Abstract
Introduction
The shortage of formal healthcare providers in urban slums has resulted in the widespread presence of untrained informal healthcare providers (IHPs), who often deliver low-quality services. Integrating these IHPs into the formal healthcare system through capacity-building initiatives could enhance the quality of services offered to clients. However, this may lead to an increase in healthcare costs, which would be passed on to the clients. This paper provides new evidence on the level of households’ willingness to pay (WTP) for improved quality of services from IHPs that will be occasioned by linking IHPs to the formal health system.
Methods
The levels of consumers’ WTP for linking IHPs to the formal health system for improved quality of services was elicited using the contingent valuation method (CVM) in eight slums from two states in southeast Nigeria, namely Anambra and Enugu. A pre-tested interviewer-administered questionnaire was used to elicit WTP from 1025 randomly selected households from the slums. Data were analyzed using R and Stata software. Univariate, bivariate, and multivariate analyses were performed. Theoretical validity of WTP was analyzed using both multiple logistic and Tobit regression analyses at 95% confidence level.
Findings
Findings showed that 64.5% of households were willing to make a one-off payment for linkages that will ensure improved quality of services from IHPs. The mean willingness-to-pay amount was Nigerian naira (₦)1353 (US $2.08), and 27.1% of the households were willing to pay ≥ ₦1353 ($2.08). Higher asset index and socioeconomic status were positively associated with willingness to pay for improved quality of health services. Each unit increase in asset index was associated with a 28% increase in the odds of willingness to pay (odds ratio (OR) 1.28; p < 0.001), while each unit increase in socioeconomic status score was associated with an 86% increase in the odds of willingness to pay (OR 1.86; p < 0.001). Compared with households that usually seek care from formal providers, those that usually seek care from informal providers (OR 0.4; p = 0.01) and those that usually seek care from both provider types (OR 0.34; p = 0.01) were significantly less likely to express willingness to pay for improved quality of health services.
Conclusions
The level of WTP for linking IHPs with the formal health system is an indication that a considerable proportion of urban slum dwellers desire to have better quality of services and are willing to support interventions that can be used to actualize the linkages. Hence, such interventions would be acceptable to slum dwellers.
Key Points for Decision Makers
| Although 64.5% of urban slum dwellers are willing to pay for improved quality of health services from informal providers that are linked to the formal health system, less than a third of them are willing to pay an amount that is ≥ ₦1353 ($2.08). |
| Wealthier households are willing to pay more money for improved health services than poorer households. |
| Linkage interventions should aim to enhance healthcare access, quality, and affordability, considering the socioeconomic diversity within slum areas |
Introduction
There are barriers to accessing appropriate health services in urban slums owing to the paucity of formal healthcare providers there. This paucity of formal healthcare providers has led to the proliferation of untrained and less knowledgeable informal healthcare providers (IHPs) that mostly provide low-quality services [1]. The uneven distribution of healthcare facilities and services in urban slums [2] has led to an increasing reliance among the urban poor on the more accessible and affordable IHPs [3].
Available studies on healthcare utilization in the slums of lower-middle income countries (LMIC) suggest that people get most advice and services from IHPs and private practitioners rather than well-organized public or private (formal) service providers [4–8]. Regardless of their capacity limitations, IHPs tend to enjoy the confidence and patronage of members of their communities on account of the relatively lower cost of care, ease of access, and in some cases, provision of services on credit [9].
In addition, informal health sector and formal sector collaborations have increasingly been recognized as important in the Nigerian health system [10]. For instance, whereas traditional birth attendants (TBAs) and patent medicine vendors (PMVs) may refer cases beyond their self-defined capacity to primary health centers (PHC), some PHC workers refer clients to PMVs to purchase medicines when they are out of stock in the health facility [11, 12].
The wide availability of IHPs presents opportunities to improve access to appropriate essential health services in underserved urban areas in many low- and middle-income countries (LMICs) [9, 11]. This is because, despite their interdependencies, IHPs are not properly integrated into the health system or even formally linked to the formal health system, hence, a call for interventions that could be in form of providing more formal health services in urban slums or linking the existing IHPs to the formal system. This will maintain continuum of care in the services that clients receive irrespective of the “levels and sites of the health system” through which these services are provided [13].
Linking the IHPs to the formal system is an innovative method of raising the quality of services that are provided by the IHPs in the urban slums and for strengthening the health system in urban areas [14]. It is envisaged that the linkage will involve capacity development of IHPs, getting them involved in health service delivery and health financing interventions of the public sector. Through the linkage, the capacity of IHPs will improve, and adequate channel that simplifies referral system will be created that creates room for continuum of care and reduces mortality [14].
However, undertaking the linkage of IHPs to the formal sector will require some costs that will be borne by the promoters of the linkage and potential beneficiaries such as the consumers and program managers. These costs will be incurred from the use of resources for developing, implementing, and sustaining such linkages. The willingness to pay for such resources is imperative for the successful implementation of the linkages.
Hence, it is important to determine the value represented by the level of willingness to pay (WTP) by the beneficiaries of the improved quality of services that are expected from such linkage for the intervention. Their level of WTP will show the value they attach to the intervention and whether it is worthwhile to undertake the activity.
The WTP that people attach to linking the IHPs to the informal system can be elicited through the contingent valuation method (CVM), which has theoretical underpinning in welfare economics (Kaldor–Hicks Pareto improvement criterion) and measures consumer surplus [15, 16]. WTP is the maximum amount of money that an individual is willing to pay for a good or service. It is practically better to implement the more valued and more worthwhile interventions, unless the government has dire needs to implement less valued and less worthwhile interventions.
WTP, elicited using the CVM, has been used to validly elicit the amounts of money that people are willing to pay for various goods and services in Nigeria and in many other countries in sub-Saharan Africa (SSA) [17–21]. It has been used to inform healthcare policy and resource allocation decisions in both developed and developing countries [19, 22].
The method is called CVM because the respondent is being asked to consider the contingency of a market existing for the thing being valued. Thus, it involves respondents evaluating, in monetary terms, goods or services with benefits that may not be directly measurable [23]. Where market failures exist or the prices charged for services do not reflect real market situations, the CVM technique through WTP can be used to determine the maximum value to the consumers of the goods and services. Asking people directly has the potential to inform about the nature, depth, and economic significance of these values [24].
This paper provides novel information on how much people living in urban slums are willing to pay for improved health services from IHPs. The elicited WTP that is described in this paper is with regard to amounts of money that people are willing to pay for utilizing improved quality of services from informal health providers that have been linked to the formal health system and consequently provided better healthcare services to the urban slum dwellers. The findings will guide policymakers to establish health insurance programs, prioritize healthcare investments, and guarantee fair access to care.
Methods
Study Area and Design
A cross-sectional survey design was utilized to collect and analyze quantitative data from households on willingness to pay for linkages that will ensure improved quality of services from IHPs in urban slums. The study was conducted in eight slums located in two states (Anambra and Enugu) in the southeast geopolitical zone of Nigeria.
Consumers were surveyed in their households. Enugu and Anambra States have estimated populations (as of 2022) of 4,690,100 and 5,953,500, respectively [25]. The two states were purposefully chosen because of their proximity to each other and the research team, and because they both contain several, large, long-term urban slums. In addition, Onitsha city in Anambra State is home to one of Nigeria’s largest open drug markets. The study was conducted in the following slums: Afia-Nine, Ngenevu, Ugbo-Oghe, and Ikilike in Enugu State; and Okpoko 4, Okpoko 5, Prison Marine, and Ibollo in Anambra State.
Study Population and Sampling Method
A modified cluster sampling technique was used to account for the informal settlement structure of slums and to ensure representative household selection within each cluster. Eligible households were those that had either one woman aged 15–49 years (i.e., of typical child-bearing age) or a child aged ≤ 5 years. The reason for the choice of these people in selected households was because of their comparatively higher propensity to consume healthcare services than others. Within eligible households, we primarily sought the main female caregiver as the respondent, but if such an individual was not available, then the male head of the household was interviewed if he stated that he was aware of his household’s health-seeking behavior and expenditure. The reason for the preference of female caregiver was based on the traditional practice of female as caregivers in the households, with the male focusing more on providing for the family. Selected household representatives were expected to be knowledgeable of the healthcare services that their households sought for inpatient and/or outpatient care in the 1 year preceding the survey.
The sample size calculation for the consumer survey was based on being able to estimate 95% confidence intervals for proportions relating to our binary outcomes, using the sample size calculation formula in the Demographic and Household Survey Sampling and Household Listing Manual [26]. For the most conservative/generalizable proportion of 0.5, assuming a design effect (deft) of 1.6 (which was typical of indicators from the Nigerian Demographic and Health Survey 2018), and targeting a 95% confidence interval width of ± 0.098 (based on a relative standard error of 0.05), we estimated that we needed a sample size of 1025 households. This was then divided approximately equally between the eight slums (approximately 128 households per slum). The rationale behind equal sample representation from each slum despite the differences in size of the slums is to allow for unbiased comparisons. This approach prevents larger populations from dominating the results and ensures smaller locations are adequately represented. Moreover, it simplifies the analysis by avoiding complex weighting and ensures statistical power across all locations. In addition, it enables a more focused comparison of localized effects without being influenced by unequal sample sizes.
Data Collection
A pre-tested interviewer-administered questionnaire was configured in Open Data Kit (ODK) and used for data collection. The questionnaire was pre-tested among ten households in a slum community in a contiguous state to ensure clarity and appropriateness of the questionnaire, as well as understanding of the WTP scenario. Following feedback from participants in the pre-test, the WTP scenario was revised with simpler (plain) language and the inclusion of more explanatory phrases, where necessary.
The administration of the questionnaire was carried out in pairs, with one of the research assistants holding the tablet with ODK installed, while the other person held the paper version of the same questionnaire. At the end of data collection each day, each pair crosschecked the data collected using the tablet and the paper to ensure the appropriate information was recorded for each household. After that, the collected data in the tablet was synchronized to the server. The questionnaire was developed by the study team for data collection from the selected households. The instrument elicited information on household demographic and socioeconomic characteristics, living conditions, preferred facility type for care, and willingness to pay for the linkage of informal providers into the formal healthcare system for improved quality of care provision among others.
Within each slum, the data collection team started at the primary healthcare (PHC) center and then located the nearest household and attempted to interview an eligible member of the household. They then moved onto the next-nearest household and repeated this process until the cluster sample size was met. If there were no eligible or willing participants in a household, the team skipped that household. The reason for using PHC as the entry point was to ensure there was a formal healthcare provider apart from the informal ones in the slum. By formal healthcare providers, we mean healthcare facilities with health workers that are formally and sufficiently trained to provide healthcare services, while informal healthcare providers are those with most often less formal education for the services they render. They include traditional birth attendants, patent medicine vendors, bone setters, and herbalists. The administration method was interviewer-administered, which enables the interviewer to ask and explain the questions in the questionnaire to the household representative. Following this method, out of the 1042 potential households that were visited, 1025 accepted to be interviewed, representing an acceptance rate of 98.4%. The data collection was carried out from November to December 2022.
Data were viewed in real time, and any missing information was flagged and addressed promptly by revisiting the households. To minimize recall bias and ensure consistency in data quality, affected households were revisited by the same data collectors within 24 h (by the next day) of data collection. In addition, data managers kept records of the missing data to check whether there was a pattern. There was no pattern in the missing data.
Eliciting WTP
The study employed the contingent valuation method (CVM) to elicit the maximum willingness to pay for improved quality of services. There are three essential elements of any CVM [27], which were covered in the study. These elements are a portrayal of the resource to be valued in our case: improved quality of care as a result of the linkage; description of the financial mechanism to be used for the resource (households’ valuation of improved quality of care availability in their slum); and the question format used to elicit WTP. First, the undervaluation of the good was presented (the scenario) with the financial mechanism described as the prevailing one-off out-of-pocket payments, and the WTP amounts were then elicited using the bidding game (BG) question format. Previous studies have shown that the BG can be used to elicit valid WTP responses in Nigeria [28]. The starting point bid for the BG in the study was chosen on the basis of a baseline qualitative interview with purposively selected slum dwellers. Focus group discussions (FGDs) were held with male heads of households and female caregivers of children under 5 years to explore their perceptions about linking informal providers to the formal health system and what this could imply for slum dwellers’ access to quality healthcare services. Participants discussed typical out-of-pocket health expenditures, perceived value of improved quality of health services from informal providers, and acceptable price ranges. The modal value identified was Nigerian naira (₦)500, which was used as the initial bid in the WTP elicitation tool. Eight sex-disaggregated FGDs were held in total—four with male participants only, and four with female participants only. One FGD was held in each slum with either male participants or female participants. The number of participants ranged from six to eight.
Comparing CVM with other methods such as revealed preference methods (RPM), several yardsticks are used, such as incentive compatibility (ensuring respondents reveal their true preferences), procedural invariance (whether results are consistent across different methods), and construct validity (whether the method accurately measures the intended preferences). Although CVM is susceptible to hypothetical bias, it is a straightforward and practical method, ensuring reliability and feasibility in resource-limited settings. It aligns with the objectives of a good elicitation method by being simple, cost-effective, and able to provide useful WTP estimates to inform policy decisions [44].
The study employed the bidding game method in eliciting the WTP. The bidding game method in CVM is preferred for estimating WTP for improved healthcare provision in urban slums as it reduces hypothetical bias by engaging households in a real or quasi-real auction environment, encouraging more truthful and accurate responses [16, 29, 30]. This method aligns incentives with truthful reporting, enhances precision through an iterative bidding process, and minimizes strategic misrepresentation. In addition, by simulating actual decision-making, it offers a more reliable estimate of WTP, better reflecting real-world behavior. Furthermore, the dynamic nature of the bidding game ensures that WTP is captured more accurately than in single-point elicitation methods. The bidding game approach improves both the validity and reliability of WTP estimates.
Scenario: Linking the informal healthcare providers to the formal system is an innovative method of raising the quality of services that are provided by the informal healthcare providers in the urban slums and for strengthening the health system in urban areas. The linkage will involve capacity development of informal providers, getting them involved in health service delivery and health financing interventions of the public sector. It will also involve sharing data, especially ensuring the data from the informal sector into the National Health Management Information System. The benefits to the informal healthcare providers will be that they will be trained to provide better quality services and will be working closely with the government to deliver some services and link to the referral systems. The PHCs and other agencies will also be providing supportive supervision of the informal sectors. However, undertaking the linkage of informal with the formal sector will require some administrative costs that will be borne by the promoters of the linkage and potential beneficiaries such as you. The ultimate benefit is that the linked informal healthcare providers will be able to provide consumers such as you with improved quality healthcare services.
This section is designed to help understand the extent that you will be willing to pay to receive improved quality of services from linked informal healthcare providers (Box 1). The payment will be made once (one-off payment) to an agreed ward health development account designed solely to help improve quality care provision that is always lacking in slums.
Data Analysis
The unit of analysis for this study was the household, and the analyses were undertaken using univariate, bivariate, and multiple-variable methods. For the univariate method, frequencies, percentages and means were majorly used to describe the sample characteristics. The bivariate method involved the use of crosstabulation to assess the association of willingness to pay additional money for improved quality of service with sociodemographic characteristics. The multivariate analysis was carried out using binary logistic regression and Tobit regression. The modeling approach has been used in previous WTP studies [31, 32]. The first step of binary regression helps to understand the relationship of the decision on whether willing or not willing to pay for the service. The use of the Tobit model is appropriate for limited dependent variables that we generated in the study. The Heckman models are also possibilities, but the Tobit model was appropriate in our study. In the Tobit regression, a logarithm + 1 (log + 1) transformation was used for WTP estimates starting from zero to ensure that zeros were not treated as missing values after long transformation [32, 33].
The choice of the methods was based on the binary nature of the dependent variable for the logistics regression, while for the topic, the Tobit regression method was selected because of the prevalence of missing values, which it treats, as explained above. The household asset index and the socioeconomic status (SES) score were assessed using the principal component analysis (PCA). The SES score was used to divide the households into quintiles.
Table 1 highlights the a priori expectations for the quantitative variables in the models estimated, which identified the determinants of household willingness to pay and the maximum amount they are willing to pay for improved quality of healthcare services. We focused on variables that are expected to explain the demand (WTP) function and have been explored mostly in previous WTP studies. In the table, the variables are continuous and the expected behavior positive. The implication is that whenever any of them is going up, the dependent variable will also increase.
Table 1.
A priori expectation for quantitative variables showing their association with willingness to pay and maximum willingness to pay
| Variables | Nature of data | Expected sign |
|---|---|---|
| Household size | Continuous | Positive (+) |
| Age | Continuous | Positive (+) |
| Asset index | Continuous | Positive (+) |
| Socioeconomic status | Continuous | Positive (+) |
Results
Table 2 presents a summary of continuous variables in the model. Average household size was 5.3 people, average age of respondents was 32.2 years, average length of time in schooling was 11.5 years, mean asset index score was −3.3, and the socioeconomic status had an average value of −7.1, ranging from −1.5 to 6.2
Table 2.
Descriptive summary of the quantitative variables in the model
| Variables | N | Mean | SD | Minimum | Maximum |
|---|---|---|---|---|---|
| Household size | 1025 | 5.3 | 1.8 | 1 | 12 |
| Age | 1025 | 32.2 | 7.4 | 16 | 65 |
| Total number of years spent schooling | 1025 | 11.5 | 2.8 | ||
| Asset index | 1025 | −3.3 | 1 | −2.2 | 2.9 |
| Socioeconomic status | 1025 | −7.1 | 1 | −1.5 | 6.2 |
SD standard deviation
Table 3 shows that out of the 1025 respondents, 95.9% were female. More than half of the respondents, 618 (61.7%), had, at least, post basic education, and 641 (62.5%) have their major source of income in petty trading, artisan works, and farming.
Table 3.
Sociodemographic characteristics of respondents (univariate)
| Characteristic | N = 1025 |
|---|---|
| Status in the household | |
| Female head of household | 491 (47.9%) |
| Wife | 459 (44.8%) |
| Representative of household | 37 (3.6%) |
| Male head of household | 36 (3.5%) |
| Grandmother | 2 (0.2%) |
| Gender of the respondent | |
| Female | 983 (95.9%) |
| Male | 42 (4.1%) |
| Highest completed education level | |
| Post basic education | 618 (61.7%) |
| Basic education | 240 (24.0%) |
| University education | 144 (14.4%) |
| Main occupation for generating income | |
| Petty trading (artisan, farming) | 641 (62.5%) |
| Employed (government, private, self, big business) | 258 (25.2%) |
| Unemployed | 126 (12.3%) |
Table 4 shows the level of respondents’ willingness to pay one-off payment to have improved health services. The majority, 64.5%, of the households were willing to make a one-off payment for a linkage intervention that will ensure improved quality of health services from IHPs. The willingness-to-pay amount ranged from ₦100 to ₦20,000. The majority (95.6%) of those who expressed a willingness to make the one-off payment were willing to pay ₦500 ($0.77). The mean WTP was ₦1353 ($2.08), and 27.1% of the households expressed WTP an amount that was up to the mean WTP.
Table 4.
Willingness to pay for improved healthcare services from the linked informal healthcare providers
| Characteristic | N = 1025 |
|---|---|
| Willing to make a payment | 661 (64.5%) |
| Not willing to make a payment | 364 (35.5%) |
| Why not willing to pay | |
| Government should pay | 225 (61.8%) |
| It should be done free of charge | 182 (50.0%) |
| The informal providers should pay | 13 (3.6%) |
| The community leaders should pay | 2 (0.5%) |
| Not bothered | 6 (1.6%) |
| Willingness by price point (among those willing to pay, N = 661) | |
| Willing to pay ₦500 | 632 (95.6%) |
| Willing to pay ₦300 | 16 (2.4%) |
| Willing to pay ₦700 | 480 (72.6%) |
| Willing to pay ₦100 | 10 (1.5%) |
| WTP amount (₦) | |
| Minimum amount | 100 |
| Maximum amount | 20,000 |
| Mean maximum WTP amount | 1352.9 |
| Median amount | 1000 |
| Number willing to pay minimum amount | 2 (0.3%) |
| Number willing to pay maximum amount | 1 (0.2%) |
| Number willing to pay mean maximum amount or higher | 179 (27.1%) |
WTP willingness to pay
Table 5 presents the results of chi-squared test of association between willingness to pay for improved quality of service and sociodemographic and household characteristics. Of the variables tested, WTP for improved quality of healthcare services had significant associations with respondent’s status/position in the household, occupation, facility where they usually visit to seek health care, household asset ownership and household socioeconomic status (p < 0.05).
Table 5.
Chi-squared test of association between WTP for improved quality of healthcare service and sociodemographic and household characteristics
| Characteristic | Not willing, N = 364 | Willing, N = 661 | P value |
|---|---|---|---|
| Status in the household | 0.00 | ||
| Female head of household | 155 (42.6%) | 336 (50.8%) | |
| Wife | 194 (53.3%) | 265 (40.1%) | |
| Representative of household | 9 (2.5%) | 28 (4.2%) | |
| Male head of household | 5 (1.4%) | 31 (4.7%) | |
| Grandmother | 1 (0.3%) | 1 (0.2%) | |
| Sex | 0.05 | ||
| Female | 355 (97.5%) | 628 (95.0%) | |
| Male | 9 (2.5%) | 33 (5.0%) | |
| Highest completed education level | 0.8 | ||
| Post basic education | 220 (61.8%) | 398 (61.6%) | |
| Basic education | 88 (24.7%) | 152 (23.5%) | |
| University education | 48 (13.5%) | 96 (14.9%) | |
| Main occupation | 0.02 | ||
| Petty trading (artisan, farming) | 249 (68.4%) | 392 (59.3%) | |
| Employed (government, private, self, big business) | 76 (20.9%) | 182 (27.5%) | |
| Unemployed | 39 (10.7%) | 87 (13.2%) | |
| Marital status | 0.09 | ||
| Living with spouse | 324 (89.0%) | 594 (89.9%) | |
| Never married | 13 (3.6%) | 36 (5.4%) | |
| Widowed | 15 (4.1%) | 22 (3.3%) | |
| Divorced/separated | 12 (3.3%) | 9 (1.4%) | |
| Type of facility household uses | 0.00 | ||
| Formal facility only | 7 (1.9%) | 44 (6.6%) | |
| Informal facility only | 50 (13.7%) | 97 (14.7) | |
| Both formal and informal facility | 307 (84.3%) | 520 (78.7%) | |
| Asset ownership quintile | 0.00 | ||
| Lowest asset ownership | 97 (26.7%) | 108 (16.3%) | |
| Low asset ownership | 66 (18.1%) | 145 (21.9%) | |
| Middle asset ownership | 64 (17.6%) | 135 (20.4%) | |
| High asset ownership | 72 (19.8%) | 142 (21.5%) | |
| Highest asset ownership | 65 (17.9%) | 131 (19.8%) | |
| SES quintile | 0.00 | ||
| Lowest SES score | 74 (25.7%) | 90 (17.0%) | |
| Low SES score | 74 (25.7%) | 89 (16.8%) | |
| Middle SES score | 62 (21.5%) | 102 (19.3%) | |
| High SES score | 44 (15.3%) | 119 (22.5%) | |
| Highest SES score | 34 (11.8%) | 129 (24.4%) |
SES socioeconomic status
Table 6 highlights results of logistic regression of factors associated with willingness to pay for improved healthcare service provision in urban slums. Respondents who identified as wives were significantly less likely to express willingness to pay for improved quality of healthcare services compared with respondents who were female heads of household (odds ratio (OR) 0.59; p < 0.01).
Table 6.
Logistic regression of factors associated with willingness to pay for improved healthcare services in urban slums
| Variables | Odds ratio | Z | P > |z| |
|---|---|---|---|
| Household status (female head of household) | |||
| Male head of household | 4.49 | 1.54 | 0.12 |
| Wife | 0.59 | −3.49 | 0.00* |
| Grandmother | 0.63 | −0.30 | 0.77 |
| Representative of household | 2.27 | 1.42 | 0.15 |
| Household size | 0.99 | −0.31 | 0.76 |
| Facility type used by households (only formal) | |||
| Only informal facility | 0.40 | −1.98 | 0.04* |
| Combination of formal and informal facilities | 0.34 | −2.51 | 0.01* |
| Sex (female) | 2.10 | 0.96 | 0.34 |
| Age | 0.98 | −1.89 | 0.06 |
| Marital status (living with spouse) | 1.96 | 1.75 | 0.08 |
| Asset index | 1.28 | 3.18 | 0.00* |
| SES score | 1.86 | 3.65 | 0.00* |
| Prob > χ2 | 0.00 | ||
| Pseudo R2 | 0.05 | ||
| N | 1017 |
*Signifies statistical significance at p < 0.05. Prob = Probability that observed relationships between variables happened by chance. Pseudo = Measure of how well the model fits that data
Compared with households that usually seek care from formal providers, those that usually seek care from only informal providers were significantly less likely to express willingness to pay for improved quality of healthcare services (OR 0.4; p = 0.01). Similarly, those that usually seek care from both formal and informal providers were significantly less likely to pay for improved quality of healthcare services (OR 0.34; p = 0.01)
Higher asset index and socioeconomic status were positively associated with willingness to pay for improved quality of healthcare services. Each unit increase in asset index was associated with a 28% increase in the odds of willingness to pay (OR 1.28; p < 0.001), while each unit increase in SES score was associated with an 86% increase in the odds of willingness to pay (OR 1.86; p < 0.001).
Table 7 presents results of Tobit regression of determinants of maximum WTP for improved quality of healthcare service in urban slums. Findings show that holding all other factors constant, on average, the latent WTP for improved quality of health services is 1.94 units lower for households that typically seek care from only informal providers compared with those that use formal providers. Similarly, latent WTP is 1.99 units lower for households that typically seek care from both formal and informal providers compared with those that use only formal providers.
Table 7.
Tobit regression result for the determinants of maximum WTP for improved healthcare services in urban slums
| Variables | Coefficient | T | P > |z| |
|---|---|---|---|
| Household status (female head of household) | |||
| Male head of household | 2.45 | 1.36 | 0.17 |
| Wife | −0.64 | −1.54 | 0.12 |
| Representative of household | 1.09 | 0.89 | 0.38 |
| Household size | −0.03 | −0.04 | 0.97 |
| Facility type used by households (only formal) | |||
| Only informal facility | −1.94 | −2.12 | 0.03* |
| Combination of formal and informal facilities | −1.99 | −2.47 | 0.01* |
| Sex (female) | 0.74 | 0.45 | 0.65 |
| Age (years) | −2.11 | −2.29 | 0.02* |
| Marital status (living with spouse) | 0.47 | 0.64 | 0.52 |
| Asset index | 0.41 | 2.03 | 0.04* |
| SES score | 0.91 | 4.45 | 0.00* |
| Constant | 11.67 | 3.16 | 0.00* |
| /Sigma | 4.79 | ||
| Prob > χ2 | 0.00 | ||
| Pseudo R2 | 0.02 | ||
|
N LL (0) |
817 |
*Signifies statistical significance at p < 0.05. Prob = Probability that observed relationships between variables happened by chance. Pseudo = Measure of how well the mod fits the data. /Sigma = Measure of the variability of the dependent variable not explained by the predictors. LL = log of the likelihood of the fitted model a
Tobit regression analysis also showed that age was negatively associated with mean WTP for improved quality of healthcare services, while asset ownership and socioeconomic status were positively associated. Holding other variables constant, a one-unit increase in age (1 year) is associated with a 2.11 unit decrease in the latent mean WTP for improved quality of health services; a one-unit increase in the asset index is associated with a 0.41 unit increase in the latent mean WTP; and for each one-unit increase in SES score, the latent mean WTP increases by 0.91 units.
Discussion
The results show that a considerable proportion of households in urban slums are willing to pay for improved quality of services from informal health providers, who are linked to the formal health system. Studies have shown that individuals, especially in low-income and underserved populations, are willing to pay for better healthcare services if they perceive an improvement in quality of care [34, 35]. A significant driver of WTP is trust in the healthcare provider [36, 37], and linking informal health providers with the formal (regulated) system can increase trust in their services [38]. This supports some of the current drive of integrating informal health providers into the formal health system, so that the formal health system can collaboratively improve the quality of services of IHPs through better training, adherence to standards, and accountability [9, 39].
The theoretical validity of the elicited WTP could be affirmed from the factors that explained it. These were variables such as respondent’s status/position in the household, occupation, asset index, SES, and facility where they usually visit to seek health care. These findings corroborate existing research, highlighting the multifaceted nature of healthcare preferences in resource-constrained urban slum areas [40, 41]. The role of occupation and SES in determining WTP for healthcare improvements add to the theoretical validity of the WTP estimates because both factors reflect income levels, job stability, and overall economic well-being, which are crucial factors influencing an individual’s capacity to pay for enhanced healthcare services [42]. Healthcare preferences among slum dwellers are often shaped by the interplay between economic constraints and the perceived quality of services, and households make trade-offs between cost, quality, and accessibility when choosing healthcare options [41]. Hence, linkage interventions that seek to enhance quality of care must remain sensitive to the diverse preferences and economic realities of slum dwellers to ensure access to healthcare for a wider segment of the slum population.
In addition, the choice of healthcare facilities emerged as a significant determinant in keeping with literature [43]. Although experiences of care, accessibility, and perceived quality of services at different facilities play a pivotal role in shaping residents’ willingness to invest in better healthcare [35], it is intuitive for people to want to invest in interventions that will improve the quality of services of their preferred providers. The finding that the type of health facility households typically use influences their willingness to pay (WTP) for improved services from informal health providers has important implications for both policy and practice. It suggests the need for more effective integration between informal providers and the formal health system by creating stronger and efficient referral pathways and protocols.
Moreover, the finding that households that use informal health facilities, or a combination of informal and formal providers, exhibit lower WTP for the proposed linkage compared with those that use only formal facilities was interesting. This could imply that households relying solely on informal providers value quality improvement less, or may be more price-sensitive, or may have different expectations about what constitutes “quality” in healthcare. The finding aligns with economic theories suggesting that individuals who rely on informal care often perceive it as more affordable and accessible, thus undervaluing potential improvements from formal integration. Similar findings corroborate that cost sensitivity and satisfaction with informal care can reduce individuals’ willingness to invest in formal system integration [44]. However, when trust in the formal system is built and clear benefits are demonstrated, individuals are more likely to value the integration of informal and formal providers [9].
Although most of the households were willing to make the one-off payment for the proposed capacity upgrade for informal healthcare providers, the mean maximum amount they were willing to pay was minimal. To be precise, household health expenditure choices is greatly influenced by their capacity to make the required payment for the choice. As expected, household SES and their asset ownership index exhibited a significant positive relationship with the maximum amount the households were willing to pay. This confirms that the maximum amount households are willing to pay increases with their wealth storage capacity measured by their asset ownership and their SES [45]. Moreover, age, household usage of only informal providers, and household usage of both formal and informal providers compared with those that use only formal providers reduces the maximum amount they are willing to pay. These connect to the determinants of their willingness to pay for the capacity upgrading linkage for better healthcare provision in urban slums.
It is generally known that females usually have varied roles in households, and it is not practical to straitjacket them into one role. However, males are usually more economically buoyant since they indulge more in income-yielding activities compared with the females. This can explain their much higher WTP amounts.
However, the finding that some respondents were not willing to pay for the linkages to improve the quality of services that they receive, although worrisome, stemmed from some valid reasons. Consumers of health care services, especially those of lower socioeconomic status, may assume that health services should be provided by the government, therefore becoming unwilling to pay for improved quality of healthcare services. Hence, this raises sustainability concerns for health financing. Onwujekwe and Uzochukwu (2004), in their study of households, stated an actual altruistic willingness to pay for insecticide-treated nets and proposed a community-based financing mechanism in which wealthier households pay the full cost of services and those who are willing to pay more can subsidize for the poor [19]. The linkage intervention could also explore models such as co-payment schemes or community-based financing that retain an element of government support but also require user buy-in.
Although this paper provides new information on what consumers are willing to invest in for quality improvements from informal providers, the study did not elicit the quality attributes that consumers are willing to pay for. Hence, the paper does not give a clear understanding of consumer preferences and falls short in recommendations of targeted improvements. A second limitation of the study is that since the associated costs of integrating informal providers into the formal healthcare system were not known, hypothetical scenarios were used to elicit willingness to pay. This may have introduced hypothetical bias and limited the generalizability of the findings to actual payment behaviors. In addition, although an earlier paper in the study area reported that there was no conclusive evidence of starting point bias [46], the existence of starting point bias cannot be completely ruled out in bidding game. To mitigate this, our starting bid was based on community inputs through FGDs.
Conclusions
This study provides valuable insights into the willingness to pay for improved health services in an urban slum context in LMICs. It underscores the importance of addressing socioeconomic disparities and the quality of healthcare facilities to enhance healthcare outcomes for vulnerable populations in such settings. Further research and policy initiatives should build upon these findings to create more effective and equitable healthcare systems in urban slums across the globe.
Recognizing the factors that influence WTP can inform the development of targeted interventions aimed at improving healthcare services in urban slums. Policymakers should focus on strategies that enhance healthcare access, quality, and affordability, considering the socioeconomic diversity within these areas. It is also vital to engage with communities to understand their specific needs and preferences. Moreover, policymakers should also explore community-based financing models, such as co-payment schemes, where wealthier households can subsidize the cost of care for poorer households. This approach would address the sustainability concerns of health financing, creating a more inclusive and sustainable system that accommodates the financial constraints of low-income populations while enhancing the overall quality of care.
Acknowledgements
The authors thank Dr. Joe Hicks for his comments on an earlier draft of the paper.
Declarations
Funding
The research leading to these results was funded by UK aid from the British people through the Community-led Responsive and Effective Urban Health Systems (CHORUS) research project.
Conflicts of interest
The authors have no competing interests to declare.
Availability of data and material
The data that support the findings of this study are available from Health Policy Research Group (HPRG), University of Nigeria, but restrictions apply to the availability of these data and so are not publicly available. The data and codes that were used to run the analysis are, however, available from the corresponding author, Chinyere Ojiugo Mbachu, upon reasonable request and with the permission of HPRG.
Ethics approval
Ethical approval for the study was obtained from University of Leeds School of Medicine Research Ethics Committee (MREC 21-009) and the Health Research Ethics Committee of University of Nigeria Teaching Hospital Enugu (NHREC/05/01/2008B-FWA00002458-1RB00002323).
Consent to participate
Written informed consent was obtained from each participant before data was collected.
Consent for publication
Not applicable.
Author contributions
Obinna Onwujekwe contributed to the study conception and design, material preparation, data collection and analysis, and drafting and review of the manuscript. Uche Godwin Ezema participated in material preparation, data collection and analysis, and review of the manuscript. Chukwudi Nwokolo participated in material preparation, data collection and analysis, and drafting of the manuscript. Chinyere Ojiugo Mbachu contributed to the study conception and design, material preparation, data collection, and drafting of the manuscript.
References
- 1.Mbachu CO, Arize I, Obi C, Ebenso B, Elsey H, Onwujekwe O. Assessing knowledge of hypertension and diabetes mellitus among informal healthcare providers in urban slums in southeastern Nigeria. Discover Public Health. 2024;21(1):21. [Google Scholar]
- 2.UN Habitat. Nigeria: Onitsha Urban Profile. In. Nairobi; 2012.
- 3.Onwujekwe O, Hanson K, Uzochukwu B. Do poor people use poor quality providers? Evidence from the treatment of presumptive malaria in Nigeria. Trop Med Int Health. 2011;16(9):1087–98. [DOI] [PubMed] [Google Scholar]
- 4.Veríssimo PF, da Silva ARA. Availability and utilization of new technologies for prevention and control of healthcare-associated infections. medRxiv 2023:2023.2003. 2009.23287055.
- 5.Siddiqi KA, Dana T, Hasan K, Haider MR. Mother’s healthcare response to child illness: a slum-based cross-sectional study in Dhaka city, Bangladesh. Columbia Univ J Glob Health. 2017;7(2):16–25. [Google Scholar]
- 6.Bhanderi MN, Kannan S. Untreated reproductive morbidities among ever married women of slums of Rajkot City, Gujarat: the role of class, distance, provider attitudes, and perceived quality of care. J Urban Health. 2010;87:254–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Adane M, Mengistie B, Mulat W, Kloos H, Medhin G. Utilization of health facilities and predictors of health-seeking behavior for under-five children with acute diarrhea in slums of Addis Ababa, Ethiopia: a community-based cross-sectional study. J Health Popul Nutr. 2017;36:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Černauskas V, Angeli F, Jaiswal AK, Pavlova M. Underlying determinants of health provider choice in urban slums: results from a discrete choice experiment in Ahmedabad, India. BMC Health Serv Res. 2018;18(1):473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Onwujekwe O, Mbachu C, Onyebueke V, Ogbozor P, Arize I, Okeke C, Ezenwaka U, Ensor T. Stakeholders’ perspectives and willingness to institutionalize linkages between the formal health system and informal healthcare providers in urban slums in southeast, Nigeria. BMC Health Serv Res. 2022;22(1):583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sieverding M, Liu J, Beyeler N. Social support in the practices of informal providers: the case of patent and proprietary medicine vendors in Nigeria. Soc Sci Med. 2015;143:17–25. [DOI] [PubMed] [Google Scholar]
- 11.Sieverding M, Beyeler N. Integrating informal providers into a people-centered health systems approach: qualitative evidence from local health systems in rural Nigeria. BMC Health Serv Res. 2016;16(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hengelaar AH, van Hartingsveldt M, Wittenberg Y, van Etten-Jamaludin F, Kwekkeboom R, Satink T. Exploring the collaboration between formal and informal care from the professional perspective—a thematic synthesis. Health Soc Care Community. 2018;26(4):474–85. [DOI] [PubMed] [Google Scholar]
- 13.World Health Organization. World Bank and WHO: Half the world lacks access to essential health services, 100 million still pushed into extreme poverty because of health expenses. Press Release 2017; 13.
- 14.Brahmachari R, Mishra M, Gotsadze G, Mandal S. Leveraging the social networks of informal healthcare providers for universal health coverage: insights from the Indian Sundarbans. Health Policy Plan. 2024;39(Supplement_2):i105–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Johannesson M. Theory and methods of economic evaluation of health care, vol. 4. Berlin: Springer Science & Business Media; 1996. [PubMed] [Google Scholar]
- 16.Onwujekwe O. Criterion and content validity of a novel structured haggling contingent valuation question format versus the bidding game and binary with follow-up format. Soc Sci Med. 2004;58(3):525–37. [DOI] [PubMed] [Google Scholar]
- 17.Agada-Amade YA, Ogbuabor DC, Obikeze E, Eborieme E, Onwujekwe OE. Willingness to pay for haemodialysis among patients with chronic kidney disease in Abuja, Nigeria. BMC Nephrol. 2024;25(1):29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Onwujekwe O, Shu E, Chima R, Onyido A, Okonkwo P. Willingness to pay for the retreatment of mosquito nets with insecticide in four communities of south-eastern Nigeria. Trop Med Int Health. 2000;5(5):370–6. [DOI] [PubMed] [Google Scholar]
- 19.Onwujekwe O, Uzochukwu B. Stated and actual altruistic willingness to pay for insecticide-treated nets in Nigeria: validity of open-ended and binary with follow-up questions. Health Econ. 2004;13(5):477–92. [DOI] [PubMed] [Google Scholar]
- 20.Wiseman V, Onwujekwe O, Matovu F, Mutabingwa TK, Whitty CJ. Differences in willingness to pay for artemisinin-based combinations or monotherapy: experiences from the United Republic of Tanzania. Bull World Health Organ. 2005;83(11):845–52. [PMC free article] [PubMed] [Google Scholar]
- 21.Ughasoro M, Udem N, Chukwudi N, Korie F, Uzochukwu B, Onwujekwe E. Caregivers f willingness to Pay for a topical anesthetic cream for minor medical procedures in children. Niger J Clin Pract. 2014;17(4):507–10. [DOI] [PubMed] [Google Scholar]
- 22.Klose T. The contingent valuation method in health care. Health Policy. 1999;47(2):97–123. [DOI] [PubMed] [Google Scholar]
- 23.Morrison GC, Gyldmark M. Appraising the use of contingent valuation. Health Econ. 1992;1(4):233–43. [DOI] [PubMed] [Google Scholar]
- 24.Mathiyazaghan K. Willingness to pay for rural health insurance through community participation in India. Int J Health Plann Manage. 1998;13(1):47–67. [DOI] [PubMed] [Google Scholar]
- 25.National Bureau of Statistics. Demographic statistics bulletin. In. Edited by Demographic Statistics Division. Abuja: National Bureau of Statistics; 2022.
- 26.ICF International. Demographic and Health Survey Sampling and Household Listing Manual. Calverton: ICF International 2012.
- 27.González-Cabán A, Loomis J. Economic benefits of maintaining ecological integrity of Rio Mameyes, Puerto Rico. Ecol Econ. 1997;21(1):63–75. [Google Scholar]
- 28.Onwujekwe O, Chima R, Shu E, Nwagbo D, Okonkwo P. Hypothetical and actual willingness to pay for insecticide-treated nets in five Nigerian communities. Trop Med Int Health. 2001;6(7):545–53. [DOI] [PubMed] [Google Scholar]
- 29.Whittington D, Briscoe J, Mu X, Barron W. Estimating the willingness to pay for water services in developing countries: a case study of the use of contingent valuation surveys in southern Haiti. Econ Dev Cult Change. 1990;38(2):293–311. [Google Scholar]
- 30.Mitchell RC, Carson RT. Using surveys to value public goods: the contingent valuation method. 1st ed. Washington: RFF Press; 1989. [Google Scholar]
- 31.Donaldson C, Jones AM, Mapp TJ, Olson JA. Limited dependent variables in willingness to pay studies: applications in health care. Appl Econ. 1998;30(5):667–77. [Google Scholar]
- 32.Onwujekwe O, Fox-Rushby J, Hanson K. Construct validity of the bidding game, binary with follow-up, and a novel structured haggling question format in determining willingness to pay for insecticide-treated mosquito nets. Med Decis Mak. 2008;28(1):90–101. [DOI] [PubMed] [Google Scholar]
- 33.Donaldson C. Valuing the benefits of publicly-provided health care: does ‘ability to pay’ preclude the use of ‘willingness to pay’? Soc Sci Med. 1999;49(4):551–63. [DOI] [PubMed] [Google Scholar]
- 34.Jeetoo J, Jaunky VC. Willingness to pay to improve quality of public healthcare services in Mauritius. Healthcare (Basel). 2021;10(1):43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Pavel MS, Chakrabarty S, Gow J. Assessing willingness to pay for health care quality improvements. BMC Health Serv Res. 2015;15(1):43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Nzowa PG, Nandonde FA, Seimu SML. Mediation effect of trust on willingness to pay for health insurance among co-operative members in Tanzania. Future Bus J. 2023;9(1):18. [Google Scholar]
- 37.Ai Y, Rahman MK, Newaz MS, Gazi MAI, Rahaman MA, Mamun AA, Chen X. Determinants of patients’ satisfaction and trust toward healthcare service environment in general practice clinics. Front Psychol. 2022;13: 856750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Christian BN, Christian NG, Keshinro MI, Olutade-Babatunde O. How to build bridges for universal health coverage in Nigeria by linking formal and informal health providers. BMJ Glob Health. 2023;8(11):e014165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Odii A, Arize I, Agwu P, Mbachu C, Onwujekwe O. To what extent are informal healthcare providers in slums linked to the formal health system in providing services in sub-Sahara Africa? A 12-year scoping review. J Urban Health. 2024;101:1248–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Park JE, Kibe P, Yeboah G, Oyebode O, Harris B, Ajisola MM, Griffiths F, Aujla N, Gill P, Lilford RJ, et al. Factors associated with accessing and utilisation of healthcare and provision of health services for residents of slums in low and middle-income countries: a scoping review of recent literature. BMJ Open. 2022;12(5): e055415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fayehun O, Madan J, Oladejo A, Oni O, Owoaje E, Ajisola M, Lilford R, Omigbodun A. What influences slum residents’ choices of healthcare providers for common illnesses? Findings of a discrete choice experiment in Ibadan, Nigeria. PLOS Glob Public Health. 2023;3(3): e0001664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Steigenberger C, Flatscher-Thoeni M, Siebert U, Leiter AM. Determinants of willingness to pay for health services: a systematic review of contingent valuation studies. Eur J Health Econ. 2022;23(9):1455–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Nugraheni DA, Satibi S, Kristina SA, Puspandari DA. Factors associated with willingness to pay for cost-sharing under universal health coverage scheme in Yogyakarta, Indonesia: a cross-sectional survey. Int J Environ Res Public Health. 2022;19(22):15017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Onwujekwe O, Hanson K, Fox-Rushby J. Inequalities in purchase of mosquito nets and willingness to pay for insecticide-treated nets in Nigeria: challenges for malaria control interventions. Malar J. 2004;3:1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Mbachu C, Okoli C, Onwujekwe O, Enabulele F. Willingness to pay for antiretroviral drugs among HIV and AIDS clients in south-east Nigeria. Health Expect. 2018;21(1):270–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Onwujekwe O, Nwagbo D. Investigating starting-point bias: a survey of willingness to pay for insecticide-treated nets. Soc Sci Med. 2002;55(12):2121–30. [DOI] [PubMed] [Google Scholar]