• What is assessed?
• How is the measurement conducted?
• When is this method used?
• How are estimates of physical activity derived?
• Strengths and limitations
• Populations
• Further considerations
• Resources required
• Instrument library
• References

Global navigation satellite systems (GNSS) provide a precise location at any point on the Earth’s surface. Global Positioning System (GPS) is the most widely used system owned by the US Government. Other systems include GLONASS (Russian) and Galileo (European).

GNSS data provide a measurement of position in three dimensions, i.e. latitude, longitude, and altitude. Time-series data of position yields speed, ascent/descent rates, and acceleration. GNSS data are most useful when combined with other behavioural exposure data in order to determine the spatial context of that behaviour. The dimensions of physical activity assessed by GNSS receivers are described in Table P.3.25.

Table P.3.25 Physical activity dimensions which can be assessed by GNSS receivers.

A dedicated GNSS receiver is carried by the participant. The monitor does not need to be attached to the body but needs to be in close proximity to the person under investigation (i.e. worn on a belt or carried in pocket or bag). A number of GNSS receivers are now available that can be worn on the wrist.

GNSS receivers estimate their position on Earth by triangulating signals sent from satellites orbiting Earth. The signals are exact time stamps of transmission, which are compared with the exact time in the GNSS receiver upon reception; the difference is the time taken for the signal to travel the distance from each satellite to the receiver, and since the speed of radio signals are a known constant, the distance can be calculated. As GNSS receivers have the satellite position schedule stored, each calculated distance identifies a number of possible positions on Earth. A minimum of three satellites is required to estimate the latitude and longitude of the receiver, and a minimum of four to also estimate altitude.

Dedicated GNSS receivers store 3-dimensional coordinates that can be exported and processed by dedicated software packages. Most smartphones also include GNSS receivers, although extracting data from them may be less straightforward.

GNSS is used when researchers want to know a participant’s location and/or route. GNSS can be combined with geographical information systems (GIS) which provide additional layers of data on the location, e.g. types of land use, crime levels, air pollution, etc. This allows examining, for example, which types of environments are more or less conducive to physical activity (e.g. in greenspace) or whether a participant is using particular geographical features of interest (e.g. new cycle routes, supermarkets, takeaways (Krenn et al., 2011)). When combined with objective time-stamped measurement of physical activity, GNSS can provide a greater understanding of where activity takes place and may even refine the classification of the type of activity being performed.

The raw measurement performed by a GNSS receiver is time differences of multiple satellites, which by triangulation will yield a 3-dimensional spatial coordinate on Earth. When such positional data are collected as a time-series, speed and ascent/descent rates (first time differentials) and acceleration (second time differentials) can be derived, which may in turn be used to make inferences on some limited activities as well as their location and whether the receiver was indoors or outdoors using the signal-to-noise ratio and an appropriate threshold (see Figure P.3.13).

Research that seeks to estimate physical activity behaviours exclusively from GNSS measures is relatively under-developed; most methods incorporate additional measurements such as accelerometry or heart rate monitoring.

Nonetheless, speed on its own can be used to make crude inference on travel modes by considering the most common speed ranges for each mode; walking trips are often identified if the 90^th percentile speed is <10km/h, cycling trips are classified as those with a 90th percentile speed between 10 and 35 km/h, and vehicle trips as those with a 90th percentile speed of ≥35 km/h (Kerr et al., 2012). Others have used slightly different statistical parameters, e.g. the average, standard deviation and maximum speeds (Miller et al., 2015).

Adding incline and land-use data to the classification scheme by projecting the GNSS track data into a GIS may further enhance the classification accuracy, e.g. if a participant was following a footpath or a railway line, it is more likely that the person is walking/running or travelling by train, respectively (Panter et al., 2014). Similarly, by combining GNSS with records from diaries/logs or time-stamped objective physical activity data, inferences on walking, running, cycling, and motor transport are enhanced, including estimation of activity energy expenditure (Costa et al., 2015, Nguyen et al., 2013, Panter et al., 2014).

Figure P.3.13 Inference scheme for use of GNSS receiver data for inferring physical activity dimensions. Adapted from: Corder et al., 2008.

Characteristics of GNSS receivers are described in Table P.3.26.

Strengths

1. Provides rich data on movement of an individual through an environment
2. No need for direct attachment to the body
3. Relatively inexpensive
4. GNSS is increasingly integrated into consumer devices which may enhance research possibilities in future

Limitations

1. Limited battery life; typically require recharging overnight
2. Some devices have relatively low data storage
3. Temporary loss/disruption of satellite signal in some environments, e.g. inside large buildings or in streets with high buildings (‘urban canyons’)
4. Most devices would need to be removed for contact or water sports
5. Potential privacy concerns (ascertainment of exactly where a participant has gone); the receiver could be switched off
6. Requires participant to switch on and off (can be forgotten or done deliberately) or charged overnight
7. Delay in signal acquisition when device is first turned on
8. People may forget to keep on person (and leave in bag or car for instance)
9. Data is considered sensitive/ identifiable so requires safe storage of data with controlled access (encryption, firewalls)
10. Complexity of analysis, including availability of time-relevant GIS data
11. When combining GNSS data with other monitor data, analysis need to allow for possibility that two devices may have become separated or contain different degrees of missingness

Table P.3.26 Strengths and limitations of GNSS receivers.

Considerations relating to the use of GNSS receivers for assessing physical activity are summarised by population in Table P.3.27.

Table P.3.27 Physical activity assessment by GNSS receivers in different populations.

1. Data security (identifiable information)
2. Potential privacy concerns
3. For smaller studies it may be possible for the researcher to collect monitors
4. Monitors can be returned either face-to-face or via post
5. Ensure time synchronisation if using other devices concurrently
6. Label chargers clearly to increase likelihood that they will be returned
7. Logs of wearing may help self-monitoring
8. Provide clear instructions on operating device
9. Investigate and overcome any barriers to wearing
10. When first turned on, signal may be delayed where the receiver acquires the required number of satellite signals to estimate position.

1. GNSS receivers
2. Chargers (for each participant)
3. Attachment mechanism (i.e. belt and pouch)
4. Written instructions for wearing/using
5. Log for on off times (potentially)
6. Secure data storage
7. Software for analysis
8. GIS data to overlay to allow fuller interpretation of GNSS data

A list of specific GNSS instruments is being developed for this section. In the meantime, please refer to the overall instrument library page by clicking here to open in a new page.

1. Corder K, Ekelund U, Steele RM, Wareham NJ, Brage S. Assessment of physical activity in youth. Journal of Applied Physiology (Bethesda, Md.:1985). 2008;105:977-87
2. Costa S, Ogilvie D, Dalton A, Westgate K, Brage S, Panter J. Quantifying the physical activity energy expenditure of commuters using a combination of global positioning system and combined heart rate and movement sensors. Preventive Medicine. 2015;81:339-44
3. Kerr J, Norman G, Godbole S, Raab F, Demchak B, Patrick K. Validating gps data with the palms system to detect different active transportation modes. Medicine and Science in Sports and Exercise. 2012;44(5S):S2529.
4. Krenn PJ, Titze S, Oja P, Jones A, Ogilvie D. Use of global positioning systems to study physical activity and the environment: a systematic review. American Journal of Preventive Medicine. 2011;41:508-15
5. Miller HJ, Tribby CP, Brown BB, Smith KR, Werner CM, Wolf J, Wilson L, Oliveira MG. Public transit generates new physical activity: Evidence from individual GPS and accelerometer data before and after light rail construction in a neighborhood of Salt Lake City, Utah, USA. Health & Place. 2015;36:8-17
6. Nguyen DM, Lecoultre V, Sunami Y, Schutz Y. Assessment of physical activity and energy expenditure by GPS combined with accelerometry in real-life conditions. Journal of Physical Activity & Health. 2012;10:880-8
7. Panter J, Costa S, Dalton A, Jones A, Ogilvie D. Development of methods to objectively identify time spent using active and motorised modes of travel to work: how do self-reported measures compare? The International Journal of Behavioral Nutrition and Physical Activity. 2014;11:116