---
title: "Clinical case definition development for Chikungunya fever in the Democratic Republic of the Congo"
id: "plos-one-23-developing-a-clinical-case-definition-for-chikungunya-fever-in-the-democratic"
canonical_url: "https://medichelpline.com/clinical-feed/plos-one-23-developing-a-clinical-case-definition-for-chikungunya-fever-in-the-democratic"
content_type: "clinical_feed_article"
specialty: "Infectious Disease"
source_name: "PLOS ONE (Medicine)"
source_url: "https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0355216"
published_at: "2026-08-06T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Clinical case definition development for Chikungunya fever in the Democratic Republic of the Congo
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-one-23-developing-a-clinical-case-definition-for-chikungunya-fever-in-the-democratic
- **Specialty:** [Infectious Disease](https://medichelpline.com/clinical-feed/infectious-disease.md)
- **Primary Source:** PLOS ONE (Medicine)
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0355216)
- **Published At:** 2026-08-06T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- The authors conducted a cross-sectional, facility-based study in Kinshasa, DRC (June–November 2019) during a chikungunya outbreak to derive an evidence-based clinical case definition suitable for resource-limited primary care. - Patients aged ≥3 years with symptoms suggestive of acute CHIKV infection were enrolled; clinical exam and blood samples were collected for CHIKV diagnosis by **RT-PCR** and **ELISA**; malaria RDTs were also performed. - Of 132 analysed participants, 40.2% had laboratory-confirmed acute CHIKV infection by PCR or serology during the acute phase. - Classification and Regression Tree (CART) analysis identified **knee pain** and **shoulder pain** as candidate predictors but the CART model classified only 67.4% correctly and missed 71.7% of true cases; cross-validated AUC was 53% (95% CI 43–63%), indicating poor discrimination. - Penalized logistic regression (lasso) produced a model with 61.1% accuracy and cross-validated AUC of 62% (95% CI 52–72%), also insufficient for reliable clinical use. - Exploratory analyses excluding malaria co-infections did not materially improve CART performance; logistic regression failed to identify robust predictors in that subgroup. - Authors conclude that simple symptom-based algorithms lack adequate sensitivity and specificity in this setting; certain joint pain patterns had some predictive value but no model reached acceptable diagnostic performance. - Recommendations emphasize expanding access to serological and molecular testing and integrating epidemiological context when diagnosing CHIKV in resource-limited settings; further research is needed to refine clinical criteria for outbreak response and surveillance.
## Clinical Analysis & Structured Key Points
Developing a clinical case definition for Chikungunya fever in the Democratic Republic of the Congo | PLOS One Browse Subject Areas ? Click through the PLOS taxonomy to find articles in your field. For more information about PLOS Subject Areas, click here . Article Authors Metrics Comments Media Coverage Reader Comments Figures Figures Abstract Introduction Chikungunya fever presents a diagnostic challenge in Sub-Saharan Africa due to the absence of a region-specific case definition, especially in settings with high disease burden and diverse aetiologies of acute febrile illnesses. Our study aimed to develop an evidence-based clinical case definition for acute Chikungunya virus (CHIKV) infection suitable for use at resource-limited primary health care level. Methods We conducted a cross-sectional study in Kinshasa, Democratic Republic of the Congo, between June and November 2019 during an ongoing CHIKV outbreak. Patients aged ≥3 years with symptoms suggestive of acute CHIKV infection were enrolled. Clinical examination data and blood samples were collected for CHIKV diagnosis via PCR and ELISA. Malaria rapid diagnostic tests were also performed. We used Classification and Regression Tree (CART) analysis with 10-fold cross-validation and predictive logistic regression models with lasso penalty, with performance assessed via the Area under the Receiver Operating Characteristics Curve (AUC). Results Of 132 analysed patients, 40.2% had acute CHIKV infection. CART analysis identified knee and shoulder pain as potential predictors, however classifying only 67.4% of patients correctly while missing 71.7% of cases (cross-validated AUC 53%; 95%CI 43%−63%). Logistic regression achieved 61.1% accuracy, with a corresponding cross-validated AUC of 62% (95%CI 52%−72%). Exploratory analysis excluding malaria co-infections yielded similar CART performance, while logistic regression failed to identify any reliable predictors. Conclusions While our study provides valuable insights into the clinical presentation of CHIKV in Sub-Saharan Africa, it highlights the challenges of developing a simple symptom-based diagnostic algorithm suitable for resource-limited primary health care settings. Certain joint pain patterns showed predictive value, however no model achieved sufficient accuracy for clinical use. Improved access to serological testing and integration of epidemiological context are essential for CHIKV diagnosis in such contexts. Further research is needed to refine clinical criteria and support outbreak response in resource-limited regions. Citation: Vogt F, Nkuba-Ndaye A, De Weggheleire A, Makiala-Mandanda S, Mbala-Kingebeni P, Mputu-Ngoyi B, et al. (2026) Developing a clinical case definition for Chikungunya fever in the Democratic Republic of the Congo. PLoS One 21(8): e0355216. https://doi.org/10.1371/journal.pone.0355216 Editor: Pierre Roques, CEA, FRANCE Received: October 4, 2025; Accepted: July 17, 2026; Published: August 6, 2026 Copyright: © 2026 Vogt 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: Data cannot be shared publicly because of confidentiality concerns. Data can be made available from the Institute of Tropical Medicine Institutional Antwerp Ethics Committee for researchers who meet the criteria for access to confidential data. Contact information: ITM Institutional Review Board, chairperson Dr Raffaella Ravinetto, email irb@itm.be or rravinetto@itg.be , url https://www.itg.be/E/institutional-review-board . Each request needs to indicate the person/institution making the request, the intended usage of the data requested and include all documents required as part of the request submission. The sharing of data is limited by the conditions outlined in the patient consent form and the requirements of the Ethics Review Board of the Kinshasa University, Kinshasa, DRC and the Institutional Review Board of ITM, Antwerp, Belgium. Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. Introduction Chikungunya fever is a mosquito-borne viral disease caused by the Chikungunya virus (CHIKV), an enveloped positive single-strain RNA alphavirus of the Togaviridae family. [ 1 ] Three genotypes have been described to date: Asian, West African and East/Central/South African (ECSA). [ 2 ] In Africa, CHIKV transmission is maintained in a sylvatic cycle involving wild non-human primates and various arboreal Aedes mosquitoes. Mosquito-to-human transmission in urban transmission cycles is dominated by A. aegypti and A. albopictus . [ 1 , 3 ] Since the first recorded epidemic in Tanzania in 1952, CHIKV outbreaks have been reported in many countries around the world and with increasing frequency in Sub-Saharan Africa during recent years. The proportion of symptomatic CHIKV infections during epidemics has been reported to be between 70–97%. [ 2 ] For people who develop symptomatic illness, the incubation period is usually between three to seven days (range one to twelve days). Patients typically present with abrupt onset of fever, and severe (mostly symmetric) joint pain followed in some cases by a rash. Other symptoms may include headache, conjunctival redness, nausea and vomiting. Severe manifestations are rare, but can include myocarditis, hepatitis, ocular and neurological disorders. [ 1 , 2 ] The viraemic period of CHIKV lasts two to ten days, which matches the acute symptomatic phase of the disease and possible detection through viral nucleic acid-based tests. Laboratory confirmation of CHIKV infection is done using CHIKV-specific real-time reverse-transcription polymerase chain reaction (RT-PCR) tests, Enzyme linked Immunosorbent Assays (ELISA), and CHIKV-specific IgM by immunofluorescence assays (IFA). [ 4 , 5 ] PCR technology and serological methods are complex, expensive, and only available at few central laboratories in most low-resource settings. A good clinical case definition for use at primary health care level is crucial to ensure timely and appropriate care, to avoid that laboratory capacity is overburdened if too many laboratory confirmations are requested by health workers during outbreaks, that the number of cases is not severely under- or overestimated during surveillance and outbreak response, or that entire outbreaks are missed. A sound balance of sufficiently high specificity and sensitivity is therefore key. However, establishing an accurate and practical case definition for acute CHIKV infection applicable to different contexts and populations is difficult due to the wide range of clinical symptoms and atypical presentations, possible overlap with other pathologies causing an acute febrile illness syndrome, differences in clinical presentations across CHIKV strains, and differences in host populations. [ 6 ] A couple of case definitions have been developed for the Americas [ 7 , 8 ] and applied to South-East Asia [ 9 ], featuring arthritis, fatigue, rash and ankle joint pain [ 7 ], and fever >38.5 °C and acute onset joint pain [ 8 ], respectively. However, no evidence-based case definition exists for Sub-Saharan Africa to date. In a context where many other severe, potentially fatal infectious diseases such as malaria, typhoid fever, and dengue co-circulate in the same population, this absence leaves clinicians with little guidance how to correctly identify, differentiate and manage patients on a clinical basis. We conducted an exploratory clinical study during a CHIKV outbreak in Kinshasa, the Democratic Republic of the Congo (DRC) in 2019 with the aim to identify symptom patterns to develop a clinical case definition for suspicion of chikungunya fever for use at primary health facility level. Methods Study design This was a cross-sectional study embedded into routine health services with prospectively-planned collection of blood samples and clinical information from patients seeking health care at the outpatient department of the health centre ‘Centre Hospitalier Lukunga’ (CH Lukunga), Binza Ozone health zone in Kinshasa, DRC between June and November 2019. Being an exploratory study, no prior null hypotheses or formal sample size estimations was done. Context In November 2018, local clinicians in the health zones of Mont-Ngafula I and Mont-Ngafula II, located in the province of Kinshasa, noted an increase in acute febrile cases associated with severe joint pain and headache. In January 2019, the Provincial Division of Health (PDH) and the National Institute of Biomedical Research (INRB) conducted a joint epidemic outbreak investigation and confirmed CHIKV infection in blood of suspect patients through RT-PCR. [ 10 ] Following further alerts, additional investigations in Matadi, Kongo Central province in collaboration with the Institute of Tropical Medicine Antwerp (ITM), showed that the outbreak in Western Kinshasa had spread further into Kongo central province. [ 11 , 12 ] Chikungunya fever is not part of the national surveillance system in DRC, and no official case definition exists for the DRC to date. [ 13 ] Laboratory confirmation of acute CHIKV infection in the DRC is limited to the Virology department of INRB. Study setting CH Lukunga was selected as study site following consultations of provincial health authorities, the INRB laboratory database to identify the health zones with the most active CHIKV alerts, and subsequent exchanges with the health management teams of the health zones of Binza Ozone and Binza Meteo. CH Lukunga is located in the municipality Ngaliema in the western part of Kinshasa and offers general preventive and curative health services at primary care level through outpatient consultations for adults and children, as well as a limited inpatient services in internal medicine, surgery, and maternity care. Staff consists of midwives, nurses, laboratory technicians, and two general physicians. CH Lukunga is a private not-for-profit entity owned and managed by the Diocesan Office of Medical Works (BDOM), but functions and reports according to Ministry of Health guidelines. It has basic laboratory infrastructure for blood sampling, blood sample storage, rapid diagnostic tests, blood grouping, and microscopy. Clinical procedures Enrolment started 25 June and continued until 28 November 2019. Eligible patients were invited consecutively to participate in the study by a trained nurse or medical doctor of the CH Lukunga during their outpatient consultation. Inclusion criteria were: being aged 3 years or above; consulting for symptoms suggestive of acute CHIKV infection (self-reported sudden-onset of fever or arthralgia, currently ongoing or during the past 7 days); being willing and able to provide written informed consent (or by a guardian for minors). Patients who were considered unsuitable for venous blood sample collection as per clinical judgement of the recruiting health care worker (e.g., very sick or fragile patients) were excluded. Standard operating procedures were developed for all study-related activities prior to start of enrolment. During medical consultations, socio-demographic characteristics, detailed clinical signs and symptoms, and previous self-medication were recorded using a paper-based case report form (see S1 Annex ). Finger prick blood was taken for on-site malaria rapid diagnostic testing (RDT), plus 5 ml venous blood were drawn into EDTA tubes for CHIKV diagnostic testing at INRB. Routine patient care continued to be offered according to the usual standards and guidance applied in CH Lukunga and was not altered by the study. The malaria RDT result was immediately shared with the treating physician or nurse. Laboratory procedures EDTA tubes were stored at 4–8 °C at the study site and transported to INRB at least once a week in a cooled transportation box. At INRB, blood samples were aliquoted and stored at −20 °C. All samples were tested in parallel for CHIKV RNA presence using RT-PCR (RNA extraction with the QIAamp Viral RNA Mini Kit (Qiagen, Germantown, MD, USA) and a CHIKV specific RT-qPCR from Bio-Rad Laboratories, Marnes-La-Coquette, France, as described previously [ 11 ], and for IgM and IgG CHIKV antibodies using the enzyme-linked immunosorbent assay from Euroimmun, Lübeck, Germany. Active CHIKV infection was defined as having a PCR cycle threshold value below 35 or showing presence of IgM antibodies (defined as optical density ratio ≥1.1). For both IgM and IgG, optical density ratios ≥0.8 and 3 joints; swelling in >3 joints; skin rash or pruritus. Most clinical and demographic data were binary or converted into categorical variables, and hence presented using percentages for descriptive analysis by acute CHIKV infection status. Borderline antibody titres were considered negative in the analysis. To identify the combination of clinical symptoms that best predict acute CHIKV infection, we used classification and regression tree (CART) analysis where the tree was pruned with cost-complexity parameter chosen by cross-validation, and logistic regression with lasso penalty chosen by 10-fold cross-validation. The performance of the prediction model was assessed by estimating the area under the Receiver Operating Characteristics (ROC) curve (AUC). Since the same data were used to build the tree and for prediction, a 10-fold cross-validation of the AUC was also done, to correct for potential over-optimization of the naïve AUC. Given the high prevalence of malaria infection in the study setting, we also conducted an exploratory secondary analysis to establish a diagnostic definition for acute CHIKV infection as differential diagnosis to malaria (i.e., after malaria is ruled out). For this we excluded all current and recent malaria infections (defined by malaria RDT pf + pan positive test result) from the analysis. Ethics This was an observational study that did not interfere with routine care. Patients received standard of care according to routine practice regardless of study participation. Individual written informed consent was obtained from each participant or guardian before any study procedures were started. The study was approved by the Institutional Review Board of ITM, Antwerp, Belgium (ref #1312/19), and the Ethics Review Board of Kinshasa University, Kinshasa, DRC. Additional information regarding the ethical, cultural, and scientific considerations specific to inclusivity in global research is included in the Supporting Information ( S4 Checklist ). Results We enrolled 134 patients into our study, of which two (1.5%) were excluded from the analysis due to missing outcome data. Among the remaining 132 patients, 86 (65.2%) were female and 34 (25.8%) were below 10 years old. One hundred and nineteen (93.2%) had an axillary temperature ≥37.5°C, 115 (87.1%) had a headache, 105 (79.5%) had arthralgia, 72 (54.5%) had myalgia, and 25 (18.9%) had either skin rash or pruritus (see Table 1 and S2 Table for further demographic and clinical characteristics). Fifty-three (40.2%) were confirmed as having acute CHIKV infection as per laboratory testing. Of those, 10 (18.9%) were positive only on RT-PCR, 8 (15.1%) on both RT-PCR and IgM, 2 (3.8%) on RT-PCR and borderline on IgM, and 33 (62.3%) only on IgM. IgG positivity, indicative of past infection, was 47.0% (n = 62) among the total study population, and 58.5% (n = 31) among those with acute CHIJKV infection. Thirty-four (25.8%) patients had a pf + pan positive malaria RDT result, indicating current or recent malaria infection, while 10 (7.6%) patients had a malaria-CHIKV co-infection ( Table 2 ). Download: PNG larger image TIFF original image Table 1. Demographic and clinical characteristics by CHIKV infection status among all included participants (N = 132). https://doi.org/10.1371/journal.pone.0355216.t001 Download: PNG larger image TIFF original image Table 2. Malaria-CHIKV co-infection status among all included participants (N = 132). https://doi.org/10.1371/journal.pone.0355216.t002 The tree from the CART analysis is shown in Figure 1 , and the comparison of the true Chikungunya fever status with the predicted status is shown in Table 3 . The predictors identified in the tree are pain in the knees and pain in the shoulders. Applying this tree would identify 67.4% of patients in our dataset correctly as either having acute CHIKV infection or not. The majority (71.7%) of the acute CHIKV cases would have been missed. While the corresponding naïve AUC was 67% (95% CI: 58%−76%) ( Figure 2 ), the 10-fold cross-validation resulted in a cross-validated AUC of 53% (95% CI: 43%−63%) ( Figure 3 ). Download: PNG larger image TIFF original image Fig 1. Classification and regression tree analysis among all included participants (N = 132). https://doi.org/10.1371/journal.pone.0355216.g001 Download: PNG larger image TIFF original image Table 3. True versus predicted CHIKV infection status from classification and regression tree analysis (N = 132). https://doi.org/10.1371/journal.pone.0355216.t003 Download: PNG larger image TIFF original image Fig 2. Area Under the Curve from classification and regression tree analysis. https://doi.org/10.1371/journal.pone.0355216.g002 Download: PNG larger image TIFF original image Fig 3. Area Under the Curve from classification and regression tree analysis after 10-fold cross-validation. https://doi.org/10.1371/journal.pone.0355216.g003 The model resulting from logistic regression analysis with lasso penalty is shown in S3 Box , and the comparison of the true Chikungunya fever status with the predicted status is shown in Table 4 . Applying this model resulted in 61.1% correctly classified individuals, and all patients without active CHIKV would have been correctly classified as such. However, 79.6% of the acute CHIKV cases would have been missed. The corresponding naïve area under the ROC curve was 75% (95% CI: 66%−85%) ( Figure 4 ), while the respective cross-validated AUC was 62% (95% CI: 52%−72%) ( Figure 5 ). Download: PNG larger image TIFF original image Table 4. True status versus predicted CHIKV infection status from lasso regression analysis (N = 126). https://doi.org/10.1371/journal.pone.0355216.t004 Download: PNG larger image TIFF original image Fig 4. Area Under the Curve from lasso regression analysis. https://doi.org/10.1371/journal.pone.0355216.g004 Download: PNG larger image TIFF original image Fig 5. Area Under the Curve from lasso regression analysis after 10-fold cross-validation. https://doi.org/10.1371/journal.pone.0355216.g005 For the exploratory analysis for chikungunya fever as differential diagnosis to malaria, the 34 participants with pf + pan positive malaria RDT result (see Table 2 ) were excluded. Among the remaining 98 patients, the respective tree from the CART analysis with cost-complexity parameter chosen by cross-validation is shown in Figure 6 , and the comparison of the true Chikungunya status with the predicted status is shown in Table 5 . This correctly identified 72.4% of patients, with the performance of the CART analysis being similar compared to the complete case analysis. The lasso logistic regression analysis on this subset with 10-fold cr
## Related Clinical Research

- [Molecular mimicry and autoimmune disease: review unavailable — summary of source limitations](https://medichelpline.com/clinical-feed/frontiers-in-immunology-15-molecular-mimicry-revisited-novel-perspectives-on-autoimmune-disease-initiation.md)
- [Avoiding Cardiac Toxicity with Immune Checkpoint Inhibitors: Safety Review (Title Only Available)](https://medichelpline.com/clinical-feed/frontiers-in-immunology-15-avoidance-of-cardiac-toxicity-during-the-application-of-immune-checkpoint.md)
- [How social networks shape STI disclosure among adolescents and young people in Lusaka: a COM-B ana](https://medichelpline.com/clinical-feed/medrxiv-2-exploring-the-role-of-social-networks-in-shaping-sti-disclosure-among.md)
- [Low antenatal hepatitis B testing and incomplete infant birth‑dose coverage in Nigeria’s MTCT prev](https://medichelpline.com/clinical-feed/medrxiv-6-contact-without-protection-high-antenatal-attendance-minimal-hepatitis-b.md)
- [Hepatitis A in Costa Rica: CDC Level 1 Travel Notice and Prevention Guidance](https://medichelpline.com/clinical-feed/cdc-travelers-health-notices-0-level-1-hepatitis-a-in-costa-rica.md)

## Navigation
- [← Back to Infectious Disease Feed](https://medichelpline.com/clinical-feed/infectious-disease.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.