Abstract

Background Hairdressers and allied occupations represent a large and fast growing group of professionals. The fact that these professionals are chronically exposed to a large number of chemicals present in their work environment, including potential carcinogens contained in hair dyes, makes it necessary to carry out a systematic evaluation of the risk of cancer in this group.

Methods We retrieved studies by systematically searching Medline and other computerized databases, and by manually examining the references of the original articles and monographs retrieved. We also contacted international researchers working on this or similar topics to complete our search. We included 247 studies reporting relative risk (RR) estimates of hairdresser occupation and cancer of different sites.

Results Study-specific RRs were weighted by the inverse of their variance to obtain fixed and random effects pooled estimates. The pooled RR of occupational exposure as a hairdresser was 1.27 (95% CI 1.15–1.41) for lung cancer, 1.52 [95% confidence interval (CI) 1.11–2.08] for larynx cancer, 1.30 (95% CI 1.20–1.42) for bladder cancer and 1.62 (95% CI 1.22–2.14) for multiple myeloma. Data for other anatomic sites showed increases of smaller magnitude. The results restricted to those studies carried out before the ban of two major carcinogens from hair dyes in the mid-1970s were similar to the general results.

Conclusions Hairdressers have a higher risk of cancer than the general population. Improvement of the ventilation system in the hairdresser salons and implementation of hygiene measures aimed at mitigating exposure to potential carcinogens at work may reduce the risk.

Introduction

Hairdressers, barbers and beauticians represent an important occupational group, with >800 000 people employed in the USA and ∼1 million in Europe.1,2 In the USA, this figure will grow 14% through 2016, faster than the average of all occupations.1 These professionals are exposed to several thousands of chemicals contained in colorants, bleaches, shampoos and hair conditioners. Furthermore, hairdressers may be exposed to volatile solvents, propellants and aerosols from hair sprays as well as to formaldehyde, methacrylates and nitrosamines contained in many hair care products.3 Several of these chemicals, essentially those contained in hair dyes, are potentially carcinogenic and some of them have been found in the urine of cosmetologists.4,5 Our previous meta-analysis did not show any marked increase in the risk of cancer for personal use of hair dyes.6 However, the fact that exposure to hair dyes is more prolonged in occupational settings, and has a higher concentration and frequency than personal exposure, in addition to the fact that they may be exposed to carcinogens contained in other products than hair dyes, puts hairdressers, barbers and beauticians at higher risk of cancer. The main social partners in the European hairdressing industry have recently expressed their growing concern on occupational diseases and emphasized the fact that hairdressers have no influence on the composition of cosmetic products, in spite of being exposed to them.7

Although some studies found an increased risk of cancer among hairdressers,8,9 others failed to find any association.10,11 The International Agency for Research on Cancer (IARC) reaffirmed recently that occupational exposure of hairdressers and barbers was ‘probably carcinogenic’ and that, globally, there was ‘limited evidence’ on this carcinogenicity.12 An increased risk of cancer among hairdressers and personal appearance workers would cause an important public health concern given the large number of people employed in this sector. To date, no comprehensive meta-analysis has focused on the relation between occupational exposure among hairdressers and allied occupations and the risk of cancer. We therefore sought to explore this issue through a meta-analysis, adhering strictly to the MOOSE guidelines for meta-analyses of observational studies.13

Methods

Search strategy

To identify eligible studies, we systematically searched Medline from 1966 to March 2009 for both English and non-English language articles by applying the following algorithm: (HAIRDRESSER* OR BEAUTICIAN* OR COSMETOLOGIST* OR BARBER*) AND (CANCER* OR NEOPLASM* OR CARCINOGEN*) both in Medical Subject Heading and in free text words. We used similar strategies to search EMBASE (1980–2009) and LILACS databases (Latin America and Caribbean) from 1982 to 2009. We searched meeting abstracts using the ISI Proceedings database from inception to 2009. Furthermore, we manually examined the bibliography of the papers that were retrieved electronically and of recent narrative reviews and monographs on carcinogenic risks of hairdressers.3,14 During a recent expert meeting on the same topic, held at IARC (Lyon, France),15 we contacted scientists working on the subject or reviewing it to identify studies that we had potentially missed in our search. We also searched for unpublished studies and results published in other supports such as internal reports, PhD dissertations and other theses. Finally, in order to complete our search and make it inclusive of those studies that list estimates for hairdressers as secondary exposure only—which may have been thus missed in previous computerized searches—we have checked the tables of every article that considers occupation at large, and not only employment as a hairdresser, as an exposure factor of cancer.

We defined a study unit (or dataset) as the analysis of the relation between occupational hairdressing exposure and cancer of a specific anatomic site. As different anatomic sites may be reported in the same article, a single publication could then report more than one study unit. For the sake of simplification, in this report we use the term ‘study’ as a synonym for study unit or dataset.

All searches were carried out independently by two epidemiologists (B.T. and A.M.M.) and discrepancies were resolved by agreement.

Inclusion criteria and data collection

Studies were included if they met the following criteria: (i) presented original data from case–control, cohort or proportional or standardized mortality or morbidity ratio studies (PMR, SMR); (ii) the outcome of interest was clearly defined as cancer of an anatomical site; (iii) the exposure of interest was occupation as a hairdresser, beautician or barber; and (iv) provided relative risk (RR) estimates and their confidence intervals (CIs) or provided enough data to calculate them (raw data, observed and expected cases in PMR and SMR studies, P-value or variance estimate). Studies on childhood cancers related to the parents’ occupation were not considered. To prevent multiple appearances of the same cohort in our analysis, we grouped the studies so that we could trace the same population. We included the most recent study as it generally provided the longest follow-up and the largest number of cases.

We carried out as many analyses as different anatomical cancer sites were available. Thus, when RRs of cancer of different anatomical sites were available in the same publication, we used this publication several times.

We recorded on a standardized questionnaire study name, year of publication, study design, type of outcome (incidence or mortality), sample size (cases and controls, or number of exposed cases for PMR, SMR and cohort studies), type of controls for case−control studies (hospital or population controls), variables used for adjustment or matching, and effect measures with their corresponding 95% CI.

Quality assessment

As no universal scale is available for measuring quality of observational studies, we followed the recommendations of the MOOSE guidelines and assessed the quality of key components of design separately rather than generate a single aggregate score.13 Following this recommendation, we assessed study quality based on the following five criteria, which contain elements of the Newcastle–Ottawa Quality Scale,16 labelled as ‘yes’ or ‘no’. For a comprehensive assessment of quality, we chose items that could apply to all designs: (i) whether the target population of the study was clearly defined or, on the contrary, the subjects were chosen at convenience; (ii) whether or not the diagnosis of cancer was reliable and included either pathology reports, direct cause on death certificates or information from an established cancer register; (iii) whether or not job assessment was reliable and included duration of employment or job matrix measurement; (iv) whether or not occupation as a hairdresser occurred clearly before cancer onset; and (v) whether or not confounding by smoking was prevented by matching or adjustment. Throughout this assessment, when the information on a specific item was not provided by the authors, we graded this item as ‘no’.

Within each item, we calculated two pooled odds ratios (ORs): one for those studies that were labelled ‘yes’ and one for those labelled ‘no’. As a secondary analysis, we carried out a pooled analysis on those studies that fulfilled more than three criteria and compared with those that scored ≤3.

Quality scoring was performed independently by two reviewers (B.T. and C.R.M.) and the results were merged by consensus. The complete protocol for quality scoring is available upon request from the first author.

Statistical analysis

We weighted the study-specific adjusted log ORs for case−control studies and log RRs for cohort studies by the inverse of their variance to compute a pooled RR and its 95% CI. ORs were considered estimates of RRs. We assumed that the person-time of the unexposed group is much larger than that of the exposed group and thus considered SMRs as equivalent to incidence rate ratios.17

For the same anatomical cancer site, we presented results for mortality and incidence studies separately. Nevertheless, in a second instance, we included incidence and mortality measures of effect together (i.e. SMRs or PMRs and RRs). The difference between mortality RRs and incidence RRs for the same cancer location and exposure is generally small.18

As recommended by methodologic experts, proportional mortality studies were considered as a variant of the case–control design.19

We presented both fixed and random effects pooled estimates, but used preferentially the latter when heterogeneity was detected.

We used a version adapted to small samples of the DerSimonian and Laird Q test to check for heterogeneity.20 The null hypothesis of this test is absence of heterogeneity. To quantify this heterogeneity we calculated the proportion of the total variance due to between-study variance (Ri statistic).20 We further explored the source of heterogeneity by restricting the analysis to subgroups of studies defined by characteristics such as type of study design (case–control or cohort), adjustment factors and quality scale.

We assessed publication bias visually through funnel plots and formal testing using the test proposed by Egger.21 We also performed sensitivity analyses, recalculating the pooled estimates under extreme conditions.

All analyses were performed with the software HEpiMA® version 2.1.320 and STATA version 8.0 (StataCorp LP, College Station, TX, USA).

Results

Our search retrieved 247 studies, published in 67 different articles and carried out in 14 countries, on cancer on 22 different anatomic sites among hairdressers and allied occupations that met our inclusion criteria (Tables 1–5). We did not find any unpublished study.

Table 1

Study-specific RRs and 95% CIs of gynaecologic cancers among hairdressers and related occupations

RR (95% CI)
ReferencesBreastCervix uteriCorpus uteriOvaryType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle430.91 (0.48–1.70)1.10 (0.51–2.24)1.52 (0.50–3.89)2.91 (1.21–6.46)HAge, sex21/719
    Kinlen et al.440.58 (0.06–2.77)H/PAge, sex, marital status, social class191/561
    Koenig et al.450.70 (0.14–3.44)PAge, sex, family history, age at 1st birth, others398/790
    Habel et al.461.5 (0.5–4.8)PAge, sex, education, alcohol, body mass index537/492
    Band et al.471.05 (0.46–2.41)PAge, sex, education, alcohol, smoking, others1018/1020
Proportionate mortality or morbidity studies
    Menck et al.481.01 (0.75–1.36)Age, sex135
    Kono et al.490.59 (0.19–1.38)1.36 (0.44–3.16)Age, sex, period141
    Spinelli et al.500.79 (0.36–15.1)2.04 (0.88–4.03)Age, sex39
    Lamba et al.511.11 (1.04–1.17)0.96 (0.80–1.14)0.94 (0.80–1.10)1.02 (0.92–1.14)Age, sex, race, region9495
Cohort studies
    Teta et al.521.02 (0.89–1.17)0.80 (0.55–1.13)1.23 (0.95–1.56)1.34 (0.99–1.78)Age, sex, calendar year688
    Gubéran et al.530.56 (0.27–1.17)1.90 (0.71–5.06)0.5 (0.07–3.55)Age, sex, matrimonial status65
    Kato et al.541.33 (1.00–1.78)Age, sex46
    Andersen et al.55 (Denmark)1.01 (0.85–1.20)1.08 (0.81–1.45)1.52 (1.12–2.06)1.33 (0.97–1.83)Age, sex, period911
    Andersen et al.55 (Finland)1.03 (0.88–1.21)1.26 (0.79–2.00)1.24 (0.91–1.68)Age, sex, period549
    Andersen et al.55 (Norway)1.00 (0.80–1.25)0.92 (0.58–1.46)1 (0.64–1.57)0.91 (0.58–1.43)Age, sex, period492
    Vasama-Neuvonen et al.81.3 (1.0–1.7)Age, sex, period, social status60
    Czene et al.561.02 (0.95–1.09)1.28 (1.12–1.47)0.91 (0.77–1.08)1.11 (0.96–1.28)Age, sex, period3901
RR (95% CI)
ReferencesBreastCervix uteriCorpus uteriOvaryType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle430.91 (0.48–1.70)1.10 (0.51–2.24)1.52 (0.50–3.89)2.91 (1.21–6.46)HAge, sex21/719
    Kinlen et al.440.58 (0.06–2.77)H/PAge, sex, marital status, social class191/561
    Koenig et al.450.70 (0.14–3.44)PAge, sex, family history, age at 1st birth, others398/790
    Habel et al.461.5 (0.5–4.8)PAge, sex, education, alcohol, body mass index537/492
    Band et al.471.05 (0.46–2.41)PAge, sex, education, alcohol, smoking, others1018/1020
Proportionate mortality or morbidity studies
    Menck et al.481.01 (0.75–1.36)Age, sex135
    Kono et al.490.59 (0.19–1.38)1.36 (0.44–3.16)Age, sex, period141
    Spinelli et al.500.79 (0.36–15.1)2.04 (0.88–4.03)Age, sex39
    Lamba et al.511.11 (1.04–1.17)0.96 (0.80–1.14)0.94 (0.80–1.10)1.02 (0.92–1.14)Age, sex, race, region9495
Cohort studies
    Teta et al.521.02 (0.89–1.17)0.80 (0.55–1.13)1.23 (0.95–1.56)1.34 (0.99–1.78)Age, sex, calendar year688
    Gubéran et al.530.56 (0.27–1.17)1.90 (0.71–5.06)0.5 (0.07–3.55)Age, sex, matrimonial status65
    Kato et al.541.33 (1.00–1.78)Age, sex46
    Andersen et al.55 (Denmark)1.01 (0.85–1.20)1.08 (0.81–1.45)1.52 (1.12–2.06)1.33 (0.97–1.83)Age, sex, period911
    Andersen et al.55 (Finland)1.03 (0.88–1.21)1.26 (0.79–2.00)1.24 (0.91–1.68)Age, sex, period549
    Andersen et al.55 (Norway)1.00 (0.80–1.25)0.92 (0.58–1.46)1 (0.64–1.57)0.91 (0.58–1.43)Age, sex, period492
    Vasama-Neuvonen et al.81.3 (1.0–1.7)Age, sex, period, social status60
    Czene et al.561.02 (0.95–1.09)1.28 (1.12–1.47)0.91 (0.77–1.08)1.11 (0.96–1.28)Age, sex, period3901

H = hospital controls; P = population controls.

Table 1

Study-specific RRs and 95% CIs of gynaecologic cancers among hairdressers and related occupations

RR (95% CI)
ReferencesBreastCervix uteriCorpus uteriOvaryType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle430.91 (0.48–1.70)1.10 (0.51–2.24)1.52 (0.50–3.89)2.91 (1.21–6.46)HAge, sex21/719
    Kinlen et al.440.58 (0.06–2.77)H/PAge, sex, marital status, social class191/561
    Koenig et al.450.70 (0.14–3.44)PAge, sex, family history, age at 1st birth, others398/790
    Habel et al.461.5 (0.5–4.8)PAge, sex, education, alcohol, body mass index537/492
    Band et al.471.05 (0.46–2.41)PAge, sex, education, alcohol, smoking, others1018/1020
Proportionate mortality or morbidity studies
    Menck et al.481.01 (0.75–1.36)Age, sex135
    Kono et al.490.59 (0.19–1.38)1.36 (0.44–3.16)Age, sex, period141
    Spinelli et al.500.79 (0.36–15.1)2.04 (0.88–4.03)Age, sex39
    Lamba et al.511.11 (1.04–1.17)0.96 (0.80–1.14)0.94 (0.80–1.10)1.02 (0.92–1.14)Age, sex, race, region9495
Cohort studies
    Teta et al.521.02 (0.89–1.17)0.80 (0.55–1.13)1.23 (0.95–1.56)1.34 (0.99–1.78)Age, sex, calendar year688
    Gubéran et al.530.56 (0.27–1.17)1.90 (0.71–5.06)0.5 (0.07–3.55)Age, sex, matrimonial status65
    Kato et al.541.33 (1.00–1.78)Age, sex46
    Andersen et al.55 (Denmark)1.01 (0.85–1.20)1.08 (0.81–1.45)1.52 (1.12–2.06)1.33 (0.97–1.83)Age, sex, period911
    Andersen et al.55 (Finland)1.03 (0.88–1.21)1.26 (0.79–2.00)1.24 (0.91–1.68)Age, sex, period549
    Andersen et al.55 (Norway)1.00 (0.80–1.25)0.92 (0.58–1.46)1 (0.64–1.57)0.91 (0.58–1.43)Age, sex, period492
    Vasama-Neuvonen et al.81.3 (1.0–1.7)Age, sex, period, social status60
    Czene et al.561.02 (0.95–1.09)1.28 (1.12–1.47)0.91 (0.77–1.08)1.11 (0.96–1.28)Age, sex, period3901
RR (95% CI)
ReferencesBreastCervix uteriCorpus uteriOvaryType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle430.91 (0.48–1.70)1.10 (0.51–2.24)1.52 (0.50–3.89)2.91 (1.21–6.46)HAge, sex21/719
    Kinlen et al.440.58 (0.06–2.77)H/PAge, sex, marital status, social class191/561
    Koenig et al.450.70 (0.14–3.44)PAge, sex, family history, age at 1st birth, others398/790
    Habel et al.461.5 (0.5–4.8)PAge, sex, education, alcohol, body mass index537/492
    Band et al.471.05 (0.46–2.41)PAge, sex, education, alcohol, smoking, others1018/1020
Proportionate mortality or morbidity studies
    Menck et al.481.01 (0.75–1.36)Age, sex135
    Kono et al.490.59 (0.19–1.38)1.36 (0.44–3.16)Age, sex, period141
    Spinelli et al.500.79 (0.36–15.1)2.04 (0.88–4.03)Age, sex39
    Lamba et al.511.11 (1.04–1.17)0.96 (0.80–1.14)0.94 (0.80–1.10)1.02 (0.92–1.14)Age, sex, race, region9495
Cohort studies
    Teta et al.521.02 (0.89–1.17)0.80 (0.55–1.13)1.23 (0.95–1.56)1.34 (0.99–1.78)Age, sex, calendar year688
    Gubéran et al.530.56 (0.27–1.17)1.90 (0.71–5.06)0.5 (0.07–3.55)Age, sex, matrimonial status65
    Kato et al.541.33 (1.00–1.78)Age, sex46
    Andersen et al.55 (Denmark)1.01 (0.85–1.20)1.08 (0.81–1.45)1.52 (1.12–2.06)1.33 (0.97–1.83)Age, sex, period911
    Andersen et al.55 (Finland)1.03 (0.88–1.21)1.26 (0.79–2.00)1.24 (0.91–1.68)Age, sex, period549
    Andersen et al.55 (Norway)1.00 (0.80–1.25)0.92 (0.58–1.46)1 (0.64–1.57)0.91 (0.58–1.43)Age, sex, period492
    Vasama-Neuvonen et al.81.3 (1.0–1.7)Age, sex, period, social status60
    Czene et al.561.02 (0.95–1.09)1.28 (1.12–1.47)0.91 (0.77–1.08)1.11 (0.96–1.28)Age, sex, period3901

H = hospital controls; P = population controls.

Table 2

Study-specific RRs and 95% CIs of haematopoietic cancers among hairdressers and related occupations

RR (95% CI)
ReferencesHodgkin's diseaseNon-Hodgkin's lymphomaMultiple myelomaLeukemiaType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.27 (0.37–3.45)6.41 (0.63–34.13)1.16 (0.03–7.78)HAge, sex139/719
    Flodin et al.583.3 (0.24–45.7)PNot given131/431
    Persson et al.592.7 (0.19–38.28)2.2 (0.17–28.71)HAge, sex, farming, fresh wood54/275
    Eriksson and Karlsson600.67 (0.13–3.64)PAge, sex, county275/275
    Pottern et al.610.7 (0.0–5.8)PAge, sex607/2596
    Cote et al.100.65 (0.3–1.3)PAge, sex, race2153/8612
    Blair et al.622.7 (0.9–8.7)PAge, sex, race, smoking, death year, others622/1245
    Herrinton et al.631.32 (0.65– 2.65)PAge, sex, race, location, education level360/933
    Figgs et al.641.7 (1.1–2.6)PAge, sex, race12 148/60 740
    Mele et al.652.25 (0.77–6.58)HAge, sex, region, education517/1161
    Miligi et al.662.1 (0.7–6.5)1.9 (0.7–5.8)11.1 (1.8–67.0)2.2 (0.7–7.1)PAge, sex, smoking, education1170/828
    Seniori Costantini et al.670.6 (0.2–1.6)2.2 (0.7–6.9))1.0 (0.3–3.2)PAge, sex2737/1779
    ‘t Mannetje et al.111.09 (0.27–4.35)PAge, sex, smoking, ethnicity, occupational status291/471
Proportionate mortality studies
    Alderson901.11 (0.36–3.44)Age, sex, period134
    Kono et al.491.37 (0.50–2.98)Age, sex, period141
    Spinelli et al.506.19 (1.27–18.11)1.11 (0.13–4.01)Age, sex39
    Gallagher et al.930.56 (0.01–3.10)0.99 (0.12–3.59)0.58 (0.07–2.11)1.69 (0.94–2.78)Age, sex288
    Shibata et al.680.78 (0.19–3.10)1.16 (0.37–3.60)Age, sex5
    Lamba et al.511.38 (1.08–1.75)1.11 (1.01–1.24)1.21 (1.06–1.39)1.1 (1.02–1.2)Age, sex, race, region9495
Cohort studies
    Guidotti et al.573.73 (2.53–5.51)AgeNot given
    Teta et al.521.29 (0.81–1.95)0.69 (0.14–2.02)1.20 (0.66–2.02)Age, sex, calendar year688
    Gubéran et al.532.5 (0.35–17.75)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.20–9.94)1.3 (0.49–3.46)2.5 (0.95–6.58)1.1 (0.53–2.30)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.61 (0.81–3.22)1.54 (1.05–2.25)0.90 (0.45–1.89)0.99 (0.63–1.55)Age, sex, period911
    Andersen et al.55 (Finland)1.08 (0.35–3.35)0.97 (0.50–1.86)1.12 (0.53–2.35)0.99 (0.54–1.85)Age, sex, period549
    Andersen et al.55 (Norway)1.16 (0.70–1.93)1.16 (0.48–2.79)1.41 (0.73–2.71)Age, sex, period492
    Czene et al.560.78 (0.50–1.20)0.93 (0.76–1.14)1.25 (0.95–1.65)1.00 (0.81–1.23)Age, sex, period3901
RR (95% CI)
ReferencesHodgkin's diseaseNon-Hodgkin's lymphomaMultiple myelomaLeukemiaType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.27 (0.37–3.45)6.41 (0.63–34.13)1.16 (0.03–7.78)HAge, sex139/719
    Flodin et al.583.3 (0.24–45.7)PNot given131/431
    Persson et al.592.7 (0.19–38.28)2.2 (0.17–28.71)HAge, sex, farming, fresh wood54/275
    Eriksson and Karlsson600.67 (0.13–3.64)PAge, sex, county275/275
    Pottern et al.610.7 (0.0–5.8)PAge, sex607/2596
    Cote et al.100.65 (0.3–1.3)PAge, sex, race2153/8612
    Blair et al.622.7 (0.9–8.7)PAge, sex, race, smoking, death year, others622/1245
    Herrinton et al.631.32 (0.65– 2.65)PAge, sex, race, location, education level360/933
    Figgs et al.641.7 (1.1–2.6)PAge, sex, race12 148/60 740
    Mele et al.652.25 (0.77–6.58)HAge, sex, region, education517/1161
    Miligi et al.662.1 (0.7–6.5)1.9 (0.7–5.8)11.1 (1.8–67.0)2.2 (0.7–7.1)PAge, sex, smoking, education1170/828
    Seniori Costantini et al.670.6 (0.2–1.6)2.2 (0.7–6.9))1.0 (0.3–3.2)PAge, sex2737/1779
    ‘t Mannetje et al.111.09 (0.27–4.35)PAge, sex, smoking, ethnicity, occupational status291/471
Proportionate mortality studies
    Alderson901.11 (0.36–3.44)Age, sex, period134
    Kono et al.491.37 (0.50–2.98)Age, sex, period141
    Spinelli et al.506.19 (1.27–18.11)1.11 (0.13–4.01)Age, sex39
    Gallagher et al.930.56 (0.01–3.10)0.99 (0.12–3.59)0.58 (0.07–2.11)1.69 (0.94–2.78)Age, sex288
    Shibata et al.680.78 (0.19–3.10)1.16 (0.37–3.60)Age, sex5
    Lamba et al.511.38 (1.08–1.75)1.11 (1.01–1.24)1.21 (1.06–1.39)1.1 (1.02–1.2)Age, sex, race, region9495
Cohort studies
    Guidotti et al.573.73 (2.53–5.51)AgeNot given
    Teta et al.521.29 (0.81–1.95)0.69 (0.14–2.02)1.20 (0.66–2.02)Age, sex, calendar year688
    Gubéran et al.532.5 (0.35–17.75)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.20–9.94)1.3 (0.49–3.46)2.5 (0.95–6.58)1.1 (0.53–2.30)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.61 (0.81–3.22)1.54 (1.05–2.25)0.90 (0.45–1.89)0.99 (0.63–1.55)Age, sex, period911
    Andersen et al.55 (Finland)1.08 (0.35–3.35)0.97 (0.50–1.86)1.12 (0.53–2.35)0.99 (0.54–1.85)Age, sex, period549
    Andersen et al.55 (Norway)1.16 (0.70–1.93)1.16 (0.48–2.79)1.41 (0.73–2.71)Age, sex, period492
    Czene et al.560.78 (0.50–1.20)0.93 (0.76–1.14)1.25 (0.95–1.65)1.00 (0.81–1.23)Age, sex, period3901

H = hospital controls; P = population controls.

Table 2

Study-specific RRs and 95% CIs of haematopoietic cancers among hairdressers and related occupations

RR (95% CI)
ReferencesHodgkin's diseaseNon-Hodgkin's lymphomaMultiple myelomaLeukemiaType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.27 (0.37–3.45)6.41 (0.63–34.13)1.16 (0.03–7.78)HAge, sex139/719
    Flodin et al.583.3 (0.24–45.7)PNot given131/431
    Persson et al.592.7 (0.19–38.28)2.2 (0.17–28.71)HAge, sex, farming, fresh wood54/275
    Eriksson and Karlsson600.67 (0.13–3.64)PAge, sex, county275/275
    Pottern et al.610.7 (0.0–5.8)PAge, sex607/2596
    Cote et al.100.65 (0.3–1.3)PAge, sex, race2153/8612
    Blair et al.622.7 (0.9–8.7)PAge, sex, race, smoking, death year, others622/1245
    Herrinton et al.631.32 (0.65– 2.65)PAge, sex, race, location, education level360/933
    Figgs et al.641.7 (1.1–2.6)PAge, sex, race12 148/60 740
    Mele et al.652.25 (0.77–6.58)HAge, sex, region, education517/1161
    Miligi et al.662.1 (0.7–6.5)1.9 (0.7–5.8)11.1 (1.8–67.0)2.2 (0.7–7.1)PAge, sex, smoking, education1170/828
    Seniori Costantini et al.670.6 (0.2–1.6)2.2 (0.7–6.9))1.0 (0.3–3.2)PAge, sex2737/1779
    ‘t Mannetje et al.111.09 (0.27–4.35)PAge, sex, smoking, ethnicity, occupational status291/471
Proportionate mortality studies
    Alderson901.11 (0.36–3.44)Age, sex, period134
    Kono et al.491.37 (0.50–2.98)Age, sex, period141
    Spinelli et al.506.19 (1.27–18.11)1.11 (0.13–4.01)Age, sex39
    Gallagher et al.930.56 (0.01–3.10)0.99 (0.12–3.59)0.58 (0.07–2.11)1.69 (0.94–2.78)Age, sex288
    Shibata et al.680.78 (0.19–3.10)1.16 (0.37–3.60)Age, sex5
    Lamba et al.511.38 (1.08–1.75)1.11 (1.01–1.24)1.21 (1.06–1.39)1.1 (1.02–1.2)Age, sex, race, region9495
Cohort studies
    Guidotti et al.573.73 (2.53–5.51)AgeNot given
    Teta et al.521.29 (0.81–1.95)0.69 (0.14–2.02)1.20 (0.66–2.02)Age, sex, calendar year688
    Gubéran et al.532.5 (0.35–17.75)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.20–9.94)1.3 (0.49–3.46)2.5 (0.95–6.58)1.1 (0.53–2.30)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.61 (0.81–3.22)1.54 (1.05–2.25)0.90 (0.45–1.89)0.99 (0.63–1.55)Age, sex, period911
    Andersen et al.55 (Finland)1.08 (0.35–3.35)0.97 (0.50–1.86)1.12 (0.53–2.35)0.99 (0.54–1.85)Age, sex, period549
    Andersen et al.55 (Norway)1.16 (0.70–1.93)1.16 (0.48–2.79)1.41 (0.73–2.71)Age, sex, period492
    Czene et al.560.78 (0.50–1.20)0.93 (0.76–1.14)1.25 (0.95–1.65)1.00 (0.81–1.23)Age, sex, period3901
RR (95% CI)
ReferencesHodgkin's diseaseNon-Hodgkin's lymphomaMultiple myelomaLeukemiaType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.27 (0.37–3.45)6.41 (0.63–34.13)1.16 (0.03–7.78)HAge, sex139/719
    Flodin et al.583.3 (0.24–45.7)PNot given131/431
    Persson et al.592.7 (0.19–38.28)2.2 (0.17–28.71)HAge, sex, farming, fresh wood54/275
    Eriksson and Karlsson600.67 (0.13–3.64)PAge, sex, county275/275
    Pottern et al.610.7 (0.0–5.8)PAge, sex607/2596
    Cote et al.100.65 (0.3–1.3)PAge, sex, race2153/8612
    Blair et al.622.7 (0.9–8.7)PAge, sex, race, smoking, death year, others622/1245
    Herrinton et al.631.32 (0.65– 2.65)PAge, sex, race, location, education level360/933
    Figgs et al.641.7 (1.1–2.6)PAge, sex, race12 148/60 740
    Mele et al.652.25 (0.77–6.58)HAge, sex, region, education517/1161
    Miligi et al.662.1 (0.7–6.5)1.9 (0.7–5.8)11.1 (1.8–67.0)2.2 (0.7–7.1)PAge, sex, smoking, education1170/828
    Seniori Costantini et al.670.6 (0.2–1.6)2.2 (0.7–6.9))1.0 (0.3–3.2)PAge, sex2737/1779
    ‘t Mannetje et al.111.09 (0.27–4.35)PAge, sex, smoking, ethnicity, occupational status291/471
Proportionate mortality studies
    Alderson901.11 (0.36–3.44)Age, sex, period134
    Kono et al.491.37 (0.50–2.98)Age, sex, period141
    Spinelli et al.506.19 (1.27–18.11)1.11 (0.13–4.01)Age, sex39
    Gallagher et al.930.56 (0.01–3.10)0.99 (0.12–3.59)0.58 (0.07–2.11)1.69 (0.94–2.78)Age, sex288
    Shibata et al.680.78 (0.19–3.10)1.16 (0.37–3.60)Age, sex5
    Lamba et al.511.38 (1.08–1.75)1.11 (1.01–1.24)1.21 (1.06–1.39)1.1 (1.02–1.2)Age, sex, race, region9495
Cohort studies
    Guidotti et al.573.73 (2.53–5.51)AgeNot given
    Teta et al.521.29 (0.81–1.95)0.69 (0.14–2.02)1.20 (0.66–2.02)Age, sex, calendar year688
    Gubéran et al.532.5 (0.35–17.75)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.20–9.94)1.3 (0.49–3.46)2.5 (0.95–6.58)1.1 (0.53–2.30)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.61 (0.81–3.22)1.54 (1.05–2.25)0.90 (0.45–1.89)0.99 (0.63–1.55)Age, sex, period911
    Andersen et al.55 (Finland)1.08 (0.35–3.35)0.97 (0.50–1.86)1.12 (0.53–2.35)0.99 (0.54–1.85)Age, sex, period549
    Andersen et al.55 (Norway)1.16 (0.70–1.93)1.16 (0.48–2.79)1.41 (0.73–2.71)Age, sex, period492
    Czene et al.560.78 (0.50–1.20)0.93 (0.76–1.14)1.25 (0.95–1.65)1.00 (0.81–1.23)Age, sex, period3901

H = hospital controls; P = population controls.

Table 3

Study-specific RRs and 95% CI of urinary tract cancers among hairdressers and related occupations

RR (95% CI)
ReferencesBladderKidneyProstateType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Wynder et al.695.07 (0.56–240.48)HAge, sex300/300
    Anthony and Thomas704.90 (0.58–222.42)HAge, sex, residence, smoking1030/741
    Cole et al.710.64 (0.27–1.54)PAge, sex461/485
    Decoufle431.28 (0.03–8.64)HAge, sex21/719
    Glashan and Cartwright720.9 (0.3–3.2)HAge, sex, year of diagnosis991/1338
    Schoenberg et al.731.27 (0.59–2.73)PAge, sex, smoking706/1392
    Vineis and Magnani740.9 (0.4–2.3)HAge, sex512/596
    Morrison et al.751.6 (0.38–6.67)PAge, sex, smoking430/397
    Baxter and McDowall762.0 (0.75–5.33)PAge, sex, borough54/108
    Risch et al.770.66 (0.19–2.23)PAge, sex, residence, smoking826/792
    Jensen et al.783.0 (0.3–33.0)HAge, sex, hospital, smoking96/288
    Silverman et al.791.3 (0.8–2.3)PAge, residence, smoking, education, source of drinking water2100/3874
    Silverman et al.801.4 (0.7–2.9)PAge, smoking, residence, education, source drinking water652/1266
    Kunze et al.811.67 (0.61–4.54)HAge, sex, smoking, beer intake, family history675/675
    Cordier et al.822.21 (0.41–11.94)HAge, sex, ethnicity, residence, smoking765/765
    Siematycki et al.831.0 (0.3–2.9)H/PAge, sex, ethnicity, smoking, economic status, coffee484/1879
    Teschke et al.843.2 (0.2–176)PAge, sex, smoking105/159
    Sorahan et al.851.75 (0.94–3.28)PAge, sex, smoking1427/2199
    Gago Dominguez et al.861.5 (0.7–3.2)PAge, sex, ethnicity, neighbourhood, smoking1514/1514
    Zheng et al.871.8 (0.4–8.0)PAge, sex, smoking, first-degree relative with bladder cancer1452/2434
    Gaertner et al.881.43 (0.67– 3.01)PAge, sex, province, smoking, diet887/2847
    Dryson et al.94.02 (1.05–15.36)PAge, sex, smoking, ethnicity, occupational status213/471
    Golka et al.894.90 (0.85–28.39)HAge, sex, smoking156/336
Proportionate mortality studies
    Alderson901.25 (0.60–2.62)Age, sex, calendar period134
    Dubrow and Wegman911.16 (0.43–3.08)Age, sex4
    Pearce and Howard9212.94 (1.45–46.7)Age, sex, social class7
    Gallagher et al.931.34 (0.69–2.34)1.03 (0.67–1.50)Age, sex288
    Lamba et al.511.16 (1.01–1.34)1.02 (0.87–1.18)0.90 (0.81–.00)Age, sex, race, region9495
Cohort studies
    Dunham et al.942.76 (1.04–7.35)Not given4
    Teta et al.521.36 (0.74–2.27)1.29 (0.62–2.37)Age, sex, calendar year688
    Gubéran et al.531.94 (1.13–3.34)1.77 (0.57–5.53)1.97 (1.12–3.47)Age, sex, matrimonial status65
    Hrubec et al.950.7 (0.23–2.17)1.0 (0.25–4.00)1.5 (0.98–2.3)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.51 (1.21–1.89)1.44 (1.00–2.07)1.03 (0.77–1.38)Age, sex, period911
    Andersen et al.55 (Finland)1.39 (0.75–2.59)0.89 (0.5–1.56)1.56 (0.65–3.75)Age, sex, period549
    Andersen et al.55 (Norway)1.57 (1.11–2.20)1.23 (0.71–2.12)0.86 (0.60–1.23)Age, sex, period492
    Czene et al.561.16 (0.96–1.39)1.17 (0.97–1.41)0.94 (0.82–1.08)Age, sex, period3901
RR (95% CI)
ReferencesBladderKidneyProstateType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Wynder et al.695.07 (0.56–240.48)HAge, sex300/300
    Anthony and Thomas704.90 (0.58–222.42)HAge, sex, residence, smoking1030/741
    Cole et al.710.64 (0.27–1.54)PAge, sex461/485
    Decoufle431.28 (0.03–8.64)HAge, sex21/719
    Glashan and Cartwright720.9 (0.3–3.2)HAge, sex, year of diagnosis991/1338
    Schoenberg et al.731.27 (0.59–2.73)PAge, sex, smoking706/1392
    Vineis and Magnani740.9 (0.4–2.3)HAge, sex512/596
    Morrison et al.751.6 (0.38–6.67)PAge, sex, smoking430/397
    Baxter and McDowall762.0 (0.75–5.33)PAge, sex, borough54/108
    Risch et al.770.66 (0.19–2.23)PAge, sex, residence, smoking826/792
    Jensen et al.783.0 (0.3–33.0)HAge, sex, hospital, smoking96/288
    Silverman et al.791.3 (0.8–2.3)PAge, residence, smoking, education, source of drinking water2100/3874
    Silverman et al.801.4 (0.7–2.9)PAge, smoking, residence, education, source drinking water652/1266
    Kunze et al.811.67 (0.61–4.54)HAge, sex, smoking, beer intake, family history675/675
    Cordier et al.822.21 (0.41–11.94)HAge, sex, ethnicity, residence, smoking765/765
    Siematycki et al.831.0 (0.3–2.9)H/PAge, sex, ethnicity, smoking, economic status, coffee484/1879
    Teschke et al.843.2 (0.2–176)PAge, sex, smoking105/159
    Sorahan et al.851.75 (0.94–3.28)PAge, sex, smoking1427/2199
    Gago Dominguez et al.861.5 (0.7–3.2)PAge, sex, ethnicity, neighbourhood, smoking1514/1514
    Zheng et al.871.8 (0.4–8.0)PAge, sex, smoking, first-degree relative with bladder cancer1452/2434
    Gaertner et al.881.43 (0.67– 3.01)PAge, sex, province, smoking, diet887/2847
    Dryson et al.94.02 (1.05–15.36)PAge, sex, smoking, ethnicity, occupational status213/471
    Golka et al.894.90 (0.85–28.39)HAge, sex, smoking156/336
Proportionate mortality studies
    Alderson901.25 (0.60–2.62)Age, sex, calendar period134
    Dubrow and Wegman911.16 (0.43–3.08)Age, sex4
    Pearce and Howard9212.94 (1.45–46.7)Age, sex, social class7
    Gallagher et al.931.34 (0.69–2.34)1.03 (0.67–1.50)Age, sex288
    Lamba et al.511.16 (1.01–1.34)1.02 (0.87–1.18)0.90 (0.81–.00)Age, sex, race, region9495
Cohort studies
    Dunham et al.942.76 (1.04–7.35)Not given4
    Teta et al.521.36 (0.74–2.27)1.29 (0.62–2.37)Age, sex, calendar year688
    Gubéran et al.531.94 (1.13–3.34)1.77 (0.57–5.53)1.97 (1.12–3.47)Age, sex, matrimonial status65
    Hrubec et al.950.7 (0.23–2.17)1.0 (0.25–4.00)1.5 (0.98–2.3)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.51 (1.21–1.89)1.44 (1.00–2.07)1.03 (0.77–1.38)Age, sex, period911
    Andersen et al.55 (Finland)1.39 (0.75–2.59)0.89 (0.5–1.56)1.56 (0.65–3.75)Age, sex, period549
    Andersen et al.55 (Norway)1.57 (1.11–2.20)1.23 (0.71–2.12)0.86 (0.60–1.23)Age, sex, period492
    Czene et al.561.16 (0.96–1.39)1.17 (0.97–1.41)0.94 (0.82–1.08)Age, sex, period3901

H = hospital controls; P = population controls.

Table 3

Study-specific RRs and 95% CI of urinary tract cancers among hairdressers and related occupations

RR (95% CI)
ReferencesBladderKidneyProstateType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Wynder et al.695.07 (0.56–240.48)HAge, sex300/300
    Anthony and Thomas704.90 (0.58–222.42)HAge, sex, residence, smoking1030/741
    Cole et al.710.64 (0.27–1.54)PAge, sex461/485
    Decoufle431.28 (0.03–8.64)HAge, sex21/719
    Glashan and Cartwright720.9 (0.3–3.2)HAge, sex, year of diagnosis991/1338
    Schoenberg et al.731.27 (0.59–2.73)PAge, sex, smoking706/1392
    Vineis and Magnani740.9 (0.4–2.3)HAge, sex512/596
    Morrison et al.751.6 (0.38–6.67)PAge, sex, smoking430/397
    Baxter and McDowall762.0 (0.75–5.33)PAge, sex, borough54/108
    Risch et al.770.66 (0.19–2.23)PAge, sex, residence, smoking826/792
    Jensen et al.783.0 (0.3–33.0)HAge, sex, hospital, smoking96/288
    Silverman et al.791.3 (0.8–2.3)PAge, residence, smoking, education, source of drinking water2100/3874
    Silverman et al.801.4 (0.7–2.9)PAge, smoking, residence, education, source drinking water652/1266
    Kunze et al.811.67 (0.61–4.54)HAge, sex, smoking, beer intake, family history675/675
    Cordier et al.822.21 (0.41–11.94)HAge, sex, ethnicity, residence, smoking765/765
    Siematycki et al.831.0 (0.3–2.9)H/PAge, sex, ethnicity, smoking, economic status, coffee484/1879
    Teschke et al.843.2 (0.2–176)PAge, sex, smoking105/159
    Sorahan et al.851.75 (0.94–3.28)PAge, sex, smoking1427/2199
    Gago Dominguez et al.861.5 (0.7–3.2)PAge, sex, ethnicity, neighbourhood, smoking1514/1514
    Zheng et al.871.8 (0.4–8.0)PAge, sex, smoking, first-degree relative with bladder cancer1452/2434
    Gaertner et al.881.43 (0.67– 3.01)PAge, sex, province, smoking, diet887/2847
    Dryson et al.94.02 (1.05–15.36)PAge, sex, smoking, ethnicity, occupational status213/471
    Golka et al.894.90 (0.85–28.39)HAge, sex, smoking156/336
Proportionate mortality studies
    Alderson901.25 (0.60–2.62)Age, sex, calendar period134
    Dubrow and Wegman911.16 (0.43–3.08)Age, sex4
    Pearce and Howard9212.94 (1.45–46.7)Age, sex, social class7
    Gallagher et al.931.34 (0.69–2.34)1.03 (0.67–1.50)Age, sex288
    Lamba et al.511.16 (1.01–1.34)1.02 (0.87–1.18)0.90 (0.81–.00)Age, sex, race, region9495
Cohort studies
    Dunham et al.942.76 (1.04–7.35)Not given4
    Teta et al.521.36 (0.74–2.27)1.29 (0.62–2.37)Age, sex, calendar year688
    Gubéran et al.531.94 (1.13–3.34)1.77 (0.57–5.53)1.97 (1.12–3.47)Age, sex, matrimonial status65
    Hrubec et al.950.7 (0.23–2.17)1.0 (0.25–4.00)1.5 (0.98–2.3)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.51 (1.21–1.89)1.44 (1.00–2.07)1.03 (0.77–1.38)Age, sex, period911
    Andersen et al.55 (Finland)1.39 (0.75–2.59)0.89 (0.5–1.56)1.56 (0.65–3.75)Age, sex, period549
    Andersen et al.55 (Norway)1.57 (1.11–2.20)1.23 (0.71–2.12)0.86 (0.60–1.23)Age, sex, period492
    Czene et al.561.16 (0.96–1.39)1.17 (0.97–1.41)0.94 (0.82–1.08)Age, sex, period3901
RR (95% CI)
ReferencesBladderKidneyProstateType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Wynder et al.695.07 (0.56–240.48)HAge, sex300/300
    Anthony and Thomas704.90 (0.58–222.42)HAge, sex, residence, smoking1030/741
    Cole et al.710.64 (0.27–1.54)PAge, sex461/485
    Decoufle431.28 (0.03–8.64)HAge, sex21/719
    Glashan and Cartwright720.9 (0.3–3.2)HAge, sex, year of diagnosis991/1338
    Schoenberg et al.731.27 (0.59–2.73)PAge, sex, smoking706/1392
    Vineis and Magnani740.9 (0.4–2.3)HAge, sex512/596
    Morrison et al.751.6 (0.38–6.67)PAge, sex, smoking430/397
    Baxter and McDowall762.0 (0.75–5.33)PAge, sex, borough54/108
    Risch et al.770.66 (0.19–2.23)PAge, sex, residence, smoking826/792
    Jensen et al.783.0 (0.3–33.0)HAge, sex, hospital, smoking96/288
    Silverman et al.791.3 (0.8–2.3)PAge, residence, smoking, education, source of drinking water2100/3874
    Silverman et al.801.4 (0.7–2.9)PAge, smoking, residence, education, source drinking water652/1266
    Kunze et al.811.67 (0.61–4.54)HAge, sex, smoking, beer intake, family history675/675
    Cordier et al.822.21 (0.41–11.94)HAge, sex, ethnicity, residence, smoking765/765
    Siematycki et al.831.0 (0.3–2.9)H/PAge, sex, ethnicity, smoking, economic status, coffee484/1879
    Teschke et al.843.2 (0.2–176)PAge, sex, smoking105/159
    Sorahan et al.851.75 (0.94–3.28)PAge, sex, smoking1427/2199
    Gago Dominguez et al.861.5 (0.7–3.2)PAge, sex, ethnicity, neighbourhood, smoking1514/1514
    Zheng et al.871.8 (0.4–8.0)PAge, sex, smoking, first-degree relative with bladder cancer1452/2434
    Gaertner et al.881.43 (0.67– 3.01)PAge, sex, province, smoking, diet887/2847
    Dryson et al.94.02 (1.05–15.36)PAge, sex, smoking, ethnicity, occupational status213/471
    Golka et al.894.90 (0.85–28.39)HAge, sex, smoking156/336
Proportionate mortality studies
    Alderson901.25 (0.60–2.62)Age, sex, calendar period134
    Dubrow and Wegman911.16 (0.43–3.08)Age, sex4
    Pearce and Howard9212.94 (1.45–46.7)Age, sex, social class7
    Gallagher et al.931.34 (0.69–2.34)1.03 (0.67–1.50)Age, sex288
    Lamba et al.511.16 (1.01–1.34)1.02 (0.87–1.18)0.90 (0.81–.00)Age, sex, race, region9495
Cohort studies
    Dunham et al.942.76 (1.04–7.35)Not given4
    Teta et al.521.36 (0.74–2.27)1.29 (0.62–2.37)Age, sex, calendar year688
    Gubéran et al.531.94 (1.13–3.34)1.77 (0.57–5.53)1.97 (1.12–3.47)Age, sex, matrimonial status65
    Hrubec et al.950.7 (0.23–2.17)1.0 (0.25–4.00)1.5 (0.98–2.3)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)1.51 (1.21–1.89)1.44 (1.00–2.07)1.03 (0.77–1.38)Age, sex, period911
    Andersen et al.55 (Finland)1.39 (0.75–2.59)0.89 (0.5–1.56)1.56 (0.65–3.75)Age, sex, period549
    Andersen et al.55 (Norway)1.57 (1.11–2.20)1.23 (0.71–2.12)0.86 (0.60–1.23)Age, sex, period492
    Czene et al.561.16 (0.96–1.39)1.17 (0.97–1.41)0.94 (0.82–1.08)Age, sex, period3901

H = hospital controls; P = population controls.

Table 4

Study-specific RRs and 95% CIs of digestive cancers among hairdressers and related occupations

RR (95% CI)
ReferencesOesophagusStomachColon–RectumLiverPancreasType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.22 (0.03–8.18)1.02 (0.34–2.59)HAge, sex43/719
Proportionate mortality or morbidity studies
    Menck et al.481.03 (0.58–1.81)Age, sex135
    Alderson901.47 (0.61–3.53)Age, sex, calendar period134
    Kono et al.490.63 (0.02–3.53)1.34 (1.02–1.72)1.69 (0.77–3.21)0.72 (0.29–1.48)0.24 (0.01–1.36)Age, sex, period141
    Gallagher et al.931.1 (0.71–1.61)1.06 (0.76–1.48)0.70 (0.08–2.53)0.60 (0.27–1.15)Age, sex288
    Lamba et al.510.81 (0.67–0.98)1.16 (1.02–1.31)1.05 (0.99–1.12)1.50 (1.36–1.64)1.08 (0.99–1.19)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.91 (0.13–6.46)0.67 (0.22–2.09)1.25 (0.72–2.15)1.37 (0.44–4.24)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.35–6.00)1.4 (0.65–2.99)0.8 (0.41–1.57)1.2 (0.30–4.80)1.0 (0.42–2.40)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)0.52 (0.20–1.38)0.87 (0.59–1.27)1.11 (0.93–1.33)1.08 (0.49–2.40)0.76 (0.49–1.18)Age, sex, period911
    Andersen et al.55 (Finland)0.97 (0.36–2.59)1.03 (0.72–1.48)1.18 (0.89–1.57)1.32 (0.59–2.94)1.60 (1.06–2.43)Age, sex, period549
    Andersen et al.55 (Norway)1.18 (0.44–3.15)0.66 (0.40–1.07)1.09 (0.87–1.37)2.07 (0.67–6.42)1.04 (0.64–1.71)Age, sex, period492
    Andersen et al.55 (Sweden)1.17 (0.38–3.63)Age, sex, period1528
    Czene et al.561.20 (0.63–2.05)0.89 (0.73–1.09)1.11 (1.00–1.23)1.00 (0.81–1.24)1.30 (1.08–1.57)Age, sex, period3901
RR (95% CI)
ReferencesOesophagusStomachColon–RectumLiverPancreasType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.22 (0.03–8.18)1.02 (0.34–2.59)HAge, sex43/719
Proportionate mortality or morbidity studies
    Menck et al.481.03 (0.58–1.81)Age, sex135
    Alderson901.47 (0.61–3.53)Age, sex, calendar period134
    Kono et al.490.63 (0.02–3.53)1.34 (1.02–1.72)1.69 (0.77–3.21)0.72 (0.29–1.48)0.24 (0.01–1.36)Age, sex, period141
    Gallagher et al.931.1 (0.71–1.61)1.06 (0.76–1.48)0.70 (0.08–2.53)0.60 (0.27–1.15)Age, sex288
    Lamba et al.510.81 (0.67–0.98)1.16 (1.02–1.31)1.05 (0.99–1.12)1.50 (1.36–1.64)1.08 (0.99–1.19)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.91 (0.13–6.46)0.67 (0.22–2.09)1.25 (0.72–2.15)1.37 (0.44–4.24)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.35–6.00)1.4 (0.65–2.99)0.8 (0.41–1.57)1.2 (0.30–4.80)1.0 (0.42–2.40)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)0.52 (0.20–1.38)0.87 (0.59–1.27)1.11 (0.93–1.33)1.08 (0.49–2.40)0.76 (0.49–1.18)Age, sex, period911
    Andersen et al.55 (Finland)0.97 (0.36–2.59)1.03 (0.72–1.48)1.18 (0.89–1.57)1.32 (0.59–2.94)1.60 (1.06–2.43)Age, sex, period549
    Andersen et al.55 (Norway)1.18 (0.44–3.15)0.66 (0.40–1.07)1.09 (0.87–1.37)2.07 (0.67–6.42)1.04 (0.64–1.71)Age, sex, period492
    Andersen et al.55 (Sweden)1.17 (0.38–3.63)Age, sex, period1528
    Czene et al.561.20 (0.63–2.05)0.89 (0.73–1.09)1.11 (1.00–1.23)1.00 (0.81–1.24)1.30 (1.08–1.57)Age, sex, period3901

H = hospital controls.

Table 4

Study-specific RRs and 95% CIs of digestive cancers among hairdressers and related occupations

RR (95% CI)
ReferencesOesophagusStomachColon–RectumLiverPancreasType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.22 (0.03–8.18)1.02 (0.34–2.59)HAge, sex43/719
Proportionate mortality or morbidity studies
    Menck et al.481.03 (0.58–1.81)Age, sex135
    Alderson901.47 (0.61–3.53)Age, sex, calendar period134
    Kono et al.490.63 (0.02–3.53)1.34 (1.02–1.72)1.69 (0.77–3.21)0.72 (0.29–1.48)0.24 (0.01–1.36)Age, sex, period141
    Gallagher et al.931.1 (0.71–1.61)1.06 (0.76–1.48)0.70 (0.08–2.53)0.60 (0.27–1.15)Age, sex288
    Lamba et al.510.81 (0.67–0.98)1.16 (1.02–1.31)1.05 (0.99–1.12)1.50 (1.36–1.64)1.08 (0.99–1.19)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.91 (0.13–6.46)0.67 (0.22–2.09)1.25 (0.72–2.15)1.37 (0.44–4.24)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.35–6.00)1.4 (0.65–2.99)0.8 (0.41–1.57)1.2 (0.30–4.80)1.0 (0.42–2.40)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)0.52 (0.20–1.38)0.87 (0.59–1.27)1.11 (0.93–1.33)1.08 (0.49–2.40)0.76 (0.49–1.18)Age, sex, period911
    Andersen et al.55 (Finland)0.97 (0.36–2.59)1.03 (0.72–1.48)1.18 (0.89–1.57)1.32 (0.59–2.94)1.60 (1.06–2.43)Age, sex, period549
    Andersen et al.55 (Norway)1.18 (0.44–3.15)0.66 (0.40–1.07)1.09 (0.87–1.37)2.07 (0.67–6.42)1.04 (0.64–1.71)Age, sex, period492
    Andersen et al.55 (Sweden)1.17 (0.38–3.63)Age, sex, period1528
    Czene et al.561.20 (0.63–2.05)0.89 (0.73–1.09)1.11 (1.00–1.23)1.00 (0.81–1.24)1.30 (1.08–1.57)Age, sex, period3901
RR (95% CI)
ReferencesOesophagusStomachColon–RectumLiverPancreasType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.22 (0.03–8.18)1.02 (0.34–2.59)HAge, sex43/719
Proportionate mortality or morbidity studies
    Menck et al.481.03 (0.58–1.81)Age, sex135
    Alderson901.47 (0.61–3.53)Age, sex, calendar period134
    Kono et al.490.63 (0.02–3.53)1.34 (1.02–1.72)1.69 (0.77–3.21)0.72 (0.29–1.48)0.24 (0.01–1.36)Age, sex, period141
    Gallagher et al.931.1 (0.71–1.61)1.06 (0.76–1.48)0.70 (0.08–2.53)0.60 (0.27–1.15)Age, sex288
    Lamba et al.510.81 (0.67–0.98)1.16 (1.02–1.31)1.05 (0.99–1.12)1.50 (1.36–1.64)1.08 (0.99–1.19)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.91 (0.13–6.46)0.67 (0.22–2.09)1.25 (0.72–2.15)1.37 (0.44–4.24)Age, sex, matrimonial status65
    Hrubec et al.951.4 (0.35–6.00)1.4 (0.65–2.99)0.8 (0.41–1.57)1.2 (0.30–4.80)1.0 (0.42–2.40)Age, sex, calendar time, smoking110
    Andersen et al.55 (Denmark)0.52 (0.20–1.38)0.87 (0.59–1.27)1.11 (0.93–1.33)1.08 (0.49–2.40)0.76 (0.49–1.18)Age, sex, period911
    Andersen et al.55 (Finland)0.97 (0.36–2.59)1.03 (0.72–1.48)1.18 (0.89–1.57)1.32 (0.59–2.94)1.60 (1.06–2.43)Age, sex, period549
    Andersen et al.55 (Norway)1.18 (0.44–3.15)0.66 (0.40–1.07)1.09 (0.87–1.37)2.07 (0.67–6.42)1.04 (0.64–1.71)Age, sex, period492
    Andersen et al.55 (Sweden)1.17 (0.38–3.63)Age, sex, period1528
    Czene et al.561.20 (0.63–2.05)0.89 (0.73–1.09)1.11 (1.00–1.23)1.00 (0.81–1.24)1.30 (1.08–1.57)Age, sex, period3901

H = hospital controls.

Table 5

Study-specific RRs and 95% CIs of respiratory cancers among hairdressers and related occupations

RR (95% CI)
ReferencesLungLarynxType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.03 (0.02–6.77)3.20 (0.07–25.32)HAge, sex35/719
    Osorio et al.960.95 (0.64–1.41)PAge, sex, race, smoking50/56
    Jahn et al.971.9 (0.79–4.32)PAge, sex, smoking, region686/712
    Bofetta et al.982.33 (1.00–5.40)PAge, sex, area, smoking, alcohol1010/2176
Proportionate mortality or morbidity studies
    Menck et al.481.76 (1.14–2.73)Age, sex135
    Garfinkel et al.996.92 (2.87–16.72)Age, sex, race25
    Alderson901.02 (0.78–1.34)0.77 (0.11–5.47)Age, sex, period134
    Kono et al.491.21 (0.56–2.31)Age, sex, period141
    Pearce and Howard922.54 (0.82–5.93)Age, sex, social class7
    Gallagher et al.930.91 (0.70–1.18)0.41 (0.01–2.31)Age, sex288
    Lamba et al.511.13 (1.08–1.18)0.94 (0.71–1.25)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.99 (0.55–1.79)2.31 (0.75–7.16)Age, sex, matrimonial status65
    Hrubec et al.951.6 (1.22–2.20)1.5 (0.21–10.65)Age, sex, calendar time, smoking110
    Leigh1001.38 (1.11–1.71)Age, sex, smoking, education96
    Andersen et al.55 (Denmark)1.14 (0.98–1.33)1.05 (0.56–1.95)Age, sex, period911
    Andersen et al.55 (Finland)1.33 (0.94–1.88)3.05 (1.15–8.13)Age, sex, period549
    Andersen et al.55 (Norway)1.22 (0.93–1.61)3.2 (1.40–7.34)Age, sex, period492
    Andersen et al.55 (Sweden)1.19 (0.64–2.21)Age, sex, period1528
    Czene et al.561.36 (1.22–1.53)Age, sex, period3901
    Ji and Hemminki1011.42 (1.20–1.66)1.78 (1.10–2.62)Age, sex, period, socio-economic status334
RR (95% CI)
ReferencesLungLarynxType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.03 (0.02–6.77)3.20 (0.07–25.32)HAge, sex35/719
    Osorio et al.960.95 (0.64–1.41)PAge, sex, race, smoking50/56
    Jahn et al.971.9 (0.79–4.32)PAge, sex, smoking, region686/712
    Bofetta et al.982.33 (1.00–5.40)PAge, sex, area, smoking, alcohol1010/2176
Proportionate mortality or morbidity studies
    Menck et al.481.76 (1.14–2.73)Age, sex135
    Garfinkel et al.996.92 (2.87–16.72)Age, sex, race25
    Alderson901.02 (0.78–1.34)0.77 (0.11–5.47)Age, sex, period134
    Kono et al.491.21 (0.56–2.31)Age, sex, period141
    Pearce and Howard922.54 (0.82–5.93)Age, sex, social class7
    Gallagher et al.930.91 (0.70–1.18)0.41 (0.01–2.31)Age, sex288
    Lamba et al.511.13 (1.08–1.18)0.94 (0.71–1.25)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.99 (0.55–1.79)2.31 (0.75–7.16)Age, sex, matrimonial status65
    Hrubec et al.951.6 (1.22–2.20)1.5 (0.21–10.65)Age, sex, calendar time, smoking110
    Leigh1001.38 (1.11–1.71)Age, sex, smoking, education96
    Andersen et al.55 (Denmark)1.14 (0.98–1.33)1.05 (0.56–1.95)Age, sex, period911
    Andersen et al.55 (Finland)1.33 (0.94–1.88)3.05 (1.15–8.13)Age, sex, period549
    Andersen et al.55 (Norway)1.22 (0.93–1.61)3.2 (1.40–7.34)Age, sex, period492
    Andersen et al.55 (Sweden)1.19 (0.64–2.21)Age, sex, period1528
    Czene et al.561.36 (1.22–1.53)Age, sex, period3901
    Ji and Hemminki1011.42 (1.20–1.66)1.78 (1.10–2.62)Age, sex, period, socio-economic status334

H = hospital controls; P = population controls.

Table 5

Study-specific RRs and 95% CIs of respiratory cancers among hairdressers and related occupations

RR (95% CI)
ReferencesLungLarynxType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.03 (0.02–6.77)3.20 (0.07–25.32)HAge, sex35/719
    Osorio et al.960.95 (0.64–1.41)PAge, sex, race, smoking50/56
    Jahn et al.971.9 (0.79–4.32)PAge, sex, smoking, region686/712
    Bofetta et al.982.33 (1.00–5.40)PAge, sex, area, smoking, alcohol1010/2176
Proportionate mortality or morbidity studies
    Menck et al.481.76 (1.14–2.73)Age, sex135
    Garfinkel et al.996.92 (2.87–16.72)Age, sex, race25
    Alderson901.02 (0.78–1.34)0.77 (0.11–5.47)Age, sex, period134
    Kono et al.491.21 (0.56–2.31)Age, sex, period141
    Pearce and Howard922.54 (0.82–5.93)Age, sex, social class7
    Gallagher et al.930.91 (0.70–1.18)0.41 (0.01–2.31)Age, sex288
    Lamba et al.511.13 (1.08–1.18)0.94 (0.71–1.25)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.99 (0.55–1.79)2.31 (0.75–7.16)Age, sex, matrimonial status65
    Hrubec et al.951.6 (1.22–2.20)1.5 (0.21–10.65)Age, sex, calendar time, smoking110
    Leigh1001.38 (1.11–1.71)Age, sex, smoking, education96
    Andersen et al.55 (Denmark)1.14 (0.98–1.33)1.05 (0.56–1.95)Age, sex, period911
    Andersen et al.55 (Finland)1.33 (0.94–1.88)3.05 (1.15–8.13)Age, sex, period549
    Andersen et al.55 (Norway)1.22 (0.93–1.61)3.2 (1.40–7.34)Age, sex, period492
    Andersen et al.55 (Sweden)1.19 (0.64–2.21)Age, sex, period1528
    Czene et al.561.36 (1.22–1.53)Age, sex, period3901
    Ji and Hemminki1011.42 (1.20–1.66)1.78 (1.10–2.62)Age, sex, period, socio-economic status334
RR (95% CI)
ReferencesLungLarynxType of controlsAdjustment, matching and restriction factorsCases/controls or exposed cases
Case–control studies
    Decoufle431.03 (0.02–6.77)3.20 (0.07–25.32)HAge, sex35/719
    Osorio et al.960.95 (0.64–1.41)PAge, sex, race, smoking50/56
    Jahn et al.971.9 (0.79–4.32)PAge, sex, smoking, region686/712
    Bofetta et al.982.33 (1.00–5.40)PAge, sex, area, smoking, alcohol1010/2176
Proportionate mortality or morbidity studies
    Menck et al.481.76 (1.14–2.73)Age, sex135
    Garfinkel et al.996.92 (2.87–16.72)Age, sex, race25
    Alderson901.02 (0.78–1.34)0.77 (0.11–5.47)Age, sex, period134
    Kono et al.491.21 (0.56–2.31)Age, sex, period141
    Pearce and Howard922.54 (0.82–5.93)Age, sex, social class7
    Gallagher et al.930.91 (0.70–1.18)0.41 (0.01–2.31)Age, sex288
    Lamba et al.511.13 (1.08–1.18)0.94 (0.71–1.25)Age, sex, race, region9495
Cohort studies
    Gubéran et al.530.99 (0.55–1.79)2.31 (0.75–7.16)Age, sex, matrimonial status65
    Hrubec et al.951.6 (1.22–2.20)1.5 (0.21–10.65)Age, sex, calendar time, smoking110
    Leigh1001.38 (1.11–1.71)Age, sex, smoking, education96
    Andersen et al.55 (Denmark)1.14 (0.98–1.33)1.05 (0.56–1.95)Age, sex, period911
    Andersen et al.55 (Finland)1.33 (0.94–1.88)3.05 (1.15–8.13)Age, sex, period549
    Andersen et al.55 (Norway)1.22 (0.93–1.61)3.2 (1.40–7.34)Age, sex, period492
    Andersen et al.55 (Sweden)1.19 (0.64–2.21)Age, sex, period1528
    Czene et al.561.36 (1.22–1.53)Age, sex, period3901
    Ji and Hemminki1011.42 (1.20–1.66)1.78 (1.10–2.62)Age, sex, period, socio-economic status334

H = hospital controls; P = population controls.

One of the most challenging tasks in this meta-analysis was to trace the same cohort in different publications and thus, to avoid including several times the same study population.

Among the studies that could have been relevant to our meta-analysis but were finally excluded, several were discarded because either they were an update of the same cohort or duplicate publications of the same results.22–36 Four studies were excluded because the corresponding CI of the effect measure was either not available,37,38 or was undefined as no case of cancer was observed among hairdressers.39,40 We also discarded two articles that presented pooled results from previously published studies, and we included the original individual studies instead.41,42 When grouped all together, the 247 studies yielded a random effects pooled RR of 1.15 (95% CI 1.12–1.19).

Gynaecologic cancers

We retrieved 41 studies on gynaecologic cancers;8,43–56 more specifically: 16 studies on breast cancer, 8 studies on cervix uteri cancer, 7 studies on corpus uteri cancer and 10 studies on ovary cancer, published between 1977 and 2003 (Table 1). Compared with other occupations, hairdressers globally present a slight increase in the risk of breast, cervix uteri and ovary, which varies between 6% for breast cancer and 12% for ovary cancer (Table 6). Except for corpus uteri cancer, heterogeneity is low, and thus, fixed and random effects pooled estimates are very similar if not identical. Stratifying the analysis by design, we found that the effect subsided among cohort studies for breast cancer. The contrary occurred in other gynaecologic cancers, for which the effect increased among cohort studies and decreased among case–control studies.

Table 6

Pooled RRs and 95% CIs of gynaecologic cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Breast
    All studies161.06 (1.02–1.10)1.06 (1.02–1.10)0.000.54
    Cohort studies71.02 (0.97–1.08)1.02 (0.97–1.08)0.000.45
    Case–control studies91.10 (1.05–1.16)1.10 (1.05–1.16)0.000.84
    Incidence-only studies121.03 (0.98–1.08)1.03 (0.98–1.08)0.000.95
Cervix uteri
    All studies81.11 (1.01–1.21)1.07 (0.92–1.25)0.490.09
    Cohort studies61.17 (1.05–1.31)1.11 (0.92–1.33)0.520.12
    Case–control studies20.97 (0.82–1.14)0.97 (0.82–1.14)0.000.72
    Incidence-only studies71.17 (1.05–1.31)1.11 (0.95–1.31)0.400.19
Corpus uteri
    All studies71.05 (0.95–1.14)1.11 (0.95–1.29)0.590.03
    Cohort studies51.10 (0.98–1.23)1.15 (0.94–1.41)0.650.03
    Case–control studies20.95 (0.82–1.11)0.95 (0.82–1.11)0.000.32
    Incidence-only studies61.10 (0.99–1.2391.16 (0.96–1.41)0.580.04
Ovary
    All studies101.12 (1.04–1.21)1.20 (1.05–1.38)0.530.06
    Cohort studies61.17 (1.06–1.30)1.17 (1.06–1.30)0.000.45
    Case–control studies41.06 (0.95–1.18)1.57 (0.92–2.68)0.940.01
    Incidence-only studies61.20 (1.08–1.33)1.24 (1.06–1.46)0.480.12
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Breast
    All studies161.06 (1.02–1.10)1.06 (1.02–1.10)0.000.54
    Cohort studies71.02 (0.97–1.08)1.02 (0.97–1.08)0.000.45
    Case–control studies91.10 (1.05–1.16)1.10 (1.05–1.16)0.000.84
    Incidence-only studies121.03 (0.98–1.08)1.03 (0.98–1.08)0.000.95
Cervix uteri
    All studies81.11 (1.01–1.21)1.07 (0.92–1.25)0.490.09
    Cohort studies61.17 (1.05–1.31)1.11 (0.92–1.33)0.520.12
    Case–control studies20.97 (0.82–1.14)0.97 (0.82–1.14)0.000.72
    Incidence-only studies71.17 (1.05–1.31)1.11 (0.95–1.31)0.400.19
Corpus uteri
    All studies71.05 (0.95–1.14)1.11 (0.95–1.29)0.590.03
    Cohort studies51.10 (0.98–1.23)1.15 (0.94–1.41)0.650.03
    Case–control studies20.95 (0.82–1.11)0.95 (0.82–1.11)0.000.32
    Incidence-only studies61.10 (0.99–1.2391.16 (0.96–1.41)0.580.04
Ovary
    All studies101.12 (1.04–1.21)1.20 (1.05–1.38)0.530.06
    Cohort studies61.17 (1.06–1.30)1.17 (1.06–1.30)0.000.45
    Case–control studies41.06 (0.95–1.18)1.57 (0.92–2.68)0.940.01
    Incidence-only studies61.20 (1.08–1.33)1.24 (1.06–1.46)0.480.12

aProportion of total variance due to between-study variance.

Table 6

Pooled RRs and 95% CIs of gynaecologic cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Breast
    All studies161.06 (1.02–1.10)1.06 (1.02–1.10)0.000.54
    Cohort studies71.02 (0.97–1.08)1.02 (0.97–1.08)0.000.45
    Case–control studies91.10 (1.05–1.16)1.10 (1.05–1.16)0.000.84
    Incidence-only studies121.03 (0.98–1.08)1.03 (0.98–1.08)0.000.95
Cervix uteri
    All studies81.11 (1.01–1.21)1.07 (0.92–1.25)0.490.09
    Cohort studies61.17 (1.05–1.31)1.11 (0.92–1.33)0.520.12
    Case–control studies20.97 (0.82–1.14)0.97 (0.82–1.14)0.000.72
    Incidence-only studies71.17 (1.05–1.31)1.11 (0.95–1.31)0.400.19
Corpus uteri
    All studies71.05 (0.95–1.14)1.11 (0.95–1.29)0.590.03
    Cohort studies51.10 (0.98–1.23)1.15 (0.94–1.41)0.650.03
    Case–control studies20.95 (0.82–1.11)0.95 (0.82–1.11)0.000.32
    Incidence-only studies61.10 (0.99–1.2391.16 (0.96–1.41)0.580.04
Ovary
    All studies101.12 (1.04–1.21)1.20 (1.05–1.38)0.530.06
    Cohort studies61.17 (1.06–1.30)1.17 (1.06–1.30)0.000.45
    Case–control studies41.06 (0.95–1.18)1.57 (0.92–2.68)0.940.01
    Incidence-only studies61.20 (1.08–1.33)1.24 (1.06–1.46)0.480.12
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Breast
    All studies161.06 (1.02–1.10)1.06 (1.02–1.10)0.000.54
    Cohort studies71.02 (0.97–1.08)1.02 (0.97–1.08)0.000.45
    Case–control studies91.10 (1.05–1.16)1.10 (1.05–1.16)0.000.84
    Incidence-only studies121.03 (0.98–1.08)1.03 (0.98–1.08)0.000.95
Cervix uteri
    All studies81.11 (1.01–1.21)1.07 (0.92–1.25)0.490.09
    Cohort studies61.17 (1.05–1.31)1.11 (0.92–1.33)0.520.12
    Case–control studies20.97 (0.82–1.14)0.97 (0.82–1.14)0.000.72
    Incidence-only studies71.17 (1.05–1.31)1.11 (0.95–1.31)0.400.19
Corpus uteri
    All studies71.05 (0.95–1.14)1.11 (0.95–1.29)0.590.03
    Cohort studies51.10 (0.98–1.23)1.15 (0.94–1.41)0.650.03
    Case–control studies20.95 (0.82–1.11)0.95 (0.82–1.11)0.000.32
    Incidence-only studies61.10 (0.99–1.2391.16 (0.96–1.41)0.580.04
Ovary
    All studies101.12 (1.04–1.21)1.20 (1.05–1.38)0.530.06
    Cohort studies61.17 (1.06–1.30)1.17 (1.06–1.30)0.000.45
    Case–control studies41.06 (0.95–1.18)1.57 (0.92–2.68)0.940.01
    Incidence-only studies61.20 (1.08–1.33)1.24 (1.06–1.46)0.480.12

aProportion of total variance due to between-study variance.

The 12 incidence studies of breast cancer yielded a pooled RR of 1.03 (95% CI 0.98–1.08), lower than that of the four mortality studies, the result of which was 1.10 (95% CI 1.05–1.16). For ovary cancer, the incidence studies yielded a pooled RR of 1.20 (95% CI 1.08–1.33) and mortality studies a pooled RR of 1.04 (95% CI 0.93–1.16).

Restricting the analysis to those eight studies that fulfilled three or more quality criteria slightly attenuated the risk increase for breast cancer (RR 1.02; 95% CI 0.97–1.07), but did not alter the results for other gynaecologic cancers. The stratification by each item of the quality scale did not alter the results for any of the gynaecologic cancers.

Haematopoietic cancers

Table 2 presents the specific RRs for the 59 studies that dealt with haematopoietic cancers.

When all haematopoietic cancers were analysed together, we observed a risk increase (random effects pooled RR 1.26, 95% CI 1.14–1.38), which was remarkably uniform across design and gender categories. The 39 incidence studies showed a pooled RR that was only slightly higher than the pooled estimate of the 21 studies, the outcome of which was mortality by haematopoietic cancer (RR 1.28, 95% CI 1.10–1.50 for incidence studies and RR 1.22 95% CI 1.10–1.35 for mortality studies).

Separate analyses by cancer site show an increase in the risk for every anatomic site (Table 7). The increase is low for leukaemia and non-Hodgkin's lymphoma, moderate for Hodgkin's disease, and high for multiple myeloma. Heterogeneity of the study-specific RRs was low for Hodgkin's disease, non-Hodgkin's lymphoma and leukaemia (Ri between 0 and 0.16), moderate for all haematopoietic cancers grouped together (Ri = 0.46) and high for multiple myeloma (Ri = 0.75). The results of effect and heterogeneity did not change substantially after we stratified the studies by outcome (incidence/mortality), design, or gender. For every cancer site, those studies that adjusted for smoking showed higher pooled effects and less heterogeneity than those studies with incomplete adjustment. Stratifying by other quality scale variables did not produce any change in the results.

Table 7

Pooled RRs and 95% CIs of haematopoietic cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
All haematopoietic cancers
    All studies591.17 (1.11–1.22)1.26 (1.14–1.38)0.460.0001
    Cohort studies241.16 (1.06–1.28)1.23 (1.04–1.45)0.590.0001
    Case–control studies351.17 (1.10–1.23)1.26 (1.13–1.40)0.310.09
    Males only341.17 (1.08–1.26)1.30 (1.12–1.50)0.550.001
    Females only401.18 (1.12–1.25)1.26 (1.13–1.41)0.450.002
    Incidence-only studies391.19 (1.09–1.30)1.28 (1.10–1.50)0.490.0008
Hodgkin's disease
    All studies81.25 (1.03–1.51)1.22 (0.96–1.55)0.160.35
    Cohort studies40.98 (0.68–1.38)1.00 (0.69–1.44)0.090.36
    Case–control studies41.39 (1.10–1.75)1.39 (1.10–1.75)0.000.60
    Males only51.30 (0.99–1.69)1.30 (0.99–1.69)0.000.81
    Females only51.18 (0.89–1.56)1.14 (0.69–1.87)0.590.08
    Incidence-only studies61.06 (0.76–1.46)1.09 (0.77–1.56)0.080.37
Non-Hodgkin's lymphoma
    All studies161.10 (1.00–1.19)1.10 (1.00–1.19)0.000.53
    Cohort studies61.08 (0.93–1.26)1.13 (0.93–1.37)0.270.26
    Case–control studies101.10 (0.99–1.22)1.10 (0.99–1.22)0.000.57
    Males only91.04 (0.90–1.21)1.04 (0.90–1.21)0.000.69
    Females only91.14 (1.03–1.26)1.14 (1.03–1.27)0.010.42
    Incidence-only studies131.08 (0.94–1.24)1.11 (0.94–1.32)0.150.32
Multiple myeloma
    All studies191.38 (1.25–1.54)1.62 (1.22–2.14)0.750.0001
    Cohort studies81.94 (1.62–2.32)1.89 (1.20–2.97)0.800.0001
    Case–control studies111.30 (1.15–1.48)1.75 (1.17–2.63)0.790.01
    Males only81.65 (1.39–1.96)1.80 (1.15–2.81)0.790.0002
    Females only141.35 (1.17–1.55)1.97 (1.32–2.93)0.780.0001
    Incidence-only studies131.60 (1.33–1.92)1.57 (1.04–2.37)0.720.0002
Leukemia
    All studies161.11 (1.03–1.19)1.11 (1.03–1.19)0.000.92
    Cohort studies61.04 (0.89–1.22)1.04 (0.89–1.22)0.000.93
    Case–control studies101.13 (1.03–1.22)1.13 (1.03–1.22)0.000.75
    Males only111.06 (0.95–1.19)1.11 (0.96–1.28)0.140.35
    Females only111.15 (1.06–1.26)1.15 (1.06–1.26)0.000.90
    Incidence-only studies101.07 (0.92–1.25)1.07 (0.92–1.25)0.000.85
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
All haematopoietic cancers
    All studies591.17 (1.11–1.22)1.26 (1.14–1.38)0.460.0001
    Cohort studies241.16 (1.06–1.28)1.23 (1.04–1.45)0.590.0001
    Case–control studies351.17 (1.10–1.23)1.26 (1.13–1.40)0.310.09
    Males only341.17 (1.08–1.26)1.30 (1.12–1.50)0.550.001
    Females only401.18 (1.12–1.25)1.26 (1.13–1.41)0.450.002
    Incidence-only studies391.19 (1.09–1.30)1.28 (1.10–1.50)0.490.0008
Hodgkin's disease
    All studies81.25 (1.03–1.51)1.22 (0.96–1.55)0.160.35
    Cohort studies40.98 (0.68–1.38)1.00 (0.69–1.44)0.090.36
    Case–control studies41.39 (1.10–1.75)1.39 (1.10–1.75)0.000.60
    Males only51.30 (0.99–1.69)1.30 (0.99–1.69)0.000.81
    Females only51.18 (0.89–1.56)1.14 (0.69–1.87)0.590.08
    Incidence-only studies61.06 (0.76–1.46)1.09 (0.77–1.56)0.080.37
Non-Hodgkin's lymphoma
    All studies161.10 (1.00–1.19)1.10 (1.00–1.19)0.000.53
    Cohort studies61.08 (0.93–1.26)1.13 (0.93–1.37)0.270.26
    Case–control studies101.10 (0.99–1.22)1.10 (0.99–1.22)0.000.57
    Males only91.04 (0.90–1.21)1.04 (0.90–1.21)0.000.69
    Females only91.14 (1.03–1.26)1.14 (1.03–1.27)0.010.42
    Incidence-only studies131.08 (0.94–1.24)1.11 (0.94–1.32)0.150.32
Multiple myeloma
    All studies191.38 (1.25–1.54)1.62 (1.22–2.14)0.750.0001
    Cohort studies81.94 (1.62–2.32)1.89 (1.20–2.97)0.800.0001
    Case–control studies111.30 (1.15–1.48)1.75 (1.17–2.63)0.790.01
    Males only81.65 (1.39–1.96)1.80 (1.15–2.81)0.790.0002
    Females only141.35 (1.17–1.55)1.97 (1.32–2.93)0.780.0001
    Incidence-only studies131.60 (1.33–1.92)1.57 (1.04–2.37)0.720.0002
Leukemia
    All studies161.11 (1.03–1.19)1.11 (1.03–1.19)0.000.92
    Cohort studies61.04 (0.89–1.22)1.04 (0.89–1.22)0.000.93
    Case–control studies101.13 (1.03–1.22)1.13 (1.03–1.22)0.000.75
    Males only111.06 (0.95–1.19)1.11 (0.96–1.28)0.140.35
    Females only111.15 (1.06–1.26)1.15 (1.06–1.26)0.000.90
    Incidence-only studies101.07 (0.92–1.25)1.07 (0.92–1.25)0.000.85

aProportion of total variance due to between-study variance.

Table 7

Pooled RRs and 95% CIs of haematopoietic cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
All haematopoietic cancers
    All studies591.17 (1.11–1.22)1.26 (1.14–1.38)0.460.0001
    Cohort studies241.16 (1.06–1.28)1.23 (1.04–1.45)0.590.0001
    Case–control studies351.17 (1.10–1.23)1.26 (1.13–1.40)0.310.09
    Males only341.17 (1.08–1.26)1.30 (1.12–1.50)0.550.001
    Females only401.18 (1.12–1.25)1.26 (1.13–1.41)0.450.002
    Incidence-only studies391.19 (1.09–1.30)1.28 (1.10–1.50)0.490.0008
Hodgkin's disease
    All studies81.25 (1.03–1.51)1.22 (0.96–1.55)0.160.35
    Cohort studies40.98 (0.68–1.38)1.00 (0.69–1.44)0.090.36
    Case–control studies41.39 (1.10–1.75)1.39 (1.10–1.75)0.000.60
    Males only51.30 (0.99–1.69)1.30 (0.99–1.69)0.000.81
    Females only51.18 (0.89–1.56)1.14 (0.69–1.87)0.590.08
    Incidence-only studies61.06 (0.76–1.46)1.09 (0.77–1.56)0.080.37
Non-Hodgkin's lymphoma
    All studies161.10 (1.00–1.19)1.10 (1.00–1.19)0.000.53
    Cohort studies61.08 (0.93–1.26)1.13 (0.93–1.37)0.270.26
    Case–control studies101.10 (0.99–1.22)1.10 (0.99–1.22)0.000.57
    Males only91.04 (0.90–1.21)1.04 (0.90–1.21)0.000.69
    Females only91.14 (1.03–1.26)1.14 (1.03–1.27)0.010.42
    Incidence-only studies131.08 (0.94–1.24)1.11 (0.94–1.32)0.150.32
Multiple myeloma
    All studies191.38 (1.25–1.54)1.62 (1.22–2.14)0.750.0001
    Cohort studies81.94 (1.62–2.32)1.89 (1.20–2.97)0.800.0001
    Case–control studies111.30 (1.15–1.48)1.75 (1.17–2.63)0.790.01
    Males only81.65 (1.39–1.96)1.80 (1.15–2.81)0.790.0002
    Females only141.35 (1.17–1.55)1.97 (1.32–2.93)0.780.0001
    Incidence-only studies131.60 (1.33–1.92)1.57 (1.04–2.37)0.720.0002
Leukemia
    All studies161.11 (1.03–1.19)1.11 (1.03–1.19)0.000.92
    Cohort studies61.04 (0.89–1.22)1.04 (0.89–1.22)0.000.93
    Case–control studies101.13 (1.03–1.22)1.13 (1.03–1.22)0.000.75
    Males only111.06 (0.95–1.19)1.11 (0.96–1.28)0.140.35
    Females only111.15 (1.06–1.26)1.15 (1.06–1.26)0.000.90
    Incidence-only studies101.07 (0.92–1.25)1.07 (0.92–1.25)0.000.85
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
All haematopoietic cancers
    All studies591.17 (1.11–1.22)1.26 (1.14–1.38)0.460.0001
    Cohort studies241.16 (1.06–1.28)1.23 (1.04–1.45)0.590.0001
    Case–control studies351.17 (1.10–1.23)1.26 (1.13–1.40)0.310.09
    Males only341.17 (1.08–1.26)1.30 (1.12–1.50)0.550.001
    Females only401.18 (1.12–1.25)1.26 (1.13–1.41)0.450.002
    Incidence-only studies391.19 (1.09–1.30)1.28 (1.10–1.50)0.490.0008
Hodgkin's disease
    All studies81.25 (1.03–1.51)1.22 (0.96–1.55)0.160.35
    Cohort studies40.98 (0.68–1.38)1.00 (0.69–1.44)0.090.36
    Case–control studies41.39 (1.10–1.75)1.39 (1.10–1.75)0.000.60
    Males only51.30 (0.99–1.69)1.30 (0.99–1.69)0.000.81
    Females only51.18 (0.89–1.56)1.14 (0.69–1.87)0.590.08
    Incidence-only studies61.06 (0.76–1.46)1.09 (0.77–1.56)0.080.37
Non-Hodgkin's lymphoma
    All studies161.10 (1.00–1.19)1.10 (1.00–1.19)0.000.53
    Cohort studies61.08 (0.93–1.26)1.13 (0.93–1.37)0.270.26
    Case–control studies101.10 (0.99–1.22)1.10 (0.99–1.22)0.000.57
    Males only91.04 (0.90–1.21)1.04 (0.90–1.21)0.000.69
    Females only91.14 (1.03–1.26)1.14 (1.03–1.27)0.010.42
    Incidence-only studies131.08 (0.94–1.24)1.11 (0.94–1.32)0.150.32
Multiple myeloma
    All studies191.38 (1.25–1.54)1.62 (1.22–2.14)0.750.0001
    Cohort studies81.94 (1.62–2.32)1.89 (1.20–2.97)0.800.0001
    Case–control studies111.30 (1.15–1.48)1.75 (1.17–2.63)0.790.01
    Males only81.65 (1.39–1.96)1.80 (1.15–2.81)0.790.0002
    Females only141.35 (1.17–1.55)1.97 (1.32–2.93)0.780.0001
    Incidence-only studies131.60 (1.33–1.92)1.57 (1.04–2.37)0.720.0002
Leukemia
    All studies161.11 (1.03–1.19)1.11 (1.03–1.19)0.000.92
    Cohort studies61.04 (0.89–1.22)1.04 (0.89–1.22)0.000.93
    Case–control studies101.13 (1.03–1.22)1.13 (1.03–1.22)0.000.75
    Males only111.06 (0.95–1.19)1.11 (0.96–1.28)0.140.35
    Females only111.15 (1.06–1.26)1.15 (1.06–1.26)0.000.90
    Incidence-only studies101.07 (0.92–1.25)1.07 (0.92–1.25)0.000.85

aProportion of total variance due to between-study variance.

Urinary tract cancers

We retrieved 52 studies on urinary tract tumours: 34 on bladder, 10 on kidney and 8 on prostate (Table 3). The 34 bladder cancer studies showed a substantial increase in the risk (RR 1.30, 95% CI 1.20–1.42) (Table 8). Cohort studies, mortality studies and studies adjusted for smoking yielded a higher pooled RR. Other variables of the quality scale did not exert any impact on the results. Heterogeneity of the study-specific RRs was low (Ri = 0.16), a characteristic that was also found in the analysis of the 10 studies on kidney cancer. For this site, the pooled RR was 1.11 (1.00–1.23). The pooled RR of the eight studies on prostate cancer did not show any effect (RR 1.02, 95% CI 0.89–1.18).

Four studies provided data on hairdressers employed for >10 years. They showed a risk increase that was higher than that corresponding to ‘ever employment’. No data were available on employment for <10 years.

For kidney and prostate cancer, we did not observe any meaningful change in the results after stratification by any of the variables of the quality scale.

Digestive cancers

We retrieved 48 studies on digestive tumours43,48,49,51,53,55,56,90,93,95 (10 on oesophagus, 10 on stomach, 11 on colon, 8 on liver and 9 on pancreas) (Tables 4 and 9). We observed an increase in the risk of colon cancer, which persisted after stratification by design, outcome measure (incidence/mortality), gender and quality variables. An increase in the risk was also observed for pancreas cancer, particularly for cohort studies (RR 1.22, 95% CI 1.22–1.42). For liver cancer, we observed an increased risk for male hairdressers.

Table 8

Pooled RRs and 95% CIs of urinary tract cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Bladder
    All studies341.30 (1.20–1.42)1.36 (1.22–1.52)0.160.24
    Cohort studies81.36 (1.21–1.53)1.41 (1.20–1.65)0.300.22
    Case–control studies261.25 (1.12–1.41)1.35 (1.14–1.59)0.160.32
    Males only261.33 (1.20–1.47)1.45 (1.21–1.72)0.450.01
    Females only101.21 (1.04–1.39)1.21 (1.04–1.39)0.000.84
    Smoking adjusted191.33 (1.07–1.67)1.33 (1.07–1.67)0.000.77
    Smoking non-adjusted151.30 (1.19–1.42)1.42 (1.22–1.66)0.490.03
    Incidence-only studies261.35 (1.21–1.50)1.35 (1.21–1.50)0.000.78
    Mortality-only studies81.24 (1.09–1.41)1.53 (1.06–2.20)0.800.009
    Employment >10 years41.89 (0.98–3.66)1.93 (0.94–3.97)0.160.32
Kidney
    All studies101.11 (1.00–1.23)1.11 (1.00–1.23)0.000.75
    Cohort studies71.20 (1.04–1.39)1.20 (1.04–1.39)0.000.84
    Case–control studies31.03 (0.89–1.19)1.03 (0.89–1.19)0.000.66
    Males only71.08 (0.92–1.28)1.08 (0.92–1.28)0.000.48
    Females only91.13 (0.99–1.30)1.13 (0.99–1.30)0.000.75
    Incidence-only studies71.20 (1.04–1.40)1.20 (1.04–1.40)0.000.81
Prostate
    All studies80.95 (0.89–1.03)1.02 (0.89–1.18)0.570.05
    Cohort studies61.01 (0.90–1.13)1.12 (0.90–1.40)0.660.04
    Case–control studies20.91 (0.82–1.01)0.91 (0.82–1.01)0.000.49
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Bladder
    All studies341.30 (1.20–1.42)1.36 (1.22–1.52)0.160.24
    Cohort studies81.36 (1.21–1.53)1.41 (1.20–1.65)0.300.22
    Case–control studies261.25 (1.12–1.41)1.35 (1.14–1.59)0.160.32
    Males only261.33 (1.20–1.47)1.45 (1.21–1.72)0.450.01
    Females only101.21 (1.04–1.39)1.21 (1.04–1.39)0.000.84
    Smoking adjusted191.33 (1.07–1.67)1.33 (1.07–1.67)0.000.77
    Smoking non-adjusted151.30 (1.19–1.42)1.42 (1.22–1.66)0.490.03
    Incidence-only studies261.35 (1.21–1.50)1.35 (1.21–1.50)0.000.78
    Mortality-only studies81.24 (1.09–1.41)1.53 (1.06–2.20)0.800.009
    Employment >10 years41.89 (0.98–3.66)1.93 (0.94–3.97)0.160.32
Kidney
    All studies101.11 (1.00–1.23)1.11 (1.00–1.23)0.000.75
    Cohort studies71.20 (1.04–1.39)1.20 (1.04–1.39)0.000.84
    Case–control studies31.03 (0.89–1.19)1.03 (0.89–1.19)0.000.66
    Males only71.08 (0.92–1.28)1.08 (0.92–1.28)0.000.48
    Females only91.13 (0.99–1.30)1.13 (0.99–1.30)0.000.75
    Incidence-only studies71.20 (1.04–1.40)1.20 (1.04–1.40)0.000.81
Prostate
    All studies80.95 (0.89–1.03)1.02 (0.89–1.18)0.570.05
    Cohort studies61.01 (0.90–1.13)1.12 (0.90–1.40)0.660.04
    Case–control studies20.91 (0.82–1.01)0.91 (0.82–1.01)0.000.49
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64

aProportion of total variance due to between-study variance.

Table 8

Pooled RRs and 95% CIs of urinary tract cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Bladder
    All studies341.30 (1.20–1.42)1.36 (1.22–1.52)0.160.24
    Cohort studies81.36 (1.21–1.53)1.41 (1.20–1.65)0.300.22
    Case–control studies261.25 (1.12–1.41)1.35 (1.14–1.59)0.160.32
    Males only261.33 (1.20–1.47)1.45 (1.21–1.72)0.450.01
    Females only101.21 (1.04–1.39)1.21 (1.04–1.39)0.000.84
    Smoking adjusted191.33 (1.07–1.67)1.33 (1.07–1.67)0.000.77
    Smoking non-adjusted151.30 (1.19–1.42)1.42 (1.22–1.66)0.490.03
    Incidence-only studies261.35 (1.21–1.50)1.35 (1.21–1.50)0.000.78
    Mortality-only studies81.24 (1.09–1.41)1.53 (1.06–2.20)0.800.009
    Employment >10 years41.89 (0.98–3.66)1.93 (0.94–3.97)0.160.32
Kidney
    All studies101.11 (1.00–1.23)1.11 (1.00–1.23)0.000.75
    Cohort studies71.20 (1.04–1.39)1.20 (1.04–1.39)0.000.84
    Case–control studies31.03 (0.89–1.19)1.03 (0.89–1.19)0.000.66
    Males only71.08 (0.92–1.28)1.08 (0.92–1.28)0.000.48
    Females only91.13 (0.99–1.30)1.13 (0.99–1.30)0.000.75
    Incidence-only studies71.20 (1.04–1.40)1.20 (1.04–1.40)0.000.81
Prostate
    All studies80.95 (0.89–1.03)1.02 (0.89–1.18)0.570.05
    Cohort studies61.01 (0.90–1.13)1.12 (0.90–1.40)0.660.04
    Case–control studies20.91 (0.82–1.01)0.91 (0.82–1.01)0.000.49
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Bladder
    All studies341.30 (1.20–1.42)1.36 (1.22–1.52)0.160.24
    Cohort studies81.36 (1.21–1.53)1.41 (1.20–1.65)0.300.22
    Case–control studies261.25 (1.12–1.41)1.35 (1.14–1.59)0.160.32
    Males only261.33 (1.20–1.47)1.45 (1.21–1.72)0.450.01
    Females only101.21 (1.04–1.39)1.21 (1.04–1.39)0.000.84
    Smoking adjusted191.33 (1.07–1.67)1.33 (1.07–1.67)0.000.77
    Smoking non-adjusted151.30 (1.19–1.42)1.42 (1.22–1.66)0.490.03
    Incidence-only studies261.35 (1.21–1.50)1.35 (1.21–1.50)0.000.78
    Mortality-only studies81.24 (1.09–1.41)1.53 (1.06–2.20)0.800.009
    Employment >10 years41.89 (0.98–3.66)1.93 (0.94–3.97)0.160.32
Kidney
    All studies101.11 (1.00–1.23)1.11 (1.00–1.23)0.000.75
    Cohort studies71.20 (1.04–1.39)1.20 (1.04–1.39)0.000.84
    Case–control studies31.03 (0.89–1.19)1.03 (0.89–1.19)0.000.66
    Males only71.08 (0.92–1.28)1.08 (0.92–1.28)0.000.48
    Females only91.13 (0.99–1.30)1.13 (0.99–1.30)0.000.75
    Incidence-only studies71.20 (1.04–1.40)1.20 (1.04–1.40)0.000.81
Prostate
    All studies80.95 (0.89–1.03)1.02 (0.89–1.18)0.570.05
    Cohort studies61.01 (0.90–1.13)1.12 (0.90–1.40)0.660.04
    Case–control studies20.91 (0.82–1.01)0.91 (0.82–1.01)0.000.49
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64

aProportion of total variance due to between-study variance.

Table 9

Pooled RRs and 95% CIs of digestive cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Oesophagus
    All studies100.87 (0.74–1.03)0.87 (0.74–1.03)0.000.78
    Cohort studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
    Case–control studies30.83 (0.69–1.00)0.83 (0.69–1.00)0.000.41
    Males only80.90 (0.73–1.10)0.90 (0.73–1.10)0.000.46
    Females only60.84 (0.65–1.09)0.84 (0.65–1.09)0.000.98
    Incidence-only studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
Stomach
    All studies101.08 (0.99–1.18)1.04 (0.91–1.19)0.420.10
    Cohort studies60.89 (0.77–1.04)0.89 (0.77–1.04)0.000.59
    Case–control studies41.19 (1.07–1.32)1.19 (1.07–1.32)0.000.76
    Males only81.02 (0.90–1.17)1.02 (0.90–1.17)0.000.64
    Females only81.11 (1.00–1.25)1.06 (0.89–1.26)0.410.15
    Incidence-only studies70.90 (0.77–1.04)0.90 (0.77–1.04)0.000.70
Colon
    All studies111.08 (1.02–1.13)1.08 (1.02–1.13)0.000.92
    Cohort studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.94
    Case–control studies51.05 (0.99–1.12)1.05 (0.99–1.12)0.000.71
    Males only81.04 (0.96–1.12)1.07 (0.95–1.22)0.430.14
    Females only91.10 (1.03–1.16)1.10 (1.03–1.16)0.000.83
    Incidence-only studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.99
Liver
    All studies81.39 (1.29–1.51)1.17 (0.90–1.53)0.800.02
    Cohort studies51.05 (0.86–1.27)1.05 (0.86–1.27)0.000.76
    Case–control studies31.48 (1.35–1.61)1.07 (0.59–1.94)0.970.08
    Males only71.49 (1.37–1.63)1.49 (1.37–1.63)0.000.50
    Females only40.96 (0.78–1.17)0.96 (0.78–1.17)0.000.81
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Pancreas
    All studies91.11 (1.02–1.20)1.08 (0.90–1.29)0.630.05
    Cohort studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
    Case–control studies31.06 (0.96–1.17)0.75 (0.40–1.41)0.960.06
    Males only80.91 (0.80–1.04)1.06 (0.74–1.52)0.810.001
    Females only81.20 (1.09–1.32)1.20 (1.09–1.32)0.000.65
    Incidence-only studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Oesophagus
    All studies100.87 (0.74–1.03)0.87 (0.74–1.03)0.000.78
    Cohort studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
    Case–control studies30.83 (0.69–1.00)0.83 (0.69–1.00)0.000.41
    Males only80.90 (0.73–1.10)0.90 (0.73–1.10)0.000.46
    Females only60.84 (0.65–1.09)0.84 (0.65–1.09)0.000.98
    Incidence-only studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
Stomach
    All studies101.08 (0.99–1.18)1.04 (0.91–1.19)0.420.10
    Cohort studies60.89 (0.77–1.04)0.89 (0.77–1.04)0.000.59
    Case–control studies41.19 (1.07–1.32)1.19 (1.07–1.32)0.000.76
    Males only81.02 (0.90–1.17)1.02 (0.90–1.17)0.000.64
    Females only81.11 (1.00–1.25)1.06 (0.89–1.26)0.410.15
    Incidence-only studies70.90 (0.77–1.04)0.90 (0.77–1.04)0.000.70
Colon
    All studies111.08 (1.02–1.13)1.08 (1.02–1.13)0.000.92
    Cohort studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.94
    Case–control studies51.05 (0.99–1.12)1.05 (0.99–1.12)0.000.71
    Males only81.04 (0.96–1.12)1.07 (0.95–1.22)0.430.14
    Females only91.10 (1.03–1.16)1.10 (1.03–1.16)0.000.83
    Incidence-only studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.99
Liver
    All studies81.39 (1.29–1.51)1.17 (0.90–1.53)0.800.02
    Cohort studies51.05 (0.86–1.27)1.05 (0.86–1.27)0.000.76
    Case–control studies31.48 (1.35–1.61)1.07 (0.59–1.94)0.970.08
    Males only71.49 (1.37–1.63)1.49 (1.37–1.63)0.000.50
    Females only40.96 (0.78–1.17)0.96 (0.78–1.17)0.000.81
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Pancreas
    All studies91.11 (1.02–1.20)1.08 (0.90–1.29)0.630.05
    Cohort studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
    Case–control studies31.06 (0.96–1.17)0.75 (0.40–1.41)0.960.06
    Males only80.91 (0.80–1.04)1.06 (0.74–1.52)0.810.001
    Females only81.20 (1.09–1.32)1.20 (1.09–1.32)0.000.65
    Incidence-only studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21

aProportion of total variance due to between-study variance.

Table 9

Pooled RRs and 95% CIs of digestive cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Oesophagus
    All studies100.87 (0.74–1.03)0.87 (0.74–1.03)0.000.78
    Cohort studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
    Case–control studies30.83 (0.69–1.00)0.83 (0.69–1.00)0.000.41
    Males only80.90 (0.73–1.10)0.90 (0.73–1.10)0.000.46
    Females only60.84 (0.65–1.09)0.84 (0.65–1.09)0.000.98
    Incidence-only studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
Stomach
    All studies101.08 (0.99–1.18)1.04 (0.91–1.19)0.420.10
    Cohort studies60.89 (0.77–1.04)0.89 (0.77–1.04)0.000.59
    Case–control studies41.19 (1.07–1.32)1.19 (1.07–1.32)0.000.76
    Males only81.02 (0.90–1.17)1.02 (0.90–1.17)0.000.64
    Females only81.11 (1.00–1.25)1.06 (0.89–1.26)0.410.15
    Incidence-only studies70.90 (0.77–1.04)0.90 (0.77–1.04)0.000.70
Colon
    All studies111.08 (1.02–1.13)1.08 (1.02–1.13)0.000.92
    Cohort studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.94
    Case–control studies51.05 (0.99–1.12)1.05 (0.99–1.12)0.000.71
    Males only81.04 (0.96–1.12)1.07 (0.95–1.22)0.430.14
    Females only91.10 (1.03–1.16)1.10 (1.03–1.16)0.000.83
    Incidence-only studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.99
Liver
    All studies81.39 (1.29–1.51)1.17 (0.90–1.53)0.800.02
    Cohort studies51.05 (0.86–1.27)1.05 (0.86–1.27)0.000.76
    Case–control studies31.48 (1.35–1.61)1.07 (0.59–1.94)0.970.08
    Males only71.49 (1.37–1.63)1.49 (1.37–1.63)0.000.50
    Females only40.96 (0.78–1.17)0.96 (0.78–1.17)0.000.81
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Pancreas
    All studies91.11 (1.02–1.20)1.08 (0.90–1.29)0.630.05
    Cohort studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
    Case–control studies31.06 (0.96–1.17)0.75 (0.40–1.41)0.960.06
    Males only80.91 (0.80–1.04)1.06 (0.74–1.52)0.810.001
    Females only81.20 (1.09–1.32)1.20 (1.09–1.32)0.000.65
    Incidence-only studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Oesophagus
    All studies100.87 (0.74–1.03)0.87 (0.74–1.03)0.000.78
    Cohort studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
    Case–control studies30.83 (0.69–1.00)0.83 (0.69–1.00)0.000.41
    Males only80.90 (0.73–1.10)0.90 (0.73–1.10)0.000.46
    Females only60.84 (0.65–1.09)0.84 (0.65–1.09)0.000.98
    Incidence-only studies71.04 (0.73–1.48)1.04 (0.73–1.48)0.000.87
Stomach
    All studies101.08 (0.99–1.18)1.04 (0.91–1.19)0.420.10
    Cohort studies60.89 (0.77–1.04)0.89 (0.77–1.04)0.000.59
    Case–control studies41.19 (1.07–1.32)1.19 (1.07–1.32)0.000.76
    Males only81.02 (0.90–1.17)1.02 (0.90–1.17)0.000.64
    Females only81.11 (1.00–1.25)1.06 (0.89–1.26)0.410.15
    Incidence-only studies70.90 (0.77–1.04)0.90 (0.77–1.04)0.000.70
Colon
    All studies111.08 (1.02–1.13)1.08 (1.02–1.13)0.000.92
    Cohort studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.94
    Case–control studies51.05 (0.99–1.12)1.05 (0.99–1.12)0.000.71
    Males only81.04 (0.96–1.12)1.07 (0.95–1.22)0.430.14
    Females only91.10 (1.03–1.16)1.10 (1.03–1.16)0.000.83
    Incidence-only studies61.11 (1.03–1.20)1.11 (1.03–1.20)0.000.99
Liver
    All studies81.39 (1.29–1.51)1.17 (0.90–1.53)0.800.02
    Cohort studies51.05 (0.86–1.27)1.05 (0.86–1.27)0.000.76
    Case–control studies31.48 (1.35–1.61)1.07 (0.59–1.94)0.970.08
    Males only71.49 (1.37–1.63)1.49 (1.37–1.63)0.000.50
    Females only40.96 (0.78–1.17)0.96 (0.78–1.17)0.000.81
    Incidence-only studies40.95 (0.85–1.07)0.95 (0.85–1.07)0.000.64
Pancreas
    All studies91.11 (1.02–1.20)1.08 (0.90–1.29)0.630.05
    Cohort studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21
    Case–control studies31.06 (0.96–1.17)0.75 (0.40–1.41)0.960.06
    Males only80.91 (0.80–1.04)1.06 (0.74–1.52)0.810.001
    Females only81.20 (1.09–1.32)1.20 (1.09–1.32)0.000.65
    Incidence-only studies61.22 (1.05–1.42)1.18 (0.95–1.46)0.380.21

aProportion of total variance due to between-study variance.

Respiratory cancers

We retrieved 18 studies on lung cancer and 12 on larynx cancer (Table 5).43,48,49,51,53,55,56,90,92,93,95–101 We observed an increase in the risk of lung cancer that rounded to 30% (Table 10). Except when restricted to male studies, this increase did not subside after stratification. The effect was higher among cohort studies, female studies, incidence studies and those studies that adjusted for smoking.

Table 10

Pooled RRs and 95% CIs of respiratory cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Lung
    All studies181.18 (1.14–1.22)1.27 (1.15–1.41)0.780.0001
    Cohort studies81.32 (1.23–1.41)1.32 (1.23–1.42)0.050.40
    Case–control studies101.13 (1.09–1.18)1.29 (1.04–1.61)0.930.0006
    Males only120.96 (0.92–1.06)1.12 (0.94–1.33)0.890.0001
    Females only141.34 (1.28–1.41)1.40 (1.24–1.59)0.700.006
    Smoking adjusted41.36 (1.16–1.60)1.36 (1.08–1.70)0.430.19
    Smoking non-adjusted141.17 (1.13–1.21)1.25 (1.12–1.40)0.810.0001
    Incidence-only studies81.32 (1.23–1.42)1.32 (1.22–1.43)0.060.37
    Mortality-only studies101.14 (1.09–1.19)1.24 (1.04–1.48)0.910.0001
Larynx
    All studies121.32 (1.10–1.59)1.52 (1.11–2.08)0.520.04
    Cohort studies71.68 (1.31–2.17)1.70 (1.25–2.32)0.230.27
    Case–control studies51.02 (0.79–1.33)1.18 (0.66–2.11)0.630.17
    Males only111.25 (1.01–1.54)1.30 (0.93–1.81)0.470.06
    Females only51.67 (1.17–2.37)2.36 (1.03–5.40)0.750.03
    Smoking adjusted22.18 (1.00–4.71)2.18 (1.00–4.71)0.000.69
    Smoking non-adjusted101.28 (1.06–1.55)1.46 (1.04–2.06)0.580.02
    Incidence-only studies91.74 (1.37–2.22)1.75 (1.36–2.26)0.050.39
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Lung
    All studies181.18 (1.14–1.22)1.27 (1.15–1.41)0.780.0001
    Cohort studies81.32 (1.23–1.41)1.32 (1.23–1.42)0.050.40
    Case–control studies101.13 (1.09–1.18)1.29 (1.04–1.61)0.930.0006
    Males only120.96 (0.92–1.06)1.12 (0.94–1.33)0.890.0001
    Females only141.34 (1.28–1.41)1.40 (1.24–1.59)0.700.006
    Smoking adjusted41.36 (1.16–1.60)1.36 (1.08–1.70)0.430.19
    Smoking non-adjusted141.17 (1.13–1.21)1.25 (1.12–1.40)0.810.0001
    Incidence-only studies81.32 (1.23–1.42)1.32 (1.22–1.43)0.060.37
    Mortality-only studies101.14 (1.09–1.19)1.24 (1.04–1.48)0.910.0001
Larynx
    All studies121.32 (1.10–1.59)1.52 (1.11–2.08)0.520.04
    Cohort studies71.68 (1.31–2.17)1.70 (1.25–2.32)0.230.27
    Case–control studies51.02 (0.79–1.33)1.18 (0.66–2.11)0.630.17
    Males only111.25 (1.01–1.54)1.30 (0.93–1.81)0.470.06
    Females only51.67 (1.17–2.37)2.36 (1.03–5.40)0.750.03
    Smoking adjusted22.18 (1.00–4.71)2.18 (1.00–4.71)0.000.69
    Smoking non-adjusted101.28 (1.06–1.55)1.46 (1.04–2.06)0.580.02
    Incidence-only studies91.74 (1.37–2.22)1.75 (1.36–2.26)0.050.39

aProportion of total variance due to between-study variance.

Table 10

Pooled RRs and 95% CIs of respiratory cancers among hairdressers and related occupations

Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Lung
    All studies181.18 (1.14–1.22)1.27 (1.15–1.41)0.780.0001
    Cohort studies81.32 (1.23–1.41)1.32 (1.23–1.42)0.050.40
    Case–control studies101.13 (1.09–1.18)1.29 (1.04–1.61)0.930.0006
    Males only120.96 (0.92–1.06)1.12 (0.94–1.33)0.890.0001
    Females only141.34 (1.28–1.41)1.40 (1.24–1.59)0.700.006
    Smoking adjusted41.36 (1.16–1.60)1.36 (1.08–1.70)0.430.19
    Smoking non-adjusted141.17 (1.13–1.21)1.25 (1.12–1.40)0.810.0001
    Incidence-only studies81.32 (1.23–1.42)1.32 (1.22–1.43)0.060.37
    Mortality-only studies101.14 (1.09–1.19)1.24 (1.04–1.48)0.910.0001
Larynx
    All studies121.32 (1.10–1.59)1.52 (1.11–2.08)0.520.04
    Cohort studies71.68 (1.31–2.17)1.70 (1.25–2.32)0.230.27
    Case–control studies51.02 (0.79–1.33)1.18 (0.66–2.11)0.630.17
    Males only111.25 (1.01–1.54)1.30 (0.93–1.81)0.470.06
    Females only51.67 (1.17–2.37)2.36 (1.03–5.40)0.750.03
    Smoking adjusted22.18 (1.00–4.71)2.18 (1.00–4.71)0.000.69
    Smoking non-adjusted101.28 (1.06–1.55)1.46 (1.04–2.06)0.580.02
    Incidence-only studies91.74 (1.37–2.22)1.75 (1.36–2.26)0.050.39
Number of studiesRR (95% CI) fixed effectsRR (95% CI) random effectsRiaQ-test P-value
Lung
    All studies181.18 (1.14–1.22)1.27 (1.15–1.41)0.780.0001
    Cohort studies81.32 (1.23–1.41)1.32 (1.23–1.42)0.050.40
    Case–control studies101.13 (1.09–1.18)1.29 (1.04–1.61)0.930.0006
    Males only120.96 (0.92–1.06)1.12 (0.94–1.33)0.890.0001
    Females only141.34 (1.28–1.41)1.40 (1.24–1.59)0.700.006
    Smoking adjusted41.36 (1.16–1.60)1.36 (1.08–1.70)0.430.19
    Smoking non-adjusted141.17 (1.13–1.21)1.25 (1.12–1.40)0.810.0001
    Incidence-only studies81.32 (1.23–1.42)1.32 (1.22–1.43)0.060.37
    Mortality-only studies101.14 (1.09–1.19)1.24 (1.04–1.48)0.910.0001
Larynx
    All studies121.32 (1.10–1.59)1.52 (1.11–2.08)0.520.04
    Cohort studies71.68 (1.31–2.17)1.70 (1.25–2.32)0.230.27
    Case–control studies51.02 (0.79–1.33)1.18 (0.66–2.11)0.630.17
    Males only111.25 (1.01–1.54)1.30 (0.93–1.81)0.470.06
    Females only51.67 (1.17–2.37)2.36 (1.03–5.40)0.750.03
    Smoking adjusted22.18 (1.00–4.71)2.18 (1.00–4.71)0.000.69
    Smoking non-adjusted101.28 (1.06–1.55)1.46 (1.04–2.06)0.580.02
    Incidence-only studies91.74 (1.37–2.22)1.75 (1.36–2.26)0.050.39

aProportion of total variance due to between-study variance.

The pooled estimates of the 12 studies on larynx cancer show a considerable increase in the risk (random effects pooled RR 1.52, 95% CI 1.11–2.08). This risk is higher when we restricted the analysis to cohort studies (RR 1.68; 95% CI 1.31–2.17) and female studies (RR 2.36, 95% CI 1.03–5.40). The two studies that adjusted for smoking showed an RR of 2.36 (95% CI 1.03–5.40).

Brain, skin and salivary gland cancers

In addition to cancers of specific sites presented above, nine studies were available for brain cancer51,52,55,93,95,102 and seven for melanoma.51,52,55,56,93 The pooled estimates were 1.12 (95% CI 1.02–1.24) for brain and 1.03 (95% CI 0.92–1.14) for melanoma. These results did not change substantially after stratification by design or gender.

We also found one study that shows that female hairdressers have an elevated risk of salivary gland cancer (RR 2.7, 95% CI 1.1–6.5).103

Assessment of publication bias

We did not find evidence of publication bias by examining the funnel plots of the anatomic sites with sufficient number of studies (data not shown, available from first author). We observed some asymmetry on the funnel plots of the studies on lung and bladder cancers. In both cases asymmetry was caused by the extreme result of one study.92,99

Except for multiple myeloma (P = 0.03), the regression test of asymmetry yielded inconclusive results for haematopoietic cancers (P = 0.08 for all sites together, 0.33 for Hodgkin's disease, 0.84 for non-Hodgkin's lymphoma, and 0.98 for leukaemia).

To further evaluate the possibility of publication bias in case–control studies, a design that is more probably disregarded by authors and editors in case of null or statistically non-significant results, we performed a sensitivity analysis by recalculating our pooled estimates under the following extreme assumptions: (i) published case–control studies are only half of the studies of occupation as hairdresser and cancer ever conducted; (ii) all unpublished studies found an RR of 1; and (iii) the unpublished studies included as many cases and controls as the average of the published studies. Under these assumptions, the pooled estimates still show a moderate increase in the risk: RR 1.23 (95% CI 1.12–1.34) for lung cancer, RR 1.29 (95% CI 1.04–1.60) for larynx cancer, RR 1.06 (95% CI 1.02–1.10) for bladder cancer, RR 1.06 (95% CI 1.02–1.09) for all haematopoietic cancers, RR 1.12 (95% CI 1.03–1.23) for multiple myeloma.

To check for any change in the pooled estimates that may have occurred after the ban of the most important carcinogenic agents (2,4-diaminotoluene and 2,4-diaminoanisole) from hair dyes in the mid-1970s,3 we restricted our analysis to those studies carried out before this date and compared their pooled estimates with those of the general analysis. We did not find any substantial difference between the results. We could not compare the pooled estimates before and after the ban of those agents as occupational exposure of studies carried out recently may have taken place before the ban.

Discussion

The high number of studies included in our meta-analysis and the consistency of results across anatomic sites and design settings provide strong evidence that cancer risk is higher among employees of the hairdressing industry.

The magnitude of the risk increase is substantial in several anatomic sites: 62% for multiple myeloma, 52% for larynx, 30% for bladder and 27% for lung cancer. The fact that the risk increase is observed for cancers of different anatomic sites may be explained by the existence of multiple exposure pathways (respiratory, dermatologic and systemic). It is also likely that some cancers, such as lung and bladder cancer, may share a common aetiology.

Nevertheless, alternative explanations for the risk increase deserve careful examination. Confounding by known or unknown factors is a non-causal explanation that is often argued to explain associations in meta-analyses. To act as a confounder, a factor must be related to exposure and disease. Besides age and sex, smoking is traditionally considered as the most important potential confounder of the association of any exposure factor with cancer. There is some evidence that smoking is more frequent among hairdressers than among the general population.54,104 However, when we restricted the analysis to those studies that adjusted for tobacco consumption, the magnitude of the association between occupation as a hairdresser and cancer did not decrease. On the contrary, the risk increased in those cancers that are known to be strongly linked to tobacco use (lung, larynx and bladder).

Potential confounding by alcohol is unlikely to occur as there is some evidence that alcohol intake is not higher among hairdressers.104 Unidentified genetic factors are not likely to introduce confounding as this would imply that the election of the hairdressing occupation by a subject is driven by genetic factors, a hypothesis that is biologically not plausible. Genotypic and phenotypic factors may, however, act as effect modifiers of the relation between occupational exposure as a hairdresser and cancer. This would lead to RRs between hairdressing and cancer that are of different magnitudes across alleles. This feature was previously described for N-acetyltransferase-2 (NAT2) phenotype and NAT1*10 allele, and bladder cancer and non-Hodgkin lymphoma among users of hair dyes.105,106

The existence of another unidentified factor associated with both the hairdresser occupation and cancer, which could explain the relation observed, is unlikely. Even if this unidentified factor could double the risk of cancer among subjects exposed to it (RR confounder–disease = 2) and, simultaneously, this factor happened to be twice more prevalent among hairdressers than among other occupations (RR confounder–exposure = 2), the adjusted RR of the relation hairdresser occupation–cancer would still be 1.13 for lung cancer, 1.35 for larynx cancer, 1.21 for bladder cancer and 1.44 for multiple myeloma (assuming one-third of people are exposed to this unknown factor).107 The existence of an unknown factor so strongly related with the hairdresser profession and with cancer, without being a chemical agent present in the occupational environment of a hairdresser, is highly improbable.

The overwhelming majority of the studies included in this meta-analysis were carried out in the USA and Europe, with the exception of Japan and New Zealand. The conclusions may then not generalize to other populations.

Our meta-analysis may be limited by the fact that we included studies that used information systems, such as death certificates, that may present incomplete information on confounders and occupational exposure.32 We overcame this caveat by restricting, in a second stage, our analysis to studies that measured incidence only, and excluding studies based on death certificates. The results did not change substantially.

Except for the very limited assessment of duration of employment in four bladder cancer studies, we could not assess the existence of a dose–response gradient as no data were available on duration of employment or exposure intensity in the studies included in this meta-analysis.

Hairdressers are chronically exposed to potentially harmful chemicals contained in hair dyes. The use of the most aggressive agents was discontinued three decades ago. A deceleration in the risk increase is then expectable in those studies carried out partially or completely after this ban. We were not able to find evidence of this change. It is then probable that hairdressers and allied occupations continue to be exposed to DNA-damaging agents other than those contained in hair dyes, such as formaldehyde, metacrylate and acetone, the concentration of which may be excessive in the air of hairdresser salons.104 IARC does not attribute specific causality of carcinogenicity among hairdressers to hair dying agents, but does mention that over 5000 chemicals are present in the environment of these workers.15 Also, it should be mentioned that due to the long latency period between exposure or its removal and the occurrence of cancer, the comparison of studies carried out between and after the ban is not straightforward. If an effect exists, only the most recent studies and, more probably, the forthcoming studies may provide evidence if the banned problematic hair dyes were involved in increased cancer risks in hairdressers or not.

The fact that cosmetics are generally exempted from the labelling information that is required for other chemicals increases potential hazard in subjects frequently exposed to them.31 In fact, detection and prevention measures related to industrial hygiene have only recently been instituted in hairdressing and nail care salons.108 A recent study found that none of the hairdresser salons surveyed in central and suburban areas of a large city had any ventilation system.109 Furthermore, it is remarkable that only about one-third of hairdressers in Nordic countries have used protective gloves while using hair dyes.110

In conclusion, our results show that hairdressers and allied occupations have a higher risk of cancer than the general population. The risk increase is substantial for lung, larynx, bladder cancer and multiple myeloma and less marked for Hodgkin's disease, non-Hodgkin's lymphoma, leukaemia, ovary, breast, kidney and colon cancer. Improvement of the ventilation system in the hairdresser salons and implementation of hygiene measures aimed at mitigating exposure to potential carcinogens at work may reduce the risk.

Funding

Takkouche, Montes-Martínez and Regueira-Méndez are employees of the University of Santiago de Compostela (Spain) and have grants from ‘CIBER en Epidemiología y Salud Pública’ (CIBER-ESP), Spain, a public centre affiliated to the Spanish Ministry of Health. Dr Takkouche's work on risk factors of haematopoietic cancers is funded by grant A/019761/08 of the International Cooperation Programme of the Spanish Ministry of Foreign Affairs. No special funding was received for this study.

Acknowledgement

The authors thank Dr Vincent Cogliano from IARC (Lyon, France) for providing studies that we could not find in our original search.

Conflict of interest: None declared.

KEY MESSAGES

  • Hairdressers and allied occupations have a higher risk of cancer than the general population.

  • The risk increase is particularly marked for lung, larynx and bladder cancer and multiple myeloma.

  • Improvement of the ventilation system in hairdresser salons and implementation of hygiene measures aimed at mitigating exposure to potential carcinogens at work are probably the best preventive tools.

References

1
BUREAU of Labor Statistics, U.S. Department of Labor.
Barbers, cosmetologists, and other personal appearance workers in: Occupational Outlook Handbook, 2008–09 edition.
19 August 2009, date last accessed 
2
CIC Europe/Uni-Europa.
Code of Conduct. Guidelines for European Hairdressers. ‘How to Get Along Code’
19 August 2009, date last accessed 
3
Anon
Occupational exposures of hairdressers and barbers and personal use of hair colourants
IARC Monogr Eval Carcinog Risks Hum
1993
, vol. 
57
 (pg. 
43
-
118
)
4
Silverman
DT
Morrison
AS
Devesa
SS
Schottenfeld
D
Fraumeni
JF
Bladder cancer
Cancer Epidemiology and Prevention
1996
New York
Oxford University Press
(pg. 
1156
-
79
)
5
Babish
JG
Scarlett
JM
Voekler
SE
Gutenmann
WH
Lisk
DJ
Urinary mutagens in cosmetologists and dental personnel
J Toxicol Environ Health
1991
, vol. 
34
 (pg. 
197
-
206
)
6
Takkouche
B
Etminan
M
Montes-Martínez
A
Personal use of hair dyes and risk of cancer: A meta-analysis
JAMA
2005
, vol. 
293
 (pg. 
2516
-
25
)
7
Coiffure EU/UNI-Europa Hair and Beauty.
Covenant on Health and Safety, in particular the Use and Handling of Cosmetic Products and their Chemical Agents, between European Social Partners in the Hairdressing Industry
19 August 2009, date last accessed 
8
Vasama-Neuvonen
K
Pukkala
E
Paakkulainen
H
, et al. 
Ovarian cancer and occupational exposures in Finland
Am J Ind Med
1999
, vol. 
36
 (pg. 
83
-
89
)
9
Dryson
E
’t Mannetje
A
Walls
C
, et al. 
Case-control study of high risk occupations for bladder cancer in New Zealand
Int J Cancer
2008
, vol. 
122
 (pg. 
1340
-
46
)
10
Cote
TR
Dosemeci
M
Rothman
N
Banks
RB
Biggar
RJ
Non Hodgkin's lymphoma and occupational exposure to hair dyes among people with AIDS
Am J Public Health
1993
, vol. 
83
 (pg. 
598
-
99
)
11
tMannetje
A
Dryson
E
Walls
C
, et al. 
High risk occupations for non-Hodgkin's lymphomas in New Zealand
Occup Environ Med
2008
, vol. 
65
 (pg. 
354
-
63
)
12
Baan
R
Straif
K
Grosse
Y
, et al. 
Carcinogenicity of some aromatic amines, organic dyes and related exposures
Lancet Oncol
2008
, vol. 
9
 (pg. 
322
-
23
)
13
Stroup
DF
Berlin
JA
Morton
SC
, et al. 
Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group
JAMA
2000
, vol. 
283
 (pg. 
2008
-
12
)
14
La Vecchia
C
Tavani
A
Epidemiological evidence on hair dyes and the risk of cancer in humans
Eur J Cancer Prev
1995
, vol. 
4
 (pg. 
31
-
43
)
15
IARC
IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Some Aromatic Amines, Organic Dyes and Related Exposures
Lyon
International Agency for Research on Cancer
 
Vol. 99
16
Wells
G
Shea
B
O’Connell
D
, et al. 
The Newcastle-Ottawa scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses.
19 August 2009, date last accessed 
17
Walker
AM
Walker
AM
Estimation
Observation and Inference: An Introduction to the Methods of Epidemiology
1991
Newton Lower Falls
Epidemiology Resources Inc
pg. 
113
 
18
Tomatis
L
Tomatis
L
Quantification of effect
Cancer: Causes, Occurrence and Control
1980
Lyon
IARC Scientific Publications
pg. 
288
 
19
Rothman
KJ
Rothman
KJ
Types of epidemiologic studies
Modern Epidemiology
1986
Boston: Little Brown and Co.
pg. 
73
 
20
Takkouche
B
Cadarso-Suarez
C
Spiegelman
D
Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis
Am J Epidemiol
1999
, vol. 
150
 (pg. 
206
-
15
)
21
Egger
M
Davey Smith
G
Schneider
M
Minder
C
Bias in meta-analysis detected by a simple, graphical test
BMJ
1997
, vol. 
315
 (pg. 
629
-
34
)
22
Clemmesen
J
Statistical studies in the aetiology of malignant neoplasms. V. Trends and risks. Denmark 1943–72
Acta Pathol Microbiol Sc and Suppl
1977
, vol. 
261
 (pg. 
1
-
286
)
23
Claude
JC
Frentzel-Beyme
R
Kunze
E
Occupation and risk of cancer of the lower urinary tract among men. A case-control study
Int J Cancer
1988
, vol. 
41
 (pg. 
371
-
79
)
24
Milne
KL
Sandler
DP
Everson
RB
Brown
SM
Lung cancer and occupation in Alameda County: a death certificate case-control study
Am J Ind Med
1983
, vol. 
4
 (pg. 
565
-
75
)
25
Malker
HS
McLaughlin
JK
Silverman
DT
, et al. 
Occupational risks for bladder cancer among men in Sweden
Cancer Res
1987
, vol. 
47
 (pg. 
6763
-
66
)
26
McLaughlin
JK
Linet
MS
Stone
BJ
, et al. 
Multiple myeloma and occupation in Sweden
Arch Environ Health
1988
, vol. 
43
 (pg. 
7
-
10
)
27
Pollán
M
Gustavsson
P
High-risk occupations for breast cancer in the Swedish female working population
Am J Public Health
1999
, vol. 
89
 (pg. 
875
-
81
)
28
Shields
T
Gridley
G
Moradi
T
Adami
J
Plato
N
Dosemeci
M
Occupational exposures and the risk of ovarian cancer in Sweden
Am J Ind Med
2002
, vol. 
42
 (pg. 
200
-
13
)
29
Robinson
CF
Walker
JT
Cancer mortality among women employed in fast-growing U.S. occupations
Am J Ind Med
1999
, vol. 
36
 (pg. 
186
-
92
)
30
Skov
T
Andersen
A
Malker
H
Pukkala
E
Weiner
J
Lynge
E
Risk of cancer of the urinary bladder among hairdressers in the Nordic countries
Am J Ind Med
1990
, vol. 
17
 (pg. 
217
-
23
)
31
Boffetta
P
Andersen
A
Lynge
E
Barlow
L
Pukkala
E
Employment as a hairdresser and risk of ovarian cancer and non-Hodgkin's lymphomas among women
J Occup Med
1994
, vol. 
36
 (pg. 
61
-
65
)
32
Lynge
E
Thygesen
L
Use of surveillance systems for occupational cancer: data from the Danish National System
Int J Epidemiol
1988
, vol. 
17
 (pg. 
493
-
500
)
33
Ji
J
Hemminki
K
Occurrences of leukemia subtypes by socioeconomic and occupational groups in Sweden
J Occup Environ Med
2005
, vol. 
47
 (pg. 
1131
-
40
)
34
Lynge
E
Danish cancer registry as a resource for occupational research
J Occup Med
1994
, vol. 
36
 (pg. 
1169
-
72
)
35
Pukkala
E
Nokso-Koivisto
P
Roponen
P
Changing cancer risk pattern among Finnish hairdressers
Int Arch Occup Environ Health
1992
, vol. 
64
 (pg. 
39
-
42
)
36
Skov
T
Lynge
E
Cancer risk and exposures to carcinogens in hairdressers
Skin Pharmacol
1994
, vol. 
7
 (pg. 
94
-
100
)
37
Doebbert
G
Riedmiller
KR
Kizer
KW
Occupational mortality of California women, 1979–1981
West J Med
1988
, vol. 
149
 (pg. 
734
-
40
)
38
Viadana
E
Bross
IDJ
Houten
L
Cancer experience of men exposed to inhalation of chemical or to combustion products
J Occup Med
1976
, vol. 
18
 (pg. 
787
-
92
)
39
Giles
GG
Lickiss
JN
Baikie
MJ
Lowenthal
RM
Panton
J
Myeloproliferative and lymphoproliferative disorders in Tasmania 1972–80: occupational and familial aspects
J Natl Cancer Inst
1984
, vol. 
72
 (pg. 
1233
-
40
)
40
Howe
GR
Burch
JD
Miller
AB
, et al. 
Tobacco use, occupation, coffee, various nutrients, and bladder cancer
J Natl Cancer Inst
1980
, vol. 
64
 (pg. 
701
-
13
)
41
′t Mannetje
A
Kogevinas
M
Chang-Claude
J
, et al. 
Occupation and bladder cancer in European women
Cancer Causes Control
1999
, vol. 
10
 (pg. 
209
-
17
)
42
Kogevinas
M
‘t Mannetje
A
Cordier
S
, et al. 
Occupation and bladder cancer among men in Western Europe
Cancer Causes Control
2003
, vol. 
14
 (pg. 
907
-
14
)
43
Decoufle
P
A Retrospective Survey of Cancer in Relation to Occupation. DHEW (NIOSH) pub. N° 77–178
1977
Cincinnati, OH
National Institute of Occupational Safety and Health
44
Kinlen
LJ
Harris
R
Garrod
A
Rodríguez
K
Use of hair dyes by patients with breast cancer: a case-control study
BMJ
1977
, vol. 
2
 (pg. 
366
-
68
)
45
Koenig
KL
Pasternack
BS
Shore
RE
Strax
P
Hair dye use and breast cancer: a case-control study among screening participants
Am J Epidemiol
1991
, vol. 
133
 (pg. 
985
-
95
)
46
Habel
LA
Stanford
JL
Vaughan
TL
, et al. 
Occupation and breast cancer risk in middle-aged women
J Occup Environ Med
1995
, vol. 
37
 (pg. 
349
-
56
)
47
Band
P
Le
ND
Fand
R
Deschamps
M
Gallagher
R
Yang
P
Identification of occupational cancer risks in British Columbia: a population-based case-control study of 995 incident breast cancer cases by menopausal status, controlling for confounding factors
J Occup Environ Med
2000
, vol. 
42
 (pg. 
284
-
310
)
48
Menck
HR
Pike
MC
Henderson
BE
Ring
JS
Lung cancer risk among beauticians and other female workers: brief communication
J Natl Cancer Inst
1977
, vol. 
59
 (pg. 
1423
-
25
)
49
Kono
S
Tokudome
S
Ikeda
M
Yoshimura
T
Kuratsune
M
Cancer and other causes of death among female beauticians
J Natl Cancer Inst
1983
, vol. 
70
 (pg. 
443
-
46
)
50
Spinelli
JJ
Gallagher
RP
Band
PR
Threlfall
WJ
Multiple myeloma, leukemia and cancer of the ovary in cosmetologists and hairdressers
Am J Ind Med
1984
, vol. 
6
 (pg. 
97
-
102
)
51
Lamba
AB
Ward
MH
Weeks
JL
Dosemeci
M
Cancer mortality patterns among hairdressers and barbers in 24 US states, 1984 to 1995
J Occup Environ Med
2001
, vol. 
43
 (pg. 
250
-
58
)
52
Teta
MJ
Walrath
J
Wister Meigs
J
Flannery
JT
Cancer incidence among cosmetologists
J Natl Cancer Inst
1984
, vol. 
72
 (pg. 
1051
-
57
)
53
Gubéran
E
Raymond
L
Sweetnam
PM
Increased risk for male bladder cancer among a cohort of male and female hairdressers in Geneva
Int J Epidemiol
1985
, vol. 
14
 (pg. 
549
-
54
)
54
Kato
I
Tominaga
S
Ikari
A
An epidemiological study on occupation and cancer risk
Jpn J Clin Oncol
1990
, vol. 
20
 (pg. 
121
-
27
)
55
Andersen
A
Barlow
L
Engeland
A
Kjaerheim
K
Lynge
E
Pukkala
E
Work-related cancer in the Nordic countries
Scand J Work Environ Health
1999
, vol. 
25
 (pg. 
1
-
116
)
56
Czene
K
Tiikkaja
S
Hemminki
K
Cancer risk in hairdressers: assessment of carcinogenicity of hair dyes and gels
Int J Cancer
2003
, vol. 
105
 (pg. 
108
-
12
)
57
Guidotti
S
Wright
WE
Peters
JM
Multiple myeloma and cosmetologists
Am J Ind Med
1982
, vol. 
3
 (pg. 
169
-
71
)
58
Flodin
U
Frederiksson
M
Persson
B
Multiple myeloma and engine exhausts, fresh wood, and creosote: a case-referent study
Am J Ind Med
1987
, vol. 
12
 (pg. 
519
-
29
)
59
Persson
B
Dahlander
AM
Fredriksson
M
Brage
HN
Ohlson
CG
Axelson
O
Malignant lymphomas and occupational exposures
Br J Ind Med
1989
, vol. 
46
 (pg. 
516
-
20
)
60
Eriksson
M
Karlsson
M
Occupational and other environmental factors and multiple myeloma: a population based case-control study
Br J Ind Med
1992
, vol. 
49
 (pg. 
95
-
103
)
61
Pottern
LM
Heineman
EF
Olsen
JH
Raffn
E
Blair
A
Multiple myeloma among Danish women: employment history and workplace exposures
Cancer Causes Control
1992
, vol. 
3
 (pg. 
427
-
32
)
62
Blair
A
Linos
A
Stewart
PA
, et al. 
Evaluation of risks for non-Hodgkins lymphoma by occupation and industry exposures from a case-control study
Am J Ind Med
1993
, vol. 
23
 (pg. 
301
-
2
)
63
Herrinton
LJ
Weiss
NS
Koepsell
TD
, et al. 
Exposure to hair-coloring products and the risk of multiple myeloma
Am J Public Health
1994
, vol. 
84
 (pg. 
1142
-
44
)
64
Figgs
LW
Dosemeci
M
Blair
A
Risk of multiple myeloma by occupation and industry among men and women: a 24-state death certificate study
J Occup Med
1994
, vol. 
36
 (pg. 
1210
-
21
)
65
Mele
A
Szklo
M
Visani
G
, et al. 
Hair dye use and other risk factors for leukemia and pre-leukemia: a case-control study. Italian Leukemia Study Group
Am J Epidemiol
1994
, vol. 
139
 (pg. 
609
-
19
)
66
Miligi
L
Seniori Costantini
A
Crosignani
P
Fontana
A
Masala
G
Nanni
O
Occupational, environmental, and life-style factors associated with the risk of hematolymphopoietic malignancies in women
Am J Ind Med
1999
, vol. 
36
 (pg. 
60
-
69
)
67
Seniori Constantini
A
Miligi
L
Kriebel
D
, et al. 
A multicenter case-control study in Italy on hematolymphopoietic neoplasms and occupation
Epidemiology
2001
, vol. 
12
 (pg. 
78
-
87
)
68
Shibata
A
Sasaki
R
Hamajima
N
Aoki
K
Mortality of hematopoietic disorders and hair dye use among barbers
Acta Haematol Jpn
1990
, vol. 
53
 (pg. 
116
-
18
)
69
Wynder
EL
Onderdonk
J
Mantel
N
An epidemiological investigation of cancer of the bladder
Cancer
1963
, vol. 
16
 (pg. 
1388
-
407
)
70
Anthony
HM
Thomas
GM
Tumors of the urinary bladder: an analysis of the occupation of 1,030 patients in Leeds, England
J Natl Cancer Inst
1970
, vol. 
45
 (pg. 
879
-
95
)
71
Cole
P
Hoover
R
Friedell
GH
Occupation and cancer of the lower urinary tract
Cancer
1972
, vol. 
29
 (pg. 
1250
-
61
)
72
Glashan
RW
Cartwright
RA
Occupational bladder cancer and cigarette smoking in West Yorkshire
Br J Urol
1981
, vol. 
53
 (pg. 
602
-
4
)
73
Schoenberg
JB
Stemhagen
A
Mogielnicki
AP
Altman
R
Abe
T
Mason
TJ
Case-control study of bladder cancer in New Jersey
J Natl Cancer Inst
1984
, vol. 
72
 (pg. 
973
-
81
)
74
Vineis
P
Magnani
C
Occupation and bladder cancer in males: a case-control study
Int J Cancer
1985
, vol. 
35
 (pg. 
599
-
606
)
75
Morrison
AS
Ahlbom
A
Verhoek
WG
, et al. 
Occupation and bladder cancer in Boston
J Epidemiol Community Health
1985
, vol. 
39
 (pg. 
294
-
300
)
76
Baxter
PJ
McDowall
ME
Occupation and cancer in London: an investigation into nasal and bladder cancer using the Cancer Atlas
Br J Ind Med
1986
, vol. 
43
 (pg. 
44
-
49
)
77
Risch
HA
Burch
JD
Miller
AB
, et al. 
Occupational factors and the incidence of cancer of the bladder in Canada
Br J Ind Med
1988
, vol. 
45
 (pg. 
361
-
67
)
78
Jensen
OM
Knudsen
JB
McLaughlin
JK
Sørensen
BL
The Copenhagen case-control study of renal pelvis and ureter cancer: role of smoking and occupational exposures
Int J Cancer
1988
, vol. 
41
 (pg. 
557
-
61
)
79
Silverman
DT
Levin
LI
Hoover
RN
Hartge
P
Occupational risks of bladder cancer in the United States: I. White men
J Natl Cancer Inst
1989
, vol. 
81
 (pg. 
1472
-
80
)
80
Silverman
DT
Levin
LI
Hoover
RN
Occupational risks of bladder cancer among white women in the United States
Am J Epidemiol
1990
, vol. 
132
 (pg. 
453
-
61
)
81
Kunze
E
Chang-Claude
J
Frentzel-Beyme
R
Life style and occupational risk factors for bladder cancer in Germany
Cancer
1992
, vol. 
69
 (pg. 
1776
-
90
)
82
Cordier
S
Clavel
J
Limasset
JC
, et al. 
Occupational risks of bladder cancer in France: a multicentre case-control study
Int J Epidemiol
1993
, vol. 
22
 (pg. 
403
-
11
)
83
Siematycki
J
Dewar
R
Nadon
L
Gérin
M
Occupational risk factors for bladder cancer: results from a case-control study in Montreal, Quebec, Canada
Am J Epidemiol
1994
, vol. 
140
 (pg. 
1061
-
80
)
84
Teschke
K
Morgan
MS
Checkoway
H
, et al. 
Surveillance of nasal and bladder cancer to locate sources of exposure to occupational carcinogens
Occup Environ Med
1997
, vol. 
54
 (pg. 
443
-
51
)
85
Sorahan
T
Hamilton
L
Wallace
DM
Bathers
S
Gardiner
K
Harrington
JM
Occupational urothelial tumours: a regional case-control study
Br J Urol
1998
, vol. 
82
 (pg. 
25
-
32
)
86
Gago-Dominguez
M
Castelao
JE
Yuan
JM
Yu
MC
Ross
RK
Use of permanent hair dyes and bladder-cancer risk
Int J Cancer
2001
, vol. 
91
 (pg. 
575
-
79
)
87
Zheng
T
Cantor
KP
Zhang
Y
Lynch
CF
Occupation and bladder cancer: a population-based, case-control study in Iowa
J Occup Environ Med
2002
, vol. 
44
 (pg. 
685
-
91
)
88
Canadian Cancer Registries Epidemiology Group
Gaertner
RR
Trpeski
L
Johnson
KC
A case-control study of occupational risk factors for bladder cancer in Canada
Cancer Causes Control
2004
, vol. 
15
 (pg. 
1007
-
19
)
89
Golka
K
Heitmann
P
Gieseler
F
, et al. 
Elevated bladder cancer risk due to colorants—a state-wide case-control study in North Rhine-Westphalia, Germany
J Tox Env Health
2008
, vol. 
71
 (pg. 
851
-
55
)
90
Alderson
M
Cancer mortality in male hairdressers
J Epidemiol Community Health
1980
, vol. 
34
 (pg. 
182
-
85
)
91
Dubrow
R
Wegman
DH
Cancer and occupation in Massachusetts: a death certificate study
Am J Ind Med
1984
, vol. 
683
 (pg. 
207
-
30
)
92
Pearce
NE
Howard
JK
Occupation, social class and male cancer mortality in New Zealand 1974–78
Int J Epidemiol
1986
, vol. 
15
 (pg. 
456
-
62
)
93
Gallagher
RP
Threlfall
WJ
Band
PR
Spinelli
JJ
Occupational mortality in British Columbia 1950–1984
1989
Vancouver (BC)
The Cancer Control Agency of British Columbia and the Workers’ Compensation Board of British Columbia
94
Dunham
LJ
Rabson
AS
Stewart
HL
Frank
AS
Young
JL
Rates, interview and pathology study of cancer of the urinary bladder in New Orleans, Louisiana
J Natl Cancer Inst
1968
, vol. 
41
 (pg. 
683
-
709
)
95
Hrubec
Z
Blair
AE
Rogot
E
Vaught
J
Mortality Risk by Occupation among US Veterans of Known Smoking Status 1954–1980
1992
Bethesda, MD
National Institutes of Health
96
Osorio
AM
Bernstein
L
Garabrant
DH
Peters
JM
Investigation of lung cancer among female cosmetologists
J Occup Med
1986
, vol. 
28
 (pg. 
291
-
95
)
97
Jahn
I
Ahrens
W
Brüske-Hohlfeld
I
, et al. 
Occupational risk factors for lung cancer in women: results of a case-control study in Germany
Am J Ind Med
1999
, vol. 
36
 (pg. 
90
-
100
)
98
Bofetta
P
Richiardi
L
Berrino
F
, et al. 
Occupation and larynx and hypopharynx cancer: an international case-control study in France, Spain and Switzerland
Cancer Causes Control
2003
, vol. 
14
 (pg. 
203
-
12
)
99
Garfinkel
J
Selvin
S
Brown
SM
Brief communication: possible increased risk of lung cancer among beauticians
J Natl Cancer Inst
1977
, vol. 
58
 (pg. 
141
-
43
)
100
Leigh
JP
Occupations, cigarette smoking, and lung cancer in the epidemiological follow-up to the NHANES I and the California Occupational Mortality Study
Bull N Y Acad Med
1996
, vol. 
73
 (pg. 
370
-
97
)
101
Ji
J
Hemminki
K
Occupation and upper aerodigestive tract cancers: a follow-up study in Sweden
J Occup Environ Med
2005
, vol. 
47
 (pg. 
785
-
95
)
102
Neuberger
JS
Brownson
RC
Morantz
RA
Chin
TDY
Association of brain cancer with dental X-rays and occupation in Missouri
Cancer Detect Prev
1991
, vol. 
15
 (pg. 
31
-
4
)
103
Swanson
GM
Burns
PB
Cancers of the salivary gland: workplace risks among women and men
Ann Epidemiol
1997
, vol. 
7
 (pg. 
369
-
74
)
104
Galiotte
MP
Kohler
P
Mussi
G
Figaro Gattas
GJ
Assessment of occupational genotoxic risk among Brasilian hairdressers
Ann Occup Hyg
2008
105
Gago-Dominguez
M
Bell
DA
Watson
MA
, et al. 
Permanent hair dyes and bladder cancer: risk modification by cytochrome P4501A2 and N-acetyltransferase 1 and 2
Carcinogenesis
2003
, vol. 
24
 (pg. 
483
-
89
)
106
Morton
LM
Bernstein
L
Wang
SS
, et al. 
Hair dye use, genetic variation in N-acetyltransferase 1 (NAT1) and 2 (NAT2), and risk of non-Hodgkin lymphoma
Carcinogenesis
2007
, vol. 
28
 (pg. 
1759
-
64
)
107
Greenland
S
Basic methods for sensitivity analyses of biases
Int J Epidemiol
1996
, vol. 
25
 (pg. 
1107
-
16
)
108
Spencer
AB
Estill
CF
McCammon
JB
Mickelson
RL
Johnston
OE
Control of ethyl methacrylate exposures during the application of artificial fingernails
Am Ind Hyg Assoc
1997
, vol. 
58
 (pg. 
214
-
18
)
109
Ronda
E
Hollund
BE
Moen
BE
Airborne exposure to chemical substances in hairdresser salons
Environ Monit Assess
2009
, vol. 
153
 (pg. 
83
-
93
)
110
Andersen
A
Barlow
L
Engeland
A
Kjaerheim
K
Lynge
E
Pukkala
E
Work-related cancer in the Nordic countries
Scand J Work Environ Health
1999
, vol. 
25
 
Suppl 2
(pg. 
1
-
116
)