F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.150743.3 Research Article Articles Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022-2023 outbreak in Liberia

[version 3; peer review: 1 approved, 3 approved with reservations]

 Dwalu Emmanuel Conceptualization Formal Analysis Methodology Validation Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0003-2875-4130 a 1 Tweya Hannock Conceptualization Formal Analysis Methodology Writing – Original Draft Preparation Writing – Review & Editing 2 Beglaryan Mher Conceptualization Formal Analysis Methodology Writing – Original Draft Preparation Writing – Review & Editing 3 Umeokonkwo Chukwuma D. Formal Analysis Methodology Writing – Review & Editing 4 Jetoh Ralph W. Conceptualization Methodology Writing – Review & Editing 1 Shobayo Bode I. Conceptualization Methodology Writing – Review & Editing https://orcid.org/0000-0002-4864-0248 1 Tarweh Fahn Conceptualization Methodology Writing – Review & Editing 1 Owiti Philip Conceptualization Methodology Writing – Review & Editing 5 Relan Pryanka Conceptualization Methodology Writing – Review & Editing https://orcid.org/0000-0002-9543-7891 6 Hassan Shermarke Conceptualization Methodology Writing – Review & Editing 7 Goteh George W. Methodology Validation Writing – Review & Editing 1 Lehyen Darius B. Conceptualization Methodology Writing – Review & Editing https://orcid.org/0009-0008-9894-7409 1 Ako-Egbe Louis Conceptualization Methodology Writing – Review & Editing https://orcid.org/0000-0002-3269-8509 8 Kamara Ibrahim Franklyn Conceptualization Methodology Writing – Review & Editing https://orcid.org/0000-0003-1454-4650 9 Akpan Godwin E. Formal Analysis Visualization Writing – Review & Editing https://orcid.org/0000-0001-8204-9219 10 Adewuyi Peter Formal Analysis Methodology Writing – Review & Editing 10 Kpanyen Patrick N. Conceptualization Methodology Writing – Review & Editing 1 Vonhm Benjamin T. Conceptualization Methodology Writing – Review & Editing 1 Gilayeneh Julius S. M. Conceptualization Methodology Writing – Review & Editing 1 National Public Health Institute of Liberia, Monrovia, Liberia International Training and Education Center for Health (I-TECH), Lilongwe, Malawi Tuberculosis Research and Prevention Centre, Yerevan 0014, Armenia African Field Epidemiology Network, Kampala, Central Region, Uganda Ministry of Health, Nairobi, Kenya Health Emergencies Programme, World Health Organization, Geneva, Switzerland Infectious Diseases Data Observatory, University of Oxford, Oxford, UK World Health Organization Country Office, Monrovia, Liberia Reproductive Maternal Newborn Child and Adolescent Health Unit, World Health Organization Country Office, Freetown, Sierra Leone African Field Epidemiology Network, Monrovia, Liberia dwaluemmanuelf@gmail.com

No competing interests were disclosed.

 10 6 2025 2024 13 661 9 6 2025 Copyright: © 2025 Dwalu E et al. 2025 This is an open access article distributed under the terms of the Creative Commons Attribution IGO Licence. Background

Lassa fever (LF) is an endemic and immediately notifiable disease in Liberia, and one laboratory confirmed case constitutes an outbreak. We described the epidemiological characteristics and hospital outcomes of LF cases hospitalized during the 2022-2023 outbreak in Liberia.

 Methods

A retrospective cohort study was conducted using routine LF surveillance data from the 2022-2023 outbreak in Liberia. Descriptive statistics were used to summarize the data and log binomial regression to assess the association between epidemiological characteristics and mortality.

 Results

A total of 439 suspected LF cases were reported. The median age was 22 (interquartile range (IQR): 10-33) years and 233 (53%) were females. The median number of days between symptom onset and admission was 4 (IQR 2-7). Of the 439 cases, 416 (95%) were tested for LF and 138 were confirmed with 33% positivity rate. The majority, 95 (69%), of confirmed cases were <30 years, 78 (57%) were females, and 81 (59%) were reported during the dry season (October – March). Contact with rodents, 95 (69%), was the commonest mode of exposure. Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%) and myalgia, 114 (83%) were the most common clinical presentations. There were 83 (19%) deaths among hospitalized suspected LF cases - 42 deaths (15%) among 278 individuals who tested negative and 41 among confirmed cases with 30% case fatality rate (CFR). Presenting CFR per age group, age 40-49 years accounted for 8/12 (67%) and those aged≥50 reported 5/8 (63%) of the deaths among the confirmed cases. There was no significant association between epidemiological characteristics and LF mortality.

 Conclusions

The outbreak highlighted a high disease burden of LF with young adults disproportionately infected, and mortality, even among those who tested negative for the virus. This underscores the urgent need for preventive measures like vaccines and health education campaigns.

 Liberia Lassa fever positivity rate case fatality rate outbreak Integrated Disease Surveillance and Response SORT IT operational research The Special Programme for Research and Training in Tropical Diseases (TDR), Geneva, Switzerland HQTDR2422924-4.1-72863. This SORT IT Programme was funded by the Special Programme for Research and Training in Tropical Diseases (TDR), Geneva, Switzerland (Grant Number HQTDR 2422924-4.1-72863. The APC was also funded by TDR. TDR is able to conduct its work thanks to the commitment and support from a variety of funders. A full list of TDR donors is available at: https://tdr.who.int/about-us/our-donors. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Amendments from Version 2

This version has carefully addressed all the reviewers’ comments and provides clarity to our readers.  We added spaces between ages and "year old group" as in "30-39-year-old group (26%, 6/23), and the 15–29-year-old group". For "Factors associated with mortality....", all the results were described in the past tense (specifically, the word compare was used in the past tense throughout the paragraph). We rephrased to "none of the variables was associated with LF mortality". We rephrased the sentence as “Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%) and myalgia, 114 (83%) were the most common clinical presentations. We added “Presenting CFR per age group, ……..age 40-49 years accounted for 8/12 (67%)  ……... Replaced “while” with “and” in the introduction. Added “home” to the sentence “Individuals with negative results after RT-PCR testing results were…… Added “The Epi Info analysis software was used for all descriptive statistics, while the Stata v18.0 (open-source alternative is R: The R Project for Statistical Computing), was used to conduct the regression analysis”. Added “while 8% (11/138) of the total confirmed cases had contact with a confirmed Lf case Rephrased sentence as” There was high CFR among the middle-aged and older adult population. Although the causes of the high mortality are unknown, this could be likely due to waning immunity among these age groups, making them less effective at fighting off infections like Lassa fever, similar to other studies…. Added “The reason for the high CFR in Nimba is unknown, considering that Nimba had a median presentation time less than 1 day, pointing out the need for further investigation”. Added to the sentence. “Furthermore, the CFR was high among the “middle-aged and older adult” population instead of the early population previously mentioned for age =>30 years.

 Introduction

Lassa fever (LF) is a viral haemorrhagic fever caused by the Lassa virus (LASV), a member of arenavirus group. 1 – 3 This LASV would be primarily transmitted to humans through direct or indirect contact with the urine or feces of the infected natal mastomys mice (Mastomys natalensis) during hunting or processing for consumption. 4 – 6 The mice are the natural reservoir for the LASV. Additionally, the virus may be transmitted from human to human through contact with an infected person's blood, faeces, or other bodily secretions. 7 , 8 Pregnant women and children have the highest risk of contracting the disease. 3 , 9 The incubation period for LF ranges from 2-21 days. A systematic review conducted in 2021 reported on clinical data, identified fever, headache, vomiting, abdominal pain, and cough as the most frequently presenting symptoms in individuals suspected of having LF and a retrospective study in 2022 identified fever, fatigue, sore throat, loss of appetite, headache, vomiting, and myalgia as presenting symptoms among confirmed cases. The disease is diagnosed through a real-time polymerase chain reaction (RT-PCR) test and intravenous (IV) ribavirin is a recommended treatment. 10 , 11 Additionally, oral ribavirin can be used for post-exposure prophylaxis. However, recent studies have reported contradictory results regarding the effectiveness of ribavirin, prompting need for further investigation. 12

LF is a significant global public health challenge, recognized by the World Health Organization (WHO) as a priority disease for surveillance, research and vaccine development efforts. 1 , 11 Endemic to various West African countries, including Guinea, Liberia, Nigeria, and Sierra Leone, 13 the disease is estimated to cause 100,000 to 300,000 new cases and 5,000-10,000 deaths annually. 6 LF exhibits a seasonal pattern, with most cases reported during the dry season each year. 3 , 14 The case fatality rate (CFR) among hospitalized patients is estimated to be between 15% and 20% but could be more during epidemics. 7

In Liberia, LF is a notifiable disease and one laboratory confirmed case constitutes an outbreak. Liberia is experiencing its most protracted LF outbreak, which started in January 2022. 15 In June 2021, the National Integrated Disease Surveillance and Response (IDSR) Technical Guidelines were revised, incorporating an updated Lassa fever case definition. This coincided with efforts to improve case detection through healthcare worker training. 16 Despite these measures, crucial information such as the positivity rate, socio-demographic and clinical characteristics, mode of exposure, CFR, and factors associated with mortality in this outbreak remain largely unknown. Therefore, this study aimed to understand the 2022-2023 LF outbreak in Liberia by investigating characteristics of suspected and confirmed LF cases, mode of exposure, healthcare access (i.e. timelines of admission and treatment), hospital outcomes, and factors associated with mortality among confirmed LF patients.

 Methods Study design

We conducted a retrospective cohort study in Liberia using routine national LF outbreak surveillance data of all suspected LF cases reported between January 2022 and December 2023.

 Study setting

General setting

Liberia has a population of over five million people. The country has a dry season (late October to March) and a rainy season (April to early October). Its capital city is Monrovia and the country is divided into 15 administrative units called counties, five health regions each comprising 3 counties, and 98 health districts. 17 The Liberian healthcare system operates at five levels: national, county, district, health facility, and community. There is a 3-tier system for service delivery: a primary level (primary health clinics), a secondary level (health center and county/regional hospitals) and a tertiary level (national specialized hospitals). Each level is staffed with designated healthcare workers including surveillance officers who coordinate disease routine surveillance activities. Liberia has 962 functional health facilities categorized as primary-level clinics, secondary-level health centers, and tertiary-level hospitals. The county hospitals have dedicated LF treatment centres. Health care services in public health facilities in Liberia are provided free of charge.

Specific setting

Case detection for LF began at the community or health facility level. At the community level, community health workers identified potential LF cases using the community case definition. Potential cases were referred to the nearest health facility for verification using the community trigger and referral form. Referrals were made immediately or within 24 hours of identification, with a phone call or written message delivered to the nearby health facility. At the health facility level, clinicians/health facility surveillance focal persons used the standard case definition for LF to identify suspected cases based on symptoms. A suspected case was defined as a patient experiencing a fever lasting for 2-21 days (above 38°C) with one or more additional symptoms such as malaise, headache, or muscle pain. Alternatively, patients who had not responded to anti-malarial treatment within 48-72 hours or had a history of rodent contact were also considered suspected cases. 16 Exposure to LASV was defined to encompass any contact with rodents, either directly (handling rodents or their food sources) or indirectly (through contaminated utensils, feces, urine, or consumption of food contaminated with rodent feces or urine). Additionally, exposure included contact with bodily fluids from a person infected with the Lassa fever virus. A probable case was defined as a suspected case who has one or more of the following complications: hearing loss, facial or neck swelling, seizures/convulsions, restlessness, confusion, hypotension, abdominal bleeding. Once verified, demographic and clinical information of the suspected cases was recorded in the IDSR ledger. The IDSR case alert and laboratory submission forms were then completed and finally whole blood samples were collected.

Samples were sent to the National Public Health Reference Laboratory (NPHRL) to confirm LF. Laboratory confirmation was performed by RT-PCR using a Real Star LASV RT-PCR kit. A confirmed case was defined as a suspected or probable case with a confirmed/positive laboratory test (positive IgM antibody, positive RT-PCR or virus isolation) or epidemiologically linked to a laboratory confirmed case. 18 Once results were available, the NPHRL notified the National Public Health Institute of Liberia (NPHIL) and county teams. The county health teams, facilitated by the county surveillance officers, then sent the results to the district level through the district surveillance officers who notified the health facilities.

While awaiting the laboratory test results, individuals were isolated and admitted to a treatment centre where IV ribavirin treatment was initiated if they were classified as “suspected Lf patients”. Ribavirin was administered only to suspected LF cases while awaiting laboratory test results, and treatment was discontinued for individuals with negative results. Daily assessments were performed (i.e. routine examinations including physical exams, patients’ vitals, malaria test, and other necessary checks, etc.) using charts to monitor the patients’ progress. On average, confirmed LF cases were treated with IV ribavirin for 10 days depending on the patient's clinical evolution and recovery. Individuals with negative results after a RT-PCR testing results were discontinued from ribavirin treatment and were discharged home immediately.

The data from the IDSR ledger and case alert and laboratory submission paper-based forms were regularly entered into an MS Excel dataset containing the line-list of LF surveillance data in each district, then submitted to the county level and onward to the National Public Health Institute of Liberia.

 Study data sources, variables, and validation

We obtained data on LF cases from the Microsoft Excel national surveillance database maintained by the Division of Infectious Disease and Epidemiology, National Public Health Institute of Liberia, for the period 2022-2023. Due to incomplete information in some patient records submitted at the national level, we reconciled the data using the county surveillance database. We opted for four sites instead of all eight because these were the sites reporting majority of the cases and reviewing records from these sites would address the incomplete record issue or data quality issues in the existing data set used for the study.

 Statistical analysis

We performed all statistical analyses using Epi Info analysis software (version 7.2.5.0.) and Stata v18.0 (open source alternative is R: The R Project for Statistical Computing), version 4.3.2. The Epi Info analysis software was used for all descriptive statistics while the Stata v18.0 (open-source alternative is R: The R Project for Statistical Computing), was used to conduct the regression analysis. Descriptive statistics were used to summarize the data. Categorical variables, such as sex, county, and symptoms, were presented as frequencies and proportions. Continuous variables including age, time from symptom onset to admission and to ribavirin treatment were presented as medians with interquartile range (IQR). We estimated LASV positivity and case fatality of LF: LF positivity were calculated by dividing the number of confirmed cases by the total number of individuals tested while the CFR was calculated by dividing the number of deaths among confirmed LF cases by the total number of confirmed LF cases. The confirmed LF cases were plotted by county using Arc Geographic Information System Pro ( ArcGIS Pro 3.2.2) (open source alternative: QGIS, version 3.36.3) to visualize the geospatial distribution of the disease. An epidemic curve was constructed using the date of symptom onset and mortality for all confirmed cases. We also estimated the death rate among individuals with negative LF test result. We used log binomial regression to assess the association between socio-demographics, clinical characteristics, and mortality. The regression model focused on a group of individuals who have been diagnosed with LF among confirmed LF patients. Models’ results were presented as risk ratios (RR) with 95% confidence intervals (CI) and p-values <0.05 were considered statistically significant.

 Results Lassa fever diagnosis, ribavirin treatment, and outcome

A total of 439 suspected LF cases were reported between January 2022 and December 2023 ( Figure 1). The median age was 22 (IQR 10-33) years. The majority (66%, 290/439) of the suspected cases were below 30 years. Women (53%, 233/439) of the cases. Occupation data was missing for 174 participants. Among the 265 with recorded occupations, students constituted 32% (141/439) followed by business persons (12%, 51/439). Geographically, Bong County reported 44% (192/439) of the suspected cases, followed by Grand Bassa County (20%, 89/439) and Nimba County (18%, 80/439) ( Table 1).

Lassa fever diagnosis, treatment and hospital outcome, Liberia 2022-2023. Characteristics and positivity rate for Lassa fever diagnosis using reverse transcription-polymerase chain reaction (RT-PCR), Liberia, 2022-2023.
Characteristics Total suspected * PT-PCR tested * Positivity rate ¥
n % n % n %
Total 439 416 95 138 33
Age (years)
≤14 143 33 133 32 45 33
15-29 147 33 140 34 50 36
30-39 78 18 73 18 23 31
40-49 46 10 45 11 12 27
≥50 25 6 25 6 8 32
Median age 22 (IQR:10-33) Years
Gender
Male 206 47 206 50 60 29
Female 233 53 233 56 78 33
Occupation
Business 51 11 51 12 15 29
Farmer 33 8 31 7 8 26
Housewife 16 4 16 4 8 50
Student 141 32 137 33 50 36
Tapper 19 4 19 5 11 58
Others β 8 2 8 2 3 38
Unknown 171 39 154 37 13 8
Reporting county
Bong 192 44 171 41 60 35
Grand Bassa 89 20 88 21 45 51
Nimba 80 18 80 19 27 34
Montserrado 56 13 56 13 5 9
Others β 22 5 21 5 1 5

Number (and column-proportion) for characteristics and RT-PCR tested.

Number (and row-proportion) positivity rate).

Others (number of suspected cases): River Gee (n=5), Grand Kru (n=7), Maryland (n=3), Bomi (n=2), Lofa (n=2), Margibi (n=2), Sinoe (n=1).

 Time from symptom onset to hospital admission

Among the 323 participants with recorded dates, the overall median number of days between symptom onset and admission was 4 (IQR 2-7). Median times by gender were similar (4.5 (IQR: 2-7) vs 4.0 (IQR 1-7) days). When stratified by county, Nimba had the shortest median time, with a median of <1 day (IQR < 1-1.5).

 Positivity rate of Lassa fever cases

Of the 439 individuals suspected of LF, 437 specimens were collected and 95% (416/439) were tested using RT-PCR test and 138 were LF positive. The overall positivity rate was 33% (138/416) ( Figure 1 & Table 1). Among suspected cases with recorded occupations, tappers accounted for 58% (11/19), followed by housewives (50%, 8/16). Grand Bassa County reported 51% (45/88) of the positivity rates followed by Bong (35%, 60/171) and Nimba Counties (34%, 27/80).

 Ribavirin treatment among Lassa fever cases

Ribavirin treatment was administered to 57% (249/439) of hospitalized suspected LF cases ( Figure 1).

Ribavirin treatment was administered to 83% (115/138) of the confirmed LF cases ( Figure 2).

Geographical distribution of confirmed Lassa fever cases by reporting county, Liberia, 2022-2023. Epi-curve of confirmed Lassa fever cases by reporting epi week and year

The 2022 calendar year accounted for over half (52%, 72/138) of the confirmed cases. Majority of the confirmed (59%, 81/138) clustered between epi-weeks 1-12 and 40-52, aligning with the dry season (October – March) ( Figure 3).

Epi curve of confirmed Lassa fever cases and deaths by reporting week and year, Liberia, 2022-2023. Type of exposure and clinical characteristics of confirmed Lassa fever cases

Rodent exposure accounted for 68% (94/138) among confirmed LF cases while 8% (11/138) of the total confirmed cases had contact with confirmed LF case (epidemiologically linked to a laboratory confirmed case). For clinical characteristics fever accounted for 93% (128/138), followed by malaise 88% (121/138), headache 83% (114/138), and myalgia 83% (114/138). Less than 1% of patients reported hearing loss ( Table 2).

Exposure and clinical characteristics of confirmed Lassa fever cases, Liberia, 2022-2023 (N=138).
Characteristics Confirmed N=138
Type of exposure n %
Exposure to rodent 95 69
Contact with confirmed Lf case 11 8
Unknown 32 23
Signs and symptoms (n=138)
Fever >38 °C 128 93
Malaise 121 88
Myalgia 114 83
Headache 114 83
Sore throat 86 62
Diarrhea 85 62
Red eyes 81 59
Vomiting 77 56
Chest pain 76 55
Nausea 70 51
Cough 55 40
Bleeding 46 33
Swollen face 43 31
Swollen neck 26 19
Hearing loss 1 0.7
Deaths and case fatality rates

There were 19% (83/439) deaths among hospitalized suspected LF; 42 deaths occurred among 278 individuals who tested negative ( Figure 2). The remaining 41 were among confirmed cases, with a 30% CFR. The CFR varied significantly by age group (p-value=0.006). The highest CFR was observed in individuals aged 40-49 years recorded (67%, 8/12) CFR, followed by those aged ≥50 years (63%, 5/8). Lower CFRs were seen in the ≤14-year-old group (27%, 12/45), the 30-39 year-old group (26%, 6/23), and the 15-29 year-old group (20%, 10/50).

 Factors associated with mortality among confirmed Lassa fever cases

In bivariate analysis, we found that LF patients ≥30 years were 1.9 times more likely to die (cRR=1.9, 95% CI=1.160-3.136, p = 0.011) compared to those < 30 years; LF patients reported from Nimba County were 2.0 times more likely to die (cRR=2.0, 95% CI=1.129-3.771, p=0.019) compared to those reported from Bong County; LF patients who did not receive ribavirin treatment were 1.8 times more likely to die (cRR=1.8, 95% CI=1.083-3.103, p = 0.024) compared to those who received ribavirin treatment. These associations were statistically significant between LF and mortality. In multivariate analysis, none of the variables was associated with LF mortality ( Table 3).

Hospital outcome and factors associated with mortality among confirmed Lassa fever cases, Liberia, 2022-2023.
Variable Confirmed (N=138) Deaths (CFR (%) (N=41) cRR (95% CI) P value aRR (95% CI) P value
Age (Yrs) N N (%)
<30 95 22 (23) Ref
>=30 43 19 (44) 1.9 (1.160- -3.136 0.011 1.2 (0.554-2.656) 0.628
Sex
Male 60 20 (33) 1.2 (0.742--2.065) 0.413
Female 78 21 (27) Ref
Reporting County
Bong 60 14 (23) Ref
Grand Bassa 45 11 (24) 1.0 (0.526-2.085) 0.895
Nimba 27 13 (48) 2.0 (1.129- -3.771) 0.019 1.1 (0.357-3.382) 0.868
Occupation (n=95)
Business 15 6 (40) Ref
Farmer 8 5 (63) 1.5 (0.688-3.547) 0.286
Healthcare workers 3 0
House wife 8 2 (25) 0.6 (0.161-2.412) 0.495
Student 50 10 (20) 0.5 (0.217-1.148) 0.102 0.6 (0.240-2.020) 0.506
Tapper 11 4 (36) 0.9 (0.335-2.465) 0.851
Duration_ onset to admission (n=103)
<7 days 59 19 (32) Ref
>=7 days 44 33 (75) 0.7 (0.412-1.460) 0.432

*CFR=case fatality rate;

cRR=crude risk ratio;

aRR=adjusted risk ratio;

CI=confidence interval;

Ref=reference point.

 Discussion

The study described the epidemiological characteristics and fatality among LF cases hospitalized during the 2022-2023 outbreak in Liberia. Findings showed a substantial burden of LF in Liberia during the outbreak with positivity rates (33%) and CFR (30%), and even higher CFR among the elderly population. Younger age groups were disproportionately affected. Furthermore, the dry season coincided with a surge in confirmed cases. Fever, headache and malaise were the most frequently reported symptoms, and rodent exposure was common among the confirmed cases. Surprisingly, there was death (13%) observed among hospitalized individuals without LF, highlighting challenges in diagnosis and LF management.

Similar to other studies, 6 , 18 the study found increased positivity with notable concentration of cases in Bong, Grand Bassa, and Nimba Counties. These counties are endemic to LF and adjacent to each other, which likely has potential influence on infected individuals migrating across the borders likely due to intermarriages, farming activities, and in search of jobs. Thus, contributing to the spread of LF in these regions. The positivity could be because LF is one of the immediately reportable priority diseases under surveillance in Liberia and an endemic disease. Additionally, strengthened surveillance systems at all healthcare delivery levels with assigned surveillance officers for the detection and reporting of suspected LF cases, a functional testing national public health reference laboratory and, most importantly, the revised 2021 case definition of LF all contributed to improved detection of LF. 16

The study highlights that younger age groups are disproportionately affected by LF. This vulnerability is likely to be due to increased exposure to rodent reservoirs. Most confirmed cases reported contact with rodents suggesting exposure to rodent droppings or urine during outdoor activities and food handling. Similar to other studies in Liberia 6 , 18 and Nigeria, 19 majority of the cases were recorded in the dry season (the latter and earlier parts of the two years in the study period: October to March with 2023 accounting for the majority burden. The seasonality of LF cases in West Africa’s dry season is likely due to a combination of factors related to the behavior of the main reservoir, the Mastomys natalensis rodent. During the dry season, with less vegetation and food sources outside, these rodents are likely to seek shelter and food inside homes, increasing their contact with humans. This could lead to frequent interactions between rodents, potentially increasing the spread of the LASV within the rodent population. Less plant cover during the dry season could make rodent burrows and movements noticeable for hunting. A clean environment both in the house and the outside environment might help to reduce the contact between mice and humans. Additionally, effective community infection prevention control programmes in endemic areas by the Ministry of Health and the National Public Health Institute of Liberia might be of help.

Our study identified a high CFR of 33%, more than double the 15% reported by WHO for the region. While this CFR is lower than those reported in Nigeria (60%) 20 and Sierra Leone (69%) 21 and slightly lower than a previous study in Liberia (40%, 41/103), it remains a significant cause for concern. Furthermore, high CFRs were observed in both treated and untreated patients. The short time between symptom onset and hospital admission suggests that healthcare-seeking behaviors alone cannot fully explain the high CFR. Previous studies have raised concerns about the effectiveness of ribavirin treatment and its potential harm. 12 The reliability of the human clinical trial data supporting ribavirin’s use has also been questioned, with pre-clinical studies suggesting that current dosing regimens may not reliably inhibit LASV replication. There was high CFR among middle-aged and older adult population. Although the causes of the high mortality are unknown, it could be likely due to waning immunity among these age groups, making them less effective at fighting off infections like Lassa fever, similar to other studies. 12 However, the reasons for this are unclear, warranting further research in understanding the occurrence. Surprisingly, CFR was higher in Nimba, an endemic LF county, a finding similar to what was reported in a previous study. 6 The reason for the high CFR in Nimba is unknown considering that Nimba had a median presentation time less than 1 day, pointing out the need for further investigation.

Historical data shows that LF has been misdiagnosed with malaria among patients at treatment facilities in Liberia, which impacts early treatment and patients’ outcome. Suspected LF cases are admitted and begin ribavirin treatment while waiting test results. Some of the LF symptoms (fever and headache) overlap with common diseases like malaria, leading to unnecessary admissions and ribavirin use as well as delays in receiving appropriate treatment. This is concerning, as a 13% death rate was observed among those who did not have LF which may be due to other undiagnosed infections. While the strategy aims to combat LF high CFR, improvements in diagnosis and management are crucial.

These findings have an important implication for the programme’s practice and policy. Addressing the screening and management challenges surrounding LF requires a proactive shift towards preventive measures. Mass vaccination campaigns, although in their developmental stages, offer a promising avenue for mitigating the burden of LF. Initiatives by the WHO and Coalition for Epidemic Preparedness Innovations (CEPI) prioritize LF for vaccine development and enhanced surveillance efforts, 22 signaling a step in the right direction. However, given the time frame for vaccine development and implementation, immediate action is warranted. Based on a clearer understanding of seasonal variations within Liberia and across West Africa, public health interventions like vector control programmes and year-round community education campaigns, intensified during the dry season when LF transmission peaks, could significantly enhance community awareness and prevention, thereby reducing mortality rates and easing the burden on healthcare systems. Implementing rodent-proofing measures could likely further reduce exposure risks.

Our findings should be viewed with the following limitations: Firstly, there were incomplete data on duration from onset of symptoms to hospital admission. Exclusion of these records might have affected the estimation of the time from onset of symptoms to hospital admission with respect to socio-demographic characteristics. Secondly, our data was focused on facility-based rather than community-based surveillance, meaning that those who did not seek healthcare were not captured, and we might therefore have underestimated the number of LF cases. Third, we could not explore the effect of ribavirin on mortality due to lack of data. Lastly, our data provides no information on the possible causes of mortality among the suspected cases that were tested negative for LF. Despite these limitations, the study used surveillance data which reflect the program setting, making the findings useful to inform policy and programmes in Liberia and other comparable settings to prevent LF outbreaks.

 Conclusions

This study showed a significant burden of Lassa fever in Liberia during the 2022-2023 outbreak, characterized by positivity, CFR, and death among those without the disease, highlighting the urgent need for proactive prevention measures such as vaccination campaigns and intensive public education. Introducing a rapid point-of-care test for LF would help reduce unnecessary admissions, avoid empirical use of ribavirin, and reduce delays in administering the appropriate treatment. Furthermore, the CFR was high among the middle-aged and older adult population, warranting further investigation. Additionally, the incompleteness of some records highlights the need to strengthen data collection practices within healthcare facilities to ensure complete and accurate data for informed outbreak response efforts, and conduct further research to investigate mortality among suspected cases tested negative for LF.

Ethics consideration

Access to data for this study was granted by the National Public Health Institute of Liberia. Ethics approvals were obtained from University of Liberia Ethics Review Board on September 27, 2023 (protocol# 23-09-390) and the Ethics Advisory Group for the International Union Against Tuberculosis and Lung Disease, Paris, France, on August 9, 2024; (EAG# 24/23). Informed consent was not obtained as we used routine programme data, which was anonymized by delinking patient identifiers from the dataset.

 Author contributions

“Conceptualization, E. D, R.W.J, B.I.S, H. T, F.T, P. O, M.B, I.F.K, G.W.G, P. K, D. B. L, L.A.E, P. R, S.H; B.T.V., J.S.M.G; methodology, E. D, H.T, P. O, M.B, C.D.U, R.W.J, B.I.S, F. T, P. A, G.W.G,D.B.L, L.A.E, P. R, S. H, B.T.V, P. K, P.A., I.F.K, J.S.M.G; software; validation, E. D, G.W.G; formal analysis, E. D, H.T, M. B, C.D.U, P.A., GEA; investigation; resources; data curation; writing—original draft preparation, E. D, H.T, M. B, writing—review and editing, E. D, H.T, F. T, P.O, C.D.U, B.T.V, M. B, B.I.S, P. K, P. A, L.A.E, P. R, S. H, I.F.K, GEA; visualization, G.E.A, E.D; supervision; project administration; funding acquisition. All authors have read and agreed to the last version of the manuscript.”

 Open access statement

In accordance with WHO’s open-access publication policy for all work funded by WHO or authored/co-authored by WHO staff members, WHO retains the copyright of this publication through a Creative Commons Attribution IGO license ( http://creativecommons.org/licenses/by/3.0/igo/legalcode) which permits unrestricted use, distribution and reproduction in any medium provided the original work is properly cited.

 Data availability Underlying data

The dataset used for this study is available at the Division of Infectious Disease and Epidemiology, National Public Health Institute of Liberia and can be accessed upon request in line with the existing data request guide, which provides opportunity for both internal and external data requests ( https://www.nphil.gov.lr/wp-content/uploads/2024/03/nphil-data-request-guide.pdf).

 Acknowledgements

This research was conducted through the Structured Operational Research and Training Initiative (SORT IT), a global partnership led by TDR, the Special Programme for Research and Training in Tropical Diseases hosted at the World Health Organization. The specific SORT IT program that led to this publication is a SORT IT partnership with the WHO Emergency Medical Teams (Geneva), WHO-AFRO (Brazzaville), WHO Country Offices and Ministries of health of Guinea, Liberia, Sierra Leone, and the Democratic Republic of the Congo, the Infectious Diseases Data Repository (IDDO); The International Union Against Tuberculosis and Lung Diseases, Paris, France and South East Asia offices, Delhi, India; The Tuberculosis Research and Prevention Center Non-Governmental Organization, Yerevan, Armenia; I-Tech, Lilongwe, Malawi; Medwise solutions, Nairobi, Kenya; All India Institute of Medical Sciences, Hyderabad, India; and the National Training and Research Centre in Rural Health, Maferinyah, Guinea; All health facility, district, and county surveillance officers in Liberia for generating the data; The District and County Surveillance Officers, Data Management Unit and the entire Division of Infectious Disease and Epidemiology, the National Public Health Institute of Liberia.

 References Ogbu O Ajuluchukwu E Uneke CJ : Lassa fever in West African sub-region: an overview. J. Vector Borne Dis. 2007;44:1. Akpede GO Asogun DA Okogbenin SA : Caseload and case fatality of Lassa fever in Nigeria, 2001-2018: A specialist center’s experience and its implications. Front. Public Heal. Frontiers Media S.A. 2019;7:450510. Merson L Bourner J Jalloh S : Clinical characterization of Lassa fever: A systematic review of clinical reports and research to inform clinical trial design. PLoS Negl. Trop. Dis. 2021;15:e0009788–e0009727. 34547033 10.1371/journal.pntd.0009788 PMC8486098 Lecompte E Fichet-Calvet E Daffis S : Mastomys natalensis and Lassa fever, West Africa. Emerg. Infect. Dis. 2006;12:1971–1974. 17326956 10.3201/eid1212.060812 PMC3291371 Yaro CA Kogi E Opara KN : Infection pattern, case fatality rate and spread of Lassa virus in Nigeria. BMC Infect. Dis. BMC. 2021 [citat 13 juliol 2023];21:1–9. Reference Source Jetoh RW Malik S Shobayo B : Epidemiological characteristics of Lassa fever cases in Liberia: a retrospective analysis of surveillance data, 2019-2020. Int. J. Infect. Dis. 122:767–774. 10.1016/j.ijid.2022.07.006 Reference Source Balogun OO Akande OW Hamer DH : Lassa Fever: an evolving emergency in West Africa. Am. J. Trop. Med. Hyg. 2020;104:466–473. 33236712 10.4269/ajtmh.20-0487 Safronetz D Mire C Rosenke K : A recombinant vesicular stomatitis virus-based Lassa fever vaccine protects guinea pigs and macaques against challenge with geographically and genetically distinct Lassa viruses. PLoS Negl. Trop. Dis. 2015;9:e0003736. 25884628 10.1371/journal.pntd.0003736 PMC4401668 Dahmane A Van GJ Van HM : Constraints in the diagnosis and treatment of Lassa Fever and the effect on mortality in hospitalized children and women with obstetric conditions in a rural district hospital in Sierra Leone. R. Soc. Trop. Med. Hyg. 2014;108:126–132. 24535150 10.1093/trstmh/tru009 PMC4023273 Eberhardt KA Mischlinger J Jordan S : Ribavirin for the treatment of Lassa fever: A systematic review and meta-analysis. Int. J. Infect. Dis. 2019;87:15–20. 31357056 10.1016/j.ijid.2019.07.015 Alli A Ortiz JF Fabara SP : Management of Lassa Fever: A current update. Cureus. 2021;13:4–11. 10.7759/cureus.14797 Cheng HY French CE Salam AP : Lack of evidence for ribavirin treatment of Lassa Fever in systematic review of published and unpublished studies. Emerg. Infect. Dis. 2022;28:1559–1568. 35876478 10.3201/eid2808.211787 PMC9328902 Buba MI Dalhat MM Nguku PM : Mortality Among Confirmed Lassa Fever Cases During the 2015-2016 Outbreak in Nigeria. Am. J. Public Health. 2018;108:262–264. 29267063 10.2105/AJPH.2017.304186 PMC5846592 Sogoba N Feldmann H Safronetz D : Lassa fever in West Africa: evidence for an expanded region of endemicity. Zoonoses Public Health. Wiley Online Library. 2012;59:43–47. 22958249 10.1111/j.1863-2378.2012.01469.x National Public Health Institute of Liberia N: Liberia Lassa Lassa fever Situational report, 2023. Monrpvia:2023. NPHIL M& W: Integrated Disease Surveillance & Response Technical Guidelines, Liberia. Liberia:2021. Liberia Institute of Statistics and Geo-Information Services (LISGIS) M of H: Liberia Demographic and Health Survey 2019-20. Monrovia, Liberia and Rockville, Maryland, USA: Liberia Institute of Statistics and Geo-Information Services (LISGIS), Ministry of Health, and ICF;2019. Dwalu E Jetoh RW Shobayo BI : Trend of Lassa fever cases and factors associated with mortality in Liberia, 2016-2021: a secondary data analysis secondary data analysis. Pan Afr. Med. J. 2021;47:2016–2021. Ilori EA Frank C Dan-Nwafor CC : Increase in lassa fever cases in Nigeria, January–March 2018. Emerg. Infect. Dis. 2019;25:1026–1027. 30807268 10.3201/eid2505.181247 PMC6478197 Buba MI Dalhat MM Nguku PM : Mortality among confirmed Lassa Fever cases during the 2015–2016 outbreak in Nigeria. Am. J. Public Health. 2018;108:262–264. 29267063 10.2105/AJPH.2017.304186 PMC5846592 Shaffer JG Grant DS Schieffelin JS : Lassa Fever in Post-Conflict Sierra Leone. Dis. PLoS Negl. Trop. 2014;8:e2748. 24651047 10.1371/journal.pntd.0002748 PMC3961205 Organization WH: An R & D Blueprint for Action To Prevent Epidemics Funding & Coordination Models for. 2016. 10.5256/f1000research.183459.r390704 Reviewer response for version 3 Hamer Davidson 1 2 Referee https://orcid.org/0000-0002-6578-1851 Department of Global Health, Boston University School of Public Health, Boston, MA, USA Boston University Center on Emerging Infectious Diseases, Boston, Massachusetts, USA

Competing interests: No competing interests were disclosed.

 14 6 2025 Copyright: © 2025 Hamer D 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

The authors have done an excellent job addressing all of our most recent suggestions.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

No

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Emerging infectious diseases, antimicrobial resistance, malaria, Chagas disease

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

 10.5256/f1000research.181955.r387846 Reviewer response for version 2 Maruyama Junki 1 Referee https://orcid.org/0000-0003-0088-4793 The University of Texas Medical Branch, Galveston, Texas, USA

Competing interests: No competing interests were disclosed.

 14 6 2025 Copyright: © 2025 Maruyama J 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

This manuscript summarizes the epidemiological characteristics of the 2022-2023 Liberia LF outbreak. The entire manuscript is well-organized. A few minor revisions would further improve it.

Please add a paragraph to discuss the possibility that qPCR did not detect Lassa viruses in the patients diagnosed as negative LF.

“Natal mastomys mice” should be “natal multimammate mice”.

“Mastomys natalensis” should be italic.

Please use “LF” after defining the abbreviation.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Virology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

 10.5256/f1000research.181955.r385516 Reviewer response for version 2 Lawal Qudus 1 Referee https://orcid.org/0000-0002-8951-7705 Irrua Specialist Teaching hospital, Irrua, Nigeria

Competing interests: No competing interests were disclosed.

 4 6 2025 Copyright: © 2025 Lawal Q 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

The study is relevant and explores an important aspect of Lassa fever control and prevention. Understanding the epidemiology of Lassa will help improve policies and guidelines.

The method of retrospective cohort using a national database helps give a reliable estimate.

A supplementary document of tool used for data collection will be helpful to improve replicability and understanding of what was done

However, the authors can improve the quality of the article by addressing the following:

1) Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%) and myalgia, 114 (83%) were the most common clinical characteristics.

Comment:    it reads better as Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%) and myalgia, 114 (83%) were the most common clinical presentation

2) Age 40-49 years accounted for 8/12 (67%) and those aged≥50 reported 5/8 (63%) of the deaths among the confirmed cases.

Comment: If the author intends to present the CFR per age group, it should be clearly stated as such.

Introduction

3) while a retrospective study in 2022 identified fever, fatigue, sore throat, loss of appetite, headache, vomiting, and myalgia as presenting symptoms among confirmed cases.

Comment: The findings of the retrospective study didn't contrast the SR rather, it added some other findings, thus the use of 'while' is not appropriate. It will be more important to mention the percentage frequency quoted in these studies, given that it was mentioned for the index study in discussion.

4) this study aimed to understand the 2022-2023 LF outbreak in Liberia by investigating characteristics of suspected 

comment: The characteristic of all suspected cases was not fully elucidated. A valuable characteristic will be the final diagnosis of the suspected cases

5) had a history of rodent contact were also considered suspected cases

Comment: The author should share how this was determined, was it from historical record, was these captures in the tool used?

6) epidemiologically linked to a laboratory confirmed case

Comment: it will be important to know the proportion of diagnosis that were based on these case definition

7) Individuals with negative results after a RT-PCR testing results were discontinued from ribavirin treatment and were discharged immediately

Comment: Where were these patients discharged to? Home or hospital, how was their outcome captured

8) We opted for four sites instead of all eight because these were the sites reporting majority of the cases and reviewing records from these sites would address the incomplete record issue or data quality issues in the existing data set used for the study.

Comment: This justification is quite unclear. were the incomplete data from the smaller sites. Interestingly the table 1 account for all the sites. Kindly explain further

9) We performed all statistical analyses using Epi Info analysis software (version 7.2.5.0.) and Stata v18.0 (open-source alternative is R: The R Project for Statistical Computing), version 4.3.2.

Comment: It would be interesting to know why two statistical package was used for analysis. Which analysis was done with each package

10A) Figure 1 will be improved if a brief description is given on

                    reason for not testing

                    reason for not treating in positive cases

10B) for the test negative ribavirin patients, what is the mean/median duration of treatment

11) Nimba had the shortest median time, with a median of <1 day (IQR < 1-1.5).

comment: This presupposes that there is a trend to admit within a day of symptoms in this county; it will be important to know the method and unit of measure (minutes/hours/days) that was used to assess this construct.

12) Positivity rate of Lassa fever cases

Comment: Positivity rate, as described in this study is a measure of sensitivity of physician suspicion of Lassa (number of suspected cases that tested positive) rather than the distribution of the disease. Using it as a estimate of distribution will be biased as it is more dependent on physician index of suspicion

13) 68% (94/138) among confirmed LF cases. For clinical characteristics fever accounted for 93% (128/138), followed by malaise (88%, 121/138), headache (83%, 114/138), and myalgia (83%, 114/138).

Comment: follow a consistent format, percent (fraction)

14) we found that LF patients ≥30 years were 1.9 times more likely to die

Comment: It will be good to know what informed the choice of age 30 for categorisation in the bivariate analysis. This is particularly important because in the discussion section, the result categorization used to explain the result was elderly.

15) Nimba County were 2.0 times more likely to die (cRR=2.0, 95% CI=1.129-3.771, p=0.019) compared to those reported from other counties

Comment: The comparison is to the ref, (Bongo). Although incidentally, Great Bassa has cRR of 1, which makes the interpretation given by the author accurate

16) Surprisingly, there was death (13%) observed among hospitalized individuals without LF, highlighting challenges in diagnosis and LF management.

Comment: This is why it is important to highlight the diagnosis of those who tested negative. Could they be having a diagnosis equally lethal or worse than LF

17) The high CFR in Nimba could be due to the fact that the county is endemic to LF and sometimes misdiagnosed with malaria, which impacts early treatment among LF patients.

Comment: This explanation isn't accurate given the study data; median presentation time is less than 1 day

18) Similar to other studies, the CFR was higher among elderly people, although the causes of high mortality are unknown, we suspect that factors such as comorbidities, waning immunity, etc may have likely contributed. 

Comment: This should be referenced.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

No source data required

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Clinical care and research

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Dwalu Emmanuel Infectious Disease & Epidemiol, National Public Health Institute of Liberia, Monrovia, Montserrado, Liberia

Competing interests: No competing interests were disclosed.

 8 6 2025

The Editor

F1000 Peer Review Journal

Ref: Response to Peer Reviewers’ Reports - F1000 Research Peer Review: Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022 – 2023 outbreak in Liberia

Dear F1000 Research Editorial Team,

Thank you for facilitating the peer review process for our manuscript. We have carefully addressed all the reviewers’ comments, and we believe that the revised manuscript now aligns with the journal’s requirements.

Below, we provide a line-by-line response to the reviewers’ comments, with the original comments highlighted in bold.

Sincerely,

Emmanuel Dwalu

Reviewer # 2: 

General comments:

The study is relevant and explores an important aspect of Lassa fever control and prevention. Understanding the epidemiology of Lassa will help improve policies and guidelines.

The method of retrospective cohort using a national database helps give a reliable estimate.

A supplementary document of the tool used for data collection will be helpful to improve replicability and understanding of what was done.

Thanks to the reviewer for this comment.

However, the data used for this study was a routinely collected LF surveillance data obtained from the National Public Health Institute of Liberia in an excel file (national line list of LF patients’ records), it was not a primary data collected from the patients using questionnaire or specific data collection tool for the purpose of this study.

However, the authors can improve the quality of the article by addressing the following:

1) Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%) and myalgia, 114 (83%) were the most common clinical characteristics.

Comment:    it reads better as Fever, 128 (93%), malaise, 121 (88%), headache, 114 (83%), and myalgia, 114 (83%) were the most common clinical presentations.

We have revised the manuscript accordingly.

2) Age 40-49 years accounted for 8/12 (67%), and those aged≥50 reported 5/8 (63%) of the deaths among the confirmed cases.

Comment: If the author intends to present the CFR per age group, it should be clearly stated as such.

The sentence has been revised in the manuscript.

Introduction

3) while a retrospective study in 2022 identified fever, fatigue, sore throat, loss of appetite, headache, vomiting, and myalgia as presenting symptoms among confirmed cases.

Comment: The findings of the retrospective study didn't contrast with the SR; rather, it added some other findings, thus the use of 'while' is not appropriate. It will be more important to mention the percentage frequency quoted in these studies, given that it was mentioned for the index study in discussion.

We appreciate the reviewer for this comment.

However, the percentages were indicated in our previous version of the manuscript, but were recommended to be removed by the previous reviewer, pointing out that we had many percentages in the manuscript, especially in the introduction section. Reducing it, especially in the introduction section, could be okay.  

Therefore, the authors have revised the manuscript, changing the “while” to “and” as recommended by the reviewer, but maintained the clinical manifestations as they are without percentages.

4) this study aimed to understand the 2022-2023 LF outbreak in Liberia by investigating characteristics of suspected

comment. The characteristic of all suspected cases was not fully elucidated. A valuable characteristic will be the final diagnosis of the suspected cases.

We thank the reviewer for these comments.

These were the available variables for analysis, and the authors believe that the demographics provided here have given some understanding of the characteristics of the suspected cases in this study before delving more into the analysis for the confirmed cases.

The final diagnosis for the suspected cases was indicated as LF positive, showing that 138 of the suspected cases tested were LF positive. The overall positivity rate was 33% (138/416) ( Figure 1 & Table 1).

5) had a history of rodent contact were also considered suspected cases

Comment: The author should share how this was determined, was it from historical record, was these captures in the tool used?

As indicated supra, “the data used for this study was routinely collected LF surveillance data”, “had a history of rodent contact” was from the historical record of the patients already collected, these were not captured in a specific tool used for primary data collection during the study.

6) epidemiologically linked to a laboratory-confirmed case

Comment: it will be important to know the proportion of diagnoses that were based on this case definition

Thanks to the reviewer for this observation.

Though the authors did not emphasize this finding, it was indicated in table 2 that 11 (8%) of the total confirmed cases had contact with a confirmed Lf case (epidemiologically linked to a laboratory-confirmed case).

We have added this to the results.  

7) Individuals with negative results after a RT-PCR testing results were discontinued from ribavirin treatment and were discharged immediately

Comment: Where were these patients discharged to? Home or hospital, how was their outcome captured

We thank the reviewer for the comment.

Individuals with negative results after a RT-PCR testing results were discontinued from ribavirin treatment and were discharged home immediately.

We have revised the manuscript.

8) We opted for four sites instead of all eight because these were the sites reporting majority of the cases and reviewing records from these sites would address the incomplete record issue or data quality issues in the existing data set used for the study.

Comment: This justification is quite unclear. were the incomplete data from the smaller sites. Interestingly the table 1 account for all the sites. Kindly explain further.

Thanks to the reviewer for this observation.

The incompleteness in some variables was across the reporting counties, not unique to one specific reporting county or reporting site. Though our visit aimed to address the incomplete record issue or data quality issues in the existing data, the secondary nature of the routinely collected data could not allow us to further correct records that were already existing at the reporting sites, but could only validate available records with the national database for analysis.

9) We performed all statistical analyses using Epi Info analysis software (version 7.2.5.0) and Stata v18.0 (open-source alternative is R: The R Project for Statistical Computing), version 4.3.2.

Comment: It would be interesting to know why two statistical packages was used for analysis. Which analysis was done with each package

Thanks to the reviewer for the comment.

The Epi Info analysis software was used for all descriptive statistics, while Stata v18.0 (the open-source alternative is R: The R Project for Statistical Computing), was used to conduct the regression analysis.

We have revised the manuscript.

10A) Figure 1 will be improved if a brief description is given on

                    reason for not testing

                    reason for not treating in positive cases

We thank the reviewer for these comments.

The routinely collected LF surveillance data used for this study did not provide these reasons, which we cannot explain at this time.

However, we pointed out that the incompleteness of some records highlights the need to strengthen data collection practices within healthcare facilities to ensure complete and accurate data for informed outbreak response efforts.

   

10B) for the test negative ribavirin patients, what is the mean/median duration of treatment

Thanks to the reviewer. This data was not available for analysis in the set of data used of the study. 

11) Nimba had the shortest median time, with a median of <1 day (IQR < 1-1.5).

comment: This presupposes that there is a trend to admit within a day of symptoms in this county; it will be important to know the method and unit of measure (minutes/hours/days) that was used to assess this construct.

We thank the reviewer for this comment.

The duration was calculated using the date of symptom onset and date of patient's admission (duration from symptom onset to the date of patient's admission) at the treatment facility, and the unit of measure used was days.

12) Positivity rate of Lassa fever cases

Comment: Positivity rate, as described in this study is a measure of sensitivity of physician suspicion of Lassa (number of suspected cases that tested positive) rather than the distribution of the disease. Using it as an estimate of distribution will be biased as it is more dependent on the physician's index of suspicion.

Thanks to the reviewer for the comment.

The sensitivity of physician suspicion of Lassa points to case detection using the suspected LF case definition at the reporting health facilities. 

Positivity as used in this study referred to the proportion of suspected LF cases positive among those tested using RT-PCR; not all suspected cases were tested.

13) 68% (94/138) among confirmed LF cases. For clinical characteristics, fever accounted for 93% (128/138), followed by malaise (88%, 121/138), headache (83%, 114/138), and myalgia (83%, 114/138).

Comment: follow a consistent format, percent (fraction)

We have revised the manuscript accordingly.

14) we found that LF patients ≥30 years were 1.9 times more likely to die

Comment: It will be good to know what informed the choice of age 30 for categorisation in the bivariate analysis. This is particularly important because in the discussion section, the result categorization used to explain the result was elderly.

Thanks to the reviewer for this very useful comment.

Pointing out mortality as our outcome variable (alive or dead) in table 3, we found in our analysis that majority of the deaths among confirmed cases were recorded among patients aged >=30 years, 44% CFR (19/43), which was nearly twice the mortality among those < 30 years, 23% (22/95). We then decided to assess if there was an association between >=30 years and LF mortality at both bivariate and multivariate levels.   

We have revised the phrase “elderly people” to middle-aged and older adult population in the entire manuscript.  

   

15) Nimba County were 2.0 times more likely to die (cRR=2.0, 95% CI=1.129-3.771, p=0.019) compared to those reported from other counties

Comment: The comparison is to the ref, (Bongo). Although incidentally, Great Bassa has cRR of 1, which makes the interpretation given by the author accurate.

Thanks to the reviewer. We have revised the manuscript.

16) Surprisingly, there was death (13%) observed among hospitalized individuals without LF, highlighting challenges in diagnosis and LF management.

Comment: This is why it is important to highlight the diagnosis of those who tested negative. Could they be having a diagnosis equally lethal or worse than LF.

We appreciate the reviewer for this comment.

The diagnosis in the data used for this study was only for Lassa fever, as was suspected. 

17) The high CFR in Nimba could be due to the fact that the county is endemic to LF and sometimes misdiagnosed with malaria, which impacts early treatment among LF patients.

Comment: This explanation isn't accurate given the study data; median presentation time is less than 1 day

Thanks to the reviewer for this observation. We have revised the manuscript accordingly.

18) Similar to other studies, the CFR was higher among elderly people, although the causes of high mortality are unknown, we suspect that factors such as comorbidities, waning immunity, etc may have likely contributed.

Comment: This should be referenced.

Thanks to the reviewer. We have revised the manuscript.

 10.5256/f1000research.181955.r385015 Reviewer response for version 2 Hamer Davidson 1 2 Referee https://orcid.org/0000-0002-6578-1851 Balogun Oluwafemi 3 Co-referee Department of Global Health, Boston University School of Public Health, Boston, MA, USA Boston University Center on Emerging Infectious Diseases, Boston, Massachusetts, USA Department of Medicine, Massachusetts General Hospital (Ringgold ID: 2348), Boston, Massachusetts, USA

Competing interests: No competing interests were disclosed.

 3 6 2025 Copyright: © 2025 Hamer D and Balogun O 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

This article is much improved. There are just a few minor formatting and grammatical issues to address.

Abstract. Results. Second to last line, need to add a space between aged and ≥50.

Results. In section entitled "Deaths and case fatality rates" need to add spaces between ages and "year old group" as in "30-39 year old group (26%, 6/23), and the 15-29 year old group".

Results. In section entitled "Factors associated with mortality....". all the results should be described in the past tense (specifically the word compare should be in the past tense throughout the paragraph).. For example "...found that LF patients ≥30 years were 1.9 times more likely to die (cRR=1.9, 95% CI=1.160-3.136, p = 0.011) compared to those < 30 years;.."

Last sentence of this section. could be rephrased to "....none of the variables was associated with LF mortality...".

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

No

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Emerging infectious diseases, antimicrobial resistance, malaria, Chagas disease

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Dwalu Emmanuel Infectious Disease & Epidemiol, National Public Health Institute of Liberia, Monrovia, Montserrado, Liberia

Competing interests: No authors declared no competing interests.

 8 6 2025

The Editor

F1000 Peer Review Journal

Ref: Response to Peer Reviewers’ Reports - F1000 Research Peer Review: Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022 – 2023 outbreak in Liberia

Dear F1000 Research Editorial Team,

Thank you for facilitating the peer review process for our manuscript. Though our manuscript has now passed peer review (approved) as was communicated to us, we have carefully addressed all the minor reviewers’ comments/suggestions, and we believe that the revised manuscript now aligns with the journal’s requirements.

Below, we provide a line-by-line response to the reviewers’ comments/suggestions, with the original comments highlighted in bold.

Sincerely,

Emmanuel Dwalu

Reviewer # 1:  

General comment.

This article is much improved. There are just a few minor formatting and grammatical issues to address.

Abstract. Results. Second to last line, need to add a space between aged and ≥50.

Thanks to the reviewer for these comments. We have revised the manuscript accordingly.

Results. In the section entitled "Deaths and case fatality rates," need to add spaces between ages and "year old group" as in "30-39-year-old group (26%, 6/23), and the 15–29-year-old group".

We have revised the manuscript accordingly.

Results. In section entitled "Factors associated with mortality....". all the results should be described in the past tense (specifically the word compare should be in the past tense throughout the paragraph).. For example "...found that LF patients ≥30 years were 1.9 times more likely to die (cRR=1.9, 95% CI=1.160-3.136, p = 0.011) compared to those < 30 years;.."

Thanks to the reviewer for these useful observations. We have revised the manuscript.

Last sentence of this section. could be rephrased to "....none of the variables was associated with LF mortality...".

We have revised the manuscript accordingly.

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Emerging infectious diseases, antimicrobial resistance, malaria, Chagas disease

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

10.5256/f1000research.165339.r296578 Reviewer response for version 1 Fichet-Calvet Elisabeth 1 Referee https://orcid.org/0000-0002-9421-3538 Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany

Competing interests: No competing interests were disclosed.

 9 9 2024 Copyright: © 2024 Fichet-Calvet E 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

The study describes cases of Lassa fever (LF) recorded in Liberia over a 2-year period (January 2022-December 2023). It presents socio-demographic variables such as the sex, age, occupation and place of origin of the patients. The clinical picture and the mortality rate complete this description.  These data are important to communicate because the epidemiology of LF is poorly documented in Liberia. I therefore encourage the authors to index these data.

However, I feel that a major revision of the manuscript and the analyses is essential in order to improve readability and highlight the important points that are currently diluted in the text. 

General comments: 

The manuscript is full of percentages; I counted over 200 in the introduction, results, discussion, figures and tables. This is far too many, because in the end you don't know which ones to pick out. There is often redundancy in the figures presented in the text and in the tables. There is no point in repeating the same information twice in a scientific text. It is also important to always mention the denominator when quoting a percentage in the text.

Percentages are often compared using terms such as "highest", "lower", "more", without any statistics to support the comparison. These formulations should be removed from the manuscript as they are not scientific. Only statistics can confirm that one figure is smaller or larger than another.

The abbreviations “LF” for Lassa fever and “LASV” for Lassa virus should be used once the term has been quoted in full for the first time. In addition, the lines should be numbered to facilitate the report of the reviewers.

Specific comments:

Introduction:

“… disease is primarily transmitted…” should be “LASV is primarily transmitted…” because a disease is never transmitted. It is the pathogen that is transmitted.

Replace “Mastomys rats” with “the Natal multimammate mice (Mastomys natalensis)”. Use the term of “mice” instead of “rats” throughout the manuscript.

Transmission of LASV by direct or indirect contact is a hypothesis. It has not been scientifically proven. Therefore, use the conditional tense in this sentence.

Update the morbidity and mortality figures according to the recent publication done by Basinski,.et.al., 2021 (Ref 1)

Some case fatality rates are presented; where do they come from? Country? Years?

 Methods:

Study design and study setting should be combined in one paragraph. The general setting is useless because the results do not use this information.

I think the authors did a case-control study, not a cohort study, because they did not analyse, the outcome variable (LF infection) over time. They compared different exposures or variables between cases (LASV positive) and controls (LASV negative).

The first sentence in “specific setting” should be shortened

Exposure to LASV: how do you know that patients touched contaminated utensils? What is a rodent-borne food? A rodent contaminated-food? It would be useful to have the questionnaire in a supplementary information in order to know how the authors evaluated/quoted the exposure to rodents.

“A confirmed case… laboratory confirmed case… or virus isolation)”. This sentence should be removed as the authors only used the RT-PCR to confirm LF in this study.

CSO and DSO: it is pointless to give abbreviation as this term is not used in the rest of the text.

Study population: this sentence is redundant with the one in Study design

Study data sources: also redundant with the information presented at the end of the Specific setting

Statistical analysis: please use the term of “LASV prevalence” instead of “positivity rate”, and throughout the manuscript. Please specify variables as dependent or independent. Why did not you perform statistics (binomial regression) on LASV prevalence and other variables? You only did statistics on CFR.

 Results

Section “Characteristics of the suspected Lassa fever cases”:

change this title in “Diagnosis, Ribavirine treatment and outcome”

Figure 1: state the CFR for patients tested negative for LASV.

The numbers presented in this paragraph are redundant with Table 1. I would remove this part and focus only on LASV-positive patients.

 Section “Positivity rate of Lassa fever cases”:

see my comment above about the positivity rate.

Use the term of “LASV prevalence” instead of “positivity rate” throughout the text, figure and table.

Again, there is a lot of redundancy between the text and the table.

“highest” see my general comment

Why not to analyse the data to build a model?

 Section “Epicurve”:

look at the “more confirmed cases”: statistics?

 Section “Type of exposure”:

data redundant with table 3, and not well analysed. If you wanted to analyse exposure to rodents, the question would be “are the LASV+ patients more exposed to rodents than the LASV- patients? To do this, you would need to build a model where the RT-PCR result as an outcome (0 or 1) is a function of rodent exposure (absence/presence according to respondent?) + other variables. You did not answer this question in your article.

I would be curious to know such a result, as I suspect that everyone was exposed to rodents in the area…

 Section “Factors associated”:

this is the most interesting part of the article. However, factors such as “occupation” and “duration of onset prior to admission” are not presented. You could say that these two variables have no effect on CFR when analysed in a bivariate model.

 Discussion:

Dry season as risk season: there is no data in this article to support this assumption.

 Furthermore, in another article (Dwalu et al. Trend of Lassa fever cases and factors associated with mortality in Liberia, 2016 - 2021: a secondary data analysis. Pan African Medical Journal. 2024;47(22). 10.11604/pamj.2024.47.22.42156) you published data clearly showing Lassa cases throughout the year, in 6 consecutive years between 2016 and 2021. Only two years showed a peak in the dry season. So, I think it is wrong to claim a seasonal risk when your data shows the opposite in 4 out of 6 years.

Rodent exposure: this is not analysed in this paper (see my comment above).

High testing: what is that?

“younger age group more affected”: I have not seen this result. Where did you do the analysis?

“direct or indirect contact with rodents”: this is not analysed, leading to over-interpretation.

CFR in other countries: give the raw data, and do not repeat this idea twice at the beginning and in the middle of the discussion.

40% (xx/xx?) in Liberia: what is the reference?

Rodent-proofing measures needs to be proposed with great caution. See our recent paper on rodent control strategies in Guinea (Marien, J., M. Sage, U. Bangura, A. Lame, M. Koropogui, T. Rieger, B. Soropogui, M. Douno, N. Magassouba and E. Fichet-Calvet (2024). "Rodent control strategies and Lassa virus: some unexpected effects in Guinea, West Africa." Emerg Microbes Infect 13(1): 2341141).  

As you published data from 2016-2021 surveillance in Dwalu et al. 2024, why don’t you discuss the results presented here with those published previously?

More generally, what is the novelty of this article when compared to Dwalu et al. 2024?

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

No

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

Lassa fever, Lassa virus evolution, phylogeography, rodent ecology, West Africa, eco-epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References : Bridging the gap: Using reservoir ecology and human serosurveys to estimate Lassa virus spillover in West Africa. PLoS Comput Biol .2021;17(3) : 10.1371/journal.pcbi.1008811 e1008811 33657095 10.1371/journal.pcbi.1008811 Dwalu Emmanuel Infectious Disease & Epidemiol, National Public Health Institute of Liberia, Monrovia, Montserrado, Liberia

Competing interests: No competing interest.

 21 3 2025

The Editor

F1000 Peer Review Journal

Ref: Response to Peer Reviewers’ Reports - F1000 Research Peer Review: Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022 – 2023 outbreak in Liberia

Dear F1000 Research Editorial Team,

Thank you for facilitating the peer review process for our manuscript. We have carefully addressed all the reviewers’ comments, and we believe that the revised manuscript now aligns with the journal’s requirements.

Below, we provide a line-by-line response to the reviewers’ comments, with the original comments highlighted in bold.

Sincerely,

Emmanuel Dwalu

Reviewer # 2: 

General comments:

The manuscript is full of percentages; I counted over 200 in the introduction, results, discussion, figures and tables. This is far too many, because in the end you don't know which ones to pick out. There is often redundancy in the figures presented in the text and in the tables. There is no point in repeating the same information twice in a scientific text. It is also important to always mention the denominator when quoting a percentage in the text.

We appreciate the reviewer’s observation. We have revised removing some percentages from the introduction section. However, we still think it is necessary to maintain the percentages in the manuscript's results section.

Percentages are often compared using terms such as "highest", "lower", "more", without any statistics to support the comparison. These formulations should be removed from the manuscript as they are not scientific. Only statistics can confirm that one figure is smaller or larger than another.

Thanks to the reviewer for this comment. We have revised the manuscript accordingly.

The abbreviations “LF” for Lassa fever and “LASV” for Lassa virus should be used once the term has been quoted in full for the first time. In addition, the lines should be numbered to facilitate the report of the reviewers.

We have revised the manuscript accordingly.

Specific comments:

Introduction:

“… disease is primarily transmitted…” should be “LASV is primarily transmitted…” because a disease is never transmitted. It is the pathogen that is transmitted.

We have revised the manuscript accordingly.

Replace “Mastomys rats” with “the Natal multimammate mice (Mastomys natalensis)”. Use the term of “mice” instead of “rats” throughout the manuscript.

We agree and have revised the manuscript.

Transmission of LASV by direct or indirect contact is a hypothesis. It has not been scientifically proven. Therefore, use the conditional tense in this sentence.

We have revised the manuscript accordingly.

Update the morbidity and mortality figures according to the recent publication done by Basinski,.et.al., 2021 (Ref 1)

We appreciate the reviewer for this observation and recommendation. However, the figure presented in the recommended study (a systematic review), has different sources of data but our study (retrospective cohort study) has all the data coming from one surveillance system is Liberia. Therefore, the authors have seen it useful to maintain the figure used in our study.

Some case fatality rates are presented; where do they come from? Country? Years?

The authors have cited the reference in the manuscript to reflect ” The case fatality rate (CFR) among hospitalized patients is estimated to be between 15% and 20% [17] but could be more during epidemics [7,18,19][7].

Methods:

Study design and study setting should be combined in one paragraph. The general setting is useless because the results do not use this information.

Thanks to the reviewer for this comment. In the general setting, we provide an overview of the Liberian health system's structure, with an emphasis on the reference system, which is later linked to LF case definition and case detection in the specific setting. Therefore, we respectfully disagree with the reviewer’s suggestion to combine the design section and remove the general setting.

I think the authors did a case-control study, not a cohort study, because they did not analyze, the outcome variable (LF infection) over time. They compared different exposures or variables between cases (LASV positive) and controls (LASV negative).

Thank you for your comment. We would like to clarify that this is a retrospective cohort study (we have now added the word retrospective) in which all suspected Lf cases were followed until diagnosis, and those confirmed with the infection were followed until discharge. The cases (LASV-positive) and controls (LASV-negative) were not selected based on their exposure status.

Exposure to LASV: how do you know that patients touched contaminated utensils? What is a rodent-borne food? A rodent contaminated-food? It would be useful to have the questionnaire in a supplementary information in order to know how the authors evaluated/quoted the exposure to rodents.

We agree with the reviewer. There was no questionnaire for data collection, as it was clearly stated that the study obtained and utilized case-based records of routine national surveillance data of all suspected Lassa fever cases reported between January 2022 and December 2023 that were reported to the National Public Health Institute of Liberia and cross-checked. However, in our study methods (specific setting), we have revised the definition of exposure to Lassa virus as “any contact with rodents, either directly (handling rodents or their food sources) or indirectly (through contact with contaminated utensils, feces, urine, or consumption of food contaminated with rodent feces or urine).

However, the form had incomplete data which necessitated grouping of the responses into the categories that are reported in the manuscript.

“A confirmed case… laboratory confirmed case… or virus isolation)”. This sentence should be removed as the authors only used the RT-PCR to confirm LF in this study.

We have revised the manuscript as: A confirmed case  was defined as a suspected or probable case with a confirmed /positive laboratory test (positive IgM antibody, positive RT-PCR) or epidemiologically linked to a laboratory confirmed case [22].

CSO and DSO: it is pointless to give abbreviation as this term is not used in the rest of the text.

We agree and have revised the manuscript.             

Study population: this sentence is redundant with the one in Study design

We appreciate the reviewer and have revised the manuscript.

Study data sources: also redundant with the information presented at the end of the Specific setting

We appreciate the reviewer’s comment. However, in the entire specific setting, the authors explained LF case detection, recording/documentation and reporting as well as case definition (suspected, probable, and confirmed).  At the end, the authors explained that data from the IDSR ledger and case alert and laboratory paper-based forms were regularly entered into an MS Excel dataset containing the line-list of LF surveillance data in each district, then submitted to the county level and onward to the National Public Health Institute of Liberia.  Information on sources was explained in the succeeding paragraph (Study data sources, variables, and validation).

Statistical analysis: please use the term of “LASV prevalence” instead of “positivity rate”, and throughout the manuscript. Please specify variables as dependent or independent.

We have revised the manuscript accordingly.

Why did not you perform statistics (binomial regression) on LASV prevalence and other variables? You only did statistics on CFR.

We thank the reviewer for the useful observation. However, the focus of our study was the characteristics and hospital outcomes of Lf patients. We had to indicate the total suspected, tested and positivity to inform our readers before delving into the characteristics of the confirmed and their outcomes. Therefore, the authors think that the analysis performed supports the study objectives' attainment.   

Results

Section “Characteristics of the suspected Lassa fever cases”:

change this title in “Diagnosis, Ribavirin treatment and outcome”

We have revised the manuscript:

The CFR for patients tested LASV negative is indicated in figure 2 (15%, 42/278)) against the 30% (41/138) CFR for confirmed cases.

The numbers presented in this paragraph are redundant with Table 1. I would remove this part and focus only on LASV-positive patients.

We appreciate the reviewer but the authors think this is relevant to be maintained.

Section “Positivity rate of Lassa fever cases”:

see my comment above about the positivity rate.

Please see response below response in addition to the succeeding comments.

Use the term of “LASV prevalence” instead of “positivity rate” throughout the text, figure and table.

We appreciate the reviewer’s detailed review and comments/recommendations. However, our analysis focused on LASV positivity rather than prevalence. This is because prevalence looks at a specific number of people who have (or had) the disease during a specific time period, which would be the number of truly positive individuals out of the entire population—not just who is tested.

The positivity rate—also referred to as “percent positive rate” or “percent positive” is the percentage of all Lf tests performed that are actually positive. Positivity rate is usually expressed as a percentage (Rogers & John Hopkins School of Public Health, 2020).

Again, there is a lot of redundancy between the text and the table.

We have revised highlighting the key findings instead of reproducing the table in text.

“highest” see my general comment

We agreed and revised the manuscript accordingly.

Why not to analyze the data to build a model?

The reviewer’s comment is appreciated. However, the authors did not set to build a model in this study. This could be considered in subsequent studies.

 Section “Epicurve”:

look at the “more confirmed cases”: statistics?

We have revised the manuscript accordingly.

 Section “Type of exposure”:

data redundant with table 3, and not well analyzed. If you wanted to analyze exposure to rodents, the question would be “are the LASV+ patients more exposed to rodents than the LASV- patients? To do this, you would need to build a model where the RT-PCR result as an outcome (0 or 1) is a function of rodent exposure (absence/presence according to respondent?) + other variables. You did not answer this question in your article.

I would be curious to know such a result, as I suspect that everyone was exposed to rodents in the area…

We appreciate the reviewer’s comments. . However, our analysis was focused on describing the characteristics of the confirmed Lf cases and their outcomes. These characteristics included socio-demographics, clinical, and exposures (including rodent exposure, etc). We did not set out to deal with factors associated with LASV positivity among the patients. 

 Section “Factors associated”:

this is the most interesting part of the article. However, factors such as “occupation” and “duration of onset prior to admission” are not presented. You could say that these two variables do not affect CFR when analyzed in a bivariate model.

Occupation and duration of onset prior to admission, along with others were presented in the bivariate analysis but were not significant. That’s why we did not present them in the multivariate analysis, but rather focused on those that were significant.

 Discussion:

Dry season as risk season: there is no data in this article to support this assumption.

Furthermore, in another article (Dwalu et al. Trend of Lassa fever cases and factors associated with mortality in Liberia, 2016 - 2021: a secondary data analysis. Pan African Medical Journal. 2024;47(22). 10.11604/pamj.2024.47.22.42156) you published data clearly showing Lassa cases throughout the year, in 6 consecutive years between 2016 and 2021. Only two years showed a peak in the dry season. So, I think it is wrong to claim a seasonal risk when your data shows the opposite in 4 out of 6 years.

The reviewer’s comment is appreciated. The two years indicated in the article mentioned above (2019 & 2020) recorded the majority of the cases (54%) but the overall majority was reported during the dry season (79%) covering October to March. Over half of the cases (59%, 81/138) clustered between epi-weeks 1-12 and 40-52, aligning with the dry season (October – March) (Figure 4).” Based on this, the authors could safely say that dry season accounted for the majority of the cases, likewise the current study.

Rodent exposure: this is not analyzed in this paper (see my comment above).

Rodent exposure was analyzed in table 2 along with clinical characteristics and outcome and 68% of the confirmed cases were exposed to rodents.

High testing: what is that?

We appreciate the reviewer and have according revised this to high detection and positivity, instead of high testing and positivity rates.

“younger age group more affected”: I have not seen this result. Where did you do the analysis?

We thank the reviewer for the observation. This analysis can be seen in tables 1 and 3 where 95 (69%) were less than 30 years. 

However, the authors have agreed to omit this finding from the discussion.

“direct or indirect contact with rodents”: this is not analyzed, leading to over-interpretation.

Our data analyzed showed exposure to rodents but did not establish direct or indirect contact. This was provided in the discussion to mean that exposure to rodents could be either by direct or indirect contact to rodents. The majority of the confirmed cases reported exposure to rodents suggesting exposure (68%, 95/138).

We have revised the manuscript to remain with exposure to rodents.

CFR in other countries: give the raw data, and do not repeat this idea twice at the beginning and in the middle of the discussion.

40% (xx/xx?) in Liberia: what is the reference?

We agree and have revised the manuscript accordingly (40 (41/103) [6] in Liberia.

Rodent-proofing measures need to be proposed with great caution. See our recent paper on rodent control strategies in Guinea (Marien, J., M. Sage, U. Bangura, A. Lame, M. Koropogui, T. Rieger, B. Soropogui, M. Douno, N. Magassouba and E. Fichet-Calvet (2024). "Rodent control strategies and Lassa virus: some unexpected effects in Guinea, West Africa." Emerg Microbes Infect 13(1): 2341141). 

We agree with the reviewer and have revised it accordingly .

As you published data from 2016-2021 surveillance in Dwalu et al. 2024, why don’t you discuss the results presented here with those published previously?

More generally, what is the novelty of this article when compared to Dwalu et al. 2024?

The reviewer’s comment is highly appreciated. However, the previous study was focused on trend, though it also highlighted factors associated with mortality. In the current study, clinical and exposure characteristics are highlighted in addition to mortality, which were missing in the previous study. 

Is the work clearly and accurately presented and does it cite the current literature?

Partly                

Is the study design appropriate and is the work technically sound?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

No

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

References

1. Basinski AJ, Fichet-Calvet E, Sjodin AR, Varrelman TJ, et al.: Bridging the gap: Using reservoir ecology and human serosurveys to estimate Lassa virus spillover in West Africa.PLoS Comput Biol. 2021; 17 (3): e1008811 PubMed Abstract | Publisher Full Text

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Lassa fever, Lassa virus evolution, phylogeography, rodent ecology, West Africa, eco-epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however, I have significant reservations, as outlined above.

10.5256/f1000research.165339.r296577 Reviewer response for version 1 Hamer Davidson 1 2 Referee https://orcid.org/0000-0002-6578-1851 Balogun Oluwafemi 3 Co-referee Department of Global Health, Boston University School of Public Health, Boston, MA, USA Boston University Center on Emerging Infectious Diseases, Boston, Massachusetts, USA Department of Medicine, Massachusetts General Hospital (Ringgold ID: 2348), Boston, Massachusetts, USA

Competing interests: No competing interests were disclosed.

 19 7 2024 Copyright: © 2024 Hamer D and Balogun O 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

F1000 Research Peer Review :  Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022 – 2023 outbreak in Liberia 

Peer Review 

[General Comment] 

Overall, this manuscript provides an insightful epidemiological analysis of the Lassa fever outbreak and its impact on hospitalized patients in Liberia from 2022 to 2023. The authors present compelling findings that address an evolving public health issue in the context of an outbreak. They have adequately described the study setting & population, exposures (sociodemographic and epidemiological characteristics), and outcome measurements.  

Fundamentally, this manuscript can contribute to the existing literature on Lassa fever by providing insights and understanding into the epidemiological dynamics and clinical outcomes associated with Lassa fever. Additionally, it expands on the substantial burden placed on hospitalized patients during the outbreak, thereby highlighting critical aspects for public health preparedness and response strategies. 

The title refers to the epidemiological characteristics and hospital outcomes of patients hospitalized with Lassa fever, but the manuscript has a moderate amount of material on the patients who were initially suspected to have Lassa fever but were not confirmed. We recommend focusing the analysis on the confirmed cases and only providing a brief description of the suspected cases at the beginning of the Results.  

[Specific Comments]

Comment 1: Rather than refer to cases in the title and the abstract, patients is a better term to use (there are sections where the term case is reasonable when summarizing statistics but not when describing the acute infection and outcomes of the pooled cohort of patients.

Comment 2: Second sentence of the Background of the abstract, outcomes should be plural as in “...and hospital outcomes of Lassa fever patients hospitalized...”.

Comment 3: Abstract, Results. Recommend introducing the abbreviation IQR in the second sentence “...was 22 (interquartile range [IQR] 10-33)...”

Comment 4: In the results section of the abstract, the authors mentioned that ‘290 (69%) of confirmed cases were <30 years.’ However, in the preceding sentence, the authors stated that there were 138 confirmed cases which equates to 33% positivity rate.  Once you start referring to confirmed cases, the denominator should be all confirmed cases (N = 138) yet the description “290 (69%) of confirmed cases were <30 years old” is incorrect. The next two percentages are based on the correct, smaller denominator.

Comment 5: Abstract, Results. Would consider removing mention of the mortality rate among the suspected but not confirmed Lassa fever cases. There is no information provided on the possible causes of febrile illness in these patients so readers will have no idea what they died from (probably a mix of malaria, bacteremia, and potentially other infectious etiologies).

Comment 6: Abstract, Results. For the last sentence, please add the denominator for each group (age 40-49 years, age > or = 50 years).

Comment 7: Abstract, Conclusions. As suggested above, we would consider removing the phrase “substantial mortality, even among those who tested negative for the virus” since the causes of acute febrile illness in this group are not described. In contrast, it might be useful to mention the empirical use of ribavirin even for patients who were ruled out for Lassa fever, as this means many patients were exposed unnecessarily to a moderately toxic antiviral.

Comment 8: Keywords. Suggest removing Liberia and Lassa fever since these are in the article title and all terms in the title will be picked up by search engines. And SORT IT may not be a searchable term. Consider adding ribavirin as a term.

[Introduction Section]

Comment 9: Mastomys natalensis in the third sentence of the ‘Introduction’ should be written in the standard scientific format as ‘ Mastomys natalensis’. This same suggestion applies to the Discussion (third paragraph).

Comment 10: Introduction. Although the prevalence of frequently presenting symptoms in suspected Lassa fever cases is interesting, the more relevant symptoms are those demonstrated in patients who have been confirmed.

Comment 11: Introduction. Suggest adding risk factors for mortality from Lassa fever as one of the objectives of this analysis.

[Methods Section]

Comment 12: It will be beneficial to the readers of the journal to know the type of cohort study conducted. Kindly specify it is retrospective or prospective ( Study Design).

Comment 12: Methods, Specific setting. Third line. Suggest rephrasing to “...for Lassa fever to identify suspect cases..”.

Comment 13: In the context of the first and second sentence of the Specific Setting of the Methods section, please include a phrase/sentence that indicates the reporting tool, means of communication, and timeliness of reporting (window period) employed by the community healthcare workers when reporting suspected cases to the nearest health facility for verification.

Comment 14: Although this sentence (In the Study Settings section), ‘The IDSR case alert and laboratory submission forms were then completed and finally samples were collected’ , is understood by the reviewer, for clarification and non-assumption by readers of the journal, the authors should indicate the specific type of sample that was collected from the patients for Lassa fever testing.

Comment 15: Generic drug names should be presented in lower case—ribavirin should not be capitalized.

Comment 16: Were there specific selection criteria for the use of ribavirin?

Comment 17: Concerning this statement, ‘Daily assessments were performed using charts to monitor the patients’ progress’, please specify ( concisely) the assessments done.

Comment 18: Please rephrase the statement ‘Individuals with negative results were discharged immediately after the Ribavirin treatment was discontinued’ to something like ‘Individuals with negative results after a repeat RT-PCR testing were discharged immediately after the ribavirin treatment was discontinued.’ If this is not what the authors are trying to convey, kindly clarify.

Comment 19: Kindly expound the phrase ‘and selected treatment centers’ in the sentence ‘Due to incomplete information in some records, we reconciled the data using the county surveillance database and selected treatment centers.’ Succinctly expand the phrase to provide more context to readers.

Comment 20: Statistical analysis. The sentence ‘We used log binomial regression to assess the association between socio-demographics, clinical characteristics, and case fatality rates’ should be revised to ‘We used log binomial regression to assess the association between socio-demographics, clinical characteristics, and mortality. ’ CFR is a population-level indicator, and it appears the authors inadvertently substituted this in place of mortality.

Comment 21: Statistical analysis. The term risk ratio should be in lower case.

[Results Section]

Comment 22: Results. Second sentence. Missing a period at the end of the sentence.

Comment 23: Results. The authors revealed that some individuals were excluded in the statement ‘Among 439 individuals, about one quarter (26%, 116/439) were excluded due to missing data on date of symptom onset or admission’ . Specify what the individuals were excluded from, e.g., from the regression analyses or Table 2. Kindly clarify.

Comment 24: Results. The utility of the data in Table 2 is relatively limited. The median time from symptom onset to hospital admission was similar across age groups (by a day or two), gender (half a day), and county (except Nimba). This table could be removed and a sentence or two added to the Results to describe the major findings.

Comment 25: In Table 4, was there any reason why Bong County was selected as the reference group? Or was the county selected randomly?

Comment 26: Results. It would be interesting and potentially informative to compare patients who had confirmed Lassa fever (PCR+) who received ribavirin vs. those who did not receive this antiviral treatment. Are there any significant differences in their clinical or demographic characteristics?

Comment 27: Table 4, Ribavirin treatment (yes or no) was not included in the bivariable and multivariable models. Historical data and corresponding studies have shown a decrease in mortality among those administered (especially within 6 days from symptom onset). For comprehensiveness, it would be important to explore and demonstrate the associations between the administration of the treatment and mortality. Additionally, the adjusted (multivariable) models with its findings would have more validity if this was also accounted for.

Comment 28: Table 4. The authors used a reference cutoff of 7 days to assess the association between duration from symptom onset to hospital presentation. However, we recommend the authors to explore the 4 days as the reference cutoff. Earlier in the results section, the authors reported the median symptom-hospital duration to be 4 days with an IQR of 2 to 7. For validity reasons, kindly conduct an exploratory analysis/sensitivity analysis to explore this new cutoff.

[Discussion Section]

Comment 29: The abbreviation CFR was introduced much earlier in the manuscript. It should be used on the first and third lines of the beginning of the Discussion. Also, we recommend using the term ‘patients’ rather than cases in the first sentence of the Discussion.

Comment 30: As noted earlier, the group with suspected Lassa fever who were not confirmed is likely a mixture of different etiological agents. Suggest removing the last sentence of the first paragraph of the Discussion for this reason.

Comment 31: The last statement of the second paragraph ‘Additionally, effective community infection prevention control programmes in endemic areas might be of help’ is quite vague. The authors (in their recommendation) should kindly be specific in who implements and ensures the effectiveness of the community IPC programmes.

Comment 32: In the third paragraph, in the third and second to last lines, might be better to state “A clean environment in both the house and outside environment...”

Comment 33: Additional information is required for this statement ‘Similar to other studies, the CFR was higher among elderly people. However, the reasons for this are unclear, warranting further research in understanding the occurrence.’ Yes, additional studies are warranted to better understand the pathophysiological mechanism contributing to the high CFR among elderly population. However, the authors should state possible reasons for high CFR being demonstrated among this population, e.g. comorbidities, waning immunity, etc.

Comment 34: The statement ‘The high CFR in Nimba could be due to the fact that the county is endemic for Lassa fever, with the majority of cases detected at the health facility rather than through early community detection, which impacts on early treatment’ is a bit conflicting with your earlier report. In the results section, the authors observed that Nimba had the shortest median time (<1 day) from symptom onset to hospital admission. Kindly reconcile the two statements and re-phrase accordingly.

Comment 35: We suggest the authors re-phrase this statement ‘leading to unnecessary admissions and ribavirin use as well as delayed appropriate treatment’ to ‘leading to unnecessary admissions and ribavirin use as well as delays in receiving appropriate treatment.’

Comment 36: The findings of the study highlighted a notable concentration of cases in Bong, Grand Bassa, and Nimba Counties. However, the authors did not provide or suggest possible explanations or hypotheses for these observations. Also, Figure 2 illustrates that these three Counties are adjacent to each other. Therefore, it would be beneficial for readers if the authors could delve into the potential influence of infected individuals migrating across County borders. This exploration could shed light on how population movements might contribute to the spread of Lassa fever in this region, offering valuable insights into disease dynamics and transmission patterns.

Comment 37: In the paragraph that describes the need for improved strategies, it might be useful to describe the need for a rapid point-of-care test for Lassa fever since this would be so helpful to triage patients for infection control purposes and treatment (especially if more effective antivirals become available).

References:

Comment 38: There is no need to include Indian Council of Medical Research in the first citation. 

Comment 39: Lassa should be capitalized in the third citation. 

Comment 40: Several of the article titles are presented in title case rather than having just the first letter capitalized (e.g., citation 7, 11, 12, and 21). 

Comment 41: There is a duplicate of Balogun OO et al. This is presented as reference 7 and 14.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

No

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Emerging infectious diseases, antimicrobial resistance, malaria, Chagas disease

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.

Dwalu Emmanuel Infectious Disease & Epidemiol, National Public Health Institute of Liberia, Monrovia, Montserrado, Liberia

Competing interests: No competing interest.

 21 3 2025

The Editor

F1000 Peer Review Journal

Ref: Response to Peer Reviewers’ Reports - F1000 Research Peer Review: Epidemiological characteristics and hospital outcomes of hospitalized Lassa fever cases during the 2022 – 2023 outbreak in Liberia

Dear F1000 Research Editorial Team,

Thank you for facilitating the peer review process for our manuscript. We have carefully addressed all the reviewers’ comments, and we believe that the revised manuscript now aligns with the journal’s requirements.

Below, we provide a line-by-line response to the reviewers’ comments, with the original comments highlighted in bold.

Sincerely,

Emmanuel Dwalu

Reviewer # 1:  

General Comment:

Overall, this manuscript provides an insightful epidemiological analysis of the Lassa fever outbreak and its impact on hospitalized patients in Liberia from 2022 to 2023. The authors present compelling findings that address an evolving public health issue in the context of an outbreak. They have adequately described the study setting & population, exposures (socio-demographic and epidemiological characteristics), and outcome measurements.

Fundamentally, this manuscript can contribute to the existing literature on Lassa fever by providing insights and understanding into the epidemiological dynamics and clinical outcomes associated with Lassa fever. Additionally, it expands on the substantial burden placed on hospitalized patients during the outbreak, thereby highlighting critical aspects for public health preparedness and response strategies.  

The title refers to the epidemiological characteristics and hospital outcomes of patients hospitalized with Lassa fever, but the manuscript has a moderate amount of material on the patients who were initially suspected to have Lassa fever but were not confirmed. We recommend focusing the analysis on the confirmed cases and only providing a brief description of the suspected cases at the beginning of the results. 

We appreciate the reviewer’s observation. However, in practice, suspected cases are initiated LF treatment while awaiting the test result, as timely intervention is crucial. By including both suspected and confirmed cases, we provide a comprehensive understanding of the complexities and real-world challenges faced in diagnosing and managing this disease. Therefore, we believe it is relevant to present both suspected and confirmed cases but went detail with the confirmed results.

[Specific Comments]

Comment 1: Rather than refer to cases in the title and the abstract, patients is a better term to use (there are sections where the term case is reasonable when summarizing statistics but not when describing the acute infection and outcomes of the pooled cohort of patients.

We have addressed this comment in the manuscript accordingly.

Comment 2: Second sentence of the Background of the abstract, outcomes should be plural as in “...and hospital outcomes of Lassa fever patients hospitalized...”.

We have updated the sentence.

 Comment 3: Abstract, Results. Recommend introducing the abbreviation IQR in the second sentence “...was 22 (interquartile range [IQR] 10-33)…”

We have added the abbreviation of interquartile range.

Comment 4: In the results section of the abstract, the authors mentioned that ‘290 (69%) of confirmed cases were <30 years.’ However, in the preceding sentence, the authors stated that there were 138 confirmed cases which equates to 33% positivity rate.  Once you start referring to confirmed cases, the denominator should be all confirmed cases (N = 138) yet the description “290 (69%) of confirmed cases were <30 years old” is incorrect. The next two percentages are based on the correct, smaller denominator.

We have revised the manuscript with the appropriate frequency and proportion.

Comment 5: Abstract, Results. Would consider removing mention of the mortality rate among the suspected but not confirmed Lassa fever cases. There is no information provided on the possible causes of febrile illness in these patients so readers will have no idea what they died from (probably a mix of malaria, bacteremia, and potentially other infectious etiologies). 

We thank the reviewer for the comment. We have revised the subheading to "deaths" instead of "mortality." We retained the presentation of deaths among unconfirmed cases due to the relatively high proportion (42/278, 15%), which warrants further investigation to improve in-hospital outcomes. We have added this recommendation to our paper to inform further research on Lassa fever in Liberia.

Comment 6: Abstract, Results. For the last sentence, please add the denominator for each group (age 40-49 years, age > or = 50 years).

We have addressed this comment accordingly.

Comment 7: Abstract, Conclusions. As suggested above, we would consider removing the phrase “substantial mortality, even among those who tested negative for the virus” since the causes of acute febrile illness in this group are not described. In contrast, it might be useful to mention the empirical use of ribavirin even for patients who were ruled out for Lassa fever, as this means many patients were exposed unnecessarily to a moderately toxic antiviral.

As noted in point 5 above, highlighting the relatively high death rates among individuals who tested negative for LF underscores the need to improve the management of suspected cases. Regarding ribavirin treatment, it was not continued for patients who tested negative for Lassa fever ("ruled out for Lassa fever").  To avoid confusion, we have revised the sentence to clearly state that individuals with negative RT-PCR results were discontinued from ribavirin treatment and discharged promptly.

We believe the main challenge regarding unnecessary antiviral exposure lies in diagnostic limitations, as LF symptoms often mimic those of other infections—a point we have discussed in detail.

Comment 8: Keywords. Suggest removing Liberia and Lassa fever since these are in the article title and all terms in the title will be picked up by search engines. And SORT IT may not be a searchable term. Consider adding ribavirin as a term.

We agree with the reviewer’s recommendation and have added ribavirin to the list of key words.

   [Introduction Section]

Comment 9: Mastomys natalensis in the third sentence of the ‘Introduction’ should be written in the standard scientific format as ‘Mastomys natalensis’. This same suggestion applies to the Discussion (third paragraph). 

We have revised the manuscript accordingly.

Comment 10: Introduction. Although the prevalence of frequently presenting symptoms in suspected Lassa fever cases is interesting, the more relevant symptoms are those demonstrated in patients who have been confirmed.

Thanks to the reviewer for the observations. We have included the symptoms reported by confirmed patients aligning with the study's scope { while a retrospective study in 2022 identified fever, fatigue, sore throat, loss of appetite, headache, vomiting, and myalgia as presenting symptoms among confirmed cases  [6,10]

Comment 11: Introduction. Suggest adding risk factors for mortality from Lassa fever as one of the objectives of this analysis.

We have added this objective to the manuscript.

[Methods Section]

Comment 12: It will be beneficial to the readers of the journal to know the type of cohort study conducted. Kindly specify it is retrospective or prospective (Study Design).

We have specified the type of cohort study as “a retrospective cohort study”.

Comment 12: Methods, Specific setting. Third line. Suggest rephrasing to “...for Lassa fever to identify suspect cases….”.

  We have changed detected to “identify”.

Comment 13: In the context of the first and second sentence of the Specific Setting of the Methods section, please include a phrase/sentence that indicates the reporting tool, means of communication, and timeliness of reporting (window period) employed by the community healthcare workers when reporting suspected cases to the nearest health facility for verification.   

We have updated the section as follows: At the community level, community health workers identified potential Lassa fever cases using the community/simplified case definition. Potential cases were referred to the nearest health facility for verification using the community trigger and referral form, detailing initial assessment to the nearby health facility. Referrals were made immediately or within 24 hours of identification, with a phone call or written message delivered to the nearby health facility.

Comment 14: Although this sentence (In the Study Settings section), ‘The IDSR case alert and laboratory submission forms were then completed and finally samples were collected’, is understood by the reviewer, for clarification and non-assumption by readers of the journal, the authors should indicate the specific type of sample that was collected from the patients for Lassa fever testing.

We have added “whole blood” to indicate the sample types.

Comment 15: Generic drug names should be presented in lower case—ribavirin should not be capitalized.

We have revised accordingly.

Comment 16: Were there specific selection criteria for the use of ribavirin? “…

Yes, ribavirin was administered only to suspected Lassa fever cases while awaiting laboratory test results and continued for confirmed cases but the treatment was discontinued for individuals with a negative result. We have clarified this in the methods section as” While awaiting the laboratory test results, individuals were isolated and admitted to a treatment center where IV ribavirin treatment was initiated if they were classified as “suspected Lassa fever patients”.

Comment 17: Concerning this statement, ‘Daily assessments were performed using charts to monitor the patients’ progress’, please specify (concisely) the assessments done.

We have revised the sentence as follows: Daily assessments were conducted, including routine examinations such as physical exams, vital signs monitoring, malaria tests, and other relevant checks, using charts to track the patients' progress.

Comment 18: Please rephrase the statement ‘Individuals with negative results were discharged immediately after the Ribavirin treatment was discontinued’ to something like ‘Individuals with negative results after a repeat RT-PCR testing were discharged immediately after the ribavirin treatment was discontinued.’ If this is not what the authors are trying to convey, kindly clarify.

There was no repeat of the RT-PCR test before discharging the patients. Once the results from the National Public Health Reference Laboratory were negative, the patient was discharged accordingly, and the ribavirin treatment was discontinued.  We have clarified this as follows: Individuals with negative RT-PCR results were discontinued from ribavirin treatment and discharged immediately.

Comment 19: Kindly expound the phrase ‘and selected treatment centers’ in the sentence ‘Due to incomplete information in some records, we reconciled the data using the county surveillance database and selected treatment centers.’ Succinctly expand the phrase to provide more context to readers.

We agree with the reviewer and have revised the manuscript as “Due to incomplete information in some patient records submitted at the national level, we reconciled the data using the county surveillance database. We opted for four sites instead of all eight because these were the sites where majority of the cases were reported from and reviewing records from these sites would address the incomplete record issue or data quality issues in the existing data set used for the study.

Comment 20: Statistical analysis. The sentence ‘We used log binomial regression to assess the association between socio-demographics, clinical characteristics, and case fatality rates’ should be revised to ‘We used log binomial regression to assess the association between socio-demographics, clinical characteristics, and mortality.’ CFR is a population-level indicator, and it appears the authors inadvertently substituted this in place of mortality.

The reviewer’s observation is highly appreciated and we have revised the manuscript.

Comment 21: Statistical analysis. The term risk ratio should be in lower case.

We agree with the reviewer and revised accordingly.

[Results Section]

Comment 22: Results. Second sentence. Missing a period at the end of the sentence.

We have added the period.

Comment 23: Results. The authors revealed that some individuals were excluded in the statement ‘Among 439 individuals, about one quarter (26%, 116/439) were excluded due to missing data on date of symptom onset or admission’. Specify what the individuals were excluded from, e.g., from the regression analyses or Table 2. Kindly clarify.

We have revised the manuscript.

Comment 24: Results. The utility of the data in Table 2 is relatively limited. The median time from symptom onset to hospital admission was similar across age groups (by a day or two), gender (half a day), and county (except Nimba). This table could be removed and a sentence or two added to the Results to describe the major findings.

We appreciate the comment/recommendation and have revised the manuscript.

Comment 25: In Table 4, was there any reason why Bong County was selected as the reference group? Or was the county selected randomly?

As a common practice in modelling, we statistically selected the county with the highest number of cases to enhance interpretability and statistical power.

Comment 26: Results. It would be interesting and potentially informative to compare patients who had confirmed Lassa fever (PCR+) who received ribavirin vs. those who did not receive this antiviral treatment. Are there any significant differences in their clinical or demographic characteristics?

Thanks for the comment. We did not compare patients with confirmed LF (PCR+) who received ribavirin versus those who did not because this was not the focus of our study. However, in our regression analysis, we assessed the effect of ribavirin treatment on outcomes. We have added a sentence that there was no significant difference between those who received treatment and those who did not. We hope this addresses the concern.

Comment 27: Table 4, Ribavirin treatment (yes or no) was not included in the bivariate and multivariable models. Historical data and corresponding studies have shown a decrease in mortality among those administered (especially within 6 days from symptom onset). For comprehensiveness, it would be important to explore and demonstrate the associations between the administration of the treatment and mortality. Additionally, the adjusted (multivariable) models with its findings would have more validity if this was also accounted for.

We have included a variable “ribavirin treatment” in the regression analysis

We could not analyze the onset of symptoms and treatment initiation because we do not have the data. We have included it as a limitation.

Comment 28: Table 4. The authors used a reference cutoff of 7 days to assess the association between duration from symptom onset to hospital presentation. However, we recommend the authors to explore the 4 days as the reference cutoff. Earlier in the results section, the authors reported the median symptom-hospital duration to be 4 days with an IQR of 2 to 7. For validity reasons, kindly conduct an exploratory analysis/sensitivity analysis to explore this new cutoff.

We appreciate the reviewer’s recommendation. We have revised the manuscript to include (1-3 days, and 4-6 days) for the duration from symptom onset to hospital admission.

  [Discussion Section]

Comment 29: The abbreviation CFR was introduced much earlier in the manuscript. It should be used on the first and third lines of the beginning of the Discussion. Also, we recommend using the term ‘patients’ rather than cases in the first sentence of the Discussion.

We have updated the manuscript.

Comment 30: As noted earlier, the group with suspected Lassa fever who were not confirmed is likely a mixture of different etiological agents. Suggest removing the last sentence of the first paragraph of the Discussion for this reason.

We appreciate the comment. However, as stated earlier, it is concerning for individuals who are suspected of having Lassa fever, test negative, and yet still die, without a clear understanding of the cause. Therefore, we aimed to use these findings to recommend further research into the causes of death among patients who test negative for Lassa fever which is not reported in literature.

Comment 31: The last statement of the second paragraph ‘Additionally, effective community infection prevention control programmes in endemic areas might be of help’ is quite vague. The authors (in their recommendation) should kindly be specific in who implements and ensures the effectiveness of the community IPC programmes.

We totally agree with the reviewer’s comment/suggestion and have specified who does the implementation as “Additionally, effective community infection prevention control programmes in endemic areas by the Ministry of Health and the National Public Health Institute of Liberia might be of help”

Comment 32: In the third paragraph, in the third and second to last lines, might be better to state “A clean environment in both the house and outside environment...”

We agree with the reviewer’s comment/suggestion and have revised the manuscript “A clean environment in both the house and the outside environment might help to reduce the contact between rats and humans”.

Comment 33: Additional information is required for this statement ‘Similar to other studies, the CFR was higher among elderly people. However, the reasons for this are unclear, warranting further research in understanding the occurrence.’ Yes, additional studies are warranted to better understand the pathophysiological mechanism contributing to the high CFR among elderly population. However, the authors should state possible reasons for high CFR being demonstrated among this population, e.g. comorbidities, waning immunity, etc.

We appreciate the reviewer’s comment and have revised the manuscript.

Comment 34: The statement ‘The high CFR in Nimba could be due to the fact that the county is endemic for Lassa fever, with the majority of cases detected at the health facility rather than through early community detection, which impacts on early treatment’ is a bit conflicting with your earlier report. In the results section, the authors observed that Nimba had the shortest median time (<1 day) from symptom onset to hospital admission. Kindly reconcile the two statements and re-phrase accordingly.

We totally agree and appreciate the reviewer for this observation and recommendation. The high CFR in Nimba could be likely due to the fact that cases presenting with fever are sometimes or most times being treated for malaria because of their presentations before suspecting Lassa fever, which impacts treatment outcome among the patients

We have revised the manuscript accordingly.

Comment 35: We suggest the authors re-phrase this statement ‘leading to unnecessary admissions and ribavirin use as well as delayed appropriate treatment’ to ‘leading to unnecessary admissions and ribavirin use as well as delays in receiving appropriate treatment.’

We agree and have accordingly revised the manuscript.

Comment 36: The findings of the study highlighted a notable concentration of cases in Bong, Grand Bassa, and Nimba Counties. However, the authors did not provide or suggest possible explanations or hypotheses for these observations. Also, Figure 2 illustrates that these three Counties are adjacent to each other. Therefore, it would be beneficial for readers if the authors could delve into the potential influence of infected individuals migrating across County borders. This exploration could shed light on how population movements might contribute to the spread of Lassa fever in this region, offering valuable insights into disease dynamics and transmission patterns.

We have revised as “These counties are endemic to Lassa fever and adjacent to each other, which likely has potential influence of infected individuals to migrate across the borders due to intermarriages, farming activities, and in search of jobs.  Thus, contributing to the spread of Lassa fever in these regions”.

   Comment 37: In the paragraph that describes the need for improved strategies, it might be useful to describe the need for a rapid point-of-care test for Lassa fever since this would be so helpful to triage patients for infection control purposes and treatment (especially if more effective antivirals become available).

We have recommended a rapid point of care test for Lassa fever.

References:

Comment 38: There is no need to include Indian Council of Medical Research in the first citation.

We have revised accordingly.

Comment 39: Lassa should be capitalized in the third citation.

The comment has been acted upon accordingly.

Comment 40: Several of the article titles are presented in title case rather than having just the first letter capitalized (e.g., citation 7, 11, 12, and 21).

We have addressed the comment.

Comment 41: There is a duplicate of Balogun OO et al. This is presented as reference 7 and 14.

We thank the reviewer for this observation and we have rectified it.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

Is the study design appropriate and is the work technically sound?

Yes 

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

No

Are the conclusions drawn adequately supported by the results?

Yes

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Emerging infectious diseases, antimicrobial resistance, malaria, Chagas disease

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.