The extensive networks of frequent population mobility in the Samoan Islands and their implications for infectious disease transmission

Abstract

Population mobility has been demonstrated to contribute to the persistent transmission and global diffusion of epidemics. In the Pacific Islands, population mobility is particularly important for emerging infectious diseases, disease elimination programs, and diseases spread by close contact. The extent of population mobility between American Samoa villages, Samoa districts and other countries was investigated based on travel data collected during community surveys in American Samoa in 2010 and 2014. Within American Samoa, workers commuted daily across the whole of the main island of Tutuila, with work hubs drawing from villages across the island. Of the 670 adult workers surveyed, 37% had traveled overseas in the past year, with 68% of trips to Samoa. Of children aged 8-13 years (n = 337), 57% had traveled overseas, with 55% of trips to Samoa. An extensive network of connections between American Samoa villages and Samoa districts was demonstrated, with most trips lasting one week to one month. Our study showed that populations in the Samoan islands are highly mobile, and quantified the extent and destinations of their travels. Our findings offer insight into the impact of population mobility on the transmission of infectious diseases and data to refine existing models of disease transmission in the Pacific islands.

Figure 1

Network A: Commuting between residence and workplace villages within American Samoa, 2010. The color of lines is an average of weight of source and target nodes. wtd degree: weighted degree.

Figure 2

Correlation between the number of workers living or working in particular villages, and village population, 2010. Each point represents a village. (a) the number of workers living in the village, (i) red dashed line: y = 5.781 + 0.00161x and (ii) blue dashed line: y = −6.927 − 2.320 ln (x − 1.963); (b) the number of workers working in the village, red dashed line: y = 4.541 + 0.00365x.

Figure 3

Network B - Overseas travel, 2014: (a) 8–13-year-old school children (travel in their lifetime); (b) adult workers (travel in the past 12 months). The size of dots for overseas countries does not reflect the country’s population, and location of dots are not to scale. The colour of lines is an average of weight of source and target nodes. wtd degree: weighted degree.

Figure 4

Overseas travel destinations 2014 - children (blue) and adults (orange). Network B.a is the overseas travelling of children (see Fig. 3(a)) and network B.b is the overseas travelling of adult workers (see Fig. 3(b)).

Figure 5

Travelling duration and frequency, 2014: (a) frequency of overseas travel (all countries and Samoa): number of trips per year for adults, and number of trips over lifetime for children; the dashed lines are the exponential fits; (b) duration of overseas trips (all countries and Samoa) and duration of stay of visitors from Samoa; (c) correlation between the number of trips from American Samoa to specific locations in Samoa and the destination district population; and (d) correlation between the number of visits from Samoa to American Samoa and the visitors’ residence district population in Samoa. Network C.a is travelling to Samoa (see Fig. 6(a)) and network C.b is visiting from Samoa (see Fig. 6(b)).

Figure 6

Network C – Connectivity between American Samoa and Samoa, 2014: (a) American Samoa residents (adult workers) travelling to specific locations in Samoa; (b) short-term visitors from specific locations in Samoa staying at homes of American Samoa residents. Note: maps of Samoa and American Samoa are not shown to scale. The colour of lines is an average of weight of source and target nodes. wtd degree: weighted degree.