Correspondence Address:
Babasola O Olugasa
Centre for Control and Prevention of Zoonoses, Department of Veterinary Public Health and Preventive Medicine, 101 Faculty of Veterinary Medicine, University of Ibadan, 200002, Ibadan, Oyo State, Nigeria

Full Text

Sir,

Liberia is in the epicenter of Lassa fever (LF), being one of the Mano River Union (MRU) countries that are most heavily affected by the disease in West Africa. [1],[2] Liberia records annual cases since 1972. [2],[3],[4],[5] During the civil conflict of 1989-2003, health care services in the country were interrupted for about a decade and half. [2],[6] This situation was expected to aggravate the incidence of LF. As a result, ministers of health of the MRU countries endorsed a sub-regional strategic plan of action for LF prevention and control, 2004-2008. [2] We assessed and mapped LF surveillance data in the country, 2008-2012. We used an integration of two sets of epidemiological tools that are fast-gaining prominence in disease mapping. [7] Global Positioning System [7] was used to capture data, and integrated into geographic information systems (GIS) to produce thematic maps of LF; an acute and sometimes fatal viral hemorrhagic illness. [1],[3] The aim of this letter is to present a map of LF spatial pattern and some proposed areas of Lassa virus (LASV) activity, which summarizes 5-year retrospective records of LF countrywide during the postconflict passive surveillance of the disease in Liberia.

A retrospective review of cases of LF among human patients was conducted on hospital records, 2008-2012. Age, gender, month and year of patients' presentation and of specimens' collection for laboratory confirmation were captured where available. Cases were selected, and enlisted based on LF definition by the World Health Organization (WHO, 2004). [8] All confirmed, probable and suspected cases were geo-referenced unto a base map of Liberia as earlier described by Olugasa et al., 2014. [9] The geographic co-ordinates of locations of LF cases were summated into thematic maps, and sub-classified into age, LASV-activity and seasonal pattern.

Seasonal index was computed based on frequency of cases during the months of the year, categorized into rainy and dry seasons. Average percentage method, which involves the expression of LF cases as a percentage of the total over the 5-year period was used. Percentages for corresponding months (rainy and dry seasons) of different years were averaged. Finally, Arc GIS 10.1® software environment (Environmental Systems Research Institute, Redlands, California, USA) was used for map design.

Thematic maps of age, gender and spatial distributions revealed strength and weaknesses of surveillance. Consistently, cases presented were relatively close to a hospital facility. A finding that may indicate a challenge for patients in remote places to access distant health facilities for treatment. [10] A little above two-thirds (152/225) of the total hospital records of suspected humans cases of LF were negative on laboratory test for LASV. Among laboratory-confirmed cases of LF (n = 73), the average age of LF patients was 28 years old (±14 years standard deviation). The modal (16 [21.9%]) age group was 20-29 years old, followed by 30-39 years old (12 [16.4%]) and 10-19 years old (10 [13.7%]). These three age groups accounted for 52% of the cases. However, age records of 22 (30.1%) cases were unspecified. Hence, a higher proportion of LF cases was found among young adults with ages between 20 and 39 years. Females 41 (56.2%) accounted for a higher proportion of cases compared with their male 32 (43.8%) counterparts.

More LF cases 38 (52.1%) were reported in the dry season, compared to 35 (47.9%) reported in the rainy season. Results of space-time cluster analysis indicated southward spread of LF cases from Nimba County during August-September, 2008 to Grand Bassa during June-December, 2009 and to Bong County during September-November, 2012 [Figure 1]. All the three space-time clusters were statistically significant (0.01 ≤ P ≤ 0.1). The presence of LF cases was not previously described in Grand Bassa County, Liberia. Hence, we have identified the first record of coastal southward spread of LF. This discovery offered an empirical basis for the investigation of Grand Bassa County for explanatory variables of LF outbreaks and spread in the local environment. [9]{Figure 1}

Time cluster of LF occurred within late August and September, corresponding to the ending of the rainy season and the peak of grains harvest in Liberia. Difference in seasonal frequencies was not statistically significant (P > 0.05). Thus, we propose that the postconflict retrospective map of LF in the fifteen counties be grouped into three categories, indicating likely activities of LASV [Figure 1], not vividly shown due to incomplete data and under-reporting. This is well-portrayed, in Sierra Leone, where a dramatic rise in suspected LF cases that were identified rose from under a 100 in 2008 to nearly 600 in 2011, following the efforts of the international collaboration between Kenema Government Hospital and Tulane University. [11] We concluded that there is a need for active surveillance and improvement in LF reporting in Liberia. To ensure this and improve research and education, we fostered collaboration with the University of Ibadan Centre for Control and Prevention of Zoonoses, Nigeria. The team has developed a time-trend model for analyzing and predicting the disease, [12] which holds a high promise for effective engagement of the MRU governments, and stakeholders, including the academia in rebuilding and strengthening LF surveillance and control systems in Liberia.

References