23 – Measure what you manage: the data quality challenge

in page functions

Administrative data systems are of highly variable quality across different GAVI eligible countries. Sources of error include numerator and denominator measurement inaccuracies.

06GAVI_Moz_20120313_Eva-LottaJanssonDSC_0141_700F

Copyright GAVI/2012/Eva-Lotta Jansson

Immunisation coverage data informs the management, performance and evaluation of EPI programmes. Coverage data sources include both administrative data reported by EPI programmes as well as coverage surveys.

While both data sources are complementary, administrative data systems are of highly variable quality across different GAVI eligible countries. Sources of error include numerator and denominator measurement inaccuracies.

Currently available tools such as the Data Quality Self-assessment and new tools such as the revised Immunisation Data Quality Assessment and the WHO denominator guidance tool are part of the GAVI Alliance’s strengthened approach to assessing and improving data quality in EPI programmes as part of a larger framework of data quality improvement.

New technologies including mobile health based platforms are being utilized to improve data recording and reporting in some GAVI-eligible countries.

The session was introduced by Anne Schuchat, Centers for Disease Control and Prevention, United States, who emphasized the importance in immunization of understanding what we are doing, and how we are doing it.

The first presenter (Peter Hansen, GAVI Alliance ) discussed the design of a Data Quality Improvement Framework and the Revised Immunisation Data Quality Assessment Tool (IDQA).

The IDQA, which has been piloted in three countries (Ghana, Bolivia, Uganda), is shifting to a “data plausibility” approach, with assessment at different levels. It promote a south-south collaboration, with cross-assessment between countries, and its implementation will start in 2013 as a substantial input in the business process, although it will not be used for PBF approach.

The second presenter (David Brown, UNICEF ) discussed the denominator issues for immunisation coverage. While a remarkable increase in immunization coverage has been observed at global level, wide inter-district variations are frequent in-country.

This raises the issue of both numerator and denominator accuracy. As regards the denominator, less that a third of GAVI eligible countries have a vital registration system. The numerator is also difficult to estimate, given that vaccination cards holding is less than 50% in 17 of 87 countries surveyed. 21 out of 33 less developed countries have a card retention rate that is less than 70%.

What is critical is to:

(1) record the immunization where and when it happens,

(2) interpret coverage data,

(3) identify problems and their causes.

Countries experiences were introduced by Thomas Cherian, from WHO.

Cuauhtemoc Ruiz Matus(WHO PAHO) presented the examples of innovative tools for data quality improvements developed by Bolivia and Honduras.

Countries in the region have started developing national computerized nominal immunization registries.

These registries have been made possible by the increase in computer availability and improved connectivity, and indispensable by the saturation of the current system with the introduction of new vaccines, and the difficulty in reliably estimating the denominator.

DQA have been conducted since 2005 in 17 countries, and they are increasingly integrated in the regular EPI evaluation, and not as a stand-alone exercise.

Dan Osei, from the Ghana Health Service, explained how Ghana introduced a number of initiatives, some based on mobile phone technology, some on computerized systems.

2 systems are dedicated on logistics (early warning system and SMS for life).

The Mobile technology for community health service (MoteCH) is assisting in the follow up of pregnant women.

The iHRS is focusing on human resource for health.

The Logistics management information system (LMIS) is a computer-based system for logistics in the MOH.

All these systems and pilots should be integrated in the DHIMS, which captures integrated data from all levels, and is used as the main report tool by the MOH. Some integration still pose problems and has to be done manually (notably SMS for life in the LMIS).

Discussion:

India reported on the project that has been started under the leadership of the Ministry of Health, based on data reported by mobile phone (SMS) or dial-in system directly from the facilities, for reporting on 10 vaccine-preventable diseases.

A participant remarked that, while card retention is mostly very poor in many countries, it is completed usually by the history report form the care-givers.

An experience from Nigeria in LQAS survey for OPV raised a discussion on the interest of the new monitoing techniques developed by the polio programs,. UNICEF is looking at integrating LQAS in RI.

A problem that was raised several time is the limit in HR/ the “human factor”. Data collection is time-consuming, not very interesting and poorly rewarded, and replacing paper by electronic systems does not change radically the problem.

However, technology does help saving staff time, with for example bar-code technology that avoid multiple manual data entry.

The Ghana system has been developed with the objective of reducing the workload at the service provider level.

The experience in Honduras, where there is an agreement between all the ministries and offices to use the same data was commended. It is a good progress from the situation observed in many countries, where the national statistics office and the MOH never work together.

A participant remarked that, in addition to good immunization coverage data, we lack disease surveillance data and systems. He observed that no session on this subject has been organized during the Forum.

Ghana was questioned about who does what: the Policy, Planning, M&E department in the MOH is in charge of the development and implementation. Districts and line managers have access online to the data, which is dated (updated daily), and feedback are possible. It is used for operational decision.

The major challenge in Ghana is in mHealth deployment, because the MOH has been struggling with technology change, etc…In general, the strategy was to develop several options, keep them flexible, and adapt. A meeting is organized each quarter to discuss feedback and corrective actions. This approach has made the progress slow but steady.

US$ 8.4 billion

By 31 August 2013, GAVI had committed US$ 8.4 billion in programme support until 2016 to the world’s poorest countries.

GAVI Alliance, 30 April 2013

More facts...

close icon

modal window here