Abstract
Objective
This study aimed to assess the sociodemographic determinants of knowledge of Lymphatic Filariasis (LF) among refugees in the Lolo and Mbile camps of Cameroon’s East region.
Methods
We conducted a community-based cross-sectional study among 528 refugees aged ≥ 18 years, selected via a multi-stage cluster sampling approach. Data were collected using pre-tested questionnaires administered by trained interviewers. Knowledge of LF was a dichotomous outcome based on recognition of the disease or its local names. We performed bivariate analyses followed by a multivariable logistic regression model to identify independent factors associated with LF knowledge, adjusting for potential confounders and testing for interactions.
Results
Overall LF knowledge was critically low (17.1%). In the adjusted multivariable model, formal education emerged as the strongest independent predictor (Adjusted Odds Ratio, aOR = 4.16; 95% CI: 1.60–10.83). A significant interaction revealed that Koranic education was a strong predictor in the Mbile camp (combined aOR = 4.1; 95% CI: 1.03–15.51) but not in Lolo. Other factors associated with higher knowledge in the adjusted model included being a household head (aOR = 1.86; 95% CI. 0.98–3.54) and shorter duration of residence.
Discussion
LF knowledge is alarmingly low in these refugee settings. The powerful effect of formal education and the context-dependent role of Koranic education highlight that awareness campaigns must be tailored to the specific educational and sociocultural landscape of each refugee camp. Strengthening both formal and context-relevant non-formal education systems should be a cornerstone of future LF elimination programs in humanitarian contexts.
Keywords: Sociodemographic characteristics, Knowledge, Lymphatic filariasis, Health education, Refugee health, Cameroon
Introduction
Lymphatic filariasis (LF), commonly known as elephantiasis, is one of the most debilitating Neglected Tropical Diseases (NTDs) and persists as a public health challenge across tropical and subtropical regions [1–3]. As a parasitic infection caused by filarial worms (Wuchereria bancrofti, Brugia malayi, and Brugia timori) and transmitted through mosquito vectors, LF leads to progressive and irreversible clinical manifestations, including lymphedema, hydrocele, and elephantiasis, which result in severe disability, social stigmatization, and substantial economic losses for affected individuals and communities [4–6]. The World Health Organization (WHO) has classified LF as a globally significant public health problem, with current estimates indicating that approximately 120 million people are infected worldwide and over 1.2 billion people are at risk across 83 endemic countries [7–9].
Within the global LF burden, sub-Saharan Africa bears a disproportionate share, accounting for approximately 38% of the global at-risk population. Recent epidemiological assessments indicate that more than 420 million people across the African region remain at risk of infection, with documented cases including 4.6 million individuals suffering from lymphedema and over 10 million affected by hydrocele [10, 11]. The disease predominantly affects impoverished populations with limited access to preventive healthcare services and health education [12, 13], creating a vicious cycle where disease-related disability further exacerbates economic hardship.
Refugee populations in humanitarian settings represent a particularly vulnerable subgroup for LF and other NTDs due to a confluence of risk factors: overcrowded living conditions with inadequate sanitation infrastructure, limited access to healthcare services, heightened exposure to vector mosquitoes due to substandard housing, and minimal health literacy levels [14–16]. The complex interplay of these factors creates an environment highly conducive to the transmission and progression of LF. Previous research has established that population displacement and migration significantly influence the epidemiology of NTDs, with refugee camps often serving as disease transmission hotspots [14, 17, 18]. Additionally, social conditions and traditional beliefs prevalent in refugee communities may further complicate the understanding of disease and appropriate health-seeking behaviors [19].
In Cameroon, the East Region hosts a significant refugee population, primarily from the Central African Republic, residing in designated camps, including Lolo and Mbile. These camps collectively shelter over 26,000 refugees under the mandate of the United Nations High Commissioner for Refugees (UNHCR) [20]. The geographical and environmental characteristics of this region, including the Guinea-Congolian climate, high humidity, extensive hydrographic networks, and dense semi-deciduous forest, create ideal conditions for the proliferation of mosquito vectors responsible for LF transmission. Despite established risk factors and documented endemicity of LF in Cameroon, a critical evidence gap exists regarding refugees baseline knowledge, attitudes, and practices regarding LF. Previous studies conducted on LF have predominantly focused on general population groups or specific endemic communities [21–24], with no dedicated research examining LF awareness among refugee populations.
Understanding the sociodemographic determinants of health knowledge represents a fundamental prerequisite for developing targeted, effective, and culturally appropriate health education interventions. Previous research in non-refugee settings has identified educational attainment, gender, age, and socioeconomic status as significant predictors of LF knowledge [21–23] However, the unique socioeconomic, cultural, and environmental context of refugee populations necessitates context-specific investigations, as findings from general populations may not directly translate to humanitarian settings. Identifying specific population subgroups with particularly low levels of LF knowledge would enable more efficient targeting of limited health education campaign resources.
The primary objective of this study is to assess the association between sociodemographic characteristics and LF knowledge among refugees in the Lolo and Mbile camps of Cameroon. The findings from this research will inform the design and implementation of culturally appropriate LF control strategies for refugee populations in Cameroon, contributing to the broader global effort to eliminate LF, as outlined in the WHO NTD roadmap 2030.
Methods
Study area and population
The survey was conducted in the Lolo and Mbile health areas, both of which are located in the Ndelele Health District. The Ndelele Health District covers the Kentzou and Ndelele subdivisions. However, the Lolo and Mbile health areas are specifically located in the Ndelele sub-division, which is part of the Kadey division in the East Region of Cameroon. The area is bordered by the Batouri sub-division to the north, the Gari gombo sub-division to the south, the Mbang sub-division to the west, the Bombette River to the east, which forms the border with the Central African Republic (CAR), and the Kadey River to the northeast, which separates it from the commune of Bombe. It covers an area of 4,000 km² and is divided into six health areas [19]. The Ndelele sub-division is home to two main ethnic groups: The Bantus, who make up a large majority of the population, and the Baka, the forest’s indigenous minority. However, due to the proximity of the border with the Central African Republic, a large influx of refugees has migrated into the study area. The localities of Lolo and Mbile host 13,614 and 12,843 refugees, respectively, under the United Nations High Commissioner for Refugees [20].
The study area experiences high humidity levels, with an average annual temperature of 28 °C. It is characterized by four distinct seasons: a short rainy season from March to June, a short dry season from July to August, a long rainy season from September to November, and a long dry season from December to March. However, climate change may disrupt this pattern. The vegetation consists mainly of dense forest and wooded savannah, with scattered gallery forests. The soil is sandy-clayey in the savannah and lateritic towards the forest. The hydrographic network is relatively extensive and includes the Kadey River, which flows through the subdivision on the northeast side and directly into the Sangha (Congo Basin). Its main tributary on the Mindourou side is the Doume River. The communities have schools and health centers where residents seek medical treatment. Many houses have unprotected windows, holes in the walls, and large open eaves, making it easy for mosquitoes to enter.
Study population and sampling technique
The source population for this study included all adult refugees (≥ 18 years) residing in the Lolo and Mbile camps who had been settled for at least six months, ensuring participants had sufficient exposure to local health information systems. Inclusion criteria comprised: (1) age ≥ 18 years; (2) minimum six months residence in the camp; (3) provision of informed consent; and (4) ability to comprehend and respond to questions in French, English, or the local Fulfulde language. Exclusion criteria included: (1) severe cognitive impairment preventing meaningful participation; (2) acute illness at the time of data collection; and (3) temporary visitors to the camps.
We determined the minimum required sample size using the single population proportion formula for cross sectional studies:
 |
where: Ζ = Ζ-score for 95% confidence level (1.96), P = assumed proportion of the population with LF knowledge (0.5, to yield maximum sample size), and e = margin of error (0.045). This calculation yielded a minimum sample of 474 participants. We adjusted this figure upward by 10% (to 528) to account for potential non-response and incomplete questionnaires, thereby enhancing the study’s statistical power.
We employed a multi-stage cluster sampling approach to ensure representative recruitment while accounting for the structural organization of the refugee camps. In the first stage, we divided each camp into geographical clusters based on existing administrative sectors, with probability proportional to population size. We randomly selected 12 clusters from Lolo camp (hosting 13,614 refugees) and 10 clusters from Mbile camp (hosting 12,843 refugees). In the second stage, we systematically sampled households within each selected cluster using a predetermined sampling interval. Finally, in the third stage, we used simple random sampling (random number table) to select one eligible adult from each recruited household. This methodological approach minimized selection bias and ensured proportional representation of both camp populations, resulting in a final sample distribution of 271 participants from Lolo and 250 from Mbile (total 521, with eventual recruitment of 528 participants accounting for replacements) (Table 1).
Table 1.
Repartition of sample
| Refugee camps | Population distribution | Sample distribution |
|---|---|---|
| Lolo Camp | 13.614 | 271 |
| Mbile Camp | 12.843 | 250 |
| Total | 26.457 | 521 |
Source: Author
Data collection instrument and procedures
We developed a comprehensive, structured questionnaire through an extensive process including: (1) systematic review of existing literature on LF knowledge assessment; (2) consultation with LF experts from the Cameroonian Ministry of Public Health; and (3) preliminary focus group discussions with refugee community representatives to ensure cultural and contextual appropriateness. The questionnaire was prepared in English and French.
The final instrument comprised four distinct sections: (1) Sociodemographic characteristics (12 items including age, sex, marital status, education, income, household characteristics); (2) LF knowledge assessment (16 items covering disease recognition, transmission modes, clinical symptoms, prevention methods, and treatment options); (3) Information sources and health-seeking behaviors (5 items); and (4) Environmental factors and vector control practices (4 items). Knowledge questions used both open- and closed-ended formats to assess spontaneous recall and recognition-based knowledge.
We conducted a comprehensive pretest of the questionnaire with 30 refugees (non-participants) from a similar camp setting to assess clarity, comprehensibility, cultural acceptability, and approximate administration time. Based on pretest findings, we refined the phrasing of questions, response categories, and translation nuances.
Data collection was conducted by 12 trained community health workers who were fluent in French, English, or Fulfulde. These data collectors received five days of intensive training covering research ethics and informed consent procedures; LF clinical and epidemiological aspects; interview techniques; questionnaire administration protocols; and data quality assurance methods. The training included role-playing exercises and one day of supervised field practice. Data collection was conducted through face-to-face interviews at participants homes, with privacy measures in place to ensure the confidentiality of responses. Each interview lasted approximately 30–45 min. We implemented rigorous supervision throughout data collection, with daily review of completed questionnaires by field supervisors and research team members to ensure completeness and consistency.
Outcome variables
In this analysis, we used one outcome variable: knowledge of LF, defined as having heard of LF. Given the scientific nature of the term “LF”, semi-structured interviews with health workers and some refugees allowed us to identify other local names for the disease: elephantiasis, Boutigo Bandou, Guilgou, and Hindou. The variable was therefore based on a question asking whether the respondent had ever heard of (i) lymphatic filariasis, (ii) elephantiasis, (iii) Boutigo Bandou, (iv) Guilgou, or (v) Hindou. A respondent who answered ‘yes’ to any of these terms received a score of 1, and a respondent who answered ‘no’ to all received a score of 0. This score represented the information recorded by the interviewer. Knowledge was thus operationalized and analyzed as a dichotomous variable (yes/no). Our study did not categorize knowledge into graded levels (e.g., low/high) beyond this fundamental awareness distinction, as the primary aim was to identify factors associated with basic disease recognition. Further questions were asked of those who had heard of the disease regarding its causes, signs and symptoms, and prevention methods.
Explanatory variables
Explanatory variables included pre-defined sociodemographic characteristics: Camp location (Lolo/Mbile); Age categories (18–35 years, 35–45 years, and ≥ 45 years); Sex (male/female); Marital status (in couple/not in couple); Household status (head/member); Household size (0–3, 4–5, ≥ five persons); Number of children (0, 1–4, ≥ 5); Religion (Muslim/Other); Country of origin (CAR/Other); Length of stay in the region (< 1 year, 1–5 years, > 5 years); Education level (no formal education, Koranic, formal); Monthly income in CFA francs (< 36,000, 36,000–200,000, > 200,000); Health insurance possession (yes/no); and Health booklet possession (yes/no). These categorizations were established a priori based on previous literature and preliminary assessments of variable distributions within the population.
Data analysis
Logistic regression was used to identify sociodemographic factors associated with knowledge of lymphatic filariasis. Measures of association were estimated using odds ratios (ORs). In the first stage, bivariate regression analyses were used to independently assess the relationship between the outcome variable and each potential predictor. Subsequently, a multivariable logistic regression model was constructed using a backward elimination approach. All variables associated with the outcome at a p-value < 0.05 in the bivariate analysis were included in the initial model. The variables “refugee camp” and “sex” were selected a priori and retained in the final model regardless of their statistical significance. Other variables were retained only if they were important confounding factors or significantly associated with the outcome. The absence of confounding was verified by observing whether the inclusion of a potential confounder altered the estimate of a key variable by more than 10%. To assess potential effect modification (interaction), an interaction term between the refugee camp and education level was added to the model. Prior to interpreting the final model, multicollinearity was assessed by calculating variance inflation factors (VIFs) for all independent variables. All VIF values ranged from 1.05 to 3.23 (mean = 1.98), indicating the absence of problematic multicollinearity. In the final model, adjusted odds ratios (aORs) and their 95% confidence intervals (95% CIs) were calculated for all variables. The analyses were performed using Stata 2015 software.
Results
Sociodemographic characteristics of the study population
As part of our study, we interviewed 528 residents of the Lolo and Mbile refugee camps, all of whom had been living there for at least 6 months. Table 2 presents the complete distribution of respondents according to their sociodemographic characteristics.
Table 2.
Sociodemographic characteristics of the study population
| Variables | Total (N = 528) | Lolo Camp (n = 313) | Mbile Camp (n = 215) |
|---|---|---|---|
| n (%) | n (% within camp) | n (% within camp) | |
| Age group | |||
| 18–34 years | 326 (61.74) | 201 (64.22) | 125 (58.14) |
| 35–64 years | 183 (34.66) | 101 (32.27) | 82 (38.14) |
| ≥ 65 years | 19 (3.60) | 11 (3.51) | 8 (3.72) |
| Mean age | 33.39 years | 32.63 years | 34.50 years |
| Sex | |||
| Men | 177 (33.52) | 141 (45.05) | 36 (16.74) |
| Women | 351 (66.48) | 172 (54.95) | 179 (83.26) |
| Marital status | |||
| In couple | 405 (76.70) | 240 (76.68) | 165 (76.74) |
| Not in a couple | 123 (23.30) | 73 (23.32) | 50 (23.26) |
| Household size | |||
| 0–3 persons | 220 (41.67) | 90 (28.75) | 130 (60.47) |
| 4–5 persons | 107 (20.27) | 80 (25.56) | 27 (12.56) |
| ≥ 5 persons | 201 (38.07) | 143 (45.69) | 58 (26.98) |
| Average household size | 4.68 persons | 5.41 persons | 3.63 persons |
| No. of children | |||
| No children | 69 (13.07) | 45 (14.38) | 24 (11.16) |
| 1–4 children | 298 (56.44) | 164 (52.40) | 134 (62.33) |
| ≥ 5 children | 161 (30.49) | 104 (33.23) | 57 (26.51) |
| Average number of children | 3.46 children | 3.62 children | 3.23 children |
| Household status | |||
| Another member | 327 (61.93) | 179 (57.19) | 148 (68.84) |
| Head of household | 201 (38.07) | 134 (42.81) | 67 (31.16) |
| Level of education | |||
| No education | 361 (68.37) | 228 (72.84) | 133 (61.86) |
| Koranic education | 137 (25.95) | 64 (20.45) | 73 (33.95) |
| Formal education | 30 (5.68) | 21 (6.71) | 9 (4.19) |
| Religion | |||
| Muslims | 517 (97.92) | 305 (97.44) | 212 (98.60) |
| Christians | 11 (2.08) | 8 (2.56) | 3 (1.40) |
| Country of origin | |||
| CAR | 523 (99.05) | 309 (98.72) | 214 (99.53) |
| Nigeria | 3 (0.57) | 3 (0.96) | 0 (0.00) |
| Chad | 2 (0.38) | 1 (0.32) | 1 (0.47) |
| Income | |||
| Less than 36,000 | 348 (65.91) | 191 (61.02) | 157 (73.02) |
| 36 000–200 000 | 175 (33.14) | 117 (37.38) | 58 (26.98) |
| More than 200,000 | 5 (0.95) | 5 (1.60) | 0 (0.00) |
| Health insurance | |||
| Yes | 507 (96.02) | 294 (93.93) | 213 (99.07) |
| No | 21 (3.98) | 19 (6.07) | 2 (0.93) |
| Health booklet | |||
| Yes | 401 (75.95) | 229 (73.16) | 172 (80.00) |
| No | 127 (24.05) | 84 (26.84) | 43 (20.00) |
Source: Author. Percentages in the Total column are column percentages based on the overall sample (N=528). Percentages in the Lolo and Mbile columns are column percentages based on the total number of respondents in each camp (n=313 and n=215, respectively). Percentages may not sum to 100% due to rounding
The study population had a mean age of 33 years. A majority of respondents were female (66.48%), in a marital union (76.70%), and reported having received no formal education (68.37%). Nearly all refugees were from the Central African Republic (99.05%) and of Muslim faith (97.9%). The detailed distribution of these characteristics, including by camp of residence, is provided in Table 2.
Level of knowledge of lymphatic filariasis (LF)
Figure 1 illustrates that the majority of the sample, comprising over 80%, lacked awareness of lymphatic filariasis, also known as elephantiasis. Conversely, a minority of the sample (17.05%) demonstrated knowledge of the disease (Fig.2).
Fig. 1.
Knowledge of Lymphatic Filariasis (LF). Source: Author (N = 528)
Fig. 2.
Information Channels on LF. Source: Author (N = 528)
Respondents’ knowledge about the causes of lymphatic filariasis
In Table 3, among the refugees who were already aware of lymphatic filariasis, 72.22% reported being unaware of its causes. However, 11.11% identified mosquito bites as the vector of transmission, with 10% of these responses coming from the Lolo and 1% from the Mbile camp.
Table 3.
Knowledge of LF
| Indicative questions | Response Categories | Total | Lolo camp | Mbile camp |
|---|---|---|---|---|
| n (%) | n (%) | n (%) | ||
| Have you ever heard about LF? | (N = 528) | (n = 313) | (n = 215) | |
| Yes | 90 (17.05) | 67 (21.41) | 23 (10.70) | |
| No | 438 (82.95) | 246 (78.59) | 192 (89.30) | |
| If ‘Yes’, how can you get LF? | (n = 90) | (n = 67) | (n = 23) | |
| Injury | 4 (4.44) | 4 (5.97) | 0 (0.00) | |
| Mosquito bites | 10 (11.11) | 9 (13.43) | 1 (4.35) | |
| Washing with contaminated water | 11 (12.22) | 8 (11.94) | 3 (13.04) | |
| Don’t know | 65 (72.22) | 46 (68.66) | 19 (82.61) | |
| If ‘Yes’, what are the signs of LF? | (n = 90) | (n = 67) | (n = 23) | |
| Pain and swelling in the feet and arms | 29 (32.22) | 25 (37.31) | 4 (17.39) | |
| Body Itching | 11 (11.11) | 11 (16.42) | 0 (0.00) | |
| Don’t know | 50 (55.55) | 31 (46.27) | 19 (82.61) | |
| If ‘Yes’, how can you prevent LF? | (n = 90) | (n = 67) | (n = 23) | |
| Washing with clean water | 13 (14.44) | 13 (19.40) | 0 (0.00) | |
| Sleep under a mosquito net | 4 (4.44) | 3 (4.48) | 1 (4.35) | |
| Getting vaccinated | 4 (4.44) | 3 (4.48) | 1 (4.35) | |
| Don’t know | 69 (76.67) | 48 (71.64) | 21 (91.30) | |
| If ‘Yes’, how is LF diagnosed? | (n = 90) | (n = 67) | (n = 23) | |
| Hospital medical examination | 26 (28.89) | 22 (32.84) | 4 (17.39) | |
| Don’t Know | 64 (71.11) | 45 (67.16) | 19 (82.61) | |
| If ‘Yes’, what can you do to treat LF? | (n = 90) | (n = 67) | (n = 23) | |
| Medical care | 43 (47.78) | 36 (53.73) | 7 (30.43) | |
| Traditional care | 1 (1.11) | 1 (1.49) | 0 (0.00) | |
| Don’t Know | 46 (51.11) | 30 (44.78) | 16 (69.57) |
Source: Author (N = 528)
For the first question (“Have you ever heard…”), percentages are based on the total sample (N = 528 for Total, n = 313 for Lolo, n = 215 for Mbile). For all subsequent conditional questions, percentages are based on the number of respondents who answered “Yes” to having heard of LF (Total: n = 90; Lolo: n = 67; Mbile: n = 23). Row percentages may not sum to 100% due to rounding
Respondents’ knowledge of lymphatic filariasis signs and symptoms
During the survey, refugees were asked if they were aware of lymphatic filariasis (LF) and to describe its symptoms. According to Table 3, nearly 55% of the surveyed refugees reported being unaware of the apparent signs and symptoms of LF. Some 32% linked pain and swelling in the feet and arms to the disease, and of these, over 27% live in Lolo. A further 11% of respondents indicated that itchy skin is a symptom of lymphatic filariasis.
Respondents’ understanding of prevention and management of lymphatic filariasis
In the survey, respondents who indicated awareness of lymphatic filariasis (LF) were queried about their knowledge of preventive measures against the disease. Over 76% of respondents reported being unaware of any such measures. Only 4% of respondents who were aware of LF recognized that impregnated mosquito nets could be used to prevent infection.
Respondents’ knowledge of lymphatic filariasis treatment
In the survey on lymphatic filariasis, among participants who had heard of the disease, 51.11% reported being unaware of treatment options. Almost half of those surveyed, 47.78%, mentioned medical treatment as a method of care, while a tiny minority, 1.1%, mentioned the use of traditional treatments.
Association between knowledge of lymphatic filariasis and some sociodemographic factors of the participants: Bivariate logistic regression analysis
Table 4 presents the bivariate logistic regression analyses examining associations between sociodemographic characteristics and LF knowledge. Several variables demonstrated statistically significant bivariate associations. The camp location was strongly associated with knowledge: Mbile camp residents were significantly less likely to know about LF than Lolo camp residents (COR = 0.43, 95% CI: 0.26–0.73; P = 0.002). Gender disparities were evident, with males having nearly twice the odds of LF knowledge compared to females (COR = 1.96, 95% CI: 1.23–3.11; P = 0.004). Household status also showed significance, with household heads demonstrating 80% higher odds of knowledge compared to other household members (COR = 1.80, 95% CI: 1.14–2.85; P = 0.011).
Table 4.
Crude Odds ratio of knowing Lymphatic Filariasis (bivariate logistic regression)
| No | Independent Variables | Don’t know FL (%) | Know FL (%) | Crude Odds Ratio (COR) | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|---|---|---|
| 1 | Refugee camp | ||||||
| Lolo (ref) | 246 (46.59) | 67 (12.69) | |||||
| Mbile | 192 (36.36) | 23 (4.36) | 0.43 | 0.26 | 0.73 | 0.002*** | |
| 2 | Sex | ||||||
| Female (ref) | 303 (57.39) | 48 (9.09) | |||||
| Male | 135 (25.57) | 42 (7.95) | 1.96 | 1.23 | 3.11 | 0.004*** | |
| 3 | Age | ||||||
| Youth (18–34) (ref) | 257 (50.57) | 59 (11.17) | |||||
| Adult (35–64) | 154 (29.17) | 29 (5.49) | 0.85 | 0.52 | 1.38 | 0.520 | |
| Elderly (≥ 65 years) | 17 (3.22) | 2 (0.38) | 0.53 | 0.11 | 2.36 | 0.408 | |
| 4 | Marital status | ||||||
| In a couple (ref) | 336 (63.64) | 69 (13.07) | |||||
| Not in a couple | 102 (19.32) | 21 (3.98) | 1.00 | 0.58 | 1.71 | 0.993 | |
| 5 | Household status | ||||||
| Member (ref) | 282 (53.41) | 45 (8.52) | |||||
| Head | 156 (29.55) | 45 (8.52) | 1.80 | 1.14 | 2.85 | 0.011** | |
| 6 | Household size | ||||||
| Small (0–3 persons) (ref) | 191 (36.17) | 29 (5.49) | |||||
| Medium (4–5 persons) | 91 (17.23) | 16 (3.03) | 1.15 | 0.59 | 2.23 | 0.663 | |
| Large (≥ 5 persons) | 156 (29.55) | 45 (8.52) | 1.89 | 1.13 | 3.17 | 0.014** | |
| 7 | Number of children | ||||||
| No | 60 (11.36) | 9 (1.70) | 0.7 | 0.33 | 1.46 | 0.345 | |
| Small (1–4 children) (ref) | 245 (46.40) | 53 (10.04) | |||||
| Large (≥ five children) | 133 (25.19) | 28 (5.30) | 1.03 | 0.63 | 1.69 | 0.889 | |
| 8 | Religion | ||||||
| Christians | 9 (1.70) | 2 (0.38) | 1.08 | 0.23 | 5.10 | 0.919 | |
| Muslims (ref) | 429 (81.25) | 88 (16.67) | |||||
| 10 | residence in the locality | ||||||
| Continuous (ref) | 257 (48.67) | 38 (7.20) | |||||
| No continuous | 181 (34.28) | 52 (9.85) | 1.94 | 1.22 | 3.07 | 0.005** | |
| 11 | Time in locality | ||||||
| Short (< 1 year) | 95 (17.99) | 20 (3.79) | 1.03 | 0.59 | 1.78 | 0.911 | |
| Medium (1–5 years) | 25 (4.73) | 10 (1.89) | 2.06 | 0.95 | 4.46 | 0.065* | |
| Long (> 5 years) (ref) | 318 (60.23) | 90 (17.05) | |||||
| 12 | Education | ||||||
| No (ref) | 314 (59.47) | 47 (8.90) | |||||
| Koranic | 108 (20.45) | 29 (5.49) | 1.79 | 1.07 | 2.99 | 0.025** | |
| Formal | 16 (3.03) | 14 (2.65) | 5.84 | 2.67 | 12.75 | 0.000*** | |
Source: Author (N = 528)
*** P < 0.01, **P < 0.05, *P < 0.1
Household size showed a gradient effect, with residents of large households (≥ 5 persons) having 89% higher odds of knowledge than those in small households (0–3 persons) (COR = 1.89, 95% CI: 1.13–3.17; P = 0.014). Educational attainment demonstrated an influential association, with formally educated participants having nearly six times higher odds of knowledge compared to those with no formal education (COR = 5.84, 95% CI: 2.67–12.75; P = 0.000), while those with Koranic education had 79% higher odds (COR = 1.79, 95% CI: 1.07–2.99; P = 0.025). Stability of residence also showed significance, with non-continuous residents having 94% higher odds of knowledge (COR = 1.94, 95% CI: 1.22–3.07; P = 0.005). The duration of residence showed a borderline significant association for medium-term residents (1–5 years) compared to long-term residents (> 5 years) (COR = 2.06, 95% CI: 0.95–4.46, p = 0.065). Variables that did not show significant associations included age, marital status, number of children, religion, and short-term residence.
Association between knowledge of lymphatic filariasis and some sociodemographic factors of the participants: multivariable logistic regression analysis
The results of the adjusted multivariable logistic regression model are presented in Table 5. This model identifies factors independently associated with knowledge of the disease while controlling for the effects of other variables and testing for potential interactions.
Table 5.
Multivariable logistic regression between Knowledge of Lymphatic filariasis and individual covariates
| Variables | Adjusted Odds Ratio (aOR) | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|
| Refugee camp | ||||
| Lolo (ref) | ||||
| Mbile | 0.17 | 0.05 | 0.54 | 0.003*** |
| Sex of refugee | ||||
| Female (ref) | ||||
| Male | 0.89 | 0.46 | 1.70 | 0.715 |
| Household status | ||||
| Member (ref) | ||||
| Head | 1.86 | 0.98 | 3.54 | 0.058* |
| Household size | ||||
| Small (0–3 persons) (ref) | ||||
| Medium size (4–5 persons) | 0.61 | 0.28 | 1.30 | 0.200 |
| Large (≥ 5 persons) | 1.10 | 0.60 | 2.02 | 0.766 |
| Locality area | ||||
| Continuous (ref) | ||||
| Not continuous | 1.54 | 0.83 | 2.86 | 0.169 |
| Time in the locality | ||||
| Short (< 1 year) | 2.34 | 0.88 | 6.21 | 0.089* |
| Medium (1–5 years) | 1.86 | 0.79 | 4.37 | 0.155 |
| Long (> 5 years) (ref) | ||||
| Education level | ||||
| No education (ref) | ||||
| Koranic education | 1.17 | 0.57 | 2.38 | 0.675 |
| Formal education | 4.16 | 1.60 | 10.83 | 0.004*** |
| Interaction | ||||
| Camp * Education 1 | ||||
| Lolo * Koranic (Ref) | ||||
| Mbile * Koranic | 4.01 | 1.03 | 15.51 | 0.044** |
| Camp education 2 | ||||
| Lolo*formal education (Ref) | ||||
| Mbile * formal education | 2.54 | 0.37 | 17.43 | 0.343 |
Source: Author (N = 528)
*** P < 0.01, **P < 0.05, *P < 0.1
The analysis reveals that the factor most strongly associated with knowledge of lymphatic filariasis is educational level. Individuals with a formal education had approximately a four-fold higher probability (aOR = 4.16, 95% CI: 1.60–10.83; P = 0.004) of being aware of the disease compared to those with no formal education.
Critically, the analysis detected a statistically significant interaction between the refugee camp and the type of education (aOR = 4.01, 95% CI: 1.03–15.51; P = 0.044). This interaction indicates that the effect of Koranic education is not uniform but is context dependent. While in the Lolo camp, Koranic education was not associated with better knowledge (aOR = 1.17, 95% CI: 0.57–2.38; P = 0.675), its effect is strong and significant in the Mbile camp. Thus, for a resident of Mbile, having a Koranic education is associated with an approximately 4.1 times higher probability of knowing about the disease compared to a resident with no formal education in the same camp.
Among other factors, household head status showed a trend toward association with higher knowledge (aOR = 1.86, 95% CI: 0.98–3.54, p = 0.058). Similarly, short duration of residence (< 1 year) was marginally associated with higher knowledge (aOR = 2.34, 95% CI: 0.88–6.21, p = 0.089). In contrast, sex, household size, and a residence duration of 1 to 5 years are not significantly associated with knowledge of filariasis in this model.
Discussion
This study provides a comprehensive assessment of LF knowledge and its sociodemographic determinants among refugee populations in Eastern Cameroon, revealing several critical findings with significant programmatic implications. The exceptionally low level of LF awareness (17.05%) identified in our study population represents a significant concern for efforts to eliminate LF in humanitarian settings. It underscores the urgent need for targeted health education interventions.
The profoundly low LF knowledge level observed in our study is substantially lower than rates reported in previous research among non-refugee populations in similar geographical regions. Studies conducted in Nigeria, Brazil, Malaysia, and Nepal have reported LF awareness ranging from 85% to 92% in endemic communities [12, 22, 24, 25]. This striking disparity highlights the extreme vulnerability of refugee populations and suggests that standard LF elimination approaches may be insufficient for reaching these marginalized communities. The low levels of knowledge likely reflect multiple intersecting vulnerabilities, such as limited access to formal education, language barriers, restricted access to mass media, and competing survival priorities that may deprioritize health education on chronic, neglected conditions like LF [22, 23]. Contextual factors, such as living conditions and access to services, may further explain this gap compared to more stable communities.
The adjusted multivariable analysis identified formal education as the strongest independent predictor of LF knowledge. Individuals with formal education had nearly five times the odds of being aware of LF compared to those with no formal education (aOR = 4.16; 95% CI: 1.60–10.83; P = 0.004). However, our analysis also revealed a significant interaction between the camp of residence and the type of education. While Koranic education was not associated with better knowledge in the Lolo camp (aOR = 1.17; 95% CI: 0.57–2.38; P = 0.675), it emerged as a strong and significant predictor in the Mbile camp (combined aOR = 4.1; 95% CI: 1.03–15.51; P = 0.044). This difference could be explained by disparities in the implementation or reception of religious education programs between the two camps, or by distinct sociocultural profiles within the refugee communities that moderate the influence of this type of education. This discovery underscores the importance of considering the local context when designing awareness interventions and suggests that health education programs integrated into Koranic educational structures could be particularly effective in specific environments, such as the Mbile camp.
It is noteworthy that while male gender was associated with higher knowledge in the bivariate analysis (COR = 1.96; CI : 1.23–3.1; P = 0.004), this association was no longer significant after adjustment for other factors, particularly education and household status (aOR = 0.89; CI : 0.46–1.70; P = 0.715). This suggests that the apparent gender effect is largely explained by these other sociodemographic determinants rather than being an independent predictor.
Other sociodemographic factors showed only weak or non-significant associations with LF knowledge in the adjusted model. Household head status (aOR = 1.86; 95% CI: 0.98–3.54; p = 0.058) and residence in a non-continuous locality area (aOR = 1.54; 95% CI: 0.83–2.86; p = 0.088) exhibited positive but non-significant trends. These findings should be interpreted with caution and do not provide sufficient evidence to support targeted programmatic recommendations. Larger studies are needed to determine whether these variables are truly associated with LF knowledge or if these trends reflect chance variation.
The finding that shorter duration of residence (< 1 year) suggested a possible association with higher odds of knowledge (aOR = 2.34; 95% CI: 0.88–6.21; p = 0.089) contrasts with some previous studies that reported positive associations between long-term residence and LF knowledge [22, 26]. However, given the wide confidence interval and borderline p-value, this association should be interpreted with caution and may reflect chance or uncontrolled confounding. It could also indicate that more recent arrivals originated from areas with better LF programming. This preliminary observation warrants further investigation but does not, on its own, support programmatic conclusions.
The overwhelming predominance of community health workers as information sources (88%) presents both a challenge and an opportunity for LF elimination programs. While this demonstrates the critical role these frontline workers play in disseminating health information in refugee settings, it also reveals an over-reliance on a single channel and the absence of a diversified communication strategy. This finding contrasts with studies in non-refugee settings where mass media typically serve as important information sources [27], highlighting the unique communication ecology of refugee camps where media access may be limited and interpersonal communication assumes greater importance.
The deficient knowledge regarding specific LF dimensions, particularly transmission modes (only 11.1% correct), prevention methods (only 4.4% correct), and clinical symptoms (only 32.2% correct) among those with basic awareness, suggests that merely hearing about a disease does not equate to functional knowledge necessary for prevention and appropriate health-seeking [28]. This superficial awareness without comprehension represents a significant programmatic challenge that must be addressed through more comprehensive, iterative, and participatory health education approaches rather than simple information dissemination.
Several limitations of our study should be acknowledged. The cross-sectional design precludes causal inference regarding the identified associations. Self-reported knowledge measures may be subject to social desirability and recall biases. The precision of our interaction estimates may be limited by small numbers in some educational subgroups. Furthermore, our knowledge assessment captured basic awareness but not functional understanding, which may underestimate the true knowledge gap. The specific context of Cameroonian refugee camps may limit the generalizability of findings to other humanitarian settings. Despite these limitations, our study offers valuable insights.
Future research should build upon the contextual insights of this study. Given our finding that the effect of education on knowledge is highly context-dependent (varying between camps and types of education), mixed-methods investigations are urgently needed to unpack the sociocultural and institutional mechanisms, such as the role of religious leaders or community structures, that explain this variation. Furthermore, to address the limitations of our cross-sectional design and basic knowledge assessment, longitudinal studies are required to establish causal pathways between education and sustained, functional health literacy, and to examine the direct link between knowledge and preventive behaviors. Finally, the critical reliance on community health workers identified here underscores the importance of implementation research focused on strategies to optimally train, support, and potentially diversify these frontline information channels within refugee camp ecosystems.
Conclusion
This study reveals an alarmingly low level of lymphatic filariasis (LF) knowledge among refugees in Cameroon, with formal education and the camp-specific context of Koranic education as the key determinants. These findings move beyond simple description to highlight that health literacy is not merely about exposure to information, but is deeply shaped by educational pathways that differ even between neighboring camps.
Consequently, programmatic recommendations must be highly specific to this context. LF elimination efforts in these camps should prioritize integrating health messaging into formal education programs while also exploring strategic partnerships with Koranic schools in the Mbile camp, where this channel proved particularly effective. Interventions must diversify beyond the heavy reliance on community health workers by developing camp-tailored, low-literacy communication materials that address critical gaps in understanding transmission and prevention.
The cross-sectional nature of this study limits causal inference, and the findings are specific to the socio-cultural context of these Central African refugee communities. Nevertheless, they provide a crucial evidence base for action. Future initiatives should test the feasibility and impact of these targeted strategies and investigate the sustainability of knowledge over time. Ultimately, overcoming the critical knowledge gap in these vulnerable populations requires moving from standardized approaches to flexible, evidence-informed programming that addresses the distinct educational and social realities of each refugee setting.
Acknowledgements
This work received financial support from the Coalition for Operational Research on Neglected Tropical Diseases (COR-NTD), which is funded by the Task Force for Global Health, primarily by the Bill & Melinda Gates Foundation, by the United States Agency for International Development through its Neglected Tropical Diseases Program, and with UK aid from the British Government. The African Research Network administered the Grant for Neglected Tropical Diseases (ARNTD).
Abbreviations
- AOR
Adjusted Odd Ratio
- LF
Lymphatic Filariasis
- NTD
Neglected Tropical Diseases
- CORs
Crude Odds Ratios
- SD
Standard Deviation
- UNHCR
United Nations High Commissioner for Refugees
- WHO
World Health Organization
Author contributions
Nama wrote the main manuscript text and Fokou analyzed data.
Data availability
The datasets generated and/or analyzed during the current study are not publicly available due to protecting the confidentiality and privacy of the refugee participants, but are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
Ethical oversight and approval for this non-interventional, community-based survey were provided by the Cameroonian Ministry of Public Health, the national authority responsible for public health research. The study received formal administrative authorization (Reference: N109/L/MINSANTE/SG/DSRPE/DSN/SSD) from the competent authorities of the Ndelele sub-division, endorsed by the district health officer. This authorization constituted the requisite ethical clearance for this research context. All procedures were conducted in accordance with core ethical principles, including those of the Declaration of Helsinki. Individual written informed consent was obtained from each participant following a detailed explanation of the study’s procedures and potential risks. It was expressly emphasized that participation was entirely voluntary and that participants could withdraw at any time without consequence.
Consent for publication
The confidentiality of all participant data was strictly guaranteed. The data presented in this article are anonymized and contain no personally identifiable information. Therefore, specific consent for publication was not required.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Christiana O, Bamidele A, Taiwo IE, Olubunmi OA. Parasitological and epidemiological studies of Wuchereria bancrofti in Imobi, Ijebu East, Local Government Area of Ogun State, South Western Nigeria. J Basic Appl Zool. 2021;82(1):49. 10.1186/s41936-021-00245-8. [Google Scholar]
- 2.Aye NN et al. Mapping and modelling the impact of mass drug administration on filariasis prevalence in Myanmar. Infect Dis Poverty. 2018; 7: 45-55. 10.1186/s40249-018-0420-9 [DOI] [PMC free article] [PubMed]
- 3.Kamgno J, Djeunga HN. Progress towards global elimination of lymphatic filariasis. Lancet Glob Health. 2020;8(9):e1108–9. 10.1016/S2214-109X(20)30323-5. [DOI] [PubMed] [Google Scholar]
- 4.Lawford H, et al. Anti-filarial antibodies are sensitive indicators of lymphatic filariasis transmission and enable identification of high-risk populations and hotspots. Int J Infect Dis. 2024;147:107194. 10.1016/j.ijid.2024.107194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Matapo BB, et al. Lymphatic filariasis elimination status: Wuchereria bancrofti infections in human populations after five effective rounds of mass drug administration in Zambia. Trop Med Infect Dis. 2023;8(7):333. 10.3390/tropicalmed8070333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.`Ofanoa R, Ofa T, Padmasiri EA, Kapa DR. Elimination of lymphatic filariasis as a public health problem from Tonga. Trop Med Health. 2019;47(1):43. 10.1186/s41182-019-0169-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Minsante, Rapport technique annuel des activités de lutte contre les Maladies Tropicales Négligées. 2017 (Additif: Plan d’action annuel 2018), Ministère de la Santé Publique, Yaounde, 2017. Consulté le: 2024. [En ligne]. Disponible sur: http://cdnss.minsante.cm/?q=fr/content/rapport-technique-annuel-des-activites-de-lutte-contre-les-maladies-tropicales-negligees
- 8.Organisation mondiale de la Santé. Intégrer les maladies tropicales négligées dans l’action pour la santé mondiale et le développement: quatrième rapport de l’OMS sur les maladies tropicales négligées. Genève: Organisation mondiale de la Santé, 2018. Consulté le: 2024. [En ligne]. Disponible sur: https://iris.who.int/handle/10665/260289
- 9.World Health Organization. Global report on neglected tropical diseases 2024: executive summary. World Health Organization; 2024. 10.2471/B09040.
- 10.Organisation Mondiale de la Sante. Stratégie régionale sur les maladies tropicales négligées dans la Région africaine 2014–2020: rapport du Secrétariat, in Stratégie régionale sur les maladies tropicales négligées dans la Région africaine 2014–2020: rapport du Secrétariat, 2013. Consulté le: 2024. [En ligne]. Disponible sur: https://iris.who.int/handle/10665/96712
- 11.Deribe K, Bakajika DK, Zoure H-G, Gyapong JO, Molyneux DH, Rebollo MP. African regional progress and status of the programme to eliminate lymphatic filariasis: 2000–2020. Int Health. 2020;13(Supplement_1):S22–7. 10.1093/inthealth/ihaa058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Amaechi EC, Ohaeri CC, Ukpai OM, Nwachukwu PC, Ukoha U. Lymphatic filariasis: knowledge, attitude and practices among inhabitants of an irrigation project community, North Central Nigeria. Asian Pac J Trop Dis. 2016;6(9):709–13. 10.1016/S2222-1808(16)61114-3. [Google Scholar]
- 13.Asiedu SO, Kwarteng A, Amewu EKA, Kini P, Aglomasa BC, Forkuor JB. Financial burden impact quality of life among lymphatic filariasis patients. BMC Public Health. 2021;21(1):174. 10.1186/s12889-021-10170-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Aagaard-Hansen J, Nombela N, Alvar J. Population movement: a key factor in the epidemiology of neglected tropical diseases: population movement and neglected tropical diseases. Trop Med Int Health. 2010;15(11):1281–8. 10.1111/j.1365-3156.2010.02629.x. [DOI] [PubMed] [Google Scholar]
- 15.Avokpaho EFGA, et al. Factors associated with soil-transmitted helminths infection in Benin: findings from the DeWorm3 study. PLoS Negl Trop Dis. 2021;15(8):e0009646. 10.1371/journal.pntd.0009646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bonniface M, Nambatya W, Rajab K. An evaluation of antibiotic prescribing practices in a rural refugee settlement district in Uganda. Antibiotics. 2021;10(2):172. 10.3390/antibiotics10020172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Dorkenoo MA, et al. Monitoring migrant groups as a post-validation surveillance approach to contain the potential reemergence of lymphatic filariasis in Togo. Parasites & Vectors. 2021;14(1):134. 10.1186/s13071-021-04644-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ramaiah KD. Population migration: implications for lymphatic filariasis elimination programmes. PLoS Negl Trop Dis. 2013;7(3):e2079. 10.1371/journal.pntd.0002079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Panigrahi SK, Gautam A. Sociocultural factors in filariasis in an endemic village. BMJ Case Rep. 2024;17(11):e255370. 10.1136/bcr-2023-255370. [Google Scholar]
- 20.UNHCR, Cameroun. Statistiques des personnes relevant de la cpmpétence du HCR, Haut-Commissariat pour les Réfugiés, Yaoundé, 2021.
- 21.Marcondes M et al. Knowledge and practices on lymphatic filariasis in patients with filarial morbidity in an endemic area in Northeastern Brazil, Front Trop Dis, 2024; 5: 1368589. 10.3389/fitd.2024.1368589
- 22.Adhikari R, Acharya D, Wagle A. Sociodemographic characteristics as predictors of knowledge regarding mode of transmission of Lymphatic Filariasis among population of Nepal. PLoS Glob Public Health. 2022;2(10):e0000082. 10.1371/journal.pgph.0000082. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Graves PM, Sheridan S, Fuimaono S, Lau CL. Demographic, socioeconomic and disease knowledge factors, but not population mobility, associated with lymphatic filariasis infection in adult workers in American Samoa in 2014. Parasit Vectors. 2020;13(1):125. 10.1186/s13071-020-3996-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cabral S et al. Knowledge, attitudes and perceptions regarding lymphatic filariasis: study on systematic noncompliance with mass drug administration, Rev Inst Med Trop São Paulo, 2017; 59, 10.1590/s1678-9946201759023 [DOI] [PMC free article] [PubMed]
- 25.Al-Abd NM, Nor ZM, Ahmed A, Al-Adhroey AH, Mansor M, Kassim M. Lymphatic filariasis in Peninsular Malaysia: a cross-sectional survey of the knowledge, attitudes, and practices of residents. Parasit Vectors. 2014;7(1):545. 10.1186/s13071-014-0545-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kwarteng A et al. oct., Knowledge and perceptions of lymphatic filariasis patients in selected hotspot endemic communities in southern Ghana, PLOS Glob Public Health, 2023; 3(10): e0002476, 10.1371/journal.pgph.0002476 [DOI] [PMC free article] [PubMed]
- 27.Tomasouw JFC, Titaley C, Malakauseya MLV, Kailola NE, Krentel A. Diverse information sources and the community’s high level of knowledge about Lymphatic Filariasis in Air Salobar and Waihaong, Ambon City, Indonesia. Natl J Community Med. 2024;15(01):11–8. 10.55489/njcm.150120243532. [Google Scholar]
- 28.Vanamail P, Gunasekaran S. Possible relationship among socio-economic determinants, knowledge and practices on lymphatic filariasis and implication for disease elimination in India. Int J Public Health. 2011;56(1):25–36. 10.1007/s00038-010-0159-y. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analyzed during the current study are not publicly available due to protecting the confidentiality and privacy of the refugee participants, but are available from the corresponding author on reasonable request.