Low level of physical activity and sedentary behaviour in elderly: a systematic review of the parameters

– Sedentary behavior tends to increase with aging and several studies have focused on analyzing the components which may influence this pattern of behavior. Indirect and direct measuring have been used to determine the relative parameters to this phenomenon while there is no consensus about which parameter should be adopted to conceptualize sedentary behavior, making difficulty to establish comparisons among the studies on this population. The aim of this study was to systematically review the characterization of low level of physical activity and sedentary behavior in studies with older people. Electronic search on Scielo, LILACS, MEDLINE, PubMed, and ISI Web of Knowledge was carried out. Selection of the studies included: original manuscripts, with elderly analyzing sedentary behavior or low-level pf physical activity through direct and indirect measure. Search initially screened 190 manuscripts yielding 10 relevant studies. Questionnaire, self-reported questionnaire and accelerometer were the instruments used in the studies. Sedentary behavior or low level of physical activity was characterized by analyzing sitting time, physical activity on leisure time, counts per minute, engaging in moderate or vigorous intensity during day or performing physical activities with MET <1.5. There is no standardization of the parameters adopted to characterize sedentary behavior on studies, generating divergent results and making it difficult to establish comparisons.


INTRODUCTION
The population worldwide is becoming old. The aging process is known for being accompanied of some conditions or illnesses, which are not necessarily results of aging; they are mostly triggered by modifiable risk factors related to lifestyle. The individual lifestyle's impact in the health outcomes has been investigated in the past years 1 . The decisions about being sedentary or physically active, eating healthy and sleeping well through the years cause an effect in some characteristics such as the body composition, blood pressure, cognitive function and glycemic levels that, if the choices were not appropriate, might lead to poor health consequences 2 .
Practicing regular physical activity influences positively the health 3,4 . In general, it has been recommended to perform at least 150 minutes of moderate to vigorous physical activity per week to increase physical activity levels and reduce sedentary behavior 5 . It may reduce body fat, blood pressure, glycemic levels and increase muscle and bone mass, preserve functional capacity and memory, ameliorate cognitive function and many others benefits 6,7 . Physical inactivity is associated with the development of many chronic diseases such as hypertension, diabetes, obesity and cancer 8,9 . Furthermore, sedentary behavior is generally characterized by any activity that requires an energetic demand between 1.0 and 1.5 METs in a sitting or reclined position, using the computer or watching television, for example, and it is a risk factor independent of physical activity practice [10][11][12] , and it is also associated with others unhealthy behaviors and negative health outcomes [13][14][15][16] .
Although physical activity has been widely explored in studies aiming to promote a more active lifestyle 5,[17][18][19] , sedentary behavior and low levels of physical activity have emerged in the last years as an important key factor to be considered when discussing about threats against physically active lifestyle on public health scenario 20 . Indeed,therehas been an increase in exposed time to sedentary behavior in the last decade 21 and prevalence of people who have sedentary behaviors is high between aged people 22 , this may be demonstrated by a report from CDC data in 2005 showing that only 37.7% of United States population have sedentary behavior or low levels of physical activity 23 . In addition, conversely to physical activity, sedentary behavior is associated to several poor health indexes in elderly people (high blood pressure, obesity and high levels of blood glucose) 24,25 . However, differently from physical activity, studies investigating sedentary behavior or low levels of physical activity in this population have adopted several and different parameters to determine their concepts, including direct measures from electronic devices (accelerometer and pedometers) [26][27][28][29] and indirect from self-report questionnaires and diaries [30][31][32] , which in turnvarybetween reporting sitting activities more than four hours a day more than five days a weekand daily activities equivalent to <1.5 Metabolic Equivalent of Task (MET) 33 . Thus, this lack of standardization between sedentary behavior concepts adopted in the several studies makes hard to establish parameters to compare the reports about this issue.
Hence, considering there is an increase in time exposed to sedentary behavior in the last decade and the necessity to standardize the characterization of it, that could help to establish more suitable parameters to oppose this phenomenon, the aim of the study was to systematically review and identify the characterization of sedentary behavior and/or low levels of physical activity in original studies with older people.

METHOD
This study is characterized as a systematic review, previously registered in PROSPERO under the protocol CRD42016038647. Information about sedentary behavior and/or low level of physical activity in elderly were analyzed, previous research was made at COCHRANE database and PROSPERO library to a better definition of the objectives and methods applied and to avoid replicate any finished or ongoing study.

Search strategy
Original studies -published between January 2006 and July 2018 in English and Portuguese -were examined about interventions with sedentary behavior or low levels of physical activity definitions in elderly. The studies search was realized in these electronic databases: Scielo, LILACS, MED-LINE, PubMed, and ISI Web of Knowledge. The following entry terms were utilized in Portuguese and English respectively: atividade motoramotor activity, exercício físico -physical exercise, atividade física -physical activity, idosos -elderly, idosos sedentários -sedentary elderly, comportamento -behavior, comportamento sedentário -sedentary behavior. The Boolean operators "and" and "or" were used to combine the entry terms in the article search. No filters were applied to the search and characteristics differing from the criteria adopted were excluded after the search was completed.

Studies selection
It was made accordingly the ensuing steps: i) search by "titles" using the entry terms and Boolean operators, applying filters to year of publication, original studies and population age; ii) selecting those presenting "sedentary behavior" or "physical activity" or "exercise" in the title; iii) reading the abstracts of those with samples that included elderly above 60 years; vi) reading full text of the articles selected from the abstracts. Two independent reviewers realized each step and a third reviewer was consulted in case of dispute.
The inclusion criteria were: published original articles from January 2006 to July 2018; studies that included participants above 60 years that presented defined criteria for low level of physical activity and/or sedentary behavior. Case reports and opinion articles were excluded. Furthermore, studies presenting individuals inactive due to serious injuries or in rehabilitation process or diagnosed with specific diseases such as mental disorders were also excluded.
The following items were obtained from the selected articles: i) title; ii) population; iii) gender iv) objectives; v) study design vi) outcomes vii) sedentary behavior description viii) physical activity description ix) results x) instruments. The description about quantitative survey from the Database to the selection of the studies included in this review is present in Figure 1.

Data management
The results obtained from the research studies were imported into Excel data management software. After analysis, the third reviewer manually removed the duplicates.

Data extraction
Two reviewers independently extracted data from the selected studies from the Excel data management foundation using a data extraction form. The form was developed considering the characteristics of studies to ensure consistency of this process. In case of disagreement, a third reviewer was consulted. The data extraction used the following categories: Author, Journal, Sample (amount of people studied), Study Design (Observational, Longitudinal, Clinical Trial, Cohort, Prospective Cohort), Instruments (direct or indirect measures), Purpose, Definition of Sedentary Behavior and/or Low-Level Physical Activity, Results and Conclusions.
The primary outcomes were parameters related to low levels of physical activity and sedentary behavior. Sedentary behavior outcomes included objectively measured sedentary behavior or sitting time using accelerometers or pedometers. Self-report sedentary behavior outcomes included time spent watching TV, computer usage, total screen time (TV, computer and phone/iPad use combined) or sitting (travel, relaxing and workplace). Selfreport outcomes of low levels of physical activity included reporting how often they took part of several ranges of physical activity intensities and objectively measuring low levels of physical activity including calculating the intensities of the physical activity bouts.

RESULTS
The present study shows that there is a high prevalence of sedentary behavior in the elderly population and a predisposition for this behavior to increase with age. The practice of physical activity occurs mostly during daytime and it is more likely to be in a light or moderate intensity. Higher levels of sedentary behavior are associated with poorer physical strength 31 ; younger age and lower body mass index may be a predictor of physical behavior 32 . Some educational interventions to change behavior may be effective to this population but yet more experimental studies are needed to verify what interventions are the most effective 30 .
Box 1 presents a summary of the overall characterization of the selected studies, in general, epidemiology and public health were the main scope for the journals, the sample size was between 20 and 7735 elderly, in the different studies design from observational to experimental (clinical trial), and the questionnaire and accelerometer were the instruments more used.

Interventions
Only two studies were categorized as clinical trials 30,34 . Hershenberg et al. 34 evaluated association between the participation of 20 older adults in weekly activities and behavioral outcomes. They carried out pre and post five weeks interventions measures. In another study, Schneider et al. 30 compared selfreported behavior outcomes over 1 year between three groups of older adults receiving different behavior education treatments. The assessments were performed in three months intervals. Moreover, one study met the inclusion criteria, however it did not describe the parameters adopted to analyze sedentary behavior or low level of physical activity 35 . Furthermore, other studies had different methodological designs that are described in Box 1.
Box 2 summarizes the descriptions about the studies definition of sedentarybehavior or physical activity, to teach change behavior, different strategies were used in the definition variable was referred the amount, duration, intensity and/or type for physical activity.   29 Participants were asked how often they took part in vigorous-intensity, moderate-intensity and lowintensity PA, using prompt cards. At each time point, PA was classified as: inactive; only light activity at least once a week (but no moderate or vigorous); moderate activity at least once a week (but no vigorous), and vigorous activity at least once a week Sartiniet al. 33 The measure used to classify behavior was counts per minute. <100 CPM for SB (<1.5 Metabolic Equivalent of Task, MET). Another measure of SB was calculated: number of sedentary bouts of at least 1 h (a period of 60 or more consecutive minutes where the accelerometer registers <100 CPM) Heseltine et al. 32 SB was defined as sitting activities for more than four hours in more than five days a week Hershenberget al. 34 There were no description of SB or PA Loginov et al. 37 PA: low-intensity, moderate-intensity, high-intensity; SB how average daily time spent in min/week Aro et al. 38 PA were grouped into three intensity categories as defined (low, medium and high) and regular exercise is engagement in exercise for at least 150 min per week; no description of SB. Information about featuring the purpose of the studies, to test hypothesis about sedentary behavior or physical activity and variables of the health conditions, frailty or fitness, presented the different results and conclusion were summarized in the Box 3. To analyze the prevalence of behavior, change stages and the main barriers or reasons that prevent or difficult the regular PA practice 64,6% of the elderly in the study did not practice PA regularly The majority of the elderly is on pre-contemplation stage and the sex does not influence on the perceived barriers Schneider et al. 30 To teach older adults to recognize and modify their thoughts, or interpretation, about exercise To explore the SB in elderly participating in an intervention test with exercise and to investigate which health, demographic and social factors are associated with SB The probability of being categorized as sedentary augmented with an abnormal BMI. Participants reporting better physical health had lower odds ratio of being sedentary In general, older participants will respond positively to join in an exercise group Hershenberget al. 34 To investigate the participation in game weekly activities in the treatment outcomes Behavioral activation was associated to a reduction in depressive symptoms. Participant's total number of reported activities was not associated with their improvements in symptoms Independent of the specific type or total mount, activation activities may be associated with improvements of symptomatology Loginov et al. 37 To establish gender-specific characteristics of PA and sedentary behavior in elderly Detected that more energy is spent on the housework and PA in the country (moderate-intensity PA for women and high-intensity one for men) Showed no statistically significant gender-specific differences in general PA. SB is more popular among men rather than women Aro et al. 38 To explore socio-demographic and clinical factors that are associated with regular exercise Participant's knowledge of the benefits of regular physical activities, opportunities to socialize, encouragement by health care workers and availability of exercise facilities and trainers promote regular physical exercise.

DISCUSSION
The purpose of this systematic review was to summarize the literature discussing sedentary behavior and low levels of physical activity in elderly. Several studies were considered, which presented various definitions for sedentary behavior and/or low levels of physical activity. Different methods of characterizing sedentary behavior included reporting sitting activities more than four hours a day more than five days a week 32 , classifying as inactive accordingly prompt cards, counting per minute (<100 CPM) for sedentary behavior equivalent to <1.5 Metabolic Equivalent of Task (MET) 33 , number of sedentary bouts equivalent to <100 CPM counted through accelerometer and the number of daily minutes with a minutely accelerometer count of <200 counts/minute 27 . The characterization of low levels of physical activity included reporting how often they took part of a vigorous-intensity physical activity, moderate intensity and low intensity through prompt cards with different pictures of the activities to help 31 , classifying physical activity at light activity (1.5-3.0 MET) and moderate to vigorous physical activity (≥3 MET) 33 , calculating moderate to vigorous physical activity bouts of at least 10 minutes 33 . Moreover, counting hours of activity per week was another method used to try to determine low levels of physical activity 30 . As reported, the methods included self-reporting and direct measures of physical activity through accelerometer. Based on this assessment, the studies provide inconsistent evidences of how sedentary behavior and low levels of physical activity in elderly are being characterized. Furthermore, there is no standardization of the sedentary behavior and level of physical activity assessment methods. Thus, to compare these results and variables becomes a difficult and maybe inaccurate task.
In this sense, this lack of standardization resulted in several outcomes related to sedentary behavior and level of physical activity. For example, Smith et al. 29 found, after investigate physical activity during 10 years, that age was associated with a lower likelihood of being physically active and physical activity levels decreased over time 29 whereas another study verified that aging was associated to lower levels of moderate to vigorous physical activity 27 . However, while the former study used prompt cards to help individuals to self-report physical activity levels, second classified subjects in sedentary or having low level of physical activity using an accelerometer. Additionally, Heseltine et al. 32 adopted a questionnaire to classify sedentary behavior thorough (defined as sitting activities for more than four hours in more than five days a week) and verified that sedentary behavior was not associated to age 32 . Similarly, a previous study, using an accelerometer, it was found that age was not a predictor for sedentary behavior 26 .
The conflicting results may be due, in part, to the different nature of the various instruments adopted to classify the sedentary behavior. Indeed, a previous study compared sedentary behavior indexes assessed through questionnaire and accelerometer 39 . It reported a small correlation with substantial variability between the methods and a low agreement between self-reported sitting category and objective sedentary time. These findings reinforce our conclusions indicating that different methods to assess sedentary behavior may promote divergent results.
Thereby, the different parameters adopted to classify sedentary behavior and low levels of physical activity may partially explain the contradictory results presented from the several studies in this population, which in turn, does not allow making comparisons between the findings. Additionally, the distinct instruments used in the investigations are another point to be considered. Direct measurement instruments (accelerometer and pedometer) used in some of the studies [26][27][28][29] give a more reliable measure about sedentary behavior patterns than indirect measurement instruments (questionnaire and self-reports) although the second are more viable to apply in larger studies [30][31][32] .
To the best of our knowledge, there were no studies focusing on reviewing sedentary behavior or low levels of physical activity in older people. A preliminary study reviewed the most effective behavior changing techniques to reduce sedentary behavior or increase level of physical activity in middle-aged adults 40 . However, as reported, the review was carried out with adults younger than 60 years old, which in turn did not allow us to make any comparisons between results.
Additionally, inactive people must be a key target since they present a higher risk of presenting negative health outcomes. As reported previously, inactive elderly has not been targeted of studies focusing on their sedentary behavior of low levels of physical activity and the several studies that analyzed sedentary behavior or levels of physical activity adopted different methods to determine these outcomes and designs. Moreover, the objectives differed between each study.
Finally, this review was the first to evaluate sedentary behavior and levels of physical activity in older people through originals studies, strengthening the need of more works targeting the development and evaluation of the interventions to increase levels of physical activity, reduce sedentary behavior and standardize the parameters of sedentary behavior assessment.

CONCLUSION
The results of the current study showed that there is no standardized method to determine sedentary behavior and/or low level of physical activity between the original studies discussed about these two variables and given the same outcomes, different instruments may elicit divergent results. Additionally, it is possible to emphasize the importance of the regular practice of moderate physical activities and the reduction of the sedentary behavior to improve physical functions and promote the health of the elderly. Thus, more studies are necessary aiming to standardize the methods to allow making comparisons about outcomes related to sedentary behavior and low level of physical activity in elderly.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. This study was funded by the authors.

Ethical approval
The article was written in accordance with standards set by the Declaration of Helsinki.