Mapping Population-Level Spatial Access to Essential Surgical Care in Ghana Using Availability of Bellwether Procedures | Global Health | JAMA Surgery | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Cost-Distance Analyses of Population-Level Spatial Access to Essential Surgery in Ghana
Cost-Distance Analyses of Population-Level Spatial Access to Essential Surgery in Ghana

Bellwether procedures include open fracture repair, emergency laparotomy (eg, splenectomy for trauma and repair of hollow viscous perforation), and cesarean section.

Figure 2.  Cost-Distance Analyses of Population-Level Spatial Access to Essential Surgery With and Without First-Level Hospitals Targeted for Capability Strengthening
Cost-Distance Analyses of Population-Level Spatial Access to Essential Surgery With and Without First-Level Hospitals Targeted for Capability Strengthening

A location-allocation model was built to identify 5 facilities in Ghana from a candidate list of first-level hospitals that performed less than 12 of each bellwether procedure per year that would have the greatest effect on population-level spatial access to essential surgery within 1 hour if strengthened. Candidate facilities are Comboni Hospital, Hawa Memorial Savior Hospital, Oda Hospital, Wiawso Hospital, and Zebilla Hospital. Population-weighted geographic centroids of districts represented demand points in the model. A second cost-distance analysis was performed after combining the candidate facilities and facilities that performed at least 12 of each bellwether procedure per year to quantify improvements in population-level spatial access after capability strengthening. Bellwether procedures include open fracture repair, emergency laparotomy (eg, splenectomy for trauma and repair of hollow viscous perforation), and cesarean section.

Table 1.  Number of Procedures at 123 First-Level Hospitals With and Without Bellwether Procedure Capability in Ghanaa
Number of Procedures at 123 First-Level Hospitals With and Without Bellwether Procedure Capability in Ghanaa
Table 2.  Percentage of First-Level Hospitals That Perform Selected Procedures and Bellwether Procedure Capability in Ghanaa
Percentage of First-Level Hospitals That Perform Selected Procedures and Bellwether Procedure Capability in Ghanaa
Table 3.  Population-Level Spatial Access to Hospitals Capable of Essential Surgery in Ghanaa
Population-Level Spatial Access to Hospitals Capable of Essential Surgery in Ghanaa
1.
GBD 2013 Mortality and Causes of Death Collaborators.  Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.  Lancet. 2015;385(9963):117-171.PubMedGoogle ScholarCrossref
2.
Alkire  BC, Raykar  NP, Shrime  MG,  et al.  Global access to surgical care: a modelling study.  Lancet Glob Health. 2015;3(6):e316-e323. doi:10.1016/S2214-109X(15)70115-4.PubMedGoogle ScholarCrossref
3.
Mock  CN, Donkor  P, Gawande  A, Jamison  DT, Kruk  ME, Debas  HT; DCP3 Essential Surgery Author Group.  Essential surgery: key messages from Disease Control Priorities, 3rd edition.  Lancet. 2015;385(9983):2209-2219.PubMedGoogle ScholarCrossref
4.
Stewart  BT, Quansah  R, Gyedu  A, Ankomah  J, Donkor  P, Mock  C.  Strategic assessment of trauma care capacity in Ghana.  World J Surg. 2015;39(10):2428-2440.PubMedGoogle ScholarCrossref
5.
Debas  HT, Donkor  P, Gawande  A, Jamison  DT, Kruk  ME, Mock  CN. Essential surgery. In: Jamison  DT, Nugent  R, Gelband  H, Horton  S, Jha  P, Laxminarayan  R, eds.  Disease Control Priorities. 3rd ed. Washington, DC: International Bank for Reconstruction and Development/The World Bank; 2015.
6.
Price  R, Makasa  E, Hollands  M.  World Health Assembly Resolution WHA68.15: “Strengthening Emergency and Essential Surgical Care and Anesthesia as a Component of Universal Health Coverage”—Addressing the Public Health Gaps Arising from Lack of Safe, Affordable and Accessible Surgical and Anesthetic Services.  World J Surg. 2015;39(9):2115-2125.PubMedGoogle ScholarCrossref
7.
Stewart  BT.  Editorial commentary on Bolkan et al: “The Surgical Workforce and Surgical Provider Productivity in Sierra Leone: A Countrywide Inventory” [published online February 23, 2016].  World J Surg. 2016.PubMedGoogle Scholar
8.
Mock  C.  Confronting the global burden of surgical disease.  World J Surg. 2013;37(7):1457-1459.PubMedGoogle ScholarCrossref
9.
Stewart  B, Wong  E, Papillon-Smith  J,  et al.  An analysis of cesarean section and emergency hernia ratios as markers of surgical capacity in low-income countries affected by humanitarian emergencies from 2008-2014 at Médecins Sans Frontières Operations Centre Brussels Projects.  PLoS Curr. 2015;7:ecurrents.dis.5e30807568eaad09a3e23282ddb41da6. doi:10.1371/currents.dis.5e30807568eaad09a3e23282ddb41da6.PubMedGoogle Scholar
10.
Gruen  R. Lancet Commission in Global Surgery. Information Management Working Group: progress report. http://www.globalsurgery.info/wp-content/uploads/2013/11/Russel-Gruen.pdf. Published June 2014. Accessed November 15, 2015.
11.
Meara  JG, Leather  AJ, Hagander  L,  et al.  Global Surgery 2030: evidence and solutions for achieving health, welfare, and economic development.  Lancet. 2015;386(9993):569-624.PubMedGoogle ScholarCrossref
12.
CIA World Factbook, Ghana. http://www.ciaworldfactbook.us/africa/ghana.html. Published 2014. Accessed April 12, 2015.
13.
The World Bank. Ghana: population growth. http://data.worldbank.org/country/ghana. Published 2015. Accessed January 14, 2015.
14.
Global Burden of Disease. Cause patterns. http://vizhub.healthdata.org/gbd-cause-patterns/. Updated September 14, 2015. Accessed May 8, 2016.
15.
The World Bank. United Kingdom. http://www.worldbank.org/en/country. Published 2014. Accessed May 6, 2015.
16.
Ghana Health Service. The health sector in Ghana: facts and figures. http://www.moh-ghana.org/UploadFiles/Publications/GHS%20Facts%20and%20Figures%202010_22APR2012.pdf. Published 2010. Accessed November 15, 2015.
17.
Amporfu  E.  Private hospital accreditation and inducement of care under the Ghanaian National Insurance Scheme.  Health Econ Rev. 2011;1(1):13.PubMedGoogle ScholarCrossref
18.
DHIS 2. Overview. https://www.dhis2.org/inaction. Accessed January 12, 2016.
19.
APRIDEC Medical Outreach Group. http://apridec.org/. Published 2010. Accessed May 12, 2016.
20.
Sanders  DL, Kingsnorth  AN.  Operation hernia: humanitarian hernia repairs in Ghana.  Hernia. 2007;11(5):389-391.PubMedGoogle ScholarCrossref
21.
Gawande  A, Weiser  T, Berry  W,  et al.  WHO Guidelines for Safe Surgery 2009. Geneva, Switzerland: World Health Organization; 2009.
22.
WorldPop. What is WorldPop? http://www.worldpop.org.uk/. Published 2015. Accessed July 30, 2015.
23.
OpenStreetMap. Welcome to OpenStreetMap! http://www.openstreetmap.org/. Published 2015. Accessed July 30, 2015.
24.
CERSGIS. Centre for Remote Sensing and Geographic Information Services, University of Ghana, Legon. http://www.cersgis.org/. Published 2015. Accessed July 30, 2015.
25.
Damsere-Derry  J, Afukaar  FK, Donkor  P, Mock  C.  Assessment of vehicle speeds on different categories of roadways in Ghana.  Int J Inj Contr Saf Promot. 2008;15(2):83-91.PubMedGoogle ScholarCrossref
26.
Derry  JD, Afukaar  FK, Donkor  P, Mock  C.  Study of vehicle speeds on a major highway in Ghana: implication for monitoring and control.  Traffic Inj Prev. 2007;8(2):142-146.PubMedGoogle ScholarCrossref
27.
Terzi  O, Sisman  A, Canbaz  S, Dündar  C, Peksen  Y.  A geographic information system-based analysis of ambulance station coverage area in Samsun, Turkey.  Singapore Med J. 2013;54(11):653-658.PubMedGoogle ScholarCrossref
28.
Polo  G, Acosta  CM, Ferreira  F, Dias  RA.  Location-allocation and accessibility models for improving the spatial planning of public health services.  PLoS One. 2015;10(3):e0119190. doi:10.1371/journal.pone.0119190.PubMedGoogle ScholarCrossref
29.
Weiser  TG, Regenbogen  SE, Thompson  KD,  et al.  An estimation of the global volume of surgery: a modelling strategy based on available data.  Lancet. 2008;372(9633):139-144.PubMedGoogle ScholarCrossref
30.
Nesbitt  RC, Gabrysch  S, Laub  A,  et al.  Methods to measure potential spatial access to delivery care in low- and middle-income countries: a case study in rural Ghana.  Int J Health Geogr. 2014;13:25.PubMedGoogle ScholarCrossref
31.
Stewart  BT, Gyedu  A, Abantanga  F, Abdulai  AR, Boakye  G, Kushner  A.  Barriers to essential surgical care in low- and middle-income countries: a pilot study of a comprehensive assessment tool in Ghana.  World J Surg. 2015;39(11):2613-2621.PubMedGoogle ScholarCrossref
32.
Asante  AD, Zwi  AB, Ho  MT.  Equity in resource allocation for health: a comparative study of the Ashanti and Northern regions of Ghana.  Health Policy. 2006;78(2-3):135-148.PubMedGoogle ScholarCrossref
33.
Gupta  S, Mahmood  U, Gurung  S,  et al.  Burns in Nepal: A population based national assessment.  Burns. 2015;41(5):1126-1132.PubMedGoogle ScholarCrossref
Original Investigation
August 17, 2016

Mapping Population-Level Spatial Access to Essential Surgical Care in Ghana Using Availability of Bellwether Procedures

Author Affiliations
  • 1Department of Surgery, University of Washington, Seattle
  • 2Department of Surgery, School of Medical Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
  • 3Department of Surgery, Komfo Anokye Teaching Hospital, Kumasi, Ghana
  • 4Department of Surgery, Dalhousie University, Halifax, Nova Scotia, Canada
  • 5Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, England
  • 6Information and Monitoring Unit, Ghana Health Service, Accra
  • 7Office of the Director General, Ghana Health Service, Accra
  • 8Pietermaritzburg Metropolitan Trauma Service, Pietermaritzburg, South Africa
  • 9Department of General Surgery, Nelson R. Mandela School of Medicine, University of Kwa-Zulu Natal, Kwa-Zulu Natal, South Africa
  • 10Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town, South Africa
  • 11Cochrane South Africa, South African Medical Research Council, Tygerberg
  • 12Harborview Injury Prevention & Research Center, Seattle, Washington
  • 13Department of Global Health, University of Washington, Seattle
JAMA Surg. 2016;151(8):e161239. doi:10.1001/jamasurg.2016.1239
Abstract

Importance  Conditions that can be treated by surgery comprise more than 16% of the global disease burden. However, 5 billion people do not have access to essential surgical care. An estimated 90% of the 87 million disability-adjusted life-years incurred by surgical conditions could be averted by providing access to timely and safe surgery in low-income and middle-income countries. Population-level spatial access to essential surgery in Ghana is not known.

Objectives  To assess the performance of bellwether procedures (ie, open fracture repair, emergency laparotomy, and cesarean section) as a proxy for performing essential surgery more broadly, to map population-level spatial access to essential surgery, and to identify first-level referral hospitals that would most improve access to essential surgery if strengthened in Ghana.

Design, Setting, and Participants  Population-based study among all households and public and private not-for-profit hospitals in Ghana. Households were represented by georeferenced census data. First-level and second-level referral hospitals managed by the Ministry of Health and all tertiary hospitals were included. Surgical data were collected from January 1 to December 31, 2014.

Main Outcomes and Measures  All procedures performed at first-level referral hospitals in Ghana in 2014 were used to sort each facility into 1 of the following 3 hospital groups: those without capability to perform all 3 bellwether procedures, those that performed 1 to 11 of each procedure, and those that performed at least 12 of each procedure. Candidates for targeted capability improvement were identified by cost-distance and network analysis.

Results  Of 155 first-level referral hospitals managed by the Ghana Health Service and the Christian Health Association of Ghana, 123 (79.4%) reported surgical data. Ninety-five (77.2%) did not have the capability in 2014 to perform all 3 bellwether procedures, 24 (19.5%) performed 1 to 11 of each bellwether procedure, and 4 (3.3%) performed at least 12. The essential surgical procedure rate was greater in bellwether procedure–capable first-level referral hospitals than in noncapable hospitals (median, 638; interquartile range, 440-1418 vs 360; interquartile range, 0-896 procedures per 100 000 population; P = .03). Population-level spatial access within 2 hours to a hospital that performed 1 to 11 and at least 12 of each bellwether procedure was 83.2% (uncertainty interval [UI], 82.2%-83.4%) and 71.4% (UI, 64.4%-75.0%), respectively. Five hospitals were identified for targeted capability improvement.

Conclusions and Relevance  Almost 30% of Ghanaians cannot access essential surgery within 2 hours. Bellwether capability is a useful metric for essential surgery more broadly. Similar strategic planning exercises might be useful for other low-income and middle-income countries aiming to improve access to essential surgery.

Introduction

Conditions that can be treated by surgery comprise more than 16% of the global disease burden.1 However, 5 billion people do not have access to essential surgical care, resulting in significant preventable death and disability.2 An estimated 90% of the 87 million disability-adjusted life-years incurred by surgical conditions could be averted by providing timely and safe surgical care in low-income and middle-income countries (LMICs).3 However, LMICs are least equipped to provide surgical care.4

Providing surgical care is cost-effective and feasible for almost all LMICs with sound planning and organization.3 To guide surgical care capacity building, the third edition of The World Bank’s Disease Control Priorities5 identified 44 surgical procedures as “essential” on the basis that the procedures (1) address a substantial burden of surgical conditions, (2) are highly cost-effective, and (3) are feasible to implement globally, regardless of national income level. Given these attributes, the package of essential surgical care has the potential to provide substantial improvements in population health if it was universally delivered.3

In 2015, the World Health Assembly ratified resolution 68.15: “Strengthening Emergency and Essential Surgical Care and Anesthesia as a Component of Universal Health Coverage.”6 The resolution calls for member states (ie, countries) to carry out regular monitoring and evaluation of the surgical care capacity of health care facilities and to collect and compile data on the number, type, and indications of surgical procedures performed to guide policy, planning, and development.7 To monitor and evaluate interventions to improve essential surgical care capacity and expand access, countries require useful benchmarks.8 Several benchmarks have been suggested, such as surgical case rate per population.9 The Lancet Commission on Global Surgery10 suggested that the performance of “bellwether procedures” (ie, open fracture repair, emergency laparotomy, and cesarean section) might be a useful proxy metric for the capability to provide essential surgical care more broadly. However, this assumption has not been well validated.11

Once essential surgical capacity has been assessed countrywide, areas and populations with good and poor access to timely care can be identified. Hospitals with large catchments that do not have the capability to provide essential surgical care could be targeted for capacity improvement. Such an assessment of essential surgical care capability and targeted planning of service expansion has not been performed in an LMIC. To address this gap, we aimed to evaluate first-level referral hospitals’ capability to perform essential surgical procedures, validate performance of the bellwether procedures as a proxy for hospitals’ capability to perform the breadth of essential surgery, and identify first-level referral hospitals that would most improve population-level spatial access to essential surgery if strengthened in Ghana.

Box Section Ref ID

Key Points

  • Question What percentage of the population in Ghana can potentially access essential surgical care (measured by hospitals that perform bellwether procedures [ie, open fracture repair, emergency laparotomy, and cesarean section]) within 2 hours?

  • Findings Population-level spatial access within 2 hours to a hospital that performed 1 to 11 and at least 12 of each bellwether procedure was assessed. Five hospitals were identified for targeted capability improvement.

  • Meaning Almost 30% of Ghanaians cannot access essential surgery within 2 hours, and similar strategic planning exercises might be useful for other countries aiming to improve access to essential surgery.

Methods
Setting

Ghana is a heavily indebted, lower-income to middle-income country in West Africa with an annual per capita income of US $1760 and a population of 26 million people.12 Although predominantly urbanized with several densely populated cities (eg, Accra, Kumasi, and Tamale), 47% of Ghana’s population live in rural areas.13 Only 13% of the country’s 109 515 km of roads are paved, and these thoroughfares are of varying quality.13 As proxies for conditions requiring essential surgery, injuries and obstetric emergencies are responsible for 45 and 9 deaths per 100 000 persons per year, respectively.14 Figures from the United Kingdom, an equivalently sized high-income country, are used for reference, where 82% live in urban centers and 100% of the 398 350 km of roads are paved.15 Furthermore, there are 36 injury deaths per 100 000 persons annually in the United Kingdom, and obstetric emergency deaths are rare (0.1 per 100 000 persons).14

First-level referral hospitals in Ghana are usually staffed by a medical officer and nurse anesthetist and have between 50 and 100 beds. Some rural districts do not have a first-level referral hospital and rely on that of a neighboring district. More densely populated districts have several first-level referral hospitals in each of their subdistricts.

Patients requiring more complex care are referred to a larger hospital (ie, 1 of 10 second-level referral hospitals or 1 of 4 tertiary hospitals). In addition, patients who reside, are injured, or become ill near a larger hospital might bypass first-level referral hospitals. Larger hospitals are staffed by a surgery-experienced medical officer, general surgeon, orthopedic surgeon, or obstetrician.

Public and private not-for-profit hospitals support 93% of the hospital beds in Ghana, and private for-profit hospitals comprise 7% of the hospital beds.16 Care at private for-profit facilities is prohibitively expensive for most of the population.17 Therefore, such facilities were excluded from this study.

Surgical Data

Surgical data were collected from January 1 to December 31, 2014. The number and type of surgical procedures performed in an operating theater at public and private non-for-profit referral hospitals in Ghana are reported to the Ministry of Health headquarters monthly using the District Health Information Software (DHIS) 2 platform.18 All procedures performed in 2014 were extracted from the Ministry of Health’s database and divided into the following 2 groups: (1) essential surgical procedures as defined by the third edition of The World Bank’s Disease Control Priorities5 and (2) other procedures. Similar data from regional and tertiary facilities are not reported. Ability to perform the bellwether procedures at larger hospitals had been previously determined by visits to each facility and by review of surgical logbooks.4

Ethics

This study was approved by the ethical committees of the Kwame Nkrumah University of Science and Technology (Kumasi, Ghana) and the University of Washington (Seattle), as well as by the Ghana Health Service and the Christian Health Association of Ghana.

Validating the Bellwether Procedures

First-level referral hospitals were categorized into the following 3 mutually exclusive groups: (1) those without capability to perform all 3 bellwether procedures, (2) those that performed 1 to 11 of each bellwether procedure per year, and (3) those that performed at least 12 of each bellwether procedure per year. The third group was included a priori by us to avoid inflating the capability in hospitals that intermittently receive surgical outreach teams but do not reliably provide these procedures to those in need19,20 and to attempt to highlight higher-volume hospitals because more frequent performance of procedures is highly correlated with safety.21

The number of essential surgical procedures performed by hospitals in each group was described. Wilcoxon rank sum test was used to identify differences between the number of essential surgical procedures performed at hospitals within each capability group. In addition, the rate of essential surgical procedures performed per 100 000 persons was calculated using district and submunicipal population data from the 2010 census adjusted to 2015 estimates.22

Geographic Information Services Data Management

The Ministry of Health provided a list of all hospitals in the country. Hospitals were geolocated using an online application (Google Earth; Google). National population data were represented by a 100-m2 gridded population surface generated by the WorldPop22 project, which was created using the 2010 census adjusted to 2015 estimates using United Nations population projections.

We created a national road network by combining data from OpenStreetMap23 and the Centre for Remote Sensing and Geographic Information Services, University of Ghana, Legon.24 Primary roads (ie, trunk roads connecting major cities) and secondary roads (ie, interregional routes) were obtained from OpenStreetMap. Tertiary roads (ie, minor roads and tracks) were obtained from the Centre for Remote Sensing and Geographic Information Services. After topological verification, travel speed was assigned to each road class based on national traffic laws as 100, 50, and 30 kmh−1 for primary, secondary, and tertiary roads, respectively.25,26

Population Access to Essential Surgical Care

Cost-distance analyses were performed to determine spatial access to essential surgery at all hospital levels countrywide (ie, first-level and second-level referral and tertiary hospitals).27 The “cost” that is implied in the model is not directly financial but is related to time or distance “spent” traveling from one point to another. The cost surface was constructed using the road network and travel speed. Background cells were assigned a value of 5 kmh−1 (ie, average walking speed). The cost surface was superimposed over the country. The value of each cell represented the time required to traverse that cell. The technique resulted in an output that represented the least time-consuming path from any point on the grid to the nearest first-level referral hospital that provided essential surgery (ie, first-level referral hospitals that performed 1-11 or those that performed ≥12 of each bellwether procedure in 2014). The proportion of the population with spatial access to care was evaluated by overlaying the gridded population surface over the cost-distance analyses and calculating the sum of the population within 1 hour and 2 hours of a first-level referral hospital capable of providing essential surgery.

To evaluate model stability, 2 additional cost surfaces were built in which travel speed for each road segment was augmented by ±20%. The analysis was repeated with each of these cost surfaces to create a lower estimate and an upper estimate of the population within each catchment (ie, uncertainty interval [UI]).

Location-Allocation Model

A location-allocation model was built using a software platform (ArcMap, version 10.0; Esri).28 The analysis identified 5 facilities from a candidate list of all first-level referral hospitals that performed less than 12 of each bellwether procedure per year that would most improve spatial access to essential surgery if strengthened. Five facilities were chosen as a reasonable starting point for capacity improvement a priori.

The population-weighted geographic centroid of each district represented demand points in the model. A second cost-distance analysis was performed after combining the candidate facilities and facilities that performed at least 12 of each bellwether procedure per year to quantify improvements in population-level spatial access after simulated capability strengthening.

Results

Of 155 first-level referral hospitals managed by the Ghana Health Service and the Christian Health Association of Ghana, 123 (79.4%) reported surgical data. Ninety-five (77.2%) did not have the capability in 2014 to perform all 3 bellwether procedures, 24 (19.5%) performed 1 to 11 of each bellwether procedure, and 4 (3.3%) performed at least 12 (Table 1).

Validating Bellwether Procedures as a Proxy for Essential Surgical Procedure Output

The median number of essential and other surgical procedures was highest in facilities that performed at least 12 of each bellwether procedure (median essential surgical procedures, 3198; interquartile range [IQR], 2365-4716 and median other surgical procedures, 384; IQR, 178-533) (Table 1). Similarly, hospitals that performed 1 to 11 of each bellwether procedure outperformed those without bellwether capability (median essential surgical procedures, 1157; IQR, 510-2569 vs 443; IQR, 0-1248 in hospitals with and without the capability to perform 1-11 of each bellwether procedure, respectively; P = .04).

Table 2 lists the percentages of first-level referral hospitals that performed selected procedures, which consist of both essential and other procedures. Hospitals that performed 1 to 11 of each bellwether procedure consistently performed more types of procedures than hospitals that did not have bellwether capability (22 of 24 [91.7%] selected procedures). Similarly, first-level referral hospitals that performed at least 12 of each bellwether procedure outperformed those that performed fewer bellwether procedures (20 of 24 [83.3%] selected procedures).

Bellwether Capability and Essential Surgical Procedure Rate

The rate of essential surgical procedures performed per 100 000 persons was greater in first-level referral hospitals capable of performing the bellwether procedures than in noncapable hospitals (median, 638; IQR, 440-1418 vs 360; IQR, 0-896 essential surgical procedures per 100 000 persons in hospitals with and without the capability to perform 1-11 of each bellwether procedure, respectively; P = .03) (eFigure in the Supplement). The rate was greater still in first-level referral hospital capable of performing at least 12 of each bellwether procedure, although not statistically significantly compared with first-level referral hospitals with capability to perform 1 to 11 of each bellwether procedure (median, 1183; IQR, 675-1603; P = .31).

Spatial Access to Essential Surgery

Population-level spatial access within 1 hour and 2 hours to a hospital that performed 1 to 11 of each bellwether procedure was 76.5% (UI, 72.2%-78.1%) and 83.2% (UI, 82.2%-83.4%), respectively (Table 3 and Figure 1). When limited to hospitals that performed at least 12 of each bellwether procedure, spatial access decreased to 49.8% (UI, 42.1%-54.2%) and 71.4% (UI, 64.4%-75.0%) of the population within 1 hour and 2 hours, respectively. Exclusion of larger hospitals (ie, second-level referral and tertiary hospitals) from the analysis of hospitals that performed at least 12 of each bellwether procedure resulted in a decrease in spatial access to essential surgical care (30.1%; UI, 25.1%-35.2% within 1 hour and 61.7%; UI, 52.7%-66.0% within 2 hours). This finding suggests that larger hospitals are often the nearest facility with essential surgical care capability for much of the population.

Effect of Strengthening First-Level Referral Hospitals

The 5 facilities identified as high-yield candidates for essential surgical care strengthening are shown in Figure 2. When the 5 candidates were incorporated into the cost-distance analysis with hospitals that performed at least 12 of each bellwether procedure, population-level spatial access to essential surgical care increased from 30.1% to 60.5% (UI, 52.2%-65.9%) and from 61.7% to 77.3% (UI, 73.4%-78.8%) of the population within 1 hour and 2 hours, respectively.

Discussion

This study aimed to evaluate first-level referral hospitals’ capability to perform essential surgery, validate the use of bellwether procedures as a proxy for essential surgical procedure capability more broadly, and identify facilities that would most improve population-level spatial access to essential surgery if strengthened in Ghana. First-level referral hospitals with capability to perform 1 to 11 of each bellwether procedure were significantly more likely to perform a greater number and higher rate of both bellwether and nonbellwether essential surgical procedures than hospitals without bellwether capability. In addition, we demonstrated that almost 30% of Ghanaians are unable to access essential surgery within 2 hours and that larger hospitals are often the nearest capable facilities (ie, not first-level referral hospitals). Last, 5 first-level referral hospitals that would most improve population-level spatial access to essential surgery if strengthened were identified.

A retrospective analysis of the World Health Organization Global Database for Emergency and Essential Surgical Care, which included data from 1357 facilities in 54 LMICs, demonstrated that at least 50% of bellwether-capable facilities frequently performed a range of other essential surgical procedures.11 While herein both groups (hospitals that performed 1-11 and hospitals that performed ≥12 bellwether procedures) of bellwether-capable first-level referral hospitals performed more procedures than non–bellwether-capable first-level referral hospitals in Ghana, less than 50% of all bellwether-capable hospitals performed a splenectomy, sequestrectomy, or burr hole or placed skeletal traction. Furthermore, the highest essential surgical procedure rate at a first-level referral hospital in Ghana was less than 4000 per 100 000 persons. This rate is significantly less than rates from health care systems that meet surgical demand.29 Therefore, it is likely that there remains a large burden of unmet surgical need in Ghana, even in populations near a bellwether-capable hospital. Thus, investment in essential surgery and capability in improving interventions are needed. Health officials can make use of bellwether capability and essential surgical procedure case rates to monitor changes in essential surgery provision over time.

From the same analysis of the World Health Organization Global Database for Emergency and Essential Surgical Care aforementioned, the estimated median distance to a bellwether-capable hospital was 35 km.11 Although we went further and described access by travel time, the findings from the present study suggest that population-level spatial access to essential surgical care in Ghana is similar or worse given that only 49% of the population could reach a bellwether-capable first-level referral hospital within 1 hour and almost 30% of the population cannot access safe essential surgical care within 2 hours. Using similar methods, Nesbitt et al30 demonstrated that the mean travel time for pregnant women in Brong Ahafo, one of Ghana’s 10 regions, to access a health care facility with comprehensive emergency obstetric care was 28 minutes. Therefore, access to essential surgery in Ghana may lag behind LMIC medians and obstetric care locally. Improving surgical capabilities in first-level referral hospitals identified by the location-allocation model would be an important first step to increasing access to essential surgical care in Ghana and averting otherwise preventable death and disability. Furthermore, the location-allocation modeling analysis could be serially performed to monitor and evaluate the effect of capability improvements on spatial access to essential surgery.

Countries planning to strengthen essential surgical care service delivery can use the modeling exercises demonstrated by this study to explore the relationship between the costs of capacity-building initiatives at 1 or more hospitals (eg, amount of money required to make essential surgical care available at a hospital when needed) and population-level spatial access to care in the countries’ context. Given limited funding for surgical care capacity building, the methods described herein may be particularly useful for identifying high-impact facilities to focus available resources.

In addition to relying on the modeling strategy to improve access to care in a cost-efficient manner, health system planners may also consider strengthening surgical care services in hospitals with some existing capacity rather than building capacity from nothing or in facilities without any surgical care services. For example, several of the non–bellwether-capable hospitals herein performed some essential surgical procedures (eg, laparotomy and suprapubic cystostomy), thus demonstrating surgical capacity. Therefore, they might be ideal targets for comparatively low-cost capability improvements (eg, trauma care training and provision of specific equipment or supplies). However, it must be noted that these findings do not validate bellwether capability as a proxy for subspecialty essential surgical procedures (eg, neurosurgical procedures, cleft lip or palate repair, obstetric fistula repair, cataract extraction, trichiasis surgery, and dentistry). These capabilities should not be overlooked when assessing and improving essential surgery capacity at hospitals, regardless of their bellwether capability. Last, the use of only population-level spatial access gaps to target hospitals for improvement neglects other important considerations, such as reducing nonspatial barriers to surgical care (eg, affordability and acceptability) and prioritizing health equity (eg, improving access to care for particularly marginalized or high-risk populations).31,32 The former (reducing nonspatial barriers to surgical care) should be done to ensure that patients are able to access care at a proximate and capable hospital when needed.

While this study describes the first-level referral hospital essential surgery capability countrywide in Ghana and used robust geographic information services methods to identify candidate hospitals for targeted capacity improvement, some limitations should be considered in interpreting the results. First, not all of the first-level referral hospitals in Ghana reported surgical data. The first-level referral hospitals that did not report surgical data were known to not routinely offer operating theater–based surgical care services; therefore, they were analyzed as if they did not have the capability to perform all 3 bellwether procedures. Second, procedure rates at first-level referral hospitals near larger hospitals may be misleading because some patients might have bypassed the first-level referral hospital tier and be effectively excluded from the population denominator. However, we demonstrated that first-level referral hospitals within the catchments of larger hospitals often do not have bellwether capabilities, evidenced by a 39% reduction in spatial access when larger hospitals were excluded from the analysis. Third, this analysis did not incorporate potential variations in prehospital transport mode (eg, private or commercial vehicle or national ambulance service), traffic patterns, or navigation errors. Although our model was stable with regard to changes in road travel speed, these additional factors might affect spatial access. Fourth, other nonspatial barriers to surgical care exist that prevent people in Ghana from seeking or accessing care.31 These other barriers require redress in parallel with strategically increasing the surgical care capacity to achieve maximum population-level benefit. Fifth, we did not consider the locoregional burden of surgical conditions alongside essential surgical capabilities at first-level referral hospitals. Some subpopulations might require a greater essential surgical case rate to meet surgical need than others.33 However, georeferenced surgical disease burden data are not available for Ghana to date. Despite these limitations, our findings allow reasonable conclusions to be drawn about the lack of essential surgical capacity in Ghana, the validity of using bellwether capability as a proxy for essential surgery capability more broadly, and the first-level referral hospitals that would most improve potential spatial access to essential surgery if strengthened.

Conclusions

Essential surgical capabilities are deficient in Ghana, as evidenced by the finding that 77.2% (95 of 123) of first-level referral hospitals were unable to perform the bellwether procedures. Furthermore, almost 30% of the population cannot access safe essential surgery within 2 hours. For the way forward, we identified 5 first-level referral hospitals where initiatives strengthening surgical care would have the greatest effect on population-level spatial access to essential surgery and should be prioritized. However, nonspatial barriers to surgical care, access to essential subspecialty procedures, and health equity must be considered when planning surgical capability improvement initiatives. Once such initiatives are under way, the frequency and rate of bellwether procedure performance can be used to monitor and evaluate essential surgery more broadly. Similar benchmarking and targeted planning exercises may be useful for other LMICs aiming to improve access to essential surgery.

Back to top
Article Information

Accepted for Publication: March 21, 2016.

Corresponding Author: Barclay T. Stewart, MD, MScPH, Department of Surgery, University of Washington, 1959 NE Pacific St, Ste BB-487, PO Box 356410, Seattle, WA 98195-6410 (stewarb@uw.edu).

Published Online: June 22, 2016. doi:10.1001/jamasurg.2016.1239.

Author Contributions: Dr Stewart had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Stewart, Tansley, Gyedu, Clarke, Mock.

Acquisition, analysis, or interpretation of data: Stewart, Tansley, Gyedu, Ofosu, Donkor, Appiah-Denkyira, Quansah, Clarke, Volmink.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: All authors.

Administrative, technical, or material support: All authors.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was funded by grants R25-TW009345 (Dr Stewart) and D43-TW007267 (Dr Mock) from the Fogarty International Center, National Institutes of Health.

Role of the Funder/Sponsor: The Fogarty International Center had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Additional Contributions: We thank the Ghana Health Service and the Christian Health Association of Ghana, as well as the hospitals and staff they represent, for their efforts and commitments to improving surgical and trauma care countrywide.

References
1.
GBD 2013 Mortality and Causes of Death Collaborators.  Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.  Lancet. 2015;385(9963):117-171.PubMedGoogle ScholarCrossref
2.
Alkire  BC, Raykar  NP, Shrime  MG,  et al.  Global access to surgical care: a modelling study.  Lancet Glob Health. 2015;3(6):e316-e323. doi:10.1016/S2214-109X(15)70115-4.PubMedGoogle ScholarCrossref
3.
Mock  CN, Donkor  P, Gawande  A, Jamison  DT, Kruk  ME, Debas  HT; DCP3 Essential Surgery Author Group.  Essential surgery: key messages from Disease Control Priorities, 3rd edition.  Lancet. 2015;385(9983):2209-2219.PubMedGoogle ScholarCrossref
4.
Stewart  BT, Quansah  R, Gyedu  A, Ankomah  J, Donkor  P, Mock  C.  Strategic assessment of trauma care capacity in Ghana.  World J Surg. 2015;39(10):2428-2440.PubMedGoogle ScholarCrossref
5.
Debas  HT, Donkor  P, Gawande  A, Jamison  DT, Kruk  ME, Mock  CN. Essential surgery. In: Jamison  DT, Nugent  R, Gelband  H, Horton  S, Jha  P, Laxminarayan  R, eds.  Disease Control Priorities. 3rd ed. Washington, DC: International Bank for Reconstruction and Development/The World Bank; 2015.
6.
Price  R, Makasa  E, Hollands  M.  World Health Assembly Resolution WHA68.15: “Strengthening Emergency and Essential Surgical Care and Anesthesia as a Component of Universal Health Coverage”—Addressing the Public Health Gaps Arising from Lack of Safe, Affordable and Accessible Surgical and Anesthetic Services.  World J Surg. 2015;39(9):2115-2125.PubMedGoogle ScholarCrossref
7.
Stewart  BT.  Editorial commentary on Bolkan et al: “The Surgical Workforce and Surgical Provider Productivity in Sierra Leone: A Countrywide Inventory” [published online February 23, 2016].  World J Surg. 2016.PubMedGoogle Scholar
8.
Mock  C.  Confronting the global burden of surgical disease.  World J Surg. 2013;37(7):1457-1459.PubMedGoogle ScholarCrossref
9.
Stewart  B, Wong  E, Papillon-Smith  J,  et al.  An analysis of cesarean section and emergency hernia ratios as markers of surgical capacity in low-income countries affected by humanitarian emergencies from 2008-2014 at Médecins Sans Frontières Operations Centre Brussels Projects.  PLoS Curr. 2015;7:ecurrents.dis.5e30807568eaad09a3e23282ddb41da6. doi:10.1371/currents.dis.5e30807568eaad09a3e23282ddb41da6.PubMedGoogle Scholar
10.
Gruen  R. Lancet Commission in Global Surgery. Information Management Working Group: progress report. http://www.globalsurgery.info/wp-content/uploads/2013/11/Russel-Gruen.pdf. Published June 2014. Accessed November 15, 2015.
11.
Meara  JG, Leather  AJ, Hagander  L,  et al.  Global Surgery 2030: evidence and solutions for achieving health, welfare, and economic development.  Lancet. 2015;386(9993):569-624.PubMedGoogle ScholarCrossref
12.
CIA World Factbook, Ghana. http://www.ciaworldfactbook.us/africa/ghana.html. Published 2014. Accessed April 12, 2015.
13.
The World Bank. Ghana: population growth. http://data.worldbank.org/country/ghana. Published 2015. Accessed January 14, 2015.
14.
Global Burden of Disease. Cause patterns. http://vizhub.healthdata.org/gbd-cause-patterns/. Updated September 14, 2015. Accessed May 8, 2016.
15.
The World Bank. United Kingdom. http://www.worldbank.org/en/country. Published 2014. Accessed May 6, 2015.
16.
Ghana Health Service. The health sector in Ghana: facts and figures. http://www.moh-ghana.org/UploadFiles/Publications/GHS%20Facts%20and%20Figures%202010_22APR2012.pdf. Published 2010. Accessed November 15, 2015.
17.
Amporfu  E.  Private hospital accreditation and inducement of care under the Ghanaian National Insurance Scheme.  Health Econ Rev. 2011;1(1):13.PubMedGoogle ScholarCrossref
18.
DHIS 2. Overview. https://www.dhis2.org/inaction. Accessed January 12, 2016.
19.
APRIDEC Medical Outreach Group. http://apridec.org/. Published 2010. Accessed May 12, 2016.
20.
Sanders  DL, Kingsnorth  AN.  Operation hernia: humanitarian hernia repairs in Ghana.  Hernia. 2007;11(5):389-391.PubMedGoogle ScholarCrossref
21.
Gawande  A, Weiser  T, Berry  W,  et al.  WHO Guidelines for Safe Surgery 2009. Geneva, Switzerland: World Health Organization; 2009.
22.
WorldPop. What is WorldPop? http://www.worldpop.org.uk/. Published 2015. Accessed July 30, 2015.
23.
OpenStreetMap. Welcome to OpenStreetMap! http://www.openstreetmap.org/. Published 2015. Accessed July 30, 2015.
24.
CERSGIS. Centre for Remote Sensing and Geographic Information Services, University of Ghana, Legon. http://www.cersgis.org/. Published 2015. Accessed July 30, 2015.
25.
Damsere-Derry  J, Afukaar  FK, Donkor  P, Mock  C.  Assessment of vehicle speeds on different categories of roadways in Ghana.  Int J Inj Contr Saf Promot. 2008;15(2):83-91.PubMedGoogle ScholarCrossref
26.
Derry  JD, Afukaar  FK, Donkor  P, Mock  C.  Study of vehicle speeds on a major highway in Ghana: implication for monitoring and control.  Traffic Inj Prev. 2007;8(2):142-146.PubMedGoogle ScholarCrossref
27.
Terzi  O, Sisman  A, Canbaz  S, Dündar  C, Peksen  Y.  A geographic information system-based analysis of ambulance station coverage area in Samsun, Turkey.  Singapore Med J. 2013;54(11):653-658.PubMedGoogle ScholarCrossref
28.
Polo  G, Acosta  CM, Ferreira  F, Dias  RA.  Location-allocation and accessibility models for improving the spatial planning of public health services.  PLoS One. 2015;10(3):e0119190. doi:10.1371/journal.pone.0119190.PubMedGoogle ScholarCrossref
29.
Weiser  TG, Regenbogen  SE, Thompson  KD,  et al.  An estimation of the global volume of surgery: a modelling strategy based on available data.  Lancet. 2008;372(9633):139-144.PubMedGoogle ScholarCrossref
30.
Nesbitt  RC, Gabrysch  S, Laub  A,  et al.  Methods to measure potential spatial access to delivery care in low- and middle-income countries: a case study in rural Ghana.  Int J Health Geogr. 2014;13:25.PubMedGoogle ScholarCrossref
31.
Stewart  BT, Gyedu  A, Abantanga  F, Abdulai  AR, Boakye  G, Kushner  A.  Barriers to essential surgical care in low- and middle-income countries: a pilot study of a comprehensive assessment tool in Ghana.  World J Surg. 2015;39(11):2613-2621.PubMedGoogle ScholarCrossref
32.
Asante  AD, Zwi  AB, Ho  MT.  Equity in resource allocation for health: a comparative study of the Ashanti and Northern regions of Ghana.  Health Policy. 2006;78(2-3):135-148.PubMedGoogle ScholarCrossref
33.
Gupta  S, Mahmood  U, Gurung  S,  et al.  Burns in Nepal: A population based national assessment.  Burns. 2015;41(5):1126-1132.PubMedGoogle ScholarCrossref
×