The impacts of cholera in Zambia, Africa 5-7 Page Research Paper (1 Reference Attached Already)

Impact of Temperature Variability on Cholera Incidence in Southeastern Africa, 1971–2006

Shlomit Paz

Department of Geography and Environmental Studies, University of Haifa, Mt. Carmel, Haifa, Israel

Abstract: Africa has a number of climate-sensitive diseases. One that remains a threat to public health is

cholera. The aquatic environment temperature is the most important ecological parameter governing the

survival and growth of Vibrio cholerae. Indeed, recent studies indicate that global warming might create a

favorable environment for V. cholerae and increase its incidence in vulnerable areas. In light of this, a

Poisson Regression Model has been used to analyze the possible association between the cholera rates in

southeastern Africa and the annual variability of air temperature and sea surface temperature (SST) at

regional and hemispheric scales, for the period 1971–2006. The results showed a significant exponential

increase of cholera rates in humans during the study period. In addition, it was found that the annual mean

air temperature and SST at the local scale, as well as anomalies at hemispheric scales, had significant impact

on the cholera incidence during the study period. Despite future uncertainty, the climate variability has to be

considered in predicting further cholera outbreaks in Africa. This may help to promote better, more efficient

preparedness.

Keywords: Cholera, Southeastern Africa, Temperature variability, Sea surface temperature, Global warming

Africa has a number of climate-sensitive diseases. One that

remains a threat to public health is cholera. Since 1970,

when the seventh pandemic reached the continent

(Swerdlow and Isaäcson, 1994), a recrudescence appeared

(CDC, 2009). In 1991, a large cholera eruption occurred in

many parts of Africa. The cause is not fully understood but

may be related to civil disturbance and poor living condi-

tions affected by drought (Colwell, 1996). Afterward, large

outbreaks became more frequent. In 2005, 95% of the

world’s cases were reported from Africa (CDC, 2009).

The pathogenic agent of cholera is Vibrio cholerae, a

Gram-negative, mobile, and facultative anaerobe (Baumann

and Schubert, 1984). Two serogroups of V. cholerae—O1

and O139—can cause outbreaks; however, V. cholerae O1

causes the majority of eruptions worldwide (WHO, 2008).

V. cholerae is an autochthonous member of the microbial

flora of brackish water that is typical of estuaries and coastal

wetland (Colwell et al., 1977) and also is found in freshwater

environments (Islam et al., 1994). The survival of V. chol-

erae in the aquatic environment is linked to abiotic and

biotic ecological factors (Colwell and Huq, 1994), which are

influenced by climatic causes.

Recent studies indicate that global warming might

create a favorable environment for V. cholerae and increase

its incidence in vulnerable areas (WHO, 2008).

Published online: December 29, 2009

Correspondence to: Shlomit Paz, e-mail: shlomit@geo.haifa.ac.il

EcoHealth 6, 340–345, 2009 DOI: 10.1007/s10393-009-0264-7

Short Communication

� 2009 International Association for Ecology and Health

TEMPERATURE VARIABILITY AND ITS IMPACT ON CHOLERA INCIDENCE

The eventual effect in terms of cholera cases is mediated by

the population immunity level (Campbell-Lendrum, 2005).

Additionally, person-to-person contact plays an important

role in the cholera incidence, such as in crowded religious

festivals or through public transportation (Paz and Broza,

2007). Several studies have shown clear evidence of the role

of climate variability in the disease transmission (Koelle

et al., 2005).

The most important climatic parameter related to

cholera outbreaks is the temperature, especially of the water

bodies and the aquatic environment (Borroto, 1998). This

factor governs the survival and growth of V. cholerae, be-

cause it has a direct influence on its abundance in the

environment, or alternatively, through its indirect influence

on other aquatic organisms to which the pathogen is found

to attach (Islam et al., 1990; Colwell, 1996). Thus, the

potential for cholera outbreaks may increase, parallel to the

increase in ocean surface temperature (Lobitz et al., 2000;

Huq et al., 2001). During the second half of the 20th cen-

tury, a large change in the ocean heat content has occurred.

For the world deep ocean (0–3000 m layer), the linear

trend of heat content was 0.33 9 1022 J year-1 (corre-

sponding to a rate of 0.20 Wm-2 per unit area of earth’s

total surface area). For the Indian Ocean, the linear increase

of heat content was 3.5 9 1022 J (Levitus et al., 2005). In-

deed, linkages between sea surface temperature (SST) in-

crease at both global and regional scales were demonstrated

for the disease outbreaks in Ghana, Africa (de Magny et al.,

2007).

TEMPERATURE VARIABILITY IN AFRICA

Africa is vulnerable to climate variability (Washington

et al., 2004). According to the IPCC report on Africa

(2007a), the air temperature has indicated a significant

warming trend since the 1960s. Although the trends seem

to be consistent over the continent, the changes are not

always uniform and are characterized by inter-annual var-

iability (Kruger and Shongwe, 2004; Malhi and Wright,

2004).

In recent years, most of the research into disease vec-

tors in Africa related to climate variability has focused on

malaria. The IPCC (2007a) indicated that the need exists to

examine the vulnerabilities and impacts of climatic factors

on cholera in Africa. In light of this, and because temper-

ature is the most important ecological parameter here, the

current study analyzed the possible association between

temperature variability and cholera outbreaks in south-

eastern Africa. Following de Magny et al. (2007), who

showed linkages at both regional and global levels, the

current research suggests possible association between

cholera and temperature at regional and hemispheric scales.

STUDY DESCRIPTION

Associations were examined between time series of cholera

cases in southeastern Africa and temperature (air temper-

ature, SST) at regional and hemispheric scales. Data

description is as follows:

1. Cholera cases: number of cholera cases per year in eight

southeastern African countries: Uganda, Kenya, Rwan-

da, Burundi, Tanzania, Malawi, Zambia, and Mozam-

bique (Fig. 1) for the period 1971–2006. These countries

were selected based on their time series sequence, with

minimum missing data along the years. Source: WHO

(2007) Global Health Atlas—cholera.

2. Seasonal and annual temperature time series for 1971–

2006, as follows:

Regional Scale

a. Air temperature for southeastern Africa (30–36�E, 5–17�S), source: NOAA NCEP-NCAR (2009)

b. Sea surface temperature, for the western Indian

Ocean (0–20�S, 40–45�E), source: NOAA, Kaplan SST dataset (2006)

Hemispheric Scale*

c. Air temperature anomaly

d. Sea surface temperature anomaly

*Both datasets cover the whole Southern Hemisphere

and interpolate at 5� 9 5� grid boxes. Source: Southern Hemispheric Anomalies (1971–2006), Climate Research

Unit, University of East Anglia.

It is an accepted assumption that the annual number of

disease events in a population has a Poisson distribution,

which was detected after the examination of the current

cholera time series distribution (CHOL). A trend in the

number of cases can be modeled by a Poisson regression

(Kuhn et al., 1994). The Poisson distribution explains the

nonstationarity of the cholera time series around its trend.

The exponential growth of V. cholerae in high temperature

Temperature Impact on Cholera in Southeastern Africa 341

conditions was noted in earlier studies (Datta and Bhadra,

2003; Kirschner et al., 2008). Based on previous works (e.g.,

Kuhn et al., 1994), the current model suggests that a linear

increase in temperature (IPCC, 2007b; Levitus et al., 2005)

will result in exponential growth of cholera. The model

assumes that the number of cholera cases each year is

determined by the mean temperature in the same and the

previous year. Consequently, the following Poisson

regression model is suggested:

log E CHOLtð Þf g¼ b0 þ b1�Xt þ b2�Xt�1 ð1Þ

where: CHOLt = the number of new cases of cholera in

year t; Xt/Xt-1 = the climate covariate measured in year

t/t-1; (b0, b1) = the coefficients.

A first order autocorrelation, AR1 = cor(Yt, Yt-1) is

taken into account in the estimation using Generalized

Estimating Equations.

b1 and b2 quantify the association of CHOL and X, i.e.

if Xt or Xt-1 increase by one unit, the mean of Yt is ex-

pected to increase in exp{b1} or exp{b2} times, respectively

(multiplicative model). In the following section, the effects

of the different climate covariates are examined by fitting

the Poisson regression model.

The majority of eastern Africa’s outbreaks coincide

with times of increased rainfall. However, the cholera

period is slightly dissimilar in different areas and in some

regions there are two cholera peaks (Emch et al., 2008). In

contrast, drought episodes reduce sanitation level and in-

crease the pathogen concentration in the water, and

therefore also can increase the disease risk (Hales et al.,

2003). Nevertheless, the cholera incidence data (WHO) is

available only for annual scale. In addition, in the current

research, correlation calculation between seasonal mean

temperatures showed high linkages between the seasons.

Therefore, the current results were calculated for the annual

scale.

With respect to results, Fig. 2 presents the cholera rates

in southeastern Africa along the period 1971–2006. A clear

increase in the cholera incidence is seen from the early

1990s. This trend in the disease incidence was found to be

significant (substituting year with Xt), with an estimate of

exp(b1) = 1.08 (P = 0.02). This implies that every year, the

cholera rate is expected to multiply by 1.08 (Table 1).

Associations have been found between the annual in-

crease of the air temperature in southeastern Africa and the

cholera incidence increase in the same area. Linkages have

Figure 1. Southeastern African coun-

tries where cholera rates served as

databases for the current study.

342 S. Paz

been detected having exp(b1) = 1.87 (P = 0.18) for the

current year and exp(b1) = 2.78 (P = 0.03)—the impact of

the previous year. As explained above, this means that

when, in a specific year, the annual mean air temperature

increases by 0.1�C, the expected annual number of cholera cases will be multiplied by 1.87 for that year and by 2.78 for

the previous year.

Linkages also were found for a wider scale, between the

cholera rate and the air temperature anomaly of the

Southern Hemisphere, with an estimate of exp(b1) = 1.18

(P = 0.04) and exp(b1) = 1.26 (P = 0.006) for the previous

year.

Significant linkages were found between the annual

cholera rate and the annual western Indian Ocean’ SST,

exp(b1) = 1.31 (P = 0.01) for the current year and

exp(b1) = 1.23 (P = 0.05)—for the previous year. The

major impact here is of the same year, when an annual

increase of 0.1�C in the SST is expected to multiply the cholera rates by 1.31, but the past year has its impact also

with a multiplication rate of 1.23 (Table 1).

Linkages were found also for the hemispheric scale,

between the SST anomaly and the cholera incidence with

an estimate of exp(b1) = 1.33 (P = 0.03) for the current

year and exp(b1) = 1.44 (P = 0.003) for the previous year,

which is more significant.

As mentioned earlier, a clear increase of cholera rates

has occurred in humans in the study area during the last

decades. Parallel with other parameters (level of population

vulnerability, human access to safe water, etc.), it seems

that the variability of annual mean air temperature and SST

in local and hemispheric scales have significant impacts on

cholera incidence in southeast Africa. Looking at Fig. 2, a

clear relationship exists between the cholera rate and the

temperature time series, especially after the large African

cholera eruption in 1991. This eruption seems to have

served as a platform for future outbreaks, which became

more frequent, and may have been encouraged by tem-

perature increase. The correspondence between tempera-

ture anomaly and cholera rate is substantial in the outbreak

of 1998 and supports the positive linkage between the

parameters (Fig. 2). This can be explained by the impact of

warm events on the bacterial populations and on their host

reservoirs, by providing new, favorable, environmental

conditions (Checkley et al., 2000; Speelmon et al., 2000).

It is important to note that the current model has been

checked after removing the anomalies of the year 1998 to

isolate their possible impact on the results. It was found

that most of the exponents are similar to those presented

above.

Figure 2. Annual standardized values of cholera incidence, air

temperature (local and hemispheric scales) and sea surface

temperature (local and hemispheric scales), 1971–2006. SH = South-

ern Hemisphere.

Table 1. Results of Poisson regressions (*multiplied by 10—so their scale has changed)

Covariate Estimate (b1) SE_mod P value Exp{b1} AR1

Time (study period) 0.07 0.03 0.02 1.08 0.53

Air temperature 0.63 0.47 0.18 1.87 0.66

Air temperature_lag1 1.02 0.47 0.03 2.78

Air temperature anomaly SH 0.17 0.08 0.04 1.18 0.49

Air temperature anomaly SH _lag1 0.23 0.08 0.006 1.26

Sea surface temperature 0.27 0.10 0.01 1.31 0.56

Sea surface temperature _lag1 0.20 0.11 0.05 1.23

Sea surface temperature SH 0.28 0.13 0.03 1.33 0.52

Sea surface temperature SH _lag1 0.37 0.12 0.003 1.44

SH = Southern hemisphere

Temperature Impact on Cholera in Southeastern Africa 343

In the current study, the SST was found to be an

important factor that impacts the cholera rate. Indeed,

earlier studies (Borroto, 1998; West, 1989) noted that the

aquatic environment temperature is the most important

ecological parameter governing the survival and growth of

V. cholerae. When the surface temperature of the ocean (or

coastal and inland lakes) rises, phytoplankton populations

tend to increase (bloom). This can indirectly influence the

viability of the V. cholerae by growing in their reservoir’s

food supply (Colwell, 1996). Moreover, previous research

(Githeko and Woodward, 2003) indicated that warming in

the southeastern African lakes may cause conditions that

increase the risk of cholera transmission.

As for the hemispheric SST scale, the potential for

cholera outbreaks may rise, parallel with the ocean surface

temperature increase (Huq et al., 2001; Lobitz et al., 2000).

Indeed, a linear increase of the Southern Hemisphere’s SST

was identified (Rayner et al., 2005). Based on the IPCC

(2007c), the global SST warming may impact water bodies

at the regional scales. Therefore, it is reasonable to assume

that there is a linkage between the hemispheric increased

SST and cholera incidence in southeastern Africa.

The increase of global temperature may influence the

temporal fluctuations of cholera, as well as potentially in-

crease the frequency and duration of its outbreaks (Emch

et al., 2008). Despite future uncertainty, the climate vari-

ability has to be considered in predicting further cholera

outbreaks in Africa. This may help to promote better, more

efficient preparedness.

ACKNOWLEDGMENT

The Author would like to thank Dr. Ronen Fluss for the

statistical advice.

REFERENCES

Baumann P, Schubert RHW (1984) Vibrionaceae. In: Bergey’s Manual of Systematic Bacteriology, Vol 1, Krieg NR, Holt GJ (editors), Baltimore: Williams & Wilkins, pp 516–550

Borroto RJ (1998) Global warming, rising sea level, and growing risk of cholera incidence: a review of the literature and evidence. GeoJournal 44(2):111–120

Campbell-Lendrum D (2005) How much does the health com- munity care about global environmental change? Global Envi- ronmental Change 15:296–298

CDC, Travelers’ Health - Yellow Book, Chapter 4: Prevention of specific infectious diseases—Cholera. http://wwwn.cdc.gov/ travel/yellowBookCh4-Cholera.aspx. Accessed 7 January 2009

Checkley W, Epstein LD, Gilman RH, Figueroa D, Cama RI, Patz JA, et al. (2000) Effect of El Nino and ambient temperature on hospital admissions for diarrhoeal diseases in Peruvian children. Lancet 355:442–450

Colwell RR, Kaper J, Joseph SW (1977) Vibrio cholerae, Vibrio parahaemolyticus, and other vibrios: occurrence and distribu- tion in Chesapeake Bay. Science 198:394–396

Colwell RR, Huq A (1994) Vibrios in the environment: viable but non-culturable Vibrio cholerae. In: Vibrio Cholerae and Cholera: Molecular to Global Perspectives, Wachsmuth IK, Blake PA, Olsvik Ø (editors), Washington DC: ASM Press, pp 117–133

Colwell RR (1996) Global climate and infectious disease: the cholera paradigm. Science 274:2025–2031

Datta PP, Bhadra RK (2003) Cold shock response and major cold shock proteins of vibrio cholera. Applied and Environmental Microbiology 69(11):6361–6369

de Magny GC, Cazelles B, Guégan JF (2007) Cholera threat to humans in Ghana is influenced by both global and regional climatic variability. EcoHealth 3:223–231

Emch M, Feldacker C, Islam MS, Ali M (2008) Seasonality of cholera from 1974 to 2005: a review of global Patterns. Inter- national Journal of Health Geographics 7:31

Githeko AK, Woodward A (2003) International consensus on the science of climate and health: the IPCC third assessment report. In: Climate Change and Human Health, McMichal AJ, Camp- bell-Lendrum DH, Corvalán CF, Ebi KL, et al. (editors), Gen- eva: World Health Organization, pp 43–60

Hales S, Edwards SJ, Kovats RS (2003) Impacts on health of cli- mate extremes. In: Climate Change and Human Health, McMichal AJ, Campbell-Lendrum DH, Corvalán CF, Ebi KL, et al. (editors), Geneva: World Health Organization, pp 79–102

Huq A, Sack RB, Colwell RR (2001) Cholera and global ecosys- tems. In: Ecosystem Change and Public Health: A Global Per- spective, Aaron JL, Patz JA (editors), Baltimore: Johns Hopkins University Press, pp 327–352

Islam SM, Drasar BS, Bradley DJ (1990) Long-term persistence of toxigenic Vibrio cholerae O1 in the mucilaginous sheath of a blue-green alga Anabaena variabilis. Journal of Tropical Medicine and Hygiene 93:133–139

Islam MS, Drasar BS, Sack RB (1994) The aquatic environment as reservoir of Vibrio cholerae: a review. Journal of Diarrhoeal Disease Research 11:197–206

IPCC Working Group II Report (2007a) Impacts, Adaptation and Vulnerability—Africa. http://www.ipcc.ch/pdf/assessment-report/ ar4/wg2/ar4-wg2-chapter9.pdf. Accessed 4 December 2008

IPCC Working Group I Report (2007b) The physical science basis. Observations: surface and atmospheric climate change. http:// www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter3. pdf. Accessed 5 November 2009

IPCC Working Group I Report (2007c) The physical science basis. Observations: oceanic climate change and sea level. http://www. ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter5.pdf. Accessed 9 January 2009

Kaplan SST dataset (2006) http://www.cdc.noaa.gov/data/time series/. Accessed 22 March 2009

Kirschner AKT, Schlesinger J, Farnleitner AH, Hornek R, Süß B, Golda B, et al. (2008) Rapid growth of planktonic vibrio cholerae Non-O1/Non-O139 strains in a Large Alkaline Lake in Austria: dependence on temperature and dissolved organic carbon quality. Applied and Environmental Microbiology 74(7):2004– 2015

344 S. Paz

Koelle K, Pascual M, Yunus M (2005) Pathogen adaptation to seasonal forcing, climate change. Proceedings of the Royal Society of London. Series B: Biological Sciences 272:971–977

Kruger AC, Shongwe S (2004) Temperature trends in South Africa: 1960–2003. International Journal of Climatology 24:1929– 1945

Kuhn L, Davidson LL, Maureen S, Durkin MS (1994) Use of Poisson regression and time series analysis for detecting changes over time in rates of child injury following a prevention pro- gram. American Journal of Epidemiology 140(10):943–955

Levitus S, Antonov J, Boyer T (2005) Warming of the world ocean, 1955–2003. Geophysical Research Letters 32:L02604

Lobitz B, Beck L, Huq A, Wood B, Fuchs G, Faruque ASG, Colwell RR (2000) Climate and infectious disease: use of remote sensing for detection of Vibrio cholerae by indirect measurement. Pro- ceeding of the National Academy of Science of the USA 97(4):1438–1443

Malhi Y, Wright J (2004) Spatial patterns and recent trends in the climate of tropical rainforest regions. Philosophical Transactions of the Royal Society of London: Series B 359:311–329

NOAA NCEP-NCAR CDAS-1 MONTHLY (2009) Diagnostic surface temperature. http://iridl.ldeo.columbia.edu/SOURCES/ .NOAA/.NCEP-NCAR/.CDAS-1/.MONTHLY/.Diagnostic/.sur face/.temp/. Accessed 10 November 2008

Paz S, Broza M (2007) Wind direction and its linkage with Vibrio cholerae dissemination. Environmental Health Perspectives 115(2):195–200

Rayner NA, Brohan P, Parker DE, Folland CK, Kennedy JJ, Vanicek M, Ansell TJ, Tett SFB (2005) Improved analyses of changes and uncertainties in sea surface temperature measured

in situ since the mid-nineteenth century: the HadSST2 dataset. Journal of Climate 19:446–469

Southern Hemispheric Anomalies (1971–2006) Climate research unit, University of East Anglia. http://www.cru.uea.ac.uk/cru/ data/temperature/#faq. Accessed 25 January 2008

Speelmon EC, Checkley W, Gilman RH, Patz J, Calderon M, Manga S (2000) Cholera incidence and El Nino-related higher ambient temperature. JAMA 283:3072–3074

Swerdlow LD, Isaäcson M (1994) The epidemiology of cholera in Africa. In: Vibrio cholerae and Cholera: Molecular to Global Perspectives, Wachsmuth IK, Blake PA, Olsvik Ø (editors), Washington DC: American Society for Microbiology Press, pp 297–308

Washington R, Harrison M, Conway D (2004) African climate report. A report commissioned by the UK government to review African climate science, policy and options for action. www. defra.gov.uk/environment/climatechange/internat/devcountry/ pdf/africa-climate.pdf. Accessed 10 July 2008

West P (1989) The human pathogenic vibrios: a public health update with environmental perspectives. Epidemiology and Infection 103(1):1–34

World Health Organization (WHO) (2008) Cholera, fact sheet N�107, November 2008. http://www.who.int/mediacentre/fact sheets/fs107/en/index.html. Accessed 10 February 2009

World Health Organization (WHO) (2007) WHO global health atlas cholera Africa 1971–2006. http://www.who.int/Global Atlas/. Accessed 12 July 2008

Temperature Impact on Cholera in Southeastern Africa 345

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.