BMJ Open Is Committed to Open Peer Review. As Part of This Commitment We Make the Peer Review History of Every Article We Publish Publicly Available

BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on September 27, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

Interpreting Lancet Surgical Indicators in Somaliland

Journal: BMJ Open

Manuscript ID bmjopen-2020-042968 ArticleFor Type: peerOriginal research review only Date Submitted by the 20-Jul-2020 Author:

Complete List of Authors: Dahir, Shukri; Edna Adan University Hospital Cotache-Condor, Cesia; Baylor University, Department of Public Health Concepcion, Tessa; Duke University, Duke University Global Health Institute Mohamed, Mubarak; Edna Adan University Hospital Poenaru, Dan; MyungSung Christian Medical Center, Pediatric Surgery; Montreal Children\'s Hospital Research Institute, Pediatric Surgery Adan Ismail, Edna; Edna Adan Hospital Leather, Andy; King’s Health Partners and King’s College London, King’s Centre for Global Health Rice, Henry; Duke University, Global Health Institute; Duke University Medical Center, Department of Surgery Smith, Emily; Baylor University, Health, Human Performance, and Recreation; Duke University, Duke University Global Health Institute

SURGERY, Health policy < HEALTH SERVICES ADMINISTRATION & http://bmjopen.bmj.com/ Keywords: MANAGEMENT, PUBLIC HEALTH

on September 27, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open

1 2

3 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from 4 5 6 7 8 9 I, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined 10 in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors 11 who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance 12 with the terms applicable for US Federal Government officers or employees acting as part of their official 13 duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd (“BMJ”) its 14 licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the 15 Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence. 16 17 The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to 18 the Submitting Author Forunless you peer are acting as review an employee on behalf only of your employer or a postgraduate 19 student of an affiliated institution which is paying any applicable article publishing charge (“APC”) for Open 20 Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and 21 intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative 22 Commons licence – details of these licences and which Creative Commons licence will apply to this Work are set 23 out in our licence referred to above. 24 25 Other than as permitted in any relevant BMJ Author’s Self Archiving Policies, I confirm this Work has not been 26 accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate 27 material already published. I confirm all authors consent to publication of this Work and authorise the granting 28 of this licence. 29 30 31 32 33 34 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 27, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

1 1 2 3 Interpreting Lancet Surgical Indicators in Somaliland 4 5 6 Shugri Dahir1, Cesia Cotache-Condor2, Tessa L Concepcion3, Mubarak Mohamed1, Dan 7 8 4 1 5 3 2,3 9 Poenaru , Edna Adan Ismail , Andrew J. M. Leather , Henry E Rice , Emily R Smith , Global 10 11 Initiative for Children’s Surgery 12 13 14 1 Edna Adan University Hospital, Hargeisa, Somaliland, 2 Department of Public Health, Baylor 15 16 University, Waco, TX,For USA; peer3 Duke Global review Health Institute, onlyDuke University, Durham, NC, 17 18 USA; 4 McGill University, Montreal, Quebec, Canada; 5 King’s Centre for Global Health and 19 20 21 Health Partnerships, King’s College London, London, UK 22 23 24 25 26 27 Shugri Dahir and Cesia Cotache-Condor contributed equally to this paper. 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 Word count: 3,099 on September 27, 2021 by guest. Protected copyright. 42 43 44 Corresponding author: 45 46 47 Emily Smith, PhD 48 49 50 Robbins College of Health and Human Sciences, Baylor University, 1621 S. 5th Street, Waco, 51 52 TX 76706 53 54 [email protected] | +1-254-710-4028 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

2 1 2 3 ABSTRACT 4 5 6 7 8 Background: The unmet burden of surgical care is high in low- and middle-income countries. 9 10 The Lancet Commission on Global Surgery (LCoGS) proposed six indicators to guide the 11 12 13 development of national plans for improving and monitoring access to essential surgical care. 14 15 This study aimed to characterize the Somaliland surgical health system according to the LCoGS 16 For peer review only 17 indicators and provide recommendations for next-step interventions. 18 19 20 Methods: In this cross-sectional nationwide study, the World Health Organization’s Surgical 21 22 23 Assessment Tool – Hospital Walkthrough and geographic mapping were used for data collection 24 25 at 15 surgically-capable hospitals. LCoGS indicators for preparedness was defined as access to 26 27 timely surgery and specialist surgical workforce density (SAO), delivery was defined as surgical 28 29 30 volume, and impact was defined as protection against impoverishment and catastrophic 31 32 expenditure. Indicators were compared to the LCoGS goals and were stratified by region.

33 http://bmjopen.bmj.com/ 34 35 Results: The health care system in Somaliland does not meet any of the six LCoGS targets for 36 37 preparedness, delivery, or impact. We estimate that only 19% of the population has timely access 38 39 to essential surgery, less than the LCoGS goal of 80% coverage. The number of specialist 40

41 on September 27, 2021 by guest. Protected copyright. 42 surgical, anesthetic, and obstetric (SAO) providers is 0.8 per 100,000, compared to a LCoGS 43 44 goal of 20 SAO per 100,000. Surgical volume is 368 procedures per 100,000 people, while the 45 46 LCoGS goal is 5,000 procedures per 100,000. Protection against impoverishing expenditures was 47 48 49 only 18% and against catastrophic expenditures 1%, both far below the LCoGS goal of 100% 50 51 protection. 52 53 54 Conclusion: We found several gaps in the surgical system in Somaliland using the LCoGS 55 56 indicators and target goals. These metrics provide a broad view of current status and gaps in 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

3 1 2 3 surgical care, and can be used as benchmarks of progress towards universal health coverage for 4 5 6 the provision of safe, affordable, and timely surgical, obstetric, and anesthesia care in 7 8 Somaliland. 9 10 11 12 13 14 Keywords: global surgery, health policy, public health, LCoGS indicators 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

4 1 2 3 Strengths and limitations of the study 4 5 6  This is the first nationwide study that comprehensively evaluates the surgical system, 7 8 9 including the public and private sectors, in Somaliland. 10 11  This study follows the Lancet Commission on Global Surgery indicators to evaluate 12 13 surgical preparedness, delivery, and impact in Somaliland. 14 15 16  Unlike previousFor studies peer based solely review on modeling, we only attempted to collect primary data 17 18 on financial costs and included lower and upper limits in our expenditure calculations to 19 20 account for a wide range of surgical costs. 21 22 23  Data on perioperative mortality rates were not included as most hospitals in Somaliland 24 25 do not collect this data. 26 27  Our reported surgical workforce density did not include data on non-specialist physicians 28 29 30 and non-physician clinicians who may provide surgical care in Somaliland. 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

5 1 2 3 INTRODUCTION 4 5 6 Over 5 billion people around the world lack safe, affordable, and timely access to surgical and 7 8 anesthesia care, with the overwhelming burden being among persons in low- and middle-income 9 10 countries (LMICs).1-3 To address this crisis, targeted and accurate measurement of surgical care 11 12 is a critical first step. The 2015 Lancet Commission on Global Surgery (LCoGS) proposed six 13 14 15 surgical indicators through an extensive and iterative process led by experts from various 16 For peer review only 17 disciplines, surgical specialties, and countries.1 These six indicators and associated target goals 18 19 define the preparedness for surgical and anesthesia care (access to essential surgery and 20 21 22 workforce density), delivery of care (quantifying the surgical volume and perioperative mortality 23 24 rates), and impact of care (protection against impoverishing and catastrophic expenditures).1 25 26 These indicators have been used as a template to assess surgical care in several countries, which 27 28 29 in turn have supported the development of national surgical, obstetrical, and anesthesia plans 30 31 (NSOAPs).4 5 32

33 http://bmjopen.bmj.com/ 34 Somaliland is a low-income country in Sub-Saharan Africa with significant health and economic 35 36 challenges. After emerging from the civil war in 1991, the government of Somaliland focused on 37 38 rebuilding its hospitals, clinics and health posts, and devoted 3% of the national budget to rebuild 39 40 6 7

41 its health sector. Somaliland has a GDP of $348 per capita (making it the fourth poorest on September 27, 2021 by guest. Protected copyright. 42 43 country in the world), has a population of approximately 4 million people, and some of the 44 45 poorest health outcomes in the Africa Region.8-14 Our previous work and that of others has 46 47 14-16 48 characterized the surgical infrastructure for children in the country. The goal of the current 49 50 study was to characterize the Somaliland surgical health system according to the LCoGS surgical 51 52 indicators. 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

6 1 2 3 METHODS 4 5 6 Setting 7 8 The study took place in Somaliland, a country in the Horn of Africa which achieved relative 9 10 11 stability and has an autonomous government since separation from Somalia in 1991. The 12 13 population estimate from the 2014 Somali census is 3.5 million people.17 Somaliland is made up 14 15 of six regions – Awdal, Maroodi Jeex (containing the capital city of Hargeisa), Sahil, Sanaag, 16 For peer review only 17 Sool, and Togdheer – with 40% of the population living in Hargeisa.18 The public health system 18 19 20 is comprised of primary health units, health centers, referral health centers, regional hospitals, 21 22 and one national referral hospital. Specialty hospitals include seven tuberculosis hospitals, one 23 24 female fistula hospital, one children’s hospital, and one mental health hospital. An estimated 25 26 27 60% of health services are provided through the private sector, consisting of non-government 28 29 organizations, private health practitioners and hospitals, and traditional practitioners.7 30 31 32 Hospital selection

33 http://bmjopen.bmj.com/ 34 Hospitals were included in the study if they had the capacity to perform surgery (defined as the 35 36 presence of at least one operating room), yielding 16 hospitals. We were not able to obtain data 37 38 39 from one private hospital, resulting in a final study cohort of 15 hospitals. Our in-country 40

41 research manager asked permission from each hospital director or senior administrator for on September 27, 2021 by guest. Protected copyright. 42 43 participation in the study, and presented the study’s letter of approval from the Ministry of 44 45 46 Health. 47 48 49 Data collection 50 51 52 We assessed hospital capacity at each facility using the World Health Organization’s (WHO) 53 54 Surgical Assessment Tool – Hospital Walkthrough and the Global Initiative for Children’s 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

7 1 2 3 Surgery (GICS) Global Assessment in Pediatric Surgery (GAPS) tool.19 20 The survey included 4 5 6 data on service delivery, the number of admissions and surgical procedures per month, whether 7 8 the facility tracked perioperative in-hospital mortality, and the number of laparotomies, cesarean 9 10 sections, and open fracture repairs (i.e., the three LCoGS “bellwether” procedures). To assess 11 12 each facility’s workforce, the survey quantified the number of staff, including surgeons, 13 14 15 anesthesiologists, non-physician anesthetists, obstetricians, nurses, and administrative staff. Each 16 For peer review only 17 staff member was counted only once even if they worked across multiple health facilities. To 18 19 assess each facility’s infrastructure, the survey asked about the availability of essential resources 20 21 22 (i.e., electricity, running water, oxygen), the number of hospital beds and operating rooms, and 23 24 the number of anesthesia machines and ventilators. 25 26 27 Indicator assessment 28 29 30 The LCoGS indicators includes 6 items pertaining to surgical care preparedness (Indicators 1 and 31 32 2), delivery (Indicators 3 and 4), and impact (Indicators 5 and 6) (Table 1). We did not assess

33 http://bmjopen.bmj.com/ 34 whether each hospital tracked perioperative mortality rates, as this metric is generally not 35 36 collected in Somaliland. Thus, Indicator 4 was not included in the analysis. Data tabulation and 37 38 39 analyses were performed using Microsoft Excel and SAS 9.3 (Cary, NC). Geographic mapping 40

41 and analyses were performed using ArcGIS (Environmental Systems Research Institute (ESRI) on September 27, 2021 by guest. Protected copyright. 42 43 2012, Redlands, CA). 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

8 1 2 3 Table 1. Lancet Commission on Global Surgery indicator definitions and targets.1 4 5 6 7 INDICATOR DEFINITION TARGET 8 1. Access to timely Proportion of the population that can A minimum of 80% coverage of 9 essential surgery access, within 2 hours, a facility that can essential surgical and anesthesia services do caesarean delivery, laparotomy, and per country by 2030 10 treatment of open fracture (the Bellwether 11 Procedures) 12 2. Specialist surgical Number of specialist surgical, anesthetic, 100% of countries with at least 20 13

Preparedness workforce density and obstetric physicians who are working, surgical, anesthetic, and obstetric 14 per 100 000 population physicians per 100 000 population by 15 2030 16 3. Surgical volumeForProcedures peer done in anreview operating theatre, only80% of countries by 2020 and 100% of 17 per 100 000 population per year countries by 2030 tracking surgical 18 volume; a minimum of 5000 procedures 19 per 100 000 population by 2030 20 4. Perioperative All-cause death rate before discharge in 80% of countries by 2020 and 100% of

21 Delivery mortality patients who have undergone a procedure countries by 2030 tracking perioperative 22 in an operating theatre, divided by the mortality; in 2020, assess global data 23 total number of procedures (%) and set national targets for 2030 24 5. Protection against Proportion of households protected 100% protection against impoverishment 25 impoverishing against impoverishment from direct out- from out-of-pocket payments for 26 expenditure of-pocket payments for surgical and surgical and anesthesia care by 2030 27 anesthesia care 28 6. Protection against Proportion of households protected 100% protection against catastrophic 29 Impact catastrophic against catastrophic expenditure from expenditure from out-of-pocket 30 expenditure direct out-of-pocket payments for surgical payments for surgical and anesthesia 31 and anesthesia care care by 2030 32

33 http://bmjopen.bmj.com/ 34 Indicator one: Access to timely essential surgery. Geographic locations of the 15 surgically- 35 36 37 capable hospitals, defined by the presence of an operating room, were recorded using a smart 38 39 phone GPS (Global Positioning System) application. Geographical location was analyzed using 40

41 ArcGIS to determine the percent of the population within a 2-hour radius of the hospitals on September 27, 2021 by guest. Protected copyright. 42 43 throughout the country. A travel rate of 30 kilometers per hour was used to determine the 44 45 46 geographical area within a two-hour drive of sites with surgical capacity using available 47 48 Somaliland road infrastructure data. Population estimates within each geographical area were 49 50 made using data from the 2014 Somalia census.17 Surgical capacity was defined as the existence 51 52 53 of an operating room within the hospital. This geospatial method did not account for 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

9 1 2 3 transportation wait time or topography of the land. However, travel times were reviewed by the 4 5 6 in-country data team to confirm validity of the estimated travel times between locations. 7 8 9 Indicator 2: Surgical workforce density. The surgical workforce density was defined as the 10 11 number of specialist surgeons, anesthesiologists, and obstetricians (SAO) per 100,000 12 13 population. Surgeons included specialties available within Somaliland (general surgeons, 14 15 colorectal surgeons, neurosurgeons, and otolaryngologists). Anesthesiologists did not include 16 For peer review only 17 18 nurse anesthetists. The surgical workforce density was calculated as the number of SAOs per 19 20 100,000 population. In addition to the national SAO workforce density, region-specific densities 21 22 were also estimated. 23 24 25 Indicator 3: Surgical volume. Surgical volume was obtained through the country-wide hospital 26 27 28 assessment. We included all procedures recorded in hospital logbooks between January 1, 2017 29 30 and December 31, 2017 using previously described methods.15 Exclusion criteria included 31 32 procedures performed outside of the operating rooms. We quantified the number of the LCoGS-

33 http://bmjopen.bmj.com/ 34 35 defined bellwether procedures (open fracture repairs, Cesarean section deliveries, and 36 37 laparotomies).1 38 39 40 Indicator 5: Protection against impoverishing expenditure. Protection against impoverishing

41 on September 27, 2021 by guest. Protected copyright. 42 expenditure was defined as 100 - % population at risk of impoverishing expenditures. 43 44 Impoverishment was defined as any income below the poverty line after out-of-pocket payments, 45 46 47 including medical and non-medical costs, for surgery. The threshold for impoverishment 48 49 expenditure for Somaliland was defined as any income below a value of: 50 51 52 [% out of pocket expenditure] * [average cost of surgery] + [Somaliland poverty line]. 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

10 1 2 3 For costs per operation, we used $200 as a proxy due to lack of information of operative costs for 4 5 6 each hospital. The $200 value was chosen as it represents the mean value between costs per 7 8 surgical procedure in a low-income country ($179) and a lower-middle income country ($219) 9 10 calculated as previously described.21 The $200 costs covers all direct costs, including 11 12 preoperative and postoperative hospitalization as well as the recurrent costs of running a surgical 13 14 15 service such as salaries, utilities, equipment, medical supplies and medicine for the procedure 16 For peer review only 17 itself. We confirmed this amount based on out-of-pocket expenses from one private hospital in 18 19 Somaliland, which used an average of $200 per procedure. Non-medical expenses, such as 20 21 22 22 transportation and lodging, were estimated at $60 based on previous literature. The expected 23 24 proportion of out-of-pocket costs for healthcare was provided by the hospital administration 25 26 during the survey data collection. The threshold for the poverty line in Somaliland was set at 27 28 8 29 $220 based on previous literature. 30 31 32 Indicator 6: Protection against catastrophic expenditure. Protection against catastrophic

33 http://bmjopen.bmj.com/ 34 expenditures was defined as 100% - population at risk of catastrophic expenditure.1 Catastrophic 35 36 expenditure was defined as direct out-of-pocket expenses greater than 40% of the gross domestic 37 38 product (GDP) per capita, excluding subsistence costs. The risk of catastrophic expenditure for 39 40

41 each hospital and region was determined by dividing all direct and indirect out-of-pocket on September 27, 2021 by guest. Protected copyright. 42 43 expenses by $348, Somaliland’s GDP per capita.22 44 45 46 Patient and Public involvement 47 48 49 Patients and/or the public were not involved in the design, or conduct, or reporting, or 50 51 dissemination plans of this research 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

11 1 2 3 RESULTS 4 5 6 The surgical system in Somaliland did not reach any of the target indicator goals as defined by 7 8 LCoGS, with each indicator lagging behind its target by 76% or more (Figure 1). The greatest 9 10 11 need was for protection against catastrophic expenditure and inadequate surgical volume. 12 13 Preparedness 14 15 16 Only an estimated 19%For of persons peer in Somaliland review live within aonly 2-hour access to a surgically- 17 18 19 capable facility, translating to 24% of the LCoGS goal of at least 80% population coverage 20 21 (Figure 2). The number of specialist SAO physicians was 0.8 per 100,000 population, or 4% of 22 23 the LCoGS goal of 20 SAOs per 100,000. Access to surgery ranged from 36% in Maroodi Jeex 24 25 to below 6% in Awdal, Sool, and Sanaag. Of the total workforce of 32 SAOs in the country, 15 26 27 28 were surgeons, 3 were anesthesiologists, and 14 were obstetricians (Table 2). The number of 29 30 surgeons ranged from 10 in Maroodi Jeex to 0 in Sool, Togdheer, or Sanaag, while the number of 31 32 obstetricians ranged from 8 in Maroodi Jeex to 0 in Sahil. Of the 3 anesthesiologists, all 3 were

33 http://bmjopen.bmj.com/ 34 35 located in Maroodi Jeex (the capital city). 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

12 1 2 3 Table 2. Surgeon, anesthesia, and obstetrics (SAO) workforce density and procedures by region. 4 5 6 Surgical Anesthesiologists (n) Obstetricians (n) Total workforce (n) 7 providers (n)* 8 All regions 15 3 14 32 9 Maroodi Jeex 10 3 8 21 10 Awdal 4 0 2 6 11 Sahil 1 12 1 0 0 13 Sanaag 0 0 1 1

14 PREPAREDNESS Sool 0 0 1 1 15 Togdheer 0 0 2 2 16 ForLaparotomies peerCesarean review Sections onlyOpen fracture Total procedures 17 (month) (month) repairs (month) (month) 18 All regions 0-11 0-100 0-40 1187 19 Maroodi Jeex 0-11 15-100 0-32 681 20 Awdal 5-20 10-15 0-40 250 21 Sahil 0-1 5-10 0 58

22 DELIVERY Sanaag 0-3 0-10 0-7 30 23 Sool 0-1 0-6 0 25 24 Togdheer 2-3 25-26 0 143 25 *Surgeons included 10 general surgeons, 1 colorectal, 3 ENT surgeons, and 1 neurosurgeon. 26 27 28 29 30 Delivery 31 32 The surgical volume was 368 procedures per 100,000 people, i.e. less than the 7% of the LCoGS

33 http://bmjopen.bmj.com/ 34 35 goal of 5,000 procedures per 100,000 population. Surgical volume ranged from 646 per 100,000 36 37 population in Sahil to 66 per 100,000 population in Sanaag (Figure 3). Total volume of 38 39 procedures ranged from 681 in Maroodi Jeex to 250 in Awdal (Table 2). Of these procedures, 40

41 on September 27, 2021 by guest. Protected copyright. 42 laparotomies ranged between 1 - 20 per month (in Sahil or Sool and Awdal, respectively), 43 44 Cesarean sections between 6 - 100 per month (in Sool and Maroodi Jeex, respectively), and open 45 46 fracture repairs between 0 - 40 per month (in Sahil, Sool, or Togdheer and Awdal, respectively). 47 48 49 Impact 50 51 52 Protection from impoverishment was only 18% of the LCoGS goal of 100% protection by 2030. 53 54 Risk of impoverishment was high throughout the country and ranged between 74.5% in Sahil 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

13 1 2 3 and 84.3% in Maroodi Jeex (Figure 3). Protection against catastrophic expenditures was 1% of 4 5 6 the LCoGS goal of 100% protection by 2030. The risk of catastrophic expenditure was greater 7 8 than 90% throughout the entire country. 9 10 11 DISCUSSION 12 13 As surgical care is increasingly recognized as an essential component of a functional healthcare 14 15 system, understanding the surgical landscape of a country is vital for health care scale-up plans 16 For peer review only 17 18 contextualized to that country. The LCoGS indicators measure several domains of surgical care 19 20 related to preparedness, delivery, and impact of care. These indicators are increasingly used 21 22 around the world as a framework to scale up surgical care in the country through a 23 24 comprehensive, system-level perspective.23 We found that Somaliland’s surgical system achieves 25 26 27 only a small fraction of these target goals, with the greatest need found in rural areas. 28 29 30 Our study found large gaps in surgical care preparedness, delivery, and impact. Preparedness, 31 32 including access to surgical care and workforce density, are far below the LCoGS targets,

33 http://bmjopen.bmj.com/ 34 consistent with other sub-Saharan African countries.3 4 23-33 The reasons for lack of access to 35 36 37 surgical care in Somaliland is multifactorial, but is largely due to the limited number of locally- 38 39 accessible surgically-capable hospitals and existing road networks. The majority of surgically- 40

41 capable hospitals are within the urban Maroodi Jeex region, with only 1 hospital located in the on September 27, 2021 by guest. Protected copyright. 42 43 44 more rural regions. We found the geographic access to timely surgical care is deficient across 45 46 Somaliland, particularly in the Northwest and Southeast areas of the country. After almost a 47 48 decade of civil conflict (1982-1991), the country was left with severe damage and deterioration 49 50 to road networks and healthcare infrastructure,34 and only 15% of the primary roads in urban 51 52 35 53 areas and networks connecting rural to urban areas are considered to be in good condition. 54 55 Workforce density is also low in the more rural parts of the country, with most of the SAOs 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

14 1 2 3 concentrated within urban areas. Delivery of surgical care as measured by the volume of 4 5 6 procedures was low particularly in rural areas and is likely related to the workforce density 7 8 within each area. For example, 57% of all procedures per month occurred in the urban Maroodi 9 10 Jeex region which has 65% of all SAOs. 11 12 13 The financial impact of surgical care is evident through the high rates of financial strain on 14 15 families, which limits Somaliland’s capacity to attain universal health coverage (UHC) as 16 For peer review only 17 18 advocated by the World Bank, UN and WHO. The provision of surgery in Somaliland is largely 19 20 dependent on a person’s financial means to pay for out-of-pocket (OOP) expenses, similar to 21 22 high rates of OOP spending in many LMICs.36 The risk of being pushed into poverty due to 23 24 16 25 OOP expenses for surgical care in Somaliland is one of the highest compared to other LMICs, 26 27 which is particularly challenging as Somaliland is the fourth poorest country in the world.36-40 28 29 Our findings highlight the need for inclusion of surgical care in the country’s UHC scheme in 30 31 32 order to prevent impoverishing and catastrophic expenditure due to OOP expenses, especially

33 http://bmjopen.bmj.com/ 34 among at-risk communities.37-39 35 36 37 As a post-conflict country, the healthcare system Somaliland experiences many structural 38 39 challenges, with just 3% of the national budget designated to the health care. In this context, 40

41 outlining a strategic national health care plan to improve access to surgery is required. Such a on September 27, 2021 by guest. Protected copyright. 42 43 44 plan should address deficiencies in the surgical system as defined by the LCoGS indicators, 45 46 including preparedness, delivery, and impact. Each LCoGS indicator should be interpreted in 47 48 conjunction with other indicator metrics to fully understand how gaps in care are interrelated .40 49 50 51 Evaluating surgical capacity begins with accessibility. However, geographical coverage is just 52 53 one dimension of access and does not ensure that people receive proper surgical care.23 41-46 54 55 Patients that eventually reach healthcare facilities might not be able to receive surgical care due 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

15 1 2 3 to deficits in infrastructure and workforce.47 48 Thus, interpreting each indicator independently of 4 5 6 the others will only provide a limited understanding of the surgical needs within a country. 7 8 9 One way to comprehensively evaluate surgical systems and gaps in care on a national level are 10 11 through country-specific National Surgical, Obstetric, and Anesthesia Plans (NSOAPs). The 12 13 purpose of NSOAPs are to improve surgical care through the six major building blocks of a 14 15 health system: infrastructure, workforce, service delivery, information management, finance, and 16 For peer review only 17 1 18 leadership/governance. One of the first steps in developing a country-specific NSOAP is 19 20 identifying gaps for each building block.5 Although Somaliland does not have a NSOAP, our 21 22 paper provides critical data regarding infrastructure, workforce, service delivery, and the need 23 24 25 for financing schemes to protect individuals from catastrophic and impoverishing expenditures. 26 27 28 Somaliland may be able to benefit from lessons learned in other LMICs during enactment of 29 30 NSOAPs. The Republic of Zambia was one of the first countries to conduct a baseline 31 32 assessment of the LCoGS indicators and utilize those in a quality improvement plan within a

33 http://bmjopen.bmj.com/ 34 49-51 49 52 53 49 54 49 55 35 NSOAP. To date, other countries including Ethiopia, India, and Tanzania, have 36 37 adopted a similar process to develop their NSOAPs. These experiences suggest that Somaliland 38 39 may benefit from upgrading geographically strategic facilities and allocating resources to high- 40

41 impact healthcare facilities to ensure equity and reinforce existing surgical capacity.56-58 Task on September 27, 2021 by guest. Protected copyright. 42 43 44 shifting and task sharing, partnerships with other countries and the private sector, culturally 45 46 embedded training, and increase of incentives are other ways to efficiently scale up workforce 47 48 density.1 59 We encourage continuous data collection as part of a NSOAP anchored within the 49 50 51 overall country’s national health strategic plan, and driven by the Somaliland Ministry of Health 52 53 to ensure standardization, integration and sustainability over time.5 23 Furthermore, engagement 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

16 1 2 3 of local champions and health care workers on the ground from the early stages will be pivotal to 4 5 49 60-62 6 ensure success and continuity. 7 8 9 Successful implementation of NSOAPs also require involving strong governmental leadership 10 11 and other stakeholders, establishing surgical priorities, and allocating funding from partnerships 12 13 and non-traditional sources.49 52 One example which is be helpful for Somaliland is that of the 14 15 Saving Lives Through Safe Surgery (SaLTS) in Ethiopia, which has shown the value of 16 For peer review only 17 18 established a comprehensive system of educational/training packages, leadership/mentoring 19 20 programs, national perioperative guidelines, implementation of the WHO Surgical Safety 21 22 Checklist, and requirements for functional operating rooms, standardized equipment, supplies, 23 24 52 25 and purchasing. In Tanzania, the NSOAP emphasizes the coverage of essential surgical care in 26 27 national health insurance schemes, implementation of electronic medical records for surgery- 28 29 related data, simultaneous strengthening of lower-level and higher-level hospitals to reinforce 30 31 32 referral pathways, and implementation of regional training to improve local retention of SAO

33 http://bmjopen.bmj.com/ 34 providers.55 35 36 37 Limitations 38 39 40 Our study has several limitations. First, our reported surgical workforce density might

41 on September 27, 2021 by guest. Protected copyright. 42 underrepresent the entire surgical workforce in Somaliland, we did not collect data on non- 43 44 specialist physicians and non-physician clinicians who may provide surgical care. Second, most 45 46 47 hospitals in Somaliland do not collect data on perioperative mortality rates. This indicator is the 48 49 most widely adopted measurement for surgical outcomes and highlight delays to receiving 50 51 care.23 Similarly, we did not measure quality of care within the hospital settings, nor were able to 52 53 54 assess the impact of surgical capacity on mortality. Third, our data on impoverishing and 55 56 catastrophic expenditures may have been based on inaccurate surgical costs. Few studies in other 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

17 1 2 3 countries have accurate primary data on surgery related costs and out-of-pocket expenditures, as 4 5 22 25 63 6 most are based on modeling. Although we collected limited primary data on financial 7 8 costs, these reported costs likely underestimate the total costs of procedures, including indirect 9 10 costs of lost earnings. However, we attempted to account for the wide range of surgical costs 11 12 depending on the procedure by including lower and upper limits in our expenditure calculations. 13 14 15 These findings emphasize the need for improved primary data collection regarding out-of-pocket 16 For peer review only 17 surgical costs in LMICs. 18 19 20 Probably the most important limitation to our analysis is the challenge among the LCoGS 21 22 indicators to correctly measure their respective components. The LCoGS indicators were 23 24 25 originally defined using data from nationally representative 194 WHO member states. However, 26 27 all indicators except for surgical workforce included data from less than half of WHO member 28 29 states.23 In addition, many countries had differing definitions of indicators, hindering 30 31 23 32 comparability and potentially leading to erroneous estimates. Taking access to care as an

33 http://bmjopen.bmj.com/ 34 example, geographic proximity is only one dimension of access and does not necessarily account 35 36 for the impact of road conditions or type of transportation. Similarly, the definition of surgical 37 38 volume did not take into account varying population needs depending on the specific country. 39 40

41 Perhaps the most potentially biased indicator is expenditure, as this indicator was defined solely on September 27, 2021 by guest. Protected copyright. 42 43 on modeling estimates with no national-level data. Although these challenges may have affected 44 45 our estimates, our data can serve to drive collection of country-wide surgical data in the future. 46 47 48 Conclusion 49 50 51 As Somaliland continues to rebuild its healthcare infrastructure, incorporating surgical care as an 52 53 54 essential component of its national health care plan is critical. We found large deficits in the 55 56 overall surgical system using the LCoGS indicators. Implementation of efforts to scale-up 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

18 1 2 3 surgery using these target goals can help Somaliland progress in improving safe, affordable, and 4 5 6 timely surgical, obstetric, and anesthesia care and achieve universal health coverage for its 7 8 population. 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

19 1 2 3 Autor contributorship: 4 5 6 SD, TC, and MM oversaw all data collection efforts. CCC, TC and ES conceptualized the 7 8 9 design, analyzed the data, and contributed to the writing and revising of the manuscript. All 10 11 authors contributed to the writing and revising of the manuscript. 12 13 14 Competing Interests: None declared. 15 16 For peer review only 17 Funding: None. 18 19 20 Ethical approval: Institutional review board (IRB) approval was granted from Duke University. 21 22 Since Somaliland does not have a national IRB, a letter of approval for the study was obtained 23 24 from the Somaliland Ministry of Health. 25 26 27 Data sharing statement: Data are available by request by emailing [email protected] 28 29 30 Acknowledgements: We want to thank the Global Initiative for Children’s Surgery (GICS) for 31 32 its support of this work. GICS (www.globalchildrenssurgery.org) is a network of children’s 33 http://bmjopen.bmj.com/ 34 35 surgical and anesthesia providers from low-, middle-, and high-income countries collaborating 36 37 for the purpose of improving the quality of surgical care for children globally. 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

20 1 2 3 REFERENCES 4 5 6 1. Meara J, Leather A, Hagander L, et al. Global Surgery 2030: evidence and solutions for 7 8 9 achieving health, welfare, and economic development. The Lancet 2015;386(9993):569- 10 11 624. doi: 10.1016/S0140-6736(15)60160-X 12 13 2. Fuller AT, Corley J, Tran TM, et al. Prevalence of Surgically Untreated Face, Head, and Neck 14 15 Conditions in Uganda: A Cross-Sectional Nationwide Household Survey. World 16 For peer review only 17 18 Neurosurgery 2018;110:e747-e54. doi: 10.1016/j.wneu.2017.11.099 [published Online 19 20 First: 2017/11/29] 21 22 3. Bearden A, Fuller AT, Butler EK, et al. Rural and urban differences in treatment status among 23 24 25 children with surgical conditions in Uganda. PloS one 2018;13(11):e0205132. doi: 26 27 10.1371/journal.pone.0205132 [published Online First: 2018/11/02] 28 29 4. Bruno E, White M, Baxter L, et al. An Evaluation of Preparedness, Delivery and Impact of 30 31 32 Surgical and Anesthesia Care in Madagascar: A Framework for a National Surgical Plan.

33 http://bmjopen.bmj.com/ 34 World Journal of Surgery 2017;41(5):1218-24. doi: 10.1007/s00268-016-3847-9 35 36 5. Sonderman KA, Citron I, Mukhopadhyay S, et al. Framework for developing a national 37 38 surgical, obstetric and anaesthesia plan. BJS Open 2019;3(5):722-32. doi: 39 40

41 10.1002/bjs5.50190 on September 27, 2021 by guest. Protected copyright. 42 43 6. Elkheir N, Sharma A, Cherian M, et al. A cross-sectional survey of essential surgical capacity 44 45 in Somalia. BMJ open 2014;4(5):1-8. doi: 10.1136/bmjopen-2013-004360 46 47 48 7. Somaliland Ministry of Health. National health policy 2011 [Available from: 49 50 http://www.nhpcsomaliland.org/documents/Somaliland%20National%20Health_Policy.p 51 52 df accessed July 16, 2019. 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

21 1 2 3 8. World Bank. New World Bank GDP and poverty estimates for Somaliland 2014 [Available 4 5 6 from: https://www.worldbank.org/en/news/press-release/2014/01/29/new-world-bank- 7 8 gdp-and-poverty-estimates-for-somaliland accessed July 16, 2019. 9 10 9. World Bank. Somaliland: poverty profile and overview of living conditions 2015 [Available 11 12 from: http://www.somalilandcsd.org/wp-content/uploads/2017/12/Somaliland-Poverty- 13 14 15 Profile..pdf accessed July 17, 2019. 16 For peer review only 17 10. United Nations International Children's Emergency Fund. Under-five mortality 2018 18 19 [Available from: https://data.unicef.org/topic/child-survival/under-five-mortality/ 20 21 22 accessed July 20, 2019. 23 24 11. United Nations International Children's Emergency Fund. Neonatal mortality 2018 25 26 [Available from: https://data.unicef.org/topic/child-survival/neonatal-mortality/ accessed 27 28 29 July 20, 2019. 30 31 12. World Bank. Africa development indicators 2012/13 2013 [Available from: 32

33 http://documents.worldbank.org/curated/en/509481467990362181/pdf/Africa- http://bmjopen.bmj.com/ 34 35 development-indicators-2012-13.pdf accessed July 20, 2019. 36 37 38 13. Krishnaswami S, Nwomeh BC, Ameh EA. The pediatric surgery workforce in low- and 39 40 middle-income countries: problems and priorities. Seminars in Pediatric Surgery

41 on September 27, 2021 by guest. Protected copyright. 42 2016;25(1):32-42. doi: 10.1053/j.sempedsurg.2015.09.007 43 44 45 14. Concepcion T, Mohamed M, Dahir S, et al. Prevalence of Pediatric Surgical Conditions 46 47 Across Somaliland. JAMA network open 2019;2(1):e186857. doi: 48 49 10.1001/jamanetworkopen.2018.6857 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

22 1 2 3 15. Concepcion TL, Smith ER, Mohamed M, et al. Provision of Surgical Care for Children 4 5 6 Across Somaliland: Challenges and Policy Guidance. World Journal of Surgery 7 8 2019;43(11):2934. doi: 10.1007/s00268-019-05079-8 9 10 16. Smith ER, Concepcion TL, Mohamed M, et al. The contribution of pediatric surgery to 11 12 poverty trajectories in Somaliland. PloS one 2019;14(7):e0219974. doi: 13 14 15 10.1371/journal.pone.0219974 16 For peer review only 17 17. United Nations Population Fund. Population estimation survey 2014 for the 18 pre-war 18 19 regions of Somalia, 2014. 20 21 22 18. Concepcion TL, Dahir S, Mohamed M, et al. Barriers to Surgical Care Among Children in 23 24 Somaliland: An Application of the Three Delays Framework. World Journal of Surgery 25 26 2020;44(6):1712-18. doi: 10.1007/s00268-020-05414-4 [published Online First: 27 28 29 2020/02/08] 30 31 19. World Bank. The Uganda Poverty Assessment Report 2016, 2016:141. 32

33 20. Yousef Y, St-Louis E, Baird R, et al. A systematic review of capacity assessment tools in http://bmjopen.bmj.com/ 34 35 pediatric surgery: Global Assessment in Pediatric Surgery (GAPS) Phase I. Journal Of 36 37 38 Pediatric Surgery 2019;54(4):831-37. doi: 10.1016/j.jpedsurg.2018.11.005 39 40 21. Verguet S, Alkire BC, Bickler SW, et al. Timing and cost of scaling up surgical services in

41 on September 27, 2021 by guest. Protected copyright. 42 low-income and middle-income countries from 2012 to 2030: a modelling study. The 43 44 45 Lancet Global Health 2015;3(S2):S28-S37. doi: 10.1016/S2214-109X(15)70086-0 46 47 22. Anderson GA, Ilcisin L, Kayima P, et al. Out-of-pocket payment for surgery in Uganda: The 48 49 rate of impoverishing and catastrophic expenditure at a government hospital. PloS one 50 51 52 2017;12(10):e0187293. doi: 10.1371/journal.pone.0187293 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

23 1 2 3 23. Holmer H, Bekele A, Hagander L, et al. Evaluating the collection, comparability and findings 4 5 6 of six global surgery indicators. British Journal of Surgery 2019;106(2):138-50. doi: 7 8 10.1002/bjs.11061 9 10 24. Albutt K, Punchak M, Kayima P, et al. Access to Safe, Timely, and Affordable Surgical Care 11 12 in Uganda: A Stratified Randomized Evaluation of Nationwide Public Sector Surgical 13 14 15 Capacity and Core Surgical Indicators. World Journal of Surgery 2018;42(8):2303-13. 16 For peer review only 17 doi: 10.1007/s00268-018-4485-1 18 19 25. Guest GD, McLeod E, Perry WRG, et al. Collecting data for global surgical indicators: a 20 21 22 collaborative approach in the Pacific Region. BMJ Global Health 2017;2(4):e000376- 23 24 e76. doi: 10.1136/bmjgh-2017-000376 25 26 26. Linden A, Sekidde F, Galukande M, et al. Challenges of Surgery in Developing Countries: A 27 28 29 Survey of Surgical and Anesthesia Capacity in Uganda’s Public Hospitals. World Journal 30 31 of Surgery 2012;36(5):1056-65. doi: 10.1007/s00268-012-1482-7 32

33 27. Knowlton L, Chackungal S, Dahn B, et al. Liberian Surgical and Anesthesia Infrastructure: A http://bmjopen.bmj.com/ 34 35 Survey of County Hospitals. World Journal of Surgery 2013;37(4):721-29. doi: 36 37 38 10.1007/s00268-013-1903-2 39 40 28. Ahuka O. La cartographie des médecins spécialistes dans les quatre grandes disciplines

41 on September 27, 2021 by guest. Protected copyright. 42 cliniques ainsi que l’anesthésie réanimation en République Démocratique du Congo 43 44 45 (RDC): État des lieux. Revue médicale des Grands Lacs 2014;3:156-69. 46 47 29. Bolkan HA, Von Schreeb J, Samai MM, et al. Met and unmet needs for surgery in Sierra 48 49 Leone: A comprehensive, retrospective, countrywide survey from all health care facilities 50 51 52 performing operations in 2012. Surgery 2015;157(6):992. doi: 53 54 10.1016/j.surg.2014.12.028 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

24 1 2 3 30. Zafar SN, Canner JK, Nagarajan N, et al. Road traffic injuries: Cross-sectional cluster 4 5 6 randomized countrywide population data from 4 low-income countries. International 7 8 Journal of Surgery 2018;52:237-42. doi: 10.1016/j.ijsu.2018.02.034 [published Online 9 10 First: 2018/02/23] 11 12 31. Davé DR, Nagarjan N, Canner JK, et al. Rethinking burns for low & middle-income 13 14 15 countries: Differing patterns of burn epidemiology, care seeking behavior, and outcomes 16 For peer review only 17 across four countries. Burns : journal of the International Society for Burn Injuries 18 19 2018;44(5):1228-34. doi: 10.1016/j.burns.2018.01.015 [published Online First: 20 21 22 2018/02/25] 23 24 32. Fuller AT, Haglund MM, Lim S, et al. Pediatric Neurosurgical Outcomes Following a 25 26 Neurosurgery Health System Intervention at Mulago National Referral Hospital in 27 28 29 Uganda. World Neurosurgery 2016;95:309-14. doi: 10.1016/j.wneu.2016.07.090 30 31 [published Online First: 2016/08/09] 32

33 33. Abdelgadir J, Punchak M, Smith ER, et al. Pediatric traumatic brain injury at Mbarara http://bmjopen.bmj.com/ 34 35 Regional Referral Hospital, Uganda. Journal of Clinical Neuroscience 2018;47:79-83. 36 37 38 doi: 10.1016/j.jocn.2017.10.004 [published Online First: 2017/10/27] 39 40 34. Leather A, Ismail EA, Ali R, et al. Working together to rebuild health care in post-conflict

41 on September 27, 2021 by guest. Protected copyright. 42 Somaliland. The Lancet 2006;368(9541):1119-25. doi: 10.1016/S0140-6736(06)69047-8 43 44 45 35. World Bank. Somali joint needs assessment. Infraestructure cluster report., 2006:90. 46 47 36. Dieleman J, Campbell M, Chapin A, et al. Evolution and patterns of global health financing 48 49 1995–2014: development assistance for health, and government, prepaid private, and out- 50 51 52 of-pocket health spending in 184 countries. The Lancet 2017;389(10083):1981-2004. doi: 53 54 10.1016/S0140-6736(17)30874-7 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

25 1 2 3 37. World Health Organization, World Bank. Global Monitoring Report on Financial Protection 4 5 6 in Health 2019. Advance copy, 2019:47. 7 8 38. Smith ER, Concepcion TL, Niemeier KJ, et al. Is Global Pediatric Surgery a Good 9 10 Investment? World Journal of Surgery 2019;43(6):1450-55. doi: 10.1007/s00268-018- 11 12 4867-4 [published Online First: 2018/12/07] 13 14 15 39. Saxton AT, Poenaru D, Ozgediz D, et al. Economic Analysis of Children's Surgical Care in 16 For peer review only 17 Low- and Middle-Income Countries: A Systematic Review and Analysis. PloS one 18 19 2016;11(10):e0165480. doi: 10.1371/journal.pone.0165480 [published Online First: 20 21 22 2016/10/30] 23 24 40. Watters AD, Guest DG, Tangi GV, et al. Global Surgery System Strengthening: It Is All 25 26 About the Right Metrics. Anesthesia & Analgesia 2018;126(4):1329-39. doi: 27 28 29 10.1213/ANE.0000000000002771 30 31 41. Smith ER, Concepcion TL, Shrime M, et al. Waiting Too Long: The Contribution of Delayed 32

33 Surgical Access to Pediatric Disease Burden in Somaliland. World Journal of Surgery http://bmjopen.bmj.com/ 34 35 2020;44(3):656-64. doi: 10.1007/s00268-019-05239-w [published Online First: 36 37 38 2019/10/28] 39 40 42. Butler EK, Tran TM, Fuller AT, et al. Quantifying the pediatric surgical need in Uganda:

41 on September 27, 2021 by guest. Protected copyright. 42 results of a nationwide cross-sectional, household survey. Pediatric Surgery International 43 44 45 2016;32(11):1075-85. doi: 10.1007/s00383-016-3957-3 [published Online First: 46 47 2016/09/12] 48 49 43. Butler EK, Tran TM, Nagarajan N, et al. Epidemiology of pediatric surgical needs in low- 50 51 52 income countries. PloS one 2017;12(3):e0170968. doi: 10.1371/journal.pone.0170968 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

26 1 2 3 44. Smith ER, Vissoci JRN, Rocha TAH, et al. Geospatial analysis of unmet pediatric surgical 4 5 6 need in Uganda. Journal of Pediatric Surgery 2017;52(10):1691-98. doi: 7 8 10.1016/j.jpedsurg.2017.03.045 9 10 45. Vissoci JRN, Ong CT, Andrade Ld, et al. Disparities in surgical care for children across 11 12 Brazil: Use of geospatial analysis. PloS one 2019;14(8):e0220959. doi: 13 14 15 10.1371/journal.pone.0220959 16 For peer review only 17 46. Abdul-Mumin A, Anyomih TTK, Owusu SA, et al. Burden of Neonatal Surgical Conditions 18 19 in Northern Ghana. World Journal of Surgery 2019 doi: 10.1007/s00268-019-05210-9 20 21 22 47. Concepcion TL, Smith ER, Mohamed M, et al. Provision of Surgical Care for Children 23 24 Across Somaliland: Challenges and Policy Guidance. World Journal of Surgery 25 26 2019;43(11):2934. doi: 10.1007/s00268-019-05079-8 27 28 29 48. Abdelgadir J, Smith E, Punchak M, et al. Outcomes and Predictors of Mortality in 30 31 Neurosurgery Patients at Mbarara Regional Referral Hospital, Mbarara, Uganda. Annals 32

33 of Global Health 2017;83(1):18-19. doi: 10.1016/j.aogh.2017.03.040 http://bmjopen.bmj.com/ 34 35 49. World Health Organization, Program in Global Surgery and Social Change. Surgical care 36 37 38 systems strengthening: Developing national surgical, obstetric and anaesthesia plans 39 40 World Health Organization, 2017.

41 on September 27, 2021 by guest. Protected copyright. 42 50. Ministry of Health of Zambia. National Surgical, Obstetric, and Anaesthesia Strategic Plan 43 44 45 (NSOASP). 46 47 51. Mukhopadhyay S, Lin Y, Mwaba P, et al. Implementing World Health Assembly Resolution 48 49 68.15: National surgical, obstetric, and anesthesia strategic plan development--the 50 51 52 Zambian experience. Bulletin of the American College of Surgeons 2017;102(6):28-35. 53 54 [published Online First: 2017/09/09] 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

27 1 2 3 52. Burssa D, Teshome A, Iverson K, et al. Safe Surgery for All: Early Lessons from 4 5 6 Implementing a National Government-Driven Surgical Plan in Ethiopia. World Journal 7 8 of Surgery 2017;41(12):3038-45. doi: 10.1007/s00268-017-4271-5 9 10 53. Burssa D, Teshome A, Iverson K, et al. Safe Surgery for All: Early Lessons from 11 12 Implementing a National Government-Driven Surgical Plan in Ethiopia. World Journal 13 14 15 Surgery 2017;41(12):3038-45. doi: 10.1007/s00268-017-4271-5 [published Online First: 16 For peer review only 17 2017/10/17] 18 19 54. Raykar N, Mukhopadhyay S, Saluja S, et al. Implementation of The Lancet Commission on 20 21 22 Global Surgery in India. Healthcare 2019;7(2):4-6. doi: 10.1016/j.hjdsi.2018.11.001 23 24 55. Citron I, Jumbam D, Dahm J, et al. Towards equitable surgical systems: development and 25 26 outcomes of a national surgical, obstetric and anaesthesia plan in Tanzania. BMJ Global 27 28 29 Health 2019;4(2) doi: 10.1136/bmjgh-2018-001282 30 31 56. Stewart BT, Tansley G, Gyedu A, et al. Mapping Population-Level Spatial Access to 32

33 Essential Surgical Care in Ghana Using Availability of Bellwether Procedures. JAMA http://bmjopen.bmj.com/ 34 35 Surgery 2016;151(8):e161239. doi: 10.1001/jamasurg.2016.1239 36 37 38 57. Esquivel MM, Uribe-Leitz T, Makasa E, et al. Mapping Disparities in Access to Safe, 39 40 Timely, and Essential Surgical Care in Zambia. JAMA Surgery 2016;151(11):1064-69.

41 on September 27, 2021 by guest. Protected copyright. 42 doi: 10.1001/jamasurg.2016.2303 43 44 45 58. Keyes EB, Parker C, Zissette S, et al. Geographic access to emergency obstetric services: a 46 47 model incorporating patient bypassing using data from Mozambique. BMJ Global Health 48 49 2019;4(Suppl 5) doi: 10.1136/bmjgh-2018-000772 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

28 1 2 3 59. Federspiel F, Mukhopadhyay S, Milsom PJ, et al. Global surgical, obstetric, and anesthetic 4 5 6 task shifting: A systematic literature review. Surgery 2018;164(3):553-58. doi: 7 8 10.1016/j.surg.2018.04.024 [published Online First: 2018/08/28] 9 10 60. Saluja S, Silverstein A, Mukhopadhyay S, et al. Using the Consolidated Framework for 11 12 Implementation Research to implement and evaluate national surgical planning. BMJ 13 14 15 global health 2017;2(2):e000269-e69. doi: 10.1136/bmjgh-2016-000269 16 For peer review only 17 61. Global Initiative for Children’s Surgery. Optimal Resources for Children’s Surgical Care: 18 19 Executive Summary. World Journal of Surgery 2019;43(4):978-80. doi: 10.1007/s00268- 20 21 22 018-04888-7 23 24 62. Global Initiative for Children's Surgery. Global Initiative for Children's Surgery: A Model of 25 26 Global Collaboration to Advance the Surgical Care of Children. World Journal of 27 28 29 Surgery 2019;43(6):1416-25. doi: 10.1007/s00268-018-04887-8 [published Online First: 30 31 2019/01/10] 32

33 63. Massenburg BB, Saluja S, Jenny H, et al. Assessing the Brazilian surgical system with six http://bmjopen.bmj.com/ 34 35 surgical indicators: a descriptive and modelling study. BMJ Global Health 2017;2(2) doi: 36 37 38 10.1136/bmjgh-2016-000226 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

29 1 2 3 FIGURE LEGEND 4 5 6 Figure 1. Somaliland Indicators Compared to the Lancet Commission on Global Surgery targets. 7 8 9 Figure 2. Catchment population within a 2-hour travel time from the closest surgically-capable 10 11 hospital. 12 13 14 Figure 3. Lancet Commission on Global Surgery indicators for Somaliland, stratified by region. 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 Figure 1. Somaliland Indicators Compared to the Lancet Commission on Global Surgery targets 29 584x381mm (96 x 96 DPI) 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 32 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Figure 2. Catchment population within a 2-hour travel time from the closest surgically-capable hospital 32

33 279x215mm (300 x 300 DPI) http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 PREPAREDNESS 15 16 For peer review only 17 18 19 20 21 22 Indicator 3: Surgical volume 23 Target: 5,000/100,000 population 24 25 800 646 26 600 544.8 27 445.6 28 400 239.5 29 DELIVERY 200 30 91.6 66.2 31 0 32

33 http://bmjopen.bmj.com/ 34 Indicator 5: protection against impoverishment Target: 100% protection 35 36 100% 37 80% 38 60% 39 40% 25.5% 40 16.3% 15.7% 20.3% 19.9% 20.8%

20%

41 on September 27, 2021 by guest. Protected copyright. 42 0% 43 44 Indicator 6: protection against catastrophic

45 IMPACT expenditure Target: 100% protection 46 100% 47 80% 48 60% 49 50 40% 51 20% 8.8% 0.1% 0.1% 0.9% 0.8% 1.2% 52 0% 53 54 55 56 *SAO: surgical, anesthetic, and obstetric providers 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

Interpreting the Lancet Surgical Indicators in Somaliland: A Cross-Sectional Study

Journal: BMJ Open ManuscriptFor ID peerbmjopen-2020-042968.R1 review only Article Type: Original research

Date Submitted by the 12-Oct-2020 Author:

Complete List of Authors: Dahir, Shukri; Edna Adan Hospital Cotache-Condor, Cesia; Baylor University, Department of Public Health Concepcion, Tessa; Duke University, Duke University Global Health Institute Mohamed, Mubarak; Edna Adan University Hospital Poenaru, Dan; MyungSung Christian Medical Center, Pediatric Surgery; Montreal Children\'s Hospital Research Institute, Pediatric Surgery Adan Ismail, Edna; Edna Adan Hospital Leather, Andy; King’s Health Partners and King’s College London, King’s Centre for Global Health Rice, Henry; Duke University, Global Health Institute; Duke University Medical Center, Department of Surgery Smith, Emily; Baylor University, Health, Human Performance, and Recreation; Duke University, Duke University Global Health Institute http://bmjopen.bmj.com/ Primary Subject Surgery Heading:

Secondary Subject Heading: Global health

SURGERY, Health policy < HEALTH SERVICES ADMINISTRATION & Keywords: MANAGEMENT, PUBLIC HEALTH, International health services < HEALTH SERVICES ADMINISTRATION & MANAGEMENT on September 27, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 29 BMJ Open

1 2

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 27, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

1 1 2 3 Interpreting the Lancet Surgical Indicators in Somaliland: A Cross-Sectional Study 4 5 6 Shukri Dahir1, Cesia Cotache-Condor2, Tessa L Concepcion3, Mubarak Mohamed1, Dan 7 8 4 1 5 3 2,3 9 Poenaru , Edna Adan Ismail , Andrew J. M. Leather , Henry E Rice , Emily R Smith , Global 10 11 Initiative for Children’s Surgery 12 13 14 1 Edna Adan University Hospital, Hargeisa, Somaliland, 2 Department of Public Health, Baylor 15 16 University, Waco, TX,For USA; peer3 Duke Global review Health Institute, onlyDuke University, Durham, NC, 17 18 USA; 4 McGill University, Montreal, Quebec, Canada; 5 King’s Centre for Global Health and 19 20 21 Health Partnerships, King’s College London, London, UK 22 23 24 25 26 27 Shukri Dahir and Cesia Cotache-Condor contributed equally to this paper. 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 Word count: 3,551 on September 27, 2021 by guest. Protected copyright. 42 43 44 Corresponding author: 45 46 47 Emily Smith, PhD 48 49 50 Robbins College of Health and Human Sciences, Baylor University, 1621 S. 5th Street, Waco, 51 52 TX 76706 53 54 [email protected] | +1-254-710-4028 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

41 on September 27, 2021 by guest. Protected copyright. 42 surgical, anesthetic, and obstetric (SAO) providers is 0.8 per 100,000, compared to a LCoGS 43 44 goal of 20 SAO per 100,000. Surgical volume is 368 procedures per 100,000 people, while the 45 46 LCoGS goal is 5,000 procedures per 100,000. Protection against impoverishing expenditures was 47 48 49 only 18% and against catastrophic expenditures 1%, both far below the LCoGS goal of 100% 50 51 protection. 52 53 54 Conclusion: We found several gaps in the surgical system in Somaliland using the LCoGS 55 56 indicators and target goals. These metrics provide a broad view of current status and gaps in 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

4 1 2 3 Strengths and limitations of the study 4 5 6  This is the first nationwide study that comprehensively evaluates the surgical system, 7 8 9 including the public and private sectors, in Somaliland. 10 11  This study follows the Lancet Commission on Global Surgery indicators to evaluate 12 13 surgical preparedness, delivery, and impact in Somaliland. 14 15 16  Data on perioperativeFor peermortality rates review were not included only as most hospitals in Somaliland 17 18 do not collect this data. 19 20  Our reported surgical workforce density did not include data on non-specialist physicians 21 22 23 and non-physician clinicians who may provide surgical care in Somaliland. 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

5 1 2 3 INTRODUCTION 4 5 6 Over 5 billion people around the world lack safe, affordable, and timely access to surgical and 7 8 anesthesia care, with the overwhelming burden being among persons in low- and middle-income 9 10 countries (LMICs).1-3 To address this crisis, targeted and accurate measurement of surgical care 11 12 is a critical first step. However, the misconception that surgery is an expensive intervention has 13 14 15 prevented it from playing a significant role in global health for a long time.4 5 Investing in 16 For peer review only 17 surgery is highly cost-effective and beneficial at a macro and micro level. Surgery promotes the 18 19 overall scaling up of national health systems and provides financial risk protection to families 20 21 22 that otherwise may be pushed into the poverty or have reduced their chances of moving out 23 24 poverty.1 4 6 Global surgery has gained recent attention due to paramount events such as the 25 26 creation of the Lancet Commission on Global Surgery (LCoGS) and the inclusion of four of the 27 28 29 six LCoGS indicators among the core 100 World Health Organization and World Development 30 31 indicators. 1 7 8 32

33 http://bmjopen.bmj.com/ 34 The 2015 Lancet Commission on Global Surgery proposed six surgical indicators through an 35 36 extensive and iterative process led by experts from various disciplines, surgical specialties, and 37 38 countries.1 These six indicators and associated target goals define the preparedness for surgical 39 40

41 and anesthesia care (access to essential surgery and workforce density), delivery of care on September 27, 2021 by guest. Protected copyright. 42 43 (quantifying the surgical volume and perioperative mortality rates), and impact of care 44 45 (protection against impoverishing and catastrophic expenditures).1 These indicators have been 46 47 48 used as a template to assess surgical care in several countries, which in turn have supported the 49 50 development of national surgical, obstetrical, and anesthesia plans (NSOAPs).9 10 51 52 53 Somaliland is a low-income country in Sub-Saharan Africa with significant health and economic 54 55 challenges. After emerging from the civil war in 1991, the government of Somaliland focused on 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

6 1 2 3 rebuilding its hospitals, clinics and health posts, and devoted 3% of the national budget to rebuild 4 5 11 12 6 its health sector. Somaliland has a GDP of $348 per capita (making it the fourth poorest 7 8 country in the world), has a population of approximately 4 million people, and some of the 9 10 poorest health outcomes in the Africa Region.13-19 Our previous work and that of others has 11 12 characterized the surgical infrastructure for children in the country.6 19 20 The goal of the current 13 14 15 study was to characterize the Somaliland surgical health system according to the LCoGS surgical 16 For peer review only 17 indicators. 18 19 20 METHODS 21 22 23 Setting 24 25 This cross-sectional study took place in Somaliland, a country in the Horn of Africa which 26 27 achieved relative stability and has an autonomous government since separation from Somalia in 28 29 21 30 1991. The population estimate from the 2014 Somali census is 3.5 million people. Somaliland 31 32 is made up of six regions – Awdal, Maroodi Jeex (containing the capital city of Hargeisa), Sahil,

33 http://bmjopen.bmj.com/ 34 Sanaag, Sool, and Togdheer – with 40% of the population living in Hargeisa.22 The public health 35 36 system is comprised of primary health units, health centers, referral health centers, regional 37 38 39 hospitals, and one national referral hospital. Specialty hospitals include seven tuberculosis 40

41 hospitals, one female fistula hospital, one children’s hospital, and one mental health hospital. An on September 27, 2021 by guest. Protected copyright. 42 43 estimated 60% of health services are provided through the private sector, consisting of non- 44 45 46 government organizations, private health practitioners and hospitals, and traditional 47 48 practitioners.12 49 50 51 Hospital selection 52 53 Hospitals were included in the study if they had the capacity to perform surgery (defined as the 54 55 56 presence of at least one operating room), yielding 16 hospitals. We were not able to obtain data 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

7 1 2 3 from one private hospital, resulting in a final study cohort of 15 hospitals. Our in-country 4 5 6 research manager asked permission from each hospital director or senior administrator for 7 8 participation in the study, and presented the study’s letter of approval from the Ministry of 9 10 Health. 11 12 13 Data collection 14 15 16 Our current study isFor built upon peer data collected review by our team and only reported in previous studies.6 19 20 17 18 19 We assessed hospital capacity at each facility using the World Health Organization’s (WHO) 20 21 Surgical Assessment Tool – Hospital Walkthrough and the Global Initiative for Children’s 22 23 Surgery (GICS) Global Assessment in Pediatric Surgery (GAPS) tool.23 24 The survey included 24 25 26 data on service delivery, the number of admissions and surgical procedures per month, whether 27 28 the facility tracked perioperative in-hospital mortality, and the number of laparotomies, cesarean 29 30 sections, and open fracture repairs (i.e., the three LCoGS “bellwether” procedures). To assess 31 32 each facility’s workforce, the survey quantified the number of staff, including surgeons,

33 http://bmjopen.bmj.com/ 34 35 anesthesiologists, non-physician anesthetists, obstetricians, nurses, and administrative staff. Each 36 37 staff member was counted only once even if they worked across multiple health facilities. To 38 39 assess each facility’s infrastructure, the survey asked about the availability of essential resources 40

41 on September 27, 2021 by guest. Protected copyright. 42 (i.e., electricity, running water, oxygen), the number of hospital beds and operating rooms, and 43 44 the number of anesthesia machines and ventilators. 45 46 47 Indicator assessment 48 49 50 The LCoGS indicators includes 6 items pertaining to surgical care preparedness (Indicators 1 and 51 52 2), delivery (Indicators 3 and 4), and impact (Indicators 5 and 6) (Table 1). We did not assess 53 54 whether each hospital tracked perioperative mortality rates, as this metric is generally not 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

8 1 2 3 collected in Somaliland. Thus, Indicator 4 was not included in the analysis. Data tabulation and 4 5 6 analyses were performed using Microsoft Excel and SAS 9.3 (Cary, NC). Geographic mapping 7 8 and analyses were performed using ArcGIS (Environmental Systems Research Institute (ESRI) 9 10 2012, Redlands, CA). 11 12 13 Table 1. Lancet Commission on Global Surgery indicator definitions and targets.1 14 15 16 INDICATORFor peerDEFINITION review only TARGET 17 1. Access to timely Proportion of the population that can A minimum of 80% coverage of 18 essential surgery access, within 2 hours, a facility that can essential surgical and anesthesia services 19 do caesarean delivery, laparotomy, and per country by 2030 20 treatment of open fracture (the Bellwether 21 Procedures) 22 2. Specialist surgical Number of specialist surgical, anesthetic, 100% of countries with at least 20 23

Preparedness workforce density and obstetric physicians who are working, surgical, anesthetic, and obstetric 24 per 100 000 population physicians per 100 000 population by 25 2030 26 3. Surgical volume Procedures done in an operating theatre, 80% of countries by 2020 and 100% of 27 per 100 000 population per year countries by 2030 tracking surgical 28 volume; a minimum of 5000 procedures 29 per 100 000 population by 2030 30 4. Perioperative All-cause death rate before discharge in 80% of countries by 2020 and 100% of

31 Delivery mortality patients who have undergone a procedure countries by 2030 tracking perioperative 32 in an operating theatre, divided by the mortality; in 2020, assess global data

33 total number of procedures (%) and set national targets for 2030 http://bmjopen.bmj.com/ 34 5. Protection against Proportion of households protected 100% protection against impoverishment 35 impoverishing against impoverishment from direct out- from out-of-pocket payments for 36 expenditure of-pocket payments for surgical and surgical and anesthesia care by 2030 37 anesthesia care 38 6. Protection against Proportion of households protected 100% protection against catastrophic Impact 39 catastrophic against catastrophic expenditure from expenditure from out-of-pocket 40 expenditure direct out-of-pocket payments for surgical payments for surgical and anesthesia and anesthesia care care by 2030 41 on September 27, 2021 by guest. Protected copyright. 42 43 44 Indicator 1: Access to timely essential surgery. Geographic locations of the 15 surgically-capable 45 46 hospitals, defined by the presence of an operating room, were recorded using a smart phone GPS 47 48 49 (Global Positioning System) application. Locations of the 15 hospitals were mapped and used as 50 51 destination points from which distance calculations were made. Network service area analysis 52 53 was used to determine the percent of the population within a 2-hour radius of the surgically- 54 55 56 capable hospitals throughout the country. A travel rate of 30 kilometers per hour was used to 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

9 1 2 3 determine the geographical area within a two-hour drive of surgically-capable hospitals using 4 5 6 available Somaliland road infrastructure data. Population estimates within each geographical 7 8 area were made using data from the 2014 Somalia census.21 Surgical capacity was defined as the 9 10 existence of an operating room within the hospital. This geospatial method did not account for 11 12 transportation wait time or topography of the land. However, travel times were reviewed by the 13 14 15 in-country data team to confirm validity of the estimated travel times between locations. The 16 For peer review only 17 service area map was combined with a dot density map to depict the relations between surgical 18 19 care coverage and population distribution at the regional level. 20 21 22 Indicator 2: Surgical workforce density. The surgical workforce density was defined as the 23 24 25 number of specialist surgeons, anesthesiologists, and obstetricians (SAO) per 100,000 26 27 population. Surgeons included specialties available within Somaliland (general surgeons, 28 29 colorectal surgeons, neurosurgeons, and otolaryngologists). Anesthesiologists did not include 30 31 32 nurse anesthetists. The surgical workforce density was calculated as the number of SAOs per

33 http://bmjopen.bmj.com/ 34 100,000 population. In addition to the national SAO workforce density, region-specific densities 35 36 were also estimated. 37 38 39 Indicator 3: Surgical volume. Surgical volume was obtained through the country-wide hospital 40

41 assessment. We included all procedures recorded in hospital logbooks between January 1, 2017 on September 27, 2021 by guest. Protected copyright. 42 43 20 44 and December 31, 2017 using previously described methods. Exclusion criteria included 45 46 procedures performed outside of the operating rooms. We collected data on demographics, 47 48 surgical diagnosis, and surgical procedure(s) performed. Data did not include admission 49 50 51 information, comorbidities, complications, or outcome. We quantified the number of the 52 53 LCoGS-defined bellwether procedures (open fracture repairs [all types], Cesarean section 54 55 deliveries, and laparotomies).1 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

10 1 2 3 Indicator 5: Protection against impoverishing expenditure. Protection against impoverishing 4 5 6 expenditure was defined as 100 - % population at risk of impoverishing expenditures. 7 8 Impoverishment was defined as any income below the poverty line after out-of-pocket payments, 9 10 including medical and non-medical costs, for surgery. The threshold for impoverishment 11 12 expenditure for Somaliland was defined as any income below a value of: 13 14 15 [% out of pocket expenditure] * [average cost of surgery] + [Somaliland poverty line]. 16 For peer review only 17 18 For costs per operation, we used $200 as a proxy due to lack of information of operative costs for 19 20 21 each hospital. The $200 value was chosen as it represents the mean value between costs per 22 23 surgical procedure in a low-income country ($179) and a lower-middle income country ($219) 24 25 calculated as previously described.25 The $200 costs covers all direct costs, including 26 27 28 preoperative and postoperative hospitalization as well as the recurrent costs of running a surgical 29 30 service such as salaries, utilities, equipment, medical supplies and medicine for the procedure 31 32 itself. We confirmed this amount based on out-of-pocket expenses from one private hospital in

33 http://bmjopen.bmj.com/ 34 35 Somaliland, which used an average of $200 per procedure. Non-medical expenses, such as 36 37 transportation and lodging, were estimated at $60 based on previous literature.26 The expected 38 39 proportion of out-of-pocket costs for healthcare was provided by the hospital administration 40

41 during the survey data collection. The threshold for the poverty line in Somaliland was set at on September 27, 2021 by guest. Protected copyright. 42 43 13 44 $220 based on previous literature. 45 46 47 Indicator 6: Protection against catastrophic expenditure. Protection against catastrophic 48 49 expenditures was defined as 100% - population at risk of catastrophic expenditure.1 Catastrophic 50 51 expenditure was defined as direct out-of-pocket expenses greater than 40% of the gross domestic 52 53 54 product (GDP) per capita, excluding subsistence costs. The risk of catastrophic expenditure for 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

11 1 2 3 each hospital and region was determined by dividing all direct and indirect out-of-pocket 4 5 26 6 expenses by $348, Somaliland’s GDP per capita. 7 8 9 Patient and Public involvement 10 11 12 Patients and/or the public were not involved in the design, or conduct, or reporting, or 13 14 dissemination plans of this research 15 16 For peer review only 17 18 19 20 RESULTS 21 22 23 The surgical system in Somaliland did not reach any of the target indicator goals as defined by 24 25 LCoGS, with each indicator lagging behind its target by 76% or more (Figure 1). The greatest 26 27 need was for protection against catastrophic expenditure and inadequate surgical volume. 28 29 30 Preparedness 31 32

33 Only an estimated 19% of persons in Somaliland live within a 2-hour access to a surgically- http://bmjopen.bmj.com/ 34 35 capable facility, translating to 24% of the LCoGS goal of at least 80% population coverage 36 37 (Figure 2). The number of specialist SAO physicians was 0.8 per 100,000 population, or 4% of 38 39 40 the LCoGS goal of 20 SAOs per 100,000. Access to surgery ranged from 36% in Maroodi Jeex

41 on September 27, 2021 by guest. Protected copyright. 42 to below 6% in Awdal, Sool, and Sanaag. Of the total workforce of 32 SAOs in the country, 15 43 44 were surgeons, 3 were anesthesiologists, and 14 were obstetricians (Table 2). The number of 45 46 47 surgeons ranged from 10 in Maroodi Jeex to 0 in Sool, Togdheer, or Sanaag, while the number of 48 49 obstetricians ranged from 8 in Maroodi Jeex to 0 in Sahil. Of the 3 anesthesiologists, all 3 were 50 51 located in Maroodi Jeex (the capital city). 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

41 on September 27, 2021 by guest. Protected copyright. 42 laparotomies ranged between 1 - 20 per month (in Sahil or Sool and Awdal, respectively), 43 44 Cesarean sections between 6 - 100 per month (in Sool and Maroodi Jeex, respectively), and open 45 46 fracture repairs between 0 - 40 per month (in Sahil, Sool, or Togdheer and Awdal, respectively). 47 48 49 Impact 50 51 52 Protection from impoverishment was only 18% of the LCoGS goal of 100% protection by 2030. 53 54 Risk of impoverishment was high throughout the country and ranged between 74.5% in Sahil 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

13 1 2 3 and 84.3% in Maroodi Jeex (Figure 3). Protection against catastrophic expenditures was 1% of 4 5 6 the LCoGS goal of 100% protection by 2030. The risk of catastrophic expenditure was greater 7 8 than 90% throughout the entire country. 9 10 11 DISCUSSION 12 13 As surgical care is increasingly recognized as an essential component of a functional healthcare 14 15 system, understanding the surgical landscape of a country is vital for health care scale-up plans 16 For peer review only 17 18 contextualized to that country. The LCoGS indicators measure several domains of surgical care 19 20 related to preparedness, delivery, and impact of care. These indicators are increasingly used 21 22 around the world as a framework to scale up surgical care in the country through a 23 24 comprehensive, system-level perspective.27 We found that Somaliland’s surgical system achieves 25 26 27 only a small fraction of these target goals, with the greatest need found in rural areas. 28 29 30 Our study found large gaps in surgical care preparedness, delivery, and impact. Preparedness, 31 32 including access to surgical care and workforce density, are far below the LCoGS targets,

33 http://bmjopen.bmj.com/ 34 consistent with other sub-Saharan African countries.3 9 27-37 The reasons for lack of access to 35 36 37 surgical care in Somaliland is multifactorial, but is largely due to the limited number of locally- 38 39 accessible surgically-capable hospitals and existing road networks. The majority of surgically- 40

41 capable hospitals are within the urban Maroodi Jeex region, with only 1 hospital located in the on September 27, 2021 by guest. Protected copyright. 42 43 44 more rural regions. Although the concentration of hospitals and SAO providers corresponds with 45 46 population density, overall, the surgical care coverage remains limited in all regions, particularly 47 48 in the Northwest and Southeast areas of the country. After almost a decade of civil conflict 49 50 (1982-1991), the country was left with severe damage and deterioration to road networks and 51 52 38 53 healthcare infrastructure, and only 15% of the primary roads in urban areas and networks 54 55 connecting rural to urban areas are considered to be in good condition.39 Workforce density is 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

14 1 2 3 also low in the more rural parts of the country, with most of the SAOs concentrated within urban 4 5 6 areas. Delivery of surgical care as measured by the volume of procedures was low particularly in 7 8 rural areas and is likely related to the workforce density within each area. For example, 57% of 9 10 all procedures per month occurred in the urban Maroodi Jeex region which has 65% of all SAOs. 11 12 However, high surgical volume, as in Sahil, may be explained by the response of other health 13 14 15 professionals to the high demand of emergency surgical care in the absence of specialized SAO 16 For peer review only 17 providers.20 18 19 20 The financial impact of surgical care is evident through the high rates of financial strain on 21 22 families, which limits Somaliland’s capacity to attain universal health coverage (UHC) as 23 24 25 advocated by the World Bank, UN and WHO. The provision of surgery in Somaliland is largely 26 27 dependent on a person’s financial means to pay for out-of-pocket (OOP) expenses, similar to 28 29 high rates of OOP spending in many LMICs.40 The risk of being pushed into poverty or having 30 31 32 reduced opportunities to move out of poverty due to OOP expenses for surgical care in

33 http://bmjopen.bmj.com/ 34 Somaliland is one of the highest compared to other LMICs,6 which is particularly challenging as 35 36 Somaliland is the fourth poorest country in the world, with an estimated 35% of households 37 38 living under $100 per month. 6 36-40 Our findings highlight the need for inclusion of surgical care 39 40

41 in the country’s UHC scheme in order to prevent impoverishing and catastrophic expenditure on September 27, 2021 by guest. Protected copyright. 42 43 due to OOP expenses, especially among at-risk communities.5 41 42 44 45 46 As a post-conflict country, the healthcare system Somaliland experiences many structural 47 48 challenges, with just 3% of the national budget designated to the health care. In this context, 49 50 51 outlining a strategic national health care plan to improve access to surgery is required. Such a 52 53 plan should address deficiencies in the surgical system as defined by the LCoGS indicators, 54 55 including preparedness, delivery, and impact. Each LCoGS indicator should be interpreted in 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

15 1 2 3 conjunction with other indicator metrics to fully understand how gaps in care are interrelated .43 4 5 6 Evaluating surgical capacity begins with accessibility. However, geographical coverage is just 7 8 one dimension of access and does not ensure that people receive proper surgical care.27 44-49 9 10 Patients that eventually reach healthcare facilities might not be able to receive surgical care due 11 12 to deficits in infrastructure and workforce.50 51 Thus, interpreting each indicator independently of 13 14 15 the others will only provide a limited understanding of the surgical needs within a country. 16 For peer review only 17 18 One way to comprehensively evaluate surgical systems and gaps in care on a national level are 19 20 through country-specific National Surgical, Obstetric, and Anesthesia Plans (NSOAPs). The 21 22 purpose of NSOAPs are to improve surgical care through the six major building blocks of a 23 24 25 health system: infrastructure, workforce, service delivery, information management, finance, and 26 27 leadership/governance.1 One of the first steps in developing a country-specific NSOAP is 28 29 identifying gaps for each building block.10 Although Somaliland does not have a NSOAP, our 30 31 32 paper provides critical data regarding infrastructure, workforce, service delivery, and the need

33 http://bmjopen.bmj.com/ 34 for financing schemes to protect individuals from catastrophic and impoverishing expenditures. 35 36 37 Somaliland may be able to benefit from lessons learned in other LMICs during enactment of 38 39 NSOAPs. The Republic of Zambia was one of the first countries to conduct a baseline 40

41 assessment of the LCoGS indicators and utilize those in a quality improvement plan within a on September 27, 2021 by guest. Protected copyright. 42 43 52-54 52 55 56 52 57 52 58 44 NSOAP. To date, other countries including Ethiopia, India, and Tanzania, have 45 46 adopted a similar process to develop their NSOAPs. These experiences suggest that Somaliland 47 48 may benefit from upgrading geographically strategic facilities and allocating resources to high- 49 50 59-61 51 impact healthcare facilities to ensure equity and reinforce existing surgical capacity. Task 52 53 shifting and task sharing, partnerships with other countries and the private sector, culturally 54 55 embedded training, and increase of incentives are other ways to efficiently scale up workforce 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

16 1 2 3 density.1 62 We encourage continuous data collection as part of a NSOAP anchored within the 4 5 6 overall country’s national health strategic plan, and driven by the Somaliland Ministry of Health 7 8 to ensure standardization, integration and sustainability over time.10 27 Furthermore, engagement 9 10 of local champions and health care workers on the ground from the early stages will be pivotal to 11 12 ensure success and continuity.52 63-65 13 14 15 Successful implementation of NSOAPs also require involving strong governmental leadership 16 For peer review only 17 18 and other stakeholders, establishing surgical priorities, and allocating funding from partnerships 19 20 and non-traditional sources.52 55 One example which is be helpful for Somaliland is that of the 21 22 Saving Lives Through Safe Surgery (SaLTS) in Ethiopia, which has shown the value of 23 24 25 established a comprehensive system of educational/training packages, leadership/mentoring 26 27 programs, national perioperative guidelines, implementation of the WHO Surgical Safety 28 29 Checklist, and requirements for functional operating rooms, standardized equipment, supplies, 30 31 55 32 and purchasing. In Tanzania, the NSOAP emphasizes the coverage of essential surgical care in

33 http://bmjopen.bmj.com/ 34 national health insurance schemes, implementation of electronic medical records for surgery- 35 36 related data, simultaneous strengthening of lower-level and higher-level hospitals to reinforce 37 38 referral pathways, and implementation of regional training to improve local retention of SAO 39 40 58

41 providers. on September 27, 2021 by guest. Protected copyright. 42 43 44 Limitations 45 46 47 Our study has several limitations. First, our reported surgical workforce density might 48 49 underrepresent the entire surgical workforce in Somaliland, we did not collect data on non- 50 51 specialist physicians and non-physician clinicians who may provide surgical care. Second, most 52 53 54 hospitals in Somaliland do not collect data on perioperative mortality rates. This indicator is the 55 56 most widely adopted measurement for surgical outcomes and highlight delays to receiving 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

17 1 2 3 care.27 There are still few studies that have successfully measured perioperative mortality rates 4 5 66 6 without relying on modelling data. Similarly, we did not measure quality of care within the 7 8 hospital settings, nor were able to assess the impact of surgical capacity on mortality. A facility 9 10 assessment tends to pick up on one of the three aspects of the Donabedian framework elements – that of 11 12 structure and not process and outcome.67 However, the presence of an operating room does not indicate a 13 14 comprehensive analysis of hospital capacity. Therefore, it was only used as an initial selection criterion 15 16 For peer review only 17 on which the Lancet Commission on Global Surgery (LCoGS) indicators were applied. Third, our data 18 19 on impoverishing and catastrophic expenditures may have been based on inaccurate surgical 20 21 costs. Few studies in other countries have accurate primary data on surgery related costs and out- 22 23 of-pocket expenditures, as most are based on modeling.26 29 66 68 Although we collected limited 24 25 26 primary data on financial costs, these reported costs likely underestimate the total costs of 27 28 procedures, including indirect costs of lost earnings. However, we attempted to account for the 29 30 wide range of surgical costs depending on the procedure by including lower and upper limits in 31 32 our expenditure calculations. Fourth, measuring household income is challenging for many 33 http://bmjopen.bmj.com/ 34 35 households in LMICs as many do not work in the formal sector or have a consistent income.69 70 36 37 Therefore, our attempts to calculate the impact indicators for Somaliland may be limited as this 38 39 40 average might be overestimated for the poorest regions of the country. These findings emphasize

41 on September 27, 2021 by guest. Protected copyright. 42 the need for improved primary data collection regarding out-of-pocket surgical costs in LMICs. 43 44 45 Probably the most important limitation to our analysis is the challenge among the LCoGS 46 47 indicators to correctly measure their respective components. The LCoGS indicators were 48 49 originally defined using data from nationally representative 194 WHO member states. However, 50 51 52 all indicators except for surgical workforce included data from less than half of WHO member 53 54 states.27 In addition, many countries had differing definitions of indicators, hindering 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

18 1 2 3 comparability and potentially leading to erroneous estimates.27 Taking access to care as an 4 5 6 example, geographic proximity is only one dimension of access and does not necessarily account 7 8 for the impact of road conditions or type of transportation. Similarly, the definition of surgical 9 10 volume did not take into account varying population needs depending on the specific country. 11 12 Perhaps the most potentially biased indicator is expenditure, as this indicator was defined solely 13 14 15 on modeling estimates with no national-level data. Although these challenges may have affected 16 For peer review only 17 our estimates, our data can serve to drive collection of country-wide surgical data in the future. 18 19 20 Conclusion 21 22 23 As Somaliland continues to rebuild its healthcare infrastructure, incorporating surgical care as an 24 25 essential component of its national health care plan is critical. We found large limitations in the 26 27 28 overall surgical system using the LCoGS indicators. Implementation of efforts to scale-up 29 30 surgery using these target goals can help Somaliland progress in improving safe, affordable, and 31 32 timely surgical, obstetric, and anesthesia care and achieve universal health coverage for its

33 http://bmjopen.bmj.com/ 34 35 population. 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

19 1 2 3 Author contributorship: 4 5 6 TC, EI, HR, and ES contributed to the conceptualization of the project. EAI and DP contributed 7 8 9 to the conceptualization of the project and writing of the manuscript. SD, TC, and MM oversaw 10 11 all data collection efforts. HR and ES oversaw the project. CCC, TC, HR, and ES conceptualized 12 13 the design, analyzed the data, and contributed to the writing and revising of the manuscript. AL 14 15 contributed to the data analysis, interpretation, and revising of the manuscript. All authors 16 For peer review only 17 18 contributed to the revising and approved the final version of the manuscript. 19 20 21 Competing Interests: None declared. For the original study on which this article is based, Ms 22 23 Concepcion and Dr Smith received grants from Duke University’s Duke Global Health Institute 24 25 and grants from Baylor University. No other disclosures were reported. 26 27 28 Funding: None. Funding for the original study on which this article is based was provided by 29 30 31 Duke University’s Duke Global Health Institute (Dr Smith) and by Baylor University’s Robbins 32

33 College of Health and Human Sciences (Dr Smith). http://bmjopen.bmj.com/ 34 35 36 Ethical approval: Institutional review board (IRB) approval was granted from Duke University. 37 38 Since Somaliland does not have a national IRB, a letter of approval for the study was obtained 39 40 from the Somaliland Ministry of Health.

41 on September 27, 2021 by guest. Protected copyright. 42 43 Data sharing statement: Data are available by request by emailing [email protected] 44 45 46 Acknowledgements: We want to thank the Global Initiative for Children’s Surgery (GICS) for 47 48 49 its support of this work. GICS (www.globalchildrenssurgery.org) is a network of children’s 50 51 surgical and anesthesia providers from low-, middle-, and high-income countries collaborating 52 53 for the purpose of improving the quality of surgical care for children globally. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

20 1 2 3 4 5 6 REFERENCES 7 8 9 1. Meara J, Leather A, Hagander L, et al. Global Surgery 2030: evidence and solutions for achieving 10 health, welfare, and economic development. The Lancet 2015;386(9993):569-624. doi: 11 10.1016/S0140-6736(15)60160-X 12 2. Fuller AT, Corley J, Tran TM, et al. Prevalence of Surgically Untreated Face, Head, and Neck 13 Conditions in Uganda: A Cross-Sectional Nationwide Household Survey. World Neurosurgery 14 2018;110:e747-e54. doi: 10.1016/j.wneu.2017.11.099 [published Online First: 2017/11/29] 15 3. Bearden A, Fuller AT, Butler EK, et al. Rural and urban differences in treatment status among children 16 For peer review only 17 with surgical conditions in Uganda. PloS one 2018;13(11):e0205132. doi: 18 10.1371/journal.pone.0205132 [published Online First: 2018/11/02] 19 4. Chao TE, Sharma K, Mandigo M, et al. Cost-effectiveness of surgery and its policy implications for 20 global health: a systematic review and analysis. The Lancet global health 2014;2(6):e334-e45. 21 doi: 10.1016/S2214-109X(14)70213-X 22 5. Smith ER, Concepcion TL, Niemeier KJ, et al. Is Global Pediatric Surgery a Good Investment? World 23 Journal of Surgery 2019;43(6):1450-55. doi: 10.1007/s00268-018-4867-4 24 6. Smith ER, Concepcion TL, Mohamed M, et al. The contribution of pediatric surgery to poverty 25 trajectories in Somaliland. PloS one 2019;14(7):e0219974. doi: 10.1371/journal.pone.0219974 26 7. World Health Organization. Global Reference List of 100 Core Health Indicators, 2015: Metadata 2020 27 [Available from: https://www.who.int/healthinfo/indicators/2015/metadata/en/ accessed 28 September 7 2020. 29 8. World Bank. World development indicators 2020 [Available from: 30 https://databank.worldbank.org/source/world-development-indicators accessed September 7 31 2020. 32 9. Bruno E, White M, Baxter L, et al. An Evaluation of Preparedness, Delivery and Impact of Surgical 33 and Anesthesia Care in Madagascar: A Framework for a National Surgical Plan. World Journal of http://bmjopen.bmj.com/ 34 Surgery 2017;41(5):1218-24. doi: 10.1007/s00268-016-3847-9 35 10. Sonderman KA, Citron I, Mukhopadhyay S, et al. Framework for developing a national surgical, 36 37 obstetric and anaesthesia plan. BJS Open 2019;3(5):722-32. doi: 10.1002/bjs5.50190 38 11. Elkheir N, Sharma A, Cherian M, et al. A cross-sectional survey of essential surgical capacity in 39 Somalia. BMJ open 2014;4(5):1-8. doi: 10.1136/bmjopen-2013-004360 40 12. Somaliland Ministry of Health. National health policy 2011 [Available from:

41 http://www.nhpcsomaliland.org/documents/Somaliland%20National%20Health_Policy.pdf on September 27, 2021 by guest. Protected copyright. 42 accessed July 16, 2019. 43 13. World Bank. New World Bank GDP and poverty estimates for Somaliland 2014 [Available from: 44 https://www.worldbank.org/en/news/press-release/2014/01/29/new-world-bank-gdp-and-poverty- 45 estimates-for-somaliland accessed July 16, 2019. 46 14. World Bank. Somaliland: poverty profile and overview of living conditions 2015 [Available from: 47 http://www.somalilandcsd.org/wp-content/uploads/2017/12/Somaliland-Poverty-Profile..pdf 48 accessed July 17, 2019. 49 15. United Nations International Children's Emergency Fund. Under-five mortality 2018 [Available from: 50 https://data.unicef.org/topic/child-survival/under-five-mortality/ accessed July 20, 2019. 51 16. United Nations International Children's Emergency Fund. Neonatal mortality 2018 [Available from: 52 https://data.unicef.org/topic/child-survival/neonatal-mortality/ accessed July 20, 2019. 53 17. World Bank. Africa development indicators 2012/13 2013 [Available from: 54 http://documents.worldbank.org/curated/en/509481467990362181/pdf/Africa-development- 55 56 indicators-2012-13.pdf accessed July 20, 2019. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

21 1 2 3 18. Krishnaswami S, Nwomeh BC, Ameh EA. The pediatric surgery workforce in low- and middle- 4 income countries: problems and priorities. Seminars in Pediatric Surgery 2016;25(1):32-42. doi: 5 10.1053/j.sempedsurg.2015.09.007 6 19. Concepcion T, Mohamed M, Dahir S, et al. Prevalence of Pediatric Surgical Conditions Across 7 Somaliland. JAMA network open 2019;2(1):e186857. doi: 10.1001/jamanetworkopen.2018.6857 8 20. Concepcion TL, Smith ER, Mohamed M, et al. Provision of Surgical Care for Children Across 9 10 Somaliland: Challenges and Policy Guidance. World Journal of Surgery 2019;43(11):2934. doi: 11 10.1007/s00268-019-05079-8 12 21. United Nations Population Fund. Population estimation survey 2014 for the 18 pre-war regions of 13 Somalia, 2014. 14 22. Concepcion TL, Dahir S, Mohamed M, et al. Barriers to Surgical Care Among Children in 15 Somaliland: An Application of the Three Delays Framework. World Journal of Surgery 16 2020;44(6):1712-18.For doi: peer 10.1007/s00268-020-05414-4 review [published only Online First: 2020/02/08] 17 23. World Bank. The Uganda Poverty Assessment Report 2016, 2016:141. 18 24. Yousef Y, St-Louis E, Baird R, et al. A systematic review of capacity assessment tools in pediatric 19 surgery: Global Assessment in Pediatric Surgery (GAPS) Phase I. Journal Of Pediatric Surgery 20 2019;54(4):831-37. doi: 10.1016/j.jpedsurg.2018.11.005 21 25. Verguet S, Alkire BC, Bickler SW, et al. Timing and cost of scaling up surgical services in low- 22 income and middle-income countries from 2012 to 2030: a modelling study. The Lancet Global 23 Health 2015;3(S2):S28-S37. doi: 10.1016/S2214-109X(15)70086-0 24 26. Anderson GA, Ilcisin L, Kayima P, et al. Out-of-pocket payment for surgery in Uganda: The rate of 25 impoverishing and catastrophic expenditure at a government hospital. PloS one 26 2017;12(10):e0187293. doi: 10.1371/journal.pone.0187293 27 27. Holmer H, Bekele A, Hagander L, et al. Evaluating the collection, comparability and findings of six 28 global surgery indicators. British Journal of Surgery 2019;106(2):138-50. doi: 10.1002/bjs.11061 29 30 28. Albutt K, Punchak M, Kayima P, et al. Access to Safe, Timely, and Affordable Surgical Care in 31 Uganda: A Stratified Randomized Evaluation of Nationwide Public Sector Surgical Capacity and 32 Core Surgical Indicators. World Journal of Surgery 2018;42(8):2303-13. doi: 10.1007/s00268- 018-4485-1 33 http://bmjopen.bmj.com/ 34 29. Guest GD, McLeod E, Perry WRG, et al. Collecting data for global surgical indicators: a collaborative 35 approach in the Pacific Region. BMJ Global Health 2017;2(4):e000376-e76. doi: 10.1136/bmjgh- 36 2017-000376 37 30. Linden A, Sekidde F, Galukande M, et al. Challenges of Surgery in Developing Countries: A Survey 38 of Surgical and Anesthesia Capacity in Uganda’s Public Hospitals. World Journal of Surgery 39 2012;36(5):1056-65. doi: 10.1007/s00268-012-1482-7 40 31. Knowlton L, Chackungal S, Dahn B, et al. Liberian Surgical and Anesthesia Infrastructure: A Survey

41 of County Hospitals. World Journal of Surgery 2013;37(4):721-29. doi: 10.1007/s00268-013- on September 27, 2021 by guest. Protected copyright. 42 1903-2 43 32. Ahuka O. La cartographie des médecins spécialistes dans les quatre grandes disciplines cliniques ainsi 44 que l’anesthésie réanimation en République Démocratique du Congo (RDC): État des lieux. 45 Revue médicale des Grands Lacs 2014;3:156-69. 46 33. Bolkan HA, Von Schreeb J, Samai MM, et al. Met and unmet needs for surgery in Sierra Leone: A 47 comprehensive, retrospective, countrywide survey from all health care facilities performing 48 49 operations in 2012. Surgery 2015;157(6):992. doi: 10.1016/j.surg.2014.12.028 50 34. Zafar SN, Canner JK, Nagarajan N, et al. Road traffic injuries: Cross-sectional cluster randomized 51 countrywide population data from 4 low-income countries. International Journal of Surgery 52 2018;52:237-42. doi: 10.1016/j.ijsu.2018.02.034 [published Online First: 2018/02/23] 53 35. Davé DR, Nagarjan N, Canner JK, et al. Rethinking burns for low & middle-income countries: 54 Differing patterns of burn epidemiology, care seeking behavior, and outcomes across four 55 countries. Burns : journal of the International Society for Burn Injuries 2018;44(5):1228-34. doi: 56 10.1016/j.burns.2018.01.015 [published Online First: 2018/02/25] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

22 1 2 3 36. Fuller AT, Haglund MM, Lim S, et al. Pediatric Neurosurgical Outcomes Following a Neurosurgery 4 Health System Intervention at Mulago National Referral Hospital in Uganda. World 5 Neurosurgery 2016;95:309-14. doi: 10.1016/j.wneu.2016.07.090 [published Online First: 6 2016/08/09] 7 37. Abdelgadir J, Punchak M, Smith ER, et al. Pediatric traumatic brain injury at Mbarara Regional 8 Referral Hospital, Uganda. Journal of Clinical Neuroscience 2018;47:79-83. doi: 9 10 10.1016/j.jocn.2017.10.004 [published Online First: 2017/10/27] 11 38. Leather A, Ismail EA, Ali R, et al. Working together to rebuild health care in post-conflict 12 Somaliland. The Lancet 2006;368(9541):1119-25. doi: 10.1016/S0140-6736(06)69047-8 13 39. World Bank. Somali joint needs assessment. Infraestructure cluster report., 2006:90. 14 40. Dieleman J, Campbell M, Chapin A, et al. Evolution and patterns of global health financing 1995– 15 2014: development assistance for health, and government, prepaid private, and out-of-pocket 16 health spendingFor in 184 countries.peer The Lancetreview 2017;389(10083):1981-2004. only doi: 10.1016/S0140- 17 6736(17)30874-7 18 41. World Health Organization, World Bank. Global Monitoring Report on Financial Protection in Health 19 2019. Advance copy, 2019:47. 20 42. Saxton AT, Poenaru D, Ozgediz D, et al. Economic Analysis of Children's Surgical Care in Low- and 21 Middle-Income Countries: A Systematic Review and Analysis. PloS one 2016;11(10):e0165480. 22 doi: 10.1371/journal.pone.0165480 [published Online First: 2016/10/30] 23 43. Watters AD, Guest DG, Tangi GV, et al. Global Surgery System Strengthening: It Is All About the 24 Right Metrics. Anesthesia & Analgesia 2018;126(4):1329-39. doi: 25 10.1213/ANE.0000000000002771 26 44. Smith ER, Concepcion TL, Shrime M, et al. Waiting Too Long: The Contribution of Delayed Surgical 27 Access to Pediatric Disease Burden in Somaliland. World Journal of Surgery 2020;44(3):656-64. 28 doi: 10.1007/s00268-019-05239-w [published Online First: 2019/10/28] 29 30 45. Butler EK, Tran TM, Fuller AT, et al. Quantifying the pediatric surgical need in Uganda: results of a 31 nationwide cross-sectional, household survey. Pediatric Surgery International 2016;32(11):1075- 32 85. doi: 10.1007/s00383-016-3957-3 [published Online First: 2016/09/12] 46. Butler EK, Tran TM, Nagarajan N, et al. Epidemiology of pediatric surgical needs in low-income 33 http://bmjopen.bmj.com/ 34 countries. PloS one 2017;12(3):e0170968. doi: 10.1371/journal.pone.0170968 35 47. Smith ER, Vissoci JRN, Rocha TAH, et al. Geospatial analysis of unmet pediatric surgical need in 36 Uganda. Journal of Pediatric Surgery 2017;52(10):1691-98. doi: 10.1016/j.jpedsurg.2017.03.045 37 48. Vissoci JRN, Ong CT, Andrade Ld, et al. Disparities in surgical care for children across Brazil: Use of 38 geospatial analysis. PloS one 2019;14(8):e0220959. doi: 10.1371/journal.pone.0220959 39 49. Abdul-Mumin A, Anyomih TTK, Owusu SA, et al. Burden of Neonatal Surgical Conditions in 40 Northern Ghana. World Journal of Surgery 2019 doi: 10.1007/s00268-019-05210-9

41 50. Concepcion TL, Smith ER, Mohamed M, et al. Provision of Surgical Care for Children Across on September 27, 2021 by guest. Protected copyright. 42 Somaliland: Challenges and Policy Guidance. World Journal of Surgery 2019;43(11):2934. doi: 43 10.1007/s00268-019-05079-8 44 51. Abdelgadir J, Smith E, Punchak M, et al. Outcomes and Predictors of Mortality in Neurosurgery 45 Patients at Mbarara Regional Referral Hospital, Mbarara, Uganda. Annals of Global Health 46 2017;83(1):18-19. doi: 10.1016/j.aogh.2017.03.040 47 52. World Health Organization, Program in Global Surgery and Social Change. Surgical care systems 48 49 strengthening: Developing national surgical, obstetric and anaesthesia plans World Health 50 Organization, 2017. 51 53. Ministry of Health of Zambia. National Surgical, Obstetric, and Anaesthesia Strategic Plan 52 (NSOASP). 53 54. Mukhopadhyay S, Lin Y, Mwaba P, et al. Implementing World Health Assembly Resolution 68.15: 54 National surgical, obstetric, and anesthesia strategic plan development--the Zambian experience. 55 Bulletin of the American College of Surgeons 2017;102(6):28-35. [published Online First: 56 2017/09/09] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

23 1 2 3 55. Burssa D, Teshome A, Iverson K, et al. Safe Surgery for All: Early Lessons from Implementing a 4 National Government-Driven Surgical Plan in Ethiopia. World Journal of Surgery 5 2017;41(12):3038-45. doi: 10.1007/s00268-017-4271-5 6 56. Burssa D, Teshome A, Iverson K, et al. Safe Surgery for All: Early Lessons from Implementing a 7 National Government-Driven Surgical Plan in Ethiopia. World Journal Surgery 8 2017;41(12):3038-45. doi: 10.1007/s00268-017-4271-5 [published Online First: 2017/10/17] 9 10 57. Raykar N, Mukhopadhyay S, Saluja S, et al. Implementation of The Lancet Commission on Global 11 Surgery in India. Healthcare 2019;7(2):4-6. doi: 10.1016/j.hjdsi.2018.11.001 12 58. Citron I, Jumbam D, Dahm J, et al. Towards equitable surgical systems: development and outcomes 13 of a national surgical, obstetric and anaesthesia plan in Tanzania. BMJ Global Health 2019;4(2) 14 doi: 10.1136/bmjgh-2018-001282 15 59. Stewart BT, Tansley G, Gyedu A, et al. Mapping Population-Level Spatial Access to Essential 16 Surgical CareFor in Ghana peerUsing Availability review of Bellwether Procedures.only JAMA Surgery 17 2016;151(8):e161239. doi: 10.1001/jamasurg.2016.1239 18 60. Esquivel MM, Uribe-Leitz T, Makasa E, et al. Mapping Disparities in Access to Safe, Timely, and 19 Essential Surgical Care in Zambia. JAMA Surgery 2016;151(11):1064-69. doi: 20 10.1001/jamasurg.2016.2303 21 61. Keyes EB, Parker C, Zissette S, et al. Geographic access to emergency obstetric services: a model 22 incorporating patient bypassing using data from Mozambique. BMJ Global Health 2019;4(Suppl 23 5) doi: 10.1136/bmjgh-2018-000772 24 62. Federspiel F, Mukhopadhyay S, Milsom PJ, et al. Global surgical, obstetric, and anesthetic task 25 shifting: A systematic literature review. Surgery 2018;164(3):553-58. doi: 26 10.1016/j.surg.2018.04.024 [published Online First: 2018/08/28] 27 63. Saluja S, Silverstein A, Mukhopadhyay S, et al. Using the Consolidated Framework for 28 Implementation Research to implement and evaluate national surgical planning. BMJ global 29 30 health 2017;2(2):e000269-e69. doi: 10.1136/bmjgh-2016-000269 31 64. Global Initiative for Children’s Surgery. Optimal Resources for Children’s Surgical Care: Executive 32 Summary. World Journal of Surgery 2019;43(4):978-80. doi: 10.1007/s00268-018-04888-7 65. Global Initiative for Children's Surgery. Global Initiative for Children's Surgery: A Model of Global 33 http://bmjopen.bmj.com/ 34 Collaboration to Advance the Surgical Care of Children. World Journal of Surgery 35 2019;43(6):1416-25. doi: 10.1007/s00268-018-04887-8 [published Online First: 2019/01/10] 36 66. Hanna JS, Herrera-Almario GE, Pinilla-Roncancio M, et al. Use of the six core surgical indicators 37 from the Lancet Commission on Global Surgery in Colombia: a situational analysis. The Lancet 38 Global Health 2020;8(5):e699-e710. doi: 10.1016/S2214-109X(20)30090-5 39 67. Donabedian A. The quality of care. How can it be assessed? Jama 1988;260(12):1743-8. doi: 40 10.1001/jama.260.12.1743 [published Online First: 1988/09/23]

41 68. Massenburg BB, Saluja S, Jenny H, et al. Assessing the Brazilian surgical system with six surgical on September 27, 2021 by guest. Protected copyright. 42 indicators: a descriptive and modelling study. BMJ Global Health 2017;2(2) doi: 10.1136/bmjgh- 43 2016-000226 44 69. Krishna A. Escaping Poverty and Becoming Poor: Who Gains, Who Loses, and Why? World 45 Development 2004;32(1):121-36. doi: https://doi.org/10.1016/j.worlddev.2003.08.002 46 70. Krishna A. Who became poor, who escaped poverty, and why? Developing and using a retrospective 47 methodology in five countries: Who Became Poor, Who Escaped Poverty, and Why? Journal of 48 49 policy analysis and management 2010;29(2):351-72. doi: 10.1002/pam.20495 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

24 1 2 3 FIGURE LEGEND 4 5 6 Figure 1. Somaliland Indicators Compared to the Lancet Commission on Global Surgery targets. 7 8 9 Figure 2. Catchment population within a 2-hour travel time from the closest surgically-capable 10 11 hospital. 12 13 14 Figure 3. Lancet Commission on Global Surgery indicators for Somaliland, stratified by region. 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

1 2 3 4 5 SOMALILAND DEFICIT 6 MET GOAL UNMET GOAL 7 8 9 10 11 12 13 14 Indicator 1: Access to timely surgery 25% 75% 15 16 17 18 19 20 21 22 23 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from 24 Indicator 2: Surgical workforce density 5% 95% 25 26 27 28 29 30 31 32 33 For peer review only 34 Indicator 3: Surgical volume 8% 9… 35 36 37 38 39 40 41 42 43 44Indicator 5: Protection against impoverishing expenditure 45 18% 82% 46 47 48 49 50 51 52 53 54 55 Indicator 6: Protection against catastrophic expenditure 1% 99% 56

57 http://bmjopen.bmj.com/ 58 59 60 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% on September 27, 2021 by guest. Protected copyright.

Indicator 1 Indicator 2 Indicator 3 Indicator 5 Indicator 6 Somaliland 18.8% coverage 0.8 SAO/100,000 population 368.8 procedures/100,000 population 18% protection 1% protection LCoGS Target 80% coverage 20 SAO/100,000 population 5,000 procedures/100,000 population 100% protection 100% protection

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

33 279x215mm (300 x 300 DPI) http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 29 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from

33 http://bmjopen.bmj.com/ 34 Indicator 5: protection against impoverishment Target: 100% protection 35 36 100% 37 80% 38 60% 39 40% 25.5% 40 16.3% 15.7% 20.3% 19.9% 20.8%

20%

41 on September 27, 2021 by guest. Protected copyright. 42 0% 43 44 Indicator 6: protection against catastrophic

1 2 STROBE Statement— “Interpreting the Lancet Surgical Indicators in Somaliland: a Cross-Sectional Study” 3 BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from 4 5 6 Item Recommendation page 7 No 8 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the 1 9 10 abstract 11 (b) Provide in the abstract an informative and balanced summary of what was 2 12 done and what was found 13 14 Introduction 15 Background/rationale 2 Explain the scientific background and rationale for the investigation being 5 16 reported 17 Objectives 3 State specific objectives, including any prespecified hypotheses 6 18 For peer review only 19 Methods 20 Study design 4 Present key elements of study design early in the paper 6 21 Setting 5 Describe the setting, locations, and relevant dates, including periods of 6 22 23 recruitment, exposure, follow-up, and data collection 24 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of 6 25 participants 26 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, 7 28 and effect modifiers. Give diagnostic criteria, if applicable 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 7 30 measurement assessment (measurement). Describe comparability of assessment methods if 31 32 there is more than one group 33 Bias 9 Describe any efforts to address potential sources of bias n/a 34 Study size 10 Explain how the study size was arrived at 7 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If 7 36 applicable, describe which groupings were chosen and why

37 http://bmjopen.bmj.com/ 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for 7 39 confounding 40 41 (b) Describe any methods used to examine subgroups and interactions n/a 42 (c) Explain how missing data were addressed n/a 43 (d) If applicable, describe analytical methods taking account of sampling n/a 44 strategy

45 on September 27, 2021 by guest. Protected copyright. 46 (e) Describe any sensitivity analyses n/a 47 Results 48 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 11 49 50 potentially eligible, examined for eligibility, confirmed eligible, included in 51 the study, completing follow-up, and analysed 52 (b) Give reasons for non-participation at each stage n/a 53 54 (c) Consider use of a flow diagram n/a 55 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, n/a 56 social) and information on exposures and potential confounders 57 (b) Indicate number of participants with missing data for each variable of n/a 58 59 interest 60 Outcome data 15* Report numbers of outcome events or summary measures 11 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted n/a

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 29

1 2 estimates and their precision (eg, 95% confidence interval). Make clear 3 which confounders were adjusted for and why they were included BMJ Open: first published as 10.1136/bmjopen-2020-042968 on 29 December 2020. Downloaded from 4 5 (b) Report category boundaries when continuous variables were categorized n/a 6 (c) If relevant, consider translating estimates of relative risk into absolute risk n/a 7 for a meaningful time period 8 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and n/a 9 10 sensitivity analyses 11 Discussion 12 Key results 18 Summarise key results with reference to study objectives 13 13 14 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias 16 15 or imprecision. Discuss both direction and magnitude of any potential bias 16 Interpretation 20 Give a cautious overall interpretation of results considering objectives, 13 17 18 Forlimitations, peer multiplicity review of analyses, results only from similar studies, and other 19 relevant evidence 20 Generalisability 21 Discuss the generalisability (external validity) of the study results n/a 21 22 Other information 23 Funding 22 Give the source of funding and the role of the funders for the present study 19 24 and, if applicable, for the original study on which the present article is based 25 26 27 *Give information separately for exposed and unexposed groups. 28 29 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 30 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 31 32 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 33 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 34 available at www.strobe-statement.org. 35 36

37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44

45 on September 27, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml