Everyone Counts – zero draft

This document has been generated from a markdown template in a github repo - the repo is private because it exposes some NS financial data, but you can ask me for access.

There are two versions of this document, with/without widgets which allow you to preview the R code. This can facilitate peer review.

You can make comments using the sidebar.

Later, we can export the output to Word.

Preparation

This section only describes the data preparation and is not part of the final document. You can skip it. A few countries still unmatched across databases #TODO note Swaziland/eswatini

Load packages

Define functions

Load data - timeseries daily/weekly

OWD (Our world in data)

This is no longer necessary for covid mortality as we have the economist data. #TODO maybe it can be removed

Read from github

Exclude the continental summaries

Transform dates

Caclulate months

How many countries total

[1] 225

Economist excess deaths data

There are datasets for both daily estimates and cumulative totals. Note that the sum of the daily estimates is not the same as the separate cumulative database because they have to be multiplied by 7

Check economist data

Create daily database

Load data - oneoffs

WB income and PP data

Risk / global-health-security

Risk / Inform

Load more data - timeseries daily/weekly

Covid indicator tracking data

Convert all data including binary to numeric

Coping with changed definitions??!!

still need to check ambulance and HF data in attachments folder #TODO

for example “OP1. Pillar 3 Community-based surveillance (CBS): is now a yes/no question and focuses on whether the National Society has volunteers actively reporting on health risks related to COVID-19 using community-based surveillance (CBS).” if it is now yes/no, then a transformation should be applied to previous datapoints to make them also yes/no.

These are probably mostly robust re ranking.

Load data - FDRS

Clean one datapoint and create ci_tracking

[29/11/2021 12:39] Parima Davachi and we are not very strict. for example in 20-month data collection, the end date must have been 31.09 but a lot of NSs put dates after Sept as well like 05.10 or dates in OCtober

This means that the Quarter dates can be roughly understood as the beginning of the quarter.

Calculate new Priorities: simply separating vaccinations as an extra priority

Calculate latest Ranks

Create clusters of Pillars? This is probably not useful.

Calculate total NSs per indicator

Dealing with changes in indicator definitions

This way of calculating a performance score for each indicator for each NS for each quarter is robust to changes in definitions (because it uses ranks) except possibly in the transition quarters where the definition changed if one NS reports using one definition and the other using the other definition.

And similarly the mean and median ranks per pillar and priority are also robust except in the same edge cases.

I guess the amount of noise introduced by those possible edge cases is tiny. But I should check.

Calculate quarters, increments and ranks

Fill in unreported data values where there was a previously reported value, except for point in time indicators

Oxford policy data

Load up, select only first day of each month

Load data - fdrs

Finance data

Check finance data

Merge some datasets

Add regions to economist database

Calculate economist latest

Create alldata database

• for some analyses

Merge indicators with oneoffs, risk data, calculate proportional data (as % of population) for all indicator types except Binary and % (note this is arguable e.g. for number of branches / funding streams).

Create alldata latest

Load preparedness data

Checks

Check every datum is of one type:

# A tibble: 0 x 2
# … with 2 variables: type <chr>, indicator_text <chr>

Check no datum is of more than one type:

Other checks #TODO

There may still be some small issues with the data e.g. matching all countries (ISO codes) in different datasets.

Chapter -1: Front matter

Everyone Counts: Outbreak

January-February 2022: Two years into the COVID-19 pandemic –

Abbreviations

RCCE = risk communication and community engagement

CBHFA Community-Based Health and First Aid

CHF Swiss Francs

ECR “Everyone Counts” Report

FDRS Federation-wide Databank and Reporting System

ICRC International Committee of the Red Cross

IFRC International Federation of Red Cross and Red Crescent Societies

Foreword

Signed by IFRC President and/or SG.

Introduction

Introduction to the report (max. 2-3 pages)¹

This Everyone Counts report (ECR) is a flagship publication of the International Federation of Red Cross and Red Crescent Societies. It is based mainly on …

• About the report, explaining its focus, structure, and caveats, including a few brief sentences on the main issues and challenges of health-related data collection (timeframe risks, patchiness of data, difficulty putting it all together, etc).

• If required, a quick explanation of any challenges, puzzles, quizzes, cartoons, etc. (i.e. ‘how to read’ this report)

Chapter 0. Introduction

Overview of what we can learn from the data about how the COVID-19 pandemic affected Red Cross and Red Crescent Societies: describing how lockdowns did not stop humanitarian frontline workers responding to existing crises, how and why the operational priorities and response framework in general were developed (description of priorities, pillars, the rationale behind them, etc.).

An introductory paragraph on the COVID-19 pandemic (i.e. when was the first confirmed case, how did spread, its global health and socio-economic impact, etc.).

Key questions

Generate curiosity about some key questions which we hope to answer with the data.

• How did people and National Societies in different countries experience the pandemic? How did they respond?

• How do differences in country contexts (in particular, lower income countries (LICs) versus higher income countries (HICs) affect the pandemic and its response?

• How much can we rely on the official Covid death statistics: are they underestimates? Are they more likely to be underestimates in LICs?

• LIC populations are younger so is the likelihood of dying given an infection much lower? Do death statistics underestimate the true scale of the pandemic in terms of infections? What does this mean for longer term challenges like “Long Covid” which are likely to affect NS work in the future?

• What about direct additional deaths because of worse preparation / worse general national mitigation / worse covid health care and worse restrictions to covid health care and slower vaccination rate? Are these more frequent in LICs?

• What about indirect additional deaths because of worse restrictions to other health care eg HIV and because of worse indirect economic effects? Are these more frequent in LICs?

• How did NSs respond? Did those in LLCs have less money to spend and can therefore not perform as well? Or are NSs in LLCs more likely to be well-prepared, and can therefore perform better?

Chapter 1. Context: The pandemic from the perspective of the NSs

Understanding the impact of the pandemic

Which countries (and NSs) did the epidemic hit the hardest?

Which regions were hit the hardest?

There are different ways to understand and answer these questions. The difference between the possible approaches is, from the perspective of 192 NSs in 192 countries across the world, so fundamental that we need to address it head-on.

Which region was affected the worst?

This seems like a simple question with a simple answer. Which region was affected the worst? Looking at the figure: The Americas.

Total confirmed deaths: comparing regions

Median confirmed deaths per country: comparing regions

But wait! Perhaps the total deaths per region is a misleading statistic. Some regions have more countries than others. Maybe it makes more sense to look at the average or typical country?

Figure 1: Mean rate per 100K per country, by region

Now the worst hit region seems to have been Europe and Central Asia, with MENA second. The typical country in those regions had the largest number of deaths per 100K.

Median deaths as a proportion of population per country: comparing regions

But some countries are of course more populated than others. The typical country in MENA is highly populated. Shouldn’t we look at the number of deaths as a proportion of the population?

Once again the story is completely different. Now Asia Pacific appears to have fared much better than MENA.

Figure 2: Mean rate per 100K per country, by region

Sometimes looking in more detail at the individual countries can be useful, for example with a violin plot

Figure 3: Mean rate per 100K per country, by region

The violin plot shows how in Africa and Asia Pacific, there a lot of countries with almost zero reported deaths.

Estimated excess deaths

But there is a problem. Most data on deaths, cases and hospitalisations comes from official sources in each country. But this data can be inaccurate. Data on deaths due to Covid depends on

• differing definitions of what death due to Covid actually means
• the ability of health institutions to report this:
  • lack of testing
  • uncertainty about people who die at home

Many reports in the global media nevertheless depend on them. But if we want to get an accurate picture across the Movement and across the globe, we have to be very careful with data which might be systematically treating countries differently.

What if the pandemic was actually much worse in places where health statistics are less reliable than these official statistics lead us to believe?

One way to avoid some of these problems is to look at excess deaths in the pandemic period: how many more people died than we might expect for the same period, given what we otherwise know about deaths in the same months in previous years?

This method has the big advantage that it also takes into account deaths which might be indirectly due to the pandemic.

Some countries report official figures for excess deaths. But these tend to be higher-income countries. There are nowhere near enough of them from lower-income countries, so they don’t help us answer the question.

The tragedy/paradox is that countries with weak health systems may appear to have escaped lightly, merely because the published mortality and case numbers are more likely to be underestimates. Basically, the poorer the country, the more likely it is to have both more covid mortality but also a less-functioning health statistics system which actually reports fewer deaths officially.

Several teams have used models have tried to estimate excess deaths (see e.g. https://www.economist.com/graphic-detail/coronavirus-excess-deaths-estimates) to geive a more accurate, estimated picture.

All told we collected data on 121 indicators for more than 200 countries and territories. We next trained a machine-learning model which used a process called gradient boosting to find relationships between these indicators and data on excess deaths in places where they were available. The finished model used those relationships to provide estimates of excess deaths in times and places for which there were no data available. A description of our methodology and the ways in which we tested it, as well as links to replication code and data, are here.

These graphics use estimated excess deaths from Economist modelling. Final estimates use governments’ official excess-death numbers whenever and wherever they are available, and the model’s estimates in all other cases. #CHECK

These models have wide confidence intervals!! #TODO

A few countries such as Australia have negative excess deaths for some periods of the pandemic, presumably reflecting the fact that public health measures such as social distancing were preventing other deaths, for example due to the flu virus.

There are alternative ways to measure the impact of the COVID-19 pandemic, for example ²:

• confirmed cases (periodic / cumulative ; raw numbers or relative to population (100,000 or million),

• officially reported deaths (see cases)

• hospitalization rate with COVID (No. of people in ICU, No. of people in hospital with (confirmed, suspected COVID-19).

All of these are sensitive to case definitions, which is influenced by the strategies of testing.

The IFRC has a duty to analyse data in a way which as far as possible makes equal sense for all countries and all NSs. Adopting only officially reported deaths means using data which is likely to treat countries in different regions and income groups very differently.

In this report, we have chosen to focus on estimated excess deaths as a measure of the extent and course of the pandemic.

Median confirmed deaths versus median estimated excess deaths as a proportion of population per country: comparing regions

Figure 4: Mean rate per 100K per country, by region

The difference between confirmed and estimated actual deaths is horrifying. Excess deaths due to Covid are estimated to be around four times as high as official figures. Around 14 million more people are estimated to have died, most of these in lower-income countries.

Total confirmed deaths versus total estimated excess deaths

Figure 5: Mean rate per 100K per country, by region

Mean versus median of estimated excess deaths as a proportion of population per country: comparing regions

Using estimated excess deaths does not solve all our problems in trying to understand the answer to basic questions like “which region was affected the most”? Although we would often expect the mean to be higher than the median, the difference for Asia-Pacific is particularly large.

This difference is mainly due to India, Russia, Indonesia, Pakistan and Bangladesh in the same region (the “Asian Big 5”) all have estimated excess death tolls over half a million people. The very high rates for these countries make an enormous difference to the mean for the region but affect the median rates hardly at all.

Examples from India, Pakistan??

Equal-area maps of reported and estimated excess deaths

These maps should have continent outlines #TODO

How many more estimated excess deaths are there, compared to reported deaths, by income group?

By income group

So looking at confirmed deaths, countries in the top two income groups report figures well over ten times as high as those in the lower two income groups. But when we look at estimated excess deaths, we see that the figures for the lower income countries are roughly equal to the reported deaths for the higher income countries. We also see that the estimated excess deaths for upper middle income countries are much higher than anywhere else.

Comparing lowest with highest income group with a baseline

This is based on the median values as a proportion of population.

This is based on the median values as a proportion of population.

Demographics: how severe was the pandemic in different regions, allowing for demographics?

Mortality vs infections …

LIC populations are younger so the likelihood of dying given an infection is much lower (up to 11 times): Age-adjusted comparisons³

We can also think of it this way: if a country has a younger population, that is a protective factor against deaths from Covid-19. But we don’t know how much it protects against other consequences of Covid-19, like “Long Covid”.

Comparing lowest with highest income group with a baseline

This is based on the median values as a proportion of population.

Comparing regions with a baseline

This is again based on the median values as a proportion of population.

Discussion: What does this mean for long covid rates and for NS activities going forward?

Figure 6: Mean rate per 100K per country, by region

Secondary effects e.g. unemployment #TODO

Excess deaths over this time period (whether official or estimated) will certainly include deaths which are not due to Covid infections but which are indirect effects of the pandemic (e.g. non availability of medical personnel) and attempts to mitigate it (e.g. the closure of clinics), as well as through devastating effects on economies due to reductions of exports, tourism and others. These effects are likely to be stronger in lower-income countries.

“According to a Global Fund survey⁴ of 32 countries in Africa and Asia, prenatal care visits dropped by two-thirds between April and September 2020; consultations for children under five dropped by three-quarters.”

Conclusion

In spite of the very large underestimation, nevertheless Africa on the one hand, and lower-income countries on the other hand, have not been as hard hit…

Time series: How did the pandemic develop over time and how does this differ by region and income? What does it mean for NSs?

Confirmed vs estimated time series total

The graphs for the development of the pandemic repeat the story we saw above:

• the number of estimated excess deaths is many times bigger than the number of confirmed deaths.
• this effect is weaker for ECA and the Americas and stronger for the other regions.

Total estimated excess deaths by region over time

The main story here is that Asia-Pacific has an increasing share of the overall death toll, mainly due to the “Asian Big 5” countries mentioned above.

The main story here is that the median Asia-Pacific country actually had negative estimated excess deaths until the middle of 2021.

Total estimated excess deaths by income group over time

The proportion of estimated excess deaths taken by the high-income countries drops continually and is replaced mostly by the lower middle-income countries.

Vaccinations are an important part of development of the pandemic: see separate chapter below

Phases: The dynamic of the pandemic across the world.

Overall and median scores over time do not capture the tragic dynamic of this pandemic. For most countries, the pandemic came in dramatic waves.

Example: one NS coping with a rapid and extreme peak

Figure 1. Graphic: Cumulative curve of reported deaths & excess deaths per region over time

Graph of pandemic over time

Need to show number of countries this is based on for each region #TODO

Summary:

• The paler lines in these graphics are the individual countries: look how often rates in individual countries exploded and receded.

• The reported/estimated distinction does not fundamentally change the shape of the curves.

• Looking at the blue lines, the shape of the pandemic seems to be arguably similar in most regions (two peaks, with an intervening one in Africa) but quite different in Europe.

• Many countries experienced at least one extreme and sudden peak.

• The twin-peak pattern which is clear for estimated excess deaths in Europe and Central Asia is not as pronounced or not even present in the other regions.

Did almost every country experience an extreme peak?

Figure 7: Figure 1. Did almost every country have an extreme peak: % which experienced at least one week with a peak rate of estimated deaths per 10K which was at least 3* the median rate for the whole pandemic?

Summary:

• Extreme peaks were less common in MENA and Africa. In Asia Pacific, not only were median rates very low but only about half of the countries experienced an extreme peak.

Relative severity

This is interesting and relevant to NSs in terms of what it means for their activities and prioritisation. However, see relativisation above re official and estimated Covid-19 mortality: Covid-19 is much worse in LICs than is apparent here. Not sure I will have time for this #TODO

Relative severity of COVID-19 mortality indicator.⁵ This helps put Covid-19 deaths into a perspective relative to more familiar causes of death. This makes Covid-19 seem relatively unimportant in lower income countries: just one of the challenges which NSs have to face.

Example: link this to actual NS / public health activities in different countries?

Policy responses

Policy over time

The stringency index⁶ systematically collects information on several different common policy responses that governments have taken to respond to the pandemic on 20 indicators:

• 8 policy indicators (containment and closure policies, School and workplace closures, Cancellation of Public Events and Gatherings, Stay at home, Face covering, Public Information Campaigns),

• 4 economic policies indicators (income support to citizens, provision of foreign aid,

• 8 health system policies indicators (COVID-19 testing regime, emergency investments into healthcare and most recently, vaccination policies).

The data from these indicators is aggregated into a set of four common indices, reporting a number between 1 and 100 to reflect the level of government action on the topics in question:

• an overall government response index (which records how the response of governments has varied over all indicators in the database, becoming stronger or weaker over the course of the outbreak);

• a containment and health index (which combines ‘lockdown’ restrictions and closures with measures such as testing policy and contact tracing, short term investment in healthcare, as well investments in vaccine)

• an economic support index (which records measures such as income support and debt relief)

Policy indices

Figure 8: Government policies over time: the first four months

Standout findings:

• The pattern of government responses seems to have been broadly very similar across the regions whereas the different developments on other indicators vary more across regions.
• Economic support was much weaker and more varied across all regions.
• It might have seemed slow at the time, but most governments responded quickly, with 90% of countries reaching at least 90% of their final highest level by xx. #TODO.

Individual policies over time

Maybe do not need this as well as the one above?

Figure 9: Government policies over time

Summary:

• Several responses were weakest in Africa, especially economic support and protection of elderly people
• Vaccination policy starts at the start of 2021 and increases in a linear way after that across all regions.
• Face covering mandates dropped off in Europe and Central Asia during 2021.

Policy: economic support

The biggest differences were on economic support.

Figure 10: Government policies over time

• There was an overwhelming lack of economic support in Africa

Policy: just look at the successes in small Asian countries / connection with mask wearing etc? #TODO

Chapter 2: The IFRC network’s COVID-19 response

About the tracking tool

Introduction on how the collection of data is organised and also include an explanation of the story behind, how it was built, with which purpose, difficulties, the complexity for NS to collect these new data in a crisis situation… A day in Parima’s life? A day in the life of one of her counterparts in an NS?

The elements of the response (indicators, pillars)

IFRC’s Covid-19 tracking is built around 44⁷ indicators which are grouped into 23 pillars which are grouped in turn into three priorities.

For the purposes of this report, the indicators concerning vaccination have been moved from the Health priority to another priority just called “Health - vaccinations” because these are quite different: they started much later and tend to involve different groups of National Societies. See chapter xx.

priority	new_priority	pillar number	pillar	type	indicator	indicator_text	N_NSs	N_time_points	first_time_point	last_time_point
Health	Health	1	Epidemic control measures	People other	testing	people tested by NS to diagnose COVID-19	69	56	2020-06-08	2021-10-27
Health	Health	1	Epidemic control measures	Staff and volunteers	screening	staff and volunteers supporting screening	79	53	2020-05-31	2021-12-31
Health	Health	1	Epidemic control measures	People other	contacts	contacts identified and/or followed disaggregated by age/sex	71	57	2020-06-30	2021-10-27
Health	Health	1	Epidemic control measures	People other	cases	cases in cohort/home isolation receiving material support from NS	104	80	2020-06-30	2021-12-31
Health	Health	2	Risk Communication, community engagement, and health and hygiene promotion	People reached	rcce	people reached through risk communication and community engagement for health and hygiene promotion activities	174	117	2020-05-31	2021-12-31
Health	Health	3	Community-based surveillance (CBS)	Staff and volunteers	cbs	staff and volunteers using CBS to report on COVID-19 signs and symptoms	78	49	2020-05-31	2021-12-31
Health	Health	3	Community-based surveillance (CBS)	Binary	cbs_binary	NS conducting community-based surveillance for COVID-19 signs and symptoms	30	6	2021-09-30	2021-10-27
Health	Health	4	Infection prevention and control and WASH (health facilities)	Health facilities	washhf	health facilities supported	98	70	2020-05-31	2021-12-31
Health	Health	5	Infection prevention and control and WASH (community)	People reached	washcom	people supported through community WASH activities	129	79	2020-05-31	2021-12-31
Health	Health	6	Mental Health and psychosocial support services (MHPSS)	People reached	mhpss	people reached with MHPSS services	150	110	2020-05-31	2021-12-31
Health	Health	7	Isolation and clinical case management for COVID-19 cases	Health facilities	hf_covid	health facilities treating COVID-19 supported	43	31	2020-06-30	2021-06-25
Health	Health	7	Isolation and clinical case management for COVID-19 cases	Health facilities	hf	# of health facilities supported with IPC, WASH or other interventions to improve COVID prevention, detection or treatment	60	12	2021-07-31	2021-10-26
Health	Health	8	Ambulance services for COVID-19 cases	People other	emt	COVID-19 cases (confirmed or suspected) who received ambulance transport	63	50	2020-06-08	2021-12-31
Health	Health	8	Ambulance services for COVID-19 cases	Binary	emt_binary	NS providing ambulance services to COVID-19 patients	43	9	2021-07-31	2021-10-25
Health	Health	9	Maintain access to essential health services (community health)	People reached	mh_com	people reached with essential community health services	75	55	2020-06-08	2021-10-31
Health	Health	10	Maintain access to essential health services (health facilities)	Health facilities	mh_hf	NS supported HFs maintaining services to pre-covid levels	61	46	2020-06-30	2021-12-31
Health	Health	12	Management of the dead	Other	burial	community burials of suspected or confirmed COVID-19 cases facilitated or directly carried out by NS volunteers	29	24	2020-07-07	2021-06-17
Health	Health - vaccinations	11	Support for Immunization (COVID-19 vaccine introduction)	Staff and volunteers	vax_covid	staff and volunteers trained on COVID-19 vaccine introduction	81	30	2020-12-31	2021-10-27
Health	Health - vaccinations	11	Support for Immunization (vaccine hesitancy)	People reached	vax_rcce	people reached through risk communication and community engagement towards addressing vaccine hesitancy	63	25	2021-01-17	2021-10-27
Health	Health - vaccinations	11	Support for Immunization (hard-to-reach persons)	People reached	vax_hardtoreach	hard-to-reach persons vaccinated	14	6	2020-12-31	2021-02-17
Health	Health - vaccinations	11	Support for immunization	People reached	vax_covid_rollout	individuals NS has supported to get vaccinated against COVID-19	72	22	2021-04-30	2021-10-15
Health	Health - vaccinations	11	Support for Immunization (routine immunization and supplimentary immunization activites)	Staff and volunteers	vax_routine_supp	staff and volunteers supporting routine immunization and supplementary immunization activities	17	7	2020-12-31	2021-02-21
Health	Health - vaccinations	11	Support for immunization	People reached	vax_routine	children under 24 months of age that were supported by NS to receive routine immunization	26	11	2021-05-31	2021-10-27
Health	Health - vaccinations	11	Support for immunization	People reached	vax_supp	children under 5 years of age that were supported by NS to receive vaccines through SIAs/campaigns	24	10	2021-04-30	2021-10-27
NS Strengthening	NS Strengthening	1	NS Readiness	People reached	drr	people reached through pandemic-proof community preparedness, response and DRR measures	175	92	2020-05-31	2021-12-31
NS Strengthening	NS Strengthening	1	NS Readiness	Binary	nsar	The role and activities of the NS are expressly included in the national government’s main plan(s) for COVID response/recovery	177	108	2020-05-31	2021-12-31
NS Strengthening	NS Strengthening	1	NS Readiness	Binary	nsr	The NS has developed contingency plans for COVID-19 response and other concomitant emergencies	176	107	2020-05-31	2021-12-31
NS Strengthening	NS Strengthening	2	NS Sustainability	Percent	nss_funding	of core organisational budget that is funded	136	78	2020-05-31	2021-12-31
NS Strengthening	NS Strengthening	2	NS Sustainability	Other	nss_is	new streams for unrestricted income	141	70	2020-07-20	2021-10-31
NS Strengthening	NS Strengthening	2	NS Sustainability	Binary	nss_fr	Unrestricted financial reserves for more than 3 months	172	80	2020-05-31	2021-10-31
NS Strengthening	NS Strengthening	2	NS Sustainability	Binary	nss_bcp	The NS has adapted its business continuity plan (BCP) for COVID-19 or developed a new one	175	97	2020-06-30	2021-10-31
NS Strengthening	NS Strengthening	3	Support to volunteers	Binary	vol_ppe	volunteers have access to the Personal Protection Equipment (PPE) necessary to safely fulfil their duty	170	58	2020-12-30	2021-10-31
NS Strengthening	NS Strengthening	3	Support to volunteers	Binary	vol_coverage	volunteers are provided with insurance that covers accidents, illness or death benefits to their families, including private, organizational (e.g. solidarity funds) or public coverage from authorities.	170	58	2020-12-30	2021-10-31
Socio-Economic	Socio-Economic	1	Livelihoods and household economic security	People reached	cash	people reached with conditional and unconditional cash and voucher assistance	94	76	2020-05-31	2021-12-31
Socio-Economic	Socio-Economic	1	Livelihoods and household economic security	People reached	inkind	people reached with food and other in-kind assistance	136	98	2020-05-31	2021-12-31
Socio-Economic	Socio-Economic	1	Livelihoods and household economic security	People reached	skills	people supported with skills development for livelihoods/economic activities	40	35	2020-07-28	2021-10-27
Socio-Economic	Socio-Economic	2	Shelter and urban settlements	People reached	shelter_covid	people reached with safe and adequate shelter and settlements under the circumstances of COVID-19	47	44	2020-07-07	2021-10-15
Socio-Economic	Socio-Economic	3	Community engagement and accountability, including community feedback mechanisms	Other	f_reports	community feedback reports produced	66	43	2020-06-30	2021-12-31
Socio-Economic	Socio-Economic	3	Community engagement and accountability, including community feedback mechanisms	Other	f_records	community feedback comments collected	94	64	2020-06-30	2021-12-31
Socio-Economic	Socio-Economic	3	Community engagement and accountability, including community feedback mechanisms	Staff and volunteers	ce_train	NS staff and volunteers trained on community engagement and accountability	124	85	2020-06-30	2021-12-31
Socio-Economic	Socio-Economic	4	Social care, cohesion and support to vulnerable groups	Other	mg_branch	branches who include an analysis of the specific needs of marginalised groups in their assessments	72	50	2020-06-30	2021-12-31
Socio-Economic	Socio-Economic	4	Social care, cohesion and support to vulnerable groups	People reached	exclusion	people reached by programmes addressing exclusion	58	56	2020-05-31	2021-12-31
Socio-Economic	Socio-Economic	4	Social care, cohesion and support to vulnerable groups	People reached	violence	people reached by programmes addressing violence	35	23	2020-06-30	2021-12-31
Socio-Economic	Socio-Economic	4	Social care, cohesion and support to vulnerable groups	People reached	education	people reached by programmes addressing education-related needs	37	36	2020-07-31	2021-10-27

Example: some examples of what these activities look like in practice.

Numbers of reported indicators per NS

As time went on, National Societies were able to implement more and more different kinds of activity.

Numbers for each indicator

could also add totals for each indicator #TODO

why aren’t medians quite the same? #TODO

This is robust to point in time indicator status because it takes the last one anyway

[[1]]

[[2]]

[[3]]

That is too much data! What counts? How shall we aggregate?

How can we combine this information to get a higher-level picture of how the National Societies responded?

Relative performance: how we created the scores

What we did is this:

• each indicator is scored in a different way, for example it might count the number of people reached or just whether an NS had adopted a certain policy or not.
  • Converted binary indicators to numbers (0/1)
  • Divided indicators which count people by the total population, in order to be fair also to smaller countries
  • For each indicator and for each NS, we derived a score for each quarter (3 months)
    • If the NS had more than one score in that quarter, we assigned the highest of the scores #TODO maybe should be the last score?
    • If the NS had not reported in that quarter but had reported in a previous quarter, we took the score from the previous quarter
  • Then for each indicator, we ranked all these scores for each quarter for each NS and derived a new standardised rank, giving 100 to the top score and 0 to the lowest score

This is a way to square the circle, to allow combinations of indicators of different types into one score, per pillar or priority.

A score of 0 on a given indicator in a given quarter means that the NS reported the lowest score, and a score of 100 means the NS achieved the highest ever score over the whole response (relative to population, where relevant) of any NS on that indicator. A missing score means the National Society did not report that indicator at all.

These scores have these properties:

• They can easily be averaged e.g. to combine several indicators within one pillar

• They can easily be totalled e.g. to sum several indicators within one pillar

• They can be used to display change over time (because the ranking is based on all-time scores, not quarterly scores)

The DNA of the response: Every NS had a different response

This graphic shows the scores for each National Societies. Each column is a national Society and each row is an indicator.

The grey squares show where an National Society did not report on that indicator at all.

Every National Society carried out a different set of activities and scored highly on at least some indicators.

[find a better way of showing the priorities than these dots xx]

Performance by Priority over time

We then combined these scores for each indicator into overall scores for each priority, by adding the performance scores for each indicator within a priority, and then rescaling these from 0 to 100 to make the priority-level scores comparable.

• the numbers on the vertical axis reflect the total performance for each priority, with 0 being the lowest and 100 the highest total performance (relative to population size) in each priority.

Summary:

• Overall, the performance across all regions was relatively similar, with all regions performing equally well overall.
• Performance in Health was particularly high in Africa and MENA, and weaker in Europe.
• Performance in NS Strengthening was particularly high in Asia Pacific.
• Performance in NS Strengthening improved strongly towards the end of 2020 in most regions.
• Performance in NS Strengthening accelerated quickly but has started to drop off again a little in most regions.
• Performance in NS Strengthening was the most evenly spread across National Societies, with most scoring around 50.
• In vaccinations on the other hand, there were a few National Societies (mostly in Europe) which scored very highly across all the constituent indicators with the majority scoring well only on one or two of the indicators. These high-scoring NSs tended also to start very quickly to reach high scores early in 2021.
• Performance in Health started quite quickly and has improved steadily.

Relative performance - latest scores by income group

Summary

• In Health⁸, overall performance in the two lower-income groups was above the two higher-income groups.
• Performance was quite high across a wide range of National Societies in both Health and NS Strenghtening.
• In vaccinations, many National Societies are still scoring zero, especially in low-income countries.
• …

The actual activities: descriptions, photos …

Look at each indicator, pillar by pillar – look at how they developed over time but also with some illustration of what they mean in practice.

[It is perhaps not good to use maps to report performance scores as it might be seen as highlighting underperformers? xx - so I hid these maps]

What does the response depend on?

Did the NSs’ response reflect / complement their government’s response?

Is there a connection between National Societies offering social support and countries’ policies?

[Analyses (not shown here) suggest not. xx could look into this more]

Examples xx from National Societies which offered social support even though this was not offered much by their governments:

# A tibble: 11 x 2
   ISO3  country                
   <chr> <chr>                  
 1 BGD   Bangladesh             
 2 BLR   Belarus                
 3 CMR   Cameroon               
 4 COG   Congo, Rep.            
 5 IND   India                  
 6 LBN   Lebanon                
 7 LKA   Sri Lanka              
 8 LSO   Lesotho                
 9 NIC   Nicaragua              
10 PSE   Palestinian territories
11 SLE   Sierra Leone           

Graph rank against GDP and National Society income

NSs in lower-income countries manage to produce a response which is almost as good as those in much richer countries.

Performance is measured as the average or total rank (expressed from 1 to 100, with 100 meaning the best) of each NS’s performance, corrected for population size where appropriate, on each indicator.

It doesn’t make much difference whether you use gdp per capita or ppp or National Society income. In either case, the poorer countries do give better value for money - just significant at 0.05.

[there are some with zero income #TODO ]

NSs in poorer countries generally perform better than NSs in richer countries!!

Chapter 2: Unity or isolation in the face of global crisis: an open and collaborative network

NSs’ collaboration with government and local authorities

Potential themes and issues to present

• The importance of auxiliary role of NSs during the response – coordinating closely with National governments

• RCRC contribution to National Pandemic Preparedness Plans and Policies

• Partnerships with local authorities – the importance of working together with local actors

• Lessons learnt about working closely with authorities:

• "You have to sit at the table before the crisis in order to sit at the table during the crises" #TODO how measure this?

In this section we will look at this indicators under Operational Priority 3: The role and activities of the NS are expressly included in the national government’s main plan(s) for COVID-19 response/recovery

[#TODO why is this broken?]

NS Sustainability

Here we look at the three time-sensitive indicators under National Society Sustainability.

The main curve shows the % of National Societies even reporting the indicator. The lighter shape and the numbers shows the % of that percentage reporting “yes” (lower two indicators) or the mean of the core organisational budget (first indicator).

Summary:

• Percentages of National Societies who have achieved these three things are particularly high in the Americas and Asia Pacific.
• The first data points involve fewer National Societies as they worked towards these achievements.
• The last data point is still incomplete as National Societies complete their reporting.
• National Societies in Africa and MENA …

Qualitative approach

• Presentation of examples (story clips or case studies) of NSs where data (interviews with Reference Centers, regional and IFRC secretariate health team and/or Federation-wide COVID-19 data report) indicates strong partnerships with national and local authorities.

Suggested NSs to select stories from:

• Congolese NS: being part of the COVID-19 response committees at national and local levels

• Kenya NS continues to co-chair with the Ministry of Health (MoH) in the mental health and psychosocial support sub-committee for the national COVID-19 response

  • Uganda NS: Uganda’s National Community Engagement Strategy for COVID-19 Response via interview with Uganda’s representative

• Selected NS from Caribbean NSs of Grenada, Trinidad and Tobago, Dominica, Belize quickly building on Zika related partnership with Ministries

• Korean NS: strong relations to various Ministries

• NS of Tajikistan is member of the National COVID-19 Task Force, National Platform for Emergency Response and Coordination Council at the Ministry of Health and Social Protection.

Collaboration inside the network, between NSs and at federation level

Social network analysis of FDRS and Covid (Marcelo)

Present analyses of FDRS and covid networks separately and then together, asking: was the “covid network” different from the pre-existing FDRS network of cooperation? Was it similar, did it build on the FDRS network, did it extend it?

I don’t think it necessarily matters that there are fewer Covid NS/NS links. Is it possible there is any duplication? (I think not, because of sources of finance?). Is it the case that total expenditure of an NS = expenditure reported for FDRS + Income reported for Covid? Or is Covid expenditure always just a part of expenditure reported in FDRS?

if there are simply fewer links for Covid, that is fine. the biggest problem would be if NSs differ in how they report expenditure as Covid or FDRS for some non-random reason which we don’t know about.

Digital transformation / Netherlands RC

• This subchapter could be introduced with an analysis made by the 510 Initiative of support link approach based on the COVID-19 field reports (and other sources) (peer-to-peer, etc.)

• The updated version of NS support networks maps can be presented

• Possible changes in the peer-to-peer support networks compared to the previous report could be analysed to detect

• Potential changes in the size of networks

• New networks, new members in already established networks

Potential themes and issues to present

• The importance of peer-to-peer collaborations and support networks between NSs

• Network spirit and values

• Agility, flexibility and adaptation

• Federation level support to NSs

• Demand and needs assessment-based approach

• Knowledge generation and a network of knowledge sharing

• Capacity building efforts

• Necessity to offer a response to the epidemic on many levels (health, psychosocial, socioeconomic…), and how these are interlaced

This chapter could present some selected examples of tools and other type of support (training activities) with visuals and with strong emphasis on the users’ experience (a pool of further potential examples can be found in the Sources table at the end of the chapter).

1. As there are a lot of strong examples here a more systematic mechanism for selection of the tools to be highlighted should be made by IFRC.

2. As we would suggest to have some feedback from tool users, check with tool creators, mainly reference centers whether they already have these feedback.

3. For the selected tools, for which there is no yet available information coming from the tool users (feedback, evaluation, user survey, etc), ask the reference centers to identify some of the users which could be NSs staff and/or volunteers which could be interviewed about: what was the usefulness of the tool, how did they adapt to it, how it changed their activities, etc.

Here is a small selection of tools/actions which could be highlighted:

• Presentation of Epidemic Preparedness and Response Package – Training modules – Africa region potentially with feedbacks
• Digital story 14. Data Collection and COVID Response Tracking Dashboard – Nepal (Source: 95)
• Short presentation clip of the report "The Greatest Need was to be listened to: The importance of mental health and psychosocial support during COVID-19
• Presentation of case study peer-to-peer relations reformed during COVID-19 in the Caribbean region
• Presentation of COVID-19 Helpdesk and its use
• Description of community of practice work in the field of WASH
• Health innovation: the case of Kenya using drones to deliver hygiene messages
• Presentation of Health Help Desk – with data on use, feedbacks
• Presentation of Atlas Adaptation BPI tools for Microbusinesses & IGAs – Pilot Ecuador

(External?) International collaboration and partnerships

This subsection should be written by IFRC, see some propositions from us below but IFRC has much more knowledge and information on this part.

Potential themes and issues to presented

• Partnerships with international actors: IFRC, ICRC and UN, other international agencies and NGOs, private sector, international donors

• Did links increase or decrease? Did they change?

• International coordination mechanisms at regional and global levels

• The role of international partnerships: national planning, locally driven priorities, international support

• The importance of strengthening international partnerships, among others with WHO to mobilize power

• Implementation standards RedCross, WHO, UNICEF

• COVID-19 pandemic as an opportunity for RCRC to be more visible externally through their response

Outline of analytical approach

• Overview of external relations related to COVID-19 pandemic and presentation of selected examples in more detail

• Examples of external relations at regional level/NS level

• Presentation of thought pieces by the Secretary General on the importance of strengthening international partnerships

• Lessons learnt from COVID for international humanitarian organizations

• Presentation of media story to how RCRC during the COVID-19 pandemic is seen from outside the network “public image”

Potential stories and case studies:

• The joint call of RCRC and the UN Secretary-General for a people’s vaccine

• Description of the story of how the international collaboration with WHO/UNICEF and international NGOs develop a community health delivery package. With additional data collection focus could also be put on some feedback on the use of this delivery package could be presented.

• RCRC European Region news: Red Cross expands COVID-19 testing in seven countries with €35.5 million EU support

• Thought pieces by the Secretary General from the March Conference of Global Health Security on the importance of strengthening international partnerships and on the global health security agenda

• A new pandemic treaty needs to be powerful not only on paper, Geneva Solutions – thought piece by the Secretary General

Chapter 3: Preparedness and prior experience: how did they help in Covid time?

Preliminary correlations between key variables (using latest datapoints for timeseries data)

225 countries

Summary:

• It seems like gdp predicts deaths. But this is because poorer countries underestimate deaths.
• Similarly it seems like risk PREVENTS deaths. But this is because riskier countries underestimate deaths. Risk especially hazard / coping is a really good predictor of estimated excess deaths and of underestimation.
• NS performance is hard to predict in terms of national indicators. The rank scores are only related to one another.
• NS performance does improve a little with increased expenditure (per head of population), especially on socio-economic indicators. And while higher expenditure is related to GDP per capita, GDP per capita and similar metrics do not predict NS performance.

National preparedness: Risk profiles of countries; Context profiles of the National Societies

See section 1.3: There have been several very serious and recent (up to 2020) attempts to assess the emergency- and pandemic preparedness of different countries, based on modern models of risk, vulnerability, etc with a wealth of data combined with the opinion of panels of renowned experts. It is quite sensational that these risk assessments seem to have been strongly negatively correlated with better covid outcomes. What can we learn from this? What does it mean for NSs?

Graphic: inverse correlation of preparedness vs outcome?

[We also need to compare with estimated excess deaths, not just official mortality. #TODO]

Chart, bar chart, histogram Description automatically generated

How do National Societies’ preparedness / experience of prior epidemics / experience of community health influence their COVID-19 response? Keep digging, CP3, #TODO

Potential themes and issues to present

• Did National Societies with past epidemics experience respond in a different, better way (type of response, quickness of response)?

• How did (if ever) previous epidemics prepare National Societies for the COVID-19 crisis.

• Lessons learned from previous epidemics.

• This section could be introduced with a map presentation of the main epidemics RCRC National Societies had to face during the 10 or 20 last years and a short explanation of the main characteristics of these epidemics like medical characteristics (effects, Incubation Period, Infective Period…), Treatment, Transmission, Geographic Distribution, Frequency, Public Health Response…. Elements from Metabiota and their Epidemic tracker might be used for this introduction (see source number 15 in the following source table).

Scaling up: Graph rank against FDRS rank

Since the beginning of the Covid 19 crisis, any NS had to face new needs and to develop new types of activities they never performed before or to do those at a much larger scale.

This result is not surprising but it is still good to see it confirmed by the data: NSs which were performing well in 2019 (breadth and depth of operations) were able to provide most impact also during the pandemic (breadth and depth of operations).

More possible analyses

Did Covid-19 mean that some National Societies began some FDRS activities in 2020 which they were not carrying out in 2019? #TODO

Dependent variable:

Also speed of ramp-up.

Independent variables:

1 epidemic experience

Comparison of the two groups of National Societies, the one with and the one without past epidemic experience. A list of the epidemic to be taken in account (Ebola, Zika, Cholera, Dengue Fever, SARS-CoV, MERS-CoV, Measles etc.) should be submitted to the Health team and/or Reference centres and/or Regional offices, as the list of National Societies with experience in these epidemics.

2 Preparedness according to PER (number of phases completed, score, etc); also, NS had plan & collaboration with government at start of 2020? …

Unfortunately I can’t yet find any strong connection.

3 Experience of community health programming

• Did National Societies with prior community health programs respond to the COVID-19 pandemic in a different, better way (type of response, quickness of response) compared to others?

• How to engage communities more in prevention and preparedness?

• The importance of epidemic/pandemic preparedness at community level for global security.

• Lessons learned from COVID-19 for community and emergency preparedness and health.

• types of measures developed (Epidemic control measures, Infection prevention and control and WASH, Isolation and clinical case management, Risk Communication, community engagement, and health and hygiene promotion…).

Narrative case studies?

Experience with contract tracing: Shoeleather epidemiology vs apps

• in what extent past experience in epidemics had a positive or negative effect on COVID-19 response and how:

• experience of the NS and other partners,

• already prepared response plan,

• existing RCCE mechanisms, established supply chains for PPE,

• staff and volunteers trained

• experience in community health and community engagement,

• relations with authorities, community / population knowledge, acceptance of measures, recognition of the NS, trust…,

• what are the similarities and differences in the responses to different epidemics,

• which lessons could be shared (these last elements would also feed the last chapter).

Presentation of the Community Pandemic and Epidemic Preparedness (CP3) program and selected stories from NSs efforts potentially from Uganda, Cameroon, Kenya, The Democratic Republic of Congo, Sierra Leone, Guinea and Mali, for example:

• Kenya and preventing anthrax outbreak in Maasai communities

• Community based surveillance case study: Indonesia

Presentation of excerpt of Uganda’s NATIONAL COMMUNITY ENGAGEMENT STRATEGY FOR COVID-19 RESPONSE and potentially an interview NSs or other relevant representative explaining the experience with strategy implementation.

Presentation of summary of main issues from publications prepared by IFRC health staff/ICRC president/IFRC Secretary General

• Article relevant for Ebola and community engagement

• Article prepared by IFRC staff Community and health security

• Thought pieces from Six essential lessons for a pandemic response in humanitarian settings. Statement to UN Security Council Open Debate: Pandemics and Security. (2020)

Presentation of some main thoughts (post it style) from IFRC preparedness stories: How a local response can halt a global crisis - Article by JAGAN CHAPAGAIN on the role of volunteers, communities and local actions

Additional case studies may be prepared to highlight the role of previous community health programme experience and capacity:

• Presentation of results from perceptions and impact survey among Caribbean communities and how the results were used to respond to community needs in relation to COVID-19

• Presentation of the story of how the healthy aging programs became an even more valuable capacity in the COVID-19 response. Materials to be collected from IFRC health team, interviews from countries of Serbia, Montenegro, Sri Lanka NSs representatives (some contact information already in Source table).

• Presentation of some findings from the report Needs Assessment for Response and Recovery in Asia Pacific related to pandemic preparedness – some interesting idea related to pandemic preparedness sensitive programming in all sectors in communities.

• Presentation of Digital story 3. Taking community engagement to a new digital level

Chapter 4: On the front line: The Story of Volunteers / maybe social media too / Simon

Solferino Academy stories / Causal mapping

What is it like for volunteers at the start of a pandemic? How do they deal with the fear and the uncertainty? How much do they rely on pre-existing skills and networks? What drives them?

These stories were submitted by volunteers around the world in April and May 2020, near the start of the pandemic. xx already made an analysis of these stories.

The stories were translated into English by xx.

188 different people from 53 National Societies made at least one causal claim⁹

These stories were analysed using causal mapping. This is a kind of qualitative data analysis, but rather than just looking for general themes (like “fear” or “contacted National Society”), analysts are instructed to identify passages of text where people talk about how one thing causally influenced another.

For example, if a volunteer says …

Volunteering in this situation made me feel truly useful as I had never felt in any previous volunteering

… this could be coded as a link from ‘volunteering in Covid situation’ to ‘feeling truly useful’.

As the analysts continue with this work, they add to an ever-growing set of “causal factors” like ‘volunteering in Covid situation’ and ‘feeling truly useful’, so the analyst tries to find common factors mentioned by several of the volunteers – as with ordinary Qualitative Data Analysis.

This approach enables the analyst to look for patterns and trends across the dataset, and to understand which stories of change are common across the set of stories, and which are specific to certain individuals, or to a particular group of respondents.

The analysts only code statements related to volunteers themselves – their thoughts, fears, actions, and so on. We code causal claims directly connected to the activity of NSs and their volunteers. So we might code a claim about independent volunteering which someone did because it then led them to volunteer for the NS, but not if it is just a story on its own.

Like all Qualitative Data Analysis, this type of analysis is subjective – it is more like asking a journalist to make a summary of the stories than, say, asking a statistician to count the frequency of different words used. The results are certainly not representative of any particular population.

Hierarchical coding

The analysts applied hierarchical coding to add extra levels of detail to more general factor labels. The different levels of a factor are separated by a semicolon, e.g., ‘’. You can read the semicolon as ‘in particular’ or ‘specifically’ - so in this example:

Volunteering in this situation made me feel truly useful as I had never felt in any previous volunteering

… this could be coded as a link from ‘volunteering; in Covid situation’ to ‘feeling good; feeling truly useful’. So zooming out to the most general point of view we can read this as “Volunteering –> Feeling good”.

The analyst can use these levels to ‘zoom’ in and out to explore and present different views of the data. Maps might be zoomed in to show only the first level of a factor (with all the sub-levels nested within it) or zoomed out to present all the levels. Most maps in this report mostly use Zoom level 1 to improve map readability.

In some cases it was hard to distinguish between the causal factor “NS action” and “Volunteer action”. We talk about volunteer action when there is no specific mention of RCRC, and we code a passage as “NS action” when there is no specific mention of volunteers or their inputs.

The point of this is not so much to look at the numbers, but rather to try to get a psychological overview of how the volunteers were thinking, what led to what, especially in terms of motivation and action.

Findings

What stands out is the rich information volunteers gave about their motivation and experiences.

[Show high-level overview and also more detail. Show separate stories/maps for different regions. Unfortunately we don’t have gender. #TODO ]

Lots of quotes …

Although volunteers certainly mentioned the way that people’s wellbeing was improved by their actions, they spoke most of all about the volunteering itself, the capacity they needed and which they developed, about their motivation to act (and sometimes how action in turn increased their motivation). They wrote in general about how volunteering was satisfying for them but also about how the different challenges they were faced with led to their own personal growth – in seeing the value of volunteering itself, the role and importance of the National Society, courage in adversity, and so on.

source_id : 217985 statement_id : 45 quote : I think like all LA has made us change our daily way we behave on safety and hygiene, appreciate what you have around, miss the hugs and have more empathy. source_id : 217920 statement_id : 6 quote : Personally, I think this problem will help me become a better person and that together, we are or not part of the medical staff we can provide our help to end the COVID-19.

Lots of quotes …

[Also possible, or as an alternative:]

• Solferino Descriptive analysis by already existing categories – visual presentation (map, colour coding, figures),

• IRaMuTeQ (R based text analysis tool). The volunteer stories will be analysed with the help of lexicometry, which aims to extract information from textual data using statistical approaches. It aims to identify the main groups of ideas or elements which are expressed in these stories by identification of proximity between stories (looking at the vocabulary used in each story or sub-sections of the stories) and then build as homogeneous groups of stories (or subsections of stories). For this we recommend the use of the software (freeware) Iramuteq. It provides users with different text analyses, from simple ones such as basic word frequency to multivariate analysis. Three tools might be in priority used: Word cloud, Descending Hierarchical analysis (DHA) and similarity analysis. See at the end if the document the Iramuteq annex with some example of first and quickly made analyses.

• Cross use of the IRaMuTeQ and Solferino Descriptive analysis to present in a synthetic way the main ideas/feelings/situations presented in the different groups of stories, selection of one or two stories to illustrate more concretely the group.

Chapter 5: Vaccinations

How are vaccinations scaling up?

Income and expenditure

Is there a connection between National Societies offering vaccination support and countries’ policies?

Analyses (not shown here) suggest not.

Find examples xx where National Societies offered vaccine support even though this was not offered much by their governments:

# A tibble: 10 x 2
   ISO3  country           
   <chr> <chr>             
 1 AFG   Afghanistan       
 2 BLZ   Belize            
 3 BWA   Botswana          
 4 GEO   Georgia           
 5 IRN   Iran, Islamic Rep.
 6 MNG   Mongolia          
 7 NGA   Nigeria           
 8 PNG   Papua New Guinea  
 9 SLV   El Salvador       
10 SOM   Somalia           

Predictive modelling of vaccinations (Parima)

Chapter 99: Looking forward, Conclusion

Innovation

• Digital innovation and Transformation have been central to the COVID-19 response

• COVID is an opportunity to accelerate some projects – an opportunity to rethink RC activities – an opportunity for change towards the 2030 Strategic Vision

• Some results from the Federation-wide COVID evaluation: some lessons from there (depending on evaluation questions)

Equity, (in)justice, vaccines

• COVID-19 uncovered injustice in society

• Vaccine inequity - How to roll out goods and services with equity

• The invisible burden of COVID in countries with less developed public health/demographic information systems

The importance of tackling epidemics and pandemics together - Resolution 3 of IC33

• A more holistic approach in the response: not just trying to break the epidemic chain, but we were able to combine medical, public health and wash and economic security and protection, all in all into one approach.

The importance of strengthening communities -remarks from IFRC Health & Care Conference - Session 2 "The future of Global Health Security: what role for Red Cross Red Crescent National Societies?

Trust

• Trust: Investing in communities trusting their authoritative’ sources and the information and services they provide

• Sustained investment in communities; The future of humanitarian response is local - On the importance of locally driven response as one learning from COVID pandemic with the help of the expert knowledge

• Define new normal – Resilient communities are the solution

• Build comprehensive health care systems at community level

• Look for the future – no building back what was before COVID-19 in a better way

• Build partnerships

Conclusion

• The RCRC added value - the only organization in the world that links global level mass scale resources with grassroots, local actors at in 192 countries - Leverage operations for longer term, more meaningful impacts

• RCRC network had the capacity to act on holistic approach and work at the highest international level up to the most local community level

• Lessons learnt about Global pandemic, local needs, priorities and responses, and the role of international support and assistance

PER lessons learned?

Technical Annexes

Indicators with less than 90 data points available:

indicator_text	N_data_points	N_NSs	N_time_points	small
# of health facilities supported with IPC, WASH or other interventions to improve COVID prevention, detection or treatment	61	60	12	TRUE
children under 24 months of age that were supported by NS to receive routine immunization	39	26	11	TRUE
children under 5 years of age that were supported by NS to receive vaccines through SIAs/campaigns	35	24	10	TRUE
community burials of suspected or confirmed COVID-19 cases facilitated or directly carried out by NS volunteers	70	29	24	TRUE
NS providing ambulance services to COVID-19 patients	44	43	9	TRUE

We will not remove these indicators with very low data, because they are mostly the start of vaccination initiatives.

Annexes

Notes for designer

Ensure regions are always in alphabetical order, consistent colours etc

Consider using consistent gender icons, colours, etc wherever feasible

• Chapter elements

  • Figure (have captions but not titles within the image. Captions are brief, most figures and tables also have bulletpoint interpretations underneath too, these could also have a separate style of their own).

  • Table (same)

  • Subtitle

  • Inline highlight

  • Quote - for actual quotes, perhaps this will be laid out with large "" symbols. Just write normal text preceded by “Quote:”

  • Sidebar - similar to a quote. Just write normal text preceded by “Sidebar:”

  • Box - bigger than a sidebar. Just write normal text preceded by “Box:”

• Numbers more than 9999 should be rounded in most cases, e.g. to nearest 1000, with a footnote „Numbers are rounded, e.g., to nearest 1000"

The stringency index

Using the COVID-19 Government Response Tracker developed by the University of Oxford (OxCGRT), see Appendix], part of OWiD. It systematically collects information on several different common policy responses that governments have taken to respond to the pandemic on 20 indicators:

• 8 policy indicators (containment and closure policies, School and workplace closures, Cancellation of Public Events and Gatherings, Stay at home, Face covering, Public Information Campaigns),

• 4 economic policies indicators (income support to citizens, provision of foreign aid,

• 8 health system policies indicators (COVID-19 testing regime, emergency investments into healthcare and most recently, vaccination policies).

The data from these indicators is aggregated into a set of four common indices, reporting a number between 1 and 100 to reflect the level of government action on the topics in question:

• an overall government response index (which records how the response of governments has varied over all indicators in the database, becoming stronger or weaker over the course of the outbreak);

• a containment and health index (which combines ‘lockdown’ restrictions and closures with measures such as testing policy and contact tracing, short term investment in healthcare, as well investments in vaccine)

• an economic support index (which records measures such as income support and debt relief)

• as well as the original stringency index (which records the strictness of ‘lockdown style’ policies that primarily restrict people’s behaviour).

Other ways to aggregate

• First date of reporting each indicator / pillar

• Number of activities overall / in each pillar

• Cumulative figures

• Latest figures

• Incremental figures per quarter

• Highest achieved performance in any one quarter

• Any of these can be relativised to population

• Relative performance scores for each NS, for each indicator and pillar

---

1. About the FDRS and authors of the report (advisory board) is to be included in the annex (see below) and referenced in the introduction↩︎

2. Refer also to “Rand corporation most common limitations of COVID-19 health indicators”↩︎

3. OSF COVerAGE-DB: A database of COVID-19 cases and deaths by age https://osf.io/mpwjq/ COVerAGE-DB: A database of COVID-19 cases and deaths by age is being developed and regularly updated where COVID-19 confirmed cases and death reported statistical offices. 121 countries covered.↩︎

4. (The Global Fund, 2021.)↩︎

5. https://blogs.worldbank.org/opendata/relative-severity-covid-19-mortality-new-indicator-world-banks-data-platform↩︎

6. Using the COVID-19 Government Response Tracker developed by the University of Oxford (OxCGRT), see Appendix, part of OWiD↩︎

7. This number changed slightly over time↩︎

8. Need to keep repeating that this excluded vaccinations xx↩︎

9. About 30 additional people did make contributions but these contributions were not judged to have made any specific causal claims.↩︎