Essential medicines for mental disorders: comparison of 121 national lists with WHO recommendations

Abstract

Background

To compare the medicines for mental disorders included in national essential medicines lists with the World Health Organization (WHO) essential medicines list and assess the extent to which economic status and WHO Region account for the differences.

Methods

We searched WHO repository and government sites for national essential medicines lists and we abstracted medicines for mental disorders. We calculated the proportion of WHO essential medicines included, the total number of differences (counting both additions and deletions) between national and WHO model list and the proportion of lists including one second-generation oral antipsychotic plus one new-generation antidepressant. Non-parametric statistics was used to investigate whether these indicators were dependent on economic status and WHO Region.

Results

Amongst the 121 identified national lists, the total number of medicines for mental disorders ranged from 2 to 63 (median: 18; IQR: 14 to 25). The median proportion of WHO essential medicines for mental disorders included was 86% (IQR: 71–93%), with 16 countries (13%, 95% CI 7.75–20.5%) including all WHO essential medicines, while the median number of differences with the WHO EML was 11 (IQR: 7 to 15). Country economic level was positively associated with both the proportion of WHO essential medicines included (Spearman's rho = 0.417, p < 0.001) and the number of differences (Spearman's rho = 0.345, p < 0.001), implying that countries with higher income level included more WHO essential medicines, but also more additional medicines. Significant differences were observed in relation to WHO Region, with the African and Western Pacific Region showing the lowest proportions of WHO essential medicines, and the European Region showing the highest median number of differences. Overall, 88 national lists (73%, 95% CI 63–80%) included at least one second-generation oral antipsychotic and new-generation antidepressant, with differences by income level and WHO Region.

Conclusions

The degree of alignment of national lists with the WHO model list is substantial, but there are considerable differences in relation to economic status and WHO Region. These findings may help decision-makers to identify opportunities to improve national lists, aiming to increase access to essential medicines for mental disorders.

Background

Since 1977 an essential medicines list (EML) has been drawn up by the World Health Organization (WHO) to provide countries with a guide in their own choices for national EMLs [1]. The WHO Model List of Essential Medicines consists of a core list of medicines considered essential for basic health-care needs, and a complementary list of additional essential medicines for which specialist diagnostic and/or monitoring facilities are required [2, 3]. Essential medicines are expected to be available for free or at affordable prices to those in need [4]. Alignment of national EMLs to the WHO Model List can facilitate access to essential medicines, particularly in the public sector and in low resource settings [5].

In the field of mental healthcare, access to evidence-based treatments remains a huge global challenge [6]. More than 75% of people in low- and middle-income countries (LMICs) does not receive any mental healthcare, and do not access essential medicines [7]. In these settings, inclusion of essential medicines for mental disorders on national EMLs has been suggested as a first, crucial step to improve global access to mental healthcare and reduce such a huge treatment gap [8]. In high-income settings, the essential medicines concept may reduce the inclusion of inappropriate medicines, ensuring appropriate medicine selection and, when combined with broader policies, better medicine access [9]. So far, whether countries are guided by the WHO EML in selecting medicines for mental disorders is largely unknown, as well as whether geographical area and income level affect national choices [10].

Against this background, the present study compared the medicines for mental disorders included in national EMLs with those included in the WHO EML, aiming to determine the degree of alignment of country choices with WHO recommendations, and whether economic status and WHO Region accounted for the differences.

Methods

Selection of medicines for mental disorders included in the WHO Model List of Essential Medicines

The 21st WHO Model List of Essential Medicines was accessed to identify essential medicines for mental health conditions. Medicines listed in the following categories of the section on mental and behavioral disorders were included: medicines used in psychotic disorders, medicines used in mood disorders (including depressive and bipolar disorders), medicines for anxiety disorders and medicines used for obsessive compulsive disorders. For each mental health condition, the listed essential medicines were extracted, recording whether they were included as individual medicines or as representatives of a specific pharmacological class. In the latter case, the WHO Model List of Essential Medicines includes an accompanying ‘square box’ symbol [3]. For each medicine, the formulation recommended by WHO was also recorded, differentiating between oral, intramuscular, and long-acting formulations.

Selection of medicines for mental disorders included in National Essential Medicines Lists and country characteristics

National EMLs were accessed from the WHO repository of National Medicines List/Formulary/Standard Treatment Guidelines. From each national EML, we abstracted all medicines for mental disorders, and we recorded which of these were also included in the WHO Model List of Essential Medicines [11]. Official country government webpages were additionally screened to check for the presence of updated national EML versions not stored in the WHO repository. When more than one national EML was found for the same country, the most recent was considered.

For identified countries with national EMLs, we collected information on WHO Region, population size, and gross domestic product (GDP) per capita. Data on WHO Region was obtained from the WHO Global Health Observatory [12], while data on country population and GDP per capita were extracted from the Central Intelligence Agency’s World Factbook [13]. World Bank criteria were used to group countries according to their income level [14].

Data analysis

In order to compare the EMLs of each country with the WHO Model List of Essential Medicines, the following indicators were calculated, in line with the methodology described by Taglione and colleagues and Persaud and colleagues [9, 15]: proportion of WHO essential medicines for mental disorders included in each national EML (for WHO essential medicines with a square box, any medicines of the corresponding pharmacological class were considered, using the Anatomical Therapeutic Chemical (ATC) codes as reference [16]); difference score: number of medicines on national EML but not the WHO Model List of Essential Medicines plus number of medicines on the WHO Model List of Essential Medicines but not on national EML. We additionally calculated the number of national EMLs including (a) at least one second-generation oral antipsychotic, and (b) at least one new-generation antidepressant, as an indicator of inclusion of medicines recently added to the WHO Model List of Essential Medicines. Using the WHO ATC classification, the following medicines were considered second-generation oral antipsychotics: amisulpride, aripiprazole, asenapine, brexiprazole, cariprazine, clozapine, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine, risperidone, sertindole, zotepine. The following medicines were considered new-generation antidepressants: bupropion, citalopram, desvenlafaxine, duloxetine, escitalopram, fluoxetine, fluvoxamine, mirtazapine, paroxetine, reboxetine, sertraline, venlafaxine, vortioxetine.

For descriptive data, medians with interquartile ranges (IQRs) were calculated for continuous variables, and proportions with 95% confidence intervals (95% CI) for categorical variables. Spearman's rho was used to determine whether continuous variables were associated with country income level, expressed as GDP per capita. For binary variables, two-sample Wilcoxon rank-sum (Mann–Whitney) test was used to investigate their association with GDP per capita. Kruskal-Wallis test was employed to assess whether continuous variables differed in relation to WHO Region (AFR = African Region, EMR = Eastern Mediterranean Region, EUR = European Region, AMR = Region of the Americas, SEAR = South-Est Asian Region, WPR = Western Pacific Region), while Pearson’s chi squared was used to investigate the association between binary variables and WHO Region.

Results

We identified 121 national EMLs posted on the WHO repository and on government sites. The total number of medicines for mental disorders extracted from the WHO model list of essential medicines was 14: chlorpromazine oral formulation, chlorpromazine injection, fluphenazine injection (decanoate or enantate), haloperidol oral formulation, haloperidol injection, risperidone oral formulation, clozapine, amitriptyline oral formulation, fluoxetine, carbamazepine, lithium carbonate, valproic acid, diazepam, clomipramine. The total number of medicines for mental disorders on each country’s list ranged from 2 to 63 (median: 18; IQR: 14 to 25). Table 1 (page 12 of the manuscript) presents the characteristics of the included countries and national EMLs. The median proportion of WHO essential medicines for mental disorders included in national EMLs was 86% (IQR: 71–93%), with 16 countries (13%, 95% CI 7.75–20.5%) including all WHO essential medicines, while the median number of differences with the WHO Model List of Essential Medicines was 11 (IQR: 7 to 15). Figure 1 shows that GDP was positively associated with both the proportion of WHO essential medicines included (Spearman's rho = 0.417, p < 0.001) and the number of differences (Spearman's rho = 0.345, p < 0.001), implying that countries with higher GDP included more WHO essential medicines, but also more additional medicines not included in the WHO Model List of Essential Medicines. However, in low-income countries the median proportion of WHO essential medicines included in national EMLs was 79% (IQR: 71–86%), with only five countries below 70%. Both the proportion of WHO essential medicines included in national EMLs (Kruskal-Wallis 17.299, p = 0.004) and the number of differences (Kruskal-Wallis 10.459, p = 0.063) differed by WHO Region, with the African and Western Pacific Region showing the lowest proportions of WHO essential medicines, and the European Region showing the highest median number of differences.

Table 1 Characteristics of 121 countries with national essential medicines lists

Fig. 1

Scatterplot of proportion of WHO essential medicines included in national EMLs(a)and difference score(b)in relation to countries’ gross domestic product. Legend. The size of the circles represents the country’s total population. GDP = gross domestic product; EML = essential medicines list

Interestingly, of the 16 countries including all WHO essential medicines for mental disorders, three countries included less than five additional medicines, namely Indonesia (three additional medicines), Lebanon (three), and Pakistan (one), resulting those with the highest alignment with the WHO list.

Overall, 88 national EMLs (73%, 95% CI 63–80%) included at least one second-generation oral antipsychotic and new-generation antidepressant, with differences by GDP (median GDP of countries with at least one second-generation oral antipsychotic and new-generation antidepressant: 17,139 US dollars [IQR 5,206 to 17,139] versus all other countries: 3,395 US dollars [IQR 2,227 to 5,535], two-sample Wilcoxon rank-sum z = -4.830, p < 0.001). The distribution of countries with at least one second-generation oral antipsychotic and new-generation antidepressant differed by WHO Region (Pearson chi2 12.507, p = 0.028), with the African and Western Pacific Region showing the highest rates of counties non including any newer medicines recommended by WHO (Fig. 2).

Fig. 2

Distribution of national EMLs including at least one essential second-generation antipsychotic and new-generation antidepressant by WHO Region. Legend. AFR = African Region, EMR = Eastern Mediterranean Region, EUR = European Region, AMR = Region of the Americas, SEAR = South-Est Asian Region, WPR = Western Pacific Region

All WHO essential medicines for mental disorders were included in more than two thirds of national EMLs (Table 2), with the exception of clomipramine and clozapine, included in 50% and 45% of national EML respectively.

Table 2 Most common medicines for mental disorders included in national EMLs

Discussion

The degree of alignment of national EMLs with the WHO Model List of Essential Medicines is substantial, implying that most national EMLs include the majority of WHO essential medicines for mental disorders. Alignment was found to be associated with income level: higher-income countries showed higher alignment, but also considered essential a number of additional medicines not included in the WHO Model List of Essential Medicines, while lower-income countries included less additional medicines, with quite high alignment with the WHO Model List of Essential Medicines, with few exceptions. Exceptions were more often geographically distributed in the African and Western Pacific WHO Region. By contrast, countries diluting the essential medicine concept by considering essential additional medicines not included in the WHO Model List of Essential Medicines were more often located in the European WHO Region. We also documented the capacity of national EMLs of being regularly updated, based on revisions to the WHO Model List of Essential Medicines. Globally, one third of countries did not include any second-generation antipsychotic and new-generation antidepressants, therefore showing low capacity of regularly updating the list, with countries belonging to the African and Western Pacific WHO Region reaching proportions of 40% and 50%, respectively. Previous studies evaluating the choice of essential medicines for other chronic noncommunicable diseases such as oncological and cardiovascular diseases revealed gaps in the appropriate selection of medicines and in the updating process of the lists [17, 18]. Moreover, as reported in some recent studies, a significant percentage of medicines withdrawn from the market at a national or even global level is still included in national EMLs, confirming the lack of proper updating [9, 19].

The present analysis has some limitations that should be considered in interpreting the data. First, of the 192 WHO member states, the WHO repository included medicine selection data for 121 countries. Considering that countries that have a national EML posted into the WHO repository may be more sensitive to the general challenge of increasing access to essential medicines, the alignment of national EMLs with WHO recommendations might have been overestimated [9, 15]. A second consideration is that countries that do not have a national EML should not be considered without any medicines for mental disorders. These countries might have followed another selection process leading to the inclusion of medicines that were considered essential for their specific geographical and environmental context. Moreover, many of the high-income countries may not have a NEML but use their positive reimbursement list as such. A third limitation is that the present study was only focused on essential medicines selection, while the other components of the access framework, namely availability, affordability and rational use, were not investigated [6].

Despite these limitations, these findings have important policy implications. For countries including several medicines in addition to the WHO Model List of Essential Medicines for mental disorders, decision-makers may want to re-examine whether some of these medicines should be removed aiming to limit the concept of being essential only to a smaller selected group. This might help to better focus national strategies to increase access to this group of medicines, without losing resources for logistic infrastructures supporting availability of other medicines that may not represent a public health priority. In addition, countries might consider conducting national or sub-national medicine access surveys, aiming to ascertain whether medicines on shorter lists are more likely to be available and affordable to the end users. The WHO and Health Action International (HAI) developed a standardized methodology to conduct such surveys [20], and regularly update the WHO/HAI global database, a repository of results of national and sub-national medicine access surveys.

Another important implication comes from the finding of high alignment with WHO recommendations in geographical areas with limited resources, as it shows that it is possible to follow WHO recommendations with a limited health care expenditure. However, in some countries of the African and Western Pacific Region, alignment was found to be still very low. In these countries decision-makers may want to consider the example of other countries with similar economic development to revise their national selection process, aiming to include and increase access to a very selected number of medicines for mental disorders. Decision-makers should consider that access to psychotropic medicines may be considered a proxy of access to mental healthcare, and increasing access to essential medicines for mental disorders may give the chance for a transformative improvement of the whole mental healthcare system, offering a unique opportunity for re-engagement in society by people suffering from mental disorders. By working at all levels of the health system, it may be possible to offer this essential component of mental health care to all who can benefit.

Overall, this study provides evidence on the global relevance of the WHO Model List of Essential Medicines as a reference standard for the pharmacological treatment of mental disorders. This evidence should urge WHO to try to keep this tool regularly updated. Over the last ten years the WHO has made a tremendous effort to produce, and regularly update, a number of evidence-based tools in the area of mental health, including recommendations [21,22,23,24,25,26], evidence-based intervention guides [27], and related implementation and operational manuals. These tools include up-to-date recommendations on selection and rational use of psychotropic medicines that not always fully match with the essential psychotropic medicines included in the WHO Model List of Essential Medicines. Some essential psychotropic medicines may no longer be essential, as they were included in the list more than 40 years ago, when the first WHO Model List of Essential Medicines was published. Aligning the WHO Model List of Essential Medicines with existing WHO recommendations and tools, and with current best evidence, would probably induce more countries to optimize adherence to the WHO list.