Global, regional and national trends and impacts of natural floods, 1990–2022

Abstract

Objective

To assess global, regional and national trends in the impact of floods from 1990 to 2022 and determine factors influencing flood-related deaths.

Methods

We used data on flood disasters from the International Disaster Database for 1990–2022 from 168 countries. We calculated the annual percentage change to estimate trends in the rates of people affected and killed by floods by study period, World Health Organization (WHO) region, country income level and flood type. We used multivariable logistic regression analysis to assess the factors associated with death from floods.

Findings

From 1990 to 2022, 4713 floods were recorded in 168 countries, which affected > 3.2 billion people, caused 218 353 deaths and were responsible for more than 1.3 trillion United States dollars of economic losses. The WHO Western Pacific Region had the most people affected by floods (> 2.0 billion), accounting for 63.19% (2 024 599 380/3 203 944 965) of all affected populations. The South-East Asia Region had the most deaths (71 713, 32.84%). The African and Eastern Mediterranean Regions had the highest number of people affected and killed by floods per 100 000 population in 2022. The odds of floods causing more than 50 deaths were significantly higher in low-income countries (adjusted odds ratio: 14.34; 95% confidence interval: 7.46 to 30.04) compared with high-income countries. Numbers of people affected and mortality due to floods declined over time.

Conclusion

Despite the decreases in populations affected and deaths, floods still have a serious impact on people and economies globally, particularly in lower-income countries. Action is needed to improve disaster risk management and flood mitigation.

Introduction

Natural disasters are catastrophic events of atmospheric, geological and hydrological origins. These disasters include earthquakes, floods, volcanic eruptions, landslides, tsunamis and droughts.¹ From 1998 to 2017, 1.3 million people were reported to have died from climate-related and geophysical disasters, and a further 4.4 billion people were injured, made homeless, displaced or in need of emergency assistance.² In the same period, disaster-hit countries also reported direct economic losses valued at 2.9 trillion United States dollars (US$).² Disasters are a major contributor to entrenched poverty in low- and middle-income countries; thus, reducing losses from disasters is key to eradicating poverty. The United Nations (UN) Office for Disaster Risk Reduction called for a substantial reduction in direct disaster losses as well as associated infectious diseases, and social and economic losses, and noted that the responsibility should be shared by the stakeholders.²⁹

Natural floods (hereafter called floods) are the most common natural hazard event and are the leading cause of death from disasters worldwide as well as serious health, social and economic consequences.^3–5 A 2013 study showed that from 1980 to 2009, floods were responsible for more than 500 000 deaths and more than 350 000 injuries, and affected nearly 3 billion people across the world.³ The World Health Organization (WHO) South-East Asia and Western Pacific Regions were the most flood-affected regions and were estimated to account for nearly 50% of flood-related deaths in the last 25 years of the 20th century.^6,7 Additionally, an estimated 1.81 billion people were directly exposed to 1-in-100-year floods, 1.24 billion of whom lived in the WHO South-East Asia Region.⁸ In the WHO European Region, about 400 floods caused the deaths of more than 2000 people, affected 8.7 million others, and resulted in at least 72 billion euros in losses during 2000–2014.⁹

A recent systematic review identified various factors that can influence flood-related deaths, which could be classified into five categories, namely hazard-related, individual, environmental, socioeconomic and managerial factors.¹⁰ However, which of these factors has a significant causal relationship with flood-related deaths is still unclear. Furthermore, the trends in the global impact of floods in the past three decades is not well understood. In this study, we aimed to summarize the trends in the global, regional and national impact of floods from 1990 to 2022, and to assess the factors associated with deaths from floods. The findings could provide a comprehensive understanding of the trends in the impact of floods and the factors influencing flood-related deaths. Such information is important to help strengthen disaster risk management and enhance preparedness for and effective responses to floods.

Methods

Study design and data sources

We conducted a worldwide observational study which covered all countries and territories that reported flood disasters from 1990 to 2022. Disaster-related data were extracted from the International Disaster Database (EM-DAT).¹¹ This database was launched in 1988 by the Centre for Research on the Epidemiology of Disasters with the support of WHO and the Belgian government. The main objective of the database is to facilitate humanitarian action at national and international levels, rationalize decision-making for disaster preparedness, and provide an objective base for vulnerability assessment and priority-setting.¹¹ EM-DAT contains essential core data on the occurrence and effects of different kinds of disasters in the world from 1900 to the present day. These data are compiled from various sources, including UN agencies, nongovernmental organizations, insurance companies, research institutes and press agencies.¹¹ The income level and number of populations of various countries in each year from 1990 to 2022 were extracted from the UN open database, UNdata.²⁸ No ethics approval was required for this analysis of publicly available data.

Variables

Flood was defined as the overflow of water from a stream channel onto normally dry land in the floodplain (riverine flooding), higher-than-normal levels along the coast and by lakes or reservoirs (coastal flooding), as well as ponding of water at or near the point where rain fell (flash flooding).¹¹ In the EM-DAT, flood disasters are one of the hydrological disasters (including flood, landslide and wave action), which are defined as hazards caused by the occurrence, movement and distribution of surface and subsurface freshwater and saltwater. We used the following variables from the database: country (countries in which the flood occurred); date (when the flood occurred and ended); flood type (riverine flooding, coastal flooding or flash flooding); deaths (number of people who died because of the flood); affected individuals (number of people who were injured or made homeless by the flood or who required immediate assistance during a period of emergency, i.e. basic survival needs such as food, water, shelter, sanitation and immediate medical assistance); and adjusted total damages in US$ (all damages and economic losses directly or indirectly related to the flood, adjusted for inflation using the consumer price index).

Statistical analysis

In the descriptive analysis, the number of affected people, the number of deaths and the total economic damages in each year were summed from 1990 to 2022 by country. The number of floods in each WHO region every month was also summed from 1990 to 2022. Given differences in population size and fluctuations of population size in different years and regions, we calculated the rates of people affected by flood per 100 000 population (affected rate) and the rates of flood-related deaths per 100 000 population (mortality) using the total number of people affected by the floods, the number of deaths caused by the floods, and the total population of each region in each year. We calculated the estimated annual percentage change to show trends. This indicator is widely used to show the trend and annual change over a specified time period.^12,13 We fitted a regression line to the natural logarithm of rates using the formula y = α + βx + ε, where y is the rate, x is the calendar year, α is the expected value of the dependent variable when all explanatory variables are 0, β is regression coefficient and ε is the error term (the part of the y variable that the model cannot explain). We calculated the estimated annual percentage change, with its 95% confidence intervals (CIs) to assess the time trend in the rates, as 100 × (e^β – 1), where e is the Euler number. If the estimated annual percentage change and its 95% CI were both > 0, we considered the trend in rates to be increasing in the given time interval. Conversely, if the change and 95% CI were both < 0, we considered the trend in rates to be decreasing.

Additionally, we divided all the flood events into four groups based on the number of deaths they caused: (i) zero deaths or no data; (ii) 1–9 deaths; (iii) 10–49 deaths; and (iv) ≥ 50 deaths. We used the χ² test to assess differences between the four groups by study period, WHO region, country income level and subtype of flood. We used multivariable logistic regression analysis to estimate the associations of factors (study period, WHO region, income level and flood type) with the risk of death at the group level compared with floods with 0 deaths or no data, expressed as adjusted odds ratios (aOR). We used R, version 4.1.1 (R Foundation, Vienna, Austria) for all analyses.

Results

Global and national impact

From 1990 to 2022, 4713 floods were recorded in 168 countries and territories, which affected more than 3.2 billion people, resulted in 218 353 deaths and caused more than US$ 1.3 trillion in economic damages. Of all the countries where floods occurred, China was the most affected, with the largest cumulative population affected (1.9 billion people), the most economic damage (US$ 442 billion) and the second largest number of deaths due to the floods (30 890 deaths). India had the second largest population affected (629 million people), the third most economic damages (US$ 115 billion) and the largest number of deaths (46 506 deaths). In Bangladesh, 159 million people were affected by floods, the third largest in the world. In the United States of America, floods caused about US$ 135 billion in economic damages, second only to China; and Bolivarian Republic of Venezuela recorded the third highest number of deaths (30 342 deaths) from floods (Fig. 1).

Fig. 1.

Cumulative impact of natural floods by country, 1990–2022

US$: United States dollars.

Floods by WHO region

Of the 4713 floods recorded, 930 (19.73%) occurred in the African Region; 1029 (21.83%) in the Regions of the Americas; 777 (16.49%) in the South-East Asia Region; 708 (15.02%) in the European Region; 485 (10.29%) in the Eastern Mediterranean Region; and 784 (16.63%) in the Western Pacific Region. In the African Region, the number of floods was highest in August (158 floods; 16.99%). In the Region of the Americas, the greatest number of floods occurred in May (122 floods; 11.86%), followed by 104 floods (10.11%) in January and 103 floods (10.01%) in October. In the European Region, floods mostly occurred from May to August, accounting for 47.60% (337/708) of the total floods. In the Eastern Mediterranean Region, most floods occurred in July and August, accounting for 29.28% (142/485) of the total floods. In both the South-East Asia and Western Pacific Regions, floods mostly occurred during June to September, accounting for 52.90% (411/777) and 54.97% (431/784) of the total floods, respectively (Fig. 2).

Fig. 2.

Number of natural floods per month, by WHO region, 1990–2022

WHO: World Health Organization.

From 1990 to 2022, most people affected by floods were residing in the Western Pacific Region (> 2 billion people), representing 63.19% of all the affected populations (2 024 599 380/3 203 944 965), followed by 890 683 355 (27.80%) affected people living in the South-East Asia Region. The European Region had the lowest number of people affected by floods from 1990 to 2022 (17 million). In Western Pacific Region, there were 24 years when more than 10 million people were affected by floods yearly and, among those years, 10 years when more than 100 million people were affected. In the South-East Asia Region, there were 25 years when more than 10 million people were affected by floods yearly, and in 1993, 146 million people were affected. More than 10 million people were affected by floods in 4 years in the Eastern Mediterranean Region: 34 million in 2022, 21 million in 2010, 13 million in 1992 and 12 million in 2019. The highest number of people affected by floods in the Region of the Americas was 16 million in 2008 (Table 1). Globally, the largest number of people affected was in 1998 (243 million people), followed by 212 million people in 1991 and 157 million people in 1996.

Table 1. Number of affected people and deaths caused by natural floods, by year and WHO region, 1990–2022.

Year	No. of people affected							No. of deaths
	Africa	Americas	South-East Asia	Europe	Eastern Mediterranean	Western Pacific		Africa	Americas	South-East Asia	Europe	Eastern Mediterranean	Western Pacific
1990	540 368	232 186	2 477 100	27 988	152 000	43 224 589		278	220	709	88	37	919
1991	2 150 214	818 922	12 604 183	57 500	402 622	211 765 828		521	96	1 511	162	1 282	2 280
1992	2 258	598 819	4 072 328	105 500	12 878 868	1 940 021		21	201	840	1363	2 109	695
1993	262 705	1 330 229	145 851 779	527 873	803 523	474 814		90	940	2 719	174	848	1 379
1994	880 256	190 425	13 686 478	848 229	1 221 641	112 956 142		232	360	2 314	209	1 208	2 448
1995	1 220 225	588 415	58 331 432	3 679 918	1 918 261	127 702 996		428	272	2 888	78	1 935	2 347
1996	493 858	717 921	22 355 210	433 190	1 737 825	156 939 009		136	146	2 428	37	767	4 533
1997	2 019 704	753 881	31 330 468	547 971	2 205 734	7 794 944		639	494	2 528	170	2 641	1 213
1998	2 274 039	486 900	45 196 907	1 534 686	1 260 140	242 910 013		169	554	3 548	285	1 360	4 737
1999	1 210 708	2 438 670	28 900 385	368 424	327 508	115 840 373		357	30 868	1 024	133	153	2 272
2000	5 338 668	442 010	57 315 993	210 006	374 850	10 223 409		1 185	495	2 849	91	16	1 389
2001	2 748 481	760 568	22 054 975	753 858	1 709 513	6 519 111		1 574	265	1 357	122	737	926
2002	897 672	985 422	47 999 183	1 081 417	330 642	116 470 387		382	372	1 152	383	349	1 535
2003	1 074 374	837 462	9 415 977	68 641	1 663 772	156 367 046		356	578	1 604	32	497	843
2004	705 234	641 062	71 797 420	868 169	118 483	42 860 447		216	3603	2 342	54	117	650
2005	566 109	1 031 306	30 464 569	153 518	7 719 199	35 092 108		380	545	2 664	181	1 130	842
2006	1 463 015	814 647	11 136 995	183 379	761 730	15 949 598		1 325	302	2 447	120	898	751
2007	4 475 712	5 878 898	55 133 479	429 457	617 426	112 344 644		745	656	4 635	85	1 118	1 357
2008	943 523	15 848 647	17 699 918	256 435	391 088	9 771 150		495	623	1 932	66	238	653
2009	2 460 622	2 454 724	7 350 598	92 765	273 329	47 102 895		590	351	1 777	132	400	377
2010	4 279 406	4 651 572	14 208 701	388 680	20 533 154	144 732 921		523	1 249	1 559	301	2 410	2 314
2011	1 443 629	5 254 513	25 193 211	39 352	5 418 366	99 098 652		672	1 679	1 941	46	588	1 237
2012	9 167 877	770 043	10 540 815	230 051	5 271 549	38 019 184		786	209	674	225	763	887
2013	1 673 743	1 622 046	8 538 777	1 437 221	2 184 625	16 543 482		632	446	6 989	62	575	1 115
2014	593 818	1 537 037	10 408 785	1 154 026	3 477 012	24 567 410		457	254	1 200	167	854	604
2015	1 846 672	1 427 334	19 559 222	241 022	2 588 226	1 759 266		961	355	1 094	112	527	430
2016	1 089 525	3 704 426	8 748 970	74 742	241 431	65 360 232		646	231	1 819	60	676	965
2017	1 394 494	2 567 331	37 179 509	52 577	123 363	14 302 014		287	350	1 876	25	358	441
2018	2 618 198	586 544	24 524 598	81 294	895 175	5 551 048		678	85	1 057	65	290	706
2019	3 648 255	1 064 920	12 381 937	96 073	11 578 581	6 034 170		837	329	2 746	113	651	473
2020	5 890 694	284 785	9 080 369	189 065	4 253 999	14 654 443		1 045	272	3 070	59	1 160	579
2021	2 096 171	3 760 254	3 956 792	433 166	1 128 779	18 520 741		262	329	1 948	333	777	517
2022	7 024 127	4 064 092	11 186 292	23 052	33 790 206	1 206 293		2 022	901	2 472	10	2 350	307
Total	74 494 354	69 146 011	890 683 355	16 669 245	128 352 620	2 024 599 380		19 927	48 630	71 713	5 543	29 819	42 721

WHO: World Health Organization.

Of the 218 353 deaths due to floods, the South-East Asia Region had the highest number (71 713; 32.84%) followed by the Region of the Americas (48 630; 22.27%). The Western Pacific Region had 42 721 deaths (19.57% of all deaths). In the South-East Asia Region, the greatest number of deaths (6989 deaths) occurred in 2013, followed by 4635 deaths in 2007. In the Region of the Americas, the greatest number of deaths (30 868 deaths) occurred in 1999, followed by 3603 deaths in 2004. In the Western Pacific Region, the greatest number deaths (4737 deaths) occurred in 1998, followed by 4533 deaths in 1996 (Table 1).

In 1990, the Western Pacific Region had the highest affected rate of floods (1120.50 per 100 000 population), followed by 335.13 per 100 000 in the African Region and 236.64 per 100 000 in the Eastern Mediterranean Region. The affected rates in the South-East Asia, American and European Regions in 1990 were all lower than 100.00 per 100 000 population. In 2022, the Eastern Mediterranean Region had the highest affected rate (3817.94 per 100 000 population), followed by 383.21 per 100 000 in the African Region. Between 1990 and 2022, the affected rates decreased significantly in the South-East Asia (estimated annual percentage change –5.75%; 95% CI: –9.00% to –2.38%) and Western Pacific (estimated annual percentage change –6.76%; 95% CI: –12.33% to –0.83%) Regions (Table 2). In certain years the affected rates of floods were particularly high in some regions, such as in 1991 in the Western Pacific and African Regions, with 5420.13 per 100 000 and 4407.24 per 100 000 population affected, respectively (Fig. 3).

Table 2. Affected population and mortality caused by natural floods in 1990 and 2022 and changes over time, by WHO region.

WHO region	Affected			Mortality
	No. of people per 100 000 population		Estimated annual percentage change (95% CI)	Deaths per 100 000 population		Estimated annual percentage change (95% CI)
	1990	2022		1990	2022
Africa	335.13	383.21	−0.70 (−4.57 to 3.32)	1.72	1.10	−2.88 (−5.41 to −0.28)
Americas	13.76	88.94	2.33 (−1.09 to 5.88)	0.13	0.20	−2.91 (−7.20 to 1.58)
South-East Asia	60.20	127.46	−5.75 (−9.00 to −2.38)	0.17	0.28	−3.18 (−5.41 to −0.91)
Europe	2.19	5.09	−3.31 (−7.87 to 1.47)	0.07	0.02	−3.53 (−6.50 to −0.47)
Eastern Mediterranean	236.64	3817.94	−0.76 (−6.08 to 4.86)	0.58	2.66	−3.74 (−7.31 to −0.03)
Western Pacific	1120.50	17.41	−6.76 (−12.33 to −0.83)	0.24	0.04	−6.99 (−9.45 to −4.47)

CI: confidence interval; WHO: World Health Organization.

Fig. 3.

Population affected by natural floods, by year and WHO region, 1990–2022

WHO: World Health Organization.

In 1990, the African Region had the highest mortality caused by floods (1.72 per 100 000 population), followed by 0.58 per 100 000 in the Eastern Mediterranean Region (Table 2). In 2022, we estimated the highest mortality in the Eastern Mediterranean Region (2.66 deaths per 100 000 population), followed by 1.10 deaths per 100 000 in the African Region. Between 1990 and 2022, mortality decreased significantly in all WHO regions except the Region of the Americas. The greatest decrease was in the Western Pacific Region (estimated annual percentage change  −6.99%; 95% CI: −9.45% to −4.47%), followed by the Eastern Mediterranean Region (estimated annual percentage change: −3.74%; 95% CI: −7.31% to −0.03%). Mortality was particularly high in 1999 in the Region of the Americas (27.24 per 100 000) and in 1991 in the African Region (10.68 per 100 000; Fig. 4).

Fig. 4.

Deaths caused by natural floods, by year and WHO region, 1990–2022

WHO: World Health Organization.

Factors affecting deaths

As shown in Table 3, the proportion of floods with more than 50 deaths decreased from 22.53% (194/861) in 1990–1999 to 12.28% (188/1531) in 2010–2022. From 1990 to 2022, the South-East Asia and Western Pacific Regions had the greatest proportion of floods with more than 50 deaths, 36.20% (236/652) and 23.01% (150/652), respectively. In the South-East Asia Region, 30.37% (236/777) of floods caused more than 50 deaths. Income level was also associated with flood-related deaths. Floods in low- and lower-middle-income countries caused significantly more deaths than in high- and upper-middle-income countries (P < 0.001). The proportion of floods with more than 50 deaths was highest for coastal floods (22.67%; 17/75), followed by flash floods (15.40%; 118/766) and riverine floods (14.60%; 363/2487; Table 3).

Table 3. Frequency of natural floods by number of deaths caused and associated variables, 1990–2022.

Variable	No. (%) of floods				Proportion of > 50 deaths group, %	P
	0 deaths or no data (n = 1267)	1–9 deaths (n = 1403)	10–49 deaths (n = 1391)	> 50 deaths (n = 652)
Time perioda						< 0.001
1990–1999	223 (20.67)	197 (16.74)	247 (19.60)	194 (32.55)	22.53
2000–2009	496 (45.97)	518 (44.01)	492 (39.05)	214 (35.91)	12.44
2010–2019	360 (33.36)	462 (39.25)	521 (41.35)	188 (31.54)	12.28
WHO region						< 0.001
Africa	276 (21.78)	281 (20.03)	280 (20.13)	93 (14.26)	10.00
Americas	318 (25.10)	375 (26.73)	273 (19.63)	63 (9.66)	6.12
South-East Asia	124 (9.79)	161 (11.48)	256 (18.40)	236 (36.20)	30.37
Europe	310 (24.47)	293 (20.88)	88 (6.33)	17 (2.61)	2.40
Eastern Mediterranean	69 (5.45)	107 (7.63)	216 (15.53)	93 (14.26)	19.18
Western Pacific	170 (13.42)	186 (13.26)	278 (19.99)	150 (23.01)	19.13
Country income level						< 0.001
High	278 (21.94)	310 (22.10)	106 (7.62)	11 (1.69)	1.56
Upper middle	331 (26.12)	342 (24.38)	284 (20.42)	92 (14.11)	8.77
Lower middle	328 (25.89)	445 (31.72)	564 (40.55)	236 (36.20)	15.00
Low	330 (26.05)	306 (21.81)	437 (31.42)	313 (48.01)	22.58
Flood type						< 0.001
Coastal	24 (1.89)	17 (1.21)	17 (1.22)	17 (2.61)	22.67
Flash	146 (11.52)	241 (17.18)	261 (18.76)	118 (18.10)	15.40
Riverine	649 (51.22)	709 (50.53)	766 (55.07)	363 (55.67)	14.60
Unknown	448 (35.36)	436 (31.08)	347 (24.95)	154 (23.62)	11.12

WHO: World Health Organization.

^a To ensure that each group had a decade-long span, the three time periods analysed were 1990–1999, 2000–2009 and 2010–2019. Thus, n = 1079, 1177, 1260 and 596 for 0 deaths or no data, 1–9 deaths, 10–49 deaths and > 50 deaths, respectively.

The odds of floods with more than 50 deaths decreased significantly over time compared with floods with no deaths or no data. Compared with the African Region, floods in the Eastern Mediterranean Region were more likely to cause 10–49 deaths (aOR: 2.83; 95% CI: 2.04 to 3.94) or > 50 deaths (aOR: 3.85; 95% CI: 2.55 to 5.83). Similar findings were observed in the South-East Asia Region (aOR: 1.87; 95% CI: 1.40 to 2.49 for 10–49 deaths; aOR: 6.72; 95% CI: 4.74 to 9.62 for > 50 deaths) and the Western Pacific Region (aOR: 1.79; 95% CI: 1.33 to 2.41 for 10–49 deaths; aOR: 4.33; 95% CI: 2.94 to 6.44 for > 50 deaths). Compared with high-income countries, the odds of more than 50 deaths from floods were significantly higher for all other incomes levels, particularly for low-income countries, (aOR: 14.34; 95% CI: 7.46 to 30.04). Compared with coastal floods, the odds of causing 10–49 deaths were higher for flash floods (aOR: 3.16; 95% CI: 1.59 to 6.41) and riverine floods (aOR: 2.16; 95% CI: 1.12 to 4.26). No significant differences were seen between flood type in causing > 50 deaths (Table 4).

Table 4. Multivariable logistic regression analysis of factors associated with natural flood-related deaths, 1990–2022.

Characteristic	aOR (95% CI)
	1–9 deaths	10–49 deaths	> 50 deaths
Time period
1990–1999	Reference	Reference	Reference
2000–2009	1.13 (0.89 to 1.43)	0.70 (0.55 to 0.90)	0.41 (0.30 to 0.57)
2010–2019	1.39 (1.09 to 1.78)	1.11 (0.86 to 1.43)	0.65 (0.47 to 0.91)
WHO region
Africa	Reference	Reference	Reference
Americas	1.01 (0.75 to 1.35)	1.05 (0.78 to 1.42)	1.25 (0.79 to 1.98)
Eastern Mediterranean	1.32 (0.93 to 1.89)	2.83 (2.04 to 3.94)	3.85 (2.55 to 5.83)
Europe	0.77 (0.57 to 1.04)	0.36 (0.25 to 0.50)	0.34 (0.18 to 0.61)
South-East Asia	1.02 (0.74 to 1.41)	1.87 (1.40 to 2.49)	6.72 (4.74 to 9.62)
Western Pacific	0.86 (0.63 to 1.18)	1.79 (1.33 to 2.41)	4.33 (2.94 to 6.44)
Income level
High	Reference	Reference	Reference
Upper middle	0.85 (0.67 to 1.07)	2.05 (1.53 to 2.76)	7.05 (3.74 to 14.54)
Lower middle	1.10 (0.86 to 1.41)	2.89 (2.15 to 3.90)	8.96 (4.81 to 18.32)
Low	0.74 (0.54 to 1.00)	2.31 (1.65 to 3.25)	14.34 (7.46 to 30.04)
Flood type
Coastal	Reference	Reference	Reference
Flash	2.02 (1.04 to 4.01)	3.16 (1.59 to 6.41)	2.02 (0.94 to 4.41)
Riverine	1.45 (0.77 to 2.80)	2.16 (1.12 to 4.26)	1.50 (0.73 to 3.14)
Unknown	1.18 (0.62 to 2.31)	1.39 (0.71 to 2.78)	0.87 (0.41 to 1.87)

aOR: adjusted odds ratio; CI: confidence interval; WHO: World Health Organization.

Note: The comparison is 0 deaths or no data during a flooding event for all the three death groups. For example, low-income countries have a higher likelihood of experiencing more than 50 deaths instead of 0 deaths than high-income countries.

Discussion

Floods have affected billions of people over the past 30 years and resulted in hundreds of thousands of deaths in the 168 countries included in our study. China, India, United States of America and Bangladesh, in that order, were the most affected countries. Of the WHO regions, the South-East Asia and Western Pacific Regions had the largest number of people affected and killed by floods. However, the Eastern Mediterranean and African Regions had the highest number of people affected and killed by floods per 100 000 population in 2022. Furthermore, countries with lower income level had a higher risk of deaths caused by floods. Our findings underscore the urgent need to manage risks and enhance preparedness measures for effective responses to floods.

China was the country most affected by natural floods, with the highest number of people affected and second-highest number of deaths. Despite significant investments in flood defences following the 1998 floods, urban flooding remains a serious concern, with more than 157 cities affected since 2006.^14,15 Catastrophic floods in Beijing (2012) and Zhengzhou (2021) underscored the need for adaptive traffic management, improved levees, enlarged reservoirs and advanced early-warning systems.^16–19 In the United States, floods also pose a great threat to lives and property, and impose a substantial financial burden on the National Flood Insurance Program, which has accrued more than US$ 20 billion in debt since 2005.^20,21 Climate change, implicated in escalating precipitation levels, contributes to growing flood-induced damages.²² Similarly, Bangladesh and India have faced increasing flood incidences due to climate change, with hasty developmental projects exacerbating the impacts.²³ Therefore, governments need to use scientific evidence to devise robust flood mitigation strategies, including successful so-called room for the river approaches, which involve enlarging the capacity of rivers and improving overall floodplain management.^24,25 Collaboration across state and international borders with relevant stakeholders is critical.

Regionally, large numbers of people in the Western Pacific and South-East Asia Regions are affected by floods. The South-East Asia, American and Western Pacific Regions, in that order of size, accounted for 74.68% (163 064/218 353) of all deaths due to floods. Of note, our study also found that although the Eastern Mediterranean and African Regions had fewer people affected, the population affected and the mortality due to floods in these regions were the highest in 2022. This situation reflects that in low-income countries, although the disaster-affected population might be smaller, the magnitude of the disaster is greater than in other areas. Thus, areas where populations are substantially affected require comprehensive flood reduction strategies that are well planned and executed to mitigate the effects of floods. Regions with high mortality need to gauge flood risks more accurately and improve emergency management of floods and other disasters to reduce fatalities from floods. To achieve this goal might require greater investment in and redistribution of resources for better flood prevention and emergency response systems.

The risk of death from floods was higher in lower-income countries than in high-income countries. Although economic development can increase disaster-related economic losses, improvements in emergency preparedness, response and coping capacity can reduce vulnerability to disasters, thereby lowering the number of affected individuals and disaster-related deaths.⁷ Therefore, economic assistance and policy support for low-income countries are important. Helping these countries to build disaster risk analysis systems could guide the development of strategies for managing disasters, and the identification of vulnerable countries and regions and the factors contributing to social vulnerability.^18,27

Our analysis showed notable time trends in the population affected and mortality due to floods, as well as reduced odds of floods resulting in more than 50 deaths. The decreasing trends suggest an improvement in our global, regional and national responses to such disasters, potentially owing to advances in predictive technology, infrastructure and disaster management strategies. However, it is important to remember that, while these declines are promising, flood events are dependent on a variety of fluctuating factors, such as changing climate patterns. Our findings warrant further in-depth investigations and discussion to fully appreciate their complex implications.

Our study had some limitations. Our findings depended on the data source. In several events, the reported number of deaths, injuries and people affected might be lower than the actual situation, and hence the impact of floods might be underestimated. Our study did not focus on personal factors, and future studies that have access to individual-level data might want to consider exploring these dimensions, given the potential influence of personal factors on the impact of floods. This information could provide a more nuanced understanding of how personal factors might modulate the effects of floods.

Our findings highlight the continued serious impact of floods on countries and populations and hence the urgent need for stronger disaster risk governance, effective flood mitigation strategies, improved emergency systems and international aid, especially for low-income countries.