WEBVTT 00:00:00.000 --> 00:00:00.500 align:middle line:90% 00:00:00.500 --> 00:00:01.980 align:middle line:90% [MUSIC PLAYING] 00:00:01.980 --> 00:00:07.443 align:middle line:90% 00:00:07.443 --> 00:00:08.860 align:middle line:84% EME OWOAGE: In this section, we'll 00:00:08.860 --> 00:00:12.490 align:middle line:84% be looking at the promise and limits of technology, 00:00:12.490 --> 00:00:16.770 align:middle line:84% and information technology for decision-making in Nigeria. 00:00:16.770 --> 00:00:19.800 align:middle line:90% 00:00:19.800 --> 00:00:23.640 align:middle line:84% So we'll be looking at auto-visual AFP detection 00:00:23.640 --> 00:00:26.120 align:middle line:90% and reporting, AVADAR. 00:00:26.120 --> 00:00:31.880 align:middle line:84% As you may know, AFP stands for acute flaccid paralysis. 00:00:31.880 --> 00:00:37.220 align:middle line:84% There's a link there regarding the WHO collaboration rollout 00:00:37.220 --> 00:00:39.860 align:middle line:84% for this mobile application, and we 00:00:39.860 --> 00:00:42.260 align:middle line:84% expect that you would have read this and reviewed 00:00:42.260 --> 00:00:46.940 align:middle line:84% this particular document before this lecture. 00:00:46.940 --> 00:00:49.760 align:middle line:90% So how does AVADAR work? 00:00:49.760 --> 00:00:51.620 align:middle line:90% What does it add? 00:00:51.620 --> 00:00:53.900 align:middle line:84% I'd like you to ponder on this questions 00:00:53.900 --> 00:00:59.110 align:middle line:84% and then answer them before we go on. 00:00:59.110 --> 00:01:02.350 align:middle line:84% So it's an alternate surveillance method 00:01:02.350 --> 00:01:06.940 align:middle line:84% relying on community members, rather than the health system. 00:01:06.940 --> 00:01:09.760 align:middle line:84% What would be the pluses and minuses 00:01:09.760 --> 00:01:12.830 align:middle line:84% of collecting surveillance information this way? 00:01:12.830 --> 00:01:17.450 align:middle line:84% It goes directly to you, but might miss a lot of AFP. 00:01:17.450 --> 00:01:22.780 align:middle line:84% And potentially cases, as well as nearly all cases of polio, 00:01:22.780 --> 00:01:27.200 align:middle line:84% will end up at the health center at some point in time. 00:01:27.200 --> 00:01:29.590 align:middle line:84% And the random community member might 00:01:29.590 --> 00:01:33.100 align:middle line:90% be less likely to see them. 00:01:33.100 --> 00:01:36.190 align:middle line:84% In the end, one person with experience with AVADAR 00:01:36.190 --> 00:01:38.230 align:middle line:84% noted, essentially it was supposed 00:01:38.230 --> 00:01:40.850 align:middle line:90% to be health facility based. 00:01:40.850 --> 00:01:43.120 align:middle line:84% Community based surveillance added 00:01:43.120 --> 00:01:47.980 align:middle line:84% to complete this at some places, but with excess reporting 00:01:47.980 --> 00:01:51.400 align:middle line:90% and bad quality. 00:01:51.400 --> 00:01:55.860 align:middle line:84% Another technology that can be used for data collection 00:01:55.860 --> 00:02:01.080 align:middle line:84% is the geographical information systems, which is GIS. 00:02:01.080 --> 00:02:05.250 align:middle line:84% So we'll be talking about GIS data in Nigeria. 00:02:05.250 --> 00:02:08.940 align:middle line:84% We have a quote from a global-level policymaker. 00:02:08.940 --> 00:02:13.140 align:middle line:84% "Kano, a city in northern Nigeria, is so dense 00:02:13.140 --> 00:02:17.780 align:middle line:84% and there was much effort and so many resources going into it, 00:02:17.780 --> 00:02:20.430 align:middle line:90% and we were still seeing cases. 00:02:20.430 --> 00:02:22.980 align:middle line:84% And it was kind of like, well heck, 00:02:22.980 --> 00:02:25.630 align:middle line:90% what else are we supposed to do. 00:02:25.630 --> 00:02:29.130 align:middle line:84% We thought, what if we tried to map out our communities 00:02:29.130 --> 00:02:34.670 align:middle line:84% and double-check that they're all getting covered." 00:02:34.670 --> 00:02:39.230 align:middle line:84% On the slide you can see hand-drawn and digital versions 00:02:39.230 --> 00:02:41.510 align:middle line:90% of maps of the same community. 00:02:41.510 --> 00:02:44.330 align:middle line:84% And you can see that the digital map 00:02:44.330 --> 00:02:47.090 align:middle line:84% is able to capture some areas that have been 00:02:47.090 --> 00:02:49.970 align:middle line:90% missed by the hand-drawn map. 00:02:49.970 --> 00:02:53.270 align:middle line:84% So a comment was made that even good hand-drawn maps 00:02:53.270 --> 00:02:55.490 align:middle line:90% have missed settlements. 00:02:55.490 --> 00:02:59.630 align:middle line:84% And not everything was where we thought it was. 00:02:59.630 --> 00:03:03.260 align:middle line:84% I'd like to ask you, what are the pluses and the minuses 00:03:03.260 --> 00:03:07.520 align:middle line:84% of using maps that are hand-drawn by campaign workers, 00:03:07.520 --> 00:03:12.950 align:middle line:84% versus maps created from aerial photography and GIS? 00:03:12.950 --> 00:03:16.430 align:middle line:84% Can you think of any pluses to the hand-drawn maps 00:03:16.430 --> 00:03:18.890 align:middle line:90% over the GIS maps? 00:03:18.890 --> 00:03:21.950 align:middle line:84% A possible answer might be that they 00:03:21.950 --> 00:03:25.750 align:middle line:84% are reflective of the health workers mental maps, 00:03:25.750 --> 00:03:27.980 align:middle line:90% and may be more useful to them. 00:03:27.980 --> 00:03:32.750 align:middle line:84% Also, the act of making maps can help local health workers 00:03:32.750 --> 00:03:35.840 align:middle line:90% learn about their areas. 00:03:35.840 --> 00:03:37.620 align:middle line:84% I'd like to ask you some more questions. 00:03:37.620 --> 00:03:40.190 align:middle line:90% What are the cost implications? 00:03:40.190 --> 00:03:44.120 align:middle line:84% Should public health programs be more cost sensitive? 00:03:44.120 --> 00:03:47.450 align:middle line:84% The work and quality has to be delivered on the ground. 00:03:47.450 --> 00:03:50.180 align:middle line:84% People may use aerial technology, 00:03:50.180 --> 00:03:53.900 align:middle line:84% but ground forces have to be aware of their area 00:03:53.900 --> 00:03:57.650 align:middle line:84% of jurisdiction and the populations there. 00:03:57.650 --> 00:04:01.610 align:middle line:84% Also, they need to take care of population movements, 00:04:01.610 --> 00:04:06.750 align:middle line:84% and temporary populations or changing dynamics there. 00:04:06.750 --> 00:04:10.180 align:middle line:84% In this slide, we are looking at improving maps 00:04:10.180 --> 00:04:13.210 align:middle line:90% in important geographies. 00:04:13.210 --> 00:04:16.660 align:middle line:84% It really doesn't require highly trained or sophisticated 00:04:16.660 --> 00:04:17.470 align:middle line:90% devices. 00:04:17.470 --> 00:04:19.579 align:middle line:84% You can see from this map that was 00:04:19.579 --> 00:04:22.750 align:middle line:84% taken from a local government Katsina State 00:04:22.750 --> 00:04:26.680 align:middle line:84% in Nigeria, the improvements in the number of settlements 00:04:26.680 --> 00:04:31.210 align:middle line:84% that were identified based on the GIS map. 00:04:31.210 --> 00:04:35.260 align:middle line:84% So on the right-hand side, you can see the final map, 00:04:35.260 --> 00:04:38.870 align:middle line:90% which is more detailed. 00:04:38.870 --> 00:04:42.050 align:middle line:84% In this slide, we'll be looking at global positioning 00:04:42.050 --> 00:04:46.130 align:middle line:84% systems, which is GPS tracking of vaccination teams. 00:04:46.130 --> 00:04:49.220 align:middle line:84% And post campaign reports include 00:04:49.220 --> 00:04:54.170 align:middle line:84% geo-coverage, settlements visited and percentage 00:04:54.170 --> 00:04:57.470 align:middle line:84% of total, target population visited 00:04:57.470 --> 00:05:00.230 align:middle line:84% and percentage of total, heat map showing 00:05:00.230 --> 00:05:03.470 align:middle line:90% visited and missed settlements. 00:05:03.470 --> 00:05:06.950 align:middle line:84% So I'd like to ask you, what are the implications of GPS 00:05:06.950 --> 00:05:10.120 align:middle line:90% tracking of vaccination teams? 00:05:10.120 --> 00:05:12.740 align:middle line:84% In your opinion, is this appropriate? 00:05:12.740 --> 00:05:17.010 align:middle line:84% One person with familiarity with this commented, technology 00:05:17.010 --> 00:05:18.950 align:middle line:84% and tracking is fine for monitoring 00:05:18.950 --> 00:05:20.660 align:middle line:90% the complete coverage. 00:05:20.660 --> 00:05:25.340 align:middle line:84% Tracking might have a negative effect on vaccinators morale, 00:05:25.340 --> 00:05:28.190 align:middle line:84% that their movements are being tracked. 00:05:28.190 --> 00:05:30.590 align:middle line:84% Finally, what matters is two drops 00:05:30.590 --> 00:05:34.270 align:middle line:84% in the mouth of every eligible child. 00:05:34.270 --> 00:05:36.640 align:middle line:84% Here's another example from Nigeria. 00:05:36.640 --> 00:05:40.630 align:middle line:84% And from the tracking, improved coverage of border settlements 00:05:40.630 --> 00:05:42.850 align:middle line:90% is highlighted. 00:05:42.850 --> 00:05:49.650 align:middle line:84% So you can see that GPS is important in the polio program. 00:05:49.650 --> 00:05:54.060 align:middle line:84% This slide was showing that the GPS can direct field workers 00:05:54.060 --> 00:05:56.250 align:middle line:90% to address gaps. 00:05:56.250 --> 00:05:58.560 align:middle line:84% In Nigeria, the chronically missed settlements 00:05:58.560 --> 00:06:01.490 align:middle line:84% identified, thanks to mapping and tracking, 00:06:01.490 --> 00:06:04.890 align:middle line:84% were located using a geographical information 00:06:04.890 --> 00:06:11.700 align:middle line:84% systems, GIS map app running on a rugged field tablet. 00:06:11.700 --> 00:06:17.010 align:middle line:84% The app has a set of online imagery and base map layers, 00:06:17.010 --> 00:06:20.640 align:middle line:84% and helps with navigating to the unknown settlement. 00:06:20.640 --> 00:06:25.200 align:middle line:84% The local polio officers are coming along on these exercises 00:06:25.200 --> 00:06:27.570 align:middle line:84% to make sure the settlements gets 00:06:27.570 --> 00:06:31.720 align:middle line:84% included in the microplanning for the fully campaigns. 00:06:31.720 --> 00:06:35.460 align:middle line:84% However, if there is a good system and ownership, 00:06:35.460 --> 00:06:39.510 align:middle line:84% vaccinators should know their area on ground. 00:06:39.510 --> 00:06:44.590 align:middle line:84% Technology use can be supplementary. 00:06:44.590 --> 00:06:47.540 align:middle line:84% So we're going to talk about a success story. 00:06:47.540 --> 00:06:50.820 align:middle line:84% And this story is given by a global-level policymaker. 00:06:50.820 --> 00:06:54.300 align:middle line:84% So I think it just brought a measure of precision 00:06:54.300 --> 00:06:57.780 align:middle line:84% to the Nigeria program in terms of microplanning 00:06:57.780 --> 00:07:02.370 align:middle line:84% and identify missed settlements, that literally had been missed 00:07:02.370 --> 00:07:07.650 align:middle line:84% by all services for years, hundreds of settlements that 00:07:07.650 --> 00:07:10.560 align:middle line:84% just had never been in microplans 00:07:10.560 --> 00:07:15.690 align:middle line:84% for the entirety of the polio eradication efforts in Nigeria. 00:07:15.690 --> 00:07:18.870 align:middle line:84% And those communities also weren't associated 00:07:18.870 --> 00:07:22.320 align:middle line:84% with any particular health center in the public health 00:07:22.320 --> 00:07:23.970 align:middle line:90% planning system. 00:07:23.970 --> 00:07:25.830 align:middle line:90% That really changed things. 00:07:25.830 --> 00:07:30.420 align:middle line:84% And I'm still working on routine immunization in Nigeria. 00:07:30.420 --> 00:07:36.120 align:middle line:84% And I still see the knock-on effects of that. 00:07:36.120 --> 00:07:38.790 align:middle line:84% They are now using the same GIS maps 00:07:38.790 --> 00:07:44.040 align:middle line:84% to create microplans in routine immunization outreach. 00:07:44.040 --> 00:07:46.050 align:middle line:90% So this is a success story. 00:07:46.050 --> 00:07:50.220 align:middle line:84% I'd like you to think about, why had these settlements been 00:07:50.220 --> 00:07:52.680 align:middle line:90% missing from the microplans. 00:07:52.680 --> 00:07:56.790 align:middle line:84% What does this tell us about local knowledge, government 00:07:56.790 --> 00:07:59.820 align:middle line:84% knowledge, international knowledge, and the efforts 00:07:59.820 --> 00:08:01.710 align:middle line:90% to eradicate disease? 00:08:01.710 --> 00:08:06.210 align:middle line:84% What would have been alternate old-school strategies 00:08:06.210 --> 00:08:08.340 align:middle line:90% to the GIS methods? 00:08:08.340 --> 00:08:14.270 align:middle line:84% What would have been the pluses and minuses of these methods? 00:08:14.270 --> 00:08:15.730 align:middle line:84% So in this slide, we'll be talking 00:08:15.730 --> 00:08:18.100 align:middle line:84% about how technology doesn't eliminate 00:08:18.100 --> 00:08:20.360 align:middle line:90% the importance of politics. 00:08:20.360 --> 00:08:24.250 align:middle line:84% We have a quotation from a global-level policymaker, 00:08:24.250 --> 00:08:29.020 align:middle line:84% and it says, "It took years, I mean, from 2011, 00:08:29.020 --> 00:08:38.230 align:middle line:84% 2012 to 2014, 2015, I'd say that finally they, the GIS data, 00:08:38.230 --> 00:08:41.289 align:middle line:90% were being used effectively. 00:08:41.289 --> 00:08:44.260 align:middle line:84% And I think had we done it another way, 00:08:44.260 --> 00:08:48.490 align:middle line:84% the government would have owned the maps a bit better. 00:08:48.490 --> 00:08:52.310 align:middle line:84% And they would have been adopted a lot easier. 00:08:52.310 --> 00:08:55.810 align:middle line:84% But that's kind of how it came about, 00:08:55.810 --> 00:08:59.170 align:middle line:84% just like, OK, what's our last-ditch effort 00:08:59.170 --> 00:09:00.850 align:middle line:90% to get this done?" 00:09:00.850 --> 00:09:03.310 align:middle line:84% This quote from a global partner is 00:09:03.310 --> 00:09:06.520 align:middle line:84% about how this information did get integrated 00:09:06.520 --> 00:09:08.530 align:middle line:90% into the government right away. 00:09:08.530 --> 00:09:13.810 align:middle line:84% The mapping was initially done by Gates Foundation and others. 00:09:13.810 --> 00:09:19.300 align:middle line:84% As this interviewee describes, it was conceived by them 00:09:19.300 --> 00:09:22.180 align:middle line:90% as part of a last-ditch effort. 00:09:22.180 --> 00:09:27.580 align:middle line:84% Think about why might it be that this scenario would 00:09:27.580 --> 00:09:32.710 align:middle line:84% lead to government resistance to using the maps. 00:09:32.710 --> 00:09:36.340 align:middle line:84% The interviewee adds, that was our fault. 00:09:36.340 --> 00:09:39.220 align:middle line:84% And the way we went about funding that 00:09:39.220 --> 00:09:44.400 align:middle line:84% through the partners meant that it was a whole public good 00:09:44.400 --> 00:09:47.230 align:middle line:90% for Nigeria as a country. 00:09:47.230 --> 00:09:50.910 align:middle line:84% I mean, they could use that for education, planning, 00:09:50.910 --> 00:09:52.260 align:middle line:90% infrastructure. 00:09:52.260 --> 00:09:55.620 align:middle line:84% But it wasn't set up within the government system. 00:09:55.620 --> 00:09:59.070 align:middle line:84% And then getting the government to trust that data 00:09:59.070 --> 00:10:02.610 align:middle line:84% and take it on, has been the whole process. 00:10:02.610 --> 00:10:03.780 align:middle line:90% It's still ongoing. 00:10:03.780 --> 00:10:07.170 align:middle line:90% 00:10:07.170 --> 00:10:08.820 align:middle line:84% So in this slide, we'll be talking 00:10:08.820 --> 00:10:13.210 align:middle line:84% about how data doesn't eliminate the importance of context. 00:10:13.210 --> 00:10:16.800 align:middle line:84% We have a quote from a global-level policymaker. 00:10:16.800 --> 00:10:20.330 align:middle line:84% "The question you had about why is it 00:10:20.330 --> 00:10:23.940 align:middle line:84% GIS mapping can't be applied elsewhere. 00:10:23.940 --> 00:10:27.570 align:middle line:84% I mean, in Pakistan and Afghanistan, 00:10:27.570 --> 00:10:31.380 align:middle line:84% the fear is around government and/or 00:10:31.380 --> 00:10:35.520 align:middle line:84% international surveillance are just too high. 00:10:35.520 --> 00:10:38.460 align:middle line:84% In Nigeria, it's accepted and it's OK. 00:10:38.460 --> 00:10:43.920 align:middle line:84% And in Pakistan and Afghanistan, the fake vaccination campaign 00:10:43.920 --> 00:10:48.630 align:middle line:84% by the CIA or the American intelligence agency 00:10:48.630 --> 00:10:51.570 align:middle line:84% and all that, just makes it too high stakes 00:10:51.570 --> 00:10:53.710 align:middle line:90% to make it possible." 00:10:53.710 --> 00:10:57.180 align:middle line:84% So I'd like you to think about, given what options you know 00:10:57.180 --> 00:11:00.870 align:middle line:84% from other portions of this class or this model, 00:11:00.870 --> 00:11:02.890 align:middle line:90% why would this be the case? 00:11:02.890 --> 00:11:05.880 align:middle line:84% Why might what worked so well in Nigeria not 00:11:05.880 --> 00:11:08.900 align:middle line:90% work in Afghanistan? 00:11:08.900 --> 00:11:14.160 align:middle line:84% What does the interviewee mean by high stakes? 00:11:14.160 --> 00:11:20.590 align:middle line:84% A culture of data, so what do we mean by culture of data? 00:11:20.590 --> 00:11:23.700 align:middle line:84% So we're looking at data for decision-making, 00:11:23.700 --> 00:11:25.630 align:middle line:90% or data for data. 00:11:25.630 --> 00:11:29.580 align:middle line:84% And then we'll also be looking at data surge. 00:11:29.580 --> 00:11:32.550 align:middle line:84% Polio has a strong culture of data. 00:11:32.550 --> 00:11:36.360 align:middle line:84% This is, as we have discussed, a strength. 00:11:36.360 --> 00:11:40.690 align:middle line:84% And it has been something other programs have drawn from. 00:11:40.690 --> 00:11:43.530 align:middle line:84% It can also lead to some complexities. 00:11:43.530 --> 00:11:47.490 align:middle line:84% First, why is data being collected and used? 00:11:47.490 --> 00:11:50.880 align:middle line:84% Data collection and analysis can and do become ends 00:11:50.880 --> 00:11:55.110 align:middle line:84% in themselves, without always enough consideration 00:11:55.110 --> 00:11:58.290 align:middle line:84% of how this data will actually be used 00:11:58.290 --> 00:12:00.720 align:middle line:90% in day to day decision-making. 00:12:00.720 --> 00:12:02.790 align:middle line:84% Sometimes a great deal of effort is 00:12:02.790 --> 00:12:06.900 align:middle line:84% expended to collect data that is never used. 00:12:06.900 --> 00:12:10.170 align:middle line:84% A related issue is the data surge. 00:12:10.170 --> 00:12:13.230 align:middle line:84% There may actually, in some cases, 00:12:13.230 --> 00:12:18.180 align:middle line:84% be too much data to use effectively. 00:12:18.180 --> 00:12:20.930 align:middle line:90% [MUSIC PLAYING] 00:12:20.930 --> 00:12:26.000 align:middle line:90%