Advertisement

Cancer Causes & Control

, 19:1391 | Cite as

Physical activity and head and neck cancer risk

  • Michael F. Leitzmann
  • Corinna Koebnick
  • Neal D. Freedman
  • Yikyung Park
  • Rachel Ballard-Barbash
  • Albert R. Hollenbeck
  • Arthur Schatzkin
  • Christian C. Abnet
Open Access
Original Paper

Abstract

Objective

To investigate the relation of physical activity to head and neck cancer.

Methods

We prospectively examined the association between physical activity and head and neck cancer in 487,732 men and women, who, at baseline in 1995–1996, were 50–71 years old and free of cancer and emphysema. Follow-up occurred through 31 December 2003.

Results

During follow-up, 1,249 participants developed head and neck cancer, of which 42.0%, 18.9%, and 32.5% were located in the oral cavity, pharynx, and larynx, respectively. In analyses adjusted for age and gender, the relative risks (RR) of head and neck cancer for increasing frequency of physical activity (0, < 1, 1–2, 3–4, and ≥5 times per week) were 1.0 (reference), 0.76, 0.66, 0.57, and 0.62 (95% CI = 0.52–0.74), respectively (p for trend < 0.001). After multivariate adjustment including smoking, the relation was attenuated and became statistically non-significant (RR comparing extreme physical activity categories = 0.89, 95% CI = 0.74–1.06; p for trend = 0.272). In analyses of head and neck cancer subtypes, the corresponding RRs for cancers of the oral cavity, pharynx, and larynx were 0.98 (95% CI = 0.75–1.29), 0.70 (95% CI = 0.45–1.08), and 0.82 (95% CI = 0.59–1.13), respectively.

Conclusions

Our findings suggest that physical activity is unlikely to play an important role in the prevention of head and neck cancer.

Keywords

Head and neck cancer Oral cavity cancer Pharynx cancer Larynx cancer Physical activity 

Introduction

Head and neck cancer is a significant global health problem, accounting for over 600,000 new cancers diagnosed each year [1]. Head and neck cancer includes tumors of the oral cavity, pharynx, and larynx [2]. The incidence rate of head and neck cancer is three- to fourfold higher among men than women [1]. Tobacco use and alcohol drinking have been consistently associated with increased risk of head and neck cancer and these two variables account for 75% of head and neck cancer cases [3]. Few other modifiable lifestyle factors have been identified that may affect this highly fatal cancer [2, 4].

Increasing evidence suggests that physical activity plays an important role in the prevention of cancer [5]. Physical activity may influence head and neck carcinogenesis specifically because physical activity modulates specific mucosal immune parameters, such as salivary immunogloblin (Ig) A [6, 7, 8, 9] and saliva composition has been linked to head and neck cancer risk due to persistent saliva exposure of the epithelial mucosa of the oral cavity, pharynx, and larynx [10].

Despite the global significance of head and neck cancer and the possibility of a preventive physical activity mechanism, little attention has been directed toward exploring the association between physical activity and head and neck cancer. Available information comes from three previous studies of squamous head and neck cancers [11, 12, 13]. Those three investigations [11, 12, 13] observed no association between physical activity and individual cancer sites within the head and neck. No study has evaluated the association between physical activity and total head and neck cancer.

We prospectively examined physical activity in relation to subsequent incidence of head and neck cancer in a large study of initially healthy middle-aged and elderly men and women from the United States (U.S.). Due to possible distinct etiologies of cancers of the oral cavity, pharynx, and larynx, we explored whether associations with physical activity varied by cancer site within the head and neck.

Material and methods

Study population

The NIH-AARP Diet and Health Study is a prospective cohort that was established in 1995–1996 when 566,402 members of AARP (formerly known as American Association of Retired Persons) aged 50–71 years and residing in one of six U.S. states (CA, FL, LA, NJ, NC, and PA) or two metropolitan areas (Atlanta, GA, and Detroit, MI) returned a mailed questionnaire on medical history, diet, and physical activity [14]. Of responding individuals, we excluded persons who reported a previous diagnosis of cancer other than non-melanoma skin cancer (n = 52,561) or emphysema (n = 13,764), those with missing information on physical activity (n = 5,705), and those with missing or inconsistent information on smoking habits (n = 6,640). The analytic cohort of the present report includes 487,732 subjects (295,253 men and 192,479 women). The study was approved by the Special Studies Institutional Review Board (IRB) of the U.S. National Cancer Institute.

Cohort follow-up and endpoint ascertainment

Cohort follow-up was performed by regular linkage to the National Change of Address database maintained by the U.S. Postal Service and through processing of undeliverable mail, other address change update services, and directly from cohort members’ notifications. Vital status was ascertained by linkage of the cohort to the Social Security Administration Death Master File in the U.S. Follow-up searches of presumed deaths in the National Death Index Plus provided verification and information on cause of death. For matching purposes, we have virtually complete data on first and last name, address history, gender, and date of birth. Study participants were followed-up through 31 December 2003.

Incident cases of head and neck cancer were identified by probabilistic linkage to the state cancer registries serving our cohort. We recently expanded our cancer registry ascertainment area by three states (TX, AZ, and NV) to capture cancer cases occurring among participants who moved to those states during follow-up. The North American Association of Central Cancer Registries certifies all eleven cancer registries [15]. We conducted a validation study comparing registry findings to self-reports and medical records and found that approximately 90% of all cancer cases in our cohort were validly identified using linkage to cancer registries [16].

The endpoints in the current analysis were classified by anatomic site and histologic code according to the International Classification of Disease for Oncology (ICD-O), third edition [17]. All newly incident cases of squamous head and neck cancer (histology code 8050–8076) were considered for analysis. Oral cavity cancers included tumors of the lips (C00.1–C00.9), tongue (C01.9–C02.9), gums (C03.0–C03.9), floor of the mouth (C04.0–C04.9), palate (C05.0–C05.9), and other parts of the mouth (C06.0–C06.9). Cancers of the pharynx included tumors of the tonsil (C09.0–C09.9), oropharynx (C10.0–C10.9), piriform sinus (C12.9), hypopharynx (C13.0–C13.9), and pharynx not otherwise specified (NOS) (C14.0). Laryngeal cancer included tumors with site codes C32.0–C32.9 and squamous histology. The overarching category of head and neck cancer included cancers of the oral cavity, pharynx, larynx, and squamous cell carcinomas at other anatomical sites of the head and neck or overlapping regions of the lip, oral cavity, and pharynx.

Physical activity assessment

At baseline, a mailed questionnaire inquired about physical activity during the previous year, defined as the frequency each week spent at activities that lasted 20 min or more and caused either increases in breathing or heart rate or working up a sweat. Six possible response options were given: never; rarely; 1–3 times per month; 1–2 times per week; 3–4 times per week; and 5 or more times per week. Our physical activity assessment corresponds to the American College of Sports Medicine (ACSM) physical activity guidelines that recommend at least 20 min of continuous vigorous exercise three times per week for improving cardio-respiratory fitness [18]. A questionnaire very similar to the one used in our cohort showed good reliability (percentage agreement = 0.76; kappa = 0.53) and reasonable validity (percentage agreement = 0.71; kappa = 0.40) as assessed by a computer science and applications (CSA) physical activity monitor [19].

In a subset of study participants we collected data on light and moderate to vigorous intensity physical activity. We used that information to evaluate associations with less vigorous forms of activity.

Statistical analysis

All statistical analyses were conducted using SAS release 9.1 (SAS Institute, Cary, NC). Cox proportional hazards regression [20] with person-time as the time scale was used to estimate hazard ratios of head and neck cancer, computed as relative risks (RR) with corresponding 95% CI. Using age as the time scale yielded similar results. We tested for and found no departures from the proportional hazards assumption. Follow-up time was calculated from the scan date of the baseline questionnaire until the first occurrence of one of the following events: diagnosis of head and neck cancer, diagnosis of esophageal or stomach cancer (as a diagnosis of one of those cancers would be associated with increased surveillance of the other sites), date moved out of the cancer registry catchment area, death, or the end of follow-up (31 Dec 2003).

Participants were divided into five categories according to their physical activity level: 0 (less than once per month), <1, 1–2, 3–4, and 5 or more times per week. The group with the lowest physical activity level served as the reference group. Tests of linear trend across increasing categories of physical activity were conducted by assigning the mean level of physical activity for categories and treating that term as a single continuous variable. We assessed head and neck cancer risk in three models, one model adjusting for age and gender, a second model adjusting for age, gender, and a combination of smoking status (never; former; current), time since quitting for former smokers (10+ years; 5–9 years; 1–4 years; <1 year), and smoking intensity for former and current smokers (1–10; 11–20; 21–30; 31–40; 41–60; 61+ cigarettes/day), and a third model additionally adjusting for body mass index (<18.5; 18.5–24.9; 25.0–29.9; 30.0–34.9; 35.0–39.9; ≥40.0 kg/m2), race/ethnicity (White; Black; Hispanic; and other race/ethnicity), education (less than high school; high school; vocational school or some college; college graduate; and postgraduate), marital status (married or living as married; other), family history of cancer (yes; no), intakes of fruit and vegetables combined (quintiles), red meat (quintiles), and alcohol (0; <1; 1–3; >3 servings/day). Risk estimates were calculated for total head and neck cancer and oral, pharyngeal, and laryngeal cancers separately.

In order to examine potential effect modification of the association between physical activity and head and neck cancer, we conducted stratified analyses. We also performed tests for interaction using cross-product terms, the statistical significance of which was evaluated using likelihood-ratio tests. In a subset of study participants, we collected information on non-steroidal anti-inflammatory drug (NSAID) use. We used those data to assess whether relations with physical activity were modified by NSAID use. All p values are based on two-sided tests.

Results

During follow-up, the 487,732 participants accrued 3,518,483 total person-years. The mean (SD) ages at entry and exit were 61.9 (5.4) and 69.1 (5.5) years, respectively. The mean durations (ranges) of follow-up in censored participants without head and neck cancer and those who developed head and neck cancer were 7.2 years (range: 1 day to 8.2 years) and 3.8 years (range: 5 days to 7.8 years), respectively.

At baseline, over half of the participants reported cigarette smoking either at present or in the past, and three-fourths of the study subjects indicated consuming alcohol on a regular basis. Specifically, participants who were current, former, and never smokers at baseline contributed 13.4%, 49.9%, and 36.7%, respectively, of the total person-time. Likewise, those who drank alcohol contributed 76% of person-time, whereas those who abstained from alcohol contributed 24% of person-time.

At study entry, 19.6% of the cohort reported engaging in a minimum of 20 min of physical activity five or more times per week, and 17.9% stated that they engaged in 20 min of continuous activity less than once per month. On average, participants who reported being physically active tended to be leaner, to be college graduates, to be married, and to have higher intakes of fruit, vegetables, and alcohol than their less active counterparts. Active individuals were also less likely to currently smoke than less active participants (Table 1). Physical activity level decreased in a stepwise fashion with increasing category of BMI (data not shown).
Table 1

Baseline characteristics according to physical activity

Characteristicsa

Physical activity (times per week)b

0

<1

1–2

3–4

≥5

Participants (n)

87,222

66,853

106,058

131,852

95,747

Gender (%)

    Men

50.6

58.5

61.4

68.9

66.6

    Women

49.4

41.5

38.6

37.1

33.4

Smoking status (%)

Current smoker

20.7

17.1

14.2

10.2

9.4

    ≤20 cigarettes/day

12.8

10.9

9.4

7.1

6.3

    >20 cigarettes/day

8.0

6.2

4.8

3.2

3.0

Former smoker

44.8

48.0

49.0

52.5

53.7

    Quit ≥ 10 years ago

32.8

36.7

38.1

41.7

43.2

    Quit 1–9 years ago

12.0

11.3

10.9

10.8

10.4

Never smoker

34.5

34.9

36.8

37.3

36.9

Age (years)

62.0

61.1

61.5

62.2

62.4

Body-mass index (kg/m2)

28.6

27.8

27.2

26.6

26.0

Race

    White

89.2

91.5

92.3

91.6

92.4

    Non-White

10.8

8.5

7.7

8.4

7.6

College education (%)

28.1

37.3

40.9

44.5

44.4

Married or living as married (%)

62.1

68.4

70.8

72.5

73.8

Family history of cancer (%)

50.6

51.8

51.4

51.2

50.9

Fruit and vegetable intakes (servings/1,000 kcal/day)

3.1

3.2

3.4

3.7

3.9

Red meat intake (grams/1,000 kcal/day)

37.8

37.2

36.1

32.4

30.6

Alcohol intake (servings/week)

6.8

6.9

6.7

6.6

7.3

NSAID user (%)

49.4

52.4

51.9

52.3

49.3

aAll values (except age) were directly standardized to the age distribution of the cohort

bPhysical activity is defined as activities that lasted 20 min or more and caused either increases in breathing or heart rate or working up a sweat

We documented 1,249 total head and neck cancer cases, of which 42.0% were located in the oral cavity, 18.9% in the pharynx, 32.5% in the larynx, and 6.6% at other locations of the head and neck. In analyses adjusted for age and gender only, we found a strong inverse association between physical activity and head and neck cancer. Participants who reported engaging in physical activity five or more times per week had a RR of 0.62 (95% CI = 0.52–0.74) compared to those who participated in physical activity less than once per month (Table 2). However, when we further adjusted for smoking the relation was substantially attenuated and became statistically non-significant (RR = 0.86; 95% CI = 0.72–1.03). Additional control for other potential confounding variables including BMI, race/ethnicity, marital status, family history of any cancer, education, intakes of fruit and vegetables, red meat, and alcohol had only minor influence on the risk estimate (RR = 0.89; 95% CI = 0.74–1.06).
Table 2

Relative risk of total head and neck cancer and head and neck cancer subtypes according to physical activity

Head and neck cancer type

Physical activity (times per week)a

p for trend

0

<1

1–2

3–4

≥5

Person-years

616,503

482,118

767,821

957,476

694,565

 

Total head and neck cancer (n = 1,249)

    No. of cases

290

178

256

289

236

 

    Age, gender-adjusted RR (95% CI)b

1.0

0.76 (0.63–0.91)

0.66 (0.56–0.78)

0.57 (0.48–0.67)

0.62 (0.52–0.74)

<0.001

    Age, gender-adjusted RR + smoking (95% CI)b,c

1.0

0.84 (0.69–1.01)

0.79 (0.67–0.94)

0.77 (0.66–0.91)

0.86 (0.72–1.03)

0.142

    Full multivariate RR (95% CI)d

1.0

0.87 (0.72–1.05)

0.84 (0.70–0.99)

0.82 (0.69–0.97)

0.89 (0.74–1.06)

0.272

Oral cavity (n = 525)

    No. of cases

119

70

111

115

110

 

    Age, gender-adjusted RR (95% CI)b

1.0

0.74 (0.55–1.08)

0.71 (0.55–0.92)

0.57 (0.44–0.74)

0.73 (0.56–0.95)

0.015

    Age, gender-adjusted RR + smoking (95% CI)b,c

1.0

0.81 (0.59–1.08)

0.83 (0.64–1.08)

0.73 (0.56–0.95)

0.95 (0.73–1.24)

0.749

    Full multivariate RR (95% CI)d

1.0

0.863 (0.61–1.11)

0.86 (0.66–1.12)

0.77 (0.59–1.00)

0.98 (0.75–1.29)

0.956

Pharynx (n = 236)

    No. of cases

57

35

49

59

36

 

    Age, gender-adjusted RR (95% CI)b

1.0

0.74 (0.49–1.13)

0.63 (0.43–0.93)

0.59 (0.41–0.85)

0.48 (0.32–0.73)

0.001

    Age, gender-adjusted RR + smoking (95% CI)b,c

1.0

0.83 (0.55–1.27)

0.77 (0.53–1.14)

0.82 (0.56–1.18)

0.68 (0.44–1.04)

0.136

    Full multivariate RR (95% CI)d

1.0

0.88 (0.58–1.35)

0.84 (0.57–1.23)

0.88 (0.61–1.29)

0.70 (0.45–1.08)

0.180

Larynx (n = 406)

    No. of cases

97

64

81

95

69

 

    Age, gender-adjusted RR (95% CI)b

1.0

0.81 (0.59–1.11)

0.61 (0.45–0.82)

0.54 (0.41–0.72)

0.52 (0.38–0.71)

<0.001

    Age, gender-adjusted RR + smoking (95% CI)b,c

1.0

0.92 (0.67–1.26)

0.77 (0.57–1.04)

0.79 (0.59–1.06)

0.79 (0.57–1.08)

0.137

    Full multivariate RR (95% CI)d

1.0

0.96 (0.69–1.32)

0.82 (0.60–1.10)

0.84 (0.63–1.12)

0.82 (0.59–1.13)

0.225

aPhysical activity is defined as activities that lasted 20 min or more and caused either increases in breathing or heart rate or working up a sweat

bRR = relative risk. CI = confidence interval

cAdjustment for smoking included the combination of smoking status (never; former; current), time since quitting for former smokers (10+ years; 5–9 years; 1–4 years; <1 year), and smoking intensity for former and current smokers (1–10; 11–20; 21–30; 31–40; 41–60; 61+ cigarettes/day)

dThe multivariate models used age as the underlying time metric and included the following covariates: gender (women; men), body mass index (<18.5; 18.5–24.9; 25.0–29.9; 30.0–34.9; 35.0–39.9; ≥40.0 kg/m2), a combination of smoking status (never; former; current), time since quitting for former smokers (10 + years; 5–9 years; 1–4 years; < 1 year), and smoking intensity for former and current smokers (1–10; 11–20; 21–30; 31–40; 41–60; 61 + cigarettes/day), race/ethnicity (White; Black; Hispanic; and other race/ethnicity), education (less than high school; high school; vocational school or some college; college graduate; and postgraduate), marital status (married or living as married; other), family history of cancer (yes; no), intakes of fruit and vegetables combined (quintiles), red meat (quintiles), and alcohol (0; <1; 1–3; >3 servings/day)

Using information from a subset of participants for whom we had a separate assessment of physical activity that included data on light and moderate to vigorous physical activity, we observed that both light activity (multivariate RR for >7-h activity per week versus no activity = 1.07; 95% CI = 0.83–1.39) and moderate to vigorous activity (multivariate RR for >7-h activity per week versus no activity = 0.81; 95% CI = 0.64–1.03) were not statistically significantly associated with head and neck cancer.

We next evaluated the relation of physical activity to cancers of the oral cavity, pharynx, and larynx separately (Table 2). Similar to the associations observed with total head and neck cancer, for each cancer site, we found inverse relations with physical activity in analyses that were adjusted for age and gender only. The age- and gender-adjusted RRs of cancers of the oral cavity, pharynx, and larynx comparing the highest to the lowest physical activity category were 0.73 (95% CI = 0.56–0.95), 0.48 (95% CI = 0.32–0.73), and 0.52 (95% CI = 0.38–0.71), respectively. After adjustment for smoking, risk estimates became considerably weaker and were rendered statistically non-significant. The impact of control for additional potential confounders was small. The corresponding RRs of cancers of the oral cavity, pharynx, and larynx were 0.98 (95% CI = 0.75–1.29), 0.70 (95% CI = 0.45–1.08), and 0.82 (95% CI = 0.59–1.13), respectively.

We also examined whether the effect of physical activity was modified by potential risk factors for head and neck cancer (Table 3). Null associations between increasing levels of physical activity and risk of total head and neck cancer were noted across subgroups defined by gender, smoking status, age, race/ethnicity, education, BMI, intakes of fruit and vegetables, red meat, alcohol, and NSAID use. Statistically significant tests for interaction were seen for the association between physical activity and head and neck cancer according to gender, education, red meat intake, and alcohol use. However, inspection of the point estimates and the tests for trend across increasing categories of physical activity among participants within strata of those variables revealed no divergent patterns. Similar results were observed for cancer sites within the head and neck (data not shown).
Table 3

Multivariate relative risk of total head and neck cancer according to physical activity in participants defined by selected variables

Variable

No. of cases

Physical activity (times per week)

p for

0

<1

1–2

3–4

≥5

trend

interaction

Gender

    Men

977

1.0

0.86 (0.69–1.07)

0.72 (0.59–0.88)

0.78 (0.65–0.95)

0.82 (0.67–1.00)

0.162

0.029

    Women

272

1.0

0.86 (0.58–1.29)

1.33 (0.96–1.86)

0.90 (0.62–1.31)

1.14 (0.77–1.71)

0.768

 

Smoking status

    Current smoker

487

1.0

0.85 (0.65–1.12)

0.87 (0.68–1.12)

0.79 (0.60–1.04)

0.93 (0.69–1.25)

0.493

0.985

    Former smoker

551

1.0

0.91 (0.67–1.22)

0.78 (0.59–1.03)

0.83 (0.64–1.08)

0.89 (0.68–1.16)

0.567

 

    Never smoker

211

1.0

0.84 (0.50–1.40)

0.89 (0.58–1.39)

0.85 (0.56–1.38)

0.88 (0.56–1.38)

0.688

 

Age at baseline (years)

    <65

718

1.0

0.93 (0.73–1.18)

0.87 (0.69–1.08)

0.75 (0.59–0.95)

0.90 (0.71–1.15)

0.202

0.392

    ≥65

531

1.0

0.77 (0.57–1.06)

0.79 (0.60–1.03)

0.89 (0.69–1.15)

0.87 (0.66–1.14)

0.829

 

BMI (kg/m2)

    <25.0

505

1.0

0.89 (0.66–1.21)

0.89 (0.68–1.18)

0.82 (0.62–1.07)

0.92 (0.69–1.21)

0.526

0.957

    25.0–29.9

512

1.0

0.85 (0.63–1.16)

0.84 (0.64–1.09)

0.87 (0.67–1.14)

0.86 (0.65–1.15)

0.571

 

    ≥30.0

232

1.0

0.85 (0.57–1.25)

0.72 (0.49–1.05)

0.70 (0.47–1.04)

0.95 (0.63–1.45)

0.593

 

Race/ethnicity

    White

1,179

1.0

0.87 (0.72–1.06)

0.84 (0.70–0.99)

0.82 (0.69–0.98)

0.89 (0.75–1.08)

0.357

0.966

    Nonwhite

70

1.0

0.83 (0.39–1.77)

0.81 (0.40–1.62)

0.70 (0.35–1.40)

0.73 (0.35–1.54)

0.359

 

Education

    Some college or less

822

1.0

1.03 (0.83–1.29)

0.90 (0.73–1.11)

0.94 (0.76–1.15)

0.92 (0.74–1.15)

0.385

0.037

    College graduate or postgraduate

427

1.0

0.57 (0.40–0.82)

0.69 (0.51–0.92)

0.61 (0.45–0.82)

0.78 (0.57–1.05)

0.518

 

Fruit and vegetable intakes

    Low

808

1.0

0.82 (0.66–1.03)

0.79 (0.65–0.97)

0.81 (0.66–0.99)

0.80 (0.64–1.01)

0.115

0.352

    High

441

1.0

1.06 (0.74–1.52)

0.99 (0.72–1.38)

0.91 (0.66–1.24)

1.11 (0.81–1.52)

0.703

 

Red meat intake

    Low

510

1.0

1.03 (0.76–1.39)

1.06 (0.80–1.39)

0.76 (0.57–1.00)

0.96 (0.73–1.27)

0.238

0.032

    High

739

1.0

0.79 (0.62–1.00)

0.72 (0.58–0.90)

0.87 (0.71–1.08)

0.85 (0.67–1.08)

0.729

 

Alcohol use

    No

315

1.0

1.08 (0.75–1.57)

1.21 (0.87–1.67)

0.95 (0.68–1.32)

0.84 (0.58–1.21)

0.156

0.031

    Yes

934

1.0

0.81 (0.65–1.00)

0.73 (0.60–0.89)

0.77 (0.63–0.94)

0.89 (0.73–1.10)

0.649

 

NSAID use

    No

329

1.0

0.72 (0.49–1.06)

0.91 (0.66–1.26)

0.77 (0.56–1.07)

0.79 (0.56–1.13)

0.291

0.182

    Yes

307

1.0

0.91 (0.63–1.32)

0.64 (0.45–0.93)

0.72 (0.51–1.01)

0.89 (0.63–1.29)

0.637

 

The multivariate models were adjusted for covariates listed in Table 2 footnote. In each case, the stratification variable was excluded from the model. Within each stratum, the category representing the lowest level of physical activity served as the reference group. NSAID = non-steroidal anti-inflammatory drug. The analysis that was stratified by NSAID use was conducted using data from a sub-cohort of study participants for whom we had collected information regarding NSAID use

Discussion

The findings of the current report—the first to our knowledge to present data on the relation of physical activity to total head and neck cancer—suggest that physical activity is unlikely to play an important role in the development of head and neck cancer. In addition, we detected no significant relationship between physical activity and individual cancer sites of the head and neck. The lack of a statistically significant association between physical activity and total head and neck cancer and its subtypes was consistent across strata of major covariates. In particular, tobacco smoking and alcohol use did not appear to modify results.

Although our risk estimates linking physical activity to head and neck cancer were in the inverse direction, our overall interpretation of a largely null association is consistent with other available studies [11, 12, 13] on the topic. One retrospective cohort study (n = 92 cases) from Denmark [11] compared physically active mail carriers with the general population and reported standardized incidence ratios (SIRs) of 0.91, 1.08, 1.16, 0.97, and 1.31 for individual cancers of the larynx, pharynx, mouth, lip, and tongue, none of which were statistically significant. Similarly, one case-control study of laryngeal cancer (n = 779 cases) from Turkey [12] (OR = 1.20; 95% CI = 0.90–1.60) and one case-control study of laryngeal cancer (n = 285 cases) and buccal cavity cancer (n = 499 cases) from the U.S. [13] observed no statistically significant association with physical activity (OR = 0.5; 95% CI = 0.3–1.0 and OR = 1.1; 95% CI = 0.8–1.7, respectively).

Despite the lack of an association with head and neck cancer observed in our study, we noted some difference in the relation of physical activity to head and neck cancer toward a stronger inverse association in men than women. Physical activity levels were greater among men than women in our study, which suggests that potentially disparate physical activity levels between genders do not explain the greater incidence rate of head and neck cancer among men compared to women [21].

Apart from the true absence of an association between physical activity and head and neck cancer, we considered several possible alternative explanations for our findings. Data on physical activity was assessed using self-report, which generally involves some extent of misclassification [22]. Any random imprecision in measuring physical activity would tend to bias the relationship between physical activity and head and neck cancer toward the null hypothesis. Also, it is possible that we did not capture physical activity at the time during which it plays an important etiologic role in head and neck carcinogenesis.

Insufficient variation in physical activity as a possible reason for the null association is improbable because our physical activity measure showed marked-variation in the expected direction across levels of BMI. Also, greater physical activity on this scale was associated with reduced risk of total mortality and death due to heart disease in our cohort [23]. In addition, a physical activity instrument comparable to the one used in our study has documented validity and reproducibility [19]. Thus, measurement error in our assessment of physical activity is not likely to fully explain the null association in our data. It is possible that our questionnaire format may have been associated with some degree of over-reporting of activity. Circumstantial data indicate that self-administered activity questions can lead to inflated estimates of the reported time spent engaging in physical activity as compared with interviewer-administered assessments [22]. Notwithstanding this potential limitation, the main possible correlates of activity over-reporting, including age and body size were accounted for in our multivariate statistical analyses.

Our study lacked information on participant income and occupation, factors that could confound the relation of physical activity to head and neck cancer. Nonetheless, we would expect uncontrolled confounding by income to result in a spurious exaggeration of a potentially inverse association between physical activity and head and neck cancer. By comparison, confounding by occupation could conceivably have obscured a possible physical activity benefit, because some occupations are associated with high activity levels but low socioeconomic status, a potential risk factor for head and neck cancer [24]. Notwithstanding these caveats, we did control for at least some potential confounding by income and occupation by adjusting for education level, a variable correlated with income and occupation as well as with head and neck cancer [24]. Strict control for tobacco and alcohol as well as for other potential risk factors for head and neck cancer further minimized the potential for confounding.

We did not collect data on infections by human papillomavirus (HPV) and Epstein-Barr virus (EBV), putative risk factors for cancer at some sites in the head and neck [25, 26], but those agents are not considered to be closely associated with physical activity [27] and are therefore unlikely to have affected our results.

Other methodologic biases are probably also not responsible for the lack of an association seen in our data. Specifically, participants with preexisting cancer and emphysema at baseline were removed from the analyses to lessen the impact that malignant or chronic disease may have had on physical activity levels at baseline. Exposure information was gathered prior to cancer diagnosis, which precluded bias ascribable to discrepant recall of physical activity level by participants with and without a diagnosis of head and neck cancer.

Due to the large size of our cohort including more than 1,200 cases of head and neck cancer, insufficient statistical power is not likely to account for the null associations observed in our study. The large number of cases facilitated a detailed exploration of the relations with physical activity across strata of major risk factors for head and neck cancer. We observed null findings within all strata, which gave us confidence that we did not miss a strong inverse association between physical activity and head and neck cancer in our analyses.

In theory, physical activity has the potential to influence head and neck carcinogenesis through its effects on immune function. The impact of physical activity on immunomodulation varies according to the level of exercise. As compared to sitting, low-intensity physical activity, such as walking increases circulating levels of immune parameters, including blood counts for neutrophils, lymphocytes, monocytes, and natural killer cells [28]. Moderate levels of exercise also enhance mucosal immune parameters, such as salivary IgA [6, 7, 9]. In contrast, vigorous levels of exercise decrease T and B cells [29] and transiently suppress natural killer cell cytotoxicity [30] and salivary IgA [8]. We observed statistically non-significant relations for head and neck cancer both with vigorous activity and with less vigorous forms of activity. Even so, the possibility of a physical activity effect on the content or composition of saliva is intriguing, because saliva continuously permeates the mucosal epithelium of the oral cavity, the pharynx and, to a lesser extent the larynx, and therefore bears potential to influence head and neck cancer risk [10].

We conclude that despite the existence of a plausible biological mechanism, physical activity is not likely to substantially impact upon total head and neck cancer risk. Since definitive conclusions cannot be drawn on the basis of findings from the limited body of currently existing data on this topic, the relation of physical activity to head and neck cancer deserves continued attention in future epidemiologic research.

Notes

Acknowledgments

We are grateful to the participants in the NIH-AARP Diet and Health Study for their outstanding cooperation. We thank Leslie Carroll at Information Management Services and Sigurd Hermansen and Kerry Grace Morrissey from Westat for data support, and Tawanda Roy at the Nutritional Epidemiology Branch for research assistance. Financial Support: This research was supported by the Intramural Research Program of the NIH, National Cancer Institute.

Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

References

  1. 1.
    Ferlay J, Bray F, Pisani P et al (2004) Globocan 2000. Cancer incidence, mortality and prevalence worldwide. IARC Cancer Base No. 5, Lyon, France,Google Scholar
  2. 2.
    Davies L, Welch HG (2006) Epidemiology of head and neck cancer in the United States. Otolaryngol Head Neck Surg 135:451–457. doi: 10.1016/j.otohns.2006.01.029 PubMedCrossRefGoogle Scholar
  3. 3.
    Hashibe M, Brennan P, Benhamou S et al (2007) Alcohol drinking in never users of tobacco, cigarette smoking in never drinkers, and the risk of head and neck cancer: pooled analysis in the International Head and Neck Cancer Epidemiology Consortium. J Natl Cancer Inst 99:777–789. doi: 10.1093/jnci/djk179 PubMedCrossRefGoogle Scholar
  4. 4.
    Sturgis EM, Wei Q, Spitz MR (2004) Descriptive epidemiology and risk factors for head and neck cancer. Semin Oncol 31:726–733. doi: 10.1053/j.seminoncol.2004.09.013 PubMedCrossRefGoogle Scholar
  5. 5.
    Westerlind KC (2003) Physical activity and cancer prevention–mechanisms. Med Sci Sports Exerc 35:1834–1840. doi: 10.1249/01.MSS.0000093619.37805.B7 PubMedCrossRefGoogle Scholar
  6. 6.
    Klentrou P, Cieslak T, MacNeil M et al (2002) Effect of moderate exercise on salivary immunoglobulin A and infection risk in humans. Eur J Appl Physiol 87:153–158. doi: 10.1007/s00421-002-0609-1 PubMedCrossRefGoogle Scholar
  7. 7.
    Akimoto T, Kumai Y, Akama T et al (2003) Effects of 12 months of exercise training on salivary secretory IgA levels in elderly subjects. Br J Sports Med 37:76–79. doi: 10.1136/bjsm.37.1.76 PubMedCrossRefGoogle Scholar
  8. 8.
    Novas AM, Rowbottom DG, Jenkins DG (2003) Tennis, incidence of URTI and salivary IgA. Int J Sports Med 24:223–229. doi: 10.1055/s-2003-39096 PubMedCrossRefGoogle Scholar
  9. 9.
    Shimizu K, Kimura F, Akimoto T et al (2007) Effect of free-living daily physical activity on salivary secretory IgA in elderly. Med Sci Sports Exerc 39:593–598. doi: 10.1249/mss.0b013e318031306d PubMedCrossRefGoogle Scholar
  10. 10.
    Reznick AZ, Hershkovich O, Nagler RM (2004) Saliva–a pivotal player in the pathogenesis of oropharyngeal cancer. Br J Cancer 91:111–118. doi: 10.1038/sj.bjc.6601869 PubMedCrossRefGoogle Scholar
  11. 11.
    Soll-Johanning H, Bach E (2004) Occupational exposure to air pollution and cancer risk among Danish urban mail carriers. Int Arch Occup Environ Health 77:351–356. doi: 10.1007/s00420-004-0510-9 PubMedCrossRefGoogle Scholar
  12. 12.
    Dosemeci M, Hayes RB, Vetter R et al (1993) Occupational physical activity, socioeconomic status, and risks of 15 cancer sites in Turkey. Cancer Causes Control 4:313–321. doi: 10.1007/BF00051333 PubMedCrossRefGoogle Scholar
  13. 13.
    Brownson RC, Chang JC, Davis JR et al (1991) Physical activity on the job and cancer in Missouri. Am J Public Health 81:639–642PubMedCrossRefGoogle Scholar
  14. 14.
    Schatzkin A, Subar AF, Thompson FE et al (2001) Design and serendipity in establishing a large cohort with wide dietary intake distributions : the National Institutes of Health-American Association of Retired Persons Diet and Health Study. Am J Epidemiol 154:1119–1125. doi: 10.1093/aje/154.12.1119 PubMedCrossRefGoogle Scholar
  15. 15.
    North American Association of Central Cancer Registries (NAACCR) (2004) Standards for completeness, quality, analysis, and management of data (vol III). North American Association of Central Disease Registries (ed), NAACCR, USAGoogle Scholar
  16. 16.
    Michaud DS, Midthune D, Hermansen S et al (2005) Comparison of cancer registry case ascertainment with SEER estimates and self-reporting in a subset of the NIH-AARP Diet and Health Study. J Regist Manag 32:70–75Google Scholar
  17. 17.
    Fritz AG, Percy C, Jack A et al (2000) International classification of diseases for oncology: ICD-O, 3rd edn. World Health Organization (WHO), Geneva, SwitzerlandGoogle Scholar
  18. 18.
    American College of Sports Medicine (1990) The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness in healthy adults. Med Sci Sports Exerc 22:265–274Google Scholar
  19. 19.
    Marshall AL, Smith BJ, Bauman AE et al (2005) Reliability and validity of a brief physical activity assessment for use by family doctors. Br J Sports Med 39:294–297. doi: 10.1136/bjsm.2004.013771 PubMedCrossRefGoogle Scholar
  20. 20.
    Cox DR (1972) Regression models and lifetables. J R Stat Soc (B) 34:187–220Google Scholar
  21. 21.
    Ragin CC, Modugno F, Gollin SM (2007) The epidemiology and risk factors of head and neck cancer: a focus on human papillomavirus. J Dent Res 86:104–114PubMedGoogle Scholar
  22. 22.
    Sallis JF, Saelens BE (2000) Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport 71:S1–14PubMedGoogle Scholar
  23. 23.
    Leitzmann MF, Park Y, Blair A et al (2007) Physical activity recommendations and decreased risk of mortality. Arch Intern Med 167:2453–2460. doi: 10.1001/archinte.167.22.2453 PubMedCrossRefGoogle Scholar
  24. 24.
    Greenberg RS, Haber MJ, Clark WS et al (1991) The relation of socioeconomic status to oral and pharyngeal cancer. Epidemiology 2:194–200. doi: 10.1097/00001648-199105000-00006 PubMedCrossRefGoogle Scholar
  25. 25.
    Badaracco G, Venuti A, Morello R et al (2000) Human papillomavirus in head and neck carcinomas: prevalence, physical status and relationship with clinical/pathological parameters. Anticancer Res 20:1301–1305PubMedGoogle Scholar
  26. 26.
    Chien YC, Chen JY, Liu MY et al (2001) Serologic markers of Epstein-Barr virus infection and nasopharyngeal carcinoma in Taiwanese men. N Engl J Med 345:1877–1882. doi: 10.1056/NEJMoa011610 PubMedCrossRefGoogle Scholar
  27. 27.
    Pottgiesser T, Wolfarth B, Schumacher YO et al (2006) Epstein-Barr virus serostatus: no difference despite aberrant patterns in athletes and control group. Med Sci Sports Exerc 38:1782–1791. doi: 10.1249/01.mss.0000230122.91264.3f PubMedCrossRefGoogle Scholar
  28. 28.
    Nieman DC, Henson DA, Austin MD et al (2005) Immune response to a 30-minute walk. Med Sci Sports Exerc 37:57–62. doi: 10.1249/01.MSS.0000149808.38194.21 PubMedCrossRefGoogle Scholar
  29. 29.
    Malm C, Ekblom O, Ekblom B (2004) Immune system alteration in response to increased physical training during a five day soccer training camp. Int J Sports Med 25:471–476. doi: 10.1055/s-2004-821119 PubMedCrossRefGoogle Scholar
  30. 30.
    Suzui M, Kawai T, Kimura H et al (2004) Natural killer cell lytic activity and CD56(dim) and CD56(bright) cell distributions during and after intensive training. J Appl Physiol 96:2167–2173. doi: 10.1152/japplphysiol.00513.2003 PubMedCrossRefGoogle Scholar

Copyright information

© The Author(s) 2008

Authors and Affiliations

  • Michael F. Leitzmann
    • 1
  • Corinna Koebnick
    • 2
  • Neal D. Freedman
    • 1
    • 3
  • Yikyung Park
    • 1
  • Rachel Ballard-Barbash
    • 4
  • Albert R. Hollenbeck
    • 5
  • Arthur Schatzkin
    • 1
  • Christian C. Abnet
    • 1
  1. 1.Department of Health and Human Services, Nutritional Epidemiology Branch, Division of Cancer Epidemiology and GeneticsNational Cancer Institute, National Institutes of HealthBethesdaUSA
  2. 2.Research and EvaluationKaiser Permanente Southern CaliforniaPasadenaUSA
  3. 3.Cancer Prevention Fellowship Program, Division of Cancer PreventionNational Cancer InstituteBethesdaUSA
  4. 4.Division of Cancer Control and Population SciencesNational Cancer InstituteBethesdaUSA
  5. 5.AARPWashingtonUSA

Personalised recommendations