Start and anticipated finishing date: 2019 - 2022

​

Funding details: This project is funded by the Australian Research Council, through a scholarship that is part of a larger grant. The grant aims to reduce health disparities in CaLD populations, regarding HIV, STIs and BBVs.

​

Project team: Nang Nge Nge Phoo (Curtin University), Associate Professor Alison Reid (Curtin University), Dr Roanna Lobo (Curtin University), and Dr Daniel Vujcich (Curtin University).

​

Background: Among people living with hepatitis B virus (HBV) infection in Australia, 38% are from CaLD groups born in North-East and South-East Asia. Gaps exist in the available information related to HBV knowledge, attitudes and practices (KAP) in these populations. Previous surveys that focus on HBV KAP in these populations used face-to-face interview and self-administered paper survey techniques. These strategies are associated with barriers such as social desirability bias, interviewer bias, cost/time for translation and interpretation, spoken-only languages, translation quality, and variable literacy levels. These barriers could be overcome by the use of audio records and pictures.

​

Project overview and aims: This study aims to answer the research questions of whether the single-item non-response rates and prevalence of self-reported sexual behaviours differ between the ACASI mode and used face-to-face interviews or self-administered paper survey mode. The study developed a web-based audio computer-assisted self-interview (ACASI) survey tool with pictures illustrating survey questions and response options. Along with the audio narration, the sound waves highlighted the corresponding texts. The pictures were hardwired to all the survey questions and response options. This application and two other survey modes (face-to-face interview and self-administered paper survey) were used to administer a hepatitis B knowledge, attitudes, and practices survey among the Myanmar-born community in Perth, Australia, a community that uses several languages and has language barriers and/or limited literacy. It was hypothesised that the item non-response rates would be lower, and self-reports would be higher when sensitive information was collected using the ACASI mode than face-to-face interview mode, while there would be no such difference between the ACASI and self-administered paper survey mode.

​

Results: ​A total of 852 participants took the survey, with 284 participants taking the survey in each of the three survey administration modes: ACASI with illustrated pictures, face-to-face interviews, and self-administered paper survey. The results indicated significantly higher item non-response rates for ‘ever had sex,’ ‘pre-marital sex,’ ‘extra-marital sex,’ ‘made payment to have sex,’ and ‘received payment to have sex’ among ACASI participants than face-to-face interview participants. The item non-response rate was significantly lower among ACASI participants than among participants in the self-administered paper survey mode for ‘number of sexual partners in the past 12 months.’ The ACASI participants self-reported ‘ever had sex’ less frequently than participants in the face-to-face interview mode. The prevalence of self-reported data for ‘multiple sexual partners,’ ‘pre-marital sex,’ and ‘extra-marital sex’ were lower among ACASI participants than participants in the self-administered paper survey mode. Apart from these stated significant differences, the differences in item non-response rates and self-reports of sexual behaviours were not statistically significantly different for the remaining comparisons. Hence, it was inconclusive whether the item non-response rates were lower or self-reports were higher for sensitive information when surveys were administered in the ACASI mode. This finding echoes the evidence in the literature.

 

The hepatitis B knowledge score was medium (5.30 +/- 1.42; range – 0 to 7). The misconception of hepatitis B transmission through sharing eating utensils was common (67.3%). The proportion of respondents in the low knowledge score range (≤ 3) was highest among the respondents originating from Malaysia (21.1%) when compared to those originating from Myanmar (9.2%) and Thailand (8.0%); and highest among those who spoke the Chin language at home (18.8%) when compared to those who spoke Karen (9.8%) and Myanmar (8.9%). The self-reported hepatitis B testing (72.8%) and vaccination (58.1%) were most common among the respondents with a high knowledge score range (6 to 7). The self-reported hepatitis B seroprevalence is 4.8%. The self-reported seropositive status was most common among the respondents aged 26 to 45 years (39.0%) and 46 to 65 years (36.6%); and more common among males (51.2%) than females (48.8%).

​

Project implications: The current study reported the process of developing a web-based ACASI application with illustrated pictures, and offered lessons learned and recommendations for future development of similar applications. The study findings added to knowledge about the efficacy of using different survey administration modes including ACASI. A strength of the current study included the finding that pre-testing of ACASI elements saved time and resources which would have been required to re-program the ACASI application at a later stage. Of note, however, the study results are generalisable only with a caution around the characteristics of study participants. One of the outcomes of this study was an increase in the understanding of the hepatitis B knowledge, attitudes, and practices among the Myanmar-born community in Perth, Western Australia. Finally, this study increased the understanding of the use of the ACASI survey mode in collecting sensitive information from a migrant community with high linguistic diversity, and such information can be used in public health survey research to obtain information from people with language barriers and limited literacy.

​

Read the project publications here.

​