1. Introduction
There is a growing attention on universal health coverage in the world, which aims to guarantee that all people have access to the needed and effective healthcare services without facing financial hardship [1]. Health insurance is one of the mechanisms for this purpose which, by periodic payments for episodes of illness, enable people to receive healthcare services when needed without high out-of-pocket payment [2]. Healthcare access is still a global challenge because many people cannot afford the costs of health services [3]. 
A National Health Insurance Scheme (NHIS) refers to a system of health insurance that covers a national population against the costs of healthcare [4]. On March 31, 2021, the Parliament of Uganda passed a national health insurance bill that outlines the general structure for a first-ever national social health insurance scheme in Uganda [5]. Private commercial health insurance exists only on a very small scale covering 1.2% of Uganda’s population, catering only to the wealthy families and individuals. Uganda’s health sector takes Community Health Insurance (CHI) as a better option for the informal sector since it gives opportunities for social participation and has relatively affordable premiums compared to CHI schemes [6]. Uganda has the poorest insurance market penetration (1%) compared to Rwanda (1%), Tanzania (2.3%), and Kenya (3.8%) [7]. 
There is low health insurance coverage for the commercial motorcyclists in Uganda despite the existence of designed health insurance packages for them in the country with affordable premiums, Boda Boda CHI schemes, and other existing products that cover the commercial motorcyclists. For example, the group life insurance policy covers the commercial motorcyclists and their passengers in case of accidents which costs 70,000 UGX per year (1$=3,500 UGX). There is also a micro-insurance product launched by Tugende which costs 67,608 UGX per year and provides medical coverage and motor third-party insurance [8, 9]. However, there is no evidence available to show the willingness of commercial motorcyclists to pay for health insurance. Therefore, this study aims to assess the determinants (individual and scheme) and level of Willingness to Pay (WTP) for CHI among commercial motorcyclists in Kampala City, Uganda. 

2. Methods
Participants and study area
In this study, participants were commercial motorcyclists in Kampala City, Uganda who are smaller size motorcycle rider groups. The poorly maintained city roads pose a serious safety threat to motorcyclists with a risk of being injured. They are more likely to die in a crash than drivers in other vehicles. 

Study design and variables
This is a descriptive study with a cross-sectional design using Contingent Valuation Method (CVM) similar to the ones used in two studies, one conducted on WTP for national health insurance fund among public servants in south Sudan, and other on WTP for CHI among Taxi drivers in Kampala City, Uganda [10, 11]. The WTP for CHI is the dependent variable and the independent variables are socio-demographic factors such as age, religious denomination, education level, marital status, duration of operation in Nakawa Division by the motorcyclist, occupation status (whether the motorcycle belongs to someone else or belongs to the rider), and having a health insurance. 

WTP assessment tool
The CVM questionnaire was used to measure WTP for CHI. The CVM questions are open-ended or closed-ended. Respondents were asked to state their maximum WTP for the benefit using the bidding game. In this game, a starting bid is made which is either accepted or rejected by the respondent. Depending on the answer, the bids were then adjusted until reaching the respondent’s maximum WTP. The respondents were first asked if they were willing to pay for CHI. Then, they were asked how much they were willing to pay. Those who mentioned a figure lower than the set premium for Boda Boda’s health insurance in Uganda, were further asked for the maximum amount of money they would pay for CHI. The respondents were then asked whether they were sensitized about CHI and its cost. 
The stated options for respondents to select were based on a monthly premium of <5,000 UGX, 5,001-10,000 UGX, 10,001-20,000 UGX, 20,001-50,000 UGX, and >50,000 UGX per person. Other stated preference was based on how often respondents were willing to pay for CHI: daily, weekly, monthly, and bi-monthly. The interviews were administered for about 30 minutes. Data were collected in April, May, and June 2019. The questionnaire was pretested among 30 Boda Boda riders at two stages in another division (Makindye) to determine the validity and reliability of the research tool.

Sampling
The sample size was determined using Krejcie and Morgan (1970)’s formula, since the estimation of sample size was done from a finite population of commercial motorcyclists in Nakawa Division (n=50,000). In this regard, a sample size of 381 was determined. The samples were selected using a random sampling method. Inclusion criteria were: Age 18 years, operating in Nakawa Division for at least half a year, and declaring a consent to participate. It was assumed that within six months, one sample had exposure to various risks and probable accidents associated with the business. Given that the commercial motorcyclists do not have specific stage points mapped out, in each Division parishes, a convenience sampling was performed by identifying a busy center within the division as a starting point, because these centers usually have more than one Boda Boda stage. 
Four research assistants were recruited and trained for two days for data collection. The training lasted five hours per session and focused on information about assessing WTP, the study methods, questionnaire and interview skills. They who were fluent in writing and speaking English and in the local language ‘Luganda` and had knowledge of commercial motorcyclists operations.

Data analysis
The data were entered and analyzed in SPSS software, v. 21. Descriptive statistics were used to check the normal distribution of data. The Chi-square was used to assess the difference between the study variables. Both bivariate and multivariate regression analyses were also conducted. The significance level was set at 0.05. The dependent variable was considered as a continuous variable in order to generate more accurate outcome for analysis. Hence, a regression model was used to predict a continuous dependent variable from a number of independent variables.

3. Results
Characteristics of participants
The socio-demographic characteristics of the participants are presented in Table 1.

The majority of them were in the age range of 29-39 years (n=224; 58.8%), and were Christians (n=265; 69.6%). Almost one-third of them had a primary education (n=120; 31.5%) and about two-third of them were cohabiting (n=237; 62.2%). Almost 20.0%, 19.7%, 20.7% and 21.1% of them were operating in Nakawa Division for 6-12 months, 12-24 months, 3-4 years, and >5 years, respectively.

WTP results
The bidding game using the CVM questionnaire revealed that most of motorcyclists were willing to enroll in a CHI scheme (n=328, 82.6%). The majority of them reported WTP for CHI if the annual insurance premium was 70,000 UGX (n=277; 72.7%). More than half of them reported that the maximum amount they were willing to pay was between 70,000 and 100,000 UGX (n=230; 60.4%). If the premium was 50,000 UGX, the majority of them were willing to pay (n=342, 89.8%). The level of WTP for CHI was found to be 70%, based on the proportion of commercial motorcyclists who were really willing to pay an amount ≥70,000 UGX which is the current average premium for all available Boda Boda CHI schemes in Uganda (Figure 1).

Individual determinants of WTP 
Few participants were in the commercial motorcycling business for a year (17.8%), and the majority had children (n=250; 65.6%) and were employed, i.e. the motorcycle belonged to someone else (n=245; 64.3%). More than half of them had not been involved in a riding accident (n=230; 60.4%), while more than three-quarter of them believed that they could be involved in a motorcycle accident anytime (n=317; 83.2%). Almost half of them believed that a riding accident would cost them 100,000 - 500,000 UGX to receive health services (n=174; 45.7%). More than three-quarter of the participants had paid for some form of insurance (n=342; 89.8%). The majority believed that CHI coverage is good for commercial motorcyclists (n=338; 88.7%). Almost half of them mentioned that insurance plans for commercial motorcyclists only covered treatment costs in case of an accident (n=184; 48.3%).
The results in Table 2 showed that work experience in the commercial motorcycle business increased the odds of WTP for CHI; the commercial motorcyclists with five years of experience in the business were 9 times more willing to pay for CHI (AOR=9.743, 95% CI: 2.118-14.820) than those with an experience less than one year (AOR=1.138, 95% CI: 0.175-7.406).

The employed participants were less likely to pay for CHI (AOR=0.597, 95% CI: 0.392-1.039). Those who had been involved in a motorcyclist accident were less likely to pay for CHI (AOR=0.786, 95% CI: 0.355-2.525). The motorcyclists who had paid for any form of insurance were 4 times more likely to pay for CHI (AOR=4.480, 95% CI: 2.250-8.919). Furthermore, those who believed that having health insurance can be good for a motorcyclist were 4 times more likely to pay for CHI (AOR=4.398, 95% CI: 1.209-5.757). Single commercial motorcyclists were also 4 times more likely to pay for CHI (AOR=4.312, 95% CI: 1.152-5.640). 

Insurance scheme-related determinants of WTP 
Other determinants of WTP included: being aware of CHI, cost of CHI, requiring to pay several times a year, coverage of CHI for high-cost treatments (e.g. surgery), affordability to pay more than 70,000 UGX for CHI, time-consuming process of getting CHI, accessibility of CHI offices, possibility of paying premiums via mobile applications, and catering for the needed health services at any time.
According to the results in Table 3, the majority of participants were not aware of CHI for commercial motorcyclists (n=246; 64.6%).

Most of them agreed that CHI was expensive (n=272; 71.4%), CHI schemes require to pay several times a year (n=229; 60.1%), and CHI covers high-cost treatments such as surgeries (n=275; 72.2%). However, the majority of them disagreed that CHI cost over 70,000 UGX was not affordable by commercial motorcyclists (n=263; 69%). Furthermore, the majority of them agreed that getting insurance is a time-consuming process (n=230; 60.4%), but disagreed that CHI outlets (offices) were accessible (n=220; 57.7%), and that CHI premiums could be paid via mobile applications (n=296; 77.7%). A half of them agreed that the commercial motorcyclists with insurance were catered for health services whenever they needed (n=191; 50.1%). Only one insurance scheme-related factor was found to be a significant determinant of WTP for CHI in Boda boda motorcyclists which was “being aware of CHI for commercial motorcyclists” (X2=7.103; P=0.008). Compared to the motorcyclists who were not aware of the CHI schemas designed for them, the motorcyclists with awareness were willing to pay for CHI (78.5%). These motorcyclists were 7 times more likely to pay for CHI (AOR=7.518; 95% CI: 2.318-11.844) compared to those who were not aware of the CHI schemas. The study population was assumed to be homogenous and, thus, heterogeneity was not considered a threat to the validity of the results. Moreover, based on the previous studies [10, 11], the authors assumed that the study variables were non-collinear.

4. Discussion
The motorcyclists require insurance to have guaranteed access to quality healthcare in case of injury; they are more prone to accidents in Uganda. The results of this study showed that 7 out 10 commercial motorcyclists in Nakawa Division, Kampala City, Uganda were willing to pay a premium amount of ≥70,000 UGX which is the current average premium for all available Boda Boda CHI schemes in Uganda [8, 9]. This indicates a fairly high rate of WTP for CHI and, thus, if Boda Boda CHI schemes are rolled out, the majority of commercial motorcyclists may pay the annual premium. However, the percentage of motorcyclists (30%) who were not willing to pay the average annual premium should not be neglected, since they most probably have little or no access to quality healthcare. 
The proportion of commercial motorcyclists who were willing to pay for CHI is consistent with the proportion that was reported in a study conducted in Malaysia (a middle-income country). In this study, by using CVM, it was revealed that more than 63.1% of participants were willing to join CHI with an average payment of 114.38 USD per month per household. Jain et al. in a study also reported that about 71% of the people in rural India were willing to pay 1500 INR (27 USD) annually for CHI [12]. An higher amount for having WTP was found in a study in India where people were willing to pay more than 30 USD per year per person for health insurance [13], and also in a study in Wuhan, China where the mean amount for having WTP was 4 USD per month per person (48 USD per person per year). 
The difference between the findings of the current study and the study conducted in China is because the participants in China were workers with higher socioeconomic status, while in our study they were commercial motorcyclists. It should be noted that the amount of money that the motorcyclists in our study were willing to pay is higher than in some studies conducted in other countries like India, Namibia, Botswana, and Nigeria. The mean premium that commercial motorcyclists in our study were willing to pay was ≥70,000 UGX (20 USD) per person per year which is greater than the amounts reported in Nigeria (11.1 and 3.26 USD) [13, 14] and in Namibia (48 NAD or 3.15 USD) [15]. Similarly, a study by Dong et al. in Burkina Faso showed a mean amount of 13 USD per year for a household with average size of 8 members [16]. This translates to 1.6 USD per person per year, a figure lower than what was found in our study. This discrepancy may be due to difference in the study population. For example, most of these studies assessed households and not exactly people who were prone to accidents in their businesses (e.g. commercial motorcyclists) and have daily income; therefore, the participants in these studies probably were not highly interested in health insurance.
The present study found that motorcyclists with at least five years of experience in commercial motorcycle business had 9 times more likely to have WTP for CHI. This finding can be due to experiencing riding accidents or observing other motorcyclists who had been involved in fatal or non-fatal riding accidents. This is consistent with the findings of Barnighausen et al. who suggested an association between the experience of riding accidents and long ridding hours [17]. Such experiences most likely change the attitude of motorcyclist with longer ridding hours, making them realize the importance of CHI for ease of access to healthcare. 
The results of this study also showed that commercial motorcyclists who had previously paid for any form of insurance were four times more likely to have WTP for CHI. In an insurance process, the first party is the owner of the motorcycle; the second party is the insurance company, and the third party is any other person who may be affected by the contract (not an employee or a family member of the first party). To get this insurance, the motorcyclist (first party) refers to an insurance company to determine the premium that should be paid to insure the third party. The process of getting multiple third-party insurance schemas exposes the motorcyclist to the insurance world and lets them somewhat appreciate the importance of insurance. Thus, the motorcyclists with a higher experience are more willing to pay for CHI if they have awareness about its schemas. We also found out that those riding a motorcycle belonged to someone else were less likely to have WTP for CHI This finding is supported by other studies who found out that people with stable income are more willing to pay compared to those with no stable income [16, 17]. In Nakawa division, the single commercial motorcyclists were found to be four times more likely to have WTP for CHI compared to those who were married or co-habiting. Single motorcyclists are likely to have lower expenditure since they may have no families to cater for; hence, they can save more money compared to married ones. Thus, the annual insurance premium is affordable for them. This is consistent with the findings of Shafie [18] and Oriakhi [13] who also showed that single participants had lower expenditures and thus could afford to pay annual health insurance premiums. 
Awareness about CHI was the only significant determinant of WTP in commercial motorcyclists. If they be given information about the usefulness of insurance, it can arouse their curiosity about its usefulness and benefits, and thus they probably consider the enrolment in a health insurance scheme. The sensitization session demystifies any myths; thus, motorcyclists receive knowledge about the benefits of CHI schemes, payment mechanisms, and processes, which can increase their WTP. This is consistent with the findings of Kamuzora and Gilson (2006) from Kenya and Tanzania, which revealed that lack of information on health insurance and registration procedures have detrimental effects on enrolment [19]. A review study by Chankova et al. identified a number of barriers that restrict potential clients from joining health insurance schemes in low- and middle-income countries. Among the barriers, the “literacy gap” or lack of knowledge about insurance (mechanism, utility etc.) was found to be important [20] especially due to low sensitization. 

Study limitations
Commercial motorcyclists were busy with transporting passengers which might not let them pay enough attention to the questions. However, the research team was patient and had to wait until they return or made an appointment in the next day. This study could have been further enriched by qualitative research. The researchers explored other existing qualitative data to add additional meaning to the quantitative results of this study. The results of this study cannot be generalized to the entire population of Boda Boda riders but can be useful for future interventions and design of similar schemes. This study did not investigate the “willingness to accept” in commercial motorcyclists which is recommended to be explored in future studies along with WTP. It can also be explained that adaptive CER srategies act as a social support mechanism and play a major role in emotional disorders such as anxiety.

5. Conclusion
The results of this study may be useful in enrolment of informal sector workers (especially those involved in public transportation) in a health insurance scheme. Significant determinants of WTP for CHI in commercial motorcyclists are work experience, awareness about CHI schemes, self-employment, previous involvement in a riding accident, history of paying for any other form of insurance, and being single. There is a need for increasing awareness and educating commercial motorcyclists about the benefits, payment and schemes of CHI. Social media and other media platforms can be used for this purpose to increase WTP among them. Government and other actors could consider to roll out or expand the motorcycle loan plan so that commercial motorcyclists can have their own motorcycles and clear the payments in installments. 

Ethical Considerations
Compliance with ethical guidelines
The study was approved by the Ethics Committee of Clarke International University. A detailed explanation about the study was given to participants prior to interviews and consent forms were signed by them. The data collected from participants were kept confidential.

Funding
This research did not receive any grant from funding agencies in the public, commercial, or non-profit sectors.

Authors' contributions
Authors had equal contribution to designing study, collecting data, revising the initial draft, and approving the final draft.

Conflict of interest
The authors declare no conflict of interest.

References

1. World Health Organization (WHO). Research for universal health coverage: World health report 2013. Geneva: World Health Organization; 2013. https://www.who.int/publications/i/item/9789240690837
2. The World Bank. World development indicators (WDI). Washington, D.C: The World Bank; 2012. https://databank.worldbank.org/home.aspx
3. Leatherman S, Dunford C. Linking health to microfinance to reduce poverty. Bulletin of The World Health Organization. 2010; 88(6):470-1. [PMID] [PMCID]
4. World Health Organization (WHO). The world health report: Health systems financing: The path to universal coverage. Geneva: World Health Organization; 2010. https://apps.who.int/iris/handle/10665/44371
5. Bukenya, M. “Second Reading of the National Health Insurance Scheme BilL, 2019.” Kampala, Uganda. Parliament House, 2021. The Hansard of the Parliament of Uganda. 10^th Parliament. ^{https://www.parliament.go.ug/documents/5017/hansards-2021-march}
6. Basaza R, Namyalo PK, Mayora C, Shepard D, Universal health insurance coverage; Where is Uganda and what are the lessons for other LMICs Uganda; 2018. ^{https://www.econbiz.de/Record/the-journey-to-universal-health-r0012260410}
7. Insurance Regulatory Authority of Uganda. 2015 Annual insurance market report Kampal Insurance Regulatory Authority of Uganda; 2015. https://ira.go.ug/download/2015-annual-insurance-market-report/
8. Adengo J. Boda boda riders to get group life insurance cover. 2016. ^{https://www.monitor.co.ug/uganda/business/markets/boda-boda-riders-to-get-group-life-ie-cover-1640028}
9. Kulabako F. Boda boda cyclists get life insurance product as micro insurance takes root [Internet]. 2017 [Updated 2019 December 20]. Available from: https://www.newvision.co.ug/news/1409454/boda-boda-cyclists-life-insurance-product-micro-insurance-takes-root
10. Basaza R, Kyasiimire EP, Namyalo PK, Kawooya A, Nnamulondo P, Alier KP. Willingness to pay for Community Health Insurance among taxi drivers in Kampala City, Uganda: A contingent evaluation. Risk Management and Healthcare Policy. 2019; 12:133-43. [DOI:10.2147/RMHP.S184872] [PMID] [PMCID]
11. Basaza R, Alier PK, Kirabira P, Ogubi D, Lako RLL. Willingness to pay for National Health Insurance Fund among public servants in Juba City, South Sudan: A contingent evaluation. International Journal for Equity in Health. 2017; 16(1):158. [DOI:10.1186/s12939-017-0650-7] [PMID] [PMCID]
12. Jain A, Swetha S, Johar Z, Raghavan R. Acceptability of, and willingness to pay for, Community health insurance in rural India. Journal of Epidemiology and Global Health. 2014; 4(3):159-67. [DOI:10.1016/j.jegh.2013.12.004] [PMID] [PMCID]
13. Oriakhi H, Onemolease EA. Determinants of rural household’s willingness to participate in community based health insurance scheme in Edo State, Nigeria. Studies on Ethno-Medicine. 2012; 6(2):95-102. [DOI:10.1080/09735070.2012.11886425.]
14. Babatunde OA, Akande TM, Salaudeen AG, Aderibigbe SA, Elegbede OE, Oyodele LM. Willingness to pay for community health insurance and its determinants among households heads in rural communities in North-Central Nigeria. International Review of Social Sciences and Humanities. 2012; 2(2):133-42. https://www.researchgate.net/publication/264030899_
15. Gustafsson-Wright E, Asfaw A, van der Gaag J. Willingness to pay for health insurance: An analysis of the potential market for new low-cost health insurance products in Namibia. Social Science & Medicine. 2009; 69(9):1351-9. [DOI:10.1016/j.socscimed.2009.08.011] [PMID]
16. Dong H, Kouyate B, Cairns J, Mugisha F, Sauerborn R. Willingness-to-pay for community-based insurance in Burkina Faso. Health Economics. 2003; 12(10):849-62. [DOI:10.1002/hec.771] [PMID]
17. Bärnighausen T, Liu Y, Zhang X, Sauerborn R. Willingness to pay for social health insurance among informal workers in Wuhan, China: A contingent valuation study. BMC Health Services Research. 2007; 7:114. [DOI:10.1186/1472-6963-7-114] [PMID] [PMCID]
18. Shafie AA, Hassali MA. Willingness to pay for voluntary community-based health insurance: Findings from an exploratory study in the state of Penang, Malaysia. Social Science & Medicine. 2013; 96:272-6. [DOI:10.1016/j.socscimed.2013.02.045] [PMID]
19. Kamuzora P, Gilson L. Factors influencing implementation of the Community Health Fund in Tanzania. Health Policy and Planning. 2006; 22(2):95-102. [DOI:10.1093/heapol/czm001] [PMID]
20. Chankova S, Sulzbach S, Diop F. Impact of mutual health organizations: Evidence from West Africa. Health Policy and Planning. 2008; 23(4):264-76. [DOI:10.1093/heapol/czn011] [PMID] [PMCID]