Article Text

Original research
Socioeconomic inequalities in underweight children: a cross-sectional analysis of trends in West Africa over two decades
  1. Habila Adamou1,2,
  2. Gregoire Naba3,
  3. Hamidou Koné4
  1. 1Center for Research in Regional Planning and Development (CRAD), Université Laval, Quebec, Quebec, Canada
  2. 2Evaluation Platform on Obesity Prevention, Institut Universitaire de Cardiologie et de Pneumologie de Quebec - Université Laval, Quebec, Quebec, Canada
  3. 3Bureau central du recensement, Institut national de la statistique, Niamey, Niger
  4. 4Institut de Formation et de Recherche Demographiques, Yaounde, Cameroon
  1. Correspondence to Habila Adamou; habila.adamou-djibo.1{at}ulaval.ca

Abstract

Objective To study trends in socioeconomic inequalities in underweight children in West Africa, and specifically to analyse the concentration index of underweight inequalities and measure inequalities in the risk of being malnourished by household wealth index.

Design Cross-sectional study.

Setting The study used 50 Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys conducted between 1999 and 2020 across 14 countries by the DHS and UNICEF.

Participants The study included 481 349 children under the age of 5 years.

Primary and secondary outcome measures The analysis used three variables: weight-for-age index, household wealth index and household residence. The inequality concentration index for underweight children and the relative risk of being underweight between 2000 and 2020 were calculated.

Results The prevalence of underweight in West Africa showed a downward trend from 2000 to 2020. Nonetheless, the prevalence of underweight children under 5 years of age is still very high in West Africa compared with other sub-Saharan African countries, and the sustainable development objective is yet to be achieved. There was a wide disparity among countries and significant socioeconomic inequalities in underweight children within countries. The proportions of underweight children were concentrated in poor households in all countries in West Africa and over all periods. Socioeconomic inequalities in underweight children were more significant in countries where the prevalence of underweight was low. These inequalities were more pronounced in urban areas in West Africa from 2000 to 2020.

Conclusions and relevance There is a high concentration of socioeconomic inequalities in underweight children in disadvantaged households in West Africa.

  • Nutrition
  • Health Equity
  • SOCIAL MEDICINE

Data availability statement

No data are available.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • This is an innovative study and the first in West Africa to investigate the concentration of socioeconomic inequalities in underweight children over two decades.

  • This study uses the household wealth index to calculate the underweight concentration index for 481 349 children under 5.

  • This study contributes to measuring the performance of West African countries in achieving the Sustainable Development Goals and identifies the populations most affected by socioeconomic inequalities.

  • This study uses a cross-sectional design, which provides a snapshot of data at specific points in time but limits the ability to establish causal relationships or assess changes over time more rigorously.

Introduction

The future of any society lies with its children, whose health, growth and development must be assured.1 Safeguarding health during childhood is necessary; compromised health at a young age can have consequences during adulthood.2 Following the Millennium Development Goals, the United Nations has developed the Sustainable Development Goals (SDGs), which focus on children’s health and reducing social inequalities in more ways than one. Like the rest of the world, African countries, particularly those in West Africa, have set targets for reducing child malnutrition and social inequalities among and within countries to be achieved by 2030.

For example, from 1990 to 2018, there has been overall progress in the fight against child mortality worldwide, with the mortality rate among children under 5 years of age declining by more than 50%3 to 390 deaths per 10 000 live births.3 However, it should be noted that this progress masks significant regional disparities. In 2018, 80% of under-5 deaths were concentrated in two regions alone: sub-Saharan Africa (54%) and South Asia (28%). This further complicates the achievement of the SDG rates in these two regions, especially in sub-Saharan Africa. The reduction of malnutrition is a significant challenge to reducing child mortality. According to the WHO,4 factors related to malnutrition contribute to about 45% of deaths in children under 5 years of age. Indeed, the risk of death is four times higher for a child with severe stunting and nine times higher for a child with severe wasting.5 Child nutritional status is thus strongly associated with under-5 mortality in developing countries,6 with a one-point decrease in the proportion of underweight reducing mortality by 1.7 points out of 1000.7 Inequalities in malnutrition exacerbate inequalities in mortality, so underweight children are at greater risk of infection, long-term health problems and death.8

The prevalence of malnutrition in children under 5 years is declining in countries of the South.9 10 However, the prevalence of underweight in sub-Saharan Africa remains very high, especially in West African countries.7 Africa and Asia alone carry the most significant burden of malnutrition, where more than 9 out of 10 children are stunted.7 More than nine out of ten children worldwide suffer from stunting, and more than nine out of ten10 suffer from wasting.9 10 The situation is of particular concern in Africa because the region has the highest rates of hunger in the world, and these rates continue to increase in almost all subregions.11 This phenomenon appears to undermine the achievement of the goals of reducing the prevalence of child stunting by 40% from its 2012 level and reducing inequalities within countries.11

Stunting and wasting declined across all wealth quintiles from 2000 to 2018,9 but inequalities persist among countries and wealth quintiles within countries.7 The objective of this study is to investigate trends in socioeconomic inequalities in underweight children in West Africa, and specifically to analyse the concentration index of underweight inequalities and measure inequalities in the risk of being underweight by household wealth index.

Data and methodologies

Data and sample

This study used national data from the Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS). Our study covered Burkina Faso, Benin, Côte d’Ivoire, Ghana, Guinea-Bissau, Gambia, Guinea, Liberia, Mali, Nigeria, Niger, Sierra Leone, Senegal and Togo. These surveys offer comprehensive insights into the health and nutritional status of women aged 15–49 and their children. They collect various information on individual, household and community characteristics. They facilitate the estimation of the nutritional status of children under 5 years of age using anthropometric measurements, specifically weight and height. Both surveys (DHS and MICS) have similar sample designs, questionnaires and collection methods. In addition, the data are comparable across countries and periods because both surveys are standardised in sampling, questionnaire and collection methods. Our study used explicitly the variables household wealth quintile, anthropometric weight-for-age index (underweight) and household residence, which are all operationalised, collected and constructed identically across countries, surveys (DHS and MICS) and periods.

Our target population was all children under 5 years of age at the time of the various surveys. The databases and the number of children by survey at the different periods are distributed as shown in table 1.

Table 1

Types of surveys and sample of children under 5

The weight-for-age index measures the nutritional status of a child. It is a combined index of malnutrition since low weight for age can be caused by wasting and stunting. This is a dichotomous variable with the modalities ‘1 - malnourished’ and ‘0 - normal’. Malnourished children are those who are underweight. For this purpose, we considered children with a weight-for-age index lower than 2 SD (−2SD) as having a malnourished state and those whose weight-for-age index was greater than or equal to 2 SD (−2SD) as having a normal nutritional state.

Analysis model

The household wealth index from the DHS and MICS identifies health inequalities. It is valuable for assessing the socioeconomic position of households in low-income countries where income measurement is difficult (DHS and UNICEF). Thus, the household wealth index was the primary analysis variable in this study, whether for calculating the concentration of inequalities index of underweight children or the relative risk (odds).

The concentration index of inequality is a relative measure that uses an approach analogous to the Gini index by ranking children according to the wealth index of their parents’ household on the x-axis and plotting the underweight status of children on the y-axis.12 Thus, if the concentration index is equal to 0, for example, in the case where each wealth quintile contains 20% of the underweight children,13 then there is no inequality in children being underweight related to the household wealth index. However, on the contrary, a harmful concentration index would mean that the percentage of underweight children is concentrated in the poorest households and the lowest wealth quintiles, and a favourable concentration index would indicate a concentration of the percentage of underweight children in the more affluent households and the top wealth quintiles. On the Stata V.17 software, we used the index package12 to calculate the Wagstaff inequality concentration index, which was formalised as follows:

Embedded Image

where C is the concentration index, y is underweight, u is the mean of underweight children and r is the fractional rank of the household wealth index.

Binary logistic regression calculates children’s relative risk of being underweight by household wealth index. This method estimates the risk or probability of an event occurring as a function of the independent variables. The dependent variable takes the modality ‘1’ when the event is realised (underweight child) and ‘o’ otherwise. Thus, logistic regression estimates the probability of a child being underweight. If p is the probability that the event under study (the underweight child) will occur, 1−p is the probability that this event does not occur. The logistic regression model allows us to put I=log (=7) in the following linear form: L=b0+b1x1+b2x2+…, where bpxp is the independent or classification variable and b0, b1, b2. and bp the regression coefficients of the model. An OR greater than 1 in a category indicates a higher probability that the child has a deficiency compared with the reference group. An OR of less than 1 means a lower likelihood that the child suffers from insufficiency in the category considered compared with the reference group.

Patient and public involvement

No patient is involved.

Results

Trends in underweight children in West Africa

Figure 1 shows a consistent decline in the underweight index of children in West Africa during the investigated period from 2000 to 2020. The available data indicate that the proportion of underweight children in West Africa decreased to 15% on average from 2015 to 2020. Notably, the decline was more significant in the eight countries that had a proportion of underweight children above 20% in 2000, except for Niger and Nigeria. In most countries, the decline in underweight was constant. Niger, Burkina Faso, Mali and Nigeria had the highest proportions of underweight children (more than 20% on average), in decreasing order, over the entire period from 2000 to 2020. In contrast, Senegal, Gambia, Liberia and Ghana had, in descending order, the lowest proportions of underweight children over the entire period from 2000 to 2020 (less than 15% on average).

Figure 1

Evolution of the prevalence of underweight in West Africa from 2000 to 2020. Source: Authors’ calculations based on DHS and UNICEF MICS data. DHS, Demographic and Health Surveys; MICS, Multiple Indicator Cluster Surveys.

According to available data, the two extremes are distributed by period as follows: 39.6% of children were underweight in Niger vs 14.2% of children in Senegal over the 2000–2005 period, and this proportion of children increased to 38.6% in Niger vs 13.9% in Ghana over the 2005–2010 period; 36.4% of children were underweight in Niger vs 11% of children in Ghana over the 2010–2015 period, and this proportion dropped to 21.6% in Nigeria vs 10.9% in Liberia over the 2015–2020 period.

The prevalence of underweight is decreasing in West Africa, but the decline varies from country to country and from 2000 to 2020.

Trends in socioeconomic inequalities in underweight children in West Africa

As shown in table 2, throughout the study period and across urban and rural settings in all countries, the inequality concentration index is globally negative and significant at p<0.01. This indicates that the proportion of underweight children was concentrated in the most socioeconomically disadvantaged wealth quintiles in West Africa. In other words, underweight children from poor households were more likely to experience underweight than those from better-off households in the 14 West African countries. On the other hand, the inequality concentration index of underweight according to household wealth varied among countries and across time. However, the inequality concentration index was decreasing in all countries. This suggests a decrease in underweight inequality among children according to the household wealth index in West Africa or a reduction in the underweight gap between the poorest and the wealthiest households, whether in urban or rural areas.

Table 2

Underweight inequalities concentration index by country

Trends in socioeconomic inequalities in underweight children at the national level

As shown in table 2, West African countries can be categorised into four groups based on the changes in underweight inequality from 2000 to 2020. Depending on available data, in the first group, the concentration index of underweight inequalities remained significant and often significant at the p<0.001 level (C>−0.10) throughout the study period.

Thus, in Côte D’Ivoire, the index fluctuated from −0.193 from 2000 to 2020. In Togo, the index varied from −0.197 to −0.21 from 2000 to 2020, respectively; in Ghana, the index ranged from −0.20 to −0.198; in Nigeria from −0.123 to −0.134; and in Senegal from −0.25 to −0.205. On the other hand, in the second group of countries, including Niger, Guinea-Bissau and Liberia, the inequality concentration index remained very low and often insignificant from 2000 to 2020. In the third group of countries, the evolution of underweight inequalities was more contrasted, such as in Burkina Faso, Benin, Gambia and Sierra Leone. In the fourth group, the inequality concentration index increased from 2010 to 2020, with inequalities increasing between households in these countries, notably Mali and Guinea Conakry. The analysis will therefore now focus on how inequalities in being underweight are distributed according to the area of residence of the households. National trends conceal major disparities between urban and rural areas.

Trends in socioeconomic inequalities in underweight children by household residence

Online supplemental table 1 shows that the socioeconomic inequalities in the underweight were more pronounced in urban areas than in rural areas in West Africa. In fact, after controlling for household residence, the concentration index of underweight inequalities in urban areas was higher than in rural areas, and the concentration index of underweight inequalities in urban areas became significant in most countries from 2000 to 2020. In addition, significant underweight inequalities at the p<0.001 threshold emerged in urban areas during the 2000–2005 period for countries such as Niger, Nigeria, Gambia, Benin, Burkina Faso or Guinea-Bissau, for which the inequality concentration index was not significant during this period. In the subgroup of countries where the inequality concentration index for underweight was elevated (Ghana and others, and now additionally Benin), the disparities in underweight inequalities concerning the socioeconomic position of households varied and the inequality concentration index for the underweight remained high in both urban and rural areas throughout the observation period from 2000 to 2020.

Our results show wide disparities in underweight in West Africa among countries and periods. Ours results show substantial inequalities in underweight between households in the lower wealth quintiles and those in the higher wealth quintiles, whether in urban or rural areas. Let us now analyse the inequalities in the risk of being underweight among children by household wealth quintile, which determines the socioeconomic position of households.

Trends in socioeconomic inequalities and the risk of a child being underweight in West Africa

Online supplemental table 2 shows that in all countries from 2000 to 2020 the household wealth index was strongly associated with being underweight (p<0.001 or p<0.005). Therefore, the socioeconomic position of the household is a determinant of underweight children. The risk of underweight children increased inversely with the household wealth index. All other things being equal, children from poor households were more likely to be underweight compared with children from affluent households, and this was true for all countries and all periods. Thus, on average, all other things being equal, a child from a very poor, poor, middle-income or wealthy household was twice as likely to be underweight as a child from a wealthy household from 2000 to 2020. This difference in risk of being underweight was even more significant between children from impoverished households and those from wealthy households.

Trends in inequality in the risk of being underweight varied across countries and periods. Notably, in Burkina Faso, Côte D’Ivoire, Ghana, Guinea-Bissau and Nigeria, the gap in the risk of being underweight between children from impoverished households and those from wealthy households decreased from 2000 to 2020. All else being equal, for example, over the 2000–2005, 2005–2010, 2010–2015 and 2015–2020 periods, respectively, children from the poorest household in Burkina Faso were 3.17, 1.71 and 2.46 times more likely to be underweight than children from the most affluent households (p<0.01), and similarly in Ghana the risk was 3.21, 2.40, 2.93 and 2.37 times more (p<0.01) for children from the poorest households.

In contrast, in the rest of the countries, the difference in the risk of being underweight between children from the poorest and wealthiest households increased. All else being equal, for example, over the 2000–2005, 2005–2010, 2010–2015 and 2015–2020 periods, respectively, children from the poorest households in Mali had 1.15, 1.81, 2.18 and 2.56 times the risk of being underweight than children from the wealthiest households, and in Sierra Leone 1.39, 2.10, 2.14 and 1.40 times the risk of being underweight than children from the most affluent households.

Trends in socioeconomic inequalities and risk of a child being underweight by household residence

Online supplemental table 2 shows that, after controlling for the risk of being underweight based on wealth index by household residence, three patterns emerged among the 14 West African countries. In the first group, inequalities in the risk of being underweight varied little by household residence. Nevertheless, the risk of being underweight remained two to three times higher among children from the poorest households than children from the wealthiest households, whether in urban or rural areas. All else being equal, for example, in Côte D’Ivoire, children from the poorest households were 2.47 and 3.40 times (p<0.001) more likely to be underweight than children from wealthy households in urban areas, as compared with rural areas where children were 2.75 and 3.78 times (p<0.001) more likely to be underweight, respectively, during the 2005–2010 and 2015–2020 periods. Overall, in Guinea-Bissau, Guinea Conakry and Niger, inequalities in the risk of being underweight between the poorest and the wealthiest households became insignificant in both urban and rural areas from 2000 to 2020.

Concerning the second trend within this group of countries, inequalities in the risk of being underweight between children from the poorest and the wealthiest households were more pronounced in urban than in rural areas. Although the differences in risk of being underweight were still significant and well to the disadvantage of children from the poorest households, all else being equal, this was the case, for example, in Ghana and Sierra Leone.

The third pattern includes countries where inequalities in the risk of being underweight between children from the poorest and wealthiest households were more remarkable in rural than in urban areas. However, significant inequalities remained in urban areas between children from the poorest and wealthiest households. All other things being equal, for example, this was the case in Burkina Faso, Benin, Gambia, Mali, Senegal and Togo.

Inequalities in the risk of being underweight persisted consistently in all periods, including urban and rural areas in Nigeria and Mali. All else being equal, extreme inequalities in the risk of being underweight among children from the poorest households were observed in Ghana, Gambia, Sierra Leone, Nigeria, Senegal and Togo. Indeed, children from the poorest households were 10.12 times (p<0.001) more likely to be underweight during the period from 2005 to 2010 in urban areas than children from the wealthiest households, 5.43 times (p<0.001) more likely to be underweight during the period from 2015 to 2020 in urban areas, and 5.72 times (p<0.001) more likely to be underweight in rural areas.

In Gambia, this risk was 6.5 higher (p<0.001) in rural areas from 2015 to 2020. In Sierra Leone, the risk was 4.1 higher (p<0.001) in urban areas from 2005 to 2010. In Nigeria, the risk was 4.66 higher (p<0.001) in rural areas from 2010 to 2015. In Senegal, the risk was 4.1 higher (p<0.001) in rural areas from 2015 to 2020. Lastly, in Togo, the risk was 12.3 higher (p<0.001) in urban areas during the 2005–2010 period and 5.99 higher (p<0.005) in rural areas during the 2015–2020 period.

Discussion

Trends in underweight children in West Africa

The mastery of regional trends in socioeconomic inequalities in malnutrition is essential for developing regional policy on the one hand. On the other, it allows countries to draw inspiration from policies implemented in countries with similar characteristics that have led to a decrease in malnutrition and a consequent reduction in socioeconomic inequalities in malnutrition. It is also a framework for monitoring malnutrition indicators towards achieving the SDGs to which West African countries have subscribed.

The underweight trends have declined in relative terms in all 14 West African countries from 2000 to 2020. This trend was observed for all malnutrition indicators (underweight, stunting and wasting) in West Africa.5 14–16 However, this decline is not enough and more is needed in most countries to reduce the growth gap by 40% from its 2012 rate by 2025.17 In addition, for an average of 15% of underweight children in the subregion in 2020. stunting rate is used to assess malnutrition in the Millennium Development Goals. There was a wide disparity between countries in the proportion of underweight children over all the observation periods, which implies a decline that is diversely distributed among countries, with significant decreases in some and shallow declines in others. Thus, six countries with a proportion of underweight children higher than 20% in 2000 reduced this proportion by more than five points in 2020 to come close to the countries with the lowest underweight rates in the subregion, Ghana and Liberia, in 2020. However, significant inequalities existed between countries and within countries from 2000 to 2020 related to the socioeconomic position of the households; a similar situation has been observed at the global level.18

Concentration of underweight children’s inequalities in West Africa

Without an income measure in West African surveys, the household wealth index, mainly classified into household wealth quintiles, is used to analyse socioeconomic inequalities in health indicators. In this logic, our study used the wealth quintile to analyse the concentration of inequalities in underweight children in West Africa from 2000 to 2020. Thus, our results highlight inequalities in being underweight in all countries and over all observation periods. Many studies on malnutrition in sub-Saharan Africa have found similar results. Consequently, malnutrition is concentrated in the lowest wealth quintiles and these inequalities persist over time.19–23 In a study on socioeconomic inequalities in developing countries, where 9 of the 14 countries in our study were included, similar inequality concentration indices were found for stunting.19 However, these inequalities varied between countries and periods, where they were greater overall in Senegal, Togo, Nigeria and Ghana than in the other countries. It was also found that the household’s place of residence during the observation period can attenuate or aggravate the inequality gaps between poor and rich households in several countries. However, overall, inequalities in being underweight were to the disadvantage of children from poor households and were more significant in urban than in rural areas.

Thus, in Niger, Benin and Burkina Faso, inequalities are significant in urban areas, whereas the concentration index of underweight inequalities remains very low at the national level and insignificant. Quantifying the individual risk of socioeconomic inequalities in the underweight, the subject of the last part of our study makes it possible to detect not only the existence of inequalities but also to measure them in terms of the socioeconomic position of the households.

Socioeconomic inequalities and risk of children being underweight in West Africa

The household wealth index determines the nutritional status of children in West Africa; the risk of malnutrition (underweight, stunting or wasting) varies according to the household wealth quintile.21 22 24 25 Thus, our results abound in this sense where the risk of being underweight in children followed the household socioeconomic gradient; indeed, the lower the household wealth quintile, the more children were exposed to a greater risk of being underweight in all countries and over all periods. The differences in socioeconomic inequalities in the risk of being underweight were even more significant between children from the poorest and the wealthiest households.26 Thus, on average, children from the poorest households in the subregion were twice as likely to be underweight as those from the richest households over all observation periods. However, these socioeconomic inequalities in the risk of malnutrition also varied between countries, household environments and over time.

Socioeconomic inequalities in the risk of being underweight between children from the poorest and wealthiest households from 2000 to 2020 were most significant in Senegal, Nigeria, Ghana, Togo, Sierra Leone and Gambia. In some countries, the risk among children from the poorest households was 10 times higher than children from the richest households. However, there was no monolithic pattern of socioeconomic inequalities in the risk of being underweight by household residents in West Africa. Three groups of countries emerged as the first group with little or no variation in risk between children from poor and rich households in urban and rural areas; in the second group, however, risk differences became more pronounced in urban areas, and in the third group the differences between children from poor and rich households increased more in rural areas.

Limitations

We used a cross-sectional design, which provides a snapshot of data at specific points in time. This design limits the ability to establish causal relationships or assess changes over time more rigorously. This study relied on data from the DHS and MICS, which are valuable surveys but may be subject to recall bias and measurement errors, impacting the accuracy of the data. Cross-sectional studies can be vulnerable to selection bias, as the participants may not represent the entire population. Sampling methods and survey participation rates can introduce biases. Despite these limitations, the results of this study can serve as a basis for assessing the performance of West African countries in achieving SDGs.

Conclusion

The prevalence of underweight in West Africa showed a downward trend from 2000 to 2020. However, the rate of underweight children under 5 years of age is still very high in the subregion compared with other countries in sub-Saharan Africa, and the SDG is yet to be achieved. There was a wide disparity between countries and significant socioeconomic inequalities in underweight children within countries. The proportions of underweight children were concentrated in poor households in all countries and over all periods in West Africa. Thus, socioeconomic inequalities in the risk of being underweight disadvantaged children from poor households. Children’s underweight socioeconomic inequalities were evident across both urban and rural areas in West Africa from 2000 to 2020, with a consistent trend: the lower the household socioeconomic status, the higher the risk of children being underweight. On the one hand, socioeconomic inequalities in underweight children, and on the other hand inequalities in the risk of being underweight, were greater in countries where the prevalence of underweight was low, except in Nigeria where underweight remained high with high socioeconomic inequalities.

To achieve SDG 2 in the subregion, special attention must be paid to children from disadvantaged households. However, actions must consider country-specific factors in the subregion because being underweight and its socioeconomic trends vary significantly despite social, political, economic and cultural similarities.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Acknowledgments

We are grateful to USAID via the Demographic Health Surveys programme and UNICEF for facilitating access to the DHS and MICS data. We thank Cheik Diallo Bountou, a doctoral student at Abdou Moumouni University in Niamey, for his valuable advice.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Twitter @Habilacogite

  • Contributors HA is responsible for the overall content. HA introduced the study concept. HA, GN and HK formulated and devised the study design. HA oversaw the data acquisition. GN authored the introduction, while HA authored the methodology section. Subsequently, HA conducted the analysis and interpretation of the results. HA, GN and HK engaged in the discussions regarding the findings. HA composed the conclusion, and GN composed the references. All authors thoroughly reviewed the manuscript. The final manuscript received scrutiny and approval from all authors.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.