Bednet Utilization and Knowledge-Attitude-Practice Survey among Selected Barangays in 40 41 Malaria Endemic Provinces in the Philippines

Survey Protocol

Introduction:

The Philippines has 80 provinces comprising 1,634 municipalities. These municipalities are further subdivided into barangays (villages) which is the smallest administrative unit. At the start of 2003, malaria was endemic in 66 provinces in the Philippines. The population at risk, defined at that time as the entire population of the province where malaria cases were found, was estimated to be 70,687,400. By 2010, malaria was endemic in 57 provinces and 843 municipalities in the Philippines. The population at most risk was 6,387,734, or 12.5% of the total national population. At this time, the population at risk was defined as the people living in barangays with stable, unstable and sporadic transmission based on the most recent stratification system of the National Program. By 2013, there was further reduction in the number of malaria endemic provinces to 53. Just 47 higher-incidence municipalities in 13 provinces contributed about 97% of the total malaria cases reported in 2011-2013

Malaria has historically been one of the 10 leading causes of morbidity and mortality in the Philippines. Over an 8-year period, and in particular, the recent years of 2010-2013 – there has been a significant reduction in malaria cases, annual parasite incidence (API) and mortality rates throughout most of the country (Malaria Program Review 2013). Incidence of malaria has declined rapidly due to the intensified strategies implemented in the endemic areas by the Philippine Department of Health (DOH) and Local Government Units with the support from The Global Fund (TGF), World Health Organization (WHO), Pilipinas Shell Foundation Inc. (PSFI) and other partners. Recent report indicates malaria cases to be less than seven thousand five hundred in 2013 down from more than twenty thousand cases at the start of the project. The national goal of malaria elimination by 2020, in the light of current developments, is attainable. Fine tuning the program strategies and activities can be further enhanced with the data gathered in this study.

Review of Literature

Insecticide treated nets (ITN) are an effective tool for preventing the transmission of malaria[Lengeler C. Insecticide-treated bed nets and curtains for preventing malaria. Cochrane Database of Systematic Reviews 2004, Issue 2.]. Efforts promoting the use of LLIN are geared towards the broader objective of universal coverage, defined at the household level as the use of insecticide-treated nets by all household members regardless of age or gender, and with a coverage at the community level reaching at least 80% of those at risk [WHO, Insecticide Treated Mosquito Nets: A WHO Position Statement].

In a bednet utilization survey conducted in 2012 by Dela Cruz et. al, ninety-eight percent (98%) of households in the areas surveyed owned a mosquito net. On the average, there were 2.72 or roughly 3 nets present in every household. The survey revealed further that eighty-four percent (84%) of total sample population have slept inside a mosquito net the previous night. Also, approximately ninety-seven per cent (97%) of children 5 years old and below in the sample population slept under a mosquito net the previous night and approximately eighty-eight (88%) of pregnant women in the same sample population did likewise. This current survey seeks to determine the current rates of use and roughly validate such practice as revealed in the previous surveys.

Objectives of the Survey:

The survey aims to:

1. To determine the coverage and usage of LLINs by general populations, by pregnant women, by children less than five years of age in the target provinces, municipalities, and barangays identified.

2. To assess the interviewees’ knowledge, attitudes, and practices with respect to malaria.

3. Generate data for program monitoring and evaluation.

The results of this survey will likewise be used as baseline for the 13 provinces to which change will be evaluated against in 2016.

Key Indicators to measure:

1. Proportion of population who slept under the mosquito net the previous night
2. Proportion of children under 5 who slept under the mosquito net the previous night
3. Proportion of pregnant women who slept under the mosquito net the previous night
4. Percentage of population at risk covered by LLINs
5. Proportion of population owning a mosquito net (ITN/ LLIN)
6. Proportion of population who know the cause of, symptoms of, treatment for, and preventive measures for malaria stratified by people owning and not owning nets

Study area and study population:

The BUS-KAP will be conducted in a total of 41 malaria endemic provinces in the country. The 41 provinces will be divided into 2 groups of provinces with each cluster having a sample obtained from them. One group will be composed of 13 provinces while the other group will be of 28 provinces. The 13 provinces in the first group represent the provinces that will be covered in 2015 to 2017 under the GFATM New Funding Model (NFM) Grant. These are the provinces with a three-year average API of more than 1 per 1,000 pop. In these provinces, there are approximately 3,352,465 people living in the stable, unstable and, sporadic transmission barangays (population-at-risk). The second group of provinces of 28 is those with a three-year average API falling below 1 per 1,000 pop. These provinces will no longer be fully covered in the NFM Grant in 2015 to 2017. These provinces have a total at-risk population of 3,094,007.

Study Methodology:

The study will be a cross-sectional survey.

The 41 provinces will be divided into 2 groups – one group with 13 provinces and the other group with 28 provinces. The group of 13 provinces are those with average 3-year API of 1 per 1,000 pop and above; the group of 28 provinces are those with a 3-year average API of less than 1 per 1,000 pop. The group of 13 provinces represents the provinces that will be covered under the NFM Grant in 2015-2017.

From each group, only the barangays with stable, unstable and sporadic transmission will be included as source of the household samples. By definition, the people residing in these barangays within the provinces are the population-at-risk.

Sample size will be computed using the estimated population-at-risk and household number in the identified stable, unstable, and sporadic transmission barangays. Both groups of provinces will have a sample of their own. The household will be the unit to be sampled or surveyed. The household number in the barangays will be obtained by dividing the population by 5, which is the estimated household size.

The barangays in each group of provinces will be clustered according to stratification – a cluster of stable transmission barangays, a cluster for unstable, and another cluster for sporadic transmission. The household samples will be taken from the clusters of the stable, unstable, and sporadic transmission barangays in the provinces. A sample will be taken from each stratification area – a number from all stable transmission barangays, another number will be taken from the unstable, and another from the sporadic areas.

The number of barangays per stratification area that are to be included in the survey is 30 (the minimum number statistically requested). and average number of households to be surveyed in each barangay is equal to total number of households per stratification area divided by 30. The barangays to be surveyed will be randomly selected. Likewise, the households to be interviewed in each barangay will be randomly selected.

Sample Size:

The sample size figures were produced using StatCalc of Epi Info 7. The factors considered in determining sample size are the following:

(1) A desired confidence level of 95%

(2) An assumed prevalence of 50% in the target population.

(3) The Design Effect (DEFF), which expresses the increase of the sample size needed over that of a simple random sample as a multiplier. A DEFF of 2 will be used.

(4) The desired precision, typically equal to one half of the confidence interval width. We will assume that an overall precision of 5% is desired.

(5) A 10% non-response rate is also assumed

Data Gathering:

A team will be trained to conduct the interviews among the selected samples. The survey questionnaire form, attached as Annex B, will be used in the interviews.

Two sets of questionnaires will be utilized. The first set is the household-level questionnaire that will be answered by the head of the household or, in the event that the head of the household is not present, the person next in authority who residing in the household will be interviewed. The questions to be answered pertains to household composition, characteristics of the house, household health seeking behavior, net ownership, net use, and others (refer to Annex A Questionnaire). The second set is the individual-level questionnaire that will be answered by all eligible adults ranging from 15 years old and above who are residing in the household and are present at the time of visit. The questions in this set pertain to knowledge and attitudes of the respondents.

The trained team of interviewers will fill-up the printed questionnaires with the answers provided by the respondents

Inclusion Criteria:

For the household level questionnaire, the household head, father or mother, will be the eligible respondent. In the absence of either, any resident of the household above 15 years old may answer.

For the individual level questionnaire, all individuals residing in the household at the time of survey who are above 15 years old are eligible respondents.

Data Management and Analysis:

The data from the completed questionnaires will be entered into excel software or other capable data management software. Means and proportions will be computed/ generated to establish the values for the identified indicators:

1. Proportion of population who slept under the mosquito net the previous night
2. Proportion of children under 5 who slept under the mosquito net the previous night
3. Proportion of pregnant women who slept under the mosquito net the previous night
4. Percentage of population at risk covered by LLINs
5. Proportion of population who know the cause of, symptoms of, treatment for, and preventive measures for malaria

Ethical Considerations:

Confidentiality will be observed and maintained at all times possible. Data gatherers and encoders will be made familiar with the questions and its implications and will be trained on delivering the questions and handling responses gathered. Wherever possible, responses are coded and known only to the research team. All these measures are to minimize the risk of breech in confidentiality that may produce negative consequences to the respondent/s reputation or status. Further, informed consent will be utilized for the respondents of the survey. The respondents will be briefed on the purpose and process of the survey and only if they fully understand and agree to what was explained will they be considered as participant. This will be documented with a signed informed consent form. The data gatherers will likewise be trained in obtaining the informed consent form. Where appropriate, the questionnaires and informed consent form will be translated to the local dialect or dominant dialect understood in the community. The translated versions will be tested for understandability or validity.

Review and clearance from a recognized Ethical Review Board will be sought. In addition, review and affirmation from the National Commission on Indigenous Peoples and other governing groups of the affected populations will be obtained as needed.

Time Frame:

The study will take about 4 months to complete and will be contracted out to a third party entity who will closely work with the malaria Technical Working Group spearheaded by the National Program Coordinator of the Department of Health.

Budget:

This conduct of this survey will be contracted out to a third party. Estimated budget and its breakdown are as follows: