Patterns of paediatric analgesic use in Africa: a systematic review
- Parvaz Madadi1,2,
- Ehijie F O Enato3,
- Shir Fulga2,
- Chinonye C Umeoduagu3,
- Stuart M MacLeod4,5,
- Gideon Koren1,2,
- Thomas R Einarson2,6
- 1Division of Clinical Pharmacology and Toxicology, Department of Paediatrics, Hospital for Sick Children, Toronto, Ontario, Canada
- 2Motherisk Program, Hospital for Sick Children, Toronto, Ontario, Canada
- 3Department of Clinical Pharmacy and Pharmacy Practice, University of Benin, Benin City, Nigeria
- 4Child and Family Research Institute, University of British Columbia, Vancouver, British Columbia, Canada
- 5Provincial Health Services Authority, Vancouver, British Columbia, Canada
- 6Department of Pharmacy, University of Toronto, Toronto, Ontario, Canada
- Correspondence to Dr Parvaz Madadi, Division of Clinical Pharmacology and Toxicology, Department of Paediatrics, Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada M5G 1X8; ,
- Received 22 May 2012
- Accepted 10 August 2012
- Published Online First 13 September 2012
We conducted a systematic literature review with two objectives: (1) to assess reported patterns of analgesic use in African children and compare these observed patterns to the analgesics given in the WHO Essential Medicines List for Children (EMLc); and (2) to summarise outcomes related to effectiveness, adverse events, cost and accessibility of these analgesics. Eligible participants were children (≤12 years) living in any African country who received an analgesic administered with the intention of relieving pain in any setting. Thirty-four peer-reviewed, observational studies representing 7772 African children were accepted. Studies were conducted in 25 different regions of 12 countries. Pain was attributed to surgery, burns, sickle cell anaemia and conditions requiring palliation in 32% of children, and was unspecified in the other 68%. Of the three EMLc analgesics, paracetamol and ibuprofen were widely employed, constituting ∼60% of all analgesics, while morphine was used in 20 children (0.2%). There were 455 suspected adverse drug reactions which included 17 deaths. Analgesic use reported in African children appears to fall short of WHO standards.
The recognition, management and treatment of pain, defined as “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage”,1 is highly problematic in both developed and developing countries. While it is clear that untreated or inadequately treated pain can result in considerable morbidity, including physiological, psychological, economic and social consequences for patients, families and society,2 ,3 how to treat pain remains a particularly contentious issue. This complexity arises from the interplay of cultural, educational, political, religious and logistical factors in a given country.4 Central to this issue is the struggle to achieve a balanced pain policy which maximises access to opioid analgesics for medical use while minimising abuse and dependence.5 To address the imbalance between the prevention of abuse of controlled substances and their use for legitimate medical purposes, the WHO proposed the Access to Controlled Medications Programme in 2010.
The need to treat pain in children has been a historically neglected issue. Despite extensive evidence-based policy guidelines and clinical recommendations, pain in children often remains untreated.6–8 The WHO has developed the Model List of Essential Medicines for Children (EMLc),9 which catalogues a core list of minimum medicine needs for a basic healthcare system. Table 1 summarises the essential analgesic list for children. While morphine, paracetamol and ibuprofen are included, the extent to which children have access to, and actually receive these analgesics for the treatment of pain on a global scale, is largely unknown. Importantly, there are no data available on how paediatric pain is treated in African countries, which comprise a region of the world with a high incidence of paediatric mortality and morbidity.
In light of the aforementioned gaps in knowledge, the objectives of this study were: (1) to systematically summarise the types of analgesics given to children in African countries and compare these observed patterns to the EMLc analgesics, and (2) to summarise the outcomes (accessibility, adverse events, cost and effectiveness) of these analgesics in Africa.
This research involved a structured review of the literature, as well as analysis and synthesis of the findings. Both quantitative and qualitative aspects of the literature have been assessed. Since observational studies were of primary interest, the review was performed in accordance with the PRISMA statement.10 All technical files are available from the authors on request.
Study inclusion criteria
We established a priori a set of criteria to define acceptable participants and types of studies to review. Eligible participants were children up to 12 years of age living in any African country, in agreement with the EMLc,9 which is intended only for that age group. These children must have received an analgesic, defined for this research as any xenobiotic (pharmaceutical medication or herbal preparation) administered with the intention of relieving pain.
We included peer-reviewed observational research studies (epidemiological, naturalistic, population-based, prospective or retrospective cohort, and cross-sectional surveys) which described analgesic use in African children. The study could have been reported in any language, and must have been conducted in part or entirely in any African country, but data from Africa must have been presented separately from those of other continents. The study must have been published between 1 January 1990 and 31 December 2010.
Two reviewers searched MEDLINE independently using the following search terms: neonate, child, p(a)ediatric(s), Africa, pain, analgesic(s), analgesia, ibuprofen, paracetamol, acetaminophen, opioid, codeine, morphine, herbal, traditional medicine, as well as the names of all African countries. Terms were translated into French and searches were repeated. Articles were retrieved and checked against predetermined inclusion criteria (listed above). Articles identified as potentially acceptable were then compared between the two reviewers. Discrepancies were noted and discussed until agreement was reached. In case of continued disagreement between the two reviewers, a third reviewer adjudicated.
Data extraction and analysis
Three data extractors reviewed each of the accepted studies independently and extracted data according to the prepared guidelines and using a pre-designed data extraction form. If needed, authors of the accepted studies were contacted for missing data. Data from studies presented in multiple publications were identified and the dataset was counted once in the sample size tally. The data from each step of this process were documented in a flow chart (figure 1) as described in the PRISMA statement. An overall summary of the number and types of articles included in the review was then prepared. The quality (ie, strength of research design) of each accepted article was assessed using the Methodological Index for Non-Randomized Studies scale.11 The scale has been validated for use with observational studies. Next, a narrative review was undertaken of accepted studies, along with quantitative summaries of data obtained. Types of analgesics used, conditions for which they were used, doses administered, reported effectiveness, adverse events, accessibility of analgesics, and associated costs were tabulated. Additionally, qualitative text related to the four latter themes was extracted, analysed and summarised. The prevalence of use of analgesics, overall and individually, was extracted from all studies reporting adequate information. The proportion of children receiving the EMLc analgesics listed in table 1 was determined. Data were categorised based on drug class. Descriptive statistics (mean, SD, median, minimum and maximum) were calculated. Data were combined across studies to provide an overview of use, when possible.
Of 1468 articles identified, 34 met our inclusion criteria and were included in the analysis (figure 1).12–45 These studies collectively reported on 7772 African children who had received one or more analgesic agents for pain relief (see online supplementary table 1 for study descriptions).12–45 The majority of studies that met the inclusion criteria were performed in the years 2000–2010, with only 17% of studies performed between 1990 and 1999 (see online supplementary figure 1). The studies were performed in 25 different regions of 12 African countries; 67% of all studies were performed in either Nigeria or South Africa (see online supplementary table 1).
For 68% of children, the cause of pain (or the reason for administering an analgesic) was not identified (see online supplementary figure 2). This ‘unspecified category’ also included ambiguous classifications of general illness. Post-operative analgesia was the most commonly specified indication (24%), followed by pain crises in children with sickle cell disease (5%), burn injury (3%) and palliative conditions (<1%).
Pharmaceuticals were used for pain relief in ∼75% of the study sample (table 2). Paracetamol was the most widely used analgesic for pain ranging from mild to severe, and constituted at least 40% of all analgesic use. Together with ibuprofen, these compounds were utilised in ∼60% of the study children. Morphine was used in 0.25% of the sample, namely 20 children from Uganda (n=11; palliative condition),33 Nigeria (n=6; burn injury)22 and South Africa (n=3; major abdominal surgery).20 Opioids in general constituted ∼8.6% of all analgesic use, the most prevalent being pentazocine (5%), followed by tramadol (1.4%), tilidine (0.6%), meperidine (0.5%), fentanyl (0.4%) and morphine (0.2%).
In addition to ibuprofen, the non-steroidal anti-inflammatory drugs acetylsalicylic acid and dipyrone (metamizole) were used in 1.3% and 0.5% of children, respectively. Additionally, over 5% of the study sample was administered anaesthetic agents post-operatively for pain relief (bupivicaine, 2.3%; ketamine, 3.4%). Table 2 also describes 455 suspected adverse drug reactions (ADRs) associated with analgesic use (one patient could have multiple episodes); amongst these 17 deaths were reported. The true incidence of ADRs is unknown, as monitoring and surveillance were not implemented across studies.
Non-pharmaceutical and/or traditional preparations were utilised in approximately one sixth of the study sample. Breast milk, with or without glucose, was most commonly administered for pain relief to neonates (14%, n=1086). There were regional differences in the utilisation of traditional medicines. In the South African rural region of Umtata, more than half of children regularly received traditional remedies. The most common reason for administering herbal medications was ‘playte’, a generic name given to all types of ill health, anticipated, real or perceived, that may befall a child.21 In Kwazulu-Natal (South Africa), almost all infants received a non-prescribed medication in the first 3 months of life: oral preparations of gripe water, Muthi Wenyoni or enema preparations of herbal/animal extracts with soap and water.39 In western Kenya, a sample of 57 Luo children were able to list and identify a total of 70 herbal remedies used in their community for pain and skin infections.32
Qualitative appraisal of the approved studies revealed several themes associated with the availability of analgesics. In many localities, ‘prescription-only’ drugs were often freely available over-the-counter, without a prescription. Opioids as a drug class were not available via local shops or community practitioners and were “generally reserved for hospital use”.28 Yet within hospitals, opioid use was limited. In Gabon, for example, “codeine is the highest step available on the WHO analgesic ladder”.35 Moreover, even in hospitals in which opioids are available to practitioners, the “non-availability of pediatric ventilators in many centers requires [physicians to] select analgesics that have no proven respiratory effects”.16 ,17
Few studies included a discussion of the cost-effectiveness of analgesics. In one South African hospital, analgesics accounted for 14% of the total cost of all products prescribed (but accounted for only 2.3% of all drugs given).31 In Nigeria, more potent analgesics were deemed too expensive for the average family to afford, which was attributed to a high rate of hospitalisation due to sickle cell pain.25 Due to the unaffordable cost of EMLA cream19 or ‘stronger’18 analgesics, non-pharmacological methods including breastfeeding, massage, eye contact, skin-to-skin contact and oral glucose were emphasised as pain-relieving measures amongst neonates undergoing surgery in several localities.18 ,19 Symptoms of persistent pain were identified in children breastfed with or without glucose14 ,16 and in children administered paracetamol12 ,16 ,18 ,27 and pentazocine (table 2).16 ,17
Increasingly, there is a movement towards the realisation of the concept of pain management as a universal human right.4 ,46 Yet within this subset of African children who were receiving analgesics, it appears that pain treatment was far from optimal.
Of the three EMLc analgesics, paracetamol and ibuprofen were widely available and given to ∼60% of children. Persistent pain was reported following the administration of these analgesics.12 ,16 ,18 ,27 Morphine, on the other hand, was administered to only 20 children (0.2%), despite indications such as major surgery, palliative pain, severe burn and pain crises requiring hospitalisation. It follows that while half of global childhood deaths occur due to AIDS and cancer in sub-Saharan Africa, palliative care services for children are limited,33 and opioid availability remains a major constraint in these areas.47
Morphine is an EMLc-recommended analgesic for severe pain because it has been established as safe, efficacious and cost-effective. Yet, our study revealed that several centres were hesitant in administering the drug to neonates due to the absence of paediatric ventilators. These authors raise an important point—the safety profile of an analgesic is not only determined by the pharmacological properties of the drug itself, but also the environment in which the drug is administered. Alternatively, dipyrone was utilised in these centres; however, this controversial medication has been long banned from over 30 countries due its association with agranulocytosis.48 It is clear that resource-poor settings may weigh the risks and benefits of analgesics differently; real or perceived risk is region-specific and needs to be considered as part of a global strategy for rational drug utilisation.
Overall, the mean opioid consumption on the African continent is 12 times lower than the global average (0.56 mg/capita vs 6 mg/capita, respectively).49 These data do not distinguish between adult and paediatric consumption. The morphine equivalency metric used to calculate opioid consumption is based on the overall use of six opioids indicated for moderate to severe pain: fentanyl, hydromorphone, methadone, morphine, oxycodone and meperidine.49 However, our study revealed that the most commonly used opioid was pentazocine, a synthetic mixed opioid agonist–antagonist administered exclusively to Nigerian children.12 ,16–18 ,22 This discovery illustrates that the global morphine equivalency metric may not reflect actual opioid use in countries utilising less popular opioid analgesics.
Fifteen pentazocine and two fentanyl-related deaths were reported; these deaths were associated with the onset of respiratory depression shortly after drug administration. The deaths occurred in neonates (mean age 8±1.2 days, mean weight 2.7±1.5 kg, with surgical indications such as colostomy, herniotomy and laparotomy) who were manually ventilated for 1–36 h after surgery and who failed to respond to naloxone. The authors noted that these infant deaths were compounded by the unavailability of paediatric ventilators and that the neonates “would have likely survived in developed countries”.16 ,17
The WHO estimates that up to 80% of some African populations depend on traditional medicine for primary healthcare. Similar to other reports,32 ,50 traditional remedies did not appear to be used to the exclusion of pharmaceutical agents such as paracetamol in our analysis. While we report a dramatic increase in the number of African articles published on the topic of paediatric analgesia over the past decade, the overall number of articles is still extremely small (34 studies in total) and reflects a limited sample of teaching hospitals and academic institutions in less than a quarter of all African countries. It could be that populations who depend on traditional medicines for primary healthcare may be less represented in this study sample. As such, the potential for both availability bias (representing systemic differences in centres who conduct and publish studies versus centres who do not publish such studies) and publication bias (the literature may reflect a biased selection of studies which have actually been undertaken) needs to be considered.
This study has several additional limitations. The appropriateness of analgesic use could not always be assessed. There was limited information pertaining to the physical, social and economic consequences of current paediatric pain practices in African countries. Furthermore, the factors influencing the choice of an analgesic for paediatric pain in a given region/institution was not studied beyond the authors’ comments. Importantly, since the starting point of analysis was the receipt of an analgesic, the unmet need for pain medication could not be determined.
As a next step, an analysis of the upstream, macrosocial factors determining patterns of paediatric analgesic use is needed in order to cultivate local and culturally relevant educational strategies for patients, families and prescribers. At the same time, barriers affecting access to and the supply and distribution of analgesic medicines51 must be identified and overcome. One concern is that in many African countries, prescription medications such as antimalarials can be freely obtained without prescriptions. It is believed that within such a framework, increased availability of opioid analgesics could become very problematic.31 Yet the North American crisis of opioid-related mortality in recent years has been attributed to widespread prescriptions of oral opioids to patients,52 illustrating that prescription-based availability of drugs alone may not be sufficient to deter opioid drug reliance and abuse. Educational strategies should encompass the principles of evidence-based and rational prescribing. In a region of the world that is disproportionately burdened with childhood morbidity and mortality, such policies and educational strategies are a vital step towards relieving the burden of pain and suffering.
What is known on this topic
Limited data are available on how paediatric pain is treated in African countries.
The WHO has developed the Model List of Essential Medicines for Children which includes morphine, paracetamol and ibuprofen.
What this study adds
Paracetamol and ibuprofen were accessible and widely used in African children with mild to severe pain.
Morphine was rarely used in African children with pain.
The authors thank Ms Justine Benevent for her assistance in translating French literature, and Ms Linda Gotlieb for library assistance.
Funding PM is or has been supported by fellowships from the Canadian Institutes for Health Research, the Quebec Training Network for Perinatal Research, and the Canadian Pain Society for the duration of this work.
Contributors PM, EFOE and TRE designed the study. SF and CCU conducted the literature search. SF, CCU and PM extracted data; PM and SF analysed the data. GK, SMM, EFOE and TRE provided methodological, pharmacological, paediatric and/or policy-related expertise. PM drafted the manuscript and all authors contributed to the final version of the report.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.