Links to Exploration step
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Offspring schooling associated with increased parental survival in rural KwaZulu-Natal, South Africa</title><author><name sortKey="De Neve, Jan Walter" sort="De Neve, Jan Walter" uniqKey="De Neve J" first="Jan-Walter" last="De Neve">Jan-Walter De Neve</name><affiliation><nlm:aff id="aff1">Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 665 Huntington Avenue, Boston, MA 02115, United States</nlm:aff></affiliation><affiliation><nlm:aff id="aff2">Institute of Public Health, Heidelberg University, Im Neuenheimer Feld 130.3, Heidelberg 69120, Germany</nlm:aff></affiliation></author><author><name sortKey="Harling, Guy" sort="Harling, Guy" uniqKey="Harling G" first="Guy" last="Harling">Guy Harling</name><affiliation><nlm:aff id="aff1">Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 665 Huntington Avenue, Boston, MA 02115, United States</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">Research Department of Infection and Population Health, University College London, off Caper Street, London, WC1E 6JB, United Kingdom</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Africa Health Research Institute, University of KwaZulu-Natal, Somkhele, 3935, South Africa</nlm:aff></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">28153751</idno><idno type="pmc">5322823</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5322823</idno><idno type="RBID">PMC:5322823</idno><idno type="doi">10.1016/j.socscimed.2017.01.015</idno><date when="2017">2017</date><idno type="wicri:Area/Pmc/Corpus">002C81</idno><idno type="wicri:explorRef" wicri:stream="Pmc" wicri:step="Corpus" wicri:corpus="PMC">002C81</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Offspring schooling associated with increased parental survival in rural KwaZulu-Natal, South Africa</title><author><name sortKey="De Neve, Jan Walter" sort="De Neve, Jan Walter" uniqKey="De Neve J" first="Jan-Walter" last="De Neve">Jan-Walter De Neve</name><affiliation><nlm:aff id="aff1">Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 665 Huntington Avenue, Boston, MA 02115, United States</nlm:aff></affiliation><affiliation><nlm:aff id="aff2">Institute of Public Health, Heidelberg University, Im Neuenheimer Feld 130.3, Heidelberg 69120, Germany</nlm:aff></affiliation></author><author><name sortKey="Harling, Guy" sort="Harling, Guy" uniqKey="Harling G" first="Guy" last="Harling">Guy Harling</name><affiliation><nlm:aff id="aff1">Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 665 Huntington Avenue, Boston, MA 02115, United States</nlm:aff></affiliation><affiliation><nlm:aff id="aff3">Research Department of Infection and Population Health, University College London, off Caper Street, London, WC1E 6JB, United Kingdom</nlm:aff></affiliation><affiliation><nlm:aff id="aff4">Africa Health Research Institute, University of KwaZulu-Natal, Somkhele, 3935, South Africa</nlm:aff></affiliation></author></analytic><series><title level="j">Social Science & Medicine (1982)</title><idno type="ISSN">0277-9536</idno><idno type="eISSN">1873-5347</idno><imprint><date when="2017">2017</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><sec><title>Background</title><p>Investing in offspring's human capital has been suggested as an effective strategy for parents to improve their living conditions at older ages. A few studies have assessed the role of children's schooling in parental survival in high-income countries, but none have considered lower-resource settings with limited public wealth transfers and high adult mortality.</p></sec><sec><title>Methods</title><p>We followed 17,789 parents between January 2003 and August 2015 in a large population-based open cohort in rural KwaZulu-Natal, South Africa. We used Cox proportional hazards models to investigate the association between offspring's schooling and time to parental death. We assessed the association separately by parental sex and for four cause of death groups.</p></sec><sec><title>Results</title><p>A one year increase in offspring's schooling attainment was associated with a 5% decline in the hazard of maternal death (adjusted Hazard Ratio [aHR]: 0.95, 95%CI: 0.94–0.97) and a 6% decline in the hazard of paternal death (aHR: 0.94, 95%CI: 0.92–0.96), adjusting for a wide range of demographic and socio-economic variables of the parent and their children. Among mothers, the association was strongest for communicable, maternal, perinatal and nutritional conditions (aHR: 0.87, 95%CI: 0.82–0.92) and AIDS and tuberculosis (aHR: 0.92, 95%CI: 0.89–0.96), and weakest for injuries. Among fathers, the association was strongest for injuries (aHR: 0.87, 95%CI: 0.79–0.95) and AIDS and tuberculosis (aHR: 0.92, 95%CI: 0.89–0.96), and weakest for non-communicable diseases.</p></sec><sec><title>Conclusion</title><p>Higher levels of schooling in offspring are associated with increased parental survival in rural South Africa, particularly for mothers at risk of communicable disease mortality and fathers at risk of injury mortality. Offspring's human capital may be an important factor for health disparities, particularly in lower-resource settings.</p></sec></div></front><back><div1 type="bibliography"><listBibl><biblStruct><analytic><author><name sortKey="Adamchak, D J" uniqKey="Adamchak D">D.J. Adamchak</name></author><author><name sortKey="Wilson, A O" uniqKey="Wilson A">A.O. Wilson</name></author><author><name sortKey="Nyanguru, A" uniqKey="Nyanguru A">A. Nyanguru</name></author><author><name sortKey="Hampson, J" uniqKey="Hampson J">J. Hampson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Andersen, P K" uniqKey="Andersen P">P.K. Andersen</name></author><author><name sortKey="Geskus, R B" uniqKey="Geskus R">R.B. Geskus</name></author><author><name sortKey="De Witte, T" uniqKey="De Witte T">T. de Witte</name></author><author><name sortKey="Putter, H" uniqKey="Putter H">H. Putter</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ansell, N" uniqKey="Ansell N">N. Ansell</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Aslam, M" uniqKey="Aslam M">M. Aslam</name></author><author><name sortKey="Kingdon, G G" uniqKey="Kingdon G">G.G. Kingdon</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Barouni, M" uniqKey="Barouni M">M. Barouni</name></author><author><name sortKey="Broecke, S" uniqKey="Broecke S">S. Broecke</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Basu, A M" uniqKey="Basu A">A.M. Basu</name></author><author><name sortKey="Stephenson, R" uniqKey="Stephenson R">R. Stephenson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bauman, L J" uniqKey="Bauman L">L.J. Bauman</name></author><author><name sortKey="Foster, G" uniqKey="Foster G">G. Foster</name></author><author><name sortKey="Johnson Silver, E" uniqKey="Johnson Silver E">E. Johnson Silver</name></author><author><name sortKey="Berman, R" uniqKey="Berman R">R. Berman</name></author><author><name sortKey="Gamble, I" uniqKey="Gamble I">I. Gamble</name></author><author><name sortKey="Muchaneta, L" uniqKey="Muchaneta L">L. Muchaneta</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Berkman, L F" uniqKey="Berkman L">L.F. Berkman</name></author><author><name sortKey="Glass, T" uniqKey="Glass T">T. Glass</name></author><author><name sortKey="Brissette, I" uniqKey="Brissette I">I. Brissette</name></author><author><name sortKey="Seeman, T E" uniqKey="Seeman T">T.E. Seeman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Berkman, L F" uniqKey="Berkman L">L.F. Berkman</name></author><author><name sortKey="Kawachi, I" uniqKey="Kawachi I">I. Kawachi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bolin, K" uniqKey="Bolin K">K. Bolin</name></author><author><name sortKey="Lindgren, B" uniqKey="Lindgren B">B. Lindgren</name></author><author><name sortKey="Lundborg, P" uniqKey="Lundborg P">P. Lundborg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bongaarts, J" uniqKey="Bongaarts J">J. Bongaarts</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bor, J" uniqKey="Bor J">J. Bor</name></author><author><name sortKey="Herbst, A J" uniqKey="Herbst A">A.J. Herbst</name></author><author><name sortKey="Newell, M L" uniqKey="Newell M">M.L. Newell</name></author><author><name sortKey="Barnighausen, T" uniqKey="Barnighausen T">T. Barnighausen</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bor, J" uniqKey="Bor J">J. Bor</name></author><author><name sortKey="Tanser, F" uniqKey="Tanser F">F. Tanser</name></author><author><name sortKey="Newell, M L" uniqKey="Newell M">M.L. Newell</name></author><author><name sortKey="Barnighausen, T" uniqKey="Barnighausen T">T. Barnighausen</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Branson, N" uniqKey="Branson N">N. Branson</name></author><author><name sortKey="Hofmeyr, C" uniqKey="Hofmeyr C">C. Hofmeyr</name></author><author><name sortKey="Lam, D" uniqKey="Lam D">D. Lam</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bravo, J" uniqKey="Bravo J">J. Bravo</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cai, F" uniqKey="Cai F">F. Cai</name></author><author><name sortKey="Giles, J" uniqKey="Giles J">J. Giles</name></author><author><name sortKey="Meng, X" uniqKey="Meng X">X. Meng</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Case, A" uniqKey="Case A">A. Case</name></author><author><name sortKey="Deaton, A" uniqKey="Deaton A">A. Deaton</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Clay, D C" uniqKey="Clay D">D.C. Clay</name></author><author><name sortKey="Vander Haar, J E" uniqKey="Vander Haar J">J.E. Vander Haar</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Coovadia, H" uniqKey="Coovadia H">H. Coovadia</name></author><author><name sortKey="Jewkes, R" uniqKey="Jewkes R">R. Jewkes</name></author><author><name sortKey="Barron, P" uniqKey="Barron P">P. Barron</name></author><author><name sortKey="Sanders, D" uniqKey="Sanders D">D. Sanders</name></author><author><name sortKey="Mcintyre, D" uniqKey="Mcintyre D">D. McIntyre</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Crews, F" uniqKey="Crews F">F. Crews</name></author><author><name sortKey="He, J" uniqKey="He J">J. He</name></author><author><name sortKey="Hodge, C" uniqKey="Hodge C">C. Hodge</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cunha, F" uniqKey="Cunha F">F. Cunha</name></author><author><name sortKey="Heckman, J" uniqKey="Heckman J">J. Heckman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cutler, D M" uniqKey="Cutler D">D.M. Cutler</name></author><author><name sortKey="Lleras Muney, A" uniqKey="Lleras Muney A">A. Lleras-Muney</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Dahl, R E" uniqKey="Dahl R">R.E. Dahl</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="De Neve, J W" uniqKey="De Neve J">J.-W. De Neve</name></author><author><name sortKey="Fink, G" uniqKey="Fink G">G. Fink</name></author><author><name sortKey="Subramanian, S V" uniqKey="Subramanian S">S.V. Subramanian</name></author><author><name sortKey="Moyo, S" uniqKey="Moyo S">S. Moyo</name></author><author><name sortKey="Bor, J" uniqKey="Bor J">J. Bor</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ding, P" uniqKey="Ding P">P. Ding</name></author><author><name sortKey="Vanderweele, T J" uniqKey="Vanderweele T">T.J. VanderWeele</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Downey, D B" uniqKey="Downey D">D.B. Downey</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Epstein, E" uniqKey="Epstein E">E. Epstein</name></author><author><name sortKey="Guttman, R" uniqKey="Guttman R">R. Guttman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Evans, R" uniqKey="Evans R">R. Evans</name></author><author><name sortKey="Becker, S" uniqKey="Becker S">S. Becker</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Field, E" uniqKey="Field E">E. Field</name></author><author><name sortKey="De La Roca, J" uniqKey="De La Roca J">J. de la Roca</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Frankenberg, E" uniqKey="Frankenberg E">E. Frankenberg</name></author><author><name sortKey="Lillard, L" uniqKey="Lillard L">L. Lillard</name></author><author><name sortKey="Willis, R J" uniqKey="Willis R">R.J. Willis</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Friedman, E M" uniqKey="Friedman E">E.M. Friedman</name></author><author><name sortKey="Mare, R D" uniqKey="Mare R">R.D. Mare</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Fritzell, J" uniqKey="Fritzell J">J. Fritzell</name></author><author><name sortKey="Lennartsson, C" uniqKey="Lennartsson C">C. Lennartsson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Frost, M B" uniqKey="Frost M">M.B. Frost</name></author><author><name sortKey="Forste, R" uniqKey="Forste R">R. Forste</name></author><author><name sortKey="Haas, D W" uniqKey="Haas D">D.W. Haas</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Gallant, M" uniqKey="Gallant M">M. Gallant</name></author><author><name sortKey="Maticka Tyndale, E" uniqKey="Maticka Tyndale E">E. Maticka-Tyndale</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Giles, J" uniqKey="Giles J">J. Giles</name></author><author><name sortKey="Wang, D" uniqKey="Wang D">D. Wang</name></author><author><name sortKey="Zhao, C" uniqKey="Zhao C">C. Zhao</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Glewwe, P" uniqKey="Glewwe P">P. Glewwe</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Government Of South Africa" uniqKey="Government Of South Africa">Government of South Africa</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Grambsch, P M" uniqKey="Grambsch P">P.M. Grambsch</name></author><author><name sortKey="Therneau, T M" uniqKey="Therneau T">T.M. Therneau</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Grepin, K A" uniqKey="Grepin K">K.A. Grépin</name></author><author><name sortKey="Bharadwaj, P" uniqKey="Bharadwaj P">P. Bharadwaj</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Grossman, M" uniqKey="Grossman M">M. Grossman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Herbst, A J" uniqKey="Herbst A">A.J. Herbst</name></author><author><name sortKey="Mafojane, T" uniqKey="Mafojane T">T. Mafojane</name></author><author><name sortKey="Newell, M L" uniqKey="Newell M">M.L. Newell</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hosegood, V" uniqKey="Hosegood V">V. Hosegood</name></author><author><name sortKey="Timaeus, I" uniqKey="Timaeus I">I. Timaeus</name></author></analytic></biblStruct><biblStruct></biblStruct><biblStruct><analytic><author><name sortKey="Kaufman, J S" uniqKey="Kaufman J">J.S. Kaufman</name></author><author><name sortKey="Cooper, R S" uniqKey="Cooper R">R.S. Cooper</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Knodel, J" uniqKey="Knodel J">J. Knodel</name></author><author><name sortKey="Friedman, J" uniqKey="Friedman J">J. Friedman</name></author><author><name sortKey="Anh, T S" uniqKey="Anh T">T.S. Anh</name></author><author><name sortKey="Cuong, B T" uniqKey="Cuong B">B.T. Cuong</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Kohler, I V" uniqKey="Kohler I">I.V. Kohler</name></author><author><name sortKey="Kohler, H P" uniqKey="Kohler H">H.P. Kohler</name></author><author><name sortKey="Anglewicz, P" uniqKey="Anglewicz P">P. Anglewicz</name></author><author><name sortKey="Behrman, J R" uniqKey="Behrman J">J.R. Behrman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Krieger, N" uniqKey="Krieger N">N. Krieger</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lau, B" uniqKey="Lau B">B. Lau</name></author><author><name sortKey="Cole, S R" uniqKey="Cole S">S.R. Cole</name></author><author><name sortKey="Gange, S J" uniqKey="Gange S">S.J. Gange</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lennartsson, C" uniqKey="Lennartsson C">C. Lennartsson</name></author><author><name sortKey="Silverstein, M" uniqKey="Silverstein M">M. Silverstein</name></author><author><name sortKey="Fritzell, J" uniqKey="Fritzell J">J. Fritzell</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lundborg, P" uniqKey="Lundborg P">P. Lundborg</name></author><author><name sortKey="Majlesi, K" uniqKey="Majlesi K">K. Majlesi</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Marmot, M G" uniqKey="Marmot M">M.G. Marmot</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Mcgarry, K" uniqKey="Mcgarry K">K. McGarry</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Mincer, J" uniqKey="Mincer J">J. Mincer</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Morris, S S" uniqKey="Morris S">S.S. Morris</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pepfar" uniqKey="Pepfar">PEPFAR</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pettifor, A" uniqKey="Pettifor A">A. Pettifor</name></author><author><name sortKey="Macphail, C" uniqKey="Macphail C">C. MacPhail</name></author><author><name sortKey="Hughes, J P" uniqKey="Hughes J">J.P. Hughes</name></author><author><name sortKey="Selin, A" uniqKey="Selin A">A. Selin</name></author><author><name sortKey="Wang, J" uniqKey="Wang J">J. Wang</name></author><author><name sortKey="G Mez Olive, F X" uniqKey="G Mez Olive F">F.X. Gómez-Olivé</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Psacharopoulos, G" uniqKey="Psacharopoulos G">G. Psacharopoulos</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Reniers, G" uniqKey="Reniers G">G. Reniers</name></author><author><name sortKey="Blom, S" uniqKey="Blom S">S. Blom</name></author><author><name sortKey="Calvert, C" uniqKey="Calvert C">C. Calvert</name></author><author><name sortKey="Martin Onraet, A" uniqKey="Martin Onraet A">A. Martin-Onraet</name></author><author><name sortKey="Herbst, A J" uniqKey="Herbst A">A.J. Herbst</name></author><author><name sortKey="Eaton, J W" uniqKey="Eaton J">J.W. Eaton</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Republic Of South Africa" uniqKey="Republic Of South Africa">Republic of South Africa</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Rowa Dewar, N" uniqKey="Rowa Dewar N">N. Rowa-Dewar</name></author><author><name sortKey="Amos, A" uniqKey="Amos A">A. Amos</name></author><author><name sortKey="Cunningham Burley, S" uniqKey="Cunningham Burley S">S. Cunningham-Burley</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Saad, P M" uniqKey="Saad P">P.M. Saad</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sabater, A" uniqKey="Sabater A">A. Sabater</name></author><author><name sortKey="Graham, E" uniqKey="Graham E">E. Graham</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sartorius, B" uniqKey="Sartorius B">B. Sartorius</name></author><author><name sortKey="Kahn, K" uniqKey="Kahn K">K. Kahn</name></author><author><name sortKey="Collinson, M A" uniqKey="Collinson M">M.A. Collinson</name></author><author><name sortKey="Sartorius, K" uniqKey="Sartorius K">K. Sartorius</name></author><author><name sortKey="Tollman, S M" uniqKey="Tollman S">S.M. Tollman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Schultz, T P" uniqKey="Schultz T">T.P. Schultz</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Steinberg, L" uniqKey="Steinberg L">L. Steinberg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sun, F" uniqKey="Sun F">F. Sun</name></author><author><name sortKey="Sebastiani, P" uniqKey="Sebastiani P">P. Sebastiani</name></author><author><name sortKey="Schupf, N" uniqKey="Schupf N">N. Schupf</name></author><author><name sortKey="Bae, H" uniqKey="Bae H">H. Bae</name></author><author><name sortKey="Andersen, S L" uniqKey="Andersen S">S.L. Andersen</name></author><author><name sortKey="Mcintosh, A" uniqKey="Mcintosh A">A. McIntosh</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Tamakoshi, A" uniqKey="Tamakoshi A">A. Tamakoshi</name></author><author><name sortKey="Tamakoshi, K" uniqKey="Tamakoshi K">K. Tamakoshi</name></author><author><name sortKey="Lin, Y" uniqKey="Lin Y">Y. Lin</name></author><author><name sortKey="Mikami, H" uniqKey="Mikami H">H. Mikami</name></author><author><name sortKey="Inaba, Y" uniqKey="Inaba Y">Y. Inaba</name></author><author><name sortKey="Yagyu, K" uniqKey="Yagyu K">K. Yagyu</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Tanser, F" uniqKey="Tanser F">F. Tanser</name></author><author><name sortKey="Hosegood, V" uniqKey="Hosegood V">V. Hosegood</name></author><author><name sortKey="Barnighausen, T" uniqKey="Barnighausen T">T. Barnighausen</name></author><author><name sortKey="Herbst, K" uniqKey="Herbst K">K. Herbst</name></author><author><name sortKey="Nyirenda, M" uniqKey="Nyirenda M">M. Nyirenda</name></author><author><name sortKey="Muhwava, W" uniqKey="Muhwava W">W. Muhwava</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Torssander, J" uniqKey="Torssander J">J. Torssander</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Torssander, J" uniqKey="Torssander J">J. Torssander</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Unesco Institute For Statistics" uniqKey="Unesco Institute For Statistics">UNESCO Institute for Statistics</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Vanderweele, T" uniqKey="Vanderweele T">T. VanderWeele</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Wilde, E T" uniqKey="Wilde E">E.T. Wilde</name></author><author><name sortKey="Batchelder, L" uniqKey="Batchelder L">L. Batchelder</name></author><author><name sortKey="Ellwood, D" uniqKey="Ellwood D">D. Ellwood</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Yahirun, J J" uniqKey="Yahirun J">J.J. Yahirun</name></author><author><name sortKey="Sheehan, C M" uniqKey="Sheehan C">C.M. Sheehan</name></author><author><name sortKey="Hayward, M D" uniqKey="Hayward M">M.D. Hayward</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Yang, L" uniqKey="Yang L">L. Yang</name></author><author><name sortKey="Martikainen, P" uniqKey="Martikainen P">P. Martikainen</name></author><author><name sortKey="Silventoinen, K" uniqKey="Silventoinen K">K. Silventoinen</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Zimmer, Z" uniqKey="Zimmer Z">Z. Zimmer</name></author><author><name sortKey="Martin, L G" uniqKey="Martin L">L.G. Martin</name></author><author><name sortKey="Ofstedal, M B" uniqKey="Ofstedal M">M.B. Ofstedal</name></author><author><name sortKey="Chuang, Y L" uniqKey="Chuang Y">Y.L. Chuang</name></author></analytic></biblStruct></listBibl></div1></back></TEI><pmc article-type="research-article"><pmc-dir>properties open_access</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-ta">Soc Sci Med</journal-id><journal-id journal-id-type="iso-abbrev">Soc Sci Med</journal-id><journal-title-group><journal-title>Social Science & Medicine (1982)</journal-title></journal-title-group><issn pub-type="ppub">0277-9536</issn><issn pub-type="epub">1873-5347</issn><publisher><publisher-name>Pergamon</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="pmid">28153751</article-id><article-id pub-id-type="pmc">5322823</article-id><article-id pub-id-type="publisher-id">S0277-9536(17)30015-1</article-id><article-id pub-id-type="doi">10.1016/j.socscimed.2017.01.015</article-id><article-categories><subj-group subj-group-type="heading"><subject>Article</subject></subj-group></article-categories><title-group><article-title>Offspring schooling associated with increased parental survival in rural KwaZulu-Natal, South Africa</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>De Neve</surname><given-names>Jan-Walter</given-names></name><email>jwdeneve@mail.harvard.edu</email><xref rid="aff1" ref-type="aff">a</xref><xref rid="aff2" ref-type="aff">b</xref><xref rid="cor1" ref-type="corresp">∗</xref></contrib><contrib contrib-type="author"><name><surname>Harling</surname><given-names>Guy</given-names></name><xref rid="aff1" ref-type="aff">a</xref><xref rid="aff3" ref-type="aff">c</xref><xref rid="aff4" ref-type="aff">d</xref></contrib></contrib-group><aff id="aff1"><label>a</label>Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 665 Huntington Avenue, Boston, MA 02115, United States</aff><aff id="aff2"><label>b</label>Institute of Public Health, Heidelberg University, Im Neuenheimer Feld 130.3, Heidelberg 69120, Germany</aff><aff id="aff3"><label>c</label>Research Department of Infection and Population Health, University College London, off Caper Street, London, WC1E 6JB, United Kingdom</aff><aff id="aff4"><label>d</label>Africa Health Research Institute, University of KwaZulu-Natal, Somkhele, 3935, South Africa</aff><author-notes><corresp id="cor1"><label>∗</label>Corresponding author. <email>jwdeneve@mail.harvard.edu</email></corresp></author-notes><pub-date pub-type="pmc-release"><day>1</day><month>3</month><year>2017</year></pub-date><pmc-comment> PMC Release delay is 0 months and 0 days and was based on .</pmc-comment>
<pub-date pub-type="ppub"><month>3</month><year>2017</year></pub-date><volume>176</volume><fpage>149</fpage><lpage>157</lpage><history><date date-type="received"><day>9</day><month>8</month><year>2016</year></date><date date-type="rev-recd"><day>9</day><month>1</month><year>2017</year></date><date date-type="accepted"><day>11</day><month>1</month><year>2017</year></date></history><permissions><copyright-statement>© 2017 The Authors</copyright-statement><copyright-year>2017</copyright-year><license license-type="CC BY" xlink:href="http://creativecommons.org/licenses/by/4.0/"><license-p>This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).</license-p></license></permissions><abstract id="abs0010"><sec><title>Background</title><p>Investing in offspring's human capital has been suggested as an effective strategy for parents to improve their living conditions at older ages. A few studies have assessed the role of children's schooling in parental survival in high-income countries, but none have considered lower-resource settings with limited public wealth transfers and high adult mortality.</p></sec><sec><title>Methods</title><p>We followed 17,789 parents between January 2003 and August 2015 in a large population-based open cohort in rural KwaZulu-Natal, South Africa. We used Cox proportional hazards models to investigate the association between offspring's schooling and time to parental death. We assessed the association separately by parental sex and for four cause of death groups.</p></sec><sec><title>Results</title><p>A one year increase in offspring's schooling attainment was associated with a 5% decline in the hazard of maternal death (adjusted Hazard Ratio [aHR]: 0.95, 95%CI: 0.94–0.97) and a 6% decline in the hazard of paternal death (aHR: 0.94, 95%CI: 0.92–0.96), adjusting for a wide range of demographic and socio-economic variables of the parent and their children. Among mothers, the association was strongest for communicable, maternal, perinatal and nutritional conditions (aHR: 0.87, 95%CI: 0.82–0.92) and AIDS and tuberculosis (aHR: 0.92, 95%CI: 0.89–0.96), and weakest for injuries. Among fathers, the association was strongest for injuries (aHR: 0.87, 95%CI: 0.79–0.95) and AIDS and tuberculosis (aHR: 0.92, 95%CI: 0.89–0.96), and weakest for non-communicable diseases.</p></sec><sec><title>Conclusion</title><p>Higher levels of schooling in offspring are associated with increased parental survival in rural South Africa, particularly for mothers at risk of communicable disease mortality and fathers at risk of injury mortality. Offspring's human capital may be an important factor for health disparities, particularly in lower-resource settings.</p></sec></abstract><abstract abstract-type="author-highlights" id="abs0015"><title>Highlights</title><p><list list-type="simple"><list-item id="u0010"><label>•</label><p>Investing in offspring's education may improve parental survival.</p></list-item><list-item id="u0015"><label>•</label><p>We follow 17,789 parents over a 13 year period in rural South Africa.</p></list-item><list-item id="u0020"><label>•</label><p>Higher offspring schooling is associated with increased parental survival.</p></list-item><list-item id="u0025"><label>•</label><p>Strongest association for mothers is for communicable disease mortality.</p></list-item><list-item id="u0030"><label>•</label><p>Strongest association for fathers is for injury mortality.</p></list-item></list></p></abstract><kwd-group id="kwrds0010"><title>Keywords</title><kwd>Offspring schooling</kwd><kwd>Survival</kwd><kwd>South Africa</kwd><kwd>Longitudinal</kwd></kwd-group></article-meta></front><body><sec id="sec1"><label>1</label><title>Introduction</title><p>Several recent studies from Asia (<xref rid="bib15" ref-type="bibr">Cai et al., 2006</xref>, <xref rid="bib28" ref-type="bibr">Frankenberg et al., 2002</xref>, <xref rid="bib33" ref-type="bibr">Giles et al., 2011</xref>, <xref rid="bib43" ref-type="bibr">Knodel et al., 2000</xref>), Latin America (<xref rid="bib14" ref-type="bibr">Bravo, 2006</xref>, <xref rid="bib58" ref-type="bibr">Saad, 2005</xref>), and sub-Saharan Africa (<xref rid="bib1" ref-type="bibr">Adamchak et al., 1991</xref>, <xref rid="bib44" ref-type="bibr">Kohler et al., 2012</xref>) suggest that a majority of individuals of ages 60 and older in low- and middle-income countries depend on the financial support provided by their children. Given the strong empirical relationship between schooling and socioeconomic status, investing in their children's human capital may thus be one of the most effective strategies for parents to improve their living conditions at older ages. This may be particularly the case in settings where public wealth transfers are limited and children play a more prominent role in supporting older generations.</p><p>The effect of children's human capital on the health of their parents is likely to be mostly protective for several reasons. First, children may function as a substitute for market institutions (e.g., employer-provided pensions and health insurance) or tax supported institutions (e.g., social security and health insurance) (<xref rid="bib5" ref-type="bibr">Barouni and Broecke, 2014</xref>, <xref rid="bib54" ref-type="bibr">Psacharopoulos, 1994</xref>, <xref rid="bib61" ref-type="bibr">Schultz, 2004</xref>). For example, children could recompense their parents for human capital investments received in childhood (‘parental repayment’), based on their increased financial means arising from their schooling (<xref rid="bib17" ref-type="bibr">Clay and Vander Haar, 1993</xref>, <xref rid="bib28" ref-type="bibr">Frankenberg et al., 2002</xref>). Additionally, more highly educated children may be better able to communicate health knowledge and skills acquired at or after school to parents, positively affecting their health behavior (<xref rid="bib8" ref-type="bibr">Berkman et al., 2000</xref>, <xref rid="bib21" ref-type="bibr">Cutler and Lleras-Muney, 2010</xref>, <xref rid="bib27" ref-type="bibr">Field and de la Roca, 2005</xref>, <xref rid="bib57" ref-type="bibr">Rowa-Dewar et al., 2014</xref>). More educated children may facilitate the use of medical information among their parents (e.g., understanding doctors' prescriptions) or may help their parents navigate the complex web of health insurance bureaucracies (<xref rid="bib38" ref-type="bibr">Grossman, 1972</xref>). Offspring with additional schooling may also be generally more familiar with modern society (<xref rid="bib34" ref-type="bibr">Glewwe, 1999</xref>), which may make them more receptive to modern medicine (<xref rid="bib4" ref-type="bibr">Aslam and Kingdon, 2012</xref>, <xref rid="bib31" ref-type="bibr">Frost et al., 2005</xref>), or the ‘hidden curriculum’ values of discipline and obedience of authority learned in school (<xref rid="bib6" ref-type="bibr">Basu and Stephenson, 2005</xref>). Finally, formal schooling may open access to careers in the health sector (e.g., as a nurse practitioner or community health worker), allowing offspring to directly provide care to their parents (<xref rid="bib7" ref-type="bibr">Bauman et al., 2006</xref>, <xref rid="bib26" ref-type="bibr">Evans and Becker, 2009</xref>, <xref rid="bib50" ref-type="bibr">McGarry, 1998</xref>).</p><p>While all of these pathways suggest that differential investment in children's human capital may be a key driver of social and health inequalities at older ages (<xref rid="bib9" ref-type="bibr">Berkman and Kawachi, 2000</xref>, <xref rid="bib42" ref-type="bibr">Kaufman and Cooper, 1999</xref>, <xref rid="bib45" ref-type="bibr">Krieger, 2001</xref>, <xref rid="bib49" ref-type="bibr">Marmot, 2003</xref>), surprisingly little evidence is available on the empirical relationship between children's human capital investment and parental survival. In the United States, one longitudinal study estimated a difference in life expectancy between parents of children with a college degree and parents of children with less than a high school diploma of about two years (age of death 71 vs 69) (<xref rid="bib29" ref-type="bibr">Friedman and Mare, 2014</xref>). Interestingly, this relationship persisted after controlling for parents' own socio-economic resources, and was more pronounced for deaths that were linked to behavioral factors (most notably, chronic lower respiratory disease and lung cancer). In Sweden, a study compared parental siblings and found that parents whose children had tertiary schooling had a 21% lower hazard of dying compared to their siblings whose children completed only nine years of compulsory schooling (<xref rid="bib66" ref-type="bibr">Torssander, 2013</xref>). A quasi-experimental study in Sweden, exploiting changes in schooling law as a ‘natural experiment’, found <italic>no evidence</italic> of a causal effect of schooling in children on parental longevity (<xref rid="bib48" ref-type="bibr">Lundborg and Majlesi, 2015</xref>). Large effects of children's schooling on parental survival might, however, seem unlikely in Sweden – where a comprehensive welfare system generously provides for the elderly, and upward intergenerational wealth transfers are uncommon (<xref rid="bib30" ref-type="bibr">Fritzell and Lennartsson, 2005</xref>, <xref rid="bib47" ref-type="bibr">Lennartsson et al., 2009</xref>, <xref rid="bib66" ref-type="bibr">Torssander, 2013</xref>). Even though children may still provide informal care in this setting (<xref rid="bib10" ref-type="bibr">Bolin et al., 2008</xref>, <xref rid="bib47" ref-type="bibr">Lennartsson et al., 2009</xref>), the marginal impact of this additional support in the Swedish setting is likely to be much smaller than the impact in a low-resource setting.</p><p>Two relevant panel studies have been conducted in middle-income countries. In a Chinese study, the adjusted hazard of parental death for those living with offspring with 10 or more years of schooling was 17% lower than that for offspring with 6 years of schooling or less (<xref rid="bib72" ref-type="bibr">Yang et al., 2016</xref>). The study suggests that children's and spousal schooling were similarly important for the health of the elderly in this setting. Another longitudinal study, using data from Mexico, assessed the impact of children's schooling on their parents' functional limitations. In adjusted analysis, parents whose children had completed high school were less likely to report functional limitations than those whose children had not (<xref rid="bib71" ref-type="bibr">Yahirun et al., 2016</xref>). To our knowledge, no studies have been conducted in sub-Saharan Africa.</p><sec id="sec1.1"><label>1.1</label><title>South Africa education system</title><p>The public education system in South Africa is divided into three phases of basic education (<xref rid="bib56" ref-type="bibr">Republic of South Africa, 2013</xref>). The ‘foundation’ phase runs from grade R (pre-primary) to grade 3; the ‘intermediate’ phase from grades 4 to 9; and the ‘senior’ phase from grades 10 to 12. Primary school would typically be considered grades 1 to 7, with the <italic>de jure</italic> primary school-age population being 7–13 years old. Lower secondary school would typically be considered grades 8 to 9; upper-secondary school would be considered grades 10 to 12. Students between the ages of 14 and 18 years old are officially being regarded as of secondary school-going age. Schooling is mandatory for children aged 7–15 years (i.e., those in the foundation and intermediate phases). Gross primary school enrollment (both sexes) was 99.7% in 2014 (<xref rid="bib68" ref-type="bibr">UNESCO Institute for Statistics, 2016</xref>). Gross lower-secondary school enrollment (both sexes) was 94.9% and gross upper-secondary school enrolment (both sexes) was 94.5%. The National Senior Certificate, commonly known as ‘matric’, is a three year qualification that signifies the end of twelve years of schooling. In principle, public schooling is free for those who cannot afford to pay school fees. Tertiary schooling in South Africa offers a wide range of options, including three-year bachelor's degrees and an honors degree, an optional fourth-year qualification, in addition to advanced masters and doctorate degrees.</p><p>The relationship between schooling attainment and health outcomes has garnered increasing attention from government, donors, and health system researchers and planners in South Africa. In particular, recent research and policy efforts have increased their focus on improving access and retention in schools as a potential HIV prevention strategy among young women. Two recent RCTs, including in rural KwaZulu Natal (KZN) and Mpumalanga provinces, have assessed the role of schooling conditional cash transfers to reduce HIV infection risk among secondary school students (<xref rid="bib41" ref-type="bibr">Karim et al., 2015</xref>, <xref rid="bib53" ref-type="bibr">Pettifor et al., 2016</xref>). Moreover, increased access to, and retention in, school is a key part of the United States President’s Emergency Plan for AIDS Relief DREAMS initiative, which is being piloted in several South African sites (<xref rid="bib52" ref-type="bibr">PEPFAR, 2016</xref>). The current study thus has important implications for research and policy, in determining whether these efforts to increase school attainment in South Africa may have additional health benefits in terms of increased survival for older generations, over and above those arising more directly from offspring's reduced HIV risk.</p><p>We therefore used longitudinal data to test the hypothesis that offspring's schooling decreases the hazard of parental death in an African setting with very high adult mortality (<xref rid="bib18" ref-type="bibr">Coovadia et al., 2009</xref>). We used one of Africa's largest cohorts, located in rural KZN, South Africa, to follow up 17,789 parents and observe their survival over the period January 2003 to August 2015. The availability of a range of demographic and socio-economic variables in our dataset allowed us to control for important determinants of adult mortality and to assess the role of offspring's schooling on the most prevalent causes of death among their parents in South Africa.</p></sec></sec><sec id="sec2"><label>2</label><title>Methods</title><sec id="sec2.1"><label>2.1</label><title>Study area</title><p>Since 2000, the Africa Centre for Population Health has collected longitudinal demographic, social, and economic data on over 100,000 people living in a 432 km<sup>2</sup> demographic surveillance area (DSA) in uMkhanyakude District, in northern KZN. The DSA includes both rural areas that were a designated Zulu ‘homeland’ area during Apartheid and urban areas that formerly constituted a black-only township. uMkhanyakude is the poorest of the 11 districts in KZN and the second most deprived district in South Africa (<xref rid="bib35" ref-type="bibr">Government of South Africa, 2015</xref>). Adult HIV prevalence in this community is very high. Prior to the roll-out of public-sector antiretroviral therapy (ART), adult life expectancy in the DSA was 49.2 (<xref rid="bib11" ref-type="bibr">Bor et al., 2013</xref>) with over half of the population deaths attributed to HIV (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). Since the roll-out of ART in 2004, however, the burden of HIV on adult mortality in this population is rapidly falling (<xref rid="bib55" ref-type="bibr">Reniers et al., 2016</xref>). By 2011, adult life expectancy had increased to 60.5 years. All-cause mortality declined by over 50% for adults aged 25 to 44 between 2003 and 2011; however, mortality reductions at older ages were much smaller (<xref rid="bib11" ref-type="bibr">Bor et al., 2013</xref>).</p></sec><sec id="sec2.2"><label>2.2</label><title>Data collection</title><p>Household surveillance in the DSA began on 1 January 2000. Data on basic household demographics, residence and vital status of DSA household members are collected at least twice a year (<xref rid="bib65" ref-type="bibr">Tanser et al., 2008</xref>). Individuals are registered if they are a member of a DSA-resident household, even if they are not themselves living in the DSA. Respondents are considered members of the same household regardless of where they live (<xref rid="bib40" ref-type="bibr">Hosegood and Timaeus, 2006</xref>). This household dataset allowed us to identify any DSA-member child of each registered parent. We had no data, however, on whether parents co-reside with their offspring, or the frequency of their contact. We linked these data from each household visit to individual-level information gathered at annual Household and Socio-Economic Survey (HSE) interviews. The HSE survey (including household wealth quintile) has been conducted annually since 2003, with the exception of 2004 and 2008. Finally, we added detailed information on time and cause of death for deceased individuals, based on verbal autopsy interviews for all notified deaths of DSA household members, which were collected for all survey years (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). Data on our main exposure, highest offspring educational attainment, was missing for 8% of observations for mothers and 7% of observations for fathers. Among parents with complete data on offspring educational attainment, data on our main covariates – marital status, employment status, and household wealth – were available for over 95% of observations, while parental age and area of residence were available for all observations. Data on deaths are based on records for each DSA-registered individual for whom a Death Outcome Notification Form has been recorded. This death notification form triggered the conduct of a verbal autopsy with the closest caregiver of the deceased, on average six months after the death. Verbal autopsy refusals in the DSA were low (2%) (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). Our analysis ran from 1 January 2003 to 16 August 2015.</p></sec><sec id="sec2.3"><label>2.3</label><title>Sample</title><p>Inclusion criteria for our analysis were that a parent had at least one DSA-linked child who was alive at the time of the first household interview and that at least one of these offspring was at least 18 years old. This age restriction was imposed so that offspring had the opportunity to complete secondary schooling or more, which runs until age 18 in South Africa, and so that they were eligible for the survey component on labor market outcomes. Many individuals reside outside the DSA, but continue to be members of a household in the DSA. Exposure continued to accumulate if the individual out-migrated from the DSA but retained membership of a resident household. The ability to track the demographic characteristics of non-resident household members is a major strength of the data. Parents were right-censored if they were lost to follow up due to non-response, lost membership of a resident household, or at the end of the study period (16 August 2015). Participation rates for household data collection were >99%. Previous studies in the DSA have found that attrition due to permanent out-migration or loss to follow-up was low, at a rate of 3.3 per 100 person-years (<xref rid="bib11" ref-type="bibr">Bor et al., 2013</xref>). Additional details on the cohort are documented elsewhere (<xref rid="bib65" ref-type="bibr">Tanser et al., 2008</xref>).</p></sec><sec id="sec2.4"><label>2.4</label><title>Exposure</title><p>In our study, we focused on the parent as unit of analysis with the schooling of their offspring as exposure. Our exposure was the highest schooling attainment of any offspring who was a member of a DSA household between 2003 and 2015. School attainment was defined as time-varying total years of schooling completed by the time of each visit. In additional models, we created time-varying indicators for having completed 0–7 years, 8–9 years, 10–12 years, or 13+ years of schooling. We compared parents whose offspring have completed 0–7 years of schooling (i.e., primary schooling or less) to parents whose offspring have completed 8–9 years, 10–12 years, or 13+ years of schooling. These comparisons are likely to be most policy-relevant in South Africa, where much of the current debate is around the benefits of increased access to secondary school and beyond.</p></sec><sec id="sec2.5"><label>2.5</label><title>Outcome</title><p>Our primary outcome was death of the parent. The ‘incidence rate’ of parental death was calculated as the number of parental deaths divided by the total number of person-years of observation between 2003 and 2015, separately for each offspring schooling attainment group. Our secondary outcome of interest was parental cause of death. Causes of death were mapped into Global Burden of Disease groups I (communicable, maternal, perinatal and nutritional conditions), II (non-communicable diseases) and III (injuries)—with the exception of deaths due to tuberculosis (TB) and AIDS. TB and AIDS were classified together into a separate group, given the considerable overlap in mortality from HIV infection and TB (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). Among women in the DSA, the three most common underlying causes of death were lower respiratory infections, meningitis, and diarrheal diseases (group I); cerebro- and cardiovascular diseases, and diabetes mellitus (group II); and violence, road traffic accidents, and other unintentional injuries (group III); in addition to AIDS or TB. The most common underlying causes of death were nearly identical for women and men. Violence, however, caused over 50% of injury-related deaths in men compared to 36% of injury-related deaths in women.</p></sec><sec id="sec2.6"><label>2.6</label><title>Covariates</title><p>To account for possible confounding, we included time-varying linear, quadratic and cubic terms in parental age, and a linear term in parental years of schooling, based on tests of functional form (<xref rid="appsec1" ref-type="sec">Supplementary Table 1</xref>) (<xref rid="bib60" ref-type="bibr">Sartorius et al., 2013</xref>). We also included time-varying parental marital status (as a binary indicator), number of children (as a continuous variable), age of the oldest child (as a continuous variable), and gender of the oldest child (as a binary indicator) (<xref rid="bib25" ref-type="bibr">Downey, 1995</xref>, <xref rid="bib63" ref-type="bibr">Sun et al., 2015</xref>, <xref rid="bib64" ref-type="bibr">Tamakoshi et al., 2010</xref>). Additionally, we included the baseline values of characteristics that were likely on the causal pathway between offspring schooling and parental survival, to capture the total effects of offspring schooling during follow-up (<xref rid="bib69" ref-type="bibr">VanderWeele, 2015</xref>). We included the baseline value of employment status of both the oldest child and parent (as a binary indicator), baseline marital status of the oldest child (as a binary indicator), as well as baseline household wealth (quintiles) (<xref rid="bib51" ref-type="bibr">Morris, 2000</xref>) and baseline household area of residence (urban, periurban, rural) of the parent and their children. We also included indicator variables for year of observation to control for unmeasured confounders that change over time (e.g., changing availability of HIV prevention and treatment services). By simultaneously controlling for year of observation and age, we implicitly also controlled for birth cohort effects. <xref rid="fig1" ref-type="fig">Fig. 1</xref> displays a simplified conceptual framework which underlies our empirical strategy.</p></sec><sec id="sec2.7"><label>2.7</label><title>Statistical analysis</title><p>To investigate the association between parental death and offspring's schooling, we used a Cox proportional hazards model (CPHM). Parents entered the study on 1 January 2003, on their oldest offspring's 18th birthday or at the date of in-migration to the DSA. Parents were followed up until death or right-censoring. In our main analyses, we assessed the incidence of maternal and paternal deaths separately, since mortality risk differs strongly by gender in our study setting, but pooled children of both sexes. In supplementary analyses, we assessed the incidence of parental deaths in subsamples by gender of the oldest child. We verified the proportional-hazards assumption using Schoenfeld residuals (<xref rid="bib36" ref-type="bibr">Grambsch and Therneau, 1994</xref>). Additionally, we estimated cause-specific models for four common groupings of the most prevalent causes of death in South Africa. The competing causes of deaths are treated as censored observations. Hazard ratios calculated using this approach can be interpreted as ‘among those who did not (yet) experience the competing causes of death’ (<xref rid="bib2" ref-type="bibr">Andersen et al., 2012</xref>, <xref rid="bib46" ref-type="bibr">Lau et al., 2009</xref>). All analyses were conducted in Stata 14 (StataCorp, College Station, Texas, USA).</p></sec><sec id="sec2.8"><label>2.8</label><title>Sensitivity analyses</title><p>We conducted a wide range of sensitivity analyses. First, we assessed time to parental death using exponential and Weibull parametric, accelerated failure time (AFT) models. Although parametric models may lead to inconsistent estimates if the baseline function is mis-specified, they will be more efficient than Cox models if the baseline hazard is correctly specified. Second, to allow the baseline hazard to vary by parental age we stratified our analyses by age, since the effect of schooling might differ depending on parental age and when offspring obtained their schooling. Stratified models used the full sample. Third, because grade repetition is high in rural South Africa (<xref rid="bib13" ref-type="bibr">Branson et al., 2013</xref>) and students may enter formal schooling late, we also assessed the robustness of our results to alternative sample specifications, including the sample of offspring ages either at least 21 or 25 years old. Fourth, we conducted two-level random-intercept Weibull survival models, allowing the effect to vary either: (i) by household membership type of the DSA; or (ii) area of residence of parents (urban, periurban, rural). Finally, we reran our main analyses after multiple imputation of missing variables in the dataset.</p></sec></sec><sec id="sec3"><label>3</label><title>Results</title><p>Baseline respondent characteristics are shown in <xref rid="tbl1" ref-type="table">Table 1</xref>. Median age at baseline was 49 for mothers and 52 for fathers. Median age at death was 63 (interquartile range [IQR]: 49–76) for mothers and 63 (IQR: 52–74) for fathers. The median years of schooling completed was similar for women and men in our sample (4 [IQR: 0–9] for mothers and 5 [IQR: 0–11] for fathers). Median highest years of schooling among offspring was substantially larger than the schooling attainment of their parents (12 [IQR: 10–12] for offspring of both mothers and fathers). The number of parents whose offspring did not complete any schooling was small (only 285 [2%] mothers and 50 [1%] fathers).</p><p>Between 2003 and 2015, we observed 1886 maternal deaths over 81,653 person-years at risk; and 1344 paternal deaths over 36,001 person-years at risk. The crude incidence of maternal death over the observation period was 2.31 deaths per 100 person-years (95% confidence interval [CI]: 2.21–2.42); whereas the crude incidence of paternal death over the observation period was 3.37 deaths per 100 person-years (95%CI: 3.53–3.93) (<xref rid="tbl2" ref-type="table">Table 2</xref>). Incidence was highest for parents of offspring with no schooling: 6.73 maternal deaths per 100 person-years (95%CI: 5.66–8.00) and 12.0 paternal deaths per 100 person-years (95%CI: 8.74–16.51). <xref rid="fig2" ref-type="fig">Fig. 2</xref> shows the cumulative hazard of parental death during follow-up, stratified by offspring's schooling attainment. The cumulative hazard of parental death does not show large deviations from linearity.</p><p>In both bivariate and adjusted analysis, parents who had offspring with more years of schooling had significantly lower hazards of death (<xref rid="tbl3" ref-type="table">Table 3</xref>). When including offspring's schooling attainment in the model as a continuous variable (model 1), a one year increase in offspring's schooling attainment was associated with a 5% lower hazard of maternal death (adjusted Hazard Ratio [aHR]: 0.95, 95%CI: 0.94–0.97) and a 6% lower hazard of paternal death (aHR: 0.94, 95%CI: 0.92–0.96). When including offspring's schooling attainment levels in the model as indicator variables (model 2), completion of 10–12 years of schooling (versus 0–7 years of schooling) among offspring was associated with a 26% reduction in the hazard of maternal death (aHR: 0.74, 95%CI: 0.64–0.85), and a 35% reduction in the hazard of paternal death (aHR: 0.65, 95%CI: 0.54–0.80). In contrast, lower levels of secondary schooling (8–9 years) were not associated with increased parental survival in adjusted models, suggesting a non-linear relationship in this setting. The relationship was strongest for fathers whose oldest child was male, although child's gender did not have a significant interactive effect (<xref rid="appsec1" ref-type="sec">Supplementary Table 3</xref>). Supplementary analyses using additional controls and AFT models (<xref rid="appsec1" ref-type="sec">Supplementary Table 4</xref>), stratified by age (<xref rid="appsec1" ref-type="sec">Supplementary Table 5</xref>), using alternative sample specifications (<xref rid="appsec1" ref-type="sec">Supplementary Table 6</xref>), and using multilevel mixed-effects parametric survival models (<xref rid="appsec1" ref-type="sec">Supplementary Table 7</xref>) yielded similar results. After multiple imputation of missing values, our results remained essentially unchanged (<xref rid="appsec1" ref-type="sec">Supplementary Tables 8 and 9</xref>).</p><p><xref rid="tbl4" ref-type="table">Table 4</xref> displays results from cause-specific models for common groupings of the most prevalent causes of death. For both mothers and fathers in our analytical sample, NCDs were the most common cause of death, followed by AIDS or TB. Parents of offspring with more schooling were less likely to die of communicable diseases (mothers) and injuries (fathers) in particular. A one-year increase in offspring's schooling attainment was associated with a 13% decline (aHR: 0.87, 95%CI: 0.82–0.92) in maternal hazard of death due to CPMN and a 13% decline (aHR: 0.87, 95%CI: 0.79–0.95) in paternal hazard of death due to injury. Similarly, one additional year of offspring schooling was associated with a 8% decline (aHR: 0.92, 95%CI: 0.89–0.96) in maternal and paternal hazard of death due to AIDS or TB. Parents of offspring with more schooling were also less likely (to a smaller extent) to die as a result of NCDs. Mothers of offspring with more schooling were not less likely to die as a result of injury, although the number of injury deaths among mothers was low (50 maternal cases vs. 92 paternal cases).</p></sec><sec id="sec4"><label>4</label><title>Discussion</title><p>Using a large prospective South African cohort, we found that additional schooling in offspring is associated with significantly increased parental survival. Overall, the association was similarly strong for mothers as for fathers. This relationship persisted after controlling for a range of known determinants of adult mortality in rural South Africa, such as age, schooling, marital status, area of residence, employment status, and household wealth of both the parent and offspring. We also controlled for unmeasured confounders that change over time, such as the roll-out of HIV prevention and treatment in South Africa. Our results were robust across a wide range of additional sensitivity analyses, including alternative functional forms of parental age and schooling, and alternative specifications of the baseline hazard function. A handful of previous studies have assessed the role of offspring's schooling in parental survival, but these studies have generally used data from developed countries (<xref rid="bib29" ref-type="bibr">Friedman and Mare, 2014</xref>, <xref rid="bib48" ref-type="bibr">Lundborg and Majlesi, 2015</xref>, <xref rid="bib66" ref-type="bibr">Torssander, 2013</xref>, <xref rid="bib73" ref-type="bibr">Zimmer et al., 2007</xref>). This study is among the first to extend this growing literature to a lower-resource setting with limited social welfare systems and very high adult mortality.</p><p>A second key finding of our study is that higher levels of schooling (i.e., upper secondary school and beyond) among offspring appeared to drive the survival benefit for parents. Late secondary and tertiary education may be particularly important for several reasons: the development of new skills (e.g., complex reasoning and future-oriented thinking) with strong returns in the labor market (<xref rid="bib19" ref-type="bibr">Crews et al., 2007</xref>, <xref rid="bib22" ref-type="bibr">Dahl, 2004</xref>, <xref rid="bib62" ref-type="bibr">Steinberg, 2005</xref>) and the decisions facing adolescents have significant path dependence, such as childbearing and planning a career path (<xref rid="bib70" ref-type="bibr">Wilde et al., 2010</xref>). These results suggest that the years of late secondary and tertiary education may be a ‘critical period’ (<xref rid="bib20" ref-type="bibr">Cunha and Heckman, 2007</xref>) to invest in the human capital of children for parental health.</p><p>While no previous studies have examined sub-Saharan Africa, our results are consistent with recent studies that found a strong protective association between offspring's schooling and: (i) parental mortality in China (<xref rid="bib72" ref-type="bibr">Yang et al., 2016</xref>); and (ii) parental functional limitations in Mexico (<xref rid="bib71" ref-type="bibr">Yahirun et al., 2016</xref>). Interestingly, all three settings share some characteristics with higher-resource countries (e.g., a relatively large elderly population) and some with lower-resource settings (e.g., limited social welfare). The role of offspring's schooling in low- and middle-income countries is likely to be of particular importance, since a larger proportion of the elderly depend on their children for financial support and co-residence of parents with offspring is more common (<xref rid="bib15" ref-type="bibr">Cai et al., 2006</xref>, <xref rid="bib33" ref-type="bibr">Giles et al., 2011</xref>). Faced with an increasingly large elderly population, human capital investments in younger populations in lower-resource settings may pay important dividends to older generations.</p><p>Our findings are particularly relevant to other sub-Saharan African countries, where supply-side education reforms have greatly increased access to schooling in recent decades. Zimbabwe, for instance, rapidly expanded access to secondary schools for black citizens in the 1980s through a government-initiated school construction program focused on rural areas (<xref rid="bib3" ref-type="bibr">Ansell, 2002</xref>). Each additional year of maternal schooling induced by the reform reduced mortality among their offspring by about 21% (<xref rid="bib37" ref-type="bibr">Grépin and Bharadwaj, 2015</xref>). Our findings suggest that additional schooling in children may not only generate such large ‘downstream’ benefits for the next generation, but could also have ‘upstream’ impacts on the preceding one. Further investigation of the impact of expanding schooling access on the health of family members in sub-Saharan Africa appears warranted.</p><p>We also assessed the association between offspring's schooling and parental causes of death, grouped by GBD categories. Additional schooling in offspring was associated with the largest reduction in parental death for communicable diseases (mothers) and injury (fathers). The strong association between offspring schooling and reduced risk of death due to injury among their fathers (but not mothers) is of particular interest. One possible explanation for this finding might be that children share information acquired at school with their fathers on the health risks associated with driving under influence of alcohol or share norms and values with their fathers around violence, possibly leading to fewer injuries. While these examples seem entirely plausible, one would expect deaths due to injury to more often occur as a result of external circumstances as opposed to individual behavioral factors such as those linked to AIDS, TB, or NCDs. Whether these results for injury are due to residual confounding, however, cannot be ascertained with the current study design.</p><p>Among both mothers and fathers, a strong association was also seen for death due to AIDS or TB. The difference seen between results for AIDS or TB compared to NCDs suggests a potentially important role for additional human capital in children in a setting where HIV is endemic. Exposure to additional school-based HIV prevention programming may have changed knowledge and attitudes (e.g., on the benefits of HIV testing, early ART initiation or adherence) (<xref rid="bib32" ref-type="bibr">Gallant and Maticka-Tyndale, 2004</xref>). Children may also have benefited from HIV education in the broader school curriculum, insofar as this latter had changed since their parents were at school. Since the government of South Africa provides an old-age pension to the elderly (<xref rid="bib16" ref-type="bibr">Case and Deaton, 1998</xref>), non-pecuniary pathways (such as HIV knowledge and attitudes) may play a more important role relative to wealth transfers than might be seen in other sub-Saharan settings. Conversely, this smaller role for wealth transfers in South Africa suggests that the protective effect of offspring education on parental health outcomes may be even stronger elsewhere in sub-Saharan Africa.</p><p>Finally, we note that parental risk-profiles and survival depend on the setting under study. Previous studies from more developed settings have examined samples of parents that were much older: at least 50 years old in the United States and Western-European countries (<xref rid="bib29" ref-type="bibr">Friedman and Mare, 2014</xref>, <xref rid="bib59" ref-type="bibr">Sabater and Graham, 2016</xref>), at least 60 years old in Sweden and Taiwan (<xref rid="bib48" ref-type="bibr">Lundborg and Majlesi, 2015</xref>, <xref rid="bib67" ref-type="bibr">Torssander, 2014</xref>, <xref rid="bib73" ref-type="bibr">Zimmer et al., 2007</xref>) and at least 65 years old in China (<xref rid="bib72" ref-type="bibr">Yang et al., 2016</xref>). In contrast, the results presented here reflect the survival of parents that are on average much younger (median of 49 for women and 52 for men) and die much earlier (median of 63 for both women and men). Nevertheless, given the lower life expectancy in South Africa relative to other countries studied, this population was not unusually early in their life course. In this study's surveillance area, between 2000 and 2009, HIV-related causes constituted the largest cause-specific mortality fraction among those ages 50–64 (42%, as opposed to 27% due to lifestyle-related NCDs) (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). These parents were at particular risk of mortality due to HIV-related causes, but relatively young to die of NCDs. An additional implication is that reduced mortality among these younger middle-aged parents is likely to have large additional positive effects on labor market outcomes (e.g., younger and healthier parents could return to work (<xref rid="bib12" ref-type="bibr">Bor et al., 2012</xref>)), and other social outcomes.</p><sec id="sec4.1"><label>4.1</label><title>Limitations</title><p>A great strength of our analysis was the availability of time and cause of death of parents, which allowed us to assess the differential role of offspring's schooling for the most prevalent causes of death in a population-based cohort in South Africa. Nevertheless, our study has some limitations. First, although we control for a wide range of covariates available in the dataset, the design of our study unfortunately does not allow for causal inference. There may be confounding by factors which we were unable to control for (e.g., unmeasured parental attitudes or resources), in which case offspring's schooling may be a strong marker of parental survival, as opposed to an exposure. Given the scarcity of evidence on the role of human capital in children and parental old-age survival, however, the aim of this study was to extend the current associational literature to low-resource settings with high mortality, to serve as a comprehensive baseline analysis for subsequent (quasi-)experimental research. Future research, for instance, could exploit schooling policy reforms in sub-Saharan Africa as a ‘natural experiment’ (<xref rid="bib23" ref-type="bibr">De Neve et al., 2015</xref>, <xref rid="bib37" ref-type="bibr">Grépin and Bharadwaj, 2015</xref>). Future research could also use mediation analysis methods to examine the proportions of the association between offspring schooling and parental survival that were mediated by employment status and household wealth (<xref rid="bib69" ref-type="bibr">VanderWeele, 2015</xref>), conduct additional sensitivity analyses to assess the robustness of our findings (<xref rid="bib24" ref-type="bibr">Ding and VanderWeele, 2016</xref>), or conduct intervention studies to test (specific pathways of) the effect of offspring schooling on parental survival.</p><p>Second, as with any long-term community-based study, the cohort suffered from attrition and non-response. However, non-response was very limited (household participation rates were >99%) and our results did not change after accounting for data missingness through multiple imputation. Third, there may be mismeasurement in our exposure and outcome. During verbal autopsies, for instance, caregivers may refuse or seek to underreport deaths due to HIV out of fear of HIV-related stigma in the community. Refusals to report deaths, however, were low (2%), and verbal autopsies were conducted by nurses with extensive training in administering the questionnaire (<xref rid="bib39" ref-type="bibr">Herbst et al., 2011</xref>). Fourth, since our study is focused on one geographically limited population, our results may not generalize widely. Educational attainment among offspring in our study, for instance, was relatively high (median years of schooling was 12) and similar for girls and boys. Our findings may not generalize to settings where educational opportunities are more limited, or more gender-unequal. Our findings are likely informative, however, of the association between offspring's schooling and parental old-age survival in low-resource settings, particularly where HIV is endemic.</p></sec></sec><sec id="sec5"><label>5</label><title>Conclusions</title><p>Additional schooling in offspring is associated with increased parental survival in rural South Africa, particularly for mothers at risk of communicable disease mortality and fathers at risk of injury mortality. These findings suggest that schooling may not only have large benefits for those enrolled, but could also be beneficial for older family members. Studies focusing only on the private returns to schooling may underestimate the societal benefits, particularly in settings where welfare sectors are underdeveloped and adult mortality is high.</p></sec></body><back><ref-list id="cebib0010"><title>References</title><ref id="bib1"><element-citation publication-type="journal" id="sref1"><person-group person-group-type="author"><name><surname>Adamchak</surname><given-names>D.J.</given-names></name><name><surname>Wilson</surname><given-names>A.O.</given-names></name><name><surname>Nyanguru</surname><given-names>A.</given-names></name><name><surname>Hampson</surname><given-names>J.</given-names></name></person-group><article-title>Elderly support and intergenerational transfer in Zimbabwe: an analysis by gender, marital status, and place of residence</article-title><source>Gerontologist</source><volume>31</volume><year>1991</year><fpage>505</fpage><lpage>513</lpage><pub-id pub-id-type="pmid">1894155</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal" id="sref2"><person-group person-group-type="author"><name><surname>Andersen</surname><given-names>P.K.</given-names></name><name><surname>Geskus</surname><given-names>R.B.</given-names></name><name><surname>de Witte</surname><given-names>T.</given-names></name><name><surname>Putter</surname><given-names>H.</given-names></name></person-group><article-title>Competing risks in epidemiology: possibilities and pitfalls</article-title><source>Int. J. Epidemiol.</source><volume>41</volume><year>2012</year><fpage>861</fpage><lpage>870</lpage><pub-id pub-id-type="pmid">22253319</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal" id="sref3"><person-group person-group-type="author"><name><surname>Ansell</surname><given-names>N.</given-names></name></person-group><article-title>Secondary education reform in Lesotho and Zimbabwe and the needs of rural girls: pronouncements, policy and practice</article-title><source>Comp. Educ.</source><volume>38</volume><year>2002</year><fpage>91</fpage><lpage>112</lpage></element-citation></ref><ref id="bib4"><element-citation publication-type="journal" id="sref4"><person-group person-group-type="author"><name><surname>Aslam</surname><given-names>M.</given-names></name><name><surname>Kingdon</surname><given-names>G.G.</given-names></name></person-group><article-title>Parental education and child health—understanding the pathways of impact in Pakistan</article-title><source>World Dev.</source><volume>40</volume><year>2012</year><fpage>2014</fpage><lpage>2032</lpage></element-citation></ref><ref id="bib5"><element-citation publication-type="journal" id="sref5"><person-group person-group-type="author"><name><surname>Barouni</surname><given-names>M.</given-names></name><name><surname>Broecke</surname><given-names>S.</given-names></name></person-group><article-title>The returns to education in Africa: some new estimates</article-title><source>J. Dev. Stud.</source><volume>50</volume><year>2014</year><fpage>1593</fpage><lpage>1613</lpage></element-citation></ref><ref id="bib6"><element-citation publication-type="journal" id="sref6"><person-group person-group-type="author"><name><surname>Basu</surname><given-names>A.M.</given-names></name><name><surname>Stephenson</surname><given-names>R.</given-names></name></person-group><article-title>Low levels of maternal education and the proximate determinants of childhood mortality: a little learning is not a dangerous thing</article-title><source>Soc. Sci. Med.</source><volume>60</volume><year>2005</year><fpage>2011</fpage><lpage>2023</lpage><pub-id pub-id-type="pmid">15743650</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal" id="sref7"><person-group person-group-type="author"><name><surname>Bauman</surname><given-names>L.J.</given-names></name><name><surname>Foster</surname><given-names>G.</given-names></name><name><surname>Johnson Silver</surname><given-names>E.</given-names></name><name><surname>Berman</surname><given-names>R.</given-names></name><name><surname>Gamble</surname><given-names>I.</given-names></name><name><surname>Muchaneta</surname><given-names>L.</given-names></name></person-group><article-title>Children caring for their ill parents with HIV/AIDS</article-title><source>Vulnerable Child. Youth Stud.</source><volume>1</volume><year>2006</year><fpage>56</fpage><lpage>70</lpage></element-citation></ref><ref id="bib8"><element-citation publication-type="journal" id="sref8"><person-group person-group-type="author"><name><surname>Berkman</surname><given-names>L.F.</given-names></name><name><surname>Glass</surname><given-names>T.</given-names></name><name><surname>Brissette</surname><given-names>I.</given-names></name><name><surname>Seeman</surname><given-names>T.E.</given-names></name></person-group><article-title>From social integration to health: Durkheim in the new millennium</article-title><source>Soc. Sci. Med.</source><volume>51</volume><year>2000</year><fpage>843</fpage><lpage>857</lpage><pub-id pub-id-type="pmid">10972429</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="book" id="sref9"><person-group person-group-type="author"><name><surname>Berkman</surname><given-names>L.F.</given-names></name><name><surname>Kawachi</surname><given-names>I.</given-names></name></person-group><chapter-title>Social Epidemiology</chapter-title><year>2000</year><publisher-name>Oxford University Press</publisher-name><publisher-loc>New York</publisher-loc></element-citation></ref><ref id="bib10"><element-citation publication-type="journal" id="sref10"><person-group person-group-type="author"><name><surname>Bolin</surname><given-names>K.</given-names></name><name><surname>Lindgren</surname><given-names>B.</given-names></name><name><surname>Lundborg</surname><given-names>P.</given-names></name></person-group><article-title>Informal and formal care among single-living elderly in Europe</article-title><source>Health Econ.</source><volume>17</volume><year>2008</year><fpage>393</fpage><lpage>409</lpage><pub-id pub-id-type="pmid">17768700</pub-id></element-citation></ref><ref id="bib74"><element-citation publication-type="journal" id="sref74"><person-group person-group-type="author"><name><surname>Bongaarts</surname><given-names>J.</given-names></name></person-group><article-title>A framework for analysing the proximate determinants of fertility</article-title><source>Popul. Dev. Rev.</source><volume>4</volume><issue>1</issue><year>1978</year><fpage>105</fpage><lpage>132</lpage></element-citation></ref><ref id="bib11"><element-citation publication-type="journal" id="sref11"><person-group person-group-type="author"><name><surname>Bor</surname><given-names>J.</given-names></name><name><surname>Herbst</surname><given-names>A.J.</given-names></name><name><surname>Newell</surname><given-names>M.L.</given-names></name><name><surname>Barnighausen</surname><given-names>T.</given-names></name></person-group><article-title>Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment</article-title><source>Science</source><volume>339</volume><year>2013</year><fpage>961</fpage><lpage>965</lpage><pub-id pub-id-type="pmid">23430655</pub-id></element-citation></ref><ref id="bib12"><element-citation publication-type="journal" id="sref12"><person-group person-group-type="author"><name><surname>Bor</surname><given-names>J.</given-names></name><name><surname>Tanser</surname><given-names>F.</given-names></name><name><surname>Newell</surname><given-names>M.L.</given-names></name><name><surname>Barnighausen</surname><given-names>T.</given-names></name></person-group><article-title>In a study of a population cohort in South Africa, HIV patients on antiretrovirals had nearly full recovery of employment</article-title><source>Health Aff.</source><volume>31</volume><year>2012</year><fpage>1459</fpage><lpage>1469</lpage></element-citation></ref><ref id="bib13"><element-citation publication-type="journal" id="sref13"><person-group person-group-type="author"><name><surname>Branson</surname><given-names>N.</given-names></name><name><surname>Hofmeyr</surname><given-names>C.</given-names></name><name><surname>Lam</surname><given-names>D.</given-names></name></person-group><article-title>Progress through school and the determinants of school dropout in South Africa</article-title><source>Dev. South. Afr.</source><volume>31</volume><year>2013</year><fpage>106</fpage><lpage>126</lpage></element-citation></ref><ref id="bib14"><element-citation publication-type="book" id="sref14"><person-group person-group-type="author"><name><surname>Bravo</surname><given-names>J.</given-names></name></person-group><chapter-title>Intergenerational Transfers and Social Protection in Latin America</chapter-title><year>2006</year><publisher-name>National Transfers Accounts Working Paper Series</publisher-name><publisher-loc>Mexico City</publisher-loc></element-citation></ref><ref id="bib15"><element-citation publication-type="journal" id="sref15"><person-group person-group-type="author"><name><surname>Cai</surname><given-names>F.</given-names></name><name><surname>Giles</surname><given-names>J.</given-names></name><name><surname>Meng</surname><given-names>X.</given-names></name></person-group><article-title>How well do children insure parents against low retirement income? An analysis using survey data from urban China</article-title><source>J. Public Econ.</source><volume>90</volume><year>2006</year><fpage>2229</fpage><lpage>2255</lpage></element-citation></ref><ref id="bib16"><element-citation publication-type="journal" id="sref16"><person-group person-group-type="author"><name><surname>Case</surname><given-names>A.</given-names></name><name><surname>Deaton</surname><given-names>A.</given-names></name></person-group><article-title>Large cash transfers to the elderly in South Africa</article-title><source>Econ. J.</source><volume>108</volume><year>1998</year><fpage>1330</fpage><lpage>1361</lpage></element-citation></ref><ref id="bib17"><element-citation publication-type="journal" id="sref17"><person-group person-group-type="author"><name><surname>Clay</surname><given-names>D.C.</given-names></name><name><surname>Vander Haar</surname><given-names>J.E.</given-names></name></person-group><article-title>Patterns of intergenerational support and childbearing in the third world</article-title><source>Popul. Stud.</source><volume>47</volume><year>1993</year><fpage>67</fpage><lpage>83</lpage></element-citation></ref><ref id="bib18"><element-citation publication-type="journal" id="sref18"><person-group person-group-type="author"><name><surname>Coovadia</surname><given-names>H.</given-names></name><name><surname>Jewkes</surname><given-names>R.</given-names></name><name><surname>Barron</surname><given-names>P.</given-names></name><name><surname>Sanders</surname><given-names>D.</given-names></name><name><surname>McIntyre</surname><given-names>D.</given-names></name></person-group><article-title>The health and health system of South Africa: historical roots of current public health challenges</article-title><source>Lancet</source><volume>374</volume><year>2009</year><fpage>817</fpage><lpage>834</lpage><pub-id pub-id-type="pmid">19709728</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal" id="sref19"><person-group person-group-type="author"><name><surname>Crews</surname><given-names>F.</given-names></name><name><surname>He</surname><given-names>J.</given-names></name><name><surname>Hodge</surname><given-names>C.</given-names></name></person-group><article-title>Adolescent cortical development: a critical period of vulnerability for addiction</article-title><source>Pharmacol. Biochem. Behav.</source><volume>86</volume><year>2007</year><fpage>189</fpage><lpage>199</lpage><pub-id pub-id-type="pmid">17222895</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal" id="sref20"><person-group person-group-type="author"><name><surname>Cunha</surname><given-names>F.</given-names></name><name><surname>Heckman</surname><given-names>J.</given-names></name></person-group><article-title>The technology of skill formation</article-title><source>Am. Econ. Rev.</source><volume>97</volume><year>2007</year><fpage>31</fpage><lpage>47</lpage></element-citation></ref><ref id="bib21"><element-citation publication-type="journal" id="sref21"><person-group person-group-type="author"><name><surname>Cutler</surname><given-names>D.M.</given-names></name><name><surname>Lleras-Muney</surname><given-names>A.</given-names></name></person-group><article-title>Understanding differences in health behaviors by education</article-title><source>J. Health Econ.</source><volume>29</volume><year>2010</year><fpage>1</fpage><lpage>28</lpage><pub-id pub-id-type="pmid">19963292</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal" id="sref22"><person-group person-group-type="author"><name><surname>Dahl</surname><given-names>R.E.</given-names></name></person-group><article-title>Adolescent brain development: a period of vulnerabilities and opportunities. Keynote address</article-title><source>Ann. N. Y. Acad. Sci.</source><volume>1021</volume><year>2004</year><fpage>1</fpage><lpage>22</lpage><pub-id pub-id-type="pmid">15251869</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal" id="sref23"><person-group person-group-type="author"><name><surname>De Neve</surname><given-names>J.-W.</given-names></name><name><surname>Fink</surname><given-names>G.</given-names></name><name><surname>Subramanian</surname><given-names>S.V.</given-names></name><name><surname>Moyo</surname><given-names>S.</given-names></name><name><surname>Bor</surname><given-names>J.</given-names></name></person-group><article-title>Length of secondary schooling and risk of HIV infection in Botswana: evidence from a natural experiment</article-title><source>Lancet Glob. Health</source><volume>3</volume><year>2015</year><fpage>e470</fpage><lpage>e477</lpage><pub-id pub-id-type="pmid">26134875</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal" id="sref24"><person-group person-group-type="author"><name><surname>Ding</surname><given-names>P.</given-names></name><name><surname>VanderWeele</surname><given-names>T.J.</given-names></name></person-group><article-title>Sensitivity analysis without assumptions</article-title><source>Epidemiology</source><volume>27</volume><year>2016</year><fpage>368</fpage><lpage>377</lpage><pub-id pub-id-type="pmid">26841057</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal" id="sref25"><person-group person-group-type="author"><name><surname>Downey</surname><given-names>D.B.</given-names></name></person-group><article-title>When bigger is not better: family size, parental resources, and Children's educational performance</article-title><source>Am. Sociol. Rev.</source><volume>60</volume><year>1995</year><fpage>746</fpage><lpage>761</lpage></element-citation></ref><ref id="bib75"><element-citation publication-type="journal" id="sref75"><person-group person-group-type="author"><name><surname>Epstein</surname><given-names>E.</given-names></name><name><surname>Guttman</surname><given-names>R.</given-names></name></person-group><article-title>Mate selection in man: evidence, theory, and outcome</article-title><source>Soc. Biol.</source><volume>31</volume><issue>3-4</issue><year>1984</year><fpage>243</fpage><lpage>278</lpage><pub-id pub-id-type="pmid">6400148</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="book" id="sref26"><person-group person-group-type="author"><name><surname>Evans</surname><given-names>R.</given-names></name><name><surname>Becker</surname><given-names>S.</given-names></name></person-group><chapter-title>Children Caring for Parents with HIV and AIDS : Global Issues and Policy Responses</chapter-title><year>2009</year><publisher-name>Policy Press</publisher-name><publisher-loc>Bristol, UK ; Portland, OR</publisher-loc></element-citation></ref><ref id="bib27"><element-citation publication-type="book" id="sref27"><person-group person-group-type="author"><name><surname>Field</surname><given-names>E.</given-names></name><name><surname>de la Roca</surname><given-names>J.</given-names></name></person-group><chapter-title>Are There Upward Intergenerational Education Spillovers? the Influence of Children's Schooling on Parents' Smoking Cessation</chapter-title><year>2005</year><publisher-name>Harvard University</publisher-name><publisher-loc>Cambridge, MA</publisher-loc></element-citation></ref><ref id="bib28"><element-citation publication-type="journal" id="sref28"><person-group person-group-type="author"><name><surname>Frankenberg</surname><given-names>E.</given-names></name><name><surname>Lillard</surname><given-names>L.</given-names></name><name><surname>Willis</surname><given-names>R.J.</given-names></name></person-group><article-title>Patterns of intergenerational transfers in Southeast Asia</article-title><source>J. Marriage Fam.</source><volume>64</volume><year>2002</year><fpage>627</fpage><lpage>641</lpage></element-citation></ref><ref id="bib29"><element-citation publication-type="journal" id="sref29"><person-group person-group-type="author"><name><surname>Friedman</surname><given-names>E.M.</given-names></name><name><surname>Mare</surname><given-names>R.D.</given-names></name></person-group><article-title>The schooling of offspring and the survival of parents</article-title><source>Demography</source><volume>51</volume><year>2014</year><fpage>1271</fpage><lpage>1293</lpage><pub-id pub-id-type="pmid">24917296</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal" id="sref30"><person-group person-group-type="author"><name><surname>Fritzell</surname><given-names>J.</given-names></name><name><surname>Lennartsson</surname><given-names>C.</given-names></name></person-group><article-title>Financial transfers between generations in Sweden</article-title><source>Ageing Soc.</source><volume>25</volume><year>2005</year><fpage>397</fpage><lpage>414</lpage></element-citation></ref><ref id="bib31"><element-citation publication-type="journal" id="sref31"><person-group person-group-type="author"><name><surname>Frost</surname><given-names>M.B.</given-names></name><name><surname>Forste</surname><given-names>R.</given-names></name><name><surname>Haas</surname><given-names>D.W.</given-names></name></person-group><article-title>Maternal education and child nutritional status in Bolivia: finding the links</article-title><source>Soc. Sci. Med.</source><volume>60</volume><year>2005</year><fpage>395</fpage><lpage>407</lpage><pub-id pub-id-type="pmid">15522494</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal" id="sref32"><person-group person-group-type="author"><name><surname>Gallant</surname><given-names>M.</given-names></name><name><surname>Maticka-Tyndale</surname><given-names>E.</given-names></name></person-group><article-title>School-based HIV prevention programmes for African youth</article-title><source>Soc. Sci. Med.</source><volume>58</volume><year>2004</year><fpage>1337</fpage><lpage>1351</lpage><pub-id pub-id-type="pmid">14759680</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal" id="sref33"><person-group person-group-type="author"><name><surname>Giles</surname><given-names>J.</given-names></name><name><surname>Wang</surname><given-names>D.</given-names></name><name><surname>Zhao</surname><given-names>C.</given-names></name></person-group><article-title>Can China's rural elderly count on support from adult Children? Implications of rural-to-urban migration</article-title><source>J. Popul. Ageing</source><volume>3</volume><year>2011</year><fpage>183</fpage><lpage>204</lpage></element-citation></ref><ref id="bib34"><element-citation publication-type="journal" id="sref34"><person-group person-group-type="author"><name><surname>Glewwe</surname><given-names>P.</given-names></name></person-group><article-title>Why does Mother's schooling raise child health in developing Countries? Evidence from Morocco</article-title><source>J. Hum. Resour.</source><volume>34</volume><year>1999</year><fpage>124</fpage><lpage>159</lpage></element-citation></ref><ref id="bib35"><element-citation publication-type="book" id="sref35"><person-group person-group-type="author"><name><surname>Government of South Africa</surname></name></person-group><chapter-title>District Health Plan 2015/16</chapter-title><year>2015</year><publisher-name>Department of Health</publisher-name><publisher-loc>Kwazulu-Natal, South Africa</publisher-loc></element-citation></ref><ref id="bib36"><element-citation publication-type="journal" id="sref36"><person-group person-group-type="author"><name><surname>Grambsch</surname><given-names>P.M.</given-names></name><name><surname>Therneau</surname><given-names>T.M.</given-names></name></person-group><article-title>Proportional hazards tests and diagnostics based on weighted residuals</article-title><source>Biometrika</source><volume>81</volume><year>1994</year><fpage>515</fpage><lpage>526</lpage></element-citation></ref><ref id="bib37"><element-citation publication-type="journal" id="sref37"><person-group person-group-type="author"><name><surname>Grépin</surname><given-names>K.A.</given-names></name><name><surname>Bharadwaj</surname><given-names>P.</given-names></name></person-group><article-title>Maternal education and child mortality in Zimbabwe</article-title><source>J. Health Econ.</source><volume>44</volume><year>2015</year><fpage>97</fpage><lpage>117</lpage><pub-id pub-id-type="pmid">26569469</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal" id="sref38"><person-group person-group-type="author"><name><surname>Grossman</surname><given-names>M.</given-names></name></person-group><article-title>On the concept of health capital and the demand for health</article-title><source>J. Political Econ.</source><volume>80</volume><year>1972</year><fpage>223</fpage><lpage>255</lpage></element-citation></ref><ref id="bib39"><element-citation publication-type="journal" id="sref39"><person-group person-group-type="author"><name><surname>Herbst</surname><given-names>A.J.</given-names></name><name><surname>Mafojane</surname><given-names>T.</given-names></name><name><surname>Newell</surname><given-names>M.L.</given-names></name></person-group><article-title>Verbal autopsy-based cause-specific mortality trends in rural KwaZulu-Natal, South Africa, 2000-2009</article-title><source>Popul. Health Metr.</source><volume>9</volume><year>2011</year><fpage>47</fpage><pub-id pub-id-type="pmid">21819602</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="book" id="sref40"><person-group person-group-type="author"><name><surname>Hosegood</surname><given-names>V.</given-names></name><name><surname>Timaeus</surname><given-names>I.</given-names></name></person-group><chapter-title>Household composition and dynamics in Kwazulu Natal, South Africa: mirroring social reality in longitudinal data collection</chapter-title><person-group person-group-type="editor"><name><surname>Van de Walle</surname><given-names>E.</given-names></name></person-group><source>African Households: Censuses and Surveys</source><year>2006</year><publisher-name>M.E. Sharpe</publisher-name><publisher-loc>Armonk, N.Y.</publisher-loc><fpage>58</fpage><lpage>77</lpage></element-citation></ref><ref id="bib41"><mixed-citation publication-type="other" id="sref41">Karim, Q.A., Leask, K., Kharsany, A., et al., 2015. Impact of conditional cash incentives on HSV-2 and HIV prevention in rural South African high school students: results of CAPRISA 007 cluster randomized trial. International AIDS Conference, Vancouver, Canada.</mixed-citation></ref><ref id="bib42"><element-citation publication-type="journal" id="sref42"><person-group person-group-type="author"><name><surname>Kaufman</surname><given-names>J.S.</given-names></name><name><surname>Cooper</surname><given-names>R.S.</given-names></name></person-group><article-title>Seeking causal explanations in social epidemiology</article-title><source>Am. J. Epidemiol.</source><volume>150</volume><year>1999</year><fpage>113</fpage><lpage>120</lpage><pub-id pub-id-type="pmid">10412955</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal" id="sref43"><person-group person-group-type="author"><name><surname>Knodel</surname><given-names>J.</given-names></name><name><surname>Friedman</surname><given-names>J.</given-names></name><name><surname>Anh</surname><given-names>T.S.</given-names></name><name><surname>Cuong</surname><given-names>B.T.</given-names></name></person-group><article-title>Intergenerational exchanges in Vietnam: family size, sex composition, and the location of children</article-title><source>Popul. Stud.</source><volume>54</volume><year>2000</year><fpage>89</fpage><lpage>104</lpage></element-citation></ref><ref id="bib44"><element-citation publication-type="journal" id="sref44"><person-group person-group-type="author"><name><surname>Kohler</surname><given-names>I.V.</given-names></name><name><surname>Kohler</surname><given-names>H.P.</given-names></name><name><surname>Anglewicz</surname><given-names>P.</given-names></name><name><surname>Behrman</surname><given-names>J.R.</given-names></name></person-group><article-title>Intergenerational transfers in the era of HIV/AIDS: evidence from rural Malawi</article-title><source>Demogr. Res.</source><volume>27</volume><year>2012</year><fpage>775</fpage><lpage>834</lpage><pub-id pub-id-type="pmid">23606809</pub-id></element-citation></ref><ref id="bib45"><element-citation publication-type="journal" id="sref45"><person-group person-group-type="author"><name><surname>Krieger</surname><given-names>N.</given-names></name></person-group><article-title>Theories for social epidemiology in the 21st century: an ecosocial perspective</article-title><source>Int. J. Epidemiol.</source><volume>30</volume><year>2001</year><fpage>668</fpage><lpage>677</lpage><pub-id pub-id-type="pmid">11511581</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal" id="sref46"><person-group person-group-type="author"><name><surname>Lau</surname><given-names>B.</given-names></name><name><surname>Cole</surname><given-names>S.R.</given-names></name><name><surname>Gange</surname><given-names>S.J.</given-names></name></person-group><article-title>Competing risk regression models for epidemiologic data</article-title><source>Am. J. Epidemiol.</source><volume>170</volume><year>2009</year><fpage>244</fpage><lpage>256</lpage><pub-id pub-id-type="pmid">19494242</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal" id="sref47"><person-group person-group-type="author"><name><surname>Lennartsson</surname><given-names>C.</given-names></name><name><surname>Silverstein</surname><given-names>M.</given-names></name><name><surname>Fritzell</surname><given-names>J.</given-names></name></person-group><article-title>Time-for-Money exchanges between older and younger generations in Swedish families</article-title><source>J. Fam. Issues</source><volume>31</volume><year>2009</year><fpage>189</fpage><lpage>210</lpage></element-citation></ref><ref id="bib48"><element-citation publication-type="book" id="sref48"><person-group person-group-type="author"><name><surname>Lundborg</surname><given-names>P.</given-names></name><name><surname>Majlesi</surname><given-names>K.</given-names></name></person-group><chapter-title>Intergenerational Transmission of Human Capital: Is it a One-way Street?</chapter-title><year>2015</year><publisher-name>Institute for the Study of Labor</publisher-name><publisher-loc>Discussion Paper No. 9280. Bonn</publisher-loc></element-citation></ref><ref id="bib49"><element-citation publication-type="journal" id="sref49"><person-group person-group-type="author"><name><surname>Marmot</surname><given-names>M.G.</given-names></name></person-group><article-title>Understanding social inequalities in health</article-title><source>Perspect. Biol. Med.</source><volume>46</volume><year>2003</year><fpage>S9</fpage><lpage>S23</lpage><pub-id pub-id-type="pmid">14563071</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal" id="sref50"><person-group person-group-type="author"><name><surname>McGarry</surname><given-names>K.</given-names></name></person-group><article-title>Caring for the elderly: the role of adult children</article-title><source>Inq. Econ. Aging</source><year>1998</year><fpage>133</fpage><lpage>166</lpage></element-citation></ref><ref id="bib76"><element-citation publication-type="book" id="sref76"><person-group person-group-type="author"><name><surname>Mincer</surname><given-names>J.</given-names></name></person-group><chapter-title>Schooling, Experience, and Earnings</chapter-title><year>1974</year><publisher-name>Columbia University Press for the National Bureau of Economic Research</publisher-name><publisher-loc>New York</publisher-loc><fpage>7</fpage><lpage>22</lpage></element-citation></ref><ref id="bib51"><element-citation publication-type="journal" id="sref51"><person-group person-group-type="author"><name><surname>Morris</surname><given-names>S.S.</given-names></name></person-group><article-title>Validity of rapid estimates of household wealth and income for health surveys in rural Africa</article-title><source>J. Epidemiol. Community Health</source><volume>54</volume><year>2000</year><fpage>381</fpage><lpage>387</lpage><pub-id pub-id-type="pmid">10814660</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="book" id="sref52"><person-group person-group-type="author"><name><surname>PEPFAR</surname></name></person-group><chapter-title>Working together for an AIDS-free future for girls and women</chapter-title><year>2016</year></element-citation></ref><ref id="bib53"><element-citation publication-type="journal" id="sref53"><person-group person-group-type="author"><name><surname>Pettifor</surname><given-names>A.</given-names></name><name><surname>MacPhail</surname><given-names>C.</given-names></name><name><surname>Hughes</surname><given-names>J.P.</given-names></name><name><surname>Selin</surname><given-names>A.</given-names></name><name><surname>Wang</surname><given-names>J.</given-names></name><name><surname>Gómez-Olivé</surname><given-names>F.X.</given-names></name></person-group><article-title>The effect of a conditional cash transfer on HIV incidence in young women in rural South Africa (HPTN 068): a phase 3, randomised controlled trial</article-title><source>Lancet Glob. Health</source><volume>4</volume><year>2016</year><fpage>e978</fpage><lpage>e988</lpage><pub-id pub-id-type="pmid">27815148</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal" id="sref54"><person-group person-group-type="author"><name><surname>Psacharopoulos</surname><given-names>G.</given-names></name></person-group><article-title>Returns to investment in education: a global update</article-title><source>World Dev.</source><volume>22</volume><year>1994</year><fpage>1325</fpage><lpage>1343</lpage></element-citation></ref><ref id="bib55"><element-citation publication-type="journal" id="sref55"><person-group person-group-type="author"><name><surname>Reniers</surname><given-names>G.</given-names></name><name><surname>Blom</surname><given-names>S.</given-names></name><name><surname>Calvert</surname><given-names>C.</given-names></name><name><surname>Martin-Onraet</surname><given-names>A.</given-names></name><name><surname>Herbst</surname><given-names>A.J.</given-names></name><name><surname>Eaton</surname><given-names>J.W.</given-names></name></person-group><article-title>Trends in the burden of HIV mortality after roll-out of antiretroviral therapy in KwaZulu-Natal, South Africa: an observational community cohort study</article-title><source>Lancet HIV</source><year>2016</year></element-citation></ref><ref id="bib56"><element-citation publication-type="book" id="sref56"><person-group person-group-type="author"><name><surname>Republic of South Africa</surname></name></person-group><chapter-title>Macro Indicator Report</chapter-title><year>2013</year><publisher-name>Department of Basic Education. Republic of South Africa</publisher-name><isbn>978-1-4315-1864-7</isbn></element-citation></ref><ref id="bib57"><element-citation publication-type="journal" id="sref57"><person-group person-group-type="author"><name><surname>Rowa-Dewar</surname><given-names>N.</given-names></name><name><surname>Amos</surname><given-names>A.</given-names></name><name><surname>Cunningham-Burley</surname><given-names>S.</given-names></name></person-group><article-title>Children's resistance to parents' smoking in the home and car: a qualitative study</article-title><source>Addiction</source><volume>109</volume><year>2014</year><fpage>645</fpage><lpage>652</lpage><pub-id pub-id-type="pmid">24304201</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="book" id="sref58"><person-group person-group-type="author"><name><surname>Saad</surname><given-names>P.M.</given-names></name></person-group><chapter-title>Informal Support Transfers of the Elderly in Brazil and Latin America. Sixty Plus: the Elderly Brazilians and Their New Social Roles</chapter-title><year>2005</year><publisher-name>Institute of Applied Economic Research</publisher-name><publisher-loc>Rio de Janeiro</publisher-loc></element-citation></ref><ref id="bib59"><element-citation publication-type="journal" id="sref59"><person-group person-group-type="author"><name><surname>Sabater</surname><given-names>A.</given-names></name><name><surname>Graham</surname><given-names>E.</given-names></name></person-group><article-title>The role of Children's education for the mental health of aging migrants in Europe</article-title><source>GeroPsych</source><volume>29</volume><year>2016</year><fpage>81</fpage><lpage>92</lpage></element-citation></ref><ref id="bib60"><element-citation publication-type="journal" id="sref60"><person-group person-group-type="author"><name><surname>Sartorius</surname><given-names>B.</given-names></name><name><surname>Kahn</surname><given-names>K.</given-names></name><name><surname>Collinson</surname><given-names>M.A.</given-names></name><name><surname>Sartorius</surname><given-names>K.</given-names></name><name><surname>Tollman</surname><given-names>S.M.</given-names></name></person-group><article-title>Dying in their prime: determinants and space-time risk of adult mortality in rural South Africa</article-title><source>Geospatial health</source><volume>7</volume><year>2013</year><fpage>237</fpage><pub-id pub-id-type="pmid">23733287</pub-id></element-citation></ref><ref id="bib61"><element-citation publication-type="journal" id="sref61"><person-group person-group-type="author"><name><surname>Schultz</surname><given-names>T.P.</given-names></name></person-group><article-title>Evidence of returns to schooling in Africa from household surveys: monitoring and restructuring the market for education</article-title><source>J. Afr. Econ.</source><volume>13</volume><year>2004</year><fpage>ii95</fpage><lpage>ii148</lpage></element-citation></ref><ref id="bib62"><element-citation publication-type="journal" id="sref62"><person-group person-group-type="author"><name><surname>Steinberg</surname><given-names>L.</given-names></name></person-group><article-title>Cognitive and affective development in adolescence</article-title><source>Trends Cognitive Sci.</source><volume>9</volume><year>2005</year><fpage>69</fpage><lpage>74</lpage></element-citation></ref><ref id="bib63"><element-citation publication-type="journal" id="sref63"><person-group person-group-type="author"><name><surname>Sun</surname><given-names>F.</given-names></name><name><surname>Sebastiani</surname><given-names>P.</given-names></name><name><surname>Schupf</surname><given-names>N.</given-names></name><name><surname>Bae</surname><given-names>H.</given-names></name><name><surname>Andersen</surname><given-names>S.L.</given-names></name><name><surname>McIntosh</surname><given-names>A.</given-names></name></person-group><article-title>Extended maternal age at birth of last child and women's longevity in the Long Life Family Study</article-title><source>Menopause</source><volume>22</volume><year>2015</year><fpage>26</fpage><lpage>31</lpage><pub-id pub-id-type="pmid">24977462</pub-id></element-citation></ref><ref id="bib64"><element-citation publication-type="journal" id="sref64"><person-group person-group-type="author"><name><surname>Tamakoshi</surname><given-names>A.</given-names></name><name><surname>Tamakoshi</surname><given-names>K.</given-names></name><name><surname>Lin</surname><given-names>Y.</given-names></name><name><surname>Mikami</surname><given-names>H.</given-names></name><name><surname>Inaba</surname><given-names>Y.</given-names></name><name><surname>Yagyu</surname><given-names>K.</given-names></name></person-group><article-title>Number of children and all-cause mortality risk: results from the Japan Collaborative Cohort Study</article-title><source>Eur. J. Public Health</source><volume>21</volume><year>2010</year><fpage>732</fpage><lpage>737</lpage><pub-id pub-id-type="pmid">21113028</pub-id></element-citation></ref><ref id="bib65"><element-citation publication-type="journal" id="sref65"><person-group person-group-type="author"><name><surname>Tanser</surname><given-names>F.</given-names></name><name><surname>Hosegood</surname><given-names>V.</given-names></name><name><surname>Barnighausen</surname><given-names>T.</given-names></name><name><surname>Herbst</surname><given-names>K.</given-names></name><name><surname>Nyirenda</surname><given-names>M.</given-names></name><name><surname>Muhwava</surname><given-names>W.</given-names></name></person-group><article-title>Cohort profile: Africa Centre demographic information system (ACDIS) and population-based HIV survey</article-title><source>Int. J. Epidemiol.</source><volume>37</volume><year>2008</year><fpage>956</fpage><lpage>962</lpage><pub-id pub-id-type="pmid">17998242</pub-id></element-citation></ref><ref id="bib66"><element-citation publication-type="journal" id="sref66"><person-group person-group-type="author"><name><surname>Torssander</surname><given-names>J.</given-names></name></person-group><article-title>From child to parent? The significance of children's education for their parents' longevity</article-title><source>Demography</source><volume>50</volume><year>2013</year><fpage>637</fpage><lpage>659</lpage><pub-id pub-id-type="pmid">23055239</pub-id></element-citation></ref><ref id="bib67"><element-citation publication-type="journal" id="sref67"><person-group person-group-type="author"><name><surname>Torssander</surname><given-names>J.</given-names></name></person-group><article-title>Adult children's socioeconomic positions and their parents' mortality: a comparison of education, occupational class, and income</article-title><source>Soc. Sci. Med.</source><volume>122</volume><year>2014</year><fpage>148</fpage><lpage>156</lpage><pub-id pub-id-type="pmid">25441327</pub-id></element-citation></ref><ref id="bib68"><element-citation publication-type="book" id="sref68"><person-group person-group-type="author"><name><surname>UNESCO Institute for Statistics</surname></name></person-group><chapter-title>Education Statistics South Africa</chapter-title><year>2016</year></element-citation></ref><ref id="bib69"><element-citation publication-type="book" id="sref69"><person-group person-group-type="author"><name><surname>VanderWeele</surname><given-names>T.</given-names></name></person-group><chapter-title>Explanation in Causal Inference: Methods for Mediation and Interaction</chapter-title><year>2015</year><publisher-name>Oxford Univ. Press</publisher-name><publisher-loc>New York</publisher-loc></element-citation></ref><ref id="bib70"><element-citation publication-type="book" id="sref70"><person-group person-group-type="author"><name><surname>Wilde</surname><given-names>E.T.</given-names></name><name><surname>Batchelder</surname><given-names>L.</given-names></name><name><surname>Ellwood</surname><given-names>D.</given-names></name></person-group><chapter-title>The Mommy Track Divides: the Impact of Childbearing on Wages of Women of Differing Skill Levels</chapter-title><year>2010</year><comment>NBER Working Paper No. 16582</comment></element-citation></ref><ref id="bib71"><element-citation publication-type="journal" id="sref71"><person-group person-group-type="author"><name><surname>Yahirun</surname><given-names>J.J.</given-names></name><name><surname>Sheehan</surname><given-names>C.M.</given-names></name><name><surname>Hayward</surname><given-names>M.D.</given-names></name></person-group><article-title>Adult childrens education and parents functional limitations in Mexico</article-title><source>Res. Aging</source><volume>38</volume><year>2016</year><fpage>322</fpage><lpage>345</lpage><pub-id pub-id-type="pmid">26966254</pub-id></element-citation></ref><ref id="bib72"><element-citation publication-type="journal" id="sref72"><person-group person-group-type="author"><name><surname>Yang</surname><given-names>L.</given-names></name><name><surname>Martikainen</surname><given-names>P.</given-names></name><name><surname>Silventoinen</surname><given-names>K.</given-names></name></person-group><article-title>Effects of individual, spousal, and offspring socioeconomic status on mortality among elderly people in China</article-title><source>J. Epidemiol.</source><volume>26</volume><issue>11</issue><year>2016</year><fpage>602</fpage><lpage>609</lpage><pub-id pub-id-type="pmid">27150012</pub-id></element-citation></ref><ref id="bib73"><element-citation publication-type="journal" id="sref73"><person-group person-group-type="author"><name><surname>Zimmer</surname><given-names>Z.</given-names></name><name><surname>Martin</surname><given-names>L.G.</given-names></name><name><surname>Ofstedal</surname><given-names>M.B.</given-names></name><name><surname>Chuang</surname><given-names>Y.L.</given-names></name></person-group><article-title>Education of adult children and mortality of their elderly parents in Taiwan</article-title><source>Demography</source><volume>44</volume><year>2007</year><fpage>289</fpage><lpage>305</lpage><pub-id pub-id-type="pmid">17583306</pub-id></element-citation></ref></ref-list><sec id="appsec1" sec-type="supplementary-material"><label>Appendix A</label><title>Supplementary data</title><p>The following is the supplementary data related to this article:<supplementary-material content-type="local-data" id="mmc1"><caption><title>Supplementary Tables</title></caption><media xlink:href="mmc1.docx"><alt-text>Supplementary Tables</alt-text></media></supplementary-material></p></sec><ack id="ack0010"><title>Acknowledgements</title><p>We thank the field and data management staff of the Africa Centre in South Africa and the respondents who gave their time. We thank Tyler VanderWeele as well as three anonymous reviewers for helpful comments. The authors report no conflicts of interest.</p></ack><fn-group><fn id="appsec2" fn-type="supplementary-material"><label>Appendix A</label><p>Supplementary data related to this article can be found at <ext-link ext-link-type="doi" xlink:href="10.1016/j.socscimed.2017.01.015" id="intref0010">http://dx.doi.org/10.1016/j.socscimed.2017.01.015</ext-link>.</p></fn></fn-group></back><floats-group><fig id="fig1"><label>Fig. 1</label><caption><p>Conceptual hierarchical framework for the relationship between offspring schooling and parental survival.</p><p>Figure shows the inter-relationship between variables in our study. The arrows represent the potential (causal) effect of the relevant explanatory factor. Offspring schooling attainment exerts its effect on parental survival through: (a) improved labor market outcomes, wealth, assortative mating with better educated spouses, changes in area of residence, and (b) more proximate determinants such as direct transfers of health knowledge and care for parents (<xref rid="bib38" ref-type="bibr">Grossman 1972</xref>; <xref rid="bib76" ref-type="bibr">Mincer, 1974</xref>, <xref rid="bib74" ref-type="bibr">Bongaarts, 1978</xref>, <xref rid="bib75" ref-type="bibr">Epstein and Guttman, 1984</xref>). In our study, we control for parental and offspring characteristics that are likely confounders, and baseline values of characteristics that are likely on the causal pathway between offspring schooling and parental survival. Our estimates can thus be interpreted as an estimate of the total effects of offspring schooling on parental survival because we allow likely mechanisms to vary during follow-up (pathways a and b).</p></caption><alt-text id="alttext0015">Fig. 1</alt-text><graphic xlink:href="gr1"></graphic></fig><fig id="fig2"><label>Fig. 2</label><caption><p>Nelson-Aalen cumulative hazard of parental death, stratified by offspring's schooling.</p></caption><alt-text id="alttext0020">Fig. 2</alt-text><graphic xlink:href="gr2"></graphic></fig><table-wrap id="tbl1" position="float"><label>Table 1</label><caption><p>Baseline characteristics of the study sample (complete case dataset).</p></caption><alt-text id="alttext0025">Table 1</alt-text><table frame="hsides" rules="groups"><thead><tr><th></th><th>Mothers</th><th>Fathers</th></tr></thead><tbody><tr><td>No. of mothers/fathers</td><td>12,105</td><td>5684</td></tr><tr><td>No. of subsequent maternal/paternal deaths</td><td>1886</td><td>1344</td></tr><tr><td>Age (years)</td><td>49 (43–61)</td><td>52 (47–62)</td></tr><tr><td>Schooling (years)</td><td>4 (0–9)</td><td>5 (0–11)</td></tr><tr><td colspan="3">Highest schooling attainment</td></tr><tr><td> 0 years</td><td>3981 (33)</td><td>1803 (32)</td></tr><tr><td> 1–7 years</td><td>4189 (35)</td><td>1749 (31)</td></tr><tr><td> 8–9 years</td><td>924 (8)</td><td>417 (7)</td></tr><tr><td> 10–12 years</td><td>2006 (17)</td><td>923 (16)</td></tr><tr><td> 13+ years</td><td>1005 (8)</td><td>792 (14)</td></tr><tr><td colspan="3">Household wealth quintile</td></tr><tr><td> Lowest</td><td>3076 (25)</td><td>1319 (23)</td></tr><tr><td> 2nd lowest</td><td>3101 (26)</td><td>1475 (26)</td></tr><tr><td> Middle</td><td>2335 (19)</td><td>1074 (19)</td></tr><tr><td> 2nd highest</td><td>1800 (15)</td><td>828 (15)</td></tr><tr><td> Highest</td><td>1793 (14)</td><td>988 (17)</td></tr><tr><td>Married</td><td>3773 (31)</td><td>3843 (68)</td></tr><tr><td colspan="3">Area</td></tr><tr><td> Rural</td><td>7802 (64)</td><td>3972 (70)</td></tr><tr><td> Peri-urban</td><td>3523 (29)</td><td>1417 (25)</td></tr><tr><td> Urban</td><td>780 (6)</td><td>295 (5)</td></tr><tr><td>No. of children</td><td>3 (2–5)</td><td>4 (2–5)</td></tr><tr><td>Offspring's highest schooling (years)</td><td>12 (10–12)</td><td>12 (10–12)</td></tr><tr><td colspan="3">Offspring's highest schooling attainment</td></tr><tr><td> 0 years</td><td>285 (2)</td><td>50 (1)</td></tr><tr><td> 1–7 years</td><td>937 (8)</td><td>291 (5)</td></tr><tr><td> 8–9 years</td><td>1061 (9)</td><td>508 (9)</td></tr><tr><td> 10–12 years</td><td>7883 (65)</td><td>3922 (69)</td></tr><tr><td> 13+ years</td><td>1939 (16)</td><td>913 (16)</td></tr></tbody></table><table-wrap-foot><fn><p>Figures for categorical data are N (%); figures for continuous data are medians and interquartile ranges in parentheses.</p></fn></table-wrap-foot></table-wrap><table-wrap id="tbl2" position="float"><label>Table 2</label><caption><p>Sample characteristics of parents followed over time.</p></caption><alt-text id="alttext0030">Table 2</alt-text><table frame="hsides" rules="groups"><thead><tr><th></th><th>Person-years</th><th>% of person-years</th><th>Parental deaths (cases)</th><th>Incidence rate (per 100 person-years)</th></tr></thead><tbody><tr><td colspan="5" align="center"><italic>Maternal mortality</italic></td></tr><tr><td>All</td><td>81,653</td><td>100.0</td><td>1886</td><td>2.31 (2.21–2.42)</td></tr><tr><td colspan="5">Offspring's highest schooling attainment</td></tr><tr><td> 0 years</td><td>1902</td><td>2.3</td><td>127</td><td>6.73 (5.66–8.00)</td></tr><tr><td> 1–7 years</td><td>5063</td><td>6.2</td><td>287</td><td>5.67 (5.05–6.36)</td></tr><tr><td> 8–9 years</td><td>4670</td><td>5.7</td><td>199</td><td>4.26 (3.71–4.90)</td></tr><tr><td> 10–12 years</td><td>53,043</td><td>65.0</td><td>1022</td><td>1.93 (1.81–2.05)</td></tr><tr><td> 13+ years</td><td>16,975</td><td>20.8</td><td>251</td><td>1.48 (1.31–1.67)</td></tr><tr><td colspan="5">
</td></tr><tr><td colspan="5" align="center"><italic>Paternal mortality</italic></td></tr><tr><td>All</td><td>36,001</td><td>100.0</td><td>1344</td><td>3.37 (3.53–3.93)</td></tr><tr><td colspan="5">Offspring's highest schooling attainment</td></tr><tr><td> 0 years</td><td>316</td><td>0.9</td><td>39</td><td>12.0 (8.74–16.51)</td></tr><tr><td> 1–7 years</td><td>1311</td><td>3.6</td><td>108</td><td>8.23 (6.82–9.94)</td></tr><tr><td> 8–9 years</td><td>1946</td><td>5.4</td><td>107</td><td>5.50 (4.55–6.64)</td></tr><tr><td> 10–12 years</td><td>24,442</td><td>67.9</td><td>865</td><td>3.54 (3.31–3.78)</td></tr><tr><td> 13+ years</td><td>7985</td><td>22.2</td><td>225</td><td>2.82 (2.47–3.21)</td></tr></tbody></table><table-wrap-foot><fn><p>The sample included 12,105 mothers and 5684 fathers followed over the years 2003–2015. Results for additional variables are shown in <xref rid="appsec1" ref-type="sec">Supplementary Tables</xref>.</p></fn></table-wrap-foot></table-wrap><table-wrap id="tbl3" position="float"><label>Table 3</label><caption><p>Survival models showing the association between offspring's schooling and hazard of parental death.</p></caption><alt-text id="alttext0035">Table 3</alt-text><table frame="hsides" rules="groups"><thead><tr><th></th><th>Bivariate analysis HR (95% CI)</th><th>Multivariable analysis aHR (95% CI)</th></tr></thead><tbody><tr><td colspan="3" align="center"><italic>Maternal hazard of death</italic></td></tr><tr><td>Model 1: Offspring highest schooling (years)</td><td>0.88 (0.87–0.89)</td><td>0.95 (0.94–0.97)</td></tr><tr><td colspan="3">Model 2: Offspring's highest schooling attainment</td></tr><tr><td> 0–7 years</td><td>1</td><td>1</td></tr><tr><td> 8–9 years</td><td>0.71 (0.60–0.84)</td><td>1.08 (0.90–1.29)</td></tr><tr><td> 10–12 years</td><td>0.32 (0.29–0.36)</td><td>0.74 (0.64–0.85)</td></tr><tr><td> 13+ years</td><td>0.25 (0.21–0.29)</td><td>0.58 (0.48–0.70)</td></tr><tr><td colspan="3">
</td></tr><tr><td>No. of mothers</td><td>12,105</td><td>12,105</td></tr><tr><td colspan="3">
</td></tr><tr><td colspan="3" align="center"><italic>Paternal hazard of death</italic></td></tr><tr><td>Model 1: Offspring highest schooling (years)</td><td>0.88 (0.87–0.90)</td><td>0.94 (0.92–0.96)</td></tr><tr><td colspan="3">Model 2: Offspring's highest schooling attainment</td></tr><tr><td> 0–7 years</td><td>1</td><td>1</td></tr><tr><td> 8–9 years</td><td>0.60 (0.48–0.78)</td><td>0.86 (0.66–1.11)</td></tr><tr><td> 10–12 years</td><td>0.39 (0.32–0.46)</td><td>0.65 (0.54–0.80)</td></tr><tr><td> 13+ years</td><td>0.30 (0.25–0.38)</td><td>0.57 (0.44–0.72)</td></tr><tr><td colspan="3">
</td></tr><tr><td>No. of fathers</td><td>5684</td><td>5684</td></tr></tbody></table><table-wrap-foot><fn><p>Output from separate multivariable survival models including model 1 (continuous variable for schooling attainment) and model 2 (indicator variables for schooling attainment levels). HR: hazard ratio; aHR: hazard ratio adjusted for a linear, quadratic and cubic term in paternal age, parental years of schooling, parental marital status, number of children, age of the oldest child, gender of the oldest child, indicators for year of observation, baseline marital status of the oldest child, baseline employment status of the oldest child and parent, and baseline household wealth and area of residence (rural, peri-urban, urban). CI: confidence interval. Results for additional variables are shown in <xref rid="appsec1" ref-type="sec">Supplementary Tables</xref>.</p></fn></table-wrap-foot></table-wrap><table-wrap id="tbl4" position="float"><label>Table 4</label><caption><p>Output from separate multivariable survival models by parental cause of death.</p></caption><alt-text id="alttext0040">Table 4</alt-text><table frame="hsides" rules="groups"><thead><tr><th rowspan="2">Cause of death</th><th>Non-communicable<hr></hr></th><th>AIDS or tuberculosis<hr></hr></th><th>Communicable, maternal, perinatal or nutritional<hr></hr></th><th>Injuries<hr></hr></th></tr><tr><th>aHR (95% CI)</th><th>aHR (95% CI)</th><th>aHR (95% CI)</th><th>aHR (95% CI)</th></tr></thead><tbody><tr><td colspan="5" align="center"><italic>Maternal hazard of death</italic></td></tr><tr><td>Model 1: Offspring highest schooling (years)</td><td>0.97 (0.95–1.00)</td><td>0.92 (0.89–0.96)</td><td>0.87 (0.82–0.92)</td><td>0.96 (0.87–1.05)</td></tr><tr><td colspan="5">Model 2: Offspring's highest schooling attainment</td></tr><tr><td> 0–7 years</td><td>1</td><td>1</td><td>1</td><td>1</td></tr><tr><td> 8–9 years</td><td>1.06 (0.81–1.38)</td><td>1.20 (0.84–1.70)</td><td>0.85 (0.45–1.63)</td><td>0.34 (0.07–1.58)</td></tr><tr><td> 10–12 years</td><td>0.85 (0.70–1.03)</td><td>0.76 (0.56–1.03)</td><td>0.38 (0.22–0.65)</td><td>0.63 (0.27–1.49)</td></tr><tr><td> 13+ years</td><td>0.75 (0.58–0.96)</td><td>0.42 (0.28–0.65)</td><td>0.21 (0.10–0.47)</td><td>0.51 (0.17–1.60)</td></tr><tr><td colspan="5">
</td></tr><tr><td>No. of mothers</td><td>12,105</td><td>12,105</td><td>12,105</td><td>12,105</td></tr><tr><td>Maternal deaths (cases)</td><td>908</td><td>564</td><td>125</td><td>50</td></tr><tr><td colspan="5">
</td></tr><tr><td colspan="5" align="center"><italic>Paternal hazard of death</italic></td></tr><tr><td>Model 1: Offspring highest schooling (years)</td><td>0.95 (0.92–0.98)</td><td>0.92 (0.89–0.96)</td><td>0.93 (0.86–1.01)</td><td>0.87 (0.79–0.95)</td></tr><tr><td colspan="5">Model 2: Offspring's highest schooling attainment</td></tr><tr><td> 0–7 years</td><td>1</td><td>1</td><td>1</td><td>1</td></tr><tr><td> 8–9 years</td><td>0.99 (0.66–1.02)</td><td>0.59 (0.36–0.95)</td><td>1.10 (0.46–2.63)</td><td>1.04 (0.39–2.76)</td></tr><tr><td> 10–12 years</td><td>0.76 (0.56–1.02)</td><td>0.62 (0.44–0.89)</td><td>0.64 (0.33–1.27)</td><td>0.38 (0.16–0.89)</td></tr><tr><td> 13+ years</td><td>0.57 (0.40–0.82)</td><td>0.46 (0.29–0.73)</td><td>0.52 (0.21–1.31)</td><td>0.65 (0.24–1.73)</td></tr><tr><td colspan="5">
</td></tr><tr><td>No. of fathers</td><td>5684</td><td>5684</td><td>5684</td><td>5684</td></tr><tr><td>Paternal deaths (cases)</td><td>569</td><td>427</td><td>88</td><td>92</td></tr></tbody></table><table-wrap-foot><fn><p>Output from separate multivariable survival models including model 1 (continuous variable for schooling attainment) and model 2 (indicator variables for schooling attainment). HR: hazard ratio; aHR: hazard ratio adjusted for a linear, quadratic and cubic term in paternal age, parental years of schooling, parental marital status, number of children, age of the oldest child, gender of the oldest child, indicators for year of observation, baseline marital status of the oldest child, baseline employment status of the oldest child and parent, and baseline household wealth and area of residence (rural, peri-urban, urban). CI: confidence interval.</p></fn></table-wrap-foot></table-wrap></floats-group></pmc></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sante/explor/SidaSubSaharaV1/Data/Pmc/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002C81 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd -nk 002C81 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Sante
|area= SidaSubSaharaV1
|flux= Pmc
|étape= Corpus
|type= RBID
|clé=
|texte=
}}
| This area was generated with Dilib version V0.6.32. |  |