UBC Faculty Research and Publications

Determinants of health insurance ownership among women in Kenya: evidence from the 2008–09 Kenya demographic… Kimani, James K; Ettarh, Remare; Warren, Charlotte; Bellows, Ben Mar 31, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12939_2013_Article_443.pdf [ 210.14kB ]
JSON: 52383-1.0221341.json
JSON-LD: 52383-1.0221341-ld.json
RDF/XML (Pretty): 52383-1.0221341-rdf.xml
RDF/JSON: 52383-1.0221341-rdf.json
Turtle: 52383-1.0221341-turtle.txt
N-Triples: 52383-1.0221341-rdf-ntriples.txt
Original Record: 52383-1.0221341-source.json
Full Text

Full Text

RESEARCH Open AccessDeterminants of health insurance ownershipamong women in Kenya: evidence from the2008–09 Kenya demographic and health surveyJames K Kimani1*, Remare Ettarh2, Charlotte Warren3 and Ben Bellows1AbstractBackground: The Government of Kenya is making plans to implement a social health insurance program bytransforming the National Hospital Insurance Fund (NHIF) into a universal health coverage program. The objectiveof this study was to examine the determinants associated with health insurance ownership among women inKenya.Methods: Data came from the 2008–09 Kenya Demographic and Health Survey, a nationally representative survey.The sample comprised 8,435 women aged 15–49 years. Descriptive statistics and multivariable logistic regressionanalysis were used to describe the characteristics of the sample and to identify factors associated with healthinsurance ownership.Results: Being employed in the formal sector, being married, exposure to the mass media, having secondary educationor higher, residing in households in the middle or rich wealth index categories and residing in a female-headedhousehold were associated with having health insurance. However, region of residence was associated with alower likelihood of having insurance coverage. Women residing in Central (OR = 0.4; p < 0.01) and North Eastern(OR = 0.1; p < 0.5) provinces were less likely to be insured compared to their counterparts in Nairobi province.Conclusions: As the Kenyan government transforms the NHIF into a universal health program, it is important toimplement a program that will increase equity and access to health care services among the poor and vulnerablegroups.Keywords: Social health insurance, National Hospital Insurance Fund, Women, KenyaBackgroundSocial health protection systems are mechanisms thatcountries use to address the challenges related to provid-ing access to health care services to their citizens, espe-cially the poor segments of the population. The benefitsof extending social protection in health include reducingfinancial barriers associated with access to health careservices and protection from financial catastrophe and im-poverishment related to health care expenditures [1-5].One of the categories of social health protection systemsis the social health insurance, which is a financing schemewhere monies are pooled into a common fund and usedfor paying for healthcare costs of members. Contributionsare usually collected from workers, self-employed indivi-duals, businesses and in some cases the government, par-ticularly where a universal coverage model is adopted [2,5].Generally, but not always, contributions have ensured thatthe rich contribute more than the poor but contributionsdo not typically vary with health status [6].In Kenya, a universal social health insurance schemehas not been implemented; however, in November 2004,the government introduced the National Social HealthInsurance Fund (NSHIF) Bill in parliament. The Bill waspassed by parliament in December 2004 [7], but thePresident declined to assent to the Bill and sent it back toparliament due to a number of concerns. One of the con-cerns was that the Bill was deemed too expensive to imple-ment and financially unsustainable [8]. As the government* Correspondence: jkimani@popcouncil.org1Population Council, General Accident Insurance House, Ralph Bunche Road,P.O. Box 17643–00500, Nairobi, KenyaFull list of author information is available at the end of the article© 2014 Kimani et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Kimani et al. International Journal for Equity in Health 2014, 13:27http://www.equityhealthj.com/content/13/1/27prepares to re-introduce the NSHIF legislation in parlia-ment, it is important to have a better understanding of fac-tors associated with participation in the current NationalHospital Insurance Fund (NHIF), particularly among thepoor, as well as a determination of the proportion of in-dividuals without access to health insurance among thisdemographic group. The NSHIF will build on the exist-ing NHIF framework and, therefore, such evidence isimperative in order to implement an effective NSHIF. Itis hoped that the proposed NSHIF will have mecha-nisms that will increase equity and access to health careservices by all population groups.Literature focusing on the determinants of participa-tion in health insurance schemes in Kenya and Africa ingeneral is limited. Studies conducted in a number ofsub-Saharan African countries showed that employmentin the formal sector was significantly associated with ac-cess to health insurance relative to being employed inthe informal sector [2,3,9]. The low participation of indi-viduals in the informal sector was attributed to a numberof factors, including low and non-regular incomes, inse-cure employment, and insurance scheme design features(e.g., inflexible payment schedules and lack of awarenessabout insurance schemes) that are not adapted to people’sneeds and preferences. In Kenya, it is estimated that 31.6%and 26.3% of the total workforce are engaged in the infor-mal and formal sectors, respectively, while 42.1% are en-gaged in small-scale farming and pastoralist activities [10].Other factors that have been cited by previous research aspredictors of health insurance ownership include income,education, household wealth status, marital status, age,place of residence [11-20].In Kenya, more than four out of 10 (46.6%) individualslive below the poverty line [21]. Data from the nationalhealth accounts show that more than a third of the poorwho were ill did not seek care compared to only 15% ofthe rich [22]. Additionally, according to the 2005/06 na-tional health accounts, 36% of funds to the health sectorcame from households and out of these, the out-of-pocketexpenditure accounted for more than 29% [23]. Thesefindings raise concern about equity and financial accessi-bility to health care by a majority of people in Kenya, par-ticularly the poor who are highly vulnerable to economicshocks that result from catastrophic out-of-pocket healthexpenditure. Existing studies show that the poor are morelikely to get sick, less likely to use preventive and curativehealth care, and consequently, have higher mortality rates.According to these studies, one of the factors responsiblefor these challenges is high out-of-pocket payments forhealth care [24-26]. The 2010 World Health Report andthe 2010 Millennium Development Goals report under-score the importance of reducing disparities in access tohealth care, particularly among the poor and margina-lized groups through universal health coverage [27,28].Extending access to health care to all segments of thepopulation, including the poor is an important objectiveof the Kenyan government’s national health sector stra-tegic plan and national development agenda as outlinedin the Kenya Vision 2030 policy framework [29-31].Besides the NHIF, in Kenya, individuals can accesshealth insurance through private insurance firms and tosome extent community-based health insurance (CBHI)organizations. Due to cost considerations, private healthinsurance is predominantly accessible to the middle andhigher-income groups [9]. Community-based health in-surance is relatively new in Kenya having been estab-lished in 1999, and, as a result it has limited coverage [3]According to the Kenya Community-Based Health Finan-cing Association (KCBHFA), currently, there are nine insti-tutions offering community health financing schemes with410,997 beneficiaries or about 1% of the population co-vered [32]. In Africa, countries such as Burkina Faso,Senegal, Tanzania and Ghana have well developed CBHIschemes that are recognized by the national governmentsas a key component in the national health financing strat-egy [33-39]. Findings from these studies suggest that CBHIschemes have the ability to reach marginalized populationgroups such as the poor, women and children, however,more support and strategies from governments are neededto enhance their development and sustainability. Existingevidence shows that in sub-Saharan Africa there arevarious types of approaches that have been used to en-sure expansion of health insurance coverage to thepopulation. A review conducted by the World Bank on theimpact of universal coverage schemes in developing coun-tries showed that Rwanda and Nigeria are examples ofcountries with more than one insurance scheme targetingdifferent population segments with aim of working towarduniversal health coverage [40]. In Namibia and South Africathey have voluntary insurance mechanisms that include pri-vate health insurance [41].The NSHIF legislation seeks to transform the currentNational Hospital Insurance Fund (NHIF) into a univer-sal health coverage program, which will ensure equityand access to healthcare services by all citizens. One ofthe criticisms of the NHIF is its failure to reach out tothe majority of Kenyans, especially the poor and those inthe informal sector [2,3,9,42]. While the NHIF has acomponent for people in the informal sector, however,some of the design features of the program act as criticalbarriers. For example, the NHIF imposes a penalty thatis five times the contribution amount for those who donot make their payments by the due date. This regula-tion particularly hurts the poor, the unemployed andcasual workers in the informal sector, who do not have asteady income that would enable them, pay their contri-butions regularly. As the government of Kenya makesplans to transform the National Hospital Insurance FundKimani et al. International Journal for Equity in Health 2014, 13:27 Page 2 of 8http://www.equityhealthj.com/content/13/1/27(NHIF) into a universal health coverage program, it isimperative to examine what factors are associated withhealth insurance ownership in Kenya, particularly amongvulnerable sub-groups in the population. The aim of thisstudy was to examine the determinants associated withhealth insurance ownership among women in Kenya.MethodsStudy design and samplingData came from the 2008–09 Kenya Demographic andHealth Survey (KDHS), a nationally representative sur-vey [43]. The sampling frame included a total of 400primary sampling units across the eight provinces. Multi-stage cluster sampling was used to select 8,444 womenaged 15 to 49 years across the eight provinces of Kenyawith stratification for rural and urban residence.MeasuresThe outcome variable was whether a woman was cov-ered by any health insurance (Yes or No). The explana-tory variables examined in the study were selected basedon factors cited in the literature as influencing health in-surance ownership and included respondents occupationgroups (categorized into three employment categories –formal, informal and not working); marital status, catego-rized into never married, married and formerly married;exposure to the mass media (grouped into frequency ofreading newspaper, listening to radio and watching televi-sion), education level, grouped into no formal education,primary education, secondary education or higher; age ofwoman in years, grouped into 15–19, 20–24, 25–29, 30–34, 35–39, 40–44 and 45–49; gender of household head(male or female); number of household members, groupedinto 1–4 members and 5 or more members (the averagenumber of household members was 5 and so the variablewas categorized as below 5 or 5 and above); householdwealth status categorized into poorest/poorer, middle andricher/richest; place of residence (urban or rural); and geo-graphical province of residence (Central, Coast, Eastern,North Eastern, Nairobi, Nyanza, Rift Valley and Western).Data analysisFor this paper, a total of 8,435 women with completedata on the key outcome variable were included in theanalyses. Descriptive statistics and multivariate logisticregression analysis were used to describe the characteris-tics of the sample and to identify factors associated withhealth insurance ownership. For the bivariate analysis,Pearson’s chi-square test (X2) was used to test the associ-ation between health insurance ownership and the explana-tory variables. Data analysis was performed using STATA®version 10 and statistical adjustments were made to get ro-bust standard errors since the sampling of respondents inthe KDHS involved stratification and clustering [44,45].Ethical considerationsThe study involved secondary analysis of data from theKDHS which excluded participant identifiers. The surveyprotocol was approved by the Scientific and Ethical ReviewCommittee of Kenya Medical Research Institute (KEMRI).ResultsDescriptive analysisTable 1 presents the results from the descriptive analysis.Only 7% of the women had health insurance and amongthese, a higher proportion were covered by employer-based health insurance (4%), while less than 1% werecovered by community-based health insurance schemes(results not shown). Many of the women were un-employed while 30% and 25% were employed in the in-formal and formal sectors, respectively. The majority ofthe women were married, listened to radio, had primarylevel of education, lived in male-headed households andresided in rural areas.The results of the bivariate analysis of the associationbetween health insurance ownership and explanatory vari-ables are shown in Table 2. A significantly higher propor-tion of women with health insurance were employed inthe formal sector (17%) while 4% were employed in the in-formal sector and a similar proportion were unemployed(p < 0.001). Having health insurance was significantlyassociated with being married (8%), listening to radioalmost every day (58%), reading newspaper almost everyday (35%) and watching television almost every day (19%),having secondary school education and higher (18%),belonging to wealthier households (14%) and residing inurban areas (15%).Multivariate analysisThe results of the multivariate logistic regression ana-lysis for determinants of health insurance coverage areshown in Table 3. Being employed in the formal sectorwas significantly associated with a higher probability ofhaving health insurance compared to being unemployed(OR = 2.2; p < 0.001). Married women were significantlyassociated with having health insurance compared tonever married women (OR = 1.8; p < 0.05). Exposure tothe mass media was significantly associated with health in-surance ownership. Specifically, women who read newspa-pers, listened to radio or watched television sometimes oralmost every day had a higher probability of having healthinsurance compared to those who never did. Educationwas a significant predictor of having insurance coverage.Women who had attained primary level of education(OR = 4.4; p < 0.01) and secondary education or higher(OR = 10.9; p < 0.001) were associated with a higher like-lihood of having health insurance compared to thosewith no formal education. Generally, controlling for allother variables, the probability of having health insuranceKimani et al. International Journal for Equity in Health 2014, 13:27 Page 3 of 8http://www.equityhealthj.com/content/13/1/27tended to increase with age although non-significant re-sults were observed for age categories 20–24 years and40–44 years. Other significant determinants of havinghealth insurance were the gender of household head andhousehold wealth status. Women living in female-headedhouseholds were significantly more likely to be insured(OR = 1.7; p < 0.01) compared to their counterparts inmale-headed households. The probability of having healthinsurance increased as the level of household wealth indexincreased. Women from wealthier households were sixtimes more likely to have health insurance coverage com-pared to those from poor households. Women residing inthe geographic provinces of Central and North Easternhad a significantly lower likelihood of having health insur-ance compared to their counterparts in Nairobi province.DiscussionThe objective of this paper was to examine the determi-nants of health insurance ownership among women inKenya. The findings showed that a high proportion ofwomen (93%) have no access to any type of health insur-ance. Our findings also showed that more women in theformal sector than informal sector had been insured.After controlling for all other variables, being employedin the formal sector was still associated with havinghealth insurance. This finding corroborates evidencefrom previous studies, which demonstrated that employ-ment in the formal sector is an important determinantof being insured [2,3,9]. The differences in insurancecoverage between the formal and informal sectors haveTable 1 Health insurance ownership and socio-demographiccharacteristics of study populationVariable N %Covered by health insuranceNo 7,831 92.8Yes 604 7.2Employment sectorFormal employment 2,116 25.1Informal employment 2,568 30.5Not working 3,739 44.4Marital statusNever married 2,540 30.1Married 5,041 59.7Formerly married 863 10.2Exposure to mediaFrequency of reading newspaperNot all 4,921 58.3Sometimes 2,949 35.0Almost everyday 566 6.7Frequency of listening to radioNot all 1,551 18.4Sometimes 2,006 23.8Almost everyday 4,881 57.8Frequency of watching televisionNot all 4,602 54.5Sometimes 1,636 19.4Almost everyday 2,204 26.1EducationNo education 1,242 14.7Primary 4,404 52.2Secondary or higher 2,798 33.1Age (Years)15-19 1,767 20.920-24 1,744 20.725-29 1,423 16.930-34 1,180 14.035-39 930 11.040-44 730 8.745-49 670 7.9Gender of household headMale 5,352 63.4Female 3,092 36.6Number of household members1-4 members 3,363 39.85+ members 5,081 60.2Table 1 Health insurance ownership and socio-demographiccharacteristics of study population (Continued)Household wealth statusPoorest/poorer 2,983 35.3Middle 1,455 17.2Richer/richest 4,006 47.4Place of residenceUrban 2,615 31.0Rural 5,829 69.0ProvinceCentral 973 11.5Coast 1,149 13.6Eastern 1,127 13.4North Eastern 608 7.2Nairobi 952 11.3Nyanza 1,318 15.6Rift Valley 1,278 15.1Western 1,039 12.3Total 8,444 100.0Note: Percentages may not add up to 100 due to rounding off.Kimani et al. International Journal for Equity in Health 2014, 13:27 Page 4 of 8http://www.equityhealthj.com/content/13/1/27important implications on the proposed plans to estab-lish a social health insurance program in Kenya. One ob-jective of comprehensive social health insurance is toensure that all population groups irrespective of theirsocio-economic status have access to quality and afford-able health care. Our findings suggest that more effortsare needed to increase health insurance coverage of indi-viduals in the informal sector. Previous research hasshown that unlike in the formal sector, it is difficult toassess incomes and collect income taxes from workersemployed in the informal sector [9] and, as a conse-quence, deduction of contributions for the proposed so-cial health insurance program can be a challenge. Thismeans that lack of suitable mechanisms for collectingcontributions from employees in the informal sectorcould hamper the implementation and sustainability ofthe proposed social health insurance program. However,evidence shows that many workers in the informal sec-tor participate in microfinance institutions such as sav-ings and credit cooperative organizations (SACCOs) andcommunity-based groups (e.g., merry-go-rounds) [46]and, therefore, these organized units can be importantplatforms through which contributions are collected andsubmitted to the social health insurance program.Table 2 Bivariate analysis for associations between healthinsurance ownership and explanatory variablesCovered by healthinsuranceVariable TotalnumberN (%) p-valuesEmployment sectorFormal employment 2,114 353 (16.7) ***Informal employment 2,565 102 (4.0)Not working 3,735 147 (3.9)Marital statusNever married 2,538 158 (6.2) ***Married 5,035 411 (8.2)Formerly married 862 35 (4.1)EducationNo education 1,239 6 (0.5) ***Primary 4,400 100 (2.3)Secondary or higher 2,796 498 (17.8)Exposure to mediaFrequency of readingnewspaperNot all 4,913 1.9 ***Sometimes 2,948 10.6Almost everyday 566 35.0Frequency of listeningto radioNot all 1,551 18.4 ***Sometimes 2,004 23.8Almost everyday 4,877 57.8Frequency of watchingtelevisionNot all 4,595 1.7 ***Sometimes 1,636 6.2Almost everyday 2,202 19.3Age (Years)15-19 1,765 53 (3.0) ***20-24 1,742 86 (4.9)25-29 1,422 123 (8.7)30-34 1,178 118 (10.0)35-39 930 91 (9.8)40-44 729 72 (9.9)45-49 669 61 (9.1)Gender of household headMale 5,345 393 (7.4) 0.368Female 3,090 211 (6.8)Table 2 Bivariate analysis for associations between healthinsurance ownership and explanatory variables(Continued)Number of householdmembers1-4 members 3,360 305 (9.1) ***5+ members 5,075 299 (5.9)Household wealth statusPoorest/poorer 2,977 18 (0.6) ***Middle 1,455 44 (3.0)Richer/richest 4,003 542 (13.5)Place of residenceUrban 2,611 395 (15.1) ***Rural 5,824 209 (3.6)ProvinceCentral 972 40 (4.1) ***Coast 1,149 67 (5.8)Eastern 1,127 62 (5.5)North Eastern 606 1 (0.2)Nairobi 951 230 (24.2)Nyanza 1,316 82 (6.2)Rift Valley 1,276 84 (6.6)Western 1,038 38 (3.7)Total 8,444 604 (7.2)*p < 0.05; **p < 0.01; ***p < 0.001; X2 was used to test the association betweenhealth insurance ownership and explanatory variables.Kimani et al. International Journal for Equity in Health 2014, 13:27 Page 5 of 8http://www.equityhealthj.com/content/13/1/27Our study findings also showed that a number of fac-tors are significant determinants of health insuranceownership including marital status (specifically, beingmarried), education, age, gender of household head andhousehold wealth status. However, geographical regionwas associated with a lower probability of having healthinsurance. Similar to previous research [11,12], our find-ings showed that being married was associated with hav-ing health insurance coverage compared to never beenmarried and formerly married. This suggests that havinga spouse/partner is beneficial possibly because of the fi-nancial support derived from being in a dual-incomehousehold, which translates into more opportunities foraccessing health insurance coverage. Another plausiblereason is that a spouse/partner can be insured throughthe other’s insurance coverage from the employer. Expo-sure to the media through reading newspapers, listening toradio or watching television was associated with havinghealth insurance. Education was also an important de-terminant of having insurance coverage. More educatedwomen were more likely to have health insurance relativeto women with no formal education. This finding corro-borates evidence from previous studies which demon-strated that education is an important predictor of havinghealth insurance [11,47,48]. Consistent with previousstudies [14,17,12], our findings demonstrated that thelikelihood of health insurance ownership tends to risewith increase in age. One possible explanation for thisoutcome is that financial security increases with age,which in turn increases the ability to purchase health in-surance policies. Another important predictor of healthinsurance ownership was the gender of the householdhead. Women residing in female-headed households weremore likely to be insured compared to their counterpartsin male-headed households. We could not find a plausibleexplanation for this observation and future researchneeds to investigate this outcome. Household wealthstatus was also an important determinant for healthTable 3 Adjusted odds ratios (ORs) and 95% confidenceintervals (CIs) for determinants of healthinsurance ownershipVariable Adjusted OR 95% CIEmployment sector (Ref = Not working)Formal employment 2.2*** [1.5-3.2]Informal employment 1.5 [0.8-3.1]Marital status (Ref = Never married)Married 1.8* [1.3-2.5]Formerly married 0.5 [0.2-1.1]Exposure to mediaFrequency of reading newspaper(Ref = Not at all)Sometimes 1.6* [1.0-2.6]Almost everyday 3.5*** [2.0-6.0]Frequency of listening to radio(Ref = Not at all)Sometimes 1.5 [0.9-2.5]Almost everyday 1.7* [1.0-3.1]Frequency of watching television(Ref = Not at all)Sometimes 2.3*** [1.4-3.5]Almost everyday 2.6*** [1.7-4.2]Education (Ref = No education)Primary 4.4** [1.6-12.2]Secondary or higher 10.9*** [3.6-33.1]Age (Years) (Ref = 15–19)20-24 1.6 [0.8-3.5]25-29 2.5** [1.3-4.9]30-34 2.0* [1.2-3.5]35-39 2.6** [1.4-4.8]40-44 1.6 [0.9-2.9]45-49 4.0*** [2.0-7.8]Gender of household head(Ref = Male)Female 1.7** [1.2-2.3]Number of household members(Ref = 1–4)5+ members 1.0 [1.0-1.1]Household wealth status(Ref = Poorest/poorer)Middle 3.2** [1.5-6.9]Richer/richest 6.3** [3.0-13.2]Place of residence(Ref = Urban)Rural 1.0 [0.6-1.4]Table 3 Adjusted odds ratios (ORs) and 95% confidenceintervals (CIs) for determinants of healthinsurance ownership (Continued)Province (Ref = Nairobi)Central 0.4** [0.2-0.8]Coast 0.8 [0.4-1.6]Eastern 1.0 [0.5-1.8]North Eastern 0.1* [0.0-0.7]Nyanza 1.1 [0.6-1.9]Rift Valley 1.4 [0.9-2.2]Western 0.6 [0.3-1.3]*p < 0.05; **p < 0.01; ***p < 0.001.Kimani et al. International Journal for Equity in Health 2014, 13:27 Page 6 of 8http://www.equityhealthj.com/content/13/1/27insurance ownership. The likelihood being insured in-creased as one moved up the household wealth index.This finding is consistent with previous studies whichshowed that wealthier households had a higher likeli-hood of being insured [13,19,12]. Region of residencewas also a significant predictor of health insuranceownership. Specifically, women residing in the geo-graphical regions of Central, North Eastern and Westernhad a lower likelihood of having health insurancecompared to Nairobi province. The geographical dif-ferential in health insurance coverage could be ex-plained by the fact that Nairobi, which is the capitalcity of Kenya, is entirely urban and has a higher propor-tion of the population in the highest wealth quintile andhigher literacy levels compared with other geographicalregions [43].The findings from our study have important policy im-plications. First, the large proportion of women withouthealth insurance and the lower likelihood of poor house-holds to have insurance coverage highlight the need by thegovernment to hasten the move towards social health pro-tection by implementing a National Social Health InsuranceFund. This is to guarantee access to quality healthcareservices for the poor and vulnerable segments of thepopulation, as well offer protection against catastrophicout-of-pocket health expenditure associated with highmedical costs. To ensure that the vulnerable and poorhave access to health care under the NSHIF, the govern-ment will need to institute targeted subsidies and exemp-tions aimed at increasing health insurance coverage,particularly for women resident outside Nairobi. Second,our study shows that people employed in the informal sec-tor are less likely to have health insurance. Consideringthat the informal sector accounts for the highest propor-tion of Kenya’s total workforce [10], reaching out to thissector is critical for the successful implementation of thesocial health insurance scheme.LimitationsOne limitation of our study is that we were unable to as-sess the association between health status and havinghealth insurance coverage due to the lack of data on re-spondents health status (for example, presence of ill-nesses, frequency of illnesses). Previous studies haveshown that health status is an important predictor ofhealth insurance coverage [20,36,39]. Also, no data werecollected on out-of-pocket payments and health careutilization; therefore, it was not possible to examine theeffect of having health insurance on these two outcomes.Another limitation is that the questionnaire did not col-lect data on the extent of insurance coverage such astype of services covered and, therefore, we were not ableto assess the association between the extent of insurancecoverage and health insurance ownership.ConclusionsAddressing disparities in access to care among the poorand marginalized demographic groups is a key agenda inthe global health debate because it’s a critical factor inaccelerating the achievement of the Millennium Develop-ment Goals (MDGs). Our study has highlighted importantissues that will inform the efforts aimed at establishing a so-cial health insurance program by transforming the NationalHospital Insurance Fund (NHIF) into a universal healthcoverage program. The large proportion of women withouthealth insurance and the lower likelihood of poor house-holds to have insurance coverage underscore the need for asocial health insurance program to ensure equitable accessto health care. Also, there is need to design and implementtargeted initiatives that will increase health insurancecoverage among people working in the informal sector. Asthe Government of Kenya moves toward transforming theNHIF into a universal health program, it is important toimplement a program that will increase equity and accessto health care services among the poor and vulnerablegroups.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsJK conceptualized the study, conducted the data analyses, participated inthe literature review, and prepared the first draft of the manuscript. RE madesubstantive contribution that informed the data analyses and reviewed themanuscript. CW and BB were involved in revising the manuscript forintellectual content and interpretation of data. All authors are aware that themanuscript is being submitted to the journal. All authors read and approvedthe final manuscript.Author details1Population Council, General Accident Insurance House, Ralph Bunche Road,P.O. Box 17643–00500, Nairobi, Kenya. 2Faculty of Medicine, University ofBritish Columbia, 2775 Laurel Street, Vancouver, British Columbia V5Z 1 M9,Canada. 3Population Council, 4301 Connecticut Avenue NW, Suite 280,Washington, DC 20008, USA.Received: 10 September 2013 Accepted: 25 March 2014Published: 31 March 2014References1. Hidayat B, Thabrany H, Dong H, Sauerborn R: The effects of mandatoryhealth insurance on equity in access to outpatient care in Indonesia.Health Policy Plan 2004, 19(5):322–335.2. Kirigia JM, Preker A, Carrin G, Mwikisa C, Diarra-Nama AJ: An overview ofhealth financing patterns and the way forward in the WHO Africanregion. East Afr Med J 2006, 83(8):S1–S27.3. Mathauer I, Schmidt JO, Wenyaa M: Extending social health insurance tothe informal sector in Kenya. An assessment of factors affectingdemand. Int J Health Plann Manage 2008, 23:51–68.4. Quayyum Z, Nadjib M, Ensor T, Sucahya PK: Expenditure on obstetric careand the protective effect of insurance on the poor: lessons from twoIndonesian districts. Health Policy Plan 2010, 25(3):237–247.5. World Health Organization: Paying for Health Services. In Social HealthProtection. Geneva, Switzerland: World Health Organization Media Center; 2007.6. Ole D, Carrin G, Evans D: Thinking of introducing social health insurance?ten questions. In World Health Report (2010) Background Paper, No 26.Geneva, Switzerland: Department of Health Systems Financing, WHO; 2010.7. World Health Organization: Health financing reform in Kenya: assessingthe social health insurance proposal. In Department of Health SystemKimani et al. International Journal for Equity in Health 2014, 13:27 Page 7 of 8http://www.equityhealthj.com/content/13/1/27Financing (HSF) Cluster- Evidence and Information for Policy (EIP). Geneva,Switzerland: WHO; 2006.8. Hakijamii Trust (Economic and Social Rights Centre): The right to socialsecurity in Kenya: the gap between International Human Rights andDomestic law and Policy. In Parallel Report to Committee on Economic,Social and Cultural Rights. Nairobi, Kenya: Hakijamii Trust; 2007.9. Kimani D, Muthaka DI, Manda DK: Healthcare Financing Through Health Insurancein Kenya. The Shift to A National Social Health Insurance Scheme. Nairobi, Kenya:Kenya Institute for Public Policy Research and Analysis; 2004:1–71.10. Kenya National Bureau of Statistics: Report of 1998/99 Labour Force Survey.Nairobi, Kenya: Kenya National Bureau of Statistics; 2003.11. Boateng D, Awunyor-Vitor D: Health insurance in Ghana: evaluation ofpolicy holders’ perceptions and factors influencing policy renewal in theVolta region. Int J Equity Health 2013, 12:1–10.12. Kirigia JM, Sambo LG, Nganda B, Mwabu GM, Chatora R, Mwase T:Determinants of health insurance ownership among South Africanwomen. BMC Health Serv Res 2005, 5:1–10.13. Kumi-Kyereme A, Amo-Adjei J: Effects of spatial location and householdwealth on health insurance subscription among women in Ghana.BMC Health Serv Res 2013, 13:221.14. Liu TC, Chen CS: An analysis of private health insurance purchasingdecisions with national health insurance in Taiwan. Soc Sci Med 2002,55(5):755–774.15. Onwujekwe O, Okereke E, Onoka C, Uzochukwu B, Kirigia J, Petu A:Willingness to pay for community-based health insurance in Nigeria:do economic status and place of residence matter? Health Policy Plan2010, 25:155–161.16. Perlman F, Balabanova D, McKee M: An analysis of trends anddeterminants of health insurance and healthcare utilisation in theRussian population between 2000 and 2004: the ‘inverse care law’ inaction. BMC Health Serv Res 2009, 9:68.17. Richardson E, Roberts B, Sava V, Menon R, McKee M: Health insurancecoverage and health care access in Moldova. Health Policy Plan 2011,27:204–212.18. Russell S: Ability to pay for health care: concepts and evidence. HealthPolicy Plan 1996, 11(3):219–237.19. Sarpong N, Loag W, Fobil J, Meyer CG, Adu-Sarkodie Y, May J, Schwarz NG:National health insurance coverage and socio-economic status in a ruraldistrict of Ghana. Trop Med Int Health 2010, 15(2):191–197.20. Schneider P: Why should the poor insure? Theories of decision-making inthe context of health insurance. Health Policy Plan 2004, 19(6):349–355.21. Poverty Data by Country. http://data.worldbank.org/country/kenya.22. Abt Associates I: Kenya National Health Accounts 2002: EstimatingExpenditures on General Health and HIV/AIDS Care. Bethesda, Maryland, USA:Abt Associates, Inc.; 2005.23. Government of Kenya Health Systems 2020 Project: Kenya National HealthAccounts 2005/2006. Bethesda, MD: Health Systems 20/20 project, AbtAssociates Inc; 2009.24. Falkingham J, Akkazieva B, Baschieri A: Trends in out-of-pocket payments forhealth care in Kyrgyzstan, 2001–2007. Health Policy Plan 2010, 25(5):427–436.25. Perkins M, Brazier E, Themmen E, Bassane B, Diallo D, Mutunga A, MwakajongaT, Ngobola O: Out-of-pocket costs for facility-based maternity care in threeAfrican countries. Health Policy Plan 2009, 24:289–300.26. Ziraba AK, Mills S, Madise N, Saliku T, Fotso JC: The state of emergencyobstetric care services in nairobi informal settlements and environs: resultsfrom a maternity health facility survey. BMC Health Serv Res 2009, 9:46.27. Evans DB, Etienne C: Health systems financing and the path to universalcoverage. Bull World Health Organ 2010, 88:402.28. Nations U: The Millennium Development Goals Report 2010. United Nations:New York, U.S.A; 2010.29. Kenya Ministry of Health: Reversing the Trends: The Second National HealthSector Strategic Plan of Kenya - NHSSP II: Midterm Review Report. Nairobi, Kenya:Ministry of Health - Sector Planning and Monitoring Department; 2007.30. Kenya Ministry of Medical Services: Ministry of Medical Services Strategic Plan,2008–2012. Nairobi, Kenya: Ministry of Medical Services; 2008.31. Ministry of Planning and National Development: Vision 2030. Nairobi, Kenya:Ministry of Planning and National Development; 2009.32. Kenya Community Based Health Financing Association: The role of theKCBHFA in the transformation of National Hospital Insurance Fund (NHIF) toNational Social Health Insurance Fund. Nairobi, Kenya: Kenya CommunityBased Health Financing Association; 2008.33. De Allegri M, Sanon M, Bridges J, Sauerborn R: Understanding consumers’preferences and decision to enrol in community-based health insurancein rural West Africa. Health Policy 2006, 76:58–71.34. De Allegri M, Sanon M, Sauerborn R: To enrol or not to enrol?”: aqualitative investigation of demand for health insurance in rural WestAfrica. Soc Sci Med 2006, 62:1520–1527.35. Bennett S: The role of community-based health insurance within thehealth care financing system: a framework for analysis. Health Policy Plan2004, 19(3):147–158.36. Dong H, Allegri De M, Gnawali D, Souares A, Sauerborn R: Drop-outanalysis of community-based health insurance membership at Nouna,Burkina Faso. Health Policy 2009, 92:174–179.37. Dong H, Mugisha F, Gbangouc A, Kouyate B, Sauerborn R: The feasibility ofcommunity-based health insurance in Burkina Faso. Health Policy 2004,69:45–53.38. Franco LM, Diop FP, Burgert CR, Kelley AG, Makinena M, Simparae CHT:Effects of mutual health organizations on use of priority health-careservices in urban and rural Mali: a case–control study. Bull World HealthOrgan 2008, 86:830–838.39. Jütting J: Health Insuarnce for the Poor?. Determinants of participation incommunity-based health insurance schemes in rural Senegal: OECD; 2003.40. Ursula G, Alfonso EA, Díaz Y: The impact of universal coverage schemes inthe developing world: a review of the existing evidence. In UNICO StudiesSeries 25. Washington DC: The World Bank; 2013.41. Spaan E, Mathijssen J, Tromp N, McBain F, Ten Haveb A, Baltussena R: Theimpact of health insurance in Africa and Asia: a systematic review.Bull World Health Organ 2012, 90:685–692A.42. Carrin G, James C, Adelhardt M, Doetinchem O, Eriki P, Hassan M, van denHombergh H, Kirigia J, Koemm B, Korte R, Krech R, Lankers C, Lente Van J,Maina T, Malonza K, Mathauer I, Okeyo TM, Muchiri S, Mumani Z, Nganda B,Nyikal J, Onsongo J, Rakuom C, Schramm B, Scheil-Adlung X, Stierle F,Whitaker D, Zipperer M: Health financing reform in Kenya – assessing thesocial health insurance proposal. South Afr Med J 2007, 97(2):130–135.43. Kenya National Bureau of Statistics, ICF Macro: Kenya Demographic andHealth Survey 2008–09. Calverton, Maryland: KNBS and ICF Macro; 2009.44. Wears RL: Advanced statistics: statistical methods for analyzing clusterand cluster-randomized data. Acad Emerg Med 2002, 9(4):330–341.45. Lumley T: Analysis of complex survey samples. J Stat Softw 2004, 9(1):1–19.46. Mputhia C: Benefits of legally registering investment clubs law. In Nation,Business Daily. Business Daily edition edition. Nation Media Group: Nairobi,Kenya; 2010.47. Dong H, Kouyate B, Cairns J, Mugisha FRS: Willingness-to-pay forcommunity-based insurance in Burkina Faso. Health Econ 2003,12(10):849–862.48. Govender V, Chersich MF, Harris B, Alaba O, Ataguba JE, Nxumalo N,Goudge J: Moving towards universal coverage in South Africa? Lessonsfrom a voluntary government insurance scheme. Glob Health Action 2013,6:19253.doi:10.1186/1475-9276-13-27Cite this article as: Kimani et al.: Determinants of health insuranceownership among women in Kenya: evidence from the 2008–09 Kenyademographic and health survey. International Journal for Equity in Health2014 13:27.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitKimani et al. International Journal for Equity in Health 2014, 13:27 Page 8 of 8http://www.equityhealthj.com/content/13/1/27


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items