Social encounter profiles of greater Melbourne residents, by location--a telephone survey

Abstract

Background: Models of infectious disease increasingly seek to incorporate heterogeneity of social interactions to more accurately characterise disease spread. We measured attributes of social encounters in two areas of Greater Melbourne, using a telephone survey.

Methods: A market research company conducted computer assisted telephone interviews (CATIs) of residents of the Boroondara and Hume local government areas (LGAs), which differ markedly in ethnic composition, age distribution and household socioeconomic status. Survey items included household demographic and socio-economic characteristics, locations visited during the preceding day, and social encounters involving two-way conversation or physical contact. Descriptive summary measures were reported and compared using weight adjusted Wald tests of group means.

Results: The overall response rate was 37.6%, higher in Boroondara [n = 650, (46%)] than Hume [n = 657 (32%)]. Survey conduct through the CATI format was challenging, with implications for representativeness and data quality. Marked heterogeneity of encounter profiles was observed across age groups and locations. Household settings afforded greatest opportunity for prolonged close contact, particularly between women and children. Young and middle-aged men reported more age-assortative mixing, often with non-household members. Preliminary comparisons between LGAs suggested that mixing occurred in different settings. In addition, gender differences in mixing with household and non-household members, including strangers, were observed by area.

Conclusions: Survey administration by CATI was challenging, but rich data were obtained, revealing marked heterogeneity of social behaviour. Marked dissimilarities in patterns of prolonged close mixing were demonstrated by gender. In addition, preliminary observations of between-area differences in socialisation warrant further evaluation.

Fig. 1

Map showing geographical extent of Hume and Boroondara Local Government Areas, within the context of Greater Melbourne. The map was generated with QGIS software using data from the Australian Bureau of Statistics [36, 37]. Red lines depict the road network [12], and shading denotes built-up areas [38]

Fig. 2

Characteristics of locations and encounters by location type. Summaries of locations and encounters from one study day for each participant, reported separately for each location type. The number of unique addresses visited by participants (top left panel). Boxplot for participants' total (weighted) number of encounters with listed individuals, by location (top right panel). Boxplot for participants' total (weighted) duration of encounters with listed individuals, by location (bottom left panel). Boxplot for participants' total duration of encounters involving physical contact with listed individuals, by location (bottom right panel). For all boxplots, boxes denote the interquartile range, interior lines shown the median and the whiskers show adjacent values. Across the boxplots, marked heterogeneity is apparent in several location types. For total duration of contact, Work (and Home to a lesser degree) is the dominant location type. For total duration of physical contact, home is the dominant location type. Boxplots use raking weights to reduce the effects of sample bias. For each location type, participants with no time spent at that location type do not contribute to those results

Fig. 3

Participants’ total number of encounters with listed individuals. Histogram for participants’ total (unweighted) number of encounters with listed individuals (left panel). Boxplot for participants’ total (weighted) number of encounters with listed individuals by participant age (right panel). Heterogeneity is apparent across participants and across age groups. In general, participants' number of encounters declines with age. The boxplot uses raking weights to reduce the effects of sample bias

Fig. 4

Age-based mixing matrices for participants’ total number of encounters with listed individuals. Values are the mean of participants’ total (weighted) number of encounters with listed individuals by age of participant and contact for male participants (left panel), and female participants (right panel). 95 % confidence intervals are shown in parentheses. Darker shading indicates larger values. Age-based assortative mixing is evident for both genders. Women notably have more numerous contacts with children less than 15 years old. Males 18–29 have the most numerous contacts within age groups. Raking weights are used to reduce the effects of sample bias

Fig. 5

Age-based mixing matrices for participants’ total duration of encounters with listed individuals. Values are the mean of participants’ total (weighted) duration of encounters with listed individuals by age of participant and contact for male participants (left panel) and female participants (right panel). 95 % confidence intervals are shown in parentheses. Darker shading indicates larger values. While age-based assortative mixing is evident, the large duration of contact between women and pre-school age children is most notable. Raking weights are used to reduce the effects of sample bias

Fig. 6

Participants’ total (weighted) number of listed encounters by type of contact individual. Boxplots for participants' total number of listed encounters by known/unknown contacts for male (left) and female (right) participants (top panel), household/non-household contacts for male (left) and female (right) participants (middle panel), and by household size, known/unknown contacts, and location of contact (Home/Outside) for male (left) and female (right) participants (bottom panel). Young men and women aged 30–49 years reported many more encounters with listed people than other participants. For young men, the majority are non-household members. For women, about half involved household members. In general, the number of encounters with known individuals increases with household size. Raking weights are used to reduce the effects of sample bias