Multi-site cholera surveillance within the African Cholera Surveillance Network shows endemicity in Mozambique, 2011–2015

Abstract

BackgroundMozambique suffers recurrent annual cholera outbreaks especially during the rainy season between October to March. The African Cholera Surveillance Network (Africhol) was implemented in Mozambique in 2011 to generate accurate detailed surveillance data to support appropriate interventions for cholera control and prevention in the country.Methodology/Principal findingsAfrichol was implemented in enhanced surveillance zones located in the provinces of Sofala (Beira), Zambézia (District Mocuba), and Cabo Delgado (Pemba City). Data were also analyzed from the three outbreak areas that experienced the greatest number of cases during the time period under observation (in the districts of Cuamba, Montepuez, and Nampula). Rectal swabs were collected from suspected cases for identification of Vibrio cholerae, as well as clinical, behavioral, and socio-demographic variables. We analyzed factors associated with confirmed, hospitalized, and fatal cholera using multivariate logistic regression models.A total of 1,863 suspected cases and 23 deaths (case fatality ratio (CFR), 1.2%) were reported from October 2011 to December 2015. Among these suspected cases, 52.2% were tested of which 23.5% were positive for Vibrio cholerae O1 Ogawa. Risk factors independently associated with the occurrence of confirmed cholera were living in Nampula city district, the year 2014, human immunodeficiency virus infection, and the primary water source for drinking.Conclusions/SignificanceCholera was endemic in Mozambique during the study period with a high CFR and identifiable risk factors. The study reinforces the importance of continued cholera surveillance, including a strong laboratory component. The results enhanced our understanding of the need to target priority areas and at-risk populations for interventions including oral cholera vaccine (OCV) use, and assess the impact of prevention and control strategies. Our data were instrumental in informing integrated prevention and control efforts during major cholera outbreaks in recent years.

Individual and Household Risk Factors for Symptomatic Cholera Infection: A Systematic Review and Meta-analysis

“…The majority of studies were both age- and sex-matched analyses, but a subset assessed risk for cholera based on these characteristics. Fifteen studies assessed differential risk by age [ 13 , 26 , 32 , 33 , 35 , 47 , 53 , 70 , 82 , 85 , 97 , 101 , 104 , 110 , 118 ], 10 of which implemented a multivariable model [ 26 , 32 , 33 , 35 , 82 , 97 , 101 , 104 , 110 , 118 ]. Age was measured heterogeneously and findings were mixed.…”

“…Four studies found that younger patients were at higher risk for cholera [ 26 , 35 , 101 , 110 ], and 1 study found that older children had higher risk than children <1 year old [ 32 ]. Other studies, however, found older people to be at higher risk [ 33 , 97 ], or no difference in risk based on age [ 82 , 104 , 118 ].…”

“…The role of sex was evaluated in 15 studies [ 13 , 26 , 32 , 33 , 35 , 37 , 43 , 70 , 72 , 73 , 90 , 101 , 104 , 116 , 118 ], of which 11 implemented a multivariable model [ 32 , 33 , 35 , 37 , 72 , 73 , 90 , 101 , 104 , 116 , 118 ]. Seven of these studies met criteria for meta-analysis ( Supplementary Figure 1 ) [ 35 , 72 , 73 , 90 , 101 , 116 , 118 ].…”

“…Fifty-six studies evaluated risk of cholera based on source of water, with many identifying a specific culprit [ 17–19 , 21 , 22 , 25–27 , 30 , 32 , 33 , 36 , 37 , 41–43 , 47 , 50 , 59–61 , 63 , 65 , 66 , 68 , 70 , 72–74 , 76 , 78 , 79 , 81–83 , 85–88 , 91 , 92 , 94 , 95 , 97 , 101–108 , 110 , 113 , 116 , 118 ]. Thirty-one studies implemented a multivariable model [ 19 , 26 , 27 , 30 , 32 , 33 , 36 , 37 , 41 , 42 , 60 , 65 , 66 , 68 , 72 , 73 , 78 , 79 , 82 , 83 , 86 , 91 , 92 , 94 , 95 , 101 , 104–106 , 116 , 118 ]. Of these, 20 measured the water source in a way that could be classified as improved (piped househol...…”

“…The SOR for risk of cholera with an unimproved water source was 3.48 (95% CI, 2.18–5.54), with I 2 = 77% ( Q test P < .00001). Of the 11 studies excluded from meta-analysis, 1 found a significant association but did not report an effect measure [ 92 ], 6 used hospital controls [ 32 , 33 , 60 , 68 , 104 , 105 ], and 4 did not define water source in way comparable to the others [ 26 , 42 , 79 , 110 ].…”