The first case of COVID-19 was reported in Nigeria on the 29th of February 2020 but there were confusions and some periods of inactivity until on the 22nd of March when a nationwide lockdown was imposed. Following this, all educational institutions, markets and places of worship were closed down. Motivations started coming on the need to visualise the trend and spread, especially as this was already going on for some countries like Australia (Krispin and Byrnes 2020).

The work by Krispin and Byrnes (2020) was a beautiful animation of COVID-19 cases in Australia at that time dis-aggregated to the States of Australia. Therefore, efforts were made to also produce the animation for Nigeria. This effort was met with disappointment as the Nigerian data (NCDC 2020) supplied to Johns Hopkins University Center for Systems Science and Engineering (JHU-CCSE) Coronavirus depository was not dis-aggregated to States. After days of re-compiling the data from Nigerian Centre for Disease Control (NCDC), the animation was produced and shared on Facebook. The outcome really indicated that the data needed further wrangling to produce good looking visuals.

First animation

Animation after dis-aggregation

There the venture to provide visual update started using rstats (R Core Team 2020) with ggplot (Wickham 2016). Here is what the raw data from NCDC looked like!

Fig. 1 Raw data as collected from NCDC

However, it was difficult to make any sense out of the raw data as you can observe in Fig. 1. Therefore, further processing of the data was required. In view of this, the data was accumulated for the whole country and presented in the following visuals. At this point, since the data points were few, elaborate analysis could not be carried out, so equal number of days forecast was invented. That is, a polynomial of two degree forecast of equal number of days since the advent of the virus in Nigeria.

CIVID19 data acquisition, wrangling, estimation, analysis and visualisation

RStats is a great tool for data management and report generation in research. It provides a very flexible environment for versatile data analysis. As indicated earlier, the work of Krispin and Byrnes (2020) provided some insights and interest in giving evidenced-based information on the growing pandemic. Therefore, their Coronavirus R package was the foundation upon which all the works reported here were built. At the end, it turned out that the package could not provide the data in a manner that would provide good visuals. The daily update based on the package commenced in April 14, 2020 with the single data stream but was abandoned as the data had to be scrapped directly from NCDC website daily, which has continued till date. The data collected were wrangled using Tidyverse ecosystem in RStat (Wickham et al. 2019). It involved filtering, grouping, re-arrangement and all processes that would produce the desired outputs and outcomes, like the visuals. The story centred confirmed cases, the analysis of deaths and those who recovered would be provided at a later date.

After it was discovered that the data from the package were not adequate, the manual compilation of the disaggregated data from February 29 when the first case was recorded to catch up with time (i.e. up to April 14 and beyond) took over a week because it had to be compiled from the records of NCDC (2020). One of the issues that arose from the data being published by NCDC (2020) was the skewed nature, which did not produce smooth curves as earlier indicated. But daily update was provided for the whole country while the lockdown persist.

As the situation worsened globally, most Nigerians were pessimistic about the claim of its existence showing nonchalant attitude to prevention guidelines being issued by the Presidential Task Force. Some totally dis-believed the existence while others believe in the superiority of the sovereign to cure them should they contract it. Some even started branding the pandemic in various Nigerian market language parlance, one of which was rich man disease. Of course, there was some period of inactivity, as if the government lacked direction as to what could be done.

As things worsened, there was need to even provide greater details and possible pathways for the pandemic in Nigeria. So various visuals were produced to show what was happening across the country on per state basis (Fig. 5 - 10) and on zonal basis (Fig. 11 - 13). One of the new analysis added to the visuals (Fig. 8) today is particularly very interesting because it gives us idea of the states in Nigeria that can be pronounced COVID19 free! based on the 14 days moratorium upon which the virus can replicate.

As the data mature, time series analysis was conducted using ensemble technique and involving:

These estimations were done to provide robust models with a view to show if the curve is flattening or not (Fig. 18).

Then, based on the estimated models, pathways to equal length of recorded cases was forecast. As at today, Nigeria has been in the pandemic for 206 days and equal length forecast ends on April 17, 2021. Based on the forecasts (Fig. 19 - 20), the accumulated cases at the end of the forecast period would be as follows:

Whatever the case, there is still work to done. For example, in the last three weeks, the confirmed cases had risen marginally which is giving rise to some of the positive forecasts recorded today. Government must be proactive when situations that could spike the number up arises. For instance, the burial ceremony in Zaria on Sunday might lead to some spike, even so that, Kaduna is the third in terms of recorded cases among the states in Nigeria, including the Federal Capital Territory (FCT).

Lastly, the possibility that either COVID19 is impacting on inflation rate and price of petroleum motor spirit (PMS) was explored. From the finding so far, it appears the COVID19 pandemic has caused a 1.1% rise in inflation. The various visuals are presented in Figs. 15 - 17.

Based on the experiences in trying to tell the story of COVID19 in Nigeria, some issues need to be examined. The first and most important is the need to operate NCDC as a research institution rather than a coordinating one. In particular, the Ministry of Health should transfer the institution to a Teaching Hospital for effective work and easy sail.

In addition, the capacity of our health facilities needs to be expanded very significantly. A situation whereby, Nigeria have tested less than 500,000 (484,051 exactly, that is 2338 tests per day) cases in 206 days when UK started testing 125,000 per day since April and recently, the US has reached one million tests per day has shown the need for the expansion. With a population of about 200m, it means it would take 85,543 years to test all the population for the virus!

Lastly, the welfare of the front-line staff in this type of pandemic is paramount. The various Unions under the health sector have embarked upon strike at least twice even in a serious emergency as COVID19 pandemic. Let nothing be left to chance!

That is my COVID19 story about Nigeria. From now on, updates of cases would be provided but the story remains same.

Fig. 1a Daily observed cases of COVID-19 in Nigeria

Fig. 2 Perecentages and previous days differences of COVID-19 in Nigeria Starting from February 29, 2020 to September 22, 2020

Fig. 3 Cases recorded in percentages Starting from February 29, 2020 to September 22, 2020 (legend as Fig. 2)

Fig. 4 Cumulative cases and Forecast of COVID-19 cases in Nigeria Starting from September 23, 2020 to April 17, 2021

Fig. 5 Number of days since average recorded cases exceeded one in each State

Fig. 6 Daily confirmed Coronavirus cases, by number of the days since 1 first daily case recorded

Fig. 7 Number of days from February 29, 2020 to September 22, 2020 that cases were recorded in the States

Fig. 8 Number of days the last COVID19 case was recorded as at September 22, 2020

Fig. 9 Diverging Bars of COVID-19 cases in the States (normalised)

Fig. 10 Monthly summary of COVID19 cases in the States

Fig. 11 Daily recorded cases of COVID19 in the States

Fig. 12 Number of days since average recorded cases exceeded one in each geo-political zone

Fig. 13 Number of days from February 29, 2020 to September 22, 2020 that cases were recorded in the Zones

Fig. 14 Monthly summary of COVID19 cases in the Zones

Fig. 15 Relationship between COVID-19, inflation and PMS price in Nigeria

Fig. 16 Relationship between PMS price, inflation and COVID-19 in Nigeria

Fig. 17 Relationship between Inflation, PMS price and COVID-19 in Nigeria

Fig. 18 Models of COVID-19 Cases using ensemble technology

Fig. 19 Equal duration forecast of COVID-19 Cases from the ensemble models using a native plotting method

Fig. 20 Forecast of COVID-19 cases in Nigeria Starting from September 23, 2020 to April 17, 2021 using a more advaced plotting method

References

Krispin, Rami, and Jarrett Byrnes. 2020. Coronavirus: The 2019 Novel Coronavirus Covid-19 (2019-nCoV) Dataset. https://CRAN.R-project.org/package=coronavirus.

NCDC. 2020. “NCDC Coronavirus Covid-19 Microsite.” Abuja, Nigeria: The Nigerian Centre for Disease Control. 2020. https://covid19.ncdc.gov.ng/.

R Core Team. 2020. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.

Wickham, Hadley. 2016. Ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. https://ggplot2.tidyverse.org.

Wickham, Hadley, Mara Averick, Jennifer Bryan, Winston Chang, and Lucy D’Agostino McGowan et. al. 2019. “Welcome to the tidyverse.” Journal of Open Source Software 4 (43): 1686. https://doi.org/10.21105/joss.01686.