 |
| About Bioline | All Journals | Testimonials | Membership | News |
|
||||||
|
||||||
East African Journal of Public Heath, Vol. 1, No. 1, Oct, 2004, pp. 18-23 Willingness to Pay for ‘Olyset’ Bednets Among Formal Employees and Rural Peasants in Korogwe and Muheza Districts, Northeastern Tanzania Godfrey M Mubyazi1, Vivien Barongo2, Kassembe Mdira2 and Kato J Njunwa3 1Dept. of Health Systems and Policy Research, 2NIMR Ubwari
Research Station Muheza, 3Kigali Health Institute, Rwanda Code Number: lp04004 Abstract Owing to their impact on reducing mosquito nuisance, morbidity and mortality due to malaria, insecticide treated bednets are promoted to be used at household level and in schools, military camps, health facility wards, and guest houses. Nevertheless, the large-scale use of such materials is constrained by demand and supply factors. The contemporary debate is that people might be willing to pay for bednets but their short supply could constrain their actual use. Also it is concerned that supplying bednets in the market is one thing but whether people are prepared to purchase them is another thing since the actual uptake of bednets is determined by personal or household’s disposable income, and their prices, their shapes and sizes as well as people’s attitudes. This paper reports on a study of willingness to pay (WTP) for Olyset bednets among formal employees and villagers predominantly engaged in subsistence farming in Korogwe and Muheza districts in Tanzania. Generally, the study found that, fixing a price of Tshs. 3,500 or more per one bednet to be paid at once is practically undesirable to many peasants even if the majority of them might have had stated their WTP that price. On the other hand, allowing prepayments or payments by instalments has the advantage of promoting sales and utilization of bednets among formal and informal employees. The study recommends demand creation mechanisms like social marketing towards changing public perceptions and attitudes for better utilization of bednets. Key Words: Willingness to pay, insecticide-treated nets, poverty, malaria, Tanzania Background Estimates indicate that over 1,000,000 people in sub-Saharan Africa (SSA) die every year due to malaria, the mostly affected groups being children under the age of five years and pregnant women. Tanzania is one of SSA countries where malaria is highly endemic, statistics provided by the Ministry of Health (1) indicating that 100,000 die of malaria every year. Malaria has negative impacts on the socio- economic development because of consequences associated with recurring episodes that reduces people’s ability to work, costs incurred on purchase of drugs, transport, admissions and funerals (in case of death) events. In recent years we have witnessed WHO/Roll Back Malaria, UNICEF national malaria control programs and other organizations increasing their advocacy for insecticide treated bednets (ITBs) for the prevention of malaria. Nevertheless, the utilization rate of bednets in rural areas is reported to be as low as 10 - 20%. One of the barriers to bednets use is the low ability to pay and the low acceptability to majority of populations in SSA (2). It is believed that the problem of affordability can be significantly reduced if payment mechanisms other than cash on delivery are explored (3). Also, Lines (2) argues that, utilization of bednets for malaria prevention purposes can be enhanced through various promotion strategies. It has been observed that, the use of ITBs can reduce malaria transmission to a remarkable degree as opposed to over-reliance on chemotherapy that seems to be inefficient in malaria control. Literature says that WTP in theory does not necessarily imply WTP in reality. WTP is defined as the maximum price that an individual is willing to dispose of in order to obtain a product or service. The technique that has popularly been used in the assessment of WTP is the contingent valuation (4-6). According to Klose (6) the CV techniques employs the use of various questions including open-ended, close-ended binary or close-ended polychotomous, bidding game, binary-with-follow-up as well as polychotomous-with follow-up questions. The binary-with-follow-up and the bidding game techniques are the mostly recommended (7-8), although it is also warned that most CV methods are still experimental, therefore need for interpreting results cautiously becomes obvious (9). This paper describes a mixture open-ended-, close-ended polychotomous (without follow-up) and binary-with-follow-up questions applied in the elicitation of WTP of people residing in three rural villages in Korogwe and Muheza districts and formal employees living in town centres of Muheza district in northeastern Tanzania. Materials and Methods Objectives, Study Areas and Populations The study commenced in March 2001. Originally, the study was targeted to household members in rural villages, most of whom being employed in small-scale (mainly subsistence) farming. Initially, there were only 1000 Olyset bednets available for this study that was targeted to cover three villages. But it was decided later to extend the study to formal employees after receiving a grant of 1000 more bednets from Sumitomo Chemical Company of Japan. Among other objectives, the study was intended to: (i) assess the utilization of bednets at household level (ii) compare individual respondents’ stated WTP with their actual WTP for Olyset bednets (iii) explore the alternative payment mechanisms for ITNs (but not specifically bednets) preferred by the respondents. The study was conducted in two malaria endemic districts – Korogwe and Muheza (both located in Tanga Region) in northeastern Tanzania. The villages studied are Mkanyageni (Muheza District) and Mgombezi and Mazinde (Korogwe District). The residents in these villages are mainly the Bondei (in Muheza) and Sambaa, Zigua and Pare (in Korogwe). The main crops grown in both districts are maize, rice, maize, cassava, rice, potatoes, oranges and coconut, and rainfall ranges from slightly less than 1200mm annually to 3000mm in a few areas. Both district have a tarmac road connecting the district capital with neighboring regions of Kilimanjaro and Dar Es Salaam, but most of the rural roads are dusty and become muddy and sleepy during rain seasons. By the time of study was started, estimates made based on previous small-scale community surveys, indicated that the rate of bednets use in Muheza District was <10% of all households in villages (Caroline Maxwell, Kato Njunwa & Chris Curtis, person communication), and was considered to be the same in Korogwe District. Sampling Village Study The three villages studied were strategically selected, classified as ‘rural’ but located very far from each other. The decision to choose two villages in Korogwe district and only one in Muheza district was made because Korogwe had so far experienced fewer studies related to bednets than Muheza where researchers from the MRC (UK e.g. Maxwell and Curtis) and from the National Institute For Medical Research (NIMR) – Amani Centre have been operating for a number of years. During the baseline survey in each village, 416 households were randomly selected whereby one head (either a male or female) or their representatives were interviewed. The assumption was that about, based on the baseline data on the estimated rate of bednets use (<10%) at village level and initially 1000 bednets available for the study, an average of 330 bednets would be purchased per study village. Although the target was to interview 140 heads of households per village, in total the householders interviewed were: Mazinde (141), Mgombezi (135), and Mkanyageni (140). In addition, 3 focus group discussions (FGDs) were held in each village with selected members of households (but not those already interviewed individually). The baseline phase also interviewed 71 retail shop/kiosk keepers on ITBs sales and utilization aspects. The information collected from households during the baseline phase (for those who stated to be willing to pay) provided the basis for selecting a number of respondents who were followed-up later to confirm their WTP in actual sense. Also, it provided the basis for identifying sales agents (shops) where the study bednets came to be supplied so that those willing to buy them at their own time could do it as conveniently as possible, as suggested by the majority of the residents interviewed. The follow-up phase adopted a systematic random sampling method to arrive at 26 households out of a total number of (140) household respondents in each village who hypothetically stated to be willing to pay for the bednets at the stated price of Tshs. 3,500 per one net. The selection 10 shops who later on acted as agents for retail selling of the study bednets during the intervention phase was made in consultation with the ward and village government leaders and involved the shop/kiosk-keepers concerned. A similar net approach was used in the KINET Project in Kilombero. Formal employees study As time went on while the study was ongoing, the research team realized formal employees had interest in Olyset bednets. Of these employees, some belonged to NIMR, particularly at the Amani Centre, others worked at the National Micro-finance Bank (NMB)- Muheza Branch, and some police at the Muheza Town Police Station. It was expected that, all these would be interviewed about their WTP before they could be offered the nets they required. Interview was just held with 59 NIMR staff (24 at Ubwari Research Station and 35 at Amani Centre). Similarly, seven staff at the NMB responded to the interview. But as the high malaria season was already confronted, the research team experienced increased requests from employees who were insisting to be provided with the bednets first and then the interview to come thereafter. Coincidently, some of the trained interviewers were not around to conduct interviews by that moment due to other official commitments. Therefore, interviews were waived and bednets were provided on loan basis to those who requested. This was done in consideration that the questionnaire would be modified and applied later to take a retrospective approach for assessing why those who took the nets were attracted to such nets and the maximum WTP price they preferred per net apart from the Tshs. 3000 that was subsidized by the project and commonly charged to everyone who required them. As time went on, the provision of bednets was not restricted to staff only, but also it involved their relatives/family members as almost all of the staff presented a request to have a relative or several of them who also were in need of such bednets. In general, 65 staff at Amani Medical Research Centre were provided with bednets as per their WTP. Also, in February 2003, the Muheza Police staff community took 5 boxes each containing 20 bednets that were shared among 100 individual police staff.Also,the police agreed to take bednets according to their needs so that they could pay later in one or two monthly instalments through their salaries after which they could be interviewed. This agreement was reached during a general staff meeting that was attended by the principal investigator and one of his research team staff, both of whom were invited to respond to some questions relating to the use of such bednets, their protective efficacy and their side-effects (if any). The meeting was also attended by the district’s police-officer-in-charge (OCD), the Station Police in-charge and other police officers. Data Collection Methods Heads of households, shopkeepers and formal employees were interviewed using a semistructured questionnaire administered by the researchers. It had a series of questions (based on experience from the pilot survey) with pre-coded answers some of which were binary, close-ended polychotomous, and others being open-ended. There was slight difference in the formulations of the questions applied between formal employees and shopkeepers. Focus group discussions (FGDs) were held (each comprising of 6-12 participants). To avoid people expressing their WTP blindly for a product they don’t see or know, there was a scenario for showing and demonstrating (to different types of respondents) the use of the Olyset bednet, and explanation given regarding its potential benefits and disadvantages. The mostly explained benefit was the efficacy of the insecticide contained in the fabrics, the durability of the nylon fabrics if carefully used, and its potential for saving lives and reduction in economic costs such as expenditures on drugs or short-lived mosquito-coils for those opting to use such products. The disadvantage mainly identified was the smell of the insecticide especially in the initial days of the bednet’s use and the re-treatment costs after (though long) while. A similar scenario was applied in Nigeria. While in FGDs the question concerning how much people would afford to pay per a bednet was left open to be discussed by the participants themselves, for the household heads and some staff a hypothetical price of Tanzanian Shillings (Tshs.) 3,500/= for a medium-sized (4 by 6 feet) square bednet was used in the elicitation. This price was determined based on the research team consulting community leaders and local retail shopkeepers as key informants. The next step the research team made was to supply bednets to each shop selected to work to shops as study agent for assessing the trend of bednets sales at different seasons of the year. Between late March and early August, 440 bednets were distributed as follows: 140 nets in 4 shops (Mazinde), 150 nets in 3 shops (Mgombezi) and 150 nets in 3 shops (Mkanyageni). A special short questionnaire was prepared to guide the net shop/kiosk keepers to record some information from all individual persons who visited their outlets for the purpose of buying the nets. Results From the Baseline Survey Concerning the desirable price per net From the open-ended question, individual shopkeepers suggested different prices. There were absolutely different prices suggested by respondents of different villages, as shown in Table 1. From Table 1 it can be seen that the mean, median, mode and range of prices suggested by shopkeepers in each study village were as follows: Local government leaders in Mazinde village viewed that bednets would be affordable to the majority of the residents if the price per net ranged between Tshs. 2000 – 3500. They initially suggested of Tshs. 4,500 – 5,000 but this came to be seen as being too high. Their fellows in Mgombezi Village proposed a price range of Tshs. 2,500 – 4,000 while those in Mkanyageni Village suggested Tshs. 1,500 – 2,500 per one bednet. With regard to the best period in which bednets would be highly purchasable, such community leaders in Mkanyageni village viewed the rain season when mosquito densities are high or during crop harvest season when cash income is easily accessible to the majority of households. Their fellow leaders in Mazinde shared the same view, although they warned that during rain season nets would not be bought if the price per net exceeded Tshs. 2,500. Similarly, leaders in Mgombezi village supported the crop harvest season on ground that during the rain season when early stages of crop cultivation such as farm preparation and weeding take place, it is difficult for most of the rural residents to access cash because the little cash they have is spent on buying fertilizers, pesticides, and food (because of having little or none in the stock by that time). As for FGD participants, who were categorized into three groups of adult men (aged 36+ years), young men (aged 18-35 years) and women (of childbearing age) in each village, the following table reveals the opinions collected regarding the most affordable price ranges suggested by individual groups of participants: A few participants suggested a price less than those suggested by the majority of their fellow discussants in the groups. For instance, in Mazinde, some participants suggested a price range of Tshs. 1000 – 1,500 while a few who realized the protective efficacy of ITNs as compared to untreated bednets proposed a price even higher than those suggested by the majority of the discussants in their individual groups. From a total of 416 household members who responded to be interviewed, the mean, modal and median family sizes were all (6) while mean and median ages of the respondents were 42 and 39 years respectively. Of all the households 320 (76.9%) were headed by men, meanwhile 312 (75%) of the total respondents were married. Moslems accounted to 295 (70.9%), Christians totaled 119 (28.6%) of all the respondents while 2 respondents reported to believe in traditional religions. Furthermore, 321 (75%) reported to have completed standard-seven primary-education while only 27 (5.6%) stated to have had an opportunity to attend secondary school. In response to a general question that, ‘if bednets were made available in areas where you can access them, would you be willing to buy any?’, answers collected from 415 respondents indicated that, 414 (85%) said ‘yes’, 54 (13%) were ‘uncertain’ while 8 (2%) said ‘no’. Nevertheless, the question did not mention the price at which bednets would be sold. Responding to an open question concerning the price they would be willing to pay, figure 1 below illustrates the different prices per net as suggested by the household respondents. Next is table 4 that gives a picture of the WTP among the study households if the price per net were Tshs. 3,500. This is followed by fig. 2 which shows respondents’ WTP if the price per net were a little < Tshs. 3,500. Opinions about the preferred payment mechanisms It was commonly suggested by all the different kinds of the respondents that payment by installment would be preferred to payment of the whole price on the spot per net out-of-pocket. In this regard, it was explained that, people should be sensitized and those interested in the bednets should register their names to any community elected local government leader and that there must be another person elected by the community to act as a treasurer so that people can be paying directly to that person by instalment until they reach the total amount required for them to be provided with the nets. Community leaders and some FGD participants also proposed some owners of shops who could act as sales agents should the community have had agreed to use that channel. In-kind transactions such as exchanging bednets with cash crops or other material from interested customers was greatly discouraged by retail shop/kiosk sellers, although some FGD participants seemed to be in support of that. From the Follow-Up [Intervention] Phase The trend of bednets sales at the study agent shops was quite discouraging. Notably there was no significant variation in bednets sales between the rain and dry seasons. In all the three villages, the highest bednets sales were noticed during May, followed by June, as shown in Householders’ WTP based on a binary with follow-up (BWFU) question Up to the end of the study, the total number of household respondents who were followed-up to elicit their true WTP was 26 in each village whereby the composition of males and female respondents was as follows: Mkanyageni (15 men, 11 women); Mgombezi (11 men, 15 women); Mazinde (10 men, 16 women). But none of the respondents were able to pay even for a single bednet in actual sense as was anticipated. Some of them (11, 12, and 19 in the each village respectively) were not even found at home on the first day the follow up was made. The researchers had to set new appointments by leaving information with relatives who were found at home. Despite efforts to make more follow-up visits to trace those who were missed, no changes were noticed for none came to actually pay for the nets. It was recognized later that it was difficult to trace some of our respondents because they deliberately gave us fake (improper) names during the baseline survey. Consulting local key informants as to why that was the case, it was revealed that, this event was mainly due to fear by such respondents of feeling ashamed if they were found with no money as they promised on the dates of appointment they would be followed-up by the researchers. NIMR staff at Amani Medical Research Centre in Muheza All the 24 staff initially interviewed about their WTP at the NIMR - Ubwari Station (in Muheza Town) and 5 staff at Bombo NIMR Station (in Tanga Municipality) stated to be willing to pay. The poor trend of bednets sales in villages prompted the research team to rethink and fix a price of Tshs. 3,000 per net in order to attract more staff to buy without regarding their employment status. Also, this was purposely done in conformity to the demand theory that the lower the price the higher the quantity demanded, ceteris paribus. This was also due to the observation that most of the respondents both in FGDs and household interviews had preferred to pay between Tshs. 2000 - 3000. The number of bednets actually taken by each staff ranged between 1- 6. The modal number of bednets that staff were willing to buy and, which they actually paid-for was 2. The mean and median actual WTP price for the nets was automatically Tshs. 3,000 equal to the fixed price per bednet. More than a half of the staff suggested to be given nets on loan basis so that they could be able to recover their debts through their monthly salaries that could be officially deducted in one or two monthly instalments depending on the number of bednets one opted to take. Those who agreed with this proposal were registered their names after which they were given the bednets. To ensure that the money would be effectively collected, the project managers liaised with the NIMR Amani Centre Finance and Administration Department through the Director who authorized staff salaries to be deducted as per agreement, and the money for the first instalment was deducted in April 2003 from their salaries. A few of the staff paid on the spot (either for some or all the bednets they decided to take) immediately after being interviewed. Interestingly, some of the staff returned one or two days later with new requests for additional bednets after having used the nets they were given one day or several days before. At the Amani Centre, a total of 65 initially registered for bednets actually took the bednets on loan basis. The total number of bednets taken was 219, thus showing an average of 3 bednets taken by each staff. At Amani Centre (located in the East Usambara Mountains) 40 staff came for the second time thereby taking 120 bednets (average of 3 bednets per staff). Those who came for the bednets for the second time at Ubwari Station in Muheza Town (in the lower parts of the mountains, about 35 km from Amani) 42 staff took a total of 147 bednets (also, average of 3 bednets per staff). Of such 42 staff at Ubwari Station, 13 (31%) were able to pay part of their loan. Police Staff at Muheza District Police Station and NMB Staff in Muheza The 7 staff at the National Micro-finance Bank (NMB) were able to pay on the spot as they promised during the interview. On average, each NMB staff took 3 bednets at a price of Tshs. 3,000. Similarly, all the 100 bednets that were supplied to the Police Station were taken by 100 police staff households (i.e. 1 bednet per each household) at a price of Tshs. 3,000 per net, and the whole amount of Tshs. 300,000 was collected in May 2003 from the OCD’s office at the Police Station. Discussion The methodological approach used in the elicitation of WTP and the findings from interviews and discussions held with various respondents provide ground for debate if one has to say exactly that this or that was the most appropriate (affordable and desirable) price per one Olyset type of bednet. It can be seen that the majority of the respondents of different categories were in favour of a price of less than or equal to Tshs. 3000 per net. The fixing of Tshs. 3,500 per bednet made by the research team based on advice from shopkeepers and local community leaders seems to have gone beyond the expectations and suggestions of the majority of the residents interviewed. In this case, it is evident that it is the consumer who should be sovereign i.e. who has to be listened more in order to make appropriate judgment of best price to be paid per the bednet. It also sounds good to use multiple techniques and sources of evidence in assessing WTP for products like bednets, but the limitation is that, it may be difficult to judge at the end of the day whose opinion or which study technique maximizes the reliability of information with regard to the most desirable (widely acceptable and affordable) WTP price. Findings from individual household respondents had shown that a considerable proportion of the respondents stated to be willing to pay in response to a hypothetical question on WTP if the price were Tshs. 3500. But none of those who were followed-up to see whether they could actually take the nets by paying that sum of money in light of their promises, actually paid. Again, this provides evidence that WTP hypothetically does not necessarily reflect one’s true WTP. As in other businesses, data from staff working in different organizations reveals that selling bednets on loan basis is one effective ways through which such materials can be promoted for their actual utilization at household and other community levels. This form of business either allows the potential customers to take the bednets in advance of payment and come to pay the whole amount later or to pay little at a time until they complete the total amount required before they could be given the bednets. The underlying challenge to latter option is that, few people (if any) especially in developing communities like those studied have little trust in disposing their money in advance of getting the commodity while they know that they might be forgoing the money required for other essential needs. Various inferences can be drawn as to why some staff requested for additional bednets after they had been interviewed and given the first number of bednets they stated to be WTP for, given the price. Possibly, the loan granted motivated them if at all it is not the quality of the nylon material that indicates a high degree of the bednets durability (for most poor households would prefer products that have a long lasting quality). The other quality things that might have had motivated them are the repellent and killing effects they might had seen after using the bednets for sometime. Probably, had the villagers also been given such an opportunity for buying bednets on a loan term for them to pay by instalment, a similar motivation would be realized. However, the problem could arise in terms of making follow up to collect the money since for rural residents (particularly peasants), the issue of cash unreliability would be reported even if people were given more time. Conclusion and public health policy implications Application of multiple methods in the assessment of community WTP is vital if more reliable information has to be established to understanding the most desirable price to guide dealers in bednets business or bednets related malaria interventions. Nevertheless, it is important to recognize that different techniques may yield diverse results that are too difficult to reconcile so as to come up with one justifiable conclusion for policy action. One possibility could be to apply systematically one or two techniques and test them, and should it reveals considerable shortcomings, then an alternative approach can be adopted. Testing one or at most two elicitation techniques at a time can be easily manageable and provide greater feasibility for analysis. Meanwhile it is important to recognize that a person’s or community’s WTP might change over time, especially when there is changes in other domain demand factors like consumer’s disposable income, preferences and expectations about the product in question that also might depend on time and market situations. As malaria preventive services are advocated for being oriented to communities with limited access, the need for continued collection of opinions from communities of different characteristics remains essential, and one way to ensure this is accomplished in terms of bednets is to continue supporting WTP and actual utilization studies on bednets, especially as the government is working to moving to a national scale with ITNs. Based on findings from different respondents and the trend of bednets sales in the study villages, it is concluded that, most Tanzanians especially those living in rural areas, fixing a price exceeding Tshs. 3,000 per net is highly undesirable (unless initial sensitization such as social marketing is instituted) and it is likely to reduce the prospects of the going to national scale with ITNs, which is the current strategy of the Tanzanian National Malaria Control Program. However, free-of-charge bednets provision not only opens the debate about who will provide such materials and who will bear the costs, but also could be crowd-out some bednets manufacturers and traders and intensifies the debate about the sustainability of such a public health policy advocacy. Acknowledgments Thanks be given to Mr. Hasnein Gulamhussein, the representative of the Sumitomo Chemical Company, country office-Dar Es Salaam,Tanzania, through whom we have been able to request and obtain additional 1000 nets to the previous ones (1000) donated by the same company, the Sumitomo Chemical Company in Japan. We are grateful to our study respondents and the district central and local government officers in Korogwe and Muheza districts, Tanzania for their cooperation during this study. The assistants in data collection (Mayunga Mayega; Mujuni Njunwa, Dorica Mubyazi; Judith Mahundi; Hamza Mohamed, Dr. Samuel Gesase) are also thanked for their active participation. The Director of Amani Centre and Mr. Julius Mhina, also of NIMR-Amani Centre are also thanked for providing support on the coordination this one type of the ENRECA supported project. This study was funded collaboratively by DANIDA that provided a grant of Tshs. 10,000 for fieldwork and the Sumitomo Chemical Company of Japan that provided 2000 Olyset bednets for study free of charge for research purposes. References
© Copyright 2004 - East African Journal of Public Heath The following images related to this document are available:Photo images[lp04004f1.jpg] [lp04004t5.jpg] [lp04004t4.jpg] [lp04004t2.jpg] [lp04004t1.jpg] [lp04004t3.jpg] [lp04004f2.jpg] |
| |||||||||
