Cost burden of neuroimaging during one-time admission for first-ever acute stroke in Nigeria ============================================================================================ * Adeleye D. Omisore * Morenikeji A. Komolafe * Oluwaseun T. Esan * Bukunmi M. Idowu * Adeniyi S. Aderibigbe * Abdul-Majeed K. Abidoye * Stephen O. Onigbinde ## Abstract **Objectives:** To determine the cost burden of Neuroimaging and its contribution to direct total hospitalization costs (HCs) during one-time admission for first-ever stroke. **Methods:** The clinical characteristics, direct itemised costs and total HCs for 170 consecutive patients with first-ever stroke, admitted at our public tertiary health facility over a 15-month period were evaluated. **Results:** The records of 170 stroke subjects were reviewed. The median total HCs for one-time admission per stroke patient was $183.30 with a median daily cost of $15.86. Median cost of radiological investigations was the highest among the categorized hospital costs. Among the radiological investigations, neuroimaging accounted for at least 99% of cost to patients. **Conclusion:** The financial burden of radiological investigations, particularly neuroimaging, is high during one-time admission of patients with first-ever stroke in our environment. The medical and economic burden of Stroke is immense.1 It is the leading cause of long term disability and the second leading cause of death globally.2 Economically, it imposes heavy financial burden on individuals and the society.2,3 In Nigeria, the prevalence of stroke is 14 per 1000 people and it has a case fatality rate of 40%. World Health Organization projects that approximately 80% of all stroke cases will occur in people living in low and middle income countries, including Nigeria.4 In the United States, there is increased interest in the economic aspects of stroke as a result of the prevailing emphasis on cost containment and managed healthcare/health insurance.5 In Nigeria, the National Health Insurance Scheme (NHIS) became fully operational in 2005.6,7 Out of a population of about 190 million, only 3% are registered in the NHIS.8 Therefore, over 90% of health financing is by private out-of-pocket expenditure.9 About 57% of the Nigerian population falls below the poverty line which is defined as an average income of $1 per day.10 The minimum monthly wage in Nigeria is 18,000 naira (57 US dollars).11 With such indices, it is imperative to establish the cost burden of diseases, especially those with long term disabilities like stroke, for more effective healthcare planning, implementation and resources allocation for a healthier nation. Few studies have evaluated the cost of in-hospital stroke care in different populations with regional differences in total and itemized hospitalization costs.12,13-27 As there could be regional and institutional variations in the in-hospital algorithms for stroke care percentage contribution of itemized hospitalization costs to the total hospitalization cost have varied among these studies. Neuroimaging is required for all stroke patients presenting in our hospital. This is similar to what obtains elsewhere.13 Undergoing neuroimaging immediately after stroke has also been shown to be associated with better outcome and a higher number of quality adjusted life years.28 There is no data on the cost burden of in-hospital care of managing stroke in Nigeria. Available data on cost burden of stroke is on the post-stroke period which showed that the minimum cost needed in a government hospital is ₦95,100 ($600 as of 2012) and the minimum cost needed in a private hospital is ₦767,900 ($4860 as of 2012), within the first 36 months of post stroke affectation.2 A local study to determine the direct cost of in-hospital stroke care and how different hospital services contribute to the total direct cost in our environment where health insurance coverage is limited is needed. This will provide data for government policy makers to develop a health system that will minimize the financial burden of in-hospital care of stroke on individuals and the society. This study was done to evaluate the contributions of itemized hospitalization costs, particularly neuroimaging expenses, to total hospitalization cost during one-time admission for first-ever stroke in a tertiary hospital in order to guide policy decisions.29 ## Methods ### Study design This was a retrospective cross sectional observational study conducted on 170 consecutive patients admitted and managed for first-ever stroke in a tertiary hospital (Obafemi Awolowo University Teaching Hospitals Complex, Ile-Ife, Osun State, Southwest, Nigeria) over a 15-month period (January 2016 - March 2017). The study protocol was approved by the institution’s Ethics and Research Committee and informed consent was obtained from the participants. Literature search for review was conducted by interrogating the Pubmed and Google Scholar databases as well as identifying relevant studies from the reference lists of the articles obtained. ### Hospital characteristics Our hospital is located in the South-Western zone of Nigeria with a land mass of 1902km2 and population of approximately 644,373 people. It is a major tertiary referral centre serving about 7.7 million people in the Southwestern region of Nigeria. The hospital has a 24-hour emergency service for stroke with a stroke team consisting of 3 Neurologists, 2 Neurosurgeons, 3 Neuro-radiologists, one Anaesthesiologist, 2 Neuro-physiotherapists, and 5 stroke Nurses. There are 10 neurology ward beds, 2 intensive care beds and 10 neurosurgery ward beds. Neuroimaging facilities available include computerized tomography (CT) and Magnetic Resonance Imaging (MRI). ### Inclusion criteria We limited the study to all stroke cases who had neuroimaging and received in-hospital care for stroke until formally discharged within the study period. ### Exclusion criteria Records of patients who had delayed medical attention/care due to financial difficulties, discharged against medical advice and those with paucity of details of management were excluded from the study. All non-hospital services and indirect costs were not included in the study. ### Data collection A bottom-top direct microcosting methodology was used to estimate cost. A documentation review of the stroke database and patients’ records in our hospital was done. Data were obtained on socio-demographic variables such as age, gender, occupation and monthly income; and clinical data such as the type of stroke, length of hospital stay and individual risk factors such as hypertension, diabetes, obesity, atrial fibrillation, chronic kidney disease, alcohol use and smoking. Data on direct costs was defined as the actual currency expenditures on services described under the following costs categories: inpatient care cost- which includes total operational charges, facility charges, bed and ward charges; diet cost- which includes cost of meals provided to patients by the hospital; radiological investigations cost- which include cost of neuroimaging (Computerized Tomography (CT) scan or Magnetic Resonance Imaging (MRI)}, Chest X-Ray (CXR), Carotid Doppler, Abdominal Ultrasound (USS), Renal Doppler; laboratory investigations- which include full blood count, serum lipids, coagulation profile (PT, PTTK), Erythrocyte Sedimentation Rate (ESR), Serum Electrolyte/Urea/Creatinine; cardiac study- which include echocardiography (ECHO) and electrocardiography (ECG); medication/drug cost- drug charges; and physiotherapy cost. Calculation of length of stay in the hospital for stroke subjects managed in the ICU included stay in the ICU and neurology ward (after they were subsequently discharged into the neurology ward). Estimated monthly income of the subjects was obtained by interview. Cost is presented in Naira and United States Dollars equivalent (1 US dollars = 457 Naira as of March 15, 2017). ### In-hospital care for stroke subjects Neuroimaging is mandatory for all patients with stroke in our hospital. Medications used included anti-hypertensives, anti-diabetics, anti-platelets, anticoagulants, and anti-lipids/ statins. The anticoagulant medications included intravenous unfractionated Heparin, Dabigatran, and Warfarin. The antiplatelet medications are Aspirin and Clopidogrel. Antidiabetic drugs included insulin and oral antidiabetic agents (Glibenclamide, Metformin). Anti-hypertensives included calcium channel blockers, angiotensin converting enzymes (ACE) inhibitors, angiotensin receptor blockers (ARB), beta-blockers (Atenolol, Labetalol), alpha-methyldopa, diuretics (Thiazides, Frusemide, Spironolactone), and alpha-adrenergic blockers. Statins/antilipids included (Atorvastatin, Fluvastatin). In addition, the subjects also underwent ancillary investigations including laboratory investigations (Serum Electrolyte/Urea/Creatinine assay, fasting blood glucose, complete blood count, fasting serum lipid profile, erythrocyte sedimentation rate, coagulation profile, abdominopelvic ultrasound, chest x-ray, lower limb Doppler, carotid Doppler, and cardiac studies (echocardiography, electrocardiography). In-hospital physical rehabilitation was individualized and provided by the Neuro-physiotherapists following a consult to them by the Neurologists. The stroke team had 2-weekly review meetings to minimize variations in patients’ care and discharge plans. ### Hospital costs Costs for intensive care unit (ICU), neurology ward, neuroimaging, abdominopelvic ultrasound, chest radiography, lower limb Doppler, physiotherapy, meals, echocardiography, electrocardiography, laboratory investigations (Serum Electrolyte/Urea/Creatinine, Fasting blood sugar, Complete blood count, Fasting serum lipid profile, ESR, and coagulation profile) are officially fixed by the hospital management. Cost of ICU admission, neuroimaging, physiotherapy, meals and echocardiography were not covered by the NHIS package while all other itemized categorized costs in this study were covered. For the costs covered by NHIS, the patients pay a tenth of the cost for drugs while ward charges, abdominopelvic ultrasound, CXR, lower limb Doppler, and laboratory investigations were totally covered by the NHIS package. ### Data analysis Statistical analysis was done using the Statistical Package for Social Sciences (IBM Corp., Armonk, NY, USA) software version 20.0 for windows. Categorical data were presented as frequencies and percentages while continuous data were presented as median, minimum, and maximum values. Kolmogorov-Smirnov’s test of normality was conducted on the cost variables and found to be skewed positively. Chi square test was used to compare the proportions of subjects with different characteristics who paid out of pocket and those on NHIS. As the cost data were highly skewed positively, Mann Whitney U test was used for bivariate analysis to explore the presence of any significant differences in cost categories among different variable groups. Spearman correlation was used to evaluate the relationship between categorized hospitalization cost and number of risk factors. The level of statistical significance was set at *p*≤0.05. ## Results The records of 170 stroke subjects with mean age 61.41±13.92 years were reviewed. There were almost as many females (48.2%) as there were males (51.8%). The ratio of subjects with ischemic stroke to those with haemorrhagic stroke was 1.5:1 with about 85.9% of all subjects being discharged home after their first admission for stroke. The modal number of risk factors in our subjects was one (for example, systemic hypertension). Estimated monthly income was less than $100 in 117 (68.8%) subjects. Out of the 170 subjects, 122 (71.8%) paid out of pocket (OOP) compared to the 48 (28.2%) on National Health Insurance Scheme (NHIS). Eighty-seven (71.3%) of these OOP subjects earn less than a 100 dollars per month as compared to 30 (62.5%) of 48 subjects on NHIS who earn the same (**Table 1**). View this table: [Table 1](http://nsj.org.sa/content/23/2/122/T1) Table 1 Subject characteristics among patients on National Health Insurance Scheme (NHIS) and those paying out of pocket (OOP). N=170 The median total hospitalization cost (HC) per patient was $183.30 (₦83,770) with a median daily cost of $15.86 (₦7, 248) (**Table 2**). View this table: [Table 2](http://nsj.org.sa/content/23/2/122/T2) Table 2 Direct medical cost for 170 patients (including insurance company inputs) Median cost of Radiological investigations ($85.45/₦39,050) was highest among the categorized itemized costs followed by laboratory investigations ($25.82/₦11,800) and feeding ($24.07/₦11,000), (**Table 2**). The median costs of the total HC and all categorised itemised costs were significantly higher (*p*<0.001) in the OOP group than the NHIS group except for feeding (*p*=0.983) and physiotherapy (*p*=0.795), (**Table 3**). Neuroimaging accounted for nearly all the cost of radiological investigations (100% of the cost of radiological investigations for NHIS and 99.8% for OOP subjects), (**Table 4**). Ischemic stroke subjects significantly paid less for radiological investigations than their haemorrhagic counterparts (*p*<0.001) for OOP subjects while the cost burden of radiological investigations was comparable (*p*=0.319) between the ischemic and haemorrhagic subjects for those on NHIS (**Table 5**). View this table: [Table 3](http://nsj.org.sa/content/23/2/122/T3) Table 3 Comparing categorized hospital costs among patients on National Health Insurance Scheme (NHIS) and those paying out of pocket (OOP). N=17 View this table: [Table 4](http://nsj.org.sa/content/23/2/122/T4) Table 4 Comparing costs of neuroimaging with other radiological investigations in US $. N=170 View this table: [Table 5](http://nsj.org.sa/content/23/2/122/T5) Table 5 Comparison of cost of radiologic investigations in US$ across stroke subtypes. N=17 The number of risk factors had very weakly positive but significant correlation with cost of drugs when both groups are combined (r=0.166, *p*=0.030) and in the OOP group (r=0.270, *p*=0.003). Similar result was obtained when number of risk factors were compared with cardiac studies in the combined group (r=0.218, *p*=0.04) and the OOP group (r=0.219, *p*=0.015). Other itemized categorized costs showed no significant association with number of risk factors in the combined group and among those paying out of pocket. A comparison of our findings to previous cost reports is shown in **Table 6**. View this table: [Table 6](http://nsj.org.sa/content/23/2/122/T6) Table 6 International comparison of first-time stroke costs. ## Discussion This study analysed the percentage contributions of categorized care costs, particularly neuroimaging, to the total hospitalization cost and evaluated differences in hospitalization costs between patients paying out of pocket and on NHIS during a single admission for first-ever stroke. Although a direct comparison is not possible due to varying degrees of heterogeneity of the studies, total and daily hospitalization costs of stroke in our study are lower than those reported in previous studies.11-26 This may be attributable to differences in available medicare resources, unit cost of hospital charges, reimbursement systems and healthcare delivery utilization pattern in the different countries. The total and daily costs in Nigeria appear far less than those in these other places. The relatively less sophisticated health care facilities available for stroke management in Nigeria may be partly responsible for the lower cost figures obtained in our study. Furthermore, another contributory factor may be the weakness of our local currency (Naira) relative to the US dollar (having lost more than 100% of its value over the last 2 years) and thus lower US dollar values are obtained on conversion. Chang et al26 studied the cost to provide acute care to 360 first-ever ischemic stroke patients in Taiwan, a more developed country than Nigeria. Hospital charges were used for cost analysis. In their study, the median cost per patient was $841 and the median cost per day was $121. In the study by Gnonlonfoun et al23 in Benin, a developing country, the average expenses were $1030.1±$101.7. A direct comparison of our study with previous studies on the economic costs of stroke is difficult because the studies differ in cost categories evaluated, number of patients assessed, study setting, and unit costs of treatment procedures. Decades ago, the diagnosis of stroke was largely dependent on bedside clinical methods. The knowledge of the location and nature of the underlying pathologic process could only be verified independently post-mortem which seriously hampered initial attempts at developing and implementing appropriate therapeutic interventions.27 There has been better understanding of the cerebrovascular and tissue pathology with neuroimaging techniques like Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound (USS) and Catheter angiography in the past 30 years which has created a wide array of opportunities for acute treatment and secondary prevention of stroke.27 Neuroimaging (47%) had the highest contribution to total hospitalization cost in our study followed by laboratory investigations and physiotherapy (14% each). This is in contrast to Japanese (69%), Greek (60%) and American (50%) studies where beds and staffs had the highest contributions to total hospitalization cost.13,14,30 In China, drugs (49%) had the highest contribution to total HC while it was room and board in Korea (31%).15,17 This variation in the highest contributing categorized cost to total HC may be due to regional differences in the unit cost of hospital charges among the studies. Cost of imaging (which ranked highest in this study) accounted for 6% of the total HC in Japan, 10% in Greece, 19% the United States of America, 9% in China and 30% in South Korea.13-15,17,30 These percentage contributions of imaging costs are lower than that in our study. Although total hospitalization costs (HCs) per stroke subject in our study is lower than those of most previous studies, imaging accounted for a much higher percentage of the total HCs. This may be due to differences in the unit hospital care charges and reimbursement systems as neuroimaging cost is being borne solely by the stroke subjects in our country unlike what may obtain in other countries. In this study, neuroimaging costs (CT or MRI) accounted for 100% of the cost burden of radiological investigations in stroke subjects on NHIS and accounted for about 99.8% of the cost of radiological investigations in those that paid out of pocket. The NHIS and OOP groups bore similar cost on neuro-imaging, but the NHIS group bore no cost on other radiological non-neuro-imaging studies, resulting in a significant difference in cost of radiological investigations between the 2 groups. Our findings showed that the OOP subjects had higher total and categorized costs, than the NHIS group; except for physiotherapy and diet, which are excluded from the NHIS package. The 28% of our stroke subjects who were on NHIS were therefore equally burdened with the cost of neuroimaging, physiotherapy and diet as the OOP subjects. The cost of radiological investigations was higher in haemorrhagic stroke patients (especially for those on NHIS) compared to their ischemic counterparts. This is similar to the findings of Gioldasis et al30 in Greece. In this study, cost of cardiac studies and drugs correlated weakly with number of risk factors. Therefore, effort to decrease the development of these risk factors in our environment may contribute to a reduction in these 2 categorized hospitalization costs. Despite being a single hospital-based analysis, this study provides data on the short-term inpatient care and costs for first-ever stroke managed in a public tertiary hospital in Nigeria which could help guide policy making on stroke financing. A future multicentre study will attempt to evaluate the general cost of acute stroke nationwide. In addition, because stroke patients are sometimes permanently dependent after the acute phase, a cost analysis of long-term care in our environment would be necessary. The main limitation of our study is that it evaluated only direct costs. Though direct costs are useful for planning healthcare funding, indirect cost estimation would further strengthen the justification for public spending on stroke care. In conclusion, the financial burden of neuroimaging is high during one-time admission of patients with first-ever stroke in our environment. Due to the high contribution of neuroimaging to the cost of radiological investigations and subsequently total HC, expansion of NHIS coverage to more people and also to include defraying neuroimaging costs will help to reduce the financial burden of stroke on patients. ## Footnotes * **Disclosure.** Authors have no conflict of interests, and the work was not supported or funded by any drug company. * Received August 27, 2017. * Accepted February 14, 2018. * Copyright: © Neurosciences Neurosciences is an Open Access journal and articles published are distributed under the terms of the Creative Commons Attribution-NonCommercial License (CC BY-NC). Readers may copy, distribute, and display the work for non-commercial purposes with the proper citation of the original work. ## References 1. Bogousslavsky J, Paciaroni M (2009) The economics of treating stroke as an acute brain attack. BMC Medicine 7, 51. 2. Birabi BN, Oke KI, Dienye P, Okafor UC (2012) Cost burden of post stroke condition in Nigeria: A pilot study. Glob J Health Sci 4, 17–22. 3. Taylor TN, Davis PH, Torner JC, Holmes J, Meyer JW, Jacobson MF (1996) Lifetime Cost of stroke in the US. Stroke 27, 1459–1466. 4. Wahab KW (2008) The burden of stroke in Nigeria. Int J Stroke 3, 290–292. 5. Dobkin B (1995) The economic impact of stroke. Neurology 45, S6–S9. 6. Campbell PC, Korie PC, Nnaji FC (2014) Risk management assessment of health maintenance organisations participating in the National Health Insurance Scheme. Niger Med J 55, 399–405. 7. Odeyemi I, Nixon J (2013) Assessing equity in health care through the national health insurance schemes of Nigeria and Ghana: a review-based comparative analysis. Int J Equity Health 12, 9. 8. By NBS estimates, Nigeria's population now 193.3 million, Available from: [http://tribuneonlineng.com/nbs-estimates-nigerias-population-now-193-3-million/](http://tribuneonlineng.com/nbs-estimates-nigerias-population-now-193-3-million/). 9. Odeyemi IA (2014) Community-based health insurance programmes and the national health insurance scheme of Nigeria: challenges to uptake and integration. Int J Equity Health 13, 20. 10. Olatunya OS, Ogundare EO, Fadare JO, Oluwayemi IO, Agaja OT, Adeyefa BS, et al. (2015) The financial burden of sickle cell disease on households in Ekiti, Southwest Nigeria. Clinico-Economics and outcomes research. Clinicoecon Outcomes Res 7, 545–553. 11. Akume A, Abdullah Y (2013) Challenges and prospects of effective industrial conflict resolution in Nigeria. J Soc Sci 36, 199–208. 12. Venketasubramanian N, Yin A (2000) Hospital costs for stroke care in Singapore. Cerebrovasc Dis 10, 320–326. 13. Yoneda Y, Uehara T, Yamasaki H, Kita Y, Tabuchi M, Mori E (2003) Hospital-based study of the care and cost of acute ischaemic stroke in Japan. Stroke 34, 718–724. 14. Diringer MN, Edwards DF, Mattson DT, Akins PT, Sheedy CW, Hsu CY, et al. (1999) Predictors of acute hospital costs for treatment of ischemic stroke in an academic center. Stroke 30, 724–728. 15. Ma Y, Liu Y, Fu HM, Wang XM, Wu BH, Wang SX, et al. (2010) Evaluation of admission characteristics, hospital length of stay and costs for cerebral infarction in a medium-sized city in China. Eur J Neurol 17, 1270–1276. 16. Gombet TR, Ellenga-Mbolla BF, Ikama MS, Ekoba J, Kimbally-Kaky G (2009) Cost of emergency cardiovascular care at the University Hospital Center in Brazzaville, Congo. Med Trop (Mars) 69, 45–47. 17. Yoon SS, Chang H, Kwon YD (2012) Itemized hospital charges for acute cerebral infarction patients influenced by severity in an academic medical center in Korea. J Clin Neurol 8, 58–64. 18. Maredza M, Chola L (2015) Economic burden of stroke in a rural South African setting. eNeurologicalSci 3, 26–32. 19. Kabadi GS, Walker R, Donaldson C, Shackley P (2013) The cost of treating stroke in urban and rural Tanzania: a 6-month pilot study. African Journal of Neurological Sciences 32, 45–53. 20. Guinhouya KM, Tall A, Kombate D, Kumako V, Apetse K, Bello M, et al. (2010) Cost of stroke in Lome (Togo). Sante. 21. Chang H, Yoon SS, Kwon YD (2013) Categorized hospitalized charges of acute ischaemic stroke according to trial of Org 10172 in acute stroke treatment classification. Neurol India 61, 633–638. 22. Toure K, Ndiaye NM, Sene Diouf F, Ndaiye M, Diallo A.K, Ndao AK, et al. (2005) Evaluation of the cost of stroke management in Dakar, Senegal. Med Trop (Mars) 65, 458–464. 23. Gnonlonfoun DD, Adoukonou T, Adjien C, Nkouei E, Houiato D, Avode DG, et al. (2013) Factors associated with stroke direct cost in francophone West Africa, Benin example. World J Neurosci 3, 287–292. 24. Mamoli A, Censori B, Casto L, Sileo C, Cesana B, Camerlingo M (1999) An analysis of the costs of ischemic stroke in an Italian stroke unit. Neurology 53, 112–116. 25. Tu F, Anan M, Kiyohara Y, Okada Y, Nobutomo K (2003) Analysis of hospital charges for ischemic stroke in Fukuoka, Japan. Health Policy 66, 239–246. 26. Chang KC, Tseng MC (2003) Costs of acute care of first-ever ischemic stroke in Taiwan. Stroke 34, e219–e221. 27. Xavier AR, Qureshi A, Kirmani JF, Yahia AM, Bakshi R (2003) Neuroimaging of stroke: A review. Southern Medical Journal 96, 367–379. 28. Wardlaw J, Farrall A (2004) Diagnosis of stroke on neuroimaging:“Scan all immediately“strategy improves outcomes and reduces costs. BMJ 328, 655. 29. Rice DP (2000) Cost of illness studies: what is good about them? Inj Prev 6, 177–179. 30. Gioldasis G, Talelli P, Chroni E, Daouli J, Papapetropoulos T, Ellul J (2008) In-hospital direct cost of acute ischemic and haemorrhagic stroke in Greece. Acta Neurol Scand 118, 268–274.