Achieving Universal Health Coverage (UHC): Dominance analysis across 183 countries highlights importance of strengthening health workforce.

BACKGROUND: Despite increasing political will to achieve Universal Health Coverage (UHC), there is a paucity of empiric data describing what health system indicators are useful surrogates of country-level progress towards UHC. We sought to determine what public health interventions were useful tracers of country-level UHC progress.
METHODS: Across 183 countries we evaluated the extent to which 16 service delivery indicators explained variability in the UHC Service Coverage Index, (UHC SCI) a WHO-validated indicator of country-level health coverage. Dominance analyses, stratifying countries by World Bank income criteria, were used to determine which indicators were most important in in predicting UHC SCI scores.
FINDINGS: Health workforce density ranked first overall, provision of basic sanitation and access to clean water ranked second, and provision of basic antenatal services ranked third. In analysis stratified by World Bank income criteria, health workforce density ranked first in Lower Middle Income-Countries (LMICs) (n = 45) and third in Upper Middle Income-Countries (UMICs) (n = 51).
CONCLUSIONS: While each country will have a different approach to achieving UHC, strengthening the health workforce will need to be a key priority if they are to be successful in achieving UHC.

Introduction

Universal Health Coverage (UHC) is a central target of the 2030 Sustainable Development Goals (SDG).[1] While there are numerous proposed strategies for achieving UHC[2] the objectives of UHC are the same, regardless of approach: improving access to health services, improving the health of individuals covered and providing financial risk protection.[3]

Recently, World Health Organization (WHO) and the World Bank proposed the use of a UHC Service Coverage Index (UHC SCI) tool to assess country-coverage with a range of different interventions.[3] This tool summarizes country-level coverage of 16 essential service indices across four core domains; (a) reproductive, maternal, newborn and child health, (b) infectious diseases, (c) non-communicable diseases and (d) service capacity and access, among the general population and the most disadvantaged populations.[3] The UHC SCI is unique insofar as it provides a means for assessing the breadth of essential services being offered by individual countries. Furthermore, it provides a simple and standardized summary of a variety of complex and heterogenous service parameters. While the index is limited by the quality of country-specific data on which it is based and the fact that much of the data has been collected asynchronously,[4] it offers valuable service-oriented insights into the implementation of UHC on a global scale.

In a separate analysis, we investigated whether country-level tuberculosis treatment coverage could serve as a useful surrogate for measuring progress towards UHC.[4] In this brief report, we describe a surprising finding relating to the importance of other indicators in achieving UHC across diverse country settings.

Methods

The rationale and methodology for how UHC SCI is derived has been described in depth elsewhere.[3] Briefly, the index is constructed from geometric means of 16 tracer indicators which measure: family planning programs, pregnancy and delivery care (measured as four or more visits to antenatal care during pregnancy), child immunization, coverage of pediatric services, HIV treatment services, TB treatment coverage, malaria prevention interventions, basic and water sanitation services, cardiovascular disease treatment coverage, cancer and diabetes screening capability, tobacco control measures, access to hospital services, health care worker density (measured as the number of physicians, surgeons and psychiatrists per person in each country), health security (measured based on International Health regulations core capacity index) and an indicator of access to essential medicines. Each indicator is calculated based on publicly reported data and a standardized approach applied across all countries. For each country, all 16 indices were aggregated and reported as a score, measured on a scale of 0–100%, with 100% representing achievement of universal health coverage.

Notably, the hospital bed density and health worker density indicators have a lower bound of 0 but do not have a clear optimal level of maximum. For these variables, a threshold value was selected based on observed minimum values across high-income Organization for Economic Co-operation and Development (OECD) countries. Countries with values above the thresholds for these indicators where held at 100 and those below were linearly rescaled between 0 and 100.

To determine the relative importance of each indicator to the overall UHC SCI score we performed a ‘dominance analysis’, using the epsilon methodology.[5] This statistical approach relies on estimating the R2 values of all possible combinations of predictors and measures the relative importance by doing pairwise comparisons of all predictors in the model as they relate to an outcome variable, such as the composite score evaluated in this analysis. [6] The relative importance of each indicator was determined by the size of the effect without any inference on the relative ‘importance’ of the other variables. Dominance analysis is well suited to answering questions of ‘importance’ among a set of variables such as this, especially since multi-collinearity would undermine use of traditional multiple regression approaches.[7] Dominance analysis was performed for all countries, collectively, and then separately for countries within strata of World Bank income categories; high income countries (HIC), upper middle-income countries (UMICs), lower middle-income countries (LMICs) and low-income countries (LICs). Indicators for malaria, cancer screening coverage, and essential medications, were not included in this analysis due to lack of consistent reporting for the majority of countries. All analyses were conducted using Stata version 14 (College Station, TX).

Results

Across 183 countries, the median service coverage index was 65 (Interquartile range [IQR]: 48, 75), with 44 countries having UHC SCI scores greater than 75 of which 35 were classified as HICs. There were 45 countries in the lowest quartile, of which 40 were classified as LICs.

The results of the dominance analysis assessing all 183 countries, collectively, indicated the most dominant interventions were density of the health workforce (ranked first), provision of basic and water sanitation services (ranked second) and provision of pregnancy and delivery care (ranked third) (Table 1). When stratified by World Bank income categories, dominance analysis ranked health workforce density fifth among LICs (n-34), first among LMICs (n = 45), third among UMICs (n = 51) and eleventh among HICs (n = 53). Across World Bank income categories, access to hospital services ranked first in LICs, family planning first in UMICs and provision of HIV treatment first in HICs.

Table 1

Dominance analysis rankings service coverage indicators based as predictors of UHC SCI score, stratified by World Bank country ranking.

	All countries (n = 183)		LICs (n = 34)		LMICs (n = 45)		UMICs (n = 51)		HICs (n = 53)
	Std. Dominance Coefficient	Rank	Std. Dominance Coefficient	Rank	Std. Dominance Coefficient	Rank	Std. Dominance Coefficient	Rank	Std. Dominance Coefficient	Rank
Family planninga	0.09	4	0.084	6	0.12	3	0.15	1	0.10	4
Pregnancy and delivery careb	0.1	3	0.05	9	0.09	4	0.04	10	0.08	5
Child immunizationc	0.06	11	0.08	7	0.04	9	0.08	7	0.01	13
Child treatmentd	0.08	7	0.05	11	0.06	7	0.3	11	0.04	9
HIV treatmente	0.06	10	0.01	12	0.04	10	0.11	4	0.21	1
Tuberculosis effective treatmentf	0.06	9	0.11	3	0.08	5	0.01	5	0.06	7
Water and sanitationg	0.12	2	0.14	2	0.16	2	0.05	9	0.03	12
Prevention of cardiovascular diseaseh	0.07	8	0.012	4	0.03	12	0.09	6	0.14	2
Management of diabetesi	0.01	13	0.01	13	0.02	13	0.02	12	0.07	6
Tobacco controlj	0.05	12	0.05	10	0.03	11	0.07	8	0.12	3
Hospital accessk	0.09	5	0.15	1	0.05	8	0.01	13	0.03	10
Health care worker densityl	0.13	1	0.11	5	0.2	1	0.13	3	0.03	11
Health securitym	0.09	6	0.06	8	0.08	6	0.15	2	0.06	8

a Family planning–measured as demand satisfied with modern methods in women aged 15–49 years who are married or in a union (%)

b Pregnancy and delivery care–measured as four or more visits to antenatal care (%)

c Child immunization–measured as children aged 1 year who have received three doses of diphtheria, tetanus and pertussis vaccine (%)

d Child treatment–measured as care-seeking behavior for children with suspected pneumonia (%)

eHIV treatment–measured as the proportion of people with HIV receiving antiretroviral treatment (5)

fTuberculosis treatment–measured as TB effective treatment, calculated as ration of rate of case detection to rate of TB treatment

g Water and sanitation—measured as the proportion of households with access to at least basic sanitation (%)

h Prevention of cardiovascular disease–measured as prevalence of non-raised blood pressure regardless of treatment status (%)

i Management of diabetes–measured as mean fasting plasma glucose measured in country-specific household surveys

j Tobacco control–measured as adults aged 15 years or older who had not smoked tobacco in the previous 30 days (%)

k Hospital access—measured as the number of hospital beds per person in each country

lHealth care worker density—measured as the number of health professionals per person, comprising physicians, psychiatrists and surgeons

m Health Security–measured based on International Health regulations core capacity index. Malaria prevention is also included in the UHC SCI for countries where malaria is prevalent. Since most countries do not collect this data, we excluded it from our analysis. Cervical cancer screening and access to essential medicines were excluded because of low data availability.

Abbreviations: HIC—High-Income Countries, UMICs—Upper Middle-Income Countries, MICs- Middle-Income Countries, LMICs—Lower Middle-Income countries, LICs—Low-Income Countries.

Discussion

In this analysis, health workforce density ranked as a critical element in determining UHC service coverage in LICs, LMICs and UMICs. Improving health service coverage in these countries depends on the availability, accessibility, and capacity of health workers to deliver high quality people-centered integrated care. While other dimensions of service provision are also important, as our dominance analysis highlights, investment must focus on investing in developing the health workforce. To meet the health workforce requirements of the Sustainable Development Goals and UHC targets outlined in the UNHLM, over 18 million additional health workers are needed by 2030.[8] The growing demand for health workers is projected to add an estimated 40 million health sector jobs to the global economy by 2030.[9] Our analysis underscores the importance of investments from both public and private sectors in health worker education, as well as in the creation and filling of funded positions for health care workers once they complete pre-service training. Optimally aligning HRH investments and developing targeted strategies to ensure UHC demands a thorough understanding of unique, country-specific labor dynamics. Policies need to take into account determinants of both the supply and the demand for health workers, how these interact and how this interaction varies in different countries.[10] While many countries have taken substantial steps to increase the numbers of skilled health workers in recent years, less attention is paid to the quality and efficiency of the health workforce, and the resulting requirements of the health system.[11] A holistic approach to improving the health workforce also demands greater attention to enhanced training and adequate mentoring; use of evidence to inform interprofessional education; a shift from predominantly hospital-based care to care in the community; and increased attention to the importance of robust management capacity, especially in low and middle income countries. [12]

In addition to highlighting HRH issues as a key element to ensuring UHC, our analysis also offers insights into the broader set of service priorities which countries should address so as to achieve universal access. Recognizing that any investment in health programs is essentially a political rather than a technical choice,[13] we assert that this dominance analysis can help inform the political economy around where and how resources are invested in the health system. In LICs and LMICs, our results underscore the importance of access to clean water and basic sanitation, and maternal and reproductive care. Absent these essential elements, it is very difficult for health policymakers and practicing healthcare workers to build a functioning system or implement change effectively. [14]

In HICs, where health workforce density and water sanitation are less important predictors of UHC SCI, HIV and cardiovascular disease programs were the most dominant interventions, highlighting the importance of disease-specific programs in settings where accesses to basic public health provisions, including availability of doctors and access to water and sanitation have often been universally secured. While HICs are able to provide a wide array of health services, most low and middle income countries only have the resources to deliver a smaller set of services, necessitating a more explicit and systematic approach to priority setting.[15] As such these findings also underscore the importance of effective and integrated public health care and preventions programs, and basic hospital access as critical elements to achieving UHC.

We note that direct hospital access is not a high-ranking parameter in HICs, but is of high importance in LICs. We posit that the importance of this parameter in LICs underscores how access to hospitals is a useful parameter for distinguishing those lower income countries that are likely to have other basic health infrastructure and those that are not. Nonetheless, the contradistinction between the importance of this variable in LICs compared to HICs may reflect the fact that this indicator is more accurate in LICs and LMICs, while the number of inpatient hospital admissions is often better documented as a surrogate for hospital availability in HICs.[3] Similarly, the fact that the family planning indicator ranks mostly highly in UMICs, but appears to be of less importance in LICs and MICs, may reflect the fact that this variable, which is calculated using a complex denominator derived from multiple survey questions, is better reported in those UMICs. As with other sub-indices in the SCI, the method for deriving this variable is highly sensitive to the methods and quality of data acquisition across countries.[3] Tracking UHC SCI scores over time, and minimizing missing data across countries would enable further exploration of the relative importance of these different variables and these discrepant findings.

This analysis has a number of important limitations. It provides no sub-national granularity and offers no insights into how human resource shortages disproportionally impacts marginalized populations,[16] even in high-income countries. In addition, the HRH sub-index does not provide any insights into the impact of training or supervision on how the workforce functions.[17] The importance of training infrastructure to ensure that health workers can respond to the changing health needs of the population while also maintaining the quality of services is vital, but not captured in the SCI. Furthermore the HRH sub-index used in calculating the UHC SCI score is based on the density of physicians per 100,000 population. It does not draw attention to differences in other cadres within the health workforce or their importance to achieving UHC. In many countries, basic information fields, such as health worker stock and distribution, are largely limited to physicians, nurses and midwifes, despite the growing role played by other cadres, such as community health workers.[8] Significant improvements in the capacity of countries to understand the conditions and opportunities to strengthen their national labor markets is essential. Crucially, this evidence should be developed through putting in place country-level mechanisms to collate, analyze, and use data on a routine basis, especially for the purpose of tracking overall progress towards UHC. Finally, another key limitation of our analysis relates to the nature of the UHC SCI, and the fact that it does not capture longitudinal data about the dynamics of health system evolution. In addition, neither the HRH sub-index nor the overall UHC SCI score capture other dimensions of UHC related to financing of health services or the quality of service provided. Metrics that track these will continue to be essential as countries plan towards UHC.[15]

In summary, this analysis represents the first attempt to define on an empirical basis the importance of health workforce requirements to UHC progress at a county level. Greater efforts to prioritize and strengthen human resources for health are necessary if countries are going to be successful in realizing the goal of UHC.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

6 Jan 2020

PONE-D-19-26015

Achieving Universal Health Coverage (UHC): dominance analysis across 183 countries highlights importance of strengthening health workforce

PLOS ONE

Dear Dr. Reid,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

This is a well written short manuscript which will further be strengthened if it has a strong discussion section. I agree that human resources for health are the prime focus of attention in LMICs to achieve UHC. However, there are other important facets of UHC like quality of care, essential services offered, access to the services and financial protection to avail the services. These facets, though not  apart of the analysis, requires to be touched upon and discussed. Apart from this, I think the reviewer has also raised some very important points which needs to be addressed in the manuscript.

We would appreciate receiving your revised manuscript by 15th January 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Manu Raj Mathur, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noticed you have some minor occurrence of overlapping text with the following previous publication(s), which needs to be addressed:

-Hogan, Daniel R., et al. "Monitoring universal health coverage within the Sustainable Development Goals: development and baseline data for an index of essential health services." The Lancet Global Health 6.2 (2018): e152-e168.

https://adra.eu/Health-For-All.pdf

In your revision ensure you cite all your sources, and quote or rephrase any duplicated text outside the methods section. Further consideration is dependent on these concerns being addressed.

3. Please amend your list of authors on the manuscript to ensure that each author is linked to an affiliation. Authors’ affiliations should reflect the institution where the work was done (if authors moved subsequently, you can also list the new affiliation stating “current affiliation:….” as necessary).

Additional Editor Comments (if provided):

This is a well written short manuscript which will further be strengthened if it has a strong discussion section. I agree that human resources for health are the prime focus of attention in LMICs to achieve UHC. However, there are other important facets of UHC like quality of care, essential services offered, access to the services and financial protection to avail the services. These facets, though not  apart of the analysis, requires to be touched upon and discussed.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a quick and elegant analysis that offers insight for understanding progress on the UHC path. However, it seemed a bit rushed to me and thus lacks depth. This is of course, fairly easily remedied - I have suggestions on this below:

For one, authors have already engaged with the strengths and weaknesses of the UHC index. Highlighting this more is desirable.

Some mention of whether an indicator centric approach to gauging progress should also be discussed. There is obviously more to this, as work on political economy by Bump, Reich, etc. have outlined and the paper is incomplete without mention of these features.

Relevant applications of DA should be referenced and selection of estimation method - least squares was used? or were adjusted methods relied on. Some sensitivity analysis would be appropriate even as conclusions tend to invariant across methods.

The discussion places emphasis on density (even as the limitations of this are later outlined) - some mention of supportive supervision and continuous training - will be especially important as a number of particularly LMIC countries transition their systems from MCH/MDG focused systems and measurement to CPHC including NCDs.

Worthwhile to discuss hospital access in LICs and family planning in UMICs? Place this in some context too.

also, what are the instrumentation and analytical reasons why other factors are lower priority. Despite the fact that the number of LMICs (is fewer in number, for eg than HICs) pregnancy care ranks higher - what is this showing us?

This method is highly sensitive to the quality of data and operationalisation of indicators, which is a big issue. This should be mentioned. There is growing work on the relevance and measurement of these indicators at the ground level that authors may also consider referencing

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Devaki Nambiar

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

23 Jan 2020

Response to Reviewer #1

(Reviewer comments are in bold italics)

Comment 1: This is a quick and elegant analysis that offers insight for understanding progress on the UHC path. However, it seemed a bit rushed to me and thus lacks depth.

Response: We are grateful for the reviewer’s statement that this is an elegant analysis and agree that it does provide valuable insights that can inform UHC policies in countries that intend to expand access to health services. As we outline below, we have expanded the discussion section highlighting in more depth the critical role of human resources for health to the broader UHC agenda. In addition, we have added more content on both the strengths and weakness of dominance analysis.

Comment 2: Authors have already engaged with the strengths and weaknesses of the UHC index. Highlighting this more is desirable.

Response: In the revised draft we have included more detail on the UHC index. In addition, we have included a reference to a more detailed description of the index in another of our papers on this topic. This additional context is on line 63:

‘The UHC SCI is unique insofar as it provides a means for assessing the breadth of essential services being offered by individual countries. Furthermore, it provides a simple and standardized summary of a variety of complex and heterogenous service parameters. While the index is limited by the quality of country-specific data on which it is based and the fact that much of the data has been collected asynchronously, it offers valuable service-oriented insights into the implementation of UHC on a global scale.’

In addition to this edit, we have added statements in the limitation section of the discussion, expanding on why the UHC SCI offers an incomplete picture of universal health coverage and how it offers no insights into subnational disparities in workforce density. On line 220, we state:

‘This analysis has a number of important limitations. The analysis provides no sub-national granularity and offers no insights into how human resource shortages disproportionally influence marginalized populations, even in high-income countries.’

On line 239, we have added the following text:

‘Finally, another key limitation of our analysis relates to the nature of the UHC SCI, and the fact that it does not capture longitudinal data about the dynamics of health system evolution. In addition, neither the HRH sub-index nor the overall UHC SCI score capture other dimensions of UHC related to financing of health services or the quality of service provided. Metrics that track these will continue to be essential as countries plan towards UHC.’

Comment 3: Some mention of whether an indicator centric approach to gauging progress should also be discussed. There is obviously more to this, as work on political economy by Bump, Reich, etc. have outlined and the paper is incomplete without mention of these features.

Response: This comment is helpful and insightful. We agree that the political economy of universal health coverage has a critical bearing on both the design and implementation of efforts to improve health system performance. Furthermore, we acknowledge that using indicators, such as those delineated within the UHC SCI is only one mechanism for determining how each country navigates a path towards UHC. In the discussion section of the paper we highlight how UHC also entails ensuring quality of service and providing mechanisms to fund healthcare, neither of which are addressed in the UHC SCI (line 242).

The purpose of our analysis is not to provide a comprehensive mapping of all the potential political economy factors and strategies related to UHC, but we do provide a structured way for thinking about the relative importance of different essential services that we hope will assist policy-makers in prioritizing what services and resources to invest in. We do not subscribe to the perspective that decisions about UHC policy should only be determined by indicators. Rather we primarily want to highlight in this manuscript how service coverage priorities differ by countries, and that HRH appears to be essential to advancing a UHC agenda in all settings, regardless of country income status. We recognize the essential role of political economy factors throughout the policy cycle, including agenda-setting, policy design, adoption, implementation and evaluation in order to achieve UHC, and have highlighted how an analysis such as ours can help inform this process. On line 181, we have added the following text:

“Recognizing that any investment in health programs is essentially a political rather than a technical choice we assert that this dominancy analysis can help inform the political economy around where and how resources are invested in the health system.”

Comment 4: Relevant applications of DA should be referenced and selection of estimation method - least squares was used? or were adjusted methods relied on. Some sensitivity analysis would be appropriate even as conclusions tend to invariant across methods.

Response: In the revised draft we have included reference to similar and relevant applications of Dominance Analysis, including citations from Azen et al and Sauceda et al. Notably, we performed the Dominance Analysis using the ‘epsilon’ approach in STATA; as proposed by the reviewer we have indicated this in the revised draft (line 95). This approach has been described in depth by Jeff Johnson in Multivariate Behavioural Research (Johnson, J. W. A heuristic method for estimating the relative weight of predictor variables in multiple regression. Multivariate Behavioral Research,35(1), 1-19.) and employs a methodology similar to least squares as part of the approach, utilizing a relative weight analysis to estimate the proportionate contribution to R2 for a set of predictors.

We did explore using other analystic tools to undertake an additional sensitivity analysis in order to determine the variable importance of multiple independent variables in accounting for variance in a single variable. In particular, we explored using variable clustering analysis, which would have enabled us to cluster countries according to which sub-indices from within the UHC SCI had maximal impact, using the beta weights to assess variable importance. We opted against using variable clustering analysis because such an analytic approach would have offered no mechanism for differentiating between relevant and irrelevant variables. In addition, we note that beta weights are also limited in their ability to determine suppression in a regression equation.

Comment 5: The discussion places emphasis on density (even as the limitations of this are later outlined) - some mention of supportive supervision and continuous training - will be especially important as a number of particularly LMIC countries transition their systems from MCH/MDG focused systems and measurement to CPHC including NCDs.

Response: We agree with the reviewer’s comment. We have included an additional clause in the limitations section of the manuscript, highlighting the importance of training and supervision as crucial elements of a well-functioning health workforce. This additional text is included from line 220:

‘The analysis provides no sub-national granularity and offers no insights into how human resource shortages disproportionally influence marginalized populations, even in high-income countries. In addition, the HRH sub-index does not provide any insights into the impact of training or supervision on how the workforce functions. The importance of training infrastructure to ensure that health workers can respond to the changing health needs of the population while also maintaining the quality of services is vital, but not captured in the SCI.’

We have also included in the discussion the importance health workforce training and mentorship to ensure workforce quality in addition increasing workforce density. On Line 175, we have added the following text:

“While many countries have taken substantial steps to increase the numbers of skilled health workers in recent years, less attention is paid to the quality and efficiency of the health workforce, and the resulting requirements of the health system.[11] A holistic approach to improving the health workforce also demands greater attention to enhanced training and adequate mentoring; use of evidence to inform interprofessional education; a shift from predominantly hospital-based care to care in the community; and increased attention to the importance of robust management capacity, especially in low and middle income countries

Comment 6: Worthwhile to discuss hospital access in LICs and family planning in UMICs? Place this in some context too.

Response: We agree that these are other parameters are worthy of discussion since they appear to have an important role in specific circumstance. To this ends we have added additional sentences highlighting the import of these findings in the discussion. These additional edits begin on line 236.

As outlined in these added sentences, direct hospital access is a fundamental component of healthcare particularly in low income countries. Furthermore, access to hospitals is also a useful parameter for distinguishing those countries that are likely to have other basic health system infrastructure and those poorest countries that are not. Hospital access is a proxy for essential inpatient services and has more data available in low-income and middle-income countries than number of inpatient hospital admissions, which is often better characterized in high-income settings. In calculating the UHC SCI a threshold value is used to capture only low capacity levels because high values might represent overcapacity or inefficient allocation of resources.

Access to family planning services is also a fundamental component of the health system. However, as discussed in the Hogan paper, the index has a relatively complex denominator derived from multiple survey questions and data collection. As a consequence, the indicator is probably better characterized in high income countries, than low-income and middle-income countries, a factor underscored by fact that the data is missing for >60 countries.

These edits are discussed starting on line 206 in the revised draft:

‘We note that direct hospital access is not a high-ranking parameter in HICs, but is of high importance in LICs. We posit that the importance of this parameter in LICs underscores how access to hospitals is a useful parameter for distinguishing those lower income countries that are likely to have other basic health infrastructure and those that are not. Nonetheless, the contradistinction between the importance of this variable in LICs compared to HICs may reflect the fact that this indicator is more accurately in LICs and LMICs, while the number of inpatient hospital admissions is often better documented as a surrogate for hospital availability in HICs. Similarly, the fact that the family planning indicator ranks mostly highly in UMICs, but appears to be of less importance in LICs and MICs, may reflect the fact that this variable, which is calculated using a complex denominator derived from multiple survey questions, is better reported in those UMICs. Tracking UHC SCI scores over time, and minimizing missing data across countries would enable further exploration of the relative importance of these different variables and these discrepant findings.’

Comment 7: What are the instrumentation and analytical reasons why other factors are lower priority. Despite the fact that the number of LMICs (is fewer in number, for eg than HICs) pregnancy care ranks higher - what is this showing us?

Response: This is a great question. The fact that pregnancy care ranks higher in LMICs compared to UMICs perhaps highlights variability in the quality of data across countries. This index measures whether women have access to four or more antenatal care visits during pregnancy and as such captures the amount of contact with the health system but not the quality of care received during pregnancy. Skilled attendance at birth is a preferred alternative for assessing pregnancy services; however insufficient standardized measurement of skilled health-care personnel makes cross-country comparisons difficult. We therefore speculate that the disparity between where this parameter ranks in lower-middle income and upper-middle countries may result because the index ceases to be as sensitive indicator of pregnancy care in settings when comparing countries that are all more likely to have such services, even if the quality of service is variable.

Comment 8: This method is highly sensitive to the quality of data and operationalisation of indicators, which is a big issue. This should be mentioned. There is growing work on the relevance and measurement of these indicators at the ground level that authors may also consider referencing

Response: We absolutely agree with this comment and have added a subclause on line 215 highlighting this challenge.

12 Feb 2020

Achieving Universal Health Coverage (UHC): dominance analysis across 183 countries highlights importance of strengthening health workforce

PONE-D-19-26015R1

Dear Dr. Reid,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Manu Raj Mathur, PhD

Academic Editor

PLOS ONE

20 Feb 2020

PONE-D-19-26015R1

Achieving Universal Health Coverage (UHC): dominance analysis across 183 countries highlights importance of strengthening health workforce

Dear Dr. Reid:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Manu Raj Mathur

Academic Editor

PLOS ONE