Measuring maternal mortality using census data in developing countries Tiziana Leone LSE Department of Social Policy.

Measuring maternal mortality using census

data in developing countries

Tiziana LeoneLSE

Department of Social Policy

Outline

Definitions Background Objectives and rationale Lesotho and Nicaragua Methods Results Discussion and future work

Definition

A maternal death is the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental causes.

WHO, 1993

Measures of Maternal Mortality

000,100#

#X

livebirths

athsmaternaldeMMRatio

000,14915#

#X

women

athsmaternaldeMMRate

Background

Pressure to get the indicators right to measure progress of MDG 5

Vital registration not sufficient to record maternal deaths

Maternal mortality ‘rare’ event: sample surveys need big sample in order to collect enough information

Census has been recommended in countries that lack complete vital registration

Advantages and Disadvantages using Census Data to Estimate

MMRateAdvantages:

• No problems with sample size• It is possible to study differentials• Evaluation methods are well developed• It’s cheap!

Disadvantages:

• Basic data need evaluation and adjustments• Only every 10 years (normally)• Methods Have Strong Assumption (e.g.:no migration) • Estimates are for intercensal period, not a specific period

Objectives

Apply methodology to two different settings : Nicaragua and Lesotho

Analyse feasibility of methods Develop methods in order to

estimate differential mortality

Lesotho and Nicaragua

Population TFR MMR GDP per capita

Net migration

HIVprevalence

Lesotho 1.8m 3.5 730 $1,300 -0.78 ‰ 23%

Nicaragua

5.7m 3.2 83-170 $2,600 -1.13‰ 0.2%

Data

Nicaragua 1995-2005 census Lesotho 1986-1996 census

Questions included in Lesotho census questionnaire

Questions included in Nicaragua census questionnaire

MethodsSeries of evaluations methods based on demographic ‘indirect techniques’ with adjustments when needed. Hill et al 2001.

Check degree of death coverage in the population

General Growth BalanceCheck quality of fertility data

P/F RatioCheck quality of information on pregnancy related deaths

No formal methods.

General growth balanceBased on the balance population equation

r = b – d or (rearranging) b – r = d

b=crude birth rater=growth rated=crude death rate.

Assumes absence of migration and that the completeness of coverage of the deaths is c, constant at all ages.The observed age-specific mortality rates are therefore equal to the true rates multiplied by c, or that the true rates are equal to the observed rates divided by c:

b(x+) – r(x+) = {1/c}*dobs(x+)

C=adjustment factor calculated using regression models

Mortality CheckGeneral Growth Balance - Lesotho, female, 1986-1996

0.0000

0.0100

0.0200

0.0300

0.0400

0.0500

0.0600

0.0000 0.0050 0.0100 0.0150 0.0200 0.0250 0.0300

Death Rate x+

En

try -

Gro

wth

Rate

x+

Observed values

Fitted values

General Growth Balance - Nicaragua, female, 1995-2005

0.0000

0.0100

0.0200

0.0300

0.0400

0.0000 0.0100 0.0200 0.0300 0.0400 0.0500

Death Rate x+

En

try

- G

row

th R

ate

x+

Observed values

Fitted values

Regression line fitted for (5+)-(65+)

Fertility Check

ASFR, Lesotho 1996

0.0000

0.0200

0.0400

0.0600

0.0800

0.1000

0.1200

0.1400

0.1600

0.1800

0.2000

15-19 20-24 25-29 30-34 35-39 40-44 45-49

Age specific fertility rates Nicaragua, 2005

0.0000

0.0200

0.0400

0.0600

0.0800

0.1000

0.1200

0.1400

15-19 20-24 25-29 30-34 35-39 40-44 45-49

Plausibility checks

Proportions of Births and Pregnancy-Related Deaths, Lesotho 1996

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0 1 2 3 4 5 6

Age Group

Pro

po

rtio

n

Births

Preg-Related Deaths

Proportions of Births and Pregnancy-Related Deaths, Nicaragua 2005

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0 2 4 6 8

Age Group

Pro

po

rtio

n

Births

Preg-Related Deaths

MMR

Census UNICEF estimate

UNDP

Lesotho 532 (1996) 550 (2000) 530 (1996)

Nicaragua 135 (2005) 170 (2005) 230 (2000)

Age specific PRMR

Age specific PRMR, Lesotho 1996

0

200

400

600

800

1000

15-19 20-24 25-29 30-34 35+

Age

PR

MR

Age Specific PRMR, Nicaragua 2005

0

500

1000

1500

2000

2500

15-19 20-24 25-29 30-34 35-39 40-44 45-49

Age

PR

MR

Age spefic PRMP, Nicaragua 2005-shorter age range

0

50

100

150

200

250

300

15-19 20-24 25-29 30-34 35-39

Series1

DiscussionWe are estimating an intercensal value and not a specific period

Careful interpretation of results

We need to explore more ways to get formal methods to estimate pregnancy related deaths

Census data give reasonable estimatesBest way forward in absence of Vital registration

Very little additional information needed

Need for more advocacy and training

Future plans

Apply the same methodology to more countries (e.g. South Africa and Zimbabwe) Two points in time

Differential (e.g.: wealth, residence, region) mortality applying smoothing functions