Original Article

Beyond Malaria — Causes of Fever in Outpatient Tanzanian Children

List of authors.
  • Valérie D'Acremont, M.D., Ph.D.,
  • Mary Kilowoko, M.P.H.,
  • Esther Kyungu, M.D., M.P.H.,
  • Sister Philipina, R.N.,
  • Willy Sangu, A.M.O.,
  • Judith Kahama-Maro, M.D., M.P.H.,
  • Christian Lengeler, Ph.D.,
  • Pascal Cherpillod, Ph.D.,
  • Laurent Kaiser, M.D.,
  • and Blaise Genton, M.D., Ph.D.*

Abstract

Background

As the incidence of malaria diminishes, a better understanding of nonmalarial fever is important for effective management of illness in children. In this study, we explored the spectrum of causes of fever in African children.

Methods

We recruited children younger than 10 years of age with a temperature of 38°C or higher at two outpatient clinics — one rural and one urban — in Tanzania. Medical histories were obtained and clinical examinations conducted by means of systematic procedures. Blood and nasopharyngeal specimens were collected to perform rapid diagnostic tests, serologic tests, culture, and molecular tests for potential pathogens causing acute fever. Final diagnoses were determined with the use of algorithms and a set of prespecified criteria.

Results

Analyses of data derived from clinical presentation and from 25,743 laboratory investigations yielded 1232 diagnoses. Of 1005 children (22.6% of whom had multiple diagnoses), 62.2% had an acute respiratory infection; 5.0% of these infections were radiologically confirmed pneumonia. A systemic bacterial, viral, or parasitic infection other than malaria or typhoid fever was found in 13.3% of children, nasopharyngeal viral infection (without respiratory symptoms or signs) in 11.9%, malaria in 10.5%, gastroenteritis in 10.3%, urinary tract infection in 5.9%, typhoid fever in 3.7%, skin or mucosal infection in 1.5%, and meningitis in 0.2%. The cause of fever was undetermined in 3.2% of the children. A total of 70.5% of the children had viral disease, 22.0% had bacterial disease, and 10.9% had parasitic disease.

Conclusions

These results provide a description of the numerous causes of fever in African children in two representative settings. Evidence of a viral process was found more commonly than evidence of a bacterial or parasitic process. (Funded by the Swiss National Science Foundation and others.)

Introduction

Digital Object ThumbnailQUICK TAKE
Beyond Malaria
 02:24

With malaria transmission declining in many parts of Africa,1,2 there is increasing awareness that most acute febrile episodes are due to other infectious diseases — some of which are life-threatening — that must be identified and treated appropriately.3,4 World Health Organization (WHO) guidelines recommend malaria testing for all patients with febrile illness in areas where malaria is endemic.5 The advent of point-of-care rapid diagnostic tests for malaria presents health care providers with a new and daunting challenge: the need to determine the causes of febrile illness and the appropriate course of action when treating children who test negative for malaria. Studies have shown that the decrease in the use of antimalarial therapy has been accompanied by a commensurate increase in antibiotic use.6-8 Stakeholders have thus recognized the need for an improved understanding of nonmalarial fever, both to optimize care and to reduce inappropriate antibiotic use.9-12 Although several studies have examined nonmalarial fever in hospitalized patients, in particular those with bacteremia,13-17 to our knowledge, none have included extensive algorithm-based screening to assess the relative distribution of bacterial, parasitic, and viral agents among acutely febrile sub-Saharan African children attending an outpatient clinic.

Methods

Study Sites and Populations

During the period from April through August 2008, we recruited children at the outpatient clinic of the Amana District Hospital in Dar es Salaam, an area of low endemicity for malaria, with infection rates in the community of roughly 1 to 4%18 and only 6 to 12% of febrile patients found to have plasmodium parasitemia.8 During the period from June through December 2008, we recruited patients at the outpatient clinic of the St. Francis Designated District Hospital in Ifakara, Kilombero District, Tanzania. Ifakara was once highly endemic for malaria, but at the time of the study, only about 7% of febrile patients had parasitemia. The Expanded Program on Immunization in Tanzania included bacille Calmette–Guérin, diphtheria–tetanus–pertussis, poliovirus, hepatitis B, and measles vaccines (but neither pneumococcal nor Haemophilus influenzae vaccine), and coverage was between 80% and 90%. A 3-day national integrated campaign of measles vaccination, vitamin A supplementation, and distribution of deworming tablets took place in August 2008. The mortality rate among children between 1 and 5 years of age in 2008 was 8.9 per 1000 person-years in the Kilombero and Ulanga districts.19

Study Participants

Consecutive patients 2 months to 10 years of age presenting with an axillary temperature of 38°C or higher were screened for study inclusion; those who had severe malnutrition or were categorized as having emergency signs (requiring immediate livesaving procedures) according to the WHO Emergency Triage Assessment and Treatment criteria20 were excluded. Infants younger than 2 months of age were also excluded, since infectious agents and thus therapeutic approaches can differ substantially between children in this age group and older children, as reflected in WHO Integrated Management of Childhood Illness guidelines. Four additional inclusion criteria were assessed: the patient's visit had to be the first for the present illness, the fever duration had to be 1 week or less, the chief reason for the clinic visit could not be injury or trauma, and the patient could not have received antimalarial agents or antibiotics during the preceding week. Details of the study were discussed with the parents or guardians of patients meeting the inclusion criteria; the parents or guardians provided written informed consent. The protocol and related documents were approved by the regional ethics committee in Basel, Switzerland, and by the national ethics committee in Tanzania. All authors vouch for the accuracy and completeness of the data and the fidelity of the study to the protocol. Patients were provided immediate supportive care and treatment for each treatable disease that was identified. The procedures followed were in accordance with the ethical standards set by the two ethics committees and with the Declaration of Helsinki.

Study Procedures

Clinical Assessment

To record clinical findings, investigators used a standardized case-report form in which items relevant to a febrile presentation were listed, among them the chief reason for the visit, 23 symptoms and their respective durations, travel history, contact with sick persons, known chronic conditions, and 49 clinical signs. If the Integrated Management of Childhood Illness clinical criteria for a suspected human immunodeficiency virus (HIV) infection were met, voluntary HIV testing was recommended to the child's guardian. If a clinical or laboratory diagnosis could not be made at the initial visit, a follow-up visit was scheduled for day 7, when an identical, full clinical and laboratory assessment was performed again for patients who had a persistent fever or new symptoms.

Microbiologic and Other Investigations

In all cases, blood samples (5 ml) and pooled nasal and throat swabs were collected for microbiologic testing (cultures and rapid tests) on site and further serologic and molecular analysis in Switzerland and the United States. A systematic set of investigations was performed for all patients (Fig. S1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). Additional tests and assessments were performed according to several non–mutually exclusive decision charts developed for the chief symptoms (Fig. S1 in the Supplementary Appendix). Methods for all laboratory analyses, as well as radiologic criteria for pneumonia, are reported in the Methods section in the Supplementary Appendix. For each pathogen, the most appropriate method of detection for the diagnosis of an acute episode of fever was selected.

Definitive Diagnosis Based on Predefined Criteria

For each patient, the final diagnosis (in some cases, more than one) was established on the basis of a set of predefined clinical and microbiologic criteria derived from WHO and Infectious Diseases Society of America guidelines as well as from systematic reviews (Table S1 in the Supplementary Appendix). To maximize consistency and to minimize subjectivity, the diagnosis was generated by means of a computer-based algorithm. A designation “severe” was added when the child's condition fulfilled WHO criteria for this state.21 An acute respiratory infection was defined as any acute infection (≤1 week in duration) that was manifested as at least one respiratory sign or symptom and was localized to the upper or lower respiratory tract. Systemic infection was defined as a febrile illness, without localizing symptoms or signs, that was due to a parasite other than malaria, a bacterium other than Salmonella enterica serovar Typhi, or a virus and was documented by serologic testing or a polymerase-chain-reaction (PCR) assay performed on a blood specimen.

Because of the high rate of viral infection among asymptomatic children,22,23 viruses detected by PCR assay performed on nasopharyngeal swab specimens were deemed the cause of the febrile illness only when the patient presented with concomitant respiratory symptoms or signs and no other infectious agent was identified. Patients who had virus detected in nasopharyngeal specimens and who had no overt respiratory symptoms or signs and no other identified disease were given a diagnosis of nasopharyngeal viral infection. Because of its high colonization rates among children,24,25 Streptococcus pneumoniae identified by PCR assay of nasopharyngeal specimens was deemed the cause of illness only in cases of radiologically confirmed pneumonia or severe acute respiratory infection. For details of the study design, see the protocol, available at NEJM.org.

Statistical Analyses

Simple proportions were used for most analyses, and differences among groups were evaluated by means of a chi-square test. Data were entered into Epi Info, version 3.5.1 (Centers for Disease Control and Prevention). All statistical analyses were performed using Stata software, version 10 (StataCorp).

Results

Study Sample

Table 1. Table 1. Demographic and Clinical Characteristics of the Children, According to Study Site.

A total of 1010 febrile children (510 in Dar es Salaam and 500 in Ifakara) were initially enrolled; 5 were subsequently excluded from the study (2 for whom consent was withdrawn, 1 who did not have blood drawn, and 2 who did not meet inclusion criteria). Demographic and clinical characteristics of the 1005 children included in the study are described in Table 1. For 63.0% of patients in Dar es Salaam and 71.5% of patients in Ifakara, fever was the main symptom cited by the parent or guardian. In total, 25,743 biologic tests were performed, and 200 chest radiographs were obtained (Fig. S1 in the Supplementary Appendix).

Diagnoses

Figure 1. Figure 1. Distribution of All 1232 Diagnoses among 1005 Febrile Children at Two Sites in Tanzania.

Numbers are percentages of all diagnoses. Percentages may not sum to 100 because of rounding.

Overall, 1232 distinct clinical or microbiologic diagnoses were made (Figure 1). Acute respiratory infection was the most frequent diagnosis, made in 625 of 1005 patients (62.2%). Infections were localized to the upper respiratory tract in 439 patients (262 with tracheobronchitis, 141 with rhinitis, 24 with acute otitis media, 11 with nonstreptococcal tonsillitis, and 1 with pharyngitis) and to the lower respiratory tract in 186 patients (35 with bronchiolitis, 120 with non–radiologically confirmed pneumonia, and 31 with radiologically confirmed pneumonia). Other diagnoses were distributed as follows: 134 of 1005 (13.3%) children had a systemic infection, 120 (11.9%) had nasopharyngeal viral infection, 105 (10.5%) had malaria, 103 (10.3%) had gastroenteritis, 59 (5.9%) had urinary tract infection, 37 (3.7%) had typhoid fever, 15 (1.5%) had skin or mucosal infection, and 2 (0.2%) had meningitis. In 32 patients (3.2%), no potential infectious cause of illness was found.

Figure 2. Figure 2. Most Common Diagnoses According to Age Group.

Figure 2 shows the distribution of diseases according to age group. Urinary tract infection was predominantly diagnosed in the youngest children (<3 years of age in 95% of cases), whereas typhoid fever was diagnosed primarily in older children (≥1 year of age in 95% of cases), with a prevalence that increased with increasing age. The prevalence of malaria was highest among children older than 3 years of age, as expected in an area where malaria is moderately endemic. Infection with human herpesvirus 6 was mainly acquired in the first year of life. Acute respiratory infection was frequent in all age groups, but the frequency decreased slightly with increasing age. Diagnoses according to month and study site and differences in diagnoses between the urban and rural sites are provided in the Supplementary Appendix (Results section and Fig. S2).

Diagnoses in Severe Cases

Figure 3. Figure 3. Distribution of All 160 Diagnoses among 133 Febrile Children with Severe Illness.

Numbers are percentages of all diagnoses. The distribution of features of severe illness was as follows: 42 cases of jaundice, 37 of respiratory distress, 27 of impaired consciousness, 19 of seizures, 14 of severe anemia (hemoglobin level, <5 g per deciliter), 9 of severe dehydration, 7 of marasmus, 5 of cardiovascular failure, 4 of plasmodium hyperparasitemia, 2 of renal failure, and 2 of meningismus. Percentages may not sum to 100 because of rounding.

Overall, the conditions of 133 of the 1005 patients (13.2%) fulfilled the criteria for severe disease. A total of 160 illnesses were diagnosed in these patients, as shown in Figure 3. Malaria, radiologically confirmed pneumonia, and typhoid fever were more common among patients with severe illness than among those with mild illness (P<0.001, P=0.008, and P=0.01, respectively). Of the 133 severely ill patients, 70 (52.6%) were subsequently hospitalized. Four children died: one from concurrent malaria and typhoid fever, one from sepsis due to H. influenzae, one from meningitis, and one from acute respiratory infection associated with coronavirus and pneumococcal infection.

Multiple Diagnoses

A total of 227 children (22.6%) had multiple conditions diagnosed. Of the 105 patients with malaria, 54 (51.4%) had a concomitant disease: 43 had acute respiratory infection (31 of whom had upper respiratory tract infection, 3 had bronchiolitis, 8 had non–radiologically confirmed pneumonia, and 1 had radiologically confirmed pneumonia), 4 had typhoid fever, 2 had gastroenteritis, 2 had urinary tract infection, 2 had purulent skin or mucosal infection, 4 had systemic viral disease, and 1 had bacteremia. The percentage of patients with multiple diagnoses was similar among those with a blood parasite density of less than 2000 parasites per cubic millimeter and those with a density of 2000 parasites per cubic millimeter or more (57.1% and 50.6%, respectively; P=0.65). Patients with malaria who had any parasite density were half as likely to have pneumonia (documented radiographically or not) as patients without malaria (P=0.05). The prevalence of typhoid fever was the same among patients with malaria and patients without malaria (4% in each group, P=0.94).

Diseases According to Type of Pathogen

Figure 4. Figure 4. Overlap among Disease Types and Pathogen Types in the 1005 Febrile Children.

Panel A shows all diseases diagnosed in the present study; disease definitions are given in Table S1 in the Supplementary Appendix. Panel B shows types of pathogens identified, regardless of the clinical and laboratory criteria used to establish final diagnoses.

The distribution of illnesses by pathogen type (bacterial, viral, or parasitic) is presented in Figure 4A. Figure 4B shows the overlap among all pathogens identified, irrespective of the clinical and laboratory criteria used to determine final diagnoses. The numbers and proportions of children infected with each individual pathogen are presented in Table S2 in the Supplementary Appendix.

Viral Diseases

Among the 625 patients with clinical diagnoses of acute respiratory infection, PCR results were positive for at least one virus in 504 (81%): 344 of the 439 children with upper respiratory tract infection (78%), 29 of the 35 with bronchiolitis (83%), 107 of the 120 with non–radiologically confirmed pneumonia (89%), and 24 of the 31 with radiologically confirmed pneumonia (77%). A virus was documented in stool specimens from 31 of 103 patients with gastroenteritis (30.1%). Evidence of recent infection with human herpesvirus 6 and parvovirus B19 was found in 79 (7.9%) and 13 (1.3%) of the 1005 children, respectively. IgM antibodies for cytomegalovirus and Epstein–Barr virus were detected in 8 and 2 of the 199 tested children, respectively, and IgM for hepatitis A virus and hepatitis E virus was present in 1 and none of the 43 children with elevated liver enzyme levels. No dengue, chikungunya, West Nile, or Rift Valley fever virus was detected in the 199 children tested by means of PCR assay. Of the 120 patients with virus detected in nasopharyngeal specimens who had no clinical or laboratory findings beyond fever (i.e., those with nasopharyngeal viral infections), the virus was deemed to be possibly causal in at least 51%, because influenza, adenovirus, and enterovirus are known to produce fever without respiratory or other localized symptoms.26-28

Bacterial Diseases

Blood culture was performed with samples from 424 (83.6%) of the 507 children enrolled at the Dar es Salaam site only. Of these patients, 18 (4.2%) were positive for noncontaminant bacteria, 6 of whom had severe disease: 5 were positive for gram-positive cocci (3 for Staphylococcus aureus, 1 for Streptococcus pneumoniae, and 1 for an undetermined streptococcus species), and 13 were positive for gram-negative rods (4 for Salmonella enterica serovar Typhi, 2 for Escherichia coli, 1 for Acinetobacter baumanii, 1 for an undetermined acinetobacter species, 1 for Aeromonas hydrophila, 1 for H. influenzae, 1 for Pseudomonas aeruginosa, 1 for non-Typhi salmonella, and 1 for Shigella flexneri). Bacteremia originated in the gastrointestinal tract of 5 patients (including 4 with typhoid fever) and in the urinary tract of 2 patients, whereas originating sites in the other 11 cases of bacteremia could not be elucidated. Urinary tract infections were identified in 59 of 144 children (41.0%) tested by urine culture. Among the 54 children tested by stool culture, Shigella flexneri was identified in 11 (20%), and Salmonella enterica serovar Typhi was identified in 1 (2%). Of the 199 children for whom serologic testing for bacteria was indicated, 15 had evidence of an acute infection (6 with Rickettsia typhi, 4 with R. conorii or R. africae, 2 with Leptospira interrogans serovar icterohaemorrhagiae, 2 with L. interrogans serovar hardjo, and 1 with Coxiella burnetii). Of the 31 children with radiologically confirmed pneumonia and the 67 children with severe acute respiratory infection, 29 (94%) and 59 (88%), respectively, had PCR results that were positive for Streptococcus pneumoniae.

Parasitic Diseases

A total of 105 children had malaria, 2 children had toxoplasmosis, and 2 children had amoebic gastroenteritis.

Discussion

Using a predefined algorithm applied to an extensive set of clinical, laboratory, and radiologic data, we were able to identify the possible cause of fever in 97% of the 1005 children included in our study. Of these children, 88% had evidence of infection with at least one microbe. More than half presented with acute respiratory infection; a virus was identified in 81% of these children.

Given the seasonal pattern of respiratory viral infections, malaria, and other vectorborne diseases, different results could have been obtained if the study had been conducted in another region or country or during another season or year. Because the two study sites are representative of urban and rural Tanzania, our findings can be generalized to the nation as a whole and, to an extent, to other countries in sub-Saharan Africa. Of course, if the rate of malaria transmission were higher in other regions, malaria would be expected to account for a higher proportion of all causes of fever. The same applies to potential epidemics of any type of disease, which would temporarily reduce the relative proportions of other diseases in that particular area. In the absence of any particular outbreak, common childhood viral diseases are as frequent in Africa as in any other continent. More studies with a similar comprehensive clinical, radiologic, and laboratory approach are required in different epidemiologic contexts to generate additional evidence.

The decision to perform diagnostic tests in predefined subsets of patients with a higher pretest probability of a given infection than in the overall study sample can be considered a potential limitation of our study. However, this increased the clinical relevance of our microbiologic investigations by increasing the probability that the identified infection was the cause of illness.

Our findings indicate that among preschool-age children, a high percentage of lower respiratory tract diseases are likely to be of viral origin29,30: 81% of patients with upper respiratory tract infection, bronchiolitis, or non–radiologically confirmed pneumonia had evidence of a virus in their nasopharyngeal samples, and this was also seen in 77% of patients with radiologically confirmed pneumonia. It is known that respiratory viruses, including rhinoviruses,31 can cause lower respiratory tract disease, which may be severe, particularly in infants and young children with preexisting medical conditions. In fact, the presence of viral–bacterial coinfection may well be a major trigger for the development of severe pneumonia.30,32 These findings indicate that less than 13% of the children with acute respiratory infection required treatment with antibiotics.

The substantial difference between Figure 4A (showing diseases diagnosed on the basis of predefined clinical and laboratory criteria) and Figure 4B (showing all pathogens identified in the children but not necessarily considered a cause of the febrile episode) underscores the limitations inherent in a diagnosis made on the basis of laboratory findings alone. It is essential to combine laboratory findings with detailed clinical information and, when available, quantitative pathogen loads in order to differentiate between colonization or infection and disease. This is especially important when new, highly sensitive molecular methods are used. However, these new tools provide the advantage of identifying agents with previously unrecognized pathogenic potential, revealing the complexity of infectious diseases in this particular clinical situation. Multiple diagnoses were made in almost a quarter of patients, and multiple infections were found in two thirds of the children, which reflects the difficulty in providing a simple diagnosis for a child's condition. The combination of multiple infections is likely to have led to pathogen interactions, which may have contributed to the clinical condition observed in many of these patients. Diagnostic tools should be considered in light of the clinical characteristics of the targeted population. A greater understanding is needed regarding the diagnostic value of different signs and tests, the disease course in patients presenting with multiple infections, and the potential diagnostic thresholds for new molecular tests, such as cutoffs for pathogen loads.

In Dar es Salaam and Ifakara, most of the febrile children probably had a viral illness and thus required neither an antimalarial agent nor an antibiotic. These findings support the suggestion that sending such children home without antimalarial treatment33-35 or antibiotic treatment35,36 is reasonable in the absence of severe clinical signs. The systematic provision of antibiotic treatment in children with nonmalarial fever is unlikely to provide a clinical benefit in most cases. Rather, it exposes the children to unnecessary adverse events and exposes the community to accelerated development of antimicrobial resistance. The diversity of the causes of fever, most of which cannot be diagnosed on clinical grounds alone, calls for the development of point-of-care tests.

Funding and Disclosures

Supported by grants from the Swiss National Science Foundation (3270B0-109696, to Dr. Lengeler; IZ70Z0-124023, to Dr. Genton; and 32003B_127160, to Dr. Kaiser), the Commission for International Medicine of the University Hospital of Geneva, and the Bill and Melinda Gates Foundation.

This article is dedicated to the late Judy Kahama, our dear friend and colleague.

Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.

We thank Drs. Elizeus Kahigwa, Bernard Vaudaux, and Mario Gehri for their advice on treatment of pediatric conditions; Hassan Mshinda, director of the Ifakara Health Institute (IHI), Honorati Urassa, head of the Ifakara branch IHI, and Pascience Kibatala, director of St. Francis Hospital, for supporting the study in Ifakara; Paschal Mgaya and Burton Ngewe for assistance with patient enrollment and assessment, including drawing blood; Constantine Mzava, Felista Ngulubayi, Hassan Njaruka, Jabir Muhsin, John Wigayi, Sebastian Kobero, and Josephine Mgaya for performing or supervising the laboratory investigations; Rehema Adam, Jane Mallya, and Walburga Ndogoti for screening the patients before inclusion; Catherine Henry for drawing blood samples; Patrick Mneney and Gwakisa Mwaipungu for performing radiography; Febronia Ndogoti for transporting samples; all the laboratory technicians involved in sample testing in Geneva, in particular Laurence Morandi, Marta Santini, and Lorena Sacco; Jorge Vidal and Keith Klugman for performing the PCR assay for S. pneumoniae and providing advice regarding how to interpret the results; Olivier Peter for performing the serologic testing for rickettsia and coxiella; Katja Reitt for performing the serologic testing for leptospira; Leonor Alamo for reading the chest radiographs; Tom Smith for his useful comments on an earlier draft of the manuscript; and Angela Huttner for assistance with editing the text of an earlier version of the manuscript. Finally, we thank all patients and caregivers who participated in the study.

Author Affiliations

From the Swiss Tropical and Public Health Institute and University of Basel, Basel (V.D., J.K.-M., C.L., B.G.), the Department of Ambulatory Care and Community Medicine, University of Lausanne (V.D., B.G.), and the Infectious Diseases Service, University Hospital (B.G.), Lausanne, and the Laboratory of Virology, Division of Infectious Diseases and Division of Laboratory Medicine, University Hospital of Geneva, and Faculty of Medicine, University of Geneva, Geneva (P.C., L.K.) — all in Switzerland; the City Medical Office of Health, Dar es Salaam City Council (V.D., J.K.M.), and Amana Hospital (M.K., W.S.), Dar es Salaam, Ifakara Health Institute, Dar es Salaam and Ifakara (B.G.), and St. Francis Hospital, Ifakara (E.K., S.P.) — all in Tanzania.

Address reprint requests to Dr. D'Acremont at .

Judith Kahama-Maro, M.D., M.P.H., is deceased.

Supplementary Material

References (36)

  1. 1. D'Acremont V, Lengeler C, Genton B. Reduction in the proportion of fevers associated with Plasmodium falciparum parasitaemia in Africa: a systematic review. Malar J 2010;9:240-240

  2. 2. Gething PW, Patil AP, Smith DL, et al. A new world malaria map: Plasmodium falciparum endemicity in 2010. Malar J 2011;10:378-378

  3. 3. Makani J, Matuja W, Liyombo E, Snow RW, Marsh K, Warrell DA. Admission diagnosis of cerebral malaria in adults in an endemic area of Tanzania: implications and clinical description. QJM 2003;96:355-362

  4. 4. Reyburn H, Mbatia R, Drakeley C, et al. Overdiagnosis of malaria in patients with severe febrile illness in Tanzania: a prospective study. BMJ 2004;329:1212-1212

  5. 5. Guidelines for the treatment of malaria. 2nd ed. Geneva: World Health Organization, 2010.

  6. 6. Msellem MI, Martensson A, Rotllant G, et al. Influence of rapid malaria diagnostic tests on treatment and health outcome in fever patients, Zanzibar: a crossover validation study. PLoS Med 2009;6:e1000070-e1000070

  7. 7. Mosha JF, Conteh L, Tediosi F, et al. Cost implications of improving malaria diagnosis: findings from north-eastern Tanzania. PLoS One 2010;5:e8707-e8707

  8. 8. D'Acremont V, Kahama-Maro J, Swai N, Mtasiwa D, Genton B, Lengeler C. Reduction of anti-malarial consumption after rapid diagnostic tests implementation in Dar es Salaam: a before-after and cluster randomized controlled study. Malar J 2011;10:107-107

  9. 9. Baiden F, Webster J, Owusu-Agyei S, Chandramohan D. Would rational use of antibiotics be compromised in the era of test-based management of malaria? Trop Med Int Health 2011;16:142-144

  10. 10. Perkins MD, Bell DR. Working without a blindfold: the critical role of diagnostics in malaria control. Malar J 2008;7:Suppl 1:S5-S5

  11. 11. English M, Scott JAG. What is the future for global case management guidelines for common childhood diseases? PLoS Med 2008;5:e241-e241

  12. 12. Lubell Y, Staedke SG, Greenwood BM, et al. Likely health outcomes for untreated acute febrile illness in the tropics in decision and economic models; a Delphi survey. PLoS One 2011;6:e17439-e17439

  13. 13. Berkley JA, Lowe BS, Mwangi I, et al. Bacteremia among children admitted to a rural hospital in Kenya. N Engl J Med 2005;352:39-47

  14. 14. Brent AJ, Ahmed I, Ndiritu M, et al. Incidence of clinically significant bacteraemia in children who present to hospital in Kenya: community-based observational study. Lancet 2006;367:482-488

  15. 15. Punjabi NH, Taylor WRJ, Murphy GS, et al. Etiology of acute, non-malaria, febrile illnesses in Jayapura, northeastern Papua, Indonesia. Am J Trop Med Hyg 2012;86:46-51

  16. 16. Mtove G, Amos B, Nadjm B, et al. Decreasing incidence of severe malaria and community-acquired bacteraemia among hospitalized children in Muheza, north-eastern Tanzania, 2006-2010. Malar J 2011;10:320-320

  17. 17. Crump JA, Ramadhani HO, Morrissey AB, et al. Invasive bacterial and fungal infections among hospitalized HIV-infected and HIV-uninfected children and infants in northern Tanzania. Trop Med Int Health 2011;16:830-837

  18. 18. Wang S-J, Lengeler C, Mtasiwa D, et al. Rapid Urban Malaria Appraisal (RUMA) II: epidemiology of urban malaria in Dar es Salaam (Tanzania). Malar J 2006;5:28-28

  19. 19. Alba S, Nathan R, Schulze A, Mshinda H, Lengeler C. Child mortality patterns in rural Tanzania: an observational study on the impact of malaria control interventions. Int J Epidemiol 2013 December 19 (Epub ahead of print).

  20. 20. Emergency Triage Assessment and Treatment (ETAT) course. Geneva: World Health Organization, 2005.

  21. 21. Hospital care for children: guidelines for the management of common illnesses with limited resources. Geneva: World Health Organization, 2005.

  22. 22. Berkley JA, Munywoki P, Ngama M, et al. Viral etiology of severe pneumonia among Kenyan infants and children. JAMA 2010;303:2051-2057[Erratum, JAMA 2010;303:2477.]

  23. 23. Feikin DR, Njenga MK, Bigogo G, et al. Viral and bacterial causes of severe acute respiratory illness among children aged less than 5 years in a high malaria prevalence area of western Kenya, 2007-2010. Pediatr Infect Dis J 2013;32:e14-e19

  24. 24. Abdullahi O, Karani A, Tigoi CC, et al. The prevalence and risk factors for pneumococcal colonization of the nasopharynx among children in Kilifi District, Kenya. PLoS One 2012;7:e30787-e30787

  25. 25. Kwambana BA, Barer MR, Bottomley C, Adegbola RA, Antonio M. Early acquisition and high nasopharyngeal co-colonisation by Streptococcus pneumoniae and three respiratory pathogens amongst Gambian new-borns and infants. BMC Infect Dis 2011;11:175-175

  26. 26. Calvo Rey C, Garcia Garcia ML, Casas Flecha I, Martin del Valle F, Centeno Jimenez M, Perez-Brena P. Influenza virus infections in infants aged less than two years old. An Pediatr (Barc) 2005;63:22-28

  27. 27. Selvaraju SB, Kovac M, Dickson LM, Kajon AE, Selvarangan R. Molecular epidemiology and clinical presentation of human adenovirus infections in Kansas City children. J Clin Virol 2011;51:126-131

  28. 28. Stalkup JR, Chilukuri S. Enterovirus infections: a review of clinical presentation, diagnosis, and treatment. Dermatol Clin 2002;20:217-223

  29. 29. Brodzinski H, Ruddy RM. Review of new and newly discovered respiratory tract viruses in children. Pediatr Emerg Care 2009;25:352-360

  30. 30. Cevey-Macherel M, Galetto-Lacour A, Gervaix A, et al. Etiology of community-acquired pneumonia in hospitalized children based on WHO clinical guidelines. Eur J Pediatr 2009;168:1429-1436

  31. 31. Peltola V, Waris M, Osterback R, Susi P, Hyypia T, Ruuskanen O. Clinical effects of rhinovirus infections. J Clin Virol 2008;43:411-414

  32. 32. Madhi SA, Ludewick H, Kuwanda L, et al. Pneumococcal coinfection with human metapneumovirus. J Infect Dis 2006;193:1236-1243

  33. 33. D'Acremont V, Malila A, Swai N, et al. Withholding antimalarials in febrile children who have a negative result for a rapid diagnostic test. Clin Infect Dis 2010;51:506-511

  34. 34. Senn N, Rarau P, Manong D, et al. Rapid diagnostic test-based management of malaria: an effectiveness study in Papua New Guinean infants with Plasmodium falciparum and Plasmodium vivax malaria. Clin Infect Dis 2012;54:644-651

  35. 35. Hamer DH, Brooks ET, Semrau K, et al. Quality and safety of integrated community case management of malaria using rapid diagnostic tests and pneumonia by community health workers. Pathog Glob Health 2012;106:32-39

  36. 36. Hazir T, Nisar YB, Abbasi S, et al. Comparison of oral amoxicillin with placebo for the treatment of World Health Organization-defined nonsevere pneumonia in children aged 2-59 months: a multicenter, double-blind, randomized, placebo-controlled trial in Pakistan. Clin Infect Dis 2011;52:293-300

Citing Articles (273)

    Letters

    Figures/Media

      Digital Object ThumbnailQUICK TAKE
      Beyond Malaria
       02:24

    1. Table 1. Demographic and Clinical Characteristics of the Children, According to Study Site.
      Table 1. Demographic and Clinical Characteristics of the Children, According to Study Site.
    2. Figure 1. Distribution of All 1232 Diagnoses among 1005 Febrile Children at Two Sites in Tanzania.
      Figure 1. Distribution of All 1232 Diagnoses among 1005 Febrile Children at Two Sites in Tanzania.

      Numbers are percentages of all diagnoses. Percentages may not sum to 100 because of rounding.

    3. Figure 2. Most Common Diagnoses According to Age Group.
      Figure 2. Most Common Diagnoses According to Age Group.
    4. Figure 3. Distribution of All 160 Diagnoses among 133 Febrile Children with Severe Illness.
      Figure 3. Distribution of All 160 Diagnoses among 133 Febrile Children with Severe Illness.

      Numbers are percentages of all diagnoses. The distribution of features of severe illness was as follows: 42 cases of jaundice, 37 of respiratory distress, 27 of impaired consciousness, 19 of seizures, 14 of severe anemia (hemoglobin level, <5 g per deciliter), 9 of severe dehydration, 7 of marasmus, 5 of cardiovascular failure, 4 of plasmodium hyperparasitemia, 2 of renal failure, and 2 of meningismus. Percentages may not sum to 100 because of rounding.

    5. Figure 4. Overlap among Disease Types and Pathogen Types in the 1005 Febrile Children.
      Figure 4. Overlap among Disease Types and Pathogen Types in the 1005 Febrile Children.

      Panel A shows all diseases diagnosed in the present study; disease definitions are given in Table S1 in the Supplementary Appendix. Panel B shows types of pathogens identified, regardless of the clinical and laboratory criteria used to establish final diagnoses.