Summary

Objective: To examine the awareness, knowledge, attitude, risk perception and practices of Hong Kong residents on dengue fever prevention.

Design: A cross-sectional telephone survey.

Subjects: Hong Kong residents aged 12 years or above.

Main outcome measures: Awareness of health education on dengue fever; knowledge, attitude, and risk perception of dengue fever; practices for dengue fever prevention.

Results: 3,163 respondents completed the interviews from 1 to 30 December 2003. The response rate was 71.2%. 97.0% of the respondents were aware of health education on dengue fever prevention. 95.8% identified that dengue fever was transmitted via a mosquito bite and 85.8% recognised high fever as a symptom. 84.4% and 64.0% regarded preventing mosquito breeding and mosquito bites as effective preventive measures. The respondents perceived higher risk of contracting dengue fever outside Hong Kong. Most respondents took some preventive measures against dengue fever. Those who were male, single, working, having higher household income, have a fair knowledge of dengue fever and perceiving low risk of catching dengue fever were less likely to take preventive measures.

Conclusion: The general public's knowledge, attitude and practices for dengue fever prevention were improved compared to previous findings in 2002.

Keywords: Dengue fever, knowledge, attitude, practices, prevention

摘要

目的：探討港人對預防登革熱的意識、理解、態度、危機感和實踐。

設計：橫面電話訪問調查。

對象：十二歲或以上的香港居民。

測量內容：對登革熱健康教育的認知；對登革熱的理解、看 法及危機感；預防登革熱的實踐。

結果：於二零零三年十二月一日至三十日間，共有3,163位被訪者完成訪問，回應率為71.2%。97.0%的被訪者曾留意有關預防登革熱的資訊，95.8%的被訪者能指出登革熱是經由蚊叮傳播及85.8%知道高燒是登革熱的病徵，分別有84.4%及64.0%的被訪者相信防止蚊子滋生和避免蚊子叮咬是有效的預防措施。被訪者認為在境外的地方感染登革熱的風險較高。多數被訪者曾採取一些預防登革熱的措施。男性、單身、在職、家庭收入較高、對登革熱有不錯的認識及認為感染機會低的被訪者都較少採取預防措施。

結論：與二零零二年的調查結果比較，公眾對預防登革熱的認識、態度和實踐都有改善。主要詞彙：登革熱，認識，態度，實踐，預防

---

Introduction

Dengue fever is an acute viral disease transmitted by the mosquito Aedes egypti and less efficiently by Aedes albopictus. It is endemic in many tropical and subtropical countries. Hong Kong discovered its first local case of dengue fever in September 2002.¹ Up to August 2005, there were 21 local cases reported in Hong Kong.²

In October 2002, a territory wide against dengue fever campaign was launched. The campaign consisted of three different components: early detection and treatment of patients, public education for the community, and eliminating mosquito breeding sites from the environment.

A survey was then carried out by the Department of Health (DH) to evaluate the effectiveness of the campaign in December 2002.³ The findings reflected that the campaign was successful. However, it was recommended that all government departments had to work together to strengthen anti-mosquito measures and related health education. In response to this, the Government of the Hong Kong Special Administrative Region brought various government departments, non-governmental organisations, academic institutions and the community together to combat the outbreak of dengue fever. Important public health measures for disease control and prevention including surveillance, border control, risk communication, public education and community prevention have been implemented. The DH has been disseminating information, health advice and guidelines on dengue fever to the community.

Although there are some case reports and surveillance studies on dengue fever conducted in Hong Kong,^4-6 there has been no local research in dengue fever prevention. The 2002 Survey remains the only available study in this area. To update the situation, the DH conducted a similar study in December 2003 to examine the awareness, knowledge, attitude, risk perception and practices of Hong Kong residents in response to prevention of dengue fever.

Methods

The target population was Hong Kong residents aged 12 years or above. A telephone survey using a structured, bilingual (Chinese and English) questionnaire was conducted from 1 to 30 December 2003. Random telephone numbers were generated from residential telephone database. Respondents were then selected by a modified "Last Birthday" method. From each contacted household, a person aged 12 years or above who had his/her birthday most recently and was at home at the time of interview was selected to be the eligible respondent. This minimised over-representation of housewives and the elderly in the sample.

Computer aided telephone interview (CATI) system was employed to carry out interviews between 4:00pm and 10:00pm on weekdays and weekends, to avoid over-representation of non-working people.

Interviews were conducted in Cantonese, English or Putonghua by trained interviewers using standardized protocols. Each interview took about 20 minutes. No contact, refusal and drop-out cases received repeated calls (a maximum of five call attempts were made).

The questionnaire was designed by medical and research professionals. The questionnaire addressed the following areas: public awareness of health education on dengue fever prevention; knowledge about and attitude towards dengue fever and its prevention; risk perception of contracting dengue fever; practices of preventive measures against dengue fever and related barriers to prevention; together with demographic information concerning the respondents.

A four-point scale was adopted for the questions on risk perception, (i.e., very likely, likely, unlikely and very unlikely). An initial survey of 61 respondents was done as a pilot test to test the questionnaire and determine the logistics for the survey. The comments from the respondents and interviewers of the initial survey were considered and the questionnaire was revised accordingly.

Among the 4,444 contacted households with eligible respondents answering the call, 3,163 respondents successfully completed their interview. The numbers of drop-out and refusal cases were 285 and 996 respectively. The response rate was thus 71.2% (3,163/4,444).

Descriptive statistics were represented as percentages. Associations between practices and different variables, including demographic characteristics, knowledge, attitude and risk perception, were analysed using chi-square test. Multiple logistic regression models with backward stepwise selection were used to identify determinants of preventive practices by adjusting confounders and estimating odds ratios (OR). Statistical significance was presumed for probability p<0.05. Data analysis was performed using SPSS for Windows version 11.0.1.

Results

Profile of the respondents

A total of 3,163 Hong Kong residents participated in the survey. Among these, 43.7% were male and 56.3% were female; 50.2% were single and 49.7% were married; 34.3% were aged 12-24, 39.3% aged 25-44 and 26.2% aged above 44 years; 12.9% attained primary education level or below, 63.8% secondary school level and 23.3% tertiary level or above; 27.2% were students, 44.1% were working and 28.7% were non-working. The respondents' demographic profile is shown in Table 1.

Table 1: Demographic profile of the respondents

Characteristic   n*   %       Gender           Male   1,381   43.7   Female   1,782   56.3 Age group (years)           12 - 14   282   9.0   15 - 24   791   25.3   25 - 34   538   17.2   35 - 44   690   22.1   45 - 54   475   15.2   55 - 64   198   6.3   > 65   147   4.7 Marital status           Married   1,558   49.7   Never married   1,473   47.0   Widowed/ divorced/ separated   101   3.2 Educational level           Primary or below   405   12.9   Secondary   2,004   63.8   Tertiary or above   733   23.3 Working status           Working   1,368   44.1   Non-working   892   28.7   Student   845   27.2 Monthly household income (HK$)           < 9,999   390   18.6   10,000 - 19,999   622   29.7   20,000 - 29,999   434   20.7   30,000 - 39,999   267   12.7   > 40,000   383   18.3             * Total is not equal to 3,163 due to missing values

Awareness of health education

Nearly all respondents (97.0%) were aware of the disseminated information about dengue fever prevention in the three months prior to the interview. Most obtained the information through television [advertisements/Government's announcements (63.7%) and news (54.7%)] followed by newspapers (42.0%), radio (17.7%), posters (11.5%), leaflets (9.9%) and websites (5.5%). Sixty percent of the respondents were satisfied with the Government's efforts in providing these information.

Knowledge of dengue fever

The majority of the respondents (95.8%) correctly identified that dengue fever is transmitted via mosquito bites. Regarding the symptoms of this disease, 85.8% of the respondents recognised high fever for three to five days as a symptom, 44.2% recognised severe headache, 24.6% recognised pain behind the eyes balls and 23.8% recognised skin rash. Many respondents perceived preventing mosquito breeding (84.4%) and mosquito bite (64.0%) as effective preventive measures. However, nearly half of the respondents (49.5%) wrongly believed that there was an effective vaccine against dengue fever. Over three-fifths (62.2%) of the respondents knew that in Hong Kong there would be a fine for allowing mosquitoes or insects to breed in stagnant water.

Attitude and risk perception towards dengue fever

Most respondents (90.2%) perceived that the consequences of mosquito bites were serious and thus regarded preventive measures as necessary. The perceived risk ("very likely" or "likely" in the questionnaire) of catching dengue fever from outside Hong Kong (89.8%) was slightly higher than that from within Hong Kong (84.5%). Nevertheless, 92.6% of the respondents perceived the chance of surviving from the illness was high.

Practices of preventive measures

The proportions of the respondents who reported taking various measures, as recommended by the Government, to prevent mosquito breeding in the three months prior to the interview are shown in Figure 1. A high percentage of the respondents had practised putting refuse such as empty lunch boxes and soft drink cans (which could collect water) into covered litter bins (96.3%) and kept their drains free from becoming blocked (85.4%). Over 60% had covered up tightly water containers, water storage tanks or wells (67.0%), changed water in flower vase or plants at least once a week (65.6%), and removed stagnant water under flower or plant containers (60.7%). Nearly 40% had levelled up defective ground surface to avoid accumulation of stagnant water (37.4%). The main reasons cited by those respondents who had not taken preventive measures were "unnecessary", "busy", "lazy", "forgotten" and "time-consuming".

Figure 1: Preventive practices against mosquito breeding

Figure 2 shows the proportions of respondents who reported taking various measures to prevent mosquito bites in the three months prior to the interview. Around 60% of the respondents had worn long-sleeved clothes and trousers (65.5%) or had avoided visiting scrubby areas (59.6%). More than two-fifths had used mosquito repellents (42.6%) and nearly one-fifth installed mosquito net in non-air-conditioned rooms (18.0%). The main reasons for not adopting any preventive measure were described as "unnecessary" and "no mosquito".

Figure 2: Preventive practices against mosquito bite

Factors associated with preventive practices

For each preventive practice against mosquito breeding, the proportion of respondents who reported "measures were not taken" was less than 8% (excluding "not applicable" and "don't remember"). Logistic regression analysis therefore was not performed for this latter group. For each preventive practice against mosquito bite, associations were individually tested with demographic characteristics, knowledge, attitude and risk perception, followed by use of logistic regression analysis.

Students were more likely than working respondents to wear long-sleeved clothes and trousers (OR=1.81, 95% CI: 1.50-2.19). Respondents who were male (OR=0.54, 95% CI: 0.46-0.63), single (OR=0.79, 95% CI: 0.68-0.91), had a fair knowledge of dengue fever (OR=0.83, 95% CI: 0.71-0.98) and perceived a low risk of catching dengue fever in Hong Kong (Unlikely: OR=0.59, 95% CI: 0.43-0.82) were less likely than others to use mosquito repellent. Single respondents (OR=0.58, 95% CI: 0.45-0.74) and those having higher household income (HK$10,000-19,999: OR=0.64, 95% CI: 0.46-0.90 and HK$30,000-39,999: OR=0.61, 95% CI: 0.39-0.95) were less likely than others to install mosquito net in non-air-conditioned rooms. Male respondents (OR=0.72, 95% CI: 0.60-0.86) and those having higher household income (HK$30,000-39,999: OR=0.68, 95% CI: 0.49-0.96 and HK$40,000 or above: OR=0.63, 95% CI: 0.46-0.86) were less likely than others to avoid visiting scrubby areas.

Comparison with the 2002 Survey

The results of this survey were compared to the findings of the 2002 Survey (Table 2). The sample sizes and the distribution of demographic characteristics of the respondents in both surveys are similar (3,163 respondents in this survey versus 3,024 in the 2002 Survey). More people were aware of dengue fever prevention measures (97.0% in this survey versus 61% in the 2002 Survey, p<0.001). Their knowledge regarding mosquito bite as the transmission route for dengue fever also improved (95.8% versus 91%, p<0.001). The percentages of respondents recognising high fever as a symptom increased (85.8% versus 83%, p<0.005). More respondents correctly answered that there was currently no effective vaccine against dengue fever (50.5% versus 43%, p<0.001), and that there would be a fine for allowing mosquito breeding in stagnant water (62.2% versus 36%, p<0.001). Attitudes about dengue fever also changed, with more people believing that mosquito bites could lead to serious consequences and that preventive measures were needed (90.2% versus 80%, p<0.001). With regard to preventive actions, more respondents had placed refuse (which could collect water such as empty lunch boxes) into covered litter bins (96.3% versus 87%, p<0.001), had covered up water containers (67.0% versus 58%, p<0.001), had changed water for flower vase at least once a week (65.6% versus 38%, p<0.001), and had removed stagnant water under flower containers (60.7% versus 45%, p<0.001). In comparing the findings of the two surveys, there were overall significant improvements in public knowledge, attitude and behaviours in relation to the prevention of dengue fever.

Table 2: Comparison between this survey and the 2002 Survey3

Statement This Survey % 2002 Survey % p-value* 1. Aware of information about dengue fever prevention 97.0 61 p<0.001 2. Know about mosquito bite as the transmission route for dengue fever 95.8 91 p<0.001 3. Know about high fever as a symptom for dengue fever 85.8 83 p<0.005 4. Know about no effective vaccine against dengue fever currently 50.5 43 p<0.001 5. Know about a fine for allowing mosquito breeding in stagnant water 62.2 36 p<0.001 6. Agree to the statement that "mosquito bites leading to serious consequences and preventive measures being needed" 90.2 80 p<0.001 7. Put refuse (which could accumulate water such as empty lunch boxes) in covered litter bins 96.3 87 p<0.001 8. Cover water containers 67.0 58 p<0.001 9. Change water for flowers at least once a week 65.6 38 p<0.001 10. Remove stagnant water under flower containers 60.7 45 p<0.001           * p-value is derived from two-sided linear by linear Chi-square test. Statistical significance is presumed for probability p<0.05.

Discussion

This study's findings reflect improvements in the Hong Kong people's knowledge, attitude and practices for dengue fever prevention. Almost all respondents were aware of the disseminated information on dengue fever, with three-fifths of the respondents being satisfied with the Government's efforts in providing the information. However, the respondents might overestimate their risk of catching dengue fever. Negative responses to some practices for dengue fever prevention might mean that the preventive measures were not appropriate to the respondents. For example, the respondents did not wear long-sleeved clothes and trousers or use mosquito repellants because they had not gone to rural area in the three months prior to the interview. Respondents with high income might have better living environment in which mosquito nets were not required. Factors related to preventive practices against mosquito bite were gender, marital status, educational level, occupation, household income, knowledge and risk perception. Females and married persons were more likely to take preventive measures. The main reasons for not adopting the preventive measures were "unnecessary", "busy" and "lazy". All these factors have to be taken into consideration in future educational and promotional activities.

The study has several limitations. Firstly, the demographic composition of the sample is slightly different from that of the 2001 Population Census,⁷ with more respondents being female, single, younger and more educated. The difference may be due to choosing the modified Last-Birthday method __ i.e. not covering those who were seldom at home and the non-respondents being the hard-to-reach population groups. Secondly, the data were collected by self-report method which might have reporting and recall bias. Thirdly, the study was cross-sectional in nature and thus the causality between variables was uncertain.

Given the above findings, the against dengue fever campaign conducted by the Government appeared successful in preventing Hong Kong from becoming endemic with dengue fever. Since dengue fever is a statutory notifiable disease in Hong Kong, surveillance has been strengthened through issuing reminders to all doctors and health care professionals, and guidelines on referring patients for timely treatment. The Government has put more emphasis on promoting the preventive measures which had been inadequately practised by the public. Publicity campaigns targeting the general public include dissemination of health information on dengue fever by various means such as television, radio, newspapers, printed materials, hotlines and websites. Education interventions targeting specific groups include mobilizing non-governmental organisations, schools and estate management committees to organise health talks and exhibitions for their respective groups. Moreover, the Government has strengthened its work in inspecting buildings for environmental hygiene and removing potential mosquito breeding sites. Oviposition traps have been set up to detect the presence and distribution of adult Aedine mosquitoes across the territory. In addition, community groups have been organised to clean up sites and to enhance community participation in these efforts to reduce risk.

Conclusion

The effectiveness of the campaign was evident by the changes in knowledge, attitude and, more importantly, behavioural changes in the public. The results of the study have helped the Government to assess the level of effectiveness of its work and to improve its health promotion based on the needs of the public.

Acknowledgement

We thank the Social Sciences Research Centre of The University of Hong Kong for collecting the data.

Key messages

1. Health education can be effective in improving general public's knowledge, attitude and practices for dengue fever prevention.
2. Government's against dengue fever campaign was successful in preventing an endemic in Hong Kong in 2002-2003.

---

---

References