This article was downloaded by: [University of Otago] On: 26 December 2014, At: 12:09 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Research Quarterly for Exercise and Sport Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/urqe20
Television Viewing and Physical Fitness in Adults Larry A. Tucker
a
a
Health Promotion , Brigham Young University , USA Published online: 08 Feb 2013.
To cite this article: Larry A. Tucker (1990) Television Viewing and Physical Fitness in Adults, Research Quarterly for Exercise and Sport, 61:4, 315-320, DOI: 10.1080/02701367.1990.10607493 To link to this article: http://dx.doi.org/10.1080/02701367.1990.10607493
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
TUCKER
REsEARCH QUARTERLY FOR EXERClSB AND SPORT
1990, VOL. 61, No.4, pp. 315-320
Television Viewing and Physical Fitness in Adults
Downloaded by [University of Otago] at 12:09 26 December 2014
LARRY A. TUCKER Brigham Young University
The purpose ofthis study was to determine the extent to which time spent watching television is associated with cardiovascular fitness among 8,885 adults. Potential confounding effects ofage, gender, smoking, length ofwork week, time reported exercising each week, and obesity were also examined. Subjects who watched IV more than 4 hours per day (frequent viewers) were 037 times as likely to be physically fit as those who watched IV less than 1 hour per day (infrequent viewers) with age and gender controlled. Similarly, adults who watched IV 3-4 hours per day (moderately frequent viewers) were 0.45 times as likely to be fit as infrequent watchers. Adjustment for potential confounders, particularly measured body fat and reported exercise duration in combination, weakened the IV viewing/fitness relation moderately. Given the findings ofthis study and the results ofprevious research, caution should be exercised regarding excessive television viewing.
Key words: fitness, exercise, television, lifestyle, obesity,
sedentaryliving Watching television is the most pervasive pastime in Americatoday.Next to sleepand work,television viewing is the nation's third most time-consuming activity. The typical child spendsmore time in front of the TV set each year than in the classroom, and the average adult watches television nearly four hours daily (Comstock, Chaffee, Katzman, McCombs, & Roberts,1978; Pearl,Bouthilet, & Lazar, 1982; Singer, 1983). Studies regarding the content of television programs indicatethat a multitudeof health-related messages are conveyedregularlytoviewers. Moreover, contemporary research suggeststhatthismediumisoneof themostpowerfulteachers in our society (pearl et al., 1982). Unfortunately, "script writersand television producersare generally unaware of the potentiality for incidental learning of health damaging-or health promoting-lifestyles as people watch their dramatic
productions" (Hamburg & Pierce, 1982,p. 289). To date, most of the concern regarding television has been connectedwith the misleading and distorted messages and modelsthat television sometimes presentsto viewers. A numberof studiesreviewedby Pearletal. (1982)indicatethat television viewing, especially in large amounts, maybe hazardous to health. According to Hamburgand Pierce (1982), "televisioncan be presumed to influencehealthattitudes and healthbehaviors"(p. 289).Moreover, Singer(1985)indicates that"widespread television viewingbythegeneralpopulation, especially youngadults and children,may be contributing to maladaptive healthhabits" (p. 668). Although manyof the "micro-lessons" to whichAmericans are regularlyexposedmaypromotemisconceptions and produce unhealthy habits, television's primary offensemay be one of omission rather thancommission. Bronfenbrenner (1973)indicatesthat"The majorimpactof television isnotthe behaviorit produces,but the behaviorit prevents" (p. 277). Repeatedly, research has shown that television has altered American leisure (Comstock, 1982; Comstocket al., 1978). When individuals are tuned in to television, activityceases. Moreover, time for physicalactivityis reducedand the heart and other muscles of the body are not strengthened during television viewing. Tucker (1986) recently studied the relation between television viewingand physicalfitness. Results showed that astelevision watchingincreased among379adolescent males, multiplemeasures of physicalfitnessdecreased markedly and systematically. Considering the tremendous amountof time American adults spend watching television, the passivity of television viewing, and the findings of past research, it is reasonable toexpecttelevision viewing to relatesignificantly with physical fitness in adults. Such research is, however, nonexistent. Hence,thepurposeof thisstudywas to determine the extent of the association between television viewing and
RESEARCH QUARTERLY FOR EXERCISBAND SPORT, VOL.
315
61, No.4
TuCKER
physicalfitness withina large sampleof adults. An ancillary objectivewas to determine the extent to which age, gender, cigarette smoking, obesity, time reported exercising, and hoursworkedper week confoundthe television viewingand fitnessrelation.
Method
Downloaded by [University of Otago] at 12:09 26 December 2014
Subjects A sampleof 8,885 adults with a meanage of 37.6 (SD = 11.4) wasstudied. Subjects wereemployeesofover75different companies thatparticipated intheHealthExaminationProgram offeredby HealthAdvancement Services,Inc.(HAS). HASis a corporation that fosters health promotion and health risk reduction intheworkplace through theuseofmultiplescreening activities, health education and lifestyle counseling, and medical referral. Approximately 85%oftheemployeeseligible for participation in the HAS programduring the data collection phase of this study completedthe screening tests.Fiftysix percentof the participants were male,68% weremarried, 84% were white, and 61% had some collegeeducation. The medianand modalannualgross familyincomewas$25,000$30,000.
Procedure Alldatawerecollectedby registerednursesemployed by HAS.Each subjectwas examinedindividually and privately for approximately 50 minafter participating in an orientation and completing an informedconsentform.Nursesemployed by HAS are trained and evaluated regarding screening and measurement skills at the time of initial employment with HASand screeningreviewsessionsare conducted regularly. Consultants areemployedperiodically to reviewandevaluate the testingmethods used by the nursesto maximize accuracy and reliability. For the present study, a written questionnaire was administered to assess demographic and lifestyle information. Timespentwatching television perdaywasself-reported, and the measure was validated indirectly in previous research (Tucker, 1985, 1986, 1987; Tucker & Friedman, 1989), as was assessment of cigarettessmokedper day, hours worked per week,and time spentexercisingper week(Tucker, 1989; Tucker,Cole, & Friedman,1987; Tucker & Friedman, 1989). The quantitative history method (LaPorte, Montoye, & Caspersen, -1985) was employed to measure the exercise duration variableusedinthepresentstudy. Subjectsresponded to multiple questions relative to their typical "exercise activities," suchasthetypesofactivities participatedin(walking, jogging, swimming, weight training, etc.), the number of years of involvement, and the total amount of time spent exercising eachweek.Thelatter variablewasusedforcontrol purposesin this investigation.
A Harpenden skinfoldcaliper was used to assesssubcutaneous fat at three body sites and the sum of the skinfold measurements alongwithage andgenderwereusedtocalculatethetotalbodyfatpercentageof eachsubject(Baun,Baun, & Raven,1981). Thesites usedfor theassessment of bodyfat wereas follows: for males (a)thigh-anterior aspectmidway between the hip and the knee (verticalfold), (b) chest-half the distancebetweenthe anterioraxillaryline and the nipple (diagonal fold),and(c)abdomen-taken laterally ata distance of2 cm from the umbilicus (vertical fold); forfemales (a) 2.5 em above the iliac crest (diagonal fold), (b) back of the arm midway betweenthe shoulderand elbow (verticalfold), and (c) thigh-anterior aspect midway between the hip and the knee (vertical fold). Indirect validation of the body fat assessment showed substantial relationships with measured cardiovascular fitness in the present study and with age, television viewing, exercisehabits,andhypercholesterolemia in other studies(Tucker, 1989; Tucker & Friedman, 1989). Physicalfitnesswasassessedusinga steptest, theKasch 3-minute Pulse RecoveryTest (Kasch & Boyer, 1968). The step test is one of the most important and cost-effective methods of assessing adult cardiovascular fitness in largescale epidemiologic studies (Astrand & Rodahl, 1986). The test exhibitsa significantdegreeof predictiveutilityrelative to other more comprehensive measures of cardiovascular fitness,suchas treadmill testing(VOzmax) andthe 12-minrun (Kasch,Phillips,Ross, Carter, & Boyer, 1966; Nagle,Balke, & Naughton, 1965),whichare not practicalwhenevaluating over 8,000 subjects. Validity of the fitness measure was confirmed and afforded additional credence by the present study, since scores on the step test were highly related to gender, age,obesity, smoking, andexercisestatus,whileother research has provided similar indirect validation (Tucker, 1989; Tucker, in review; Tucker, Cole, & Friedman, 1986; Tucker & Friedman, 1989).
Data Analysis Subjectswhoreportedlessthan 1hourofTVviewingper day were classified as infrequent viewers, and those who reported 1-2 hours, 3-4 hours, or more than 4 hours per day were categorized as moderateviewers, moderately-frequent viewers, and frequent viewers, respectively. Physicalfitness wasrecordedas excellent, verygood,good,fair,poor,or very poor, based on subjects' age, gender, and exercise-recovery heart rates. Regarding the control variables, obesity was defmedas body fat of 20% or greater for males and 30% or more for females (Sharkey, 1984). Males and females were classified as lean when body fat was 10% or less and 16% or less, respectively. Moderate bodyfatwasconsidered 11%-19% for males and 17%-29% for females. The smoking and exercise variables were grouped as follows: never smoked, ex-smoker, light-moderate smoker (1-20cigarettes perday),or heavysmoker(over20 cigarettes perday)andnonexerciser, moderateexerciser(30-200minof
RESEARCH QUARTERLY fOR EXERCISE AND SPOIIT, VOL.
316
61, No.4
TuCKER
Downloaded by [University of Otago] at 12:09 26 December 2014
exerciseper week), and heavy exerciser (more than 200 min of exercise per week), respectively. Hours worked per week (40 hours or less, 41-50, 51 hours or more) and age (19-29 years,3Q-39,4Q-49, 50-59,60 yearsor more)werealsotreated as categorical variables. The relation of physical fitness to time spent watching television was measured by the odds ratio with infrequent viewersas the referencegroup (Kelsey,Thompson,& Evans, 1986).Subjects were consideredphysicallyfit if they scored in the excellent or very good categories on the step test. To adjust statistically for the potential confounders, MantelHaenszelsummaryrisk estimateswereused,and to ascertain the extentto which thecontrol variableswerelinearlyassociated withtheTV and fitnessmeasures, Mantel-Haenszel Chisquare statistics were calculated(Landis, Heyman, & Koch, 1978; Mantel, 1963; Mantel & Haenszel,1959).
401 (4.5%) indicated more than 4 hours of daily television viewing. Regardingphysical fitness,423(4.8%)wereclassified as excellent,777 (8.7%)wereverygood, 1,671 (18.8%)were good, 1,544(17.4%)werefair, 1,300(14.6%)werepoor,and 1,816(20.4%)were classifiedas very poor. A total of 1,354 (15.2%)adults elected not to take the step test The association between physical fitness and television viewing is shown in Table 1. As television viewing time increased, the prevalence of physically fit adults decreased significantly (19.1% to 9.5%), reflecting a strong and systematic relation. Frequent television watchers showed the lowest prevalence of fitness, followed by moderately-frequent viewers, then moderate viewers, and lastly infrequent TV watchers. Comparedto infrequentTVviewers,prevalenceofphysically fit adults(ratingof excellentor verygoodon thesteptest)was 27.8%loweramongmoderateviewers,41.2%reducedamong moderatelyfrequentviewers, and50.3%loweramongfrequent television viewers.Similarly,while only 10.5%of the infrequent viewers elected not to take the step test, 14.6% of the moderateviewers,18.0%of themoderately-frequent viewers, and20.5%of the frequentviewerschosenot to taketheexam. Compared to infrequent watchers, frequent television viewerswere .37 times (95% CI =0.25, 0.54) as likely to be
Results Of the 8,885 subjects, 955 (10.7%)reported less than 1 hourof TV viewingper day (referencegroup),5,300(59.7%) indicated 1-2 hours, 2,229 (25.1%) reported 3-4 hours, and
Table 1 Relation Between Television Viewing and Physical Fitness Physically Fit"
n
%
RRmhb
none
182
19.1
1.00
none age & sex age, sex, exercise, body fat allb
730
13.8
0.68 0.62
0.57-0.81 0.52-0.75
0.74 0.81
0.61-0.91 0.64-1.00
none age & sex age, sex, exercise, body fat allb
250
0.54 0.45
0.44-0.66 0.36-0.56
0.64 0.76
0.50-0.81 0.57-1.02
none age & sex age, sex, exercise, body fat allb
38
0.44 0.37
0.31-0.64 0.25-0.54
0.46 0.50
0.29-0.70 0.27-0.95
TV Group
Variable Controlled
4 hours/day (N= 401)
11.2
9.5
95% CI
Note. Subjects watching TV less than 1 hour/day (n = 955) were used as the reference group; RR mh = Mantel-Haenszel summary estimate of relative risk. "Subjects were classified as physically fit if they scored within the excellent or very good categories on the step test. b All includes control of age, sex, exercise duration, smoking, length of work week, and body fat. RESEARCH QUARTERLY FOR EXERCISE AND SPORT, VOL.
317
61, No.4
Downloaded by [University of Otago] at 12:09 26 December 2014
TuCKER
physically fit withage and sexcontrolled,while moderatelyfrequent viewers were .45 times (95% CI = 0.36, 0.56) as likelyand moderateviewerswere .62 times (95% CI =0.52, 0.75) as likely to be fit after adjustingfor differences in age and sex (see Table 1). Controlof the other potentialconfounders individually, inadditiontoageandsex,hadsundryeffectson theTV/fitness relations. Specifically, adjustmentfor body fat, in additionto age and sex, increasedthe odds ratios 18.2% on the average, whileadjustmentfor time reportedexercising, smoking, and hoursworkedper week,in additionto age and sex, increased the odds ratios 12.2%, 14.9%, and 1.8%, respectively. Adjustment for exercise duration and body fat, along with age andsex,increasedtheestimateofrelativeriskby 28.2%onthe average. Controlfor all of the potentialconfounders in combinationincreasedthe odds ratios an averageof 44.9% compared to the estimates with only age and sex controlled. To better understand the sources of the differences in physical fitness among the TV groups, the associations between the control variables, TV viewing, and fitness were examined. FromtheresultsinTable 2, it isclearthattelevision viewing was related to all of the potential confounders, particularly cigarettesmokingand hoursworkedper week.In short, frequent television viewers tended to be relatively youngor old, male, more likely to smoke,more obese, more sedentary, and workfewer hours than thosewho viewedless television. Regardingthe relationsbetween the control variables and physical fitness, several significant associations wereidentified,asrevealedinTable2.Specifically,physically fit adultstended to be male, younger, leaner,and morelikely to exercise, not smoke, and work more hours than less fit subjects.
Discussion Consistent with the results of Tucker's (I986) earlier study with a sample of adolescents, the present findings indicatedthat durationof dailytelevision viewingis strongly andinversely associated withcardiovascular fitnessin adults. In brief, frequentTV viewers were .37 times as likely to be physically fit as infrequentviewers. The significantassociations of this study between television watching andcardiovascularfitness couldbethefunction of one of the following conditions: (a) lack of fitness could cause increasedtelevision watching, (b) physicalfitness and television viewing could each be related to other factors resultingin spuriouscausal associations betweenfitness and TV viewing,or (c) excessivetelevision viewingcould cause reducedphysical fitness. The first conditionis based on the conceptof selection. Adultswholackfavorable levelsoffitnessmaybe attractedto watching TV as a primary source of recreation or entertainment, whereas physically fit adults may prefer more active
pastimes. Since mostleisureactivitiesrequiremorephysical exertion than television watching and wouldtherefore be less enticingandenjoyableto theunfit,it isconceivable thatadults whoare notphysically fit chooseto remainpassiveand watch more TV than their counterparts. The second competing explanation of the TV/fitness association involves the indirect relation or third variable issue. To help protectagainst this potentialproblem,several important variables werestatistically controlled(seeTable 1). Moreover, therelationof eachof thepossibleconfounders to TV viewing and cardiovascular fitness was evaluated (see Table2). Fewof thepotentialconfounders influenced theTV/ fitnessrelations significantly whenconsideredindividually. Although self-reportedexerciseduration washighlyrelatedto measured physicalfitness inthepresentinvestigation, control of the exercisevariableby itselfhadonly a modestimpacton the TV/fitness association. This may have occurredbecause of thesubjective natureof theexerciseduration measurement. A less crude variable, perhaps daily logs or diary reports regardingexerciseactivities, mayprovidebettervalidation of the relationship. In combination, measured bodyfat andreportedexercise duration, along with age and gender, increasedthe risk estimates by more than 25% compared to control for age and gender alone. Evidently, infrequent television viewers are significantly more fit than frequent watchers partly because infrequentviewers exercise more and are less fat than their peers. Adjustment for all of the potentialconfounders together may result in the most valid association between television viewing and fitness since all of these factors make their contributions simultaneously in real life. Becausecontrolfor age, gender,exerciseduration, smoking, body fat, andlength of work weekin combination weakenedthe relationby more than 40%, it seemsplausibleto concludethat the TV/fitness association is partly a function of these measures. The TV/fitness relation could be the result of factors unassessed in the present study. Physically fit adults may simply watch less television than less fit adults because of differences in personality, health status, leisure preferences, athletic skill development, health maintenance motivation, and the like. In 1986,Tuckerproposeda specificpathwayto explain theTV/fitness linkindicating that"as television viewing time increases, physical activity tends to decrease. As physical activitydeclines, physicalfitnesstendstodecline.Asphysical fitness declines, attraction to passive recreation [such as television watching] tends to increase" (p. 803). Findingsof the present study seem to supportTucker's thesis, because (a) as TV viewing time increased, exercise activity tendedtodecrease significantly, (b) asexerciseactivity declined, measured physical fitness tended to decline significantly, and (c) as physical fitness declined, attraction to passiverecreation, in this case television viewing, tendedto increasesignificantly. Futureresearchwillneedtoconsiderfactors notassessed
REsEARCH QUAIITERLY FOR EXEROSB AND SPOIIT, VOl..
318
61, No.4
TUCKER
in the present study to further clarify the intricacies, mediators, and direction of the TV/fitness relation. Moreover, becausesubjectsinthisstudywereemployedandhadrelatively high family incomesand education levels, generalization to unemployed, poorer,less educatedgroups will require additional study. Recommendations regardinghealthy levelsof television viewingfor adults based on scientific evidencepresentlydo not exist. If a causalrelationis assumedbetweenTV viewing andfitness, thentimespentwatching television shouldprobably be reduced in most adults, especially those who are in poor physicalcondition. Sinceexcessivetelevision viewingis also
linkedwithobesity(Dietz& Gortmaker,1985; Tucker,1989), increased alcohol use (Tucker, 1985),emotional instability, insensitivity, introversion, amoral behavior, and reduced imagination, IQ, and self-esteem (pearl et al., 1982; Tucker, 1987), cautionshouldbe exercisedregardingthis behavior. It is apparentfrom contemporary research and the findings of this study that the potential effectof television on the lifestyles and well-being of adults cannot be ignored. Since physical fitness iscloselylinkedwithcardiovascularmortality, and television viewing is highly related to fitness, it is conceivable that excessive TV viewing may be a health-risk behavior.
Table 2 Television Viewing Hours and Physical FitnessAccording to the Control Variables Fitness (row %)
Downloaded by [University of Otago] at 12:09 26 December 2014
Television Hours/Day (row %) Control Variable
col %
Research Quarterly for Exercise and Sport Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/urqe20
Television Viewing and Physical Fitness in Adults Larry A. Tucker
a
a
Health Promotion , Brigham Young University , USA Published online: 08 Feb 2013.
To cite this article: Larry A. Tucker (1990) Television Viewing and Physical Fitness in Adults, Research Quarterly for Exercise and Sport, 61:4, 315-320, DOI: 10.1080/02701367.1990.10607493 To link to this article: http://dx.doi.org/10.1080/02701367.1990.10607493
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
TUCKER
REsEARCH QUARTERLY FOR EXERClSB AND SPORT
1990, VOL. 61, No.4, pp. 315-320
Television Viewing and Physical Fitness in Adults
Downloaded by [University of Otago] at 12:09 26 December 2014
LARRY A. TUCKER Brigham Young University
The purpose ofthis study was to determine the extent to which time spent watching television is associated with cardiovascular fitness among 8,885 adults. Potential confounding effects ofage, gender, smoking, length ofwork week, time reported exercising each week, and obesity were also examined. Subjects who watched IV more than 4 hours per day (frequent viewers) were 037 times as likely to be physically fit as those who watched IV less than 1 hour per day (infrequent viewers) with age and gender controlled. Similarly, adults who watched IV 3-4 hours per day (moderately frequent viewers) were 0.45 times as likely to be fit as infrequent watchers. Adjustment for potential confounders, particularly measured body fat and reported exercise duration in combination, weakened the IV viewing/fitness relation moderately. Given the findings ofthis study and the results ofprevious research, caution should be exercised regarding excessive television viewing.
Key words: fitness, exercise, television, lifestyle, obesity,
sedentaryliving Watching television is the most pervasive pastime in Americatoday.Next to sleepand work,television viewing is the nation's third most time-consuming activity. The typical child spendsmore time in front of the TV set each year than in the classroom, and the average adult watches television nearly four hours daily (Comstock, Chaffee, Katzman, McCombs, & Roberts,1978; Pearl,Bouthilet, & Lazar, 1982; Singer, 1983). Studies regarding the content of television programs indicatethat a multitudeof health-related messages are conveyedregularlytoviewers. Moreover, contemporary research suggeststhatthismediumisoneof themostpowerfulteachers in our society (pearl et al., 1982). Unfortunately, "script writersand television producersare generally unaware of the potentiality for incidental learning of health damaging-or health promoting-lifestyles as people watch their dramatic
productions" (Hamburg & Pierce, 1982,p. 289). To date, most of the concern regarding television has been connectedwith the misleading and distorted messages and modelsthat television sometimes presentsto viewers. A numberof studiesreviewedby Pearletal. (1982)indicatethat television viewing, especially in large amounts, maybe hazardous to health. According to Hamburgand Pierce (1982), "televisioncan be presumed to influencehealthattitudes and healthbehaviors"(p. 289).Moreover, Singer(1985)indicates that"widespread television viewingbythegeneralpopulation, especially youngadults and children,may be contributing to maladaptive healthhabits" (p. 668). Although manyof the "micro-lessons" to whichAmericans are regularlyexposedmaypromotemisconceptions and produce unhealthy habits, television's primary offensemay be one of omission rather thancommission. Bronfenbrenner (1973)indicatesthat"The majorimpactof television isnotthe behaviorit produces,but the behaviorit prevents" (p. 277). Repeatedly, research has shown that television has altered American leisure (Comstock, 1982; Comstocket al., 1978). When individuals are tuned in to television, activityceases. Moreover, time for physicalactivityis reducedand the heart and other muscles of the body are not strengthened during television viewing. Tucker (1986) recently studied the relation between television viewingand physicalfitness. Results showed that astelevision watchingincreased among379adolescent males, multiplemeasures of physicalfitnessdecreased markedly and systematically. Considering the tremendous amountof time American adults spend watching television, the passivity of television viewing, and the findings of past research, it is reasonable toexpecttelevision viewing to relatesignificantly with physical fitness in adults. Such research is, however, nonexistent. Hence,thepurposeof thisstudywas to determine the extent of the association between television viewing and
RESEARCH QUARTERLY FOR EXERCISBAND SPORT, VOL.
315
61, No.4
TuCKER
physicalfitness withina large sampleof adults. An ancillary objectivewas to determine the extent to which age, gender, cigarette smoking, obesity, time reported exercising, and hoursworkedper week confoundthe television viewingand fitnessrelation.
Method
Downloaded by [University of Otago] at 12:09 26 December 2014
Subjects A sampleof 8,885 adults with a meanage of 37.6 (SD = 11.4) wasstudied. Subjects wereemployeesofover75different companies thatparticipated intheHealthExaminationProgram offeredby HealthAdvancement Services,Inc.(HAS). HASis a corporation that fosters health promotion and health risk reduction intheworkplace through theuseofmultiplescreening activities, health education and lifestyle counseling, and medical referral. Approximately 85%oftheemployeeseligible for participation in the HAS programduring the data collection phase of this study completedthe screening tests.Fiftysix percentof the participants were male,68% weremarried, 84% were white, and 61% had some collegeeducation. The medianand modalannualgross familyincomewas$25,000$30,000.
Procedure Alldatawerecollectedby registerednursesemployed by HAS.Each subjectwas examinedindividually and privately for approximately 50 minafter participating in an orientation and completing an informedconsentform.Nursesemployed by HAS are trained and evaluated regarding screening and measurement skills at the time of initial employment with HASand screeningreviewsessionsare conducted regularly. Consultants areemployedperiodically to reviewandevaluate the testingmethods used by the nursesto maximize accuracy and reliability. For the present study, a written questionnaire was administered to assess demographic and lifestyle information. Timespentwatching television perdaywasself-reported, and the measure was validated indirectly in previous research (Tucker, 1985, 1986, 1987; Tucker & Friedman, 1989), as was assessment of cigarettessmokedper day, hours worked per week,and time spentexercisingper week(Tucker, 1989; Tucker,Cole, & Friedman,1987; Tucker & Friedman, 1989). The quantitative history method (LaPorte, Montoye, & Caspersen, -1985) was employed to measure the exercise duration variableusedinthepresentstudy. Subjectsresponded to multiple questions relative to their typical "exercise activities," suchasthetypesofactivities participatedin(walking, jogging, swimming, weight training, etc.), the number of years of involvement, and the total amount of time spent exercising eachweek.Thelatter variablewasusedforcontrol purposesin this investigation.
A Harpenden skinfoldcaliper was used to assesssubcutaneous fat at three body sites and the sum of the skinfold measurements alongwithage andgenderwereusedtocalculatethetotalbodyfatpercentageof eachsubject(Baun,Baun, & Raven,1981). Thesites usedfor theassessment of bodyfat wereas follows: for males (a)thigh-anterior aspectmidway between the hip and the knee (verticalfold), (b) chest-half the distancebetweenthe anterioraxillaryline and the nipple (diagonal fold),and(c)abdomen-taken laterally ata distance of2 cm from the umbilicus (vertical fold); forfemales (a) 2.5 em above the iliac crest (diagonal fold), (b) back of the arm midway betweenthe shoulderand elbow (verticalfold), and (c) thigh-anterior aspect midway between the hip and the knee (vertical fold). Indirect validation of the body fat assessment showed substantial relationships with measured cardiovascular fitness in the present study and with age, television viewing, exercisehabits,andhypercholesterolemia in other studies(Tucker, 1989; Tucker & Friedman, 1989). Physicalfitnesswasassessedusinga steptest, theKasch 3-minute Pulse RecoveryTest (Kasch & Boyer, 1968). The step test is one of the most important and cost-effective methods of assessing adult cardiovascular fitness in largescale epidemiologic studies (Astrand & Rodahl, 1986). The test exhibitsa significantdegreeof predictiveutilityrelative to other more comprehensive measures of cardiovascular fitness,suchas treadmill testing(VOzmax) andthe 12-minrun (Kasch,Phillips,Ross, Carter, & Boyer, 1966; Nagle,Balke, & Naughton, 1965),whichare not practicalwhenevaluating over 8,000 subjects. Validity of the fitness measure was confirmed and afforded additional credence by the present study, since scores on the step test were highly related to gender, age,obesity, smoking, andexercisestatus,whileother research has provided similar indirect validation (Tucker, 1989; Tucker, in review; Tucker, Cole, & Friedman, 1986; Tucker & Friedman, 1989).
Data Analysis Subjectswhoreportedlessthan 1hourofTVviewingper day were classified as infrequent viewers, and those who reported 1-2 hours, 3-4 hours, or more than 4 hours per day were categorized as moderateviewers, moderately-frequent viewers, and frequent viewers, respectively. Physicalfitness wasrecordedas excellent, verygood,good,fair,poor,or very poor, based on subjects' age, gender, and exercise-recovery heart rates. Regarding the control variables, obesity was defmedas body fat of 20% or greater for males and 30% or more for females (Sharkey, 1984). Males and females were classified as lean when body fat was 10% or less and 16% or less, respectively. Moderate bodyfatwasconsidered 11%-19% for males and 17%-29% for females. The smoking and exercise variables were grouped as follows: never smoked, ex-smoker, light-moderate smoker (1-20cigarettes perday),or heavysmoker(over20 cigarettes perday)andnonexerciser, moderateexerciser(30-200minof
RESEARCH QUARTERLY fOR EXERCISE AND SPOIIT, VOL.
316
61, No.4
TuCKER
Downloaded by [University of Otago] at 12:09 26 December 2014
exerciseper week), and heavy exerciser (more than 200 min of exercise per week), respectively. Hours worked per week (40 hours or less, 41-50, 51 hours or more) and age (19-29 years,3Q-39,4Q-49, 50-59,60 yearsor more)werealsotreated as categorical variables. The relation of physical fitness to time spent watching television was measured by the odds ratio with infrequent viewersas the referencegroup (Kelsey,Thompson,& Evans, 1986).Subjects were consideredphysicallyfit if they scored in the excellent or very good categories on the step test. To adjust statistically for the potential confounders, MantelHaenszelsummaryrisk estimateswereused,and to ascertain the extentto which thecontrol variableswerelinearlyassociated withtheTV and fitnessmeasures, Mantel-Haenszel Chisquare statistics were calculated(Landis, Heyman, & Koch, 1978; Mantel, 1963; Mantel & Haenszel,1959).
401 (4.5%) indicated more than 4 hours of daily television viewing. Regardingphysical fitness,423(4.8%)wereclassified as excellent,777 (8.7%)wereverygood, 1,671 (18.8%)were good, 1,544(17.4%)werefair, 1,300(14.6%)werepoor,and 1,816(20.4%)were classifiedas very poor. A total of 1,354 (15.2%)adults elected not to take the step test The association between physical fitness and television viewing is shown in Table 1. As television viewing time increased, the prevalence of physically fit adults decreased significantly (19.1% to 9.5%), reflecting a strong and systematic relation. Frequent television watchers showed the lowest prevalence of fitness, followed by moderately-frequent viewers, then moderate viewers, and lastly infrequent TV watchers. Comparedto infrequentTVviewers,prevalenceofphysically fit adults(ratingof excellentor verygoodon thesteptest)was 27.8%loweramongmoderateviewers,41.2%reducedamong moderatelyfrequentviewers, and50.3%loweramongfrequent television viewers.Similarly,while only 10.5%of the infrequent viewers elected not to take the step test, 14.6% of the moderateviewers,18.0%of themoderately-frequent viewers, and20.5%of the frequentviewerschosenot to taketheexam. Compared to infrequent watchers, frequent television viewerswere .37 times (95% CI =0.25, 0.54) as likely to be
Results Of the 8,885 subjects, 955 (10.7%)reported less than 1 hourof TV viewingper day (referencegroup),5,300(59.7%) indicated 1-2 hours, 2,229 (25.1%) reported 3-4 hours, and
Table 1 Relation Between Television Viewing and Physical Fitness Physically Fit"
n
%
RRmhb
none
182
19.1
1.00
none age & sex age, sex, exercise, body fat allb
730
13.8
0.68 0.62
0.57-0.81 0.52-0.75
0.74 0.81
0.61-0.91 0.64-1.00
none age & sex age, sex, exercise, body fat allb
250
0.54 0.45
0.44-0.66 0.36-0.56
0.64 0.76
0.50-0.81 0.57-1.02
none age & sex age, sex, exercise, body fat allb
38
0.44 0.37
0.31-0.64 0.25-0.54
0.46 0.50
0.29-0.70 0.27-0.95
TV Group
Variable Controlled
4 hours/day (N= 401)
11.2
9.5
95% CI
Note. Subjects watching TV less than 1 hour/day (n = 955) were used as the reference group; RR mh = Mantel-Haenszel summary estimate of relative risk. "Subjects were classified as physically fit if they scored within the excellent or very good categories on the step test. b All includes control of age, sex, exercise duration, smoking, length of work week, and body fat. RESEARCH QUARTERLY FOR EXERCISE AND SPORT, VOL.
317
61, No.4
Downloaded by [University of Otago] at 12:09 26 December 2014
TuCKER
physically fit withage and sexcontrolled,while moderatelyfrequent viewers were .45 times (95% CI = 0.36, 0.56) as likelyand moderateviewerswere .62 times (95% CI =0.52, 0.75) as likely to be fit after adjustingfor differences in age and sex (see Table 1). Controlof the other potentialconfounders individually, inadditiontoageandsex,hadsundryeffectson theTV/fitness relations. Specifically, adjustmentfor body fat, in additionto age and sex, increasedthe odds ratios 18.2% on the average, whileadjustmentfor time reportedexercising, smoking, and hoursworkedper week,in additionto age and sex, increased the odds ratios 12.2%, 14.9%, and 1.8%, respectively. Adjustment for exercise duration and body fat, along with age andsex,increasedtheestimateofrelativeriskby 28.2%onthe average. Controlfor all of the potentialconfounders in combinationincreasedthe odds ratios an averageof 44.9% compared to the estimates with only age and sex controlled. To better understand the sources of the differences in physical fitness among the TV groups, the associations between the control variables, TV viewing, and fitness were examined. FromtheresultsinTable 2, it isclearthattelevision viewing was related to all of the potential confounders, particularly cigarettesmokingand hoursworkedper week.In short, frequent television viewers tended to be relatively youngor old, male, more likely to smoke,more obese, more sedentary, and workfewer hours than thosewho viewedless television. Regardingthe relationsbetween the control variables and physical fitness, several significant associations wereidentified,asrevealedinTable2.Specifically,physically fit adultstended to be male, younger, leaner,and morelikely to exercise, not smoke, and work more hours than less fit subjects.
Discussion Consistent with the results of Tucker's (I986) earlier study with a sample of adolescents, the present findings indicatedthat durationof dailytelevision viewingis strongly andinversely associated withcardiovascular fitnessin adults. In brief, frequentTV viewers were .37 times as likely to be physically fit as infrequentviewers. The significantassociations of this study between television watching andcardiovascularfitness couldbethefunction of one of the following conditions: (a) lack of fitness could cause increasedtelevision watching, (b) physicalfitness and television viewing could each be related to other factors resultingin spuriouscausal associations betweenfitness and TV viewing,or (c) excessivetelevision viewingcould cause reducedphysical fitness. The first conditionis based on the conceptof selection. Adultswholackfavorable levelsoffitnessmaybe attractedto watching TV as a primary source of recreation or entertainment, whereas physically fit adults may prefer more active
pastimes. Since mostleisureactivitiesrequiremorephysical exertion than television watching and wouldtherefore be less enticingandenjoyableto theunfit,it isconceivable thatadults whoare notphysically fit chooseto remainpassiveand watch more TV than their counterparts. The second competing explanation of the TV/fitness association involves the indirect relation or third variable issue. To help protectagainst this potentialproblem,several important variables werestatistically controlled(seeTable 1). Moreover, therelationof eachof thepossibleconfounders to TV viewing and cardiovascular fitness was evaluated (see Table2). Fewof thepotentialconfounders influenced theTV/ fitnessrelations significantly whenconsideredindividually. Although self-reportedexerciseduration washighlyrelatedto measured physicalfitness inthepresentinvestigation, control of the exercisevariableby itselfhadonly a modestimpacton the TV/fitness association. This may have occurredbecause of thesubjective natureof theexerciseduration measurement. A less crude variable, perhaps daily logs or diary reports regardingexerciseactivities, mayprovidebettervalidation of the relationship. In combination, measured bodyfat andreportedexercise duration, along with age and gender, increasedthe risk estimates by more than 25% compared to control for age and gender alone. Evidently, infrequent television viewers are significantly more fit than frequent watchers partly because infrequentviewers exercise more and are less fat than their peers. Adjustment for all of the potentialconfounders together may result in the most valid association between television viewing and fitness since all of these factors make their contributions simultaneously in real life. Becausecontrolfor age, gender,exerciseduration, smoking, body fat, andlength of work weekin combination weakenedthe relationby more than 40%, it seemsplausibleto concludethat the TV/fitness association is partly a function of these measures. The TV/fitness relation could be the result of factors unassessed in the present study. Physically fit adults may simply watch less television than less fit adults because of differences in personality, health status, leisure preferences, athletic skill development, health maintenance motivation, and the like. In 1986,Tuckerproposeda specificpathwayto explain theTV/fitness linkindicating that"as television viewing time increases, physical activity tends to decrease. As physical activitydeclines, physicalfitnesstendstodecline.Asphysical fitness declines, attraction to passive recreation [such as television watching] tends to increase" (p. 803). Findingsof the present study seem to supportTucker's thesis, because (a) as TV viewing time increased, exercise activity tendedtodecrease significantly, (b) asexerciseactivity declined, measured physical fitness tended to decline significantly, and (c) as physical fitness declined, attraction to passiverecreation, in this case television viewing, tendedto increasesignificantly. Futureresearchwillneedtoconsiderfactors notassessed
REsEARCH QUAIITERLY FOR EXEROSB AND SPOIIT, VOl..
318
61, No.4
TUCKER
in the present study to further clarify the intricacies, mediators, and direction of the TV/fitness relation. Moreover, becausesubjectsinthisstudywereemployedandhadrelatively high family incomesand education levels, generalization to unemployed, poorer,less educatedgroups will require additional study. Recommendations regardinghealthy levelsof television viewingfor adults based on scientific evidencepresentlydo not exist. If a causalrelationis assumedbetweenTV viewing andfitness, thentimespentwatching television shouldprobably be reduced in most adults, especially those who are in poor physicalcondition. Sinceexcessivetelevision viewingis also
linkedwithobesity(Dietz& Gortmaker,1985; Tucker,1989), increased alcohol use (Tucker, 1985),emotional instability, insensitivity, introversion, amoral behavior, and reduced imagination, IQ, and self-esteem (pearl et al., 1982; Tucker, 1987), cautionshouldbe exercisedregardingthis behavior. It is apparentfrom contemporary research and the findings of this study that the potential effectof television on the lifestyles and well-being of adults cannot be ignored. Since physical fitness iscloselylinkedwithcardiovascularmortality, and television viewing is highly related to fitness, it is conceivable that excessive TV viewing may be a health-risk behavior.
Table 2 Television Viewing Hours and Physical FitnessAccording to the Control Variables Fitness (row %)
Downloaded by [University of Otago] at 12:09 26 December 2014
Television Hours/Day (row %) Control Variable
col %
