3 5 PART 1: SUB-SAHARAN AFRICA AND BURDEN OF

Transcription

1

2 TABLE OF CONTENTS 1. ABSTRACT...Page 3 2. INTRODUCTION...Page 5 3. PART 1: SUB-SAHARAN AFRICA AND BURDEN OF DISEASE...Page 6 a. Contribution of non-communicable diseases and injury to burden of disease in the region...page 6 b. Burden of surgical disease, trauma and injury in the region...page 9 c. NCDs, surgical disease and global health funding....page 12 d. Access to surgical care in LMICs, and brain drain of healthcare workers in the region...page PART II: REGION-SPECIFIC SOLUTIONS TO BRAIN DRAIN AND SURGICAL TRAINING...Page 17 a. Policy proposals and delivery platforms...page 18 b. Foreign surgical care provision in sub-saharan Africa.Page 18 c. Non-surgical providers delivering surgical care...page 19 d. Development of future physicians and training of sub-saharan African surgeons....page PART III: MALAWI..Page 25 a. Background:.Page 25 b. Health indicators, health funding and expenditures..page 26 c. Malawi burden of trauma and surgical disease.....page 26 d. Healthcare system and delivery page 27 e. Access to surgical care and surgical training model in Malawi Page PART IV: MALAWI: TRAINING GENERAL SURGEONS REDUCES TRAUMA MORTALITY. Page ACKNOWLEDGEMENTS...Page REFERENCES...Page 36

3 ABSTRACT BACKGROUND: Many factors contribute to the burden of surgical diseases, such as trauma, in sub-saharan Africa, including a paucity of international funding, a shortage of trained clinicians, and underdeveloped infrastructure. Training surgical providers is one solution that has been proposed to reduce the mortality and morbidity due to these diseases in the region. Kamuzu Central Hospital in Lilongwe, Malawi, in partnership with the University of North Carolina at Chapel Hill, established a general surgery residency program in The purpose of this study is to evaluate changes in trauma-related mortality following the institution of this residency program. METHODS: We review data related to the burden of noncommunicable and surgical disease in sub-saharan Africa, as well as access to surgical providers and the shortage of healthcare workers in the region. We perform a literature review of proposals aimed at improving the access to surgical services in sub-saharan Africa. We then present Malawi-specific data, including economic and health indicators, healthcare structure, access to surgical providers, and traumarelated burden of disease. Finally, to determine the influence of the general surgery residency on trauma related mortality at Kamuzu Central Hospital, we perform a retrospective analysis of patients in the hospital s trauma surveillance database, from 2009 to A logistic regression model was constructed to compare the odds ratio of in-hospital death. RESULTS: In 2009 there were 3 general surgery residents at KCH; this number increased to eleven residents in During the period of study, 82,534 patients were recorded into the trauma registry database at KCH; the majority were male (72.1%), and mean age was 23.1 years (SD: 15.7). Trauma volume increased for each of the years under study, with 8725 patients 3

4 recorded in 2009 and 15,998 patients in Odds of in-hospital death decreased every year as compared to the referent year, when adjusted for primary injury type (severity); age; and gender. CONCLUSION: While surgical burden of disease in sub-saharan Africa is a complicated issue, training general surgeons in a resource-poor setting represents an effective intervention to help reduce trauma mortality in the region. 4

5 INTRODUCTION The United Nations defines sub-saharan Africa as all of Africa except northern Africa. Northern Africa includes Algeria, Egypt, Libya, Morocco, Tunisia and West Sahara; Sudan, geographically in the North, is considered sub-saharan. The area is further subdivided into southern, eastern, middle, and western Africa. Larger than China, Japan, the UK, India, United States and Eastern Europe combined, it includes fifty-two countries, from the relatively well-developed South Africa to some of the world s poorest countries, such as Malawi, Niger and Rwanda. 1 Although communicable diseases receive the bulk of attention and funding, the burden of noncommunicable and surgical disease in sub-saharan Africa is great. Beyond its immense effect on human suffering, disease also hampers development in the region and maintains the cycle of poverty and economic depression. Data does suggest that investment in the training and development of a surgical workforce in the region represents a worthwhile and cost effective intervention to lessen the mortality and morbidity of surgical disease. Many rural district hospitals currently do not have access to a surgical provider. Several surgical residencies have been established in collaboration with or input from developed countries, and there are some published studies on how the volume of procedures and the case mix of these residencies compares to the Western standards. However, to our knowledge, no study has evaluated whether the institution of a general surgery residency meaningfully reduces mortality from a given surgical disease. The Department of Surgery at the University of North Carolina at Chapel Hill has collaborated with Kamuzu Central Hospital in Lilongwe, Malawi, since A general surgery residency was established in 2009, in collaboration with Haukeland University in Norway. The 5

6 purpose of the present study is to evaluate whether the presence of general surgery trainees has reduced trauma mortality at Kamuzu Central Hospital. This paper is divided into three sections. In the first part, I will discuss the burden of noncommunicable disease and injury in sub-saharan Africa, including mortality, morbidity, and funding challenges. I will then discuss the brain drain of healthcare workforce in the region. Part two focuses on a review of proposed solutions to the brain drain, along with the models of training surgical providers that have been attempted in the region. The third part of this paper will describe Malawi in terms of its economic and health indicators, its healthcare delivery structure, and surgical workforce training models. I will finish by discussing the establishment of a general surgery residency at one of the central hospitals in Malawi, and the reduction in trauma-related mortality at this hospital following the establishment of this residency. PART 1: SUB-SAHARAN AFRICA AND BURDEN OF DISEASE a) Contribution of non-communicable diseases and injury to burden of disease in the region. HIV/AIDS, malaria, TB, malnutrition, and maternal and childhood deaths have traditionally focused the world s attention in sub-saharan Africa, and continue to affect a large number of people. Whereas globally, non-communicable diseases (NCDs) and injuries caused 78.47% of total deaths in 2013, in sub-saharan Africa NCDs and injuries killed 37.25%, with the remainder of deaths coming from communicable, nutritional, maternal and neonatal diseases (CNMNDs) combined. But although CNMNDs cause such a significant proportion of overall deaths, NCDs and injuries contribute more significantly to the mortality of individual groups in the region. For example, 32% of global deaths, and 19.43% of sub-saharan deaths, among males aged resulted from injuries; in contrast, the total percentages of deaths due to injuries in all 6

7 age groups were 8.72% globally and 7.24% in sub-saharan Africa. 2 Furthermore, while HIV/AIDS caused the largest proportion of deaths in this group, at 33.88%, road injuries outpaced TB as the second most common cause of death, at 8.97%. And, while deaths attributable to almost all communicable, maternal and nutritional diseases, with the exception of HIV/AIDS, decreased annually, deaths due to road injuries increased. This group represents the most economically productive individuals, and the disease burden in this group thus has important implications for country s economic viability. Diseases and injuries affect a country s workforce by more than simple subtraction: disability may remove individuals from the employment pool as effectively as death. And, individuals with disability further strain a healthcare delivery system in ways that dead individuals do not. For this reason, measures such as Disability-Adjusted Life Years (DALYs), Years of Life Lost (YLLs), and Years Lived with Disability (YLDs) represent important measurements to determine a burden of disease in a region. According to the World Health Organization, YLL is calculated by multiplying the standard life expectancy for an age at which death occurs by the number of deaths: YLL=N*L, where N is the number of deaths, and L is the standard life expectancy in years at the age of death. YLDs are calculated by taking into perspective the degree of disability that can be expected from a particular disease. This is accomplished by assigning a weight to disease severity, ranging from 0 (perfect health) to 1 (dead), as follows: YLD=I * DW * L, where I=number of cases, DW is the disability weight (0-1), and L is the average duration of the disease (defined as time to remission or death, in years). 7

8 Finally, DALYs take into account both YLLs and YLD: DALY= YLL +YLD; the World Health Organization (WHO) states one DALY can be thought of as one lost year of healthy life. 3 The weight of disease severity is a subjective measurement, and the disability weights in the Global Burden of Disease study are calculated based, in part, on questionnaires from Australia and the Netherlands. The cross-applicability of these values from a developed to a developing society, therefore, may not be complete; nonetheless, DALYs are currently the most commonly employed method of valuing the morbidity incurred from a disease. In order to minimize their subjectivity, the WHO assigns standardized weights, regardless of an individual s race, country of origin, or economic status, and uses an ideal rather than actual life expectancy when calculating them. 4 The weights are further adjusted to discount (at 3%) the years lived at the very young and the very old ends of the spectrum. 4 In practice, this means that the weight assigned to cancer is lower than adult-onset hearing loss, for example. Injury and non-communicable diseases tend to carry a higher weight than do communicable diseases; for example, the weight assigned to tuberculosis is the same as that assigned to a patellar fracture at and the weight assigned to even untreated AIDS is less than that assigned to a spinal cord injury (0.505 vs 0.725). 4 Because of the discounting for age extremes, a given disease can cause a disproportionate amount of DALYs if it is very common, enables the sufferer to live a long but disabled life (like depression); or is extremely disabling (for example, the disability weight for stroke approaches 1, and thus cardiovascular disease contributes to 4.62% of DALYs worldwide, whereas HIV accounts for 2.84% of worldwide DALYs). 2 The overall burden of DALYs in HICs and LMICs mirrors that of overall deaths: in HICs, NCDs contribute 79-88% of total DALYs for all ages and sexes, with injuries contribute 8

9 7.58%-10.67%. In sub-saharan Africa, the bulk of disability comes from communicable, maternal, neonatal and nutritional diseases, while NCDs cause between 26.59% and 29.04%, and injuries contribute 6.38% to 7.74%. However, adults who survive to an age where NCDs become more prevalent (>50) suffer the majority of DALYs from these causes even in sub-saharan Africa. And, in this region males years old suffer double the DALYs due to injury of the overall population in the region, at 15-19%. 2 b) Burden of surgical disease, trauma and injury in the region. Although trauma and injury are surgical diseases, NCDs cannot be as easily conflated with surgical disease, and vice versa. However, patients suffering from NCDs commonly require surgical attention, and thus the burden of disease from NCDs in sub-saharan Africa also indirectly denotes surgical need. Surgical diseases are estimated to contribute 11-15% to the total global burden of disease 5,6 A 2012 systematic review of the burden of surgical disease in sub-saharan Africa found 39 studies, from which the authors extrapolated the incidence of fatal and nonfatal injury, peritonitis, open bone fractures and inguinal hernia; the admissions needed and operations performed to address these diseases; and the unmet need (calculated by subtracting the latter from the former). The authors estimated that the overall incidence of nonfatal injury is at least 1,690/100,000 population per year, with morbidity up to 190/100,000 population per year, and the annual mortality from injury is 53-92/100, They estimated that the unmet need for open fracture treatment is 194/100,000 population per year. Because there was no data on acute abdomen incidence in sub-saharan Africa, the authors used a study from Pakistan to compare the incidence with the laparotomy rate of 14/100,000, to arrive at an unmet need of approximately 1,318/100,000 population per year. There were wide differences in the estimated incidence of inguinal hernias in the region, with one study reporting 205/100,000 population per year and 9

10 another 1400/100,000 per year. Using the conservative estimate, the authors report an unmet need of inguinal herniorrhaphy at 175/100,000 population per year, although they argue the real need is likely much higher. 7 Weiser et al used modeling to develop estimates of the global volume of surgery. 8 The authors queried WHO data to determine death rates per country, physician and other healthcare worker density, GDP and other statistics. They gathered data from 29% of the surveyed countries (56 out of 192) to estimate surgical volumes, and calculated their yearly surgical rates per 100,000 people. 16% of the countries with available data were African, but 27% of the countries without data were also from this region. The study showed that 30.2% of the world s population received 73.6% of all surgeries, whereas the poorest 34.8% received 3.5% of all surgeries. Globally, 38DALYs per 100 people are lost in sub-saharan Africa from conditions treatable by surgery, in contrast to less than 9 in the developed world. 6 Trauma-related mortality and morbidity in sub-saharan Africa deserves a separate mention. Trauma comprises a diverse set of injury causes, including road traffic injuries (RTIs), falls, animal bites, interpersonal violence, and burns, and its contribution to the worldwide and regional burden of disease is considerable. In 2010, RTIs and falls caused 17.5% and 12.2% of the total burden of DALYs worldwide. Of those, 94% were lost in LMICs. 9 A 2013 WHO Global Report on Traffic Safety showed that, although LMICs have 54% of the world s automobiles, they bear the burden of 90% of the RTI deaths. In 2015 in sub-saharan Africa, estimated deaths attributable to RTIs represented 1.77% of all deaths and 1.63% of total DALYs, with the age group having double those rates, at 3.09% of deaths, and 2.79% of DALYs. 4.66% of all males in this group died from road injuries, and 6.6% from self-harm or interpersonal violence. 10 And, although HIV continued to cause the largest amount of deaths in 10

11 this age and sex group, malaria-related deaths were only half of those due to RTIs. 2 Pediatric injury deaths, caused mainly by RTIs and drowning, number close to 1 million, 11 and children surviving to the age of 5 are most likely to die from trauma until they reach the age of A full 25% of all external cause deaths in the under-15 year old population in sub-saharan Africa are due to trauma, with the most common ones (in order) being RTIs, suicides and assaults. 11 According to the INDEPTH study, which follows residents of selected global sites longitudinally and collects epidemiologic data, RTIs have the highest age, time, and sex-standardized mortality rates per 1000 person-years for adults at all sites in Africa 11,13 RTI injuries in LMICs cause up to 5% of annual GDP loss. 14 In a 2002 study of surgical admissions in a Kenyan hospital for , the authors found that of 5907 admissions, 3411 were trauma-related (73.5%), 1499 (25.4%) were emergent surgical non-trauma admissions; and 997 (16.8%) were elective surgical admissions. 73.5% of all trauma admissions were male, and 57.6% were in the year olds. Here again, RTIs represented the highest proportion of trauma admissions, at 32.7%; they were followed by assaults at 23.8%; falls at 15.5%, and burns at 13%. 51.7% of all surgeries were trauma-related, and trauma was the leading cause of death for all surgical patients, at 6.6%. 15 A 2007 retrospective review of deaths in South Africa found 33,484 injury-related deaths out of a total of 601,113 (5.57%), of which almost 36% were violence-related and 32% were due to RTIs. 16 In a different Kenyan study, 17 reviewing causes of death for a rural district that is part of the Health and Demographic Surveillance system from the INDEPTH network, the authors performed Verbal Autopsies on deaths from 358 villages encompassing approximately 220,000 people, from They found 11,147 deaths, with 447 due to trauma (4%), with a male preponderance (71%); trauma caused 17% of deaths in males aged The study also found 11

12 that trauma deaths increased from 2.5% of total to 5.9% between 2003 and Absence of care was the strongest predictor of trauma death, with an adjusted risk ratio of 12.2 (95% CI: ); age, sex, alcohol ingestion and place of death were also significant in adjusted multivariable analysis. 17 In a Kenyan analysis 18 of 22 hospitals which provide surgical internship training, a review 956 records during June-July of 2012 found a prevalence of admissions due to injuries to be 736/956, or 77%. There were 178 head injuries or 24% of injury admissions; 136 chest injuries or 18.5%; 237 long bone fractures 237, or 32.2%; and 185 burns 185, or 25.1%. The remainder of admissions was for acute abdomen (220, or 23% of total admissions). Overall operative rate was 18%, with the highest being acute abdomen admissions, of which 38.2% went to the operating room; and the lowest for head injury (3.9%). As in other trauma related studies, males predominated admissions for all injuries, with the highest proportion in head injury (81%), chest injury (71.7%) and long bone fractures (70.5%). 18 In Maputo, Mozambique, 19 of the total 517 patients admitted over a 5-month period in an urban hospital, RTIs represented 39.2% of all trauma admissions, with falls adding a 9.3% and burns 8.2%. The majority of patients were male, and aged This study also reported that, in Maputo, RTIs caused 43.7% of all registered deaths in 2000, followed by firearms (8.7%) and burns (7.8%), meaning that at least 60.2% of all deaths in that year were caused by surgical diseases. 19 c) NCDs, surgical disease and global health funding. In spite of their large contribution to mortality and morbidity in the region, NCDs and injuries receive almost no attention from global funders. This is in spite of many calls for an increase in government and international funding for non-communicable and surgical diseases, and in spite of evidence showing that investment in surgical disease is very cost-effective For example, the $8.7 billion spent by 12

13 the US on global health funding in 2013 was divided into HIV, global fund, maternal/child health, family planning/reproductive health, tuberculosis, nutrition, neglected tropical diseases, and other, 27 with no mention of NCDs. Globally, although funding for NCDs between 2004 and 2015 increased, growing by $260 million or 124%, funding for HIV/AIDS in the same period grew by $7.1 billion, or 192% 28 In 2015, funding for all noncommunicable diseases represented 1% of all global health financing. 29 And, with the exception of the Democratic Republic of Congo, Sierra Leone, Zambia, Botswana and Lesotho, where funding for NCDs per population approaches 1%, in other sub-saharan countries it is on the order of 0.001%. 28 Gutnik et al analyzed data from the NIH, USAID, Foundation Center, and registered US charitable organizations to determine funding for surgical disease in LMICs. 30 During , the authors found 22 NIH grants totaling $33.1 million; during the same time, there were 6 funded USAID projects, all having to do with obstetric fistula care. These expenditures represented less than 1% of the total operating budgets for both NIH and USAID. The same authors examined the databases of registered charitable organizations in the US, Canada, Australia, the UK, and New Zealand to find those devoted exclusively to surgical financing, and studied the fiscal data of these organizations from The authors found 160 organizations devoted to 15 surgical specialties, with total revenue from totaling $3.3 billion. The bulk of these organizations addressed specific surgical disease, with 45% focused on ophthalmology, and 26% on cleft lip and palate repairs. Only 14% of organizations funded diverse surgical specialty services. 31 The imbalance of funding versus contribution to mortality and morbidity has implications beyond the outright dollars and cents. Foreign aid represents the bulk of financing for many sub- Saharan health ministries and individual hospitals. Until at least 2015, foreign aid assistance was 13

14 tied to the achievement of objectives defined based on the Millennium Development Goals (MDGs), none of which addressed trauma, injury or NCDs. Because of this, aspects of the healthcare system not explicitly focused on MDGs, such as basic surgical care, were frequently compromised or neglected in favor of projects and evaluations that conformed to international funders targets. 22 Furthermore, international funding preferences strongly influence the recipient country s National Health Strategic Plans, which play an essential role in defining a country s vision, priorities, budgetary decisions and course of action for improving and maintaining the health of its people. 32 In a 2016 review of the available 43 of 55 African country National Health Strategic Plans, the authors found that, while all discussed targets for reduction of HIV and tuberculosis, and 42/43 had targets for reduction of malaria, maternal mortality, and child mortality, only 2.3% of all the targets dealt with surgical conditions, and 14 had no mention of surgical targets at all. Eight of the plans did not mention surgery. HIV and malaria received ten times the mentions that surgery did across all the documents (3801 to 379 times). 33 d) Access to surgical care in LMICs, and brain drain of healthcare workers in the region. The Lancet Commission identified access to surgical care as a pressing issue for up to 5.3 billion people, although its methods of estimation were somewhat generous. According to the modeling methodology, of the estimated million surgeries worldwide in 2012, only 6.3% were done in countries comprising the poorest 37.3% of the world s population. 34 Using complex methodologies, the commission estimated the minimum need for surgical procedures in order to achieve desirable outcomes, and found that 70% of the world s population lived in areas that did not have a large enough volume of procedures, defined as 5,000/100,000 population. According to a WHO 2006 report, a minimum threshold of 228 skilled health professionals per 100,000 population is needed for acceptable patient outcomes. Worldwide, 83 14

15 countries were below this number in 2012, with only a fifth of the world s specialist surgeons, a sixth of anesthesiologists, and a third of specialist obstetricians attending to the poorest half of the world s population. 12% of the world s surgeons practice in Africa, with a third of the world s population. In 2008, there were 39 pediatric surgeons in all of sub-saharan Africa. 35 Whereas in North America, one cardiac center serves approximately 120,000 people, in sub- Saharan Africa it serves 33,000, There are many reasons for the shortage of healthcare professionals in LMICs. 23,36 Brain drain is one of them, and has been studied in some detail. Highly skilled labor emigration happens because of wars, political unrest, social deprivation, lack of personal and professional advancement opportunities, recruitment from high income countries, and remuneration incentives, but the most commonly cited reasons for medical personnel emigration are professional, including the lack of opportunity for continuing medical education, advancement, and interaction with colleagues. Whereas globally in the 1970s, 6% of physicians and 5% of nurses lived abroad, 37, 39 in % of UK and 23.3% of US practicing physicians were foreign-trained. 37, 45 These numbers did not include the additional brain waste of medical professionals who emigrate but cannot or do not wish to practice their profession in their destination country. 45 In 2012, the US, UK and Canada employed 23-28% of all international medical graduates, and LMICs contributed 40-75% of their graduates to these countries, with the US and the UK attracting the highest raw numbers. Sub-Saharan Africa, India and the Caribbean had the highest numbers of migrants. 41 In terms of relative proportion of the workforce represented by the extremely vulnerable countries of sub-saharan Africa, in 2007 Portugal had the highest proportion of physicians from the region in its foreign-trained physician force, followed by 15

16 Canada, the UK, Australia, France, US and Germany. 47 Nigeria, Sudan, South Africa, Ghana, Uganda, Madagascar, Senegal, Sudan, Angola, Guinea and Mozambique sent the most physicians abroad. Surgeons followed the same trend. A study of US practicing general surgeons in 2008 showed that approximately 11% of them were foreign born international medical graduates, and a survey of foreign born surgeons from 2013 revealed that 56% of respondents came from countries the WHO considers particularly threatened by healthcare worker migration. 45, 46 Because in 2008, 24 of 47 sub-saharan African countries had only 1 medical school, and eleven had no schools at all, the loss of even one physician represented a serious loss for the health systems of these countries. 37 Inversely, HICs saved the costs of training their medical cadre. The investment in training a healthcare professional that then leaves the country strained the economies of LMICs, which lost an estimated $500 million invested in medical education annually because of healthcare worker migration. Ghana specifically lost approximately 35 million in physician training investment between 1998 and 2005, while at the same time the UK saved approximately 65 million by recruitment from the country; in 1999, 43% of Ghanaian physician posts were vacant. 47 The loss of medical workforce affects not only the access of a patient to a physician. It also has a profound effect on the workload and the isolation felt by physicians who remain behind. Furthermore, it affects rural areas at disproportionate rates, since migration takes place preferentially from these areas; and it negatively affects the training of future professionals. One study estimated that one physician will be expected to deliver HIV care to 26,000 patients per year, instead of the current 9,000, if migration of healthcare workers continues at current rates for the next five years. 44 The shortage of HCW in developing countries means that well-funded 16

17 initiatives such as PEPFAR and Global Fund malaria programs cannot impact the maximum numbers of people because they do not have the personnel to do so. 41 Emigration also exacerbates the losses from HIV/AIDS that do not spare the medical profession: between 1999 and 2005, for example, Botswana had lost 17% of its healthcare workforce to the disease, and in Lesotho and Malawi death due to AIDS is the most common cause of healthcare worker loss. 47 The loss of positive role models means that students are less likely to choose surgery as a career, as evidenced by a survey of 177 Nigerian senior medical students and interns, in which exposure to a positive surgical role model was the only modifiable factor associated in choosing surgery as a career (the other being sex and age). 49 Faculty surveyed in an Ethiopian hospital named deficient surgical education as the leading cause of limited surgical care in the country, with emigration seen as contributing significantly to that problem. 50 PART II. REGION-SPECIFIC SOLUTIONS TO BRAIN DRAIN AND SURGICAL NEED a) Policy proposals and delivery platforms Many solutions have been proposed to combat the loss of medical professionals and to address the pressing needs of the region in terms of surgical disease. Proposals range from policies aimed at improving retention (such as fiscal and professional incentives, requirements of local service in return for training, and outright prohibitions on migration) to international laws aimed at discouraging the active recruitment of healthcare workers by developed nations, to the improvement of working conditions by investment in infrastructure and training, to the training of non-physician clinicians without internationally-transferrable skills. 23,35-47,

18 b) Foreign surgical care provision in sub-saharan Africa. Supplementary surgical services provided by foreign medical professionals take place in three delivery models. The most common are surgical missions, short blitzkrieg endeavors that exploit existing infrastructure for the delivery of a high volume of specific surgeries by a visiting specialist, with little or no follow-up. The second platform involves bringing infrastructure and specialists to an area, and providing surgical care for varying amounts of time; this is the model used by Mercy Ships, for example. The third model involves the construction of specialized hospitals for the treatment of one particular problem, which are staffed by local or foreign specialists trained in the provision of specialized care for this one condition. Examples include obstetric fistula and cataract centers. In a 2015 systematic review of these models to determine their cost-effectiveness, sustainability, and role training, Shrime et al conducted a review of literature from , finding 104 pertinent publications. The authors concluded that short-term surgical missions are not costeffective and do not improve outcomes; in fact, because of their lack of follow-up and disruption of local infrastructure, many have unacceptably high complication rates. The specialty surgical hospitals and temporary complete platforms (such as Mercy Ships) showed promise, both in terms of outcomes and cost-effectiveness. Importantly, the specialty hospitals are the only platform that provides role training, which makes them attractive models for both the provision of surgical care and the training of future physicians. 40 c) Non-surgical providers delivering surgical care. Surgical task shifting has been advocated as an effective means of relieving the shortage of trained clinicians. In contrast to task sharing, which involves joint cooperation between the physician and a non-physician provider, task-shifting is defined as the delegation of certain medical responsibilities to less specialized healthcare workers. 23,53 An estimated half of all sub-saharan African countries use some form 18

19 of non-physician clinicians (NPCs) to perform surgical procedures. 23,53 NPCs are regarded as highly cost-effective and, because their skills cannot transfer across borders, they are more likely to remain in the district hospitals where basic surgical care is most needed. Since these hospitals are meant to be the first line of defense, and are located closer to the patients than are large referral centers, they may represent the only access to Western medicine for a population that is frequently too poor to travel long distances. Studies have shown that introducing even basic surgical training at the district hospital level significantly decreases referrals to large hospitals. 55 In an opinion essay, Médecins sans Frontiers published a list of procedures that could be safely performed by a non-surgical provider in resource-poor countries, including incision and drainage, burn care, skin grafting, bowel resection, cholecystectomies and splenectomies, among others. The authors called for more international research and policy development to guide the employment of NPCs in surgical care. 53 Although the use of NPCs for surgical procedures has met with some resistance and ethical questions, studies of outcomes reveal no difference between procedures performed by NPCs versus those done by surgeons. 23 It is important to note, however, that the majority of those studies report on obstetric and gynecologic procedures, and not those routinely considered in the domain of general surgeons. Several other models for surgical training exist, notably in Niger and Ghana. In Niger, the Ministry of Health launched a surgical training initiative for district hospitals in As part of the training, medical officers who were practicing generalists attended an additional 12-month program of instruction in basic surgical diseases, with 3 months of theoretical lectures and 9 months of hands-on practical instruction. Areas covered included basic general surgical procedures, such as herniorrhaphies and splenectomies; orthopedics (treatment of open and closed fractures, and amputations); obstetrics and gynecology (e.g., C sections); and urologic 19

20 emergencies (e.g., testicular torsion). The program showed mortality rates comparable to those of the regional hospital, where cases were performed by trained general surgeons and obstetricians. 55 In one study of the surgical resources of Ghanaian district hospitals, the authors found that the 14 medical officers who staffed the 10 district hospitals had graduated from medical school and had two years of additional housemanship. Together, they had performed 6629 surgeries in 1 year, or approximately two operations addressing abortion complications and one herniorrhaphy per week, and one exploratory laparotomy and one appendectomy per two months. Only 1 medical officer had had any dedicated surgical training (1 year of surgical residency); seven others had variable levels of supervision, and six received no further training beyond housemanship. 56 d) Development of future physicians and training of sub-saharan African surgeons. The training of future physicians represents both the most cost-intensive and the most promising and vital solution to solving the burden of surgical disease and the issue of brain drain in sub- Saharan Africa. To determine and evaluate the models of training general surgeons in sub- Saharan Africa, and to determine the factors affecting the supply and distribution of general surgeons in the area, I searched PUBMED, Medline, Global Health Database, and Google Scholar, using the MESH term surgery residency sub-saharan Africa. Of the 99 results from PUBMED and Medline, 63 were removed because they described global surgical electives in low income countries; were duplicates; or because they concerned obstetrics and gynecology, orthopedics, emergency care, urology or neurosurgery. This left 36 abstracts, which were reviewed. Global Health Database returned 2 hits, which were already included in the PUBMED results; both were kept. Google Scholar search with surgery residency and sub-saharan Africa linked with and, returned 104 results. Of those, 32 were relevant, and 16 were 20

21 duplicates, leaving 16 additional publications. A search of references yielded an additional 15 articles, which were reviewed and 15 were kept. Personnel training can be the most difficult and resource-intensive hurdle to improving delivery of care. 57 Many methods of training general surgeons and/or supplementing the curriculum were reported in literature; they ranged from workshops for existing personnel to the development of new training platforms, the provision of curricula, and the formation of collaborative efforts between developed and developing countries. Basic skills and focused educational initiatives aim to improve the professional stagnation felt by healthcare workers who consider emigration, as well as to address the gaps in education inherent in some medical education models in the area. Ezeome et al reported a basic skills workshop in Nigeria that assisted in the teaching of basic surgical skills using simulation. 58 Bergman et al evaluated a trauma team-training program, developed by the Canadian Network for International Surgery, and implemented in Tanzania. It involved a 3-day course with lectures, skills (BLS, intubation, thoracostomy, surgical airway). The authors found that the majority of providers involved in the course reported improved comfort with the trauma management process. Their post-test scores also improved significantly. 59 Dreyer et al reported the development of a surgical emergency course for residents in East/Central Africa to improve management of these issues. The report concerned resident satisfaction with the training, but did not evaluate outcomes. 60 Finally, the World Health Organization holds training workshops based on the Integrated Management of Emergency and Essential Surgical Care toolkit and the Surgical Care at the District Hospital manual

22 Partnerships between developed and developing nations take many forms to address the shortage of qualified physicians and the burden of surgical disease. In their least integrated form, they involve the transfer of patients with complex disease for surgical care in the developed country. 61 Some involve foreign surgeons in the provision of surgical services while relying on local specialists for screening of eligible individuals and follow-up. Others conduct needs assessments, participate in curriculum development, provide instruction, or host exchange programs that help to develop new surgeons. For example, Addis Ababa, Ethiopia, partners with a hospital in Toronto, Canada. 62 The first report from their collaboration involved conducting a needs assessment of educational and work realities for surgeons and residents, as well as a review of operative procedures at the training hospital to evaluate gaps in training and needs. Both faculty and residents believed that improving resident education would improve surgical care in Ethiopia, specifically in the areas of managing emergent surgical situations, education in subspecialty areas, and disease-specific knowledge. Faculty desired assistance from high-income country partnerships in supervision in the operating room, topic-specific lectures, and the supervision of residents in clinic assessments. A similar assessment was conducted in a Rwandan hospital, where Rickard et al reported on the surgical curriculum and the unmet training needs by a review of patients operated on by surgical residents in 2013 at the University Teaching Hospital in Kigali. 63 It found 2780 procedures in which a general surgery resident was the primary operator or assistant. A Rwandan partnership with McGill University Health Centre in Canada involved the development of a curriculum in which an invited surgeon from Canada participated as a consultant, but local faculty mainly provided the training and implementation. The idea behind this model was to increase local capacity for postgraduate education, as well as to reduce dependence on foreign 22

23 programs for the training of Rwandan surgeons. 43 Rwanda was also the site of Team Heart, a partnership between its ministry of health, the King Faisal Hospital in Kigali, and Partners in Health, with expatriate surgeons traveling to the area to provide surgical care and supportive services for rheumatic heart disease, while integrating local clinicians in screening and follow-up of selected patients. 21 Involvement from surgical trainees and faculty trained in the developed world serves to improve education and address some gaps in operative and clinical experience. Such a program is in place at a Tanzanian teaching hospital in Dar es Salaam, where the Pacific Coast Surgical Association collaborates with the Muhimbili National Hospital by sending American surgeons and general surgery residents (PGY3 or above) to the hospital for 1-2 months at a time. The Tanzanian surgeons and residents reported support for the program. In surveys, they also indicated a desire of further instruction in evidence-based practice, formal didactics, supervision in the OR, longer presence in the teaching hospital, and reciprocal rotations. 64 Some organizations engage in collaborative efforts with more than one recipient country. For example, the Pan-African Academy of Christian Surgeons (PAACS), formed in 1997, by 2011 had established six training programs in 4 countries, modeled on the five-year general surgery residency American model. During the academic year, the programs collectively trained 35 residents, with 18 general surgery and 1 pediatric surgery graduate. 65 Another multi-country collaboration is modeled by the Association of Surgeons of Great Britain and Ireland (ASGBI), which has been providing courses to surgeons in sub-saharan Africa since the early 2000s, and has held courses in 14 countries. An example of the services it provides is the development and implementation of a country-specific training course on surgical emergencies and critical care in Zambia

24 Workshops and other adjunct instruction address the gaps in clinical exposure inherent in practicing in some of the sub-saharan hospitals. In Botswana, a general surgery program was initiated in 2009, and a curriculum was developed to take into account local conditions. The authors reviewed surgical logs to determine the repertoire of cases in their hospital, contrasting it with the SCORE curriculum. They found relatively low levels of certain categories, such as basic and advanced laparoscopy, thoracic surgery, and liver/pancreatic cases. 66 The authors reported on the lack of these procedures but believed that they reflected local conditions and thus may not represent problematic gaps. Access to online curricula appears to be a promising adjunct to the training of surgeons in LMICs, although this model has not been objectively evaluated. One study described the self-reported levels of satisfaction with two online curricula: one developed by the Royal College of Surgeons in Ireland specifically for COSECSA countries, and the other the standard SCORE curriculum used by American resident surgeons. Both curricula were well received and appeared to be acceptable adjuncts in sub-saharan surgeon training. 57 PART III. MALAWI a) Background: Malawi, a small, densely-populated, landlocked country in Southern sub- Saharan Africa, is one of the poorest countries in the world. In 2014 it had a population of million people, and GDP of $4.258 billion dollars, with a gross national income per person of $250. In contrast to the 2010 low-income country average of 52.4% of the population living on less than $1.90 per day, the corresponding Malawian proportion was 70.1%, with 66.7% living at less than $1.25/day Out of 186 countries, Malawi ranked 170 th in the Human Development Index. 3 Its economy is based on agriculture, with tobacco providing the main export, and 80% of the population is employed in agricultural activities

25 The country is divided into 28 districts, which are further subdivided into authorities, made up of villages. Authorities and villages honor the country s tribal underpinnings, and each is ruled by a chief (authority) or a headman (villages). As of 2013, the majority of Malawians lived in villages (84%), although the precision of this data may be questionable, given that the percentage hasn t changed between 2006 and The country s major cities are: Lilongwe (capital), located in the center of the country, which almost doubled in population from 2008 to 2015 (from 669,021 to 1,077,116); 70 Blantyre in the South at 661,444 and Mzuzu and Zomba in the North, at 175,345 and 88,314, respectively. 5 Poverty in Lilongwe is approximately 25% b) Health indicators, health funding and expenditures: The 2014 Malawian life expectancy was 62.7 years, 63.7 for females and 61.7 for males; all those were slightly above the LMIC and sub-saharan averages. 67 With a female fertility rate of 5.1 births per woman, in 2014 Malawi s population grew at an annual rate of 3.1%, with almost half (45.4%) of its population aged 0-14 years. Infant mortality rate was 66.9/1000 live births, and the prevalence of HIV in adults aged was 10%. 2 In 2012, 7.1% of deaths in Malawi were caused by injury and 27.8% by non-communicable diseases. 67 The per capita yearly expenses on health were $24.4, much lower than the LMIC average of $36.8, and sub-saharan African average of $96.5. This low number is most likely related to the fact that Malawi relied much more heavily on public support for its healthcare financing; in 2014, public sources funded 62.7% of healthcare expenditures, when in other LMICs the average was 39.1%, and sub-saharan Africa 42.2%. Relatedly, Malawi spent 9.6% of its GDP on healthcare, when other LMICs spent 5.7%, and sub- Saharan countries spent 5.4%. c) Malawi burden of trauma and surgical disease. According to the WHO s Data Repository country-specific data, in 2013 Malawi had the third highest death rate from traffic 25

26 accidents in the world, at 35 per 100,000 population, behind only Libya and Thailand. 73 In comparison, this rate stands at 10.6/100,000 in the US, and in most of European countries at below 10 per 100,000 population. The average LMIC death rate from road traffic accidents (RTI) is 27.4, and in sub-saharan Africa it stands at 26.6 per 100,000 population. 74 In a 2008 study of a tertiary care center in Malawi, 1474 patient encounters for trauma occurred in a 5-month period, with an overall admission rate of 26.8%; a predominance of males (75.5%), and the majority of patients being under % of injuries were due to RTIs. 5 A 2012 retrospective chart review of four rural hospitals in Malawi found that 3.5% of admissions were related to trauma using WHO s ICD classification, and that animal bites (34.7%) and road injuries (32.8%) were the most common causes of trauma presentations to the hospital. 75 In , children represented 30.6% of all trauma patients, with the majority of injuries occurring at home (65.6%) and on the road (15%). In fact, in children older than 5, trauma was responsible for more deaths than malaria, HIV and tuberculosis combined. 12 d) Healthcare system and delivery: Healthcare in Malawi is delivered predominantly through government-run facilities (63%), followed by those run by the Christian Health Association of Malawi (26%), and a low percentage of private for-profits. 68 Public sector facilities are under the auspices of the Ministries of Health, Local Government and Rural Development, and Forestry; and the police, prisons and the army. Public health is free of charge. 69 Delivery takes place in a three-tier system, with primary care addressed in dispensaries, community initiatives, health posts, community and rural hospitals. District hospitals make up the second layer, and are meant to provide specialized services to patients referred from the primary health care level, through outpatient and inpatient services and community health services. They are located in the rural health districts. Additionally, there are four central 26

27 hospitals in the four major urban centers, which together with other private specialist hospitals represent the third level of care, serving specialized disease conditions or specialized groups of patients. 52,69 The main cities of Malawi, Blantyre, Lilongwe, and Mzuzu have regional or referral hospitals, larger complexes designed in theory to serve as tertiary referral centers able to deal with complex medical and surgical diagnoses. Kamuzu Central Hospital in Lilongwe is variously reported as having anywhere between 750 and 1200 beds, with real numbers likely above the upper limit, given multiple occupancy for most of the beds. Its catchment area is the central region of Malawi, with a population of about 3 million people, and a referral net of 13 hospitals. In spite of its designation as a tertiary center, KCH offers a mix of primary, secondary and tertiary services because Lilongwe has no district hospital. 77 The University of North Carolina at Chapel Hill, Baylor University in Texas, and Norway s Haukeland University all have a presence in the hospital, participating in the delivery of care to pediatric HIV patients, burn survivors, surgical patients, obstetric and gynecologic patients, and others. Poorly developed infrastructure hampers healthcare delivery in Malawi the same way it does in the rest of sub-saharan Africa. There is no central emergency response system and no formal ambulance service. Local private clinics run private ambulance services in the major cities, with private vehicles remaining the main means of transport to a hospital for the majority of the population. In cities most roads are paved, but in the villages dirt roads predominate, and the majority of the population commutes by foot or bicycle. Patients needing a higher level of care or living farther from a district hospital thus face major delays in receiving care. 27

28 The availability of diagnostic and treatment modalities varies widely among the hospitals. At KCH basic lab work and radiography are available, along with anesthesiology providers trained to staff the majority of surgical procedures. Antibiotics are usually available, though their limited classes and poor stewardship result in a high proportion of drug-resistant organisms. Post-surgical recovery services, such as physical and occupational therapy, are very limited. Currently, ten physical therapists and four interns service the entire patient population of KCH. For burn patients, for example, this dearth of rehabilitative services further compounds the lack of surgical care and dressing supplies, and can mean the difference between lifelong disability from burn contracture and a return to a productive life. e) Access to surgical care and surgical training model in Malawi: The majority of surgical care in Malawi is provided by non-physicians, the Clinical Officers (COs). CO training began in 1979, initially conceived as a measure of temporarily relieving the country s shortage of physicians, but has since been solidified as a viable option for clinical training. Clinical officers currently receive 3+1 years of medical training, with the first three spent in didactics and the last year a surgically-focused internship at one of the district hosptials. 78 District hospitals normally work under the leadership of a district health officer, who receives internship training and thereafter is responsible for the management of the district hospital, with little time for clinical duties. Formal general surgery training, on the other hand, follows the US model and requires completion of medical school plus a dedicated residency. Formally trained general surgeons in Malawi work at the central hospitals or in private practice. To our knowledge there are no formally trained surgeons on the district hospital level in the entire country. A 2007 report studied Malawian hospital logbooks in 2003, collected from all of the 21 district and 4 central hospitals. The authors found that the former performed 25,053 surgical 28