1 of 25

Blood Lead Level Estimates for Low- and Middle-Income Countries�The 32nd Annual ISEE Conference �August 24-27, 2020.�Virtual Presentation �

Bret Ericson, PhD(1); Howard Hu, MD(2); Emily Nash, MPH(3); Greg Ferraro, MA(3); Julia Sinitsky, MA(3); and Mark Patrick Taylor, PhD(1)

(1)Department of Earth and Environmental Sciences, Macquarie University, Sydney, Australia.;

(2)University of Washington School of Public Health, Seattle, USA;

(3)Pure Earth, New York, USA

​

bret.ericson@students.mq.edu.au

​

2 of 25

Presentation Outline

  • Introduction
    • trends in lead exposure; previous studies; gaps
  • Methods
    • PubMed search; data extraction; pooling of values
  • Results
    • background exposures; major sources
  • Discussion & Conclusions

​

3 of 25

Introduction

  • Lead (Pb) is a naturally occurring bluish gray metal with a range of applications
    • Corrosion resistant, malleable, excellent conductor of electricity
    • > 85 % of Pb used in lead-acid batteries
  • Low-levels (< 5 µg/dL) of lead exposure can result in adverse health impacts
    • Neurological decrement
    • Cardiovascular disease
    • No known safe level of exposure

Budtz-Jørgensen, E., Bellinger, D., Lanphear, B., Grandjean, P., Lanphear, B. P., Hornung, R., Khoury, J., Yolton, K., Baghurst, P., Bellinger, D. C., Canfield, R. L., Dietrich, K. N., Bornschein, R., Greene, T., Rothenberg, S. J., Needleman, H. L., Schnaas, L., Wasserman, G., Graziano, J., & Roberts, R. (2013). An international pooled analysis for obtaining a benchmark dose for environmental lead exposure in children. Risk Analysis, 33(3), 450–461. https://doi.org/10.1111/j.1539-6924.2012.01882.x

Lanphear, B. P., Rauch, S., Auinger, P., Allen, R. W., & Hornung, R. W. (2018). Low-level lead exposure and mortality in US adults: a population-based cohort study. The Lancet Public Health, 3(4), e177–e184. https://doi.org/10.1016/S2468-2667(18)30025-2

​

4 of 25

Introduction

  • Leaded petrol phase-outs resulted in substantial decreases in blood lead levels
    • High Income Countries 1970s–1990s
    • Low and Middle Income Countries (LMICs) 1990s–2005

https://www.epa.gov/transportation-air-pollution-and-climate-change/accomplishments-and-success-air-pollution-transportation

5 of 25

Introduction

  • France, Germany, Japan, Sweden, United States
    • BLLs ~1 µg/dL
  • BLLs in LMICs lowered to a point, but remain substantially higher

​

Nakayama SF, Espina C, Kamijima M, et al. Benefits of cooperation among large-scale cohort studies and human biomonitoring projects in environmental health research: An exercise in blood lead analysis of the Environment and Child Health International Birth Cohort Group. Int J Hyg Environ Health 2019; 222: 1059–67.

Strömberg U, Lundh T, Skerfving S. Yearly measurements of blood lead in Swedish children since 1978: The declining trend continues in the petrol-lead-free period 1995-2007. Environ Res 2008; 107: 332–5.

Centers for Disease Control and Prevention (CDC). (2013). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. https://wwwn.cdc.gov/nchs/nhanes/search/datapage.aspx?Component=Laboratory&CycleBeginYear=2013

6 of 25

Introduction

  • Previous studies
    • Olympio, et al (2017) – 56 studies from 2000–2014 on Latin America
    • Attina and Trasande (2013) – 68 studies from 2000-2012 on all LMICs
    • IHME annual burden of disease estimates – 88 studies from 2010–2017 on all LMICs
  • Gaps
    • Low number of studies reviewed
    • Pooled BLL estimates not published
    • Limited efforts to identify possible sources of exposure

​

​

Olympio KPK, Gonçalves CGCG, Salles FJ, et al. What are the blood lead levels of children living in Latin America and the Caribbean? Environ Int 2017; 101: 46–58.

Attina TM, Trasande L. Economic costs of childhood lead exposure in low- and middle-income countries. Environ Health Perspect 2013; 121: 1097–102.

IHME. Global Burden of Disease Study 2017 (GBD 2017) Data Input Sources Tool | GHDx. 2018. http://ghdx.healthdata.org/gbd-2017/data-input-sources (accessed Feb 16, 2019).

7 of 25

Methods

  • PubMed search
    • Conducted November 2019
    • ‘[country name]’ ‘blood’ ‘lead’
      • e.g. India blood lead
    • 1 January 2010 and 31 October 2019
  • 137 LMIC names (World Bank)

8 of 25

Methods

  1. Literature review
  2. Bias assessment
  3. Data extraction
  4. Imputing missing data
  5. Pooling sample means
  6. Estimating the number of children above thresholds
  7. Sensitivity testing

9 of 25

Methods

  1. Literature review
  • Title review for relevance
  • Abstract review
  • Full review
  • Bias assessment

​

​

10 of 25

Methods

  1. Literature review
  • Inclusion Criteria
    1. contained BLL data from human populations residing in [country];
    2. comprised at least 30 participants;
    3. presented BLL data derived from venous, capillary, or umbilical cord samples of whole blood (serum and plasma samples were excluded);
    4. data must have been collected after 2005;
    5. published in English

11 of 25

Methods

  1. Bias assessment
    • US Office of Health Assessment and Translation (OHAT) Handbook
    • 7 of 11 criteria were applied
    • studies with high risk of bias were excluded (tier 3)
  2. Data extraction
    • title; author; year; location; population characteristics (sex, age); BLL statistics (central tendency, dispersion, sample size); sources of exposure; analysis method; and the nature of exposure (i.e. background, occupational, elevated non-occupational)
    • child & adult subgroups
    • background & exposed (occupational or elevated non-occupational) subgroups

​

OHAT. (2019). Handbook for Conducting a Literature-Based Health Assessment Using OHAT Approach for Systematic Review and Evidence Integration; March 4, 2019. https://ntp.niehs.nih.gov/ntp/ohat/pubs/handbookmarch2019_508.pdf

12 of 25

Methods

  1. Imputing missing data
    • mean BLLs (µg/dL) and standard deviations were required to pool BLLs
    • guided by Weir et al., (2018); relied heavily on Wan, et al. (2014)
  2. Pooling sample means
    • use of log-transformed sample-weighted values
    • followed Fewtrell, et al. (2003) and Attina and Trasande (2013)

 

Weir, C. J., Butcher, I., Assi, V., Lewis, S. C., Murray, G. D., Langhorne, P., & Brady, M. C. (2018). Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: A systematic review. BMC Medical Research Methodology, 18(1), 1–14. https://doi.org/10.1186/s12874-018-0483-0

Wan, X., Wang, W., Liu, J., & Tong, T. (2014). Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Medical Research Methodology, 14(1), 1–13. https://doi.org/10.1186/1471-2288-14-135

Fewtrell, L., Kaufmann, R., & Prüss-Üstün, A. (2003). Lead: Assessing the environmental burden of disease at national and local levels. In WHO Environmental Burden of Disease Series, No. 2. World Health Organization. http://www.who.int/quantifying_ehimpacts/publications/en/leadebd2.pdf?ua=1

​

13 of 25

Methods

  1. Estimating the number of children above thresholds
    • assumed log-normal distribution of BLLs
    • calculated # children > 5 & 10 µg /dL
  2. Sensitivity testing
    • leave one out approach; ANOVA testing of results

15

5

0

10

14 of 25

Results�

15 of 25

Results�literature review

16 of 25

Results�background BLLs

  • Children (background)
    • 34 countries with 305 subsamples
  • Adults (background)
    • 37 countries with 365 subsamples

​

  • 631 million children (49%) >5 µg/dL
  • 413 million children (32%) > 10 µg/dL

17 of 25

18 of 25

Country

Background children

n subsamples

total sample size

pooled mean BLL (sd) (µg/dL)

Bangladesh

7

3,460

7·87 (5·69)

Benin

3

1,092

5·27 (3·02)

Brazil

18

6,384

2·45 (3·17)

Cameroon

1

147

8·7 (3·9)

China

111

591,463

4·17 (4·54)

Colombia

3

866

3·06 (0·6)

Dem. Rep. of the Congo

3

314

7·46 (4·03)

Ecuador

1

69

3·17 (2·54)

Egypt

11

1,100

8·24 (4·78)

Ethiopia

1

132

1·66 (3·31)

Haiti

1

273

6 (2·38)

India

22

5,593

5·22 (6·66)

Indonesia

3

387

5·2 (3)

Iran

8

928

3·62 (4·62)

Iraq

1

207

5·3 (1·9)

Jamaica

5

651

2·9 (2·98)

Kosovo

1

53

2·3 (0·07)

Mexico

45

4,509

3·62 (3·46)

Mongolia

2

338

3·82 (2·55)

Morocco

12

770

4·57 (3·14)

Nepal

1

312

6·69 (4·22)

Nigeria

6

844

7·67 (5·89)

Pakistan

33

2,382

9·27 (3·17)

Palestine

2

1,883

9·3 (11·73)

Romania

2

144

2·6 (2·61)

Russia

4

1,088

5·17 (4·18)

Senegal

1

32

8·22 (3·16)

Serbia

1

54

7·8 (4·27)

South Africa

4

2,185

5·59 (3·61)

Tanzania

1

43

2·26 (0·96)

Thailand

5

1,813

5·12 (3·03)

Turkey

4

462

3·23 (2·53)

Uganda

2

263

6·68 (4·43)

Vietnam

1

311

4·97 (5·5)

19 of 25

Results�sources

EXPOSURE SOURCE

N SUBSAMPLES

TOTAL SAMPLE SIZE

N BACKGROUND SUBSAMPLES

N NON-OCCUPATIONAL SUBSAMPLES

N OCCUPATIONAL SUBSAMPLES

automobile repair

13

781

2

2

9

battery manufacture or recycling

124

14,757

4

39

82

bullets

8

1,277

2

5

1

ceramics

13

1,485

5

5

4

contaminated site

4

1,676

0

4

0

dietary sources

15

2,533

10

5

0

dumpsite

5

410

2

2

2

ewaste

40

7,306

6

27

7

industry (lead)

8

818

0

3

5

industry (other)

70

8,209

24

13

34

lead-based paint

7

940

4

0

3

mining

32

4,998

1

31

3

other

36

5,927

11

18

6

petrol

7

1,252

4

0

3

smelting

52

21,778

10

21

22

tobacco products

58

3,771

7

51

0

20 of 25

Discussion and Conclusions

  • Limited data on BLLs in LMICs
  • BLLs appear to remain elevated
  • Sources appear to be largely industrial in nature

​

21 of 25

Discussion and Conclusions

  • Results generally consistent with other studies
    • Téllez-Rojo, et. al (2019) studied 1,457 urban Mexican children aged 1–4.
      • Median BLL of 3.3 compared with mean of 3.62 in this study

​

    • Yan, et al. (2020) pooled 95 studies representing 297,000 Chinese children finding a
      • mean BLL of 5·34 µg/dL (SD: 3·09) compared to the 4·17 µg/dL (SD: 4·54) in this study of 591,463 children

​

22 of 25

Discussion and Conclusions

  • Limitations
    • Discrete samples for national estimates
    • Imputed values

​

    • Limitations shared by previous studies
      • This study is significantly larger than preceding studies (538 publications reviewed vs 88 for IHME over similar time period)

​

​

23 of 25

Discussion and Conclusions

  • Importance of industrial emissions
  • 631 (95%CI: 395–778) million children > 5 µg/dL.
  • Issue likely much larger
    • only 44 countries (34 with child subgroups) of 137 LMICs could be included
  • To facilitate collection of BLL data, an international registry could be established to house the anonymized results of testing carried out by researchers and improve surveillance of BLLs in LMICs.

​

24 of 25

Discussion and Conclusions

  • Pure Earth/ UNICEF efforts in reducing lead exposure in LMICs
  • GBD – Pollution, Climate and Health Initiative

​

https://www.unicef.org/reports/toxic-truth-childrens-exposure-to-lead-pollution-2020

Shaffer, R. M., Sellers, S. P., Baker, M. G., de Buen Kalman, R., Frostad, J., Suter, M. K., Anenberg, S. C., Balbus, J., Basu, N., Bellinger, D. C., Birnbaum, L., Brauer, M., Cohen, A., Ebi, K. L., Fuller, R., Grandjean, P., Hess, J. J., Kogevinas, M., Kumar, P., … Hu, H. (2019). Improving and expanding estimates of the global burden of disease due to environmental health risk factors. In Environmental Health Perspectives (Vol. 127, Issue 10, pp. 105001-1-105001–105016). Public Health Services, US Dept of Health and Human Services. https://doi.org/10.1289/EHP5496

​

​

25 of 25

Thank you

bret.ericson@students.mq.edu.au

​