Keywords
Universal Health Coverage, pandemic preparedness, maternal emergencies, health system resilience, SDGs, SORT IT, Operational Research, access to care
-
Views
-
Downloads
How to cite this article
Kamara J, Moses F, Thekkur P et al. The effect of the COVID-19 pandemic on emergency maternal and under-five referrals in Sierra Leone: A cohort study [version 1; peer review: peer review discontinued]. F1000Research 2024, 13:703 (https://doi.org/10.12688/f1000research.149455.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
Export Citation
Sciwheel
EndNote
Ref. Manager
Bibtex
ProCite
Sente
▬
✚
Research Article
The effect of the COVID-19 pandemic on emergency maternal and under-five referrals in Sierra Leone: A cohort study
[version 1; peer review: peer review discontinued]
PUBLISHED 27 Jun 2024
OPEN PEER REVIEW
PEER REVIEW DISCONTINUED
This article is included in the TDR gateway.
This article is included in the TDR: Ebola and Emerging Infections in West and Central Africa collection.
Abstract
Corresponding author:
Jonta Kamara
Competing interests:
No competing interests were disclosed.
Grant information:
This SORT IT Programme was funded by the Special Programme for Research and Training in Tropical Diseases (TDR), Geneva, Switzerland (Grant Number HQTDR 2422924-4.1-72863. The APC was also funded by TDR. TDR is able to conduct its work thanks to the commitment and support from a variety of funders. A full list of TDR donors is available at: https://tdr.who.int/about-us/our-donors
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Copyright:
© 2024 World Health Organisation.
This is an open access article distributed under the terms of the Creative Commons Attribution IGO License, which permits copying, adaptation and distribution in any medium or format for any purpose, provided the original work is properly cited, a link is provided to the license, and any changes made are indicated. Any such copying, adaptation and distribution must not in any way suggest that World Health Organisation endorses you or your use.
How to cite: Kamara J, Moses F, Thekkur P et al. The effect of the COVID-19 pandemic on emergency maternal and under-five referrals in Sierra Leone: A cohort study [version 1; peer review: peer review discontinued]. F1000Research 2024, 13:703 (https://doi.org/10.12688/f1000research.149455.1)
First published: 27 Jun 2024, 13:703 (https://doi.org/10.12688/f1000research.149455.1)
Latest published: 27 Jun 2024, 13:703 (https://doi.org/10.12688/f1000research.149455.1)
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Introduction
Globally, the COVID-19 pandemic disrupted essential health services in about 90% of countries. The adverse impact of the COVID-19 pandemic on essential health services was disproportionately higher in low-and-middle-income countries (LMICs), which have limited health system resilience.1 The World Health Organization’s (WHO) pulse survey (2022) reported that ambulance services and 24-hour emergency room services were disrupted in 37% and 33% of countries respectively.
A systematic review (2022) reported a significant increase in stillbirths and maternal deaths in LMICs during the COVID-19 pandemic, hindering the progress made towards achieving the United Nations Sustainable Development Goal 3.2,3 The delay in seeking care due to anxiety or fear of contracting COVID-19 increased the risk of emergency and intensive care unit admissions.4–7 Furthermore, the difficulties in accessing emergency services due to inadequate transportation, movement restrictions and closure or diversion of health facilities for COVID-19 care contributed to poor outcomes among maternal and under-five children.3,8–10
Sierra Leone is a country with a high maternal mortality ratio of 443 deaths per 100,000 live births in 2020 and one of the top five countries with the highest under-five mortality rates. Disruption in the health system during the decade-long civil war (1991 to 2002) and the West African Ebola outbreak (2014 to 2016) contributed to high maternal and under-five mortality.11 Acknowledging this, studies on the impact of the Ebola outbreak recommended building health system resilience and preparedness to counter the adverse impacts of future outbreaks on health services for maternal and under-five children.12,13
In Sierra Leone, managing maternal and under-five emergencies at peripheral health units (PHU) is challenging due to shortages of essential medicines, medical commodities, and healthcare workers.14–16 Thus, the majority of the maternal and under-five emergencies are referred to secondary or tertiary hospitals for further care. However, patients in an emergency experience delays in being transferred to secondary or tertiary hospitals, especially in rural areas.17 Lack of transportation infrastructure, non-functional referral systems and long geographical distances lead to prehospital delays which increase the risk of maternal and perinatal mortality.17
To improve the maternal and under-five referral system, the Ministry of Health (MoH) introduced the National Emergency Medical Service (NEMS) in 2018. Referral coordinators were placed in secondary and tertiary hospitals to coordinate referrals including facilitation of NEMS ambulance deployment for transfer.18 A study during the pre-COVID-19 period reported that the introduction of the NEMS led to a 54% increase in hospital admissions of pregnant women by reducing challenges with geographical accessibility and transport unavailability, especially in the rural areas of Sierra Leone.18
In Sierra Leone, a study on the impact of COVID-19 on admissions to district and tertiary hospitals reported that maternal and paediatric admissions did not significantly change.19 However, there is no published literature on the extent of the impact of COVID-19 on the referral system, utilization of the NEMS ambulances and referral outcomes. This information is pivotal for strengthening the resilience of the established referral system to withstand the adverse impacts of the future outbreaks and public health emergencies. We aimed to assess the impact of COVID-19 on the maternal and under-five referral system by comparing, 1) the number of weekly maternal and under-five referrals, 2) the demographic and clinical characteristics of those referred, 3) the duration from referral to reaching the receiving facility (prehospital delay) and 4) referral and hospital exit outcomes between the pre-COVID-19 period (week one of January 2019 to week three of March 2020) and the COVID-19 period (week four of March 2020 to week four of December 2021).
Methods
Study design
This was a cohort study using secondary data routinely collected by the NEMS of the Ministry of Health (MoH), Sierra Leone.
Study setting
General setting
Sierra Leone is a low-income country in West Africa, sharing borders with Guinea and Liberia. The country is divided into five provinces and has 16 districts, with Freetown as the capital city. The population is estimated at 8 million (2022), of which about 59% live in poverty.
Sierra Leone has a three-tier public healthcare system with PHUs providing primary care, regional (4) and district (8) hospitals providing secondary care and three hospitals in Freetown providing tertiary care. The PHUs, in increasing order of case management capacity, include Maternal and Child Health Posts (616), Community Health Posts (320) and Community Health Centres (227) spread across the country. In 2010, the Government of Sierra Leone launched the Free Health Care Initiative (FHCI) to provide free health services to pregnant women, lactating mothers, and children under-five, and this was later extended to Ebola survivors in 2015.20,21
Specific setting
Maternal and child health services in Sierra Leone
The Maternal and Child Health Posts and Community Health Posts provide antenatal, intrapartum and postpartum care to all pregnant women without high-risk pregnancy or obstetric emergencies. Community Health Centres provide Basic Emergency Obstetric and Neonatal Care (BEmONC) services. The district and tertiary hospitals provide Comprehensive Emergency Obstetric and Neonatal Care (CEmONC) services managing high-risk pregnancies or obstetric emergencies with additional facilities for caesarean sections and blood transfusion services.22
The facilities designated as BEmONC and CEmONC in rural districts of Sierra Leone do not provide a full range of services due to inadequate infrastructure, trained healthcare workers, equipment, medicines and consumables. Patients are therefore referred from lower level facilities to higher levels of care. Sierra Leone has national protocols for emergency obstetric care, including referral indications. Common reasons for maternal referrals include obstetric haemorrhage, eclampsia and prolonged obstructive labour. Like maternal care, even under-five children with health emergencies are referred to district hospitals for further management. Common reasons for under-five referrals include severe malaria, severe pneumonia, diarrhoea with severe dehydration, and sepsis.
Referral between public health facilities
Maternal and under-five referrals from the PHU level are initiated when the healthcare worker at the PHU determines that the patient needs to be referred to a higher level of care in line with the national guidelines. Referrals are classified subjectively by the health worker as urgent when the patient needs a life-saving intervention within 24 hours of presentation. The healthcare worker contacts NEMS via a toll-free phone call to initiate the process. In addition, the health worker calls the referral coordinator at the receiving district or tertiary hospital by phone to inform them about the need for referral and to ensure arrangements are in place to receive the patient. For all the incoming referrals, the referral coordinator collects demographic and other details like age, gender, facility name and type, the referral origin, the reason for referral, urgency of referral and time of call for referral (Figure 1).
Figure 1. Process of referral between public health facilities for managing maternal and under-five emergencies in Sierra Leone.
Transportation of a referred patient
When an ambulance is required to transfer patients to a higher level of care, the healthcare worker or the referral coordinator at the referring facility calls the NEMS using the 117 toll-free emergency number or the designated NEMS toll-free number. Once the call is made, NEMS deploys the ambulance if available. The NEMS ambulance services are provided free of cost to pregnant women, lactating mothers and under-five children under the FHCI. If the NEMS ambulance is unavailable, a non-NEMS ambulance is deployed if available. A patient can also decide to use a private vehicle when they are notified that they need to be referred to a higher level of care or if ambulances are unavailable, incurring out-of-pocket expenditure.
Referral outcomes
When the patient reaches the receiving facility, the referral coordinator records the arrival date and time at the hospital triage unit. The referral coordinator also records the referral outcomes, including admission, death on arrival and onward referral (referral to a higher level of care). If the patient dies in transit, the outcome is recorded as ‘death before arrival’. If the patient never reaches the receiving facility, the outcome is noted as ‘patient not arrived’ or sometimes the outcome details remain unrecorded. In case the patient requires referral to a higher-level facility, the referral coordinator arranges for the onward referral. The referral coordinators follow up on all the admitted referrals and update the hospital exit outcomes every week as ongoing admission, discharge, discharge against medical advice and death.
Recording and reporting
Referral coordinators document the demographic and clinical details of each incoming and outgoing referral using a standardized paper-based case record form. The NEMS uses the Referral Coordinator database on Epi Info™ to monitor the referrals and bed occupancy in the district and tertiary hospitals. The database was fully functional in all the facilities from January 2019.
Impact of COVID-19 on maternal and under-five referrals in Sierra Leone
On 24th March 2020, the President of Sierra Leone declared a state of emergency due to the COVID-19 pandemic. The first reported case of COVID-19 was on 31st March 2020. Multiple national lockdowns were introduced during April and May 2020 as a mitigating measure. Inter-district travel was banned from 14th April to 4th July 2020.19 The Ola During Children’s Hospital, Sierra Leone’s sole paediatric hospital, closed for four months during the pandemic. Some of the NEMS ambulances which were not operational prior to the COVID-19 pandemic were used for COVID-19 response. All these changes may have collectively impacted timely access to healthcare, particularly for maternal and under-five referrals.19
Study population
The study included all maternal and under-five referrals made between the public health facilities of Sierra Leone from January 2019 to December 2021. The maternal referrals include all the referrals of women at any time during the antepartum (during pregnancy), intrapartum (during labor and delivery) and up to 42 days postpartum (after delivery) periods. The under-five referrals include all the referrals of children below the age of five years. All the referrals from week one to January 2019 to week three of March 2020 were considered as pre-COVID-19 period referrals and those from week four of March 2020 to work four of December 2021 as COVID-19 period referrals.
Data collection and variables
All the variables required for the study were available in the Referral Coordinator Database maintained by the NEMS. The NEMS shared the database as a Microsoft Excel Workbook in July 2023. The unique referral ID was used to remove duplicate entries. The database of maternal and under-five referrals with variables like age, gender, province and type of referral facility, diagnosis, urgency of referral, date of referral, time of referral, district, mode of transportation used, date of reaching receiving facility, time of reaching receiving facility, province and type of receiving facility, referral and hospital outcomes were created. The referral origin district was used to create a referral province variable.
Operational definitions
Duration to reach referred facility: The time between referral initiation, which is the documented time the referral coordinator receives a phone call from a PHU healthcare worker, and the time the patient arrived at the receiving facility
Referred out of district: All those with a different referral origin district different from that of the referral destination district
Patient did not arrive: All those who had referral outcomes reported as ‘Patient did not arrive’ and those for whom referral outcomes were not recorded.
Data analysis
Data available in the Microsoft Excel Workbook was analysed using Stata 16.1 software (College Station, TX, USA: StataCorp LLC). R programming is an open-access alternative that can be used. The number of maternal and under-five referrals, urgent referrals and referrals on NEMS ambulances per week during the study reference period (January 2019 to December 2021) were deduced. The median and interquartile range (IQR) were used to summarize the weekly number of referrals, emergency referrals and referrals on ambulances during the pre-COVID-19 and COVID-19 periods. The Mann–Whitney U test was used to assess significant differences in the number of weekly referrals during the pre-COVID-19 and COVID-19 periods.
Interrupted time series (ITS) analysis was used to quantify the immediate and long-term effects of COVID-19 on the number of weekly referrals. The third week of March 2020, when the President of Sierra Leone declared a state of emergency due to the COVID-19 pandemic, was taken as an interruption in ITS analysis. Using the Newey–West regression methods, the predictive linear model for the pre-COVID-19 period (segment 1) with data from 65 weeks (first week of January 2019 to the third week of March 2020) was developed to reflect the counterfactual during the COVID-19 period (fourth week of March 2020 to the fourth week of December 2021). The predictive linear model accounted for secular trends, autocorrelation, and seasonality. Similarly, the predictive linear model for the COVID-19 period (segment 2) with data from 91 weeks (fourth week of March 2020 to the fourth week of December 2021) was developed. The beta coefficients (β) with a 95% confidence interval (CI) were obtained for the intercept (starting point of the model) and the weekly trend (average increase or decrease in the numbers during consecutive months) in the number of referrals during the pre-COVID-19 and COVID-19 period. The immediate effect of the COVID-19 interruption was quantified using level change in the intercept in the COVID-19 period compared to counterfactual for the interruption week obtained from the pre-COVID-19 period. A negative β coefficient for level change indicated the immediate decline in the number of referrals with the interruption.
Frequencies and percentages were used to describe the demographic and clinical characteristics, mode of transportation and outcomes of the referrals made during the pre-COVID-19 and COVID-19 periods. The demographic and clinical characteristics, mode of transportation and outcomes were compared across two time points using the Chi-squared test. The Mann–Whitney U test was used to assess whether there was a statistical difference in the median (IQR) duration of arrival at the referred facility between the pre-COVID-19 and COVID-19 periods.
Ethical considerations
Permission to use the data was obtained from the Chief Executive Officer of the National Emergency Medical Services, Ministry of Health in Sierra Leone. This study received ethical approval from the Sierra Leone Ethics and Scientific Review Committee, Freetown, Sierra Leone (SLESRC No. 010/11/2023) on 9th November, 2023. As part of the standard procedure of SORT IT programme, the International ethical approval was received from the Union Ethics Advisory Group of the International Union against Tuberculosis and Lung Disease, Paris, France (EAG Number 25/23) on 8th September, 2023.
Results
Differences in the number of weekly referrals (Tables 1 and 2)
Table 1. Number of maternal and under-five referrals to district or tertiary hospitals from other public health facilities in Sierra Leone during the pre-COVID-19 (January 2019 to week 3 of March 2020) and COVID-19 period (week 4 of March 2020 to December 2021).
| Parameters | Pre-COVID-19 | COVID-19 | p value* | ||
|---|---|---|---|---|---|
| Median | (IQR) | Median | (IQR) | ||
| Number of maternal and under-five referrals per week | |||||
| Total | 406 | (378-437) | 317 | (255-364) | <0.001 |
| Maternal | 227 | (205-243) | 205 | (153-240) | 0.008 |
| Under-five | 177 | (149-202) | 104 | (89-123) | <0.001 |
| Number of referrals per week by province | |||||
| Western Area | 127 | (113-137) | 82 | (71-91) | <0.001 |
| Southern Province | 96 | (87-110) | 73 | (47-99) | <0.001 |
| North West Province | 40 | (33 – 45) | 33 | (27-41) | <0.001 |
| Northern Province | 76 | (65-87) | 52 | (37-71) | <0.001 |
| Eastern Province | 70 | (59-78) | 65 | (54-76) | 0.186 |
| Number of referrals per week during | |||||
| Dry (December to April) | 380 | (333-413) | 317 | (223-358) | <0.001 |
| Rainy (May to November) | 417 | (399-465) | 308 | (252-347) | <0.001 |
| Number of referrals per week from | |||||
| Maternal Child Health Posts | 74 | (64-79) | 58 | (40-72) | <0.001 |
| Community Health Posts | 61 | (56-70) | 48 | (37-57) | <0.001 |
| Community Health Centres | 226 | (205-245) | 163 | (133-186) | <0.001 |
| Private/NGO Facility | 26 | (20-31) | 13 | (10-18) | <0.001 |
| District hospitals | 12 | (9-14) | 10 | (6-14) | 0.008 |
| Number of urgent referrals per weekly | 396 | (362-424) | 306 | (243-349) | <0.001 |
| Number of referrals per week on NEMS ambulance | 348 | (311-372) | 269 | (180-324) | <0.001 |
Table 2. Number of maternal and under-five referrals to district or tertiary hospitals from other public health facilities in Sierra Leone during the pre-COVID-19 (January 2019 to week 3 of March 2020) and COVID-19 period (week 4 of March 2020 to December 2021).
The median (IQR) number of weekly referrals decreased from 406 (378 – 437) in the pre-COVID-19 period to 317 (255 – 364) in the COVID-19 period (p<0.001). There was a similar significant decrease in the median (IQR) number of weekly referrals of maternal (227 (205 – 243) to 205 (153 – 240)) and under-five (177 (149 – 202) to 104 (89 – 123)) patients. The statistically significant decrease in the referrals during the COVID-19 period compared to the pre-COVID-19 period was seen in all the provinces of Sierra Leone except the Eastern province, and across all the referring facilities, as seen in Table 1. The number of weekly referrals with a NEMS ambulance also decreased from 348 (311 – 372) in the pre-COVID-19 period to 269 (180 – 324) in the COVID-19 period (p<0.001) (Table 1 and Figure 2).
Figure 2. Clustered area diagram showing the number of weekly maternal and under-five referrals to district or tertiary hospitals from other public health facilities in Sierra Leone from January 2019 to December 2021.
The interrupted time series (ITS) analysis showed a significant decrease in the number of under-five referrals (β coefficient = -38.7 (95% CI -70.7 – -6.7)) in the fourth week of March 2020 as shown in Figure 3C. However, the maternal referrals increased significantly with interruption (β coefficient = 23.1 (95% CI (1.6 – 44.7)) as seen in Figure 3B. There was no significant immediate decline in the number of referrals with the NEMS ambulances (β coefficient -6.3 (95% CI -50.5 – 38.0)) as shown in Figure 3D. During the entire COVID-19 period, there was a significant declining trend in total referrals (β coefficient = -3.0 (95% CI -4.5 – -1.5), maternal referrals (β coefficient = -2.8 (95% -3.6 – -2.1), and referrals on the NEMS ambulances (β coefficient = -5.3 (-6.8 – -3.8)), as shown in Table 2.
Figure 3. Interrupted time series showing the unadjusted linear trend in the number of maternal and under-five referrals to district or tertiary hospitals from other public health facilities in Sierra Leone during the Pre-COVID-19 (segment-1) and COVID-19 period (segment-2).
(A) Total referrals; (B) Maternal referrals; (C) Under-five referrals; (D) Referrals on NEMS ambulance started in January 2019 as this is when the service was fully launched; Segment-1: Pre-COVID-19 (January 2019 to week 4 March 2020), Segment-2: COVID-19 period (week 4 March 2020 – December 2021); Interruption: Month the first of case of COVID-19 was reported (March 2020). Abbreviation: NEMS: National Emergency Medical Service.
Differences in demographic and clinical characteristics of those referred (Tables 3 and 4)
Table 3. Demographic and clinical characteristics of maternal referrals to district or tertiary hospitals from other public health facilities in Sierra Leone during the pre-COVID-19 (January 2019 to week 3 of March 2020) and COVID-19 period (week 4 of March 2020 to December 2021).a
| Characteristics | Pre-COVID period | COVID period | p Value$ | ||
|---|---|---|---|---|---|
| n | (%)* | n | (%)* | ||
| Total | 14,057 | (100) | 18,027 | (100) | |
| Age in years | <0.001 | ||||
| ≤ 18 | 2,484 | (18) | 2,869 | (16) | |
| 19-24 | 4,737 | (34) | 6,456 | (36) | |
| 25-29 | 3,016 | (21) | 4,069 | (23) | |
| 30-34 | 2,177 | (16) | 2,659 | (15) | |
| 35 and above | 1,639 | (12) | 1,968 | (11) | |
| Missing | 3 | (<1) | 16 | (<1) | |
| Referral origin province | |||||
| Western Area | 4,067 | (29) | 4,543 | (25) | <0.001 |
| Southern Province | 3,379 | (24) | 4,473 | (25) | |
| North West Province | 1,759 | (13) | 2,193 | (12) | |
| Northern Province | 2,421 | (17) | 3,896 | (16) | |
| EasternProvince | 2,431 | (17) | 3,922 | (22) | |
| Diagnosis | <0.001 | ||||
| Abortion | 72 | (<1) | 216 | (1) | |
| Pre-Eclampsia/Eclampsia | 190 | (1) | 802 | (4) | |
| Haemorrhage | 301 | (2) | 1,487 | (5) | |
| High-risk pregnancy | 539 | (4) | 2,184 | (8) | |
| Malaria | 15 | (<1) | 53 | (<1) | |
| Obstetric - other | 151 | (1) | 1,683 | (10) | |
| Other | 127 | (<1) | 773 | (3) | |
| Missing | 12,652 | (90) | 10,829 | (60) | |
| Referred from | |||||
| Community Health Centre | 7,420 | (53) | 9,080 | (50) | <0.001 |
| Community Health Posts | 2,774 | (20) | 3,157 | (18) | |
| District Hospital | 115 | (1) | 291 | (2) | |
| Maternal and Child Health Post | 3,164 | (22) | 4,099 | (23) | |
| Private/NGO Facility | 306 | (2) | 258 | (1) | |
| Other (including tertiary facility) | 273 | (2) | 1,124 | (6) | |
| Missing | 5 | (<1) | 18 | (1) | |
| Referred to | |||||
| Tertiary hospitals | 4,117 | (30) | 4,095 | (23) | <0.001 |
| District hospitals | 9,486 | (67) | 13,086 | (73) | |
| Other (including CHC, CHP, MCHP) | 454 | (3) | 837 | (5) | |
| Referred within the province | |||||
| Yes | 13,984 | (99) | 17,789 | (98) | <0.001 |
| No | 85 | (<1) | 238 | (2) | |
| Referral classification | |||||
| Urgent | 13,867 | (99) | 17,148 | (95) | <0.001 |
| Non-Urgent | 190 | (1) | 879 | (5) | |
Table 4. Demographic and clinical characteristics of under-five referrals to district or tertiary hospitals from other public health facilities in Sierra Leone during the pre-COVID-19 (January 2019 to week 3 of March 2020) and COVID-19 period (week 4 of March 2020 to December 2021).a
| Characteristics | Pre-COVID period | COVID period | p Value$ | ||
|---|---|---|---|---|---|
| n | (%)* | n | (%)* | ||
| Total | 11,677 | 100.00 | 9,886 | 100.00 | |
| Age in years | <0.001 | ||||
| ≤1 | 7,990 | (68) | 6,496 | (66) | |
| 2-5 | 3,687 | (32) | 3,390 | (34) | |
| Sex | 0.602 | ||||
| Female | 5,373 | (46) | 4,584 | (46) | |
| Male | 6,304 | (54) | 5,302 | (54) | |
| Referral origin province | <0.001 | ||||
| Western Area | 3,856 | (33) | 2,781 | (28) | |
| Southern Province | 2,992 | (26) | 2,459 | (25) | |
| North West Province | 801 | (7) | 863 | (9) | |
| Northern Province | 2,193 | (19) | 1,8272 | (19) | |
| Eastern Province | 1,835 | (16) | 2,132 | (19) | |
| Diagnosis | <0.001 | ||||
| Diarrohea | 29 | (<1) | 43 | (<1) | |
| Malaria | 435 | (4) | 2,586 | (26) | |
| Malnutrition | 137 | (1) | 617 | (6) | |
| Sepsis | 19 | (<1) | 42 | (<1) | |
| Pneumonia | 165 | (1) | 487 | (5) | |
| Other | 52 | (<1) | 169 | (2) | |
| Missing | 10,840 | (93) | 5,951 | (60) | |
| Referred from | <0.001 | ||||
| Community Health Centre | 6,698 | (57) | 5,596 | (57) | |
| Community Health Posts | 1,179 | (10) | 1,114 | (11) | |
| District Hospital | 668 | (6) | 639 | (7) | |
| Maternal and Child Health Post | 1,517 | (13) | 1,177 | (12) | |
| Private/NGO Facility | 1,327 | (11) | 915 | (10) | |
| Other (including tertiary facility) | 287 | (3) | 433 | (5) | |
| Other | 1 | (<1) | 12 | (<1) | |
| Referred to | <0.001 | ||||
| Tertiary hospitals | 4,017 | (35) | 2,554 | (27) | |
| District hospitals | 7,443 | (64) | 6,910 | (70) | |
| Other (including CHC, CHP, MCHP) | 217 | (2) | 416 | (4) | |
| Missing | 0 | (0) | 6 | (<1) | |
| Referred within the province | <0.001 | ||||
| Yes | 11,454 | (98) | 9,488 | (96) | |
| No | 223 | (1) | 398 | (4) | |
| Referral classification | <0.001 | ||||
| Urgent | 10,935 | (94) | 9,736 | (99) | |
| Non-Urgent | 742 | (6) | 150 | (1) | |
In total there were 14,057 and 18,027 maternal referrals during the pre-COVID-19 and COVID-19 periods, respectively as shown in Table 3. The mean (SD) age of the referrals during both the pre-COVID-19 and COVID-19 period was 25(7) years. The details of diagnosis were missing in 90% and 60% of referrals during the pre-COVID-19 and COVID-19 period, respectively (p<0.001). Of all the maternal referrals, 99% were ascertained as urgent during the pre-COVID-19 period, while 95% were ascertained as urgent during the COVID-19 period (p<0.001) (Table 3).
In total there were 11,677 and 9,886 under-five referrals during the pre-COVID-19 (91 weeks) and COVID-19 (65 weeks) periods, respectively as shown in Table 4. Of all the under-five referrals, 68% and 66% were infants (<1 year) during the pre-COVID-19 and COVID period, respectively (p<0.001). The details of diagnosis were missing in 93% and 60% of referrals during the pre-COVID-19 and COVID-19 period, respectively (p<0.001). Among under-five referrals, 94% and 99% were ascertained as urgent during the pre-COVID-19 period and COVID-19 period, respectively (p<0.001).
Mode of transportation and prehospital delay (Table 5)
Table 5. Mode of transportation, time to reach the referred facility and outcomes of maternal and under-five referrals to district or tertiary hospitals of Sierra Leone during the pre-COVID-19 (January 2019 to week 3 of March 2020) and COVID-19 period (week 3 of March 2020 to December 2021).a
| Characteristics | Maternal | Under-five children | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pre-COVID-19 | COVID-19 | P Value | Pre-COVID-19 | COVID-19 | P Value | |||||
| n | (%)* | n | (%)# | n | (%)* | n | (%)* | |||
| Total | 14,057 | (100) | 18,057 | (100) | 11,677 | (100) | 9,886 | (100) | ||
| Mode of transportation | <0.001 | <0.001 | ||||||||
| NEMS ambulance | 12,654 | (90) | 16,319 | (90) | 8,447 | (72) | 6,649 | (68) | ||
| Non-NEMS ambulance | 385 | (3) | 355 | (2) | 2,004 | (17) | 1,172 | (12) | ||
| Others@ | 1,018 | (8) | 1,353 | (8) | 1,226 | (11) | 2,065 | (21) | ||
| Median (IQR) duration from referral to reaching receiving facility (mins) | 72 | (45-123) | 86 | (54-142) | <0.001# | 75 | (49-117) | 90 | (59-141) | <0.001# |
| Referral outcomes | ||||||||||
| Patient did not arrive | 160 | (1) | 81 | (<1) | <0.001 | 90 | (<1) | 35 | (<1) | <0.001 |
| Death in Ambulance | 2 | (<1) | 9 | (<1) | 9 | (<1) | 18 | (<1) | ||
| Death on arrival | 30 | (<1) | 3 | (<1) | 160 | (1) | 59 | (<1) | ||
| Onward referral | 102 | (1) | 116 | (<1) | 121 | (1) | 142 | (1) | ||
| Admitted at receiving Hospital | 13,763 | (98) | 17,818 | (98) | 11,297 | (96) | 9,632 | (96) | ||
| Hospital exit outcomes | <0.001 | <0.001 | ||||||||
| Discharged | 13,536 | (96) | 16,935 | (93) | 9,715 | (83) | 7,768 | (79) | ||
| Death | 165 | (1) | 191 | (1) | 1,307 | (11) | 1,297 | (13) | ||
| Missing | 356 | (<1) | 901 | (5) | 655 | (6) | 821 | (8) | ||
Table 5 highlights that in all of the maternal referrals 90% were through a NEMS ambulance during both the pre-COVID-19 and COVID-19 period. For maternal referrals, the median (IQR) duration from referral to arrival at the receiving facility was 72 (45-123) minutes and 86 (54-142) minutes in the pre-COVID-19 and COVID-19 periods respectively (p<0.001).
Of all the under-five referrals, 72% were through a NEMS ambulance during the pre-COVID-19 period whereas this was 68% in the COVID-19 period. For under-five referrals, the median (IQR) duration from referral to arrival at the receiving facility was 75 (49-117) minutes and 90 (59-141) minutes during the pre-COVID-19 and COVID-19 periods respectively.
Referral outcomes and hospital exit outcomes (Table 5)
For maternal referrals, in both the pre-COVID-19 and COVID-19 periods, 98% of patients referred were successfully admitted in the receiving facilities. For under-five referrals, 96% of patients referred were admitted to the receiving facilities in both the pre-COVID-19 and COVID-19 periods.
Of all the maternal referrals, about 1% had death as a hospital outcome in both pre-COVID-19 and the COVID-19 periods. Of all the under-five referrals 11% had death as a hospital outcome in the pre-COVID-19 period and this was 13% in the COVID-19 period (p<0.001).
Discussion
This is the first study to assess the impact of the COVID-19 pandemic on the maternal and under-five referral system including prehospital delays and referral outcomes in Sierra Leone. This study has five key findings. Firstly, there was a significant decrease in the number of total maternal and under-five referrals per week during the COVID-19 period compared to the pre-COVID-19 period. Secondly, the number of under-five referrals decreased immediately after the President declared a state of emergency on March 24th, 2020, due to the COVID-19 pandemic, while the number of maternal referrals increased for a short period. Thirdly, the number of weekly referrals with the NEMS ambulances declined during the COVID-19 period, especially for under-five referrals. Fourthly, it took longer for both under-five and maternal referrals to reach a facility during the COVID-19 period compared to the pre-COVID-19 period. Finally, during the COVID-19 period there were no significant differences in the maternal and under-five referral outcomes.
This study reinforces the WHO Emergency Medical Teams (EMT) recommendation to establish an effective patient referral process during emergencies. The study findings are relevant to all LMICs that are progressing towards building resilient maternal and under-five referral systems to achieve the Sustainable Development Goal 3 targets.23,24 These findings are particularly relevant to the Directorate of Reproductive and Child Health and the NEMS of Sierra Leone, as strengthening maternal and under-five referral systems is one of their operational priorities.
This study has a number of strengths. First, we used routinely collected country-wide programmatic data reflecting the ground realities of the NEMS implementation in Sierra Leone. Second, a robust analysis was conducted using an interrupted time series (ITS) analysis, which accounted for seasonal variations and auto-correlation of the variable time periods (65 and 91 weeks) of the pre-COVID-19 and COVID-19 period, respectively. The ITS provided insights into the level of change from pre-COVID-19 to the COVID-19 period and trends during both periods. Finally, we adhered to STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines while conducting this study.25
The study has some limitations. First, we only conducted a quantitative assessment of the impact of COVID-19 on the referral system. However, in-depth qualitative interviews with the referral coordinators and the NEMS functionaries would have provided a deeper understanding of the reasons behind the quantitative differences found between the pre-COVID-19 and the COVID-19 periods. Second, we were unable to assess the duration of various stages from referral to reaching the receiving facility due to the absence of variables such as the time of deployment of the NEMS ambulance, the time of the NEMS ambulance reaching the referral facility and the time of picking up the patient from the referral facility. Thirdly, this study is unable to make a causal link between COVID-19 and changes in utilization of the referral systems, as other factors such as the functioning and financing of the NEMS may have contributed to the changes seen. Finally, the Referral Coordinator database had some deficiencies in recording and reporting variables such as diagnosis and urgency of referrals. The diagnosis variable was missing for over 60% of the referrals, and more than 90% of the referrals were marked as urgent due to lack of standard criteria for ascertaining the urgency of the referral. Despite these limitations, there are some important implications and recommendations from this study.
First, this study revealed that the performance of the NEMS referral system was maintained during the COVID-19 pandemic, despite the lower numbers of referrals. This shows the effectiveness of the NEMS engagement in the national preparedness response plan.26 The referral system was resilient as patients were able to access care and there were no significant differences in outcomes during the pre-COVID-19 and COVID-19 periods. There were low numbers of maternal deaths, highlighting efficiencies in referring maternal patients. Deaths of under-five referrals remained relatively the same during the pre-COVID-19 and COVID-19 periods, although there was a higher percentage of under-five deaths compared to maternal deaths which needs attention.
Second, this study demonstrated a significant decline in the number of referrals per week, especially for children under-five during the COVID-19 period compared to the pre-COVID-19 period. A study in Sierra Leone during the pandemic showed that about 60% of the population had poor knowledge of COVID-19, which would have adversely impacted their healthcare seeking behaviour and utilization.27 Like other neighbouring countries, Sierra Leone also reported a significant decline in the utilization of under-five healthcare services during the COVID-19 pandemic.28–31 Along with underutilization, some of the secondary/tertiary hospitals were either closed or diverted entirely for COVID-19 patient care. For example, the Ola During Children’s Hospital, which is the largest paediatric hospital in Sierra Leone, remained closed during the initial COVID-19 period and did not receive any referrals. This calls for better health emergency preparedness in outbreak prone countries like Sierra Leone. Health emergency responses should avoid the closure or diversion of entire health facilities to ensure continuity of essential health services defined in the Sierra Leone Basic Package of Essential Services, with priority for maternal and under-five children services. Steps to consider to avoid the closure of essential services include reinforcement of preventive services and health promotion activities, careful application of public health measures for healthcare workers, and regular monitoring of essential health services.32 Also, multi-media campaigns and engagement with community and religious leaders can be used to generate awareness and build confidence among the general public to utilise essential and emergency healthcare services.
Third, there was a significant increase in maternal referrals immediately after the declaration of the COVID-19 emergency in Sierra Leone. The closure of some of the PHUs and the lack of personal protective equipment (PPEs) for delivering healthcare services during the initial phase of COVID-19 compelled the referral of pregnant women to secondary or tertiary hospitals for care.33–35 A study also revealed the resilience of hospitals reporting increases in the number of caesarean sections and deliveries in secondary and tertiary hospitals of Sierra Leone during the initial phase of the COVID-19 pandemic.19,36 To avoid such unnecessary referrals and to avert the risks associated with such transfer during health emergencies, the PHUs should be well equipped with necessary PPEs for continuing the essential and emergency health services.
Fourth, during the COVID-19 pandemic, the number of weekly referrals with NEMS ambulances declined, and there were especially lower proportions of under-five children who received the NEMS ambulance services. This concurs with a study in Liberia which revealed a decline in ambulance referrals due to their diversion to support the COVID-19 response.37 As some of the NEMS ambulances and their operators were deployed for the COVID-19 response, all the requests from the maternal and under-five referral system could not be attended to.38 This required maternal and under-five referrals to use alternate modes of transportation incurring out-of-pocket expenditures. We also observed that the time to reach the receiving facility increased during the COVID-19 period compared to the pre-COVID-19 periods. The unavailability of ambulances in the face of increased demand for the COVID-19 response could have impacted the duration of referrals during the COVID-19 period.26,39 Also, mobility restrictions, checkpoints and diversions due to curfews could have contributed to an increase in the duration of transfer.39
Possible ways to address the challenge of reduced transfers through the NEMS ambulance and increased delays during pandemic, include: a) increasing the fleet of ambulances by immediate repurposing of available government vehicles and developing partnerships with the private and commercial sectors for mobilisation of transport resources; b) as part of emergency preparedness in pandemic prone countries such as Sierra Leone, having a reserve fleet of ambulances within the health services as a “safety net” to cater for increased demand; c) establishing a human resource surge capacity to increase the number of ambulance drivers and operators; and d) ensuring free and prioritised passage of ambulances at border and control checkpoints. There should also be considerations to include innovation and community engagement in referral transportation methods which has been done in Liberia by including poda podas.40
Lastly, some deficiencies were noted in the Referral Coordinator database which need urgent attention. The current database does not capture time at various stages between referral to reaching the receiving facility. There is a need to introduce these variables in the database to allow identification of the critical path in the referral process and eventually reduce the overall duration during referral and transfer. There is a need to improve the completeness of data, especially on the diagnosis at the point of referral and institute standard criteria for categorisation of urgent referrals. Improving the completeness of diagnosis and correctness in ascertaining the urgency of referral, would aid in auditing the appropriateness of the referrals. Additionally, the existing NEMS databases should be used for real-time monitoring and for making evidence-based actions to improve the referral system.
Conclusions
In conclusion, this cohort study found that the NEMS system performance was maintained in terms of referrals reaching the receiving health facilities, despite the lower number of weekly referrals during the COVID-19 period. There is a need to sustain this performance while also making more efforts to increase utilization and reduce delays during outbreaks/pandemics. This study reveals an increase in the prehospital delay although these differences are unlikely to contribute to mortality. It is important to ensure that there is not a prolonged delay given the associations between the delay to reach care and maternal mortality and as such it is critical to address the timeliness to reach care. We recommend that efforts should be made to build a resilient maternal and under-five referral system that can adapt to health emergencies.
Download
Export To
metrics
| Views | Downloads | |
|---|---|---|
| F1000Research | - | - |
|
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Citations
CITE
how to cite this article
Kamara J, Moses F, Thekkur P et al. The effect of the COVID-19 pandemic on emergency maternal and under-five referrals in Sierra Leone: A cohort study [version 1; peer review: peer review discontinued]. F1000Research 2024, 13:703 (https://doi.org/10.12688/f1000research.149455.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
track
receive updates on this article
Track an article to receive email alerts on any updates to this article.
Peer review discontinued
At the request of the author(s), this article is no longer under peer review.
What does this mean?
Comments on this article Comments (0)