HUMAN DEVELOPMENT SECTOR UNIT Central Asia Country Unit | Europe and Central Asia Region Republic of Tajikistan Improving Statistics for Children’s Births and Deaths THE WORLD BANK World Bank Report Number 62871-TJ ON THE COVER Tajik mother and her newborn baby HUMAN DEVELOPMENT SECTOR UNIT Central Asia Country Unit Europe and Central Asia Region June, 2011 Republic of Tajikistan Improving Statistics for Children’s Births and Deaths THE WORLD BANK 66094_World_Bank_Txt.indd 1 9/30/11 6:22:06 AM 66094_World_Bank_Txt.indd 2 9/30/11 6:22:06 AM Table of Contents Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Acronyms, Abbreviations and Tajik Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Tajikistan Demographic Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Health Statistics in Tajikistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Vital Statistics: Child Birth & Death Registration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Chapter 2. Main Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Health Information system in Tajikistan – Structure and Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Tajikistan Laws and Regulations on Vital Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Efforts to Improve Tajikistan’s Health Information System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Chapter 3. Factors Affecting Vital Statistics in Tajikistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Civil Registration System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Health System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Community/Population Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Chapter 4. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Summary of Main Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK iii 66094_World_Bank_Txt.indd 3 9/30/11 6:22:06 AM Annexes Annex 1: Annex 1: Study Objective and Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Annex 2: Health Statistical Reporting Forms for Vital Registration in Tajikistan . . . . . . . . . . . . . . . . 17 Annex 3: Summary of Projects to Improve HIS in Tajikistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 List of Tables Table 1: Children aged 0-59 months that have not been registered . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Table of Figures Figure 1: Tajikistan Infant Mortality Rate Trends 1974-2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Figure 2: Tajikistan Under 5 Mortality Rate Trends 1974-2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 iv Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 4 9/30/11 6:22:06 AM Acknowledgments A World Bank team prepared this report with support from The report has benefited greatly from the peer reviewer the Global Alliance for Vaccines Initiative (GAVI) Trust Fund. comments (on both the concept note and the draft report) The preparation of the report was led by Wezi Msisha (Task provided by Ed Bos (Lead Population Specialist, HDNHE), Team Leader, Health Specialist, ECSH1), and the study was Nargis Rakhimova (UNFPA Tajikistan), Mustafa Dinc conceptualized by Anne Bakilana (Economist, ECSH1). (Senior Economist/Statistician, DECDG), Utkir Umarov The team members from the World Bank included Sarvinoz (Economist, PREM), and Son Nam Nguyen (Senior Health Barfieva (Operations Officer, ECSH1), Gabriel Francis Specialist, ECSHD). Charles Griffin (Sr. Advisor, ECAVP) (Program Assistant, ECSHD), Gulnora Kamilova (Program also provided very useful guidance and comments. Assistant, ECCUZ), Shoira Zukhurova (Team Assistant, ECCTJ), Ivdity Chikovani, Ketevan Chkhatarashvili, and The task was undertaken under the guidance of Tamar Natia Rukhadze (Curatio International Foundation). Editorial Manuelyan Atinc (former Sector Director, ECSHD), services were provided by Rosemarie Esber. Mamta Murthi (Acting Sector Director, ECSHD ), Abdo Yazbeck (former Sector Manager, ECSH1), Daniel Dulitzky We also extend our thanks to Dr. S. Rhakmatuloev (Head of (Sector Manager, ECSH1), Motoo Konishi (Country Maternal & Child Health Department [MCH], Ministry of Director, ECCU8), Chiara Bronchi (former Country Health, Tajikistan), Dr. S. Rakhmatullaeva (Deputy Head of Manager, ECCUTJ), and Marsha Olive (Country Manager, MCH Department, Ministry of Health, Tajikistan), Dr. Z. ECCUTJ). Kasimova (CBHP Project Implementing Unit), and Dr. Lola Rajabova, (Medstat), officials and staff of Goskomstat and the Ministry of Justice Civil Registration Offices. Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK v 66094_World_Bank_Txt.indd 5 9/30/11 6:22:06 AM Acronyms, Abbreviations and Tajik Terms ADB Asian Development Bank MOH Ministry of Health CBHP Community and Basic Health Project MOH Ministry of Justice CMSI Center for Medical Statistics and Information MMR Maternal Mortality Rate CNHS Comprehensive National Health Strategy Oblast A province DHIS2 District Health Information Software 2 PHC Primary Health Care EC European Commission Rayon A rural district ECA Europe and Central Asia RCMSI Republican Center for Medical Statistics and Information FGD Focus Group Discussion RRS Rayons of Republican Subordination HIS Health Information System TLSS Tajikistan Living Standards Survey HMN Health Metrics Network U5MR Under Five Mortality Rate ILBD International Live Birth Definition UNFPA United Nations Population Fund IMR Infant Mortality Rate UNICEF United Nations Children’s Fund Jamoat A local self-government unit, usually a rural sub district USAID United States Agency for International Development Khukumat An executive body that governs an oblast, rayon, or town WHO World Health Organization MCH Maternal and Child Health ZAGS Zapis Aktov Grazhdanskogo Sostoyaniya (Offices of Civil Registration) MDG Millennium Development Goals MICS Multiple Indicator Cluster Survey vi Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 6 9/30/11 6:22:06 AM Executive Summary 1. The Government of Tajikistan has identified improving 4. The analysis reveals a number of issues that are key con- maternal and child health (MCH) as key priorities in its new straints to the further development of the vital statistics system Health Sector Strategy for 2010-2020. The Government rec- in Tajikistan, particularly in the specific area of registration of ognizes that improving MCH outcomes is critical to achieve births and deaths. Most of these go well beyond the health the Millennium Development Goals (MDG) for maternal sector’s span and call for broader action by the Government and child health over the next four years. The fourth MDG is in order to be effectively and comprehensively addressed. The a two-thirds reduction in child mortality, which requires the most important is the absence of clear leadership and owner- under-five mortality rate to decrease from 79 to 39.3 deaths ship among the Government agencies for the issues related to per 1,000 live births, and for the infant mortality rate (IMR) accurate vital statistics. The lack of a coordinated approach to to decrease from 65 to 29.6 deaths per 1,000 live births in vital statistics improvements between the agencies as well as Tajikistan. in the efforts of the main international development partners supporting the Government in this area also limits the effec- 2. Tajikistan’s new National Health Sector Strategy also identi- tiveness of actions taken. Another key issue is the complex and fies the importance of strengthening health information sys- often time-consuming procedures for births and deaths to be tems to facilitate evidence-based policy decision making. Lack registered in the system, which serves as a disincentive to the of data or poor quality data on births and deaths affects the population. Inconsistencies between the different agencies planning of programs critical to the achievement of improved involved in the collection of data further add to the difficul- child health, such as immunization services. Inadequate data ties. Other problems identified relate to lack of knowledge of may also hinder children’s access to health, education and the personnel responsible for collecting and reporting data on other services, influence the ability to monitor the impact of new reporting definitions as well as of the population on the interventions, and hinder the ability of the MOH to advo- importance of timely registration of births. cate for an intensified focus on issues affecting child health. Besides this, vital statistics records serve a number of impor- 5. In view of this, there are several priority actions that have to tant purposes, including providing legal status and identity to be taken over the short to medium term by the Government individuals; proof of identity, citizenship and age to name a and its key agencies to improve this situation. The most few. Improving data quality therefore is one key area required critical action is to clearly establish one Government agency to strengthen the stewardship function of the MOH and to undertake the overall coordination, responsibility and other Ministries working on issues to improve child and ownership on the issue of vital statistics. The Ministry of maternal health outcomes and development in general. Justice which is responsible for the entire civil registration system including birth and death registration is well placed to 3. Tajikistan’s data on most of the MDG indicators for take on this role. This would also require clarification of the maternal and child health can be improved significantly. The roles of Goskomstat and the RCMSI, including establishment Government’s ability to track its progress and to take action of unified and standardized databases, procedures, informa- to ensure the achievement of its MDGs will be consider- tion sharing and clear delineation of lines of accountability. ably bolstered by access to reliable data on childbirths, child Coordination of efforts among the various donor agencies mortality and others factors that affect these outcomes. To supporting the vital statistics system also needs to be bet- improve data reporting, data collection needs to be consistent ter enhanced and formalized to ensure that use of available in its methods and sources, which is currently not the case. resources is maximized. As a result, health facilities and national agencies’ reports diverge significantly from the results of nationally representa- 6. Furthermore, birth registration should be brought closer tive surveys. With this in mind, the study therefore set out to women, i.e., in maternity hospitals to facilitate registration to identify the main factors affecting two specific areas of where most births occur. This will eliminate a key time barri- Tajikistan’s Health Information System-namely the child birth er. The procedures for registering births need revising urgently and death registration system as well as the possible steps to to make them more friendly and accessible to the population address them. Document reviews, in depth interviews and and to encourage early registration. Included in this revision, focus group discussions with key personnel as well as users of should be efforts to enable parents to register their children’s the civil registration system were used to elicit the informa- births where they live. This should include changing the legal tion sought. frameworks and procedures as necessary. Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 1 66094_World_Bank_Txt.indd 1 9/30/11 6:22:06 AM 6. All agencies responsible for collecting and reporting births monthly within their catchment areas to guarantee health statistics should begin to use the same internationally that all births, particularly home deliveries, are captured by accepted, and WHO recommended, definition of what con- the system. Finally, there should be greater efforts made to stitutes a live birth. The Ministry of Health should continue sensitize the public about the importance of early registration to ensure that all relevant health workers are trained to use of births and deaths, as well as the correct procedures to do so, the WHO definition within the near future. Additionally, including the associated financial costs. community health nurses should be mobilized to report new 2 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 2 9/30/11 6:22:06 AM Chapter 1. Introduction TAjIKISTAN DEMOgRAPHIC reporting, data collection needs to be consistent in its method SITUATION and source, which currently is not the case (see Figures 1 and 1.1 Tajikistan is demographically a young country compared 2). Health facilities and national agencies reports diverge sig- to other countries in the Europe and Central Asia (ECA) nificantly from the results of nationally representative surveys, region. Of its estimated 7 million population, approximately as well as the type of local statistical agency reporting the data. 3 million (42%) are aged 18 and under. The median age is just For example, child health indicators reported by the National 20.7; however, this is projected to rise to 26 years by 2025. Statistical Agency Goskomstat are on average 30 percent Although fertility has declined in recent years, the total fertil- higher than indicators reported by the Republican Centre ity rate remains above three. Tajikistan’s population growth for Medical Statistics and Information (RCMSI). And, child rate is around 20 per 1,000 population per year. Mortality mortality data reported by Goskomstat/ Medstat is almost rates remain relatively low at approximately seven per 1,000. 50 percent lower than determined through the Multiple Life expectancy at birth is estimated at 67 years for the total Indicator Cluster Survey (MICS) or the Tajikistan Living population and is slightly higher for women compared to Standards Survey (TLSS). males. In comparison, Russia currently has a median age of 1.5 There are also differences in statistics produced by the about 37, and is projected to have a median age of about 42 two main household surveys that are regularly fielded in by 2025, when Kyrgyzstan’s median age will be about 31. Tajikistan, the MICS and the TLSS. To improve outcomes, 1.2 Unlike other countries in ECA, the proportion of the better quality survey data on child health and especially for Tajik population of working and reproductive ages is expected the direct and indirect determinants of child mortality is to remain large. For example, the proportion of the popula- necessary. The MICS and TLSS data are not ideal to analyze tion 15-64 was about 57 percent in 2005 and is expected to be the determinants of child mortality. These two surveys present approximately 66 percent in 2030. Therefore, the dependency the following challenges: ratio is projected to remain favorable for economic growth • The MICS collects statistics for children ever born and for decades. children who have died but does not collect birth his- tories for children born; HEALTH STATISTICS IN TAjIKISTAN • While the TLSS collected data on children, it contains 1.3 The Government of Tajikistan has identified improv- very few experiences of child deaths, which constrains ing maternal and child health (MCH) as key priorities analyses on child mortality due to the number of in its recently approved Comprehensive National Health observations/ cases available for analyses. The number Strategy (CNHS) for 2010-2020. The Government recog- of births available for fertility and childhood mortality nizes that improving MCH outcomes is critical to achieve estimation is also quite small. For instance, out of the the Millennium Development Goals (MDG) for maternal 15,000 birth records collected in the 2007 TLSS, there and child health over the next four years. The fourth MDG is were 800 deaths in total; some of these deaths per- a two-thirds reduction in child mortality, which requires the tained to children who died after their fifth birthdays under-five mortality rate to decrease from 79 to 39.3 deaths and so could not be included in analyses. per 1,000 live births, and for the Infant Mortality Rate (IMR) • There is a lack of consistency between the two main to decrease from 65 to 29.6 deaths per 1,000 live births in surveys, which makes them incomparable across time. Tajikistan. The Government also identified strengthening the Questions about child mortality are asked differently health information systems (HIS) as a priority of the CNHS in the two surveys. The MICS relies on children ever to enable evidence-based decision making. born and children that died to indirectly estimate child mortality rates, while the TLSS asks questions on dates 1.4 Tajikistan’s data on most of the MDG indicators for of births and dates of deaths for children who have maternal and child health can be improved significantly. The died. Measures of child mortality produced by these Government’s ability to track its MDG progress and to take different data sources can lead to different conclusions action to ensure their achievement will be considerably bol- on child mortality. stered by access to reliable data on childbirths, child mortality, and others factors that affect these outcomes. To improve data Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 3 66094_World_Bank_Txt.indd 3 9/30/11 6:22:06 AM Figure 1: Tajikistan Infant Mortality Rate Trends 1974-2006 100 WHO WHOSIS: 91 90 MICS 2000: IMR = 89 80 LSMS 1999: 79 WHO WHOSIS: IMR = 75 70 MICS 2005: 65 60 50 40 WHO WHOSIS: 38 30 20 10 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 MOH Estimates HFA WHO WHOSIS MICS 2000-2005 LSMS 1999/2003/2007 Source: Inter-agency Group for Child Mortality Estimation [IGME], (2011) 1 1 IGME. (2011). Child Mortality Estimates. Available: http://www.childmortality.org/cmeMain.html • The TLSS mainly aims to collect variables that mea- consumption, but the MICS has much fewer socioeco- sure economic status and poverty. While the TLSS col- nomic variables. lects information on health care utilization and access, questions about child mortality and their determinants, • isaggregated data for child health by region or socio- D economic characteristics is lacking (or weak) in both such as birth intervals and durations of breastfeeding, surveys. are not to the level as detailed as in typical demograph- ic and health surveys. 1.6 Due to these weaknesses, caution is required when ana- • The other fundamental difference between the TLSS lyzing mortality data for Tajikistan, since the magnitude of and the MICS is that, with the TLSS, it is possible to the problem appears very different depending on the source derive measures of poverty from data on income and of infant and child mortality data used. Several explanations 4 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 4 9/30/11 6:22:06 AM exist for the wide gap between rates produced by different before 28 weeks of pregnancy, or those who do not manifest agencies and those produced by the Ministry of Health vital signs other than breath. Another contributing factor is (MOH). The variance may be due partly to different method- the still significant proportion of deliveries that take place at ological approaches. Although the MOH has adopted the use home, which means these births are inconsistently captured of the International Live Birth Definition (ILBD) since 2008, by the health system. More importantly this variance may also some health facilities (as well as the Goskomstat) still use the arise from the low birth registration rate, particularly for the definition of infant mortality established during the Soviet first six months of a child’s life, which affects the denominator era. Under the Soviet definition, newborns weighing less than used to calculate the official infant mortality rate (Aleshina & 999 grams are not considered live births, nor are those born Redmond, 2003). Figure 2: Tajikistan Under 5 Mortality Rate Trends 1974-2006 2009 2010 ARI, 20% ARI, 18% Other, 36% Other, 35% Pneumonia, 19% Pneumonia, 21% Congenital Congenital Malformations, 7% Acute Diarrhea, 18% Malformations, 7% Acute Diarrhea, 19% Total Cumulative Reported Deaths = 1,535 Total Cumulative Reported Deaths = 1,604 Source: Inter-agency Group for Child Mortality Estimation [IGME], (2011) 2 2 IGME. (2011). Child Mortality Estimates. Available: http://www.childmortality.org/cmeMain.html Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 5 66094_World_Bank_Txt.indd 5 9/30/11 6:22:06 AM Table 1: Children aged 0-59 months that have not been registered SeLecTed reASonS why birTh wAS noT regiSTered Don’t know Costs Didn’t know Unregistered if birth is too much Must travel child must Births unregistered to register too far be registered Residence % Urban 15.1 1.7 40.5 0.4 2.4 Rural 10.5 2.3 42.2 9.6 3.8 Region Dushanbe 17.3 1.2 24.1 1.2 2.4 Khatlon 10.5 3.0 59.3 4.8 6.2 Sogd 6.4 1.3 (18.3) (21.0) (3.4) DRD 19.1 2.2 42.4 2.9 1.2 GBAO 9.2 0.9 (41.0) (14.4) (4.0) Age of Child % 0-11 17.9 2.9 25.3 4.9 4.5 12-23 13.2 2.0 47.7 5.1 3.2 24-35 10.0 1.5 70.7 7.7 4.8 36-47 9.5 2.3 58.6 9.3 0.4 48-59 8.2 2.1 51.0 5.4 2.1 Total 11.7 2.2 41.6 6.2 3.3 Source: Tajikistan 2005 MICS. VITAL STATISTICS: CHILD BIRTH RRS and 21 percent in rural RRS during 1998–2002. In Sogd & DEATH REgISTRATION oblast and RRS, 53 percent of respondents did not consider registration of deaths as important; rather, they considered 1.7 Incomplete registration of births and deaths is one of the it as a significant expense. More recent data shows that the main causes for the different data reported. Research under- proportion of unregistered births is still high, which makes taken by the MOH in 2002/2003 showed that in the period tracking childhood mortality rates and progress in meeting between 1998 and 2002, as many as 20 percent of respon- set goals more difficult. dents in the Sogd region had not registered the births of their children. A larger proportion of respondents in the Rayons 1.8 Data from the 2005 MICS also corroborate the MOH of Republican Subordination (RRS), about 33 percent, had study. It showed that as many as 18 percent of children aged not registered their births either. The research also found 0-11 months were unregistered. Like the MOH study, 42 that residents of the RRS generally did not register births percent of respondents who had not registered the births of because some did not consider it important, and a significant their children said that the registration costs were too high, proportion of respondents (68.1%) lacked the money to do and 6 percent mentioned that distance was a factor. The pro- so. The research further found that failure to record births portion of respondents who had not registered their children’s and deaths can lead to large distortions in the rates calculated. births was higher among respondents who had no education For example, the study found that in Khatlon oblast and and slightly higher in rural areas. Registration was lowest Dushanbe, only 10.4 percent of deaths were reported in 2001. in Dushanbe and varied by the mother’s level of education, The share was only 44 percent in Sogd, 25 percent in urban almost 21 percent of children 0-59 months were unregistered 6 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 6 9/30/11 6:22:06 AM and about 7 percent for births to women with secondary or intensifying and increasing resources in support of MCH. higher levels of education. Although crucial for health, poor quality vital registration sys- tems have impacts on government and development activities 1.9 Civil registration records serve a number of important well beyond the health sector. purposes, including providing legal status and identity to individuals; proof of existence and identity 3, citizenship, age 1.10 A better understanding of the factors affecting the col- and establishing rights based on age (e.g. school enrollment, lection of accurate childbirth and death registration data pension etc); establishing family relationships and inheritance will provide the Government and its development partners rights; proof of marriage, divorce and the right to marry, as with the needed evidence to improve the situation. More well as evidence of death. Of equal importance is the fact that importantly, accurate data will complement the Government’s this data forms the basis of national vital statistics systems, ongoing efforts to strengthen the country’s health information which in turn are an integral part of country health infor- systems and improve the collection and quality of health, mation systems.4 The lack of data or poor quality data on population and especially MDG indicators in Tajikistan, as childbirths and deaths affects the planning of programs criti- well as the overall national vital statistics system. cal to improving child health such as immunization services, 1.11 The main objective of this study therefore was to exam- and it may also hinder children’s access to health, education ine the childbirth and death registration aspects of Tajikistan’s and other services. Furthermore, insufficient data influences Health Information System to identify problem areas and the MOH’s ability to monitor the impact of interventions propose possible ways to improve the systems reporting on child health status and impedes its advocacy in favor of accuracy. This was done through, document reviews, in depth interviews and focus group discussions with key personnel of 3 Plan International (2009). Count every child: the right to birth the Ministry of Justice, Goskomstat, the Ministry of Health, registration. as well as users of the civil registration system in Dushanbe, 4 Danel & Bortman (2008). An assessment of LAC’s vital statistics Khatlon, Sogd and the RRS. Further details of the methodol- system: the foundation of maternal and infant mortality monitoring. ogy used by the team are provided in Annex 1. Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 7 66094_World_Bank_Txt.indd 7 9/30/11 6:22:06 AM Chapter 2. Main findings HEALTH INfORMATION SySTEM epidemiological surveillance data (communicable and IN TAjIKISTAN – STRUCTURE AND non-communicable diseases). Goskomstat publishes this information annually. REPORTINg State Committee of Statistics (Goskomstat) Office of Civil Registration - Zapis Aktov 2.1 The Health Information System of Tajikistan was estab- Grazhdanskogo Sostoyaniya (ZAGS) lished during the Soviet period, and involves several minis- 2.4 The activities of the Offices of Civil Registration (ZAGS) tries and Government institutions. According to the 2003 are supported by the Law of the Republic of Tajikistan, par- amendment to the Law on Statistics of the Republic of ticularly based on the legal code: “On State of Registration of Tajikistan (May 1997)5, Goskomstat (The State Committee Acts of Civil Status� and the Family Code of the Republic of of Statistics) is the main body responsible for the production Tajikistan. ZAGS is an independent entity within the Ministry and dissemination of statistical information. The law also of Justice. It is responsible for collecting information concern- describes in detail the organizational structure and responsi- ing births, deaths, marriages and divorces, restoration of age, bilities of Goskomstat, which administratively is independent adoption, changing family etc. ZAGS organizational structure and whose head reports to the President and Government. is similar to that of the Goskomstat. ZAGS collects informa- Goskomstat is responsible for the population census, house- tion at the local level through its rayon and city branches and hold surveys, demographic statistics and a wide range of Jamoats (local government authorities) who then send it to economic information. The law also requires Goskomstat to the Oblast ZAGS. The information is checked (comparing use recommended international methodologies to produce data on paper with registration books), aggregated and sent to objective and reliable statistics. the central office in Dushanbe. Procedures at the central level are similar, i.e. checking and aggregating information. ZAGS 2.2 The architecture of Goskomstat includes the Head Office then provides monthly reports to Goskomstat. in Dushanbe, the main computer center, and local offices. The rayon (district) statistical offices collect primary data from reporting units, aggregate it and provide the aggregated Republican Center for Medical Statistics data to Oblast Statistical Offices. At the oblast (province) and Information level, further aggregations are performed and the data is sent 2.5 The Republican Center for Medical Statistics and to the Central Office. The Oblast Statistical Offices also pro- Information (RCMSI) is an institution under the Ministry vide summaries and analyses to the regional governments. of Health. The RCMSI is in charge of providing health sta- tistics. According to its charter, the objective of the RCMSI 2.3 Goskomstat obtains health related statistical information is to define a unified health information system by gathering, from several Ministries: processing, storing and transmitting information to provide a (i) The Ministry of Justice’s Office of Civil Registration dynamic assessment of the health status of the population. (ZAGS) provides vital statistics (number of births and deaths), as well as the number of marriages and divorces 2.6 The main activities of the RCMSI are to: (i) coordinate the (all these numbers represent cases which are registered health statistics services across the Republic; (ii) provide meth- at the ZAGS).6 odological and organizational guidance (including reporting forms); (iii) provide national health statistics reports; and (iv) (ii) The Ministry of Health’s Republican Center for select and support a set of indicators to monitor health. Medical Statistics and Information provides annual information concerning number of doctors by special- 2.7 The RCMSI has a vertical structure similar to the other ties, other medical personnel, number of beds in the two agencies. Health statistics are collected from all medical hospital sector, utilization data (number of patients), facilities at the rayon, or local level, aggregated and then sent to the upper or oblast level. This data includes child births and deaths occurring in the health facilities and their catch- 5 A new Law on Statistics became effective in January 2010. ment areas. The Oblast RCMSI collects the information, 6 Goskomstat also receives information from other Ministries and aggregates it and sends it to the Dushanbe central office. Departments; however, this information is not crucial for the purpose of There are five main forms used in the health system for births this report, and is therefore not described. 8 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 8 9/30/11 6:22:06 AM and death. These were reviewed as part of this study, and their 2.13 The law also defines procedures to register stillbirths main characteristics are summarized in Annex 2. and perinatal deaths . To register a perinatal death, a medical certificate of death should be presented. In case of a stillbirth, a medical certificate is not issued. Also, ZAGS does not issue a TAjIKISTAN LAWS AND birth certificate for a stillborn child, but can issue a document REgULATIONS ON VITAL confirming the stillbirth case, if requested by at least one of STATISTICS parents. In case of perinatal death, both the birth and death 2.8 The regulations and procedures concerning vital statistics are registered . Perinatal death is registered based on medical in Tajikistan are included mainly in the law “On State of certificates of birth and death issued by a medical facility Registration of Acts of Civil Status.� where the event occurred or by a doctor who witnessed the fact (including private practitioners). In the case of perina- a. Law On State of Registration of Acts of tal death, ZAGS issues only a death certificate . Upon the Civil Status parents’ requests, ZAGS can issue a document confirming registration of birth, but not a birth certificate. 2.9 The team carefully reviewed the clauses concerning birth and death registration from the law “On State of Registration 2.14 The responsibility to declare a stillbirth and/or perina- of Acts of Civil Status� for this study. The law’s main aspects tal death lies with (i) the management of the health facility are outlined below. where the death occurred, and (b) the management of the health facility/medical personnel who witnessed the stillbirth Birth Registration or death of a child, or the attending medical practitioner in Chapter 2 of the law describes procedures related to birth the case of a home delivery. Stillbirth or perinatal death has to registration as follows: be declared no later than three days after its occurrence. The registration of death is free of charge. 2.10 To register a birth at ZAGS (or with the Jamoats in the 2.15 The law also defines procedures for registration of birth case of small villages) the parents must present the Medical when a child is aged one or more. If parents and/or caregiver Certificate of Birth, issued by the medical facility where the have a medical certificate of birth, the child can be registered. delivery took place. In the case of home deliveries and/or The parents have to present a document from their local delivery outside of a medical facility, the medical certificate authorities stating that ZAGS or the Jamoat had not previ- can or should be issued by the doctor or nurse/midwife who ously issued a birth certificate. Late registration can only be attended the delivery. If medical personnel did not attend the done at the city or rayon ZAGS offices . delivery, then registration can be done based on a statement given by the person present at the delivery. If a birth fits none 2.16 If the parents cannot present a medical certificate of of these stipulations, then registration is determined by a birth, the registration will be done based on a court decision. Court decision, which must confirm that a child was delivered In cases of late registration, rules concerning the parents’ mar- by the identified woman. riage certificate and all other procedures are still the same. 2.11 ZAGS is responsible for registering births at the place of 2.17 A birth certificate will be issued only when the registra- residence of one of the child’s parents. In exceptional cases, tion process at ZAGS is completed. The parents’ request, defined by the law, birth registration can be done according to together with the birth certificate, will enable them to get the actual place of birth and not that of the parents’ residence. the document, which allows them to receive the childbirth One or both parents have to declare the birth of the child in allowance. Currently the fee for the birth certificate itself is writing at ZAGS. For the registration process, the parent(s) 4.1 Tajik Somoni (equivalent to USD 0.90), but the overall must present the Medical Certificate of Birth to ZAGS no fees for the process of registering a birth is approximately 18 later than three months after the birth . However, according Tajik Somoni (equivalent USD 3.90). to the law, ZAGS may not refuse to register a birth due to delayed registration . Death Registration 2.12 The marriage certificate of the parents must also be 2.18 Chapter 3 of the law describes procedures concerning presented for birth registration. In case the marriage is not death registration. This chapter provides general procedures registered, the father has to confirm his paternity. If this is for deaths of all ages and not only children. not possible, then the newborn will be registered with the The basis for official registration of death can only be: (i) a mother’s family name. medical certificate of death issued by a medical facility or a Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 9 66094_World_Bank_Txt.indd 9 9/30/11 6:22:06 AM private medical practitioner; (ii) a court decision confirming Civil Registration- ZAGS (Clause 159, p.2). The registration the death case or declaring the person as dead according to of birth and death in the cities and rayon centers is done the law. by the offices of civil registration, city or rayon khukumats. While in small villages, this procedure is the responsibility 2.19 Death registration is done by ZAGS according to (i) of the Jamoats. Outside the Republic, this registration takes the person’s last place of residence; (ii) place where the death place at consulate offices. occurred; (iii) place where the body was found; or (iv) the location of the court which has confirmed the death. EffORTS TO IMPROVE TAjIKISTAN’S 2.20 Notifying ZAGS about a death case is compulsory. HEALTH INfORMATION SySTEM A death must be reported by: (i) the spouse or other fam- 2.22 Over the past decade, several projects have been imple- ily members, or any person who has witnessed the death or mented to support improvements to the Health Information learned about the case; (ii) the medical facility or social care System (HIS) of Tajikistan. Different donor agencies have facility if the death occurred there; and (iii) army, prison worked with different players in the HIS on various aspects management, or investigating agency, in case the death occurs of reform. However, the efforts have not always been well in any of those facilities. A statement about the death should coordinated with different levels of government. Some efforts be made no later than three days after death. According to were with central agencies and others directly with local health existing legislation, only in the event of the death of an adult, facilities. In many ways this affects the pace of further refining a one-time death compensation of 200 Tajik Somoni (equiva- the system as there is no single comprehensive and unified lent to USD 43) is paid to the family of the deceased. approach to addressing the deficiencies. Details on the main 2.21 According to the Family Code of the Republic of projects that have been undertaken are provided in Annex 3. Tajikistan, birth and death are registered at the Offices of 10 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 10 9/30/11 6:22:07 AM Chapter 3. factors Affecting Vital Statistics in Tajikistan A. System sharing is diverse across the system. In addition, the manual data management, and the absence of a unified database, also 3.1 The civil registration and health systems each have their contributes to data inaccuracy. strengths. In the case of birth data collection, the civil registra- tion system registers more births than the health system. On 3.5 Lack of an established format for data exchange. The the other hand, the health system registers more child death format, frequency, and timing for data exchange with health cases than the civil registration system. Goskomstat presents facilities is undefined, and no measures exist to correct data birth numbers based on issued birth certificates and makes a discrepancies. Rural health facilities usually are well informed 30 percent upward adjustment to reflect the birth cases that about birth cases within their catchment areas. Children the civil registration office does not capture. delivered at home therefore are issued a medical certificate, 3.2 Demographic 7 (Goskomstat produced) and medical sta- which is obligatory for issuing the birth certificate. Rural tistics 8 (from RCSMI) yearbooks were compared for infant health facilities routinely share information on births with mortality data for 2008. Medical statistics presented an infant the local civil registration offices (Jamoats). However, chil- mortality rate of 15.2 per 100,000 live births, which trans- dren from home deliveries, which are unknown to the health lated into absolute numbers is 3,090. This is 600 more infant system, e.g. migrated households or pregnant women not yet deaths than the 2,480 deaths reported by Goskomstat. About captured by the health facility, or children born in remote one-fifth of cases are not included in the civil registration sys- villages without permanent health workers, may register later tem. Likewise, in Khatlon Oblast, health facilities registered at the civil registry. Also, perinatal deaths and stillbirths that 1,244 death cases, while ZAGS captured only 1,072 infant occur at home without attendance by health personnel, more deaths. It is clear therefore that there are gaps in each of these often than not, are not captured by the civil registration or systems. Improving data collection and reporting will ensure health system. that the overall national vital statistics are accurate. The main 3.6 Regulations require that child deaths occurring more than issues identified in the civil registration and health systems are six days previously be registered by place of residence. Health summarized in the following section. authorities are required to notify the corresponding (child domicile rayon) civil registration office about the child death CIVIL REgISTRATION SySTEM case. However, the lack of regulated information sharing and different practices in various rayons leads to under-registration 3.3 Registration Procedures. The current system does not of such cases. allow births to be registered at the place of birth and then to be linked to the parent’s place of residence, which is essential 3.7 Lack of resources. One of the main obstacles to improv- for proper population estimates and projections by admin- ing the civil registration system is the inadequate level of istrative units. Currently, births are registered only at the consistent financial resources available to the various levels of place of residence of one of the child’s parents. If the place of the system. This deficiency is reflected in the shortage of moti- delivery and the parents registered domicile differ, obtaining vated and technically qualified human resources at different a birth certificate in a short period is difficult for the parents. levels of the system, and the absence of modern technologies, Therefore, due to the time and cost involved, the birth may methodologies, and quality control systems. Internationally end up unregistered. funded projects have focused on limited parts of the system and not its integrity. 3.4 Lack of a unified data exchange system between dif- ferent oblast health administrations and ZAGS. There are 3.8 Other issues. Child death cases are not recorded when no normative documents or procedures to regulate horizontal a child dies without prior birth registration. There is also information sharing between different rayon civil registra- a shortage of human resources in the Jamoats dedicated to tion offices and health authorities. Therefore, information registration of acts of civil status. Also affecting timely birth registration is a lack of strong regulations to ensure that birth certificates for children are issued before discharge from 7 Republic of Tajikistan demographic yearbook, State Committee on Statistics, 2009. maternity hospitals. 8 Health and healthcare in the Republic of Tajikistan in 2008, Center for Medical Statistics and Information, Dushanbe, 2009. Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 11 66094_World_Bank_Txt.indd 11 9/30/11 6:22:07 AM HEALTH SySTEM 3.14 Information transfer in the system, in some cases, is done by telephone. Although the information on births and 3.9 Accurate reporting of deaths. The Ministry of Health child deaths is not officially used for national level statistics, (MoH) is paying greater attention to accurate reporting of this reporting method may result in data distortion and inac- maternal and child deaths in health facilities, through close curate reporting. monitoring of the mortality rates. However, punitive action is still largely used in instances where facilities misreport the 3.15 Insufficient human resources to undertake routine number of deaths that occurred. Districts with high maternal door-to-door visits to record births and deaths within the and child death rates undergo a peer review process to analyze communities is also an issue. In remote locations, the lack of the cases and if the deaths did not occur due to the doctor’s human resources is a problem, and some villages have no per- fault no punitive measures are instituted. manent health workers. The timely registration of birth/death data in remote areas is even more challenging. 3.10 Large volume of paperwork to register/report data for different vertical programs. Many of the respondents believe that the quality of health statistical data may be affected by COMMUNITy/POPULATION the amount of paperwork doctors have to complete. Doctors fACTORS have only 15 minutes for each patient (which has not changed 3.16 There are a number of issues on the population side since the Soviet period), while approximately double that time that affect vital statistics registration. The team identified the is spent on the same patient’s paperwork. Doctors have to following as main factors contributing to childbirth under complete different registration and reporting forms for each registration. system level, especially for the vertical national programs (e.g. immunization, tuberculosis etc). As a result, doctors have very 3.17 Home deliveries. Children born at home are more likely little time to actually do clinical work. Respondents stressed to get a birth certificate late. Overall registration is much bet- the importance of simplifying the forms. ter in Sogd Oblast and Dushanbe City, where the majority of deliveries take place in maternity facilities, compared with 3.11 Lack of material resources allocated for information other districts. In other oblasts, a large proportion of deliver- exchange. The lack of a computerized system is a major ies happen at home, e.g., in Davidarinsky rayon and Rasht problem. Computers are only available from the oblast level Valley districts, 50 percent to 70 percent of deliveries occur upwards, and data is transferred by paper from the lower lev- at home. els. In addition, there are no standardized forms for reporting from a medical facility to the central level. 3.18 Low motivation of community members for timely registration of childbirth. Respondents reported a main 3.12 Lack of established formats for data exchange (for- problem with birth registration is the parents’ lack of incen- mat, frequency, and timing). Because of a lack of consistent tives to register early. In the majority of cases, a birth certifi- formats for data exchange with the civil registration office, cate is obtained when children are taken to kindergarten or there are no defined measures to be taken if data discrepancies school, or the family is migrating from the country. According are identified. Furthermore, perinatal death registration is not to a UNFPA-supported survey on the accuracy of registration performed countrywide according to the WHO standards, records, birth registration dramatically increases from July to even though the international live birth definition (ILBD) August every year. This period directly correlates with school was introduced in the country in 2004. enrolment prior to the start of the school year in September. 3.13 Health personnel knowledge on international live 3.19 Women wait for their migrant husbands to register birth definition. Since the introduction of the ILBD, the their child. Many heads of households (fathers) are work- MoH with the support of its international development part- ing in Russia. Absent the fathers’ consent, a child would ners has held several sessions to train health personnel on it. be registered under the mother’s name. Therefore, mothers However these trainings have been limited to a few districts in usually wait for their husbands to return before registering the country, as such many health personnel are not yet aware their child. of the ILBD and the new procedures of registering perinatal deaths. Furthermore even in those districts where training has 3.20 Women are reluctant to register a child in their name occurred, there is little monitoring being done to ensure the in the case of unregistered marriages. Official marriage reg- use of the new definitions, and this is partly due to resource istration has decreased in the country, as religious marriages constraints. 12 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 12 9/30/11 6:22:07 AM increased. Such marriages, mainly among rural populations, in registration delays. This is particularly the case for rural while accepted by society, remain unrecognized by the formal poor families. Additionally, the population is misinformed civil registration system. Polygamous unions also contribute about the documentation required to register births at the to the number of unregistered marriages. civil registry. 3.21 Poor understanding of the importance of a birth 3.22 Local traditions. Respondents emphasized the impor- certificate. Community members have a poor understand- tance of local birth traditions and perceptions, especially in ing of the purpose and importance of timely registration of rural parts of Tajikistan. For instance, it may take some time birth and/or death. Although the fee for the birth certificate for a child to be named- some parents may not name their itself was not identified as a leading barrier, the additional child until 40 days after birth. fees required for the entire registration process do play a role Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 13 66094_World_Bank_Txt.indd 13 9/30/11 6:22:07 AM Chapter 4. Conclusions SUMMARy Of MAIN CONSTRAINTS 4.8 The above action would also require clarification of the roles of Goskomstat and the RCMSI, including establishment 4.1 This assessment revealed a number of issues that are key of unified and standardized databases, procedures, informa- constraints to the further development of the vital statistics tion sharing and clear delineation of lines of accountability. system in Tajikistan, particularly in the specific area of reg- istration of births and deaths. Most of these go well beyond 4.9 Coordination of efforts among the various donor agencies the health sector’s span and call for broader action by the supporting the vital statistics system also needs to be bet- Government in order to be effectively and comprehensively ter enhanced and formalized to ensure that use of available addressed. resources is maximized. As a starting point the various World Bank units that work on the issue of statistics in Tajikistan 4.2 Firstly, while the three main agencies- Ministry of Justice will plan to have a series of roundtable discussions with key (ZAGS), Goskomstat and Ministry of Health all make efforts Government policymakers and international development to collect this data, it is not clear where the main leadership partners over the next six months to a year on possible joint and ownership for the issues related to accurate vital statistics actions on the vital statistics system, including exploring sup- lies. Consequently this affects the effectiveness of the entire port provided by the Health Metrics Network on improving system and is manifested through the inconsistent procedures national vital statistics systems. and limited information sharing between the agencies at vari- ous levels. 4.10 All procedures introduced to the civil registration and health care systems should be formalized by changes in the 4.3 The lack of a coordinated approach to vital statistics normative and legal documentation, and a standardized improvements between the agencies as well as in the efforts information-sharing format (form, timing, frequency) should of the main international development partners supporting be established between the health system and civil registra- the Government in this area also limits the effectiveness of tion system. Alongside this, greater efforts should be made actions taken. to engage with the community members (through commu- 4.4 Another key issue is the complex and often time-consum- nity leaders and health care workers, mass communication ing procedures for births and deaths to be registered in the campaigns, schools etc) to increase understanding of the system, which serves as a disincentive to the population. importance of timely registration of birth and death cases, as well as to provide the correct information on the process of 4.5 Also of importance is the lack of knowledge on the new registration and the financial cost. reporting definitions among the personnel responsible for collecting and reporting data, as well as low awareness of the Civil Registration System Changes population of the importance of early registering of births. 4.11 It is critical that changes should be made to the civil reg- 4.6 The various issues identified as problems from the system istration system to allow registering of birth and child death as well as population side on the issue of birth and death reg- by place of occurrence of the vital event with an indication of istration, can serve as a roadmap to complement the existing the parent(s) place of residence. efforts by the different Government agencies and other stake- holders to improve statistics in general in Tajikistan. 4.12 Although this will require a large amount of resources, steps should be taken to begin to move toward an electronic information system within the civil registration system to RECOMMENDATIONS allow easier and more reliable transmission of information 4.7 The most critical action is to clearly establish one between the oblast, rayon levels and with the health system Government agency to undertake the overall coordination, and Goskomstat. responsibility and ownership on the issue of vital statistics. This would include the final verification and triangulation of 4.13 Financial and human resources should be strengthened data on births and deaths from Goskomstat and the RCMSI. within the civil registration system. This should include The Ministry of Justice which is responsible for the entire training of civil registry staff on the updated live birth defi- civil registration system including birth and death registration nitions and procedures in a manner consistent with that of would be the ideal agency. Goskomstat and health sector personnel. 14 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 14 9/30/11 6:22:07 AM 4.14 Goskomstat should take into account infant death cases and families on this starting from the antenatal through to provided by the medical statistics system and make relevant the postnatal period. adjustments to their annual infant mortality rate. 4.19 Primary health care nurses who already make home 4.15 The maternal death definition used by Goskomstat visits within their communities can play a great role in should be updated (including on the reporting forms) to be ensuring early birth and death registration, by following up consistent with the recommended WHO definition and that with mothers with newborns. The job descriptions of family used by the health system. and community health nurses should explicitly define this responsibility of confirming childbirths and deaths within Health System Changes their catchment areas. 4.16 Health administrators should place greater emphasis 4.20 Once civil registration offices are allowed to register on the achievement of health results and understanding and birth and death by place of occurrence of the event, health addressing factors within the continuum of health care that administrations/ facilities/workers should inform only local contribute to mortality or poor outcomes, as opposed to puni- civil registration offices. tive measures towards providers. Supportive supervision and not punitive measures should be used to encourage accurate 4.21 Health care facilities should report on birth and death reporting of deaths and also to identify reasons for the death, cases in the system according to the actual place of occurrence and measures to prevent the deaths from occurring in the first of the birth or death. This will avoid missing a case during place. This will help to foster a positive environment where data transmission from one administrative unit to another. head doctors and health personnel focus on providing quality This data should be published as health statistical data. care to the population. 4.23 It is very important that the Ministry of Health expands 4.17 Once civil registration offices are allowed to register birth the training on the international live birth definition to all by place of birth, further changes should be introduced to health personnel in the country. This should be supported enable maternity facilities to actually register births prior to with supervision and monitoring to make sure that providers hospital discharge. report births and stillbirths according to this definition. 4.18 Health personnel should be sensitized on the importance of early birth registration, and they should educate mothers Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 15 66094_World_Bank_Txt.indd 15 9/30/11 6:22:07 AM ANNEx 1: STUDy OBjECTIVE AND METHODOLOgy Objective urban areas of the following rayons: Tursunzade, Varzob, The objective of the study was to identify deficiencies in spe- (RRP); Qabodyon, Nurek, Kurgan-Tyube city (Khatlon); cific areas of Tajikistan’s Health Information System, as well Isfara (Sogd); Dushanbe city. as recommend steps to address them. Improving the HIS will contribute to building Tajikistan’s capacity to collect and use Study Limitations reliable and timely health indicators. The study has a number of limitations: 1. Almost all the respondents refused to be tape-recorded, Methodology thus detailed notes had to be handwritten during the The study followed the methodology outlined in the WHO’s interviews. Health Metrics Network (HMN) Framework9 (2008), which 2. In the majority of the communities, the research facilitates the procedures for analysis of subsystems in a health team was accompanied by a representative of the local information system. Specifically, two out of the six areas of a government or health authority. In almost all the com- national HIS (civil registration and population surveys) were munities, the women were informed for the reason of assessed as follows: the visit, and they did not feel comfortable to speak • Assessment of the content of the civil registration sys- candidly in the presence of the local authorities (who tem and population surveys; also offered translation from Russian to Tajik and vice • Assessment of capacity and practices in collecting data versa in some instances). However, to minimize bias for these two components of the system; and ensure the presence of a neutral person during the FGD with the community, the research team had an • Assessment of dissemination of data from these two independent interpreter. components of the HIS; and 3. The research team experienced difficulties obtaining • Assessment of integration and use of data from the two some data, concerning childbirth and death, even components of the HIS. though those numbers were officially published later. In addition to the assessment of the data sources, qualitative data on the major factors affecting collection of accurate health data and particularly on childbirths and deaths was also collected. This was done through a series of: 1) In-depth interviews with key informants such as MOH officials, health administration staff, donor organiza- tion representatives, staff responsible for demographic data management at different administrative levels and health care providers. 2) Focus group discussions (FGD) with health care pro- viders and community members. In-depth interviews and FGDs with health care providers, rural/urban civil registration officials were conducted in seven districts in three regions; Khatlon, RRS, and Sogd. The FGD’s with community members (women of reproductive age, men and elderly women) were conducted in rural and 9 The Health Metrics Network (HMN) was launched in 2005 to help countries and other partners improve global health by strengthening the systems that generate health-related information for evidence-based deci- sion-making. 16 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 16 9/30/11 6:22:07 AM ANNEx 2: HEALTH STATISTICAL REPORTINg fORMS fOR VITAL REgISTRATION IN TAjIKISTAN a. Form # 14 10 on Hospital Performance and presented to the local health authorities every January 5. The local health authorities aggregate the information from This is an annual reporting form; approved by Goskomstat of all the rural Health Houses and send the aggregated report the Republic of Tajikistan by Decree #2 on February 14, 2006. (svodni) to the Republican Center for Medical Statistics, by This form is completed by hospitals and presented to the local the date defined by the RCMSI. Information concerning the health authorities every January 5. The health authorities number of children under one, and pregnant women who aggregate information from all hospital facilities and send the were registered is entered on this form. Also, information aggregated report (svodni) to the RMCI by the date defined about death cases, including maternal deaths and children by RCMSI. The aggregated report from the RCMSI is then under one, as well as number of death certificates issued is presented to the Goskomstat on March 20 annually. documented. Information concerning home deliveries and Among the information entered in the form is the special outcomes of those deliveries both for mother and newborn section devoted to the number of newborns (0- 6 days) is also included. transferred from other facilities for further treatment, and the treatment outcome. Information is given by diagnoses c. Form #3213 on Health Services to Pregnant (International Classification of Diseases [ICD-10]) and birth Women, Women in Labor and Delivery weight (less than and more than 1000 grams). At the end of This annual reporting form was approved by Goskomstat each form, there are detailed instructions on how to complete of the Republic of Tajikistan by Decree #2 on February 14, the tables and how to do the calculations. Notably, instruc- 2006. This form is completed by all medical facilities (all tions for the aggregated table on death cases stipulate that specialties) that provide services to adults and children. The newborn death cases do not include those with birth weight completed form is presented to local health authorities (city, less than 1000 grams and less than 28 weeks of gestation rayon) and Oblast CMSI (Center for Medical Statistics & if the death occurred within the first 6 days of life. This Information) on January 5 annually. The local health authori- instruction reflects the old live birth definition prior to intro- ties (city, rayon) and Oblast CMSI present the aggregated ducing the WHO international live birth definition. information to the RCMSI on a date defined by the Center, According to the WHO (2011),11 a live birth refers to the and the RCMSI present the final summary to the Goskomstat complete expulsion or extraction from its mother of a product on March 20 annually. of conception, irrespective of the duration of the pregnancy, Section 5 of this form contains information on live births, which, after such separation, breathes or shows any other stillbirths and perinatal death cases by birth weight from 500 evidence of life, e.g. beating of the heart, pulsation of the gram and upwards. This section also provides information on umbilical cord or definite movement of voluntary muscles, the cause of death of live-born and stillborn children with whether or not the umbilical cord has been cut or the pla- birth weight of 500-999 grams. A separate Table 6 contains centa is attached. Each product of such a birth is considered information about maternal deaths.14 Maternal death15 live born. b. Form #16 12 on Rural Health House Reporting 13 Ministry of Health Republic of Tajikistan, Republican Center for Medical Statistics and Information, Project Reforming Health Care System This is another annual reporting form, approved by the in Tajikistan (Asian Development Bank, Credit ADB# 2054 Taj-SF). Ministry of Health of Tajikistan by Decree # 342 on Catalog of the Primary Medical Documentation/Forms. Dushanbe 2009. December 7, 2000. This form is completed by Health Houses 14 At the end of the statistical form #32, maternal death is defined as death caused by pregnancy, during pregnancy or during 42 days after delivery, with an exception of cases caused by HIV and Maternal Tetanus. 10 Ministry of Health Republic of Tajikistan, Republican Center for This is important because RCMSI defines maternal death differently; it Medical Statistics and Information, Project Reforming Health Care System includes only death directly caused by pregnancy and delivery. in Tajikistan (Asian Development Bank, Credit ADB# 2054 Taj-SF). 15 Maternal death is defined as the death of a woman while pregnant or Catalog of the Primary Medical Documentation/Forms. Dushanbe 2009. within 42 days of termination of pregnancy, irrespective of the duration 11 WHO (2011). Health Statistics and Health Information Systems. and site of the pregnancy, from any cause related to or aggravated by the Available online: http://www.who.int/healthinfo/statistics/indneonatalmor- pregnancy or its management but not from accidental or incidental causes tality/en/ (Accessed June 6, 2011). (WHO, 2004). 12 Ibid. Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 17 66094_World_Bank_Txt.indd 17 9/30/11 6:22:07 AM includes all pregnancy and delivery complications,16 which of Health of the Republic of Tajikistan has approved several might cause the death. forms of death certificates. (i) Medical Certificate of Death issued by Doctors (Form# d. Medical Certificate of Birth 106/y approved by MoH decree #98, on March 27th, (Form #103/y-06) 17 2006); This form was approved by the Ministry of Health of the (ii) Medical Certificate of Death issued by Feldshers Republic of Tajikistan by Decree #98 on March 27, 2006. (Form# 106-1/y approved by MoH decree #98, on This form has two sections: one is left at the facility issuing March 27th, 2006); the certificate; another has to be presented to ZAGS for regis- tration. The Medical Certificate of Birth is issued by a doctor (iii) Medical Certificate of Perinatal Death (Form# 106- or other medical staff of the facility where the delivery took 2/y approved by MOH decree #98, on March 27th, place. The certificate is issued when the mother is discharged 2006); from the facility. It has to be issued by all medical facilities, even if they do not have obstetric wards. In facilities, where All the statistical forms have detailed instructions on how doctors are not present (e.g., Health Houses), the certificate is to complete them, how to make the calculations as well as issued by nurses/midwives/feldshers. references to the appropriate clauses and chapters of The Family Code and Law On State of Registration of Acts of e. Medical Certificate of Death (Form #106/y) 18 Civil Status. It has to be emphasized, that the procedures for issuance of each of the certificates described above are in full To ensure the registration of deaths at ZAGS, the Ministry compliance with the corresponding laws and regulations of Tajikistan. 16 Pregnancy related death is defined as the death of a woman while pregnant or within 42 days of termination of pregnancy irrespective of the cause (WHO, 2004). 17 Ministry of Health Republic of Tajikistan, Republican Center for Medical Statistics and Information, Project Reforming Health Care System in Tajikistan (Asian Development Bank, Credit ADB# 2054 Taj-SF). Catalog of the Primary Medical Documentation/Forms. Dushanbe 2009. 18 Ibid 18 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 18 9/30/11 6:22:07 AM ANNEx 3: SUMMARy Of PROjECTS TO IMPROVE HIS IN TAjIKISTAN 1) The Asian Development Bank (ADB) supported the General Vital Registration Offices (GVRO) in terms of Health Sector Reform Project (2004-2008), which expertise and equipment. The following activities were developed the Strategy of Health Management conducted: (i) a number of on-site seminars with the Information System for 2006-2015. The strategy plans involvement of the GVRO staff to provide assistance to to reinforce human and technical capabilities of the local level offices were conducted to strengthen capac- RCMSI, and computerization of the system is seen as ity of VROs and Jamoat staff; (ii) 68 Vital Registration a main way to improve the quality of data collected. Offices (VRO), the GVRO and Jamoats were equipped The ADB-funded project also supported the adaptation with computers, copying machines, and typewriters; and piloting of District Health Information Software (iii) a 2-day seminar with involvement of the bod- 2 (DHIS2), which is an open source, flexible software ies dealing with statistical data (Ministry of Justice, used in a number of Asian countries. Ministry of Health and GKS) to identify single policy to address current issues to allow for producing reliable 2) The USAID funded ZdravPlus project (2004-2009), statistical data was conducted. in preparation for the implementation of a case-based hospital payment system, worked with the MOH to 5) The European Commission has supported the develop the Form 66 Hospital Computer program, Assessment of the Health Information System in which improves routine health statistics. Tajikistan, by using the Health Metrics Network (HMN) methodology. The second and third phas- 3) The World Bank funded Community and Basic Health es (2010-2013) of the EC supported project will Project includes an HMIS component (2005-2010). be focused on supporting Goskomstat, RCMSI, the During this project, four software applications were Sanitary & Epidemiological Services (SES) and ZAGS. developed to support financing reforms. The software The project will cover all the main players in HIS in was developed to assist in more effective management Tajikistan. The project will include developing a road and improvement of quality of services. map and strategy following the HMN s 4) The World Bank funded TAJSTAT project included framework and standards, a plan to implement the a sub-component to improve the quality of vital reg- strategy and costing it, and modernizing the civil regis- istration and statistics, especially at the municipality tration system. level. This sub-component provided support to the Human Development Sector Unit | Central Asia Country Unit | Europe and Central Asia Region THE WORLD BANK 19 66094_World_Bank_Txt.indd 19 9/30/11 6:22:07 AM References 1 Aleshina, N., & Redmond, G. (2003). How high is 5 Plan International (2009). Count every child: the right infant mortality in Central & Eastern Europe & and to birth registration. Available online: http://plan-inter- the CIS? Florence: UNICEF Innocenti Research Centre. national.org/birthregistration/files/count-every-child- 2009 (Accessed June 25, 2011). 2 Danel & Bortman (2008). An assessment of LAC’s vital statistics system: the foundation of maternal and infant 6 State Committee on Statistics (2009). Republic of mortality monitoring. Tajikistan demographic yearbook. 3 Ministry of Health Republic of Tajikistan, Republican 7 Center for Medical Statistics and Information (2009). Center for Medical Statistics and Information, Project Health and healthcare in the Republic of Tajikistan in Reforming Health Care System in Tajikistan (Asian 2008. Development Bank, Credit ADB# 2054 – Taj-SF). 8 UNICEF (2011). State of the World’s Children Report Catalog of the Primary Medical Documentation/Forms. 2011. Dushanbe 2009. 9 WHO Health Metrics Network (2008). Framework and 4 Ministry of Health Republic of Tajikistan, Republican standards for country health information systems. Center for Medical Statistics and Information, Project Reforming Health Care System in Tajikistan (Asian 10 WHO (2011). Health Statistics and Health Information Development Bank, Credit ADB# 2054 – Taj-SF). Systems. Available online: http://www.who.int/health- Catalog of the Primary Medical Documentation/Forms. info/statistics/indneonatalmortality/en/ (Accessed June Dushanbe 2009. 6, 2011). 20 Improving Statistics for Children’s Births and Deaths 66094_World_Bank_Txt.indd 20 9/30/11 6:22:07 AM 66094_World_Bank_Txt.indd 21 9/30/11 6:22:07 AM 66094_World_Bank_Txt.indd 22 9/30/11 6:22:07 AM