UBC Faculty Research and Publications

Health system determinants of infant, child and maternal mortality: A cross-sectional study of UN member… Muldoon, Katherine A; Galway, Lindsay P; Nakajima, Maya; Kanters, Steve; Hogg, Robert S; Bendavid, Eran; Mills, Edward J Oct 24, 2011

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

Item Metadata

Download

Media
52383-12992_2011_Article_139.pdf [ 811.39kB ]
Metadata
JSON: 52383-1.0223211.json
JSON-LD: 52383-1.0223211-ld.json
RDF/XML (Pretty): 52383-1.0223211-rdf.xml
RDF/JSON: 52383-1.0223211-rdf.json
Turtle: 52383-1.0223211-turtle.txt
N-Triples: 52383-1.0223211-rdf-ntriples.txt
Original Record: 52383-1.0223211-source.json
Full Text
52383-1.0223211-fulltext.txt
Citation
52383-1.0223211.ris

Full Text

RESEARCH Open AccessHealth system determinants of infant, child andmaternal mortality: A cross-sectional study of UNmember countriesKatherine A Muldoon1,2, Lindsay P Galway3, Maya Nakajima3, Steve Kanters3, Robert S Hogg2,3, Eran Bendavid4 andEdward J Mills2,5*AbstractObjective: Few studies have examined the link between health system strength and important public healthoutcomes across nations. We examined the association between health system indicators and mortality rates.Methods: We used mixed effects linear regression models to investigate the strength of association betweenoutcome and explanatory variables, while accounting for geographic clustering of countries. We modelled infantmortality rate (IMR), child mortality rate (CMR), and maternal mortality rate (MMR) using 13 explanatory variables asoutlined by the World Health Organization.Results: Significant protective health system determinants related to IMR included higher physician density(adjusted rate ratio [aRR] 0.81; 95% Confidence Interval [CI] 0.71-0.91), higher sustainable access to water andsanitation (aRR 0.85; 95% CI 0.78-0.93), and having a less corrupt government (aRR 0.57; 95% CI 0.40-0.80). Out-of-pocket expenditures on health (aRR 1.29; 95% CI 1.03-1.62) were a risk factor. The same four variables weresignificantly related to CMR after controlling for other variables. Protective determinants of MMR included access towater and sanitation (aRR 0.88; 95% CI 0.82-0.94), having a less corrupt government (aRR 0.49; 95%; CI 0.36-0.66),and higher total expenditures on health per capita (aRR 0.84; 95% CI 0.77-0.92). Higher fertility rates (aRR 2.85; 95%CI: 2.02-4.00) were found to be a significant risk factor for MMR.Conclusion: Several key measures of a health system predict mortality in infants, children, and maternal mortalityrates at the national level. Improving access to water and sanitation and reducing corruption within the healthsector should become priorities.BackgroundA working definition of a health system, as proposed bythe World Health Organization (WHO) is a system“whose primary purpose is to promote, restore, or main-tain health” [1]. In 2007, with the purpose of promotinga common understanding of what a health system isand action areas for strengthening health systems, theWHO developed a framework composed of six buildingblocks of a health system: 1) health service coverage, 2)human health resources, 3) health information systems,4) medical products, vaccines and technology, 5) healthfinancing, and 6) leadership and governance [2]. Thesebuilding blocks aim to support a health system that canprevent, treat and manage illness and to preserve mentaland physical well-being for all individuals equitably andefficiently, within a specified geographic area. Healthsystem activities range from direct service provisionthrough clinics and hospitals to community level pre-vention strategies and health education. Over the pastdecade there has been renewed interest in the horizontalrole of health systems in the promotion and mainte-nance of health [3]. Additionally, the robustness of apublic health system has been highlighted as a necessarycomponent to achieve the Millennium DevelopmentGoals (MDG) [4,5], however the indicators to measurehealth system strengthening are less understood.* Correspondence: Edward.mills@uottawa.ca2British Columbia Centre for Excellence in HIV/AIDS, St. Paul’s Hospital, 1081Burrard Street, Vancouver, British Columbia, CanadaFull list of author information is available at the end of the articleMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42© 2011 Muldoon 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 cited.There is an on-going debate about global health ‘geo-metry’ of the vertical or horizontal approaches to healthas both have strengths and limitations [6-8]. Both pri-vate and public systems can employ vertical and hori-zontal approaches to health care and programming andsome have even used the term ‘diagonal’ to describecombining the two approaches to optimize processesand outcomes [9]. A notable trend is that private orga-nizations tend to have a more narrow focus and employa more vertical approach. For example, in many low-income countries (LIC), externally led, donor drivenprojects have met with some success, especially with theestablishment of care centres for the treatment and pre-vention of HIV/AIDS, immunization coverage, TB con-trol, and Roll Back Malaria Campaigns, all typicallyconsidered a vertical approach to health. These disease-focused initiatives are intensive, may avoid the bureau-cracies and inefficiencies of a national health system,and are typically implemented to either respond to anemergency (as in the case of HIV/AIDS) or meet donorspecific requirements (such as vaccines through GAVI,the Global Alliance for Vaccines and Immunizations).However, investments aimed at the overall strength andfunctioning of a health system (i.e. horizontalapproaches to health) are grounded in the expectationthat a functioning, efficient health care system will con-tribute most effectively to improving the health of apopulation [10].Although some countries have made substantialimprovements in infant, child and maternal mortalityrates (IMR, CMR, MMR respectively) during the lastcentury, improvements have slowed and even reversedin some nations during the last few decades [11]. Anestimated 9.7 million children under-five die worldwideeach year [12]. Additionally, mortality rates are highlyvariable across nations highlighting health inequities andlarger social and environment determinants that predis-pose some nations to higher rates of mortality [13]. Dif-ferences in all-cause mortality rates across nations may,in part, be explained by the strength and functioning ofa national health system’s ability to safe-guard healthbeyond the disease specific approach. Important fundingagencies such as the US Global Health Initiative, arenow directing their financial contributions to health sys-tem strengthening at the expense of disease focusedinitiatives, even though validated indicators to determineand monitor health systems strength are not well deter-mined or understood [14]. We aimed to develop anexploratory analysis to examine the strength of associa-tion between important public health endpoints (IMR,CMR, MMR) and potential indicators of health systemstrength and functioning as theorized by the WHOusing publicly available data.MethodsData and variablesVariable selection was informed a priori by the WHObuilding block framework. The goal was to select vari-ables that could represent each of the 6 building blocksand then to investigate how well they explain the varia-bility in global mortality rates. All data was publiclyaccess so variable selection was constrained by dataavailability. Data on ten indicators categorized into fiveof the six main building blocks of a health system asoutlined by the WHO, and four relevant demographicvariables were used as explanatory variables. Nursingand midwife density and physician density measuredavailable human health resources. Vaccines coveragewas indicated by the percentage of children receivingmeasles immunizations annually. Health service deliverywas represented by the percentage of the populationwith sustainable access to water and sanitation and thepercentage of births attended by skilled attendants.Health financing was assessed by total, out-of-pocket,government, and private expenditures on health. Thehealth finance data was gathered from WHOSIS. Theycite that all financial measurements are made using the“International dollar rate [which] is a common currencyunit that takes into account differences in relative pur-chasing power annual average”.Finally, The Corruption Perception Index, a metricdesigned to measure the perceived levels of public sectorcorruption published annually by Transparency Interna-tional, was used to measure the governance and leader-ship category [15]. Although the CPI focuses onperceptions of corruption rather than the actual extentof corruption, the index has been assessed to be a reli-able and consistent measure [16]. The final buildingblock of a health system is health information systemsthat can be captured by the presence of a functioningsurveillance system, however multinational data was notavailable for this building block. Together these indica-tors act as a proxy representing the robustness ofnational health systems to finance, staff, and providehealth services to their citizens. Demographic variablesincluded fertility rate, national population growth, urbanpopulation growth, and female labour force participationand were used to capture demographic heterogeneityacross countries.We extracted all data from our prospectively main-tained archive of publicly accessible health statistics,named the Globally Accumulated health IndicatorArchive (GAIA). Source data for the outcome andexplanatory variables originated from UN and WHOdata, with the exception of the CPI, which originatedfrom Transparency International; all publicly availablesources. The outcome variables are based on 2008 dataMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 2 of 10while the explanatory variables were collected over aseven-year span from 2001-2008 using the most recentdata available. Of 192 UN member countries, 136 coun-tries provided sufficient data for the chosen variables.Eight of the 136 countries would have been excludeddue to lack of data on sustainable access to water.Rather than excluding these countries, we assumed 95%value for Poland and Portugal and assumed 100% forBelgium, France, Ireland, Italy, New Zealand, and theUnited Kingdom (the median value for Australia, andWestern European and North American countries).Without this assumption the countries from Westernand Southern Europe were under-represented.Statistical AnalysisDescriptive statistics were used to display the dispersionof the outcome and explanatory variables. A linearmixed effect model was chosen to account for the nat-ural geographic clustering of the countries according toUN sub-region classification. In order to comply withthe strict conditions of linear modeling, some transfor-mations were required. Each outcome required a loga-rithmic transformation. Nursing and midwife density,total government spending, out-of- pocket expenditures,government expenditures and fertility rate were trans-formed via logarithm. Measles immunization and skilledbirth attendants were dichotomized as 90% or more andunder 90% based on the scatter plot indicating a cleardrop-off after 90%.Multicollinearity was an issue as the variance inflationfactors (VIF) was high for government health expendi-tures. Upon removing government expenditures, the VIFwere moderate in size, reaching a maximum value of6.21 when considering the full model prior to modelselection. Model conditions were assessed through ana-lysis of marginal and conditional residuals. Model selec-tion was achieved by minimizing the AkaikeInformation Criterion (AIC) while keeping all type III p-values for covariates below 0.20. Unadjusted results con-sider the association between the outcome and eachexplanatory variable individually. Adjusted risk ratiosconsider the association between the outcome and anexplanatory variable simultaneous to all variablesselected in the model. Variables selected in the multi-variate models are considered the strongest predictorsbecause the non-selected variables are no longer infor-mative with respect to the outcome. All analyses weredone by SK using SAS 9.1.3 [17].Ethics approval for this project was not requiredbecause it uses publicly available data.ResultsThe descriptive statistics for each of the outcome mea-sures (IMR, CMR, MMR) and the explanatory variablesare included in Table 1. The median IMR across allnations was 21.5 deaths per 1,000 live births (IQR 10.0 -60.0), median CMR was 24.5 deaths per 1,000 live births(IQR 11.0-80.0) and median MMR was 81.5 deaths per100,000 live births (IQR 26.0-350.0). The geographicclassification of the 136 countries included in this studyis shown in Table 2. Of the 136 countries, 46 (33.8%) ofthe countries are located in Sub-Saharan Africa; 39(28.7%) in Asia; 25 (18.4%) in Europe; 21 (15.4%) inLatin America and the Caribbean; 2 (1.8%) in NorthAmerica; and 3 (2.2%) in Oceania. The proportion ofcountries included in the model varies between regions,where over 80% of all Sub-Saharan countries areincluded but only 12% of Oceanic countries had suffi-cient data available for inclusion in this model. Thecountries included in the analysis and the mortalityrates are represented in Figure 1, Figures 2, 3, and 4show the global distribution of mortality rates in 2008.All selected health system indicators were significantlyassociated with IMR at the bivariate level except forpopulation growth and female labour force participation,and were therefore included in the multiple regressionanalysis. When controlling for the effects of other vari-ables in the model, four variables remained significantlyassociated with IMR. Health system determinants asso-ciated with lower IMR are higher physician density(adjusted rate ratio [aRR] 0.81; 95% CI 0.71-0.91), highersustainable access to water and sanitation (aRR 0.85;95% CI 0.78-0.93), and having a less corrupt government(aRR 0.57; 95% CI 0.40-0.80). Out-of-pocket expenditureon health (a-RR 1.29; 95% CI 1.03-1.62) was associatedwith higher for IMR (see Table 3).The same four variables that were significantly asso-ciated with IMR were also significant for CMR aftercontrolling for other factors (see Table 4). Higher physi-cian density (aRR 0.80; 95% CI 0.70-0.92), higher sus-tainable access to water and sanitation (aRR 0.82, 95%CI 0.75-0.91), and having a less corrupt government (a-RR 0.58; 95% CI 0.40-0.84) were associated with lowerCMR. Out-of-pocket expenditures on health (aRR 1.29;95% CI 1.01, 1.65) was significantly associated withhigher CMR.Finally, higher sustainable access to water and sanita-tion (aRR 0.88; 95% CI 0.82-0.94), having a less corruptgovernment (aRR 0.49; 95% CI 0.36-0.66), and highertotal expenditures on health per capita (a-RR 0.84; 95%CI 0.77-0.92) were associated with lower MMR. Itshould be noted that higher fertility rate (aRR 2.85; 95%CI 2.02-4.00) is a significant risk factor for MMR (seeTable 5).InterpretationThis ecological analysis explores how the WHO buildingblocks of a health system are associated with infant,Muldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 3 of 10child and maternal mortality rates across 136 UN mem-ber countries. Service coverage as measured by sustain-able access to water is associated with decreasedmortality. Leadership and governance as measured bythe corruption index (i.e. less government corruption)are associated with decreased mortality. Human healthresources as measured by physician density, and healthfinancing as measured by less out-of-pocket paymentsare associated with decreased mortality but only forinfants and children.Stewardship is a neglected function in most healthsystems [18]. Murray & Frenck (2000) have describedhealth system stewardship as involving three key aspects“setting, implementing and monitoring the rules for thehealth system; assuring a level playing field for all actorsin the system; and defining strategic directors for thehealth system as a whole”. Currently there is no onemetric to measure health stewardship at the nationallevel, we used the Corruption Index as a measure ofnational governance and a proxy for health system stew-ardship because the general functioning of the govern-ment can strongly influence stewardship and regulation.Corruption is broadly defined by Transparency Interna-tional as the misuse of public office for private gain[19]. As a result, our findings are limited to corruptionwithin the public sphere although we do acknowledgeTable 1 Descriptive statistics for all outcome and explanatory variables sub-divided into the WHO framework for thebuilding blocks of a health system (n = 136 countries)Variables Median (IQR) RangeOutcomeInfant mortality rate (per 1,000 births) 21.5 (10.0 - 60.0) 2.0 - 165.0Child mortality rate (per 1,000 births) 24.5 (11.0 - 80.0) 3.0 - 257.0Maternal mortality ratio (per 100,000 births) 81.5 (26.0 - 350.0) 3.0 - 1400.0ExplanatoryI. Human health resourcesNursing/midwife density (per 10,000 population) 18.5 (7.0 - 51.0) 2.0 - 158.0Physician density (per 1,000 population) 11.0 (2.0 - 25.0) 0.3 - 64.0II. Health service coverage% Of population with sustainable access to water and sanitation 87.50 (59.0 - 98.5) 24.0 - 100.0% Of births attended by skilled staff 93.0 (57.0 - 100.0) 6.0 - 100.0III. Medical products, vaccines and technology% Measles immunization coverage 91.0 (79.0 - 97.0) 23.0 - 99.0IV. Health financingTotal health expenditure per person (USD) 153.0 (35.5 - 441.0) 4.0 - 6714.0Out-of-pocket expenditure on health (as a % of total health expenditure) 33.1 (19.8 - 48.4) 4.2 - 82.7Government health expenditure (USD) 148.0 (41.0 - 457.5) 4.0 - 3074.0Private share of total health expenditure (%) 44.8 (27.9 - 58.5) 9.3 - 83.6V. Leadership and governanceCorruption Index 3.0 (2.4 - 4.5) 1.3 - 9.4Demographic variablesFertility rate (average number of children per woman) 2.5 (1.8 - 4.1) 1.2 - 6.6Population growth value (annual %) 1.42 (0.72 - 2.29) -1.17 - 5.32Urban population value (annual %) 2.23 (1.16 - 3.35) -1.02 - 5.90Female labour force participation (%) 59.8 (48.5 - 68.1) 14.9 - 90.2Lower value of Corruption Index on a scale of ten indicates higher perceived corruptionTable 2 Descriptive classification of the study countries (n = 136 countries)Region N (%) Total number of countries by region, % included in the analysis by regionAfrica 46 (33.8) 57 (80.7)Asia 39 (28.7) 50 (78.0)Europe 25 (18.4) 51 (49.0)Latin America and the Caribbean 21 (15.4) 48 (43.8)North America 2 (1.5) 5 (40.0)Oceania 3 (2.2) 25 (12.0)Total 136 (100.0) 236Muldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 4 of 10that corruption is present in the private and non-gov-ernmental arena. In our study we have found that themore corrupt a government is perceived to be (i.e. lowerCPI score) the stronger the association with increasedrates of infant, child and maternal mortality.As health systems are publicly administered andrequire strong national commitment and resources, acorrupt government runs the risk of diverting publichealth resources for private gains. Our findings suggestthat transparent governance is an essential componentof health system strengthening and an important path-way to improve population health. Three quarters of thecountries in the world have a CPI score less than five,translating to a serious level of corruption [20], as aresult it has been recognized by the UN that anti-cor-ruption should be a central approach to global aid anddevelopment [21]. Corruption is systemic and existswithin and across scales and sectors of the governmentand thus requires anti-corruption efforts that are bothbroad and sector-specific. Private vertical programs areoften fast and effective because they often operate out-side the public sphere, however an unintended conse-quence of this approach could be enabling a cycle ofcorruption within the public sphere. Public health existsand is implemented within the larger public system, andtherefore must incorporate wherever possible policiesthat buttress transparency among participating stake-holders from multiple disciplines [22].Sustainable access to water and sanitation was signifi-cantly associated with IMR, CMR and MMR when con-trolling for other variables presumably for severalreasons. Elevated incidence and prevalence rates of diar-rhoeal diseases are commonly observed in settings withlimited access to improved and sustainable water andsanitation services. Foreign aid is associated withincreased access to water, but not necessarily sanitation[23]. Water-borne diarrhoeal diseases alone account for17% of deaths in children under-five and 1% of neonataldeaths [12]. Other ecological level studies have alsoshown that MMR is strongly associated with sustainableFigure 1 Countries included in analysis (n = 136) .Figure 2 Infant mortality rate per 1000 live births across countries (n = 136).Muldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 5 of 10access to water and sanitation because access to safedrinking water is a fundamental pillar for maternalhealth [24]. Unhygienic birthing practices and facilitiesthat are not properly equipped to provide a sterile envir-onment for a post-partum mother commonly contributeto elevated rates of maternal mortality. Mothers who areunable to breast-feed are at risk of using unsafe waterfor formula-feeding especially in low income countriesas a mode of prevention of mother-to-child transmissionof HIV [25].Health financing was a central finding across all threemodels. Each financial variable with the exception ofprivate share of total health expenditure was signifi-cantly related to mortality outcomes, but once weincluded them within the multivariate model out-of-pocket best explained IMR and CMR, and total healthexpenditure best describes the MMR. This finding is notindicative that out-of-pocket is not important for MMR,or that total health expenditure is not important forIMR and CMR, but rather that the model selected thevariable that described the strongest association. Out-of-pocket expenditure is a commonly cited barrier tohealth care especially if out-of-pocket costs exceedhousehold income. In many African countries, thehealth financing system is too weak to function withoutthe cushion of out-of-pocket costs. In a study of 15African countries investigating household coping beha-viours in the face of health expenditures, it was foundthat between 23-68% of households would resort to bor-rowing and selling their assets [26]. Households in thissituation are often affected by both the cost of medicalcare, but also the loss of income from sick family mem-bers that cannot work [26]. This contributes to a highlyinequitable system that puts infants and children atincreased risk for adverse health outcomes and death.Although we cannot tell the temporality of this rela-tionship, we observe that as per capita spending onhealth increases mortality rates decrease. Others haveshown that total health expenditures is a significant pre-dictor of IMR in their bivariate analysis, however, this isno longer significant in the multivariate model, afterincluding Gross National Income per capita [11]. Thiswas the same for our analysis and probably points tothe larger influence of a countries economic status (i.e.Figure 3 Child mortality rate per 1000 live births across countries (n = 136) .Figure 4 Maternal mortality rate per 100,000 live births across countries (n = 136).Muldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 6 of 10GNI) rather than the amount of funding earmarked forhealth care.Physician density significantly reduces infant and childmortality but does not appear to reduce maternal mor-tality after controlling for other health system indicators,nor does nursing and midwife density. There have beenat least six cross-national studies that have investigatedhuman health resources, indicated by either physician ornurse densities as predictors of infant mortality[4,27-32]. Of these studies, four found no relationshipbetween human health resources while two of the morerecent studies have indicated that both physician andnurse densities are significant in accounting for varia-tions in rates of infant mortality across countries. Inter-estingly, Farahani et al. (2009) have shown, usinglongitudinal panel data to examine both the short- andlong-term effects of human health resources, thathuman health resources may have greater long-termbenefits than previously estimated. We chose not to usean amalgamated measure (i.e. nurses, doctors, skilledbirth attendants) for human health resources and foundthat physician density was significant yet nurse densityand % of births with a skilled attendant was not signifi-cant. This could be due to the fact that some countriesonly include professional nurses while associate profes-sion such as nursing assistants are not included [33].This would under-represent the role that nurses play inhuman health resources.The MDG #6 was designed to improve maternalhealth because it is estimated that in some areas of theworld a woman has a 1 in 16 chance of dying in preg-nancy. High infant, child and maternal mortality areoften observed concurrently with high fertility, howeveronly MMR was positively and significantly associatedwith fertility in our analysis. It is widely supported thata high fertility rate is observed in settings where chil-dren are not surviving and families need to replace thelost children. If a woman has had a complication duringa previous pregnancy or her health becomes compro-mised this can lead to a vicious circle that puts mothers(and children) at risk for malnourishment and healthcomplications [34].In 1990, The World Summit for Children called for areduction in infant mortality to below 70 deaths per1000 live births (or a one third reduction if this resultedin a lower mortality rate) by the year 2000 [12,35]. Thisgoal was attained by discouragingly few nations; a failurethat some suggest may be rooted in inadequateTable 3 Linear mixed effect regression analysis results for IMR, 2008 sub-divided into the WHO framework for thebuilding blocks of a health system (n = 136 countries)Explanatory Variables Unadjusted Risk Ratio (95% CI) Adjusted Risk Ratio (95%CI)I. Human health resourcesNursing/midwife density (per 10,000 population) 0.82 (0.71, 0.94) -Physician density (per 1,000 population) 0.72 (0.63, 0.83) 0.81 (0.71, 0.91)II. Health service coverage% Of population with sustainable access to water and sanitation(for a 10% increase)0.74 (0.68, 0.80) 0.85 (0.78, 0.93)% Of births attended by skilled staff 0.28 (0.20, 0.39) -III. Medical products, vaccines and technology% Measles immunization coverage 0.71 (0.52, 0.98) -IV. Health financingTotal health expenditure per person (USD) 0.74 (0.67, 0.82) -Out-of-pocket expenditure on health (as a % of total health expenditure) 1.60 (1.28, 2.01) 1.29 (1.03, 1.62)Government health expenditure (USD) 0.65 (0.58, 0.71) -Private share of total health expenditure (%) 1.01 (1.00, 1.02) -V. Leadership and governanceCorruption index (log of) 0.37 (0.26, 0.53) 0.57 (0.40, 0.80)Demographic variablesFertility rate (average number of children per woman) 3.07 (2.04, 4.62) -Population growth value (annual %) 1.20 (1.01, 1.43) -Urban population value (annual %) 1.26 (1.12, 1.43) -Female labour force participation (%) 1.00 (0.99, 1.01) -- : Not selected in final modelCI: Confidence intervalA Risk Ratio below 1 corresponds to a protective variableA Risk Ratio above 1 corresponds to a risk factorMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 7 of 10investments in health and limited improvements in thestrength and functioning of health systems [35]. Resultsfrom our analyses show that more up-stream determi-nants such as sustainable access to water and sanitation,health financing, and transparent governance are impor-tant pathways to reducing mortality rates. Health finan-cing is not currently listed within the MDGs howeverthe latest WHO report [36] focuses exclusively on sus-tained economic and social development to movetowards universal coverage and improved health out-comes. Studies such as this are needed to strengthenour current understanding of the role of health systemsas a societal safety net in achieving the MDGs andimproving health worldwide.LimitationsSeveral limitations should be considered when interpret-ing these results. Data selection was constrained primar-ily by data availability and therefore does not includethe most comprehensive list of health system indicators.Our sample size (n = 136 countries) also constrains ourchoices for the number of variables that we can includein the model. As a result we have a relatively smallnumber of variables used to describe the variability andcomplex nature of a health system. This study is across-sectional analysis at the country level and there-fore we cannot draw causal inferences from the results.As we have used countries as the unit of analysis, thisdoes not provide any information about variation withinthe nation state. This is an important point to stressbecause health status throughout a country may varytremendously and these differences will be masked bycountry-level data. While many studies have controlledfor female education as an important variable related toinfant mortality, we did not include this as an explana-tory variable because the data was not adequately popu-lated [11]. In place, we used the indicator for femalelabour involvement. While our study focused on out-comes of maternal and child health we recognize thatmen are one of the highest risk groups for early mortal-ity, yet are not the focus of any large directed fundinginitiatives, with the possible exception of male circumci-sion [37].ConclusionIn conclusion, our analysis identifies the importance ofseveral key indicators of health system strength andfunctioning that are significantly associated with infant,Table 4 Linear mixed effect regression analysis results for CMR, 2008 sub-divided into the WHO framework for thebuilding blocks of a health system (n = 136 countries)Explanatory Variables Unadjusted Risk Ratio (95% CI) Adjusted Risk Ratio (95%CI)I. Human health resourcesNursing/midwife density (per 10,000 population) 0.80 (0.69, 0.93) -Physician density (per 10,000 population) 0.71 (0.61, 0.82) 0.80 (0.70, 0.92)II. Health service coverage% Of population with sustainable access to water and sanitation(for a 10% increase)0.71 (0.65, 0.77) 0.82 (0.75, 0.91)% Of births attended by skilled staff 0.48 (0.32, 0.72) -III. Medical products, vaccines and technology% Measles immunization coverage 0.67 (0.48, 0.94) -IV. Health financingTotal health expenditure per person (USD) 0.73 (0.66, 0.82) -Out-of-pocket expenditure on health (as a % of total health expenditure) 1.64 (1.28, 2.10) 1.29 (1.01, 1.65)Government health expenditure (USD) 0.63 (0.56, 0.70) -Private share of total health expenditure (%) 1.01 (1.00, 1.02) -V. Leadership and governanceCorruption index ( log of) 0.35 (0.24, 0.52) 0.58 (0.40, 0.84)Demographic variablesFertility rate (average number of children per woman) 3.54 (2.28, 5.49) -Population growth value (annual %) 1.25 (1.04, 1.52) -Urban population value (annual %) 1.31 (1.15, 1.50) -Female labour force participation (%) 1.00 (0.99, 1.01) -- : Not selected in final modelCI: Confidence intervalA Risk Ratio below 1 corresponds to a protective variableA Risk Ratio above 1 corresponds to a risk factorMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 8 of 10child and maternal survival at the national aggregatelevel and after controlling for other health system deter-minants and demographic factors. The strength of ahealth system offers an important and sustainablemechanism to influence key population level indicatorsof health. There is now an important need to under-stand indicators of health system strength at the locallevel and how to improve health system strength andfunctioning in practice.EthicsEthics approval for this project was not required becauseit uses publicly available data.AcknowledgementsThe authors would like to acknowledge Erin Ding, Anya Shen, andChristopher Au-Yeung for contributions to the preliminary analysis. Nofunding was received for this work, no funding bodies played any role inthe design, writing or decision to publish this manuscript.Author details1School of Population and Public Health, University of British Columbia, 2206East Mall, Vancouver, British Columbia, Canada. 2British Columbia Centre forExcellence in HIV/AIDS, St. Paul’s Hospital, 1081 Burrard Street, Vancouver,British Columbia, Canada. 3Faculty of Health Sciences, Simon FraserUniversity, 888 University Drive, Burnaby, British Columbia, Canada. 4Divisionof General Internal Medicine, Stanford University, Palo Alto, California, USA.5Faculty of Health Sciences, University of Ottawa, Roger Guindon Hall 451,Smyth Road, Ottawa, Ontario, Canada.Authors’ contributionsKAM, LPG and MN contributed equally to the drafting, interpretation andincorporation of critical feedback from co-authors. SK conducted thestatistical analysis and assisted with interpretation. EB, RSH and EJMsupervised, drafted, and provided critical feedback at all stages of themanuscript. All authors read and approved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 10 June 2011 Accepted: 24 October 2011Published: 24 October 2011References1. WHO: World Health Report: Health Systems Improving Performance.Geneva, Switzerland; 2000.2. WHO: Everybody’s Business: Strengthening Health Systems to ImproveHealth Outcomes. Geneva: World Health Organization; 2007, 44, pp. 44.3. Leipziger D, Fay M, Wodon Q, Yepes T: Achieving the MillenniumDevelopment Goals. World Bank Policy Working Paper 2003, 3163.4. Farahani M, Subramanian SV: The Effect of Changes in Health SectorResources on Infant Mortality in the Short-run and the Long-run: Alongitudinal econometric analysis. Social Science & Medicine 2009,68:1918-1925.5. UN: UN Millennium Project: Who’s Got the Power? Transforming healthsystems for women and children. Summary version of the report of theTask Force on Child Health and Maternal Health 2005.6. Levine R: Should All Vertical Programs Just Lie Down? Centre for GlobalDevelopment 2007.Table 5 Linear mixed effect regression analysis results for MMR, 2008 sub-divided into the WHO framework for thebuilding blocks of a health system (n = 136 countries)Explanatory Variables Unadjusted Risk Ratio (95% CI) Adjusted Risk Ratio (95%CI)I. Human health resourcesNursing/midwife density (per 10,000 population) 0.76 (0.66, 0.87) -Physician density (per 1,000 population) 0.68 (0.58, 0.79) -II. Health service coverage% Of population with sustainable access to water and sanitation(for a 10% increase)0.67 (0.61, 0.73) 0.88 (0.82, 0.94)% Of births attended by skilled staff 0.28 (0.20, 0.39) -III. Medical products, vaccines and technology% Measles immunization coverage 0.55 (0.40, 0.74) -IV. Health financingTotal health expenditure per person (USD) 0.60 (0.55, 0.65) 0.84 (0.77, 0.92)Out-of-pocket expenditure on health (as a % of total health expenditure) 1.32 (1.04, 1.66) -Government health expenditure (USD) 0.53 (0.48, 0.58) -Private share of total health expenditure (%) 1.01 (1.00, 1.02) -V. Leadership and governanceCorruption index (log of) 0.18 (0.13, 0.23) 0.49 (0.36, 0.66)Demographic variablesFertility rate (average number of children per woman) 9.93 (6.96, 14.16) 2.85 (2.02, 4.00)Population growth value (annual %) 1.07 (0.89, 1.28) -Urban population value (annual %) 1.33 (1.17, 1.51) -Female labour force participation (%) 1.00 (0.99, 1.02) -- : Not selected in final modelCI: Confidence intervalA Risk Ratio below 1 corresponds to a protective variable.A Risk Ratio above 1 corresponds to a risk factorMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 9 of 107. Mills A: Vertical vs. Horizontal Health Programmes in Africa: Idealism,pragmatism, resources and efficiency. Social Science & Medicine 1983,17:1971-1981.8. Elzinga G: Vertical-Horizontal Synergy of the Health Workforce. Bulletin ofthe World Health Organization 2005, 83:242.9. Ooms G, Van Damme W, Baker BK, Zeitz P, Schrecker T: The ‘Diagonal’Approach to Global Fund financing: A cure for the broader malaise ofhealth systems? Globalization and Health 2008, 4:1.10. De Savigny D, Kasale H, Mbuya , Reid G: Fixing Health Systems (2ndEdition). 2008.11. Schnell CO, Reilly M, Rosling H, Peterson S, Skstrom AM: SocioeconomicDeterminants of Infant Mortality: A world-wide study of 152 low-,middle-, and high-income countries. Scandinavian Journal of Public Health2007, 35:288-297.12. UNICEF: The State of the World’s Children. New York, NY, USA; 2009.13. Marmot M: Achieving health equity: From root causes to fair outcomes.The Lancet 2007, 370:1153-1163.14. PEPFAR: Health System Strengthening. [http://www.pepfar.gov/strategy/ghi/134854.htm].15. International Transparency: Global Corruption Report. 2010 [http://www.transparency.org/], (accessed November 20, 2010)..16. Ko K, Samajdar A: Evaluation of International Corruption Indexes: Shouldwe believe them or not? The Social Science Journal 47:508-540.17. Littell RC: SAS for Mixed Models. SAS Institute Inc 2006.18. Murray CJL, Frenk J: A Framework for Assessing the Performance ofHealth Systems. Bulletin of the World Health Organization 2000, 78:717-731.19. International Transparency: Corruption Perception Index 2010. 2010.20. Paldam M: The Cross-Country Patterns of Corruption: Economics, cultureand the seesaw dynamics. European Journal of Political Economy 2002,18:215-240.21. UN: United Nations Convention Against Corruption. United Nations; 2004.22. Vian T: Review of Corruption in the Health Sector: Theory, methods andinterventions. Health Policy Plan 2008, 23:83-94.23. Botting MJ, Porbeni EO, Joffres MR, Johnston BC, Black RE, Mills EJ: Waterand Sanitation Infrastructure for Health: The impact of foreign aid.Global Health 2010, 6:12.24. Alvarez JL, Gil R, Hernandez V, Gil A: Factors Associated with MaternalMortality in Sub-Saharan Africa: An ecological study. BMC Public Health2009, 9:462.25. Piwoz E, Ross JS: Use of Population-Specific Infant Mortality Rates toInform Policy Decisions Regarding HIV and Infant Feeding. Journal ofNutrition 2005, 145:1113-1119.26. Leive A, Xu K: Coping with Out-of-pocket Health Payments: Empiricalevidence from 15 African countries. Bull World Health Organ 2008,86:849-856.27. Anand S, Barnighausen T: Human Resources and Health Outcomes: Cross-country econometric study. Lancet 2004, 364:1603-1609.28. El-Jardali F, Jamal D, Abdallah A, Kassak K: Human Resources for HealthPlanning and Management in the Eastern Mediterranean Region: Facts,gaps and forward thinking for research and policy. Human Resources forHealth 2007, 5:9-20.29. Hertz E, Hebert JR, Landon J: Social and Environmental Factors and LifeExpectancy, Infant Mortality and Maternal Mortality Rates: Results of across-national comparison. Social Science & Medicine 1994, 39:105-114.30. Kim K, Moody P: More Resources Better Health? A cross-nationalperspective. Social Science & Medicine 1992, 34:837-842.31. Kruppa K, Madhivanan P: Leveraging Human Capital to Reduce MaternalMortality in India: Enhanced public health system or public-privatepartnerships? Human Resources for Health 2009, 7:18-25.32. Robinson J, Wharrad H: Invisible Nursing: Exploring health outcomes at aglobal level. Relationships between infant and under-5 mortality ratesand the distribution of health professionals, GNP per capita, and femaleliteracy. Journal of Advanced Nursing 2000, 32:288-297.33. Adams O, Buchan J, DP MR: Human Resources for Health (internaldocument). Department of Health Services Geneva: World HealthOrganization; 2003.34. Zakir M, Wunnava P: Factors Affecting Infant Mortality Rates: Evidencefrom cross-sectional data. Applied Economic Letters 1999, 6:271-273.35. Black R, Morris S, Bryce J: Where and Why are 10 million children dyingevery year? Lancet 2003, 361:2226-2234.36. WHO: The World Health Report 2010: Health system financing: the pathto universal coverage. Geneva: WHO; 2010, (WHO ed.).37. Wakabi W: Uganda Steps up Efforts to Boost Male Circumcision. Lancet2010, 376:757-758.doi:10.1186/1744-8603-7-42Cite this article as: Muldoon et al.: Health system determinants ofinfant, child and maternal mortality: A cross-sectional study of UNmember countries. Globalization and Health 2011 7:42.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/submitMuldoon et al. Globalization and Health 2011, 7:42http://www.globalizationandhealth.com/content/7/1/42Page 10 of 10

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223211/manifest

Comment

Related Items