| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
There is virtually no published data on these questions. It is important to distinguish between net ownership (possession of a net/ITN) and net use (actual use of nets owned). Most studies of "net use" actually report on net ownership. Of studies providing data on actual use, almost all report the percent of vulnerable groups sleeping under a net among all households sampled. The African Summit on Roll Back Malaria (RBM) held in Abuja, Nigeria, on April 25, 2000, set the target of having at least 60% of each vulnerable group sleep under an ITN. To monitor progress toward Abuja targets, RBM developed indicators using the number of under-5s or pregnant women from all households, including those that do not own nets, as the denominator.1 This indicator is appropriate for looking at nationwide progress toward Abuja targets, but the resulting percentage is necessarily constrained by the percent of households owning a net. (For example, if 20% of households own an ITN, there will be a maximum of 20% of under-5s sleeping under an ITN.) Furthermore, the indicators do not reveal intra-household net-use patterns.
Few studies look only among net-owning households and analyze if and by whom nets are used, which requires measuring use by family members besides under-5s or pregnant women to determine how use by vulnerable groups compares to that of other household members. A few studies do address some aspects of intra-household net use, with one in The Gambia concluding that vulnerable groups were more likely than other family members to use a net2; another from The Gambia finding that, by a small margin, adults used nets more than children (but defined children as under 10 years of age) and girls were more likely than boys to use the net3; and another from Burkina Faso finding that adult men were most likely to use the net (though the sample size was small at 103 households, all from 1 province).4 A study based on secondary analyses of the Demographic and Health Survey in Uganda concluded that young children were sleeping under a net only because their mothers were using the net.5 Several other studies focusing on who uses the household net were intervention studies where nets were given free to those living in a research area in Kenya6 or to pregnant women attending antenatal clinics in Kenya,7 or where nets were acquired via vouchers distributed to pregnant women in Tanzania.8 The first found that adults were slightly more likely than young children to be using the net; and the last 2 found that nets were being used by the groups targeted by the intervention: pregnant women and infants. Because these are intervention sites, however, we do not know whether these findings apply to the general population.
None of the studies indicates whether the data were collected during the rainy or dry season, although the Demographic and Health Surveys usually collect data during the dry season to avoid the logistical problems of collecting data during the rainy season. Net use is subject to seasonal variation and it is important to know the context of the findings.
Several characteristics of the present study distinguish it from other studies on net use:
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
NetMark conducted household surveys in Nigeria, Senegal, and Zambia in 2000 and again in 2004. In addition, NetMark conducted a survey in Mali in 2003 and in Ghana and Ethiopia in 2004. This paper compares results among the 6 countries surveyed in 2003/2004 and reports on changes over time in the 3 countries where both 2000 and 2004 data were collected.
The survey used population-based, multistage sampling. In each country, the sample was drawn from and divided equally among 5 primary sites, listed in Table 1
. A site consists of an urban center along with surrounding rural areas, up to 200 km from the urban center. The sites were located in malarious areas and were purposively selected to reflect the geoethnic diversity of the country. The capital or main commercial city was included as a site in each country except in Ethiopia, because Addis Ababa is located at too high an altitude for malaria to be a significant problem.
|
Respondents were women of reproductive age (15–49) who were mothers or guardians of children under 5 years of age. Each potential respondent was read an informed consent statement that included the purpose of the project, the benefit to the community, the approximate amount of time it would take, and an assurance of confidentiality. The respondent was asked if she would like to participate, and if she responded affirmatively, the interviewer signed a statement saying the informed consent had been read and that the respondent agreed to participate.
In Nigeria, Senegal, and Zambia there were 2 rounds of data collection that were structured to enable a valid measure of change over time in the sites sampled. The same sites and the same sampling procedures within sites were used for all survey administrations, but the number of clusters was doubled in 2004 to permit more robust analyses. In each country, the fieldwork was conducted at the same time of year for each data collection round to control for seasonal variation. In all countries except Zambia, the data were collected during the rainy season. (All baseline data had to be collected before program start-up, which was just before the rainy season in Zambia.) The same basic instrument was used across countries and data collection waves.
The survey asked each respondent whether her household owned any hanging nets used for sleeping and, if so, how many. Information such as treatment status, size, shape, and brand was collected for up to 4 nets per household. Each respondent was also asked whether the household owned any baby nets. A baby net was defined as a small, umbrella-like net with a built-in frame.
The questionnaire enumerated household members and recorded the age and gender of each, as well as the pregnancy status of women of reproductive age. A household was defined as the interviewee, the children she is responsible for, her spouse, and any other dependents, such as parents, siblings, etc. This excludes co-wives and their children, although it may include other women of reproductive age, such as a sister living in the household. Table 1
shows, for each country, the number of households sampled and the resultant number of household members and nets, which constitute the denominators for the analyses here. The large increase in the number of people in net-owning households between 2000 and 2004 is due not only to the increased number of households sampled but also to increases in the percent of households owning a net.
The questionnaire section on household members and net use was organized into a grid of nets and family members, providing the means to indicate which family members, if any, slept under each net owned "last night." This allowed us to identify nets that had not been used the prior night. In the main data file, the unit of analysis was the household. The data were organized and coded in such a way as to permit generation of another data file with each family member as the unit of analysis as well as a third data file with each net as the unit of analysis.
Although this means of gathering data is more time consuming and complex than asking only about the youngest child and pregnant woman or only asking "who slept under the net," the systematic enumeration of household members and individual nets is likely to yield more accurate data about who slept under the net(s) and allows comparison of use by vulnerable groups with use by those less vulnerable. When only use by vulnerable groups is reported, there is no way of knowing if this is higher or lower than use by other household members. The application of an identical methodology for all surveys provides an opportunity to produce reliable and valid comparisons among the 6 countries and, for 3 of them, changes that have taken place between 2000 and 2004.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
shows the percent of nets used the prior night for each country and data collection wave.
|
Between 2000 and 2004, the percent of nets used the prior night stayed about the same in Senegal and Zambia but declined considerably in Nigeria: from 91% to 56%. Because, in each country, the data were collected at the same time of year in both 2000 and 2004, seasonal variation would not explain Nigeria’s substantial drop in the percent of nets used.
Number of months per year nets were used. Another measure of net use is consistency of use during the year. Most families do not use their nets year-round because of seasonal variation in mosquito density and malaria frequency. However, nets should be used year-round because there can be substantial risk of transmission even when there is low vector density.9,10
Respondents were asked how many months per year the family slept under the nets they owned. In 2004, Ethiopia and Nigeria reported using nets for the shortest amount of time during the year (5.8 and 5.9 months, respectively), and Zambia, Ghana, and Mali the longest (7.1, 7.2, and 7.6, respectively). Since 2000, there was an increase in the mean number of months nets were used in Senegal from 5.6 to 6.3 months (P = 0.002) and Zambia from 6.4 to 7.1 months (P = 0.009). However, in Nigeria the number of months of net use declined considerably, from 7.6 to 5.9 months of the year (P = 0.000).
Number of people sleeping under 1 net.
Each household net that was used protected on average just over 2 people (see Table 2). In 2004, the average number of people sleeping under a net ranged from a low of 2.03 in Nigeria to a high of 2.37 in Mali. In Senegal and Zambia, the average number of people sleeping under a net was about the same in both 2000 and 2004, but in Nigeria, the average number of people under a net increased from 1.83 to 2.03 (P = 0.013). Across countries, rural households were somewhat more likely than urban ones to have a third person under the net.
|
Some nets owned were single size, some double, and some triple/king. The number of people under a net increased with the size of the net. Ethiopia deviates from the pattern slightly in that the average number of people under a single and double net is about the same.
Percent of household members sleeping under a net.
The percent of household members sleeping under a net ranged from 32% in Nigeria to 54% in Mali (see Figure 2). The percent under a net in Zambia was also rather low (34% in 2000 and 38% in 2004), but Zambia was the only country where the data were collected during the dry season, and rainy season data would show a higher use rate. The 32% for Nigeria is from 2004 and represents an unexplained drop from 2000, when 45% of family members slept under a net. Ethiopia is also rather low at 39%.
|
Children under 5
Looking across countries in 2004, rather consistent and positive patterns emerged, as depicted in Figure 2. Generally, children under 5 as well as pregnant women—the two most vulnerable groups—were most likely to be under the net. The possible exception was Ethiopia, where a smaller percentage of pregnant than nonpregnant women of reproductive age slept under a net, although the difference was not statistically significant. In all countries, children under 2 were more likely than any other family member to be under a net (see Figure 3).
|
In all countries, the 2 groups least likely to sleep under a net were children aged 5–14 and adult males. Adult males were much less likely to sleep under a net than adult females (both pregnant and nonpregnant), although the difference is least pronounced in Ethiopia.
In the 2000 baseline in Zambia, pregnant women were least likely to be under the net; but in 2004 the situation greatly improved. The proportion of pregnant women under a net in Zambia increased from 17.6% in 2000 to 50.5% in 2004.
Nigeria presents a mixed picture. Although children under 5 and pregnant women are most likely to be using the household net, the percent of these groups under a net in net-owning households decreased between 2000 and 2004 due to the large increase in unused nets.
Sleeping groups.
In all countries, the most common sleeping group by far was a woman of reproductive age and a child under 5 together under 1 net, often along with another child or the spouse (see Table 3). This pattern characterized both urban and rural households. The likelihood of various other family members sleeping under a household net depended principally on the number of nets the household owned. We ran analyses of the percent of various family member groupings under the net for households owning only 1 net and for households owning multiple nets. We also ran analyses looking at sleeping groups under a hanging net when the youngest slept under a baby net. Note that in these analyses, the figures represent the percent of nets rather than the percent of people.
|
When the family owned more than 1 net, others besides the mother and young child were more likely to be protected by a net. Most frequently, these were children aged 5–14 and, in Ghana only, adult men.
The larger proportion of nets in Nigeria having only a woman of reproductive age sleeping under it is at least partly explainable by the much higher use of baby nets in Nigeria than in the other countries.
Adult males were among the least likely to use a net and rarely slept under a net by themselves. From 1% to 3% of nets in 1-net households (but 8% in Senegal) were occupied by adult male(s) only. In multiple-net households, the percent of nets covering only adult male(s) ranged from 6% (Zambia) and 7% (Ethiopia) to 17% (Nigeria). It was also uncommon for an adult man and woman to share the net without also having a child under the net, even in multiple-net households, where the percent of nets with only a couple under it ranged from 5% in Ghana and Senegal to 12% in Ethiopia. Even in 2000, before intensive ITN-promotion efforts, men were among the least likely family members to be using the net.
Effect of baby net use on who sleeps under the household net.
Baby nets—small umbrella-like nets with a built-in frame that are only large enough to cover an infant or small 1-year old—are common in some countries and not in others. Baby nets are typically used to protect children < 1 year of age from flies, and they are virtually never treated. The AED/NetMark 2004 surveys found that baby nets were owned by 
40% of households in Nigeria, 25% in Ghana, 10% in Senegal, 8% in Mali, 6% in Ethiopia, and 1% in Zambia. Here we report the impact of baby nets only in Nigeria and Ghana, the 2 countries surveyed with a sufficient number of baby nets to potentially affect intra-household net-use patterns. (Note that baby nets are only large enough to cover an infant and were therefore excluded from prior analyses unless otherwise specified.)
In both Nigeria and Ghana, 27% of households owning a baby net had a child sleep under it the night prior to the survey. Having many baby nets unused the prior night is not surprising. Only an infant fits under a baby net, so the net cannot be used after the child reaches 12 months of age. Furthermore, some families use a baby net mainly to keep flies off the child during the day rather than for mosquito protection at night. We compare use of a family net in households where a baby net was used the prior night, with households where one was not used—either because one was not owned or because it was owned but not used.
When the infant slept under a baby net, the average number of people sleeping under the family net decreased: from 2.12 to 1.65 in Nigeria, and from 2.31 and 2.04 in Ghana (see Table 4). This means that another family member was less likely to "substitute" for the infant under the family net in Nigeria than in Ghana. In Ghana, an adult male was much more likely to join the mother under the net when the infant slept under a separate baby net. When a baby net was used, there was no change in the number of older children under the family net in either Nigeria or Ghana. Few older adults used nets in Nigeria or Ghana, regardless of whether a baby net was used for the family infant.
|
|
DISCUSSION AND CONCLUSIONS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Within the age segment "children under 5," the younger the child, the more likely he or she was to sleep under a net. Those under 2 years had the highest net-use rates of any family member—a positive finding because younger under-5s are more vulnerable to malaria than older under-5s. It is also good that young boys and girls in the African countries surveyed are equally likely to be protected by a net. This may not be the case in other regions of the world.
Within the segment "women of reproductive age," those who were pregnant were more likely to be under a net than those who were not (with the possible exception of Ethiopia, although the small number of net-owning households with a pregnant woman prohibits definitive conclusions). It is especially encouraging that pregnant women were much more likely to use a net in 2004 than they were in 2000, suggesting that ITN programs have been successful in educating people about the susceptibility of pregnant women to severe malaria and in allaying fears about the insecticide. Analysis of other variables in the AED/NetMark data sets shows that in 2000, a much greater proportion of women expressed fears about the safety of the insecticide than in 2004. Zambia made particularly impressive strides in this regard: in 2000, 27.3% of respondents (all women of reproductive age) said that the chemical on the net could be dangerous to the fetus or pregnant woman; in 2004 only 6.5% said this.11 The priority given to pregnant women conforms to public health recommendations, as pregnant women are more vulnerable to malaria than nonpregnant women.
Although it is pregnant women specifically rather than women of reproductive age generally who are most at risk of severe malaria, it is nonetheless extremely desirable to have all women of reproductive age sleep under treated nets nightly. Because a woman may not know exactly when her pregnancy begins, or may not wish to make her pregnancy public for several months or more, nightly ITN use ensures that she will be protected from the beginning of her pregnancy. Furthermore, a woman may be especially vulnerable to malaria for 2 months post-partum12 and need the protection an ITN affords even after her pregnancy ends. Thus, high net use rates among women of reproductive age constitute another positive finding.
The consistency of net-use patterns across the 6 African countries surveyed lends strong credence to the results and is highly encouraging.
These findings run counter to a popular belief that the man of the house often takes the net for his own use.13,14 It is possible that in other countries not surveyed, or in particular localized areas, adult men are most likely to use the net; however, in the countries studied here, it is clear that men are actually least likely to be using the family net. Nor do the data here support the idea that the young child is put under the net principally because the mother is sleeping there.5 In all countries, children under 5 were as likely as or more likely than a woman of reproductive age to be sleeping under a net.
Ideally, there would be a sufficient number of nets to cover all family members, although the data suggest that as additional nets are owned, fewer people sleep under 1 net. Given limited nets, it is good to know that most nets are being used to directly protect 2–3 people. If the nets are treated, they will be indirectly protecting additional family members not under the net by repelling mosquitoes away from the general vicinity of the net. Therefore, families that use nets, especially ITNs, are gaining protection for multiple household members.
The data also point to practices that need to be changed to maximize the public health impact of expanded net coverage. Although the aggregated national data used here show that the most vulnerable family members are given priority for net use, subanalyses by site show that this is not the case in all areas of each country. For example, in the Maiduguri site in Nigeria, few pregnant women slept under a net. Other analyses on the AED/NetMark data show that Maiduguri respondents had high levels of fear about the insecticide for pregnant women. Work will be needed in targeted parts of some countries to assuage fears and encourage pregnant women to use treated nets.
The baby net analyses appear to have several implications. When a baby net is used, fewer people use the household hanging net. Baby nets separate mothers and infants who would normally sleep together. The separation can have detrimental effects, such as limiting breastfeeding and preventing the regulation of infant body temperature via skin-to-skin contact. Baby nets have not been tested for effectiveness against malaria, and they are rarely treated. The money a family spends on a baby net could be better used for a larger hanging net that would serve the child for a longer period than infancy, allow other family members to sleep under it, and be treated. In countries where baby nets are popular, notably Nigeria and Ghana, malaria control committees may want to consider whether to discourage use of baby nets at least for use at night.
Perhaps the most important net-use issue that needs to be addressed by ITN programs is that of nets that go unused. When nets go unused, the impact of increased net coverage is diminished. Other studies also found that use can be lower than ownership, even in areas where nets were given free and there were intensive efforts to get people to sleep under them.15,16 We need a better understanding of why some households do not use their nets even during the rainy season when there are many mosquitoes. The finding of a sizable proportion of unused nets in some countries was unexpected, and we are conducting separate analyses exploring factors related to use and non-use of nets that are owned. Preliminary bivariate analyses show that treated nets were used more than untreated nets; newer nets were used more than older ones; and nets that were paid for were used more than free nets. NetMark qualitative research as well as other variables in the data indicate that some people find nets to be hot, difficult to hang or ill-suited to the sleeping place, or believe themselves protected by other means such as coils or aerosols. Among the 6 countries included in this study, these perceptions as well as the proportion of unused nets was most marked in Nigeria. ITN programs are likely to find even rapid qualitative research helpful in identifying and overcoming the net-use barriers specific to the populations they serve.
These findings point to the overall conclusion that ITN programs can focus on increasing access to and ownership of hanging, treated nets; on getting people to use them, and on encouraging year-round use. Families are already giving priority to their most vulnerable members, so this practice needs only periodic reinforcement rather than major emphasis in an ITN program strategy.
Received March 29, 2007. Accepted for publication August 9, 2007.
Acknowledgments: The authors thank the United States Agency for International Development for funding the NetMark project, including the collection of the data used for these analyses. We also thank AED’s Global Health, Population, and Nutrition Group for allocating Leadership Funds to cover staff time for writing this article. Special appreciation goes to Ms. Roshelle Payes who assisted with data analysis of all surveys and with preparation of the document. We also thank those who reviewed the manuscript: Martin Alilio, Lawrence Barat, Mr. David McGuire, Ms. Lonna Shafritz, and Susan Zimicki. We also thank Mr. Joseph Boniaszczuk of Research International (RI), South Africa, who gave valuable input on various aspects of the survey and worked with local RI personnel and local field teams to collect the data. Finally, we thank the mothers who gave their time to provide us with the information for this research.
* Address correspondence to Carol Baume, Academy for Educational Development, 1825 Connecticut Avenue NW, Washington, DC 20009-5721. E-mail: cbaume{at}aed.org 
Authors’ addresses: Carol Baume and Celeste Marin, Academy for Educational Development, 1825 Connecticut Avenue NW, Washington, DC 20009, Telephone: +1 (202) 884-8980, Fax: +1 (202) 884-8844, E-mails: cbaume{at}aed.org and cmarin{at}aed.org.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. P. Eisele, J. Keating, M. Littrell, D. Larsen, and K. Macintyre Assessment of Insecticide-Treated Bednet Use Among Children and Pregnant Women Across 15 Countries Using Standardized National Surveys Am J Trop Med Hyg, February 1, 2009; 80(2): 209 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Harvey, M. P. Olortegui, E. Leontsini, C. B. Pezo, L. M. O. Pezantes, and P. J. Winch The Whole World Will Be Able to See Us: Determining the Characteristics of a Culturally Appropriate Bed Net Among Mestizo Communities of the Peruvian Amazon Am J Trop Med Hyg, December 1, 2008; 79(6): 834 - 838. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
