Bloomberg Philanthropies Data For Health Initiative Associate Professor Deirdre McLaughlin Associate Professor Deirdre McLaughlin Associate Professor Deirdre McLaughlin Global CRVS LANDSCAPE CRVS is fundamental ".. Without these fundamental health data, that is, counting births and deaths and [accurately]
recording the cause of death, we are working in the dark. We may also be shooting in the dark. Without these data, we have no reliable way of knowing whether interventions are working, and whether development aid is producing the desired health outcomes." Dr Margaret Chan, WHO Director General, Sept 2007 Whos counting? 1 in 3 births are not
registered 1 in 2 deaths are unregistered Global death registration has increased from 28% in 1970 to 45% in 2013, 17% in 43 years! Common challenges: registration Births Lack of registration facilities and/or poor
distribution across communities Lack of knowledge about importance and benefits Disincentives to register, including fees Deaths Disinterest due to perceived absence of individual/societal benefits Lack of registration facilities and/ or poor distribution across
communities Poor coordination between health and registration/statistical sectors Lack of strategic thinking and use of IT advances for deaths in communities Common challenges: cause of death Poor medical records certification by doctors in health facilities Lack of awareness about importance of correct CoD certification
Insufficient training of medical students in CoD certification Lack of trained coders and inadequate use of automated coding Use of lay personnel to determine CoD for home deaths Insufficient capacity to critically appraise CoD data and to analyze and interpret data in policy-relevant ways Addressable challenges Incomplete registration of deaths (less so for births) Poor certification of the cause of death Poor availability/timeliness of data
Poor use of data Poor understanding of the true policy value of reliable birth, death and CoD data Enormous policy value of GOOD cause of death data from CRVS TIME: Track emergence/re-emergence/decline of epidemics in the population, e.g. HIV
TB Lung cancer Coronary heart disease/stroke Traffic fatalities SPACE: Geographical breakdown helps isolate populations most in need of services or to assess comparative performance of health systems (e.g. Hospitals) RESEARCH: Guide epidemiological studies into the causes of disease/injuries in the population (e.g. tobacco and lung cancer) 5000 3750 2500 1250
0 19001910192019311941195119611971198219922002 Year Lung cancer death rate per 100,000 Number of cigarettes per capita Tobacco control in America: lung cancer rates mirror cigarette consumption The Bloomberg Philanthropies Data for Health
Initiave Rationale Without data, governments, donors, and NGOs essentially have to guess how to best target their resources to prevent deaths and diseases. Better data allows for better targeting of resources, which can save lives. Michael Bloomberg / Julie Bishop (Australian Foreign Minister) Objectives: -
Strengthen birth and death registration systems (CRVS) - Help policy makers better use data to make better decisions - Improve risk factor surveillance surveys with accessible mobile phone technology - Budget: US$100 Million for first 4 years (3/2015- 3/2019). Jointly funded by Bloomberg Philanthropies and the Australian Department of Foreign Affairs and Trade (DFAT). CRVS D4H Initiative To help countries improve public health and save lives, the initiative will build and strengthen civil registration and vital statistics (CRVS) systems: This includes a) strengthening systems that generate
data; b) adopting innovative practices to fill data gaps cost-effectively; and c) building country analytical capacity Program partners in the CRVS Initiative include: The University of Melbourne (overall technical direction) Vital Strategies (country implementation) CDC Foundation Overview of the CRVS D4H Initiative
20 countries and cities worldwide. Brazil, Peru, Colombia, Ecuador, Solomon Islands, PNG, Indonesia, Philippines, Myanmar, Sri Lanka, Bangladesh, Ghana, Turkey, Morocco, Tanzania, Malawi, Rwanda, Zambia; cities - Mumbai, Shanghai Main area of focus is strengthening CRVS systems in low to middle income countries, particularly causes of death. GBD Group at MSPGH are the overall technical lead for CRVS component, with Vital Strategies and US CDC (NCHS) key collaborating partners. Primarily a technical assistance program with multiple partner organizations providing technical assistance, access to networks and additional resources. Catalytic, rather than financial focus. Main areas of activity
Direct technical assistance to improve mortality and cause of death data for policy Translating research findings on population health measurement (e.g verbal autopsy methods) into country CRVS systems using IT advances Capacity building in CRVS data analysis/system strengthening through UoM Fellowships/courses Knowledge generation and management through University of Melbourne products: CRVS Knowledge Gateway, Courses, CRVS Development series, RoadMaps for Action, Technical series, etc.
What is CRVS? Civil Registration: continuous recording and registration of all births, deaths and causes of death, including key socio-demographic information on all vital events Vital Statistics: statistical compilation of data on births and deaths in a population derived from a Civil Registration system and used to calculate trends, patterns and differentials in birth & death rates and causes of death Monitoring: 10 CRVS
milestones Notification of VE Validation of VE & Quality Registration of VE Certification of VE Share VE information Storage and archiving of VE
Civil Registra tion Compilation of VS Quality control of VS Vital Generation of VS Statist ics Dissemination of VS
What do we need vital statistics for? Develop key health indicators nationally and for small areas (fertility, mortality, disease specific rates) Generate reliable and timely information for policy-making nationally and sub-nationally, particularly on burden of disease Help guide efficient resource allocation in the health sector Identify regional health inequalities Provide information on patterns and trends for leading diseases of public health concern Identify emerging health problems Monitor and evaluate the effectiveness of health programs
and policies Why Civil Registration systems are the preferred source of vital statistics Only source that provides data for small areas Only source that provides detailed cause of death information to inform policy Only source that produces annual data Only source that provides policy relevant characteristics of both the events and the persons involved
In addition, a Civil Registration system Provides individuals with legal documents of identity, proof of age, civil status, and death Provides government with key health intelligence and administrative data for economic and social development Promotes the development process in societies no developed country today could function without some form of civil registration. Policy requires accurate and timely cause of death data: just how bad are they?
Classification of global CRVS systems based on the VSPI ( Vital Statistics Performance Index) Figure 1: Typology of CRVS systems based on the VSPI scores, best possible available year, 2005 -12 Fraction of deaths assigned to garbage codes, latest ICD-10 year since 2000 Main challenges with vital registration data Incomplete registration of deaths (less so for births) Poor certification of the cause of death Poor availability/timeliness of data Poor use of data
Poor understanding of the true value of reliable birth, death and CoD data Vast and unacceptable uncertainty in rankings of leading causes of death and how they are changing Improving data for health (policy) CRVS systems characterized by decades of neglect ; single most important failure of development in last 30 years Richard Horton, Lancet, 2007 CRVS Series Bloomberg (as Mayor of NYC) used CoD data extensively to justify, agitate for and monitor public health programs for NY
Willing to invest to support similar data environment (and use) elsewhere. Julie Bishop agreed to partner. Hence D4H Initiative. Huge need, recognizing strength and role of universities as development partners. Key challenges in improving CRVS systems and capacity in LMICs. What are we doing? Assess and plan: Develop (SwissTPH) and apply BPM/EA methods to understand and develop strategies to correct system inefficiencies Increase registration: Increase the registration of births and deaths using knowledge about incentives and advances in IT/systems approach/methodological innovations (e.g. SMS)
Conduct certification training: Improve the quality of cause of death assignment in hospitals through targeted training of doctors and medical students to correctly certify deaths cont.. Introduce coding software: Introduction of automated CoD coding software (IRIS) Increase use of verbal autopsy: Incorporate automated verbal autopsies into routine CR systems to rapidly improve information on causes of death for home deaths Expand sample sites: Support transition of HDSS & SRS into functioning CRVS nodes
Create and apply data quality assessment tools: Build capacity and tools for critical appraisal of data quality Increase public demand/workforce CRVS knowledge: Increase knowledge about the public health importance of good vital registration data at all levels of the CRVS system; build community awareness on need and benefits of registration Enterprise Architecture and Business Process Mapping Enterprise Architecture - Conceptual blueprint that defines the structure and operation of any enterprise. Shows how an organization can most
effectively achieve current and future objectives. Business Process Analysis - The effort to understand an organization (or system) and its purpose while identifying the activities, participants, and information flows that enable the organization to do its work. Business Process Mapping and Modelling - A method for representing the processes in a system and predicting performance EA applied to CRVS systems EA can be the organizing logic for understanding how CRVS processes and CRVS information
technology & information flows work together. A system analysis using an EA approach is extremely useful to describe, understand, and compare national CRVS systems in a systematic way. EA can be used to monitor change over time Automated Verbal Autopsy Three stage process: Interview of the deceaseds family using a standard, brief questionnaire to gather information on signs and symptoms experienced before death Diagnosis of cause of death based on this information
Compilation of VA data to derive population-level cause-specific mortality fractions (CSMFs) Why automate the process of assigning cause of death? Historically, VA questionnaires were reviewed by a physician. However, this approach suffers from many disadvantages; it is expensive, time consuming, non-standardised (due to differences in physician training and diagnostic skills), and takes physicians away from essential health care provision. Cause of Death (CoD) Certification Why should doctors be trained to accurately report cause of death?
In general, there is little emphasis on the sequence of events leading to death. No culture of support for or professional understanding of the importance of medical certification leading to underlying COD. Most junior doctor delegated the task of certification. Complete lack of oversight. Conclusion: this is a trivial routine of little concern to the practising clinician. Feeling that the needs of the living have greater claim on time than the description of the dead. ICD-10 coding manual and automated (IRIS)
The ICD-10 was developed by WHO to systematically record morbidity and mortality data It is now the international standard diagnostic classification for epidemiological and may health management purposes The ICD consists of alphanumeric codes and these are used to translate diagnoses of diseases and other health problems from words
Although primarily designed for the classification of diseases and injuries with a formal diagnosis, it also provides for a wide variety of signs and symptoms, abnormal findings, complaints and social circumstances Can also be automatically coded using IRIS Analysis of Causes of National Death for Action - ANACONDA Quality assessment of causes of death Policy should be informed by accurate and timely data Poor quality data poor decisions lost opportunities to improve population health Availability of data however does NOT guarantee
quality COD data are complex. Potential for errors, biases, omissions etc. exist at many levels of a CRVS system. We need to identify these and correct them Hence ANACONDA Welcome to the ANACONDA Tool: Getting started Distribution of deaths by broad cause groups Country example: what exactly are we doing in/with countries?
Myanmar CRVS situation in Myanmar at baseline 450,000 deaths a year (high 5q0; 75/1000) 200,000 registered with CSO; 250,000 in HMIS system in MoH, but just fact of death 100,000 urban deaths (15 million pop); 40,000 in hospitals Midwives in rural villages responsible for reporting deaths, assigning cause and sending monthly paper reports to RHC
Midwives collect extensive information on each birth and death; possible reason for low reporting Phase 1 strategy: Overhaul system to improve efficiency of birth and death registration practices and accuracy of CoD using VA
Systematic System Review with recommended changes to forms based on international standards, and use of tablets for reporting events and doing VA on all reported deaths. Phase 1 roll-out in 14 townships. Established TWGs & engaged with the strategic plan of Central Committee on Birth and Death Registration (CCBDR) Strengthened collaboration and awareness of the stakeholders & community on the importance of birth and death registration CSO staff trained on birth and death completeness methods to be applied to future CRVS data to monitor improvements Trained 10 coders in ICD coding and 119 master trainers in McCoD CRVS: Verbal Autopsy methods to assign COD for non-facility deaths (84% of all
deaths) Pilot completed in 3 townships with evaluation Implementation in 14 townships (pop ~2.2 M) between Jan-May 2017 Collaboration (training and analysis) with Global Fund National Mortality survey in nationally representative sample of deaths in 34 townships ( 2 per region) Evaluation in June 2017 and Phase 2 activities
Plan to roll-out to 48 townships across Myanmar during 2 years (using existing capacity through the Global Fund mortality survey) CRVS: Training on Medical certification of Cause of Death (MCCOD) and ICD coding Trained 119 Master trainers and 1,000 doctors (out of 10,000 nationally) from 4 states and regions using D4H assessment tool for MCCOD
ICD-10 coder training for 10 Master trainers and data management staff in 14 townships Phase 2 a) Scale up MCCOD to a national level b) Incorporate COD certification training into medical school curricula choose 20 largest hospitals and train all doctors c) One single integrated coding system is established to obtain ICD coded COD data for all deaths Anticipated skills/system changes with long-term impact ( Myanmar)
Awareness among doctors of the public health importance of correctly certifying death A cadre of trainers in sentinel hospitals who can train other hospitals, with D4H tools in local language A VA method and data collection/reporting mechanism that has been shown to work for community deaths and can be further
rolled out nationally Established nationwide integrated mortality surveillance system incorporating both VA and MCCoD data Core staff with confidence and competence to analyse and interpret these data for monitoring national development goals Ability to use tools such as ANACONDA to routinely monitor and critically assess data quality to accelerate system improvements Bloomberg Philanthropies D4HI (CRVS): expected outcomes/impact in all 20 countries/cities
Substantial improvement in availability and quality of data on births, deaths and causes of death, irrespective of initial state of the VS system More efficient and rational CRVS systems, including integration of HDSS/ SRS sites into CRVS system Improved capacity in countries for critical data quality assessment and data analysis/interpretation New manuals/tools/software packages (e.g. SmartVA)/CRVS knowledge resources (UoM CRVS Knowledge Gateway, Learning Centre) Much improved understanding of the critical role of reliable and timely
VS data to guide health and development policies & monitor progress with goals
Systemic arterial hypoxemia may stimulate development of bronchpulmonary collaterals. Usually in TGA with solely a restrictive inter-atrial communication. Prolonged survival of such infants may be due to this extra-cardiac site of shunting/mixing. Bronchial veins are also a source of LA...
They are the number one leader of printing products in the world. 20% of its revenue came from the printing and imaging group in the 2011 fiscal year, making it the most sold peripheral group (Ibis World & Bueno, 2012)....
The Foot . A muscular organ, shaped and used differently by different species. Used for digging, grasping, or creeping. Examples: Pelecypoda (Bivalvia) - "hatchet - shaped foot". Clams & Oysters - use the foot for digging
Causality reasoning Enablement, causation, prevention relations between events Knowledge about words for access to content The verb "acquire" is a hypernym of the verb "buy" The verbs "get to" and "reach" are synonyms Inferential properties of "manage", "fail", "avert", "not"...
Adjustable Pension Plan (APP) Richard Hudson, Principal Consulting Actuary CHEIRON, Inc * The General Concept Adjustable Pension Plan (APP) 100% employee 100% employer Shared risk Traditional DC (eg 401k) Traditional DB APP Like a DB Plan Employers' bear mortality risk,...
need to be carefully marked, if BB designation missed at any point will go entirely to DMF. Can set up automatic giving for BB. Direct gifts to Disciples Women - online, card reader, check **Pat and Chesla have card readers...
Rule 1 Most nouns become plural by adding "s". boy book desk school boys books desks schools Rule 2 Nouns that end in "ch", "sh", "x", add "es". bush box church bushes boxes churches Rule 3 Nouns that end in...
Ready to download the document? Go ahead and hit continue!