acknowledged, [3,4] while effective civil registration system - - PDF document

SLIDE 1

1

Completeness of Death Registration under CRS and Construction a Life Table with Adjusted ASDR for India and States

Anurag Verma1*, Gyan Prakash Singh1, Abhinav Singh2

• 1. Dept. of Community Medicine, IMS, BHU, Varanasi, India
• 2. Dept. of Statistics, University of Allahabad, Allahabad, India

*presenting author, e-mail:-imsbhuanurag@gmail.com

Abstract

The main objective of the present study was to estimate the completeness of deaths registration under Civil Registration System in India and its major states during 2001-2010. In the present project methods proposed by Bennett and Horiuchi (1983) have been used to estimate the completeness of death registration. The next aim of this study is to construct a Life Table for Adjusted ASDR. The analysis of life table was done using adjusted ASDR which gives an accurate measure of life expectancy in comparison to another method. The completeness of death registration in India is 49% and 39% in males and females respectively. The adjusted Life Expectancy at birth as 66 and 69 years for males and females respectively in India. Finally, The National mortality statistics from civil registration were rated unsatisfactory for coverage and completeness of death registration in India. Keywords: Mortality Statistics; Vital Statistics; Life Expectancy; Indirect Estimation Technique; MortPak.

Introduction

The value of good-quality mortality data for public health is widely acknowledged. WHO comparative assessments rated the quality of Indian mortality data as present is low. Since then, focused initiatives were introduced to improve Civil Registration and Vital Statistics. Furthermore, Indian mortality data are widely used by researchers and international development agencies as the basis for making policy and program. It is hence important to assess the quality of more recent Indian mortality data. Evaluating the quality of national mortality statistics from civil registration is important [1] not only for its role in the estimation of intercensal populations but also in the construction of life table which is used in many other demographic estimation procedures. [1,2]. The importance of reliable and valid as well as comparable mortality data for measuring and improving population health with the value of good quality mortality data for public health is widely

SLIDE 2

2

acknowledged, [3,4] while effective civil registration system remains the “gold standard” source for continuous mortality measurement [5]. Kingsley Davis talked of the “Amazing Decline” of mortality in underdeveloped countries [6]. Since the developing countries continue in their effects to enhance their Civil Registration & Vital Statistics are particularly beneficial to identify biases in mortality data and it is important for them to have the ability to evaluate their progress by periodic evaluations of mortality statistics. A couple of low and middle revenue countries have such data [3, 9]. Among the few countries which have carried

• ut such as exercise in the developing world are Brazil [7], china [8], & India, [9]. According to WHO

report, India rank third, next to just Myanmar and Nepal, among all south Asian countries ordered by adult death rate in 2002[10]. One of the greatest human achievements in the decline in mortality that’s

• ccurred during the modern era [11].

Civil Registration System(CRS), a unified process of continuous, permanent and compulsory recording of the vital events (livebirth, deaths, fetal deaths, marriages, and divorces) and characteristics thereof, as provide through the legal requirements of the country. [12] In India, the Registration of Births and Deaths act, 1969, provides for the compulsory registration of births &

• deaths. Registrar general of India initiated a scheme of Sample Registration of birth and death in India,

rural in 1964-65 on pilot basis. The scheme of sample registration become operational on full scale from 1969-70 and is popularly known as SRS. [13, 14] The difference in approach of data collection between CRS and SRS, a comparison of vital rates based on these two sources helps in evaluation the performance of CRS over SRS. The level of registration helps in reviewing the registration system and defining measures that would be necessary to improve registration levels across the country. [15] According to Bhat [16], the omission rate is common is the registration system and even in census of developing countries and therefore it is necessary to check their completeness. The assessment of the mortality level of a population is often based on information including the number of registered deaths. In many developing nations, like India, however, deaths are under registered by a significant margin, which in turn many lead to a biased estimate of the level of mortality [16,17]. Various methods have been developed for estimating the completeness of death registration. Each of them assumes that the population is closed to migration and that ages at death were correctly

• reported. In addition, Brass’s sectional growth balance method, 1975[18] biased by declining

mortality; Preston and Coale method,1980[19], sensitive to growth rate chosen and biased by

• verstatement of age at death; Bourgeais-Pichat method,1984, Limited experience with method seems

to overestimate completeness because of age overstatement at high ages; the other methods which relax the assumption of stability are, Forward projection method, The modified growth balance

SLIDE 3

3

method, 1979[20], Preston and Hill method, 1980, [21]; United Nation method,1979, and Bennett and Horiuchi method, 1981[22-23]; Excellent summary of these techniques has been given by Preston, 1984. The method that do not assume stability are a generalization of the methods assuming stability. One of these non-stable methods, the United Nation method, has been much used and has shown to give erratic results [24].In the modified growth balance method, Martin, 1980[25] sought the Brass estimate of completeness, for mortality decline. The procedure requires the knowledge of the rate and duration of mortality decline. Estimation of these factors involves some circularity because the estimate of the mortality condition is the aim of the exercise in the first place. The Preston and Hill method and the forward projection methods are related to the size of cohorts in two censuses. They become awkward to use irregular intercensal periods. For every developing countries where census are conducted irregular as well as regular intercensal periods. For such a purpose, one is left with the Bennett-Horiuchi method. In India, the assessment of mortality level in population can be done by CRS and SRS. Since 2002[16], there has not been any attempt to examine the quality of CRS due to unavailability of age specific mortality data. In this context, present paper is an attempt to assess the quality of national as well as states level mortality statistics from Civil Registration in India. The next aim of this study is to construct the life table with Adjusted ASDR which is given by Bennett-Horiuchi Method.

Materials and Methods

Data Sources

The analysis is based on the data available from four sources. The first data source is the annual statistical report of the Civil Registration system for the period 2000 through 2010. The second data source is the annual statistical report of the Simple Registration system for the period 2000 through 2010. The third data source used in the present analysis is the abridged life tables for the period 2002-2006 which are based on the age specific death rates available through the sample registration system. Finally, the fourth data source used is the population by age and sex enumerated at 2001 and 2011 population census. The Bennett- Horiuchi method is basically the extension of the methodology given by Preston and Coale [19] with the important development which does not require the assumption of stability of the population. Beside the advantage of discards the assumption of stability, this method can also be used for irregular intercensal periods. For stable population Preston et al. (1980) employed the relationship:

SLIDE 4

4

N a = 𝐸∗ 𝑦 exp 𝑠 𝑦 − 𝑏 𝑒𝑦

∞ 𝑏

, (1) Where N(a) denotes the number of persons aged a, r is the growth rate for stable population and 𝐸∗ 𝑦 is the true number of deaths experienced by persons aged x in the current population. If the completeness of the deaths registration is constant at age a and above, then 𝐸∗ 𝑦 = 𝑙𝐸 𝑦 , 𝑔𝑝𝑠 𝑏𝑚𝑚 𝑦 ≥ 𝑏, (2) Where D(x) is the number of registered deaths to persons aged x and k is the inverse of the completeness of death registration. There for from (1) & (2), we obtain: N a = k 𝐸 𝑦 exp 𝑠 𝑦 − 𝑏 𝑒𝑦

∞ 𝑏

, (3) If we define, N a = 𝐸 𝑦 exp 𝑠 𝑦 − 𝑏 𝑒𝑦

∞ 𝑏

, (4) Then the estimates of completeness are derived from the median of a series of age specific completeness of death which are ratio of estimated to observed population (N a / N a ), when the number of registered deaths by age, the number of living person by age and the growth rate of the population are provided. This median is then used to calculate an adjusted set of age specific death rates(ASDR) and the Life expectancy for ages 5 and above [22-23,37-38]. Bennett-Horiuchi method is based on four assumptions. First, the population was not exposed to migration during the intercensal period. Second, both censuses have the same degree of

• completeness. Third, age misreporting occurs only after age 50 years, and Fourth, degree of

completeness of death registration is uniform above age 5 years. The following procedure was adopted to apply the Bennett-Horiuchi method:  The total number of registered deaths by sex was taken from the CRS report [26].  The percentage distribution of deaths by sex and 5-years age-group was taken from SRS report [27].  Then the total number of registered deaths by sex and 5-year age group in CRS was estimated by using above SRS data. Similarly, it is done for all Annual Reports on Vital Statistics of India and its states based on CRS 2001-2010.  Summing the number of registered death by sex and age group for 2001-2010 which is

• btained by above procedure for calculating the number of intercensal deaths between 2001-

2010.

SLIDE 5

5

 Total population by sex available for the years 2001[28] and 2011[29] was used to estimate population of India and selected states for different years of the intercensal period using Newton forward and backward interpolation methods. This procedure provided estimates of population for years 2010.  Finally, to estimate the completeness of death registration, population of two censuses 2001 and 2010 with age groups and total deaths during the period has been used in BENHR program in MORTPAK software [30] which gives completeness of death registration for each age group, median value for all age groups and life expectancy at age 5 and above.

Construction of life table

 To constructing a life table, the usual method, the death rates for 0-1, 1-5, and other remaining five years’ age group are required. But, the Bennett-Horiuchi provided adjusted ASDR starting by age 5 and above. Then m0-5 values are converted into the m0-1 & m1-5 with the help of qn-x by employing the following steps:  First, construct a life table based on average SRS ASDR from the period 2001 to 2010, and then we get three estimated values q0-1, q1-5, q0-5.  Next, we have (m0-5) adjusted ASDR given by Bennett-Horiuchi method. Now estimated Adjusted ASDR m0-1 and m1-5 with the help of above SRS estimated qn-x values as follows: The adj. ASDR in 𝑛0−1 is =

𝑟0−1 𝑟0−5 ∗ 𝑛0−5

(5) The adj. ASDR in 𝑛1−5 is =

𝑟1−5 𝑟0−5 ∗ 𝑛0−5

(6)



Finally, we have Adj. ASDR with starting 0-1, 1-5, and other remaining five years’ age group and up to age 75+. Then LIFTB program used for the construction of a life table based on a set

• f adjusted ASDR in the MORTPAK software package for mortality measurement.

Results and Discussions

Table 1 provides the estimated value of completeness of death registration in India and major

• states. The national level coverage of median deaths completeness is 49% male and 39% in female.

The major states have been categorized into 5 group as per their level of registration of death for year, 2001-2010. The states, Bihar (15.3%) show that the minimum level of completeness and Haryana is the maximum (95.8%) as well as on states where the level of completeness is above 90% in Total (both male and female). The reasons behind, the Haryana states took bold step and shifted the registration of birth and death from police to Primary Health Center (PHC) with encouraging trends

SLIDE 6

6

though the states have not yet achieved 95% registration of births and deaths but substantial impartments have been made. Table 1: Paste Here Table 2 provides the adjusted ASDR and Life Expectancy in India and is selected major states. With the help of theses Abridged adjusted ASDR can construct a complete life table by using MORTPAK or other software package. The national life expectancy at birth is 66 years and 69 years in males and females respectively. The state, Andhra Pradesh show that the minimum 60 years and states Kerala show that the maximum 70 years of life expectancy at birth in male. The state, Assam shows that the minimum 64 years and states Kerala show that the maximum 74 years of life expectancy at birth in females. The 10-years age difference between the states in males as well as females, it shown that wide variation present in mortality situation in India. Table 2: Paste Here Figure 1 gives the Completeness of death registration in India and its major selected states. The data are arranged in ascending order in the Completeness of death registration. The national level

• f completeness in India is 45% which show very less amount of completeness which means the half
• f the death is under registered. Haryana is an only one state in India where the death registration

(95.8%) is above 90%. In Haryana, to ensure better implementation of various provisions of RBD Act, it was decided to get the work of registration done through PHC instead of police stations [31]. This policy change was also meant to increase the number of registration center so as to make registration more accessible to people. Bihar, is states where the minimum level of completeness of death registration and it is only 15.3 %, in this state as well as other states in India where the completeness is less than 90%, it is very important to improve the process of death registration by taking bold steps, getting the example of Haryana state.

Figure 1. Estimates of Completeness of death registration for India and States, 2001-2010. The data are arranged in ascending order in the Completeness of death registration. 15.3 30.2 45 50.1 54.5 55.3 60.3 67.4 70.4 80.9 80.9 85.1 86.3 86.8 95.8 20 40 60 80 100 Level of Completeness Place

Level of Completeness

SLIDE 7

7

Life expectancy at birth reflects the overall mortality of a population. It summarizes the mortality pattern that prevails across all age group- children and adolescents, adults and the elderly [32]. The Life Table has been constructing for India and states separately for male and female for the period 2001-2010. Figure 2 gives the estimates of Life Expectancy at birth by sex for India and States from the period 2001 to 2010. The data are arranged for female life expectancy at births. The national level, India Life Expectancy at birth is 66 year in males and 69 year in females. Bihar is a state where the mortality condition in male and female more are less equal that means both are equal risk. On the

• ther hand, Andhra Pradesh is a state where the difference in Life Expectancy in males and female is

high, it show that the males is higher mortality risk condition.

Figure 2. Estimates of Life Expectancy at birth by sex, India and States, 2001-2010. The data are arranged in descending order in the Life Expectancy at birth in females.

Life Expectancy is regularly used to measure the overall health of the society. Life expectancy at age 65 is frequently used as a measure of a health status of adult population. Changes in life expectancy are repeatedly used to explain trends in mortality. Then we able to predict how populations will age have enormous implications for the planning and provision of services and supports. As the life expectancy of population lengthens, the numbers of people living with chronic illnesses are more common among older person. Figure 3 gives the estimates of Life Expectancy at age 65+ by sex for India and States from the period 2001 to 2010. The data are arranged for female life expectancy at births.

KERALA HARYA NA MAHAR ASTHRA RAJAST HAN PUNJAB TAMIL NADU KARNAT AK MADHY A PRADES H WEST BANGAL GUJRAT INDIA ANDHR A PRADES H BIHAR ODISSA ASSAM Male 69.69 67.89 68.3 67.89 67.62 67.79 66.95 67.59 67.38 66.4 66.66 60.19 67.79 65.54 62.51 Female 74.33 71.8 71.65 71.39 71.36 70.6 70.3 69.8 69.45 69.34 69 68.34 66.57 65.75 64.14

10 20 30 40 50 60 70 80 Life Expectancy at Birth

SLIDE 8

8

Figure 3. Estimates of Life Expectancy at age 65 by sex, India and States, 2001-2010. The data are arranged in descending order in the Life Expectancy at age 65+ in females.

In developed countries, male’s riskier unhealthy behaviors are a major reason they die younger. [33] Their higher rates of cigarette smoking, heavy drinking, gun use, employment in hazardous

• ccupations, and risk taking in recreation and driving are responsible for male’s higher death rate due

to lung cancer, accidents, suicide, and homicide. [34] Making a difference in Life Expectancy have helped decrease deaths related to unhealthy behaviors, particularly those associated with male’s

• deaths. In the, US life expectancy difference in males for female is 5.01 years, whereas in France it

was 7.8 years and in the U.K., 4.3 years, difference in Russia reaching more than 12 years, whereas in China with 3-year difference[35]. The diversity in worldwide longevity alone indicates that the difference in mortality between the sexes is not purely biological and that there are intervening social

• factors. Also from, Figure 4, show that the estimates difference in female and male Life Expectancy at

birth for India and States from the period, 2001-2010. It helps us to measure the difference in mortality condition by sex. Wide gap in life expectancy lead to need more focus in health policy and try to improvement in health situation to reduce wide gap. In the addition, from the Figure 4, it is clearly

• bserved that Bihar is on state where male life expectancy is higher than female, it so that in the state
• f Bihar where female is much higher mortality risk condition. On the other hand, more than 8 years’

difference in life expectancy in the state Andhra Pradesh in show that a male’s is much higher risk mortality stage due to social factor with also unhealthy life style.

HARYA NA PUNJAB KERALA MAHAR ASTHRA KARNA TAK RAJAST HAN MADHY A PRADES H GUJRAT TAMIL NADU INDIA WEST BANGA L ANDHR A PRADES H ODISSA ASSAM BIHAR Male 17.35 16.91 15.83 16.45 16.42 15.64 16.12 15.5 16.19 15.81 15.46 15.2 16.09 14.56 15.62 Female 19.1 18.1 17.89 17.76 17.69 17.49 17.35 17.08 16.93 16.79 16.7 16.44 16.11 15.91 15.11

5 10 15 20 25 Life Expectancy at age 65

SLIDE 9

9

Figure 4. Estimates the difference in female and male Life Expectancy at birth, India and States, 2001-2010. The data are arranged in ascending order in the difference in Life Expectancy at birth in females and males.

Figure 5 show that the Intercensal and Adjusted ASDR for total (Both male and female) for three selected states of India with their level of completeness of death registration. Its show that, form

• fig. 5(a) the state Bihar where the difference in death rates is increases with age it shows that that

death registration is highly under registered with age increase which leads to biasness is present in Bihar mortality data. In the states, Haryana more than 90 % coverage of death registration is indicates the good quality of mortality statistics. Also from fig. 5(c) which show that there is no difference in death rates. Whereas in the Gujarat is less biasness in mortality data form fig. 5(b) it is clear observed.

Figure 5. Adjusted ASDR & Intercensal Death Rates for selected state Bihar, Gujarat & Haryana, 2001-2010

• 1.22

0.21 1.63 2.07 2.21 2.34 2.81 2.94 3.35 3.35 3.5 3.74 3.91 4.64 8.15

• 2

2 4 6 8 10

• Diff. in

Life Expectancy at birth Difference in years

0.02 0.04 0.06 0.08 0.1 0.12 0.14

0- 5 5

• 10

10

• 15

15

• 20

20

• 25

25

• 30

30

• 35

35

• 40

40

• 45

45

• 50

50

• 55

55

• 60

60

• 65

65

• 70

70

• 75 75+

Fig.5(a). Estimates of Adj. ASDR and ASDR (Bihar)

Intercensal Death Rate

• Adj. Death Rate

0.02 0.04 0.06 0.08 0.1 0.12

0- 5 5

• 10

10

• 15

15

• 20

20

• 25

25

• 30

30

• 35

35

• 40

40

• 45

45

• 50

50

• 55

55

• 60

60

• 65

65

• 70

70

• 75 75+

Fig.5(b). Estimates of Adj. ASDR and ASDR (Gujarat)

Intercensal Death Rate

• Adj. Death Rate

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

0- 5 5

• 10

10

• 15

15

• 20

20

• 25

25

• 30

30

• 35

35

• 40

40

• 45

45

• 50

50

• 55

55

• 60

60

• 65

65

• 70

70

• 75 75+

Fig.5(c). Estimates of Adj. ASDR and ASDR (Haryana)

Intercensal Death Rate

• Adj. Death Rate

SLIDE 10

10

Figure 6 provides that the difference between estimated (based on Adjusted ASDR for 2001- 2010) and SRS (2004-2008) [36] Life Expectancy at birth. Figure shows that for some states SRS give

• ver estimates and for some states SRS give under estimates compare estimated of Life Expectancy at

age birth.

Figure 6. Estimates of difference between (Estimated 2001-2010 & SRS 2004-2008) Life Expectancy at birth, India and States. The data are arranged in ascending order in the difference in Life Expectancy at birth in females and males between (Estimate & SRS).

Figure 7 provides that the difference between estimated (based on Adjusted ASDR for 2001- 2010) and SRS (2004-2008) Life Expectancy at age 5. In this figure its so that for all most every states SRS give over estimates compare to estimate of Life Expectancy at age 5 based on adjusted ASDR. Also from the figure 6 & 7 it shows there is a need to improvement in SRS procedure and reviewing the registration system and defining measure that would be necessary to improve registration levels across the country.

Figure 7. Estimates of difference between (Estimated 2001-2010 & SRS 2004-2008) Life Expectancy at age 5, India and States. The data are arranged in ascending order in the difference in Life Expectancy at age 5 in females and males between (Estimate & SRS).

KERALA WEST BANGA L MAHAR ASHTR A ANDHR A PRADES H PUNJA B TAMIL NADU GUJRAT KARNA TAKA ASSAM BIHAR ODISHA HARYA NA INDIA RAJAST HAN MADHY A PRADES H Male

• 1.51

0.68 0.7

• 3.31

0.12 1.19 2.1 2.55 2.51 3.59 3.44 2.59 2.06 3.59 7.29 Female

• 3.07
• 0.85

0.05 0.24 0.26 0.5 0.64 1.1 1.84 1.87 2.75 2.8 3.05 3.99 7.1

• 4
• 2

2 4 6 8

• Diff. in Life Expectancy

KERALA BIHAR RAJAST HAN ANDHR A PRADE SH GUJRA T WEST BANGA L PUNJA B INDIA MAHA RASHT RA KARNA TAKA TAMIL NADU ASSAM HARYA NA ODISH A MADH YA PRADE SH Male

• 2.27
• 1.1
• 2.46
• 2.52
• 0.95
• 1.39
• 2.14
• 1.05
• 1.13
• 0.15
• 0.88
• 0.96
• 0.75
• 0.56

0.72 Female

• 3.79
• 3.47
• 3.13
• 2.8
• 2.71
• 2.71
• 2.53
• 2.37
• 1.9
• 1.6
• 1.57
• 1.51
• 1.36
• 1.12
• 0.15
• 4.5
• 4
• 3.5
• 3
• 2.5
• 2
• 1.5
• 1
• 0.5

0.5 1

• Diff. in Life Expectancy

Male Female

SLIDE 11

11

Conclusion

In India, CRS give reliable estimate of vital events, but this has some problem regarding completeness in adult deaths. As a result, the death rate implied by the reported deaths is usually an underestimate of the true death rate then some method of adjustment is required to transform the reported death rate into a better estimate of true mortality conditions. Therefore, it is necessary to check the completeness of death registration over time. This study finds the gap in actual death and registered death of by CRS. Therefore from this study, it is can be concluded that there is a need to review the CRS procedure and updating of adults deaths registration system. That will give better and more reliable estimates of deaths at national and states level by sex. On the whole it may be concluded that there is a need of incredible change in completeness of deaths registration process. Thus the paper suggests for considering death completeness while estimating completeness of deaths situation or representing gender differential in completeness of death at any place. However, it is necessary to implement policy changes to reform in CRS.

References

• 1. Joubert J, Rao C, Bradshaw D, Vos T, Lopez AD. Evaluating the quality of national

mortality statistics from civil registration in South Africa, 1997–2007. PLoS One. 2013 May 27;8(5):e64592.

• 2. Bah S. The evaluation of the completeness of death registration in the presence of high net
• ut-migration: the case example of Mauritius.
• 3. Mathers CD, Ma Fat D, Inoue M, Rao C, Lopez AD. Counting the dead and what they died

from: an assessment of the global status of cause of death data. Bulletin of the world health

• rganization. 2005 Mar;83(3):171-7c.
• 4. United Nations. Dept. of Economic and Social Affairs. Population Division (2006) World

Mortality Report 2005.New Your United Nation.

• 5. Joubert J, Rao C, Bradshaw D, Dorrington RE, Vos T, Lopez AD. Characteristics,

availability and uses of vital registration and other mortality data sources in post- democracy South Africa. Global health action. 2012;5.

• 6. Davis K. The amazing decline of mortality in underdeveloped areas. The American

Economic Review. 1956 May 1;46(2):305-18.

• 7. França E, de Abreu DX, Rao C, Lopez AD. Evaluation of cause-of-death statistics for

Brazil, 2002–2004. International journal of epidemiology. 2008 Aug 1;37(4):891-901.

SLIDE 12

12

• 8. Rao C, Lopez AD, Yang G, Begg S, Ma J. Evaluating national cause-of-death statistics:

principles and application to the case of China. Bulletin of the World Health Organization. 2005 Aug;83(8):618-25.

• 9. Mahapatra PR, Rao PC. Cause of death reporting systems in India: a performance analysis.

National Medical Journal of India. 2001 May 1;14(3):154-62.

• 10. World Health Organization. Cancer control: knowledge into action: WHO guide for

effective programmes. World Health Organization; 2007.

• 11. Krishnan P. Mortality decline in India, 1951–1961: Development vs public health program
• hypothesis. Social Science & Medicine (1967). 1975 Aug 1;9(8-9):475-9.
• 12. Hartnett C. Handbook on civil registration and vital statistics systems: Developing

information, education and communication.

• 13. Census
• f

India: Sample Registration. Available from: http://censusindia.gov.in/Vital_Statistics/SRS/Sample_Registration_System.aspx.

• 14. Mahapatra P. An overview of the sample registration system in India. InPrince Mahidol

Award Conference & Global Health Information Forum 2010 Jan.

• 15. Registrar General I. Vital Statistics of India 1994 based on the Civil Registration System.
• 16. Bhat PM. Completeness of India's sample registration system: an assessment using the

general growth balance method. Population Studies. 2002 Jan 1;56(2):119-34.

• 17. Mahapatra P, Shibuya K, Lopez AD, Coullare F, Notzon FC, Rao C, Szreter S. Civil

registration systems and vital statistics: successes and missed opportunities. The Lancet. 2007 Nov 16;370(9599):1653-63.

• 18. Brass W. Methods for estimating fertility and mortality from limited and defective data.

Methods for estimating fertility and mortality from limited and defective data.. 1975.

• 19. Preston S, Coale AJ, Trussell J, Weinstein M. Estimating the completeness of reporting of

adult deaths in populations that are approximately stable. Population Index. 1980 Jul 1:179- 202.

• 20. Brass W. A procedure for comparing mortality measures calculated from intercensal

survival with the corresponding estimates from registered deaths. InAsian and Pacific Census Forum 1979 Nov 1 (Vol. 6, No. 2, pp. 5-7).

• 21. Preston S, Coale AJ, Trussell J, Weinstein M. Estimating the completeness of reporting of

adult deaths in populations that are approximately stable. Population Index. 1980 Jul 1:179- 202.

• 22. Bennett NG, Horiuchi S. Estimating the completeness of death registration in a closed
• population. Population index. 1981 Jul 1:207-21.
• 23. Bennett, Neil G. and Shiro Horiuchi. 1984. "Mortality estimation from registered deaths in

less developed countries", Demography21(2): 217-234.

SLIDE 13

13

• 24. Preston SH. Use of direct and indirect techniques for estimating the completeness of death

registration systems.

• 25. Martin LG. A modification for use in destabilized populations of Brass's technique for

estimating completeness of death registration. Population Studies. 1980 Jul 1;34(2):381-95.

• 26. Registrar General for India (2001-2010), " Annual Report on Vital Statistics of India based
• n CRS". Ministry of Home Affairs, New Delhi.
• 27. Registrar General for India (2001-2010), "Sample Registration System Bulletins". Ministry
• f Home Affairs, New Delhi.
• 28. Registrar General I. Census of India, Provisional Population Totals, Series I, Paper I. 2001.
• 29. Registrar General I. Census of India, Provisional Population Totals, Series I, Paper I. 2011.
• 30. United Nations. MORTPAK-LITE: the United Nations Software Package for Mortality

Measurement.

• 31. Singh PK, Kaur M, Jaswal N, Kumar R. Impact of policy initiatives on civil registration

system in Haryana. Indian Journal of Community Medicine. 2012 Apr 1;37(2):122.

• 32. World Health Organization (WHO). (WHO) Statistical Information System (WHOSIS).
• 33. Carl Haub, 2007 World Population Data Sheet (Washington, DC: Population Reference

Bureau, 2007)

• 34. Alter G, Manfredini M, Nystedt P. Gender differences in mortality.
• 35. "CIA - The World Factbook Life Expectancy". Cia.gov. Retrieved 2012-03-22.
• 36. (2004) Registrar General of India SRS-based abridged life tables (2004-2008) India:

Registrar General, New Delhi.

• 37. Ph F. Manual X. Indirect techniques for Demographic estimations. Population.

1984;39(3):628-9.

• 38. Shryock HS, Siegel JS, Larmon EA. The methods and materials of demography. US

Bureau of the Census; 1973.

• 39. Brass. William. 1967. Methods of Estimating Basic Demographic Measures from

Incomplete Data. Manual IV. Population Studies No. 42. New York: United Nations.

SLIDE 14

14

Appendix Table 1. Estimated of death registration (Application of Bennett-Horiuchi Technique) India and States, 2001-2010

Groups

(as per their

level of death registration)

Bennett-Horiuchi technique Total Male Female Above 90 1 States (Haryana) 5 States (Haryana, Kerala, Punjab, Karnataka, Tamil Nadu) No States 80-90 5 States (Kerala, Punjab, Karnataka, Maharashtra, Tamil Nadu) 1 State (Maharashtra) 2 States (Haryana, Punjab) 50-80 6 States (Orrisha, Gujarat, Rajasthan, Madhya Predesh, Andhrya Pradesh, West Bengal) 6 States (Orrisha, Gujarat, Rajasthan, Madhya Predesh, Andhrya Pradesh, West Bengal) 7 States (Kerala, Maharashtra, Tamil Nadu, Karnataka, Orrrisha, Gujarat, Andhra Pradesh) 25-50 1 States (India*, Assam) 1 States (India, Assam) 3 States (India, Madhya Pradesh, Rajasthan, West Bengal) Below 25 1 States (Bihar) 1 States (Bihar) 2 States (Assam, Bihar) * Uttar Pradesh, Tripura, Uttarakhand excluded from the measure of the level of completeness of death registration in India due to unavailability of registered death.

SLIDE 15

15

Table 2. Estimates of Adjusted ASDR and Life Expectancy in India and States, 2001-2010.

INDIA ANDHRA PRADESH ASSAM AGE Male Female Male Female Male Female GROUP Adj. ASDR E(X)

• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)

0-1 0.0134 66.06 0.0011 69.95 0.0548 60.19 0.0148 68.34 0.0157 62.51 0.0156 64.14 1-5 0.0035 65.94 0.0004 69.03 0.0076 62.51 0.0020 68.35 0.0055 62.49 0.0061 64.15 5 - 10 0.0013 62.85 0.0015 65.13 0.0008 60.38 0.0007 64.90 0.0021 59.84 0.0019 61.69 10 - 15 0.0010 58.23 0.0010 60.59 0.0008 55.62 0.0007 60.13 0.0013 55.43 0.0014 57.25 15 - 20 0.0015 53.50 0.0019 55.88 0.0017 50.83 0.0020 55.33 0.0023 50.78 0.0030 52.63 20 - 25 0.0021 48.87 0.0025 51.39 0.0027 46.24 0.0024 50.85 0.0025 46.33 0.0031 48.39 25 - 30 0.0026 44.36 0.0023 47.01 0.0036 41.83 0.0023 46.43 0.0033 41.89 0.0031 44.10 30 - 35 0.0033 39.91 0.0024 42.52 0.0050 37.55 0.0025 41.93 0.0036 37.54 0.0040 39.75 35 - 40 0.0042 35.54 0.0026 38.01 0.0061 33.42 0.0025 37.42 0.0045 33.18 0.0041 35.50 40 - 45 0.0056 31.23 0.0035 33.48 0.0076 29.37 0.0039 32.85 0.0073 28.88 0.0053 31.19 45 - 50 0.0077 27.04 0.0046 29.02 0.0102 25.41 0.0045 28.45 0.0100 24.86 0.0070 26.95 50 - 55 0.0110 22.99 0.0082 24.63 0.0151 21.60 0.0093 24.03 0.0144 21.00 0.0120 22.82 55 - 60 0.0191 19.14 0.0134 20.55 0.0253 18.09 0.0153 20.05 0.0274 17.38 0.0212 19.07 60 - 65 0.0238 15.81 0.0175 16.79 0.0274 15.20 0.0177 16.44 0.0294 14.56 0.0187 15.91 65 - 70 0.0409 12.49 0.0319 13.09 0.0475 12.06 0.0353 12.72 0.0524 11.47 0.0433 12.22 70 - 75 0.0525 9.75 0.0499 9.91 0.0557 9.62 0.0540 9.67 0.0581 9.17 0.0647 9.56 75+ 0.1009 6.93 0.0968 7.00 0.1031 6.92 0.1008 6.89 0.1138 6.42 0.1001 7.28 BIHAR GUJARAT HARYANA AGE Male Female Male Female Male Female GROUP Adj. ASDR E(X)

• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)

0-1 0.0013 67.79 0.0018 66.57 0.0124 66.40 0.0139 69.34 0.0122 67.89 0.0141 71.80 1-5 0.0005 66.88 0.0009 65.69 0.0032 66.22 0.0044 69.30 0.0020 67.72 0.0037 71.81 5 - 10 0.0018 63.00 0.0026 61.93 0.0011 63.05 0.0010 66.49 0.0007 64.25 0.0014 68.84 10 - 15 0.0012 58.55 0.0013 57.72 0.0008 58.38 0.0009 61.79 0.0007 59.48 0.0007 64.31 15 - 20 0.0020 53.88 0.0024 53.08 0.0012 53.60 0.0014 57.07 0.0014 54.67 0.0014 59.52 20 - 25 0.0023 49.40 0.0038 48.70 0.0019 48.91 0.0020 52.46 0.0021 50.04 0.0018 54.93 25 - 30 0.0026 44.94 0.0029 44.59 0.0024 44.34 0.0018 47.95 0.0021 45.55 0.0018 50.42 30 - 35 0.0026 40.48 0.0029 40.21 0.0030 39.84 0.0021 43.36 0.0039 41.02 0.0017 45.86 35 - 40 0.0034 35.98 0.0035 35.75 0.0040 35.40 0.0022 38.78 0.0036 36.78 0.0019 41.22 40 - 45 0.0043 31.55 0.0045 31.33 0.0052 31.06 0.0028 34.19 0.0062 32.40 0.0026 36.59 45 - 50 0.0066 27.18 0.0057 26.98 0.0070 26.81 0.0043 29.62 0.0084 28.34 0.0032 32.04 50 - 55 0.0093 23.01 0.0104 22.68 0.0116 22.67 0.0065 25.21 0.0111 24.45 0.0061 27.51 55 - 60 0.0191 18.97 0.0162 18.75 0.0190 18.87 0.0121 20.95 0.0173 20.70 0.0077 23.29 60 - 65 0.0206 15.62 0.0237 15.11 0.0248 15.50 0.0172 17.08 0.0171 17.35 0.0087 19.10 65 - 70 0.0404 12.03 0.0395 11.69 0.0431 12.20 0.0315 13.38 0.0247 13.67 0.0184 14.83 70 - 75 0.0529 9.16 0.0677 8.69 0.0539 9.54 0.0437 10.23 0.0392 10.13 0.0335 11.00 75+ 0.1147 6.17 0.1200 6.19 0.1055 6.71 0.0928 7.10 0.0908 6.76 0.0773 7.53 KARNATAKA KERALA MADHYA PRADESH AGE Male Female Male Female Male Female GROUP Adj. ASDR E(X)

• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)

0-1 0.0112 66.95 0.0120 70.30 0.0026 69.69 0.0028 74.33 0.0016 67.59 0.0015 69.80 1-5 0.0021 66.70 0.0024 70.15 0.0006 68.87 0.0006 73.54 0.0005 66.70 0.0006 68.90 5 - 10 0.0008 63.25 0.0007 66.80 0.0004 65.03 0.0004 69.71 0.0020 62.82 0.0020 65.05 10 - 15 0.0006 58.50 0.0007 62.04 0.0003 60.14 0.0003 64.83 0.0012 58.42 0.0013 60.68 15 - 20 0.0012 53.66 0.0015 57.24 0.0007 55.23 0.0006 59.92 0.0018 53.75 0.0025 56.04 20 - 25 0.0020 48.96 0.0023 52.65 0.0012 50.42 0.0010 55.08 0.0022 49.23 0.0033 51.71 25 - 30 0.0026 44.42 0.0019 48.22 0.0018 45.71 0.0009 50.34 0.0030 44.74 0.0026 47.53 30 - 35 0.0038 39.98 0.0028 43.66 0.0022 41.09 0.0010 45.55 0.0028 40.38 0.0025 43.11 35 - 40 0.0046 35.68 0.0022 39.24 0.0030 36.51 0.0014 40.77 0.0041 35.90 0.0029 38.62 40 - 45 0.0063 31.46 0.0033 34.65 0.0042 32.01 0.0018 36.03 0.0049 31.59 0.0036 34.14 45 - 50 0.0083 27.38 0.0043 30.18 0.0061 27.64 0.0025 31.34 0.0074 27.31 0.0043 29.72 50 - 55 0.0127 23.44 0.0071 25.78 0.0106 23.41 0.0040 26.70 0.0115 23.24 0.0075 25.31 55 - 60 0.0180 19.80 0.0111 21.61 0.0147 19.54 0.0071 22.18 0.0188 19.46 0.0123 21.17 60 - 65 0.0224 16.42 0.0148 17.69 0.0216 15.83 0.0108 17.89 0.0236 16.12 0.0172 17.35 65 - 70 0.0376 13.07 0.0272 13.85 0.0340 12.34 0.0191 13.73 0.0407 12.82 0.0317 13.67 70 - 75 0.0474 10.25 0.0436 10.49 0.0567 9.16 0.0401 9.84 0.0538 10.15 0.0454 10.58 75+ 0.0916 7.32 0.0867 7.43 0.1116 6.33 0.0999 6.44 0.0934 7.51 0.0864 7.63

SLIDE 16

16

MAHARASHTRA

ODISHA PUNJAB AGE Male Female Male Female Male Female GROUP Adj. ASDR E(X)

• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR

E(X) Adj. ASDR E(X) 0-1 0.0073 68.30 0.0079 71.65 0.0172 65.54 0.0181 65.75 0.0096 67.62 0.0118 71.36 1-5 0.0014 67.80 0.0017 71.21 0.0049 65.66 0.0059 65.94 0.0019 67.27 0.0031 71.20 5 - 10 0.0007 64.17 0.0007 67.70 0.0015 62.94 0.0019 63.48 0.0008 63.76 0.0007 68.07 10 - 15 0.0008 59.39 0.0006 62.94 0.0012 58.40 0.0015 59.06 0.0007 59.01 0.0005 63.29 15 - 20 0.0011 54.62 0.0013 58.13 0.0016 53.74 0.0022 54.50 0.0013 54.19 0.0013 58.44 20 - 25 0.0016 49.91 0.0015 53.48 0.0024 49.17 0.0030 50.06 0.0023 49.54 0.0014 53.81 25 - 30 0.0024 45.28 0.0015 48.86 0.0029 44.73 0.0029 45.77 0.0035 45.08 0.0017 49.18 30 - 35 0.0032 40.79 0.0016 44.21 0.0033 40.34 0.0027 41.40 0.0036 40.82 0.0014 44.58 35 - 40 0.0045 36.41 0.0022 39.55 0.0041 35.97 0.0032 36.93 0.0044 36.51 0.0019 39.87 40 - 45 0.0051 32.18 0.0025 34.95 0.0054 31.66 0.0043 32.48 0.0056 32.27 0.0025 35.24 45 - 50 0.0073 27.95 0.0038 30.36 0.0071 27.45 0.0048 28.13 0.0076 28.12 0.0034 30.65 50 - 55 0.0100 23.89 0.0067 25.89 0.0109 23.34 0.0097 23.75 0.0106 24.11 0.0064 26.13 55 - 60 0.0162 19.98 0.0111 21.67 0.0180 19.50 0.0147 19.79 0.0176 20.27 0.0121 21.89 60 - 65 0.0232 16.45 0.0147 17.76 0.0231 16.09 0.0192 16.11 0.0187 16.91 0.0120 18.10 65 - 70 0.0324 13.16 0.0224 13.92 0.0392 12.75 0.0360 12.47 0.0289 13.32 0.0238 14.05 70 - 75 0.0466 10.03 0.0423 10.26 0.0522 9.97 0.0544 9.43 0.0433 9.99 0.0388 10.50 75+ 0.0948 6.99 0.0897 7.06 0.0966 7.19 0.1065 6.59 0.0947 6.78 0.0861 7.19 RAJASTHAN TAMIL NADU WEST BANGAL AGE Male Female Male Female Male Female GROUP Adj. ASDR E(X)

• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR E(X)
• Adj. ASDR

E(X) Adj. ASDR E(X) 0-1 0.0012 67.89 0.0012 71.39 0.0081 67.79 0.0083 70.60 0.0098 67.38 0.0099 69.45 1-5 0.0003 66.97 0.0004 70.47 0.0014 67.34 0.0012 70.19 0.0022 67.04 0.0024 69.14 5 - 10 0.0014 63.04 0.0015 66.57 0.0007 63.72 0.0005 66.53 0.0009 63.61 0.0010 65.79 10 - 15 0.0009 58.47 0.0010 62.06 0.0007 58.94 0.0006 61.69 0.0009 58.90 0.0008 61.10 15 - 20 0.0015 53.73 0.0017 57.36 0.0012 54.14 0.0015 56.88 0.0013 54.14 0.0017 56.35 20 - 25 0.0020 49.11 0.0021 52.82 0.0020 49.46 0.0020 52.29 0.0017 49.48 0.0019 51.80 25 - 30 0.0030 44.58 0.0022 48.34 0.0026 44.92 0.0018 47.78 0.0018 44.88 0.0017 47.27 30 - 35 0.0031 40.20 0.0020 43.84 0.0035 40.48 0.0020 43.19 0.0024 40.27 0.0019 42.65 35 - 40 0.0037 35.79 0.0021 39.26 0.0039 36.15 0.0018 38.60 0.0033 35.73 0.0024 38.03 40 - 45 0.0049 31.41 0.0032 34.64 0.0055 31.82 0.0033 33.92 0.0049 31.27 0.0033 33.46 45 - 50 0.0069 27.13 0.0038 30.16 0.0071 27.63 0.0038 29.43 0.0074 26.98 0.0040 28.97 50 - 55 0.0102 22.99 0.0064 25.69 0.0111 23.54 0.0076 24.94 0.0097 22.89 0.0079 24.50 55 - 60 0.0185 19.05 0.0109 21.43 0.0163 19.73 0.0120 20.80 0.0184 18.90 0.0143 20.38 60 - 65 0.0235 15.64 0.0149 17.49 0.0227 16.19 0.0187 16.93 0.0280 15.46 0.0189 16.70 65 - 70 0.0410 12.27 0.0269 13.64 0.0379 12.83 0.0304 13.33 0.0446 12.40 0.0347 13.09 70 - 75 0.0535 9.49 0.0417 10.23 0.0521 9.98 0.0492 10.10 0.0587 9.87 0.0513 10.09 75+ 0.1063 6.63 0.0918 7.01 0.0960 7.21 0.0934 7.20 0.0988 7.40 0.0939 7.31