- Email: [email protected]
Physical activity among adults with intellectual disabilities living in community settings
Available online at www.sciencedirect.com
Preventive Medicine 47 (2008) 101 – 106 www.elsevier.com/locate/ypmed
Physical activity among adults with intellectual disabilities living in community settings Jana J. Peterson a,⁎, Kathleen F. Janz b , John B. Lowe c a
Child Development and Rehabilitation Center, Oregon Health and Science University, Portland, OR, USA b Health and Sport Studies, The University of Iowa, Iowa City, IA, USA c School of Health and Sport Studies, University of the Sunshine Coast, Maroochydore, QLD, Australia Available online 26 January 2008
Abstract Objective. The aim was to objectively monitor steps accrued by a sample of adults with intellectual disabilities and to describe physical activity patterns by monitoring steps taken across weekdays, weekends, and hours of the day using time-stamped technology. Method. This study used pedometers with time-stamped recording capabilities to measure physical activity behavior of 131 adults with mild to moderate levels of intellectual disabilities living in community-based supported living group settings in Iowa. Participants wore a pedometer for seven consecutive days. All data was collected in 2006 and analyzed in 2007. Results. The participants accrued 6508 ± 3296 steps/day. Controlling for age, participants with mild intellectual disability were more active than participants with moderate intellectual disability (F = 7.03, p b .01). A total of 14.1% accumulated 10,000 steps/day. Participants were more active on weekdays than on weekends (Z = −7.36, p b .01), and least active during the evening period compared to the morning and afternoon hours (Fr = 103.3, p b .01). Conclusion. Physical activity (steps/day) achieved by the majority of this population is insufficient for health benefits, particularly among individuals with moderate intellectual disability. Evenings and weekends are especially inactive time periods. © 2008 Elsevier Inc. All rights reserved. Keywords: Adult; Mentally disabled persons; Exercise; Health behavior; Physical fitness
Introduction Intellectual disability (ID) is a state of functioning beginning in childhood, characterized by limitation in both intelligence and adaptive skills (Schalock et al., 2007; Luckasson et al., 2002). People with ID are a very heterogeneous group, and ID is usually not associated with a known biological etiology. This group experiences high rates of chronic diseases associated with insufficient physical activity (PA) (Draheim, 2006; Janicki et al., 1999; Sutherland et al., 2002), contributing to increased premature and preventable morbidity (Beange et al., 1995; LantmandeValk et al., 1997). Interventions to promote PA are indicated, and information on PA patterns and trends is needed to inform intervention development. ⁎ Corresponding author. Oregon Health and Science University — CDRC, PO Box 574, Portland, OR 97207-0574, USA. Fax: +1 503 494 6868. E-mail address: [email protected] (J.J. Peterson). 0091-7435/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2008.01.007
Researchers are beginning to characterize the PA among adults with ID, including activity levels, patterns, and trends among population subgroups. A recent literature review of PA among adults with ID states that existing evidence indicates only a small proportion of adults with ID (17.5% to 33%) meet PA guidelines (Temple et al., 2006). Studies suggest no PA gender differences among this group, more activity on weekdays than on weekends (Stanish, 2004; Stanish and Draheim, 2005), and a large amount of inter-individual variation in activity volume (Temple and Walkley, 2003). A study that used Health Survey for England data to examine PA trends among adults with ID in England observed negative relationships between PA and age, ID level, and attendance at sheltered day centers (Emerson, 2005). Relationships between age, ID level, employment variables, and objectively monitored PA for this population have not been examined. In general, continued work is needed to establish the coherence and generalizability of previous findings.
102
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Pedometers are a practical and accurate method for gathering PA data (Bassett et al., 1996; Le Masurier and Tudor-Locke, 2003). Their validity has been demonstrated among individuals with ID (Stanish, 2004). To relate pedometer-determined steps to health outcomes, Tudor-Locke and Bassett (2004) have developed indices to classify PA by step counts. According to the indices, b 5000 steps/day indicates a sedentary lifestyle; 5000−7499 steps/day indicates typical daily activity (excluding volitional sports/exercise); 7500−9999 indicates a somewhat active lifestyle; and ≥ 10,000 steps/day indicates an active lifestyle. The active lifestyle category roughly approximates usual daily activity plus 30 min of moderate-intensity PA, equivalent to the public health guideline for PA necessary for optimum health benefits (U.S. Department of Health and Human Services (USDHHS), 1996). The largest study of adults with ID using pedometers was conducted by Stanish and Draheim (2005), who studied walking behavior among 103 individuals with mild−moderate ID in varied residential settings in Eastern Canada using Digiwalker pedometers. Mean daily counts ranged between 5828−9548 steps, numbers similar to general population levels. Approximately 21% met the 10,000 steps/day guideline (TudorLocke and Bassett, 2004). The goal of the present study was to build upon previous work characterizing PA of adults with ID to aid future intervention work. Specifically, using pedometers with time-stamped technology, this study objectively monitored steps accrued by a sample of adults with mild−moderate ID and described PA patterns across weekdays, weekends, and hours of the day. Based on the literature, we hypothesized that a majority of the sample would take b 10,000 steps/day and that individuals with mild ID and younger adults would be more active than those with moderate ID and older adults. Methods Participants Participants were 131 adults with mild−moderate ID. The sample was 51.9% female. Ages ranged from 18−60 yrs, with a sample mean of 37.2 yrs (SD 11.6). The majority of participants (69.5%) did not have a specific etiology associated with ID, but 28 participants had Down syndrome (21.4%). The sample is further described in Table 1. Participants were from 13 agencies that provide supported living services in rural and urban Iowa communities. Ambulatory individuals aged 18−60 yrs with mild−moderate ID receiving ≥10 h/week of group supported living services were eligible. All participants were non-institutionalized and lived outside the family home. Environments and activities in the supported living settings are restricted as little as possible to maximize choice according to individual abilities. Participants usually lived with other adults with ID and some had fulltime residential staff. This investigation was part of a study that included an interview, so individuals unable to meaningfully complete the interview were ineligible. Recruitment letters were sent to guardians or the individuals, according to guardianship status. After guardian consent, the individual was invited to participate. Individuals who agreed signed a simply worded assent form indicating agreement to volunteer. Of those approached, 39.0% consented (30.6% with guardians; 52.0% without guardians), for 162 eligible participants in the initial pool. Of these, 20 participants with insufficient pedometer usage and 11 participants with comorbid physical disabilities were excluded, for a total sample of 131. Data were collected in summer and fall 2006. Study procedures were approved by the University Institutional Review Board.
Demographics and descriptive data Demographics were reported by agency administration. These data, including gender, age, any etiology associated with ID, ID level, and employment status were collected for descriptive purposes, and to examine group differences. ID level was reported by 11 of 13 agencies. Moderate ID was defined as IQ of 35−54, and mild ID was defined as IQ of 55−74 (American Psychiatric Association, 2000). For individuals without a current IQ on record, the agency reported whether their classification was mild or moderate ID in their agency record. The disability classification in the agency record was based on IQ tests administered for U.S. eligibility criteria for Supplemental Security Income benefits. All participant IQ levels were dichotomized into mild and moderate ID. Employment categories included unemployed, workshop/ day program (sheltered) placement, and community employment. Participants employed simultaneously in sheltered and community settings were coded as having community employment. Given small numbers of unemployed individuals, employment was dichotomized (community or sheltered) for testing group differences.
Pedometry Omron pedometers (Model HJ-700IT, Omron Healthcare, Kyoto, Japan) were used to collect step counts. Pedometers were chosen for this study because they are practical and affordable, and research indicates walking as the most popular PA for this population (Draheim et al., 2002). While pedometers cannot produce detailed PA intensity and pattern data, they are accurate step counting devices. The Omron pedometer has an accelerometer-type internal mechanism, constructed of a horizontal beam and piezo-electric crystal. Walking motion generates a sinusoidal curve of vertical acceleration versus time, with a step counted when the curve crosses zero (Schneider et al., 2004). The pedometer stores steps for 41 days and time-stamps hourly steps using an internal clock. The Omron is accurate at varied walking speeds and during running (Hasson et al., 2004, 2005). Compared to Yamax Digiwalker SW pedometers, Omron has been shown to have lower variability in steps at self-selected speeds (Doyle et al., 2007) and slow/moderate speeds, which enhances its usefulness for populations with slower ambulation (Haller et al., 2005). To reduce non-step movement noise, Omron does not record movement of b4 s duration, so a few steps taken in isolation are not recorded. The Omron can be worn at varied attachment sites (Doyle et al., 2007; Lee et al., 2007) with equal accuracy in step counts among optimal weight and obese individuals, regardless of attachment site (Roberts et al., 2005). Accuracy at varied attachment sites benefits community-based studies, where daily placement cannot be monitored.
Table 1 Description of participants, Iowa, summer and fall 2006 (N = 131) (n, %) Sample Gender Age (yrs) 18−30 31−40 41−50 51−60 Down syndrome Intellectual disability Mild Moderate Unknown Employment Unemployed or student Workshop or day program Community employment
Males
Females
63 (48%)
68 (52%)
48 (37%) 25 (19%) 38 (29%) 20 (15%) 28 (21%)
27 (43%) 8 (13%) 19 (30%) 9 (14%) 12 (19%)
21 (31%) 17 (25%) 19 (28%) 11 (16%) 16 (24%)
73 (56%) 41 (31%) 17 (13%)
35 (56%) 18 (29%) 10 (16%)
38 (56%) 23 (34%) 7 (10%)
7 (5%) 72 (55%) 52 (40%)
2 (3%) 36 (57%) 25 (40%)
5 (7%) 36 (53%) 27 (40%)
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Pedometer protocol Pedometer displays were covered to reduce reactivity. Protocol instructions were given to participants and a caregiver. Participants were instructed to wear the pedometer, clipped to their pants waist or in either pants pocket, from waking until bedtime (except during bathing or swimming) for seven consecutive days. The pedometer was considered worn during an hour when the Omron detected any movement, triggering the “wearing” stamp. Consistent with previous ID research (Frey, 2004), all participants wearing a pedometer for ≥10 h on ≥5 days, including a weekend day, were included in the data set.
Statistical analyses Statistical analysis was performed using SPSS 14.0. Significance for statistical analyses was set at the .05 alpha level. Gender-specific percentages of descriptive variables were calculated. Down syndrome was the only etiologic disorder examined because it was the only disability besides primary ID reported for N10% of participants, and it is associated with decreased peak heart rate and respiratory capacity (Fernhall et al., 1996, 2001). Means and standard deviations (SD) of pedometry-derived steps/day, steps/weekday, and steps/weekend day were constructed for the total sample and by ID level. Percentage recording ≥10,000 steps/day was calculated. Base 10 logarithmic transformations of pedometry variables were calculated because distributions were non-normal and positively skewed. A paired t-test was conducted to test for differences in weekday and weekend day steps, and a repeated-measures analysis of variance (ANOVA) with post-hoc Bonferroni comparisons was conducted to examine differences between morning, afternoon, and evening time segments. Significant contributions of age, gender, ID level, Down syndrome, and employment were examined using independent t-tests or one-way ANOVA's of the logarithm of mean steps/day. A minority of participants with unknown ID level were excluded from any analysis that included ID level. Unemployed participants were excluded from employment analyses. To examine a possible interaction between age and ID, a 2 × 2 factorial ANOVA was conducted. Independent sample t-tests were conducted to examine ID level differences for logarithms of all pedometry variables.
Results Participants were cooperative with the protocol, with only 12% of recruited individuals not wearing pedometers sufficiently for inclusion. An inter-day intraclass correlation coefficient (ICC) was calculated to confirm adequate stability across days for daily PA estimation. The ICC = .843 (95% CI .791–.885), indicating that the inclusion criterion of ≥10 h on ≥ 5 days produced sufficiently reliable data.
103
Table 2 Mean daily step counts of participants, by level of intellectual disability, Iowa, summer and fall 2006 a, b Sample (n = 131) Steps/day⁎⁎ Steps/weekday⁎ Steps/weekend day⁎
Mild (n = 73)
Moderate (n = 41)
Mean
SD
Mean
SD
Mean
SD
6621 7194 5083
3366 3551 3950
7260 7859 5597
3653 3815 4430
5133 5747 3573
1702 2053 1592
⁎p b .05, ⁎⁎p b .01. a Significant differences are between mild and moderate levels of intellectual disability for each pedometry variable. b Analysis controls for categorical age.
Steps per day Participants wore the pedometers for a mean of 14.5 h/day (SD 2.2). Mean steps/day for the week ranged from 1703–24,369; mean steps/weekday ranged from 1796–21,744; and mean steps/ weekend day ranged from 1189–30,931. Only 20 participants (15.3%) achieved the public health guideline of 10,000 steps/day. The distribution of steps/day for the sample is displayed in Fig. 1. Differences between groups The logarithm of steps/day was analyzed by each descriptive study variable (gender, Down syndrome, ID level, dichotomous employment category, and age). Age (F = 4.81, p b .01) and ID (t = 3.71, p b .01) were both significant factors. A 2-way factorial ANOVA examining differences in the logarithm of steps/day by ID and age was conducted as follow-up. The ANOVA revealed a significant association for ID on the logarithm of mean steps/ day (F = 9.01, p b .01), but no primary age effect or interaction effect. Thus, all subsequent analyses include examination of differences by ID level, controlling for age. Steps by level of ID Mean steps per day, weekday, and weekend day for the sample and by ID level are summarized (Table 2). ANOVA tests controlling for age revealed significant differences by ID level Table 3 Mean steps taken by time of day, for all participants and by level of intellectual disability, Iowa, summer and fall 2006 a, b, c Sample (n = 131) Morning⁎ Afternoon⁎⁎ Evening
Fig. 1. Mean steps/day by sample participants, Iowa, summer and fall 2006.
Mild (n = 73)
Moderate (n = 41)
Mean
SD
Mean
SD
Mean
SD
2040 2613 1475
1267 1552 1256
2154 2833 1653
1373 1754 1426
1739 2008 1039
863 799 561
⁎p b .05, ⁎⁎p b .01. a Morning, afternoon, and evening are 5-hour blocks of time. Morning = 7:00 AM –11:59 AM hours; Afternoon = 12:00 PM – 4:59 PM hours; Evening = 5:00 PM – 9:59 PM hours. b Significant differences indicated are between mild and moderate levels of intellectual disability for time period. c Analysis controls for categorical age.
104
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Fig. 2. Patterns of mean weekday steps by hour by entire sample, and participants with mild and moderate levels of intellectual disability, Iowa, summer and fall 2006.
for all variables, including log10 steps/day (F = 10.88, p b .01), log10 steps/weekday (F = 8.81, p b .01), and log10 steps/weekend day (F = 7.00, p b .01). Day of week and time of day Paired t-tests revealed significantly greater weekday than weekend day steps for the entire sample (t = 9.98, p b .01), participants with mild ID (t = 7.35, p b .01), and those with moderate ID (t = 7.04, p b .01). Mean steps/day, weekday, and weekend day were divided into three 5-hour time blocks (Table 3). These blocks were morning (7:00AM–11:59AM), afternoon (12:00PM– 4:59PM), and evening (5:00PM–9:59PM), dividing the day into segments likely reflecting primarily work or non-work activities. ANOVA tests of the logarithm of steps per each time segment, controlling for age, revealed participants with moderate ID took fewer morning steps/day (F = 6.77, p b .05) and afternoon steps/ day (F = 9.28, p b .01). The repeated-measures ANOVA indicated significant difference between time periods for the week (F = 57.55, p b .01). Post-hoc Bonferroni comparisons showed significant differences between all pairs of time period (p b .01). The same pattern was observed for individuals of mild ID (F = 28.89, p b .01). For individuals with moderate ID, there was also an overall difference observed between time periods for the week (F = 34.18, p b .01). While this group was significantly less active in the evenings compared to other time periods, there was no difference between morning and afternoon time periods for this group. The patterns of weekday steps for the sample and each ID group are displayed graphically in Fig. 2. Steps are displayed by hour to optimize detail. Discussion The primary study aim was to objectively monitor steps of adults with mild–moderate ID. Although the ICC indicated good intra-individual reliability for steps/day, the range of 1,703–24,369 steps/day for the entire sample demonstrated
large variation in PA among participants (Fig. 1). The sample SD values were relatively large compared to the means, further indicating the dissimilarity of PA levels among individuals. Comparison to the 10,000 steps/day criterion gives a useful approximation of the U.S. PA guideline (USDHHS, 1996) recommended for health benefits (Tudor-Locke and Bassett, 2004). As observed in other studies of this population (Stanish and Draheim, 2005; Temple et al., 2006), a proportion of the population obtains a sufficient daily steps. Also important, however, is the proportion of the sample that is sedentary (TudorLocke and Bassett, 2004), with approximately 39% achieving fewer than 5000 steps/day. Only 15% of the current study sample took ≥ 10,000 steps/ day. The Stanish and Draheim (2005) study, which used the Digiwalker pedometer, found that 21% of participants reached the 10,000 steps/day mark (Stanish and Draheim, 2005). Although studies examining the Omron indicate it is a reliable and valid step counter (e.g. Hasson et al., 2004), the Omron foursecond step-recording delay feature would be expected to result in an undercount of total daily steps compared to a Digiwalker pedometer. We assume the number of participants reaching 10,000 steps/day would be somewhat higher if Digiwalker pedometers were used. While the 10,000 steps criterion is an indication of PA volume necessary for health benefits, future research should also examine the intensity and duration of PA bouts among this population. The mean of 6,508 steps/day indicates that adults with ID take a similar number of steps as general population adults (Chan et al., 2003; Sequeira et al., 1995). The mean is slightly less than some general population studies, but when the Omron four-second feature is considered, the mean would likely fall in the range often observed in the general population. Comparing steps to general population levels is an insufficient comparison when considering criterion-based interventions, since the general population mean is insufficient for health. Additionally, the subset with moderate ID had lower step counts than those with mild ID or general population levels, constituting a subgroup especially at risk. A second study aim was to describe PA patterns across weekdays, weekends, and hours of the day. Steps varied significantly, with greater activity on weekdays, and during morning and afternoon time periods. Weekends were quite inactive for both ID groups. One possible explanation is that fewer work and organized activity options are available. Weekends and evenings appear to be an appropriate focus for health promotion intervention. Individuals with mild ID are more active than individuals with moderate ID during all days/time periods except evenings. Individuals with mild disabilities were more active during the day but not the evening, suggesting that employment activities, rather than leisure pursuits, differ for the two groups. Research should examine the contribution of employment and leisure activities to PA behavior. The current investigation cannot elucidate whether physical inactivity during these times is due to personal preference or lack of supports, resources, and opportunities. Data about the periodicity of activity will allow further exploration into these issues. This study also examined step differences among population subgroups. Differences in several variables were observed by ID
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
level, a variable not examined in previous PA studies (Temple et al., 2006). Two factors that could account for these differences are residential and work environments and activities. All participants in this study received supported living services. Within this environment, perhaps individuals with higher ability have, in general, fewer restrictions, less staff supervision, and more independence to pursue leisure activities of interest. Research into the specific causes of this behavioral disparity is needed. Because individuals with severe and profound ID and individuals with co-existing physical disabilities were not included in this sample our study results likely overestimate total PA compared to the total population with ID. Consistent with other studies in this population (Stanish, 2004; Stanish and Draheim, 2005), no step differences were observed for gender. This trend conflicts with general population PA patterns, where men are more active than women (Trost et al., 2002). This may be due to the structured daily patterns of individuals with ID (Stanish and Draheim, 2005), and perhaps this population experiences fewer gender expectations than adults in the general population. This finding suggests that intervention programs targeted at this group may successfully reach both men and women. Study limitations This research had several limitations. Study participants were volunteers, with a higher response rate among individuals serving as their own guardians, and few individuals with limited communication participating due to the interview requirement. These factors suggest that participants are skewed toward a more able, and therefore possibly more active, segment of the population with ID. A second limitation is that all participants were from Iowa. Although an attempt was made to include individuals from varying sized communities, results may not be generalizable to geographical regions outside the Midwest or larger urban centers. Another limitation is the novelty of the Omron HJ-700IT. Although the technology was deemed beneficial for this study and population, the four-second delay inhibits direct comparisons to data from other pedometers. An additional limitation of using pedometers to study PA is that they only measure ambulatory activity. Conclusions In recent decades, adults with disabilities have experienced many advances in education, work parity, and human rights. Despite these developments, adults with ID experience clear health disparities. The present study examines PA among adults with ID utilizing objective monitoring. Results indicate that the majority of this population does not achieve sufficient PA for all health benefits, and individuals with moderate ID are less active than adults with mild ID. Given the risk for obesity and chronic disease in this population, research into PA promotion for is imperative. Results of this study indicate that PA promotion programs do not necessarily need to be gender-specific, but that those with greater level of impairment may especially need support to perform and maintain PA.
105
Acknowledgments This study was supported by a grant from the Special Olympics Health Professions Student Grants program (Federal project grant # U59/CCU 321826 through the U.S. Centers for Disease Control and Prevention). The pedometers were provided by Hiroshi Ogawa, Omron Health Corporation. The authors are indebted to the study participants, as well as staff at the agencies that provide the participants services, for the support of their participation. References American Psychiatric Association, 2000. Diagnostic and statistical manual of mental disorders, 4th ed. American Psychiatric Press, Inc., Washington, DC. (text revision (DSM-IV-TR)). Bassett, D.R., Ainsworth, B.E., Leggett, S.R., Mathien, C.A., Main, J.A., Hunter, D.C., Duncan, G.E., 1996. Accuracy of five electronic pedometers for measuring distance walked. Med. Sci. Sports Exerc. 28, 1071–1077. Beange, H., McElduff, A., Baker, W., 1995. Medical disorders of adults with mental retardation: a population study. Am. J. Ment. Retard. 99, 595–604. Chan, C.B., Spangler, E., Valcour, J., Tudor-Locke, C., 2003. Cross-sectional relationship of pedometer-determined ambulatory activity to indicators of health. Obes. Res. 11, 1563–1570. Doyle, J.A., Green, M.S., Corona, B.T., Simone, J., Dennison, D.A., 2007. Validation of an electronic pedometer in a field-based setting. New Orleans, LA: American College of Sports Medicine 54th Annual Meeting. Med. Sci. Sports Exerc., vol. 39, p. S186. Draheim, C.C., 2006. Cardiovascular disease prevalence and risk factors of persons with mental retardation. Ment. Retard. Dev. Disabil. Res. Rev. 12, 3–12. Draheim, C.C., Williams, D.P., McCubbin, J.A., 2002. Prevalence of physical inactivity and recommended physical activity in community-based adults with mental retardation. Ment. Retard. 40, 436–444. Emerson, E., 2005. Underweight, obesity and exercise among adults with intellectual disabilities in supported accommodation in Northern England. J. Intellect. Disabil. Res. 49, 134–143. Fernhall, B., Pitetti, K.H., Rimmer, J.H., et al., 1996. Cardiorespiratory capacity of individuals with mental retardation including Down syndrome. Med. Sci. Sports Exerc. 28, 366–371. Fernhall, B., McCubbin, J.A., Pitetti, K.H., et al., 2001. Prediction of maximal heart rate in individuals with mental retardation. Med. Sci. Sports Exerc. 33, 1655–1660. Frey, G.C., 2004. Comparison of physical activity levels between adults with and without mental retardation. J. Phys. Act. Health 1, 235–245. Haller, J., Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Validation of the Omron HJ-112 pedometer at various walking speeds. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Hasson, R.E., Pober, D.M., Freedson, P.S., 2004. Validation of new pedometer during running and walking. Indianapolis, IN, American College of Sports Medicine 51st Annual Meeting. Med. Sci. Sports Exerc., 36, p. S31. Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Evaluation of the Omron HJ112 and Yamax Digiwalker SW-701 pedometers during variable speed walking. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Janicki, M.P., Dalton, A.J., Henderson, C.M., Davidson, P.W., 1999. Mortality and morbidity among older adults with intellectual disability: health services considerations. Disabil. Rehabil. 21, 284–294. LantmandeValk, H.M.J.V., vandenAkker, M., Maaskant, M.A., et al., 1997. Prevalence and incidence of health problems in people with intellectual disability. J. Intellect. Disabil. Res. 41, 42–51. Lee, M., Zhu, W., Yang, L., Bendis, K., Hernandez, J., 2007. Position invariance of Omron-BI pedometers in older adults. New Orleans, LA, American College of Sports Medicine 54th Annual Meeting. Med. Sci. Sports Exerc., 39, p. S187.
106
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Le Masurier, G., Tudor-Locke, C., 2003. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc 35, 867–871. Luckasson, R., Borthwick-Duffy, S., Buntinx, W.H.E., et al., 2002. Mental Retardation: Definition, Classification, and Systems of Supports, 10th ed. American Association on Mental Retardation, Washington, DC. Roberts, D.E., Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Effect of pedometer position on step counts accuracy in normal weight and obese individuals. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Schalock, R.L., Luckasson, R.A., Shogren, K.A., 2007. The renaming of “mental retardation”: understanding the change to the term “intellectual disability”. Intellect. Dev. Disabil. 45, 116–124. Schneider, P.L., Crouter, S.E., Bassett, D.R., 2004. Pedometer measures of freeliving physical activity: comparison of 13 models. Med. Sci. Sports Exerc. 36, 331–335. Sequeira, M.M., Rickenbach, M., Wietlisbach, V., Tullen, B., Schutz, Y., 1995. Physical activity assessment using a pedometer and its comparison with a questionnaire in a large population survey. Am. J. Epidemiol. 142, 989–999.
Stanish, H.I., 2004. Accuracy of pedometers and walking activity in adults with mental retardation. Adapt. Phys. Act. Q. 21, 167–179. Stanish, H.I., Draheim, C.C., 2005. Walking habits of adults with mental retardation. Ment. Retard. 43, 421–427. Sutherland, G., Couch, M.A., Iacono, T., 2002. Health issues for adults with developmental disability. Res. Dev. Disabil. 23, 422–445. Temple, V.A., Walkley, J.W., 2003. Physical activity of adults with intellectual disability. J. Intellect. Dev. Disabil. 28, 342–353. Temple, V.A., Frey, G.C., Stanish, H.I., 2006. Physical activity of adults with mental retardation: review and research needs. Am. J. Health Promot. 21, 2–12. Trost, S.G., Owen, N., Bauman, A.E., Sallis, J.F., Brown, W., 2002. Correlates of adults' participation in physical activity: review and update. Med. Sci. Sports Exerc. 34, 1996–2001. Tudor-Locke, C., Bassett, D.R., 2004. How many steps/day are enough? Sports Med. 34, 1–8. U.S. Department of Health and Human Services, 1996. Physical Activity and Health: A Report of the Surgeon General. U.S. Department of Health and Human Services, Atlanta, GA.
Preventive Medicine 47 (2008) 101 – 106 www.elsevier.com/locate/ypmed
Physical activity among adults with intellectual disabilities living in community settings Jana J. Peterson a,⁎, Kathleen F. Janz b , John B. Lowe c a
Child Development and Rehabilitation Center, Oregon Health and Science University, Portland, OR, USA b Health and Sport Studies, The University of Iowa, Iowa City, IA, USA c School of Health and Sport Studies, University of the Sunshine Coast, Maroochydore, QLD, Australia Available online 26 January 2008
Abstract Objective. The aim was to objectively monitor steps accrued by a sample of adults with intellectual disabilities and to describe physical activity patterns by monitoring steps taken across weekdays, weekends, and hours of the day using time-stamped technology. Method. This study used pedometers with time-stamped recording capabilities to measure physical activity behavior of 131 adults with mild to moderate levels of intellectual disabilities living in community-based supported living group settings in Iowa. Participants wore a pedometer for seven consecutive days. All data was collected in 2006 and analyzed in 2007. Results. The participants accrued 6508 ± 3296 steps/day. Controlling for age, participants with mild intellectual disability were more active than participants with moderate intellectual disability (F = 7.03, p b .01). A total of 14.1% accumulated 10,000 steps/day. Participants were more active on weekdays than on weekends (Z = −7.36, p b .01), and least active during the evening period compared to the morning and afternoon hours (Fr = 103.3, p b .01). Conclusion. Physical activity (steps/day) achieved by the majority of this population is insufficient for health benefits, particularly among individuals with moderate intellectual disability. Evenings and weekends are especially inactive time periods. © 2008 Elsevier Inc. All rights reserved. Keywords: Adult; Mentally disabled persons; Exercise; Health behavior; Physical fitness
Introduction Intellectual disability (ID) is a state of functioning beginning in childhood, characterized by limitation in both intelligence and adaptive skills (Schalock et al., 2007; Luckasson et al., 2002). People with ID are a very heterogeneous group, and ID is usually not associated with a known biological etiology. This group experiences high rates of chronic diseases associated with insufficient physical activity (PA) (Draheim, 2006; Janicki et al., 1999; Sutherland et al., 2002), contributing to increased premature and preventable morbidity (Beange et al., 1995; LantmandeValk et al., 1997). Interventions to promote PA are indicated, and information on PA patterns and trends is needed to inform intervention development. ⁎ Corresponding author. Oregon Health and Science University — CDRC, PO Box 574, Portland, OR 97207-0574, USA. Fax: +1 503 494 6868. E-mail address: [email protected] (J.J. Peterson). 0091-7435/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2008.01.007
Researchers are beginning to characterize the PA among adults with ID, including activity levels, patterns, and trends among population subgroups. A recent literature review of PA among adults with ID states that existing evidence indicates only a small proportion of adults with ID (17.5% to 33%) meet PA guidelines (Temple et al., 2006). Studies suggest no PA gender differences among this group, more activity on weekdays than on weekends (Stanish, 2004; Stanish and Draheim, 2005), and a large amount of inter-individual variation in activity volume (Temple and Walkley, 2003). A study that used Health Survey for England data to examine PA trends among adults with ID in England observed negative relationships between PA and age, ID level, and attendance at sheltered day centers (Emerson, 2005). Relationships between age, ID level, employment variables, and objectively monitored PA for this population have not been examined. In general, continued work is needed to establish the coherence and generalizability of previous findings.
102
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Pedometers are a practical and accurate method for gathering PA data (Bassett et al., 1996; Le Masurier and Tudor-Locke, 2003). Their validity has been demonstrated among individuals with ID (Stanish, 2004). To relate pedometer-determined steps to health outcomes, Tudor-Locke and Bassett (2004) have developed indices to classify PA by step counts. According to the indices, b 5000 steps/day indicates a sedentary lifestyle; 5000−7499 steps/day indicates typical daily activity (excluding volitional sports/exercise); 7500−9999 indicates a somewhat active lifestyle; and ≥ 10,000 steps/day indicates an active lifestyle. The active lifestyle category roughly approximates usual daily activity plus 30 min of moderate-intensity PA, equivalent to the public health guideline for PA necessary for optimum health benefits (U.S. Department of Health and Human Services (USDHHS), 1996). The largest study of adults with ID using pedometers was conducted by Stanish and Draheim (2005), who studied walking behavior among 103 individuals with mild−moderate ID in varied residential settings in Eastern Canada using Digiwalker pedometers. Mean daily counts ranged between 5828−9548 steps, numbers similar to general population levels. Approximately 21% met the 10,000 steps/day guideline (TudorLocke and Bassett, 2004). The goal of the present study was to build upon previous work characterizing PA of adults with ID to aid future intervention work. Specifically, using pedometers with time-stamped technology, this study objectively monitored steps accrued by a sample of adults with mild−moderate ID and described PA patterns across weekdays, weekends, and hours of the day. Based on the literature, we hypothesized that a majority of the sample would take b 10,000 steps/day and that individuals with mild ID and younger adults would be more active than those with moderate ID and older adults. Methods Participants Participants were 131 adults with mild−moderate ID. The sample was 51.9% female. Ages ranged from 18−60 yrs, with a sample mean of 37.2 yrs (SD 11.6). The majority of participants (69.5%) did not have a specific etiology associated with ID, but 28 participants had Down syndrome (21.4%). The sample is further described in Table 1. Participants were from 13 agencies that provide supported living services in rural and urban Iowa communities. Ambulatory individuals aged 18−60 yrs with mild−moderate ID receiving ≥10 h/week of group supported living services were eligible. All participants were non-institutionalized and lived outside the family home. Environments and activities in the supported living settings are restricted as little as possible to maximize choice according to individual abilities. Participants usually lived with other adults with ID and some had fulltime residential staff. This investigation was part of a study that included an interview, so individuals unable to meaningfully complete the interview were ineligible. Recruitment letters were sent to guardians or the individuals, according to guardianship status. After guardian consent, the individual was invited to participate. Individuals who agreed signed a simply worded assent form indicating agreement to volunteer. Of those approached, 39.0% consented (30.6% with guardians; 52.0% without guardians), for 162 eligible participants in the initial pool. Of these, 20 participants with insufficient pedometer usage and 11 participants with comorbid physical disabilities were excluded, for a total sample of 131. Data were collected in summer and fall 2006. Study procedures were approved by the University Institutional Review Board.
Demographics and descriptive data Demographics were reported by agency administration. These data, including gender, age, any etiology associated with ID, ID level, and employment status were collected for descriptive purposes, and to examine group differences. ID level was reported by 11 of 13 agencies. Moderate ID was defined as IQ of 35−54, and mild ID was defined as IQ of 55−74 (American Psychiatric Association, 2000). For individuals without a current IQ on record, the agency reported whether their classification was mild or moderate ID in their agency record. The disability classification in the agency record was based on IQ tests administered for U.S. eligibility criteria for Supplemental Security Income benefits. All participant IQ levels were dichotomized into mild and moderate ID. Employment categories included unemployed, workshop/ day program (sheltered) placement, and community employment. Participants employed simultaneously in sheltered and community settings were coded as having community employment. Given small numbers of unemployed individuals, employment was dichotomized (community or sheltered) for testing group differences.
Pedometry Omron pedometers (Model HJ-700IT, Omron Healthcare, Kyoto, Japan) were used to collect step counts. Pedometers were chosen for this study because they are practical and affordable, and research indicates walking as the most popular PA for this population (Draheim et al., 2002). While pedometers cannot produce detailed PA intensity and pattern data, they are accurate step counting devices. The Omron pedometer has an accelerometer-type internal mechanism, constructed of a horizontal beam and piezo-electric crystal. Walking motion generates a sinusoidal curve of vertical acceleration versus time, with a step counted when the curve crosses zero (Schneider et al., 2004). The pedometer stores steps for 41 days and time-stamps hourly steps using an internal clock. The Omron is accurate at varied walking speeds and during running (Hasson et al., 2004, 2005). Compared to Yamax Digiwalker SW pedometers, Omron has been shown to have lower variability in steps at self-selected speeds (Doyle et al., 2007) and slow/moderate speeds, which enhances its usefulness for populations with slower ambulation (Haller et al., 2005). To reduce non-step movement noise, Omron does not record movement of b4 s duration, so a few steps taken in isolation are not recorded. The Omron can be worn at varied attachment sites (Doyle et al., 2007; Lee et al., 2007) with equal accuracy in step counts among optimal weight and obese individuals, regardless of attachment site (Roberts et al., 2005). Accuracy at varied attachment sites benefits community-based studies, where daily placement cannot be monitored.
Table 1 Description of participants, Iowa, summer and fall 2006 (N = 131) (n, %) Sample Gender Age (yrs) 18−30 31−40 41−50 51−60 Down syndrome Intellectual disability Mild Moderate Unknown Employment Unemployed or student Workshop or day program Community employment
Males
Females
63 (48%)
68 (52%)
48 (37%) 25 (19%) 38 (29%) 20 (15%) 28 (21%)
27 (43%) 8 (13%) 19 (30%) 9 (14%) 12 (19%)
21 (31%) 17 (25%) 19 (28%) 11 (16%) 16 (24%)
73 (56%) 41 (31%) 17 (13%)
35 (56%) 18 (29%) 10 (16%)
38 (56%) 23 (34%) 7 (10%)
7 (5%) 72 (55%) 52 (40%)
2 (3%) 36 (57%) 25 (40%)
5 (7%) 36 (53%) 27 (40%)
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Pedometer protocol Pedometer displays were covered to reduce reactivity. Protocol instructions were given to participants and a caregiver. Participants were instructed to wear the pedometer, clipped to their pants waist or in either pants pocket, from waking until bedtime (except during bathing or swimming) for seven consecutive days. The pedometer was considered worn during an hour when the Omron detected any movement, triggering the “wearing” stamp. Consistent with previous ID research (Frey, 2004), all participants wearing a pedometer for ≥10 h on ≥5 days, including a weekend day, were included in the data set.
Statistical analyses Statistical analysis was performed using SPSS 14.0. Significance for statistical analyses was set at the .05 alpha level. Gender-specific percentages of descriptive variables were calculated. Down syndrome was the only etiologic disorder examined because it was the only disability besides primary ID reported for N10% of participants, and it is associated with decreased peak heart rate and respiratory capacity (Fernhall et al., 1996, 2001). Means and standard deviations (SD) of pedometry-derived steps/day, steps/weekday, and steps/weekend day were constructed for the total sample and by ID level. Percentage recording ≥10,000 steps/day was calculated. Base 10 logarithmic transformations of pedometry variables were calculated because distributions were non-normal and positively skewed. A paired t-test was conducted to test for differences in weekday and weekend day steps, and a repeated-measures analysis of variance (ANOVA) with post-hoc Bonferroni comparisons was conducted to examine differences between morning, afternoon, and evening time segments. Significant contributions of age, gender, ID level, Down syndrome, and employment were examined using independent t-tests or one-way ANOVA's of the logarithm of mean steps/day. A minority of participants with unknown ID level were excluded from any analysis that included ID level. Unemployed participants were excluded from employment analyses. To examine a possible interaction between age and ID, a 2 × 2 factorial ANOVA was conducted. Independent sample t-tests were conducted to examine ID level differences for logarithms of all pedometry variables.
Results Participants were cooperative with the protocol, with only 12% of recruited individuals not wearing pedometers sufficiently for inclusion. An inter-day intraclass correlation coefficient (ICC) was calculated to confirm adequate stability across days for daily PA estimation. The ICC = .843 (95% CI .791–.885), indicating that the inclusion criterion of ≥10 h on ≥ 5 days produced sufficiently reliable data.
103
Table 2 Mean daily step counts of participants, by level of intellectual disability, Iowa, summer and fall 2006 a, b Sample (n = 131) Steps/day⁎⁎ Steps/weekday⁎ Steps/weekend day⁎
Mild (n = 73)
Moderate (n = 41)
Mean
SD
Mean
SD
Mean
SD
6621 7194 5083
3366 3551 3950
7260 7859 5597
3653 3815 4430
5133 5747 3573
1702 2053 1592
⁎p b .05, ⁎⁎p b .01. a Significant differences are between mild and moderate levels of intellectual disability for each pedometry variable. b Analysis controls for categorical age.
Steps per day Participants wore the pedometers for a mean of 14.5 h/day (SD 2.2). Mean steps/day for the week ranged from 1703–24,369; mean steps/weekday ranged from 1796–21,744; and mean steps/ weekend day ranged from 1189–30,931. Only 20 participants (15.3%) achieved the public health guideline of 10,000 steps/day. The distribution of steps/day for the sample is displayed in Fig. 1. Differences between groups The logarithm of steps/day was analyzed by each descriptive study variable (gender, Down syndrome, ID level, dichotomous employment category, and age). Age (F = 4.81, p b .01) and ID (t = 3.71, p b .01) were both significant factors. A 2-way factorial ANOVA examining differences in the logarithm of steps/day by ID and age was conducted as follow-up. The ANOVA revealed a significant association for ID on the logarithm of mean steps/ day (F = 9.01, p b .01), but no primary age effect or interaction effect. Thus, all subsequent analyses include examination of differences by ID level, controlling for age. Steps by level of ID Mean steps per day, weekday, and weekend day for the sample and by ID level are summarized (Table 2). ANOVA tests controlling for age revealed significant differences by ID level Table 3 Mean steps taken by time of day, for all participants and by level of intellectual disability, Iowa, summer and fall 2006 a, b, c Sample (n = 131) Morning⁎ Afternoon⁎⁎ Evening
Fig. 1. Mean steps/day by sample participants, Iowa, summer and fall 2006.
Mild (n = 73)
Moderate (n = 41)
Mean
SD
Mean
SD
Mean
SD
2040 2613 1475
1267 1552 1256
2154 2833 1653
1373 1754 1426
1739 2008 1039
863 799 561
⁎p b .05, ⁎⁎p b .01. a Morning, afternoon, and evening are 5-hour blocks of time. Morning = 7:00 AM –11:59 AM hours; Afternoon = 12:00 PM – 4:59 PM hours; Evening = 5:00 PM – 9:59 PM hours. b Significant differences indicated are between mild and moderate levels of intellectual disability for time period. c Analysis controls for categorical age.
104
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Fig. 2. Patterns of mean weekday steps by hour by entire sample, and participants with mild and moderate levels of intellectual disability, Iowa, summer and fall 2006.
for all variables, including log10 steps/day (F = 10.88, p b .01), log10 steps/weekday (F = 8.81, p b .01), and log10 steps/weekend day (F = 7.00, p b .01). Day of week and time of day Paired t-tests revealed significantly greater weekday than weekend day steps for the entire sample (t = 9.98, p b .01), participants with mild ID (t = 7.35, p b .01), and those with moderate ID (t = 7.04, p b .01). Mean steps/day, weekday, and weekend day were divided into three 5-hour time blocks (Table 3). These blocks were morning (7:00AM–11:59AM), afternoon (12:00PM– 4:59PM), and evening (5:00PM–9:59PM), dividing the day into segments likely reflecting primarily work or non-work activities. ANOVA tests of the logarithm of steps per each time segment, controlling for age, revealed participants with moderate ID took fewer morning steps/day (F = 6.77, p b .05) and afternoon steps/ day (F = 9.28, p b .01). The repeated-measures ANOVA indicated significant difference between time periods for the week (F = 57.55, p b .01). Post-hoc Bonferroni comparisons showed significant differences between all pairs of time period (p b .01). The same pattern was observed for individuals of mild ID (F = 28.89, p b .01). For individuals with moderate ID, there was also an overall difference observed between time periods for the week (F = 34.18, p b .01). While this group was significantly less active in the evenings compared to other time periods, there was no difference between morning and afternoon time periods for this group. The patterns of weekday steps for the sample and each ID group are displayed graphically in Fig. 2. Steps are displayed by hour to optimize detail. Discussion The primary study aim was to objectively monitor steps of adults with mild–moderate ID. Although the ICC indicated good intra-individual reliability for steps/day, the range of 1,703–24,369 steps/day for the entire sample demonstrated
large variation in PA among participants (Fig. 1). The sample SD values were relatively large compared to the means, further indicating the dissimilarity of PA levels among individuals. Comparison to the 10,000 steps/day criterion gives a useful approximation of the U.S. PA guideline (USDHHS, 1996) recommended for health benefits (Tudor-Locke and Bassett, 2004). As observed in other studies of this population (Stanish and Draheim, 2005; Temple et al., 2006), a proportion of the population obtains a sufficient daily steps. Also important, however, is the proportion of the sample that is sedentary (TudorLocke and Bassett, 2004), with approximately 39% achieving fewer than 5000 steps/day. Only 15% of the current study sample took ≥ 10,000 steps/ day. The Stanish and Draheim (2005) study, which used the Digiwalker pedometer, found that 21% of participants reached the 10,000 steps/day mark (Stanish and Draheim, 2005). Although studies examining the Omron indicate it is a reliable and valid step counter (e.g. Hasson et al., 2004), the Omron foursecond step-recording delay feature would be expected to result in an undercount of total daily steps compared to a Digiwalker pedometer. We assume the number of participants reaching 10,000 steps/day would be somewhat higher if Digiwalker pedometers were used. While the 10,000 steps criterion is an indication of PA volume necessary for health benefits, future research should also examine the intensity and duration of PA bouts among this population. The mean of 6,508 steps/day indicates that adults with ID take a similar number of steps as general population adults (Chan et al., 2003; Sequeira et al., 1995). The mean is slightly less than some general population studies, but when the Omron four-second feature is considered, the mean would likely fall in the range often observed in the general population. Comparing steps to general population levels is an insufficient comparison when considering criterion-based interventions, since the general population mean is insufficient for health. Additionally, the subset with moderate ID had lower step counts than those with mild ID or general population levels, constituting a subgroup especially at risk. A second study aim was to describe PA patterns across weekdays, weekends, and hours of the day. Steps varied significantly, with greater activity on weekdays, and during morning and afternoon time periods. Weekends were quite inactive for both ID groups. One possible explanation is that fewer work and organized activity options are available. Weekends and evenings appear to be an appropriate focus for health promotion intervention. Individuals with mild ID are more active than individuals with moderate ID during all days/time periods except evenings. Individuals with mild disabilities were more active during the day but not the evening, suggesting that employment activities, rather than leisure pursuits, differ for the two groups. Research should examine the contribution of employment and leisure activities to PA behavior. The current investigation cannot elucidate whether physical inactivity during these times is due to personal preference or lack of supports, resources, and opportunities. Data about the periodicity of activity will allow further exploration into these issues. This study also examined step differences among population subgroups. Differences in several variables were observed by ID
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
level, a variable not examined in previous PA studies (Temple et al., 2006). Two factors that could account for these differences are residential and work environments and activities. All participants in this study received supported living services. Within this environment, perhaps individuals with higher ability have, in general, fewer restrictions, less staff supervision, and more independence to pursue leisure activities of interest. Research into the specific causes of this behavioral disparity is needed. Because individuals with severe and profound ID and individuals with co-existing physical disabilities were not included in this sample our study results likely overestimate total PA compared to the total population with ID. Consistent with other studies in this population (Stanish, 2004; Stanish and Draheim, 2005), no step differences were observed for gender. This trend conflicts with general population PA patterns, where men are more active than women (Trost et al., 2002). This may be due to the structured daily patterns of individuals with ID (Stanish and Draheim, 2005), and perhaps this population experiences fewer gender expectations than adults in the general population. This finding suggests that intervention programs targeted at this group may successfully reach both men and women. Study limitations This research had several limitations. Study participants were volunteers, with a higher response rate among individuals serving as their own guardians, and few individuals with limited communication participating due to the interview requirement. These factors suggest that participants are skewed toward a more able, and therefore possibly more active, segment of the population with ID. A second limitation is that all participants were from Iowa. Although an attempt was made to include individuals from varying sized communities, results may not be generalizable to geographical regions outside the Midwest or larger urban centers. Another limitation is the novelty of the Omron HJ-700IT. Although the technology was deemed beneficial for this study and population, the four-second delay inhibits direct comparisons to data from other pedometers. An additional limitation of using pedometers to study PA is that they only measure ambulatory activity. Conclusions In recent decades, adults with disabilities have experienced many advances in education, work parity, and human rights. Despite these developments, adults with ID experience clear health disparities. The present study examines PA among adults with ID utilizing objective monitoring. Results indicate that the majority of this population does not achieve sufficient PA for all health benefits, and individuals with moderate ID are less active than adults with mild ID. Given the risk for obesity and chronic disease in this population, research into PA promotion for is imperative. Results of this study indicate that PA promotion programs do not necessarily need to be gender-specific, but that those with greater level of impairment may especially need support to perform and maintain PA.
105
Acknowledgments This study was supported by a grant from the Special Olympics Health Professions Student Grants program (Federal project grant # U59/CCU 321826 through the U.S. Centers for Disease Control and Prevention). The pedometers were provided by Hiroshi Ogawa, Omron Health Corporation. The authors are indebted to the study participants, as well as staff at the agencies that provide the participants services, for the support of their participation. References American Psychiatric Association, 2000. Diagnostic and statistical manual of mental disorders, 4th ed. American Psychiatric Press, Inc., Washington, DC. (text revision (DSM-IV-TR)). Bassett, D.R., Ainsworth, B.E., Leggett, S.R., Mathien, C.A., Main, J.A., Hunter, D.C., Duncan, G.E., 1996. Accuracy of five electronic pedometers for measuring distance walked. Med. Sci. Sports Exerc. 28, 1071–1077. Beange, H., McElduff, A., Baker, W., 1995. Medical disorders of adults with mental retardation: a population study. Am. J. Ment. Retard. 99, 595–604. Chan, C.B., Spangler, E., Valcour, J., Tudor-Locke, C., 2003. Cross-sectional relationship of pedometer-determined ambulatory activity to indicators of health. Obes. Res. 11, 1563–1570. Doyle, J.A., Green, M.S., Corona, B.T., Simone, J., Dennison, D.A., 2007. Validation of an electronic pedometer in a field-based setting. New Orleans, LA: American College of Sports Medicine 54th Annual Meeting. Med. Sci. Sports Exerc., vol. 39, p. S186. Draheim, C.C., 2006. Cardiovascular disease prevalence and risk factors of persons with mental retardation. Ment. Retard. Dev. Disabil. Res. Rev. 12, 3–12. Draheim, C.C., Williams, D.P., McCubbin, J.A., 2002. Prevalence of physical inactivity and recommended physical activity in community-based adults with mental retardation. Ment. Retard. 40, 436–444. Emerson, E., 2005. Underweight, obesity and exercise among adults with intellectual disabilities in supported accommodation in Northern England. J. Intellect. Disabil. Res. 49, 134–143. Fernhall, B., Pitetti, K.H., Rimmer, J.H., et al., 1996. Cardiorespiratory capacity of individuals with mental retardation including Down syndrome. Med. Sci. Sports Exerc. 28, 366–371. Fernhall, B., McCubbin, J.A., Pitetti, K.H., et al., 2001. Prediction of maximal heart rate in individuals with mental retardation. Med. Sci. Sports Exerc. 33, 1655–1660. Frey, G.C., 2004. Comparison of physical activity levels between adults with and without mental retardation. J. Phys. Act. Health 1, 235–245. Haller, J., Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Validation of the Omron HJ-112 pedometer at various walking speeds. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Hasson, R.E., Pober, D.M., Freedson, P.S., 2004. Validation of new pedometer during running and walking. Indianapolis, IN, American College of Sports Medicine 51st Annual Meeting. Med. Sci. Sports Exerc., 36, p. S31. Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Evaluation of the Omron HJ112 and Yamax Digiwalker SW-701 pedometers during variable speed walking. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Janicki, M.P., Dalton, A.J., Henderson, C.M., Davidson, P.W., 1999. Mortality and morbidity among older adults with intellectual disability: health services considerations. Disabil. Rehabil. 21, 284–294. LantmandeValk, H.M.J.V., vandenAkker, M., Maaskant, M.A., et al., 1997. Prevalence and incidence of health problems in people with intellectual disability. J. Intellect. Disabil. Res. 41, 42–51. Lee, M., Zhu, W., Yang, L., Bendis, K., Hernandez, J., 2007. Position invariance of Omron-BI pedometers in older adults. New Orleans, LA, American College of Sports Medicine 54th Annual Meeting. Med. Sci. Sports Exerc., 39, p. S187.
106
J.J. Peterson et al. / Preventive Medicine 47 (2008) 101–106
Le Masurier, G., Tudor-Locke, C., 2003. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc 35, 867–871. Luckasson, R., Borthwick-Duffy, S., Buntinx, W.H.E., et al., 2002. Mental Retardation: Definition, Classification, and Systems of Supports, 10th ed. American Association on Mental Retardation, Washington, DC. Roberts, D.E., Hasson, R.E., Pober, D.M., Freedson, P.S., 2005. Effect of pedometer position on step counts accuracy in normal weight and obese individuals. Nashville, TN, American College of Sports Medicine 52nd Annual Meeting. Schalock, R.L., Luckasson, R.A., Shogren, K.A., 2007. The renaming of “mental retardation”: understanding the change to the term “intellectual disability”. Intellect. Dev. Disabil. 45, 116–124. Schneider, P.L., Crouter, S.E., Bassett, D.R., 2004. Pedometer measures of freeliving physical activity: comparison of 13 models. Med. Sci. Sports Exerc. 36, 331–335. Sequeira, M.M., Rickenbach, M., Wietlisbach, V., Tullen, B., Schutz, Y., 1995. Physical activity assessment using a pedometer and its comparison with a questionnaire in a large population survey. Am. J. Epidemiol. 142, 989–999.
Stanish, H.I., 2004. Accuracy of pedometers and walking activity in adults with mental retardation. Adapt. Phys. Act. Q. 21, 167–179. Stanish, H.I., Draheim, C.C., 2005. Walking habits of adults with mental retardation. Ment. Retard. 43, 421–427. Sutherland, G., Couch, M.A., Iacono, T., 2002. Health issues for adults with developmental disability. Res. Dev. Disabil. 23, 422–445. Temple, V.A., Walkley, J.W., 2003. Physical activity of adults with intellectual disability. J. Intellect. Dev. Disabil. 28, 342–353. Temple, V.A., Frey, G.C., Stanish, H.I., 2006. Physical activity of adults with mental retardation: review and research needs. Am. J. Health Promot. 21, 2–12. Trost, S.G., Owen, N., Bauman, A.E., Sallis, J.F., Brown, W., 2002. Correlates of adults' participation in physical activity: review and update. Med. Sci. Sports Exerc. 34, 1996–2001. Tudor-Locke, C., Bassett, D.R., 2004. How many steps/day are enough? Sports Med. 34, 1–8. U.S. Department of Health and Human Services, 1996. Physical Activity and Health: A Report of the Surgeon General. U.S. Department of Health and Human Services, Atlanta, GA.