Financial Risk Protection, Decomposition and Inequality Analysis of Household Out-of-Pocket Health Payments

English: This research examines equity trends in financing health care through out-of-pocket payments (OOP) using South African Income and Expenditure Surveys for the periods 1995, 2000, 2005-06 and 2010-11. South Africa is interesting to examine for a variety of reasons. In 1994, South Africa removed user charges at public health facilities (clinics) for children aged below six years, pregnant and nursing mothers and the elderly (as long as they were not covered by any medical aid scheme) with the aim of increasing access to public health care facilities. The policy was extended to the entire population in 1996. These initiatives, even though they were targeted at promoting access, were also an effort on the part of policy makers to cushion households against the financial costs associated with the consumption of medical care – something that is likely to influence the distribution of household OOP. Whether, this indeed has been the case remains relatively unknown. Within the scope of the investigation, this thesis tries to answer three broad questions: (i) What is the incidence of catastrophic health care expenditures (CHE) arising from OOP health care financing in South Africa from 1995 to 2011? (ii) What are the factors influencing the incidence of CHE among male and female headed households? and (iii) Who pays for health care in South Africa? In investigating the incidence of catastrophic health expenditure, the research has employed two approaches, which are: the financial burden approach and the income approach – the income approach is derived from the equity measures of public finance where progressivity is the main concern, while the financial burden approach argues that the burden should be equally distributed across all households (see Carrin et al., 2009). Both approaches relate health payments incurred by households to households’ capacity (ability) to pay and not to households’ risks of illness, albeit with different definitions of the capacity (ability) to pay. The research has found that in 1995, around 0.03 percent of households incurred health expenses that are likely to force them to cut back on consumption of other basic needs, while for the years 2000, 2005-06 and 2010-11, the incidence is 0.06 percent, 0.09 percent and 0.07 percent, respectively. Given such a low incidence of CHE, the research evaluated the utilisation of health care facilities by households when confronted with illness. This was only done for the year 1995, as it is only year in which data was collected on the illness status of each household member, whether or not they consulted when ill and where they consulted. The results suggest that a negligible percentage of households did not seek treatment when ill. Of those who consulted, it was found that a relatively higher percentage sought treatment in public health care facilities (0.21 percent) than in private facilities (0.13 percent). Having established the incidence of CHE, the second analysis examined the factors associated with CHE and then decomposed the difference between male-headed and female-headed households to establish whether the gap between the two groups had widened or narrowed. The results suggest that the gender gap in the incidence of CHE narrowed by 0.4 percent between 1995 and 2010-11. This reduction in the gender gap is attributable to education, access to piped water and residing in urban areas. Across the different surveys (as well as over the entire time period) education, having access to piped water and residing in urban areas narrowed the gender gap. These results are consistent with existing evidence documenting the important role played by access to basic amenities, such as water and sanitation, as well as human capital (education), in explaining gendered inequalities in health care. Finally, the research examined the distribution of health payments relative to income, focusing on who incurs OOP for their health care needs to establish OOP concentration and quantify its magnitude. The levels of concentation were compared over time, and decomposed to see if it was possible to attribute changes in social determinants of health to the level of concentration in OOP payments for health care. In general, health care payments are concentrated among non-poor households, suggesting that there is progressivity in health care financing, at least as it pertains to OOP. Such results are corroborated by the corresponding concentration indices. When the analysis occurs across the 15-year time period from 1995 to 2010-11, the research finds that changing inequalities across age groups, racial groups, education (particularly completion of secondary education), well-being quintiles and type of toilet used, as well as water source for drinking, explained changes in OOP concentration. It was also found that changing elasticities with respect to OOP payments also play a crucial role in explaining differences over time. Overall, most of the changes in OOP payment inequality are attributable to inequality in the social determinants. Sepedi: Dinyakišišo tše di lekola mekgwa ya tekatekano go thekga ka ditšhelete tlhokomelo ya maphelo ka go šomiša ditefelo go tšwa ka potleng (OOP) ka go šomiša Diphatišišo tša ka Afrika Borwa tša Letseno le Ditshenyegelo tša mengwaga ya 1995, 2000, 2005-06 le 2010-11. Afrika Borwa ke naga ye e gogago šedi yeo e ka lekolwago ka mabaka a mehutahuta. Ka 1994, Afrika Borwa e tlošitše ditefišo tša bašomiši ka mafelong a setšhaba a tša maphelo (ditleliniking) go bana ba mengwaga ya ka fase ga ye tshela, go bomma bao ba imilego le bao ba nyantšhago bana letswele le go batšofadi (ge fela e le gore ga ba na setlamo sefe goba sefe sa kalafo) ka maikemišetšo a go oketša phihlelelo go mafelo a setšhaba a tlhokomelo ya maphelo. Molawana wo o ile wa katološetšwa go setšhaba ka moka ka 1996. Matsapa a, le ge e le gore a be a phethagatšwa ka nepo ya go tšwetša pele phihlelelo go maphelo, gape a be e le nepo ya bahlami ba melao go fokoletša boima go malapa kgahlanong le ditshenyegelo tša ditšhelete tše di amanago le tšhomišo ya tlhokomelo ya maphelo - e lego seo go nago le kgonagalo ya gore se ka tšwela pele go huetša phatlalatšo OOP ya ka malapeng. Go sa kgathale, se ka nnete ebile seemo seo se bego se dula se sa tsebje ka nnete. Ka gare ga mollwane wa dinyakišišo, thesese ye e leka go araba dipotšišo tše tharo tša kakaretšo: (i) Naa tiragalo ya ditshenyegelo tša tlhokomelo ya mapehlo tša go šiiša ke efe (CHE) yeo e bakwago ke thekgo ya ditšhelete tša tlhokomelo ya maphelo ya OOP ka Afrika Borwa go thoma ka 1995 go fihla ka 2011? (ii) Naa ke mabaka afe ao a huetšago tiragalo ya CHE gareng ga malapa ao a laolwago ke banna le basadi? le (iii) Naa ke bomang bao ba lefelago tlhokomelo ya maphelo ka Afrika Borwa? Ge go nyakišišwa tiragalo ya tshenyegelo ya maphelo ye e šiišago, dinyakišišo di dirišitše mekgwa ye mebedi, ye e lego: mokgwa wa morwalo wa ditšhelete le mokgwa wa letseno - mokgwa wa letseno o tšwa go magato a tekatekano a ditšhelete tša setšhaba fao tšwetšopele e lego bothata bjo bogolo, mola e le gore mokgwa wa morwalo wa ditšhelete o bolela gore morwalo o swanetše go abaganywa ka go lekana go ralala le malapa ka moka (bona Carrin et al., 2009). Bobedi mekgwa ye e amana le ditefelo tša maphelo tšeo di lefelwago ke malapa ka bokgoni (go kgona) bja malapa bja go lefa le bja go se kgone go lefa ga malapa go dikotsi tša malwetši, le ge e le gore go na le ditlhalošo tše di fapanego tša bokgoni (go kgona) bja go lefa. Dinyakišišo di hweditše gore ka 1995, dipersente tše di ka bago tše 0.03 tša malapa di bile le ditshenyegelo tša mabapi le maphelo tšeo go nago le kgonagalo ya gore di ka ba gapeletša gore ba fokotše tšhomišo ya dinyakwa tše dingwe tša motheo, mola go mengwaga ya bo 2000, 2005-06 le 2010-11, tiragalo ye e le dipersente tše 0.06, dipersente tše 0.09 le dipersente tše 0.07, ka go latelana. Ka lebaka la tiragalo ye nnyane ya CHE, dinyakišišo di sekasekile tšhomišo ya dinolofatši tša tlhokomelo ya maphelo ka malapa ge a lebane le malwetši. Se se dirilwe fela ka ngwaga wa 1995, bjalo ka ge e le ngwaga fela wo ka ona tshedimošo e kgobokeditšwego mabapi le maemo a malwetši a leloko le lengwe le le lengwe la lapa, ba ka be ba boledišitšwe ge ba babja goba aowa le ge ba be ba boledišwa. Dipoelo di šišinya gore persente ye nnyane ya malapa ga se ya nyaka kalafo ge e babja. Go bao go boledišanwego le bona, go hweditšwe gore persente ya godimo kudu e nyakile kalafo ka mafelong a setšhaba a tlhokomelo ya maphelo (dipersente tše 0.21) go feta ka mafelong a phraebete (dipersente tše 0.13). Ka ge go hweditšwe tiragalo ya CHE, tshekatsheko ya bobedi e lekotše mabaka ao a amanago le CHE gomme gwa abja phapano magareng ga malapa ao a laolwago ke banna le ao a laolwago ke basadi ka nepo ya go tseba ge eba sekgoba sa magareng ga dihlopha tše ka bobedi se oketšegile goba se fokotšegile. Dipoelo di šišinya gore sekgoba sa mabapi le bong ka tiragalong ya CHE se fokotšegile ka dipersente tše 0.4 magareng ga 1995 le 2010-11. Phokotšego ye ya sekgoba sa tša bong e bakwa ke thuto, phihlelelo go meetse a thepe le go dula metsesetoropong. Go ralala diphatišišo tše di fapafapanego (gammogo le mo lebakeng le ka moka) thuto, go ba le phihlelelo go meetse a thepe le go dula metsesetoropong go fokoditše sekgoba sa tša bong. Dipoelo tše di sepelelana le bohlatse bjo bo lego gona bjo bo laetšago tema ye bohlokwa ye e ralokilwego ke phihlelelo go dinyakwa tša motheo, tša go swana le meetse le kelelatšhila, gammogo le letlotlo la batho (thuto), ge go hlalošwa tlhokego ya tekatekano ya bong ka go tlhokomelo ya maphelo. Sa mafelelo, dinyakišišo di lekotše kabo ya ditefelo tša maphelo ge go bapetšwa le letseno, go lebeletšwe kudu gore ke bomang bao ba itemogelago OOP go dinyakwa tša bona tša maphelo mabapi le go tlala ga OOP le go tseba bontši bja yona. Maemo a go tlala fa a bapeditšwe ge nako e dutše e sepela, gomme a aroganywa go bona ge eba go be go kgonagala go bona gore diphetogo tše tša leago tša maphelo di bakilwe ke go tlala ga ditefelo tša OOP tša tlhokomelo ya maphelo. Ka kakaretšo, ditefelo tša tlhokomelo ya maphelo di tletše ka malapeng ao a hlokago, gomme se se šišinya gore go na le kgatelopele ka go thekgo ya ditšhelete tša tlhokomelo ya maphelo, bonnyane ge e amana le OOP. Ka kakaretšo, ditefelo tša tlhokomelo ya maphelo di tletše ka malapeng ao a hlokago, gomme se se šišinya gore go na le kgatelopele ka go thekgo ya ditšhelete tša tlhokomelo ya maphelo, bonnyane ge e amana le OOP. Ge tshekatsheko e direga go ralala le lebaka la mengwaga ye 15 go thoma ka 1995 go fihla ka 2010-11, dinyakišišo di hwetša gore tlhokego ya tekatekano ye e fetogago go ralala le dihlopha tša mengwaga, le dihlopha tša merafe, thuto (kudukudu go phetha+D8 ga dithuto tša sekolo seo se phagamego), dikhwinthaele tša go phela gabotse le mehuta ya dintlwana tša boithomelo tše di šomišwago, gammogo le methopo ya meetse a go nwewa, ditlhalošo tše di hlalošwago ka go tlaleng ga OOP. Go hweditšwe gape gore go fetogafetoga ga diphapano mabapi le ditefelo tša OOP le gona go raloka tema ye bohlokwa go hlalošeng ga diphapano mo nakong ye telele. Ka kakaretšo, bontši bja diphetogo ka go tlhokego ya tekatekano go ditefelo tša OOP go bonwa e bakwa ke tlhokego ya tekatekano ka go seemo sa setšhaba. Afrikaans: Hierdie navorsing ondersoek billikheidstendense in die finansiering van gesondheidsorg deur middel van kontantuitgawes (KU's) aan die hand van Suid-Afrikaanse inkomste- en uitgawe-opnames vir die periodes 1995, 2000, 2005–06 en 2010–11. Suid-Afrika is is om verskeie redes interessant om te ondersoek. In 1994 het Suid-Afrika gebruikersfooie by openbaregesondheidsfasiliteite (klinieke) afgeskaf vir kinders onder ses jaar, swanger en borsvoedende moeders en bejaardes (mits hulle nie deur enige mediese hulpskema gedek was nie) met die doel om toegang tot openbaregesondheidfasiliteite te verhoog. Binne die bestek van die ondersoek poog hierdie proefskrif om drie breë vrae te beantwoord: (i) Wat is die voorkoms van katastrofiese gesondheidsorguitgawes (KGU's) wat voortspruit uit KU-gesondheidsorgfinansiering in Suid-Afrika van 1995 tot 2011? Laastens het die navorsing die verspreiding van gesondheidsbetalings relatief tot inkomste ondersoek, met die fokus op wie KU's vir hul gesondheidsorgbehoeftes aangaan om KU-konsentrasie vas te stel en die omvang daarvan te kwantifiseer. Die vlakke van konsentrasie is oor tyd vergelyk, en ontleed om te sien of dit moontlik was om veranderinge in sosiale determinante van gesondheid toe te skryf aan die vlak van konsentrasie in KU's vir gesondheidsorg. Oor die algemeen is gesondheidsorgbetalings gekonsentreer onder nie-arm huishoudings, wat daarop dui dat daar progressiwiteit in gesondheidsorgfinansiering is, ten minste wat USB's betref. Wanneer die ontleding oor die 15-jaar tydperk van 1995 tot 2010-11 plaasvind, bevind die navorsing dat veranderende ongelykhede tussen ouderdomsgroepe, rassegroepe, in onderwys (veral voltooiing van sekondêre onderwys), welstandskwintiele en tipe toilet wat gebruik word, sowel as bron van drinkwater, die veranderinge in KU-konsentrasie verklaar. Daar is ook bevind dat veranderende elastisiteite met betrekking tot KU's ook 'n deurslaggewende rol speel in die verklaar van verskille oor tyd. In die geheel gesien, is die meeste van die veranderinge in KU-ongelykheid toe te skryf aan ongelykheid in die maatskaplike determinante.

Grantor dc:publisher

University of Pretoria

Year dc:date.issued

2020

Advisor dc:contributor.advisor

• Koch, Steven F.

• UCTD

Statement dc:rights

• © 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.

Language dc:language.iso

en

Dc Identifier Other

A2020

OAI identifier oai:identifier

oai:repository.up.ac.za:2263/74559