Global, regional and national availability of essential medicines for children, 2009–2020: a systematic review and meta-analysis

Background Access to essential medicines is a vital component of universal health coverage. The low availability of essential medicines for children (EMC) has led the World Health Organization (WHO) to issue a number of resolutions calling on member states on its improvement. But its global progress has been unclear. We aimed to systematically evaluate the progress of availability of EMC over the past decade across economic regions and countries. Methods We searched eight databases from inception to December 2021 and reference lists to identify included studies. Two reviewers independently conducted literature screening, data extraction and quality evaluation. This study was registered with PROSPERO, CRD42022314003. Results Overall, 22 cross-sectional studies covering 17 countries, 4 income groups were included. Globally, the average availability rates of EMC were 39.0% (95%CI: 35.5-42.5%) in 2009–2015 and 43.1% (95%CI: 40.1-46.2%) in 2016–2020. Based on the World Bank classification of economic regions, income was not proportional to availability. Nationally, the availability rate of EMC was reasonable and high (> 50%) in only 4 countries, and low or very low for the rest 13 countries. The availability rates of EMC in primary healthcare centers had increased, while that for other levels of hospitals slightly declined. The availability of original medicines decreased while that of generic medicines was stable. All drug categories had not achieved the high availability rate. Conclusion The availability rate of EMC was low globally, with slight increase in the last decade. Continuous monitoring and timely reporting of the availability of EMC are also needed to facilitate targets setting and inform relevant policy making. Supplementary Information The online version contains supplementary material available at 10.1186/s12889-023-15820-7.


Background
Access to essential medicines is a vital component to the fulfilment of the right to the highest attainable standard of health [1]. As one of the Sustainable Development Goals of the United Nations, access to safe, effective, quality, and affordable essential medicines is important to health coverage for children by 2030 [2]. In the last ten years, to improve availability of essential medicines for children (EMC), the World Health Organization (WHO) has issued some resolutions calling on member states to focus on it and also has regularly updated the model list of EMC. [3][4][5] The first Essential Medicines List for Children was published in 2007. The current versions, updated in September 2021, are the 8th Essential Medicines List for Children (EMLc). [6].
To promote and standardize investigation on the availability, price, and affordability of essential medicines, WHO and Health Action International (HAI) have developed "Measuring medicine prices, availability, affordability and price components" in 2003. [5] Previous studies found that the availability and affordability of EMC was generally low in low-and middle-income countries (LMCs). [7,8] However, it is unclear whether there is a progress on the availability of EMC over the last decade globally, to what extent, and its variations among countries. Therefore, this systematic review and meta-analysis was conducted to examine the global trend on the availability of EMC, its variation among economic regions and countries, to provide benchmarking and inform evidence-based health policy-making.

Search strategy and selection criteria
We conducted a systematic review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. [9] The protocol of this systematic review was registered on PROS-PERO, CRD42022314003.
We systematically searched PubMed, Embase, the Cochrane Library, Web of Science, CNKI, China Biology Medicine (CBM) from inception to December 2021. We also searched websites of WHO and Health Action International, reference lists of included studies, and published reviews for more eligible studies. The search strategy included a combination of medical subject headings and free text terms for ("child*" or "pediatri*") and ("essential medic*" or "essential drug*") and was adapted for each database.
The studies were included if they investigated medical institutions or pharmacies for availability of EMC, including vaccines, and they were designed as cross-sectional study. The studies published in all languages were included. Studies were excluded if they were: editorials, conference abstracts, and any unobtainable full-texts.
Two reviewers (SYQ and CZ) selected studies according to the inclusion and exclusion criteria independently. Discrepancies were resolved by discussion or consultation with a third reviewer.

Data extraction and analysis
Two reviewers (SYQ and CZ) independently extracted data using predefined excel data extraction form. The extracted data included the first author, published year, survey area, survey time, methods, the availability of EMC. Joanna Briggs Institute (JBI) was used to evaluate the risk of bias of included studies. [10] This tool has nine items and categorized each item with "yes", "unclear" and "no", and gave them "1" and "0" points. The total score > 6 points was considered as high quality, between 4 and 6 points was considered as medium quality; < 4 was considered as low quality. [11].

Statistical analysis
The availability rate was defined as the percentage of facilities with stock of the required on the day survey data were collected. We extracted the number of the facilities that had the medicine and total number of facilities on the day of data collection. Data were meta-analyzed using Stata 15.1 (Stata Corp, College Station, TX). The final availability rates were presented as weighted average and their 95% confidence intervals (CIs).
They were estimated by world, economic levels (The World Bank groups -high-income, upper-middleincome, lower-middle-income and low-income countries), country, types of medical institutions (ownership -public and private; institutional level -primary healthcare and other levels hospital), and medicine classification using the Anatomical Therapeutic Chemical (ATC) code. We also analyzed results across the six WHO geographical regions, but were excluded due to the small number of studies and lack of representativeness.
The United Nations Millennium Development Goals (MDGs) are goals that UN Member States have agreed to try to achieve by the year 2015. Therefore, we estimated the availability rate for 2009-2015 years and 2016-2020 years, respectively, to quantify its progress over time. [2].

Economic regional availability of essential medicines by world bank country-income groups
By World Bank classification of income countries, the availability rate of EMC in LMC countries was the highest (48.1%, 95%CI: 41.9-54.4%) from 2009 to 2015. (Fig. 2) From 2016 to 2020, the availability rate of EMC was highest in HIC countries (95.8%, 95%CI: 93.5-97.7%). (Figs. 3 and 4) As was shown in the Fig. 5, there was no proportional between income and availability rates, and the regularity    (Fig. 3) There were studies conducted in India, China and Ethiopia in both of the two periods, and the availability rates of India and China increased by 2.6% and 2.0%, respectively. While the availability rates of Ethiopia decreased by 1.0%. (Fig. 6)

Availability of essential medicines for children by types of medical institutions
By ownership of medical institutions, the availability rate of EMC was 31.

Availability of medicine categories according to ATC
The results showed that there were significant differences in the availability rate of different categories of medicines.

Discussion
In this study, we comprehensively assessed the global availability of EMC from 2009 to 2020. The global availability rate of EMC in 2016-2020 (43.1%) increased, but not significantly compared with that in 2009-2015 (39.0%). By World Bank country income groups, income was not proportional to availability. By country, the availability rate varied substantially across countries in both 2009-2015 and 2016-2020. By ownership of medical institutions, the availability rate of EMC in public medical institutions was slightly lower than that of in private medical institutions in both 2009-2015 and 2016-2020. By level of medical institutions, the availability rates in primary healthcare centers had increased, while that for higher level hospitals had slightly declined. In terms of generic and original medicines, the availability of generic medicines was stable but that of original medicines decreased. The availability rate of EMC also varied largely for drug categories, from low to reasonable availability. Furthermore, our findings showed that access to essential medicines for major noncommunicable diseases hadn't reached the target of 80% [35]. As far as we know, this study is the first to quantify the availability rate of EMC and its trend with time globally, nationally and regionally for the last decade. The global availability rate of EMC had slightly increased in the last decade. There was substantial variability between countries and economic regions. However, there was a relatively limited number of economic regions and countries involved in surveys, indicating substantial research gaps globally, regionally and nationally. Caution is also needed in generalizing the availability rate of a country as representative of the whole economic region. We recommend that countries continue to conduct surveys on the availability of EMC to fill the current lack of data.
In terms of relationship of availability and income, there are several possible reasons for lower levels of EMCs in UMCs or than LMCs. Firstly, LMCs may focus on essential medicines from the WHO EMLc (covered by the survey); that UMCs often use other branded/ combination medicines not on the EML, so their score were low on WHO EMLc medicines. Besides, an article surveyed availability of essential cytotoxic medicines for treating children with cancers in 50 countries also concluded that income is not proportional to availability, and analyzed that was attributed to a narrower range of clinical protocols or manage lower stages of disease and therefore rely on a smaller number of agents in LICs. [27] Finally, national policies may facilitate access to medicines. Sri Lanka, though a lower-middle income country, had a high availability rate, which might be related to its national policies. It had established a national pharmaceutical company to distribute drugs to the public and private sectors, and the public sector provided free essential medicines to residents. [13] China, a upper-middle income country, had officially implemented the essential medicine system since 2009, but there was a low availability of EMC, due to the high cost of development and low pricing, leading to lose motivation for enterprises. Secondly, the implementation of the policy of China's list of essential medicine for children has been slow. Thirdly, the varieties of essential medicines procured by medical institutions at all levels are not fully consistent with EMLc. Therefore, China also needs to further implement the policy, formulate incentive measures, and encourage enterprise to invest. [24] In addition, there are other countries that increased availability rate through reforms. Uganda had improved access to cancer medicines from 28.5 to 85.8% within a space of 2 fiscal years by two main factors: first, redefining cancer medicines as highly specialized drugs and legalizing an independent procurement in 2016; second, by streamlining the procurement and supply chain to eliminate or minimize the role of middlemen. [36] The Mexico government had implemented pooled procurement to improve availability. [37].
We also found that the availability rate of EMC in public medical institutions was slightly lower than that of in private medical institutions in both 2009-2015 and 2016-2020, which were similar to a systematic review of essential medicines for asthma. [38] While private sector is important to facilitate access to EMC, more measures are warranted to increase the availability of EMC in public health institutions, for they are the main providers of public health and primary healthcare services and pivotal in pursuing universal health coverage in many countries.
The gap between the availability of original and generic medicines is growing. On the one hand, countries encourage the research and development of generic medicines, promoting higher availability of generic medicines. On the other hand, they also need to encourage the research and development of original medicines, establish an early reliable patent warning system of medicines, and protect intellectual property rights. [39] Therefore, introducing incentives for production infrastructure and suppliers improvements is crucial, including financial incentives and policy support to address the economic causes of manufacturing issues. [40].
There is also a large variation of availability rates among medicine categories, ranging from 24.1 to 87.1% in 2009-2020. Among them, the anti-neoplastic and immunomodulating agents had the highest availability rate. Among the survey institutions with an availability rate of > 80%, the type of institution and the level of the medical institution were not clearly indicated in part of studies due to the small number of available studies. Therefore, it could not be explained from this perspective. Analyzing income levels > 80% of the countries, high-income and upper-middle-income countries accounted for more than half, similar to the findings of a global cross-sectional survey of 82 countries. [41] Although availability was also high in low-income countries, the data are all from one study with only eight institutions, making them unrepresentative. Policy support may be a possible reason for the increased availability, indicating the important and positive role of pharmaceutical policies in improving the availability of EMC. A survey of 37 European countries from 2016 to 2018 showed that children and adolescents with cancer still experienced lack of access to essential medicines. [42] To address this issue, the Expert Committee in developing the WHO EMLc included several additional supportive care agents for cancer and the Essential Medicines Working Group of the International Society of Pediatric Oncology (SIOP) proposed a list of anti-neoplastic drugs, [43]- [44]. In our study, carboplatin, methotrexate, ifosfamide, cyclophosphamide, etoposide, vincristine, doxorubicin, dactinomycin, cytarabine, cisplatin are the top ten anticancer drugs in the research frequency and belong to the SIOP core catalog.
There were several limitations of this study. Firstly, limited by the number of studies included, economic regional and national availability rates estimated may not be stable or accurate. Secondly, limited by the quality of some studies, such as the medicines selected for investigation inconsistent, there were some bias in the results. However, this study provides a comprehensive map of the global, regional and national availability of EMC and that by medical institutions, original and generic medicines and medicine categories from 2009 to 2015 using evidence-based methodology, and there is nothing better for the moment, which shed light on the needs in practice and in research in the future.

Conclusions
In conclusion, the availability rate of EMC is still low globally, with only marginal increase in the last decade and only a few categories reached reasonable and high availability (> 50%). Income is not always proportional to EMC. No country had reached the target of high availability rate (> 80%) of EMC. It is suggested to carry out relevant studies to fill the data gaps in the children's essential medicine survey and improve the accessibility of children's medicines. Substantial efforts from all stakeholders are warranted to improve the availability of EMC globally and its equity among countries. Continuous monitoring and timely reporting of the availability of EMC are also needed to facilitate targets setting and inform relevant policy making.