Over half of births and newborn care occur in primary healthcare facilities in Nigeria, but information on activities of personnel working there is scarce.
To assess the knowledge and practices relating to neonatal jaundice (NNJ) among community health workers (CHWs) and community birth attendants (CBAs) in Nigeria.
We conducted a cross-sectional survey of all 227 CHWs and 193 registered CBAs in Ibadan, Nigeria.
Knowledge and practices regarding NNJ were measured using a pretested questionnaire. Knowledge and practices were assessed on a 33-point scale and a 13-point scale, respectively. Scores ≤ 17 and ≤ 9 was regarded as poor knowledge and as wrong practice, respectively.
Many (64.5%) of the respondents could not correctly describe examination for NNJ (CHWs: 49.4%; CBAs: 50.6%). Of the 200 (47.6%) who treated NNJ 3 months prior to the study, 62.5% (CHWs: 66.9% and CBAs: 53.7%) treated NNJ with orthodox drugs. Drugs prescribed included: antibiotics (93.3%), antimalarials (5.3%), multivitamins (28.0%), paracetamol (6.2%) and phenobarbitone (7.1%). Significantly more CHWs than CBAs practiced exposure to sunlight (33.1% versus 16.4%) and administration of glucose water (28.6% versus 14.9%), while 58.0% of all respondents referred cases to secondary health facilities. Overall, 80.2% had poor knowledge (CHWs: 78.9%; CBAs: 81.9%) and 46.4% engaged in wrong practices (CHWs: 57.3%; CBAs: 33.7%). CHWs were more likely to indulge in wrong practices than CBAs (OR = 2.22, 95% CI = 1.03, 4.79).
Primary Health Workers in Ibadan had poor knowledge and engaged in wrong practices about NNJ. The needs to organise regular training programmes were emphasised.
In sub-Saharan Africa, though the annual rate of reduction of under-five mortality was faster than the rest of the world, Africa continues to experience higher child morbidity and mortality rates as well as greater numbers of developmentally deprived children.
Jaundice may be clinically significant for many reasons. The high levels of unconjugated bilirubin may cause irreversible brain damage, referred to as kernicterus, and bilirubin toxicity (kernicterus) causes nerve deafness, choreoathetoid cerebral palsy and mental retardation.
According to the World Health Organization (WHO), achieving the MDG 4, that is reducing child mortality by two-thirds, will require universal coverage with key effective, affordable interventions including care for newborns.
Knowing that majority of newborns are cared for at home and at the primary healthcare facilities where community health workers (CHWs) are working underscores the need to find out what they know and current practices in order to proffer solutions to identified gaps. At the moment, there is dearth of state-wide data relating to practices of health workers on treatment of NNJ at the primary care level in Nigeria. This information is necessary for the purpose of formulating strategies and policies geared towards effective care in this respect. Often, the obstacle to change in practice may be a lack of knowledge of the benefits of more effective methods, or lack of knowledge of the problems and severity of associated effect the existing practice may be having. To address the United Nations MDG 4 on reducing childhood mortality, there is the need for better understanding of the levels of knowledge and roles played by CHWs who attend to a remarkable number of ill neonates in resource-limited countries like Nigeria. Therefore, this study was designed to assess the knowledge and practices relating to NNJ among CHWs and community birth attendants (CBAs) in Ibadan, Nigeria.
This research employed a cross-sectional study design. The primary health facilities and community health posts, which include the CBAs’ homes, were the main centres of this study. The study involved a one-time interaction with CHWs and CBAs while at their duty post. The study was carried out in Ibadan, located in the south west of Nigeria. Ibadan is the third largest metropolitan area in Nigeria, by population, after Lagos and Kano, with a population of over 3 million, and the largest metropolitan geographical area. The principal inhabitants of Ibadan city are the Yoruba-speaking people of south-western Nigeria. Healthcare delivery in Ibadan spans the three tiers of government, that is, the federal, the state and the local government, managing the tertiary, secondary and primary health facilities, respectively. These primary health facilities are located within the communities, at readily accessible locations in order to serve as the first ‘point-of-call’ for healthcare services. These facilities open every day and are manned by CHWs; they also have affiliated homes that are manned by trained CBAs. The primary health clinics (PHCs) operate using the standing order of the federal ministry of health while the CBAs are trained locally to take births and refer to the affiliated PHC which is usually the Comprehensive PHC of the local government area (LGA). On an average, most PHCs open during the day while the CBA homes are open to clients 24 hours of the day.
In all, there were 167 officially registered primary health facilities covered in this study. The target population for the study included CHWs (community health officers and community health extension workers) at the primary health centres and CBAs in the 11 local government areas that made up Ibadan. At the time of the study, there were a total of 433 eligible health workers in the 11 LGAs selected for this study. However, only those who were not on leave constituted the sampling frame.
All CHWs present at work and all available trained CBAs in the CBAs’ home were selected for the study. The investigator first visited the facility, explained to the head of the facility the study requirements, after which all CHWs, health extension workers and CBAs present at the facility at the time of visit were interviewed. Thus, CHWs that were not at their duty post at the time of recruitment and unregistered CBAs were excluded from the study.
Efforts were made to enrol all eligible health workers during the 8-month study period unless consent was declined. There were 433 health workers eligible to participate in the study, but complete responses were obtained from 420 participants.
The data for the study were collected using an interviewer-administered semi-structured questionnaire. The items in this questionnaire were adapted from the study by Ogunfowora and Daniel.
The data were entered into and analysed using SPSS 18.0 statistical software (SPSS Inc., Chicago, Illinois, USA). The main outcome variables were knowledge score and practice. The knowledge score for each respondent was determined by allotting a score of ‘1’ to correct response and ‘0’ (zero) to wrong response to questions on the definition of NNJ, colour of urine of neonates with NNJ, dangers signs, causes, treatments and perceived adverse events associated with the management of NNJ. Thus, maximum obtainable knowledge score was 33. A knowledge score of ≤ 17 was considered as poor knowledge while > 17 was categorised as good knowledge. Similarly, the practice score for each respondent was determined by allotting a score of ‘1’ to correct response and ‘0’ (zero) to wrong response to questions on ‘what primary health workers would offer for a case of NNJ?’ and whether or not they would refer cases. These gave a total of 13 maximum obtainable knowledge scores. A primary health worker with a practice score of ≤ 9 was regarded as engaging in wrong practice while a score of > 19 was considered correct practice. The association between practices (wrong or correct) and demographic variables were assessed using odds ratio (OR) and 95% confidence intervals. Logistic regression was carried out to identify variables that independently predicated wrong practices among the three (age, workers’ category and education) found to show significant association at bivariate level of analysis. The inferential statistics were considered statistically significant only if
The protocol for this study was reviewed and approved by the UI/UCH Institutional Review Committee (Approval Number: UI/EC/12/0007). Each respondent’s consent was obtained using a standardised informed consent form after provision of adequate, clear and complete information about what the study entails. Participation in the study was voluntary. Confidentiality was ensured by using a serial number on the information collected rather than a name. Only the researcher knew the identification, and this information was kept.
Characteristics of study participants: The gender distribution, mean age and duration of practice of the respondents are given in
Respondents’ age, gender and duration of practice.
Characteristics | All respondents ( |
CHWs ( |
CBAs ( |
|
---|---|---|---|---|
Male, |
23 (5.5) | 21 (9.3) | 2 (1.0) | 0.000 |
Female, |
397 (94.5) | 206 (90.7) | 191 (99.0) | |
Range | 22 – 76 | 22 – 60 | 25 – 76 | |
Mean | 45.45 ± 9.3 | 42.07 ± 7.6 | 49.41 ± 9.5 | 0.000 |
Range | 1–49 | 1–34 | 1–49 | |
Mean | 16.12 ± 9.1 | 16.38 ± 8.6 | 16.12 ± 9.6 | 0.526 |
CHW, community health workers; CBAs, community birth attendants.
Though all the 420 respondents had heard and seen cases of NNJ before the study visit, only 5.2% correctly defined it as ‘yellowish discolouration of the skin and other tissues of a newborn infant’. Also, only 21.9% of respondents reported that they had ever received training on NNJ management. Significantly fewer CBAs (
Perceived danger signs associated with neonatal jaundice.
Characteristics | Yes/No | All participants | CHWs | CBAs | ||||
---|---|---|---|---|---|---|---|---|
% | % | % | ||||||
Fever | No | 91 | 21.7 | 50 | 22.0 | 41 | 21.2 | 0.846 |
Yes | 329 | 78.3 | 177 | 78.0 | 152 | 78.8 | ||
Refusal to feed | No | 85 | 20.2 | 47 | 20.7 | 38 | 19.7 | 0.790 |
Yes | 335 | 79.8 | 180 | 79.3 | 155 | 80.3 | ||
High pitch cry | No | 205 | 48.8 | 116 | 51.1 | 89 | 46.1 | 0.300 |
Yes | 215 | 51.2 | 111 | 48.9 | 104 | 53.9 | ||
Arching of the back | No | 400 | 95.2 | 217 | 95.6 | 183 | 94.8 | 0.710 |
Yes | 20 | 4.8 | 10 | 4.4 | 10 | 5.2 | ||
Convulsion in neonate | No | 357 | 85.0 | 193 | 85.0 | 164 | 85.0 | 0.989 |
Yes | 63 | 15.0 | 34 | 15.0 | 29 | 15.0 | ||
‘Down turning of the eye’ | No | 383 | 91.2 | 205 | 90.3 | 178 | 92.2 | 0.489 |
Yes | 37 | 8.8 | 22 | 9.7 | 15 | 7.8 | ||
Severe hypotonic/hypertonia | No | 396 | 94.3 | 214 | 94.3 | 182 | 94.3 | 0.990 |
Yes | 24 | 5.7 | 13 | 5.7 | 11 | 5.7 | ||
Hard to wake/sleep | No | 333 | 79.3 | 177 | 78.0 | 156 | 80.8 | 0.472 |
Yes | 87 | 20.7 | 50 | 22.0 | 37 | 19.2 |
CHW, community health workers; CBAs, community birth attendants.
Perceived causes of neonatal jaundice among respondents.
Characteristics | Yes/No | All participants | CHWs | CBAs | ||||
---|---|---|---|---|---|---|---|---|
% | % | % | ||||||
Disparity between blood groups | No | 345 | 82.1 | 195 | 85.9 | 150 | 77.7 | 0.029 |
Yes | 75 | 17.9 | 32 | 14.1 | 43 | 22.3 | ||
Sepsis | No | 288 | 68.6 | 152 | 67.0 | 136 | 70.5 | 0.441 |
Yes | 132 | 31.4 | 75 | 33.0 | 57 | 29.5 | ||
Malaria | No | 79 | 18.8 | 35 | 15.4 | 44 | 22.8 | 0.054 |
Yes | 341 | 81.2 | 192 | 84.6 | 149 | 77.2 | ||
Mosquito bite | No | 362 | 86.2 | 198 | 87.2 | 164 | 85.0 | 0.505 |
Yes | 58 | 13.8 | 29 | 12.8 | 29 | 15.0 | ||
Breast milk or germs in the breast | No | 407 | 96.9 | 223 | 98.2 | 184 | 95.3 | 0.087 |
Yes | 13 | 3.1 | 4 | 1.8 | 9 | 4.7 | ||
Prematurity | No | 377 | 89.8 | 208 | 91.6 | 169 | 87.6 | 0.171 |
Yes | 43 | 10.2 | 19 | 8.4 | 24 | 12.4 | ||
G6PD defficiency | No | 397 | 94.5 | 213 | 93.8 | 184 | 95.3 | 0.500 |
Yes | 27 | 5.5 | 14 | 6.2 | 9 | 4.7 | ||
Haemolysis in newborn | No | 405 | 96.4 | 218 | 96.0 | 187 | 96.9 | 0.638 |
Yes | 15 | 3.6 | 9 | 4.0 | 6 | 3.1 |
G6PD, Glucose-6-Phosphate Dehydrogenase; CHW, community health workers; CBAs, community birth attendants.
Perceived effective treatment for neonatal jaundice stated by respondents.
Characteristics | Yes/No | All participants | CHWs | CBAs | ||||
---|---|---|---|---|---|---|---|---|
% | % | % | ||||||
Exposure to sunlight | No | 140 | 33.3 | 58 | 25.6 | 82 | 42.5 | 0.000 |
Yes | 280 | 66.7 | 169 | 74.4 | 111 | 57.5 | ||
Glucose water | No | 138 | 32.9 | 74 | 32.6 | 64 | 33.2 | 0.903 |
Yes | 282 | 67.1 | 153 | 67.4 | 129 | 66.8 | ||
Phototherapy | No | 348 | 82.9 | 181 | 79.7 | 167 | 86.5 | 0.066 |
Yes | 72 | 17.1 | 46 | 20.3 | 26 | 13.5 | ||
Exchange blood transfusion | No | 375 | 89.3 | 192 | 84.6 | 183 | 94.8 | 0.001 |
Yes | 45 | 10.7 | 35 | 15.4 | 10 | 5.2 | ||
Unripe paw-paw | No | 366 | 87.1 | 201 | 88.5 | 165 | 85.5 | 0.351 |
Yes | 54 | 12.9 | 26 | 11.5 | 28 | 14.5 |
CHW, community health workers; CBAs, community birth attendants.
Adverse events and morbidities relating to neonatal jaundice stated by respondents.
Characteristics | Yes/No | All participants | CHWs | CBAs | ||||
---|---|---|---|---|---|---|---|---|
% | % | % | ||||||
Death of neonate | No | 14 | 3.3 | 8 | 3.5 | 6 | 3.1 | 0.813 |
Yes | 406 | 96.9 | 219 | 96.5 | 187 | 96.9 | ||
Brain damage | No | 282 | 67.1 | 134 | 59.0 | 148 | 76.7 | 0.000 |
Yes | 138 | 32.9 | 93 | 41.0 | 45 | 23.3 | ||
Mental retardation | No | 332 | 79.0 | 175 | 77.1 | 15.7 | 81.3 | 0.288 |
Yes | 88 | 21.0 | 52 | 22.9 | 36 | 18.7 | ||
Physical handicap | No | 326 | 77.6 | 183 | 80.6 | 143 | 74.1 | 0.110 |
Yes | 94 | 22.4 | 44 | 19.4 | 50 | 25.9 | ||
Convulsions later in life | No | 358 | 85.2 | 193 | 85.9 | 163 | 84.5 | 0.677 |
Yes | 62 | 14.8 | 32 | 14.1 | 30 | 15.5 | ||
Abnormal behaviours | No | 379 | 90.2 | 202 | 89.0 | 177 | 91.7 | 0.349 |
Yes | 41 | 9.8 | 25 | 11.0 | 16 | 8.5 | ||
Ever had about a light meter | No | 407 | 96.9 | 220 | 96.9 | 187 | 96.9 | 0.988 |
Yes | 13 | 3.1 | 7 | 3.1 | 6 | 3.1 |
CHW, community health workers; CBAs, community birth attendants.
Assessing the knowledge of the respondents based on their responses to questions on the definition of NNJ, colour of urine of neonates with NNJ and items listed in
Only 35.5% of all respondents could correctly describe the examination of a neonate for NNJ; 96.7% of all participants will check the eyes of the neonate, 71.1% said they would examine the skin for jaundice, 52.1% of respondents would check the palm or foot when examining a neonate for jaundice and 96.7% acknowledge that checking the eyes for a yellowish colour is important as part of examination for NNJ.
Respondents’ responses to examination at presentation with neonatal jaundice (NNJ).
Characteristics | Yes/No | All participants | CHWs | CBAs | ||||
---|---|---|---|---|---|---|---|---|
% | % | % | ||||||
Eyes of a neonate | No | 14 | 3.3 | 4 | 1.8 | 10 | 5.2 | 0.052 |
Yes | 406 | 96.7 | 223 | 98.2 | 183 | 94.8 | ||
Looking at the skin | No | 119 | 28.3 | 67 | 29.5 | 52 | 26.9 | 0.560 |
Yes | 301 | 71.7 | 160 | 70.5 | 141 | 73.1 | ||
Looking at the palm, sole, or footsore | No | 201 | 47.9 | 98 | 43.2 | 103 | 53.4 | 0.037 |
Yes | 219 | 52.1 | 129 | 56.8 | 90 | 46.6 | ||
Colour of urine | No | 341 | 81.2 | 171 | 75.3 | 170 | 88.1 | 0.001 |
Yes | 79 | 18.8 | 56 | 24.7 | 23 | 11.9 | ||
Colour of stool | No | 405 | 96.4 | 218 | 96.0 | 187 | 96.9 | 0.638 |
Yes | 15 | 3.6 | 9 | 4.0 | 6 | 3.1 | ||
Follow up birth | No | 74 | 7.6 | 55 | 24.2 | 19 | 9.8 | 0.000 |
Yes | 346 | 82.4 | 172 | 75.8 | 174 | 90.2 | ||
Include NNJ in focused antenatal care | No | 130 | 31.0 | 65 | 28.6 | 65 | 33.7 | 0.265 |
Yes | 290 | 69.0 | 162 | 71.4 | 128 | 66.3 |
With respect to the treatment they would offer, a significantly higher proportion of CHWs than CBAs reportedly practice treatment of NNJ with drugs (50.2% versus 21.2%), exposure to sunlight (23.8% versus 9.8%) and glucose water (22.0% versus 8.8%);
The overall practice score ranged from 4 to 12 with a mean practice score of 9.5 ± 1.7. Overall, slightly over half of the respondents (53.6%) engaged in the right practices and significantly more CHWs (57.3%) than CBAs (33.7%) engaged in wrong practices (
Predictors of wrong practices relating to neonatal jaundice.
Variable | UOR | 95% CI of UOR | AOR | 95% CI of UOR | ||
---|---|---|---|---|---|---|
CHWs | 2.64 | 1.77, 3.92 | 0.000 | 2.23 | 1.03, 4.80 | 0.043 |
CBAs |
1 | - | - | - | - | - |
≤ 40 | 1.51 | 0.99, 2.27 | 0.051 | 1.18 | 0.76, 1.82 | 0.464 |
> 40 |
1 | - | - | - | - | - |
≤ Secondary |
1 | - | - | 1 | - | - |
Tertiary education | 2.43 | 1.61, 3.66 | 0.000 | 1.18 | 0.54, 2.58 | 0.685 |
, reference category; CI, confidence interval; UOR, unadjusted odds ratio; AOR, adjusted odds ratio.
Three main conclusions may be made from the findings in this study. Firstly, the level of awareness about NNJ among primary health care workers in Ibadan was high, but overall assessment of level of knowledge about the disease was below average. Secondly, primary health care workers reported ominously incorrect causes of and treatments for NNJ. Thirdly, unexpected wrong practices relating to NNJ was more prevalent among CHWs than birth attendants, and the higher level of education among CHWs did not impact positively on their practice.
Though all the primary health workers who participated in this study claimed they were aware of NNJ, only 5.2% of them could accurately describe the condition. This finding corroborates findings from a previous study which revealed that primary health workers’ awareness of NNJ may not necessarily translate to good practices.
The overall assessment of the primary health workers’ level of knowledge of NNJ relating to cause, presentations and treatment approaches showed that approximately four out of five (80.2%) of respondents had poor knowledge. This finding suggests that most of the primary health workers were not adequately informed about NNJ. This was corroborated by the fact that only 21.9% of respondents reported that they had training on NNJ management. The finding of low knowledge among the primary health workers in the study area is at variance with an earlier report by Ogunfowora and Daniels
The data from this study suggest that knowledge was not influenced by primary health workers’ level of education. In all, the observed lack of difference in the knowledge of CHWs and CBAs raises questions on the quality of the training curriculum for CHWs and perhaps implies a need for review. Consistent with the finding in this study, previous studies also found that there was no significant association between age and knowledge about NNJ
About half (53.1%) of the respondents reported that orthodox drugs are effective in the treatment of NNJ. However, responses provided on the names of such useful drugs indicated a gross lack of knowledge on the use of medication for treatment. Whereas many respondents mentioned different drugs as being effective, only a few (7.1%) mentioned phenobarbitone. For reasons unknown, most (93.3%) respondents mentioned antibiotics as being effective; an action that amounts to abuse of drugs when the cause of NNJ is not sepsis. This finding is consistent with that of a previous study in which the authors considered it a dangerous trend as reliance on unproven medication in the management of NNJ is very hazardous.
Another important finding from this study is that some myths relating to treatment modalities for NNJ exist among health workers. For instance, the primary health workers mentioned the use of ‘glucose water’, ‘exposure to sunlight’ and ‘unripe paw-paw’ for treatment of NNJ. Only a few of the respondents mentioned phototherapy and exchange blood transfusion, which are standard treatments for NNJ. Interestingly, more CHWs mentioned that ‘exposure to sunlight’ is an effective treatment of jaundice in neonates. Similarly, more CHWs than CBAs reported exchange transfusion as an effective treatment for NNJ. This is consistent with an earlier finding which revealed that 60.4% of respondents thought that ‘exposure of neonate to sunlight’ is an effective treatment.
Notably, 46.4% of the primary health workers in this study engaged in wrong practice. This finding is similar to that of an earlier study which revealed that 56% of the study respondents, had right practices.
As a whole, this study has generated data that provide insight into what is currently being done, why and how newborns with NNJ are being managed by CHWs and CBAs. With these data, interventions that would enhance the knowledge, attitude and practice of health workers regarding NNJ in the light of the local circumstances can be proffered. Furthermore, these data provide useful background for planning training activities that better suit the CHWs and CBAs practicing in Ibadan, Nigeria. Accessible literature suggests that this study is the first to be conducted on the knowledge and practices relating to NNJ among CHWs and CBAs in Ibadan, Nigeria. Also, the fact that an attempt was made recruit all eligible primary health workers on duty during the study period lends credence to the generalisability of our findings. In addition, the number of participants in the study is far above the number recruited in a similar study by Ogunfowora and Daniels.
However, a few factors may limit the generalisability of our findings. Firstly, the extent to which health workers responded accurately to the items in the questionnaire and whether their responses truly reflect the actual level of knowledge and practices relating to NNJ could not be verified. Nonetheless, the findings are correct within the limit of the validity and reliability of the questionnaire. It is important to note that the questionnaire was designed with items drawn from a previous study and pretested for validity in a population relatively similar to the study population. In addition, the questionnaire used in this study was translated into the local language (Yoruba) for the purpose of enhancing the understanding of respondents. These should expectedly improve the quality of the data. Secondly, the similarities or otherwise of the health workers who did not participate in the study remains unknown. The researchers made effort to capture all primary health workers at their duty post during the study period.
The findings of this study suggest that there is a need to intensify efforts at improving the knowledge of NNJ among primary health workers in Ibadan through in-service training on a scheduled basis in order to address specific needs such as referral and correct treatments for NNJ. This training should be all-inclusive and should involve the CBAs using teaching and learning methods that are tailored to their level of education. Also, the LGAs health department should scale up efforts at monitoring the activities of CBAs to protect the lives of their clients who invariably ‘must’ patronise them. Synergy between the birth homes and the primary health centres is also vital to reducing the impact of NNJ on child health and well-being.
The authors would like to express their gratitude to colleagues in the Institute of Child Health, College of Medicine, University of Ibadan, for their critiques at various stages of the study and drafting of the manuscript. We further acknowledge the invaluable support rendered by the research assistants. Special thanks go to the Medical Officers of Health in charge of the
The authors declare that they have no financial or personal relationships which may have inappropriately influenced them in writing this article.
A.E.O. was the project leader responsible for the conceptualisation and design of the study. He also analysed and wrote the first draft of the manuscript. A.O.O. contributed to the planning of the study, supervised data collection and analysis, and reviewed the manuscript.