|Home | About | Journals | Submit | Contact Us | Français|
Surgery is a foundational component of health care systems. However, previous efforts to integrate surgical services into global health initiatives do not reflect the scope of surgical need and many health systems do not provide essential interventions. We estimate the minimum global volume of surgical need to address prevalent diseases in 21 epidemiological regions from the Global Burden of Disease Study 2010 (GBD).
Prevalence data were obtained from GBD 2010 and organized into 119 disease states according to the World Health Organization’s Global Health Estimate (GHE). These data, representing 187 countries, were then apportioned to the 21 GBD epidemiological regions. Using previously defined values for the incident need for surgery for each of the 119 GHE disease states, we calculate minimum global need for surgery based on the prevalence of each condition in each region.
We estimate that at least 321·5 million surgical procedures would be needed to address the burden of disease for a global population of 6.9 billion in 2010. Minimum rates of surgical need vary across regions, ranging from 3,383 operations per 100,000 in Central Latin America to 6,495 operations per 100,000 in Western Sub-Saharan Africa. Global surgical need also varied across sub-categories of disease, ranging from 131,412 procedures for Nutritional Deficiencies to 45.8 million procedures in Unintentional Injuries.
The estimated need for surgical procedures worldwide is large and addresses a broad spectrum of disease states. Surgical need varies between regions of the world according to disease prevalence and many countries do not meet the basic needs of their populations. These estimates may be useful for policy makers, funders, and ministries of health as they consider how to incorporate surgical capacity into health systems.
Surgical care has a role in treating a broad spectrum of diseases in the alleviation of human suffering.1 It is required at all ages; from neonates with congenital anomalies to the elderly with cataracts. Surgery can be preventative, as in reducing HIV transmission through circumcision, or curative as in many cancers. It is often a component of acute emergency care, such as bowel perforations and trauma, as well as the treatment of chronic diseases such as osteoarthritis and inflammatory bowel disorders. Additionally, surgical care is important in the diagnosis and supportive care of numerous conditions. For example, patients with renal failure require dialysis access and may eventually be candidates for kidney transplantation. In recognition of these many roles, researchers and economists now acknowledge that surgical care is a fundamental component of healthcare and contributes to overall social and economic development.2–4
Previous efforts to integrate surgical services into global health have failed to recognize the scope of surgical need. Traditionally, surgical initiatives in global health were implemented as disease-specific vertical interventions to meet targeted needs in resource-poor settings of the world.5–7 More recent efforts to expand the breadth of such services include the World Health Organization’s (WHO) Emergency and Essential Surgical Care program and the World Bank’s Disease Control Priorities project; both of which promote the implementation of essential packages of interventions at first-level hospitals in Low- and Middle-Income Countries (LMICs).8,9 However, research now shows that surgical care is required in the treatment of nearly all disease states of the global burden of disease, strengthening the argument that these services must be integrated into health systems at all levels and for diverse clinical problems.1,10 This complicates the existing lack of access to surgical care estimated to affect 4.8 billion people worldwide.11
Defining the role of surgery in health systems remains a difficult problem in global health. Specifically, there is a need to define its role across a spectrum of diseases and what minimum levels of surgical intervention might be. In this report, we use prevalence data for diseases and conditions described by the Global Burden of Disease, Injuries, and Risk Factors Study 2010 (GBD 2010) to estimate the minimum need for surgery in each of the GBD regions of the world.
We obtained prevalence data from the Global Burden of Disease Study (GBD) from the Institute for Health Metrics and Evaluation (IHME).12 From its beginnings in 1990, the complexity of GBD has grown to include 291 diseases and injuries organized into 21 disease sub-categories, 1160 sequelae of these conditions, and 67 risk factors in 187 countries.13 The IHME is the custodian of this comprehensive database and performs periodic updates and reports, most recently published in the Lancet with a rigorous description of methods and data-sources.13,14 They combine data from many sources including: vital registration, demographic surveillance systems, household surveys, verbal autopsies, and other sources.
We extracted all-age population prevalences from GBD 2010 and standardized our disease taxonomy according to the WHO Global Health Estimate (GHE).15 This was accomplished by disaggregating GBD disease states into their ICD-10-based coding definitions and reorganizing them according to the GHE framework. The GHE framework was chosen because of the ability to link any analysis to a variety of other health systems analysis tools available through the WHO. While both frameworks rely on the same prevalence data, they differ in classification of some disease states.16 In the GHE framework there are 22 disease sub-categories whereas in the GBD there are 21 disease sub-categories. Both GBD and GHE taxonomies are publicly available for comparison of ICD-10 codes.15,17
During this study there were multiple disease states with missing data due to ongoing updates of the GBD dataset. These include: Other Maternal, Other Neonatal, Other Nutritional, Other Respiratory, Other Digestive, Other Genitourinary, Encephalitis, Intestinal Nematode Infections, Otitis Media, Vitamin A Deficiency, Other Nutritional Deficiencies, Exposure to Forces of Nature, Collective Violence and Legal Intervention. These missing data are shown in tables with an asterisk and were excluded from analysis.
The final dataset extracted from the GBD 2010 consisted of all-age population prevalence for 119 GHE disease states organized into 22 disease subcategories. These data, representing 187 countries, were grouped according to geographic and epidemiologic similarity into 21 regions, according to methods described in GBD 2010.14 Listed alphabetically, these regions are: Andean Latin America, Australasia, Caribbean, Central Asia, Central Europe, Central Latin America, Central sub-Saharan Africa, East Asia, Eastern Europe, Eastern sub-Saharan Africa, High-income Asia Pacific, High-income North America, North Africa and Middle East, Oceania, South Asia, Southeast Asia, Southern Latin America, Southern sub-Saharan Africa, Tropical Latin America, Western Europe, Western sub-Saharan Africa.
In order to estimate the global need for essential surgical procedures based on prevalence of diseases in each GBD region. we required national or multinational data to benchmark the relationship between the prevalence of discrete disease states and utilization of surgical services. This required surgical encounters to be rigorously and consistently documented in a national dataset with ICD coding for diagnoses and procedures. Another requirement was a well-financed national healthcare system with universal access for an entire population where it could be generally assumed that surgical care is provided when needed. Lastly, we needed a setting and strategy that would minimize the amount of unnecessary surgery captured in the analysis.
New Zealand satisfied the above criteria while also achieving excellent health outcomes with efficient use of resources. It has a strong, nationalized healthcare system with universal access and a publicly-supported national repository of hospital discharge information from all public sector hospitals.18 In 2010, overall life expectancy was 80.7 years in New Zealand and 80.2 in high-income countries (HICs).19–21 Similarly, maternal mortality was 12 per 100,000 live births while in HICs it was 15 per 100,000 live births. Under-five mortality was also comparable at 6.4 per 1,000 live births in New Zealand and 5.5 in HICs. These outcomes were also achieved with comparatively low cost. In 2010, total expenditure for health per capita in New Zealand was 25% lower ($3,260 USD) in comparison to other high-income countries ($4,325 USD). Lastly, the national rate of surgery is among the lowest of all HICs even when including estimates of operations from the private sector (6,270 per 100,000 in 2012), minimizing the effect of boutique or potentially unnecessary surgery.22 For these reasons, New Zealand was an optimal setting from which to derive a standard for surgical care.
We used previously-defined index rates of procedure for each disease state of the Global Health Estimate from New Zealand.10 Rates of procedure were defined as the likelihood of each disease state requiring an inpatient surgical procedure in one calendar year. Rates of procedure were calculated by compiling International Classification of Diseases (ICD), Version 10-AM, codes from the New Zealand National Minimum Dataset. Using the primary cause of admission, hospitalizations were aggregated into disease states according to the WHO GHE and determined whether or not the admission was associated with an operation in a binary fashion (i.e. present or absent). Surgical procedures were defined as any procedure requiring general or neuroaxial anesthesia. Using these results and New Zealand disease and condition prevalence from the GBD 2010 study, we determined the incident rate of procedures per disease prevalence. A more detailed description of methods is available in Hider et al.10
The analysis of index rates of surgical procedures in New Zealand was structured to err on the conservative side by restricting the analysis to the inpatient setting. In addition, the analysis did not include inpatient surgical procedures performed outside the public sector, procedures performed under local anesthesia, outside the operating room, or multiple procedures (i.e. reoperations) during the same hospitalization. This was done because there may be a nontrivial proportion of surgical volume that is not clinically indicated and supply-sensitive surgery may contribute to wide variation in rates of surgery between countries.23 In some settings the component of overall surgical volume that is performed on an outpatient basis approximates 40–50% of total surgical procedure volume.24 Collectively, the above precautions ensure the most conservative analysis possible for minimum rates of procedures.
To estimate the number of procedures needed in each epidemiologic region we extrapolated the need for surgical care for all disease states in our dataset. We accomplished this in two steps, outlined in the following equation. First, we multiplied the unique standardized incident rate of procedure (‘ROP’) for each disease state (‘i’) by the prevalence counts (‘Prev’) of each disease state (‘i’). Second, we summed the absolute volume of surgical procedures needed for all disease states within a sub-category and then for all sub-categories (‘n’) within a region. Confidence intervals for procedure volumes were not calculated because index rates of procedure were calculated at the population level from New Zealand.
Lastly, we normalized the cumulative volume of procedures needed in each region by that region’s population in 2010 and report the overall rate of surgery per 100,000 population. Populations for each region in 2010 were extracted from publicly available descriptive statistics at the World Bank.25
This study was supported by grants from the United States National Institutes of Health. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
In 2010, we estimate that 321.5 million inpatient surgical procedures were needed to address the global burden of disease. This number of operations amounts to a global per capita surgical need of 4,664 per 100,000. The volume of need across the three broad categories of disease was: 64·2 million for Communicable, Maternal, Perinatal and Nutritional Conditions; 208·8 million for Non-Communicable Diseases; and 48·8 million for Injuries.
Absolute volumes of surgical procedure need varied by several orders of magnitude between epidemiologic regions. (Table 1) Surgical need ranged from 447,554 in Oceania to 72,919,681 in Southern Asia. The regions with the largest volume of need were also the regions with the largest populations, namely: Southern Asia (72,919,681 operations; 1,613,194,931 inhabitants), Eastern Asia (57,821,123 operations; 1,397,940,667 inhabitants), and Southeast Asia (25,794,258 operations; 610,448,769 inhabitants). This relationship is displayed graphically in Figure 1A.
Minimum rates of surgical need per population also varied between regions, After normalizing the absolute volume according to population size in each region, the lowest rate of surgical need was in Central Latin America (3,384 operations per 100,000 inhabitants) and the highest was in Western Sub-Saharan Africa (6,495 operations per 100,000 inhabitants); almost a two-fold difference. The highest rates of surgical need were in the following regions: Western Sub-Saharan Africa (6,495 operations per 100,000 inhabitants), Central Sub-Saharan Africa (6,255 operations per 100,000 inhabitants), and Eastern Sub-Saharan Africa (6,145 operations per 100,000 inhabitants). The median rate of need per capita at the regional level was 4,669 operations per 100,000 in Australasia.
The need for surgical procedures also varied between disease subcategories and individual disease states. (Table 2) Global volume of procedure need for the 22 disease subcategories ranged from the lowest in Nutritional Deficiencies (131,412) to the highest in Unintentional Injuries (45,809,537). The subcategories with the greatest absolute need were Unintentional Injuries (45,809,537), Musculoskeletal Diseases (40,493,940), and Maternal Conditions (38,720,343). There was also variation within disease subcategories. For example, within the subcategory of Malignant Neoplasms, the need for surgical procedures differed by as much as two degrees of magnitude between types of cancers. (Table 3) A full description of need for 119 disease states is available in Appendix 1.
The need for surgical procedures within regions was also heterogeneous and differed substantially from one region to the next. (Figure 1B) For example, surgical procedures for Infectious and Parasitic Diseases accounted for 0.32% of procedures in Western Europe (lowest), 3.0% in Andean Latin America (median), and 35.3% in Western Sub-Saharan Africa (highest). Similarly, surgical procedures for Maternal Conditions varied from 2.1% in Western Europe (lowest) to 7.5% in Andean Latin America (median) to 20.4% in Eastern Sub-Saharan Africa (highest).
Heterogeneity between regions is also evidenced in reviewing the single disease subcategory accounting for the greatest proportion of surgical procedures within each region. Accordingly, the disease subcategory with greatest surgical need was Unintentional Injuries in the Caribbean (32.1%), Infectious and Parasitic Diseases in Western Sub-Saharan Africa (35.3%), and Maternal Conditions in Southeast Asia (15.7%). Figure 1B displays the heterogeneity for all 22 disease subcategories across the 21 GBD regions.
The global estimated need for surgical procedures is large and cuts across a broad spectrum of GHE disease states. Surgery is essential for addressing basic health needs globally, although the degree varies between epidemiological regions of the world and between disease subcategories. Our findings provide the first estimation of need for surgical services that is sensitive to each region’s unique epidemiologic profile. This fundamentally challenges the notion that surgical need can be met through vertical programming and reinforces the fact that a versatile surgical core should lie at the center of health systems.
These results demonstrate how the development of surgical capacity might vary according to region. For example, in Western Sub-Saharan Africa, Central Sub-Saharan Africa, and Eastern Sub-Saharan Africa two disease subcategories combined (Infectious and Parasitic Diseases and Maternal Conditions) account for over 50% of surgical procedure need. The same two subcategories account for less than 20% of procedure need in Southern Asia, Eastern Asia, and Southeastern Asia and less than 10% in Latin American regions. The single disease subcategory accounting for the greatest proportion of surgical procedures within each region also varied; Unintentional Injuries in the Caribbean, Infectious and Parasitic Diseases in Western Sub-Saharan Africa, and Maternal Conditions in Southeast Asia. These results are an opportunity for ministries of health to repeat the analysis at the country level and consider how current priorities align with population needs in regards to workforce capacity, regionalization of service delivery, referral patterns, and financing. Without considering these divergent needs, prioritization frameworks for up-scaling surgical capacity will inevitably fail to maximize population health.
It is important to note that our findings of surgical need are of comparable magnitude with other research and may be helpful in identifying initial targets for up-scaling surgical capacity. Weiser et al use multiple imputations of counts of surgery to estimate that the global volume of surgical output is large and growing, estimated to be 234·3 million procedures in 2004 and 312·9 million procedures in 2012.22,26 According to their results, rates of surgery at the country level range from 53 per 100,000 in Chad to 29,399 per 100,000 in the United States. Our estimation of global minimum surgical need at 4,664 per 100,000 globally, is a conservative estimate which rounds to 5,000 procedures per 100,000 as a possible target for the initial development of surgical infrastructure. For some regions, such as Western Sub-Saharan Africa with a rate of surgical need estimated at 6,495 per 100,000, this global rate will be insufficient. However, we consider it a modest initial goal that, if adopted by funding agencies and ministries of health, could lead to profound improvements in population health.
This approach has multiple policy implications for monitoring and evaluation at the country level. The WHO previously endorsed a package of six standardized metrics to monitor the delivery of surgical services and amongst them was the volume of surgical procedures (per operating room or population).27 While this metric is useful in evaluating service utilization and efficiency, it does little to evaluate whether current services are meeting the needs of the population when significant proportions of patients do not access healthcare services. Our methodology links hospital utilization with prevalence of diseases to bridge this gap and may be useful as a supplemental metric of coverage. Furthermore, as social and economic conditions improve in LMICs, demographic changes and epidemiologic transitions inevitably ensue. As the prevalence of Non-Communicable Disease rises, ministries of health can use this methodology to follow trends in prevalence of diseases and anticipate impending surgical needs within the healthcare system.
Another potential utility of this methodology is to compare estimates of surgical output with estimates of overall surgical need to ascertain the unmet component of surgical need. Based on global data from 2012, Weiser et al estimate that surgical output at the regional level ranges from 392,358 per year in Oceania to 63.4 million per year in High-Income North America, with dramatic disparities between countries and regions even after normalizing for population size.22 (Table 4) Some regions perform as many as four times the minimum rates of surgical need (i.e. High-income North America, with 15.8 million procedures needed and 63.4 million procedures performed) while others fall far behind (Eastern Sub-Saharan Africa, with 21.9 million procedures needed and 4.3 million procedures performed). (Table 4) We roughly calculate the global unmet need by identifying 12 of 22 regions with current surgical output falling below estimated minimum need and tabulating the surgical gap, or additional volume of procedures needed to meet the need, to be 143·1 million operations.
Of note, Weiser et al use the following definition in surgical counts, “any intervention occurring in a hospital operating theatre involving the incision, excision, manipulation, or suturing of tissue, usually requiring regional or general anesthesia”.22 This is quite similar to the definition in our modeling; differing in that Weiser et al include private practice procedures, outpatient procedures, and procedures performed under regional anesthesia where data were available. Considering that the availability of these data was restricted to HICs in Weiser’s analysis, the effect of any mismatch between his estimations and ours would artificially inflate the excess volume of surgical output in HICs but have little or no effect on the estimations of surgical output in LMICs. As such, our calculations are less helpful in determining excess in HICs but should be reliable estimates of surgical gap (or minimum unmet need) in LMICs. As there continue to be advances in medical informatics in LMICs these rough approximations of unmet need can and should be improved.
These results also carry an important message pertinent to global health governance. There is a potential for policymakers to perceive a tension between greatest need in absolute numbers and greatest need in rates per 100,000. (Table 1) The two regions with greatest absolute volume of surgical procedure need (Southern Asia and Eastern Asia) are disproportionately swayed by the impact of two very populous Middle-Income Countries (China and India, respectively). However, the two regions with the greatest surgical need per 100,000 are Western Sub-Saharan Africa and Eastern Sub-Saharan Africa, both replete with Low-Income Countries. A utilitarian approach might prioritize capacity building in the direction of the ‘greatest good for the greatest number’; while an approach rooted in social justice might prioritize development for underserved populations with the greatest disadvantage. These data make it clear that the need for surgical procedures worldwide affects many people in diverse settings and that if surgical capacity is developed in an either/or manner it is unlikely that the sustainable development goals will lead to a grand convergence in health status by 2035.28
This study has multiple limitations that merit attention. One is that prevalence-based estimates of disease are not ideally suited to encapsulate the complexity of surgical need.29 A simple example is circumcision – shown to reduce HIV transmission by as much as 60% – which is not performed for a diagnosis of disease at all, but rather to prevent the occurrence of disease.30 Another example is bariatric surgery – effective in the treatment of obesity – which is performed to treat a highly prevalent epidemic in HICs that is not included within the GHE taxonomy. Due to examples such as these, we expect subsequent estimates of surgical need to increase as frameworks are adapted to reflect surgical services. The greatest limitation to exploring the implications of such discrepancies is the paucity of data. In order to overcome this barrier, we recommend that national registries be developed to capture clinical encounters in LMICs that are consistent with burden of disease categories. An ideal starting point is the framework of four digestive diseases, four maternal-neonatal conditions, and treatable injuries recommended by the Disease Control Priorities, Volume 3.31
Another limitation is that our definition of surgery limits the interpretability of results. The presence of a surgical procedure was defined as a binary variable and does not differentiate between the myriad types of surgical procedures, resource utilization, complexity of care delivery, or health consequences. While our methods provide a broad estimation of the scope of the surgical need across disease states and geographic regions, they do not have the granularity to help policy-makers and healthcare administrators make concrete decisions about financing and resource allocation in their current form. Furthermore, our definition of surgery excludes non-operative management of many disease states (i.e. bowel obstruction) that occurs on surgical wards. In order to contribute to the up-scaling of surgical services in underserved populations, the current methodology should be applied to country-level data and explored in closer detail to make estimates relevant for the allocation of human and physical resources.
There is also an inherent limitation to benchmarking rates of procedure in a single country and then extrapolating those rates to other regions because this intrinsically reproduces key assumptions and bias. The association between a disease state and the provision of surgical care depends on contextual factors that may change the safety profile of procedures between settings such as the availability of banked blood, post-operative ventilators, or the ability to transfer to higher levels of care. Pertinent to our analysis, index rates of procedure were calculated in New Zealand exclusively from the public sector in an effort to produce conservative results. However, if a hypothetical Procedure A is performed in the private sector 5% of the time and hypothetical Procedure B is performed in the private sector 50% of the time, our modeling strategy based exclusively in the public sector would be unevenly conservative between disease states. In other scenarios, the need for surgery may be tied to disease severity and medical management. For example, hospital admissions for Communicable Diseases of the GBD required an operation 2·1% of the time in Sweden but 17.0% of the time in South Africa, reflecting underlying differences in disease severity.32 Due to the fact that patients in LMICs are more likely to have undertreated infections than New Zealand, this aspect our modeling strategy will underestimate extrapolations of global surgical need, but the extent and variation by disease state is unknown. Future efforts to estimate surgical need should capitalize on ongoing improvements in data access to evaluate the effect of benchmarks from diverse settings on this modeling strategy’s results and explore the variation in rates of procedure between regions.
Finally, our estimations do not ensure the appropriateness of individual surgical services. We extrapolate the frequency that GHE disease sub-categories require a surgical procedure without specifying what procedure that might be or whether the treatment decision was justified according to clinical guidelines and standards of care. This is especially important when one disease state has multiple surgical treatment options, such as internal versus external fixation in a context of long-bone fracture or in complicated presentations such as poly-trauma and intra-abdominal catastrophes where multiple procedures must be planned and coordinated. Appropriateness criteria are especially important because we do not account for unethical disparities in surgical care within the index country or potentially unnecessary surgery in LMICs.33 While appropriateness criteria have been developed by expert consensus panels for a limited number of surgical procedures in HICs, it is beyond the scope of this analysis to methodologically address this concern beyond the careful consideration of a benchmark country already described in our methods.23
In light of these limitations, it is worth reiterating that this analysis errs on the side of conservative estimates in multiple ways. For example, we did not account for hospitalizations where more than one procedure was performed (6% of total procedure volume in New Zealand).10 In addition, ICD-10 codes that were missing in the GBD 2010 prevalence data were highly associated with inpatient surgical procedures, accounting for 25% of all surgical procedures in the index country, but these procedures were not included in global estimations due to unavailable GBD 2010 prevalence data in each region. We also know from independent clinical databases that procedures performed in private hospitals in the index country may account for up to 25% of procedures in select disease states (i.e. knee arthroplasty); because private hospitals do not report procedures in New Zealand’s NMDS, these procedures were also omitted from global extrapolations. We also excluded outpatient procedures, bedside procedures, and surgical procedures under local anesthesia. For these reasons, it is not surprising that current counts of surgical procedures in New Zealand from Hider et al (7,840 procedures/100,000 population) are nearly double our current estimates for Australasia (4,669 procedures/100,000 population).10
Furthermore, we performed an almost identical sensitivity analysis in the United States using the Nationwide Inpatient Sample and the GBD 2010 framework and found comparable inpatient rates of procedure for the broad GBD disease categories. In Communicable-Maternal-Neonatal Diseases (24%, versus 16% in New Zealand), Non-Communicable Diseases (34%, vs 32% in New Zealand), and Injuries (35%, versus 30% in New Zealand).1 Despite the fact that national procedure output differs significantly between the two countries (29,399 per 100,000 in US; 6,270 per 100,000 in New Zealand), the comparable scale of these rates of procedure validates the assumption that excess surgery was largely eliminated by our extensive exclusion criteria.22
In conclusion, we report a large volume of surgical need, estimated at one procedure per 21 people alive today, with a global rate of surgery of 4,664 per 100,000. While the estimated need varies according to disease epidemiology, there is a remarkable consistency in our estimates across different countries and settings. This methodology may be useful in advocating for initial targets to meet procedure need. Future research should focus on improving data quality and availability, validating rates of procedure in different settings, and incorporating measures of population health into policy-oriented frameworks for monitoring and evaluation of surgical capacity in underserved populations.
Surgical care plays a critical role in advancing population health through facility-based treatment of acute and chronic diseases. In fact, the movement towards universal healthcare access will ultimately fail to achieve sustainable development goals without the integration of surgical services.
The global burden of disease is a useful framework to evaluate the comparative impact of discrete disease entities on population health in the form of death and disability. However, the current iteration of the global burden of disease study does not make reference to treatment modalities and the role of surgery is neglected. This is especially problematic when the comparative burden of conditions is used to justify the allocation of resources in global health.
The preparation of this manuscript included an extensive PubMed literature review in the areas of global burden of disease, clinical indicators relevant to surgery, and frameworks for monitoring and evaluation of health systems. This review contributed directly to the current manuscript, the Disease Control Priorities Project, Volume 3, and the Lancet Commission on Global Surgery. As such, details of these reviews are published within these documents, as appendices to the same, or as stand-alone manuscripts.
To our knowledge, this study is the first estimation of global need for surgical procedures. These data suggest that the need for operations is large and that surgical care is indicated across a broad spectrum of conditions. Future efforts should focus on explaining variation in rates of surgical need between regions and improving the quality of data available for such analyses at the country level.
Acknowledgements: The authors acknowledge the assistance of the Institute for Health Metrics and Evaluation in provision of prevalence data from the Global Burden of Disease Study 2010 and the Perioperative Mortality Review Committee and Health Quality and Safety Commission, Wellington, for provision of New Zealand data.
Funding: United States National Institutes of Health.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Authors Contributions: JR, TGW, PH, and SWB were responsible for the study conception and design. JR, TGW, and PH analyzed and interpreted the data. JR drafted the report, which was subsequently revised by TGW, SWB, PH, LW, RG, and SWB. All authors read and approved the final report.
Conflict of Interest: The authors have no conflicts of interest.