IMHO it's rather common for civil wars in Africa to have a poor counting of victims. Often, one or more combatant sides are not too interested in a detailed accounting, including of civilian casualties. See for another example the recent war[s] in Ethiopia.
In this Sudan conflict, in particular, it looks like neither the gov't nor the RSF are too interested that kind of accounting, e.g.
A health ministry report circulated to aid agencies and seen by Reuters put the death toll in Khartoum State at 234 people as of July 5 [2023]. The report specifies that the data is collected only from civilian hospitals.
But across Khartoum State, which includes the capital and its sister cities Omdurman and Bahri, activist and volunteer groups have recorded at least 580 civilian deaths through July 26 as a result of air strikes, artillery and gunfire.
The disparity in the figures for Khartoum State suggests that the official nationwide death toll, which the health ministry puts at 1,136 people as of July 5, may also be an undercount.
An official in Sudan’s health ministry told Reuters the official figure was “the tip of the iceberg.” That’s because many civilians have died in their neighborhoods or at home – not in hospital – so their deaths wouldn’t have been recorded, he said.
Representatives for the army and RSF did not respond to requests for comment, including on the civilian death toll and the May 7 attack.
If that happens in the capital and large cities, you can imagine how it's in the more remote parts of the country.
It's also hard for me to cut through the conflicting claims of who controls how much of the country, in this war, BTW.
The stuff I said in the beginning about Africa is not limited to wars though. Only 8 countries there have a (reasonably working) registration system for the dead.
All but two countries in Europe - Albania and Monaco - have a universal death registration system, and in Asia, just over half, analysis of UN data shows.
But in Africa it is only Egypt, South Africa, Tunisia, Algeria, Cape Verde, São Tomé and Príncipe, Seychelles and Mauritius that have what are called functioning, compulsory and universal civil registration systems - known as CRVS systems - which record deaths.
So they don't know who and how dies even in normal times.
The main way we know something about deaths in that region is through [random sampling] surveys, e.g. in the CAR (last year, April 2023)
Families described malaria or fever, and diarrhea as the primary reported causes of death with violence accounting for 6 percent of all deaths. Most people expressed that they were desperate for assistance, asking for health care (26.9 percent) and food (25.2 percent). A disturbing 82.3 percent of households reported adults ate ≤ 1 meals per day at the time of interview. Similarly, 72.6 percent of children, who have greater nutritional needs and are at higher risk for malnutrition and its associated morbidities, ate ≤ 1 meals per day at the time of interview.
The survey finds an acute dearth of children under age 3 in a population with low contraceptive use, implying either a recent elevation of infant mortality, a decrease in birth rates due to extreme stress, or both. Corroborating the likelihood of a reduced birth rate, 25.5 percent of known pregnancies resulted in a loss. This is higher than the fraction recorded in neighboring Eastern DRC in 2002 during a period of extreme conflict. These findings align with a 2022 UNICEF report that food insecurity and malnutrition are widespread in CAR and imperil the lives of tens of thousands of children.
The authors write: “CAR is experiencing a severe health emergency, with the highest measured nationwide mortality in the world to our knowledge. UN-published death rate estimates appear to be less than one-fourth of reality. [...]
Not that there's that much agreement between such (CAR surveys), e.g. an older 2020 paper noted that "[a]vailable mortality estimates vary and differ in methodology." (That one also found that violence accounted for 16.7% of deaths there. And that after excluding some areas from the survey due to insecurity.)
I'm not even sure there was survey like that done/possible in Sudan in recent times, due to the widespread fighting and targeting of the aid/int'l workers.
