7.3 Cambodia Mine Victims Information System (CMVIS)
Handicap International (HI), in collaboration with Cambodian Red Cross (CRC) maintains this database. The database unit is dedicated to maintaining this database and to production of regular reports that are widely distributed to the mine action community.
Initially the survey planned to compile all victims’ information from the CMVIS and not to collect any new victims information as part of the questionnaire. At the conclusion of our discussions with UNMAS in September 2000, concerning certification requirements, we were directed to ensure the quality of any information imported to the survey database from any external source. Amongst other actions this caused us to perform a quality assurance review of information available from CMVIS, and in particular the accuracy of information on the location of accidents
to victims identified in the database.
Our first review of victims’ data from the CMVIS suggested that it was going to be difficult to confirm exactly where accidents had taken place due to the methods used for describing these locations. It needs to be stressed that location of accidents is not a primary
concern of HI/CRC as they are primarily concerned with victim assistance and their requirement is to know where the victim is now that (s)he needs support.
Victims’ data is important for the survey as accidents that have occurred in the two years immediately prior to the survey are given the most weight of any factor in the priority assessment scoring system adopted by the survey Action Centre (SAC). The numbers of victims of accidents that have occurred during the last two years have to be corroborated with victims’ names, and a minimum of information about the victim, commonly known as their bio-data. We had assumed that the CMVIS bio-data would be adequate to satisfy this requirement, but when we had difficulty confirming the location of CMVIS accidents, we decided it would be prudent to start collecting victims’ bio-data as part of the survey.
Between August 2000 and December 2000 the survey questionnaire dealing with victims’ numbers and bio-data was amended four times, as follows:
||Victims’ data collected
||Numbers of Victims in last two years
||Numbers and names of victims in last two years
(names appended to form)
|Pursat, Battambang and Pailin
||Numbers and names, with names included in the questionnaire, but no QA consistency checks established
||Banteay Meanchey, Otdar Meanchey, Siem Reap and Preah Vihear
||Final methodology including QA checks
||All other Provinces
It is important to remember that victims’ information is only collected from villages in which a suspected contaminated area is reported.
To ensure consistency in survey methodology, all contaminated villages in the provinces listed above for which questionnaire versions 1, 2 and 3 were used, were revisited in the period December 2001 – February 2002.
Throughout this period the survey interacted intensively with officials from HI/CRC in an effort to resolve these accident location identification issues. As a result the survey amended its procedures and changes were made to the reporting structure used by CRC/HI.
Two conclusions are of most importance:
- The numbers reported by both the L1S and the CMVIS for each of the last two years are almost identical, and
- Therefore, the number to use from the survey is the number of victims’ reported by village chiefs for the last two years.
As a consequence we recommend that the numbers of victims for which the survey was able to collect bio-data should NOT be used.
At the conclusion of the survey, the database and maps created by the survey have been supplied to the CRC/HI database staff. The intent is that the database staff will examine each victim data record for the last two years and attempt to plot on survey maps the reported location of each accident. This will allow CRC/HI to determine the extent to which these locations correspond to suspected contaminated areas. Over time the continued supply of accurate accident location information will be plotted on survey maps and help to identify previously undocumented suspected contaminated areas.
This work will assist CMAA in maintaining an up to date record of suspected areas and assist CMVIS in accurately recording accident locations.
The following statement summarises the conclusions jointly reached by the two projects.
Joint Statement by CRC/HIB and GeoSpatial/L1S on Victims’ Information
Over the last year a very large amount of time has been spent cross-checking CMAC/GeoSpatial International Level One survey (L1S) casualty information with that held in the CMVIS database. While the overall number of casualties reported by each project is quite similar, there are significant differences in casualty information reported, particularly in terms of the location of accidents
and personal information about casualties,
circumstances of their accidents. Overall, it was the opinion of both agencies that:
CMVIS seems to have more complete information concerning personal information
. This personal information includes: the nature of injuries suffered by casualties, information relating to activities and the situation leading to accidents, information on the present location of casualties and services they may have received from rehabilitation agencies.
L1S seems to have more accurate information about the location of accidents
than does CMVIS. The L1S has devoted considerable time and effort to ensure that the number of incidents and their locations, and the number of victims associated with these incidents is reported as accurately as possible. For the L1S the most important consideration is the number of victims and the location of the mined area in which each accident occurred.
While the L1S did collect personal information on casualties, the number of casualties reported by the L1S for which this personal information is available is much lower than reported by CMVIS.
Overall it is the opinion of HIB that the reasons for these differences are a result of the different methodologies employed by both projects, and can be summarised as follows:
- CMVIS deploys permanent data gatherers in mined areas through a system of surveillance, while the L1S utilised mobile teams to collect information utilising survey methods. Consequently, CMVIS spends longer trying to identify casualty bio-data (i.e. personal information about the casualty) and records more casualties with bio-data than the L1S. Moreover, CMVIS more often uses primary sources of information to collect casualty information (from the casualty, witnesses to an accident or immediate relatives of a victim), whereas the L1S gathers casualty information through interviews with village chiefs, which is verified by independent informants in the village. Consequently, CMVIS has a personal information report for each casualty, whereas the L1S has a summary report of all casualties in a particular area.
- CMVIS is dependent on information from the Geography Department about the location of villages and uses approximate measures of direction and distance for the location of accidents relative to the village centre, while the L1S utilises GPS technology. Consequently, CMVIS is less precise than the L1S about locating the accident site.
Importantly, the result of cross-checking the information from the two projects has reinforced the suitability of the methods employed by each to reach their stated objectives:
The main objective of the completed Level One survey is to facilitate planning and priority setting for humanitarian demining in terms of socio-economic impact, requiring the location of contaminated areas to be of the highest importance; whereas CMVIS strives to provide comprehensive ongoing information on the nature and situation of mine and UXO casualties in Cambodia to facilitate planning and monitoring of mine action and victim assistance in terms of casualty reduction and survivor assistance, requiring casualty information to be of the highest importance.
An agreement was reached between CMAC/GeoSpatial and HI, in early 2002, to include the two data sets in an integrated database. It is expected that this integration will occur sometime in 2002. The integrated database will make available to users the advantages of the data available from each project.