Wide Variability in Physical Activity Environments and Weather-Related Outdoor Play Policies in Child Care Centers Within a Single County of Ohio FREE

For preschool-aged children, daily physical activity is essential for obesity prevention and for learning age-appropriate gross motor, fine motor, cognitive, and social-emotional skills.^1- 7 More than half of US children aged 3 to 6 years are in child care centers,⁸ and many spend long hours there.⁹ Unfortunately, in these settings, children are sedentary 70% to 83% of the time (excluding nap), and most are not getting adequate amounts of physical activity,^10- 12 perhaps contributing to the increasing rates of obesity among US children.^13- 16

Preschool children's physical activity levels vary widely among child care centers. Indeed, there is substantial variability (27%-47%) in the physical activity of children depending on the center that they attend,^10- 11,17 whereas child factors such as age, race, sex, body mass index, and socioeconomic status all combined explain only a small part (3%-10%)^10- 11 of such variability. Because this is a new and emerging area of inquiry,¹⁸ little is known about why physical activity levels vary among centers or about the physical activity environments in these centers (including facilities and outdoor play policies) that might contribute to the variability in physical activity levels. A better understanding of current conditions and the extent to which these conditions vary according to the demographic characteristics of the centers (eg, for-profit status and race and/or ethnicity and socioeconomic status of population served) is essential to ensure that all children have adequate opportunities to achieve recommended levels of physical activity.^19- 20

Evidence from a few studies^21- 23 suggests that variations (including size and variety of fixed and portable play equipment) in playgrounds and indoor rooms for the development of gross motor skills (hereafter referred to as gross motor rooms) may explain the variability in children's activity among centers. Yet recent surveys of physical activity environments in child care centers have not included measures of playground size, access, surfaces, topography, or quality.^24- 25 Only one published study²⁴ included a measure of shade and access to playground for children with special needs, both of which were recently cited as model playground amenities for child care settings.²⁶

One recent cross-sectional study^21,23 found that higher levels of activity occurred in child care settings that offered more daily occasions of outdoor play and increased total daily time allotted for active play.²³ However, another experimental study²⁷ found no increase in activity levels when outdoor playtime was increased. Both studies were conducted in areas where cold temperatures and/or snow (which have both been linked to lower activity levels^6,17,28- 29) would rarely preclude children's outdoor active opportunities. Our prior research of child care providers in southwestern Ohio^30- 31 has identified clothing- and weather-related policies for outdoor play as novel, but critical, barriers to children's physical activity. In Ohio, centers and/or individual teachers must decide which specific temperatures and precipitation conditions will safely permit children's outdoor play, because guidelines are not explicitly stated by state licensing (Ohio Department of Jobs and Family Services rule 5101:2-12-14). To our knowledge, the variability in playground offerings or in clothing- and weather-related policies for outdoor play among centers has not been systematically studied, despite their potentially significant effect on children's opportunities for physical activity.

The purpose of this study was to better understand the breadth and variability of physical activity environments in child care centers, specifically the (1) indoor and outdoor facilities, including playground size, perceived quality, surfaces, access, and topography, and (2) the weather and clothing policies for outdoor play. Because a center's demographic characteristics may reflect access to resources and center programmatic priorities, we also aimed to understand how physical activity environment and outdoor play policies may vary by child care center demographic characteristics.

Hamilton County, Ohio (greater Cincinnati area), is an urban area and demographically similar to the United States overall (according to the 2000 census), with median household income of $60 646, 10.8% of families below poverty level, and 6.6% of employed workers (with children aged <6 years), but it has a higher percentage of black or African American residents (24.2% vs 12.4%) and a lower proportion of Asian residents (1.7% vs 4.4%) and Hispanic or Latino residents (1.5% vs 14.8%). The mean maximum and minimum temperatures for the months of January (37°F [2.8°C] and 21°F [−6.1°C], respectively) and July (88°F [31.1°C], 66°F [18.8°C]) and the year (63°F [17.2°C], 44°F [6.6°C]) reflect moderate but varied temperatures. Annually, Cincinnati receives an average of 42.6 in of rainfall, 22.5 in of snow, and has 132 days with precipitation of 0.01 in or more (National Weather Service).

All directors (n = 196) of licensed child care centers in Hamilton County that offer full-time care to children aged 3 to 6 years (not in kindergarten) were eligible to participate. These directors received a letter of introduction and a follow-up telephone call. After 8 unsuccessful attempts to conduct an interview with a director, the child care center was considered unreachable. Interviews were completed between August 2008 and March 2009. Participants completing the survey received a $15 gift certificate. The protocol was reviewed and determined to be exempt by the Cincinnati Children's Hospital institutional review board.

The final sample (Table 1) included 162 of 185 eligible centers (88%). Eleven centers were ineligible because they were no longer in service (n = 5), had no valid telephone number (n = 3), did not serve preschool-aged children (n = 1), were not full-time centers (n = 1), or were not located in Hamilton County (n = 1). Thirteen centers refused to participate in our survey because senior administrative staff (eg, the owner) declined to participate (n = 5), they were not interested (n = 4), they had no time to participate (n = 2), or participation was not mandatory (n = 2).

The only existing validated instrument to measure physical activity environments in child care centers, the Nutrition and Physical Activity Self-Assessment for Child Care (NAP SACC),^33- 35 does not include measures of weather- or clothing-related policies for active play or measures of playground size, access, surfaces, shade, topography, or quality. Thus, we developed a new instrument, the Early Learning Environments for Physical Activity and Nutrition Environments Telephone Survey (ELEPHANTS), which targeted these specific constructs and also several concepts related to program structure, staff training and behavior, and sedentary environment that have been linked to children's physical activity levels²⁵ and recently cited as model regulations for promoting physical activity in child care (although ELEPHANTS was developed and deployed prior to the publication of the model regulations).²⁶ This survey included items from NAP SACC that were previously shown to have good reliability and validity.³⁵ Items that had low agreement with directly observed behaviors were modified. The wording and content of both the new and modified items were developed from the results of a series of focus groups³¹ that assessed barriers to physical activity in child care.³¹ This survey was reviewed for content validity by experts from diverse fields, including early education, child care playground equipment sales, landscape architecture, physical activity and parks assessment, nutritional epidemiology, and scale development. Additionally, the instrument was assessed for face validity through pretesting and cognitive interviews^36- 37 with 13 child care providers. The final ELEPHANTS instrument consists of 65 questions designed to be administered by phone to administrative staff by a trained interviewer. Approximately half the items were open-ended and/or volunteer responses, which required the interviewer to choose from preselected response options only visible to the interviewer. If a volunteered response did not fit clearly into any of the response options, the interviewer was instructed to probe with clarifying questions. For multiple response questions (eg, “What fixed structures are on your playground?”), participants were prompted to supply an exhaustive list. Phone interviews were audio-recorded, and survey responses were recorded on paper. This survey took a mean (SD) time of 33 (9) minutes to complete. A copy of the final ELEPHANTS instrument is available from the primary author on request.

Descriptive statistics identified the prevalence and variability of environmental features, policies, and child care center characteristics. We developed 2 summary scores based on the presence (scored as 1) or absence (scored as 0) of 12 facility criteria (Table 2) and 7 weather-related and/or outdoor play policies (Table 3) that have been linked to children's physical activity. Continuous criteria, such as the number of winter clothing items kept for loan and the number of types of portable play equipment, were dichotomized at their median value²¹ (3 and 5, respectively, for these examples), with centers that reported values greater than or equal to the median receiving a score of 1 for that criterion. Because the amount of fixed equipment has been found to be inversely related to children's physical activity,^21,23 centers that reported having a number of fixed pieces of equipment less than or equal to the median number (ie, 4) were assigned a score of 1 for this criterion. Table 4 lists the center characteristics considered as candidate predictors for the summary scores. We used a t test and 1-way analysis of variance to compare summary scores by level of the independent variables, and we performed post hoc Tukey pairwise comparisons for all statistically significant categorical predictor variables with 3 or more levels. A P value of less than .05 was considered statistically significant.

Of the 162 child care centers that responded to our survey, 151 (93%) reported having an on-site outdoor playground, and 96 (59%) reported having more than 1 playground (range, 0-5 playgrounds) dedicated to different age groups. On playgrounds, centers reported a mean (SD) number of 3.8 (1.9) pieces of fixed equipment (range, 0-13 pieces), usually a climber (143 of 151 centers [95%]), dramatic play structures such as a playhouse (56 of 151 centers [37%]), basketball hoop or other aiming structure (49 of 151 centers [32%]), a place to sit and do quiet activities (46 of 151 centers [30%]), and tunnels (44 of 151 centers [29%]). Similarly, centers reported a mean (SD) number of 4.6 (1.9) types of outdoor portable play equipment (range, 0-9 types), usually balls (138 of 151 centers [91%]), riding toys (107 of 151 centers [71%]), art materials (91 of 151 centers [60%]), jump ropes (81 of 151 centers [54%]), and sand or water toys (67 of 151 centers [44%]). The typical playground had 2 surfaces: usually mulch (102 of 161 centers [63%]) and grass (63 of 161 centers [39%]) or concrete (55 of 161 centers [34%]). Of 158 centers, 135 (85%) had special-needs access, 63 (40%) had wheelchair access to playground, 48 (30%) had surfaces that accommodate a wheelchair, 48 (30%) had specialized adaptive equipment, and 39 (25%) had a place to sit and do quiet activities. Playground size varied from “very large” (27 of 158 centers [17%]), “large” (61 of 158 centers [39%]), “medium to average” (52 of 158 centers [33%]), to “small” (18 of 158 centers [11%]) or, alternatively, from about the size of a football field (5%), half of a football field (21%), a basketball court (41%), a tennis court (16%), to half of a basketball court (17%). Common improvements that directors wanted to make were to add more climbing or fixed equipment (75 of 162 centers [46%]), change or add surfaces (55 of 162 centers [34%]), and add more portable play equipment (36 of 162 centers [22%]).

Of the 162 centers that participated in our survey, 82 (51%) reported having a dedicated indoor gross motor room. Room size varied from about the size of a basketball court (19 of 78 centers [24%]), a tennis court (9 of 78 centers [12%]), half a basketball court (22 of 78 centers [28%]), a large classroom (24 of 78 centers [31%]), to a small classroom (4 of 78 centers [5%]). All but 1 of the 80 centers without a dedicated indoor gross motor room reported that they used either a classroom or hallway for indoor activity. The Figure shows the distribution of facility and weather scores.

Weather-related policies permitting outdoor play varied among centers (Table 3). Of the 137 centers that cited a minimum temperature for outside play, 74 (54%) cited it as 32°F [0°C], 32 (23%) reported lower minimum temperatures (ie, <32°F [<0°C]), and 31 (23%) reported higher minimum temperatures (>32°F [>0°C]) (range, 20°F-65°F [−6.6°C to 18.3°C]); 125 of 160 centers (78%) reported that the minimum temperature was listed in their parent handbook. Centers reported a wide range in total time (17.5-120 minutes; mean [SD], 77 [35] minutes) that children were scheduled outdoors, usually consisting of 2 outdoor sessions lasting a mean (SD) time of 38 (17) minutes each. However during inclement weather, directors reported curtailing these sessions to a mean (SD) time of 17 (9) minutes each. When deciding whether the weather conditions safely permit outdoor play, 55% of directors indicated that they made this decision, 41% left it to the individual teacher's discretion, and 4% decided through teacher consensus. Centers reported keeping a mean (SD) number of 2.8 (1.4) additional clothing items (range, 0-5 items), such as coats (76%), hats (71%), and mittens (75%), for loan as needed.

Higher facility scores were significantly related to fewer children receiving tuition assistance, having primarily white children enrolled, and a greater percentage of staff with a bachelor's degree (Table 4). Physical activity–promoting weather-related policies were associated with centers being not for profit, having fewer children receiving tuition assistance, and being accredited by the National Association for the Education of Young Children (NAEYC).

To our knowledge, this is the first systematic study of child care centers' physical activity facilities and outdoor play policies. In this study, we found that there is considerable variability in facilities (even within a single county of Ohio) and that, depending on a center's policies, children's active opportunities can often be curtailed because of inclement weather. This is problematic because only half of the centers had an indoor gross motor room where children could be active during inclement weather. Most centers had an accessible playground on site, but only half had a large space, shade, and a variety of portable play equipment. Less than a third had a hard surface to permit all types of portable play equipment. Compared with other studies of playground environments, a similar proportion of centers had playgrounds,²⁴ and the playgrounds were of comparable size,²¹ had less fixed equipment^21,25 and shade,²⁴ and more portable play equipment^21,25 and special needs access.²⁴ Insufficient playground and indoor facilities may help explain why preschool-aged children are not getting sufficient amounts of physical activity.^10- 12

Weather-related policies permitting playground use also varied considerably, with multiple weather conditions potentially curtailing children's opportunities for physical activity. No center reported allowing children outside in wind chills of 15°F (−9.4°C) or below, the minimum safe temperature per recent guidelines.^20,38- 39 Specifically, only 20% of centers permitted children to go outside in subfreezing temperatures (<32°F [<0°C]). This would mean that children in this area could not go outside approximately 1.5 months per year (National Weather Service). Only 43% of centers reported allowing children outdoors during light rain, adding an additional 132 days in which children would not be allowed outdoors (National Weather Service). In summary, the typical weather-related policies at child care centers in Cincinnati would restrict outdoor play for approximately 179 days (46%) of the year. The 32°F [0°C] minimum temperature is much higher than the minimum safe temperature for play according to newly released physical activity guidelines for child care (−15°F [−26.1°C] windchill)²⁰ but is congruent with the local elementary school system's outdoor play policy. It is not known how these outdoor play policies were developed. To our knowledge, this is the first study of weather-related outdoor policies of child care centers. More research is needed to understand how these policies were developed and how parents' and teachers' attitudes about weather (including concerns about children getting dirty or sick when exposed to cold air)³¹ may influence decisions about outdoor play. This highlights the interaction between policy, provider, and parent attitudes. Reducing the minimum safe temperature for outdoor play and providing training about safe conditions for outdoor play could substantially increase opportunities for physical activity. Given that we have previously reported that parents do not dress children appropriately for weather and active play,³¹ pediatric clinicians should consider discussing with parents during well-child checks the importance of (1) outdoor play for healthy growth and development, (2) dressing children appropriately for play, and (3) working with child care center staff to ensure that children are given adequate opportunities for physical activity.

Centers having more low-income children (receiving subsidized child care) also reported having poorer quality facilities and more stringent outdoor play policies. Centers with primarily nonwhite children reported poorer quality facilities. This is of concern because nonwhite and low-income children may not have safe opportunities for outdoor play near or around their homes,^40- 43 so the child care center may be their only opportunity for outdoor play. The tuition rate for full-paying attendees was not found to be significantly associated with quality of physical activity facilities or practices, yet most children (65%) were receiving some form of subsidized child care. A potential way to increase children's physical activity and reduce income and race disparities in physical activity opportunities may be to provide small grants to centers that serve minority and/or low-income children; these grants would enable centers to establish or enrich safe places for children to be active. Many of the improvements would not be costly (eg, reconfiguring an indoor space to allow gross motor play during inclement weather or the addition of shade trees, more portable play materials, or a hard surface to the playground). The grants should be directive, though, because the most common change that respondents wanted to make to the playground was to add more fixed play equipment, which is both very costly and inversely related to children's physical activity levels.^21,23 Such grants could offer opportunities to educate staff about cost-effective strategies to increase children's physical activity. The reason for the relation between lower income child care centers and more stringent outdoor policies is unclear, but perhaps better training or subsidies to provide warm or waterproof clothing to children could help these centers have a less strict weather-related policy regarding outdoor play.

Centers that were NAEYC-accredited reported having more practices associated with physical activity promotion, which may be related to the NAEYC requirement that directors be trained in early childhood development. Centers that were not for profit also reported a greater number of weather conditions under which children were permitted to play outdoors. These associations may be helpful in identifying which types of centers may benefit from policy training about safe outdoor weather conditions for play.

Our results may not be generalizable to other regions and were based on self-report rather than observation of actual practices. However, our study's strengths are that it was a census of all licensed full-time child care centers within this region, not just centers of a particular type (eg, Head Start), and that it achieved a high response rate. This allowed us to examine demographic predictors of physical activity policies and facilities within the sample. We focused on facilities and outdoor play policies because directors may not reliably report their staff's behavior. Our survey was designed to use open-ended questions and volunteer responses to minimize social desirability bias. Lastly, because there have been few previous studies of the quality of child care physical activity facilities or weather-related policies for playground use, we developed summary scores for these 2 domains de novo. The individual criteria in the scores were selected from predictors identified in the literature and our formative research. Further research is needed to identify the factors associated with higher levels of physical activity.

Our study contributes to the sparse but growing literature of physical activity environments in child care centers by reporting on the prevalence of “typical” child care policies and environments, and the extent to which they vary even within a single county. Additionally, it highlights the potential large effect of weather, outdoor play policies, and caregiver behaviors related to weather in influencing children's opportunities for active play.

Correspondence: Kristen A. Copeland, MD, General and Community Pediatrics, Cincinnati Children's Hospital Medical Center, 3333 Burnet Ave, MLC 7035, Cincinnati, OH 45229 (kristen.copeland@cchmc.org).

Accepted for Publication: October 14, 2010.

Published Online: January 3, 2011. doi:10.1001/archpediatrics.2010.267

Author Contributions: Dr Copeland had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Copeland, Saelens, and Kalkwarf. Acquisition of data: Copeland, Sherman, and Foster. Analysis and interpretation of data: Copeland, Sherman, Khoury, Saelens, and Kalkwarf. Drafting of the manuscript: Copeland and Foster. Critical revision of the manuscript for important intellectual content: Sherman, Khoury, Saelens, and Kalkwarf. Statistical analysis: Copeland and Khoury. Obtained funding: Copeland. Administrative, technical, and material support: Sherman and Foster. Study supervision: Kalkwarf.

Financial Disclosure: None reported.

Funding/Support: This study was supported in part by a grant under The Robert Wood Johnson Foundation Physician Faculty Scholars Program, a career development award from the National Institutes of Health (K23 HL088053), and the National Center for Research Resources, National Institutes of Health (US Public Health Service grant UL1 RR026314).

Additional Contributions: We thank 4C, the child care resource and referral agency for the greater Cincinnati area, for their assistance with this study.