Hamlet areas vaccination – GIS micro planning: A case study from Kano...

Hamlet areas vaccination – GIS micro planning: A case study from Kano State, Nigeria


Much is known about the use of GIS drawn maps in western countries. However, its empirical use to promote immunization activities in under developed nations, particularly Nigeria, is still obscure.  Because Polio still remains in the country, it is imperative that whatever will promote quality campaigns (increase immunisation coverage) needs to be done. One such way is the use of GIS drawn maps to conduct IPDs micro plans.

Analysis of GIS dashboard information over time revealed poor coverage of hamlet areas during IPDs campaigns. In Tudun Wada LGA of kano state for instance; after June 2013 IPDs, only 62.2% of the hamlet areas were visited by vaccinators. Cumulative geographic coverage after 4 days implementation of IPDs in the hamlet areas was a dismal 49.9%. Indeed a total of about 17 hamlet areas with machine generated names were discovered to be uncovered at all by teams in Dalawa ward of Tudun Wada LGA alone.

Map below shows Dalawa before & after the activity.

Machine generated or named hamlet areas are tiny hamlets not reflected in the updated ward micro plan list of settlements but sighted through satellite imagery. Despite repeated update of micro plans in every round, these hamlet areas have not been identified by any local names, hence difficulty in tracing them for vaccinations. After 4 days implementation of 2013 IPDs in Dalawa ward, missed hamlet areas (including machine generated and locally named ones) were line-listed and shipped back to the LGA for 2 days mop-up. The immunization coverage remained the same after the mop-up activity. Many consultants opine that, vaccination teams very well know where to find the hamlet areas but for attitudinal problems do not visit them for vaccinations. However, studying the pattern of GPS tracks generated by vaccinators in Dalawa ward over many rounds suggests the contrary.

Recent (June 2013) GIS map of Dalawa Ward shows a wholly rural ward with over 90% of the inhabitants living in hamlet areas. From arc GIS, it has a total land mass of about 45 sq km. There are 15 vaccination teams operational in the ward during IPDs. Quick arithmetic suggests that, a team area for 4 days of IPDs implementation is probably 3 sq km. If this is divided into 4 days, a team will be ‘roaming’ a land mass of less than a km square. If this is the case, it should not be difficult to ‘roam’ into the machine generated hamlet areas during a day’s assignment. This is the logic of GIS micro-planning!

The activity is a quasi-experiment. Two set of wards were selected, ‘Test and Control’. GIS drawn maps were used to conduct micro-plans for implementation of September 2013 IPDs in the Test ward only. During ward level trainings, relevant personnel including ward focal persons, group supervisors, ward heads, monitors received training/and generated micro-plans in the GIS manner. This was the ‘treatment’. Prior to implementation (2days), GIS consultant led a vaccination team in a field rehearsal of using the GIS micro-plans so produced. The Control ward operated with their normal conventional micro-plans. Both wards were supported by VTS (vaccination tracking system) throughout the campaign. The activity was piloted in 3 wards namely Dalawa, Nata’ala and Baburi of Tudun Wada LGA. The wards were assigned thus: Dalawa (Test ward, where GIS maps were used & supervision very intensive), Nata’ala (Control 1, where no GIS maps were used but supervision very intensified) and Baburi (Control 2 ward, where neither GIS maps were used nor supervision very intensive).

The study suggests that the Test ward where GIS drawn maps was used to support micro-planning (GIS micro-planning) fared much better than other wards in terms of everything including, coverage of HAs, overall coverage of all settlements, vaccine utilization, report of AFP etc. Even the trends within the Test ward showed remarkable improvement this round (September) when GIS micro-planning was done compared to previous rounds when it was not.

During the September round, the geographic coverage of HAs alone increased by more than 50% in the test. In Control 1 (which benefited the same kind of high level supervision as the Test), it was a dismal 16% rise. Precisely, 65 out 100 HAs were well covered (65% geographic coverage) in the Test ward, while only 43 out of 100 HAs (43% geographic coverage) were reached in Control 1. Using the Chi square statistics and at 0.05 level of significance, calculated x2 = 8.08. As this value is greater than tabulated value of 3.84, the null hypothesis is rejected. Thus, the coverage of hamlet areas is contingent on the ward of study and hence type of micro-plan used.

Statistical comparison of Control 1 & 2 revealed no association between coverage of HAs and supervision. At 0.05 level of significance, it is clear that, supervision no matter how intensified has no impact on coverage of HAs. This is not unexpected; you only supervise what & where you know!

In the study, there was a precipitous drop in OPV utilization in the Test ward compared to the Controls. The general trend in OPV utilization appears to be increasing vaccine use with each and every round of IPDs. This jinx was broken in the Test ward. Indeed, close to 3,000 doses of OPV were saved in the Test ward compared to previous rounds.

Overall geographic coverage of all settlements (BUAs, SS & HAs) September round when compared to vaccine utilization in the last 3 rounds for the Test ward appear paradoxical and tend to suggest that, the real immunization coverage in terms of children vaccinated (seen having little fingers marked during end process monitoring by surveyors) is not as high as previous rounds. LQAS result (September round) of 90% coverage in the Test ward however, is a testimony that real immunization coverage is truly good in September round and that probably in the past rounds vaccines were simply flagrantly wasted!

Only the test ward reported an AFP case. It is zero dose with respect to OPV. It was also seen among the machine named HA reached (Mallam Bello’s camp, lat 11°18”2.47″N, long 8°38”42.57″E). The case seemed to be clinically compatible with Polio and suggests that, machine named HAs are a safe heavens to WPV where nobody ever goes for vaccinations!