Results of survey on infant, child and maternal mortality in the Republic of Tajikistan in 2010

Transcription

1 Results of survey on infant, child and maternal mortality in the Republic of Tajikistan in 2010 Agency on Statistics under the President of the Republic of Tajikistan

2 In some cases, minor discrepancies in the results from the sum of the terms explained with rounding during the weighing procedure Editorial board: Mukhammadieva B. Z. Chairwoman of the editorial board Asoev A. A., Norov Q. D., Gukasova T. P., Kislitsyna Е.А. Agency on Statistics under the President of the Republic of Tajikistan, 2010 AGENCY ON STATISTICS, 2010

3 FOREWORD Agency on Statistics under the President of the Republic of Tajikistan is pleased to present the analytical report on the results of "Survey on infant, child and maternal mortality", which was held in Tajikistan in This work was funded under the project "Further development of national statistics in the Kyrgyz Republic, Turkmenistan and Tajikistan - Statistics 11" and is the first survey on infant, child and maternal mortality conducted by the statistical authorities of the Republic of Tajikistan. Agency on Statistics under the President of the Republic of Tajikistan is grateful for the provision of consulting services and advice in the development of survey methodology and sample design to experts of ICON-INSTITUT and DESTATIS Mr. Lars Lundrgen and Mr. Bart de Bruijn. Grateful acknowledgment is also to the leadership of the European Commission on behalf of the Bureau on Cooperation of EuropeAid for funding, the coordinating group of the project and for timely support in the implementation of this work. And, of course, to all employees of Statistics of the Republic of Tajikistan for their active participation and making a contribution in this work. In carrying out of field work was attended about 100 experts of the Agency on Statistics under the President of the Republic of Tajikistan, members of the Main Computer Centre, Department of Statistics statistical divisions of cities and regions. Huge appreciation is expressed to the Government of the Republic of Tajikistan, representatives of local authorities for their support and assistance in conducting the survey We want also to thank all the people who have kindly agreed to take part in this work. The Report was prepared by Mrs. Kyslitsyna Е.А. Head of the Department of Demography, Employment and Social statistics and Mr Kulov A. N. Head of the Section on Demography and population Employment, under the general management and editorship of Mrs. Mukhammadieva B. Z., the Project Leader, Director of the Agency on Statistics under the President of the Republic of Tajikistan. Dushanbe, Tajikistan August, 2010 Project Leader Mukhammadieva Bakhtiya Ziyodullaevna Agency on Statistics under the President of the Republic of Tajikistan AGENCY ON STATISTICS,

4 CONTENT Content Symbolic notation Introduction Part 1. Survey methodology 1.1. Aim and targets of survey System of indexes and definitions Sampling methodology Core survey components Interview and data processing Socio-demographic population characteristics Part II. Analysis of infant, child and mmaternal mortality in the Republic of Tajikistan. Estimation of the data coverage and data quality 2.1. System of statistical data collection on population growth (birth and death) Analysis of data on birth and deth for the period of Part III. Results of survey on infant, child and maternal mortality 3.1. Results of survey on birth registration in the Civil Registry Office (here and after register offices) Estimation of the birth level Results of survey on dead children in the register offices Estimation of infant and child mortality by standard method and Brass method Estimation of maternal mortality by indirect sisterhood method Conclusions and suggestions ANNEXES: ANNEX А: List of tables 176 Table 1. Results of the households interview by regions, by area type by data of survey Table 2. Data about number of livebirth and dead, which are registered in the register offices Table 3. Coefficients of birth rate, infant, child and maternal mortality (calculated based on registrations data in the register offices) Table 4. Children born by women in the age of years by regions, sex, registration and year of birth, in % AGENCY ON STATISTICS, 2010

5 Table 5. Children born by women in the age of years by area type, sex, sex, registration and year of birth, in % Table 6. Children born by women in the age of years by reasons of non-registration, by regions and sex according to the survey data Table 7. Children born by women in the age of years by reasons of non-registration and year of birth according to the survey data Table 8. Children born by women in the age of years by reasons of non-registration and educational level of mother Table 9. Children born by women in the age of years by birth registration, life status (alive, dead), regions and sex Table 10. Children born by women in the age of years by reasons of nonregistration, life status (alive, dead),regions and age of death Table 11. Children born by women in the age of years and dead by area type, sex, registration and age at death, in % Table 12. Children born by women in the age of years and dead by registration of daeth and birth Table 13. Children born by women in the age of years and dead by registration, by case of non-registration, area type and sex Table 14. Children born by women in the age of years and dead, which doesn t have birth registration, by case of non-registration and mother s education ANNEX B: The questionnaire of the survey on infant, child and maternal mortality in Tajikistan in ANNEX С: List of participants of the survey on infant, child and maternal mortality in Tajikistan in ANNEX D: Bibliography 200 AGENCY ON STATISTICS,

6 Following symbolic notation were used in the Report: thsnd. thousand - not applicable data not available WB World Bank DRS (RRP) Districts of Republican Subordination GBAO Gorno-Badakhshan Autonomous Oblast CRO Civil Registry Office (here and after register offices) MCC Main Computing Center TLSS Living Standards Survey 2007 LSST Living Standards Survey of Tajikistan 2002 MICS Multi-Indicutor Cluster Survey PPS Probability proportional to size PSU Primary sampling units PEC Personal electronic computer IMR Infant mortality rate CMR Child mortality rate SFR Summary fertility rate TFR Total fertility rate 142 AGENCY ON STATISTICS, 2010

7 INTRODUCTION The political and economic transformations in the former Soviet Union and Central Asian region, including in Tajikistan, in the 90 years of the XX century and the first decade of the XXI century have had a negative impact not only on the social, economic and political development, but also on demographic. The decrease in fertility and life expectancy, increased mortality and migration activity. The situation has changed and with registration in the registry offices (CRO), Ministry of Justice, wwhich are dealing with registration of births, deaths, marriages and divorces. In recent years, a decline of registration of births and deaths by registry offices, and this situation has a negative impact on the quality of statistical data on demography,which in turn put in turn put beside statisticians certain tasks, such as in demographic statistics area are assessment of the infant, child and maternal using the new international methods. Data on the number of deaths are a major component in calculating the population by age groups and regions of the republic in the period between censuses and demographic indicators to evaluate the development of the demographic situation in the country the average of life expectancy, the level of general mortality, infant, child and maternal mortality. Child and maternal mortality are important indicators that reflect the activity of health system and it is important to assess and monitor their condition. In Tajikistan, mortality study by constant statistical observation, based on data obtained from the registry offices on these data can be traced in the dynamics of change in performance on multiple criteria by sex, age, nationality and cause of death, region. However, the situation with the registration of both births and deaths, especially regarding children under 5 years, a more in-depth study of this issue with the using of new methods to assess the situation. In recent years there has been a rapid decline in child, infant and maternal mortality, but in a low register is difficult to determine what specifically caused this decline - actually improved the work of health authorities, or is it still due to low registration of causes of deaths. According to the registration of register offices number of children who died at the age of 5 years from 1990 to 2009 decreased by 3,5 times and amounted to 3488 people in 2009, the number of women dying from pregnancy, delivery, and in consequence of the postpartum period - 2,2 times and amounted to 40 cases. Infant mortality rate (deaths of children under 1 year) during this period the level declined from 40,4 deaths per 1000 live births to 12,0, the level of Child mortality rates (deaths of children aged AGENCY ON STATISTICS,

8 under 5 years) - with 61 per 1000 live births-excited to 14,8. The maternal mortality rate declined from 24,8 per live births to 17. However, data from surveys conducted in the republic on issues of the position of women and children (MICS) and living standards suggest that rates of infant, child and maternal mortality rates are 2 4 times higher than data obtained from the register offices and made according to the Multi- Indicator Cluster Survey 2005: Infant mortality rate - 65 per 1000 live births, child - 79 per 1000 live births, maternal - 97 per 100,000 live births. According to survey of living standards in 2007 the infant mortality rate was 46 per 1000 live births, child mortality rate Also, data MICS 2005 showed that the level of registration of the birth by register offices in the first year of life is 82%. In the MICS and living standards have been included the registration of births and estimation of infant and child mortality, but not all ask for the registration of children who died at the age of 5 have not been studied. Therefore, for a more in-depth study of this problem it was necessary to conduct a special examination, which would highlight the problems that emerged in the country in recent years. To this end, the Agency for Statistics of the President of the Republic of Tajikistan decided to pro-conduct a special survey on infant, child and maternal mortality, as well as problems with the registration in the registry offices. The survey was conducted within the project "Further development of national statistics in the Kyrgyz Republic, Turkmenistan and Tajikistan - Statistics 11" with funding support from the European to the Commission and the Office for Cooperation EuropeAid. Gathering information on births and deaths by interviewing people in households yielded comprehensive data on both a half-nomu number of dead children, and on the record by the registry office, which can be used to estimate infant and child mortality rates, using standard demographic methods and new methods of assessment. A survey of infant, child and maternal mortality conducted by statistical agencies in the country in October-November 2010 made it possible to obtain data to study issues of fertility, infant and maternal mortality, as well as issues related to the study of the completeness and quality of the data and registering them in the bodies registrar. Information on these mortality rates and causes of mortality are not registered will be used to refine the current calculations of the population, estimates of life expectancy, fertility and mortality. In an analytical report used data from a multi-indicator of cluster Survey (MICS) of women and children, 2000 and 2005, the survey data the standard of living of 2007, as well as current statistical data. The survey provided an opportunity to work out the statistics authorities inquestionnaire and use it later as an additional module in other surveys, as well as use in evaluating mortality, new international methods - Brass method (for estimating infant and child mortality) and "Sisterhood method " (to evaluate maternal mortality). 144 AGENCY ON STATISTICS, 2010

9 PART I. METHODOLOGY OF SURVEY INFANT, CHILD AND MATERNAL MORTALITY 1.1. Aim and targets of survey The purpose of this survey, infant, child and maternal mortality is to get more information on the survival of children between the ages of 5 years and pregnant women, as well as the level of registration of births and infant mortality in the registry offices of the Republic of Tajikistan. Survey targets are following: 1. Assessment of risk of death of small children in the first year or the first five years of their lives. This is measured by calculating, respectively, in infant mortality and mortality in children under 5 years old. For the calculations will be applied two methods - the standard method of estimating infant and child mortality and the method of Brass. 2. Assessment of risk of women deaths due to complications during pregnancy and childbirth. The measurements were made by calculating the rate of maternal mortality (with the use of nursing methods). 3. Estimation of the proportion of children whose death was registered by public authorities (Civil Registry Office), and the reasons for not registering. 4. Prepare an analytical report on the survey of infant, child and maternal mortality. 5. Provision of the Government of Tajikistan necessary information on the survey System of indexes and definitions The methodology and system of indicators included in the questionnaire on the component "A survey of infant, child and maternal mortality, developing-las on international standards and taking into account the situation in the Republic of Tajikistan. The unit of analysis and unit of observation in the survey of infant, child and maternal mortality is a private household, its members aged (men and women) and children, Ro-divshiesya in The survey used the following terms and definitions: The head of household. Usually, people have a clear understanding of who is the head of household, and may themselves determine the title. If there is doubt, then, most elderly people who can clearly answer the questions, can be regarded as the head of the AGENCY ON STATISTICS,

10 household. Head of household can be male or female.в обследовании использовались следующие термины и определения: Household. This is a group of people living together and dividing the total food. Household members usually eat food cooked in a hearth, and they distribute the cost of collection and cooking of food. A household may include people not related to him, such as servants. This concept differs from the concept of "family" which consists of a family - either blood ties or by marriage. Household can also consist of one person or one family, or be a big family. The household also differs from the concept of "home": the first - a group of people, the second - this building. Authorised person. A person who is responsible on behalf of another person who is absent or unable (unwilling) to answer. Respondent. The person answering for the survey. In this survey on mortality, can usually be different respondents within the household: people in the age group years. Habitual residence. Households consist of people who usually live in one dwelling, are unified budget and eat together. There may be situations where it is difficult to determine if a person lives usually people in the household or not: temporary visitors are not considered as household members and should not be included in the poll, but the permanent members of the household who is temporarily visiting other households (temporarily unavailable) should be included; servant - a member of the household if he / she usually lives in the household; a person who lives alone - a household with one person Sampling methodology The project is a pilot, so the sampling design was developed in such a way as to generate representative data on key indicators (fertility, infant, child and maternal mortality, the registration of births and deaths) of the population and the urban and rural areas of Tajikistan. A survey of mortality conducted in five major administrativeterritorial units: the city of Dushanbe, RRP, Sogd, Khatlon, GBAO, which allows us to obtain sufficiently reliable results. Construction of the sample for the survey was conducted separately for urban and rural areas. A base for construction of a sample survey served as the database is registered with the registry office of births in In urban area used one-step standard procedure for systematic sampling with a random selection from the beginning 146 AGENCY ON STATISTICS, 2010

11 of a database registered with the registry office of births in 2004, followed by a sample of addresses from the documentary records of births in Through the countryside used a two-stage probability sampling. In the first stage formed the sample of primary sampling units (PSUs - jamoats and villages) from a database of rural settlements as of 01/01/2010, drafted on the basis of current population statistics in proportion to population in the regions. Table 1. Population by urban and rural area, by regions on Population, thsnd people Frequency total urban rural total urban rural Republic 7529,6 1987,5 5542,1 1,000 1,0000 1,0000 GBAO 220,6 29,3 191,3 0,029 0,0147 0,0345 Sorg 2216,9 560,4 1656,5 0,294 0,2820 0,2989 Khatlon 2700,2 464,5 2235,7 0,359 0,2337 0,4034 Dushanbe 706,1 706,1 0,094 0,3553 0,0000 RRP 1685,8 227,2 1458,6 0,224 0,1143 0,2632 In the second stage is carried out the sample of household addresses from the documentary records of registered births in 2004, within the selected PSUs in the first stage (sub-districts). In general, rural areas of the republic was formed by 144 PSUs. There were surveyed 20 households in each PSU. Table 2. Final sample distribution by regions Number of selected Frequency Number of clusters by rural area total households urban rural total urban rural Republic ,0000 1,0000 1,0000 GBAO ,0221 0,0116 0,0347 Sorg ,2902 0,2832 0,2986 Khatlon ,3091 0,2312 0,4028 Dushanbe ,2114 0,3873 0,0000 RRP ,1672 0,0867 0,2639 The total sample size was 6340 households in urban areas households (54,6%) in rural areas households (45,4%). By region sample was distributed as follows: GBAO households (2,2%), Sogd households (29,2%), Khatlon oblast households (30,9%), Dushanbe households (21,1%), RRP households (16,7%). (Table 2). Geographically, the structure of the survey covers the entire country, and has an evaluation information at the level of urban and rural areas. To obtain sufficiently reliable results of the entire population of Tajikistan has developed a special procedure AGENCY ON STATISTICS,

12 for sampling (scheme), which resulted by random sampling was selected 48 districts of 58 and 48 of the 71 urban settlement and 142 rural and village jamoat with subordinate rural settlement (144 cluster) of 381. Feature representation in the sample of territorial units at various levels up to the selection of PSUs is shown in Table 3. Table 3. Distribution of represened in the sample of territorial units dy different levels Districts Towns and villages Jamoats dekhot Regions Sample Total Sample Total Sample Total Tajikistan GBAO Sorg Khatlon Dushanbe RRP Condition survey was not resolved replacement households - it was necessary to interview exactly a household where the child lives in the sample. In general, the poll received 71% fully or partially completed the interview. Due to the fact that the population of urban settlements is very actively involved in all types of migration, the survey results in full or partially completed amounted to 60,7% (4498 households), the city of Dushanbe - 47,7% (640 households). In the countryside their share amounted to 83,3% (2399 households). At number of partially completed questionnaires to households include those households in which someone from the members to be questioned absent (dead or temporarily moved), refused to answer questions one of the modules, or at the time of the survey were sent on mission to him more than 49 years. 1 1 During the sample on the Districts of Republican Subordination (RRP) the districts of Rasht area were excluded 148 AGENCY ON STATISTICS, 2010

13 Table 4. Survey results Total Interview finished completely Interview finished partly Survey results Household was not found Was out of home Refusal Other Number of interviewed Total Urban Rural Dushanbe Khatlon Sogd RRP GBAO % Total 100,0 69,7 1,2 0,6 0,3 23,8 4,4 Urban 100,0 59,4 1,3 0,4 0,5 33,8 4,6 Rural 100,0 82,2 1,1 0,8 0,2 11,7 4,0 Dushanbe 100,0 46,9 0,8 0,3 0,5 44,6 6,9 Khatlon 100,0 67,5 1,1 0,1 0,2 26,6 4,5 Sogd 100,0 84,6 1,6 0,2 0,2 11,6 1,8 RRP 100,0 77,4 1,2 2,6 0,6 14,8 3,4 GBAO 100,0 66,4 2, ,1 18,6 The survey results for the calculation of infant, child and maternal mortality rates were weighted by assigning an appropriate weight of each individual unit of observation - the person. Using the method of iterative weighting sample, using as all the data population on 1 January Core survey components The main component of the survey questionnaire is designed by international consultants. Also developed Guidelines on how to conduct surveys of infant, child and maternal mortality. The questionnaire is divided into four main sections or modules, each of which contains a specific topic. Module "Data of the household" is the title page of the questionnaire. Information on this module is filled regardless of the final results of the interview. It provides the information necessary to determine the household as well as some of the key results of the interviews. All question numbers on household begin with the letter "H". Module "List of members of the household." This module contains information on all household members who usually live in it. For each person there is a line number in the first column (R1). Other modules can relate to this li- AGENCY ON STATISTICS,

14 ne number to identify the person in the household. All questions of the module begin with the letter «R». Module "Reproduction" Interviewed all the women members of households aged 15-49, each of them filled with some form of this module on reproduction. The module contains information on all births and deaths of children who were at the woman throughout her life. This information is necessary for a complete list of all children and for the calculation of infant and child (age 5) mortality. All questions of the module begin with the letter «F». Module "Child mortality". This module is filled with all women, members of the household aged who have ever had live births. This information can be obtained from the module "Reproduction", and only for those women who had children, and for which the module is filled with "Reproduction" is filled with a module on"child mortality". Based on the information presented in this module - possible to calculate infant and child (age 5) mortality. In addition, the module provides information on registering of births and deaths, and the reasons for non-registering of births and deaths. All questions of the module begin with the letter "C". Module "Maternal mortality". Interviewed all household members, aged years (women and men), and information on each respondent to fill in a separate form. Module requests information on all the brothers and sisters of the respondent, who were born by his mother. All questions of the module begin with the letter "M " Interview and data processing Survey on infant, child and maternal mortality was conducted by the Agency on Statistics under the President of the Republic of Tajikistan within 19 calendar days from October 23 to November 10, 2010 by specially trained experts. There were created 18 groups: 4 in Dushanbe, 1 - in GBAO, 3 - in the RRP, the 5 in the Sogd and Khatlon regions. Each group includes 1 supervisor and 3 interviewers. Total in the survey attended by 18 supervisors and 55 interviewers. On average, every interviewer must conduct 115 interviews (six interviews a day). The survey was conducted by interviewing people in selected households and record answers to questions in the questionnaire survey. The survey was conducted by the interviewer through direct visits to the household. In the absence of the household at the time of the survey include a visit to the household to three times. Supervisors were given lists of children in the sample of urban and jamoats and addresses where they lived in It was necessary to clarify the address of the residence of the child at the time of the survey and pro-poll lead is in the household, 150 AGENCY ON STATISTICS, 2010

15 where he currently resides rebbe, films, ensnared in the sample. Replace the household was not allowed. Table 5. Burden distribution per interviewer by regions Numbers of households Numbers Numbers of Burden per interviewer total urban rural of group interviewers total per day Republic ,3 6,1 GBAO ,0 3,7 Sorg ,0 6,1 Khatlon ,5 6,4 Dushanbe ,7 5,9 RRP ,8 6,2 Program data input on the survey was designed using the software CSPro 3.3, which among other things, include the necessary range checking and consistency of data. When entering data was placed under the control of the variables on the basis of existing rules set out in the questionnaire. After collecting and entering data, the survey was created a database containing a list of indicators. The database is in the latter was exported to the format of software for statistical analysis SPSS-15. Tabs output tables are also carried out by software SPSS- 15. Based on survey data of about 20 output tables, taking into account the weight of the total population, which presents data on urban and rural areas of the country. Scheme for processing the survey data provides: record preliminary information from the questionnaire forms on computers and its logical control; calculation of individual weights; formation of aggregated data in tabular form using of software SPSS; data analysis; preparation and issue a report; presentation of the survey to users; dissemination of results; archiving of documents for inspection. AGENCY ON STATISTICS,

16 1.6. Socio-demographic population characteristics In 4501 all or part of the households surveyed had received responses from people in urban area people (44,8%) in rural areas people (55,2%). Among the respondents, men were people (49,0%), women men (51,0%). Distribution of surveyed age groups of population was as follows: the population aged years amounted to persons, men persons (47,7%), women - 6,582 persons (52,3%), children under 1 year 675 people (boys - 356, Girls persons), children under 5 years people (male people, girls people) Table 6. Population of survyed persons by age Persons % Total under years under years year years year years Total ,3 14,2 42,2 Men ,4 15,1 41,1 Women ,1 13,4 43,3 Urban ,2 11,5 42,7 Men ,3 11,3 41,6 Women ,1 11,6 43,7 Rural ,3 7,5 41,9 Мужчины ,5 13,6 40,7 Женщины ,1 11,1 43,0 The average size of households surveyed was 6,6 persons: in urban areas persons, in rural areas 6,9 persons Table 7. Average household size by survey s data, persons Urban Rural Total Tajikistan 6,4 6,9 6,6 GBAO 5,6 6,4 6,2 Sorg 6,2 6,4 6,3 Khatlon 6,3 7,1 6,8 Dushanbe 6,5-6,5 RRP 6,6 7,2 7,1 Have different levels of education 97.8% of respondents aged 15 and older, men - 98,2%, women - 97,5%. 152 AGENCY ON STATISTICS, 2010

17 Table 8. Educational level of interviewed population in the age of 15 years and elder Persons % men women total men women total Total ,0 100,0 100,0 high ,4 6,4 11,6 secondary technical ,2 7,0 9,5 secondary completed ,7 57,2 55,1 secondary general ,7 23,4 18,8 primary ,2 3,6 2,9 doesn t have ,4 2,4 1,9 not standard ,3 0,0 0,2 don t know ,1 0,1 0,1 When analyzing and comparing the survey data of infant, child and maternal mortality data of current statistics and other surveys should consider the following specific points of inspection: It was a contemporaneous survey which was conducted in October-November. Method of survey. The survey was conducted with the words of the respondents without the requirement documents. The sample did not include 7 districts of republican subordination. AGENCY ON STATISTICS,

18 Part II. ANALYSIS OF INFANT, CHILD AND MMATERNAL MORTALITY IN THE REPUBLIC OF TAJIKISTAN. ESTIMATION OF THE DATA COVERAGE AND DATA QUALITY Carrying out a qualitative analysis of the demographic situation in Tajikistan is complicated by deterioration in the quality of population data. Reliability of the information affects the accuracy of calculations of many demographic indicators, particularly life expectancy, population size, fertility and mortality. Therefore, in the situation assessment of the quality and completeness of data on the population and, above all, data on fertility and mortality is important. To assess the quality of data on population statistics authorities of the republic is constantly carried out some work. Various methods of analysis, evaluation and comparison data. This report will be submitted to the survey results in comparison with other studies and data of current statistics on births and deaths 2.1. System of statistical data collection on population growth (birth and death) The country collection, development and compilation of vital statistics (births, deaths, marriages and divorces) are assigned to the statistical authorities. Registration of births, deaths, marriages and divorces performed in the departments of civil registry (Registry Office), Ministry of Justice and jamoats (village administration) of rural settlements, where filled Documents on marital status and given the population of the birth, death, marriage and divorce. Law (Family Code) of the Republic of Tajikistan provides deadlines for registering births - within 1 month, death - within three days. Registration of births and deaths by parents (or relatives), and the death of children in the first weeks of life - health care providers. Second copies of the documentary records are received each month in the regional statistical offices. For an assembly accounts of death and stillbirth attached medical certificate of death (stillbirth), indicating the cause of death (stillbirth). Further Documents on marital status are transferred to the Main Computing Centre of the Agency for Statistics of the President's Office) for further development. Preliminary results are published in the quarterly report on the socio-economic situation of the Republic of Tajikistan. Complete development of the Ito-gov vital statistics (gender, age, nationality, place of residence, cause of death and other indicators) is the statistical authorities once a year. Such a system of registration, and dissemination of data on vital statistics in Tajikistan, established in 1924 and exist to this day. 154 AGENCY ON STATISTICS, 2010

19 Documents on marital status and medical death certificates are basic and the most comprehensive source for information on fertility and mortality. The main requirements for this system to the sources of data on births and deaths for the statistics are: timeliness of registration; quality of submissions - the completeness and accuracy of filling the data in the documentary records and medical certificates of death. 2.2 Analysis of data on birth and deth for the period of (by registration in register ofiices CRO) Over 20 years of transition the demographic situation in the Republic of Tajikistan has changed and developed it under the influence not only economic crisis but also the demographic transition from high fertility to low fertility, low mortality rates. Currently, the overall mortality rate of 4,2-4,4 per 1000 population, total fertility rate per 1000 population. Declining birth rates in Tajikistan started in the 70-ies of the XX century. Studying the demographics of the data developed by statistical services on the basis of coming from the registry office of the documentary records, it is necessary on the one hand, to assess the direct impact of the demographic transition in the figures, on the other hand, how these changes relate to the completeness and quality of registration of births and deaths. During these years, the completeness and timeliness of registration of births and deaths registry offices within the period prescribed by law, particularly acute, as registry offices following situation: decreased registration of births and deaths (especially children died at the age of 5 years), marriage and divorce; violated terms of registration of births and deaths, established by law; decreased quality of entries in the documentary records and medical certificates of death. All this has led to the fact that fertility, mortality, particularly infant, child and maternal mortality (women who died during pregnancy, childbirth and postpartum period), beginning in 1994, an annual decline. Thus, according to data presented in Annexx Table 2, the registration of births from 1989 to 2000 decreased by 1,8 times (from 200,400 in 1989 to 110,500 in 2000). AGENCY ON STATISTICS,

20 Number of registered births in the register offices, perons However, besides the general decline in birth registration is a particular problem is "Late Registration" that is birth registration in the registry offices in violation of the terms established by law. During these years, we have a situation where the parents have started to register their children and receive a birth certificate, when the child is more than three years old when the child draw up a kindergarten or school. In the years , "Late Registration" by various estimates, up to 40%. But the biggest problem has developed with the registration of deaths of children age 5 and maternal mortality. If the whole period from 1989 to 2000, registration of deaths has decreased by 20,7% (from 33,395 to 26,492 cases), the registration of deaths in children under 1 year decreased by 4 times (from 8673 to 2102 cases), children 1-4 years - in 3,4 times (from 4364 to 1267 cases). Number of registered children who were die, under 5 years age, persons children, died in the age of under 1 year children died in the age of 1-4 years 156 AGENCY ON STATISTICS, 2010

21 Registration of women dying during pregnancy, childbirth and after postpartum period decreased by 25,6% (from 78 to 58 cases). Number of women, who were die during pregnancy, delivering and postpartum period, by information from registry offices, persons The main reasons for this situation were: introduction to the registry offices registration fee and birth certificate in the amount of $ 3 for a death certificate - $ 2; decline registration of births in the period prescribed by law, which led to a decrease in the registration of children who died at the age of 5 years; increase in the number of home deliveries, often without medical staff (their share was more than 40%); decline registration of marriages in registry offices - children born of undocumented marriage on registered "with the establishment of paternity." If the father is absent (in labor migration), the wife is waiting for her husband to enroll the child in his name. Registration of marriages from 1989 to 2000 decreased by 1,8 times (from 47,616 to 26,597). This situation has created many problems and this issue in subsequent years was paid enough attention on the part of stakeholders bathrooms ministries and agencies and from the Government. Was reduced fee for registration of births to 1 dollar, adopted the Law "On regulation of traditions, celebration and ceremonies" (June 8, 2007, 272), decreased the proportion of home births (13%), increased registration of marriages (in 3,8 times ). All this has led to positive changes and impact on the growth of birth registration in the registry offices. After 2000, the registration of births annually grown and in 2009 co-put up to 235,400 children, have increased by 2 times. From 2000 to 2009 was detected births in the period prescribed by law has grown in 1,6 times. However, the proportion of "Late AGENCY ON STATISTICS,

22 Registration" is still high and on data for 2009 was 26,3% or children were registered in the registry offices of breach of terms of registration. The situation in the registration of deaths continues to be critical. Pla-ta for the registration of the death of both adults and children is $ 2, which is not turns out to be a positive influence on the growth of registration of deaths of children. At the burial of the deceased adult allowance is issued, and the funeral of a child - no. As a result, indicators of registration of deaths of adults (aged 18 and older) for the period increased by 19,6% and sotavili in 2009, 26,700 people died at the age of children under 1 year increased by 33,9% (from 2102 to 2814), and children who died at the age of 1-4 years have decreased in 1,9 times (from 1,267 to 674). Registration of women dying from pregnancy, childbirth and after delivery period also declined over this period, 1,5 times (from 58 to 40 cases). These data show that there are some positive results in the registration, especially birth and not weighty enough to register the death of children. Is it correctly reflect the real situation - down, stabilize or grow? In these circumstances it was necessary to conduct additional analysis and to this end by statistical survey was conducted in infant, child and maternal mortality, whose results are presented in the next part of the analytical report. Reference: In the Republic from by data from the register offices (CRO) children were die in the following ages: under 1 year years years AGENCY ON STATISTICS, 2010

23 Part III. RESULTS OF SURVEY ON INFANT, CHILD AND MATERNAL MORTALITY OF POPULATION IN THE REPUBLIC OF TAJIKISTAN 3.1. Results of survey on birth registration in the Civil Registry Office The question of birth registration included in the survey, conducted in the republic since 2000: Multi-Indicator Cluster Survey of Women and Children (2000 and 2005) and surveys of living standards (2003 and 2007). This question was included to the survey of infant, child and maternal mortality rates in Data from these surveys indicate positive changes in the issues of birth registration. The level of birth registration The following table according to surveys from 2000 to 2010 enjoyed a steady growth in the birth registration of children. So, if in 2000 the proportion of children registered in the registry offices under the age of 5 years was 74,6% in ,3%, in ,3%, in ,3 %. Data about registration of children in the age of under 5 years, by number of months (survey data), in % Less then 6 monts Total MICS ,9 60,6 74,4 78,2 84,8 87,6 74,6 LSST ,3 MICS ,1 86,8 90,0 90,5 91,8 88,3 Survey on Infant, Child and Maternal Mortality ,1 94,4 95,6 96,1 96,9 96,4 95,3 urban 83,0 94,6 93,9 96,0 97,4 96,5 94,7 rural 90,3 94,2 96,7 96,1 96,6 96,3 95,7 boys 85,8 95,6 95,1 96,4 97,5 95,8 95,1 girs 88,6 93,2 96,1 95,7 96,3 96,9 95,5 Dushanbe 66,0 88,9 81,7 91,0 97,1 94,7 88,9 Khatlon 89,1 95,1 98,3 96,8 96,7 96,0 96,0 Sogd 96,1 100,0 99,2 97,6 97,9 97,8 98,0 RRP 83,3 85,1 94,4 95,6 96,7 95,7 94,0 GBAO 66,7 100,0 88,9 90,9 82,4 94,1 88,8 However, despite the steady growth in registration of births, the proportion of children registered during the first year of life saved at a lower level than in later ages - by the 2000 survey to first year of life was reported only 60,6% of children in 2005 year - 82,1%, in ,4%. The highest level of registration of births is observed in Sogd - to 5 years in the proportion of registered births was 98.0%. The lowest rates of birth registration in the city of Dushanbe and GBAO, and in 2010 amounted to 88,9% and 88,8%. AGENCY ON STATISTICS,

24 Data about registration of children in the age of under 5 years, by regions, in % Republic Dushanbe Sogd Khatlon GBAO RRP MICS ,6 62,2 92,0 71,1 74,1 62,9 LSST ,3 82,1 94,3 82,4 76,2 73,1 MICS ,3 82,7 93,6 89,5 90,8 80,9 Survey on Infant, Child and Maternal Mortality ,3 88,9 98,0 96,0 88,8 94,0 By statistical demographic indicators of fertility and mortality are estimated for the year. These estimates are used in calculating the population, mortality tables and life expectancy, general and age-specific fertility and mortality, so it was important for the survey to explore these issues by year of birth. If we analyze the data registration of births for 2010, the level of birth registration in the registry offices for the whole country is 89,6% in urban areas 86,8% in rural areas - 91,9%. By regions level of birth registration in the first year of life ranges from 75% in Dushanbe, to 97,6% - in Sogdiana. N other regions the level of registration is: the RRS and Badakhshan - 85% in Khatlon - 90,6%. With age, the child is not registered share is reduced, but continues to persist. The level of registration of children born in 2009 amounts to 95,6% in urban areas 94,3% in rural areas - 96,5%. But the lowest birth registration continues to persist in Dushanbe. The problem with the registration of births exists at a later age children - even to 10 years (children born 2001) is about 3% of children having no registration in the registry office and organh not obtain a birth certificate (in Dushanbe about 6% of children). The question is - and whether the children are learning in school without birth certificate? In schools, for admission of students, must appear on the birth certificate. The survey revealed a fairly low rate of registration of children born in 2003 (90,9%) in all regions. 160 AGENCY ON STATISTICS, 2010

25 Birth registration level, by year of birth and residence, by survey data, in % percentage, % 100,0 98,0 96,0 94,0 92,0 90,0 88,0 86,0 84,0 82,0 80, and earlier urban rural Large differences in the registration of births of boys and girls are not observed, but even so the level of birth registration of girls by about 0,5-1 percentage points higher. Reasons for non-registering births When the survey took into account a list of reasons, which was adopted in other surveys: high cost; far; do not know what necessary to record; did not know where to register; no time; other; i don t know. The greatest share among the reasons have to answer "other" and "I do not know" - 38,7% and 31,1% respectively. Among other reasons are such as "lack of documentation","late in registering", "did not want to register". The high specific weight of answere don t know is about low awareness of parents about theirs and child s rights. And as a result, parents are not going to register the birth of the child, believing that this is no longer necessary as the registration of birth in later, than the time prescribed by law, the fee is higher than the period specified by law. Therefore, parents postpone this process until the moment when it will be necessary, for example, give the child to school. For other reasons, the share of non-registration of births is differentiated by region, by area, age of the child. As shown by the survey data the high cost, as it did in previous years, is no longer the main reason for non-registering and reached 4,5%, while for children aged under 5 AGENCY ON STATISTICS,

26 years - 8,9%. According to MICS , the high cost of registration, as one of the reasons for not registering the birth, was for children under 5 years of 41,6%. In total, the reasons for not registering the birth "no time" reduces arrive at the 11,5% of the answers, "do not know what to register" - 7,1%,"far away"- 6,5%,"did not know where to register"- 0,7%. It should be noted that in urban areas the proportion of reasons "high cost" and "no time" is higher than in rural areas is 5,3% and 13,5% against 3,7% and 9,7%. At the same time, the share of reasons "far" and "do not know what to record in rural areas than in urban areas and rural soctavili 7,5% and 8,1% against 5,3% and 6,0% in urban areas. The largest proportion of births are not registered because of the "high cost" is marked in Dushanbe and Khatlon region - 5,6% and 5,2%, due to "far away" - in Khatlon (7,6%) and Sogd (14,8 ) areas, "do not know what to record - in Sogd (13,6%) and RRP (10,8%). In GBAO, these issues are not reasons for not registering. In Badakhshan has the highest proportion of the reasons for "no time" - 23,8%, in Dushanbe - 18,1%, in RRP - 16,5%. In Khatlon to share this cause has 6,7% and in Sogd - only 1,1% Reasons for non-registration of borth in urban area, in % don't know 31% high cost 5% far 5% no time 14% don't know what necessary to record 6% didn't know where to register 0% other 39% 162 AGENCY ON STATISTICS, 2010

27 Reasons for non-registration of borth in rural area, in % don't know 31% high cost 4% far 7% no time 10% don't know what necessary to record 8% didn't know where to register 1% other 39% Regarding reasons for not registering the birth year of birth of the child (or age), then the younger the child, the higher the proportion of the causes of "high cost" and "no time". For children born between , these causes accounted for 8,9% and 24,1%, whereas for children born ,4% and 8.3% respectively. With age of child the specific weight of the causes of "far" and "do not know what to register" is growing. For children born between in the proportion of those reasons have to 8,3% and 11,5%, for children born before ,0%, and for children born between ,4% and 3,9%. Also with the child's age increases the share of answers "other" and "I do not know" and a total of 58,1% for children born cities, 66,3% - for children born between and 81,8% - for children born before Mother's education influences the level of birth registration. The lowest rate of registration is observed in mothers with no primary education (92,9%) and highest among mothers with secondary vocational and secondary education total - 97%. The level of birth registration of children of mothers with primary and secondary education amounted to 96,3% and 96,5% respectively. Sufficiently low levels of birth registration of children were seen in women with higher education - 96,1%. Birth registration also depends on whether the child is alive or dead. At the time of the survey the number of children ever born to women aged years amounted to persons. Of these, were alive people or 94,4%, died children (5,6%). Of the children alive at the time of the survey birth was registered in children (98,9%) who died - at 618 people (60,2%). That is, 40% of deceased children AGENCY ON STATISTICS,

28 birth was not registered. In urban areas the proportion of these children was 42,1% in rural areas - 38,1%. In Dushanbe, the proportion of such children is 56,1% in GBAO - 48,4%. Even in Sogd, where the share register of births is the highest, not the birth was registered in 27,7% of children died. This situation with the registration, especially of children who died directly affects the quality and complete data provided in the statistical offices of the Registry Office and as a result, affects the rates of infant and child mortality. Given that parents do not always register the birth of the child, the more they have no incentive to register the birth of a child is already dead and still pay for it. If among the reasons for not recording a live child answers "other" and "not know" are 56,5%, "no time" - 26,9% and "high cost" - 13,0%, the reasons for not registering the birth of the deceased child's answers "Other" and "Do not know" are 76,1%, "no time" - 4,2%, "he did not know that it is necessary to regist" - 10,3% and "far" - 8,8% Estimation of the birth level According shown in the previous part survey of births registration of children can be seen that the annual underestimation is about 10%, that is, if the data of registry office, in 2009, was registered within the period prescribed by law thousand children, in fact, this number should be thousand more. To assess the level of fertility statistical bodies using several indicators, among them the most frequently used: the total fertility rate, age-specific fertility rates and total fertility rate. The total fertility rate was calculated by the formula: N TFR = *1000 where, N- number of born children, P population P According to data presented in Annex in Table 3 shows that the overall birth rate coefficient, calculated on the basis of births registration from 1989 to 2000 declined from 38.8 per 1000 population to 17.9 in subsequent years have seen its growth to 2009 to 23.3 per 1000 population. By survey data the general fertility rates, calculated on a two-year period from 2010 to 2006, fluctuating between 25 to 30 per 1000 population, which is average for this period is 27,5 per 1000 of the population, in urban areas 25,7, in rural 28,2 per 1000 population. And in dynamics is traced its decline. 164 AGENCY ON STATISTICS, 2010

29 Birth rate index, per 1000 population Period total urban rural Calculated based on data from birth registration in term established by law ,8 28,4 43, ,4 24,9 31, ,8 13,9 20, ,3 22,0 23,7 Data of survey 2010 года ,6 23,3 27, ,4 28,0 31, ,5 25,6 26, ,4 25,1 28, ,0 23,4 25,6 In average for ,5 25,7 28,2 From the data shows that in comparison with 1989 the total fertility rate fell by 10 promille points, but its rate of decline was lower than the registration data. And in general, the dynamics can be traced its decline. 45,0 40,0 Estimated fertility rates according to birth registration and survey, per 1000 population 35,0 30,0 25,0 20,0 15,0 10,0 5,0 0, Total birth registration Registration at term establish by law Estimation of birth rate index by survey data Also according to the survey it was possible to estimate age-specific fertility rates (fertility) and the total fertility rate. Age-specific fertility rates were calculated using the formula: Nb К = i пр х1000 Nw where, Р i number of birth in age i; i Nw i number of women in age i AGENCY ON STATISTICS,

30 Summary fertility rate were calculated using the formula: Рi К ср = Ч i As seen in the table on survey data, significant differences in the levels of the age of fertility (fertility) of women on urban and rural areas is observed - the total fertility rate-the SFP 2 for the whole country for the period is 3,6 child in urban areas 3,3 and the rural - 3,7. In comparison with 1989 shows that setting the number of children in the family changed to average child index (3 4 of the Child in the family). And especially the reduction in the number of children among women who are living in the rural area - from 6,1 in 1989 to 3,7 in In urban areas the number of children per 1 woman declined by only 0,16 points. Age-specific coefficients of birth per 1000 women Mother s age Summary coefficient ,50 302,13 281,66 208,15 124,51 55,35 9,09 172,63 5,104 urban 38,59 233,24 200,62 128,14 61,15 21,33 2,52 116,28 3,430 rural 38,46 332,52 322,45 257,47 171,91 83,75 14,23 203,08 6, ,74 269,29 237,47 169,67 98,90 34,71 5,27 147,81 4,385 urban 56,22 238,81 199,81 135,62 66,79 17,88 2,84 120,93 3,592 rural 62,35 281,46 253,49 184,75 114,31 43,67 6,69 159,35 4, ,77 207,50 188,19 140,70 85,11 31,10 3,67 110,52 3,493 urban 46,42 205,94 166,80 117,26 62,45 17,37 2,02 98,85 3,092 rural 40,24 208,05 196,91 151,01 95,08 37,21 4,44 115,13 3,669 Data of survey ,6 121,9 265,4 169,4 99,8 43,6 11,6 103,3 3,576 urban 4,6 134,7 241,3 152,9 74,6 35,5 10,8 94,6 3,272 rural 3,2 117,6 273,5 175,6 110,7 47,1 11,9 106,5 3,698 From the data we can conclude that in the early 90-ies of XX century with the beginning of economic and political reforms fertility declined fairly rapidly, but the last 10 years (since 2000) in-group decline slowed somewhat. According to the survey data the totalfertility rate varies from per 1000 population, the SFR is on the level of 3,4 children per a woman of fertile age (15-49 years). 166 AGENCY ON STATISTICS, 2010

31 3.3. Results of survey on dead children in the register offices The question of registration of children deaths in the preliminary surveys conducted in the Republic has not been studied. This is the first survey on this issue. There were included to questionnaire a special module on infant mortality, on of results of which answer were received data on the level of registration in the register offices deceased children by year of birth, age at death and the reasons are not registered. According to the responses, the number of deaths of children in surveyed women aged years was 1026, of whom 584 boys and 442 girls. Of the total number of deaths of children are registered in registry offices only 45,5%. Among the dead boys registration level is 46,2%, among girls 44,5%. Registration in urban areas is higher than in rural and reached 47,3%, in rural - 44,3%. That is more than half of infant deaths in urban areas and rural areas, parents are not registered in registry offices, and so data on them are not available in the statistical agencies for the design and analysis. Moreover, if the birth of the deceased child was not registered in orgsnah registry office, the chances are that his death would not also registered very high and is 88,2%. Registration level on dead child by age of death and are type, in % 80,0 75,0 70,0 60,0 50,0 51,252,4 48,0 64,8 61,9 40,0 30,0 28,7 25,0 35,433,8 20,0 10,0 0,0 first week of life month less then year urban rural under 2 years 2 years+ If we consider the data on registration and the age at which child died, the lowest registration level of occurs during the first week of life - registration level is only 26.6%. With age, the level of registration of child s death increases - the first month of life registration level is 34,5%, in the first year - 48,1%, in under two years old - 42,6% and the age of 2 years or more - 63, 0%. As for the distribution of responses for reasons no recording of the dead children, the situation is similar with the causes are not registering the birth of that is the largest AGENCY ON STATISTICS,

32 share falls on the answers "other" (29,3%) and "I do not know" (41,7%). For other reasons, the largest number of respondents referred to the fact that "he did not know what to register" - 20,8%, "far away" - 4,5% and "no time" - 2,3%. And in urban areas the specific weight of these answers higher than in rural and it was composed as following - 22,3%, 6,7% and 2,7% in rural areas - 19,7%, 3,0% and 2,1 %, respectively. Depends directly on the level of registration of death of the child from the mother's education level in the survey was not revealed. The highest level of registration among mothers with secondary complete education - 46,8%, slightly lower than in mothers with secondary special and higher education - 45,5% and the lowest rate of registration among mothers with secondary and primary education %. Survey data showed that if the situation with the registration of births in the country improves, the registration of deaths of children remains very low and during , less than 52%. Registration level on dead child during the first year of life,in % 3.4. Estimation of infant and child mortality by standard method and Brass method According shown in the previous section data on survey of registration of deaths of children shows that the annual underestimation is about 50%, that is, if the data of registry office, for example, in 2009 died 2814 in children under 1 year, in fact, this number should be 2 times larger. According to the registration of deaths of children observed an annual decline of infant 2 and child mortality 3. According to Table 3 in Annex infant mortality from 1989 to 2000 decreased with 43,3 per 1000 live births to 19,0, and in 2009 was 12,0. By 2 Infant mortality rate number of children who were die in the age of under 1 year per 1000 livebirths 3 Child mortality rate number of children who were die in the age of under 5 years per 1000 livebirths 168 AGENCY ON STATISTICS, 2010

33 analogy with the situation with the infant mortality rate - in 1989 it was equal to 65,0 per 1000 live births in ,5 in ,8. The question naturally arises, and whether the infant and child mortality declined over 20 years in 2,5 times? And if not, at what level at the present time is the country on these indicators. To analyze the dynamics of the infant and the infant mortality rate - down at the situationwith the quality of data registration in by the current data statistics used data from the surveys - in 2000 and 2005 was conducted Multi-Indicator Cluster Survey of the position of women and children, 2003 and survey of living standards of the population. However, the calculation of mortality rates according to the survey-tionsand the current statistics in different ways. In assessing the infant and child mortalitystatistics for the current direct method is used to estimate the mortality and survey data- a method of Brass. Infant and child mortality by direct distribution used to read the following formulas: 0 М М IMR = х CMR = 1000 Nb Nb х where, М 0 number of children who were die in the age of under 1 year; М 0-4 number of children who were die in the age of under 5 years; Р number of children was born during the year. If the infant and child mortality, calculated by the direct method show the mortality rate for the analyzed year, the coefficients calculated by the Brass are estimates for earlier years. In the survey on infant, child and maternal mortality rates, we estimate mortality rates by two methods direct method and the Brass. Data for the calculations by these methods are obtained modules "infant mortality". Infant and child mortality rate, calculated by direct method per 1000 livebirths Years Registration data from register offices Survey data 2010 IMR CMR IMR CMR ,3 65,0 51,1 75, ,6 49,0 58,9 84, ,0 30,5 50,0 75, ,0 16,7 16,4 25, ,4 15,5 38,0 70, ,4 13,6 32,5 49, ,2 14,5 22,2 37, ,0 14,8 21,4 33,9 Infant and child mortality, calculated by the direct method according to the survey as well as indicators calculated from the data on registration, illustrate the decline and infant and child mortality rate, yet their level in almost 2 times higher AGENCY ON STATISTICS,

34 and amounts to infant mortality rates of per 1000 livebirths by child In recent years, by urban areas, infant mortality rates higher and the child martality is lower than in rural areas (survey data are shown in table). Infant and child mortality rate, calculated by direct method by area type, by survey data per 1000 livebirths Years Infant mortality rates Child mortality rates urban rural urban rural ,2 61,0 52,8 83, ,4 68,7 117,4 101, ,0 57,9 85,9 83, ,8 45,9 82,5 73, ,5 40,6 71,4 70, ,7 32,4 48,0 49, ,7 21,0 30,9 40, ,7 21,3 27,9 35, ,2 15,7 38,1 31,7 Infant and child mortality in Tajikistan, calculated by the Brass method, also show a decline in infant and child mortality, but they are several times higher than rates calculated by the direct method. Infant and child mortality rate, calculated by Brass method, by survey data, per 1000 livebirths Infant mortality rates Child mortality rates Total urban rural boys girls Total urban rural boys girls According to the Multi-Indicator Cluster Survey, held in 2000 the infant mortality rate (under 1 year) was 89 per 1000 live births, child mortality rate (under 5 years) This is 2 times higher than the figures calculated by the direct method. According to the Census2005, these rates decreased and amounted to 65 and 79, respectively, according to a survey of living standards in and 53, according to a survey on infant, child and maternal mortality rates in and 50. Over the past 11 years, infant and child mortality rates, calculated by the Brass, fell more than 2 times. 170 AGENCY ON STATISTICS, 2010

35 Estimation of infant mortality rate, per 1000 livebirths 100,0 90, ,0 70, ,0 50, , ,0 20,0 10,0 0, direct method by survey data direct method by data of CRO Brass method (survey) Estimation of child mortality rate, per 1000 livebirths 140,0 120,0 100,0 80,0 60, direct method by survey data direct method by data of CRO 40, Brass method (survey) 20,0 0, From all the above we can conclude that the coefficients of infant mortality are fluctuating between 25 to 45 per 1000 live births and infant mortality - from 35 to 55. AGENCY ON STATISTICS,

36 3.5. Estimation of maternal mortality by indirect sisterhood method Maternal mortality is estimated by statistical bodies for the following reasons of death among women: during pregnancy; during delivery; in the period after childbirth. Maternal mortality is estimated by the direct method when the number of women dying per year from these causes, correlates with the number of live births by the following formula: Md MMR = N The level of registration of maternal mortality conditions of the survey could not be assessed because the number of women who are die during pregnancy, delivery and postpartum reriode quite small and the probability to hit of these deaths in a sample survey of almost equal "zero." So from the data in Table 3 in Annex shows that in the last decade, the number of deaths of women for these reasons that annual rate of cases, and maternal mortality rates is per live births. To estimate the maternal mortality survey using the direct method of calculation was not possible and has been used Sisterhood indirect method where maternal mortality is estimated from the answers of respondents aged years of all their brothers and sisters living and dead, but for sisters who died in age of 15 years and older separately identifiable cause of death during pregnancy, delivery and the postpartum period (within 6 weeks after birth). In comparing the maternal mortality data obtained by the indirect method of nursing must be understood that these figures correspond to the level of mortality, existed years ago, so the calculation used the average for the period , total fertility rate, equal to 4.0. Also please be aware that by this method estimates the number of maternal deaths could be included deaths sisters during pregnancy and after birth period is not associated with pregnancy, such as accidents, etc. This method was first applied in the country to assess the maternal mortality rate according to Multi-Indicator Cluster Survey of women and children conducted by UN- ICEF and the SSC in While maternal mortality was estimated at 97 per 100,000 live births, which corresponds to the level of maternal mortality According to current statistics in this period, maternal mortality rate is per live births. According to our survey maternal mortality rate is 86 per 100,000 live births, and complies with mortality According to current statistics in this period, maternal mortality rate is per live births. The calculations of maternal mortality is given in the following table (ata are weighted): 172 AGENCY ON STATISTICS, 2010

37 Aged Age R(i) S(i) k(i) SE(i) D(i) LTR(i) groups index (4)*(5) (7)/(6) (1) (2) (3) (4) (5) (6) (7) (8) , , , , , , , , , , , , , ,004 Total ,003 Sx 2,76 TFR: 4,01 MMR: 86 Where, R(i) - Number of respondents aged 15-49, by 5-year age group S(i) - Number of sisters who reached age 15+, by 5-year age groups of respondents Sx - Average number of sisters who reached age 15+ to respondents aged k(i) - Adjustment factors for number of sisters 15+, by age group of respondent (standard for all calculations) SE(i) - Sister-unit of exposure to mortality, by 5-year age group of respondents; representing the women-years of exposure to the risk of mortality in the reproductive age D(i) - Number of sisters of respondents who died during pregnancy, delivery or within six weeks of delivery, by 5-year age groups of respondents LTR(i) The fraction of adolescent females who would die eventually from maternal causes when competing causes of death are taken into account: TFR Summary fertility rate MMR Maternal mortality rate Based on the above data shows that the maternal mortality ratio from 1990 to 2000 and decreased its level in the range of 17 to 86. AGENCY ON STATISTICS,

38 Maternal mortality rate per livebirths 120,0 100,0 80,0 60,0 97,0 86 by date of registration in CRO 40,0 Sistrehood method (survey) 20,0 0, AGENCY ON STATISTICS, 2010

39 Conclusions Changes in the data on fertility and mortality over 20 years. Marked decrease in registration in the registry offices of births and deaths (especially children under 5 years), causing: Poor quality of statistical data on demography (fertility, mortality) Failure to conduct reliable demographic calculations (population, mortality tables and life expectancy, fertility tables, mortality rates) Failure to analyze the demographic situation in the republic The survey results found that: The level of birth registration in the registry offices increased from 74,6% in 2000 to 95,3% in 2010 Continuing decline in fertility, but the rate of decline slowed - Factor of birth rate dropped from 38,8 per 1000 population in 1989 to 27,5 Registration with the registry offices of deaths of children aged under 5 years is very low and does not exceed 50% The level of infant and child mortality is decreasing and subject to assessment by the direct method and the method of Brass is within for infant mortality and for child mortality per 1000 live births. The maternal mortality rate, calculated using the indirect method of nursing decreased from to , the year from 97 to 86 per 1000 livebirths, and according to current statistics from 58,2 to 52,5 per 1000 livebirths Suggestions To improve the situation with the registration of died children, it is necessary to cancel registration fee There is need to use the data to adjust of corrections for Demographic statistics - population, fertility and mortality (particular attention should be paid to assessing the number of children without birth and death registration) Use the survey data for re-calculating the population during the intercensus period Necessary to conduct such surveys at least 1 time in five years. AGENCY ON STATISTICS,

40 ANNEX А Table 1. Results of households interview by regions and area type, by survey data Total Interview finished completely Interview finished partly Survey results Household was not found Was out of home Refusal Other Total Total urban rural Dushanbe Khatlon urban rural Sogd urban rural RRP urban rural GBAO urban rural in % Total 100,0 66,1 4,9 0,6 1,2 23,0 4,4 urban 100,0 57,9 2,8 0,4 0,5 33,8 4,7 rural 100,0 75,9 7,4 0,8 2,0 9,9 4,0 Dushanbe 100,0 46,6 1,1 0,3 0,5 44,6 6,9 Khatlon 100,0 56,3 12,1 0,1 0,2 26,7 4,6 urban 100,0 50,1 6,1 0,0 0,3 39,1 4,4 rural 100,0 60,6 16,3 0,1 0,2 18,1 4,7 Sogd 100,0 84,3 1,9 0,2 3,0 8,8 1,8 urban 100,0 77,4 2,8 0,1 0,4 16,5 2,8 rural 100,0 92,1 0,9 0,2 6,0 0,0 0,7 RRP 100,0 77,1 1,5 2,6 0,6 14,8 3,4 urban 100,0 63,3 1,3 2,7 1,0 31,0 0,7 rural 100,0 82,5 1,6 2,6 0,4 8,4 4,5 GBAO 100,0 65,0 4,3 0,0 0,0 12,1 18,6 urban 100,0 70,0 2,5 0,0 0,0 15,0 12,5 rural 100,0 63,0 5,0 0,0 0,0 11,0 21,0 176 AGENCY ON STATISTICS, 2010

41 Table 2. Data on born and died children, registered in the register offices, persons Registration years Registration of children who was born Registration of children who was born at statutory term Children died in the age under 1 year Children died in the age 1-4 years Number of registered death causes of women during pregnancy, delivering and postpartum period AGENCY ON STATISTICS,

42 Years of registration Table 3. Birth rate and infant, child and maternal mortality rate (calculated based on data of register offices Birth rate index (with adjustment on underestimation), per 1000 population Birth rate index per 1000 population Total ferti lity rate Infant mortality rate per 1000 livebirths Child mortality rate per 1000 livebirths Maternal mortality rate per l ivebirths ,8 38,8 5,104 43,3 65,0 38, ,8 38,8 5,091 40,4 61,0 24, ,1 39,1 5,094 40,4 57,7 53, ,4 32,4 4,205 48,1 72,4 69, ,5 33,5 4,320 46,5 82,5 74, ,2 28,9 4,355 42,4 74,9 87, ,1 29,4 4,385 30,6 49,0 58, ,0 22,9 3,931 33,6 55,8 45, ,6 20,0 4,064 29,6 53,7 47, ,3 20,0 4,068 24,5 44,5 48, ,8 18,7 3,852 20,6 33,2 52, ,0 17,8 3,852 19,0 30,5 52, ,2 19,6 3,493 18,0 27,9 34, ,3 19,4 3,487 17,3 27,5 45, ,1... 3,471* 11,8 18,4 21, ,8 3,420* 12,5 17,8 22, ,4 3,354* 12,0 16,7 16, ,7 3,274* 11,4 15,5 21, ,1 3,266* 10,4 13,6 17, ,9 11,2 14,5 14, ,8 23,3 12,0 14,8 17,0 178 AGENCY ON STATISTICS, 2010

43 Table 4. Children was born by women in age of years, by regions, sex, registration and birth year, in % Birth year Total Dushanbe Khatlon boys girls both boys girls both boys girls both ,1 90,2 89,6 76,1 73,8 75,0 90,6 90,6 90, ,1 96,1 95,6 84,9 80,0 82,8 97,5 100,0 98, ,3 95,6 96,0 95,7 84,9 90,0 98,1 94,5 96, ,4 96,7 97,0 100,0 94,6 97,2 97,1 96,1 96, ,2 96,6 96,4 93,0 95,8 94,6 96,4 95,8 96, ,4 97,7 97,5 94,1 95,0 94,5 98,8 97,7 98, ,8 99,3 99,0 98,6 99,7 99,1 97,3 98,2 97, ,6 93,8 90,9 96,2 95,5 95,7 87,9 88,2 88, ,4 96,6 96,5 91,6 97,1 94,1 95,6 95,2 95, ,6 98,1 97,3 96,4 91,7 94,2 95,1 99,2 96, and early 95,9 96,6 96,2 94,4 94,3 94,3 95,9 97,3 96,6 Continuation Birth year Sogd RRP GBAO boys girls both boys girls both boys girls both ,0 97,1 97,6 79,6 89,7 85,0 88,9 80,0 85, ,7 100,0 99,3 92,5 93,7 93,0 90,9 85,7 88, ,6 98,5 97,5 94,1 98,8 96,2 83,3 100,0 92, ,8 97,6 98,2 94,4 100,0 96,7 90,9 71,4 83, ,5 98,0 97,8 95,7 95,5 95,6 85,7 100,0 93, ,0 100,0 99,4 95,5 98,0 96,6 100,0 83,3 90, ,8 99,8 99,8 99,6 99,7 99,7 96,8 100,0 98, ,8 98,0 91,7 86,4 96,7 90,5 66,7 100,0 83, ,7 97,7 98,2 98,4 96,5 97,5 90,0 100,0 95, ,4 100,0 99,7 96,2 96,5 96,3 50,0 100,0 81, and early 97,7 97,5 97,6 95,3 95,7 95,5 94,1 98,2 96,0 AGENCY ON STATISTICS,

44 Table 5. Children was born by women in age of years, by area type, sex, registration and birth year, in % Birth year Total Urban Rural boys girls both boys girls both boys girls both ,1 90,2 89,6 86,9 86,7 86,8 90,8 93,1 91, ,1 96,1 95,6 94,4 94,3 94,3 95,6 97,5 96, ,3 95,6 96,0 98,0 93,4 95,8 95,1 97,3 96, ,4 96,7 97,0 98,3 96,6 97,4 96,8 96,7 96, ,2 96,6 96,4 95,9 97,0 96,5 96,3 96,2 96, ,4 97,7 97,5 97,7 97,0 97,4 97,2 98,1 97, ,8 99,3 99,0 98,8 99,5 99,1 98,7 99,1 98, ,6 93,8 90,9 86,2 90,7 88,7 88,9 97,7 93, ,4 96,6 96,5 93,8 96,9 95,4 98,4 96,4 97, ,6 98,1 97,3 96,1 96,9 96,5 96,9 98,8 97, and early 95,9 96,6 96,2 96,4 96,3 96,3 95,6 96,8 96,2 Total 96,4 97,0 96,7 96,5 96,6 96,5 96,4 97,4 96,9 180 AGENCY ON STATISTICS, 2010

45 Table 6. Children was born by women in age of years, by reasons of nonregistration, by regions and sex, by survey data, in % Total urban rural Dusnabe Khatlon Sogd RRP GBAO high price far don t know necessary don t know where don t have time other don t know Total boys 3,3 7,2 7,8 0,9 12,3 39,3 29,1 100,0 girls 5,9 5,6 6,3 0,4 10,4 37,9 33,5 100,0 both 4,5 6,5 7,1 0,7 11,5 38,7 31,1 100,0 boys 2,8 5,6 7,6 0,0 11,8 40,3 31,9 100,0 girls 8,0 5,1 4,4 0,0 15,3 38,0 29,2 100,0 both 5,3 5,3 6,0 0,0 13,5 39,1 30,6 100,0 boys 3,7 8,5 7,9 1,6 12,7 38,6 27,0 100,0 girls 3,8 6,1 8,3 0,8 5,3 37,9 37,9 100,0 both 3,7 7,5 8,1 1,2 9,7 38,3 31,5 100,0 boys 1,4 4,2 11,1 0,0 15,3 44,4 23,6 100,0 girls 9,7 2,8 6,9 0,0 20,8 36,1 23,6 100,0 both 5,6 3,5 9,0 0,0 18,1 40,3 23,6 100,0 boys 6,0 10,3 1,7 1,7 6,8 29,9 43,6 100,0 girls 4,3 4,3 1,1 1,1 6,5 30,1 52,7 100,0 both 5,2 7,6 1,4 1,4 6,7 30,0 47,6 100,0 boys 2,1 17,0 19,1 2,1 2,1 36,2 21,3 100,0 girls 4,9 12,2 7,3 0,0 0,0 46,3 29,3 100,0 both 3,4 14,8 13,6 1,1 1,1 40,9 25,0 100,0 boys 2,4 1,2 8,5 0,0 19,5 48,8 19,5 100,0 girls 5,3 7,0 14,0 0,0 12,3 43,9 17,5 100,0 both 3,6 3,6 10,8 0,0 16,5 46,8 18,7 100,0 boys 0,0 0,0 0,0 0,0 33,3 46,7 20,0 100,0 girls 0,0 0,0 0,0 0,0 0,0 66,7 33,3 100,0 both 0,0 0,0 0,0 0,0 23,8 52,4 23,8 100,0 AGENCY ON STATISTICS,

46 Table 7. Children was born by women in age of years, by reasons of nonregistration and birth year, by survey data, in % high price far don t know necessary don t know where don t have time other don t know Total ,8 0,0 1,6 0,0 31,3 37,5 21,9 100, ,6 0,0 8,6 2,9 31,4 34,3 14,3 100, ,7 2,8 5,6 0,0 25,0 30,6 19,4 100, ,8 7,7 7,7 0,0 15,4 42,3 23,1 100, ,1 14,3 0,0 0,0 11,9 40,5 26,2 100, ,0 0,0 7,1 0,0 7,1 50,0 35,7 100, ,8 6,4 8,5 2,1 14,9 25,5 29,8 100, ,0 5,9 8,8 0,0 2,9 35,3 47,1 100, ,6 13,2 18,4 2,6 7,9 36,8 18,4 100, ,0 12,5 12,5 0,0 4,2 45,8 25,0 100, and early 0,8 7,0 7,0 0,4 2,9 42,1 39,7 100,0 Total 4,5 6,5 7,1 0,7 11,5 38,7 31,1 100,0 By birth years of children ,9 4,4 3,9 0,05 24,1 36,9 21,2 100, ,4 8,3 11,5 1,3 8,3 35,7 30,6 100, and early 0,8 7,0 7,0 0,04 2,9 42,1 39,7 100,0 Data on MICS 2005 For children under 5 years 41,6 6,2 3,3 4,0 16,4 11,6 16,9 100,0 Table 8. Children was born by women in age of years, by reasons of nonregistration and mother s educational level, by survey data, in % Don t have education Birth registration yes no Reasons of non-registration don t know necessary don t know where don t have time high price far other don t know Total Total 96,7 3,3 2,6 13,2 18,4 2,6 7,9 36,8 18,4 100 Don t have education 92,9 7,1 28,6 14,3 0,0 0,0 14,3 14,3 28,6 100 Primary 96,3 3,7 11,1 0,0 0,0 0,0 11,1 22,2 55,6 100 Secondary general 96,5 3,5 4,5 5,5 10,0 2,7 13,6 36,4 28,2 100 Secondary complete 96,9 3,1 4,3 8,0 6,6 0,3 10,4 38,8 31,6 100 Secondary technical 97,0 3,0 4,5 6,5 7,2 0,7 11,6 38,5 31,3 100 High 96,1 3,9 0,0 0,0 4,3 0,0 13,0 52,2 32, AGENCY ON STATISTICS, 2010

47 Table 9. Children was born by women in age of years by birth registration, life status (alive, dead), regions and sex, in % Specific weight by life status Birth was registered Birth was not registered Total Specific weight of registration to number of born children Birth was registered Birth was not registered AGENCY ON STATISTICS, Total Total ,7 3,3 100 alive 96,5 32,0 94,4 98,9 1,1 100 dead 3,5 68,0 5,6 60,2 39,8 100 Urban ,5 3,5 100 alive 96,8 35,8 94,7 98,7 1,3 100 dead 3,2 64,2 5,3 57,9 42,1 100 Rural ,9 3,1 100 alive 96,3 28,7 94,2 99,1 0,9 100 dead 3,7 71,3 5,8 61,9 38,1 100 Boys ,4 3,6 100 alive 96,0 32,5 93,7 98,8 1,2 100 dead 4,0 67,5 6,3 61,6 38,4 100 Girls ,0 3,0 100 alive 97,0 31,3 95,1 99,0 1,0 100 dead 3,0 68,7 4,9 58,4 41,6 100 Dushanbe ,5 5,5 100 alive 97,2 39,2 94,0 97,7 2,3 100 dead 2,8 60,8 6,0 43,9 56,1 100 Khatlon ,5 3,5 100 alive 96,6 39,2 94,6 98,6 1,4 100 dead 3,4 60,8 5,4 60,8 39,2 100 Sogd ,4 1,6 100 alive 96,2 9,1 94,9 99,9 0,1 100 dead 3,8 90,3 5,1 72,3 27,7 100 RRP ,2 3,8 100 alive 96,4 28,8 93,9 98,8 1,2 100 dead 3,6 71,2 6,1 56,4 43,6 100 GBAO ,0 6,0 100 alive 95,1 28,6 91,1 98,1 1,9 100 dead 4,9 71,4 8,9 51,6 48,4 100

48 Table 10. Children was born by women in age of years by reasons of nonregistration of birth, life status (alive, dead) and age of death, in % high price far don t know necessary don t know where don t have time other don t know Total Total born and not registered of birth 4,5 6,5 7,1 0,7 11,5 38,7 31,1 100,0 Child alive 13,0 1,6 0,5 1,6 26,9 29,0 27,5 100,0 Child dead 0,5 8,8 10,3 0,2 4,2 43,3 32,8 100,0 on the 1 st week of life 0,6 9,7 14,8 0,6 2,8 44,3 27,3 100,0 on the 1 st month of life 1,8 7,1 3,6 0,0 7,1 44,6 35,7 100,0 on the 1 st year of life 0,0 4,6 6,9 0,0 6,2 43,8 38,5 100,0 in the age of 2 years 0,0 17,2 6,9 0,0 0,0 37,9 37,9 100,0 in the age of 2 years and more 0,0 22,2 16,7 0,0 0,0 33,3 27,8 100,0 Total born and not registered of birth 100,0 100,0 100,0 100,0 100,0 100,0 100,0 100,0 Child alive 92,6 7,7 2,3 75,0 75,4 24,0 28,3 32,1 Child dead 7,4 92,3 97,7 25,0 24,6 76,0 71,7 67,9 on the 1 st week of life 100,0 100,0 100,0 100,0 100,0 100,0 100,0 100,0 on the 1 st month of life 50,0 47,2 61,9 100,0 29,4 44,1 35,8 43,0 on the 1 st year of life 50,0 11,1 4,8 0,0 23,5 14,1 14,9 13,7 in the age of 2 years 0,0 16,7 21,4 0,0 47,1 32,2 37,3 31,8 in the age of 2 years and more 0,0 13,9 4,8 0,0 0,0 6,2 8,2 7,1 Total born and not registered of birth 0,0 11,1 7,1 0,0 0,0 3,4 3,7 4,4 184 AGENCY ON STATISTICS, 2010

49 Table 11. Children was born by women in age of years and died, by area type, sex, registration and age of death, in % Age of death Total Urban Rural boys girls both boys girls both boys girls both Registered death of child 46,2 44,5 45,5 48,5 45,7 47,3 44,6 43,8 44,3 Age of dead child: on the 1 st week of life 27,7 25,0 26,6 30,6 26,4 28,7 25,8 23,7 25,0 on the 1 st month of life 39,4 26,2 34,5 41,9 23,5 35,4 37,5 28,0 33,8 on the 1 st year of life 53,2 50,3 51,9 49,5 53,5 51,2 56,2 48,1 52,4 in the age of 2 years 54,2 60,7 57,4 84,2 66,7 75,0 40,0 57,1 48,0 in the age of 2 years and more 63,0 63,2 63,0 66,7 62,5 64,8 60,8 63,6 61,9 Table 12. Children was born by women in age of years and died, by sex, registration of death and birth Birth registration Death registration In % to those who have or not birth registration In % to the total number of number of dead children exist not total exist not total Total dead 45,5 54,5 100,0 100,0 100,0 100,0 have registration 67,8 32,2 100,0 89,7 35,6 60,2 don t have registration 11,8 88,2 100,0 10,3 64,4 39,8 Boys 46,2 53,8 100,0 100,0 100,0 100,0 have registration 68,6 31,4 100,0 91,5 36,0 61,6 don t have registration 10,3 89,7 100,0 8,5 64,0 38,4 Girls 44,6 55,4 100,0 100,0 100,0 100,0 have registration 66,7 33,3 100,0 87,3 35,1 58,4 don t have registration 13,6 86,4 100,0 12,7 64,9 41,6 AGENCY ON STATISTICS,

50 Table 13. Children was born by women in age of years and died, by registration, reason of non-registration, by type area and sex, in % No registration From numbers of non-registered by reasons don t don t don t Total high know know have don t price far necessary where time other know Total 54,5 100,0 0,2 4,5 20,8 1,3 2,3 29,3 41,7 boys 53,8 100,0 0,3 5,7 21,0 1,0 2,5 28,3 41,1 girls 55,4 100,0 0,0 2,9 20,4 1,6 2,0 30,6 42,4 Urban 52,7 100,0 0,0 6,7 22,3 0,9 2,7 31,3 36,2 boys 51,5 100,0 0,0 9,8 25,2 0,0 3,3 28,5 33,3 girls 54,3 100,0 0,0 3,0 18,8 2,0 2,0 34,7 39,6 Rural 55,7 100,0 0,3 3,0 19,7 1,5 2,1 28,1 45,4 boys 55,4 100,0 0,5 3,1 18,3 1,6 2,1 28,3 46,1 girls 56,3 100,0 0,0 2,8 21,5 1,4 2,1 27,8 44,4 Table 14. Children was born by women in age of years and died without birth registration by reasons of non-registration and mother s education, in % No registration don t don t don t From numbers of non-registered by reasons Total high know know have don t Education price far necessary where time other know Total 54,4 100,0 0,2 4,5 20,8 1,3 2,3 29,3 41,7 Primary and secondary general 59,1 100,0 0,0 0,9 23,4 0,9 8,0 29,0 43,9 Secondary completed 53,2 100,0 0,3 5,6 20,9 1,6 10,1 28,8 40,7 Secondary technical and high 54,5 100,0 0,0 4,2 16,7 0,0 25,0 31,9 43,1 186 AGENCY ON STATISTICS, 2010

51 ANNEX B: AGENCY ON STATISTICS,

52 188 AGENCY ON STATISTICS, 2010

53 AGENCY ON STATISTICS,

54 Reproduction This module should be asked to all women aged Check for eligible women in question R7 (Household roster) and use for each a separate form F1 Household number F2 Woman's line number Now I would like to ask about all the deliveries you have had during your life. F3 Have you ever given birth? Yes 1 No. 2 F8 F4 Do you have any sons or daughters to whom you have Yes 1 given birth who are now living with you? No. 2 F6 F5 How many sons live with you? And how many daughters live with you If none, record '00' Sons at home Daughters at home F6 Do you have any sons or daughters to whom you have Yes 1 given birth who are alive but do not live with you? No. 2 F8 F7 How many sons are alive, but do not live with you? And how many daughters are alive, but do not live with you? If none, record '00' Sons elsewhere Daughters elsewhere F8 Have you ever given birth to a boy or girl who was born Yes 1 alive but later died? No. 2 F10 If 'No', probe: Any baby who cried or showed signs of life but did not survive? F9 How many boys have died? And how many girls have died? If none, record '00' Sons dead Daughters dead F10 Sum answers to F5, F7 and F9, and enter total If none, record 00 Total F11 Check F10 Just to make sure that I have this right: Probe you have had in TOTAL births Yes No and during your life. Is that correct? correct F3-F10 as necessary Check F10 One or more births No births Check Household roster, question R7 No other eligible Another eligible woman Continue with Module 'Child mortality' with this woman woman in the household in the household Continue with Module 'Maternal mortality' with this woman Continue with Module 'Reproduction' with that woman 190 AGENCY ON STATISTICS, 2010

55 AGENCY ON STATISTICS,

56 192 AGENCY ON STATISTICS, 2010

57 AGENCY ON STATISTICS,

58 194 AGENCY ON STATISTICS, 2010

59 AGENCY ON STATISTICS,