Hrev_master [Healthcare in Low-resource Settings 2018; 6:7464] [page 27] Impoverishing effect of household healthcare expenditure in semi-rural communities in Yenagoa, Nigeria Adedotun Daniel Adesina, Daprim Samuel Ogaji Department of Preventive and Social Medicine, University of Port Harcourt, Nigeria Abstract Inequity in the payment mechanism for healthcare systematically affects poorer households more than the rich. This article examines the impoverishing effect of healthcare expenditure on households in Yenagoa. Data was obtained from a cross- sectional survey of households in 2 commu- nities in Yenagoa selected by simple ran- dom sampling. A pretested, structured, interviewer-administered questionnaire was used to obtain information on household (HH) income, general expenditures and financing for healthcare. Two international poverty lines designed by World Bank were employed to classify households as poor, extremely poor and to determine the impov- erishing effects of households’ healthcare expenditures. Responses were received from 525 HHs with 9.2% of HHs falling below poverty line, another 9% pushed deeper into poverty after healthcare spend- ing. A 12.3% and 16% increase in the poverty and extreme poverty gaps respec- tively were attributable to health payment. A significant percentage of households who were non-poor were pushed into poverty after healthcare spending. There is need for increased public spending and implementa- tion of innovative pre-payment mechanisms and social insurance that assures financial- risk protection and equity in health financ- ing in Yenagoa. Introduction Inequities in health financing systemat- ically place households (HHs), especially the poor ones, who are already socially dis- advantaged at further disadvantage with respect to their health. Hence, health financ- ing options which ensure achievement of the core objectives of ‘goodness’ and ‘fair- ness’ of health systems should be the focus of policy makers and stakeholders in devel- oping equitable distribution of qualitative healthcare goods and services. However, many health systems especially in develop- ing countries are mainly financed privately through out-of-pocket payments for health- care at the point of access.1 The direct out-of-pocket (OOP) pay- ment for health services is an inequitable way to finance a health system as it places great financial burden on households,1 excludes financial solidarity2 and could compel many households to forgo basic needs such as education, food, and housing in order to pay for healthcare.2-4 The usual consequence is that they suffer financial catastrophe or even impoverishment while seeking healthcare. They may also totally avoid or delay to seek necessary healthcare where the cost is perceived to exceed their ability to pay.3-5 Nigeria, like many lower middle-income countries (LMIC), relies on OOP payments for financing health servic- es. Fund from private sources is responsible for 75% of total expenditure on health (THE) and 90% of this is OOP payments.6,7 This payment modality prevent people from seeking or continuing care, while some who do seek care incur catastrophic financial burdens that push them into poverty.8 The ensuing vicious cycle of poverty further magnifies the need for healthcare while shrinking the capacity of household to pay for it.9 Household spending on health in set- tings like Nigeria, can also disrupt their budget, making it impossible to meet some essential expenditure in the home.10-16 Catastrophic health expenditure which occurs when healthcare expenditures exceed pre-defined proportions of house- hold income and/or non-food expenditure does not completely demonstrate the extent of hardship bore by household after such expenses.11-13,17,18 The concept of impover- ishment after healthcare spending paints a clearer picture of this financial burden as it demonstrates how expenditure on health- care could push households into poverty or further down the poverty line.19 A study done in Kenya reported that 3.5% of house- holds and 4% of households were impover- ished by health spending in 2003 and 2007 respectively. Outside the continent of Africa, study done in Brazil also revealed an increasing trend like in Kenya as poverty headcount increased from 6.8% in 2002/2003 to 11.6% in 2008/2009.15 Though a multifaceted social menace, poverty can be measured by the poverty line which defines a monetary threshold below which it becomes difficult for individuals or households to afford basic needs. Poverty lines are commonly defined in relation to average household subsistence spending or ‘food share’20 and household per capita income. The World Bank had developed the international poverty lines using per capita income of households, adjusted for purchas- ing power parity. The most recent thresh- olds are US$1.9 per capita per day and US$3.1 per capita per day for extreme poverty and poverty respectively.9,11,15 A non-poor household that becomes poor after paying for healthcare is said to be impover- ished by such health expenditures with ref- erence to any of the defined poverty lines.12- 15,20 There is a paucity of studies that quan- tified the impoverishing effect of health spending on households in this setting. This study aims to bridge this gap by investigat- Healthcare in Low-resource Settings 2018; volume 6:7464 Correspondence: Adedotun Daniel Adesina, Department of Preventive and Social Medicine, University of Port Harcourt, Nigerian Law School, Yenagoa Campus, PMB 60, Yenagoa, Bayelsa State, Nigeria. Tel.: +234.8034469945 E-mail: adeshinadedotun@yahoo.com Key words: Out-of-pocket; Impoverishing effect; Poverty line; Healthcare financing; Yenagoa. Acknowledgements: the authors are grateful to Johnson Igoniderigha and Ipusi Ikiobho staff of Bayelsa Geographic Information System who assisted them in identifying the geographical zones of Yenagoa and assisted in data collection in conjunction with Ibarakumo Sokare and Amani Samuel. They are also grateful to the study communities for agreeing to participate in the study. Finally, they thank Adesina Ileola, Emudiaga-Ohwerhi, Mabel, and Diete-spiff Kaine for their contributions to the project design and overall management of the study. Contributions: DAA was involved in the con- ceptualization and design of the study, super- vised data collection, data processing and analysis and the initial write up of the manu- script; DSO was involved in the conceptual- ization and design of the study and the study instrument; he reviewed the data processing and analysis and finally reviewed the manu- script. Both authors approved the final manu- script for publication. Conflict of interest: the authors declare no potential conflict of interest. Funding: none. Received for publication: 5 April 2018. Revision received: 8 July 2018. Accepted for publication: 30 July 2018. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright A.D. Adesina and D.S. Ogaji, 2018 Licensee PAGEPress, Italy Healthcare in Low-resource Settings 2018; 6:7464 doi:10.4081/hls.2018.7464 No n- co mm er cia l u se on ly ing the incidence and depth of poverty due to healthcare spending among households in Yenagoa where OOP mechanism of pay- ment is the predominant payment method for healthcare. Materials and Methods Study setting The study was conducted in Yenagoa, one of the traditional homes of the Ijaw peo- ple and the capital of Bayelsa state. It is located on the banks of Epie and Ekole Creeks, major tributaries of the Nun River between 4o 47’ 15” and 5o 11’ 55” North of the equator and Long. 6o 07’ 35” and 6o 24’ 00” East of the Greenwich meridian.21,22 Yenagoa is a semi-rural settlement made up of 21 communities linearly arranged along both sides of the Mbiama- Yenagoa road22 inhabited by people who engage in fishing, farming, palm oil produc- tion, palm wine tapping, local gin making, lumbering, craving and weaving. Communities in Yenagoa are organized as compounds with representative family heads.23 Yenagoa has at least one primary health care centre in each of its ward and major communities. It is also served by two terti- ary health institutions, a number of private hospitals/clinics, patent medicine dealers, and a wide range of non-formal healthcare providers (including traditional medicine practitioners). Study design The study is a cross-sectional survey of households in two randomly selected com- munities in Yenagoa (Akenfa and Kpansia). Sampling The 2 study communities were selected from the existing 21 communities by simple random sampling. Households were recruit- ment with the help of the Bayelsa Geographic Information System (BGIS). The three geographical zones of these 2 communities were demarcated and zones 2 and 3 in Kpansia and Akenfa communities respectively were chosen for the study. All roads in the 2 selected zones were identified as clusters from which houses were chosen by systematic random sampling, using the new BGIS numbering system. The inter- views were conducted in households select- ed by simple random sampling (balloting) from the chosen houses. The number of households needed for this study was calculated using the estima- tion formula for calculating the required sample size for household survey (nh) which is suitable for international use given by the Department of Economic and Social Affairs, Statistics Division, United Nations:24 Eq. 1 where nh is the parameter to be calculated and is the sample size in terms of number of households to be selected; z is the statistic that defines the level of con- fidence desired; r is an estimate of a key indicator to be measured by the survey; f is the sample design effect, Deff, assumed to be 2.0 (default value); k is a multiplier to account for the anticipat- ed rate of non-response; p is the proportion of the total population accounted for by the target population and upon which the parameter, r, is based; n is the average household size (number of persons per household); e is the margin of error to be allowed. Data collection Data was collected by trained data col- lectors over a period of 5 weeks in July and August 2017 using an extensive question- naire adapted from previous studies.7,12,25,26 The questionnaire investigated household sociodemographic profile, household income, total consumption expenditure, healthcare expenditure and household assets. The study considered: Household income included all earn- ings, welfare package or financial benefits accruing to the household from all members of the household not just the income of household head. Household total consumption expendi- ture to include spending on health, food and non-food items like rents, transportation, school fees, cable television and mobile phone subscription bills, fuel for generator, clothing, religious contributions and expenses at social events. Household healthcare expenditure includes expenditures on drug and medi- cines, consultation fees, hospital bed charges, transport charges to the treatment facilities and daily living cost, including food and lodging for the purpose of caring for the ailing household member. It also included expenditure made on self-medica- tion for minor illnesses and other services sought from alternative/traditional medical practitioners (e.g. TBAs, TBSs and Spiritual healing homes). The study explicitly explored healthcare payments during episodes of chronic ill- nesses, hospitalizations, childbirths in the preceding 12-month period (July 2016 – June 2017) and minor illnesses over 4-week period. Chronic illness was defined as a condi- tion that is long-lasting (e.g., more than 6 weeks), in many cases lifelong, which needs to be managed on a long-term basis.27 Minor illnesses were considered as non- severe health conditions of less than 6-week duration for which affected household members were treated on outpatient basis. Hospitalization care were similarly con- sidered as in-patient care received by household members either in the formal health sector or with the alternative/tradi- tional health practitioner.26 The study instrument was pre-tested among 30 households in Yenegwe, a small community on the outskirt of Yenagoa. The results obtained were used to improve the different aspects of the questionnaire. The pattern of some of the questions, the arrangement of the sections and the coding of some responses were revised after the pre-test. Data analysis Data generated from the field was directly entered into IBM SPSS 22.0 ver- sion which was also used for the analysis. Analyses were done to uncover the demo- graphics, earnings and expenditures of the households. Data was presented as frequen- cy distribution tables and descriptive statis- tics like means, standard deviation and range were calculated. A principal compo- nent analysis was done to group household into different socioeconomic groups. We estimated household impoverish- ment by calculating poverty estimates using international poverty and extreme poverty lines of N1,095 (US$3) and N730 (US$2) per person per day respectively before and after households made healthcare pay- ments.14 The calculated estimates are the poverty headcount, poverty gap and the nor- malized poverty gap before and after house- holds made health expenditures. These poverty estimates were operationalised as follows: The poverty headcount represented the percentage of households living below the defined poverty lines.2,5,12,15 The poverty gap represented the mean deficit from the poverty line among the study population. It is the average amount by which resources fall short of the defined poverty line.2,5,15 The normalized poverty gap was com- puted by dividing the estimated poverty gap by the defined poverty line. This is useful for international comparisons.2,5,15 All financial estimates were made in Naira which is the Nigerian currency (con- version: US$1 = 365 Naira). Article [page 28] [Healthcare in Low-resource Settings 2018; 6:7464] No n- co mm er cia l u se on ly The difference between the poverty estimates before and after healthcare expen- diture represented the impoverishing effect of health payment.28 Thus the differences in poverty headcounts, poverty gaps and nor- malized poverty gaps before and after health payment represents the impoverish- ment attributable to health spending. Ethics and Permission Ethical approval was obtained from the University of Port Harcourt research ethics committee. The standard protocol for com- munity entrance was applied and data was obtained from respondents only after the study objectives were explained and a writ- ten consent obtained from them. Results Sociodemographic profile of Household Responses were received from 525 Households of which majority were headed by males (77%), had married/cohabiting partners (70.7%) and over 90% of house- hold heads had post-primary education (Table 1). In total, two thousand five and twenty- eight (2,528) persons were studied in the five hundred and twenty-five (525) house- holds with a median household size of 5. About 17% of households made health expenditures related to childbirth while oth- ers did same when they sought care for long-term health condition (16.2%) and hospitalization (13%) (Table 2). Household income, total consump- tion expenditure and health expen- diture Table 3 shows that the household mean monthly income from all sources is N160, 785 with a high level of variability (SD N148, 871). The mean total expenditure on consumption in the study was almost N150, 000 (SD N128, 087) while healthcare gulped on the average N19,520 monthly from the households’ income, this corre- sponds to a mean percentage of 15.9% of household income spent on health. Impoverishing effect of healthcare payment The mean household income per capita per day was estimated as N1, 220 (SD = 1,073; SE = 47). This estimate reduced to N1, 038 after health spending was discount- ed from household total income. The proportion of households (poverty headcount) whose members live on less than the poverty line of N1,095 (US$3) per day was 58.7% and 67.9% before and after discounting household income by house- hold health expenditure respectively, increasing the prevalence of poverty by 9.2% (Table 4). The results also show that 9% of poor households were further pushed deeper into extreme poverty by health pay- ment. Other impoverishing impact attribut- able to health spending are presented in Table 4. Discussion The study demonstrated the impover- ishing effect of healthcare expenditure on households in Yenagoa. It showed that a substantial proportion of households are liv- ing on the margin of poverty. Almost 10% of households who were hitherto non-poor were pushed below the poverty line and another 9% who were poor were further pushed deeper into extreme poverty by healthcare expenditure. A 12.3% increase in the poverty gap and 16% increase in the extreme poverty gap were attributable to health payment. The average per capita deficit of N480 suffered by households without healthcare spending increased to approximately N540 after accounting for health spending. Studies done in India, Kenya and Brazil, reported 3.3%, 2.7% and 2.6% increase in poverty headcount respectively Article Table 1. Sociodemographic of household heads. Characteristics Frequency (n =525) Percentage (%) Sex Male 404 77.0 Female 121 23.0 Age of household heads (in years) 18 - 24 9 1.7 25 - 34 98 18.7 35 - 44 197 37.5 45 - 54 120 22.9 55 - 64 48 9.1 65 and above 53 10.1 Marital status Single 77 14.6 Married 371 70.7 Divorced/Separated 51 9.7 Widowed 26 5.0 Educational status No formal education 9 1.7 Primary education 35 6.7 Secondary education 158 30.1 Post-secondary education 323 61.5 Occupation Unemployed 14 2.7 Student/Apprentice 18 3.4 Farming/Fishing 27 5.1 Company worker/Artisan 97 18.5 Civil servants 181 34.5 Business owner/ contractor 127 24.2 Professionals 32 6.1 Pensioner 29 5.5 Socioeconomic status (n = 475) Q1 (Poor) 150 31.6 Q2 (Middle) 273 57.5 Q3 (Wealthy) 40 8.4 Q4 (Wealthiest) 12 2.5 Household Assets (Ownership) Radio 300 57.3 Television 508 96.9 Fridge 449 85.7 Car 173 33.0 Phone 519 99.0 House 213 40.7 Stocks/Equities 40 7.6 [Healthcare in Low-resource Settings 2018; 6:7464] [page 29] No n- co mm er cia l u se on ly after health expenditures.15,29,30 The analysis of household income and expenditure from 11 countries in Asia showed an increase of 3.8% and 3.6% in extreme poverty and poverty headcounts respectively after household health spending deductions in Bangladesh which had the most significant proportional variation in the study.4 Our finding shows an increase of approximately 9% in both extreme poverty and poverty headcounts which is higher than these quot- ed percentages from India, Kenya, Brazil and Bangladesh. However, a direct compar- ison of the estimates from different studies and countries can be misleading because the different methods that might have been employed in constituting health expenditure and the cutoff to define poverty vary in time and place. Nonetheless, all these studies showed that healthcare spending especially through OOP mechanism have an impover- ishing effect on households. The high proportion of households impoverished in Yenagoa, Bayelsa state due to healthcare spending provides additional support for the recently established Bayelsa health insurance scheme (BHIS). It is expected that stakeholders would galvanize efforts towards the success of this mandato- ry social health insurance scheme in the state. However, there are other important considerations which the operators would need to consider. Notable among these are the provisions for funding premiums for the poor, near poor and other vulnerable groups as contributory mechanisms alone will not ensure universal health coverage in situa- tions where the population is largely poor and/or in the informal sector.14,31 It is pertinent to note that from the multi-country study involving 11 Asian countries, Indonesia had the lowest inci- dence of impoverished households attribut- able to healthcare payments.4 This arose from the country’s ability to protect poor households from high healthcare cost through targeted exemptions with the use of a health card.2 Even in developed setting like the UK where hospital services are free at the point of access to all, similar exemp- tions from co-payment exist for prescribed drugs, dental treatment and eyesight exami- nation for vulnerable population including those with long-term conditions.32 Indeed, there are further lessons to learn from the scenario in the UK32 as 16% and 13% of households in Yenagoa had at least a member living with at least one long-term condition or hospitalized in the last one year respectively. The enormous financial bur- den associated with these events can be Article Table 4. Impoverishment estimates before and after health expenditure. Before discounting After discounting Difference (Absolute) Difference health payment (1) health payment (2) (3)= (2) – (1) (Relative)[(3)/(1)*100] Assessment using the $3.00 (1,095 Naira) capita/day poverty line Poverty Headcount (%) 58.7 67.9 9.2 15.7 Poverty gap (Naira) 482.4 541.6 59.2 12.3 Normalized Poverty Gap (%) 44.1 49.5 5.4 12.2 Assessment using the $2.00 (730 Naira) capita/day extreme poverty line Poverty Headcount (%) 37.8 46.8 9.0 23.8 Poverty Gap (Naira) 280.2 324.9 44.7 16.0 Normalized Poverty Gap (%) 38.4 44.5 6.1 15.9 Table 2. Morbidity pattern of households. Characteristics Frequency Percent (%) Morbidity pattern in households (n = 525) HHs with members having long term health condition 85 16.2 HHs with members hospitalized 68 13.0 HHs with members that had minor illness 265 50.5 HHs with childbirth 87 16.6 HHs with nonspecific medical conditions 169 32.2 HHs without health expenditure in last 1 year 115 22.0 Table 3. Household income, total consumption expenditure and healthcare expenditure. Variable Mean Value (in Naira) Standard Deviation (in Naira) Range HH mean monthly income Primary income 150,970 140,079 (10,000 – 750,000) Collective income (all sources) 160,785 148,871 (10,000 – 1,010,000) HH mean monthly expenditure Total consumption expenditure 149,597 128,087 (12,000 – 771,925) Food expenditure 60,900 32,625 (7,000 – 195,300) Non-food expenditure 73,729 80,391 (3,450 – 550,000) Total healthcare expenditure 19,510 44,899 (0 – 683,330) Breakdown of HH mean health care expenditure Long-term medical condition 4,515 15, 475 (0 – 200,000) Minor illness 8,940 3,709 (0 – 35,000) Childbirth 3,150 6,980 (0 – 46,000) Hospitalization 6,770 28,630 (0 – 333,330) Non-specific health payments 3,065 7,278 (0 – 86,000) [page 30] [Healthcare in Low-resource Settings 2018; 6:7464] No n- co mm er cia l u se on ly [Healthcare in Low-resource Settings 2018; 6:7464] [page 31] ameliorated with expanded funding options for public health services that can guarantee improved access to hospital care for all and exemption of poorer households from all co-payments as part of a broader social security system. Although the relative increase of 15.7% and 12.3% in the headcount and depth of poverty respectively are worrisome, this may still represent an underestimation of the impoverishing effect of healthcare cost on households in Yenagoa as indirect costs and lost earnings by households with sick members were not accounted for in this study. Like the Kenyan study,18 underesti- mation could also arise from the 22% of households that reported zero spending on healthcare in the one-year recall period. The zero spending may reflect non-recall or denial of past illness episodes which are often given negative connotations or because they had completely forgone care due to lack of resources, not necessarily because they do not need healthcare. Interestingly, while impoverishment as a result of health expenditures occurs in all countries irrespective of income levels, its prevalence is higher in countries which depend predominantly on OOP payment mechanisms like Nigeria.6,33,34 A quick rec- ommendation would be urging all leaders in Africa to increase public spending on health to at least achieve the target of 15% endorsed at the Abuja declaration.35 This would appear difficult in view of other for- midable challenges, dwindling public rev- enue and lack of political will in these coun- tries. However, widening the sources of funding may just well be the way out. In this regard, a range of innovative prepay- ment methods including the National Health Insurance Scheme (NHIS) should be introduced, strengthened and expanded to achieve national coverage. Although this community-based study quantified and deduced the impoverishing effect of healthcare expenditure, its limita- tions arise from the fact that only the total OOP expenditures was reported without categorizing them into healthcare expendi- ture subheads like fees for drug, consulta- tion, investigation, transportation, accom- modation as was done in similar studies.33-34 This categorization would have helped identified the specific spending that influ- enced household impoverishment the most in our setting. Furthermore, applying a longitudinal approach is more apt and reliable in study- ing expenditures and their impact on house- hold impoverishment. The paucity of longi- tudinal data on the subject may not be unre- lated to the difficulties in implementing such research protocols, hence researchers often resort to cross sectional designs.2,11,13,15,18,33-34,36,37 Despite these limi- tations, findings from this study would be a useful guide in the on-going implementa- tion of the state-wide health financing model that would minimize systematic dis- parities while ensuring the achievement of universal health coverage for the popula- tion. Conclusions A significant percentage of households who are marginally non-poor were pushed into poverty because of healthcare expendi- ture. There is need for increased public spending on healthcare, implementation of innovative and progressive pre-payment mechanisms as well as exemption from payment by vulnerable households that would assure financial risk protection, guar- antee equity in health financing and univer- sal coverage for households in Yenagoa, Bayelsa state. References 1. World Health Organization. The world health report 2000: health systems: improving performance. World Health Organization; 2000. 2. Kimani DN, Mugo MG, Kioko UM. Catastrophic health expenditures and impoverishment in Kenya. ESJ. 2016;12:434-52. 3. Baeza C, Packard T. Beyond survival: protecting households from health shocks in Latin America. Stanford University Press; 2006. 4. Van Doorslaer E, O'Donnell O, Rannan- Eliya RP, et al. Effect of payments for health care on poverty estimates in 11 countries in Asia: an analysis of house- hold survey data. Lancet 2006;368:1357-64. 5. Wagstaff A, Doorslaer EV. Catastrophe and impoverishment in paying for health care: with applications to Vietnam 1993–1998. Health Econ 2003;12:921-33. 6. World Health Organization. World health statistics 2015. World Health Organization; 2015. 7. Onwujekwe OE, Uzochukwu BS, Obikeze EN, et al. Investigating deter- minants of out-of-pocket spending and strategies for coping with payments for healthcare in southeast Nigeria. BMC Health Serv Res 2010;10:67. 8. Xu K, Evans DB, Carrin G, et al. Protecting households from catastroph- ic health spending. Health Affairs 2007;26:972-83. 9. World Bank Group. World Development Report-A Better Investment Climate for Everyone; 2005. Available from: siteresources.worldbank.org/INTWDR 2005/Resources/complete_report.pdf 10. Su TT, Kouyaté B, Flessa S. Catastrophic household expenditure for health care in a low-income society: a study from Nouna District, Burkina Faso. Bull World Health Org 2006;84:21-7. 11. Rashad AS, Sharaf MF. Catastrophic economic consequences of healthcare payments: Effects on poverty estimates in Egypt, Jordan, and Palestine. Economies 2015;3:216-34. 12. Xu K, Evans DB, Kawabata K, et al. Household catastrophic health expendi- ture: a multicountry analysis. Lancet 2003;362:111-7. 13. Mchenga M, Chirwa GC, Chiwaula LS. Impoverishing effects of catastrophic health expenditures in Malawi. Int J Equity Health 2017;16:25. 14. Hoang VM, Oh J, Tran TA, et al. Patterns of health expenditures and financial protections in Vietnam 1992- 2012. J Korean Med Sci 2015;30:S134- 8. 15. Boing AC, Bertoldi AD, Posenato LG, Peres KG. The influence of health expenditures on household impoverish- ment in Brazil. Rev Saude Publ 2014;48:797-807. 16. Elgazzar H, Raad F, Arfa C, et al. Who pays? Out-of-pocket health spending and equity implications in the Middle East and North Africa. 2010 Health, Nutrition and Population (HNP) Discussion Paper. Washington DC: The World Bank; 2010. 17. Kusi A, Hansen KS, Asante FA, Enemark U. Does the National Health Insurance Scheme provide financial protection to households in Ghana? BMC Health Serv Res 2015;15:331. 18. Buigut S, Ettarh R, Amendah DD. Catastrophic health expenditure and its determinants in Kenya slum communi- ties. Int J Equity Health 2015;14:46. 19. Wagstaff A. Measuring financial protec- tion in health. World Bank, Washington, DC; development research group. Policy research working paper# WPS 2008;4554. 20. Xu K, World Health Organization. Distribution of health payments and cat- astrophic expenditures methodology. Discussion paper No. 2. HSF, World Health Organization; 2005. 21. Koinyan AA, Nwankwoala HO, Eludoyin OS. Water resources utiliza- Article No n- co mm er cia l u se on ly [page 32] [Healthcare in Low-resource Settings 2018; 6:7464] tion in Yenagoa, Central Niger Delta: Environmental and health implications. Int J Water Res Environ Engin 2013;5:177-86. 22. Iyorakpo J. Impact of Rapid Urbanization on Environmental Quality in Yenagoa Metropolis, Bayelsa State- Nigeria. ESJ 2015;11:255-68. 23. National Population Commission Abuja, Nigeria. 2006 Housing popula- tion Census: Population on distribution by age and sex: State and Local Government Area, Priority Table. National Population Commission 2010;4:54-8. 24. Department of Economic and Social Affairs, Statistics Division, United Nations. Designing Household Survey Samples: Practical Guidelines. Studies in Methods Series F No.98 New York, 2005. pp 44-46. 25. Musoke D, Boynton P, Butler C, Musoke MB. Health seeking behaviour and challenges in utilising health facili- ties in Wakiso district, Uganda. Afr Health Sci 2014;14:1046-55. 26. Mondal S, Kanjilal B, Peters DH, Lucas H. Catastrophic out-of-pocket payment for health care and its impact on house- holds: Experience from West Bengal, India. Future Health Syst Innov Equity 2010. Available from: http://www.chronicpoverty.org/uploads /publication_files/mondal_et_al_health. pdf 27. Choi JW, Choi JW, Kim JH, et al. Association between chronic disease and catastrophic health expenditure in Korea. BMC Health Serv Res 2015;15:26. 28. Berki SE. A look at catastrophic med- ical expenses and the poor. Health Affairs 1986;5:138-45. 29. Garg CC, Karan AK. Reducing out-of- pocket expenditures to reduce poverty: a disaggregated analysis at rural-urban and state level in India. Health Policy Plann 2008;24:116-28. 30. Chuma J, Maina T. Catastrophic health care spending and impoverishment in Kenya. BMC Health Serv Res 2012;12:413. 31. Somanathan A, Tandon A, Dao HL, et al. Moving toward universal coverage of social health insurance in Vietnam: assessment and options. World Bank; 2014. 32. World Health Organization. World Health Report, 2010: health systems financing the path to universal cover- age. World Health Organization; 2010. 33. Ogaji DS, Nwi-ue LB, Agalah HN, et al. Impact and contributors to cost of managing long term conditions in a uni- versity hospital in Nigeria. J Commun Med Primary Health Care 2015;27:30- 40. 34. Ogaji DS, Mark OC, Oghenetega EP, et al. Cost burden for Accessing Paediatric Emergency Services at a Tertiary Health Facility. Niger Health J 2015;15:103- 10. 35. Birch S. Health care charges: lessons from the UK. Health Policy 1989;13:145-57. 36. Barasa EW, Maina T, Ravishankar N. Assessing the impoverishing effects, and factors associated with the inci- dence of catastrophic health care pay- ments in Kenya. Int J Equity Health 2017;16:31. 37. Levie A, Xu K. Coping with out-of- pocket health payments: empirical evi- dence from 15 African countries. Bull World Health Org 2008;86:849-56C. Article No n- co mm er cia l u se on ly