Free distribution of insecticide treated bed nets to pregnant women in Kinshasa: an effective way to achieve 80% use by women and their newborns

Authors


Corresponding Author Audrey Pettifor, University of North Carolina, Epidemiology, McGavran Greenberg Bldg, Campus Box 7435, Chapel Hill, NC 27599, USA. Tel.: +1 919 966 7439; E-mail: apettif@email.unc.edu

Summary

Objective  To determine whether long lasting insecticide treated bed nets (LLINs) distributed free of charge to pregnant women at their first antenatal clinic visit in Kinshasa, DRC are used from the time of distribution to delivery and 6 months after delivery.

Methods  Women were enrolled into a cohort study at their first antenatal care (ANC) visit and provided LLINs free of charge. Reported use of these nets was then measured at the time of delivery (n = 328) and in a random sample of women (n = 100) 6 months post-delivery using an interviewer administered, structured questionnaire.

Results  At baseline, only 25% of women reported having slept under a bed net the night before the interview. At the time of delivery, after being provided an LLIN for free, this increased to 79%. Six months post-delivery (n = 100), 80% of women reported sleeping under a net with a child under the age of 5 the night before the interview.

Conclusions  Freely distributed bed nets are acceptable, feasible and result in high usage. Free distribution of bed nets during antenatal clinic visits may be a highly effective way to rapidly increase the use of bed nets among both pregnant women and their newborn infants in areas with high levels of ANC attendance.

Abstract

Objectif:  Déterminer si les moustiquaires traitées aux insecticides durables distribués gratuitement aux femmes enceintes lors de leur première visite dans une clinique prénatale à Kinshasa en RDC, sont utilisées à partir du moment de la distribution jusqu’à l’accouchement et 6 mois après l’accouchement.

Méthodes:  Les femmes ont été recrutées dans une étude de cohorte lors de leur première visite à la clinique et des moustiquaires leur ont été fournies gratuitement. L’usage rapporté de ces moustiquaires a ensuite été estimé au moment de l’accouchement (n = 328) et dans un échantillon aléatoire de femmes (n = 100), 6 mois après l’accouchement à l’aide d’un questionnaire d’interview structuré.

Résultats:  Au départ, seules 25% des femmes ont déclaré avoir dormi sous une moustiquaire la nuit avant l’interview. Au moment de l’accouchement, après avoir obtenu une moustiquaire gratuitement, ce chiffre est passéà 79%. Six mois après l’accouchement (n = 100), 80% des femmes ont rapporté avoir dormi sous une moustiquaire avec un enfant de moins de 5 ans la nuit précédant l’interview.

Conclusions:  La distribution gratuite des moustiquaires est acceptable, réalisable et procure des résultats d’utilisation élevée. La distribution gratuite de moustiquaires au cours de visites prénatales pourrait être un moyen efficace d’accroître rapidement l’usage des moustiquaires chez les femmes enceintes et leurs nouveau-nés dans les régions avec des niveaux élevés de fréquentation de cliniques prénatales.

Abstract

Objetivo:  Determinar si las mosquiteras impregnadas de larga duración (MILDs) distribuidas gratuitamente a mujeres embarazadas durante su primera visita a la clínica prenatal de Kinshasa, son utilizadas desde el momento de su entrega hasta 6 meses después del parto.

Métodos:  Las mujeres fueron reclutadas en un estudio de cohortes durante su primera visita a la clínica prenatal y se les entregó una MILD sin coste alguno. Se midió el uso reportado de las mosquiteras al momento del parto (n = 328) y, en una muestra de mujeres elegida al azar (n = 100), 6 meses después de haber dado a luz, mediante un entrevistador con un cuestionario estructurado.

Resultados:  Al comienzo del estudio solo un 25% de las mujeres reportaban haber dormido bajo una mosquitera la noche antes de la entrevista. En el momento del parto, después de haber recibido un MILD de forma gratuita, esta cifra aumentó a un 79%. Seis meses después del parto (n = 100), un 80% de las mujeres reportaron haber dormido bajo una mosquitera con un niño menor de 5 años la noche antes de la entrevista.

Conclusiones:  La distribución gratuita de mosquiteras impregnadas es aceptable, posible y resulta en un alto uso. La distribución gratuita de MILD durante las visitas a clínicas prenatales, en áreas con altos niveles de asistencia a estos centros, podría ser una forma altamente eficaz de aumentar rápidamente el uso de mosquiteras entre mujeres embarazadas y sus recién nacidos.

Introduction

Long lasting insecticidal bed nets (LLINs) are one of the most effective ways to prevent malaria and associated morbidity and mortality resulting from the disease (Lengeler 2004; Fegan et al. 2007). Use of insecticide treated bed nets (ITNs) was associated with a 44% reduction in mortality among young children in Kenya (Fegan et al. 2007). The Roll Back Malaria (RBM) Campaign seeks to increase LLIN coverage among pregnant women and children under 5 to 80% by 2010 (Roll Back Malaria 2005). Coverage of LLINs in most malaria endemic countries in Africa does not meet this goal, but numbers are increasing in some areas where programmes are working aggressively to increase access to LLINs (Baume & Marin 2007; Noor et al. 2007).

How best to achieve 80% coverage of vulnerable populations with LLINs has been debated (Curtis et al. 2003; Lengeler & deSavigny 2007; Teklehaimanot et al. 2007). One programme that demonstrated significant increase in bed net coverage in Kenya attributed this primarily to massive free distribution of nets (Noor et al. 2007). We undertook this study in 2005 to evaluate use of LLINs distributed free of charge to pregnant women at their first antenatal clinic visit in Kinshasa, Democratic Republic of the Congo (DRC) from the time of distribution to delivery and in the 6 months after delivery.

Methods

Details of the study are published elsewhere (Pettifor et al. 2008), but beginning in November 2005, 855 LLINs (PermaNet 2.0) were distributed to pregnant women attending the Bomoi maternity in Kinshasa, DRC. LLINs were provided free of charge to all pregnant women attending their first antenatal care (ANC) visit. Women attending the Bomoi maternity for their first ANC visit, who were 18 years old, or older, were eligible to participate in the study. Three hundred and sixty two women were randomly selected to take part in the study. Women who provided informed consent to participate were given an appointment to complete the baseline interview; bed nets were provided after the baseline interview was completed. All women who were screened for the study, regardless of whether they enrolled in the study, received one LLIN, material to hang the net, an educational leaflet demonstrating how to properly hang the net, and a brief educational session on effective malaria prevention (including transmission and symptoms of malaria, and use of bed nets). Women who enrolled in the study provided contact details so that they could be located at the time of delivery, if not found at the hospital.

All participants who completed the baseline interview were to be interviewed after delivery. The estimated due date of delivery was collected at baseline so that study staff would have an approximate time to conduct the delivery interview. When deliveries were expected, study staff contacted Bomoi Maternity daily for a list of study participants who had delivered in the previous 24 h. Interviews were conducted at Bomoi Maternity at the participant’s earliest convenience and prior to her release from the clinic (normally within 72 h of delivery). Participants who delivered at another clinic or hospital were interviewed during their stay, when possible, or in their home following their release. Staff contacted participants at their home to arrange an interview if they did not deliver at Bomoi Maternity or contact Bomoi Maternity (or UNC study staff) about their delivery. Staff conducted a total of 328 delivery interviews.

At the time of the delivery interviews, a random sample of 115 women out of the first 205 were invited to participate in a 6-month post-delivery interview and a date was set for the interview at that time. Addresses were confirmed with women selected for the home interview at the time of delivery. Participants were interviewed in their home to determine the continued use of the nets post-delivery. There were 103 in-home interviews, but analysis was restricted to 100 women due to missing data.

Study participants were interviewed in Lingala by trained interviewers using a pre-tested, structured questionnaire. The questionnaire covered socio-demographics; bed net ownership; information about individuals using bed nets; compliance with net use; knowledge and attitudes about malaria and bed net use; and perceived control over malaria prevention. The delivery interviews were conducted from April through October and home interviews occurred in October.

Institutional review board (IRB) approval was obtained from University of North Carolina and the Kinshasa School of Public Health.

Analysis

In this study the term bed nets is used when referring to bed nets prior to our intervention and LLINs to refer to the nets that we distributed. At delivery and the home interview we measured the use of the LLIN provided at baseline, as well as use of other bed nets in the home. For the purposes of the analysis, we used having slept under a net and did not distinguish between the LLIN and other nets. When we refer specifically to an LLIN we are referring to the net we provided, otherwise we analyzed data concerning our nets and other nets women may have used.

We examined the frequency and distribution of baseline socio-demographic characteristics for women completing the interview at delivery. Frequencies and chi-square tests were used to determine differences in net usage among baseline, delivery and home interviews, as well as attitude changes toward malaria from baseline to delivery. Potential factors associated with bed net use at the time of delivery (including social support, net position and satisfaction with the net) were also examined. To identify factors we hypothesized to be associated with net usage the night before delivery, we constructed separate multivariate logistic regression models using a backward selection process. Based on a priori knowledge, we included socio-demographic variables in the model and then used backward elimination to identify other covariates. An alpha level of 0.05 was used to determine which variables remained in the model. Statistical analyses were conducted using STATA version 9.

Context in Kinshasa, DRC

This study was conducted at Bomoi Maternity which is managed by the Salvation Army. It is located in a densely populated, downscale neighbourhood in eastern Kinshasa. The Bomoi catchment area is fairly typical of many neighbourhoods in Kinshasa in that most inhabitants get by with modest to severely deficient incomes, living conditions and infrastructure. At Bomoi, approximately 250 new ANC patients are seen per month. Overall, HIV prevalence in this population was 2–3% per month for the later part of 2004 (Musuamba et al. 2006). In 2004, Bomoi clinic registered approximately 149–200 deliveries per month. A review of 47 ANC records at Bomoi conducted in January of 2005 found that 28% of women came for their first ANC visit at 6 months gestation, 25% of women came at 7 months and 25% at 8 months, 17% came at 5 months and about 4% came at less than 5 months (unpublished data). In 2001, approximately 85% of pregnant women living in Kinshasa reported at least one ANC visit (Ministre du plan et de la reconstruction RDC 2002). According to the 2007 Demographic and Health Surveys, 95.7% of pregnant women in Kinshasa received some ANC provided by qualified staff (Measure DHS2007).

Malaria in the DRC is endemic with stable transmission of malaria year round. Seasonal fluctuations in malaria transmission occur during the rainy season from September/October to May. Malaria is one of the primary causes of morbidity and mortality in the DRC particularly among young children and pregnant women.

Distribution of LLINs is one of the main components of the RBM strategy in the DRC. At the start of this study, bed nets were available in Kinshasa through social marketing and commercial outlets. Shortly after, LLINs began to be distributed through organizations in DRC working in partnership with the Global Fund to fight AIDS, Tuberculosis and Malaria (Global Fund) and RBM.

Results

The majority of the women participating were 25–34 years old (53.4%), married (78.5%) and cohabiting (78.8%) (Table 1). About one third of the women (32.2%) experienced their first pregnancy. Most women (59.2%) had some secondary school and 46% had never been employed. The median gestational age of women at receipt of the bed net was 5.0 months [interquartile range (IQR) = 2.0].

Table 1.   Demographic characteristics of women seeking antenatal care in malaria-endemic Kinshasa, DRC enrolled in bed net study (n = 326)
 % (n)*
  1. *Percents may not sum to 100 due to missing values.

Age
 18–24 years31.9 (104)
 25–34 years53.4 (174)
 35 years and over14.7 (48)
Marital status
 Single21.5 (70)
 Married78.5 (256)
Lives with partner
 No13.8 (45)
 Yes78.8 (257)
First pregnancy
 No67.8 (221)
 Yes32.2 (105)
Highest level of education attained
 Completed less than primary school7.1 (23)
 Completed primary school2.5 (8)
 Some secondary school59.2 (193)
 Completed secondary school25.8 (84)
 Completed university education4.6 (15)
Employment status
 Never employed46.0 (150)
 Not currently employed18.1 (59)
 Employed35.9 (117)
Gestational age in months at baseline
 Median (interquartile range, IQR)5.0 (2.0)
 Min, max 2, 8

At baseline, only 25% of women enrolled in the study reported sleeping under a bed net the night before and one third reported owning a net. At the time of delivery, after being provided an LLIN free of charge, this increased to 79% of women reporting that they had slept under a bed net the night before they came to the hospital to deliver (Table 2). For the subset of women (n = 100) who were visited 6 months post-delivery in their homes, 85% reported sleeping under a bed net the night before. At the home interview, 80% of women reported sleeping under a net with a child under the age of 5, most likely the newborn baby. When asked at delivery about use in the last week, 72% reported sleeping under a bed net all seven nights (100% use), 19% reported that they had slept under it at least one time but not every night and 9% reported not sleeping under it at all. Use in the past month as reported at delivery was slightly lower, 62% reported always sleeping under a bed net, 30% reported sometimes sleeping under a net and 7% reported never sleeping under it. Use in the last week and month were fairly similar at the 6 month post-delivery home visit (Table 2). Eighty-two percent of women reported thinking it was very easy to use the LLIN we gave them.

Table 2.   Use of bed nets† at baseline, delivery (n = 326) and 6 months post-partum (n = 100) among women in Kinshasa, DRC
 Baseline n‡ (%) Delivery n (%)Home n (%)
  1. *Percents may not sum to 100 due to missing values.

  2. †Long lasting insecticidal bed nets at delivery and 6 months post-partum, unspecified bed nets at baseline.

  3. ‡Includes only women who owned a net prior to the baseline interview.

  4. §P ≤ 0.001 compared to baseline.

  5. ¶P ≤ 0.01 compared to baseline.

Use the night before82 (25)258 (79)§85 (85)§
Use in the last week
 Never18 (16)30 (9)¶11 (11)
 More than 0 but   less than 100%31 (30)61 (19)¶15 (15)
 100%66 (57)234 (72)¶74 (74)
Use in last month*
 Always49 (15)203 (62)¶68 (68)§
 Sometimes56 (17)99 (30)¶15 (25)§
 Never220 (67)24 (7)¶17 (7)§

The median number of bed nets owned by women was 1.0 nets (IQR = 2.0) and the median number of individuals sleeping under the LLIN we gave them was 2.0 (IQR = 2.0) (Table 3). Other individuals that women reported sleeping under the net the night before going to the hospital for delivery included: husband (61%), child under the age of five (40%), a child age 5–18 (17%), an adult relative (7%) and an adult guest (1%). At the home interview, in homes where a child under-five was reported to live (97% of homes), 85% of children under-five slept under a bed net the night before the home interview.

Table 3.   Use of long lasting insecticidal bed nets distributed free of charge as reported by women at the time of delivery in Kinshasa, DRC
 % (n)
  1. *Other includes: Indoor heat leads us to stay out for very long before sleeping under a mosquito net; Slept at a mourning/prayer wake; Husband said no; I have nothing to tie a mosquito net with; Waiting until we had our own home; Gave the net away.

  2. †Other includes: I can sleep without being disturbed; White colour; Registered trademark.

  3. ‡Other includes: Net was not killing mosquitoes; Net was too small; Had a reaction to net (cough or itch); Net was too soft.

  4. #Participants were allowed to select more than one option.

Do you still have the mosquito net we gave you?
 No4.0 (13)
 Yes96.0 (313)
Number of bed nets owned (median, IQR)1.0 (1.0)
Number of individuals sleeping under LLIN (median, IQR)2.0 (2.0)
Reasons for not sleeping under a mosquito net the night before coming to the hospital (unprompted)# (n = 68)
 Don’t have a net13.2 (9)
 It’s hot to sleep under a mosquito net20.6 (14)
 It’s difficult to breathe under a mosquito net17.6 (12)
 There isn’t enough place in the bedroom to tie a mosquito net8.8 (6)
 Experiencing labour pains32.4 (22)
 Saving net for use after delivery14.7 (10)
 Net was lost or stolen10.3 (7)
 Net was dirty7.4 (5)
 Slept in a place other than my home7.4 (5)
 Other*22.1 (15)
Has anyone in your household slept under the mosquito net since we gave it to you?
 No9.5 (31)
 Yes90.5 (295)
Reasons for no one sleeping under the mosquito net (unprompted)# (n = 31)
 I do not have a net6.5 (2)
 There isn’t enough place in the bedroom to tie a mosquito net29.0 (9)
 Waiting for the newborn/It was a gift for the baby58.1 (18)
 Net was stolen16.1 (5)
 I already had a bed net6.5 (2)
 I don’t have my own home/I live with my parents9.7 (3)
Who decided the net would not be used (unprompted, n = 30)
 Respondent60.0 (18)
 Partner/Husband23.3 (7)
 Respondent and husband together23.3 (7)
How easy was it for you to use the mosquito net we gave you? (n = 295)
 Very easy82.0 (242)
 Not really easy11.9 (35)
 A bit difficult6.1 (18)
Was your spouse/husband supportive to the idea of sleeping under a mosquito net? (n = 295)
 He really supported me79.0 (233)
 He supported me anyway11.2 (33)
 He didn’t really support me2.0 (6)
 He didn’t support me at all3.7 (11)
 I no longer have a partner/My partner-husband does not live with me4.1 (12)
Who selected those who slept under the mosquito net? (n = 295) (unprompted)
 I, the respondent50.9 (150)
 My partner/husband25.1 (74)
 My parents-in-law0.7 (2)
 My husband and I23.1 (68)
 Other people0.3 (1)
Is your mosquito net tied permanently or do you remove and tie it every night? (n = 295)
 I left it tied since I tied it95.9 (283)
 I remove and tie it every night4.1 (12)
Do the same people sleep under the net each night or does it change from night to night? (n = 295)
 Same people every night98.6 (291)
 It changes from night to night0.7 (2)
Now that you are no longer pregnant, how likely is it that you still sleep under the bed net regularly? (n = 236)
 Really likely52.2 (170)
 Likely44.2 (144)
 Quite likely3.1 (10)
 Not likely0.6 (2)
Have you noticed that there are fewer, more, or the same number of mosquitoes in your house from the time we gave you a mosquito net until now?
 Fewer83.4 (272)
 Same10.7 (35)
 More5.5 (18)
What pleased you the most with the mosquito net we gave you? (n = 326)
 It protected me from getting malaria70.3 (229)
 It was for free8.6 (28)
 It protected my baby from getting malaria8.3 (27)
 It killed mosquitoes/insects7.1 (23)
 Protection2.5 (8)
 Other†2.1 (7)
What pleased you least about the mosquito net we gave you? (n = 326)
 It smells bad2.8 (9)
 It is hot to sleep under it8.3 (27)
 It’s difficult to tie it2.8 (9)
 It’s difficult to breathe under it2.5 (8)
 The house is too small to have a place to tie it2.8 (9)
 You didn’t give me many mosquito nets9.5 (31)
 Other‡71.5 (233)

When examining the use of a bed net at delivery, by ownership of a net at enrolment into the study, 83% of women who owned a net reported sleeping under a bed net the night before compared to 77% of women who did not own a net at enrolment. Among women who owned a net at enrolment, 68% reported always sleeping under a net in the past month compared to 59% of women who did not own a net at enrolment. Among women who reported sleeping under a bed net at delivery (79%), 76.7% reported sleeping under the LLIN and 23.3% slept under another bed net (83% of women reported that these nets were treated with insecticide).

During the home visit, interviewers verified whether the net was in the home; they observed a bed net in 87% of the homes. In 69% of the homes the net was hanging, and in 19% holes were observed in the bed net. Among women who reported sleeping under a net the night before the home interview, 74% had a net that was observed hanging in the home.

Only 13 (4%) of women reported no longer owning the LLIN we gave them at their enrolment visit (first ANC visit) (Table 3). Of these women, 10 (77%) reported the net had been stolen, one (8%) reported it was lost and the remaining women reported other reasons.

The main reasons reported for not sleeping under the net the night before delivery included discomfort due to labour pains (32.4%), being too hot under the net (20.6%), finding it difficult to breathe under the net (17.6%) and saving the net for use after delivery (14.7%) (Table 3). In the instances where no one used the net, the majority of women (58.1%) reported they were waiting for the baby to be born before using it or it was a gift for the baby (Table 3). Twenty-nine percent of women reported that there was not enough space in the bedroom to tie the net (Table 3).

Among the 30 women who reported not using the LLIN, 60% reported that they were the ones who decided not to use the net, 23.3% reported that their husband or partner decided and 10% reported another person decided it would not be used. The majority of women (79%) reported that their partner was very supportive of using the LLIN (Table 3).

At the delivery interview, 96% of the women reported that they left the LLIN tied permanently to the same place and 99% reported that the same people slept under the net every night (i.e. it did not change from night to night). Eighty-three percent of women reported noticing fewer mosquitoes in their home since they were given the LLIN (Table 3).

Overall, women perceived bed net use and malaria prevention positively at baseline and at delivery (Table 4). In some instances there were substantial changes in attitudes between baseline and delivery, for example, at baseline 22.1% of women reported that it was useless to use a mosquito net because you can suffer from malaria anyway, this declined to only 5.2% at delivery (P < 0.001). At baseline 8.1% of women reported thinking it was difficult to use a mosquito net compared to only 2.2% at delivery (P = 0.001) (Table 4). Fewer women at delivery reported that they would prefer to use a net they had bought compared to one that was given for free, and that it is more difficult to use a net than to take drugs when one suffers from malaria (Table 4).

Table 4.   Attitudes of women towards malaria and bed net use at baseline and delivery, Kinshasa, DRC 2005/06
 Baseline %Delivery %Chi-square (P-value)
  1. *Significant at the 0.05 level.

In your own opinion, do you believe
 That you are worried about getting malaria99.499.71.86 (0.2)
 It is important and beneficial to sleep under a mosquito net every night99.499.70.27 (0.6)
 That sleeping under a mosquito net is a good way to protect yourself from malaria98.998.50.01 (0.9)
 You would prefer using a mosquito net you have bought better than one given to you for free13.97.74.56 (0.03)*
 That you couldn’t afford to buy a mosquito net if you did not receive one for free15.816.60.10 (0.7)
 It is useless to use a mosquito net because you can suffer from malaria anyway22.15.239.92 (<0.001)*
 That obtaining a mosquito net in the community where you live is easy28.528.81.35 (0.9)
 That malaria is a serious disease97.298.51.35 (0.2)
 That it is more difficult to use a mosquito net than taking drugs when you suffer from malaria6.12.26.38 (0.01)*
 Hat children suffer from malaria more than adults do79.175.20.32 (0.6)
 It’s difficult to use a mosquito net8.12.211.89 (0.001)*
 That mosquito nets that you have to buy are better quality (more effective in preventing malaria) than nets that are given out for free5.12.81.62 (0.2)

We examined factors associated with using a bed net at the delivery interview (Table 5). In multivariable models, women who reported that this was their first pregnancy were less likely to have slept under a bed net the night before delivery (AOR 0.46, 95% CI 0.22–0.97) and women who reported owning more than one net were more likely to have slept under a bed net (AOR 1.9, 95% CI 1.0–3.8). At the home interview, primiparity did not appear to influence whether a woman had slept under a net the night before the interview (OR 1.0; 95% CI 0.3–3.3), although the measure is imprecise due to small sample size (data not shown). We could not examine the influence of multiple net ownership at the home interview due to the small sample.

Table 5.   Factors associated with reported LLIN use the night before delivery
VariableOR (95% CI)AOR (95% CI)
Age
 17–240.70 (0.39,1.3)1.1 (0.53, 2.4)
 25–341.01.0
 35 and over1.0 (0.45, 2.3)0.83 (0.35, 2.0)
First pregnancy
 No1.01.0
 Yes0.52 (0.30, 0.90)0.46 (0.22, 0.97)
Highest level of education completed
 Less than secondary school1.01.0
 Secondary school or beyond1.4 (0.75, 2.6)1.8 (0.88, 3.6)
Employment status
 Never had a job1.01.0
 Currently unemployed0.88 (0.42,1.8)0.94 (0.42, 2.1)
 Currently employed0.92 (0.51, 1.7)1.0 (0.51, 2.0)
Living with partner
 No1.01.0
 Yes2.0 (0.97, 4.0)1.8 (0.82, 4.0)
Marital status
 Single1.01.0
 Married1.4 (0.77, 2.6)0.78 (0.32, 1.9)
Number of nets owned
 <=11.01.0
 >11.8 (1.0, 3.4)1.9 (1.0, 3.8)

Discussion

This study found that reported use of bed nets in this population after free distribution was high at the time of delivery and 6-months post-delivery. Use of the bed nets the night before the interview ranged between 79 and 85% at the two survey time points meeting the Roll Back Malaria 2010 targets of 80% coverage for pregnant women. Provision of the LLINs to women free of charge at their first ANC visit increased the reported use of bed nets dramatically, with 79% of women reporting sleeping under a net the night before delivery compared to 25% the night before enrolment. Importantly, reported use of the LLINs did not wane post-delivery and was slightly higher post-partum. At the 6-month post-partum interview, 80% of women reported sleeping under a bed net with a child under the age of 5, most likely the newborn which is customary in Kinshasa.

Efforts to increase net use among pregnant women and their children by the RBM Campaign and Global Fund have recently led to significant increases in net use among these populations. In a cohort study in Kenya, bed net coverage in children under 5 was 7.1% in 2004 when the predominant source of nets was the commercial retail sector (Noor et al. 2007). This increased to 23.5% at the end of 2005 after the expansion and distribution of heavily subsidized LLINs through clinics. In 2006, LLIN coverage increased to 67.3% after large-scale mass distribution of LLINs free of charge to children. In an evaluation of RBM activities in Ghana, net use by pregnant or nursing mothers increased to 58% in 2003 from 29% in 2000 after RBM activities (Owusu-Agyei et al. 2007).

Overall study participants in Kinshasa reported few barriers to net use, attitudes towards net use were overwhelmingly positive and male partners in general were reported to be supportive of use. Surprisingly, most women in this population reported playing a role in decision making over use of bed nets. In contrast to a study conducted on net use in Kenya (Alaii et al. 2003a), most women in our study reported that their husbands were supportive of them sleeping under the nets and, of the women who did not sleep under the net the night before the interview, only 23% reported that their husband made this decision. In addition, few of our participants reported problems such as disruption of sleeping arrangements and the need to hang and re-hang nets as described by Alaii et al. (2003a,b) in Kenya. Overall, attitudes about using bed nets and the importance of malaria prevention were positive in this population, even at baseline. Importantly, use of nets only further increased positive attitudes towards nets, particularly, reducing concerns that nets were difficult to use. But, the educational messages given to pregnant women regarding LLINs should be further improved because several women who did not slept under the LLIN reported saving the net for use after delivery, or waiting for the newborn because the net was a gift for the baby.

In multivariable analyses, we found that women who owned more than one bed net were more likely to have slept under a net the night before the interview. Given that the mean household size in this study was 5 (range 0–20) and that the average number of people sleeping under a bed net was 2, it makes sense that more nets would increase coverage. Other studies on bed net use in Africa have found similar bed net coverage of between 1.7–2.0 persons covered per net (Baume & Marin 2007; Teklehaimanot et al. 2007). It has been estimated that for a typical family of five people approximately three nets will be needed to cover the entire household (Teklehaimanot et al. 2007).

We found that primiparous women were less likely to have slept under a net at the delivery interview than women who were multiparous. Similar results have been found in other studies (Marchant et al. 2002; ter Kuile et al. 2003; van Eijk et al. 2005). It is not clear why primiparous women were less likely to use the nets. It is possible that they were not as knowledgeable of the risks of malaria during pregnancy as women who had been pregnant before; but all women in our study received the same information on malaria and pregnancy risk, irrespective of parity. Other studies have hypothesized that younger women are more likely to be primiparous and that their living situations may be more unstable and they may move to other households during pregnancy (Marchant et al. 2002; van Eijk et al. 2005). It has also been hypothesized that younger women may also have less priority with regard to net use due to lower social status in the family (Alaii et al. 2003a; van Eijk et al. 2005), this does not appear to be the case in this population: very few women reported that other family members slept under the nets, and most women reported making the decision about who would sleep under the net, either independently or jointly with their spouses. Interestingly, in the home interview we did not observe a difference in net use by parity, but the measure was imprecise. It could be that once primiparous women have a newborn infant at home, they have a greater incentive to sleep under the net.

While informative, the current study is not without limitations. The study sample was drawn from a small, clinic-based sample, which limits the statistical power of the results. Nevertheless, we achieved high rates of reported LLIN use during pregnancy and demonstrated sustained use at delivery and 6 months post-delivery. Use of a clinic-based sample also limits the generalizability of the study results; however, the most recent statistics from the Demographic Health Survey (DHS) indicate that 96% of women in Kinshasa attend ANC, thus distribution of bed nets through ANC services would reach a large proportion of pregnant women in Kinshasa. Although we cannot generalize these results to the DRC population, they are likely to be relevant to many urban and peri-urban neighbourhoods in Kinshasa. Another limitation is that this data is based on self-report. At the home interview, interviewers were asked to record whether they observed the net hanging in the home. Only 74% of interviewers reported observing a hanging net in the homes of women who reported sleeping under a net the night before. This may mean that women over-reported use of the net but it may also be that women did not leave the net hanging-up. It should also be noted that 19% of nets at the home interview were observed to have holes in them which diminishes the effectiveness of the nets and emphasizes the importance of repairing or replacing nets when they are damaged.

Another potential limitation to this study is that women were asked to return to the clinic for a separate appointment to complete the baseline interview. The appointment to complete the baseline interview and receive the LLIN was scheduled 3–4 days after the initial antenatal visit for logistical reasons. It is unlikely that this arrangement had any impact on reported bed net use. Recall bias may have played a role in this study as all information was self-reported by the women. But, women were asked specifically about use the night before delivery, keeping recall bias at a minimum.

In theory, seasonality could have impacted the study results. It is important to note that the LLINs were distributed between November and January, delivery interviews were conducted between April to September, and home interviews from July to October. Malaria is endemic in Kinshasa year round, but the rainy season from October to April exacerbates this. Thus some women’s reported LLIN use at the delivery interview reflected use during the rainy reason, when mosquitoes may be more abundant and use thought to be greater; however, many of the follow-up interviews were conducted during the ‘dry season’ and net use remained equally high.

In many African countries, bed net ownership and usage is low (less than 10 percent in some areas) (UNICEF, WHO 2003; Wiseman et al. 2007), but coverage and use is increasing in the wake of the RBM and Global Fund efforts. ITN distribution practices range from sale to free distribution. When we began this study, bed nets were available through the commercial sector and through social marketing in Kinshasa. We undertook this study due to the belief, locally as well as globally, that nets distributed free of charge would not be valued or used by participants. Contrary to beliefs of local stakeholders, the vast majority of women reported sleeping under the bed net and reported positive attitudes towards use of bed nets. Two recent studies have found that free-distribution of LLINs has resulted in substantial increases in net coverage compared to subsidized, ‘social marketing’ approaches (Maxwell et al. 2006; Noor et al. 2006). Even in areas where LLINs are subsidized, bed nets remain out of reach for those who are the poorest and often at highest risk for malaria (D’Alessandro & Coosemans 2003).

Our data add to a growing body of evidence finding that LLIN distribution free of charge is acceptable, feasible, results in high use (Guyatt & Ochola 2003; Noor et al. 2007) and may be an extremely effective way to reach the high coverage goals set by RBM (Lengeler & deSavigny 2007; Noor et al. 2007). It is likely that distribution strategies utilizing a range of approaches, both subsidized sales as well as free distribution, will be most successful in reaching mass coverage (Lengeler & deSavigny 2007). But, free distribution during ANC visits may be among the most effective ways to provide bed nets to pregnant women and their newborn infants in areas where ANC use is high (Noor et al. 2007). Antenatal care might provide a special opportunity to reach pregnant women and their partners when they might be most open to using LLINs to protect the health of the mother-offspring dyad.

Acknowledgements

We thank Ms Karen Hawkins Reed for her support. This study was funded by the United States Centers for Disease Control and Prevention Global AIDS Program U62/CCU422422.

Ancillary