University of Ghana http://ugspace.ug.edu.gh SCHOOL OF PUBLIC HEALTH COLLEGE OF HEALTH SCIENCES UNIVERSITY OF GHANA DEPARTMENT OF HEALTH POLICY PLANNING AND MANAGEMENT HOUSEHOLD COSTS OF INJURIES: A CASE STUDY OF THE CASUALTY AND ACCIDENT CENTRE OF KORLE-BU TEACHING HOSPITAL BY PAA-KWESI BLANKSON (10198895) THIS DISSERTATION IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MASTER OF PUBLIC HEALTH DEGREE JULY, 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that excluding precise references which have been duly acknowledged, this submission is my own work towards my MPH dissertation and that, to the best of my knowledge, it contains no material previously published by another person nor material which has been accepted for the award of any other degree of the University or elsewhere. PAA-KWESI BLANKSON (STUDENT) ……………………………………………. SIGNATURE ……………………………………………. DATE CERTIFIED BY: PROF MOSES AIKINS (ACADEMIC SUPERVISOR) ……………………………………………. SIGNATURE …………………………………………… DATE i University of Ghana http://ugspace.ug.edu.gh DEDICATION To Naa, Mom, Dad and the entire family. For your vision, genius, dedication, friendship and love, I am daily grateful. For Mama Maggie, who thankfully survived the upsetting event of a road accident but not without hefty injuries. This one too shall pass, and speedily. ii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT I am indebted to God for the success of this project, without whose providence I would not have come this far. To Prof. Moses Aikins and Dr. Justice Novignon who have been quintessential in support and teaching, guiding me through this work, every step of the way, I am truly grateful. My deep gratitude also goes to my heaven-sent research assistants, as well as the staff and administration of the Korle-Bu Teaching Hospital who embraced this project with encouraging attention and kindness. My parents as usual took an unfathomably interest in this entire course while Naa Afia deftly read the chapters over my shoulder. I am very grateful. Also, it seems so that the pursuance of MPH degree will come with not just relevant knowledge and a certificate, but perpetual friends and family in class and staff. Thank you all very much. iii University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS Content Page DECLARATION ................................................................................................................... i DEDICATION ...................................................................................................................... ii ACKNOWLEDGEMENT .................................................................................................. iii TABLE OF CONTENTS ..................................................................................................... iv LIST OF TABLES .............................................................................................................. vii LIST OF FIGURES .......................................................................................................... viii LIST OF ABBREVIATIONS .............................................................................................. ix ABSTRACT .......................................................................................................................... x CHAPTER ONE ................................................................................................................... 1 1.0 INTRODUCTION .......................................................................................................... 1 1.1 Background ................................................................................................................. 1 1.2 Problem statement ....................................................................................................... 3 1.3 Justification ................................................................................................................. 4 1.4 Conceptual framework of injuries ............................................................................... 4 1.5 Objectives .................................................................................................................... 5 1.6 Research Questions ..................................................................................................... 6 CHAPTER TWO .................................................................................................................. 7 2.0 LITERATURE REVIEW ............................................................................................... 7 2.1 Definition and scope of injuries .................................................................................. 7 2.2 Incorporated metric for injuries ................................................................................... 7 2.3 Types and pattern of injuries ....................................................................................... 8 2.4 Cost analysis in Health ................................................................................................ 9 2.4.1 Evolution of Cost-of-Illness studies ..................................................................... 9 2.4.2 Constituents of Economic Costs of Disease ......................................................... 9 2.4.3 Approaches to Cost-of-Illness studies ................................................................ 11 2.4.4 Cost-of-Illness studies......................................................................................... 12 2.4.5 Discounting and sensitivity analysis ................................................................... 13 2.5 Applications and benefits of Cost-of-Illness studies ................................................. 14 2.6 Costs of injuries ......................................................................................................... 14 iv University of Ghana http://ugspace.ug.edu.gh 2.7 Cost-of-Illness studies of other diseases.................................................................... 16 2.8 Conclusion ................................................................................................................. 17 CHAPTER THREE ............................................................................................................. 19 3.0 METHOD ...................................................................................................................... 19 3.1 Study Design ............................................................................................................. 19 3.2 Study area .................................................................................................................. 20 3.3 Study population ........................................................................................................ 20 3.4 Sample size Calculation ............................................................................................ 21 3.5 Sampling Method ...................................................................................................... 22 3.6 Study Variables ......................................................................................................... 23 3.7 Data Collection Techniques and Tools...................................................................... 23 3.8 Quality Control .......................................................................................................... 24 3.9 Data Analysis ............................................................................................................ 25 3.9.1 Background characteristics of respondents......................................................... 25 3.9.2 Estimation of direct costs .................................................................................... 25 3.9.3 Estimation of indirect costs ................................................................................ 26 3.9.4 Total household cost estimation ......................................................................... 27 3.9.5 Estimation of annual total household cost of injuries by aetiology .................... 28 3.9.6 Determination of intangible costs ....................................................................... 28 3.9.7 Sensitivity analysis of cost of Injuries ................................................................ 29 3.10 Study Limitations .................................................................................................... 29 3.11 Ethical Considerations ............................................................................................. 29 CHAPTER FOUR ............................................................................................................... 31 4.0 RESULTS ..................................................................................................................... 31 4.1 Background characteristics of respondents ............................................................... 31 4.2 Total household cost of injuries ................................................................................ 33 4.3 Estimated annual household cost of injuries by aetiology ........................................ 36 4.4 Sensitivity analysis of cost of Injuries ....................................................................... 38 v University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE ................................................................................................................. 40 5.0 DISCUSSION ............................................................................................................... 40 5.1 Background Characteristics ....................................................................................... 40 5.2 Direct cost .................................................................................................................. 42 5.3 Indirect cost ............................................................................................................... 44 5.4 Intangible cost ........................................................................................................... 46 CHAPTER SIX ................................................................................................................... 47 6.0 CONCLUSION AND RECOMMENDATIONS .......................................................... 47 6.1 Conclusion ................................................................................................................. 47 6.2 Recommendations ..................................................................................................... 47 REFERENCES .................................................................................................................... 48 APPENDICES .................................................................................................................... 56 vi University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1: Estimation of household indirect cost of Injuries ................................................. 27 Table 2: Composite intangible score ranges ....................................................................... 28 Table 3: Background characteristics of respondents ........................................................... 32 Table 4: Total household cost of injuries ............................................................................ 35 Table 5: Total annual household cost of injuries for the year ............................................. 37 Table 6: Sensitivity analysis of total cost ............................................................................ 39 vii University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 1: Conceptual framework of household costs of injuries .......................................... 5 Figure 2: Schematic diagram of the study design ............................................................... 19 Figure 3: Proportions of direct and indirect cost of injuries ............................................... 33 Figure 4: Intangible cost of injured patients ....................................................................... 36 viii University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS COI -Cost of Illness CVM -Contingent Valuation Method DALY -Disability-Adjusted Life Year DCPP - Disease Control Priorities Project FCM - Friction Cost Method HCM - Human Capital Method ICECI -International Classification of External Causes of Injuries ICD - International Classification of Diseases KBTH -Korle-Bu Teaching Hospital NHIS -National Health Insurance Scheme RTA -Road Traffic Accident SA -Sensitivity Analysis SDG -Sustainable Development Goals UG -University of Ghana UN -United Nations WHO -World Health Organisation WTP - Willingness To Pay YLL -Years of Life Lost YLD -Years Lost to Disability ix University of Ghana http://ugspace.ug.edu.gh ABSTRACT BACKGROUND: Accidents, and casualties in general remain a major cause of morbidity and mortality worldwide. There seems to be changing trends in causes of death over the past Century with gradual increasing importance of injuries in particular. It has been suggested that injury is one of the leading causes of death in many developing countries, accounting for more deaths than HIV, tuberculosis, and malaria combined. While the World Health Organization (WHO), lists road injury as the tenth commonest cause of death after ischaemic heart disease, stroke, lower respiratory infections, chronic obstructive pulmonary disease, respiratory system cancers, diabetes mellitus, Alzheimer’s disease, diarrhoeal diseases and tuberculosis worldwide, the use of cost estimates for evidence-based advocacy on injury prevention policies and strategies is crucial. OBJECTIVE: To determine the economic burden to patients with injuries at the Casualty and Accident centre of Korle-Bu Teaching Hospital METHODS: A cross-sectional study of patients at the Casualty and Accident centre of the Korle-Bu teaching hospital was undertaken. Using a sample size of 264, the direct cost was estimated. Indirect cost estimates were determined by use of the Human Capita approach to determine the burden of injuries. Intangible cost was assessed with a descriptive approach by use of a Likert scale. The overall cost was then reported, estimated by the summation of the total direct cost and total indirect cost. RESULTS: This study found the total direct and indirect costs to the respondents to be GHS 1,384,548.83 (US$ 318,287) and GHS 306,728.03 (US$ 70,512.19) respectively, representing 82% and 18% of the total cost of treatment. The estimated annual cost to all injured patients reporting to the unit was GHS 48,729,294.01 (US$ 11,202,136.55), with an average total cost of GHS 5,551.27 (US$ 1,276.15). The study also found that the highest x University of Ghana http://ugspace.ug.edu.gh proportion of injured patients experienced a moderate levels of intangible cost while a 32% experienced a high level of intangible cost. CONCLUSION: In addition to ongoing efforts in prevention, this study provides further evidence on the social limitations in protecting patients and their family from the high costs of severe injuries. xi University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE 1.0 INTRODUCTION 1.1 Background Injuries have over time become a critical consideration in any nation’s health system. Injuries caused by violence, road traffic accidents, falls, drowning, burns, among others, kill more than 5 million people every year, accounting for 9% of the world’s deaths (WHO, 2016a). With its increasing incidence, the United Nations, recognizes injuries in its health program for development, including it in its health targets for the 2030 Agenda for Sustainable Development Goals (SDG) (United Nations, 2014). Worldwide, accidents and casualties, in general, remain one of the leading causes of morbidity and mortality. There seems to be changing trends in causes of death over the past century with the gradually increasing importance of accidents in particular (House, Nyabera, Yusi, & Rusyniak, 2014). This may largely be attributed to rising numbers and sophistication of high-velocity vehicles, accounting for 24% of global injury-related deaths. Another 16% is accounted for by suicides; 14% from falls; 10% from homicide; and 7% from drowning and 2% from war and conflict (WHO, 2016b). The social and economic impact of injuries is also tremendous requiring much attention by countries of all economic levels. Aside from direct deaths resulting from accidents and trauma, tens of millions of people suffer directly from the injuries as well as the cost of hospitalization and treatment. Other areas of direct cost to the individual and family may include transportation, first aid, property damage and funeral expenses. There is also an inevitable area of indirect cost comprising the loss of productive working time for the patient and relatives. The intangible costs are the cost of pain, grief, and suffering endured by the victims and family members as a result of the injury. Indeed many are left with temporary 1 University of Ghana http://ugspace.ug.edu.gh or permanent disabilities, accounting for an estimated 6% of all years lived with disability (DALY) according to WHO (2016). Likewise, for the sub-Saharan Africa region, figures regarding cost as well as the pattern of injuries vary from place to place, depending on what is pertinent to the location. Such injury aetiology may include road traffic accidents (RTAs), falls, assaults, firearm injuries, burns, sports injuries, animal bites, drowning, domestic violence, industrial accidents, terrorism and civil conflicts. The United Nations estimated annual costs of road traffic crashes in low income and middle-income countries to be between US$ 65 billion and US$ 100 billion, more than the total annual amount received in development aid (United Nations, 2007). Generally, economic analysis or evaluation of injuries give insight into and informs costs resulting from targeted interventions. In many countries, it plays a role in decision-making on reimbursement as well (Polinder, Toet, Panneman, & Van Beeck Iii, 2011). In Ghana, 72 persons out of every 100 000 population, suffered from grievous bodily injury and close to 8 of the same population died from RTAs over the past decade while more than 60% of road traffic fatalities occur in children and young persons under 35 years of age (Hesse & Ofosu, 2014). Information relating to costs of various injuries, however, is scarce for Ghana. In Ghana, 72 persons out of every 100,000 population, suffered from grievous bodily injury and close to 8 persons of the same population died from Road Traffic Accidents (RTA) over the past decade. In this same period, more than 60% in of road traffic fatalities occurred in children and young persons under 35 years of age (Hesse & Ofosu, 2014). A community-based study of incidence and injury outcome in Kumasi, an urban city and a rural town in the Brong-Ahafo region of Ghana reported an 83 per 100,000 death resulting from injuries. The study further suggested the overall rate of injury-related disability was 2 University of Ghana http://ugspace.ug.edu.gh higher in the rural areas as compared to the urban areas. This difference is primarily accounted for by high levels of agriculture related injuries in the rural area (Mock, Abantanga, Cummings, & Koepsell, 1999). Kudebong et al. (2011) also found motor cycle accident costs in the Bolgatanga Municipality to be US D 411, 204.40 in social cost annually. The researchers found 64% of this figure due to the direct cost to be borne mostly by the victims and relations, and quoted road traffic crashes to cost Ghana 1.6% of GDP which translates to US$ 165 million. In addition to the scarcity of information with regards to economic evaluation of injuries, there seems to be insufficient data on costs of other non-motor injuries, as well as a hospital based description of injury pattern which this study seeks to address. 1.2 Problem statement There is a tremendous economic impact of injuries on households as WHO lists road injury as the tenth commonest cause of death after ischaemic heart disease, stroke, lower respiratory infections, chronic obstructive pulmonary disease, respiratory system cancers, diabetes mellitus, Alzheimer’s disease, diarrhoeal diseases and tuberculosis worldwide. WHO also predicted a 40% increase in global deaths owing to injuries in general between 2002 and 2030 (WHO, 2016a). It has also been suggested that injury is one of the leading causes of death in many developing countries, accounting for more deaths than HIV, tuberculosis, and malaria combined (WHO, 2008). Approximately half of injury-related deaths occur in individuals aged 15 - 44 years. Since this is during their most economically productive years (Nilsen and Hudson, 2006), it makes the burden of injuries exceed by far, the perceived immediate medical costs associated with the injury (Hadley, Wesson, Boikhutso, Bachani, Hofman, & Hyder, 2014). This impact is much more for a developing 3 University of Ghana http://ugspace.ug.edu.gh country like Ghana and others in the sub-region where individuals could have a relatively higher number of dependents. WHO and Disease Control Priorities Project (DCPP) have encouraged the use of Cost of Illness (COI) studies to assess and implement injury prevention policies and strategies (WHO-CHOICE, 2003). However, there seems to be a conspicuous lack of recent studies regarding the cost of injuries to households in Ghana. This study, therefore, harnesses this method to bring more focus to the economic burden of injuries, while providing more recent and tangible basis for planning for injuries in the country. 1.3 Justification Given limited available data regarding cost of injuries in Ghana, this study will come to fill in the knowledge gap concerning cost of injuries at a typical tertiary hospital in the country. Several authors have emphasized the need for attention to be drawn to injuries as a major pertinent health problem, showing the relative cost of different injury types (Mock, Gloyd, Adjei, Acheampong, & Gish, 2003). This study, aside from filling in this gap, will also give a current perspective and inform the economic framework for program evaluation regarding injuries. It would, also, be a necessary tool for policy and institutional assessment. Likewise, cost analyses will highlight the relative magnitude of burden injuries have on the society thus informing the need for regulatory measures regarding road injuries, sports injuries, falls among others. 1.4 Conceptual framework of injuries As illustrated in Figure 1, the conceptual framework generally describes the components of household cost. The costs associated with the management of injury are divided into three major components; direct, indirect and intangible costs. With the occurrence of an injury, 4 University of Ghana http://ugspace.ug.edu.gh the individual and family incur household expenditure in the form of transportation, first aid, communication and cost of intervention. These constitute the direct cost. Indirect costs refer to productivity loss incurred by an illness and are significant in cost of illness studies as they can be substantial, sometimes even being more than the Direct cost. Intangible costs, on the other hand, are defined as pain and sufferings of patients because of a disease, which is usually determined by the reduction in quality of life. These in its totality make up the household cost of injuries, also referred to as the social cost. Di rect cost (i.e. medical & non- medical cost) Indirect cost Household INJURIES (i.e. lost productivity total cost of & income, assets lost Injuries Intangible cost i.e. pain, anxiety, stress, isolation etc. Figure 1: Conceptual framework of household costs of injuries 1.5 Objectives General objective: To determine the cost borne by household for injuries of patients attending the Casualty and Accident Centre of Korle-Bu Teaching hospital Specific objectives: 1. To determine the direct cost of injuries to households of patients 2. To determine the indirect cost of injuries to household of patients 3. To determine intangible costs associated with injuries to household of patients 5 University of Ghana http://ugspace.ug.edu.gh 1.6 Research Questions 1. What is the direct cost of injuries to households of patients? 2. What is the indirect cost of injuries to households of patients? 3. What are the intangible costs associated with injuries to households of patients? 6 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Definition and scope of injuries The role of Injuries in morbidity and mortality has become a major concern as WHO reports that injuries are the principal cause of death among people 5–44 years old (WHO, 2016a). The definition of physical injury according to WHO, refers to damage caused by the acute transfer of energy, be it physical, chemical, radiant or thermal that surpasses the physiological threshold (WHO, Regional Office for Europe 2014). The World Health Organization proposed the International Classification of External Causes of Injuries (ICECI) to help researchers and practitioners to describe, monitor and evaluate the occurrence of injuries, a system which augments the International Classification of Diseases (ICD 10) in describing injuries. This takes into consideration, among other things, the mechanism of injury, place of occurrence, objects/substances producing injury, activity when injured, the role of human intent and the use of alcohol, and other psycho-active drugs (WHO, 2010). Intentional injuries refer to those that are deliberately caused by violence, use of threat or physical force against another person, community or oneself, that results in injury, death, mental harm, maldevelopment or deprivation. This includes suicide or self- harm, interpersonal violence, child maltreatment and wars. Unintentional injuries, on the other hand, refers to the category caused by road traffic injuries, poisonings, falls, fire, heat and hot substances, or drownings or submersion (WHO Regional Office for Europe 2014). 2.2 Incorporated metric for injuries The first Global burden of disease study in 1993, reinforced by the most recent (WHO, 2008) introduced an internationally consistent measurement for the burden of disease and injury to a country. Disability-adjusted-life-years (DALYs) may be considered as a measure 7 University of Ghana http://ugspace.ug.edu.gh of the gap between current health status and represents an absolute ideal healthy state where the entire population lives to an advanced age, free of disease and disability (WHO, 2014). The DALYs for injury or any disease incorporates mortality (years of life lost-YLL) and disability (years lost due to disability-YLD) for the incident cases of the disease or injury (WHO, 2008). According to Haagsma et al. (2016), injury DALY rates declined by 30.9% between 1990 and 2013. However, the decline was slow in the sub-Saharan Africa region. The study also observed that the burden of disease due to road injury had decreased significantly since 1990 in high-income areas, with the reverse trend occurring in low-income and middle-income countries. The study also indicated that falls are the dominant cause of injury DALY rates among the elderly. Murray et al. (2012) also suggested a 34% increase in DALYs from road injuries, moving it from 12th to 10th in a ranking of causes of DALYs. 2.3 Types and pattern of injuries While it seems quite evident that mortality rates resulting from injuries are 2.4 times higher in low- and middle-income as compared to high-income countries (WHO Regional Office for Europe 2014), distribution of its presentation seems to be closely linked to the political and social dynamism of the particular location. In a descriptive study of an emergency centre in Western Kenya, the authors reported a mean age of 35.6 years and 52.6% of patients being female (House et al., 2014). In their multifunctional emergency centre, they reported 20.2% (highest number) of the patients reporting due to injury. The gender distribution is however in contradiction to many assertions that mortality rates due to injuries are higher (2.7 times) in males than in females (WHO Regional Office for Europe 2014). This has been accounted for by several theories, popular among them being the fact that men are considered to be risk-takers (Harris, Jenkins, & Glaser, 2006). 8 University of Ghana http://ugspace.ug.edu.gh Another descriptive study of injuries reporting to the national hospital in Kampala, Uganda suggested that except for those under 5, road traffic injuries were the commonest causes of injury for all age groups, accounting for 49% of all injuries. Blunt injuries in the same study accounted for 15% and constituted the second largest cause of injury in patients 15 years and above. Penetrating injury was seen more often in patients between the ages of 15 and 64. Poisonings, choking or hanging, drowning, and sexual assault were rarely reported (all less than 1%). Among traffic injuries, more passengers (44%) and pedestrians (30%) were injured than drivers (27%) (Hsia et al., 2010a). 2.4 Cost analysis in Health 2.4.1 Evolution of Cost-of-Illness studies Based on the supposition that the economic costs of illness represent the economic benefits of a medical intervention should the illness be eradicated, Cost of Illness (COI) analyses has been suggested to be the earliest form of economic evaluation in the health care sector (Tarricone, 2006). The COI referred to by some authors as Burden of Disease (BOD) (Jo, 2014), is “to itemize, value, and sum the costs of a particular problem with the aim of giving an idea of its economic burden" (Jefferson et. al 2000). The first detailed description of COI analysis is attributed to the health economist Rice (1985), with his work reviewing previous COI study done in 1963 and 1972. It, however, seems to be evident that some COI analysis were in regular use well before the 1960s (Tarricone, 2006). 2.4.2 Constituents of Economic Costs of Disease While the cost of disease may be borne by either the household or the institution/state as a whole, cost on COI studies are broadly categorized into direct, indirect and intangible costs. 9 University of Ghana http://ugspace.ug.edu.gh Direct Cost: Direct costs refers to the use of resources of a particular value to directly influence the provision of treatment (Anders, Ommen, Pfaff, Lüngen, Lefering, Thüm 2013). This may be healthcare and non-healthcare costs. The former refers to the medical care expenditures for diagnosis, treatment, continuing care, rehabilitation, and terminal care. The non-healthcare direct cost, however, refers to the consumption of non-healthcare resources, such as transportation to and from health providers, certain household expenditures, costs of relocating and individual property losses, legal and court costs, and informal care from family or volunteers of the injured (Tarricone, 2006) Indirect Cost: Indirect costs refer to productivity loss incurred by an illness (Tarricone, 2006). The calculation of this category assumes and incorporates the incomes of the patient or the relative during the course of the disease corresponds to the marginal product of labor. This has been suggested to also include sickness benefits (Anders et al., 2013). Indirect cost could be very substantial as several studies have suggested that estimated indirect costs could be three times higher than direct costs, accounting for up to 80% of total costs in patients (Xie et al., 2016). Direct cost has conventionally been assumed to be higher than indirect cost. However, patterns of some disease costs suggest otherwise, as reported by (Joo, George, Fang, & Wang, 2014), (Tajima-Pozo, de Castro Oller, Lewczuk, & Montañes- Rada, 2015) and (Yang et al., 2011). Intangible Cost: Intangible costs traditionally referred to patients’ psychological pain and discomfort (Tarricone, 2006). Intangible costs are seldom quantified in COI studies due to measurement difficulties and related controversies (Jo, 2014). However Xie et al. (2016) employed a contingent valuation method (CVM), based on the willingness to pay (WTP) approach to quantify intangible cost. 10 University of Ghana http://ugspace.ug.edu.gh 2.4.3 Approaches to Cost-of-Illness studies The concept of COI has evolved, with several approaches to cost determination having been described. Tarricone (2006) conceptualized the approaches to COI studies into three models namely, prevalence versus incidence approach, top-down versus bottom-up and retrospective versus prospective studies. Prevalence versus incidence approach: Byford et al. (2000) recognizes these as the two fundamental approaches to COI studies, with the prevalence method, the commonest, estimating the total cost of a disease incurred in a given year. The incidence approach however involves estimating the lifetime costs of the new cases of a condition or group of conditions which have their onset in a specified period. The working principle for the prevalence approach is that disease costs be assigned to the years in which they are endured whereas the incidence approach associates the cost of illness, both direct and indirect, to the year in which the disease first appears (Tarricone, 2006). Top-down versus bottom-up: The top-down and bottom up methods of cost estimation are associated with the prevalence and incidence approaches respectively. The top-down approach involves allocating total national health care expenditures by use of pre-existing estimates, whereas the bottom-up approach firstly estimates the quantity of health inputs used followed by calculating the unit costs of the inputs. The costs are then obtained by multiplying unit costs by the quantities (Tarricone, 2006). Retrospective versus Prospective studies: This perspective looks at the temporal relationship between the initiation of the study and data collection. In the retrospective approach, all the relevant events, i.e. the disease or illness for cost assessment would have occurred before the commencement of the study. A prospective approach would, on the other hand, involve a follow up of the target individuals in time to ascertain detailed 11 University of Ghana http://ugspace.ug.edu.gh estimates of their costs incurred. While the retrospective approach is relatively much simpler and cost effective, the elements of participant recall bias and inadequate data record must be overcome (Tarricone, 2006). 2.4.4 Cost-of-Illness studies Two methods of direct costing have been described, micro-costing and gross-costing, which are quite identical in course to bottom-up and top-down approaches respectively. The more preferred and precise micro-costing involves summing up each single cost component that has contributed to the provision of the service, while gross-cost is obtained by dividing total costs of the service unit by the total number of services produced in a period (Tarricone, 2006). While direct cost estimation involves direct recording of actual monetary resources, indirect and intangible cost estimation may, on the other hand, be associated with more difficulty and uncertainty. The following have been described in standardizing the cost estimation process. Human Capital Method (HCM): This is the method adopted in most COI studies (Jo, 2014), having been used at least since the 17th century (Tarricone, 2006). This method tries to put monetary value to an individual on the premise that society would have benefitted from the productivity of the individual should he/she have continued to live in full health. It thus assumes calculates the human capital per their current socio-economic status. (Jo, 2014). Friction cost method (FCM): In this method, which is an alternative to HCM, there is a hypothetical ‘friction period’ which is essential in employing a new previously unemployed worker to replace the sick or injured individual until his return. The human capital is then calculated for this transition period which would incorporate the disruption of work as well 12 University of Ghana http://ugspace.ug.edu.gh as the training of the new employee. FCM assumes that productivity is not affected the ‘friction period’ has elapsed (Jo, 2014). Willingness to pay method (WTP): The willingness to pay (WTP) method estimates productivity loss by attempting to measure the amount that an individual is eager to pay to reduce or eliminate the morbidity associated with the illness. Several methods have been suggested to achieve these. However the commonest is the contingency valuation method (CVM). Though this method has generally been used in the estimation of indirect cost (Jo, 2014), Xie et al. (2016) used CVM in the estimation of intangible cost. In their study, patients who were managed with Osteoarthritis of the knees were provided with an initial bid, asked whether they would like to pay that specified amount as one-time payment for the absolute cure of the condition. This amount was further bargained until they expressed a WTP. 2.4.5 Discounting and sensitivity analysis Like many other costing studies, COI studies requires some form of robustness and safeguarding of values. Discounting and sensitivity are thus applied to give the values obtained, the heftiness required. Discounting has been described as an economic method that considers an individual's preference for income or payment today rather than that in the future. Opportunity cost is often used to account for this time preference. Also, one-way (1- way) and multi-way (M-way) sensitivity analysis (SA) have been described as methods to test for robustness of costing studies by varying relevant costs components (Jo, 2014). Polinder et al., (2011) in their description of approaches to costing studies in health, used varying figures of 3% and 5% in their sensitivity analysis. 13 University of Ghana http://ugspace.ug.edu.gh 2.5 Applications and benefits of Cost-of-Illness studies COI is over time becoming a standardized gauge for policy makers. Aside reimbursement decision making (Polinder et al., 2011), their estimates are invaluable in: (1) Defining the magnitude of disease or injury in monetary terms; (2) Justifying intervention programs; (3) Assisting in the allocation of research resources on specific conditions; (4) Providing a basis for policy and planning relative to prevention and control initiatives; and (5) Providing an economic framework for program evaluation (Rice, 2000) 2.6 Costs of injuries Outcomes of cost of injuries studies seem to vary from country to country since different social factors such as insurance and cost of living affect the depth of out-of-pocket expenditure. The average cost per patient for hospital treatment of severely injured patients in Germany, for instance, was found to be 31,478 (SD: 18,591) (Anders et al. 2013). On the other hand, with an estimated annual costs of between US$ 65 billion and US$ 100 billion (United Nations, 2007), the mean costs of road traffic crashes in low income and middle- income countries was found to range from US$14 to US$17 400 (Wesson et al., 2014). With the overwhelming world statistics for the incidence of injuries and injury related mortality, Ghana is not left out of its shadow as Hesse & Ofosu (2014) demonstrated that mortality rate for every 100 accidents between 1991 and 2011, rose from 11.0 to 20.2, representing an increase of 83.6%. In the same study, road traffic accidents accounted for a far higher mortality among men, by an approximate ratio of 3:1 when compared to females. In a community-based survey by Mock et al. (1999) where a comparative descriptive study to ascertain the incidence of household injuries, the investigators found prevalent injuries to consist of transport related injuries, burns, assault, lacerations, falls and others in decreasing frequency. This pattern was similar for both rural and urban communities. Comparing an 14 University of Ghana http://ugspace.ug.edu.gh urban and a rural area in the Kumasi municipality, Mock et al., (2003) found the mean out of pocket expenditure for treatment in the urban area to be US$ 31 ± 105 (S.D.) per injury. This was significantly lower than the US$ 11 ± 58 found for the rural area. Aside the staggering statistic of more than two thousand people dying annually from road injuries in Ghana, the highest fatalities during the period 1991 to 2011 was in the 26 – 35- year-old age group (Hesse & Ofosu, 2014). This indicates the vulnerability of the productive age group of the country. Implementation of policies thus ceases to be a luxury but a matter of necessity and urgency. The National Road Commission (2007), in consolidating this fact, attributed most road traffic accidents to drunk driving, deficiencies in law enforcement agencies, among others. Another community-based study conducted in the Ejisu-Juabeng municipality in the Ashanti region of Ghana to determine the household cost of injuries of children under five showed that the total direct cost of injury to be GHS 3,922.60 (US$ 1028.00), averaging GHS 6.67 (US$ 1.87) per injury for each household. The total indirect cost for this study was found to be GHS 4,808,938.00 (US$ 1,260,534.81). This averaged GHS 814.89 (US$ 213.60) incurred on injury per household (Amissah, 2016). In a study by Kudebong et al. (2011) who did a retrospective cross-sectional cost study on victims of motorcycle accidents, data revealed the social cost of motor cycle accidents to be GHS1,630,979.60 per annum. Direct cost accounted for 64% of the total cost while indirect cost accounted 36%. While this trend of lower comparative indirect cost with direct cost was in sharp contrast to many-a-studies (Hadley K H Wesson et al., 2014; Xie et al., 2016), the difference could be accounted for by the relatively lower socioeconomic status of the country. Kudebong et al. (2011) also found average out-of-pocket expenses to be US$ 42.72. 15 University of Ghana http://ugspace.ug.edu.gh 2.7 Cost-of-Illness studies of other diseases In a review study to determine the economic burden of illness for households in developing countries, the authors, focusing on malaria, tuberculosis (TB), and HIV found the mean household spending on TB to range from about D50 to more than US D100 over the treatment period with the exception of Malawi. This imposed a cost burden of 8–20% of annual income in impoverished settings (Russell, 2004). The same study established that health financing and delivery arrangements as well as the quality of care influenced patients’ interactions with public and private providers and therefore direct costs. Also, a mean fixed costs ranging from US$ 21.00 to US$ 88.00 was found in a COI due to Cholera in Zanzibar. This burden, the authors claimed to primarily consist of direct and indirect human resources costs (Schaetti et al., 2012). In a descriptive cross sectional study to estimate the COI of diabetes management in Ghana, Quaye et al. (2015) found the mean annual financial cost of managing one diabetic case at the clinics to be US$ 372.65. Service charge in the study constituted 22% of the cost, whereas direct medical cost constituted 78% of the cost. Measures of direct, as well as indirect burden of chronic diseases at the household level estimated from the last World Health Survey for Ghana, found mean household health expenditure was US$ 9.09, with a median of US$ 3.10 and maximum of US$ 442.25 (Tagoe, 2012). Asante and Asenso- Okyere, (2003) in a Cost-of-illness study for malaria in Ghana found that considerable amount of time was expended in seeking malaria care and taking care of the affected individuals, which made the indirect cost represent 79% of the total cost of seeking treatment in the survey areas. The average cost of treatment from their study amounted to US$ 8.67. In a more recent study, Nonvignon et al., (2016) showed that businesses in Ghana lost about US$6.58 million to malaria in 2014, 90 % of which were direct costs, while the 16 University of Ghana http://ugspace.ug.edu.gh total annual costs (in millions) were estimated to be up to US$ 131.9 nationwide in Ghana (Sicuri, Vieta, Lindner, Constenla, & Sauboin, 2013). Generally, values for different cost studies have varied across countries and regions, which aside from the methods and perspectives used, could also be attributed to the economic status of the countries. For instance Rappaport, Bonthapally, Aller, & Ther, (2012) estimated the total direct cost of illness associated with asthma in the United States in 2007 to be US$ 43.57 billion, with a mean cost of US$ 3,100. The mean cost was however reported to be US$ 1,900 in Europe (Mukherjee et al., 2014). Even average total costs of injuries have been reported to range among LMIC from US$ 25 to US$ 15,400. (Wesson et al., 2014). 2.8 Conclusion While Quaye et al. (2015) in their COI analyses recommended that more studies be done in other diseases for comparison, Mock et al. (2003) who studied different types of injuries in two different communities in Ghana, used a community approach in their survey as opposed to a hospital survey. Asenso-Okyere & Dzator (1997) outlined some challenges with the community approach to be recall bias and inconsistency with information gathering. The burden of injuries on society cannot be overemphasized as current literature demonstrates the trend across different times and regions. As endorsed by the WHO Commission on Macroeconomics and Health (Conteh et al., 2002), much more investment is needed to increase access to treatment while reducing the direct and indirect costs of illness to households. Protection against high direct treatment costs for serious illnesses is of particular importance, for example, through tax- or insurance-based financing systems (Russell, 2004). In Ghana, there seems to be insufficient data with regards to household cost 17 University of Ghana http://ugspace.ug.edu.gh of severe injuries, more so, for those receiving care from a tertiary institution which accounts for a significant number of incidents. 18 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 3.0 METHOD 3.1 Study Design The study was retrospective, using Cost-of-illness analytic approach of injuries and records review. As described in Figure 2, the study adopted a fixed design and a quantitative method to determine direct, indirect and intangible costs of candidate patients. Cost-of-Illness study Ethics Quantitative Desk review of study Study approval data Annual Random sampling record of of injured patients injuries Informed Cost data collection consent Quality control Data processing Data analysis: No scientific -Type of injuries Fraud/Falsification -Estimate injury Figure 2: Schematic diagram of the study design 19 University of Ghana http://ugspace.ug.edu.gh 3.2 Study area The study was undertaken at the Casualty and Accident centre of the Korle-Bu Teaching Hospital, Ghana. Established on October 9, 1923, the Korle Bu Teaching Hospital is presently the third largest Hospital in Africa and the premier national referral centre in Ghana. The 2000-bed capacity Hospital currently has over 4,000 medical and paramedical staff with an average daily attendance of 1,500 patients, about 250 of which are admitted daily for further management (KBTH, 2014). Korle-Bu Teaching Hospital is located in the Ablekuma locality of Accra Metropolis, in the Greater Accra region of Ghana. The hospital has 21 clinical and diagnostic departments/units which include the Casualty and Accident Centre where the study is to take place (KBTH, 2014). The Casualty and Accident Unit of the hospital sees both referral and walk-in trauma cases in the Metropolis as well as the regions and sometimes from the West African Sub region and works in relation with other Sub-BMCs in the Hospital. This makes it an ideal study area to capture a good representation of the Ghanaian population. It runs a twenty-four-hour service and is equipped for the management of varying degrees of trauma. According to the most recent annual report of the hospital, the department saw a total of 2914 patients in the 2013 calendar year (KBTH, 2014). The highest cause of mortality for the same period was reported by the department to be Head injuries as a result of Road traffic injuries, followed by assaults and fall from height. 3.3 Study population The study population consisted of patients who had been managed at the Casualty and Accident Centre of the Korle-Bu Teaching Hospital. The patients were interviewed at their review clinic days. 20 University of Ghana http://ugspace.ug.edu.gh Inclusion Criteria: Review injured patients who reported to the Casualty and Accident Unit of Korle-Bu Teaching hospital and consented to be part of the study. Exclusion Criteria 1. Patients who did not consent to be interviewed 2. Patients whose injury severity did not allow them to talk 3.4 Sample size Calculation The sample size was determined at a confidence level of 95% using the single sample for infinite population formula: Ss = (z2 x σ2)/d2 (Kadam & Bhalerao, 2010), where Z = Z value (e.g. 1.96 for 95% confidence level); σ = Standard Deviation; d = distance on either side of mean in confidence interval. In a similar study by Mock et al. (2003), though employing a community-based approach, estimated the mean household out-of- pocket cost of injuries in rural and urban areas to be US$ 11 ± 58 for one of the communities studied in Ghana. Using the standard deviation from Mock et al. (2003), and assuming d of 7, the sample size was calculated as follows: [1.962 x 582] / 72 = 263.74, which approximated to 264. Adding 15% of this figure to accommodate for possible fall-outs, a resulting 303 patients was obtained as sample size. However, a total of 301 patients were involved in the study. 21 University of Ghana http://ugspace.ug.edu.gh 3.5 Sampling Method Review of Casualty and Accident Centre records: Firstly, record books for the Casualty and Accident Centre were used to capture all patients seen within the past year from January 2016 to December 2016. A complete count was then done from these records books. Available folders were retrieved for verification of incomplete recorded information. Selection of participants: Following management of injuries at the Accident and Casualty center, patients are required to return to the same center or appropriate unit for review. The reviews are done at the Out-Patient Department (OPD) area of the Orthopaedics department or appropriate units, as well as the theatre waiting area, in the case of those who require wound inspection. These are the review patients from whom participants for the study were selected. Four (4) days out of the working week were randomly selected with the use of an online sampling software (random.org). On each of those selected days, before the clinic started, a list was compiled, consisting of the names of patients who had turned up for review following management at the Accident center for a sustained injury. At the waiting areas, patients from the compiled list reporting for review were then selected randomly. This selection was made by first randomly selecting a starting point out of 3 by use of the same software (random.org). The random scheme of 3 was then followed with replacement to obtain the patients for the study for each selected day. The patients were spoken to, and with their consent, interviewed before or after their review session. Patients were given the option of having the interview done at a later review date, in which case, their contacts were taken and followed up to the rescheduled date of review where the interview was done. 22 University of Ghana http://ugspace.ug.edu.gh 3.6 Study Variables The outcome variable in the quantitative aspect of the study was total household cost whereas the independent variables were direct, indirect and intangible costs. Direct cost: This was made up of surgical, medical and non-medical costs (i.e. cost of transport, property damage cost, food and consumables, communication) Indirect cost: This included the number of days the patient had been absent from work, the number of days caretakers had absented themselves from work, the occupation of caretakers, waiting and travel times to and fro the Casualty and Accident Centre. Intangible cost: Variables for intangible cost consisted of patients’ fear, pain, and emotional sufferings. The sociodemographic characteristics of each participant were also obtained, including marital status, monthly income, NHIS beneficiary status, and age. Others included the highest education level and the nature of the injury. 3.7 Data Collection Techniques and Tools Structured questionnaires, as well as records review were employed for data collection. The questionnaire was made up of both open and closed ended questions covering relevant data on patients’ socio-demographic information. The questionnaire also focused on identifying the type of cost incurred by injured patients. It further elicited information on fear, emotional suffering and pain endured by patients as a measure of the intangible cost. The questionnaire (Appendix 2) was in five (5) sections: The first part capturing the sociodemographic characteristics of the patient and coding information. The second section described the nature of the injury obtained. The third part, in ascertaining direct cost, 23 University of Ghana http://ugspace.ug.edu.gh obtained estimate for property damage, medical cost, and out-of-pocket expenditure cost. The fourth and fifth sections assessed variables for estimating indirect and intangible costs respectively. 3.8 Quality Control Adequate mechanisms were put in place to safeguard and guarantee data accuracy and quality, thus minimizing bias. The measures included training of research assistants, pretesting of questionnaires, editing of completed questionnaires, and data entry. Research assistants were also monitored on a daily basis. All completed data were validated and entered on a daily basis. Also, the dataset was cleaned before analysis. Completed questionnaires were kept under key and lock to prevent unauthorized people from gaining access to them. Again, the questionnaires will be discarded six months after publication of the findings of the study. Training of field workers: Three (3) research assistants were recruited to help code and enter the information obtained into Microsoft Excel (version 2010). These individuals were clinical year medical students of University of Ghana School of Medicine and Dentistry. They were fluent in English, Akan, and Ga and were trained for two (2) days on the questionnaires and how to obtain consent, as well as handle the information collected. Their work was additionally supervised and reviewed daily. Pre-testing of questionnaires: The questionnaire was pre-tested prior to the final administration to the candidate participants. Pre-testing was conducted on injured patients and their accompanying household members at the Casualty and Accident Unit of Korle-Bu Teaching Hospital. However, this sample was not part of the sample for the main study. This stage also offered the interviewers better understanding of the questionnaire and 24 University of Ghana http://ugspace.ug.edu.gh organize appropriate answers for likely questions that could have been asked by some members of the study population. Editing completed questionnaires: All exposed errors and inconsistencies from the conducted pre-testing were corrected and reviewed before actual commencement of the study. Data entry and processing: The information collected were screened thoroughly, validated, serialized and coded within 24 hours before entry into Microsoft Excel 2010. After entry, the data set was cross-checked for errors with individual hard copies to ensure every defined variable was in its right place. The data was analyzed in Microsoft Excel 2010 and Stata version 12. 3.9 Data Analysis 3.9.1 Background characteristics of respondents Sociodemographic information was extracted, with their corresponding proportions counted and tabulated by use of Microsoft Excel 2010. These included age, sex, marital status and highest educational level. 3.9.2 Estimation of direct costs This was made up of surgical intervention cost, medical cost, transportation, property damage cost, cost of food and consumables, cost on first-aid and additional medication. Surgical costs: This was the cost of surgical interventions and items carried out from admission to discharge. This was obtained from the discharge bill in the folders of the patients. An average surgical cost per injury type for patients who had procedures done was 25 University of Ghana http://ugspace.ug.edu.gh then estimated and reported. This was computed separately for patients with and without access to the National Health Insurance (NHIS). Medical cost: This consisted of non-surgical costs incurred by the patients and relatives. This included the cost of medication, laboratory investigations, first aid, the cost of folders, radiographs, allied treatment such as physiotherapy and dietherapy. These were summed up for each patient and an average determined. Likewise, this was computed separately for patients with, and without the national Health insurance (NHIS). The non-medical related costs included the following: Transport cost: This was the sum of the cost of conveying the patient to and from the hospital on the day of injury occurrence as well as subsequent reviews, where applicable. Property damage cost: Where applicable, this was estimated from repair or replacement of lost or damaged property. Food and Consumables cost: This category consisted of the sum the cost of food and drinks as well as other consumables that the injured patient consumed in the period of admission in the hospital where applicable. Miscellaneous: These consisted of the sum of unaccounted for, but significant cost which arose from legal and court procedures, communication, photocopies, etc. The total direct cost was ascertained by a summation of all components of medical-related and non-medical expenses. 3.9.3 Estimation of indirect costs Indirect cost was calculated by multiplying the daily minimum wage rate by the number of workdays lost for the patient as well as his/her care giver(s). Productivity loss was therefore 26 University of Ghana http://ugspace.ug.edu.gh valued using the 2017 national minimum wage in the country (that is GH¢ 8.80 per day). The estimation of indirect household cost is shown in Table 1: Table 1: Estimation of household indirect cost of Injuries No Category Cost estimation approach 1. Days lost to patients This is the summation of days lost to patients who are due to injury per month 2. Days lost to those who accompany the patient to This is the summation of the hospital. days lost to household members as a result of injury per month 3. Productivity loss due to traveling and waiting This is the summation of the times total number of hours spent by the patient as well as household members as traveling time to seek treatment for injuries per month 4. Total Indirect cost This is the overall aggregation of the total valued productivity, obtained by multiplying the total time spent by the minimum daily wage. 3.9.4 Total household cost estimation The overall cost was estimated by the summation of the total direct cost and total indirect cost. The average cost per patient was determined by dividing the overall cost by the number of injured patients sampled. 27 University of Ghana http://ugspace.ug.edu.gh 3.9.5 Estimation of annual total household cost of injuries by aetiology The total household cost was calculated for the year of review. This was done for each aetiology of injury studied by summing the direct and indirect cost and determining the average total cost for each injury category by dividing this total by the corresponding number of patients in the sample size. The average cost per aetiology was then multiplied by the total number of patients seen at the facility for the period for each category of aetiology. 3.9.6 Determination of intangible costs The intangible cost was assessed by a descriptive approach. This was done using a Likert scale. A five dimension Likert scale was used, in which patients and family members were asked to rate statements under each dimension as (1) ‘not at all’ (2) ‘a little’ (3) ‘moderately’ (4) ‘quite a bit’ (5) ‘extremely’ in respect of the fear, pain and emotional sufferings. The composite intangible score was obtained by adding the dimensions in each domain and multiplying by the number of questions. This score was then reclassified into low, moderate and high intangible cost with their corresponding ranges using descriptive tertile statistic as shown in table 2. Cross tabulations and Chi-square tests were done to ascertain associations in intangible cost between males, females and among different educational levels and age groups. Table 2: Composite intangible score ranges No. Dimension Upper range limit 1. Low 1/3 (11-26) 2. Moderate 2/3 (27-37) 3. High 3/3 (38-55) 28 University of Ghana http://ugspace.ug.edu.gh 3.9.7 Sensitivity analysis of cost of Injuries To test for robustness of costs estimated, one-way (1-way) and multi-way (M-way) sensitivity analysis (SA) were done by varying relevant costs components. The components on which the sensitivity tests were conducted were medication and wage. These components were selected due to the presence of uncertainty associated with those items. The test was performed by increasing the two cost components by 3% and 5% respectively. The various cost components were then evaluated to ascertain proportional changes or otherwise. 3.10 Study Limitations The study limitations were: 1. Recall of patients may not have been entirely accurate for those whose injury occurred a longer time before the study. This could have led to self-reporting of past cost data, thus, misreporting cannot be ruled out. 2. The small sample size may also not have been representative enough for all individual injury types and thus was minimized by randomization. 3. The study site, being a referral hospital may have influenced the nature of injuries reporting to the hospital to more severe injuries. 4. Some of the folders had incomplete records. In such cases, the patients were asked to recall and approximate details of their injury and medical expenditure. 3.11 Ethical Considerations The study was carried out at the Casualty and Accident center of the Korle-Bu Teaching Hospital. Ethical approval was thus sought from the IRB-KBTH. Approval was also sought from the administration hierarchy of the Orthopaedics and Trauma Department. 29 University of Ghana http://ugspace.ug.edu.gh Participants involved in the study were patients who had reported to the Accident centre of the Hospital, had been managed, discharged and had returned for review. Though the participants may not have had any immediate or direct benefits from the study, their responses would be helpful in policy planning and formulation of recommendations to appropriate authorities and to inform reimbursement procedures. There was no anticipated risk or harm from the study. The only inconvenience, being the time spent in the interview. In view of this, the design of the questionnaire (Appendix 2) was well structured to facilitate the discourse. The respondents were told about the general nature of the study and assured of no potential harm as well as the confidentiality of the data to be collected. They were told of their freedom to decline participation, with no consequences should they decide to do so. Respondents had to sign to indicate consent (Appendix 1) while guardians of minors were also required to sign a consent form. For injured patients who happen to be minors, assent was sought after guardians had signed the consent form. All patients were allowed to ask any questions they might have had about the study. A written informed consent was sought from study participants and their accompanying family members before data collection. Regarding privacy and anonymity, the information provided was, (and still is being) treated with strict confidentiality. Filled questionnaires are being kept under key and lock to prevent unauthorized people from gaining access to them. Also, the questionnaires will be discarded six months after publication of the findings of the study. Again, no participant’s name shall appear or be mentioned in any report that will come out from this study. No payments were made for the time spent by the respondents. There is no conflict of interest declared. This study was self-sponsored in partial fulfillment for a Masters of Public Health Degree from the University of Ghana. 30 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR 4.0 RESULTS 4.1 Background characteristics of respondents A total of 301 patients were interviewed, representing 99.3% response rate. Of these, 59.5 % (179) were males and 40.5% (122) were females. The mean age was 35.2 years (SD: 18.3), ranging from 1 to 85 years. The mean age, however, varied for males and females, being 34.7 and 36.0 years respectively. Table 3 shows other socio-demographic characteristics of respondents. The modal age group was the 30-39 year group, followed by those between 20-29 years. However, 9.6% of the injured population were below the age of 10 years while 7% consisted of people five years or younger. Those above seventy years also constituted 3.7% of the study population. Of the 50.2% who were employed, 40.4% worked in the formal sector while the majority worked in the informal sector. Also, the majority of the respondents were NHIS beneficiaries (about 67%). Again, as demonstrated in Table 3, most of the respondents, (91.4%) have had at least basic education, with those in the tertiary education category having the highest average cost among the education level category. Among the unemployed, 8% had indicated that they had unsuccessfully been looking for a job within the past six months. Furthermore, of those who were unemployed, 44.3% of them indicated that they could not work because of the injuries they had sustained. Majority of the respondents were non-workers, while among those who worked, 59.6% worked in the non-formal sector. Also, most of the respondents (56.1%) had to access additional source of income during the period of their injury management. 31 University of Ghana http://ugspace.ug.edu.gh Table 3: Background characteristics of respondents Background characteristics Number Percent (% Sex: Male 179 59.5 Female 122 40.5 Age (years): 1-9 29 9.6 10-19 29 9.6 20-29 60 19.9 30-39 64 21.3 40-49 54 17.9 50-59 29 9.6 60-69 25 8.3 70 + 11 3.7 Marital Status: Married 123 40.9 Not married 178 59.1 Employment: Employed 151 50.2 Unemployed 88 29.2 Student/ Apprentice 42 13.9 Housewife 2 0.7 Retiree 18 6.0 Employment sector: Formal sector 61 19.9 Informal sector 90 33.6 Non-workers 150 46.5 NHIS Beneficiary: Beneficiary 203 66.8 Non-beneficiary 98 32.6 Education: No education 26 8.6 Primary 55 18.3 Middle/JSS/JHS 79 26.3 SSS/SHS/Vocational/Technical 92 30.6 Tertiary 49 16.3 Additional source of income: None 132 43.9 Other relatives 85 28.2 Friends 37 12.3 Welfare fund from workplace 19 6.3 Church 12 4.0 1Others 16 5.3 Total 301 100 1 Includes: Loan/Savings/Sale of property, motorist who reportedly caused the injury, Media, NGO, Member of Parliament, Private insurance 32 University of Ghana http://ugspace.ug.edu.gh 4.2 Total household cost of injuries The total household cost of injury was GHS 1,691,273.85 (US$ 388,798.59) of which the direct cost constituted 82% and the indirect cost, 18% as shown in Figure 3. Indirect cost 18% Direct cost 82% Figure 3: Proportions of direct and indirect cost of injuries Direct cost: The total direct cost spent by the respondents of this study was GHS 1,384,548.83 (US$ 318,287.09). The direct cost as shown in Table 4 depicts, in all, a total direct cost of GHS 1,384,548.83 (US$ 318,287.09), with an average direct household spending of GHS 4,599.82, SD: 4,710.94 (US$ 1,057.43) on medical and non-medical areas relating to injuries. As demonstrated in Table 4, most of the direct cost of injuries was accounted for by medical component, of which cost of intervention took the greatest proportion, followed by the cost of medications. Food and drinks consumed cost nearly as much as travel cost. Other items such as implants for surgery, crepe bandages, and cervical collars accounted for 2.7% of the direct cost of injuries. Similarly, miscellaneous items constituted 1.1% of the direct cost of injuries. These included costs such as phone credit for communication, legal fees, consumables, and processing fees paid due to the injuries sustained. 33 University of Ghana http://ugspace.ug.edu.gh The medical cost contributed more to the direct cost compared to the non-medical cost by a ratio of about 3:1. The surgical cost (the cost of the main intervention) accounted for most of the total cost, and invariably to the direct cost of injuries among the respondents in this study. For the direct cost of injuries, the cost of the surgical intervention was followed by the cost of medications, food and drinks, property loss, laboratory investigations, travel cost and imaging, respectively as the highest contributors to the direct cost. More money was found to have been spent on food and drinks during the period of admission, as compared with that of traveling to and from the hospital, as well as property loss. Loss of property contributed 7.8% of the direct cost, while traveling time accounted for 7.8% of the direct cost of injuries. Indirect cost: As shown in Table 4, the greatest proportion of the indirect cost of injuries was lost to the days absented from work by both the patient and the caregiver(s), with the time spent in traveling and waiting taking the least proportion. As demonstrated in Figure 3, the direct cost was nearly four times the indirect cost. 34 University of Ghana http://ugspace.ug.edu.gh Table 4: Total household cost of injuries Cost Components Total Cost Cost (GHS) profile (US$)1 (%) DIRECT COST Surgical 550,772.81 126,614.44 32.5 Registration/ Consultations 33,329.73 7,662.01 2.0 Medication 225,954.68 51,943.60 13.4 Laboratory investigations 104,046.67 23,918.77 6.2 Imaging 64,724.03 14,879.09 3.8 Allied treatment 24,284.68 5,582.69 1.4 Others 38,088.54 8,755.99 2.3 Medical 1,041,201.14 239,356.58 61.6 Property loss 107,842.28 24,791.33 6.4 Travel cost 102,153.38 23,483.54 6.0 Food/ Drinks 118,238.82 27,181.34 7.0 Miscellaneous 14,240.31 3,373.97 0.9 Non-medical 343,347.69 78,930.50 20.3 SUB-TOTAL 1,384,548.83 318,287.09 81.9 INDIRECT COST Valued productive time lost 230,647.00 53,022.30 13.5 Valued travel and waiting 8,825.32 2,028.81 0.5 time Valued caregiver’s time 69,567.12 15,992.44 4.1 lost 306,728.03 70,512.19 18.1 SUB-TOTAL TOTAL 1,691,273.85 388,798.59 100 1US$1.00 equivalent to GHS 4.35 (Bank of Ghana average monthly interbank exchange rate, May 2017) Intangible cost: Figure 4 shows that 32% of the injured individuals had high intangible cost while 41% were in the moderate range. There was, however, no significant difference in intangible cost by sex (p=0.137), education (p=0.269), employment status (p=0.506), nor aetiology of the various injuries (p=0.871). The proportions of the degree of fear, pain and emotional suffering perceived by the respondents are demonstrated in Figure 4. 35 University of Ghana http://ugspace.ug.edu.gh Low High 27% 32% Moderate 41% Figure 4: Intangible cost of injured patients 4.3 Estimated annual household cost of injuries by aetiology Table 5 shows the total annual household cost of injuries to all patients attending the unit in 2016 to be GHS 48,729,294.01 (US$ 11,202,136.55). Table 5 indicates that the highest proportion of household cost was contributed by road traffic accidents (RTAs) followed by falls, interpersonal violence and play respectively. The cost accrued by RTAs alone was found to be GHS 29,279,253.56 (US$ 6,730,862.89) which was more than the cost of all other injuries combined. Other injuries had a total of GHS 803,286.40 (US$ 184,663.54) accounting for 1.6% of the total household cost of injuries. In all injury aetiologies, the direct cost contributed the greatest proportion of the cost of the injuries. While road traffic accidents, falls and assaults had the highest direct costs in respective order, the highest contributors to indirect cost were RTAs, domestic violence and falls respectively. The highest average cost of injuries for the year was however accrued by RTAs followed by burn 36 University of Ghana http://ugspace.ug.edu.gh Table 5: Total annual household cost of injuries for the year Total cost category of Total Cost Average Cost in GHS Cost Injuries (GHS) (US$)1 (SD)2 profile (%) RTA (3964): Direct 23,921,788.64 5,499,261.76 6,034.76 (5,624.62) 49.1 Indirect 5,357,464.92 1,231,601.13 1,351.53 (4,052.30) 11.0 Sub-total 29,279,253.56 6,730,862.89 7,386.29 (7,919.38) 60.1 Occupational Injuries (294): 946,062.60 217,485.66 3,217.90 (3,053.38) 1.9 Direct 185,096.52 42,550.92 629.58 (750.67) 0.4 Indirect 1,131,159.12 260,036.58 3,847.48 (3,398.84) 2.3 Sub-total Sports (80): 227761.60 52,358.99 2,847.02 (2,615.95) 0.5 Direct 104,623.20 24,051.31 1,307.79 (3,035.41) 0.2 Indirect 332,384.80 76,410.30 4,154.81 (4,820.84) 0.7 Sub-total Domestic Violence (559): 1,745,779.36 401,328.59 3,123.04 (2,969.60) 3.6 Direct 115,550.89 26,563.42 206.72 (346.89) 0.2 Indirect 1,861,330.25 427,892.01 3,329.75 (3,156.15) 3.8 Sub-total Assault (1123): 3,919,348.61 900,999.68 3,490.07 (4,793.05) 8.1 Direct 1,177,218.44 270,624.93 1,048.27 (2,518.66) 2.4 Indirect 5,096,567.05 1,171,624.61 4,538.35 (5,775.20) 10.5 Sub-total Play (257): 996,923.56 229,177.83 3,879.08 (3,723.48) 2.0 Direct 46,959.04 10,795.18 182.72 (274.30) 0.1 Indirect 1,043,882.60 239,973.01 4,061.80 (3,722.17) 2.1 Sub-total Fall (1606): 5,740,020.66 1,319,544.98 3,574.11 (3,305.52) 11.8 Direct 1,418,933.12 326,191.52 883.46 (1,975.54) 2.9 Indirect 7,158,953.78 1,645,736.50 4,457.63 (3,998.84) 14.7 Sub-total Burns (293): 1,698,860.88 390,542.73 5,798.16 (5,082.14) 3.5 Direct 323,615.57 74,394.38 1,104.49 (1,939.50) 0.7 Indirect 2,022,476.45 464,937.11 6,902.65 (5,871.99) 4.2 Sub-total Others (503): 702,565.25 161,509.25 1,396.75 (1,049.05) 1.4 Direct 100,721.15 23,154.29 1,627.18 (3,203.02) 0.2 Indirect 803,286.40 184,663.54 1,596.99 (1,246.64) 1.6 Sub-total 48,729,294.01 5,551.27 (7,056.88) 100 Total 11,202,136.55 1US$1.00 equivalent to GHS 4.35 (Bank of Ghana average monthly interbank exchange rate, May 2017) 2Standard deviation 37 University of Ghana http://ugspace.ug.edu.gh 4.4 Sensitivity analysis of cost of Injuries As shown in Table 6, a one-way SA conducted by varying the cost of medication by 3%, and 5% yielded respectively 0.4% and 0.7% increases in total treatment cost of type two treatment cost. However, same analysis conducted on wage rate yielded percentage increases of 2.0 and 2.3 respectively in total treatment cost. Also, while the 3% and 5% variations in medication respectively resulted in 0.1 percent each increase in direct cost, the same level of variations in wage rate respectively resulted in 1.6 and 1.9 percentage increases in indirect cost. Furthermore, concurrent variations in both medication and wage rate by 3% and 5% resulted in a percentage fall in direct cost in proportions to total treatment cost and thus a percentage rise in indirect cost in proportions to total treatment cost. However, there was 2.4 and 3.0 percentage increases in total treatment cost respectively. The results of the sensitivity analysis (Tables 6) showed that this study’s cost estimates were sensitive to changes in wage and medicine cost variables. 38 University of Ghana http://ugspace.ug.edu.gh Table 6: Sensitivity analysis of total cost Scenario Cost Percentage Total cost Percentage Proportion of total Percentage component change in change in cost change in parameter total cost proportions of cost GHS US$ Direct Indirect Direct Indirect Base scenario 0 1,693,151.82 389,230.56 0.0 81.7 18.3 0 0 Variation (One- Medication 3 1,699,930.46 390,788.86 0.4 81.8 18.2 0.1 -0.1 way Sensitivity Analysis)1 5 1,704,449.55 391,827.74 0.7 81.9 18.1 0.1 -0.1 Variation (One- Wage rate2 3 1,726,261.69 396,842.02 2.0 80.2 19.8 -1.6 1.6 way Sensitivity Analysis)1 5 1,732,905.37 398,369.31 2.3 79.9 20.1 -1.9 1.9 Multi-variation Medication 3 1,733,040.33 398,400.33 2.4 80.3 19.7 -1.5 1.5 (Multi-way and Wage Sensitivity rate 5 1,744,203.10 400,966.49 3.0 80.0 20.0 -1.7 1.7 Analysis)1 US$1.00 equivalent to GHS 4.35 (Bank of Ghana average monthly interbank exchange rate, May 2017) 1The cost of medication and wage rate was independently and concurrently varied by 3% and 5% increment. 2The national minimum wage per day of GHS8.80 as at May, 2017 was used to value productivity days and time lost to patients 39 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE 5.0 DISCUSSION With the objective of determining the direct, indirect and intangible cost of injuries to households using the Casualty and Accident centre of the Korle-Bu Teaching Hospital, this study found the total cost of injuries to the respondents to be GHS 1,691,273.85 (US$ 388,798.59). The constituent direct and indirect costs were found to be GHS 1,384,548.83 (US$ 318,287) and GHS 306,728.03 (US$ 70,512.19) respectively. This figures respectively represented 82% and 18% of the total cost of treatment. The estimated annual cost however to all injured patients reporting to the unit was found to be GHS 48,729,294.01 (US$ 11,202,136.55), with an average total cost of GHS 5,551.27 (US$ 1,276.15). The study also found that the highest proportion of injured patients experienced a moderate amount of intangible cost (41%) while 32% experienced a high amount of intangible cost. 5.1 Background Characteristics There was a greater occurrence of injuries among males, as compared to females. The higher occurrence of injuries among males in this study is consistent with the thought that the existence of gender differences is as a result of the propensity of males to take more risks (Harris et al., 2006), hence increasing the chances for involvement in injurious activities and situations. With a mean age of 35.2 years, this study finds about 59.1% of all injuries to have occurred between the age groups of 20 and 50 years. Also, 41.2% of the injuries occurred between the ages of 20 and 39 years. Though limited to road traffic accidents, Hesse & Ofosu, (2014) similarly found the highest fatalities among casualties during the period 1991 to 2011 to be in the 26 – 35-year-old age group from their epidemiological study of road traffic accidents in Ghana. The Global Burden of Disease likewise ascertained the injuries to be commonest in the 15-44 year group range (Murray & Lopez, 1996). This 40 University of Ghana http://ugspace.ug.edu.gh age-group is coincident with the most productive year group of the citizenry, and thus, probably indicates that health promotion strategies related to injuries should target this same group. Also, such strategies, as per this study, should not discriminate between employment categories, as well as well as educational levels. Promotion and informational services must be tailored to meet all social groups among the productive year groups. This study found the majority (43.8%) of all injuries presenting to the Casualty and Accident centre to be as a result of road traffic accidents. The Global Burden of Disease suggests road traffic accidents to be the leading cause of death among men, and the fifth commonest among women in the productive year group (Murray & Lopez, 1996). Our finding is also consistent with findings of 43.9% in Tanzania (Casey et al., 2012) and 49% in Kampala, Uganda (Hsia et al., 2010b). After road traffic accidents, falls and interpersonal violence were respectively the next highest causes of injuries. Also, aside RTAs, falls were the highest source of household cost. Falls, in addition, had the third highest average cost of injury by aetiology. In favourable comparison, Mock et al., (2003) in their Ghanaian community survey of the incidence injuries also found falls and play to be significant aetiologies for trauma. It has been indicated that different environmental risks exist for play and falls. Findings of this study add to the assertion that prevention strategies need to prioritize falls as well as play. Implications for this would be to focus on the extremes of age, reining in interventions such as making playgrounds safer, use of safety gates, grab bars, improving general geriatric care and screening programs among others. This study also found that people with no formal education have 37% less odds of sustaining an injury in an RTA as compared to people with at least, primary level education. 41 University of Ghana http://ugspace.ug.edu.gh 5.2 Direct cost The study established the household cost of injuries in a typical referral centre in Ghana, with the hope of contributing to the provision of concrete and reliable foundation for the development of injury management in Ghana. Though a good number of cost studies have generally been done on the subject of injuries (Wesson et al., 2014), very few have explored the area of injuries in Ghana, albeit other significant cost studies for physiotherapy care, chronic diseases (Tagoe, 2012), Diabetes (Quaye et al., 2015), motorcycle accidents (Kudebong et al., 2011), malaria (Nonvignon et al., 2016) among others. The medical cost in this study, consisting of the cost of surgical/medical intervention of the injury as well as the medication, allied treatment, and consultation fees constituted a greater proportion of the direct household cost, i.e. average of GHS 4,410.58 and the non-medical cost accounting for the remaining average of GHS 1,140.69. Though these costs generally varied among the different injury aetiologies, a significant component of the household non-medical direct cost was property loss or damage. Wesson et al., (2014) in their review of injury cost in LMIC reported that in 22 studies reporting only direct costs per hospitalization, costs ranged from US$ 14 to US$ 17,400. Regarding specific types of injuries, the highest reported median direct medical costs were for burn injuries in this review, while of the six studies reporting direct medical and non- medical costs, costs ranged from less than US$ 25 to US$ 15 400. The finding of Wesson et. al, (2014) favourably compares to that of this study which determined the direct cost of injury management to be US$ 1,276.15. While one study in Nigeria reports the average direct medical cost of road injuries to be US$ 35.64, some in major hospitals reported a spending of US$ 444 per month (Juillard, Labinjo, Kobusingye, & Hyder, n.d.), which could compare to this study’s finding per annum. In comparison to the former, the Korle-Bu Teaching Hospital being a tertiary institution, and one of the few major referral centres in 42 University of Ghana http://ugspace.ug.edu.gh the country and West Africa, is likely to have comparatively more severe cases, though the accident centre in particular welcomes ‘walk-in’ patients without referral. Also, by virtue of the hospital’s status, it is more likely to administer more sophisticated and standard investigations and treatment options which may invariably be more costly. This is evident in the study by the fact that laboratory investigations, imaging, and medication alone comprised 23.4% of the total cost. Nguyen, Ivers, Jan, & Pham (2017) also ascertained the average out of pocket expenditure of injuries during hospitalization in Vietnam to be over US$ 270 while indicating that costs were higher for severe injuries and those that required a more complex surgery. Among the estimated annual household cost, the highest average cost was accounted for by road traffic accidents followed by burns injuries. Burns related injuries may present with a range of systemic malfunctions which after their initial correction, may require a long-term period of wound management, often rather expensive procedures. Road traffic accidents, similarly come with a very wide spectrum of injuries, which may range from non-serious bruises and lacerations to severe morbid states requiring extensive management. Thus, direct costs associated with these injuries are expectedly the highest among the different aetiologies. Hesse & Ofosu (2014) reported an annual mean of 6,259.8 injuries from their epidemiological review of road traffic accidents in Ghana. It could, therefore, be extrapolated from this study’s average household spending of GHS 5,551.27 that the annual social cost of road traffic accident in Ghana is GHS 34,749,839.95 (US$ 8,692,473.40). This therefore implies that the household cost alone of road traffic injuries accounts for 0.2% of GDP and 92.4% of GDP per capita. Wesson et. al, (2014) in their review of LMIC similarly reported the cost of injuries to be 97% of GDP per capita. However, in 2007, the National Road Safety Commission of Ghana estimated road traffic accidents to cost 1.6% of GDP. 43 University of Ghana http://ugspace.ug.edu.gh This variance with our estimated figure is likely because while this study solely focuses on household cost, the latter might have included other cost components such as institutional costs. Evident from this study, in conjunction with indirect and intangible cost, direct cost justifiably in the conceptual framework, constitutes a major component of the total household cost of injuries. Incidentally, the average direct medical cost for NHIS subscribers was higher than that of non-subscribers. This however could have been accounted for by the fact that the NHIS subscribers had greater leverage to opt for more extensive treatment plans, as compared to non-subscribers. There was no significant difference in cost borne to injuries between NHIS subscribers and non-subscribers. This suggests a limitation of the scheme to address severe injuries. There was, in addition, a significant 56.1% of the study population requiring help from other sources to enable them undergo management for all severe injury types. Since most of this same population was employed, albeit informal, governmental social service institutions, as well as other financial institutions, could explore means of accessing such eligible individuals for financial assistance. 5.3 Indirect cost This measurement for households included time spent away from work as well as additional time (not included in the days spent away from work) spent waiting and traveling. The human capita method employed here considered the national minimum daily wage as of the period of the study, which was GHS 8.80, implying a GHS 1.10 minimum wage per hour. This method was however applied to get a monetary value of indirect cost for only the employed and the unemployed individuals who had been looking for a job within the past six months prior to the occurrence of the injury. The most number of hours were lost to the households of road traffic accidents, followed by sports injuries. This again reflects the 44 University of Ghana http://ugspace.ug.edu.gh severity of injuries which were attended to at the accident centre of the Korle-Bu Teaching Hospital. Mock et al., (2003) found the mean times absent from work in Ghana for injured people from urban and rural areas who had been working before their injury to be 43.3±61.6 days and 32.6±53.5 days, respectively. The difference, in comparison to our finding of 87.08±316.80 days, could be accounted for by the fact that while the former conducted the study with a whole community approach of which not all participants sought formal health care, this study is based on a specialist hospital approach, saw a wider range of morbidities associated with injuries. Anders, et al., (2013) also comparatively had an average of 185.2 days spent away from work in his cost assessment of similarly severely injured patients in Germany which favourably compares with findings from this study. The average total cost from this study was comparable to Wesson et al. (2014) who in their review of literature performed to explore costing data available for injury and/or trauma care in LMICs found the median cost of US$ 4,085.00 among thirteen studies which reported direct medical, direct non-medical and indirect costs. The indirect cost from this study constituted 18% of the total household cost. Though this is consistent with many COI evaluations, several others have also reported a higher proportion of indirect cost, as demonstrated by studies on obesity (Ricci et al., 2005), schizophrenia (Tajima-Pozo et al., 2015), and stroke (Joo et al., 2014), while that of smoking in Vietnam showed a proportionate 49.5% of total cost being accounted for by indirect cost (Yang et al., 2011). This may be accounted for by the fact that, unlike injuries, these diseases have a longer disease progression. It has also been suggested that the varying weights and proportions of indirect cost could be as a result of associated comorbidities (Rappaport et al., 2012). The average waiting time for hospital consultations was found to be 3.58 hrs (SD: 2.33) while that of traveling to and from the hospital was found to be 3.36 hours, (SD: 3.42). Hospital waiting time is a major index for quality of care. While negligible waiting time is 45 University of Ghana http://ugspace.ug.edu.gh nearly unattainable, necessary steps should be taken to minimize the near 4-hour mean waiting period for injured patients. 5.4 Intangible cost Some studies have previously attempted to cost emotional and psychological suffering. Xie et al., (2008), for instance, used willingness-to-pay (WTP) to ascertain the intangible cost of Osteoarthritis. This study, however, describes the intangible component without costing, where it was found that the greatest proportion of the patients had a high proportion of fear, pain and emotional suffering. This finding highlights the importance of incorporating psychological and counseling programs into the holistic management of injuries in Ghana, more especially, for severe injuries in tertiary institutions like Korle-Bu Teaching Hospital where this study was conducted. Furthermore, though there was no significant difference in intangible cost among sex, likewise occupations, females had a higher overall intangible cost score compared to males. This finding seems to be consistent with the concept that there is higher prevalence of stress, anxiety and pain perception among females as compared to males (Abdel Wahed & Hassan, 2017). In summary, injuries are a huge source of burden to society at large, with Ghana not being an exception to this. This study achieved its aims of measuring the direct, indirect and intangible costs of injuries to the patients attending the Casualty and Accident center of the Korle-Bu Teaching Hospital. It found that while averages varied for different aetiologies, the annual total household cost to injuries was GHS 48,729,294.01 (US$ 11,202,136.55) with 41% of the injured patients incurring moderate level of pain, fear, and emotional suffering. 46 University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX 6.0 CONCLUSION AND RECOMMENDATIONS 6.1 Conclusion The findings of this study are relevant for health policy making in Ghana. This study, among other findings, provides estimates and distribution of costs for injuries in a specialist hospital in Ghana. In addition to ongoing efforts in prevention, this study provides further evidence on the limited benefits of health insurance in shielding patients and their family from the high costs of severe injuries. Also, the overall moderate intangible cost of injuries further suggests the need to institute and integrate proper counseling sessions with severe injury management, as is currently being done with HIV/AIDS management and with injury management in other countries. Direct cost averages found in this study also serves as an important basis for reimbursement policies by insurance agencies for people who sustain severe injuries. It may also serve as reliable data to inform workers compensation and employers liability policies in Ghana. 6.2 Recommendations The following recommendations are made from the study: 1. For costing and budgeting: a. The unit costs from this study could be used in costing interventions in injury management. b. The unit costs could be used to provide budgets for similar injury units in health facilities in the country. 2. Accountability: The study results provides benchmark estimates for accountability purposes with regards to spending on injuries 47 University of Ghana http://ugspace.ug.edu.gh REFERENCES Abdel Wahed, W. Y., & Hassan, S. K. (2017). Prevalence and associated factors of stress, anxiety and depression among medical Fayoum University students. Alexandria Journal of Medicine, 53(1), 77–84. https://doi.org/10.1016/j.ajme.2016.01.005 Amissah, J. (2016). Household Cost Of Injuries in Children Under Five Years in the Ejisu- Juabeng Municipality of the Ashanti Region, Ghana: A Cross Sectional Study. Value in Health, 19(7), A907. https://doi.org/10.1016/j.jval.2016.08.165 Anders B, Ommen O, Pfaff H, Lüngen M, Lefering R, Thüm S, et al. (2013). Direct , indirect , and intangible costs after severe trauma up to occupational reintegration – an empirical analysis of 113 seriously injured patients Direkte , indirekte und intangible Kosten nach einem schweren Trauma. GMS Psychosoc Med, 10(2), 1–15. https://doi.org/10.3205/psm000092 Asenso-Okyere, W. K., & Dzator, J. A. (1997). Household cost of seeking malaria care. A retrospective study of two districts in Ghana. Social Science & Medicine (1982), 45(5), 659–67. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9226789 Byford, S., Torgerson, D. J., & Raftery, J. (2000). Economic note: cost of illness studies. BMJ (Clinical Research Ed.), 320(7245), 1335. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10807635 Casey, E. R., Muro, F., Thielman, N. M., Maya, E., Ossmann, E. W., Hocker, M. B., & Gerardo, C. J. (2012). Analysis of traumatic injuries presenting to a referral hospital emergency department in Moshi, Tanzania. International Journal of Emergency Medicine, 5(1), 28. https://doi.org/10.1186/1865-1380-5-28 Conteh, L., Corrao, M., Hanson, K., Kumaranayake, L., Libatique, D., Kurowski, C., & Galvez Tan, J. Z. (2002). Improving Health Outcomes of the Poor. World Health 48 University of Ghana http://ugspace.ug.edu.gh Organisation. Felix Ankomah Asante Kwadwo Asenso-Okyere, B. (2003). Economic Burden of Malaria in Ghana. Retrieved from http://www.who.int/countries/gha/publications/Economic_Burden_of_Malaria_in_Gh ana_Final_Report_Nov03.pdf Haagsma, J. A., Graetz, N., Bolliger, I., Naghavi, M., Higashi, H., Mullany, E. C., … Vos, T. (2016). The global burden of injury: incidence, mortality, disability-adjusted life years and time trends from the Global Burden of Disease study 2013. Injury Prevention : Journal of the International Society for Child and Adolescent Injury Prevention, 22(1), 3–18. https://doi.org/10.1136/injuryprev-2015-041616 Harris, C. R., Jenkins, M., & Glaser, D. (2006). Gender Differences in Risk Assessment : Why do Women Take Fewer Risks than Men ? Judgment and Decision Making, 1(1), 48–63. Health Organization Regional Office for Europe, W. (2014). Injuries in Europe : A call for public health action. Hesse, C. A., & Ofosu, J. B. (2014). Epidemiology of Road Traffic Accidents in Ghana. European Scientific Journal, 10(9), 1857–7881. https://doi.org/10.1177/146642408310300401 House, D. R., Nyabera, S. L., Yusi, K., & Rusyniak, D. E. (2014). Descriptive study of an emergency centre in Western Kenya: Challenges and opportunities. African Journal of Emergency Medicine, 4(1), 19–24. https://doi.org/10.1016/j.afjem.2013.08.069 Hsia, R. Y., Ozgediz, D., Mutto, M., Jayaraman, S., Kyamanywa, P., & Kobusingye, O. C. (2010a). Epidemiology of injuries presenting to the national hospital in Kampala, Uganda: Implications for research and policy. International Journal of Emergency 49 University of Ghana http://ugspace.ug.edu.gh Medicine, 3(3), 165–172. https://doi.org/10.1007/s12245-010-0200-1 Hsia, R. Y., Ozgediz, D., Mutto, M., Jayaraman, S., Kyamanywa, P., & Kobusingye, O. C. (2010b). Epidemiology of injuries presenting to the national hospital in Kampala, Uganda: implications for research and policy. International Journal of Emergency Medicine, 3(3), 165–72. https://doi.org/10.1007/s12245-010-0200-1 Jo, C. (2014). Cost-of-illness studies: concepts, scopes, and methods. Clinical and Molecular Hepatology, 20(4), 327–37. https://doi.org/10.3350/cmh.2014.20.4.327 Joo, H., George, M. G., Fang, J., & Wang, G. (2014). A literature review of indirect costs associated with stroke. Journal of Stroke and Cerebrovascular Diseases : The Official Journal of National Stroke Association, 23(7), 1753–63. https://doi.org/10.1016/j.jstrokecerebrovasdis.2014.02.017 Juillard, C., Labinjo, M., Kobusingye, O., & Hyder, A. A. (n.d.). Socioeconomic impact of road traffic injuries in West Africa: exploratory data from Nigeria. https://doi.org/10.1136/ip.2009.025825 Kadam, P., & Bhalerao, S. (2010). Sample size calculation. International Journal of Ayurveda Research, 1(1), 55–7. https://doi.org/10.4103/0974-7788.59946 KBTH. (2014). KORLE BU TEACHING HOSPITAL ANNUAL REPORT 2013. Kudebong, M., Wurapa, F., Nonvignon, J., Norman, I., Awoonor-Williams, J. K., & Aikins, M. (2011). Economic burden of motorcycle accidents in Northern Ghana. Ghana Medical Journal, 45(4), 135–42. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22359418 Mock, C. N., Abantanga, F., Cummings, P., & Koepsell, T. D. (n.d.). Incidence and outcome of injury in Ghana: a community-based survey. 50 University of Ghana http://ugspace.ug.edu.gh Mock, C. N., Abantanga, F., Cummings, P., & Koepsell, T. D. (1999). Incidence and outcome of injury in Ghana: a community-based survey. Bulletin of the World Health Organization, 77(12), 955–64. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10680242 Mock, C. N., Gloyd, S., Adjei, S., Acheampong, F., & Gish, O. (2003). Economic consequences of injury and resulting family coping strategies in Ghana. Accident Analysis and Prevention, 35, 81–90. Mukherjee, M., Gupta, R., Farr, A., Heaven, M., Stoddart, A., Nwaru, B. I., … Sheikh, A. (2014). Estimating the incidence, prevalence and true cost of asthma in the UK: secondary analysis of national stand-alone and linked databases in England, Northern Ireland, Scotland and Wales—a study protocol. BMJ Open, 4(11), e006647. https://doi.org/10.1136/bmjopen-2014-006647 Murray, C. J. L., & Lopez, A. D. (1996). The global burden of disease: a comprehensive assessment of mortality and disability from deceases, injuries and risk factors in 1990 and projected to 2010. Harvard University Press, 1, 1–35. https://doi.org/10.1186/1471-2458-13-863 Murray, C. J. L., Vos, T., Lozano, R., Naghavi, M., Flaxman, A. D., Michaud, C., … Lopez, A. D. (2012). Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet, 380(9859), 2197–2223. https://doi.org/10.1016/S0140-6736(12)61689-4 Nguyen, H., Ivers, R., Jan, S., & Pham, C. (2017). Analysis of out-of-pocket costs associated with hospitalised injuries in Vietnam. BMJ Global Health, 2(1). Retrieved from http://gh.bmj.com/content/2/1/e000082 51 University of Ghana http://ugspace.ug.edu.gh Nilsen P, Hudson D, L. K. (2006). Economic analysis of injury prevention-applying results and methodologies from cost-of injury studies. International Journal of Injury Control and Safety Promotion. Nonvignon, J., Aryeetey, G. C., Malm, K. L., Agyemang, S. A., Aubyn, V. N. A., Peprah, N. Y., … Aikins, M. (2016). Economic burden of malaria on businesses in Ghana: a case for private sector investment in malaria control. Malaria Journal, 15(1), 454. https://doi.org/10.1186/s12936-016-1506-0 Polinder, S., Toet, H., Panneman, M., & Van Beeck Iii, E. (2011). Methodological approaches for cost–effectiveness and cost–utility analysis of injury prevention measures Editors. Copenhagen, Denmark. Retrieved from http://www.euro.who.int/__data/assets/pdf_file/0007/144196/e95096.pdf Quaye, E. A., Amporful, E. O., Akweongo, P., & Aikins, M. K. (2015). Analysis of the Financial Cost of Diabetes Mellitus in Four Cocoa Clinics of Ghana. Value in Health Regional Issues, 7, 49–53. https://doi.org/10.1016/j.vhri.2015.08.005 Rappaport, Bonthapally, J., Aller, & Ther. (2012). The Direct Expenditures and Indirect Costs Associated with Treating Asthma in the United States, 3(2). https://doi.org/10.4172/2155-6121.1000118 REPUBLIC OF GHANA 2007 ANNUAL REPORT. (n.d.). Retrieved from http://s3.amazonaws.com/zanran_storage/www.nrsc.gov.gh/ContentPages/42627715. pdf Ricci, J. A., Chee, E., O’Neill, C., Sharp, L., Staines, A., O’Dwyer, V., … Brugha, R. (2005). Lost Productive Time Associated With Excess Weight in the U.S. Workforce. Journal of Occupational and Environmental Medicine, 47(12), 1227–1234. https://doi.org/10.1097/01.jom.0000184871.20901.c3 52 University of Ghana http://ugspace.ug.edu.gh Rice, D. P. (2000). Cost of illness studies: what is good about them? Injury Prevention : Journal of the International Society for Child and Adolescent Injury Prevention, 6(3), 177–9. https://doi.org/10.1136/IP.6.3.177 Rice, D. P., Hodgson, T. A., & Kopstein, A. N. (1985). The economic costs of illness: a replication and update. Health Care Financing Review, 7(1), 61–80. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10311399 Russell, S. (2004). The Economic Burden of Illness for Households in Developing Countries: A Review of Studies Focusing on Malaria, Tuberculosis, and Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome. Schaetti, C., Weiss, M. G., Ali, S. M., Chaignat, C. L., Khatib, A. M., Reyburn, R., … Hutubessy, R. (2012). Costs of Illness Due to Cholera, Costs of Immunization and Cost-Effectiveness of an Oral Cholera Mass Vaccination Campaign in Zanzibar. PLoS Neglected Tropical Diseases, 6(10). https://doi.org/10.1371/journal.pntd.0001844 Sicuri, E., Vieta, A., Lindner, L., Constenla, D., & Sauboin, C. (2013). The economic costs of malaria in children in three sub-Saharan countries: Ghana, Tanzania and Kenya. Malaria Journal, 12(1), 307. https://doi.org/10.1186/1475-2875-12-307 Tagoe, H. A. (2012). Household burden of chronic diseases in Ghana. Ghana Medical Journal, 46(2 Suppl), 54–8. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23661818 Tajima-Pozo, K., de Castro Oller, M. J., Lewczuk, A., & Montañes-Rada, F. (2015). Understanding the direct and indirect costs of patients with schizophrenia. F1000Research, 4, 182. https://doi.org/10.12688/f1000research.6699.2 Tarricone, R. (2006). Cost-of-illness analysis What room in health economics? Health 53 University of Ghana http://ugspace.ug.edu.gh Policy, 77, 51–63. https://doi.org/10.1016/j.healthpol.2005.07.016 United Nations. (2014). Sustainable Development Goals Open Working Group proposal for Open Working Group proposal for Sustainable Development Goals. Retrieved from http://undocs.org/A/68/970 Wesson, H. K. H., Boikhutso, N., Bachani, A. M., Hofman, K. J., & Hyder, A. a. (2014). The cost of injury and trauma care in low- and middle-income countries: a review of economic evidence. Health Policy and Planning, 29(6), 795–808. https://doi.org/10.1093/heapol/czt064 Wesson, H. K. H., Boikhutso, N., Bachani, A. M., Hofman, K. J., & Hyder, A. A. (2014). The cost of injury and trauma care in low- and middle-income countries: a review of economic evidence. Health Policy and Planning, 29(6), 795–808. https://doi.org/10.1093/heapol/czt064 WHO. (2014). WHO | Metrics: Disability-Adjusted Life Year (DALY). WHO. WHO. (2016a). WHO | Road traffic injuries. WHO. WHO. (2016b). WHO | World’s experts gather to exchange knowledge and practice on preventing violence and injuries and saving lives. WHO. WHO-CHOICE. (2003). Making choices in health: WHO guide to cost-effectiveness analysis. Global Programme on Evidence for Health Policy ,World Health Organization, Geneva, 71. https://doi.org/10.1590/S1135-57272004000300012 WHO, I. (2010). WHO | International Classification of External Causes of Injury (ICECI). WHO. World Health Organization. (2008). The Global Burden of Disease: 2004 update. 2004 Update, 146. https://doi.org/10.1038/npp.2011.85 54 University of Ghana http://ugspace.ug.edu.gh Xie, F., Thumboo, J., Edin, F., Fong, K., Edin, F., Lo, N., & Edin, F. (2016). A Study on Indirect and Intangible Costs for Patients with Knee Osteoarthritis in Singapore A Study on Indirect and Intangible Costs for Patients with Knee Osteoarthritis in Singapore, (October). https://doi.org/10.1111/j.1524-4733.2008.00371.x Xie, F., Thumboo, J., Fong, K.-Y., Lo, N.-N., Yeo, S.-J., Yang, K.-Y., … Li, S. (2008). A study on indirect and intangible costs for patients with knee osteoarthritis in Singapore. Value in Health : The Journal of the International Society for Pharmacoeconomics and Outcomes Research, 11 Suppl 1(s1), S84-90. https://doi.org/10.1111/j.1524-4733.2008.00371.x Yang, L., Sung, H.-Y., Mao, Z., Hu, T., Rao, K., & Minh, N. T. (2011). Economic costs attributable to smoking in China: update and an 8-year comparison, 2000-2008. Tobacco Control, 20(4), 266–72. https://doi.org/10.1136/tc.2010.042028 55 University of Ghana http://ugspace.ug.edu.gh APPENDICES APPENDIX 1: PARTICIPANTS’ CONSENT FORM HOUSEHOLD COSTS OF INJURIES: A CASE STUDY OF THE ACCIDENT AMD CASUALTY CENTRE OF KORLE-BU TEACHING HOSPITAL Introduction and Purpose: My name is PAA-KWESI BLANKSON, a post-graduate student from the school of Public Health, University of Ghana. I am carrying out a study in this department to find out the general cost of injuries which report to this hospital and I would be very glad if you could participate in it. Requirement: Your participation would involve a 30-minute interview with to answer some questions about yourself and how much you have spent on the injury incurred. Benefits: You may not have any immediate or direct benefits from my interview but your responses would be helpful in policy planning and formulation of recommendations to appropriate authorities to help reduce injuries in general and to inform reimbursement procedures. Risks: The only inconvenience, if any that you would face by accepting to take part in this study perhaps is your time. Voluntary withdrawal: If you indeed decide to take part, you are allowed to withdraw whenever you wish to, and are also allowed to skip answering any of the questions that you are not very comfortable with. Confidentiality: The information you would provide is going to be treated with strict confidentiality. Apart from my research team and members of the Ethics Committee of this hospital, no body shall have access to the information since it shall be under lock and key. We also assure you that your name shall not appear or be mentioned in any report that will come out from this study. 56 University of Ghana http://ugspace.ug.edu.gh Ethical approval: As part of our duty to conform with standard practice and to ensure your safety, ethical approval has been sought from the Institutional Review Board of Korle-Bu Teaching Hospital (KBTH-IRB) in order to carry out this study. Before taking Consent Do you have any questions you wish to ask about the study? Yes |____| No |____| If yes, please, indicate the questions below)…………………………………………………………………………………. ………………………………………………………………………………………… In case you have any questions later please, do not hesitate to contact Paa-Kwesi Blankson, Department of Health Policy, Planning and Management, School of Public Health, University of Ghana. (Tel: 0208887878) Email: pkblankson@yahoo.com. Also, if you need further clarifications about this study please, kindly contact Mr. Victor Nortey, Research Office Secretary, Korle-Bu Teaching Hospital. (0277743365) 57 University of Ghana http://ugspace.ug.edu.gh PARTICIPANT’S CONSENT I have read, or I have let somebody read or translated all the necessary information that I need to know concerning this study and have fully understood it. I have decided on my own accord without any coercion to take part in this study. However, by deciding to participate in this study, I am not waiving any of my personal rights by signing or thumb printing this consent form. Signature: OR ................................... L/R Thumb Print Interviewers Statement I, the undersigned, have explained this consent to the participant in English language/Ga/Twi, and that she/he understands the purpose of the study, procedures to be followed, as well as the risks and benefits of the study. The participant has fully agreed to participate in the study. Signature of Interviewer…………………………………………………………….... Date…………………………………………………………………………………… Address………………………………………………………………………………. 58 University of Ghana http://ugspace.ug.edu.gh APPENDIX 2 QUESTIONNAIRE HOUSEHOLD COSTS OF INJURIES: A CASE STUDY OF THE CASUALTY AND ACCIDENT CENTRE OF KORLE-BU TEACHING HOSPITAL Dear Respondent, This is a research being carried out on Household costs of injuries: A case study of the Casualty and Accident centre of Korle-Tu Teaching Hospital. I will, therefore, like to take some minutes of your precious time to answer these questions. You are assured that the answers you give will be strictly confidential and your name will not be mentioned in our research reports. Thank you. Qn. No. Questions Response Respondent ID: | | | | Section 1 SOCIO-DEMOGRAPHIC INFORMATION 1 What is your sex 1. Male 2. Female 2 What is your age in years? | | | years 3 What is the highest level of school you attended? 1. No education 2. Primary 3. Middle/JSS/JHS 4. SSS/SHS/Secondary/Vocational/Technical 5. Tertiary 4 What is your current marital status? 1. Married | 2. Not married 5 What is your employment status? 1. Employed 2. Unemployed (If Unemployed, answer Qs. 6 & 7) 3. Student/Apprentice 4. Housewife 5. Retiree 6 If Unemployed, why are you not working now? 1. Unable to work due to illness 2. Other (please specify)...…………………………… 7 If Unemployed, have you been looking for a job in the last 6 months? 1. Yes 2. No 59 University of Ghana http://ugspace.ug.edu.gh 8 If employed, in which sector you are employed? 1. Formal sector 2. Informal sector 9 If Employed, what is your average monthly income? (i.e., salary plus other monies from other sources) GHS …………. 10 Are you an NHIS beneficiary? 1. Yes 2. No Section 2 DETAILS OF INJURIES SUSTAINED 11 What was the cause of the injury sustained? 01. RTA- Motorist 02. RTA- Passenger 03. RTA- Pedestrian 04. Industrial/ Occupational 05. Sports 06. Domestic violence 07. Assault/Interpersonal violence 08. Sexual assault 09. Gunshot 10. Play 11. Fall 12. Burns 13. Foreign body 14. Other: Specify Section 3 DIRECT COST INFORMATION 12 Medical cost: how much have you spent during your hospital GH¢ visit or stay in the management of the injuries you sustained? (a) Surgical intervention(s) if applicable (b) Registration (c) Consultation (d) Laboratory investigations (e) Medicines/drugs (f) Other diagnostic tests (such as scan etc.) (g) Any allied procedures such as dietherapy, physiotherapy (h) Other, specify: 13 Non-m edical cost: how much did you spend/pay for (you and GH¢ accom panying relative)? (a) E stimate of property loss or damage/repair if applicable. Kindly specify: (b) Estimate of total travel cost (to and from the facility) 60 University of Ghana http://ugspace.ug.edu.gh (c) Food cost (d) Drinks/water cost (e) Other miscellaneous costs (i.e., phone calls/phone credits, other consumables used due to this illness) (f) Others, kindly specify: 14 Did you rely on financial help from other source(s) for treatment, apart from normal income? 1. Yes 2. No 15 What are the sources (multiple responses possible)? 1. Relative 2. Friend 3. Savings 4. Loan/Grant 5. Other (Specify) 16 How much money did you receive from the identified source(s)? GH¢……………… Section 4 INDIRECT COST INFORMATION 17 How many days in total have you absented from work (if |___|___|___| days applicable) because of the injury sustained? 18 How many hours in total do you estimate to have spent traveling |___|___|___|hrs to and from the health facility, aside the days lost (if applicable)? X days 19 How many minutes did you spend waiting before you were called |___|___|___|hrs to see the doctor or health officer for treatment aside the days lost X (if applicable)? days 20 Did anyone from your household accompany you from home to the health facility? 1. Yes 2. No 21 If anyone did accompany you to the health facility, what is his or her employment status? 1. Employed 2. Unemployed 3. Student 4. Housewife 5. Retired 6. Other, specify: 61 University of Ghana http://ugspace.ug.edu.gh 22 Did the person who accompany you, come with you from the house and stay with you for treatment and take you back home? 1. Yes 2. No If yes in Q22 above, use same time as the patient’s in Q 17 & 18 for Q 23 & 24 23 How many hrs/days in total did he/she travel to and fro to be with |___|___|___| days you in the health facility? 24 How many hours/days in total did he/she spend with you when |___|___|___| hrs you were receiving treatment in the health facility? 25 Total time spent Section 5 INTANGIBLE COST INFORMATION Please, select from the following statements concerning FEAR due to the injury sustained, from “not at all” to ‘extremely” depending on how it applies to you. 26 I think about my health now more than before since I sustained the injury. 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 27 B ecause of the injury sustained, my future is of concern to me 1. No t at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 28 I am always worried about sustaining another injury 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 62 University of Ghana http://ugspace.ug.edu.gh When I think about my future health, I feel some uneasiness 29 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely N ow kindlys select again, from the following statements concerning PAIN due to the injury sustained, from “not at all” to ‘extremely” depending on how it applies to you. 30 I always feel burning pain around the site of the injury 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 31 I have lost the sense of temperature at the region of injury and am not able to differentiate between warm and cold 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 32 I feel a dull aching discomfort when the site of injury is in function 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely Now kindly select again, from the following statements concerning EMOTIONAL SUFFERING due to the injury sustained, from “not at all” to ‘extremely” depending on how it applies to you. 33 I suffer emotionally because I constantly worry about my condition 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 63 University of Ghana http://ugspace.ug.edu.gh 34 I think I am a burden to others because of my illness 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 35 I am constantly worried about exhibiting the loss of function resulting from the injury down in public 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 36 I feel embarrassed in social situations 1. Not at all 2. A little 3. Moderately 4. Quite a bit 5. Extremely 64