Skip to main content

Potentially Avertable Child Mortality Associated with Surgical Workforce Scale-up in Low- and Middle-Income Countries: A Global Study



Expansion of access to surgical care can improve health outcomes, although the impact that scale-up of the surgical workforce will have on child mortality is poorly defined. In this study, we estimate the number of child deaths potentially avertable by increasing the surgical workforce globally to meet targets proposed by the Lancet Commission on Global Surgery.


To estimate the number of deaths potentially avertable through increases in the surgical workforce, we used log-linear regression to model the association between surgeon, anesthetist and obstetrician workforce (SAO) density and surgically amenable under-5 mortality rate (U5MR), infant mortality rate (IMR), and neonatal mortality rate (NMR) for 192 countries adjusting for potential confounders of childhood mortality, including the non-surgical workforce (physicians, nurses/midwives, community health workers), gross national income per capita, poverty rate, female literacy rate, health expenditure per capita, percentage of urban population, number of surgical operations, and hospital bed density. Surgically amenable mortality was determined using mortality estimates from the UN Inter-agency Group for Child Mortality Estimation adjusted by the proportion of deaths in each country due to communicable causes unlikely to be amenable to surgical care. Estimates of mortality reduction due to upscaling surgical care to support the Lancet Commission on Global Surgery (LCoGS) minimum target of 20–40 SAO/100,000 were calculated accounting for potential increases in surgical volume associated with surgical workforce expansion.


Increasing SAO workforce density was independently associated with lower surgically amenable U5MR as well as NMR (p < 0.01 for each model). When accounting for concomitant increases in surgical volume, scale-up of the surgical workforce to 20–40 SAO/100,000 could potentially prevent between 262,709 (95% CI 229,643–295,434) and 519,629 (465,046–573,919) under 5 deaths annually. The majority (61%) of deaths averted would be neonatal deaths.


Scale up of surgical workforce may substantially decrease childhood mortality rates around the world. Our analysis suggests that scale-up of surgical delivery through increase in the SAO workforce could prevent over 500,000 children from dying before the age of 5 annually. This would represent significant progress toward meeting global child mortality reduction targets.

This is a preview of subscription content, access via your institution.


  1. Meara JG, Leather AJM, Hagander L et al (2015) Global surgery 2030: evidence and solutions for achieving health, welfare, and economic development. Lancet 386:569–624.

    Article  PubMed  Google Scholar 

  2. Mullapudi B, Grabski D, Ameh E et al (2019) Estimates of number of children and adolescents without access to surgical care. Bull World Health Organ 97:254–258

    Article  Google Scholar 

  3. Roa L, Jumbam DT, Makasa E et al (2019) Global surgery and the sustainable development goals. Br J Surg 106:e44-52

    Article  CAS  Google Scholar 

  4. Rocha TAH, Vissoci J, Rocha N et al (2020) Towards defining the surgical workforce for children: a geospatial analysis in Brazil. BMJ Open 10:e034253

    Article  Google Scholar 

  5. Leeming M, Holmer H, Marks IH et al (2018) Defining global pediatric surgical and anesthesia workforce densities for optimal child health outcomes. J Am Coll Surg 227:S128–S129

    Article  Google Scholar 

  6. Hamad D, Yousef Y, Caminsky NG et al (2020) Defining the critical pediatric surgical workforce density for improving surgical outcomes: a global study. J Pediatr Surg 55:493–512

    Article  Google Scholar 

  7. Kruk ME, Gage AD, Joseph NT et al (2018) Mortality due to low-quality health systems in the universal health coverage era: a systematic analysis of amenable deaths in 137 countries. Lancet 392:2203–2212

    Article  Google Scholar 

  8. Martinez R, Lloyd-Sherlock P, Soliz P et al (2020) Trends in premature avertable mortality from non-communicable diseases for 195 countries and territories, 1990–2017: a population-based study. Lancet Glob Health 8:e511–e523

    Article  Google Scholar 

  9. Citron I, Jumbam D, Dahm J et al (2019) Towards equitable surgical systems: development and outcomes of a national surgical, obstetric and anasthesia plan in Tanzania. BMJ Glob Health 4:e001282

    Article  Google Scholar 

  10. Truché P, Shoman H, Reddy CL et al (2020) Globalization of national surgical, obstetric and anesthesia plans: the critical link between health policy and action in global surgery. Global Health 16:1

    Article  Google Scholar 

  11. Goodman LF, St-Louis E, Yousef Y et al (2018) The global initiative for children’s surgery: optimal resources for improving care. Eur J Pediatr Surg 28:51–59

    Article  Google Scholar 

  12. Wright N, Jensen G, St-Louis E et al (2019) Global initiative for children’s surgery: a model of global collaboration to advance the surgical care of children. World J Surg 43:1416–1425

    Article  Google Scholar 

  13. World Bank Country and Lending Groups (2020) World Bank Data Help Desk. Accessed 1 Jun 2020

  14. Global Health Observatory (GHO) data (2020) WHO | The data repository. Accessed 3 Sep 2020

  15. Alkema L, New JR (2013) Global estimation of child mortality using a Bayesian B-spline Bias-reduction model. arXiv [stat.AP].

  16. Levels and Trends in Child Mortality Report 2018 (2018). Accessed 3 Sep 2020

  17. GHO | By category | Proportion of deaths by country - HIV/AIDS. Accessed 3 Sep 2020

  18. Anand S, Bärnighausen T (2004) Human resources and health outcomes: cross-country econometric study. Lancet 364:1603–1609

    Article  Google Scholar 

  19. Schell CO, Reilly M, Rosling H et al (2007) Socioeconomic determinants of infant mortality: a worldwide study of 152 low-, middle-, and high-income countries. Scand J Public Health 35:288–297

    Article  Google Scholar 

  20. Houweling TAJ, Caspar AEK, Looman WN et al (2005) Determinants of under-5 mortality among the poor and the rich: a cross-national analysis of 43 developing countries. Int J Epidemiol 34:1257–1265

    Article  Google Scholar 

  21. Atiyeh BS, Gunn SWA, Hayek SN (2010) Provision of essential surgery in remote and rural areas of developed as well as low and middle income countries. Int J Surg 8:581–585

    Article  Google Scholar 

  22. White H (1980) A Heteroskedasticity-consistent covariance matrix estimator and a direct test for Heteroskedasticity. Econometrica 48:817–838

    Article  Google Scholar 

  23. van Buuren S (2018) Flexible imputation of missing data, 2nd edn

  24. Josse J, Husson F (2012) Handling missing values in exploratory multivariate data analysis methods. Published Online First: 2012. Accessed 22 Aug 2020

  25. van Buuren S, Groothuis-Oudshoorn K (2011) Mice: multivariate imputation by chained equations in R. J Stat Softw 45:1–67

    Article  Google Scholar 

  26. Rubin DB, Schenker N (1991) Multiple imputation in health-care databases: an overview and some applications. Stat Med 10:585–598

    Article  CAS  Google Scholar 

  27. Rubin DB (2004) Multiple imputation for nonresponse in surveys. Wiley, London

    Google Scholar 

  28. R Core Team (2012) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. ISBN 3-900051-07-0.

  29. Piburn J (2016) wbstats. Oak Ridge National Lab.(ORNL), Oak Ridge, TN (United States).

  30. Fitzgerald TN, Rice HE (2019) Investing in all of our children: global pediatric surgery for the twenty-first century. World J Surg 43:1401–1403

    Article  Google Scholar 

  31. Child Mortality - UNICEF DATA. Accessed 30 Aug 2020

  32. Chao F, You D, Pedersen J et al (2018) National and regional under-5 mortality rate by economic status for low-income and middle-income countries: a systematic assessment. Lancet Glob Health 6:e535–e547

    Article  Google Scholar 

  33. Golding N, Burstein R, Longbottom J et al (2017) Mapping under-5 and neonatal mortality in Africa, 2000–15: a baseline analysis for the Sustainable Development Goals. Lancet 390:2171–2182

    Article  Google Scholar 

  34. Liu L, Oza S, Hogan D et al (2016) Global, regional, and national causes of under-5 mortality in 2000–15: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet 388:3027–3035

    Article  Google Scholar 

  35. Boyle B, Addor M-C, Arriola L et al (2018) Estimating global burden of disease due to congenital anomaly: an analysis of European data. Arch Dis Child Fetal Neonatal Ed 103:F22–F28

    Article  Google Scholar 

  36. Toobaie A, Yousef Y, Balvardi S et al (2019) Incidence and prevalence of congenital anomalies in low- and middle-income countries: a systematic review. J Pediatr Surg 54:1089–1093

    Article  Google Scholar 

  37. Daniels KM, Riesel JN, Meara JG (2015) The scale-up of the surgical workforce. Lancet 385(Suppl 2):S41

    Article  Google Scholar 

  38. Saxton AT, Poenaru D, Ozgediz D et al (2016) Economic analysis of children’s surgical care in low- and middle-income countries: a systematic review and analysis. PLoS ONE 11:e0165480

    Article  Google Scholar 

  39. Ng-Kamstra JS, Greenberg SLM, Abdullah F et al (2016) Global surgery 2030: a roadmap for high income country actors. BMJ Glob Health 1:e000011

    Article  Google Scholar 

  40. Gajewski J, Bijlmakers L, Brugha R (2018) Global surgery—informing national strategies for scaling up surgery in Sub-Saharan Africa. Int J Health Policy Manag 7:481–484

    Article  Google Scholar 

  41. Fatima I, Shoman H, Peters AW et al (2020) Pakistan’s national surgical, obstetric, and anesthesia plan: an adapted model for a devolved federal-provincial health system. Can J Anaesth 67:1212–1216

    Article  Google Scholar 

Download references


We want to thank the Global Initiative for Children's Surgery (GICS) for its support of this work. GICS ( is a network of children's surgical and anesthesia providers from low-income, middle-income and high-income countries collaborating for the purpose of improving the quality of surgical care for children globally. There was no external funding source for this study. This work was conducted with support from Harvard Catalyst | The Harvard Clinical and Translational Science Center (National Center for Advancing Translational Sciences, National Institutes of Health Award UL 1TR002541) and financial contributions from Harvard University and its affiliated academic healthcare centers. The content is solely the responsibility of the authors and does not necessarily represent the official views of Harvard Catalyst, Harvard University and its affiliated academic healthcare centers, or the National Institutes of Health.

Author information

Authors and Affiliations



Corresponding author

Correspondence to Paul Truche.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

On behalf of The Global Initiative for Children’s Surgery (GICS): a network of children's surgical and anasthesia providers from low-income, middle-income and high-income countries collaborating for the purpose of improving the quality of surgical care for children globally


Appendix 1: Imputation Density Plots for Imputed Variables

figure a

Appendix 2: Methods for Estimation of Mortality

We used the following formula to calculate deaths averted where Y is the under-5 mortality rate (deaths for 1000 live births) for a given country, X the surgeon/anaesthesiologist/obstetrician (SAO) density, and Z the remaining covariates in the model, and where E denotes expected value (i.e., average).

$${\mathbb{E}}(\log Y|X,W,{\varvec{Z}}) = \alpha X + \gamma W + \beta^{T} {\varvec{Z}}$$

The parameter α represents the mean change in log Y associated with an increase of X by one unit (number of SAOs per 100,000 population), keeping the remaining variables Z fixed. Thus, for any baseline SAO density x, and potential increase r according to the formula:

$$(1 - e^{r\alpha + s\gamma } ){\mathbb{E}}(Y|X = x,W = w,{\varvec{Z}}) \approx {\mathbb{E}}(Y|X = x,W = w,{\varvec{Z}}) - {\mathbb{E}}(Y|X = x + r,W = w + s,{\varvec{Z}})$$

To estimate potentially avertable deaths across countries, every country that has SAO below 20 (or 40), we can take r = 20 − X (or 40 − X), and plug in the estimated α into the formula above, and the observed Y in place of the baseline E(Y | X = x, Z), to obtain an expected decrease in under-5 mortality rate associated with increasing SAO up to 20 (or 40).

To convert this to a raw number of annual deaths prevented by such an increase in SAO, we can replace Y with the raw number of observed annual deaths, Y*.

$$(1 - e^{{r\widehat{\alpha } + s\widehat{\gamma }}} )Y^{*} \approx Y^{*} - {\mathbb{E}}(Y^{*} |X = x + r,W = w + s,{\varvec{Z}})$$

Note that Y* = TY/1000, where T is the total number of live births in a year. Multiplying the above approximation by T/1000, we can use the following formula where the right hand side expresses the expected decrease in under-5 deaths associated with an increase in SAO density by r. The same method was used substituting for infant and neonatal mortality estimates. We adjusted our predictions for the potential change in procedures associated with scale up of surgical workforce in order to take into account non workforce related factors.

$${\mathbb{E}}[\log W|X] = a_{0} + a_{1} \log X$$

Appendix 3: Sensitivity Analysis

Model scenario at extremes of surgically amenable child mortality



Lower CI of mortality*

Upper CI of mortality*

Under 5 Mortality rsate

 − 0.0045 (0.0027)

 − 0.00437 (0.0037)

 − 0.0048 (0.006)

Infant mortality Rate

 − 0.0046 (0.0017)

 − 0.0044 (0.002)

 − 0.0047 (0.005)

Neonatal Mortality Rate

 − 0.0048 (0.0076)

 − 0.0046 (0.01)

 − 0.005 (0.012)

  1. *Sensitivities based on use of upper and lower confidence intervals compared to point estimates for mortality estimates by the UN Agency for Child Mortality. 2018 estimates were used for all calculations. Model beta coefficients and p values are reported

Delta adjustment technique In order to test if the imputed values can be considered to be missing at random conditional on the variables included in the imputation model, each value for SAO density was adjusted by 0, − 10, − 20 and − 50 and the mean imputed values across 5 imputations was compared. The mean SAO density on complete case analysis was 37.4 with a difference of − 1.77 among imputed values for a delta adjustment of 0 (MAR). There were no significant increases in the difference for deltas up to 20 SAO density, suggesting that missing SAO variables can be considered missing at random conditional on the variables included in the imputation model (“Appendix 3”).

Delta Adjustment technique

Realized difference in means of the observed and imputed SAO density data under various delta adjustments. Based on 5 imputations

Delta adjustment

Mean SAO density


0 (MAR)


− 1.77

− 10


− 2.118

− 20


− 4

− 50


− 7.155

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Truche, P., Botelho, F., Bowder, A.N. et al. Potentially Avertable Child Mortality Associated with Surgical Workforce Scale-up in Low- and Middle-Income Countries: A Global Study. World J Surg 45, 2643–2652 (2021).

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: