The 3-year evolution of a preschool physical activity intervention through a collaborative partnership between research interventionists and preschool teachers.

Despite evidence that preschoolers spend the majority of their time in sedentary activities, few physical activity interventions have focused on preschool-age children. Health promotion interventions that can be integrated into the daily routines of a school or other setting are more likely to be implemented. The Study of Health and Activity in Preschool Environments employed a flexible approach to increasing physical activity opportunities in preschools' daily schedules through recess, indoor physical activity and physical activity integrated into academic lessons. Eight preschools were randomly assigned to receive the study's physical activity intervention. Teachers in these schools partnered with university-based interventionists across 3 years to design and implement a flexible and adaptive intervention. The intervention approach included trainings and workshops, site visits and feedback from intervention personnel, newsletters, and physical activity equipment and materials. Teachers reported a high acceptability of the intervention. The purpose of this article is to describe the evolution of a multi-component physical activity intervention in preschools, including (i) a description of the intervention components, (ii) an explanation of the intervention process and approach, and (iii) a report of teachers' perceptions of barriers to implementation.

Association between objectively measured sedentary behavior and body mass index in preschool children.

OBJECTIVE: To determine the association between accelerometry-derived sedentary behavior and body mass index (BMI) z-score in preschool children, and to determine whether the association changed when applying three different accelerometry cutpoints for sedentary behavior. DESIGN AND SUBJECTS: Cross-sectional design. Data came from two completed studies: Children's Activity and Movement in Preschool Study (CHAMPS) and the Environmental Determinants of Physical Activity in Preschool Children (EDPAPC) study. Children of ages 3-5 years with complete data on sedentary behavior, BMI z-score, physical activity and other covariates were included in the analyses (N=263 in CHAMPS and N=155 in EDPAPC). Accelerometry data were summarized as time spent in sedentary behavior (min h(-1)) using three different cutpoints developed specifically for preschool children (<37.5, <200 and <373 counts per 15 s). Linear mixed regression models were used to determine the association between time spent in sedentary behavior and BMI z-score; age, gender, race, parental education, preschools and moderate-to-vigorous physical activity (MVPA) were included as covariates. RESULTS: In both CHAMPS and EDPAPC studies, no independent association between time spent in sedentary behavior and BMI z-score was observed after adjusting for MVPA. The observed null association between sedentary behavior and BMI z-score was maintained even with different sedentary behavior cutpoints. CONCLUSIONS: Regardless of cutpoints used, accelerometry-derived sedentary behavior was not independently associated with BMI z-score in two independent samples of preschool children. Longitudinal studies addressing this research question are needed.

Time spent in sedentary behavior and changes in childhood BMI: a longitudinal study from ages 9 to 15 years.

OBJECTIVE: To determine if time spent in objectively measured sedentary behavior is associated with a change in body mass index (BMI) between ages 9 and 15 years, adjusting for moderate-to-vigorous physical activity (MVPA). DESIGN: Prospective observational study of children at ages 9 (2000), 11 (2002), 12 (2003) and 15 years (2006). Longitudinal quantile regression was used to model the influence of predictors on changes at the 10th, 25th, 50th, 75th and 90th BMI percentiles over time. SUBJECTS: Participants were enrolled in the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care and Youth Development and include both boys and girls (n=789). MEASUREMENTS: Objectively measured BMI (kg m(-2)) was the outcome variable and objectively measured sedentary behavior was the main predictor. Adjustment was also made for MVPA, gender, race, maternal education, hours of sleep and healthy eating index. RESULTS: Increases in BMI were observed at all percentiles, with the greatest increase observed at the 90th BMI percentile. Spending more time in sedentary behavior (h per day) was associated with additional increases in BMI at the 90th, 75th and 50th BMI percentiles, independent of MVPA and the other covariates (90th percentile=0.59, 95% confidence interval (95% CI): 0.19-0.98 kg m(-2); 75th percentile=0.48, 95% CI: 0.25-0.72 kg m(-2); and 50th percentile=0.19, 95% CI: 0.05-0.33 kg m(-2)). No associations were observed between sedentary behavior and changes at the 25th and 10th BMI percentiles. CONCLUSION: Sedentary behavior was associated with greater increases in BMI at the 90th, 75th and 50th BMI percentiles between ages 9 and 15 years, independent of MVPA. Preventing an increase in sedentary behavior from childhood to adolescence may contribute to reducing the number of children classified as obese.

Assessing sustainability of Lifestyle Education for Activity Program (LEAP).

Sustained intervention effects are needed for positive health impacts in populations; however, few published examples illustrate methods for assessing sustainability in health promotion programs. This paper describes the methods for assessing sustainability of the Lifestyle Education for Activity Program (LEAP). LEAP was a comprehensive school-based intervention that targeted change in instructional practices and the school environment to promote physical activity (PA) in high school girls. Previous reports indicated that significantly more girls in the intervention compared with control schools reported engaging in vigorous PA, and positive long-term effects on vigorous PA also were observed for girls in schools that most fully implemented and maintained the intervention 3 years following the active intervention. In this paper, the seven steps used to assess sustainability in LEAP are presented; these steps provide a model for assessing sustainability in health promotion programs in other settings. Unique features of the LEAP sustainability model include assessing sustainability of changes in instructional practices and the environment, basing assessment on an essential element framework that defined complete and acceptable delivery at the beginning of the project, using multiple data sources to assess sustainability, and assessing implementation longitudinally.

After-school interventions to increase physical activity among youth.

Most children and adolescents do not meet the recommended 60 minutes or more of moderate to vigorous physical activity per day. One attractive approach to increasing physical activity in young people is providing activity through structured after-school programmes. This paper provides a review of the scientific literature on the effects of after-school programmes on physical activity in children and adolescents. After-school physical activity interventions provided mixed results; some increased children's physical activity, others did not. Although after-school programmes have the potential to help children and adolescents engage in regular, enjoyable physical activity, the research on these programmes is limited and, in some cases, methodologically weak. Additional, well-controlled studies are needed to identify the components of after-school programmes that promote physical activity and to determine the level of activity that can be attained when children and adolescents participate in these programmes.

Regional comparisons of walking or bicycling for fun or exercise and for active transport in a nationally distributed sample of community-based youth.

BACKGROUND: Physical activity helps prevent excessive weight gain in youth. Few studies have examined regional variations in common activities, walking and bicycling. OBJECTIVE: The primary aim was to examine participation in previous 7-day walking/bicycling for active transport and for fun/exercise by region, gender and age groups. METHODS: Children (n = 2693) from 130 diverse, geographically distributed US communities reported participation in previous 7-day walking/bicycling. Mixed model ANOVAs were used to compare walking/bicycling by region, gender and age groups using Healthy Communities Study data. RESULTS: Walking/bicycling for fun/exercise was higher in the Midwest, South and West compared with the Northeast, with the same regional patterns for boys, girls and 9-11-year olds. Walking/bicycling to non-school destinations was higher in the Midwest compared with the South and Northeast, with the same pattern for girls and 9-11-year olds. Walking/bicycling to school was higher Midwest, Northeast and West compared with the South, with similar patterns for girls and 12-15-year olds. CONCLUSIONS: Walking/bicycling for fun/exercise and to non-school destinations had higher prevalence than active transport to school. The Northeast reported lower walking/bicycling for fun/exercise; the South and Northeast reported lower walking/bicycling to non-school destinations. Program/policy initiatives should consider contextual demographic influences, as well as purpose/destination of walking/bicycling.

Relationship of objective street quality attributes with youth physical activity: findings from the Healthy Communities Study.

BACKGROUND: Residential environments may significantly influence youth physical activity (PA). However, few studies have examined detailed street quality attributes via observational audits in relation to context-specific PA among youth. OBJECTIVES: The objective of this study was to explore whether the overall quality of street environments, as well as specific attributes, was associated with neighbourhood-based and street-based PA within a national sample of youth in the Healthy Communities Study. METHODS: Data were collected from 4616 youth from 130 communities across the USA. Youth PA in the neighbourhood and on the participant's street was captured using 7-d recall interviews. Windshield survey observational audits documented five street quality variables: burned, boarded up or abandoned residential units, litter, overall condition of residences, street type and presence of sidewalks in good condition. RESULTS: Youth with no litter on their street reported significantly lower neighbourhood-based PA and youth living on a side street, cul-de-sac, dead-end or one-way street reported greater neighbourhood-based PA. No significant associations were detected for the overall street quality index or with street-based PA. CONCLUSIONS: Specific street quality attributes may be associated with youth PA. Further research and collaboration between diverse disciplines and agencies should focus on understanding and improving street quality to promote youth PA and health.

The prevalence of community programmes and policies to prevent childhood obesity in a diverse sample of US communities: the Healthy Communities Study.

INTRODUCTION: This manuscript describes the prevalence and attributes of community programmes and policies (CPPs) to address childhood obesity documented as occurring in 130 diverse US communities. METHODS: Key informant interviews (N = 1420) and document abstraction were used to identify and characterize CPPs to promote physical activity and healthy nutrition occurring during a 10-year retrospective study period. Data were collected in 2013-2015 and analysed in 2016. RESULTS: Across all 130 communities, 9681 distinct CPPs were reported as occurring by key informants. Of these, 5574 (58%) focused on increasing physical activity, 2596 (27%) on improving nutrition and 1511 (16%) on both behaviours. The mean number of CPPs per community was 74.0, with a range of 25 to 295 across all communities. Most CPPs occurred more than once (63%) and on average lasted 6.1 years. The greatest number of reported CPPs occurred in school settings (44%). CONCLUSIONS: Communities showed a wide range of investment in the amount of CPPs occurring in settings that affect opportunities for children to engage in physical activity and healthy nutrition. The pattern of implementation of CPPs showed variation over time, with an increase in more recent years. This observational study provides new and valuable information about what US communities are doing to prevent childhood obesity.

Associations between community programmes and policies and children's physical activity: the Healthy Communities Study.

BACKGROUND: Community initiatives to promote physical activity in children are common, but evidence supporting their effectiveness is limited. OBJECTIVES: The objective of this study is to examine the relationships between community programmes and policies and children's physical activity in a large and diverse sample of US communities. METHODS: Programmes and policies to promote children's physical activity were assessed in 130 communities by key informant interviews, and physical activity behaviours were measured by self-report and parental report in samples of children in each community (total n = 5138). Associations between composite indices of community programmes and policies and indicators of total and moderate-to-vigorous physical activity were examined without and with adjustment for demographic factors. RESULTS: An index reflecting the 6-year history of the number of behaviour change strategies used in community programmes and policies was positively associated with children's moderate-to-vigorous physical activity. This association was attenuated with adjustment for demographic factors. Effect modification analyses found that the association was positive among non-Hispanic children but was negative for Hispanic children. CONCLUSIONS: Community initiatives to promote physical activity in children were positively associated with children's physical activity in non-Hispanic children. Such initiatives were negatively associated with physical activity in Hispanic children, suggesting that future research should consider unique cultural factors when designing community initiatives to promote activity in this population sub-group.

The longitudinal relationship between community programmes and policies to prevent childhood obesity and BMI in children: the Healthy Communities Study.

BACKGROUND: Although a national epidemic of childhood obesity is apparent, how community-based programmes and policies (CPPs) affect this outcome is not well understood. OBJECTIVES: This study examined the longitudinal relationship between the intensity of CPPs in 130 communities over 10 years and body mass index (BMI) of resident children. We also examined whether these relationships differ by key family or community characteristics. METHODS: Five thousand one hundred thirty-eight children in grades K-8 were recruited through 436 schools located within 130 diverse US communities. Measures of height, weight, nutrition, physical activity and behavioural and demographic family characteristics were obtained during in-home visits. A subsample of families consented to medical record review; these weight and height measures were used to calculate BMI over time for 3227 children. A total of 9681 CPPs were reported during structured interviews of 1421 community key informants, and used to calculate a time series of CPP intensity scores within each community over the previous decade. Linear mixed effect models were used to assess longitudinal relationships between childhood BMI and CPP intensity. RESULTS: An average BMI difference of 1.4 kg/m2 (p-value < 0.01) was observed between communities with the highest and lowest observed CPP intensity scores, after adjusting for community and child level covariates. BMI/CPP relationships differed significantly by child grade, race/ethnicity, family income and parental education; as well as community-level race/ethnicity. CONCLUSIONS: These results indicate that, over time, more intense CPP interventions are related to lower childhood BMI, and that there are disparities in this association by sociodemographic characteristics of families and communities.

Predicting lower-extremity injuries among habitual runners.

This prospective study of 583 habitual runners used baseline information to examine the relationship of several suspected risk factors to the occurrence of running-related injuries of the lower extremities that were severe enough to affect running habits, cause a visit to a health professional, or require use of medication. During the 12-month follow-up period, 252 men (52%) and 48 women (49%) reported at least one such injury. The multiple logistic regression results identified that running 64.0 km (40 miles) or more per week was the most important predictor of injury for men during the follow-up period (odds ratio = 2.9). Risk also was associated with having had a previous injury in the past year (odds ratio = 2.7) and with having been a runner for less than 3 years (odds ratio = 2.2). These results suggest that the incidence of lower-extremity injuries is high for habitual runners, and that for those new to running or those who have been previously injured, reducing weekly distance is a reasonable preventive behavior.

Effect of exercise duration on plasma endothelin-1 concentration.

AIM: Endothelin-1 (ET-1) is a potent vasoconstricting peptide released mostly from vascular endothelial cells. Isolated exercise sessions of relatively long duration (=or>30 min) have produced increases in plasma ET-1 concentration while shorter exercise sessions usually have not. The purpose of the present study was to verify an effect of exercise duration at a steady work rate on plasma ET-1 concentration. METHODS: Eleven endurance-trained males (age 27+/-6 years; maximal oxygen consumption--VO2max--56+/-7 mLxkg-1xmin-1, body fat 11+/-5%; mean+/-SD) exercised on a treadmill at 70% VO2max on 2 occasions separated by at least 2 weeks. During a short-duration session, subjects expended approximately 3,360 kJ (60+/-2 min). During a long-duration session, subjects expended approximately 6,300 kJ (112+/-4 min). Six of the subjects performed the 3,360 kJ session before the 6,300 kJ session while the other 5 subjects performed the 6,300 kJ session first. RESULTS: The short-duration session did not cause plasma ET-1 concentration to change immediately after exercise (0.23+/-0.01 pmolxL-1 before exercise, 0.22+/-0.02 pmolxL-1 after exercise, mean+/-SE). However, 10 of 11 subjects had increased ET-1 after the long-duration session (0.28+/-0.02 pmolxL-1 before exercise, 0.32+/-0.02 pmolxL-1 after exercise, P=0.0004). A treatment-by-time effect was present (P=0.003). CONCLUSION: These results demonstrate an effect of exercise duration on plasma ET-1 concentration. Exercise duration is, therefore, an essential consideration when investigating exercise's effect on ET-1.

Moderate-intensity physical activity and fasting insulin levels in women: the Cross-Cultural Activity Participation Study.

OBJECTIVE: The purpose of this study was to determine the association between moderate-intensity physical activity (PA) and fasting insulin levels among African-American (n = 47), Native American (n = 46), and Caucasian women (n = 49), aged 40-83 years, enrolled in the Cross-Cultural Activity Participation Study. Associations by race/ethnicity, levels of central obesity, and cardiorespiratory fitness were also examined. RESEARCH DESIGN AND METHODS: Physical activity scores were obtained from detailed PA records that included all PA performed during two consecutive 4-day periods scheduled 1 month apart. Using MET intensity (the associated metabolic rate for a specific activity divided by a standard resting metabolic rate), PA was expressed as MET-min (the product of the minutes for each activity times the MET intensity level) per day of energy expended in moderate (3-6 METs) and moderate/vigorous (> or = 3 METs) PA. Fasting insulin levels were determined by radioimmunoassay. Data were analyzed by multiple linear regression analysis. RESULTS: After adjusting for race/ethnicity, age, educational attainment, and site, an increase of 30 min of moderate-intensity PA was associated with a 6.6% lower fasting insulin level (P < 0.05). The association was similar among races/ethnicities, centrally lean and centrally obese women, and women with low and high cardiorespiratory fitness levels. CONCLUSIONS: These findings lend support to the 1995 Centers for Disease Control and Prevention and American College of Sports Medicine recommendations for an accumulation of 30 min/day in moderate-intensity PA. They also contribute to the growing literature suggesting that moderate amounts of PA have a significant role in reducing the burden of hyperinsulinemia and diabetes among ethnic populations at highest risk for these conditions.

Carbohydrate-electrolyte drinks: effects on endurance cycling in the heat.

On three occasions cyclists completed, as fast as possible, two exercise tasks (T1 and T2) separated by 30 min rest. T1 and T2 were equivalent to the work performed during 2 h cycling at 75% VO2max and 30 min at 75% VO2max, respectively. Every 20 min subjects drank 275 mL of a 6% (MC) or 2.5% (LC) carbohydrate-electrolyte beverage or a water placebo (P). The initial drink during both T1 and the rest period contained 20 g D2O as a marker for entry of ingested fluid into blood. No differences in drink effects were found for heart rate, sweat rate, change in plasma volume, rectal temperature, or D2O accumulation in blood. Blood glucose and respiratory exchange ratios were higher and T2 was performed faster with MC than with P. Ingestion of MC can help maintain blood glucose and enhance performance of prolonged cycling exercise without compromising fluid replenishment.

The role of stress hormones in exercise-induced suppression of alveolar macrophage antiviral function.

We hypothesized that a previously observed exercise-induced suppression of alveolar macrophage antiviral resistance results from increases in corticosterone and/or epinephrine. Mice (CD-1) were run to fatigue on a treadmill (exercise), or placed in Plexiglas lanes above the treadmill (control). The role of corticosterone was assessed by further dividing mice into groups receiving one of the following treatments; sham surgery, adrenalectomy, or adrenalectomy plus corticosterone replacement. Macrophage antiviral function was suppressed in the exercised mice compared to the control mice. However, macrophage antiviral function was not suppressed in the exercised mice that underwent adrenalectomy or adrenalectomy plus corticosterone replacement. We tested whether another adrenal factor (epinephrine) may be involved by dividing mice into exercise and control groups treated with either saline or propranolol. Macrophage antiviral function was again suppressed in the saline-treated exercised mice compared to saline-treated control mice, but no differences were found between the exercised mice receiving propranolol, control mice receiving propranolol, or saline-treated control mice. Isoproterenol, when added to alveolar macrophages in culture, also suppressed antiviral resistance. These findings suggest that decreased macrophage antiviral function following exercise may be due to increased release of adrenal catecholamines.

The use of proportional hazards regression in investigating dropout rates in a longitudinal study.

One difficulty with the interpretation of data from longitudinal studies is the bias associated with those who do not complete the study. In a 12-month study on the occurrence of musculoskeletal problems in 966 runners (583 of whom completed the study), a proportional hazards model with time-dependent covariates was used to assess factors associated with dropout at the various stages of the study. This approach allowed examination of baseline factors as well as the effect of change in mileage, the occurrence of a musculoskeletal problem, or the occurrence of another health problem on the rate of dropout. Those most likely to drop out of the study were younger and heavier at baseline and, prior to drop out, were less likely to experience general health problems and more likely to show a 40% decline in weekly running mileage in the month before dropout. Examination of factors associated with dropout is important since factors influencing dropout may also affect the study outcome for the risk factor analysis (a musculoskeletal problem severe enough to be seen by a physician). The results of the dropout analysis can be used to guide in the choice of analytic methods and to aid in the interpretation of the risk factor analyses.

Exaggerated blood pressure response to dynamic exercise and risk of future hypertension.

This study examined the association between an exaggerated blood pressure response to treadmill exercise and the risk of developing hypertension. Subjects were healthy normotensive men (n = 5386) who had a baseline graded maximal exercise test between 1971 and 1982, and completed a mailed follow-up questionnaire. At follow-up in 1986, cases (n = 151) reported physician diagnosed hypertension and controls (n = 201) reported normotension. Those who had developed hypertension at follow-up were more likely to have had an exaggerated blood pressure response to exercise (OR = 2.4, 1.4-4.3). In multiple logistic regression analysis an exaggerated response was significantly associated (OR = 3.0, 1.5-6.1) with future hypertension after controlling for sitting systolic and diastolic blood pressure, weight change from age 21 to follow-up, entry age, family history of hypertension, body mass index, treadmill time, alcohol consumption, and years of follow-up. These results suggest that an exaggerated blood pressure response to exercise is independently associated with increased risk of future hypertension, and therefore, may be an important factor in determining hypertension risk.

Factors affecting fibrinolytic potential: cardiovascular fitness, body composition, and lipoprotein(a).

The purpose of the study was to determine the factors that affect basal (resting) and poststressor fibrinolytic activity or potential. Variables of interest included cardiovascular fitness (maximal oxygen consumption [Vo2max]), body fat, body mass index (BMI), and lipids/lipoproteins, including lipoprotein(a) [Lp(a)]. Blood was collected from 46 middle-aged men before and after a maximal exercise test. Pearson and Spearman correlation coefficients were calculated to determine associations between the variables of interest and tissue plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) activities in the basal state and after stimulation with maximal exercise. Multiple regression analyses were also conducted to determine independent predictors of the fibrinolytic variables. Maximal exercise produced significant increases in t-PA activity and decreases in PAI-1 activity. Postexercise t-PA activity was inversely related to basal PAI-1 activity (r = -.34). Vo2max was positively correlated with t-PA activity (basal, r = .39; postexercise, r = .67) and inversely related to PAI-1 activity (basal, r = -.41; postexercise, r = -.42). Body fat was correlated with postexercise t-PA activity (r = -.60) and both basal and postexercise PAI-1 activity (r = .42), but the correlation with basal t-PA activity was not significant (P = .058). Postexercise t-PA activity was positively correlated (r = .37) with high-density lipoprotein cholesterol (HDL-C) and negatively correlated (r = -.42) with low-density lipoprotein cholesterol (LDL-C). Basal PAI-1 activity was negatively correlated with HDL-C (r = -.37), Lp(a) was not correlated with any fibrinolytic variable or fitness. Multiple regression analyses showed that Vo2max was an independent predictor of both basal and postexercise t-PA activity (R2 = .16 and .34, respectively). Triglyceride (TG) levels and Vo2max were significant independent predictors of PAI-1 activity (R2 = .31). In conclusion, cardiovascular fitness was a strong independent predictor of fibrinolytic potential. In addition, poststressor measures of fibrinolytic potential may provide more information about the fibrinolytic system than basal values.

Sports participation and health-related behaviors among US youth.

OBJECTIVE: To examine the relationship between sports participation and health-related behaviors among high school students. DESIGN: Cross-sectional design using data from the 1997 Centers for Disease Control and Prevention Youth Risk Behavior Survey. PARTICIPANTS: A nationally representative sample of 14,221 US high school students. MAIN OUTCOME MEASURES: Prevalence of sports participation among males and females from 3 ethnic groups and its associations with other health behaviors, including diet, tobacco use, alcohol and illegal drug use, sexual activity, violence, and weight loss practices. RESULTS: Approximately 70% of male students and 53% of female students reported participating on 1 or more sports teams in school and/or nonschool settings; rates varied substantially by age, sex, and ethnicity. Male sports participants were more likely than male nonparticipants to report fruit and vegetable consumption on the previous day and less likely to report cigarette smoking, cocaine and other illegal drug use, and trying to lose weight. Compared with female nonparticipants, female sports participants were more likely to report consumption of vegetables on the previous day and less likely to report having sexual intercourse in the past 3 months. Among white males and females, several other beneficial health behaviors were associated with sports participation. A few associations with negative health behaviors were observed in African American and Hispanic subgroups. CONCLUSION: Sports participation is highly prevalent among US high school students, and is associated with numerous positive health behaviors and few negative health behaviors.

Systolic blood pressure response to exercise in black and white preadolescent and early adolescent boys.

OBJECTIVE: To investigate differences in the response of systolic blood pressure (SBP) to exercise in black and white boys while controlling for the possible confounding effects of relative body weight, body surface area, physical work capacity index, preexercise SBP, and average power output. DESIGN: Comparative and correlational. PARTICIPANTS: Eighty-seven black and 52 white boys between the ages of 5 and 16 years. Participants were recruited from day camps, community centers, and summer recreation programs in and near Augusta, Ga. INTERVENTIONS: None. MEASUREMENTS/MAIN OUTCOMES: The slope of the SBP response to exercise was not significantly different between groups. Analysis of covariance revealed race, age, relative body weight, body surface area, preexercise SBP, and average power output to be significant univariate predictors of SBP at each power output. With multiple regression analyses, the effect of race was removed, and only preexercise SBP and average power output were found to be significant predictors of exercise SBP. CONCLUSION: There were no significant differences between black and white boys in the SBP response to exercise after controlling for the effects of preexercise SBP and average power output.