Nationwide study of factors associated with public’s willingness to use home self-test kit for dengue fever in Malaysia

Interviews were conducted between February 2012 and June 2013 using a computer-assisted telephone interview (CATI) system. The telephone numbers were randomly generated by the computer from the latest electronic residential telephone directory (2012/2013) of all 13 states in Malaysia. To qualify for the telephone interview, participants had to be Malaysians, between 18 and 70 years old, have heard of dengue fever and residing in the contacted household. The computer randomly generated telephone numbers from the electronic residential telephone directory. Only one person per household was surveyed. If more than one qualified person was found in a household, one person was randomly selected using a random number table. Interviews were conducted between 5:30 pm and 10:00 pm on weekdays and from 12:00 p.m. to 7:00 p.m. on weekends or public holidays to avoid over-representation of unemployed participants. Unanswered calls were attempted at least two more times on separate days before being regarded as non-responses.

The questionnaire (Additional file 1) about dengue fever consisted of: socio-demographic characteristics, dengue fever knowledge, a health belief questionnaire from the HBM, perceived uptake of a self-test kit and characteristics of the test kits if one should become available.

Dengue fever knowledge consists of 44 items. For each statement, participants could choose between three response categories: 'yes', 'no' and 'don't know'. For analyses, participants were scored as 1 for a correct response and 0 for an incorrect or 'don't know' response. Several negatively worded items were reversed and re-coded during the data analysis process. Possible scores ranged from 0 to 44. Higher scores indicate greater dengue knowledge.

Belief questions were based on several HBM constructs. Severity perception regarding dengue fever consisted of two parts: 1) perceived severity and perceived susceptibility of dengue fever, where the perceived severity assesses feelings concerning the seriousness of dengue fever and 2) perceived susceptibility assesses one's subjective perception of the risk of contracting dengue fever. Perceived severity of dengue fever was measured on a scale of 0–10 with higher score indicating higher severity. Likewise, perceived susceptibility of dengue fever was measured on a scale of 0–10 with a higher score indicating higher susceptibility. Perceived barriers to dengue prevention examine perceptions of barriers to prevent dengue fever among participants. This was also measured on a scale of 0–10 with a higher score indicating greater barriers. Self-efficacy was the opinion of the respondent regarding their ability to successfully manage the dengue prevention behaviour; self-efficacy was measured by a 4-point Likert scale that ranged from 1 (strongly agree) to 4 (strongly disagree).

Perceived uptake of the dengue self-test kit consisted of three questions. First, “would you consider using a home self-test kit for dengue if the kit is available?” Second, participants were asked, “Would you be able to prick your finger to do self-testing for dengue?” Thirdly, characteristics of the test kits which influence the use of dengue test kit if available, consist of five questions; easy to obtain, easy to use and instructions easy to understand, accuracy of test kit, price and recommendation by health care providers.

Demographic questions (10-item) were asked after completion of the survey questions. To avoid the problems inherent in translation, two bilingual experts translated the instruments from English to Bahasa Malaysia, after which they were again blindly back-translated by two other bilingual experts. Content validity of the questionnaire was assessed by groups of experts to ensure that the items have acceptability content validity. The final draft version of the questionnaire was pilot tested.

All statistical analyses were performed with the Statistical Package for the Social Sciences Version 20.0 (SPSS; Chicago, IL, USA). Non-responses and irrelevant answers were treated as missing values and therefore excluded from the analyses. Values of p < 0.05 were considered significant. In addition to descriptive analyses, the chi-square test was used to test the significance of differences in percentages. Multivariable logistic regression was used to identify factors associated with the intent to use the dengue home test kit if it becomes available. In the modeling strategy, the independent variables were included in the model if they had a p of < 0.05 on univariate analysis using the 'Enter' method.

Figure 1 shows the flowchart of the CATI process. A total of 15,644 call attempts were made, resulting in 2,512 responding households. The response rate computed as the number of completed interviews (2,512) divided by the number of eligible and contactable households (5,354) was 46.0 %. Participants who declined to participate in the survey were asked for their reasons for not wanting to participate. The most common reasons for not wanting to participate were “busy” (35.2 %) and “not interested” (18.0 %). A total of 11 (0.3 %) participants said that they did not know about dengue, and declined to be interviewed.

As Table 1 shows, the mean age of the sample was 40.18 (16.08) years, age range 18 to 70 years old. There were more females than males in this study. Most of the study participants had secondary school education. The majority of participants were Malay (55.5 %). Less than half of participants 31.5 % (n = 790) had a monthly average income below RM 2,000.

More than half of participants (n =1612, 64.2 %) knew a virus causes that dengue fever. Many of the participants (n =2287, 91 %) reported that Aedes mosquito breeds in clean and stagnant water. Most of the participants (n = 2330; 92.8 %) were aware that the Aedes aegypti mosquito has black and white stripes on its legs and body. More than one-third of the participants (n = 917; 36.5 %) indicated that dengue haemorrhagic fever (DHF) usually occurs in people who had several dengue infections. Less than half of participants (n = 1205; 48.0 %) reported that the Aedes spp. mosquito could live in places with a lot of plants. About half of participants (n = 1282; 51.0 %) indicated that dengue usually appears four to seven days after a mosquito has bitten someone. Most participants (n = 2261; 90.0 %) were aware that fever is a symptom of dengue infections. However, only 39.2 % (n = 985) of participants knew that pain in the eyes is a symptom of dengue. Many of the participants (n = 1930; 76.8 %) were aware that there is no vaccine to prevent dengue infection (Fig. 2). The mean total dengue fever knowledge score for the overall sample was 27.49 (SD ± 8.34), out of a possible score of 42.

Socio-demographic characteristics were not significantly associated with the perceived severity of dengue scores. Many participants (71.4 %) had scores below 5 (out of possible 10 score) for perceived susceptibility of dengue fever. Marital status and gender were significantly associated to perception of susceptibility of dengue fever. A significantly higher proportion of married participants (30.3 %) had scores above 5 for perceived susceptibility of dengue fever compared to single participants (25.3 %), χ2 = 5.90, df 1, p < 0.05. A significantly higher proportion of male participants (31.9 %) had scores above 5 for perceived susceptibility of dengue fever compared to female participants (27.3 %), χ2 = 5.40, df 1, p < 0.05. More than two-third of participants (79.7 %) had scores above 5 for perceived barriers to prevent dengue fever. All socio-demographic characteristics were not significantly associated with the perceived barriers to prevent dengue fever scores. More than half (69.5 %) indicated they agreed or strongly agreed that they lacked the self-efficacy to prevent dengue. Significantly higher proportions of female participants (72.7 %) indicated that they agreed or strongly agreed to lacking self-efficacy in preventing dengue compared to male participants (61.8 %), χ2 = 5.40, df 1, p < 0.05.

As shown in Table 2, those who were single and participants with a tertiary educational level were more likely to consider using a home self-test kit for dengue if the kit were available. Participants with higher mean total dengue fever knowledge scores (23–44) had significantly higher proportions of using the self-testing dengue kit for dengue if the kit were available (47.6 %, n = 816) compared to lower total dengue fever knowledge scores (range 0-22; 38.9 %, n = 310). Among those willing to use the dengue test kit, 45.5 % of them have the perceived severity of dengue fever score of 6–10 compared to only 38.7 % with score of 0–5. A total of 48.6 % of those willing to use dengue test kit have the perceived severity score of 6–10, compared to only 43.9 % with score of 0–5. In the multivariate logistic regressions analysis, when all the significant associations in the univariate analyses were entered into the model, the results showed that participants with Malay [odd ratio (OR) 0.55; 95 % confidence interval (CI) 0.33–0.93; p = 0.001], Chinese [OR: 0.23; 95 % CI: 0.13–0.41; p = 0.01] and Indian [OR:0.35; 95 % CI: 0.19–0.64; p = 0.02] ethnicity indicated that they were less likely, compared to other ethnicities, to use a home self-testing dengue kit for dengue if the kit were available. Participants with primary educational levels were less likely than participants with a tertiary educational level to use a home self-testing dengue kit for dengue if the kit were available [OR: 0.65; 95 % CI: 0.43–0.89; p = 0.02]. Participants with lower perceived barriers to dengue prevention (level of barriers 0–5) were less likely (OR: 0.67, 95 % CI, 0.53–0.85, p < 0.001) to use a home self-testing dengue kit for dengue if the kit were available compared to the reference level of barriers 6–10. Participants with lower total dengue fever knowledge scores (range 0–22) were less likely to use a home self-testing dengue kit for dengue if the kit were available [OR: 0.75; 95 % CI: 0.61–0.91; p = 0.001] compared to the reference group (participants with higher total dengue fever knowledge scores; range 23–44).

As shown in Table 3, the following 5 characteristics influenced participants’ overall acceptance of the dengue home test kit. The majority (n = 114, 63.7) mentioned ease of use and easy to understand instructions, followed by the price of the test kit, the accuracy of the test, recommendation by healthcare providers and ease of accessibility. Participants that noted ease of use and easy to understand instructions as important to the use of the dengue home test kit [OR: 2.42; 95 % CI: 1.76–3.32; p = 0.001] were more likely to use the home test kit if the kit were available (Table 3).