Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon. A case-finding retrospective observational study: 2006-2012

RESEARCH ARTICLE Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon. A casefinding retrospective observational study: 2006– 2012 John Njuma LibweaID 1,2 *, Sandrine Rachel Bebey Kingue 3,4 , Nadesh Taku AshukemID 3,5 , Marie Kobela 2,6 , Angeline Boula 3 , Koulla-Shiro Sinata 5,7 , Paul Koki Ndombo 3,6 1Faculty of Social Sciences, Health Sciences Unit, Tampere University, Tampere, Finland, 2Expanded Programme on Immunization, Yaounde ´, Cameroon, 3Mother & Child Centre (MCH), Chantal Biya Foundation, Yaounde ´, Cameroon, 4Faculty of Sciences, Department of Medical Microbiology, University of Yaounde ´1, Yaounde ´, Cameroon, 5Faculty of Sciences, Department of Microbiology, University of Buea, Buea, Cameroon, 6Faculty of Medicine and Biomedical Sciences, University of Yaounde ´1, Yaounde ´, Cameroon, 7Ministry of Public Health, Yaounde ´, Cameroon *[email protected]om Abstract Background Vital registration data outlining causes of deaths (CoD) are important for a sustainable health system, targeted interventions and other relevant policies. There is data paucity on vital registration systems in developing countries. We assessed the leading causes and proportions of under-five deaths, and particularly those related to pneumococcal infections in Yaounde ´, Cameroon, using hospital registration data. Methods A retrospective case-finding observational study design was used to access and identify data on 817 death cases in children under-five years of age recorded in health facilities in Yaounde ´, within the period January 1, 2006 and December 31, 2012. Patients’ files were randomly selected and needed information including demographic data, date of admission, clinical and laboratory diagnosis, principal and/or underlying causes of death were abstracted into structured case report forms. The International Classification of Diseases and Clinical Modifications 10 th revision (ICD-10-CM) codes (ICD10Data.com 2017 edition) were used to classify the different CoD, retrospectively. Ascertainment of CoD was based on medical report and estimates were done using the Kaplan-Meier procedure and descriptive statistics. Results Of the 817 death records assessed, malaria was the leading CoD and was responsible for 17.5% of cases. Meningitis was the second largest CoD with 11.0%; followed by sepsis PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 1 / 16 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Njuma Libwea J, Bebey Kingue SR, Taku Ashukem N, Kobela M, Boula A, Sinata K-S, et al. (2019) Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon. A casefinding retrospective observational study: 2006– 2012. PLoS ONE 14(4): e0212939. https://doi.org/ 10.1371/journal.pone.0212939 Editor: Yap Boum, II, Epicentre, CAMEROON Received: September 21, 2018 Accepted: February 12, 2019 Published: April 17, 2019 Copyright: ©2019 Njuma Libwea et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. (10.0%), Streptococcus pneumoniae infections (8.3%), malnutrition (8.3%), gastro-enteritis / diarrhoea (6.2%), anaemia (6.1%) and HIV (3.5%), respectively. Conclusion The main CoD in this population are either treatable or vaccine-preventable; a trend consistent with previous reports across developing countries. Besides, the health effects from non-communicable infections should not be neglected. Therefore, scaling-up measures to reduce causes of under-five deaths will demand sustainable efforts to enhance both treatment and disease prevention strategies, to avoid a decline in the progress towards reducing under-five deaths by 2/3 from the 1990 baseline. Background Accurate civil registration systems are essential in the documentation on the distribution of CoD in children as well as in the general population. This is important in the planning of sustainable health policies and needed interventions in concordance with the millennium development goal (MDG 4) [1], with a target of reducing by two-thirds between 1990 and 2015 the under-five mortality rate [2]. We have exceeded the year 2015 deadline and globally, reports suggest the number of under-five deaths dropped from 12.7 million in 1990 to 6.3 million in 2013 [3]. However, most of the 6.3 million reported under-five deaths occurred in developing countries, with porous data on child health and mortality [4]. In these countries, reaching the MDG4 target will require an acceleration of essential, effective and affordable interventions against diarrhea, sepsis, the human immuno-deficiency virus / acquired immuno-deficiency syndrome (HIV/AIDS), malaria and pneumonia including improved nutrition and vaccines access. The reduction of under-five mortality remains a major priority in developing countries, considering the high number of deaths resulting from preventable conditions [5]. However, to achieve the MDG4 target, reliable data are needed on under-five mortality to guide health planners and to scale up prevention and treatment strategies [5,6]. Acute Lower Respiratory Infections (ALRI) account for over 6.0% of morbidity and mortality of children worldwide, with Streptococcus pneumoniae reported as one of the principal causes of illness and death in children younger than five years of age [2]. About 75% of all cases occur in only 15 countries, with Sub-Saharan Africa and Southeast Asia representing the vast majority of cases [7]. Pneumococcal pneumonia is more frequent than can be confirmed by positive blood cultures and up to half of pneumonia deaths in children is attributed to pneumococcus [8]. In the absence of recent research data in the country, up to 19% of deaths in children under-five years old, have been estimated to result from pneumococcal infections in Cameroon with a total under-five mortality rate of 84 /1000 live births [9–11]. Addressing major risk factors such as malnutrition, breastfeeding and indoor air pollution are essential in prevention of pneumonia, but vaccination remains the cornerstone [12]. In this paper, we present data on hospital-based case finding of CoD in children abstracted into structured case report forms (CRF) in four health districts including the mother and child hospital (MCH) in Yaounde ´, Cameroon. The MCH is one of the largest children’s hospitals in the country and it is accessible and affordable to all strata of the population. It keeps records of hospital visits, admissions, and deaths, in addition to specific clinical, laboratory and serotype data on invasive diseases including Haemophilus influenzae and Streptococcus pneumoniae. It Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 2 / 16 has a capacity of 300 beds and had over 85000 admissions with 1816 reported under-five deaths between January 2006 and the end of December 2012. The principal causes of under-five mortality in the country have not been assessed and there is no baseline data for vaccine effectiveness evaluation, as is the case with the pneumococcal conjugate vaccine introduced in July 2011. Our primary goal was to access and identify all cases of under-five deaths in the study area and define the most probable causes of underfive deaths based on available data sources. Secondarily, we estimated the proportion of deaths possibly or definitely due to pneumococcus. Moreover, it is expected the findings may be a useful baseline with information on disease pattern needed to re-scale appropriate public health targets and indicators to measure their progress and achievements in the country, with respect to the MDG4. Material and methods Study design and study site We applied a retrospective case-finding observational study design using hospital registration data from the infectious disease surveillance sites hosted at the MCH in Yaounde ´, Cameroon (Fig 1). As earlier described [13], the sites involved hospitals in both an urban and rural/semiurban zones around Yaounde ´, with a population of over 3.5 million inhabitants out of which 18% are children under-five [14]. In 2013, we conducted a sample survey on over 1800 cases of death registered within the MCH, representing less than 2% of an estimated 117000 expected deaths in Yaounde ´[9], among children aged under-five between 2006 and 2012. Trained study personnel randomly selected 817 of these cases and information on the socio-demographic and CoD were extracted from medical reports and keyed into structured case report forms (CRF). This facilitated the identification of the direct, intermediate and underlying causes of CoD from medical reports. Besides, it avoided duplicate or repeated counting of cases from the different hospital registers. Ascertainment of CoD was based on medical declaration and the International Classification of Diseases and Clinical Modifications 10 th revision (ICD-10-CM) codes were used to retrospectively classify the different CoD. Ethical considerations Death registration data for the general population are held at Civil Status Registries (CSR) in municipalities and local population offices in each municipality in Cameroon, and permission for their use was obtained from the National Ethics Committee (CNE) No. 234/CNE/SE/2012, written and signed on May 2, 2012.(Because we needed to effectively link patients’ records in registers to their respective files and to avoid duplicate entries,they were accessed unencoded.All patient records abstracted into CRF were later encoded and analysed anonymously). This study was approved as one of the specific objectives in a broader protocol entitled "Estimating pneumococcal disease burden and evaluating the impact of introducing the pneumococcal conjugate vaccine (PCV13) into the Expanded Programme of Immunization in Cameroon." Data sources for death identification It is mandatory in the country that, in the case of death in a hospital or other medical institution or in a prison, the head of the establishment must declare the death within fifteen days [15]. The certification by medical personnel of the CoD is an essential step in the series of processes in the construction of vital registration data. In this setting, the customary practice in the registration of deaths commences with the issuance of a death declaration form (DDF) to Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 3 / 16 the family of the deceased by a qualified medical staff. The family is expected to transmit the DDF to the municipality for registration and be issued a death certificate. However, this last step is hardly respected and this generates an inherent problem with the quality of data in vital registrations systems witnessed in most of resource-low settings. The recommended practice is Fig 1. Flow chart on data identification and collection processes. N.B: For the sample size estimation, we assumed that, 18% of study population (630000) were children under-five years old; a = 0.05; power = 80%, and proportion of death cases with missing data on cause-specific death = 10%. Using the computer-based Creative Research Systems Survey software (http://www.surveysystem.com), we assumed a desired confidence level of 95% and a confidence interval of 4% units on each side; the estimated sample size for this study is 600 deaths. Therefore, a minimum of 660 cases of death in children aged 29days to 59 months was targeted as the sample size in this study. DH = District Hospital; ICD-10-CM = the International Classification of Diseases and Clinical Modifications 10th revision. https://doi.org/10.1371/journal.pone.0212939.g001 Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 4 / 16 to respect the World Health Organisation’s (WHO) criteria for medical certification; where the CoD are outlined in a sequential format beginning with the direct CoD through the intermediate causes, while the underlying CoD is registered in the lowest line of Part ׀of the medical certificate of cause of death [10]. The data search was limited only to hospital registers because we could not find sufficient information on under-five deaths from the civil status registers (vital registration systems) in municipalities and local population offices. Generally, typical hospital setting in the country consists of independent units and wards such as the female, male, paediatric, HIV/AIDS and oncology wards. Each keeps a register of patients’ medical records (usually manual but in cities both manual and electronic formats maybe available). Our search to identify under-five hospital deaths relied principally on inpatient paediatric hospital registers, as well as on other outpatient and emergency units registers (Fig 1). This included all causes of hospital documented under-five deaths within the study period. From these, further verifications were made for any cases of death with pneumococcal aetiology. Case definitions The CoD as described in hospital registers were retrospectively coded using the ICD-10-CM codes. Cause-specific mortality proportions were derived as the fraction of total deaths possibly associated to specific conditions, using hospital-based ICD-10-CM list diagnoses. Here, owing to the complexity associated with defining infections with pneumococcal aetiology, we limit our perspectives only on pneumococcal-related causes of deaths defined as follow: A): Clinical pneumococcal infections 1. ALRI: Diagnosis of pneumonia based on medical declaration as cause of death (i.e. any Acute Lower Respiratory Tract Infection (ALRI) or pneumonia including all diagnosis of ALRI); or 2. Clinically severe pneumonia: Death resulting from cough or difficult breathing as admission symptoms for child 1–59 months old residing in study area, And with either - A respiratory rate �40/minute, or temperature >38.5˚C, or refusing to feed, or vomiting and/or lower chest in-drawing B): Laboratory confirmed infections • Culture-positive invasive pneumococcal disease • Culture-negative polymerase chain reaction (PCR) or antigen test positive invasive pneumococcal disease C): Radiologically confirmed infections • Radiologically confirmed pneumonia i.e. Chest X-ray-Community Acquired Pneumonia compatible with endpoint consolidation • Radiologically confirmed pneumonia i.e. Chest X-ray-Community Acquired Pneumonia with any radiologic abnormality Clinical cases of pneumococcal infections, which did not meet these criteria, were considered as non-confirmed pneumococcal diseases (unspecified). Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 5 / 16 D): Causes of deaths not related to pneumococcus These included other biologically related causes of under-five deaths apart from those due to pneumococcal infections e.g. Tuberculosis (TB), HIV or malaria and others. E): Deaths due to injuries This consisted of under-five deaths with non-biological causes such as those resulting from injuries or accidents. Statistical considerations Data analysis. Cases included in this analysis were children aged between 29 days and 59 months (for uniformity we have used 29 days = 1 month, since no cases were registered as 29 or 30 days old), whose deaths were recorded between 1 st January 2006 and 31 st December 2012 in the study area. Deaths occurring in the neonatal period (i.e. 28 days following birth) were excluded. The date of birth (age in months) was considered as the entry point to the study in 2006. Follow-up time was up to 2012, which was the end of the study period, or at time of death reported. The proportion and causes of deaths were estimated from data collected on the total deaths using descriptive measures and Kaplan-Meier method. Data has been analysed using the software package SPSS 25.0 version. Results Distribution and main causes of death in children under-five years old A total of 817 randomly selected cases of under-five deaths from hospital registers that were recorded between January 2006 to December 2012 in Yaounde ´were assessed. The median ages at death were 11months for males and 15 months for females (Fig 2).Table 1 shows the age groups and distribution of the principal causes of death. The most vulnerable were those in the age group 1–11 months old, harbouring 45% of all-cause mortality. Malaria (17.5%) was the leading CoD followed by meningitis (11.0%), sepsis (10.0%), and pneumococcal diseases (8.3%). Causes of death registered with corresponding ICD-10-CM codes Table 2 shows a summary distribution of the 817 cases of under-five deaths classified by ICD10-CM coding. Communicable infections resulting from infectious and parasitic diseases, respiratory infections, nutritional deficiencies and perinatal conditions were accountable for most deaths (71.5%). Non-communicable infections including those from malignant disorders, cardiovascular disorders and congenital abnormalities contributed for some 11.6%, anaemia contributed to 6.2% of deaths and injuries were responsible for 0.2%. Comparison of leading CoD in our study with estimates from SubSaharan Africa (SSA) In Fig 3, are presented percentages of cause-specific deaths of children age 1–59 months obtained in our study compared to estimates from the SSA region [3]. Meningitis and sepsis were contributing more cause-specific deaths in our study population compared to the rest of the sub-region. The statistics were similar for malaria and HIV/AIDS. Some positives are observed with a decline in CoD resulting from tetanus, measles and perinatal conditions in our study. Deaths attributed to other and ill-defined causes were as much as threefold higher in our study compared to that of the entire SSA region [3]. Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 6 / 16 Clinical and laboratory diagnosed pneumococcal disease associated cause of death Table 3 presents the distribution of clinically diagnosed and laboratory confirmed pneumococcal disease associated CoD. Overall, more than 55% of cases (n = 158) diagnosed with clinically Fig 2. Cumulative proportion of age at death by gender among children under-five years registered at the Infectious Diseases Surveillance Sites in Yaounde ´, Cameroon: 2006–2012. https://doi.org/10.1371/journal.pone.0212939.g002 Table 1. Distribution of the main CoD in children 1–59 months old in Yaounde ´, 2006–2012 (N = 817). CoD Age groups 1–11 12–23 24–35 36–47 46–49 All (N = 368) (N = 187) (N = 104) (N = 55) (N = 103) (N = 817) % % % % % % (95%CI) Malaria 11.7 22.5 21.2 32.7 17.5 17.5 (15.0–20.3) Meningitis 11.7 10.2 16.3 5.5 7.8 11.0 (09.0–13.4) Sepsis 13.0 11.2 6.7 0.0 5.8 10.0 (08.1–12.3) Pneumonia 9.8 7.0 8.7 3.6 3.9 8.3 (06.5–10.4) Malnutrition 7.6 12.8 7.7 0.0 3.9 8.3 (06.5–10.4) Diarrhoea/gastro-enteritis 8.7 5.3 4.8 1.8 2.9 6.2 (04.7–08.1) Others 37.5 31.0 34.6 56.4 58.3 38.6 (35.2–42.0) CoD = Causes of death; N = number; % = percentage; CI = Confidence Interval https://doi.org/10.1371/journal.pone.0212939.t001 Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 7 / 16 severe pneumonia had been coughing, had a respiratory rate �40/minute and a body temperature of over 38˚C. Seventeen percent of these were confirmed as culture-positive invasive pneumococcal disease, and 73.4% were confirmed as antigen test-positive. Only 1.3% of these had a radiological confirmation. Discussion This study aimed to determine the leading CoD in children aged 1–59 months in the MCH and health facilities within the infectious disease surveillance network in Yaounde ´, Cameroon, using hospital-based registers. Vital statistics regarding the CoD among children based on death certificates are porous in resource-limited settings like Cameroon, which remain amongst the highest contributors to the number of under-five reported deaths [16]. In this study, we have assessed only about 2% of deaths of an estimated 117000 deaths expected to have occurred within the study period in Yaounde ´[9], justified by our intention to capture CoD with clinical and laboratory-culture specific data to depict deaths from pneumococcal related aetiology. Thus, we could not ascertain that these deaths were representative of all deaths within this age group at that period of time, but we remain optimistic they mirrored the major causes of morbidity and mortality in this community since they were medically declared. Based on the data we assessed, infectious diseases and malnutrition were the leading CoD in children aged 1–59 months within the study period in this population. Additionally, 71.5% of the deaths occurred within the first 24 months of life. However, our findings are in agreement with previous reports on the trends in the geographical distribution of CoD in the under-fives globally [16,17], although these studies had explored various methodologies than what we have used. This is partly due to the absence of a standardized universal algorithm to measure CoD. In a related study on the causes and circumstances of death in a district hospital in Kolofata, Northern Cameroon between 1993 to 2009; malaria (15.9%), infectious diseases including Table 2. List of ICD-10-CM codes collected for this study and percent of diagnostic CoD in children 1–59 months old in Yaounde ´, 2006–2012. No. ICD-10-CM�code Clinical diagnosis % 1. B54 Unspecified malaria 17.5 2. G00.1 Pneumococcal meningitis 11.0 3. A41.9 / B96.29 Sepsis 10.0 4. B95.3 / J13 Streptococcus pneumoniae as the cause of diseases classified elsewhere 8.3 5. C83.7/C95/D17.9/R59.1 Malignancy / Tumours/ Generalized lymph nodes 8.2 6E46 Unspecified protein-calorie malnutrition 8.3 7. K52.1/K52.89/R19.7 Gastro-enteritis / Diarrhoea 6.2 8. D61.2 Aplastic anaemia due to other external agents 6.1 9. P96.89 Other specified conditions originating in the perinatal period 4.4 10. B20 Human immunodeficiency virus (HIV) disease 3.5 11. A15.5 Tuberculosis of larynx, trachea and bronchus 2.3 12. I25.5 Ischemic heart diseases 2.3 13 D57.1 Sickle-cell disease without crisis 1.1 14. T14.90 Injury 0.2 15 R99/A35/B05.9/K75.9 Ill-defined and unknown cause of mortality and others 10.0 �ICD codes specific for diseases and compatible with clinical diagnosis as primary cause of death used in the study; No. = Serial number based on leading cause of death; S = Streptococcus; % = percent https://doi.org/10.1371/journal.pone.0212939.t002 Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 8 / 16 ALRI (15.1%) and diarrhoeal diseases (13.3%) were the leading CoD in children before the age of 5 years [18], a similar trend to that obtained in our study. The Kolofata study included 1281 inpatient deaths of which 46.9% were deaths before the age of five years. Although there was a decline in the under-five mortality rate in the country from 144/1000 live births reported in 2004 to 122/1000 live births in 2011 [9], country-specific data on the major causes of death are scarce [18]. Vital statistics with notifications on CoD in children are not maintained in Cameroon, despite legislative provisions. However, the 2011 Cameroon Demographic Health Survey (CDHS) documented important information on symptoms and other characteristics leading to the death for children aged under-five years old nationwide [9]. Based on the CDHS data, ALRI, particularly those with pneumococci aetiology had 7% prevalence in Yaounde ´, similar to the 8.3% we have obtained but differ with the 16% reported for the SSA region where diarrhoea (21%), malaria (30%), malnutrition (33%) and anaemia (60%) were the predominant CoD in children aged under-five years (Fig 3). In the CDHS questionnaires, to ascertain pneumococcal-related causes of deaths caretakers were asked the following; if the child was coughing in the last two weeks prior to death and if Fig 3. Frequent causes of deaths in our study with comparative percentages for the Sub-Saharan Africa region. https://doi.org/10.1371/journal.pone.0212939.g003 Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 9 / 16 28. Gessner BD, Mueller JE, Yaro S. African meningitis belt pneumococcal disease epidemiology indicates a need for an effective serotype 1 containing vaccine, including for older children and adults. BMC Infect Dis. 2010; https://doi.org/10.1186/1471-2334-10-22 PMID: 20146815 29. Al Khorasani A, Banajeh S. Bacterial profile and clinical outcome of childhood meningitis in rural Yemen: A 2-year hospital-based study. J Infect. 2006; https://doi.org/10.1016/j.jinf.2005.12.004 PMID: 16434101 30. Grimwood K, Anderson P, Anderson V, Tan L, Nolan T. Twelve year outcomes following bacterial meningitis: further evidence for persisting effects [Internet]. Arch Dis Child. 2000. www.archdischild.com 31. AAP 2015. 2015 Report of the Committee on Infectious Diseases RED BOOK. 2015. 32. Barreiro P, Tiziano G, Fano H, Yohannes T, Gosa A, Reyes F, et al. Malaria and severe anemia over eight years at Gambo Rural Hospital, southern Ethiopia. Pathog Glob Health. 2017; https://doi.org/10. 1080/20477724.2017.1322262 PMID: 28502227 33. Calis JCJ, Phiri KS, Faragher EB, Brabin BJ, Bates I, Cuevas LE, et al. Severe anemia in Malawian children. Malawi Med J. 2016; 28. https://doi.org/10.1056/NEJMoa072727 PMID: 18305266 34. Kortz TB, Sawe HR, Murray B, Enanoria W, Matthay MA, Reynolds T. Clinical Presentation and Outcomes among Children with Sepsis Presenting to a Public Tertiary Hospital in Tanzania. Front Pediatr. 2017; https://doi.org/10.3389/fped.2017.00278 PMID: 29312910 35. Kissoon N, Carcillo JA, Espinosa V A A et al. World Federation of Pediatric Intensive Care and Critical Care Societies: Global Sepsis Initiative. Pediatr Crit Care Med. 2011; 12: 494–503. https://doi.org/10. 1097/PCC.0b013e318207096c PMID: 21897156 36. Maitland K, George EC, Evans JA, Kiguli S, Olupot-Olupot P, Akech SO, et al. Exploring mechanisms of excess mortality with early fluid resuscitation: Insights from the FEAST trial. BMC Med. 2013; https:// doi.org/10.1186/1741-7015-11-68 PMID: 23496872 37. Bang AT, Bang RA. Diagnosis of causes of childhood deaths in developing countries by verbal autopsy: suggested criteria. 1992. 38. Roca A, Sigauque B, Quinto L, Mandomando I, Valles X, Espasa M, et al. Invasive pneumococcal disease in children less than 5 years of age in rural Mozambique. Trop Med Int Heal. 2006; https://doi.org/ 10.1111/j.1365-3156.2006.01697.x PMID: 16930265 39. Rice AL, Sacco L, Hyder A, Black RE. Malnutrition as an underlying cause of childhood deaths associated with infectious diseases in developing countries. Bull World Health Organ. 2000; 40. Delpeuch F, Martin-Pre ´vel Y, Foue ´re ´T, Traissac P, Mbemba F, Ly C, et al. L’alimentation de comple ´- ment du jeune enfant après la de ´valuation du franc CFA: deux e ´tudes de cas en milieu urbain, au Congo et au Se ´ne ´gal. Bull World Health Organ. 1996; 74: 67–75. 41. Tambe A, Nzefa L, Nicoline N. Childhood Diarrhea Determinants in Sub-Saharan Africa: A Cross Sectional Study of Tiko-Cameroon. Challenges. 2015; https://doi.org/10.3390/challe6020229 42. Tadesse M, Getachew T, Amenu K, Asfaw E, Bekele A, Kebede A et al. Countdown To 2015: Ethiopia ‘ S Progress Towards Reduction in Under-Five Mortality. 2015. Assessing the causes of under-five mortality and proportion associated with pneumococcal diseases in Cameroon PLOS ONE | https://doi.org/10.1371/journal.pone.0212939 April 17, 2019 16 / 16