Remember me
ORIGINAL ARTICLE
Year : 2022 | Volume
: 29
| Issue : 3 | Page : 198-205
Caregivers' perception and determinants of delayed presentation of children with severe malaria in an emergency room in Benin City, Nigeria
Moses Temidayo Abiodun1, Oluwatosin Ruth Ilori2
1 Department of Child Health, University of Benin Teaching Hospital, and School of Medicine, University of Benin, Benin City, Nigeria
2 Department of Community Medicine, LAUTECH, Ogbomoso, Oyo State, Nigeria
Correspondence Address:
Moses Temidayo Abiodun
Department of Child Health, University of Benin Teaching Hospital, Benin City
Nigeria
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/npmj.npmj_80_22
Introduction: Severe malaria is a leading cause of mortality due to late presentation to health facilities. Hence, there is a need to identify and mitigate factors promoting delayed presentation with severe malaria. Objective: This study aimed to evaluate determinants of delayed presentation of children with severe malaria in a tertiary referral hospital. Methods: This study adopted a descriptive, cross-sectional design. The participants were children with a diagnosis of severe malaria, based on WHO diagnostic criteria. Delayed presentation was defined as presentation at the referral centre at >3 days of illness. Inferential analyses were done to identify factors associated with delayed presentation. P < 0.05 was considered statistically significant. Results: A total of 126 children with severe malaria participated in the study; their mean (standard deviation) age was 4.2 (5.3) years. The prevalence of delayed presentation in this study is 37.3%. Socio-economic class (P = 0.003); marital status (P = 0.015) and the number of health facilities visited before admission in the referral centre (P = 0.008) were significantly associated with delayed presentation. Children from upper socio-economic class were thrice more likely to present late, compared to those from lower social class (odds ratio [OR] = 3.728, 95% confidence interval [CI]: 1.694–8.208; P = 0.001). Likewise, the Yorubas were more delayed than the Binis (OR = 0.408, 95% CI: 0.180–0.928; P = 0.033). There was a negative correlation between caregivers' perception of treatment (r = −0.113, P = 0.21) of convulsion in severe malaria and timing of presentation. Conclusions: Delayed presentation is common with multifactorial determinants in the setting. Health education of caregivers on the consequences of delayed presentation in severe malaria is desirable.
Keywords: Children, delayed presentation, determinants, emergency room, severe malaria
Severe malaria is acute malaria with signs of organ dysfunction and/or hyper-parasitaemia. Under-five children and pregnant women are at increased risk for severe malaria in endemic regions while others acquire partial immunity following repeated infections reducing their susceptibility to complicated or severe malaria.[1],[2],[3] Severe malaria is a leading cause of morbidity and mortality in sub-Saharan Africa, and Nigeria in particular. Severe malaria accounted for 25% of infant mortality and 30% of under-5 mortality.[1],[4] Complications and poor outcomes of malaria are often due to late presentation to health facilities because delayed treatment of malaria has been shown to result in severe forms of malaria with attendant morbidities.[5] However, early presentation will significantly reduce blood transfusions, prolonged hospital stay and death in children with severe malaria.[3],[6]
Globally, <50% of severe malaria cases were able to reach health facilities despite the high case-fatality rate of severe malaria at home. About 90% of the world's severe and fatal malaria is estimated to affect young children in sub-Saharan Africa.[2] This is so because severe disease with rapid progression to death occurs in young children without acquired immunity, highlighting the danger of delayed treatment which increases the likelihood of progression to severe stages and mortality in all settings. In eastern India between 1995 and 2001, 526 patients aged >12 years with cerebral malaria had 23% mortality. Of the fatal cases, 107 (61%) died within the first 24 h of hospitalisation, indicative of late presentation.[7] Likewise, researchers in Zimbabwe found that independent risk factors for severe malaria were distance >10 km to the nearest health facility, duration of symptoms before seeking medical care >2 days and the presence of comorbidities.[8] Undernourishment and female gender were associated with mortality in the study.[8] Moreover, in a recent meta-analysis of 13 studies, the odds of complications were significantly higher in children with delayed presentation (P = 0.009). Furthermore, delayed presentation was a strong predictor of presenting with severe malarial anaemia in children in the multi-centre study (P < 0.001), with nearly a half of such anaemia cases being potentially avoidable with prompt treatment.[9]
A study in four rural communities in south-eastern Nigeria showed that 22% of participants sought treatment within 24 h for their children with malaria but 51.5% delayed treatment partly due to financial constraints.[10] In a cross-sectional study in Eastern Nigeria in 2020, Udujih et al.[11] found that over one-third of their participants sought treatment in pharmacy stores while 27.1% were self-medicating. Some researchers in south-west Nigeria have reported high cost, poor access to health facilities and preference for herbal concoctions as reasons for late presentation with severe malaria.[12] A prior study amongst caregivers of under-five children in Edo State showed also that cost and long waiting time were major reasons for not seeking care in health facilities.[13] This is consistent with a report by Nwaneri and Sadoh[14] in 2020 that less than a third (29.0%) of caregivers had good health-seeking behaviour for their under-five children with severe malaria in our study setting.
Hence, in order to reduce childhood deaths from severe malaria, there is a need to identify specific factors influencing delayed presentation to health facilities. This is a step towards achieving malaria-related targets in the Sustainable Development Goal. We hypothesise that clinical-demographic factors and caregivers' perception of severe malaria symptoms will influence the timing of presentation of the children. Therefore this study evaluates determinants of delayed presentation of children with severe malaria in our emergency unit.
MethodsStudy area
This study was carried out at the Children Emergency Room (CHER) of the University of Benin Teaching Hospital in southern Nigeria from July 2021 to January 2022. It opens 24 hours every day, including weekends and public holidays. About 30–40 children are seen daily in CHER with an average daily admission rate of 8, and severe malaria is a leading indication for admission in the unit.[15]
Study design
This study adopted a descriptive, cross-sectional design.
Study population
This study enrolled children aged between 3 months and 18 years admitted into the CHERs with severe malaria based on the inclusion criterion.
Inclusion criteria
All children aged between 3 months and 18 years with a diagnosis of severe malaria and whose parents consent to the study were included in the study. Severe malaria is defined based on the presence of its clinical/laboratory diagnostic criteria.[2]
Exclusion criteria
Children with malaria who did not meet the standard diagnostic criteria for severe malaria and also children aged <3 months were excluded from the study (being partly protected from severe malaria by subsisting foetal haemoglobin and transplacental antibodies).
Sample size determination
The minimum sample size was determined using the formula for cross-sectional study:[16]
where
Z1-α = normal standard deviation for confidence level of 95% = 1.96
P = Prevalence of delayed malaria treatment-seeking time (92.4% amongst parents with primary education in southeastern Nigeria reported by Chukwuocha et al.[10]).
d = margin of error to be tolerated (5% or 0.05).
Therefore, substituting the values,
A non-response rate of 10% was considered. Hence,
N = 114+ (114 × 0.10)
=114 + 12 = 126
A total of 126 children were recruited during the study period.
Data collection
This was a total population study of all eligible children (and their caregivers) recruited consecutively during the study period; they were purposively selected. Delayed presentation (as the outcome variable of this study) was defined as presentation at the referral centre at >3 days of illness.[10] Data were collected using a semi-structured questionnaire comprising: Baseline features and socio-demographics, caregivers' perception of severe malaria symptoms, clinical features and investigation results (to identify clinical-laboratory criteria of severe malaria), diagnosis and outcome. Caregivers' perception was rated on a 4-point Likert scale with a reliability rating Cronbach's alpha 0.60. Scoring ranged from 1 (strongly disagree) to 4 (strongly agree). Perception score for each item on the Likert scale was derived from the mean score of the participants' responses. A mean perception score of 2.5 and above was rated as adequate. The questionnaire was administered to the respondents by the researcher and a trained research assistant. Participants' socio-economic status were classified based on their parental educational levels and occupations.[17]
Malaria parasite tests
Malaria rapid diagnostic test was done using standard procedures. Malaria parasite microscopy was done at the main laboratory.
Statistical analysis
The data were analysed using the IBM Statistical Package for the Social Sciences (SPSS) version 26.0 for Windows. Armonk, NY, USA. Frequencies and percentages were calculated for the socio-demographic features and other categorical variables. Mean scores were computed for the variables assessing caregivers' perception of severe malaria morbidities on each sub-scale (reverse scoring applied where appropriate). Adequate perception was defined with a cut-off point of 2.5. Chi-square was done to detect significant difference between proportions of early and late presenters. Multivariate logistic regression was done to identify variables that independently predict delayed presentation. Pearson's correlation analysis was done to determine the relationship between caregivers' perception of severe malaria morbidities and duration of illness before presentation in the emergency unit. P < 0.05 was considered statistically significant.
Ethical consideration
Ethical clearance was obtained from the Research Ethics Committee of the College Medicine, University of Benin, and the participating institutions (CMS/REC/01/VOL. 2/225).
ResultsBaseline characteristics of the participants
A total of 126 children/caregivers pairs participated in the study. The mean (standard deviation [SD]) age of these children with severe malaria was 4.2 (5.3) years, mean weight (SD) 17.0 (±14.6) kg, male:female ratio 1:1. One hundred and twenty-two participants (96.8%) were Christians (including 88 [69.8%] Pentecostals) while 4 (3.2%) were Muslims. They were mainly from upper and middle socio-economic classes, 51.6% and 40.5% respectively. Sixty-nine (54.8%) participants were of Bini ethnicity; 119 (94.4%) were from a monogamous family setting; 20.6% of them had an extended family member lining with them. The mean (SD) family size was 4.7 (1.5) and 27 (21.4%) of participants had a history of sibling loss. Over two-thirds of (68.35) of participants had an insecticide-treated net at home, but only 39 (31.0%) of them had slept under net in the preceding 3 weeks.
One hundred and fourteen (90.5%) participants lived in Benin City while the rest of them lived in sub-urban communities and villages. Twenty-eight (22.2%) of the participants have not been admitted at any health facility before presentation, 69 (54.8%) have been treated in one health facility during this illness while the rest (23.0%) have been treated in 2 or 3 health facilities before presentation at the referral centre. On the day of admission at the referral centre, a majority of the participants (56.3%) presented directly from home, 21.4% from other public hospitals and 20.6% from private clinics. The means transit time of the children to the referral centre was 1.3 (1.1) hours. Further details of the baseline characteristics of the participants are shown [Table 1].
Caregivers' perceived causes of severe malaria symptoms
[Table 2] shows the perceived causes of severe malarial symptoms by their caregivers. Nearly all of them perceived fever and malaria parasite as causes of convulsion while 32 (25.4%) of them said convulsion is caused by evil spirit. Furthermore, most of the caregivers rightly identified common causes of coma and severe pallor but 27.8% and 20.6% respectively said they can be caused by spleen [Table 2].
Table 2: Caregivers' perceived causes of severe malaria morbidities (n=126)Their weighted mean scores for perceived causes of convulsion, coma and severe pallor were adequate 3.21 (0.38), 2.78 (0.40) and 3.19 (0.47) respectively. Overall, their grand mean score for perception of causes of severe malaria symptoms was adequate, 3.06 (0.24).
Caregivers' perceived treatment of severe malaria symptoms
[Table 3] shows the caregivers' perceived treatments of severe malaria symptoms. Nine out of every 10 of the caregivers identified diazepam as a treatment of convulsion and 25.4% agreed with oxygen therapy during convulsion. Harmful practices including scarification (42.1%), fire (19.1%) and cow urine (14.3%) were acceptable to some caregivers. The details of their perceived treatment of severe pallor are also shown in [Table 3].
Table 3: Caregivers' perceived treatments of severe malaria morbidities (n=126)One hundred and three (81.7%) of respondents had adequate perception of the treatments of convulsion and 75 (59.5%) adequately perceived severe anaemia treatments and 10 (7.9%) of the participants agreed to concurrent use of traditional and modern drugs. Overall, participants' grand means score for perceived treatment options was adequate, 2.69 (0.11).
Prevalence of delayed presentation amongst the participants
The prevalence of delayed presentation in this study is 37.3%. The mean (SD) duration of illness of the participants was 3.73 (±3.15) days; 79 (62.7%) of them were ill for ≤3 days while 47 (37.3%) delayed >3 days before presentation in the referral centre for admission. Their leading presenting complaints at presentation were fever (53.2%), cough (38.9%) and convulsion (7.9%).
Determinants of delayed presentation with severe malaria
[Table 4] shows the socio-demographic determinants of delayed presentation amongst the participants. Socio-economic class was significantly associated with delayed presentation (χ2 = 11.456, P = 0.003); marital status (cohabiting and single parenthood) was also associated with a significant increase in delayed presentation (χ2 = 8.466, P = 0.015). Likewise, the number of health facilities visited before admission in the referral centre was significantly associated with delayed presentation (χ2 = 11.899; P = 0.008).
Table 4: Socio-demographic determinants of delayed presentation amongst the participantsParticipants with a prior history of sibling loss show a tendency towards delayed presentation but this did not attain statistical significance (X2 = 3.111, P = 0.078). Gender, tribe, family size and presence of an extended family member in the household were not associated with delayed presentation of these children with severe malaria (P > 0.05). Other socio-demographic variables did not show a significant association with delayed presentation [Table 4].
Furthermore, there was no clinical variable that showed a significant association with delayed presentation in this study. The place of treatment (χ2 = 2.505; P = 0.644), complaints before admission (χ2 = 0.038; P = 0.981) and high temperature (38.5°C) at presentation (χ2 = 44.75; P = 0.067) were not significantly associated with late presentation at the referral centre. Likewise, the type of pre-hospital care received (scarification, herbal concoction, ACT antimalarial, etc.) and severity criteria of malaria did not affect the timing of presentation of the participants (P > 0.05).
Perception of severe malaria symptoms and delayed presentation
The relationship between perception of severe malaria symptoms and delayed presentation was not statistically significant, adequate perception score (≥2.5) on the cause or treatment of convulsion, coma and anaemia did not result to early presentation in the study setting [P > 0.05; [Table 5]]. There was a negative correlation between caregivers' scores on perceived causes (r = −0.042, P = 0.64) and treatment (r = −0.113, P = 0.21) of convulsion as well as perceived therapies of severe pallor (r =−0.12, P = 0.17); however, this relationship was not statistically significant [Figure 1]. Likewise, caregivers' perception of other severe malaria symptoms did not significantly influence timing of presentation at the referral centre.
Table 5: Comparison between perception of severe malaria symptoms and presentation
Figure 1: Correlation of weighted mean score of perceived treatment of severe pallor with duration of illnessPredictors of delayed presentation amongst the participants
On multiple logistic regression analysis, social class and ethnicity were the only demographic variables that independently predict delayed presentation amongst the participants: Children from upper socio-economic class were thrice more likely to present late in the emergency unit, compared to those from lower social class (odds ratio [OR] = 3.728, 95% confidence interval [CI]: 1.694–8.208; P = 0.001). Likewise, the Yorubas were more delayed compared to the Binis (OR = 0.408, 95% CI: 0.180–0.928; P = 0.033) Parental marital status and other baseline variables were not independent predictors of delayed presentation in this study [P > 0.05; [Table 6]].
Table 6: Independent predictors of delayed presentation amongst the participants Discussion This study shows that delayed presentation of children with severe malaria is common (37.3%) in the referral centre, consistent with earlier reports of late presentation of children with severe malaria and poor health-seeking behaviours of their caregivers.[9],[18] Chukwuocha et al.[10] in eastern Nigeria reported the prevalence of delayed presentation as 22% amongst children with malaria, with 92.4% of their participants in lower social class presenting after 3 days of illness. This delayed presentation of critically ill children to referral centres has contributed significantly to the high mortality associated with childhood severe malaria and the poor child health indices in the setting.[6],[10] The transit time (1.3 ± 1.1 h) to the referral centre was significantly higher than the expected time for a life-threatening emergency such as severe malaria and recommended prompt access to health facilities during critical illnesses.[8],[10],[19]
The social-demographic determinants associated with delayed presentation in this study are social class, ethnicity and parental marital status. Interestingly, participants in high social class were more likely to present late in this study compared to those in low socio-economic class, possibly arising from their preference for treatment in private health facilities before presentation for admission at the referral centre. A recent research in the setting shows that patients arriving the CHER from a private clinic are more critically-ill than those coming directly from home, perhaps due to transit delays.[20] The referral threshold of some private clinics in the setting may be sub-optimal; this highlights the need for feedbacks on cases received at tertiary centres.[21],[22] Patients in low social class may prefer to present directly to public hospitals, hoping to receive free immediate emergency care. Nonetheless, a poor health-seeking behaviour for childhood severe malaria has been previously noted in the study setting, potentiating poor outcomes.[14] Furthermore, dysfunctional parental marital status predisposed to delayed presentation, possibly related to financial hardship in the affected households.
Furthermore, Yoruba ethnicity is significantly associated with delayed presentation in this study. The Yorubas are a minority group in the study locale but this cannot be possibly due to tribal discrimination. The clinical services provided at the study centre are generally based on a fee-for-service model which preclude the possibility of tribal sentiments that can occur during enrolments of non-indigenes into free health programs.[23] Cultural and belief systems may contribute to the preponderance of delayed presentation amongst minority groups.[12],[23] Although this study is not designed primarily for tribal comparison of presentation timing in severe malaria, frontline clinicians should pay cognizance to the possibility of delayed presentation of children from minority groups in various settings, and community-based interventions should focus on such children.
There is a need to identify additional factors that may contribute to delayed presentation in the setting. Clinical factors such as the symptoms and signs manifested by the children with severe malaria did not significantly predict delayed presentation in this study. This can be due to caregivers' inability to recognise or interpret these features, as being potentially life-threatening to the children.[13],[24] Hence, they were not frightened by such symptoms to present early at the referral centre. Nevertheless, the causes of delayed presentation are multi-factorial as highlighted in the three delays model.[25],[26] It remains pertinent to train caregivers/parents to recognise emergent signs in acutely-ill children and present early to health facilities to avert a poor outcome. Participants with a prior history of sibling loss show a tendency towards delayed presentation; this points to an 'ignorance-disease-death cycle' in communities.[12],[14]
It is noteworthy that the overall perception of causes and treatment of severe malaria morbidities by caregivers were adequate but there were still some misconceptions about the efficacy of 'scent leaf' and 'malt drink' in the treatment of malarial neuro-morbidity and anaemia respectively. This highlights the need for sustained health education at the community level to dissuade against potentially harmful therapies in malaria.[12],[27] The negative correlation between caregivers' perception of severe malaria symptoms and duration of illness suggests that this is a potential area of intervention to reduce delayed presentation in our setting.[28] Every time gained towards early presentation in childhood emergencies can avert dire consequences, such as progression to irreversible shock in childhood severe malaria with circulatory failure.[20]
The strength of this study includes the recruitment of patients that were actually suffering from severe malaria in an emergency unit, documenting their real perception of the symptoms and actions regarding treatments during the illness. Sometimes, there can be a discrepancy between what respondents claim they will do in event of an illness and what they actually do during the illness but this research overcomes such a drawback. A limitation of this study is that it did not assess participants' baseline perception of uncomplicated malaria which may have implications on their malaria prevention practices.[29] Furthermore, the relatively small sample size hinders sub-analysis for perception of less common severe malaria phenotypes amongst the participants. In addition, a purposive sampling technique was used in this study.
ConclusionsDelayed presentation in childhood severe malaria is rampant in the study setting and its determinants are multifactorial. As caregivers' perception of severe malaria morbidities increases, duration of illness before presentation in the referral centre decreases. There is a need to health educate caregivers on the causes, prevention and consequences of delayed presentation in childhood severe malaria. Furthermore, an increased access to emergency medical services/transport to shorten the transit time to referral centres in cases of severe malaria is desirable.
Availability of data and materials
The study data are available on request to the corresponding author.
Prior presentation
This research is a part of a Masters in Public Health project presented at the Department of Community Medicine, LAUTECH, March 2022.
Acknowledgements
The authors thank Atunde Joseph Adekunle (clinical data analyst).
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
Comments (0)