Abstract
The objectives of this study were to determine if neighborhood measures were associated with physical activity cross-sectionally during late pregnancy (27–30 weeks’ gestation), 3 months postpartum, and 12 months postpartum, and longitudinally with an increase in physical activity from late pregnancy to 12 months postpartum. Data are from the Pregnancy, Infection, and Nutrition (PIN3) and Postpartum Prospective Cohort Study. Dichotomized self-reported recreation and total physical activity hours/week were explored cross-sectionally at three time points, and as an increase over time. Four factors from a neighborhood environmental audit were examined: arterial or thoroughfare, walkable neighborhood, physical incivilities, and decoration. Secondary spatial data included population density, hilliness, intersection density, distance to nearest major road, distance to nearest park, distance to nearest physical activity facility, and distance to nearest bus stop. Multilevel mixed-effects logistic regression models were used to assess the association between environmental variables and physical activity measures. A number of environmental variables were associated with total and recreation physical activity at the three time points in cross-sectional models. For increase in recreation physical activity over time, a moderate distance to nearest major road was significantly associated with increased recreation physical activity from 3 to 12 months postpartum (tertile 2 OR 2.13, 95% CI 1.08, 4.22). Living the furthest distance from the nearest park was inversely associated with an increase in recreation physical activity from pregnancy to 3 months postpartum (tertile 3 OR 0.50, 95% CI 0.29, 0.85). The findings of this study indicate that several aspects of the neighborhood environment, such as walkability, access to transit, distance to recreation facilities, and road networks, are associated with physical activity during different stages of pregnancy and postpartum. Since physical activity may result in long-term health benefits for both the woman and child, environments that support this activity should be encouraged.
Keywords: Recreation, GIS, Audit, Built environment, Leisure-time physical activity, Women
Introduction
Physical activity provides health benefits during pregnancy, delivery, and postpartum for both the mother and child [1, 2]. Because of the known benefits, many countries have specific guidelines in place for recommending physical activity during pregnancy [3, 4] and postpartum [5, 6]. Evidence suggests that women reduce their levels of physical activity when they are pregnant [7–9], a reduction that can extend to the postpartum period [10, 11]. These changes may lead to a lifetime of insufficient physical activity and related adverse health consequences, such as postpartum weight retention and obesity [12, 13].
Physical activity is influenced by factors within and across levels of the ecologic model, including intrapersonal, interpersonal, environmental, and institutional factors [14]. Neighborhood environmental factors include aspects of the physical built environment (e.g., street connectivity), as well as the social environment (e.g., visible people on the streets). In the general population, evidence suggests that physical activity is influenced by aspects of the built environment in which children and adults live, work, and play [15, 16]. It is of particular importance to examine the influence of neighborhood environmental factors on physical activity in pregnancy and postpartum, as it is a period in the life course where women’s physical activity is apt to change [11]. With their growing family and associated responsibilities, these women in particular need places to be physically active that are close to home and accessible [17].
Aspects of the neighborhood environment have been examined in relation to physical activity in pregnant and postpartum women, but only to a limited degree [18–20]. Findings from studies on access to green space or recreation facilities among pregnant and postpartum women have been mixed. For example, a study by Richardsen and colleagues [18] indicated an association between access to recreational areas and higher levels of physical activity, while Nichani and colleagues [19] showed no association. Other features of the physical environment, such as population density and land use mix, are also important for physical activity in the general population and may be important for women during and post-pregnancy as well [16].
Research is needed to understand what aspects of the neighborhood environment are associated with physical activity during pregnancy and postpartum, as well as changes in physical activity across these time points. The first objective of this study was to determine if environmental audit and spatially derived neighborhood measures were associated with self-reported physical activity (total and recreation) during late pregnancy (27–30 weeks’ gestation), at 3 months postpartum, and at 12 months postpartum. The second objective of this study was to determine if neighborhood measures were associated with an increase in self-reported physical activity (total and recreation) from late pregnancy to 12 months postpartum.
Methods
Study Sample
Data for this analysis are from the Pregnancy, Infection, and Nutrition (PIN3) and PIN Postpartum Study. A full description of the study had been published previously [21]; more information can be found at http://www.cpc.unc.edu.libproxy.lib.unc.edu/pin. PIN3 was a prospective cohort study that enrolled 2006 pregnant women at 20 weeks’ gestation between January 2001 and June 2005 from prenatal clinics at the UNC Hospitals. Of the original PIN3 participants, 688 women agreed to participate in the PIN Postpartum Study and completed the 3 months postpartum interview (73.3% of those invited to participate). Five-hundred and fifty (79.9%) of the participants that completed the 3 months postpartum visit also completed an interview at 12 months postpartum [11, 22]. Description of drop-out and assessment of selection bias can be found elsewhere [11, 22, 23]. Data used in this analysis were collected via telephone and in-home interview at four time points: 17–22 weeks’ gestation (phone #1), 27–30 weeks’ gestation (phone #2), 3 months postpartum (in-home #1), and 12 months postpartum (in-home #2). If a participant moved out of the 4-county area between study visits, they were excluded from further follow-up. This study was reviewed and approved by the Institutional Review Board at the University of North Carolina at Chapel Hill (UNC), and each participant provided informed consent prior to participation in these studies.
Physical Activity
Self-reported physical activity was measured at each of the interview time points using the PIN3 physical activity survey with evidence for validity and reliability [24]. This interviewer administered survey assessed mode-by-mode physical activity over a 1-week recall period; modes assessed included recreation, workplace, transportation, childcare, adult care, and both indoor and outdoor household activities. For each mode, participants reported (a) each type of activity that was performed in the past week, (b) the number of sessions performed over the past week for each reported activity, (c) duration of each session, and (d) perceived intensity of each session (“fairly light,” “somewhat hard,” or “hard or very hard”). Data were converted to hours/week within each mode and summed for total physical activity.
For the purposes of this analysis, physical activity data from phone #2, in-home #1, and in-home #2 are used. Recreation and total physical activity hours/week were explored, with all perceived intensities included. Recreation physical activity hours/week was dichotomized into none/any, whereas total physical activity hours/week was dichotomized using distribution-based cut-points (low/high). Increase in recreation physical activity was defined as ≥ 1-h/week difference in either measure of physical activity between late pregnancy and 3 months postpartum, as well as between 3 months postpartum and 12 months postpartum. Conversely, increase in total physical activity was defined as ≥ 2-h/week difference in either measure of physical activity between late pregnancy and 3 months postpartum, as well as between 3 months postpartum and 12 months postpartum. These cut-points were determined based on the increase in domain-specific physical activity from pregnancy to 3 months postpartum at the 75th percentile rounded up to the nearest hour (e.g., the 75th percentile of recreation physical activity hours per week was 1.29 at phone #2 and 2.25 at in-home #1, for an increase of approximately 1 h).
Built Environment
Participants’ home addresses were identified at three time points: 27–30 weeks’ gestation (phone #2), 3 months postpartum (in-home #1), and 12 months postpartum (in-home #2). Addresses collected via telephone interview were geocoded using an address database; global position system (GPS) was used to identify and geocode the home location during the in-home interviews. Using a customized road network developed for this study [25], all roads within ¼ mile driving distance of the participants residential address, defined as the participant’s neighborhood, were examined using on-the-ground environmental audit and secondary spatial data. Each road within the neighborhood was divided up into segments, defined as the length between two intersections, between an intersection and a cul-de-sac, or an intersection and a dead-end. This process was done for the addresses collected at each time point, so that all spatial data are representative of the neighborhood in which the participant lived at that time. Table 1 provides a full description of all built environment variables used in this study. For the purposes of analyses, all built environment variables were categorized into distribution-based tertiles (tertile 1 [T1], tertile 2 [T2], tertile 3 [T3]).
Table 1.
Objectively measured built environment variables
| Variables | Definitions | Median (IQR) 27–30 weeks’ gestation |
Median (IQR) 3 months postpartum |
Median (IQR) 12 months postpartum |
Min | Max | Data source |
|---|---|---|---|---|---|---|---|
| Arterial | Aspects of the road network and infrastructure, such as speed limits, 3- and 4-way intersections | − 0.06 (0.3) | − 0.06 (0.3) | − 0.1 (0.3) | − 0.96 | 0.65 | Neighborhood environmental audit |
| Walkable neighborhood | Social and accessibility aspects of the neighborhood, such as visible children and adults, pedestrian-oriented lighting | 0.07 (0.3) | 0.07 (0.3) | 0.07 (0.3) | − 0.30 | 1.10 | Neighborhood environmental audit |
| Physical incivilities | Visually or socially distressing aspects of the neighborhood, such as poor condition of residential units, presence of litter | − 0.3 (0.9) | − 0.3(0.9) | − 0.3 (0.8) | − 1.61 | 1.58 | Neighborhood environmental audit |
| Decoration | Aspects of the neighborhood that may be visually appealing, such as porches and decorations | 0.01 (0.3) | 0.01 (0.3) | 0.02 (0.3) | − 1.19 | 0.55 | Neighborhood environmental audit |
| Population density | The number of people per square mile over non-water area within the area of the ¼-mile network neighborhood | 2004.6 (3004.9) | 2007.1 (3091.3) | 1923.9 (2827.9) | 0 | 12,351 | 2000 Census Block Data |
| Hilliness | Percent of 100-ft road segment lengths within a ¼-mile network neighborhood with slope > 5% | 47.9 (48.1) | 46.4 (49.3) | 47.8 (50.4) | 0 | 98.30 | Alamance County GIS website; NCDOT; UNC GIS database; NC General Assembly Redistricting Data website |
| Distance to nearest major road | Network distances, in miles, from each respondent location to the nearest major road segment, defined as either Interstate, US Highway, or State Highway | 0.8 (0.9) | 0.8 (0.9) | 0.8 (0.9) | 0 | 5.92 | Alamance County GIS website; NCDOT; UNC GIS database; NC General Assembly Redistricting Data website |
| Distance to nearest park | Network distances, in miles, from each respondent location to the nearest park access point | 0.8 (0.9) | 1.5 (2.0) | 1.6 (1.9) | 0 | 12.19 | |
| Distance to nearest physical activity facility | Network distances, in miles, from each respondent location to the nearest physical activity facility | 1.3 (1.4) | 1.3 (1.4) | 1.4 (1.8) | 0.04 | 8.50 | Reference USA |
| Intersection density | The total number of 3-way and 4-way intersections per square mile, excluding area of water features, within each of the respondent ¼ mile network neighborhood | 6 (7) | 6 (7) | 6 (6) | 0 | 35 | Alamance County GIS website; NCDOT; UNC GIS database; NC General Assembly Redistricting Data website |
| Distance to nearest bus stop | Network distances, in miles, from each respondent location to the nearest bus stop | 0.6 (3.8) | 0.7 (4.6) | 0.7 (3.7) | 0.00 | 28.8 | Chapel Hill Transit; Durham Area Transit Authority; Triangle Transit Authority |
| Median parcel tax value | Median tax value, US dollars, of the parcels for each audit road segment | 205.0 (4727.5) | 204.7 (7686.1) | 212.7 (490.0) | 0 | 10,985,960 | Tax parcel datasets for Alamance County, Chatham County, Durham County, and Orange County |
Abbreviations: IQR, interquartile range; GIS, geographic information system; NCDOT, North Carolina Department of Transportation; UNC, University of North Carolina at Chapel Hill; NC, North Carolina; USA, United States of America
PIN3 Neighborhood Audit
The PIN3 Neighborhood Environmental Audit Instrument, including development and data collection procedures [25, 26], have been previously described. The PIN3 Neighborhood Audit Instrument included 43 measures that made up five domains hypothesized to influence outdoor physical activity: residential land use; non-residential land use; public, residential and non-residential space/esthetics; mobility (walking and bicycling) amenities; and transit and road characteristics. Trained auditors, in teams of two, rated all road segments within the participant’s neighborhood over two summers in 2005 and 2006. Road segments were assessed by driving up and down the segment several times and then stopping to complete the audit instrument. Auditors then moved on to the next segment and repeated the process. A total of 10,770 street segments were assessed. The data collected were used to create four factors for analysis: arterial or thoroughfare, walkable neighborhood, physical incivilities, and decoration [26]. The PIN3 Neighborhood Environmental Audit Instrument shows evidence for reliability and stability in urban and rural environments [27].
Secondary Spatial Data
All secondary spatial data were created in ArcGIS 9.1 (Esri, 2005), with further processing in Python. A network neighborhood, or buffer, was a polygon that encompassed all accessible streets within a specified distance using the road network. Each participant’s ¼ mile network neighborhood was used to create individually based estimates for several spatial variables under study at each of 3 time points, including population density, hilliness, and intersection density. Other spatial proximity variables, such as distance to nearest major road, distance to nearest park, distance to nearest physical activity facility, and distance to nearest bus stop, were calculated as the network distance along the shortest route from the participant’s residence, regardless of distance.
Covariates
Covariates were chosen a priori based of previous literature [21]. These include age in years at the start of pregnancy (≤ 24, 25–29, 30–34, 35+), race (white, black, other), marital status at phone #1 (married or unmarried), years of education at phone #1 (≤ 12, 13–15, 6, and 17+), employment status at each time point (employed yes or no), poverty status at phone #1 (≤ 185% or > 185% of the poverty line), parity (number of previous live births and stillbirths: 0, 1, or 2+), chronic hypertension (yes/no), pregnancy-induced hypertension (yes/no), vaginal bleeding (any/none), and median parcel tax value along the participant’s residential address road segment (see Table 1 for the source of data for this variable).
Statistical Analysis
Descriptive statistics included median and interquartile range (IQR) for continuous variables, and proportion for categorical and nominal variables. As participants were nested within neighborhoods, multilevel mixed-effects logistic regression models were used to assess the association between built environment variables (T2 and T3 vs. T1 [referent]) and the following outcomes: (1) dichotomized total physical activity (high vs. low [referent]) at (a) pregnancy, (b) 3 months postpartum, and (c) 12 months postpartum; (2) dichotomized recreation physical activity (any vs. none [referent]) at (a) pregnancy, (b) 3 months postpartum, and (c) 12 months postpartum; (3) dichotomized increase in total physical activity from (a) pregnancy to 3 months postpartum and (b) 3 months postpartum to 12 months postpartum; and (4) dichotomized increase in recreation physical activity from (a) pregnancy to 3 months postpartum and (b) 3 months postpartum to 12 months postpartum. Increase in total physical activity hours/week was compared to decrease or no change [referent] at the two change points—pregnancy to 3 months postpartum, and 3 months to 12 months postpartum. Change in recreation physical activity was similarly assessed. Bivariate and multivariable analyses controlling for covariates (age, race, marital status, years of education, employment status, poverty status, parity, and median parcel tax value along the participant’s residential address road segment) were conducted for all models. To account for clustering by neighborhood, grid cells at a 0.25-mile cell size were laid over the network neighborhood polygons; participants who fell within each grid cell were considered to be nested in the same neighborhood. Within each time point, participants with missing values for any variables were excluded. Statistical significance was set at p < 0.05; all analyses were conducted in STATA 15.1 (STATA Corp., College Station, TX, USA) in 2018.
Results
Descriptive Characteristics of the Study Population
Descriptive characteristics of the study population at the three examined time points are reported in Table 2. Of those that participated in the PIN Postpartum Study, 415 participants had complete spatial and interview data at 27–30 weeks’ gestation (missing n = 38), 437 participants at 3 months postpartum (missing n = 16), and 324 participants at 12 months postpartum (missing n = 26); 68% (n = 309) had complete data for all three time points. At 27–30 weeks’ gestation, the majority of participants were aged 30–34, white, married, had 17+ years of education, and lived above the poverty line. The median self-reported hours spent in total physical activity in the past week increased across the three time points, from 1.0 h at 27–30 weeks’ gestation, to 1.7 h at 12 months postpartum.
Table 2.
Characteristics of the study population from pregnancy to postpartum
| 27–30 weeks’ gestation (n = 415) | 3 months postpartum (n = 437) | 12 months postpartum (n = 324) | |
|---|---|---|---|
| Collected at 17–22 weeks’ gestation | |||
| Age; n (%) | |||
| ≤ 24 | 49 (11.8) | 58 (13.3) | 36 (11.1) |
| 25–29 | 124 (29.9) | 130 (29.8) | 93 (28.7) |
| 30–34 | 171 (41.2) | 175 (40.1) | 133 (41.1) |
| 35+ | 71 (17.1) | 74 (16.9) | 62 (19.1) |
| Race; n (%) | |||
| White | 333 (80.2) | 341 (78.0) | 262 (80.9) |
| Black | 52 (12.5) | 62 (14.2) | 38 (11.7) |
| Other | 30 (7.2) | 38 (11.7) | 24 (7.4) |
| Marital status; n (%) | |||
| Married | 357 (86.0) | 365 (83.5) | 285 (88.0) |
| Single, widowed, divorced, or separated | 58 (14.0) | 72 (16.5) | 39 (12.0) |
| Education; n (%) | |||
| ≤ 12 | 46 (11.1) | 56 (12.8) | 32 (9.9) |
| 13–15 | 45 (10.8) | 53 (12.1) | 34 (10.5) |
| 16 | 136 (32.8) | 139 (31.8) | 111 (34.3) |
| 17+ | 188 (45.3) | 189 (43.3) | 147 (45.4) |
| Income (percent of the poverty line); n (%) | |||
| ≤ 185 | 68 (16.4) | 81 (18.5) | 45 (13.9) |
| > 185 | 347 (83.6) | 356 (81.5) | 279 (86.1) |
| Parity; n (%) | |||
| 0 | 211 (50.8) | 219 (50.1) | 160 (49.4) |
| 1 | 144 (34.7) | 153 (35.0) | 115 (35.5) |
| 2+ | 60 (14.5) | 65 (14.9) | 49 (15.1) |
| Chronic hypertension | |||
| Yes | 22 (5.3) | 26 (6.0) | 16 (4.9) |
| No | 393 (94.7) | 411 (94.1) | 308 (95.1) |
| Pregnancy-induced hypertension | |||
| Yes | 89 (21.5) | 94 (21.5) | 71 (21.9) |
| No | 326 (78.6) | 343 (78.5) | 253 (78.1) |
| Vaginal bleeding | |||
| Yes | 102 (24.6) | 106 (24.3) | 80 (24.7) |
| No | 313 (75.4) | 331 (75.7) | 244 (75.3) |
| Time varying | |||
| Employment status; n (%) | |||
| Yes | 298 (71.8) | 242 (55.4) | 208 (64.2) |
| No | 117 (28.2) | 195 (44.6) | 116 (35.8) |
| Moved residence during the study period | |||
| Yes | – | 30 (6.9) | 20 (6.2) |
| No | – | 407 (93.1) | 304 (93.8) |
| Total physical activity, hours/week | |||
| Median (IQR) | 1 (2.5) | 1.5 (4.3) | 1.7 (3.8) |
| Low physical activity; n (%) | 236 (56.9) | 215 (49.2) | 147 (45.4) |
| High physical activity; n (%) | 179 (43.1) | 22 (50.8) | 177 (54.6) |
| Total recreation physical activity, hours/week | |||
| Median (IQR) | 0 (1.3) | 0.3 (2.3) | 0.3 (2.2) |
| None; n (%) | 259 (62.4) | 209 (47.8) | 160 (49.4) |
| Any; n (%) | 156 (37.6) | 228 (52.2) | 164 (50.6) |
| Increase in total physical activity, hours/week; n (%)a, c | |||
| No | – | 268 (61.3) | 226 (69.8) |
| Yes | – | 169 (38.7) | 98 (30.3) |
| Increase in recreation physical activity, hours/week; n(%)a, b | |||
| No | – | 301 (68.9) | 251 (77.5) |
| Yes | – | 136 (31.1) | 73 (22.5) |
Excludes individuals with missing data for outcomes and covariates (n = 80)
aIncrease from the previous interview—at 3 months postpartum, increase from 27 to 30 weeks’ gestation; at 12 months postpartum, increase from 3 months postpartum
bIncrease in at least 1 h of recreation physical activity between time points
cIncrease in at least 2 h of total physical activity between time point
Cross-Sectional Associations
The bivariate and multivariable multilevel mixed-effects logistic regression models examining the independent associations between each built environment variable and total and recreation physical activity at each time point are reported in Tables 3 and 4, respectively. Considering total physical activity, in multivariable models, having a high level of neighborhood walkability was positively associated with the odds of reporting high total physical activity (T3 OR 2.30; 95% CI 1.29, 4.12), and distance to nearest physical activity facility was inversely associated with the odds of reporting high total physical activity (T3 OR 0.52; 95% CI 0.29, 0.92) (Table 3) at 27–30 weeks’ gestation. Living the furthest distance from the nearest bus stop was inversely associated with the odds of reporting high total physical activity at 12 months postpartum (OR 0.50; 95% CI 0.26, 0.96) (Table 3). Having a moderate level of decoration in the participant’s neighborhood was inversely associated with the odds of reporting high total physical activity at 3 months postpartum (T2 OR 0.54; 95% CI 0.31, 0.95) (Table 3).
Table 3.
Multilevel mixed-effects logistic regression estimates, cross-sectional association between built environment variables and total physical activity
| 27–30 weeks’ gestation (n = 417) OR (95% CI) |
3 months postpartum (n = 439) OR (95% CI) |
12 months postpartum (n = 325) OR (95% CI) |
||||
|---|---|---|---|---|---|---|
| Bivariate | Multivariable | Bivariate | Multivariable | Bivariate | Multivariable | |
| Environmental audit variables | ||||||
| Arterial | ||||||
| T2 |
1.56 (0.89, 2.73) |
1.64 (0.94, 2.87) |
0.96 (0.56, 1.64) |
0.95 (0.56, 1.62) |
1.00 (0.54, 1.83) |
1.02 (0.56, 1.87) |
| T3 |
1.30 (0.74, 2.27) |
1.38 (0.78, 2.41) |
1.32 (0.77, 2.27) |
1.33 (0.77, 2.29) |
1.23 (0.65, 2.31) |
1.25 (0.66, 2.36) |
| Walkable neighborhood | ||||||
| T2 |
1.88 (1.06, 3.33)* |
1.70 (0.96, 3.01) |
2.20 (0.70, 2.06) |
1.09 (0.64, 1.87) |
0.75 (0.40, 1.41) |
0.72 (0.39, 1.35) |
| T3 |
2.52 (1.41, 4.51)** |
2.30 (1.29, 4.12)** |
1.34 (0.77, 2.33) |
1.19 (0.69, 2.07) |
1.72 (0.90, 3.29) |
1.59 (0.83, 3.02) |
| Physical incivilities | ||||||
| T2 |
0.69 (0.40, 1.21) |
0.72 (0.41, 1.25) |
0.90 (0.52, 1.55) |
0.90 (0.52, 1.55) |
0.85 (0.46, 1.58) |
0.90 (0.49, 1.65) |
| T3 |
0.84 (0.48, 1.48) |
0.89 (0.49, 1.60) |
0.95 (0.55, 1.63) |
1.04 (0.58, 1.84) |
1.03 (0.55, 1.95) |
1.15 (0.60, 2.22) |
| Decoration | ||||||
| T2 |
0.83 (0.47, 1.46) |
0.78 (0.44, 1.38) |
0.61 (0.35, 1.05) |
0.54 (0.31, 0.95)* |
1.38 (0.72, 2.62) |
1.29 (0.67, 2.46) |
| T3 |
0.72 (0.41, 1.27) |
0.71 (0.40, 1.26) |
0.99 (0.57, 1.71) |
0.93 (0.53, 1.63) |
1.03 (0.54, 1.92) |
0.97 (0.51, 1.86) |
| Secondary spatial variables | ||||||
| Population density | ||||||
| T2 |
1.23 (0.70, 2.17) |
1.22 (0.69, 2.15) |
1.13 (0.65, 1.94) |
1.07 (0.62, 1.84) |
1.44 (0.78, 2.67) |
1.39 (0.75, 2.56) |
| T3 |
1.70 (0.96, 3.02) |
1.58 (0.89, 2.82) |
0.90 (0.52, 1.55) |
0.81 (0.47, 1.40) |
1.94 (1.01, 3.73)* |
1.77 (0.92, 3.41) |
| Hilliness | ||||||
| T2 |
1.24 (0.71, 2.17) |
1.23 (0.71, 2.15) |
1.15 (0.68, 1.97) |
1.15 (0.67, 1.96) |
0.76 (0.40, 1.44) |
0.72 (0.38, 1.36) |
| T3 |
0.80 (0.45, 1.42) |
0.77 (0.44, 1.37) |
1.46 (0.84, 2.53) |
1.43 (0.82, 2.49) |
1.00 (0.54, 1.88) |
0.99 (0.53, 1.84) |
| Distance to nearest major road | ||||||
| T2 |
0.81 (0.47, 1.42) |
0.80 (0.46, 1.40) |
0.80 (0.46, 1.37) |
0.79 (0.46, 1.36) |
1.08 (0.57, 2.06) |
1.09 (0.58, 2.07) |
| T3 |
0.74 (0.42, 1.31) |
0.75 (0.42, 1.32) |
0.88 (0.51, 1.52) |
0.89 (0.52, 1.55) |
0.64 (0.34, 1.21) |
0.67 (0.35, 1.26) |
| Distance to nearest park | ||||||
| T2 |
1.27 (0.73, 2.21) |
1.29 (0.74, 2.24) |
0.91 (0.53, 1.57) |
0.94 (0.55, 1.61) |
1.23 (0.65, 2.31) |
1.25 (0.66, 2.34) |
| T3 |
0.78 (0.44, 1.38) |
0.84 (0.48, 1.49) |
0.64 (0.37, 1.10) |
0.73 (0.42, 1.25) |
0.72 (0.38, 1.34) |
0.80 (0.43, 1.50) |
| Distance to nearest physical activity facility | ||||||
| T2 |
0.98 (0.57, 1.69) |
0.94 (0.54, 1.61) |
0.81 (0.47, 1.38) |
0.77 (0.45, 1.32) |
0.88 (0.46, 1.66) |
0.84 (0.45, 1.59) |
| T3 |
0.51 (0.28, 0.90)* |
0.52 (0.29, 0.92)* |
0.70 (0.40, 1.21) |
0.74 (0.43, 1.28) |
0.69 (0.37, 1.28) |
0.72 (0.39, 1.35) |
| Intersection density | ||||||
| T2 |
1.36 (0.78, 2.35) |
1.34 (0.77, 2.32) |
1.15 (0.68, 1.96) |
1.09 (0.64, 1.85) |
1.31 (0.71, 2.41) |
1.27 0.69, 2.33) |
| T3 |
1.36 (0.78, 2.40) |
1.29 (0.73, 2.25) |
0.67 (0.39, 1.15) |
0.61 (0.36, 1.05) |
1.09 (0.58, 2.05) |
1.02 (0.54, 1.90) |
| Distance to nearest bus stop | ||||||
| T2 |
0.88 (0.51, 1.52) |
0.88 (0.51, 1.52) |
1.15 (0.67, 1.98) |
1.15 (0.67, 1.97) |
0.85 (0.45, 1.59) |
0.79 (0.42, 1.47) |
| T3 |
0.52 (0.29, 0.92)* |
0.56 (0.31, 1.00) |
0.88 (0.51, 1.51) |
0.98 (0.56, 1.71) |
0.47 (0.25, 0.90)* |
0.50 (0.26, 0.96)* |
Multivariable models controlled for mother’s age in years at the start of pregnancy, race, marital status, years of education, employment status, poverty status, parity, and median parcel tax value along the participant’s residential address road
Total physical activity operationalized as a median split, > 1.4 h/week vs. <= 1.4 h/week (ref.)
Abbreviations: OR, odds ratio; CI, confidence interval; T2, tertile 2; T3, tertile 3
Italicized values indicate statistical significance; *p < 0.05, **p < 0.01, ***p < 0.001
Table 4.
Multilevel mixed-effects logistic regression estimates, cross-sectional association between built environment variables and recreation physical activity
| 27–30 weeks’ gestation (n = 417) OR (95% CI) |
3 months postpartum (n = 439) OR (95% CI) |
12 months postpartum (n = 325) OR (95% CI) |
||||
|---|---|---|---|---|---|---|
| Bivariate | Multivariable | Bivariate | Multivariable | Bivariate | Multivariable | |
| Environmental audit variables | ||||||
| Arterial | ||||||
| T2 |
1.07 (0.62, 1.85) |
1.14 (0.67, 1.95) |
0.80 (0.47, 1.35) |
0.79 (0.48, 1.33) |
1.39 (0.77, 2.50) |
1.43 (0.80, 2.55) |
| T3 |
0.87 (0.50, 1.51) |
1.08 (0.63, 1.86) |
1.01 (0.60, 1.70) |
1.22 (0.72, 2.07) |
0.93 (0.50, 1.72) |
1.04 (0.57, 1.92) |
| Walkable neighborhood | ||||||
| T2 |
2.05 (1.16, 3.62)* |
1.71 (0.97, 3.01) |
2.06 (1.23, 3.47)** |
1.77 (1.05, 2.99)* |
1.11 (0.61, 2.02) |
0.99 (0.55, 1.80) |
| T3 |
2.70 (1.53, 4.78)** |
2.09 (0.19, 3.67)* |
2.07 (1.22, 3.51)** |
1.52 (0.90, 2.58) |
1.79 (0.97, 3.29) |
1.39 (0.76, 2.56) |
| Physical incivilities | ||||||
| T2 |
0.83 (0.48, 1.42) |
0.85 (0.50, 1.45) |
0.68 (0.40, 1.16) |
0.67 (0.40, 1.14) |
0.73 (0.41, 1.32) |
0.78 (0.44, 1.40) |
| T3 |
0.71 (0.41, 1.23) |
1.02 (0.58, 1.79) |
0.45 (0.27, 0.76)** |
0.71 (0.41, 1.23) |
0.67 (0.37, 1.24) |
0.99 (0.53, 1.85) |
| Decoration | ||||||
| T2 |
0.99 (0.57, 1.72) |
0.71 (0.41, 1.23) |
1.27 (0.75, 2.14) |
0.85 (0.50, 1.45) |
2.25 (1.21, 4.21)** |
1.70 (0.91, 3.18) |
| T3 |
1.01 (0.58, 1.74) |
0.76 (0.43, 1.32) |
1.88 (1.11, 3.19)* |
1.32 (0.77, 2.26) |
1.87 (1.01, 3.46)* |
1.44 (0.77, 2.68) |
| Secondary spatial variables | ||||||
| Population density | ||||||
| T2 |
1.51 (0.86, 2.63) |
1.26 (0.72, 2.18) |
1.02 (0.61, 1.72) |
0.81 (0.48, 1.36) |
1.41 (0.78, 2.54) |
1.25 (0.69, 2.24) |
| T3 |
1.81 (1.03, 3.17)* |
1.67 (0.95, 2.92) |
1.04 (0.62, 1.75) |
0.95 (0.56, 1.60) |
1.07 (0.58, 1.97) |
0.96 (0.52, 1.79) |
| Hilliness | ||||||
| T2 |
1.14 (0.65, 1.98) |
1.06 (0.61, 1.82) |
1.41 (0.84, 2.36) |
1.31 (0.78, 2.20) |
0.79 (0.43, 1.47) |
0.68 (0.37, 1.25) |
| T3 |
1.49 (0.85, 2.60) |
1.20 (0.69, 2.07) |
1.36 (0.80, 2.32) |
1.10 (0.65, 1.86) |
1.57 (0.86, 2.87) |
1.29 (0.71, 2.33) |
| Distance to nearest major road | ||||||
| T2 |
1.11 (0.64, 1.90) |
0.95 (0.55, 1.62) |
1.04 (0.61, 1.75) |
0.86 (0.51, 1.46) |
1.87 (1.01, 3.48)* |
1.70 (0.92, 3.14) |
| T3 |
0.90 (0.52, 1.58) |
0.77 (0.44, 1.33) |
0.91 (0.54, 1.54) |
0.74 (0.44, 1.25) |
1.21 (0.66, 2.24) |
1.13 (0.61, 2.07) |
| Distance to nearest park | ||||||
| T2 |
1.10 (0.64, 1.90) |
0.99 (0.58, 1.69) |
1.07 (0.63, 1.81) |
0.98 (0.58, 1.66) |
1.15 (0.63, 2.12) |
1.03 (0.57, 1.89) |
| T3 |
1.02 (0.59, 1.78) |
1.05 (0.61, 1.82) |
0.41 (0.24, 0.71)** |
0.45 (0.26, 0.77)** |
0.91 (0.49, 1.66) |
0.95 (0.52, 1.74) |
| Distance to nearest physical activity facility | ||||||
| T2 |
0.88 (0.52, 1.50) |
0.77 (0.45, 1.31) |
0.78 (0.46, 1.31) |
0.69 (0.41, 1.17) |
0.90 (0.49, 1.67) |
0.84 (0.45, 1.54) |
| T3 |
0.56 (0.32, 0.99)* |
0.58 (0.33, 1.02) |
0.46 (0.27, 0.79)** |
0.49 (0.29, 0.85)* |
0.71 (0.39, 1.29) |
0.81 (0.44, 1.47) |
| Intersection density | ||||||
| T2 |
1.18 (0.69, 2.01) |
1.11 (0.65, 1.89) |
1.50 (0.90, 2.50) |
1.35 (0.81, 2.25) |
1.06 (0.59, 1.90) |
1.00 (0.56, 1.78) |
| T3 |
1.15 (0.66, 1.98) |
1.07 (0.62, 1.83) |
0.94 (0.56, 1.57) |
0.83 (0.50, 1.39) |
0.95 (0.52, 1.74) |
0.91 (0.50, 1.66) |
| Distance to nearest bus stop | ||||||
| T2 |
0.89 (0.53, 1.52) |
0.80 (0.48, 1.36) |
1.15 (0.68, 1.94) |
0.99 (0.59, 1.67) |
1.52 (0.84, 2.77) |
1.25 (0.69, 2.25) |
| T3 |
0.60 (0.35, 1.05) |
0.78 (0.45, 1.36) |
0.47 (0.28, 0.79)** |
0.64 (0.37, 1.09) |
0.81 (0.44, 1.49) |
1.04 (0.56, 1.91) |
Multivariable models controlled for mother’s age in years at the start of pregnancy, race, marital status, years of education, employment status, poverty status, parity, and median parcel tax value along the participant’s residential address road
Recreation physical activity operationalized as any vs. none (ref)
Abbreviations: OR, odds ratio; CI, confidence interval; T2, tertile 2; T3, tertile 3
Italicized values indicate statistical significance; *p < 0.05, **p < 0.01, ***p < 0.001
For recreation physical activity, in multivariable analysis, having a high level of neighborhood walkability was positively associated with the odds of reporting high total physical activity at 27–30 weeks’ gestation (T3 OR 2.09; 95% CI 1.09, 3.67), and a moderate level of walkability was positively associated at 3 months postpartum (T2 OR 1.77; 95% CI 1.05, 2.99) (Table 4). Living the furthest distance from the nearest park was inversely associated with the odds of reporting recreation physical activity at 3 months postpartum (T3 OR 0.45; 95% CI 0.26, 0.77), as was living the furthest distance from the nearest physical activity facility (T3 OR 0.49; 95% CI 0.29, 0.85) (Table 4).
Longitudinal Associations
In multivariable models, no objectively measured neighborhood-based built environment variables were associated with increased total physical activity (Table 5). For recreation physical activity, in multivariable models, a moderate level of an arterial road network was positively associated with increased recreation physical activity from 3 months postpartum to 12 months postpartum (T2 OR 1.91, 95% CI 1.00, 3.68) (Table 5). A moderate distance to nearest major road was positively associated with increased recreation physical activity from 3 months postpartum to 12 months postpartum (T2 OR 2.13, 95% CI 1.08, 4.22) (Table 5). Finally, living the furthest distance from the nearest park was inversely associated with an increase in recreation physical activity from pregnancy to 3 months postpartum (T3 OR 0.50, 95% CI 0.29, 0.85) (Table 5).
Table 5.
Multilevel mixed-effects logistic regression estimates, association between built environment variables and increase in physical activity
| Total physical activity hours/week | Recreation physical activity hours/week | |||||||
|---|---|---|---|---|---|---|---|---|
| Increase from pregnancy to 3 months postpartum OR (95% CI) |
Increase from 3 months postpartum to 12 months postpartum OR (95% CI) |
Increase from pregnancy to 3 months postpartum OR (95% CI) |
Increase from 3 months postpartum to 12 months postpartum OR (95% CI) |
|||||
| Bivariate | Multivariable | Bivariate | Multivariable | Bivariate | Multivariable | Bivariate | Multivariable | |
| Environmental audit variables | ||||||||
| Arterial | ||||||||
| T2 |
1.00 (0.62, 1.61) |
0.99 (0.61, 1.60) |
0.89 (0.50, 1.57) |
0.89 (0.50, 1.59) |
0.97 (0.59, 1.59) |
0.94 (0.57, 1.56) |
1.95 (1.03, 3.72)* |
1.91 (1.00, 3.68)* |
| T3 |
1.21 (0.76, 1.93) |
1.20 (0.74, 1.95) |
1.14 (0.64, 2.03) |
1.13 (0.62, 2.04) |
0.96 (0.58, 1.57) |
0.98 (0.59, 1.64) |
1.59 (0.80, 3.13) |
1.57 (0.78, 3.13) |
| Walkable neighborhood | ||||||||
| T2 |
0.93 (0.59, 1.49) |
0.87 (0.54, 1.40) |
0.95 (0.53, 1.71) |
0.90 (0.50, 1.64) |
1.59 (0.97, 2.61) |
1.44 (0.85, 2.40) |
0.98 (0.50, 1.90) |
0.88 (0.45, 1.75) |
| T3 |
0.98 (0.61, 1.57) |
0.88 (0.54, 1.44) |
1.32 (0.74, 2.35) |
1.27 (0.71, 2.30) |
1.21 (0.73, 2.03) |
1.02 (0.60, 1.73) |
1.66 (0.88, 3.11) |
1.49 (0.78, 2.85) |
| Physical incivilities | ||||||||
| T2 |
1.07 (0.67, 1.72) |
1.13 (0.70, 1.83) |
1.03 (0.58, 1.82) |
1.06 (0.60, 1.90) |
1.01 (0.62, 1.65) |
1.05 (0.64, 1.74) |
1.17 (0.63, 2.14) |
1.24 (0.66, 2.30) |
| T3 |
1.00 (0.63, 1.61) |
1.15 (0.69, 1.90) |
1.05 (0.58, 1.88) |
1.16 (0.63, 2.13) |
0.73 (0.45, 1.21) |
0.93 (0.54, 1.61) |
0.76 (0.39, 1.48) |
0.89 (0.45, 1.79) |
| Decoration | ||||||||
| T2 |
0.96 (0.60, 1.54) |
0.88 (0.54, 1.43) |
1.54 (0.85, 2.78) |
1.43 (0.78, 2.62) |
1.06 (0.64, 1.76) |
0.89 (0.52, 1.51) |
2.14 (1.09, 4.20)* |
1.92 (0.96, 3.85) |
| T3 |
0.89 (0.56, 1.43) |
0.83 (0.50, 1.36) |
1.03 (0.56, 1.88) |
0.95 (0.51, 1.78) |
1.42 (0.86, 2.32) |
1.20 (0.71, 2.03) |
1.31 (0.65, 2.63) |
1.16 (0.56, 2.39) |
| Secondary spatial variables | ||||||||
| Population density | ||||||||
| T2 |
1.01 (0.93, 2.96) |
1.02 (0.63, 1.65) |
0.81 (0.45, 1.47) |
0.80 (0.44, 1.48) |
1.03 (0.64, 1.68) |
0.94 (0.57, 1.56) |
1.14 (0.61, 2.12) |
1.08 (0.57, 2.04) |
| T3 |
0.79 (0.49, 1.26) |
0.80 (0.49, 1.30) |
1.66 (0.93, 2.96) |
1.75 (0.96, 3.18) |
0.67 (0.40, 1.10) |
0.64 (0.38, 1.08) |
0.96 (0.50, 1.87) |
0.95 (0.48, 1.88) |
| Hilliness | ||||||||
| T2 |
1.02 (0.63, 1.63) |
1.02 (0.63, 1.64) |
0.79 (0.44, 1.45) |
0.79 (0.43, 1.44) |
1.03 (0.63, 1.68) |
1.01 (0.61, 1.68) |
0.53 (0.30, 1.03) |
0.51 (0.26, 1.00) |
| T3 |
1.24 (0.77, 1.99) |
1.21 (0.74, 1.96) |
1.02 (0.58, 1.79) |
1.03 (0.58, 1.82) |
1.13 (0.69, 1.87) |
1.03 (0.62, 1.73) |
0.75 (0.41, 1.38) |
0.71 (0.40, 1.32) |
| Distance to nearest major road | ||||||||
| T2 |
0.94 (0.59, 1.50) |
0.88 (0.54, 1.43) |
1.64 (0.91, 2.98) |
1.52 (0.83, 2.80) |
1.16 (0.70, 1.92) |
1.05 (0.62, 1.76) |
2.36 (1.21, 4.60)* |
2.13 (1.08, 4.22)* |
| T3 |
0.94 (0.59, 1.51) |
0.90 (0.56, 1.47) |
1.21 (0.66, 2.21) |
1.15 (0.62, 2.14) |
1.21 (0.74, 1.99) |
1.09 (0.65, 1.83) |
1.24 (0.61, 2.52) |
1.15 (0.56, 2.38) |
| Distance to nearest park | ||||||||
| T2 |
1.08 (0.70, 1.71) |
1.12 (0.70, 1.80) |
1.15 (0.64, 2.08) |
1.23 (0.67, 2.25) |
0.74 (0.46, 1.20) |
0.73 (0.44, 1.20) |
1.72 (0.88, 3.39) |
1.79 (0.90, 3.58) |
| T3 |
0.59 (0.37, 0.96)* |
0.61 (0.37, 1.00) |
1.29 (0.72, 2.32) |
1.33 (0.73, 2.42) |
0.48 (0.29, 0.80)** |
0.50 (0.29, 0.85)* |
1.87 (0.96, 3.66) |
2.00 (0.99, 3.98) |
| Distance to nearest physical activity facility | ||||||||
| T2 |
0.65 (0.41, 1.03) |
0.62 (0.39, 1.01) |
1.11 (0.61, 2.03) |
1.07 (0.58, 1.97) |
0.96 (0.59, 1.56) |
0.92 (0.56, 1.52) |
0.68 (0.34, 1.35) |
0.67 (0.33, 1.36) |
| T3 |
0.74 (0.46, 1.19) |
0.74 (0.45, 1.19) |
1.30 (0.73,2.30) |
1.33 (0.74, 2.38) |
0.79 (0.48, 1.30) |
0.81 (0.48, 1.37) |
1.24 (0.66, 2.28) |
1.40 (0.74, 2.61) |
| Intersection density | ||||||||
| T2 |
0.84 (0.53, 1.33) |
0.86 (0.53, 1.37) |
0.89 (0.51, 1.57) |
0.89 (0.50, 1.58) |
1.00 (0.62, 1.61) |
0.96 (0.59, 1.58) |
0.85 (0.47, 1.56) |
0.81 (0.44, 1.51) |
| T3 |
0.69 (0.43, 1.11) |
0.65 (0.40, 1.06) |
1.06 (0.60, 1.90) |
1.03 (0.57, 1.87) |
0.69 (0.41, 1.14) |
0.62 (0.37, 1.04) |
0.64 (0.33, 1.25) |
0.59 (0.30, 1.17) |
| Distance to nearest bus stop | ||||||||
| T2 |
1.25 (0.79, 2.00) |
1.17 (0.73, 1.89) |
0.96 (0.54, 1.72) |
0.91 (0.51, 1.65) |
1.11 (0.68, 1.81) |
0.97 (0.59, 1.61) |
1.43 (0.74, 2.79) |
1.28 (0.65, 2.52) |
| T3 |
0.82 (0.51, 1.32) |
0.86 (0.52, 1.43) |
1.05 (0.58, 1.89) |
1.10 (0.60, 2.02) |
0.86 (0.52, 1.42) |
1.04 (0.61, 1.76) |
1.70 (0.87, 3.30) |
1.98 (0.99, 3.94) |
Multivariable models controlled for mother’s age in years at the start of pregnancy, race, marital status, years of education, employment status, poverty status, parity, and median parcel tax value along the participant’s residential address road segment
Italicized values indicate statistical significance; *p < 0.05, **p < 0.01, ***p < 0.001
Abbreviations: OR, odds ratio; CI, confidence interval; T2, tertile 2; T3, tertile 3
Discussion
This study explored the association between objectively measured neighborhood-based built environment variables and total and recreation physical activity from pregnancy to postpartum. Several environmental variables explored were associated with total and recreation physical activity and during pregnancy, 3 months, and 12 months postpartum. Additionally, associations between environmental variables and an increase in recreation and total physical activity from pregnancy to postpartum were identified.
In cross-sectional analyses, more favorable walkability was found to be associated with both total and recreational physical activity during pregnancy, and partially with recreational physical activity at 3-months postpartum. Further, as expected, distance to the nearest bus stop was inversely associated with total physical activity at 12 months postpartum. However, no association was identified between either walkability or distance to the nearest bus stop and increase in physical activity from pregnancy to postpartum. Components of walkability, including access to and availability of public transit, have been identified as important for physical activity in adults generally [15], and results here indicate that they may be important for physical activity in pregnant and postpartum women as well. In this study, walkability was assessed using the PIN3 environmental audit tool, and included 11 separately measured items such as visible/active child/youth, presence of trails, and pavement markings, crosswalk [26].
Environments that support individuals to be out in their neighborhood being physically active are surely important for all persons regardless of age or condition but may be particularly salient for women during the pregnancy and postpartum period. It is important to consider the needs of the most vulnerable populations, including pregnant and postpartum women, when considering mobility alternatives. In terms of public transportation, it is important to ensure that those most in need have some basic level of access by considering how these users access transit systems and their particular needs for accommodations at stops and on vehicles. Transportation physical activity was not independently assessed for the purposes of this analysis due to low reported participation [28], but it is possible that the association between distance to the nearest bus stop and total physical activity is driven by transportation-based physical activity during pregnancy, particularly because more women were working at this time. Conversely, increased distance to transit may indicate less urbanicity and increased social isolation, factors that can contribute to lower physical activity [29].
Distance to the nearest park was found to be associated with a decreased odds of recreational physical activity at 3 months postpartum, and also with increase in recreational physical activity from pregnancy to 3 months postpartum. Further, increased distance to the nearest physical activity facility was associated with a decreased odds of total physical activity during pregnancy, and of recreational physical activity 12 months postpartum. Comparably, Nichani and colleagues [19] found no association between proportion of green space within a census unit and physical activity in a population of pregnant women living in New Zealand. One difference between these two studies is that Nichani and colleagues’ measure of greenspace included blue spaces (beaches) and areas which may or may not be suitable for physical activity [19]. It is also possible that access to parks and recreation areas are more essential to women after giving birth and in that transition from pregnancy to postpartum, rather than during pregnancy when they may feel less comfortable being pregnant and active outside or in recreation facilities. It is also possible that parks and recreation areas are places where new mothers feel comfortable visiting with a baby.
In contrast, Richardsen and colleagues [18] explored the association between access to recreational spaces and moderate to vigorous physical activity in pregnant and postpartum women living in Oslo, Norway, and found an association during pregnancy, but not postpartum. These differential findings between the work by Richardsen and colleagues and the current study may be in part due to their particular consideration of access to these recreation areas, which included consideration of an eligible walking route, whereas the current study examined distance only [18]. Further, Richardsen and colleagues examined objectively measured total moderate to vigorous physical activity [18], while the present study examined self-reported total and recreation physical activity regardless of intensity. More generally, the association between proximity to parks and measures of physical activity has been largely mixed, with few studies finding an association using objective measures of the environment [30].
Other environmental factors found to be associated with increase in recreational physical activity from 3 to 12 months postpartum included an arterial road network (increased odds of increasing recreational physical activity) and distance to major roads (increased odds of increasing recreational physical activity). Having a neighborhood that allows for multiple route options with lower traffic speeds may be particularly conducive to recreation physical activity in postpartum women. Women report being encouraged to walk during pregnancy [31], and it may be a feasible activity for women to return to once giving birth, particularly in neighborhoods with these characteristics.
Many of the environmental factors explored here were not associated with physical activity during pregnancy and postpartum, despite evidence of their association with physical activity in other populations [16]. Pregnancy causes substantial change to a woman’s body, particularly during the third trimester and after birth, timeframes examined in this study. Indeed, women often report that physical discomfort and fatigue associated with fetal development prohibit them from being physically active, particularly during the third trimester, which may be a contributing reason for decreasing physical activity regardless of many aspects of the neighborhood environment [32]. However, physical activity is recommended for women throughout pregnancy and the postpartum period in the absence of medical or surgical complications [33], and the results of this study suggest that certain environmental factors in the neighborhood, such as walkability and access to recreation areas, can help support women in pursuing physical activity during this pivotal time.
Strengths and Limitations
As far as the authors are aware, this is the first study to examine environmental characteristics measured via an environmental audit and secondary spatial data in relation to physical activity in pregnancy and postpartum both cross-sectionally and longitudinally. One strength of this study is the use of a large prospective cohort, which allowed for the examination of both cross-sectional and longitudinal change in physical activity, leading to a more robust understanding of the association between aspects of the built environment and physical activity during pregnancy and postpartum. Another strength is the use of objective measures of the built environment. Further, the use of diverse methods to assess the built environment, including neighborhood auditing and secondary data sources, allows for a more comprehensive assessment of the environment at both the micro- and macro-scale.
This study does have limitations. Participation in the PIN Postpartum Study was lower than PIN3 since not all of the initial participants were eligible to participate due to timing of birth, subsequent pregnancy, or moving out of the geographic area [22]. The majority of participants of this study were white, married, and financially stable, so generalizability of these findings may be limited. Physical activity, while measured comprehensively, was assessed via self-report, which has inherent bias. Exploration of transportation physical activity is warranted in any exploration of the built environment and physical activity, but volume of transportation physical activity reported by participants in this study was limited [28]. Information on the environmental setting in which the various types of physical activity took place was not collected as part of the PIN Postpartum Study, which limits the predictive ability and interpretability of models [34]. Neighborhood self-selection was not considered in the PIN Postpartum Study, so reverse causation is a possibility. In some cases, the associations were in the middle group—such that there was no dose-response relationships. This supports that, at least in some cases, associations may be by chance because of multiple statistical testing. The use of secondary spatial data has inherent limitations, including questions of data accuracy and accessibility. In particular, parcel information was not available for the entire study area, so potentially important measures such as access to destinations or land use mix were not assessed in this analysis. More studies with larger sample sizes in diverse populations are needed to further explore these associations.
Conclusions
This study examined the association between environmental audit and spatially derived neighborhood measures and self-reported physical activity during late pregnancy and postpartum, as well as assessing the association with longitudinal change in physical activity from pregnancy to postpartum. Taken together, the findings of this study indicate that several aspects of the environment are associated with physical activity in women during pregnancy and postpartum. These associations may result in a number of benefits to both the woman and child, including postpartum measures of adiposity [35] and depression [36] and fetal growth and birth weight [37]. However, many of the built environment measures examined in this study were not associated with physical activity during and across pregnancy and postpartum, so it is important to consider what other factors, in the context of the built environment, can encourage and support pregnant and postpartum women to be physically active.
Acknowledgments
The PIN Study was a joint effort of many investigators and staff members whose work is gratefully acknowledged. In particular, the authors thank Molly Wen and Amy Herring for their invaluable assistance with preparing the data and advising on analyses.
Funding
Funding for this study was provided by the National Institutes of Health (NIH)/National Cancer Institute (#CA109804). Data collection was supported by NIH/National Institute of Child Health and Human Development (#HD37584), NIH General Clinical Research Center (#RR00046), and NIH/National Institute of Diabetes and Digestive and Kidney Diseases (#DK 061981). AKP was supported by a National Research Service Award T32 post-doctoral research fellowship, funded by the NIH, National Heart, Lung, and Blood Institute (#T32HL007055). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Nascimento SL, Surita FG, Cecatti JG. Physical exercise during pregnancy: a systematic review. Curr Opin Obstet Gynecol. 2012;24(6):387–394. doi: 10.1097/GCO.0b013e328359f131. [DOI] [PubMed] [Google Scholar]
- 2.Committee PAGA. Physical activity guidelines advisory committee scientific report. In: Services USDoHaH, ed. Washington, DC 2018.
- 3.Evenson KR, Barakat R, Brown WJ, Dargent-Molina P, Haruna M, Mikkelsen EM, Mottola MF, Owe KM, Rousham EK, Yeo SA. Guidelines for physical activity during pregnancy: comparisons from around the world. Am J Lifestyle Med. 2014;8(2):102–121. doi: 10.1177/1559827613498204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bø K, Artal R, Barakat R, Brown W, Davies GAL, Dooley M, Evenson KR, Haakstad LAH, Henriksson-Larsen K, Kayser B, Kinnunen TI, Mottola MF, Nygaard I, van Poppel M, Stuge B, Khan KM. Exercise and pregnancy in recreational and elite athletes: 2016 evidence summary from the IOC expert group meeting, Lausanne. Part 1-exercise in women planning pregnancy and those who are pregnant. Br J Sports Med. 2016;50(10):571–589. doi: 10.1136/bjsports-2016-096218. [DOI] [PubMed] [Google Scholar]
- 5.Evenson KR, Mottola MF, Owe KM, Rousham EK, Brown WJ. Summary of international guidelines for physical activity after pregnancy. Obstet Gynecol Surv. 2014;69(7):407–414. doi: 10.1097/OGX.0000000000000077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bø K, Artal R, Barakat R, Brown WJ, Davies GAL, Dooley M, Evenson KR, Haakstad LAH, Kayser B, Kinnunen TI, Larsén K, Mottola MF, Nygaard I, van Poppel M, Stuge B, Khan KM. Exercise and pregnancy in recreational and elite athletes: 2016/17 evidence summary from the IOC expert group meeting, Lausanne. Part 3-exercise in the postpartum period. Br J Sports Med. 2017;51(21):1516–1525. doi: 10.1136/bjsports-2017-097964. [DOI] [PubMed] [Google Scholar]
- 7.Bacchi E, Bonin C, Zanolin ME, Zambotti F, Livornese D, Donà S, Tosi F, Baldisser G, Ihnatava T, di Sarra D, Bonora E, Moghetti P. Physical activity patterns in normal-weight and overweight/obese pregnant women. PLoS One. 2016;11(11):e0166254. doi: 10.1371/journal.pone.0166254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rousham EK, Clarke PE, Gross H. Significant changes in physical activity among pregnant women in the UK as assessed by accelerometry and self-reported activity. Eur J Clin Nutr. 2006;60(3):393–400. doi: 10.1038/sj.ejcn.1602329. [DOI] [PubMed] [Google Scholar]
- 9.Liu J, Blair SN, Teng Y, Ness AR, Lawlor DA, Riddoch C. Physical activity during pregnancy in a prospective cohort of British women: results from the Avon longitudinal study of parents and children. Eur J Epidemiol. 2011;26(3):237–247. doi: 10.1007/s10654-010-9538-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Symons Downs D, Hausenblas HA. Women’s exercise beliefs and behaviors during their pregnancy and postpartum. J Midwifery Womens Health. 2004;49(2):138–144. doi: 10.1016/j.jmwh.2003.11.009. [DOI] [PubMed] [Google Scholar]
- 11.Borodulin K, Evenson KR, Herring AH. Physical activity patterns during pregnancy through postpartum. BMC Womens Health. 2009;9:32. doi: 10.1186/1472-6874-9-32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schmitt NM, Nicholson WK, Schmitt J. The association of pregnancy and the development of obesity—results of a systematic review and meta-analysis on the natural history of postpartum weight retention. Int J Obes. 2007;31(11):1642–1651. doi: 10.1038/sj.ijo.0803655. [DOI] [PubMed] [Google Scholar]
- 13.Nascimento SL, Pudwell J, Surita FG, Adamo KB, Smith GN. The effect of physical exercise strategies on weight loss in postpartum women: a systematic review and meta-analysis. Int J Obes. 2014;38(5):626–635. doi: 10.1038/ijo.2013.183. [DOI] [PubMed] [Google Scholar]
- 14.Sallis JF, Cervero RB, Ascher W, Henderson KA, Kraft MK, Kerr J. An ecological approach to creating active living communities. Annu Rev Public Health. 2006;27:297–322. doi: 10.1146/annurev.publhealth.27.021405.102100. [DOI] [PubMed] [Google Scholar]
- 15.Smith M, Hosking J, Woodward A, et al. Systematic literature review of built environment effects on physical activity and active transport—an update and new findings on health equity. Int J Behav Nutr Phys Act. 2017;11(1):14–158. doi: 10.1186/s12966-017-0613-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McCormack GR, Shiell A. In search of causality: a systematic review of the relationship between the built environment and physical activity among adults. Int J Behav Nutr Phys Act. 2011;8:125. doi: 10.1186/1479-5868-8-125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.O'Brien OA, Lindsay KL, McCarthy M, et al. Influences on the food choices and physical activity behaviours of overweight and obese pregnant women: a qualitative study. Midwifery. 2017;47:28–35. doi: 10.1016/j.midw.2017.02.003. [DOI] [PubMed] [Google Scholar]
- 18.Richardsen KR, Mdala I, Berntsen S, et al. Objectively recorded physical activity in pregnancy and postpartum in a multi-ethnic cohort: association with access to recreational areas in the neighbourhood. Int J Behav Nutr Phys Act. 2016;13:78. doi: 10.1186/s12966-016-0401-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Nichani V, Dirks K, Burns B, Bird A, Morton S, Grant C. Green space and physical activity in pregnant women: evidence from the growing up in New Zealand study. J Phys Act Health. 2016;13(12):1341–1350. doi: 10.1123/jpah.2016-0013. [DOI] [PubMed] [Google Scholar]
- 20.Laraia B, Messer L, Evenson K, Kaufman JS. Neighborhood factors associated with physical activity and adequacy of weight gain during pregnancy. J Urban Health. 2007;84(6):793–806. doi: 10.1007/s11524-007-9217-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Demissie Z, Siega-Riz AM, Evenson KR, Herring AH, Dole N, Gaynes BN. Physical activity and depressive symptoms among pregnant women: the PIN3 study. Arch Womens Ment Health. 2011;14(2):145–157. doi: 10.1007/s00737-010-0193-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Siega-Riz AM, Herring AH, Carrier K, Evenson KR, Dole N, Deierlein A. Sociodemographic, perinatal, behavioral, and psychosocial predictors of weight retention at 3 and 12 months postpartum. Obesity (Silver Spring) 2010;18(10):1996–2003. doi: 10.1038/oby.2009.458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Borodulin KM, Evenson KR, Wen F, Herring AH, Benson AM. Physical activity patterns during pregnancy. Med Sci Sports Exerc. 2008;40(11):1901–1908. doi: 10.1249/MSS.0b013e31817f1957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Evenson KR, Wen F. Measuring physical activity among pregnant women using a structured one-week recall questionnaire: evidence for validity and reliability. Int J Behav Nutr Phys Act. 2010;7:21. doi: 10.1186/1479-5868-7-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Frizzelle BG, Evenson KR, Rodriguez DA, Laraia BA. The importance of accurate road data for spatial applications in public health: customizing a road network. Int J Health Geogr. 2009;8:24. doi: 10.1186/1476-072X-8-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Evenson KR, Sotres-Alvarez D, Herring AH, Messer L, Laraia BA, Rodríguez DA. Assessing urban and rural neighborhood characteristics using audit and GIS data: derivation and reliability of constructs. Int J Behav Nutr Phys Act. 2009;6:44. doi: 10.1186/1479-5868-6-44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Porter AK, Wen F, Herring AH, Rodríguez DA, Messer LC, Laraia BA, Evenson KR. Reliability and one-year stability of the PIN3 neighborhood environmental audit in urban and rural neighborhoods. J Urban Health. 2018;95(3):431–439. doi: 10.1007/s11524-018-0243-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Evenson KR, Herring AH, Wen F. Self-reported and objectively measured physical activity among a cohort of postpartum women: the PIN Postpartum Study. J Phys Act Health. 2012;9(1):5–20. doi: 10.1123/jpah.9.1.5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Josey MJ, Moore S. The influence of social networks and the built environment on physical inactivity: a longitudinal study of urban-dwelling adults. Health Place. 2018;54:62–68. doi: 10.1016/j.healthplace.2018.08.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bancroft C, Joshi S, Rundle A, Hutson M, Chong C, Weiss CC, Genkinger J, Neckerman K, Lovasi G. Association of proximity and density of parks and objectively measured physical activity in the United States: a systematic review. Soc Sci Med. 2015;138:22–30. doi: 10.1016/j.socscimed.2015.05.034. [DOI] [PubMed] [Google Scholar]
- 31.Ferrari RM, Siega-Riz AM, Evenson KR, Moos MK, Carrier KS. A qualitative study of women’s perceptions of provider advice about diet and physical activity during pregnancy. Patient Educ Couns. 2013;91(3):372–377. doi: 10.1016/j.pec.2013.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Duncombe D, Wertheim EH, Skouteris H, Paxton SJ, Kelly L. Factors related to exercise over the course of pregnancy including women’s beliefs about the safety of exercise during pregnancy. Midwifery. 2009;25(4):430–438. doi: 10.1016/j.midw.2007.03.002. [DOI] [PubMed] [Google Scholar]
- 33.ACOG ACOG Committee Opinion No. 650: Physical activity and exercise during pregnancy and the postpartum period. Obstet Gynecol. 2015;126(6):e135–e142. doi: 10.1097/AOG.0000000000001214. [DOI] [PubMed] [Google Scholar]
- 34.Giles-Corti B, Timperio A, Bull F, Pikora T. Understanding physical activity environmental correlates: increased specificity for ecological models. Exerc Sport Sci Rev. 2005;33(4):175–181. doi: 10.1097/00003677-200510000-00005. [DOI] [PubMed] [Google Scholar]
- 35.Keller C, Todd M, Ainsworth B, et al. Overweight, obesity, and neighborhood characteristics among postpartum Latinas. J Obes. 2013;2013:916468. doi: 10.1155/2013/916468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Saligheh M, Hackett D, Boyce P, Cobley S. Can exercise or physical activity help improve postnatal depression and weight loss? A systematic review. Arch Womens Ment Health. 2017;20(5):595–611. doi: 10.1007/s00737-017-0750-9. [DOI] [PubMed] [Google Scholar]
- 37.Dadvand P, Wright J, Martinez D, Basagaña X, McEachan RRC, Cirach M, Gidlow CJ, de Hoogh K, Gražulevičienė R, Nieuwenhuijsen MJ. Inequality, green spaces, and pregnant women: roles of ethnicity and individual and neighbourhood socioeconomic status. Environ Int. 2014;71:101–108. doi: 10.1016/j.envint.2014.06.010. [DOI] [PubMed] [Google Scholar]