or eyebrows before
the reference date and as a “non-user” if she had never used any hair dye products. A
“permanent” hair dye product was defined as a product that dyes one’s hair permanently
by penetrating the hair shaft and all other types of hair dyes were defined as “non-
permanent” hair dyes. The color of the hair dye product was defined as “black” or “non-
black” for each type of hair dye used.
The cumulative amount of exposure to each hair dye type and color was
calculated by multiplying the number of times used per year by the total years the
participant had used that type and color of hair dye. The cumulative lifetime exposure to
all hair dyes was estimated by adding up all the cumulative amounts. Consistent with
previous studies, initial calendar year of hair dye use was categorized by two groups:
before 1980, and in or after 1980.
Socioeconomic status (SES) was categorized into lower SES, middle SES, higher
SES, based on the classification of address into census tracts, which were designed to be
homogeneous in terms of population characteristics, economic status, and living
conditions; and were weighted according to the median household income and average
level of education among adult residents 25 years or older.
Smoking history was grouped into never, former or current smoking. Participants
who had smoked more than 100 cigarettes over their lifetimes and smoked at least once a
8

month for six months or longer were “ever-smokers”. All other participants were defined
as “non-smokers”. Among ever smokers, those who continued smoking until the
reference date or quit smoking less than 12 months prior to the reference date were
defined as “current smokers”, while those who quit smoking more than one year prior to
that date were defined as “former smokers”.
Alcohol history was also categorized into three groups. Participants who never
drank, or drank alcoholic beverages such as beer, wine, or hard liquor no more than once
a month and for fewer than 6 months were “non-drinkers.” Those who drank more than 1
drink of alcoholic beverage such as beer (12oz.), wine (4 oz.), or hard liquor (1.5oz.) per
month for at least 6 months were defined as “ever drinkers”. And ever drinkers were
grouped into “former drinkers” and “current drinkers” based on whether they had stopped
drinking more than 12 months before the reference date.
Statistical Analysis
We compared participant characteristics between cases and controls using Chi-
square tests for differences in frequency distributions and paired t-test for differences in
means.
The relative risk estimates for the development of B-cell NHL were calculated as
odds ratios with 95% confidence intervals using multivariable conditional logistic
regression analysis. We categorized continuous variables (age at first use, lifetime
duration and total number of applications) into three groups based on the tertile
distribution of each variable among controls. We then combined middle (26-40 years)
and younger (≤25 years) groups of age at first hair dye use since they had similar
9

estimates. Likelihood ratio chi-square tests were employed to evaluate the dose-response
effects (linear trends) using ordinal variables.
Our multivariable models were adjusted for the following factors, selected a
priori, as potential confounders: alcohol (never, former, current), smoking (never, former,
current), and education (high school or lower, college or higher). We also performed
stratified analysis to examine whether associations of B-cell NHL with hair dye use
varied by race/ethnicity (Hispanic white, non-Hispanic white, African American,
Asian/other), or by the three major B-cell NHL subtypes (DLBCL, follicular lymphoma
and CLL/SLL).
In reporting the results of trend tests, we considered a two-sided p value less than
0.05 as statistically significant. All analyses were performed using the SAS statistical
package (Version 9.2, SAS Institute, Cary, NC).
10

CHAPTER III
RESULTS

Participant Characteristics
Table 1 shows the selected characteristics of case participants and control
participants. The mean age at diagnosis was 59.2 years (standard deviation, SD=13.9) for
B-cell NHL case participants, and the mean age at reference date was 58.7 years
(SD=13.8) for control participants. Case participants with histologic subtype of CLL/SLL
were older than those with other subtypes (mean age=64.5, SD=10.6). Among both case
and controls participants, 63.8% were non-Hispanic white; 18.0% were Hispanic white;
10.0% were African-American; and 8.2% were Asian women or women with other
ethnicities. 64.8% of the participants were at high level of socioeconomic status.
Compared with control participants, case participants were more likely to have a lower
level of education (p<0.001), to be former or current smokers (p=0.03), to have never
consumed alcohol (p=0.02), and to have ever used hair dyes (p=0.01). Hair dye users in
cases were more likely to initiate the use at younger ages (p=0.03), to have a longer
duration of use (p=0.01) and a greater number (p=0.03) of hair dye use episodes,
compared with control participants.

11

Table 1. Participant characteristics
Cases Controls p-value
a

Number of participants 986 986

Mean age at reference date (SD), years 59.2 (13.9) 58.7 (13.8) -
b

Diffuse large B-cell lymphoma 57.4 (15.0) 56.8 (14.7)
Follicular lymphoma 58.8 (12.2) 58.5 (12.2)
Chronic lymphocytic leukemia/Small lymphocytic
lymphoma
64.5 (10.6) 64.0 (10.6)
Other subtypes 57.5 (15.1) 56.9 (15.0)

Race (%)
non-Hispanic white 629 (63.8) 629 (63.8)
Hispanic white 177 (18.0) 177 (18.0) -
b

African-American 99 (10.0) 99 (10.0)
Asian/Others 81 (8.2) 81 (8.2)

Socioeconomic status (%)
Higher 625(63.4) 653 (66.2) -
b

Middle 198 (20.1) 176 (17.9)
Lower 163 (16.5) 157 (15.9)

Education (%) <0.001
≤High school 347 (35.2) 278 (28.2)
≥College 639 (64.8) 708 (71.8)

Smoking status in year prior to reference date (%) 0.03
Never 574 (58.2) 615 (62.4)
Former 307 (31.1) 290 (29.4)
Current 105 (10.7) 81 (8.2)

Alcohol drinking status in year prior to reference date
(%)

0.02
Never 475 (48.2) 443 (44.9)
Former 188 (19.1) 165 (16.7)
Current 323 (32.8) 378 (38.3)

Ever use of any hair dye (%) 879 (89.2) 842 (85.4) 0.01

Among hair dye users
Age at first use (SD), years
d
33.2 (14.2) 34.3 (14.7) 0.03
c

Duration of use (SD), years
e
19.7 (14.8) 18.6 (13.9) 0.01
c

Total number of applications (SD), times
f
182.2(235.9) 165.0 (218.0) 0.03
c

a
p ascertained from Chi-square test, except where otherwise noted.
b
tests were not performed due to matching.
c
p ascertained from paired t-test.
d
7 cases with unknown/missing age started and corresponding controls were excluded from the analysis.
e
12 cases and 2 controls with unknown/missing data and corresponding pairs were excluded from the
analysis.
f
10 cases and 3 controls with unknown/missing data and corresponding pairs were excluded from the
analysis.
12

Personal hair dye use and risk of B-cell NHL
After adjustment for education, smoking status, and alcohol consumption status,
women who reported having ever used personal hair dyes had a 38% (adjusted OR=1.38,
95%CI=1.05-1.31) greater risk for B-cell NHL than those who had never used any hair
dye products (Table 2).
Age at first hair dye use was negatively associated with the risk of B-cell NHL
(adjusted p for trend=0.007, Table 2). Women whose age at first hair dye use was 40
years or younger had a 42% (adjusted OR=1.42, 95%CI=1.07-1.88) greater risk of B-cell
NHL compared with those who had never used any hair dye products.
The duration of hair dye use was positively associated with the risk of B-cell NHL
(adjusted p for trend=0.006, Table 2). Women who reported at least 25 years of hair dye
use had a 53% (adjusted OR=1.53, 95%CI=1.12-2.08) greater risk than women who had
never used any hair dye products.
The total number of hair dye applications was also positively associated with the
risk of B-cell NHL (adjusted p for trend=0.025, Table 2). Women who reported the
largest numbers of uses (≥177) had a 48% (adjusted OR=1.48, 95%CI=1.09-2.00) greater
risk than never users.

13

Table 2. Odds ratio (OR) (95% confidence interval, CI) of B-cell non-Hodgkin lymphoma associated
with hair dye use in Los Angeles women

Cases
(986)
Controls
(986)
Crude OR
(95% CI)
Adjusted OR
a

(95% CI)
Ever use of any hair dye
Never use 107 144 1.00 1.00
Ever use 879 842 1.40 (1.07,1.83) 1.38 (1.05,1.81)

Age at first use
b
,years
Never use 107 143 1.00 1.00
≥41 265 274 1.27 (0.94,1.73) 1.28 (0.94,1.75)
≤40 607 562 1.45 (1.10,1.92) 1.42 (1.07,1.88)
p for trend 0.003 0.007

Duration of use
c
,years
Never use 107 142 1.00 1.00
≤10 306 302 1.33 (0.99,1.79) 1.32 (0.98,1.78)
11-24 257 264 1.29 (0.95,1.76) 1.27 (0.92,1.73)
≥25 302 264 1.56 (1.15,2.11) 1.53 (1.12,2.08)
p for trend 0.003 0.006

Total number of
applications
d
, times

Never use 107 143 1.00 1.00
≤49 291 278 1.39 (1.03,1.87) 1.37 (1.02,1.86)
50-176 267 280 1.27 (0.94,1.73) 1.26 (0.92,1.71)
≥177 308 272 1.51 (1.12,2.04) 1.48 (1.09,2.00)
p for trend 0.017 0.025
a
Adjusted for education (high school or lower, college or higher), smoking status (never, former, current),
and alcohol status (never, former, current).
b
7 cases with unknown/missing age started and corresponding controls were excluded from the analysis.
c
12 cases and 2 controls with unknown/missing data and corresponding pairs were excluded from the
analysis.
d
10 cases and 3 controls with unknown/missing data and corresponding pairs were excluded from the
analysis.

Type, color, initial calendar year of personal hair dye use and risk of B-cell NHL
The risk of B-cell NHL varied with different type, color, and initial calendar year
of hair dye use (Table 3). Compared with women who had never used any hair dye
products, the higher risk of B-cell NHL was observed among those who had ever used
permanent hair dyes (adjusted OR=1.46, 95%CI=1.11-1.93), who had ever used black
hair dyes (adjusted OR=1.76, 95%CI=1.18-2.63), and who initiated hair dye use prior to
1980 (adjusted OR=1.45, 95% CI=1.08-1.96).
14

Table 3. Odds ratio (OR) (95% confidence interval, CI) of B-cell non-Hodgkin lymphoma associated
with different type/color/ initial calendar year of hair dye use in Los Angeles women
Cases
(986)
Controls
(986)
Crude OR
(95% CI)
Adjusted OR
a

(95% CI)
Type of hair dye
Permanent dye
Never use 107 144 1.00 1.00
Only use non-permanent 278 303 1.24 (0.92,1.67) 1.24 (0.91,1.67)
Ever use permanent 601 539 1.49 (1.13,1.96) 1.46 (1.11,1.93)

Color of hair dye
Black hair dye
Never use 107 144 1.00 1.00
Only use non-black 779 767 1.36 (1.04,1.78) 1.34 (1.02,1.77)
Ever use black 100 75 1.81 (1.22,2.69) 1.76 (1.18,2.63)

Initial calendar year of hair
dye use
b

Never use 107 143 1.00 1.00
≥1980 474 474 1.32 (0.99,1.75) 1.31 (0.98,1.74)
<1980 398 362 1.49 (1.11,2.00) 1.45 (1.08,1.96)
a
Adjusted for education (high school or lower, college or higher), smoking status (never, former, current),
and alcohol status (never, former, current).
b
7 cases with unknown/missing data and corresponding controls were excluded from the analysis.

Personal hair dye use and risk of B-cell NHL by race/ethnicity
Overall, ever use of hair dyes was associated with an increased risk of B-cell
NHL among all racial/ethnic groups although the 95% CIs included 1 for all groups
except for non-Hispanic white women (OR=1.46, 95% CI=1.03-2.09, Table 4); non-
Hispanic white women accounted for 63.8% of participants. We also found that the
associations between different type, color, and initial calendar year of hair dye use and
NHL were similar across different racial/ethnic groups (Table 4). Few women in the
Asian/other group used black hair dyes (case participants, n=7; control participants, n=9)
and no association with B-cell NHL risk among those women was observed (adjusted
OR=0.98, 95% CI =0.27, 3.56).

15
Table 4. Adjusted odds ratio (OR) (95% confidence interval, CI) of B-cell non-Hodgkin lymphoma associated with different type/color/initial calendar
year of hair dye use by race/ethnicity in Los Angeles women
Hispanic white Non-Hispanic white African-American Asian/other
Cases/Controls
(177/177)
OR
a

(95% CI)
Cases/Controls
(629/629)
OR
a

(95% CI)
Cases/Controls
(99/99)
OR
a

(95% CI)
Cases/Controls
(81/81)
OR
a

(95% CI)
Ever use of any
hair dye

Never 18/22 1.00 60/86 1.00 17/21 1.00 12/15 1.00
Ever 159/155 1.26(0.66,2.41) 569/543 1.46(1.03,2.09) 82/78 1.33(0.62,2.87) 69/66 1.29(0.56,2.95)

Type of hair dye
Permanent dye
Never 18/22 1.00 60/86 1.00 17/21 1.00 12/15 1.00
Only use
non-
permanent
31/33 1.16(0.53,2.52) 186/210 1.28(0.86,1.89) 37/39 1.21(0.52,2.81) 24/21 1.39(0.54,3.62)
Ever use
permanent
128/122 1.29(0.66,2.51) 383/333 1.56(1.08,2.25) 45/39 1.44(0.64,3.26) 45/45 1.23(0.51,2.96)

Color of hair dye
Black hair dye
Never 18/22 1.00 60/86 1.00 17/21 1.00 12/15 1.00
Only use
non-black
132/135 1.21(0.63,2.34) 536/515 1.46(1.02,2.09) 49/60 0.99(0.44,2.23) 62/57 1.33(0.57,3.06)
Ever use
black
27/20 1.72(0.72,4.12) 33/28 1.57(0.85,2.91) 33/18 2.23(0.89,5.59) 7/9 0.98(0.27,3.56)

Initial calendar year
of hair dye use
b

Never 18/22 1.00 60/86 1.00 17/20 1.00 12/15 1.00
≥1980 98/105 1.16(0.59,2.30) 271/264 1.44(0.98,2.11) 44/42 1.30(0.57,2.99) 61/63 1.30(0.57,2.97)
<1980 60/49 1.45(0.69,3.03) 293/274 1.50(1.03,2.18) 37/36 1.21(0.51,2.91) 8/3 3.37(0.70,16.30)
a
Adjusted for education (high school or lower, college or higher), smoking status (never, former, current), and alcohol status (never, former, current).
b
Unknown/missing data and corresponding pairs were excluded from the analysis. Hispanic white: 1 case; African-American: 1 case; Non- Hispanic white: 5
cases.

16

Personal hair dye use and subtypes of B-cell NHL
Stronger adverse associations for follicular lymphoma and CLL/SLL were
consistently observed in the analysis overall, by type of hair dyes, by color of hair dyes,
and by initial calendar year of use of hair dyes, whereas the association with DLBCL was
weak (Table 5). The highest risk observed was for CLL/SLL which was strongly
associated with ever having used black hair dyes (adjusted OR=4.61, 95%CI=1.58-13.42).

17
Table 5. Adjusted odds ratio (OR) (95% confidence interval, CI) of B-cell non-Hodgkin lymphoma associated with different type/color/ initial calendar
year of hair dye use by three major histologic subtypes in Los Angeles women
DLBCL
a
Follicular lymphoma CLL/SLL
a

Cases
(277)
Controls
(277)
OR
b

(95% CI)
Cases
(214)
Controls
(214)
OR
b

(95% CI)
Cases
(210)
Controls
(210)
OR
b

(95% CI)
Ever use of any hair dye
Never 36 42 1.00 24 35 1.00 20 31 1.00
Ever 241 235 1.12(0.67,1.85) 190 179 1.51(0.84,2.72) 190 179 1.62(0.89,2.94)

Type of hair dye
Permanent dye
Never use 36 42 1.00 24 35 1.00 20 31 1.00
Only use non-permanent 88 88 1.10(0.63,1.89) 49 56 1.21(0.61,2.40) 62 72 1.34(0.69,2.60)
Ever use permanent 153 147 1.13(0.67,1.92) 141 123 1.63(0.90,2.97) 128 107 1.75(0.95,3.22)

Color of hair dye
Black hair dye
Never use 36 42 1.00 24 35 1.00 20 31 1.00
Only use non-black 215 208 1.13(0.68,1.88) 177 163 1.53(0.85,2.75) 167 170 1.50(0.82,2.74)
Ever use black 26 27 0.98(0.47,2.06) 13 16 1.24(0.50,3.12) 23 9 4.61(1.58,13.42)

Initial calendar year of hair
dye use
c

Never use 36 42 1.00 24 34 1.00 20 31 1.00
≥1980 132 140 1.02(0.60,1.75) 103 95 1.56(0.83,2.94) 85 90 1.37(0.72,2.60)
<1980 107 93 1.27(0.72,2.24) 84 82 1.33(0.70,2.51) 105 89 1.92(1.01,3.66)
a
DLBCL: diffuse large B-cell lymphomas, CLL/SLL: chronic lymphocytic leukemia/small lymphocytic leukemia
b
Adjusted for education (high school or lower, college or higher), smoking status (never, former, current), and alcohol status (never, former, current).

c
Unknown/missing data and corresponding pairs were excluded from the analysis. DLBCL: 2 cases; Follicular: 3 cases.

18

CHAPTER IV
DISCUSSION

In this large, population-based, individually matched case-control study, increased
risk of B-cell NHL was observed among women who reported having ever used any hair
dye products, especially among those who started use at a younger age, had longer
duration of use, or had a greater number of hair dye applications. Permanent hair dyes,
black hair dyes, and using hair dye products prior to 1980 had strong adverse effects on
B-cell NHL, particular on follicular lymphoma and CLL/SLL.
Previous experimental studies suggested hair dye products contain chemicals or
ingredients that may be carcinogenic and/or mutagenic (Corbett, 1999; Nohynek, et al.,
2010). As a result, many ingredients (eg.2,4-diaminoanisole, DA; 2,4-toluenediamine,
TDA) of coloring products classified as “carcinogenic” or “probably carcinogenic” are no
longer used in coloring products (Nohynek, et al., 2010). However, due to inadequate
evidence on systemic toxicity or carcinogenicity in previous studies, some carcinogenic
aromatic amines are still found in current products. For example, 4-aminobiphenyl (4-
ABP) is a potential urinary bladder carcinogen, but is still used in hair dyes; little is
known about the association between this chemical and NHL risk (Zahm et al., 1992;
Turesky et al., 2003; Miligi, et al., 2005; Bailey & Skare, 2008).
The association between personal hair dye use and NHL risk has been explored in
several studies. A case-control study in eastern Nebraska reported a 50% increased NHL
risk among women who ever used hair dyes (OR=1.5, 95%CI=1.1-2.2) (Zahm et al.,
1992). Another case-control study in Europe found a slight increased risk of lymphoid
19

neoplasms for female hair dye users (OR=1.24, 95%CI=1.01-1.53) (de Sanjose et al.,
2006). Consistent with those studies, we observed a moderately increased risk of B-cell
NHL for women who were personal hair dye users (adjusted OR=1.38, 95%CI=1.05-
1.81). Furthermore, clear dose-response relationships between the risk of B-cell NHL and
earlier age at first use of hair dye, longer duration of hair dye use, and greater total
number of hair dye applications were observed in this study (adjusted p for trend: 0.013,
0.006, 0.025 respectively), whereas previous studies fail to detect such risk pattern with
increasing exposure levels. Providing additional evidence on the increased risk of NHL in
hair dye users, these dose-response relationships indicate that the findings in our study
are probably not due to chance. Our finding for ever use of hair dye products is also
consistent with a recent meta-analysis, which indicated a 23% increased risk of NHL in
hair dye users (Takkouche, Etminan, & Montes-Martinez, 2005).
With the individually matched case-control design and relative large sample size,
one advantage of our study is greater statistical power to detect moderate risks and more
precise confidence intervals on relative risk estimating. Several epidemiological studies
did not find an elevated risk of NHL among hair dyes users (Holly, Lele, & Bracci, 1998;
Miligi, et al., 2005; Tavani et al., 2005; Mendelsohn et al., 2009; Wong, et al., 2010).
One case-control study conducted in the San Francisco Bay area (female cases:328,
female controls:615) found no evidence for association between hair dyes and NHL risk
in women (OR=1.0,95%CI=0.77,1.3) (Holly, Lele, & Bracci, 1998). A hospital-based
case-control study in northern Italy (female cases: 190, female controls: 504) also
suggested no evidence of positive association between NHL risk and ever use of hair
dyes (OR=0.95, 95%CI=0.55-1.65), or permanent dyes (OR=1.14, 95%CI=0.65-1.99)
20

(Tavani et al., 2005). Similarly, no associations between hair dye use in women and risk
of NHL were observed in another Italian study (female cases: 715, female controls: 861)
(OR=1.0, 95%CI=0.8-1.2)(Miligi, et al., 2005). With the limited sample sizes in these
unmatched case-control studies, the power to detect the association between hair dye and
B-cell NHL can be poor. Nevertheless, a recent hospital-based case-control study with
individually matched controls conducted in China (Wong, et al., 2010) failed to identify
an association between hair dyes and NHL risk (OR=0.93, 95%CI=0.75-1.16). However,
this result was based only on 257 female cases and 514 female controls.
Several cohort studies also failed to confirm the association between personal hair
dye use and NHL risk (Thun et al., 1994; Mendelsohn et al., 2009). For example, a recent
cohort study in China (total 70,366 women, 99 hematopoietic cancer, 51 NHL cases)
suggests no statistically significant association between hair dyes and overall
hematopoietic cancer risk(RR=0.89, 95%CI=0.59-1.35), or NHL risk (RR=1.09,
95%CI=0.61-1.92), and no dose-response relationship between duration and NHL risk
was observed (Mendelsohn et al., 2009). This could be due to the small number of NHL
cases involved in the studies, resulting in low statistical power.
Permanent hair dyes account for more than 75% of the hair coloring market
(Burnett et al., 1975); thus it raises great public health significance to investigate the
potential health effects of personal permanent hair dyes. We found a statistically
significantly moderately increased risk of B-cell NHL among women who had ever used
permanent hair dyes irrespective of the colors of hair dye products, compared to women
who had never used hair dyes.
21

The association between permanent hair dyes and NHL risk has been investigated
in many studies. A pooled study (based on four case-control studies) by Zhang et al.
suggest that having use of permanent dyes was associated with a small increased risk of
NHL in women (OR=1.3, 95%CI=1.1-1.5) (Zhang, et al., 2008). Another study by Zahm
et al. reported a 70% increase in NHL risk associated with ever use of permanent dyes
among women (OR=1.7, 95%CI=1.1-2.8) (Zahm, et al., 1992). Results from our study
are consistent with these findings.
In contrast, some studies have reported no evidence of an association between
permanent hair dye use and NHL (Miligi et al., 1999). A population-based case-control
study by Miligi et al. did not find an association between permanent dyes and NHL and
CLL combined (OR=1.1, 95%CI=0.9-1.4) (Miligi, et al., 1999). Their analytic approach
may be flawed, as these authors combined never use with only use of non-permanent
dyes as a single reference group, and it is known that non-permanent hair dye can also
contain carcinogenic ingredients. In a case-control study conducted in 2004, Zhang et al.
(Zhang et al., 2004) did not find a statistically significant association between permanent
dyes and NHL (OR=1.2, 95%CI=0.9-1.5), although their effect estimate indicated a
positive association. Likewise, no association was observed in the prospective Nurses'
Health Study cohort in whom 99,067 women were followed for up to 14 years (1976-
1990) (Grodstein, Hennekens, Colditz, Hunter, & Stampfer, 1994). In this study, 244
women were newly diagnosed with hematopoietic cancer, including 144 women with
NHL; the association between permanent hair dyes and risk of all hematopoietic cancers
combined was null (RR = 0.9, 95% CI = 0.7-1.2) as was that for NHL specifically (RR=
1.1, 95% CI = 0.8-1.6). An important limitation of this study, as the authors noted, was
22

that exposure misclassification might occur since information on hair dye use was not
updated after 1982, which could have biased the estimates towards the null.
Different colors contain different concentrations of coloring chemicals. Compared
to other color dyes, black hair dye has the greatest p-phenylenediamine (PPD)
concentrations (Zahm, et al., 1992; Bailey & Skare, 2008; Zahm et al., 2008). Although
PPD has not been confirmed as a carcinogen by the International Agency for Research on
Cancer (IARC Working Group, in 1993) (Rollison, Helzlsouer, & Pinney, 2006), some
epidemiological studies have reported positive associations between permanent, black
hair dyes and cancer, including NHL. For example, after 7 years of follow-up the
American Cancer Society prospective cohort study (CPS-II) reported a fourfold increased
risk of death from NHL among women who used black hair dyes for at least 20 years.
(RR=4.37, 95%CI=1.3-15.2) (Thun et al., 1994). In the same cohort, extending to 12
years of follow-up, a twofold increased risk of NHL mortality was observed among
women who had used black or brown permanent dyes for 10-19 years (RR=2.5,
95%CI=1.1-5.8 for black; RR=1.4, 95%CI=1.0-1.9 for brown), although no statistically
significant trend was found with longer duration of use ( ≥20 years) (Altekruse, Henley,
& Thun, 1999). A case-control study by Zahm et al. also reported a fourfold increase of
NHL among women who ever used permanent, black hair dyes (OR=4.1, 95%CI=0.9-
18.8) (Zahm, et al., 1992). On the other hand, Holly et al found a reduced risk among
women who ever used dark color products (medium/dark brown/black) with increasing
frequency of use per year (Holly, et al., 1998). However as the conclusion is based on
only two exposed cases, this effect estimate in the reverse direction may be unreliable.
23

In our study, we analyzed black and non-black hair dyes separately, and our
findings are generally consistent with the hypothesis that black hair dyes confer greater
risk of B-cell NHL than non-black hair dyes. However, we believe the cumulative effects
of non-black hair dyes on NHL risk should not be ignored. In addition, as noted in
Mendelsohn et al, most studies conducted among Whites lacked sufficient power to
assess long-term use of black hair dyes because less than 1% of subjects used black hair
dyes (Mendelsohn, et al., 2009).
It should be noted that the ingredients of hair dye products have changed over
time. Between 1978 and 1982, the hair coloring industry reformulated all permanent
coloring products to eliminate some chemicals identified in bioassays as potentially
carcinogenic (Miligi, et al., 2005; de Sanjose, et al., 2006 ). Thus it is of importance to
investigate whether the risk of hair dye use before 1980 differs from that after 1980.
Several studies have yielded positive results for hair dye use among women who started
use prior to 1980. Zhang et al found an increased risk among women who began using
hair dyes before 1980 (OR=1.3, 95%CI=1.0-1.8); in particular for permanent, dark color
dyes in long-term users (>25 years, OR=2.1, 95%CI=1.0-4.0), and among women with
more episodes of use (>200, OR=1.7, 95%CI=1.0-2.8) (Zhang, et al., 2004). Similarly, a
study in Europe found significant increased risks of lymphoma among users who began
using hair dyes before 1980(OR=1.37, 95%CI=1.09-1.72), or who only used hair dyes
before 1980 (OR=1.62, 95%CI=1.10-2.40), compared to never users (de Sanjose, et al.,
2006). Interestingly, the authors also observed a slight elevated risk among users who
started using hair dyes before 1980, but who ended use during or after 1980 (OR=1.32,
95%CI=1.03-1.69), while a non-significant association was found among users who used
24

hair dyes after 1980 (OR=1.11, 95%CI=0.92-1.34) (de Sanjose, et al., 2006). Another
case-control study found an increased risk of NHL for use of hair dyes before 1980,
especially for permanent, intense tones (black, dark brown, dark blonde) dye (OR=1.6,
95%CI=0.9-2.7) (Morton, et al., 2007).
In our study, we observed a significant elevated risk of B-cell NHL in women
who began using hair dyes prior to 1980 (adjusted OR=1.45, 95%CI=1.08-1.96), which
confirmed that risk was elevated for when use was initiated before changes were made in
the composition of the dyes. However, we observed a marginally significant increased
risk for hair dye use among women who began using hair dyes during or after 1980
(adjusted OR=1.31, 95%CI=0.98-1.74). Therefore, we cannot conclude that the risk of
NHL is associated only with first use of hair dye products prior to 1980. Further research
is warranted to evaluate the effects after that time period and to identify potentially
harmful components of the dyes as currently formulated.
Earlier studies regarded NHL as one disease entity with differing distributions by
NHL subtype; due to the potentially different etiological mechanisms underlying each
subtype of NHL, this could be a reason for conflicting results between studies on the
association between hair dye use and NHL. In the current investigation, we used the latest
WHO 2001 classification to evaluate the associations with three major histologic
subtypes (DLBCL, follicular lymphoma, CLL/SLL). Our results suggest that personal
hair dye use affects only follicular lymphoma and CLL/SLL, but not DLBCL.
In our study the associations between hair dye use and follicular lymphoma are
not statistically significant, largely due to the limited sample size. Elevated risks are
observed for most of the exposures, although none of the confidence intervals for the OR
25

excluded 1. Zhang et al reported a slight increased follicular lymphoma risk with any
type of hair dye use initiated prior to 1980 (OR=1.4, 95%CI=1.1-1.9), and with dark
color dye use initiated after 1980 (OR=1.5, 95%CI=1.1-2.0) (Zhang, et al., 2008). Morton
et al. found an increased risk of follicular lymphoma, especially among users of
permanent, intense color (black, dark brown, dark blonde) dyes when use was initiated
prior to 1980 (OR=2.0, 95%CI=0.9-4.4) (Morton, et al., 2007). Similar findings of non-
significant increased risks of follicular lymphoma associated with permanent dyes were
also reported by Grodstein et al.(Grodstein, et al., 1994), Zhang et al.(Zhang, et al., 2004),
Miligi et al.(Miligi, et al., 2005), and Sanjose et al.(de Sanjose, et al., 2006). Further
studies or pooled analyses are needed to explore the effects for this specific subtype.
The association between hair dye use and risk of CLL/SLL was also found in
previous studies. Wong et al. reported an inverse risk of CLL/SLL associated with hair
due use; however, this finding was based on a limited number of cases (n=15) (Wong, et
al., 2010). In contrast with this study, most other studies of subtypes found positive
associations with CLL/SLL (Miligi, et al., 2005; de Sanjose, et al., 2006; Zhang, et al.,
2008). A case-control study in Europe observed an increased risk of CLL among hair dye
users, and a two-fold elevated risk of CLL when use started prior to 1980
(OR=1.43,95%CI=1.01-2.03; OR=2.02, 95%CI=1.31-3.11 respectively) (de Sanjose, et
al., 2006). Miligi et al reported black hair dyes were statistically significantly related to a
three-fold higher risk of CLL (OR=3.0, 95%CI=1.1-7.5)(Miligi, et al., 2005). In addition,
Zhang et al also found women who started using hair dyes before 1980 had a 50% higher
risk of CLL/SLL (OR=1.5, 95%CI=1.1-2.0) (Zhang, et al., 2008). Our results are
generally consistent with these three studies. The association between hair dyes and
26

CLL/SLL risk was slightly stronger than that observed for follicular lymphoma, DLBCL,
or NHL overall. We found a four-fold elevated risk associated with black hair dyes (with
9 exposed controls) (adjusted OR=4.61, 95%CI=1.58-13.42), and an approximately two-
fold increased risk associated with initiation before 1980 (adjusted OR=1.92,
95%CI=1.01-3.66).
Even though it is known that the incidence of NHL varies by race (with highest
incidence in Whites and lowest in American-Indians/Alaska Natives), no prior studies
have investigated the association between personal hair dye use and NHL risk in different
racial/ethnic groups. Our results indicate that personal hair dye uses affects all
racial/ethnic groups fairly equally. In general, relative risk estimates were higher for
those who had ever used permanent dyes, black hair dyes, and also higher for those who
initiated hair dye use prior to 1980. However, there were relatively few subjects in some
categories; further studies with larger sample sizes are needed to explore the effects of
hair dyes in different racial/ethnic groups.
Several lines of evidence indicate that hair dyes may be linked to genetic variation
in the development of NHL. Several genes variants (CYP2C9 Ex3-52C>T TT/CT,
CYP2E1 -332T>A AT/AA, GSTP1 Ex5-24A>G AA, NAT2 genotypes, et al) involved in
xenobiotic metabolism have been found to modify the association between hair dye use
and NHL risk (Morton, et al., 2007; Zhang et al., 2009). Both Morton et al and Zhang et
al have reported that women who carried the NAT2 rapid/intermediate acetylator
phenotype have higher risk of NHL particularly those who began using hair dyes before
1980 (Morton, et al., 2007; Zhang, et al., 2009). NAT2 rapid/intermediate acetylation
may lead to the formulation of DNA adducts by activating the xenobiotic metabolism of
27

aromatic and heterocyclic amines in hair dye products (Morton, et al., 2007; Zhang, et al.,
2009). Further research is needed to examine whether the association of NHL risk with
hair dye use is modified other genetic variants and whether this modification varies by
subtype of NHL.
Our study has many advantages. First of all, with the relatively large sample size
in this matched case-control study, greater statistical power to detect moderate risks and
more precise confidence intervals estimates can be expected. Also the use of population-
based case and control selection reduces the likelihood of selection bias, in comparison
with hospital-based case-control studies. To reduce survival bias, we used the RCA
procedure within the CSP to identify the B-cell NHL patients recently diagnosed in Los
Angeles County within 30-35 days, on average. For each patient, we then reviewed the
pathology reports and bioassay results to confirm the NHL subtype classification and to
exclude ineligible cases, which minimized the potential for misclassification of the cases.
Also we applied the newly developed WHO classification system to define our NHL
histologic subtypes; this classification has the advantage over prior classifications of
being based on epidemiological considerations. In addition, we considered potential
confounding factors in our study. In contrast, some previous studies fail to consider
confounding factors (eg. smoking) in their analysis, which might bias their results
(Takkouche, et al., 2005).
There are also limitations in our study. First, the study is restricted to women and
our findings are not generalizable to men. However, this restriction actually benefits
assessment of hair dye usage as we obtained substantially increased statistical power for
the same level of study effort since hair dye usage is much more common in women than
28

in men. Second, although the sample size of our study was larger than most other studies,
the statistical power to examine the associations within different subtypes or in different
racial/ethnic groups was limited. Studies with larger sample sizes are warranted to
replicate our findings in subgroup analyses. Thirdly, although we used a well-tested and
standardized interview process and a list of hair dye products to help participants recall
their hair dye use history, the retrospective recall of exposure could introduce bias if
recall were more accurate among cases than among controls. Fourth, participation rate
among cases (54.8%) were relatively low, suggesting the possibility of bias if the
prevalence rate of hair dye use among participants differ from that of non-participants.
However, it seems unlikely since clear dose-response relationships were observed in our
study.
In conclusion, our study suggests that use of personal hair dye is associated with
greater risk of B-cell NHL, and particularly risk of follicular lymphoma and CLL/SLL.
Further studies are needed to confirm these findings by subtype and race or ethnicity and
to address the biological/genetic mechanisms involved in these associations.
29

REFERENCES

ACS. (2010). Cancer Facts & Figures 2010. Atlanta: American Cancer Society, from
http://www.cancer.org/acs/groups/content/@nho/documents/document/acspc-
024113.pdf

Altekruse, S. F., C.L., K., M, K., N, N., R, A., W, W., et al. (2009). SEER Cancer
Statistics Review, 1975-2007. National Cancer Institute, from
http://seer.cancer.gov/csr/1975_2007/

Altekruse, S. F., Henley, S. J., & Thun, M. J. (1999). Deaths from hematopoietic and
other cancers in relation to permanent hair dye use in a large prospective study
(United States). Cancer Causes Control, 10(6), 617-625.

Bailey, J. E., & Skare, J. A. (2008). Reply: letter to the editor in response to Morton et al.
(2007) Carcinogenesis, 28, 1759-1764. Carcinogenesis, 29(9), 1851.

Baris, D., & Zahm, S. H. (2000). Epidemiology of lymphomas. Current Opinion in
Oncology, 12(5), 383-394.

Bolduc, C., & Shapiro, J. (2001). Hair care products: waving, straightening, conditioning,
and coloring. Clin Dermatol, 19(4), 431-436.

Burnett, C., Lanman, B., Giovacchini, R., Wolcott, G., Scala, R., & Keplinger, M. (1975).
Long-term toxicity studies on oxidation hair dyes. Food Cosmet Toxicol, 13(3),
353-357.

Corbett, J. F. (1999). An historical review of the use of dye precursors in the formulation
of commercial oxidation hair dyes. Dyes and Pigments(41), 127-136.

de Sanjose, S., Benavente, Y., Nieters, A., Foretova, L., Maynadie, M., Cocco, P. L., et al.
(2006). Association between personal use of hair dyes and lymphoid neoplasms in
Europe. Am J Epidemiol, 164(1), 47-55.

Fisher, S. G., & Fisher, R. I. (2004). The epidemiology of non-Hodgkin's lymphoma.
Oncogene, 23(38), 6524-6534.

Grodstein, F., Hennekens, C. H., Colditz, G. A., Hunter, D. J., & Stampfer, M. J. (1994).
A prospective study of permanent hair dye use and hematopoietic cancer. J Natl
Cancer Inst, 86(19), 1466-1470.

Holly, E. A., Lele, C., & Bracci, P. M. (1998). Hair-color products and risk for non-
Hodgkin's lymphoma: a population-based study in the San Francisco bay area. Am
J Public Health, 88(12), 1767-1773.
30

Jemal, A., Siegel, R., Xu, J. Q., & Ward, E. (2010). Cancer Statistics, 2010. Ca-a Cancer
Journal for Clinicians, 60(5), 277-300.

La Vecchia, C., & Tavani, A. (1995). Epidemiological evidence on hair dyes and the risk
of cancer in humans. Eur J Cancer Prev, 4(1), 31-43.

Mendelsohn, J. B., Li, Q. Z., Ji, B. T., Shu, X. O., Yang, G., Li, H. L., et al. (2009).
Personal use of hair dye and cancer risk in a prospective cohort of Chinese
women. Cancer Sci, 100(6), 1088-1091.

Miligi, L., Costantini, A. S., Benvenuti, A., Veraldi, A., Tumino, R., Ramazzotti, V., et al.
(2005). Personal use of hair dyes and hematolymphopoietic malignancies. Arch
Environ Occup Health, 60(5), 249-256.

Miligi, L., Seniori Costantini, A., Crosignani, P., Fontana, A., Masala, G., Nanni, O., et al.
(1999). Occupational, environmental, and life-style factors associated with the
risk of hematolymphopoietic malignancies in women. Am J Ind Med, 36(1), 60-69.

Morton, L. M., Bernstein, L., Wang, S. S., Hein, D. W., Rothman, N., Colt, J. S., et al.
(2007). Hair dye use, genetic variation in N-acetyltransferase 1 (NAT1) and 2
(NAT2), and risk of non-Hodgkin lymphoma. Carcinogenesis, 28(8), 1759-1764.

Morton, L. M., Wang, S. S., Devesa, S. S., Hartge, P., Weisenburger, D. D., & Linet, M.
S. (2006). Lymphoma incidence patterns by WHO subtype in the United States,
1992-2001. Blood, 107(1), 265-276.

Nohynek, G. J., Antignac, E., Re, T., & Toutain, H. (2010). Safety assessment of
personal care products/cosmetics and their ingredients. Toxicol Appl Pharmacol,
243(2), 239-259.

Rollison, D. E., Helzlsouer, K. J., & Pinney, S. M. (2006). Personal hair dye use and
cancer: a systematic literature review and evaluation of exposure assessment in
studies published since 1992. J Toxicol Environ Health B Crit Rev, 9(5), 413-439.

Takkouche, B., Etminan, M., & Montes-Martinez, A. (2005). Personal use of hair dyes
and risk of cancer: a meta-analysis. JAMA, 293(20), 2516-2525.

Tavani, A., Negri, E., Franceschi, S., Talamini, R., Serraino, D., & La Vecchia, C. (2005).
Hair dye use and risk of lymphoid neoplasms and soft tissue sarcomas. Int J
Cancer, 113(4), 629-631.

Thun, M. J., Altekruse, S. F., Namboodiri, M. M., Calle, E. E., Myers, D. G., & Heath, C.
W., Jr. (1994). Hair dye use and risk of fatal cancers in U.S. women. J Natl
Cancer Inst, 86(3), 210-215.

31

Turesky, R. J., Freeman, J. P., Holland, R. D., Nestorick, D. M., Miller, D. W.,
Ratnasinghe, D. L., et al. (2003). Identification of aminobiphenyl derivatives in
commercial hair dyes. Chem Res Toxicol, 16(9), 1162-1173.

USC. (2002). Cancer Surveillance Program. from
http://www.usc.edu/schools/medicine/departments/preventive_medicine/divisions
/epidemiology/research/csp/index.html

Wong, O., Harris, F., Wang, Y., & Fu, H. (2010). A hospital-based case-control study of
non-Hodgkin lymphoid neoplasms in Shanghai: analysis of personal
characteristics, lifestyle, and environmental risk factors by subtypes of the WHO
classification. J Occup Environ Med, 52(1), 39-53.

Zahm, S. H., Morton, L. M., Bernstein, L., Wang, S. S., Colt, J. S., Severson, R. K., et al.
(2008). Re: Hair dye use, genetic variation in N-acetyltransferase 1 (NAT1) and 2
(NAT2), and risk of non-Hodgkin lymphoma, author response. Carcinogenesis,
29(5), 1084-1085.

Zahm, S. H., Weisenburger, D. D., Babbitt, P. A., Saal, R. C., Vaught, J. B., & Blair, A.
(1992). Use of hair coloring products and the risk of lymphoma, multiple
myeloma, and chronic lymphocytic leukemia. Am J Public Health, 82(7), 990-997.

Zhang, Y., Holford, T. R., Leaderer, B., Boyle, P., Zahm, S. H., Flynn, S., et al. (2004).
Hair-coloring product use and risk of non-Hodgkin's lymphoma: a population-
based case-control study in Connecticut. Am J Epidemiol, 159(2), 148-154.

Zhang, Y., Hughes, K. J., Zahm, S. H., Holford, T. R., Dai, L., Bai, Y., et al. (2009).
Genetic variations in xenobiotic metabolic pathway genes, personal hair dye use,
and risk of non-Hodgkin lymphoma. Am J Epidemiol, 170(10), 1222-1230.

Zhang, Y., Sanjose, S. D., Bracci, P. M., Morton, L. M., Wang, R., Brennan, P., et al.
(2008). Personal use of hair dye and the risk of certain subtypes of non-Hodgkin
lymphoma. Am J Epidemiol, 167(11), 1321-1331.

Abstract (if available)

Abstract Study Aim: There have been inconsistent reports regarding personal hair dye use and B-cell non-Hodgkin’s lymphomas (NHL) risk. This study aims to evaluate the relationship between women’s personal hair dye use and the risk of B-cell NHL. Methods: A 1:1 matched population-based case-control study of women conducted in Los Angeles County recruited 986 case-control pairs. Case participants were diagnosed with histologic confirmed primary B-cell NHL between 2004 and 2008. Control participants were individually matched to the index cases by age, gender, race, and neighborhood of residence. Exposure information was obtained by in-person interviews. Multivariable conditional logistic regression was performed to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of B-cell NHL associated with personal hair dye use overall, by race, and by NHL subtypes, adjusting for potential confounders (alcohol status, smoking status, and education levels).

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Reproductive factors and risk of B-cell non-Hodgkin's lymphomas among women in Los Angeles

PDF

Study of risk factors of B-cell non-Hodgkin lymphoma (NHL) in the California Teachers Study

PDF

Associations between exercise and quality of life in adult survivors of non-Hodgkin's lymphoma and colorectal cancer

PDF

Common immune-related factors and risk of non-Hodgkin lymphomy

PDF

Effect of soy isoflavones on anthropometric and metabolic measurements in postmenopausal women

PDF

Non-steroidal anti-inflammatory drugs and risk of esophageal and gastric adenocarcinomas in Los Angeles county

PDF

Association of carotid artery stiffness with measures of cognition in older adults

PDF

Incidence and survival rates of the three major histologies of renal cell carcinoma

PDF

A cross-sectional examination: impact of diet and physical activity on plasma glucose metabolism, insulin sensitivity and pancreatic β-cell function in Hispanic women with histories of gestationa...

PDF

Genetic epidemiological approaches in the study of risk factors for hematologic malignancies

PDF

Age related macular degeneration in Latinos: risk factors and impact on quality of life

PDF

The effect of vitamin D supplementation on the progression of carotid intima-media thickness and arterial stiffness in elderly African American women: Results of a randomized placebo-controlled trial

PDF

Pharmacogenetic association studies and the impact of population substructure in the women's interagency HIV study

PDF

Chronic eye disease epidemiology in the multiethnic ophthalmology cohorts of California study

Asset Metadata

Creator Xu, Xinxin (author)

Core Title Personal hair dye use and risk of B-cell non-Hodgkin’s lymphomas among adult women in Los Angeles County

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics

Publication Date 05/16/2011

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag case-control study,hair dyes,non-Hodgkin's lymphoma,OAI-PMH Harvest,risk factors,Women

Place Name California (states), Los Angeles (counties)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bernstein, Leslie (committee chair), Azen, Stanley Paul (committee member), Mack, Wendy J. (committee member)

Creator Email reranxin@hotmail.com,xinxinxu@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3531

Unique identifier UC1271246

Identifier etd-Xu-4189 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-397460 (legacy record id),usctheses-m3531 (legacy record id)

Legacy Identifier etd-Xu-4189.pdf

Dmrecord 397460

Document Type Thesis

Rights Xu, Xinxin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu