Literature DB >> 32746843

The association between recreational screen time and cancer risk: findings from the UK Biobank, a large prospective cohort study.

Ruth F Hunter1, Jennifer M Murray2, Helen G Coleman3.   

Abstract

BACKGROUND: Evidence is suggestive of sedentary behaviour being associated with an increased risk of endometrial cancer, but the evidence base is too limited to draw any conclusions for other cancers. The aim of the study was to investigate the association between recreational screen time and site-specific cancer risk.
METHODS: We analysed data from the prospective UK Biobank cohort study. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for associations between daily recreational screen time (including television (TV) viewing time, computer use time and total screen time) and site-specific cancer risk. Partition models and isotemporal substitution models investigated the impact of substituting recreational screen time with physical activity.
RESULTS: During a mean follow-up of 7.6 years, 28,992 incident cancers were identified among 470,578 adults. A 1-h increase in daily TV viewing time was associated with higher risks of oropharyngeal, oesophago-gastric and colon cancer in fully adjusted models. Participants who reported ≤1, compared with 1- ≤ 3, hours/day of TV viewing time had lower risks of lung, breast, and oesophago-gastric cancer. Findings were inconsistent for daily recreational computer use and daily total recreational screen time. The majority of observed associations were small, and were attenuated after excluding cancers diagnosed within the first two years of follow-up, except for oesophago-gastric and colon cancers (HR 1.05, 95% CI: 1.01, 1.10; and HR 1.04, 95% CI: 1.01, 1.07 per 1-h increase in daily TV viewing time, respectively). However, isotemporal substitution models showed reduced risk of some site-specific (oropharyngeal, lung, breast and colorectal) cancers when replacing 1-h/day of TV viewing with 1-h of moderate-intensity physical activity or walking.
CONCLUSIONS: Our findings show that daily recreational screen time, particularly TV viewing, was associated with small increased risks of oesophago-gastric and colon cancer. Replacing 1-h/day of TV viewing with 1-h of moderate-intensity physical activity or walking was associated with lower risk of oropharyngeal, lung, breast and colorectal cancers. Further research from other large prospective cohort studies is required, while mechanistic research is warranted to enhance the biological plausibility of these findings.

Entities:  

Keywords:  Cancer; Cohort study; Epidemiology; Screen time; Sedentary behaviour

Mesh：

Year:  2020        PMID： 32746843      PMCID： PMC7398067          DOI： 10.1186/s12966-020-00997-6

Source DB:  PubMed          Journal:  Int J Behav Nutr Phys Act        ISSN： 1479-5868            Impact factor:   6.457


Introduction

Research in sedentary behaviours has grown rapidly over recent years [1]. Such behaviours are seen as distinct from physical inactivity or sleep, and have been defined as “any waking behaviour characterised by an energy expenditure ≤1.5 metabolic equivalents (METs), while in a sitting, reclining or lying posture” [1, 2]. This definition is typically operationalised as self-reported sitting (including in recreational and occupational activities), television (TV) viewing or other screen-time. The most recent UK Chief Medical Officers’ Physical Activity Guidelines lists behaviours such as TV viewing and computer-use as examples of sedentary behaviour, highlighting that self-reported screen time is among the most common measures of sedentary behaviour cited in the literature [3]. Screen-time can take many forms including social media use, internet use, gaming, general Smartphone use, watching TV and computer use (regardless of what these devices are used for) [4]. The UK Government guidance on sedentary behaviours, published in 2011 and 2019, suggests that we should minimise time spent in prolonged sedentary behaviours for health benefits [3, 5]. However, owing to the relative early stage of the evidence base, no further recommendations were provided around a timeframe for what would be deemed a ‘harmful’ level of sedentary time exposure. Even the most recent US guidance published in 2018 does not provide more specific recommendations for minimising sedentary time [6]. Evidence demonstrates that prolonged sedentary time is associated with increased risk of non-communicable diseases (NCDs). Mechanistically, sedentary behaviour is thought to impact particularly on cardio-metabolic diseases through adverse effects on lipid and glucose metabolism [7, 8]. Recent evidence from a meta-analysis has demonstrated a significant direct association between 6 and 8 h daily sedentary time and increased all-cause mortality, cardiovascular disease mortality and Type 2 Diabetes Mellitus risk [9]. Prolonged sedentary behaviour is therefore a significant burden on our healthcare systems. In 2016–2017, for example, it was estimated to cost the UK National Health Service £0.8 billion [10]. However, much less is known about sedentary behaviour and cancer, and known biological mechanisms are less well understood [11]. The World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) global report in 2018 stated that evidence on sedentary behaviours is limited but is suggestive as being associated with an increased risk of endometrial cancer (pooled risk estimate from three studies comparing the highest versus lowest levels of sitting time was 1.46, 95% CI: 1.21, 1.76, cases = 1579) [11-14]. The evidence base was deemed to be too limited to draw any conclusions for other cancers [11]. However, in a 2018 meta-analysis, Patterson et al., demonstrated significant linear associations of TV viewing with cancer mortality (N = 4 studies; relative risk [RR] 1.02, 95% CI: 1.01, 1.03 per 1-h increase in TV viewing/day) [9]. More recent evidence from analyses of the large, prospective UK Biobank cohort shows mixed evidence for an association between sedentary behaviour and cancer outcomes [15]. Celis-Morales et al. (2018) found significant associations of discretionary (or recreational) screen-time (time spent in TV viewing or computer screen use during leisure time) exposure and all-cause mortality (hazard ratio [HR] 1.06, 95% CI: 1.05, 1.07), and cancer incidence (HR 1.04, 95% CI: 1.03, 1.04). This study also found that the associations between longer recreational screen time and adverse health outcomes were strongest for participants with low physical activity, cardiorespiratory fitness and grip strength and markedly attenuated for participants with the highest levels of physical activity, cardiorespiratory fitness and grip strength [15]. Our research group have previously found no evidence for an association between recreational screen time and oesophago-gastric cancer risk within the UK Biobank cohort [16]. In contrast, higher levels of TV viewing time were associated with a greater risk of colon cancer in the same study population (HR for ≥5 h/day vs ≤ 1 h/day = 1.32, 95% CI: 1.04, 1.68) [17], although time spent using computers (excluding using a computer at work) was not associated with colorectal cancer risk in the UK Biobank cohort [17]. The findings of a 2017 meta-analysis including six studies also demonstrated significant associations between the highest compared with the lowest levels of occupational sedentary behaviour, and risk of colon cancer (pooled RRs 1.44, 95% CI: 1.28, 1.62) [18]. On the other hand, there was little evidence of an association between sedentary behaviour and rectal cancer risk [18]. Many of the previous studies investigating the association between sedentary behaviour and health outcomes have attempted to adjust for physical activity levels in their analysis. A recent US Government report has highlighted limited evidence on the role of physical activity in displacing the mortality risks associated with sedentary behaviour [6]. Previous research has demonstrated that high levels of physical activity can attenuate the risks associated with sedentary behaviour [19, 20]. An improved understanding of these interactive effects would enable more specific recommendations to be made regarding quantifying prolonged sedentary time. Much of the previous research has modelled the joint effects of physical activity and sedentary behaviour [20]. However, other analytical approaches such as isotemporal substitution and partition models [19], enable us to model replacing sedentary behaviour with physical activity which may be a more time efficient method of promoting healthy behaviour. Analytical techniques such as partition models and isotemporal substitution models [21] could help to model such predictions, but have yet to be extensively applied in large cohort analyses. Therefore, this study aimed to add to the relatively scant evidence base [11] by investigating daily recreational screen time (including TV viewing, computer use and total screen-use) in relation to the risk of site-specific cancers in the large UK Biobank cohort study. Partition and isotemporal substitution models were also used to investigate the impact of substituting recreational screen time with physical activity, in relation to site-specific cancer risk.

Methods

Study design

Between 2006 and 2010, UK Biobank recruited a cohort of 502,619 adults (5.5% response rate) aged 40–69 years from the general population [22, 23]. Approximately 9.2 million invitations were mailed to potential participants who were registered with the National Health Service (NHS) and living within a 25-mile radius of one of the 22 assessment centres across England, Scotland and Wales. From this overall cohort, we excluded participants if: [1] they had been diagnosed with malignant cancer (excluding non-melanoma skin cancer) at baseline (n = 26,868); and [2] they did not complete the self-report assessments of their daily TV viewing time (n = 5078), daily recreational computer time (n = 8000) or daily total recreational screen time (n = 11,232) [3]; they requested to be removed from the UK Biobank dataset as per General Data Protection Regulation (GDPR) (n = 95). This resulted in 470,578 participants being included in the analysis for daily TV viewing time, 467,656 participants being included in the analysis for daily recreational computer time and 464,424 participants being included in the analysis for daily total recreational screen time. All participants provided informed consent.

Screen time assessment

We have used daily TV viewing time as our primary exposure. Firstly, TV viewing time was almost three times more prevalent as a recreational sedentary behaviour than computer use within this UK population. Secondly, we were concerned that using daily total recreational screen time as the primary analysis may overestimate total screen time through double counting (if participants watched TV and used computers at the same time). Therefore, we did not feel that total screen time was an appropriate focus for our primary analysis. Thirdly, as acknowledged in a recent study [24], a large body of research has focussed on TV viewing as a primary exposure representing a type of sedentary behaviour, demonstrating consistent associations, particularly with CVD risk, in population cohort studies akin to the UK Biobank. Patterson et al., (2018) also highlights that daily TV viewing time may show stronger associations with health outcomes, and also may be one of the most amenable types of sedentary behaviour [9]. Relevant screen-time exposure variables were assessed by self-reported time spent watching TV, and time spent using the computer outside of work, which were used to derive daily total recreational screen time. Self-reported TV viewing time was assessed for all participants by asking the following question: “In a typical DAY, how many hours do you spend watching TV? (Put 0 if you do not spend any time doing it)?” Self-reported daily recreational computer use time was assessed for all participants by asking the following question: “In a typical DAY, how many hours do you spend using the computer? (Do not include using a computer at work; put 0 if you do not spend any time doing it).” Durations of < 0 h were set to missing, as were responses of “Do not know” or “Prefer not to answer”. If the respondent replied “Less than an hour a day”, this was recoded to 0.5 h. Daily total recreational screen time was then computed as the sum of hours spent watching TV and hours spent using the computer. If the summation of total hours spent watching TV and hours spent using the computer was greater than 24, this was set to missing (n = 35).

Physical activity assessment

Self-report physical activity was assessed for all participants using the validated short-form International Physical Activity Questionnaire (IPAQ) [25] on which participants reported the frequency (i.e. days/week) and duration (i.e. minutes/day) of walking, moderate- and vigorous-intensity physical activity in the past seven days. For each domain (walking, moderate, vigorous), durations of < 10 min/day were recoded to 0 and durations of > 180 min were truncated at 180 min/day in line with IPAQ processing rules. This was used to derive hours/day spent in walking, moderate- and vigorous-intensity physical activity. A categorical variable was derived representing participants’ IPAQ physical activity category and participants were classified as ‘low’ (not meeting criteria for the ‘moderate’ or ‘high’ categories), ‘moderate’ (at least 20 min of vigorous-intensity physical activity on three or more days/week; at least 30 min of moderate-intensity physical activity or walking on five or more days/week; five or more days/week spent in any combination of walking, moderate- or vigorous-intensity physical activity achieving at least 600 MET minutes/week) or ‘high’ (at least three days/week of vigorous-intensity physical activity achieving at least 1500 MET minutes/week; at least seven days/week spent in any combination of walking, moderate- or vigorous-intensity physical activity achieving at least 3000 MET minutes/week) activity levels. All data processing was carried out according to official IPAQ rules [26].

Assessment of covariates

Height (m), weight (kg), and waist and hip circumference (cm) were measured by staff at the UK Biobank study centre. Body mass index (BMI) was then calculated from the weight and height measurements (kg/m2). Waist circumference measurements were taken from the level of the umbilicus and regarded as a measure of central obesity, using official cut-off values established by the International Diabetes Federation (> 94 cm in men and > 80 cm in women) [27]. Age, sex and postcodes were acquired from a central registry for all participants and updated by the participant. Participants also self-reported their ethnicity, educational attainment, lifestyle behaviours (smoking status, alcohol consumption, dietary intake, and sunscreen/ultraviolet (UV) protection use) and medical history using electronic questionnaires. Townsend deprivation scores were derived from postcodes [28]. Core confounders for all models included socio-demographic factors (i.e. age, sex, ethnicity, educational attainment and deprivation index), smoking status, alcohol consumption, fruit and vegetable consumption, BMI, height and waist-hip ratio. Cancer site-specific confounders included use of sun/UV protection (melanoma), self-reported oesophageal reflux (oesophagus cancer), diabetes at baseline (pancreatic and colorectal cancers), aspirin use (colorectal cancers), red and processed meat intake (colorectal cancers), hormone replacement therapy (HRT) use (breast, uterus and colorectal cancers), oral contraceptive use (breast and uterus cancers), number of live births (breast and uterus cancers), age at menarche (breast and uterus cancers), age at menopause (breast and uterus cancers), hysterectomy status (breast and uterus cancers) and self-reported family history of cancer (lung, prostate, and breast cancers), based on known aetiological risk factors for these tumours. Proportions of missing data were less than 1% for all variables apart from aspirin use (1.9%), red meat intake (1.1%), age at menarche (1.6%), age at menopause (2.1%), education (1.5%), fruit and vegetable consumption (2.6%), hysterectomy status (5.9%), family history of cancer (1.5%), daily moderate-intensity physical activity (14.0%), daily vigorous-intensity physical activity (10.5%) and daily walking time (12.7%). In multivariable models adjusting for the specific factors listed, we conducted a complete-case analysis restricted to individuals who did not have missing information.

Cancer ascertainment

For the present analysis, the main outcomes were incident site-specific cancers. Incident cancers for participants in the UK Biobank cohort were identified through records maintained at national cancer registries (Health and Social Care Information Centre and the NHS Central Register) and identified from the International Classification of Diseases, 9th and 10th revisions (ICD-9 and ICD-10 [29]). Cancer outcomes were coded according to ICD-9 and ICD-10 as follows: melanoma (ICD-10: C43; ICD-9: 172), oropharyngeal cancers (ICD-10: C00-C14; ICD-9: 140–149), lung (ICD-10: C33-C34; ICD-9: 162), breast: female only (ICD-10: C50; ICD-9: 174), uterus (ICD-10: C54; ICD-9: 182), ovary (ICD-10: C56; ICD-9: 183), prostate (ICD-10: C61; ICD-9: 185), oesophagus (ICD-10: C15; ICD-9: 150), stomach (ICD-10: C16; ICD-9: 151), hepatobiliary tract (ICD-10: C22-C24; ICD-9: 155–156), pancreatic (ICD-10: C25; ICD-9: 157), kidney (ICD-10: C64-C65; ICD-9: 189.0–189.1), bladder (ICD-10: C66-C67; ICD-9: 188, 189.2), colorectal (ICD-10: C18-C21; ICD-9: 153–154), colon (ICD-10: C18; ICD-9: 153), rectum (ICD-10: C19-C20; ICD-9: 1540–1541), brain tumours (ICD-10: C71; ICD-9: 191), thyroid (ICD-10: C73; ICD-9: 193), and haematological malignancies (ICD-10: C81-C96; ICD-9: 200–208), including separate analysis of non-Hodgkin’s lymphoma (ICD-10: C82-C85; ICD-9: 200, 202).

Statistical analyses

Our statistical analyses addressed the following research questions: 1. What is the association between daily recreational screen time (i.e. TV viewing, computer use and total screen time) and site-specific cancers (including endometrial, colorectal, pre- and post-menopausal breast, prostate, lung, and other cancers)? 2. How do these associations vary by gender, age, socio-economic status, smoking and excess body weight? 3. What is the effect of replacing TV viewing time with physical activity on site-specific cancer risk? Descriptive statistics for all covariates are presented according to participants’ total daily TV viewing time. Categorical variables are presented as participant numbers and percentages. Means and standard deviations (SDs) are presented for continuous variables. Follow-up time in days from baseline was used as the timescale, and for each participant end of follow-up occurred at: [1] cancer diagnosis date [2]; date of emigration; (3) date of death; or (4) end of follow-up (14th December 2016), whichever came first. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) showing the relationship between a 1-h increase/day in TV viewing time and cancer. All analyses were adjusted for age and sex in the baseline model. Additional covariates were added in the second adjusted model and included ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, Other, None), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]), alcohol intake (never, former, current [Cancer site-specific covariates were included in the third adjusted model for each type of cancer (details included in the footnotes of Tables 2-6). These included use of sun/UV protection, HRT use, oral contraceptive use, number of live births, age at menarche, age at menopause, hysterectomy status, diabetes at baseline, aspirin use, red meat intake, and processed meat intake. For analyses including gender-specific covariates (e.g. colorectal cancer, colon cancer and rectum cancer), separate models were run for males and females and HRs were combined using inverse variance meta-analysis and a fixed-effects model [30-32]. Participants were excluded from the analysis if they did not have the complete exposure and covariate data required for each model. We did not adjust for total dietary energy intake as the large amount of missing data (for 57.6% of participants) made this unfeasible. Further analyses were conducted to investigate the role of central adiposity by running all models with and without adjustment for waist-hip ratio. Models including incident breast cancer, prostate cancer and lung cancer were run with and without adjustment for self-report family history (mother, father, siblings). The oesophageal cancer model was also run with and without adjustment for self-reported gastro-oesophageal reflux disease (GORD).
Table 2

Results of Cox proportional hazards analyses investigating the association between self-report daily TV viewing time and cancer incidence

1 h increase in TV viewing timep-value≤1 h1- ≤ 3 h (reference)3- ≤ 5 h> 5 h
Person-years3,526,324736,5371,781,542818,674189,571
Skin, melanomaCases163531583140485
HR (95% CI)*0.98 (0.95 1.01)0.24**0.99 (0.87 1.12)1.000.96 (0.85 1.08)0.84 (0.67 1.06)
HR (95% CI)†1.01 (0.97 1.04)0.740.99 (0.87 1.13)1.001.004 (0.89 1.14)1.01 (0.80 1.29)
HR (95% CI)a1.004 (0.97 1.04)0.841.01 (0.88 1.15)1.001.001 (0.88 1.13)1.02 (0.81 1.30)
OropharyngealCases5578626314860
HR (95% CI)*1.12 (1.08 1.17)< 0.0010.83 (0.65 1.06)1.001.17 (0.96 1.43)1.99 (1.50 2.63)
HR (95% CI)†1.06 (1.02 1.11)0.0090.83 (0.64 1.07)1.001.07 (0.87 1.32)1.48 (1.09 2.01)
HR (95% CI)1.06 (1.02 1.11)0.0090.83 (0.64 1.07)1.001.07 (0.87 1.32)1.48 (1.09 2.01)
LungCases2076236901656283
HR (95% CI)*1.17 (1.14 1.19)< 0.001**0.74 (0.64 0.86)1.001.29 (1.17 1.43)2.28 (1.99 2.61)
HR (95% CI)†1.02 (0.995 1.04)0.12**0.87 (0.75 1.01)1.000.98 (0.88 1.09)1.09 (0.93 1.26)
HR (95% CI)h1.02 (0.997 1.05)0.09**0.85 (0.73 0.997)1.000.98 (0.88 1.09)1.09 (0.94 1.27)
Breast (female only)Cases5702109729031386316
HR (95% CI)*1.01 (0.99 1.02)0.430.93 (0.87 1.002)1.000.97 (0.91 1.03)1.003 (0.89 1.13)
HR (95% CI)†1.003 (0.98 1.02)0.77**0.94 (0.88 1.01)1.000.96 (0.90 1.03)0.99 (0.87 1.12)
HR (95% CI)b, h1.01 (0.99 1.03)0.59**0.92 (0.85 0.996)1.000.95 (0.88 1.02)1.01 (0.87 1.16)
UterusCases87215141126446
HR (95% CI)*1.04 (0.999 1.08)0.0530.97 (0.81 1.17)1.001.21 (1.03 1.41)0.95 (0.70 1.28)
HR (95% CI)†0.97 (0.93 1.02)0.211.05 (0.86 1.27)1.001.05 (0.89 1.24)0.63 (0.44 0.88)
HR (95% CI)c0.97 (0.93 1.02)0.241.03 (0.84 1.27)1.001.04 (0.87 1.24)0.61 (0.42 0.88)
OvaryCases57810528715531
HR (95% CI)*1.002 (0.95 1.05)0.930.95 (0.76 1.19)1.001.02 (0.84 1.24)0.91 (0.63 1.32)
HR (95% CI)†1.02 (0.96 1.08)0.530.90 (0.71 1.15)1.001.04 (0.84 1.27)0.93 (0.63 1.38)
HR (95% CI)1.02 (0.96 1.08)0.530.90 (0.71 1.15)1.001.04 (0.84 1.27)0.93 (0.63 1.38)
ProstateCases5979111629571562344
HR (95% CI)*0.96 (0.95 0.98)< 0.001**1.08 (1.01 1.15)1.000.96 (0.90 1.02)0.81 (0.73 0.91)
HR (95% CI)†0.99 (0.97 1.004)0.12**1.05 (0.98 1.13)1.001.01 (0.94 1.07)0.94 (0.83 1.06)
HR (95% CI)h0.99 (0.97 1.01)0.17**1.04 (0.97 1.12)1.001.01 (0.95 1.08)0.95 (0.84 1.07)
OesophagusCases5417024617649
HR (95% CI)*1.10 (1.05 1.15)< 0.0010.80 (0.61 1.04)1.001.30 (1.07 1.58)1.44 (1.06 1.96)
HR (95% CI)†1.03 (0.98 1.08)0.31**0.85 (0.64 1.13)1.001.09 (0.89 1.34)1.02 (0.73 1.42)
HR (95% CI)f1.02 (0.97 1.08)0.34**0.86 (0.65 1.14)1.001.09 (0.88 1.34)1.02 (0.73 1.42)
StomachCases3563616412135
HR (95% CI)*1.14 (1.08 1.20)< 0.0010.61 (0.43 0.88)1.001.34 (1.06 1.70)1.55 (1.08 2.24)
HR (95% CI)†1.06 (1.001 1.13)0.0450.66 (0.45 0.97)1.001.12 (0.87 1.44)1.03 (0.69 1.53)
HR (95% CI)1.06 (1.001 1.13)0.0450.66 (0.45 0.97)1.001.12 (0.87 1.44)1.03 (0.69 1.53)
Oesophagus and stomachCases89110540529784
HR (95% CI)*1.12 (1.08 1.15)< 0.0010.73 (0.59 0.90)1.001.33 (1.15 1.55)1.50 (1.19 1.90)
HR (95% CI)†1.04 (1.005 1.09)0.030.78 (0.62 0.98)1.001.12 (0.95 1.31)1.04 (0.81 1.34)
HR (95% CI)1.04 (1.005 1.09)0.030.78 (0.62 0.98)1.001.12 (0.95 1.31)1.04 (0.81 1.34)
Hepatobiliary tractCases4567420313049
HR (95% CI)*1.08 (1.03 1.14)0.0021.02 (0.78 1.33)1.001.15 (0.93 1.44)1.77 (1.30 2.43)
HR (95% CI)†1.01 (0.96 1.07)0.621.08 (0.82 1.43)1.000.98 (0.78 1.24)1.26 (0.90 1.77)
HR (95% CI)1.01 (0.96 1.07)0.621.08 (0.82 1.43)1.000.98 (0.78 1.24)1.26 (0.90 1.77)
PancreaticCases6159728318748
HR (95% CI)*1.07 (1.02 1.11)0.0040.96 (0.76 1.21)1.001.19 (0.99 1.43)1.25 (0.92 1.70)
HR (95% CI)†1.04 (0.99 1.09)0.150.99 (0.78 1.27)1.001.12 (0.92 1.36)1.07 (0.77 1.49)
HR (95% CI)d1.03 (0.98 1.08)0.200.996 (0.78 1.27)1.001.11 (0.92 1.36)1.03 (0.74 1.44)
KidneyCases77911339020670
HR (95% CI)*1.06 (1.02 1.10)0.0070.79 (0.64 0.98)1.000.99 (0.83 1.17)1.37 (1.06 1.76)
HR (95% CI)†0.996 (0.95 1.04)0.86**0.92 (0.74 1.14)1.000.88 (0.74 1.05)1.08 (0.82 1.42)
HR (95% CI)0.996 (0.95 1.04)0.86**0.92 (0.74 1.14)1.000.88 (0.74 1.05)1.08 (0.82 1.42)
BladderCases6779229522169
HR (95% CI)*1.10 (1.05 1.14)< 0.0010.90 (0.71 1.14)1.001.32 (1.10 1.57)1.62 (1.25 2.11)
HR (95% CI)†1.04 (0.99 1.09)0.13**1.04 (0.81 1.32)1.001.21 (1.002 1.45)1.29 (0.97 1.73)
HR (95% CI)1.04 (0.99 1.09)0.13**1.04 (0.81 1.32)1.001.21 (1.002 1.45)1.29 (0.97 1.73)
ColorectalCases33585381643936241
HR (95% CI)*1.03 (1.01 1.05)0.001**0.90 (0.82 0.99)1.001.05 (0.97 1.14)1.11 (0.97 1.28)
HR (95% CI)†1.02 (0.999 1.04)0.07**0.93 (0.84 1.03)1.001.03 (0.94 1.12)1.05 (0.90 1.22)
HR (95% CI)e, g, f (males)1.02 (0.995 1.04)0.13**0.93 (0.83 1.03)1.001.02 (0.93 1.11)1.03 (0.88 1.20)
ColonCases21553291041614171
HR (95% CI)*1.05 (1.02 1.08)< 0.001**0.87 (0.77 0.99)1.001.08 (0.98 1.19)1.24 (1.05 1.45)
HR (95% CI)†1.04 (1.01 1.07)0.007**0.92 (0.81 1.05)1.001.05 (0.94 1.17)1.19 (1.003 1.42)
HR (95% CI)e, g, f (males)1.04 (1.01 1.06)0.02**0.93 (0.81 1.06)1.001.05 (0.94 1.16)1.17 (0.98 1.41)
RectumCases112719655630768
HR (95% CI)*1.01 (0.98 1.05)0.53**0.96 (0.81 1.13)1.001.04 (0.90 1.20)0.94 (0.73121)
HR (95% CI)†0.996 (0.96 1.04)0.84**0.98 (0.82 1.16)1.001.04 (0.89 1.20)0.84 (0.63 1.11)
HR (95% CI)e, g0.99 (0.95 1.03)0.670.96 (0.81 1.14)1.001.01 (0.87 1.18)0.82 (0.62 1.10)
Brain tumoursCases4638223711430
HR (95% CI)*1.03 (0.98 1.08)0.290.92 (0.72 1.19)1.000.93 (0.74 1.16)1.003 (0.69 1.47)
HR (95% CI)†1.04 (0.98 1.10)0.200.87 (0.66 1.13)1.000.92 (0.73 1.17)0.96 (0.63 1.46)
HR (95% CI)1.04 (0.98 1.10)0.200.87 (0.66 1.13)1.000.92 (0.73 1.17)0.96 (0.63 1.46)
ThyroidCases242481245713
HR (95% CI)*0.99 (0.91 1.07)0.750.95 (0.68 1.32)1.000.97 (0.70 1.33)0.97 (0.55 1.73)
HR (95% CI)†1.001 (0.92 1.09)0.980.92 (0.65 1.30)1.000.93 (0.66 1.32)1.14 (0.63 2.06)
HR (95% CI)1.001 (0.92 1.09)0.980.92 (0.65 1.30)1.000.93 (0.66 1.32)1.14 (0.63 2.06)
Haematological malignanciesCases24684381208652170
HR (95% CI)*1.01 (0.98 1.03)0.520.995 (0.89 1.11)1.000.99 (0.90 1.09)1.06 (0.91 1.25)
HR (95% CI)†1.002 (0.98 1.03)0.890.97 (0.87 1.09)1.000.97 (0.88 1.08)0.97 (0.82 1.16)
HR (95% CI)1.002 (0.98 1.03)0.890.97 (0.87 1.09)1.000.97 (0.88 1.08)0.97 (0.82 1.16)
Non-Hodgkin’s lymphomaCases119319758633773
HR (95% CI)*1.01 (0.98 1.05)0.440.92 (0.78 1.08)1.001.06 (0.93 1.21)0.95 (0.74 1.21)
HR (95% CI)†1.01 (0.98 1.05)0.480.89 (0.75 1.05)1.001.08 (0.93 1.24)0.85 (0.65 1.11)
HR (95% CI)1.01 (0.98 1.05)0.480.89 (0.75 1.05)1.001.08 (0.93 1.24)0.85 (0.65 1.11)

*Models adjusted for age and sex (total observations = 470,578)

†Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [

aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine)

bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no)

eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week)

fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women)

f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men)

gResults for males and females combined using meta-analysis as covariates are different

hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

**Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05)

Table 6

Results of Cox proportional hazards analyses investigating the association between self-report daily total screen time and cancer incidence

1 h increase in total screen timep-value≤1 h1- ≤ 4 h (reference)4- ≤ 8 h> 8 h
Person-years3,474,425254,1472,111,765997,699110,815
Skin, melanomaCases161410196050449
HR (95% CI)*0.99 (0.97 1.02)0.64**0.93 (0.76 1.14)1.001.02 (0.91 1.13)0.98 (0.73 1.30)
HR (95% CI)†1.01 (0.98 1.03)0.620.97 (0.78 1.19)1.001.07 (0.96 1.20)1.06 (0.78 1.44)
HR (95% CI)a1.01 (0.98 1.03)0.700.99 (0.80 1.22)1.001.07 (0.95 1.19)1.07 (0.79 1.45)
OropharyngealCases5522431219125
HR (95% CI)*1.05 (1.01 1.09)0.0090.69 (0.46 1.05)1.001.18 (0.99 1.42)1.39 (0.92 2.09)
HR (95% CI)†1.02 (0.98 1.06)0.410.69 (0.45 1.04)1.001.10 (0.91 1.32)1.09 (0.71 1.67)
HR (95% CI)1.02 (0.98 1.06)0.410.69 (0.45 1.04)1.001.10 (0.91 1.32)1.09 (0.71 1.67)
LungCases2014119995774126
HR (95% CI)*1.11 (1.09 1.13)< 0.001**1.13 (0.93 1.37)1.001.37 (1.25 1.50)2.49 (2.07 3.00)
HR (95% CI)†1.02 (1.003 1.04)0.03**1.13 (0.92 1.38)1.001.05 (0.95 1.16)1.42 (1.16 1.72)
HR (95% CI)h1.03 (1.004 1.05)0.02**1.12 (0.91 1.37)1.001.05 (0.95 1.16)1.45 (1.19 1.77)
Breast (female only)Cases560941835261522143
HR (95% CI)*1.01 (0.996 1.02)0.16**0.91 (0.82 1.004)1.000.97 (0.92 1.04)1.08 (0.91 1.27)
HR (95% CI)†1.003 (0.99 1.02)0.64**0.94 (0.85 1.04)1.000.95 (0.90 1.02)1.08 (0.91 1.29)
HR (95% CI)b, h1.01 (0.99 1.02)0.37**0.93 (0.83 1.04)1.000.97 (0.90 1.04)1.11 (0.92 1.34)
UterusCases8567050426418
HR (95% CI)*1.03 (0.99 1.06)0.141.11 (0.87 1.43)1.001.12 (0.96 1.30)0.99 (0.62 1.58)
HR (95% CI)†0.97 (0.93 1.004)0.081.15 (0.88 1.50)1.000.93 (0.80 1.09)0.57 (0.33 0.97)
HR (95% CI)c0.97 (0.94 1.01)0.171.16 (0.89 1.53)1.000.93 (0.79 1.09)0.66 (0.38 1.12)
OvaryCases5614435414914
HR (95% CI)*0.99 (0.95 1.04)0.750.99 (0.72 1.35)1.000.90 (0.74 1.09)1.08 (0.63 1.85)
HR (95% CI)†0.997 (0.95 1.04)0.910.98 (0.71 1.35)1.000.90 (0.74 1.11)1.14 (0.66 1.95)
HR (95% CI)0.997 (0.95 1.04)0.910.98 (0.71 1.35)1.000.90 (0.74 1.11)1.14 (0.66 1.95)
ProstateCases589833533402032191
HR (95% CI)*0.98 (0.97 0.99)< 0.001**1.07 (0.95 1.19)1.000.96 (0.91 1.01)0.86 (0.74 0.99)
HR (95% CI)†0.99 (0.98 1.004)0.16**1.06 (0.94 1.19)1.001.01 (0.95 1.07)0.94 (0.81 1.10)
HR (95% CI)h0.99 (0.98 1.005)0.21**1.05 (0.93 1.18)1.001.01 (0.95 1.07)0.94 (0.81 1.10)
OesophagusCases5282827221612
HR (95% CI)*1.05 (1.02 1.10)0.0061.02 (0.69 1.50)1.001.34 (1.12 1.60)0.75 (0.42 1.34)
HR (95% CI)†1.003 (0.96 1.05)0.891.13 (0.77 1.68)1.001.13 (0.94 1.36)0.54 (0.29 0.99)
HR (95% CI)f1.001 (0.96 1.04)0.951.15 (0.77 1.69)1.001.12 (0.93 1.36)0.54 (0.29 0.99)
StomachCases3481417714116
HR (95% CI)*1.08 (1.04 1.13)< 0.0010.77 (0.45 1.33)1.001.37 (1.09 1.71)1.58 (0.95 2.63)
HR (95% CI)†1.03 (0.98 1.08)0.190.75 (0.43 1.32)1.001.12 (0.89 1.42)1.07 (0.61 1.85)
HR (95% CI)1.03 (0.98 1.08)0.190.75 (0.43 1.32)1.001.12 (0.89 1.42)1.07 (0.61 1.85)
Oesophagus and stomachCases8704244435628
HR (95% CI)*1.07 (1.04 1.10)< 0.0010.93 (0.68 1.28)1.001.36 (1.18 1.57)1.09 (0.74 1.59)
HR (95% CI)†1.02 (0.98 1.05)0.310.99 (0.72 1.36)1.001.14 (0.98 1.32)0.76 (0.50 1.14)
HR (95% CI)f1.02 (0.98 1.05)0.340.99 (0.72 1.37)1.001.13 (0.98 1.31)0.76 (0.50 1.14)
Hepatobiliary tractCases4462224915619
HR (95% CI)*1.05 (1.01 1.10)0.020.84 (0.54 1.29)1.001.11 (0.90 1.35)1.45 (0.91 2.31)
HR (95% CI)†1.01 (0.96 1.05)0.730.82 (0.52 1.30)1.000.95 (0.77 1.18)1.05 (0.65 1.72)
HR (95% CI)1.01 (0.96 1.05)0.730.82 (0.52 1.30)1.000.95 (0.77 1.18)1.05 (0.65 1.72)
PancreaticCases6043033322021
HR (95% CI)*1.04 (1.004 1.08)0.030.85 (0.58 1.23)1.001.17 (0.99 1.39)1.22 (0.78 1.89)
HR (95% CI)†1.02 (0.98 1.06)0.370.86 (0.58 1.26)1.001.10 (0.92 1.31)1.01 (0.64 1.60)
HR (95% CI)d1.02 (0.98 1.06)0.450.86 (0.58 1.26)1.001.08 (0.91 1.30)0.99 (0.63 1.57)
KidneyCases7714341727932
HR (95% CI)*1.04 (1.01. 1.08)0.010.98 (0.71 1.34)1.001.19 (1.02 1.38)1.38 (0.96 1.98)
HR (95% CI)†1.01 (0.97 1.04)0.67**1.03 (0.74 1.44)1.001.07 (0.91 1.25)1.14 (0.79 1.66)
HR (95% CI)1.01 (0.97 1.04)0.67**1.03 (0.74 1.44)1.001.07 (0.91 1.25)1.14 (0.79 1.66)
BladderCases6622435125928
HR (95% CI)*1.06 (1.02 1.09)0.0010.69 (0.46 1.04)1.001.21 (1.03 1.42)1.36 (0.93 2.00)
HR (95% CI)†1.01 (0.98 1.05)0.50**0.76 (0.50 1.16)1.001.10 (0.93 1.30)0.99 (0.65 1.53)
HR (95% CI)1.01 (0.98 1.05)0.50**0.76 (0.50 1.16)1.001.10 (0.93 1.30)0.99 (0.65 1.53)
ColorectalCases329018018961096118
HR (95% CI)*1.02 (1.002 1.04)0.03**0.88 (0.76 1.03)1.001.04 (0.97 1.12)1.16 (0.97 1.40)
HR (95% CI)†1.01 (0.99 1.03)0.41**0.92 (0.79 1.08)1.001.003 (0.93 1.08)1.11 (0.91 1.35)
HR (95% CI)e, g, f (males)1.01 (0.99 1.02)0.58**0.89 (0.76 1.04)1.000.99 (0.91 1.07)1.08 (0.88 1.32)
ColonCases2110111120272176
HR (95% CI)*1.03 (1.01 1.05)0.003**0.86 (0.70 1.04)1.001.08 (0.99 1.19)1.21 (0.96 1.53)
HR (95% CI)†1.02 (0.998 1.04)0.08**0.91 (0.75 1.11)1.001.04 (0.95 1.15)1.16 (0.91 1.48)
HR (95% CI)e, g, f (males)1.02 (0.99 1.04)0.15**0.90 (0.73 1.10)1.001.04 (0.94 1.15)1.12 (0.88 1.44)
RectumCases11076464535840
HR (95% CI)*0.998 (0.97 1.03)0.89**0.94 (0.73 1.21)1.000.99 (0.87 1.13)1.10 (0.80 1.51)
HR (95% CI)†0.99 (0.96 1.02)0.380.95 (0.73 1.25)1.000.95 (0.83 1.09)1.04 (0.74 1.46)
HR (95% CI)e, g0.98 (0.95 1.02)0.310.91 (0.69 1.20)1.000.92 (0.80 1.06)1.01 (0.72 1.43)
Brain tumoursCases4582826914516
HR (95% CI)*1.01 (0.97 1.06)0.500.96 (0.65 1.42)1.000.99 (0.81 1.22)1.07 (0.65 1.77)
HR (95% CI)†1.02 (0.98 1.07)0.310.98 (0.66 1.46)1.001.001 (0.81 1.24)1.13 (0.67 1.91)
HR (95% CI)f1.03 (0.98 1.07)0.280.97 (0.65 1.45)1.001.01 (0.81 1.25)1.13 (0.67 1.92)
ThyroidCases23615154625
HR (95% CI)*0.99 (0.93 1.06)0.860.78 (0.46 1.32)1.000.87 (0.65 1.18)0.71 (0.29 1.72)
HR (95% CI)†0.999 (0.93 1.07)0.970.76 (0.44 1.32)1.000.88 (0.64 1.20)0.70 (0.28 1.72)
HR (95% CI)0.999 (0.93 1.07)0.970.76 (0.44 1.32)1.000.88 (0.64 1.20)0.70 (0.28 1.72)
Haematological malignanciesCases2427142139680683
HR (95% CI)*1.02 (0.998 1.04)0.090.95 (0.80 1.12)1.001.04 (0.96 1.14)1.12 (0.90 1.40)
HR (95% CI)†1.01 (0.99 1.03)0.430.96 (0.80 1.15)1.001.02 (0.93 1.12)1.07 (0.85 1.34)
HR (95% CI)1.01 (0.99 1.03)0.430.96 (0.80 1.15)1.001.02 (0.93 1.12)1.07 (0.85 1.34)
Non-Hodgkin’s lymphomaCases11746867539239
HR (95% CI)*1.02 (0.99 1.04)0.280.93 (0.72 1.19)1.001.06 (0.93 1.20)1.10 (0.80 1.52)
HR (95% CI)†1.01 (0.98 1.04)0.430.93 (0.72 1.20)1.001.08 (0.95 1.23)1.06 (0.75 1.48)
HR (95% CI)1.01 (0.98 1.04)0.430.93 (0.72 1.20)1.001.08 (0.95 1.23)1.06 (0.75 1.48)

*Models adjusted for age and sex (total observations = 464,424)

†Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [

aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine)

bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no)

eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week)

fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women)

f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men)

gResults for males and females combined using meta-analysis as covariates are different

hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

**Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05)

These analyses were repeated separately to investigate the relationship between a 1-h increase/day in [1] daily recreational computer time; (2) daily total recreational screen time; and site-specific cancer risk. To characterise the dose-response relationships [9, 33], we repeated these analyses with categorised independent variables as follows: daily TV viewing time (1- ≤ 3 h [reference category]; ≤1 h; 3- ≤ 5 h; > 5 h), daily recreational computer use time (≤1 h [reference category]; none; 1- ≤ 3 h; > 3 h) and daily total recreational screen time (1- ≤ 4 h [reference category]; ≤1 h; 4- ≤ 8 h; > 8 h) categorised based on previously published categories [17]. A series of partition models and isotemporal substitution models [21] were used for each type of cancer to examine the associations of daily TV viewing time, time spent walking/day, time spent in moderate-intensity physical activity/day, time spent in vigorous-intensity physical activity/day and cancer incidence [21, 34–37]. Partition models examined all behaviours simultaneously, without adjusting for total physical activity time. Therefore, the HR for one type of physical activity represented the effect of increasing this type of physical activity (by 1-h/day) while holding the other physical activities constant. Since total physical activity time is not included in the model (and thus is not held constant), these results represent the effect of adding a behaviour (i.e. walking, moderate-activity, vigorous-activity, TV screen time) whilst holding the others constant. The effects of substituting one behaviour type by another for the same amount of time (i.e. replacing 1-h/day of TV screen time for 1-h/day of walking, moderate-intensity physical activity or vigorous-intensity physical activity) was investigated using isotemporal substitution models which adjusted for time spent walking/day, time spent in moderate-intensity physical activity/day, time spent in vigorous-intensity physical activity/day and total activity time/day (i.e. the summation of walking, moderate activity, vigorous activity and TV viewing time). In this case, since total activity is included the model (and thus is held constant), these results represent the effect of replacing daily TV viewing time with the same amount of another physical activity type (i.e. walking, moderate- or vigorous- activity) while holding the others constant. Sensitivity analyses were conducted by confining the analysis to cancers diagnosed at least two years following baseline to examine the impact of removing prevalent disease. Subgroup analyses were conducted by selected baseline characteristics (supplement 1/Table 1.2–1.9). These included sex, age, deprivation index, smoking status, BMI (with reference to obese/non-obese thresholds defined for various ethnic groups by gender in a previous UK Biobank study [38], assuming that participants with mixed backgrounds or ‘other’ ethnicities had the same obesity thresholds as white participants since cut-off points were not available for this group), and IPAQ physical activity category. Further analyses were conducted by creating four categories based on body fat percentage and physical activity levels defined according to the IPAQ (High/moderate physical activity and low/optimal body fat percentage; Low physical activity and low/optimal body fat percentage; High/moderate physical activity and high body fat percentage; Low physical activity and high body fat percentage), in order to investigate any differential associations between sedentary behaviour and cancer risk according to the ‘fat but fit’ hypothesis [39]. Body fat percentage cut-points were derived from previously established thresholds defined by age, gender, ethnicity and BMI [40]. We assumed participants with mixed backgrounds or ‘other’ ethnicities had the same body fat percentage thresholds as white participants. Subgroup analyses were also conducted by menopausal status for female-specific cancers (i.e. breast, uterus, ovary cancers). Interactions were tested using the Wald test for homogeneity and declared significant if p < 0.01 in line with previous studies [41]. Baseline characteristics by self-report daily TV viewing time. Values are numbers and percentages unless otherwise stated CSE: Certificate of Secondary Education; GCSE: General Certificate of Secondary Education; HNC: Higher National Certificate; HND: Higher National Diploma; HRT: hormone-replacement therapy; IDF: International Diabetes Federation; MVPA: moderate-vigorous intensity physical activity; NVQ: National Vocational Qualifications; UV: ultraviolet aDefined in terms of pack-years: light (< 20 pack-years), heavy (≥20 pack-years) bBased on IDF criteria (waist circumference > 94 cm in men; > 80 cm in women) cDiagnosed by doctor dSelf-reported eBased on self-reported illnesses of father, mother and siblings fRegular use defined as most days of the week for the last 4 weeks gFemale participants only The proportional hazards assumption was tested for each model formally using Schoenfeld residuals (p < 0.05 indicated potential violation of the proportional hazards assumption), and by visual inspection of scaled Schoenfeld residual plots [42] and log-log plots (parallel curves indicated that there was no evidence for violation of the proportional hazards assumption). We used restricted cubic splines analyses to examine the associations between the continuous exposure variables (daily TV viewing time, daily recreational computer time, daily total recreational screen time, daily moderate-intensity physical activity, daily vigorous-intensity physical activity, and daily walking time) and cancer risk at each site for potential violations of linearity assumptions. No serious violation of the linearity assumption was observed. Details are reported in supplement 2. Analyses were carried out using Stata 13 [43].

Results

Participant characteristics according to total daily TV viewing time are shown in Table 1. Among the 470,578 participants included in this analysis, 53.8% were women and the mean age was 56.3 years. Most participants reported that they spent between 2 and 8 h/day watching TV or using the computer. During a mean follow-up time of 7.6 (SD 1.4) years (median 7.8 years, interquartile range 7.0–8.5), 28,992 incident cancers were identified.
Table 1

Baseline characteristics by self-report daily TV viewing time. Values are numbers and percentages unless otherwise stated

Total TV viewing time
Total≤1 h per day1- ≤ 3 h per day3- ≤ 5 h per day> 5 h/day
No./mean%/SDNo./mean%/SDNo./mean%/SDNo./mean%/SDNo./mean%/SD
Total participants470,578100.0%97,41920.7%236,98850.4%110,33423.5%25,8375.5%
Self-report total screen time (hours/day; mean/SD)3.92.12.01.53.51.45.31.47.92.4
Time spent watching TV (hours/day; mean/SD)2.81.70.70.42.50.54.30.56.91.8
Time spent using computers (hours/day; mean/SD)1.11.41.21.51.11.31.01.31.11.7
IPAQ physical activity (mean/SD)
 Hours/day of walking0.750.780.710.740.760.790.770.790.670.74
 Hours/day of moderate-intensity physical activity0.560.720.540.700.570.730.580.740.460.67
 Hours/day of vigorous-intensity physical activity0.200.350.220.340.200.350.180.350.140.33
Age at baseline (mean/SD)56.38.154.28.055.98.158.57.659.07.6
Height (m) (mean/SD)1.70.11.70.11.70.11.70.11.70.1
Sex
 Female253,18853.8%53,50054.9%127,13553.7%59,55354.0%13,00050.3%
 Male217,39046.2%43,91945.1%109,85346.4%50,78146.0%12,83749.7%
Ethnicity
 White443,48494.6%90,40593.3%224,14294.9%104,92295.4%24,01593.3%
 Black75051.6%15491.6%33581.4%18461.7%7522.9%
 South Asian93952.0%25822.7%47212.0%16391.5%4531.8%
 Chinese15010.3%4650.5%7170.3%2660.2%530.2%
 Mixed background or others70691.5%19492.0%32991.4%13541.2%4671.8%
Townsend deprivation quintile
 1 (Least deprived)94,59020.1%19,86020.4%51,16421.6%20,49718.6%306911.9%
 293,95020.0%18,85419.4%49,80421.0%21,69119.7%360114.0%
 394,16620.0%18,85719.4%48,70620.6%22,37920.3%422416.4%
 494,11820.0%20,20920.8%46,50519.7%22,08120.0%532320.6%
 5 (Most deprived)93,16519.8%19,53020.1%40,49517.1%23,56121.4%957937.1%
Education
 University degree153,22333.1%52,25054.3%79,04133.9%19,25717.8%267510.6%
 A-levels/HNC/HND/NVQ83,31518.0%15,93416.6%44,64919.1%18,92117.5%381115.1%
 GCSE/O-level/CSE124,76526.9%17,51818.2%66,18028.3%34,25731.6%681027.0%
 Other24,0185.2%42304.4%12,5305.4%60915.6%11674.6%
 None78,02816.8%63466.6%31,12513.3%29,80027.5%10,75742.7%
Smoking statusa
 Never257,69655.0%58,98160.7%132,97656.3%54,85849.9%10,88142.4%
 Former light smoker119,08525.4%24,55625.3%61,14725.9%27,89125.4%549121.4%
 Former heavy smoker42,2519.0%53505.5%19,2568.2%13,52112.3%412416.1%
 Current light smoker27,7945.9%55355.7%13,7065.8%66466.1%19077.4%
 Current heavy smoker22,0824.7%27352.8%90943.9%69726.3%328112.8%
Alcohol intake
 Never20,7494.4%48735.0%94284.0%48684.4%15806.1%
 Former drinker16,6593.5%33303.4%71283.0%43423.9%18597.2%
 Current drinker: <once/week106,02022.6%19,46620.0%50,95821.5%27,93925.3%765729.7%
 Current drinker: ≥once/week326,75969.5%69,67671.6%169,32471.5%73,08966.3%14,67056.9%
Dietary intake (mean/SD)
 Fruits and vegetables (portion/day)4.73.15.13.24.73.04.53.04.23.3
 Red meat (portion/week)2.11.52.01.42.11.42.21.52.41.7
 Processed meat (portion/week)1.51.41.31.41.51.41.61.41.91.6
Body Mass Index (Kg/m2) (mean/SD)27.44.826.04.327.34.628.54.929.75.8
Body Mass Index (Kg/m2)
  < 18.524180.5%8250.9%11130.5%3520.3%1280.5%
 18.5- < 25152,53332.6%44,07545.5%77,50732.9%26,15723.8%479418.8%
 25- < 30199,21242.6%37,52838.7%103,14143.7%48,58644.3%995739.1%
 30+113,92224.3%14,48915.0%54,19723.0%34,62631.6%10,61041.6%
Body fat percentage (mean/SD)31.38.528.98.331.18.433.28.434.38.8
Waist:hip ratiob
 Waist:hip ratio (mean/SD)0.90.10.90.10.90.10.90.10.90.1
 Below IDF guideline202,54543.2%54,75056.4%104,48244.2%36,71733.4%659625.7%
 Above IDF guideline266,44356.8%42,33343.6%131,82955.8%73,24966.6%19,03274.3%
Health status
 Diabetesc24,3475.2%30853.2%10,4044.4%76877.0%317112.4%
 Gastro-oesophageal refluxd22,4954.8%32333.3%10,6724.5%66486.0%19427.5%
Family historye
 Prostate cancer37,2258.0%84318.8%18,6078.0%83327.7%18557.3%
 Breast cancer49,52410.7%10,52010.9%24,98610.7%11,36010.5%265810.5%
 Lung cancer59,04212.7%959610.0%29,21812.5%16,10714.9%412116.3%
 Bowel cancer52,10911.2%10,18110.6%25,85111.1%12,94311.9%313412.4%
Use of sun/UV protection
 Never/rarely/sometimes203,96843.7%43,45044.9%99,12142.1%48,28644.1%13,11151.4%
 Most of the time/always260,24155.7%52,69954.5%135,03357.4%60,49355.3%12,01647.1%
 Do not go out in sunshine27700.6%5380.6%11360.5%7170.7%3791.5%
Aspirin use
 Regularly uses aspirinf64,82214.0%971110.1%29,90812.9%19,24617.8%595723.7%
HRT useg
 Ever used HRT95,36937.8%14,79127.8%46,58736.8%27,63946.6%635249.1%
Oral contraceptive useg
 Ever taken oral contraceptive pill205,52881.4%44,28583.0%104,77282.7%46,69578.7%977675.6%
Number of live births (0, 1, 2, 3+ live births)g(mean/SD)1.81.21.81.21.81.21.91.22.01.3
Age at menarche (mean/SD)g13.01.613.01.613.01.613.01.713.01.7
Age at menopause (mean/SD)g49.85.150.04.749.95.049.65.449.05.8
Menopausal statusg
 Had menopause151,10159.8%27,73651.9%74,07558.4%40,39968.0%889168.6%
 Not had menopause62,57024.8%18,65934.9%33,17426.1%907515.3%166212.8%
 Unsure39,06515.5%700213.1%19,68415.5%996916.8%241018.6%
Hysterectomy statusg
 Had hysterectomy17,5307.8%24585.0%81937.2%548310.6%139612.7%
 Not had hysterectomy/unsure207,95392.2%46,84695.0%105,23892.8%46,23289.4%963787.4%

CSE: Certificate of Secondary Education; GCSE: General Certificate of Secondary Education; HNC: Higher National Certificate; HND: Higher National Diploma; HRT: hormone-replacement therapy; IDF: International Diabetes Federation; MVPA: moderate-vigorous intensity physical activity; NVQ: National Vocational Qualifications; UV: ultraviolet

aDefined in terms of pack-years: light (< 20 pack-years), heavy (≥20 pack-years)

bBased on IDF criteria (waist circumference > 94 cm in men; > 80 cm in women)

cDiagnosed by doctor

dSelf-reported

eBased on self-reported illnesses of father, mother and siblings

fRegular use defined as most days of the week for the last 4 weeks

gFemale participants only

Association of site-specific cancer risk and daily TV viewing time

Table 2 and Fig. 1 show the association between daily TV viewing time and site-specific cancer risk. A 1-h increase in daily TV viewing time was associated with higher risk of oropharyngeal cancer (HR 1.06, 95% CI: 1.02, 1.11), stomach cancer (HR 1.06, 95% CI: 1.001, 1.13), oesophagus and stomach cancer (HR 1.04, 95% CI: 1.005, 1.09), and colon cancer (HR 1.04, 95% CI: 1.01, 1.06) in fully adjusted models. In addition, the categorical analysis showed that participants who reported > 5 h/day of TV viewing time had a higher risk of oropharyngeal cancer (HR 1.48, HR: 1.09, 2.01) and a lower risk of uterus cancer (HR 0.61, 95% CI: 0.42, 0.88) compared to participants who reported 1- ≤ 3 h/day of TV viewing time. Participants who reported 3- ≤ 5 h/day of TV viewing time had a higher risk of bladder cancer (HR 1.21, 95% CI: 1.002, 1.45) compared to participants who reported 1- ≤ 3 h/day of TV viewing time, but no dose-response association was evident for greater duration of TV viewing time.
Fig. 1

Summary hazard ratio estimates by cancer site

Results of Cox proportional hazards analyses investigating the association between self-report daily TV viewing time and cancer incidence *Models adjusted for age and sex (total observations = 470,578) †Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [ aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine) bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no) eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week) fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women) f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men) gResults for males and females combined using meta-analysis as covariates are different hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history) **Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05) Summary hazard ratio estimates by cancer site Participants who reported ≤1 h/day of TV viewing time had a lower risk of lung cancer (HR 0.85, 95% CI: 0.73, 0.997), breast (female only) cancer (HR 0.92, 95% CI: 0.85, 0.996), stomach cancer (HR 0.66, 95% CI: 0.45, 0.97), and oesophagus and stomach cancer (HR 0.78, 95% CI: 0.62, 0.98) compared to participants who reported 1- ≤ 3 h/day of TV viewing time. After excluding cancers diagnosed within the first two years following baseline, all associations were attenuated except those for oesophagus and stomach cancers, and colon cancers (Table 3). Whilst the results of the Schoenfeld residual tests indicated that some of our models may not have been in line with the proportional hazards assumption, our visual inspection of log-log plots and Schoenfeld residual plots showed no serious violations. Therefore, we proceeded with the analyses as planned. Further, we have included our analyses by both continuous and categorical variables side-by-side in the main outcomes tables which facilitated stratification by categories of potentially violating variables.
Table 3

Results of Cox proportional hazards analyses investigating the association between self-report daily TV viewing time and cancer incidence (excluding cancers diagnosed within the first 2 years following baseline)

1 h increase in TV viewing timep-value≤1 h1- ≤ 3 h (reference)3- ≤ 5 h> 5 h
Skin, melanomaCases119222261329958
HR (95% CI)a1.02(0.98,1.06)0.39**0.93(0.80,1.09)1.001.04(0.90,1.20)1.004(0.76,1.33)
OropharyngealCases4106919710539
HR (95% CI)1.04(0.99,1.10)0.120.90(0.68,1.19)1.001.02(0.80,1.30)1.27(0.89,1.84)
LungCases1461165638467191
HR (95% CI)h1.02(0.99,1.05)0.21**0.87(0.73,1.03)1.000.97(0.86,1.10)1.10(0.93,1.30)
Breast (female only)Cases32886571724756151
HR (95% CI)b, h1.004(0.98,1.03)0.76**0.90(0.82,0.99)1.000.96(0.87,1.04)0.94(0.79,1.11)
UterusCases56710227016530
HR (95% CI)c0.98(0.93,1.04)0.56**1.05(0.83,1.33)1.001.06(0.87,1.29)0.75(0.50,1.10)
OvaryCases4046920410922
HR (95% CI)1.02(0.95,1.09)0.600.88(0.67,1.16)1.001.01(0.80,1.29)0.93(0.59,1.46)
ProstateCases423580421271069235
HR (95% CI)h0.99(0.97,1.01)0.34**1.02(0.94,1.10)1.000.98(0.91,1.06)0.94(0.82,1.08)
OesophagusCases3924618212539
HR (95% CI)f1.04(0.98,1.10)0.20**0.79(0.57,1.10)1.001.08(0.86,1.37)1.11(0.77,1.60)
StomachCases250241257823
HR (95% CI)1.06(0.99,1.14)0.090.57(0.37,0.89)1.000.996(0.74,1.33)0.96(0.60,1.53)
Oesophagus and stomachCases6387030320362
HR (95% CI)1.05(1.011.10)0.03**0.71(0.54,0.92)1.001.07(0.89,1.28)1.07(0.80,1.42)
Hepatobiliary tractCases3485115610041
HR (95% CI)1.03(0.97,1.10)0.290.9999(0.72,1.38)1.001.001(0.77,1.29)1.36(0.94,1.95)
PancreaticCases4637521514033
HR (95% CI)d1.01(0.95,1.06)0.851.02(0.78,1.33)1.001.09(0.88,1.36)0.91(0.62,1.34)
KidneyCases5839028016152
HR (95% CI)0.99(0.94,1.04)0.751.02(0.80,1.30)1.000.95(0.77,1.15)1.11(0.81,1.51)
BladderCases4615720815541
HR (95% CI)1.03(0.98,1.09)0.25**0.88(0.65,1.19)1.001.19(0.96,1.47)1.10(0.77,1.55)
ColorectalCases22813831118621159
HR (95% CI)e, g, f (males)1.03(0.998,1.05)0.07**0.95(0.85,1.07)1.001.03(0.93,1.14)1.10(0.93,1.31)
ColonCases1478246712407113
HR (95% CI)e, g, f (males)1.04(1.01,1.07)0.02**0.97(0.84,1.13)1.001.05(0.92,1.19)1.22(0.99,1.50)
RectumCases75413437320146
HR (95% CI)e, g1.000(0.96,1.05)0.9950.98(0.80,1.20)1.001.03(0.87,1.23)0.99(0.72,1.36)
Brain tumoursCases333581788017
HR (95% CI)1.03(0.96,1.11)0.380.85(0.62,1.15)1.000.89(0.68,1.17)0.81(0.48,1.35)
ThyroidCases1613285359
HR (95% CI)0.99(0.89,1.09)0.780.91(0.60,1.38)1.000.88(0.59,1.32)0.998(0.49,2.03)
Haematological malignanciesCases1786315888470113
HR (95% CI)1.003(0.97,1.03)0.86**0.96(0.84,1.10)1.000.98(0.87,1.10)0.95(0.78,1.17)
Non-Hodgkin’s lymphomaCases88613843125443
HR (95% CI)1.02(0.98,1.07)0.260.84(0.69,1.02)1.001.12(0.95,1.31)0.78(0.57,1.08)

aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine)

bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no)

eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week)

fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women)

f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men)

gResults for males and females combined using meta-analysis as covariates are different

hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

**Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05)

Results of Cox proportional hazards analyses investigating the association between self-report daily TV viewing time and cancer incidence (excluding cancers diagnosed within the first 2 years following baseline) aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine) bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no) eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week) fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women) f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men) gResults for males and females combined using meta-analysis as covariates are different hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history) **Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05) Subgroup analyses and tests of effect modification (supplement 1, Tables 1.2–1.9) showed that HR estimates between daily TV viewing time and lung cancer risk differed according to an area-based measure of deprivation (p for interaction =0.004). For example, for participants who ranked in the top 20% least deprived (i.e. most affluent) area of residence, a 1-h increase in daily TV viewing time was associated with a lower risk of lung cancer (HR 0.87, 95% CI: 0.79, 0.95), whereas risk was increased (HR 1.06, 95% CI: 1.004, 1.12) or null (HR 1.02, 95% CI: 0.99, 1.06) for individuals residing in the top 40 and 20% most deprived areas, respectively (Supplementary Table 1.4). The results of other subgroup analyses were mostly non-significant according to our threshold (p for interaction > 0.01), however some differential associations were evident when comparing magnitudes of hazard ratios or precision of confidence intervals. For example, in analyses stratified by sex (Supplementary Table 1.2), 1 h increases in TV viewing time were associated with significant increased oropharyngeal, lung and colon cancer risk in males only, and increased kidney cancer risk in females only. No clear dose-response associations were observed between increased TV viewing and cancer risk across age groups (Supplementary Table 1.3), or categories of smoking status (Supplementary Table 1.5). The increased risk for oesophago-gastric cancer with each 1 h increase in daily TV viewing did appear to be strongest in never smokers (HR 1.11, 95% CI: 1.04, 1.19) although the p for interaction wasn’t significant (p = 0.27). Supplementary Tables 1.6, 1.7, 1.8 illustrate results from analyses stratified by body composition, physical activity levels, or a combination of these (to reflect the ‘fat but fit’ hypothesis). The magnitude of increased oropharyngeal and oesophago-gastric cancer risks for each 1 h increase in daily TV viewing were small, but strongest in individuals who were non-obese (Supplementary Table 1.6), but did not differ across categories of physical activity (Supplementary Table 1.7), although tests for interaction were not significant. Significant interactions were observed according to physical activity level for rectum cancers amongst males (p for interaction = 0.004). According to strata of combined body fat percentage and physical activity level categories, the associations between daily TV viewing and risk of lung cancer (p for interaction = 0.003) and haematological malignancies (p for interaction = 0.0004), as shown in Supplementary Table 1.8. This is difficult to interpret, as the risk estimates do not differ considerably between categories, however are slightly lower in individuals with high physical activity but high body fat percentages. Finally, menopausal status did not alter the null associations between daily TV viewing time and risk of female cancers (Supplementary Table 1.9). Overall, these stratified analyses generate some interesting hypotheses, but results should be interpreted with caution due to small numbers and multiple testing.

Results of partition models and isotemporal substitution models

Partition models showed there was an association between a 1-h increase in daily TV viewing time and a higher risk of oropharyngeal cancer (HR 1.11, 95% CI: 1.05, 1.17) and lung cancer (HR 1.04, 95% CI: 1.01, 1.07), when holding daily time spent in moderate-intensity physical activity, vigorous-intensity physical activity and walking constant. There was an association between a 1-h increase in daily time spent in moderate-intensity physical activity and a lower risk of breast (female only) cancer (HR 0.91, 95% CI: 0.86, 0.96), and colon cancer (HR 0.89, 95% CI: 0.81, 0.97) when holding daily TV viewing time, and time spent in vigorous-intensity physical activity and walking constant (supplement 1, Table 1.1). Isotemporal substitution models showed there was an association between replacing 1-h of daily TV viewing time with 1-h of moderate-intensity physical activity and a lower risk of breast (female only) cancer (HR 0.90, 95% CI: 0.85, 0.96), colorectal cancer (HR 0.92, 95% CI: 0.86, 0.99) and colon cancer (HR 0.87, 95% CI: 0.79, 0.95), when holding time spent in vigorous-intensity physical activity and walking constant. There was an association between replacing 1-h of daily TV viewing time with 1-h of walking and a lower risk of oropharyngeal cancer (HR 0.79, 95% CI: 0.67, 0.92), and lung cancer (HR 0.89, 95% CI: 0.82, 0.97) when holding time spent in moderate- and vigorous-intensity physical activity constant (Table 4).
Table 4

Results of isotemporal substitution models showing the impact on cancer incidence of replacing a 1-h of total daily TV viewing time with the same amount of daily moderate activity, daily vigorous activity or daily walking time, holding the other activities constant

1-h increase in daily moderate activity1-h increase in daily vigorous activity1-h increase in daily walking time
HR (95% CI)HR (95% CI)HR (95% CI)

Skin melanoma

[cases = 1256]a

0.98 (0.89, 1.09)0.97 (0.81, 1.17)1.03 (0.94, 1.12)

Oropharyngeal

[cases = 411]

0.91 (0.77, 1.08)0.86 (0.63, 1.18)0.79 (0.67, 0.92)

Lung

[cases = 1355]h

1.0003 (0.92, 1.09)0.84 (0.71, 1.004)0.89 (0.82, 0.97)

Breast (female only)

[cases = 3454]b, h

0.90 (0.85, 0.96)1.02 (0.89, 1.16)0.99 (0.94, 1.05)

Uterus

[cases = 570]c

1.001 (0.86, 1.17)1.05 (0.76, 1.46)0.99 (0.86, 1.13)

Ovary

[cases = 405]

1.09 (0.93, 1.28)1.12 (0.81, 1.55)0.97 (0.83, 1.13)

Prostate

[cases = 4629]h

1.01 (0.96, 1.06)1.05 (0.97, 1.15)0.9997 (0.95, 1.05)

Oesophagus

[cases = 386]f

1.09 (0.93, 1.28)1.06 (0.80, 1.42)0.91 (0.77, 1.06)

Stomach

[cases = 264]

1.06 (0.87, 1.29)0.77 (0.52, 1.15)0.91 (0.76, 1.10)

Oesophagus and stomach

[cases = 644]

1.08 (0.95, 1.22)0.94 (0.74, 1.18)0.90 (0.80, 1.02)

Hepatobiliary tract

[cases = 331]

0.84 (0.69, 1.02)1.01 (0.71, 1.43)1.03 (0.87, 1.21)

Pancreatic

[cases = 467]d

1.07 (0.92, 1.24)0.92 (0.69, 1.23)0.95 (0.82, 1.09)

Kidney

[cases = 559]

1.01 (0.88, 1.17)1.12 (0.87, 1.44)0.95 (0.83, 1.09)

Bladder

[cases = 502]

0.98 (0.85, 1.13)0.83 (0.62, 1.09)1.03 (0.90, 1.17)

Colorectal

[cases = 2405]e, g, f (males)

0.92 (0.86, 0.99)0.997 (0.87, 1.14)1.01 (0.95, 1.08)

Colon

[cases = 1530]e, g, f (males)

0.87 (0.79, 0.95)0.96 (0.81, 1.14)1.001 (0.92, 1.09)

Rectum

[cases = 821]e, g

0.99 (0.88, 1.12)1.06 (0.86, 1.30)1.01 (0.90, 1.12)

Brain tumours

[cases = 345]

0.85 (0.70, 1.03)0.85 (0.59, 1.23)1.04 (0.88, 1.23)

Thyroid

[cases = 181]

0.94 (0.72, 1.23)0.80 (0.46, 1.40)1.07 (0.85, 1.35)

Haematological malignancies

[cases = 1794]

0.98 (0.90, 1.06)1.07 (0.93, 1.24)0.99 (0.92, 1.07)

Non-Hodgkin’s lymphoma

[cases = 864]

0.99 (0.88, 1.11)1.07 (0.87, 1.32)0.95 (0.85, 1.06)

All models were adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [

aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine)

bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no)

eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week)

fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women)

f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men)

gResults for males and females combined using meta-analysis as covariates are different

hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

Results of isotemporal substitution models showing the impact on cancer incidence of replacing a 1-h of total daily TV viewing time with the same amount of daily moderate activity, daily vigorous activity or daily walking time, holding the other activities constant Skin melanoma [cases = 1256]a Oropharyngeal [cases = 411] Lung [cases = 1355]h Breast (female only) [cases = 3454]b, h Uterus [cases = 570]c Ovary [cases = 405] Prostate [cases = 4629]h Oesophagus [cases = 386]f Stomach [cases = 264] Oesophagus and stomach [cases = 644] Hepatobiliary tract [cases = 331] Pancreatic [cases = 467]d Kidney [cases = 559] Bladder [cases = 502] Colorectal [cases = 2405]e, g, f (males) Colon [cases = 1530]e, g, f (males) Rectum [cases = 821]e, g Brain tumours [cases = 345] Thyroid [cases = 181] Haematological malignancies [cases = 1794] Non-Hodgkin’s lymphoma [cases = 864] All models were adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [ aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine) bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no) eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week) fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women) f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men) gResults for males and females combined using meta-analysis as covariates are different hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

Association of site-specific cancer risk and daily recreational computer time

Table 5 shows the association between a 1-h increase in daily recreational computer time and site-specific cancer risk. A 1-h increase in daily recreational computer time was associated with lower risk of oropharyngeal cancer (HR 0.93, 95% CI: 0.87, 0.998). The categorical analysis showed that participants who reported that they spent no hours using computers had a higher risk of oropharyngeal cancer (HR 1.27, 95% CI: 1.03, 1.56), and ovary cancer (HR 1.23, 95% CI: 1.01, 1.50) compared to participants who reported ≤1 h of daily recreational computer time.
Table 5

Results of Cox proportional hazards analyses investigating the association between self-report daily computer use time and cancer incidence

1 h increase in computer use timep-valueNone≤1 h (reference)1- ≤ 3 h> 3 h
Person-years3,498,487969,7211,744,785582,168201,813
Skin, melanomaCases162140485227689
HR (95% CI)*1.01 (0.98 1.05)0.43**0.77 (0.68 0.86)1.000.90 (0.79 1.04)0.92 (0.74 1.14)
HR (95% CI)†1.01 (0.97 1.05)0.720.90 (0.79 1.02)1.000.97 (0.85 1.12)0.99 (0.79 1.24)
HR (95% CI)a1.01 (0.97 1.05)0.730.91 (0.80 1.03)1.000.98 (0.85 1.12)0.996 (0.79 1.25)
OropharyngealCases5612092398825
HR (95% CI)*0.90 (0.84 0.97)0.0041.56 (1.29 1.88)1.000.97 (0.76 1.23)0.81 (0.54 1.23)
HR (95% CI)†0.93 (0.87 0.998)0.041.27 (1.03 1.56)1.000.91 (0.71 1.17)0.77 (0.51 1.17)
HR (95% CI)0.93 (0.87 0.998)0.041.27 (1.03 1.56)1.000.91 (0.71 1.17)0.77 (0.51 1.17)
LungCases2040894700316130
HR (95% CI)*0.97 (0.93 1.003)0.08**1.84 (1.66 2.03)1.001.17 (1.03 1.34)1.68 (1.39 2.03)
HR (95% CI)†1.02 (0.99 1.06)0.16**1.10 (0.99 1.23)1.001.01 (0.88 1.16)1.33 (1.10 1.62)
HR (95% CI)h1.02 (0.99 1.06)0.16**1.11 (0.99 1.24)1.000.996 (0.87 1.15)1.36 (1.12 1.65)
Breast (female only)Cases565017282931762229
HR (95% CI)*1.01 (0.99 1.03)0.27**0.93 (0.88 0.99)1.001.02 (0.94 1.10)0.997 (0.87 1.14)
HR (95% CI)†1.003 (0.98 1.03)0.83**0.97 (0.91 1.04)1.000.999 (0.92 1.08)1.001 (0.87 1.15)
HR (95% CI)b, f, h1.01 (0.98 1.03)0.57**0.96 (0.90 1.04)1.000.998 (0.91 1.09)1.03 (0.89 1.20)
UterusCases86331538912732
HR (95% CI)*1.01 (0.95 1.07)0.741.18 (1.02 1.38)1.001.25 (1.03 1.53)1.12 (0.78 1.60)
HR (95% CI)†0.96 (0.90 1.02)0.201.18 (0.999 1.39)1.001.04 (0.84 1.28)0.91 (0.63 1.33)
HR (95% CI)c0.98 (0.92 1.04)0.471.16 (0.98 1.37)1.001.08 (0.87 1.35)0.95 (0.64 1.40)
OvaryCases5672112666327
HR (95% CI)*0.98 (0.91 1.05)0.511.16 (0.96 1.39)1.000.91 (0.69 1.19)1.36 (0.91 2.02)
HR (95% CI)†0.96 (0.89 1.04)0.361.23 (1.01 1.50)1.000.91 (0.69 1.21)1.36 (0.91 2.04)
HR (95% CI)0.96 (0.89 1.04)0.361.23 (1.01 1.50)1.000.91 (0.69 1.21)1.36 (0.91 2.04)
ProstateCases5933154326991298393
HR (95% CI)*1.005 (0.99 1.02)0.61**0.91 (0.85 0.97)1.000.97 (0.91 1.03)0.97 (0.87 1.07)
HR (95% CI)†0.998 (0.98 1.02)0.85**0.98 (0.92 1.06)1.000.99 (0.93 1.06)0.9998 (0.90 1.12)
HR (95% CI)h0.997 (0.98 1.02)0.73**0.99 (0.92 1.06)1.000.99 (0.93 1.06)0.99 (0.89 1.11)
OesophagusCases53017422110827
HR (95% CI)*0.97 (0.90 1.03)0.321.20 (0.98 1.47)1.001.13 (0.90 1.42)0.94 (0.63 1.41)
HR (95% CI)†0.97 (0.90 1.04)0.370.99 (0.79 1.23)1.001.02 (0.81 1.30)0.78 (0.51 1.19)
HR (95% CI)f0.97 (0.90 1.04)0.350.99 (0.79 1.23)1.001.02 (0.80 1.30)0.77 (0.51 1.18)
StomachCases3491331336122
HR (95% CI)*0.98 (0.90 1.06)0.601.50 (1.18 1.92)1.001.09 (0.80 1.47)1.30 (0.83 2.05)
HR (95% CI)†0.98 (0.91 1.07)0.711.16 (0.88 1.51)1.000.98 (0.72 1.35)1.04 (0.64 1.69)
HR (95% CI)0.98 (0.91 1.07)0.711.16 (0.88 1.51)1.000.98 (0.72 1.35)1.04 (0.64 1.69)
Oesophagus and stomachCases87330535216848
HR (95% CI)*0.97 (0.92 1.02)0.251.31 (1.12 1.53)1.001.11 (0.93 1.34)1.06 (0.79 1.44)
HR (95% CI)†0.97 (0.92 1.03)0.331.05 (0.89 1.25)1.001.01 (0.83 1.22)0.86 (0.62 1.19)
HR (95% CI)f0.97 (0.92 1.03)0.321.05 (0.89 1.25)1.001.01 (0.83 1.22)0.86 (0.62 1.19)
Hepatobiliary tractCases4511701689122
HR (95% CI)*0.97 (0.90 1.04)0.421.49 (1.20 1.85)1.001.38 (1.07 1.79)1.14 (0.73 1.77)
HR (95% CI)†0.99 (0.92 1.06)0.741.21 (0.95 1.53)1.001.30 (0.997 1.69)1.02 (0.65 1.61)
HR (95% CI)0.99 (0.92 1.06)0.741.21 (0.95 1.53)1.001.30 (0.997 1.69)1.02 (0.65 1.61)
PancreaticCases60618927611427
HR (95% CI)*0.98 (0.92 1.05)0.621.01 (0.83 1.21)1.001.07 (0.86 1.34)0.87 (0.59 1.29)
HR (95% CI)†0.98 (0.92 1.05)0.620.90 (0.73 1.11)1.001.05 (0.84 1.31)0.76 (0.50 1.15)
HR (95% CI)d0.98 (0.92 1.05)0.610.90 (0.73 1.10)1.001.05 (0.84 1.31)0.76 (0.50 1.14)
KidneyCases78325133314950
HR (95% CI)*1.01 (0.96 1.07)0.601.17 (0.995 1.39)1.001.12 (0.92 1.36)1.23 (0.91 1.66)
HR (95% CI)†1.02 (0.97 1.08)0.39**1.04 (0.87 1.25)1.001.05 (0.86 1.29)1.19 (0.88 1.61)
HR (95% CI)1.02 (0.97 1.08)0.39**1.04 (0.87 1.25)1.001.05 (0.86 1.29)1.19 (0.88 1.61)
BladderCases67022727114230
HR (95% CI)*0.98 (0.93 1.04)0.541.22 (1.02 1.46)1.001.18 (0.96 1.44)0.85 (0.59 1.25)
HR (95% CI)†0.97 (0.92 1.04)0.41**1.09 (0.89 1.32)1.001.08 (0.87 1.33)0.76 (0.51 1.13)
HR (95% CI)0.97 (0.92 1.04)0.41**1.09 (0.89 1.32)1.001.08 (0.87 1.33)0.76 (0.51 1.13)
ColorectalCases331210591512556185
HR (95% CI)*0.99 (0.96 1.02)0.45**1.07 (0.99 1.16)1.000.95 (0.87 1.05)1.05 (0.90 1.23)
HR (95% CI)†0.99 (0.96 1.01)0.31**1.08 (0.99 1.18)1.000.95 (0.86 1.05)1.03 (0.88 1.21)
HR (95% CI)e, g, f (males)0.98 (0.96 1.01)0.28**1.06 (0.97 1.16)1.000.95 (0.86 1.05)1.02 (0.87 1.20)
ColonCases2124681980348115
HR (95% CI)*0.99 (0.96 1.03)0.63**1.04 (0.94 1.15)1.000.94 (0.83 1.06)1.04 (0.86 1.26)
HR (95% CI)†0.99 (0.95 1.02)0.50**1.04 (0.93 1.15)1.000.93 (0.82 1.06)1.02 (0.83 1.24)
HR (95% CI)e, g, f (males)0.99 (0.95 1.02)0.42**1.03 (0.92 1.14)1.000.93 (0.82 1.06)1.02 (0.83 1.25)
RectumCases111535450119565
HR (95% CI)*0.98 (0.94 1.03)0.42**1.12 (0.98 1.29)1.000.97 (0.82 1.14)1.04 (0.80 1.35)
HR (95% CI)†0.97 (0.93 1.02)0.281.20 (1.03 1.39)1.000.97 (0.82 1.15)1.03 (0.79 1.35)
HR (95% CI)e, g0.97 (0.93 1.02)0.281.16 (0.999 1.36)1.000.96 (0.81 1.15)0.999 (0.76 1.32)
ThyroidCases237821063514
HR (95% CI)*1.02 (0.92 1.12)0.761.31 (0.98 1.76)1.001.10 (0.75 1.61)1.32 (0.76 2.31)
HR (95% CI)†1.01 (0.91 1.11)0.861.36 (0.99 1.87)1.001.08 (0.73 1.59)1.28 (0.73 2.25)
HR (95% CI)1.01 (0.91 1.11)0.861.36 (0.99 1.87)1.001.08 (0.73 1.59)1.28 (0.73 2.25)
Brain tumoursCases4631302218725
HR (95% CI)*1.02 (0.95 1.09)0.620.95 (0.77 1.19)1.001.02 (0.79 1.31)0.93 (0.61 1.40)
HR (95% CI)†1.03 (0.96 1.10)0.390.92 (0.72 1.17)1.001.03 (0.80 1.34)0.96 (0.63 1.47)
HR (95% CI)f1.03 (0.96 1.10)0.380.92 (0.72 1.17)1.001.04 (0.80 1.34)0.97 (0.63 1.48)
Haematological malignanciesCases24467141137445150
HR (95% CI)*1.03 (0.998 1.06)0.060.95 (0.87 1.05)1.001.02 (0.92 1.14)1.14 (0.96 1.36)
HR (95% CI)†1.02 (0.99 1.05)0.240.95 (0.86 1.06)1.000.997 (0.89 1.12)1.11 (0.93 1.32)
HR (95% CI)1.02 (0.99 1.05)0.240.95 (0.86 1.06)1.000.997 (0.89 1.12)1.11 (0.93 1.32)
Non-Hodgkin’s lymphomaCases118234954522563
HR (95% CI)*1.02 (0.98 1.06)0.370.97 (0.85 1.11)1.001.09 (0.94 1.28)1.02 (0.78 1.32)
HR (95% CI)†1.01 (0.97 1.06)0.650.996 (0.86 1.15)1.001.07 (0.91 1.26)1.03 (0.79 1.34)
HR (95% CI)1.01 (0.97 1.06)0.650.996 (0.86 1.15)1.001.07 (0.91 1.26)1.03 (0.79 1.34)

*Models adjusted for age and sex (total observations = 467,656)

†Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [

aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine)

bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure)

dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no)

eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week)

fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women)

f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men)

gResults for males and females combined using meta-analysis as covariates are different

hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history)

**Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05)

Results of Cox proportional hazards analyses investigating the association between self-report daily computer use time and cancer incidence *Models adjusted for age and sex (total observations = 467,656) †Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [ aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine) bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no) eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week) fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women) f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men) gResults for males and females combined using meta-analysis as covariates are different hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history) **Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05) Participants who reported > 3 h using computers had a higher risk of lung cancer (HR 1.36, 95% CI: 1.12, 1.65) compared to participants who reported ≤1 h of daily recreational computer time.

Association of site-specific cancer risk and daily total recreational screen time

Table 6 shows the association between a 1-h increase in daily total recreational screen time and site-specific cancer risk. A 1-h increase in daily total recreational screen time was associated with a higher risk of lung cancer (HR 1.03, 95% CI: 1.004, 1.05). Results of Cox proportional hazards analyses investigating the association between self-report daily total screen time and cancer incidence *Models adjusted for age and sex (total observations = 464,424) †Models adjusted for age, sex, ethnicity (white/other), deprivation index (quintiles), education (University degree, A-levels/HNC/HND/NVQ, GCSE/O-level/CSE, OTHER, None), fruit and vegetable intake (< 5 portions/day, ≥5 portions/day), BMI (kg/m2), height (m), smoking status (never, former light smoker [< 20 pack-years], former heavy smoker [≥20 pack-years], current light smoker [< 20 pack-years], current heavy smoker [≥20 pack-years]) and alcohol intake (never, former, current [ aAdditional site-specific covariates in the final model include use of sun/UV protection (Never/rarely/sometimes; most of the time/always; do not go out in sunshine) bAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) cAdditional site-specific covariates in the final model include HRT use (ever used/never used), oral contraceptive use (ever used/never used), number of live births (0, 1, 2, 3+ live births), age at menarche (early menarche [< 12 years], menarche at 12–14 years, late menarche [≥15 years]), age at menopause (< 40 years, 40–44 years, 45–49 years, 50–54 years, 55–59 years, 60–64 years, ≥65 years, not had menopause/unsure), hysterectomy status (had hysterectomy, not had hysterectomy/unsure) dAdditional site-specific covariates in the final model include diabetes at baseline (yes/no) eAdditional site-specific covariates in the final model include diabetes at baseline (yes/no), aspirin use (regular use/non-regular use or no use), HRT use (ever used/never used; females only), red meat intake (portion/week), processed meat intake (portion/week) fFinal model also adjusted for waist-hip ratio (> 94 cm in men, > 80 cm in women) f(males)For cancer sites which were adjusted for different sets of covariates for males and females (colorectal, colon, rectum), this indicates that the final model for male participants was also adjusted for waist-hip ratio (> 94 cm in men) gResults for males and females combined using meta-analysis as covariates are different hFinal model also adjusted for family history of cancer (mother/father/sibling had cancer, no family history) **Schoenfeld test indicated potential violation of the proportional hazards assumption (p < 0.05) Participants who reported > 8 h of daily total recreational screen time had a higher risk of lung cancer (HR 1.45, 95% CI: 1.19, 1.77) but a lower risk of oesophagus cancer (HR 0.54, 95% CI: 0.29, 0.99) compared to participants who reported 1- ≤ 4 h of daily total recreational screen time.

Discussion

Overview of key findings

This large, prospective cohort study indicates that daily recreational screen time was associated with some site-specific cancers (notably oropharyngeal, oesophagus and stomach, colon, and lung cancer), particularly for TV viewing time, albeit mainly small associations were found. Results for oesophagus and stomach cancers, and colon cancers were robust to the omission of cancers occurring within the first two years of follow-up. However, for many of the other cancer sites the associations were attenuated after eliminating cancers diagnosed within two years, suggesting reverse causation. The results of our isotemporal substitution models revealed a benefit in terms of reduced risk of several site-specific cancers when replacing 1-h/day of TV viewing with 1-h/day of moderate-intensity physical activity or walking. Results were less consistent for daily recreational computer time and daily total recreational screen time, and were often in the opposite direction to daily TV viewing time. This may suggest that the mechanism of action is more nuanced and complex than the act of being sedentary, but that the specific activity undertaken during sedentary time (i.e. watching TV or using the computer) is an important mechanistic driver. Indeed, Patterson et al., (2018) suggested that sedentary behaviour was not a homogeneous behaviour and found that different sedentary behaviours had different determinants [44]. This will be explored further below.

Daily TV viewing time and site-specific cancer risk

Television viewing was the most common recreational screen time in this population. Our results showed that a 1-h/day increase in TV viewing time was associated with higher risk of oropharyngeal, stomach, oesophagus and stomach, and colon cancers. Compared with our reference group of 1–3 h/day of TV viewing, reporting less than 1-h/day TV viewing was associated with decreased risk of lung, breast, stomach, and oesophagus and stomach cancers. Thus our analytical approach (setting the reference group to 1–3 h/day as opposed to zero hours/day TV viewing time) enables the exploration of the possible benefits of zero TV viewing time hours for these cancers. There is some evidence in the literature that higher levels of physical activity may reduce lung cancer risk. Mechanistically, this is likely to be due to increased respiratory ventilation, which has the effect of reducing the concentration of carcinogenic agents in the lungs [45]. Previous research also provides evidence for a relationship between higher levels of physical activity and lower risk of incident breast cancer due to decreased sex and metabolic hormone levels, decreased adiposity, reductions in insulin resistance and reduced inflammation [41, 46–49]. It is plausible that similar mechanisms could be applied to the relationship between these cancers and recreational screen time. Previous research has suggested that individuals who have increased TV viewing time tend to have poor lifestyle behaviours, such as being more likely to smoke, eating a poor diet, doing little, if any, physical activity, and being overweight or obese [7]. Further, Ogden et al. (2013) discussed the concept of ‘mindless eating’, where the distraction of watching the TV led to individuals consuming more calories [50]. A review of the literature on sedentary behaviour and biological pathways by Lynch (2010) supported the hypothesised role of adiposity and metabolic dysfunction as mechanisms operant in the association between sedentary behaviour and cancer [7]. Our findings and other evidence would suggest that recreational sedentary behaviour (including screen time) is much more than an act of not being ‘active’ or being in a stationary position for a prolonged period, but rather a range of sedentary behaviours where the ‘activity’ being undertaken while sedentary is very important. Subsequently, the mechanisms of action for the association between sedentary behaviours and cancer risk are likely to act via a number of complex pathways. For example, TV viewing has been associated with increased risk of being obese or overweight [51], and there is also a strong evidence base associating being overweight or obese to increased cancer risk [7, 52]. However, we adjusted for BMI in our models to try to account for this. Known mechanisms associated with body fatness, such as sex hormones, insulin, and inflammation, may explain part of the association between recreational screen time and cancer risk. The association between prolonged TV viewing time and lower levels of vitamin D has also been hypothesised as a possible mechanistic pathway [7, 11]. However, the association between TV viewing and cancer risk may also be explained by unmeasured confounders, as people who do not watch TV are likely to be different from the broader population in a number of ways.

Daily recreational computer time and site-specific cancer risk

The mean recreational computer use time was 1.1 h/day, which is almost three times less prevalent as a recreational screen time than daily TV viewing time within this UK population. Paradoxically, our findings showed that a 1-h/day increase in recreational computer use was associated with lower risk of oropharyngeal cancer and the results of the categorical analysis showed that 0 h/day of recreational computer use was associated with higher risk of oropharyngeal and ovary cancers compared with ≤1 h/day. Reporting > 3 h/day of recreational computer use was also associated with increased risk of lung cancer compared with ≤1 h/day. It is difficult to compare the findings for computer use with other literature given the explicit exclusion of ‘using a computer at work’ from our measure. Most of the previous literature is focused on occupational sedentary time which largely encompasses computer use [17].

Daily total recreational screen time and site-specific cancer risk

The mean daily total recreational screen time was almost 4 h/day, reflecting combined TV viewing and recreational computer use time. The most notable associations were observed for an increased risk of lung cancer in both continuous and categorical analysis. Previous literature has demonstrated that household air pollution exposure from solid fuel is associated with high rates of lung cancer, especially in low- and middle-income countries, such as China [53]. However, this seems an unlikely mechanistic pathway in the UK. It is plausible that indoor sedentary behaviour may be linked to increased residential radon exposure which is known to be associated with an increased risk of lung cancer, particularly in European populations [54]. Results were somewhat mixed for other cancers which may be due to the combined nature of essentially two different behaviours (i.e. TV viewing and recreational computer use).

Findings in relation with other literature

Our observations are somewhat mixed to those previously reported for oesophago-gastric cancer risk [16] and colon cancer risk [17] in relation to sedentary behaviour. However, it is difficult to draw direct comparisons between these studies and our current analysis, since each of those used the lowest category of screen-time exposure as their reference category. Due to our a priori hypothesis that individuals with less than 1-h/day of screen time may have different characteristics, we chose 1–3 h of screen-time as our reference category. This revealed some novel associations not previously identified, such as protective associations for oesophageal and stomach cancers in individuals with the lowest screen-time exposure.

Strengths and limitations

This study provides a comprehensive overview of recreational screen time for site-specific cancers. By investigating all cancer sites within the same analytical population, using the same measurement tool for recreational screen time, we hoped to reduce the likelihood of differential measurement error explaining any inconsistencies in the association between screen time and site-specific cancer risk. However, we appreciate that investigating a large number of associations in one analysis may have led to spurious findings. The findings from the partition and isotemporal substitution models are the first, to our knowledge, to model the impact of displacing 1-h/day of TV viewing time with more physically active behaviours for site-specific cancer risk. The UK Biobank has previously been criticised for not being a representative sample for physical activity levels, obesity prevalence and other co-morbidities, indicating a healthy volunteer bias. However, the cohort is representative of the UK population in terms of age, sex, ethnicity and deprivation for the targeted age group [15, 55] and a recently published generalisability study suggests that the results of UK Biobank studies can be generalised to England and Scotland [56]. All models were adjusted for important socio-demographic, health and behavioural variables, including BMI, which is hypothesised to be on the causal pathway between screen time and cancer incidence. Some have argued that this may lead to over-adjustment and therefore underestimation of the strength of the tested associations [15]. Due to the large amount of missing data, the analyses were not adjusted for total calorie consumption or dietary habits other than total fruit and vegetable intake, red and processed meat consumption. Further, we have interpreted our effect modification results with caution owing to the number of cancer sites and number of subgroups which have been investigated. The analysis uses self-report recreational screen time data, which may be subject to social desirability and recall bias, and the measure has not been investigated for criterion validity [15]. However, the estimates are in line with previous population estimates [57, 58]. Although the UK Biobank cohort does measure sedentary behaviour using accelerometers, we were unable to use this data to examine the association with cancer incidence as the follow-up time was too short (mean follow-up time 1.9 years). The nature of the observational study means that we cannot attribute causal interpretations to our results owing to the potential for residual confounding, particularly for alcohol and tobacco-related cancers. During peer-review, our statistical analysis approach was critiqued for not applying causal inference methodologies. It is our opinion that further biological understanding of the associations shown is required to draw such conclusions. However, we accept that collider bias is possible due to potential bi-directional relationships between screen time, the covariates included in our statistical models, and cancer risk. This would affect the precision of the risk estimates shown. Finally, some associations were attenuated when excluding cancers diagnosed within the first two years of follow-up, suggesting that our results could have been affected by a possible reverse causation bias. An alternative explanation is that the results for some sites became non-significant due to the drop in the number of cancer cases resulting from excluding cancers diagnosed over the 2-year period.

Future research

Due to the small and inconsistent associations demonstrated in this study, further research is needed to explore the varying and possible trivial associations, which may be due to the large sample size of the UK Biobank cohort. Given the contrasting findings for TV viewing time and recreational computer use time, future research should take a more nuanced approach to exploring recreational screen time. This might help provide a better understanding of the underlying mechanisms of action. The literature to date is dominated by daily and weekly duration of sedentary behaviours. Increasing our knowledge about the role of bouts of sedentary behaviour and the impact of breaks in sedentary behaviour could help us develop more specific time-based recommendations and contribute to the development of much needed cancer prevention strategies. Analysing accelerometer data in large prospective cohorts in future will allow such analyses to be conducted. Accelerometer data has been assessed in UK Biobank during secondary waves of data collection, and so this will be possible given longer follow-up in due course. In addition, the current analysis focussed on site-specific cancer risk, but much remains unknown about the interactive effects of physical activity and recreational screen time on cancer mortality. These areas of research have been highlighted as important evidence gaps in the US 2018 physical activity guidelines [6].

Conclusions

In summary, our findings show that daily recreational screen time was associated with some site-specific cancers (including oesophagus and stomach, and colon cancers), particularly for daily TV viewing time. Our findings were less consistent for daily recreational computer time and daily total recreational screen time. Substitution models showed that replacing 1-h/day of TV viewing with 1-h/day of moderate-intensity physical activity or walking was associated with lower risk of several site-specific cancers (including oropharyngeal, lung, breast, and colorectal). However, further research from large prospective cohort studies are required to replicate these findings. Supplementary file 2: Testing linearity assumptions using restricted cubic splines. Additional file 1. Supplementary file 1: Subgroup analyses and sensitivity analyses.
  42 in total

1.  Healthy percentage body fat ranges: an approach for developing guidelines based on body mass index.

Authors:  D Gallagher; S B Heymsfield; M Heo; S A Jebb; P R Murgatroyd; Y Sakamoto
Journal:  Am J Clin Nutr       Date:  2000-09       Impact factor: 7.045

2.  Physical activity and risk of endometrial cancer: a population-based prospective cohort study.

Authors:  Emilie Friberg; Christos S Mantzoros; Alicja Wolk
Journal:  Cancer Epidemiol Biomarkers Prev       Date:  2006-10-20       Impact factor: 4.254

3.  Fatness, fitness, and cardiometabolic risk factors in middle-aged white men.

Authors:  Gary O'Donovan; Edward Kearney; Roy Sherwood; Melvyn Hillsdon
Journal:  Metabolism       Date:  2011-08-04       Impact factor: 8.694

Review 4.  Sedentary behavior and cancer: a systematic review of the literature and proposed biological mechanisms.

Authors:  Brigid M Lynch
Journal:  Cancer Epidemiol Biomarkers Prev       Date:  2010-09-10       Impact factor: 4.254

5.  Physical activity and breast cancer risk by pathological subtype.

Authors:  Virginia Lope; Miguel Martín; Adela Castelló; Soraya Casla; Amparo Ruiz; Jose Manuel Baena-Cañada; Ana Mª Casas; Lourdes Calvo; Begoña Bermejo; Montserrat Muñoz; Manuel Ramos; Ana de Juan-Ferré; Carlos Jara; Antonio Antón; Mª Ángeles Jimeno; Ana Lluch; Silvia Antolín; José Ángel García-Sáenz; Purificación Estévez; Esperanza Arriola-Arellano; Joaquín Gavilá; Beatriz Pérez-Gómez; Eva Carrasco; Marina Pollán
Journal:  Gynecol Oncol       Date:  2017-01-03       Impact factor: 5.482

6.  Association between TV viewing and heart disease mortality: observational study using negative control outcome.

Authors:  Mark Hamer; Ding Ding; Josephine Chau; Mitch J Duncan; Emmanuel Stamatakis
Journal:  J Epidemiol Community Health       Date:  2020-01-20       Impact factor: 3.710

7.  Isotemporal substitution analysis for physical activity, television watching, and risk of depression.

Authors:  Rania A Mekary; Michel Lucas; An Pan; Olivia I Okereke; Walter C Willett; Frank B Hu; Eric L Ding
Journal:  Am J Epidemiol       Date:  2013-06-19       Impact factor: 4.897

8.  Letter to the editor: standardized use of the terms "sedentary" and "sedentary behaviours".

Authors: 
Journal:  Appl Physiol Nutr Metab       Date:  2012-04-27       Impact factor: 2.665

9.  Sedentary Behavior Research Network (SBRN) - Terminology Consensus Project process and outcome.

Authors:  Mark S Tremblay; Salomé Aubert; Joel D Barnes; Travis J Saunders; Valerie Carson; Amy E Latimer-Cheung; Sebastien F M Chastin; Teatske M Altenburg; Mai J M Chinapaw
Journal:  Int J Behav Nutr Phys Act       Date:  2017-06-10       Impact factor: 6.457

10.  Ethnic-specific obesity cutoffs for diabetes risk: cross-sectional study of 490,288 UK biobank participants.

Authors:  Uduakobong E Ntuk; Jason M R Gill; Daniel F Mackay; Naveed Sattar; Jill P Pell
Journal:  Diabetes Care       Date:  2014-06-29       Impact factor: 19.112
View more
  2 in total

1.  Risk Factors for Female Breast Cancer: A Population Cohort Study.

Authors:  Yu-Chiao Wang; Ching-Hung Lin; Shih-Pei Huang; Mingchih Chen; Tian-Shyug Lee
Journal:  Cancers (Basel)       Date:  2022-02-03       Impact factor: 6.639

2.  Unhealthy Lifestyle, Genetics and Risk of Cardiovascular Disease and Mortality in 76,958 Individuals from the UK Biobank Cohort Study.

Authors:  Katherine M Livingstone; Gavin Abbott; Joey Ward; Steven J Bowe
Journal:  Nutrients       Date:  2021-11-27       Impact factor: 5.717
  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.