For this cross-sectional study, we will first identify low, medium, and high walkability areas of different size cities. Afterwards, we will randomly select a sample of observation routes (1000 foot long street segments) from each walkability and city strata. The BWM will then be conducted along each observation route on 2 different days and at 6 different times. A total of 2 observers will perform the BWM simultaneously. A total of 1 observer will follow the traditional BWM procedures, whereas the other walks side-by-side with this observer and records video using the WVD. Later, 2 investigators will review the videos and, based on the BWM criteria for counting individuals, derive independent counts of individuals being physically active on sidewalks and streets. Comparative analyses will be conducted to determine the equivalence of the 2 approaches.

We are stratifying our sample to observe PAs occurring along sidewalks and streets given a wide range of conditions related to city size and walkability. We selected 3 cities: West Chester, Pennsylvania; Wilmington, Delaware; and Philadelphia, Pennsylvania that are small, medium, and large in terms of population, respectively (Multimedia Appendix 1) [45]. City size was considered a strata because it is associated with factors (eg, local norms, population density, and types of destinations) that could influence how humans use sidewalks and streets. In essence, our goal is to increase the generalizability of this study’s outcomes.

Drawing from our familiarity with the study cities and examinations of aerial maps, we will identify 3 neighborhoods per city we estimate as being low, medium, and high walkability. This is being done to streamline the process because there are 44, 92, and 160 defined neighborhoods in West Chester, Wilmington, and Philadelphia, respectively. Afterwards, we will actually measure walkability for each selected neighborhood using WalkScore. As WalkScores can vary across neighborhoods, we will base a neighborhood’s WalkScore on the average of WalkScores for 10 randomly selected addresses drawn from a list of all addresses in the neighborhood. This process will be repeated until 1 low (WalkScore ≤33), 1 medium (WalkScore 33 to ≤66), and 1 high (WalkScore >66) walkable neighborhood is located in each city giving us a total of 9 neighborhoods. We are using WalkScore because it is a valid measure for estimating walkability [46-49]. It is significantly correlated with geographical information system–derived indicators of neighborhood walkability such as the availability of retail destinations, intersection density, amenities, street connectivity, residential density, and access to public transit provisions [46-48]. In addition, a higher Walk Score is significantly associated with minutes/week of transport and leisure walking independent of sociodemographic and health variables [49]. WalkScore uses publicly available data from various sources (Google, Open Street Map, and Localeze) and an algorithm to assign a score to a location based on the straight-line distance to various categories of amenities (eg, schools, stores, parks, and libraries) weighted by facility type priority and a distance decay function [50]. The result is a walkability score between 0 and 100, with 0 being the least walkable and 100 being the most walkable. The location can be entered as geographic coordinates or as an address which is then geolocated using Google Geolocation [50].

The total linear length of sidewalks and streets in the 9 neighborhoods will be estimated using the ruler tool in Google Earth (a geobrowser that accesses satellite, aerial imagery, and other geographic data to represent the Earth as a 3-dimensional globe). The ruler tool is a geographical information systems-based application with submeter resolution. We have found the ruler tool to be accurate to within ±1.5% for measuring street segment lengths. Based on our previous work, we expect an average of 180,000 total linear ft. of sidewalks and streets per neighborhood [32-34]. The total linear ft./neighborhood will be divided into 1000 ft. routes, and a sample of these routes representing 20% of the total number of routes in a neighborhood will be randomly selected for study, which is an adequate percentage to obtain a representative sample [51]. Given our expectations, this would equate to an average of 36 observation routes per neighborhood or a total of 324 observation routes.

Each observation route in a neighborhood will be observed 3 times on a weekday and 3 times on a weekend day, which will give us a stable estimate of the outcome variable [52]. Each observation period will last 10 min and occur during each of the following time periods: 8 to 9 am, 12 to 1 pm, and 5 to 6 pm (note: all observations will occur during daylight hours). We will be able to complete observations of approximately 6 to 8 observation routes per week meaning a total of 5 to 6 weeks will be needed to assess all observation routes in a given neighborhood. To reduce ordering and seasonal effects, observations will be conducted in only 1 neighborhood per day with each day randomly selected from the pool of days available for the 12-month period when the BWM will be conducted. To reduce the effects of ordering within a time period, observation routes will be numbered consecutively and then placed in a random order for the observation schedule. Observations will not be conducted on days having an event that would affect counts (eg, parade and marathon) or during times when it is raining or snowing.

During an observation period, 2 trained observers (1 wearing a WVD and the other not wearing a WVD) will traverse an observation route at a pace of 100 ft/min (50 steps/min [largo]; stride width 2 ft; pace set by a battery-powered metronome). The observer without the WVD will record the number of individuals engaging in the targeted activities within an observation field. The observation field will be defined as a line extending to the left and right of the observer’s shoulders, linear and perpendicular from the observer’s plane of motion. The observation fields are expected to range in width from 30 to 70 ft. and include both sidewalks (if present) and the streets associated with an observation route. Individuals will be counted only if they cross a parallel plane of motion with the observer (Figure 1). For example, individuals walking down the sidewalk toward the observer (from ahead or from behind the observer) are counted if they continue to walk past the observer. An individual will be counted only once in an observation route on a given day of observation. When an observer encounters a street intersecting an observation route being observed, they will cease observing, cross the street, and then resume observations. An observation recording instrument was previously developed specifically for the procedure [32]. The instrument was designed so that an observer could record the PA observed, the street name where the PA occurred, and the number of individuals engaged in the PA. The instruments will contain information specific to each neighborhood including detailed walking directions and a map (note: We decided not to use a single observer to conduct the BWM while wearing the WVD because the BWM requires an observer to look away from the observation field while entering data on the BWM instrument, and we have found this to be a source of error especially with larger groups. This is a deficit we expect the WVD to rectify).

The primary outcome variable for aim 1 is the number of individuals observed walking, cycling, running, and standing/sitting along each observation route/50 min of observation.

The 2 study’s principal investigators will conduct independent evaluations of the videos obtained during the BWM. This will be done over a 1-year period beginning after the first week of BWMs are completed. They will use the BWM criteria to count individuals walking, cycling, running, and sitting/standing on sidewalks and streets along the observation routes.

All observers will participate in 2 training sessions before beginning data collection. During the first training session, they will be given detailed instructions on the BWM and procedures to be used. The second training session will involve mock field observations.

Data on meteorological conditions (rainfall, relative humidity, temperature, wind speed, and barometric pressure) for the exact time of day observations are conducted will be obtained from an automated weather sensor system located at the local airport.

The Pivothead Smart (Pivothead, Denver, CO) is a state-of-the-art, noninvasive WVD indistinguishable from a pair of normal sunglasses (Figures 2 and 3). The camera is discretely centered in the bridge of the glasses for the truest first-person perspective possible, and it features an 8 MP Sony complementary metal-oxide-semiconductor sensor for capturing full 1080 p high definition 4 mega-pixel video at 30 frames per second as well as 8 mega-pixel stills (Figure 4). The glasses accept a 32 GB memory card allowing up to 8 hours of video recording per card at 1080 p. They have a self-contained battery providing 6 to 8 hours of recording time; a 77 degree field of vision, which approximates the human 90 degree field of vision; and they can be fitted with polarized prescription lenses. The Pivothead also allows for audio recordings (helpful for obtaining auxiliary information), time and date stamp, and geolocation capabilities, which can be used to create and retain precise maps of the observation routes.

We will use the videos of observation routes assessed in aim 1.

The WVD video data, along with annotated ground truth for each human and feature of interest, will be analyzed automatically using multiple deep CNNs. The first deep CNN will be used in collaboration with the Simple, Online, and Real-time Tracking algorithms to determine the number of humans in BWM videos who cross the path of the observer (criteria for being counted) and the distance they traveled per unit time before crossing paths with the observer. For each human in the video, a bounding box will be drawn around their pixels, with identifying information such as faces blurred automatically. Once the humans in the scene are identified, activity recognition will be the next step. Activities will include standing/sitting, walking, cycling, and running. For bicycle riders, the answer is already given by the detection algorithm. For other activities, a new, separate deep network can be applied to classify the target behavior. An activity is a temporal event that is defined across many frames, so a recurrent neural network will need to be designed to handle this. These networks must be tested and fine-tuned for ground-level views. There are several state-of-the-art networks to choose from, but because of the dynamic nature and heterogeneous viewpoints, a new network architecture may be necessary. The output of the automatic methods can be compared against ground truth to give an accuracy score for how reliable the automatic methods are.

The primary outcome variable for aim 2 is the number of individuals observed walking, cycling, running, and sitting/standing along each observation route/50 mins of observation.

Before developing statistical models, an examination of the univariate distribution of variables will be conducted (eg, scatter plots). Statistics such as means or proportions, SEs, ranges, and estimates of skewness and kurtosis will be derived. Data transformation procedures (eg, logarithmic) may be applied to quantitative variables whose distribution shows considerable departure from normality. Bland-Altman plots will be used to assess agreement on quantitative measures between the traditional BWM and WVD manual video analysis, the WVD manual video analysis, and the automated video analysis [53,54]. The difference between 2 methods for a variable of interest will be plotted against the average of the 2 methods for that variable. Horizontal lines representing the bias between 2 methods will be drawn at the mean difference. Additional horizontal lines will be drawn at the 95% limits of agreement (mean difference ± 1.96 [SD of the difference]). Two-way random effects intraclass correlation coefficients (ICCs) will also be calculated to examine agreement on quantitative measures between methods [55]. To test whether environmental variables (eg, obstructions such as trees) moderated outcome measure associations between methods, unadjusted and adjusted (for covariates) regression models that include a methods variable (eg, traditional BWM vs WVD-BWM), potential moderators, and interaction terms will be created to predict the outcome variable of interest (eg, number observed). In addition, we also will stratify the data by the levels of the moderator and re-examine effects. Simple effects analyses will be used to deconstruct significant interactions by examining associations between method and outcomes in separate subsamples stratified by levels of the moderator variables [56]. All statistical analyses will be performed using the SPSS statistical software package (IBM Corp Released 2015. IBM SPSS Statistics for Windows, version 23.0. IBM Corp).

We expect the WVD to experience technical difficulties at times. In recent months, we have been working with the Pivothead Smart, and on a few occasions there were issues with the recording device stopping during use and uploading videos from the device to a computer, which was because of a faulty cable. To correct or minimize these issues, we will provide observers with a reserve pair of glasses and keep additional cables on hand.

It is probable that some observations will be conducted in high-crime areas, making it unsafe for data collectors. We have encountered this in previous studies and addressed this by having a law enforcement officer accompany data collectors when necessary.

The Hawthorne effect is the alteration of behavior by the subjects of a study because of their awareness of being observed. Although this is a valid concern, in our past studies using the BWM we have not found any noticeable reaction to the observers. This is likely because of a couple of reasons such as the observers not standing out and appearing simply as individuals walking down the sidewalk. If people do react to the observers, it would most likely be because the observers walk at a slow pace and periodically write in a notebook while walking. We expect this concern to be eliminated with the use of the video glasses that are indistinguishable from regular glasses.