Cost analysis of a primary health care centre in Bangladesh

Bulletin ofthe WorldHealth Organization, 61(3): 477-483 (1983) i World Health Organization 1983 Cost analysis of a primary health care centre in Bangladesh A. M. ZAKIR HUSSAIN' This report describes the first-ever cost analysis of a primary health care centre in Bangladesh. The aim was to provide information on costs incurred on various health care activities available in a thana health complex, in relation to the number ofpatients and the intensity of use of services. Overall costs were estimated by examining records and papers and by interviewing staff. Cost apportionment was carried out in three stages. In the first stage, all direct expenditure was calculatedfor each activity, e.g., wards, outpatient department, tubercu- losis services, etc. In the second stage, general service costs, covering water, sanitation, security, and administration were dividedamong the various services. In the thirdstage, the costs of laboratory services were distributed according to the number of examinations carried outfor each service. The capital costs for 1979 were US$ 36 382, of which 84.6% were incurred on the buildings. Recurrent costsfor 1979 were US$ S9 556, i.e., 62% of overall costs, reflecting the labour intensiveness of the health complex. The distribution of the costs among the various health care activities was detailed. It wasfound that, in general, the costper unit of activity depended mainly on the intensity ofuse ofthe resources, e.g., unit costs incurred in the outpatient department, maternal and child health services, and subcentres were relatively low because of the high rate of utilization of services. This study has demonstrated that it is possible to estimate unit costs in a thana health complex. It is hoped that the methods used here will provide a basisforfurther work ofa similar nature. A study was carried out in 1979 to ascertain the costsa involved in providing different types of services in a primary health care centre in Bangladesh. The aims of the study were: 1. to assist the health administrators in controlling the costs and expendituresb in the health complex; 2. to assist in formulating the health complex budget on the basis of performance; 3. to contribute to decision-making about the health complex programmes and services in the light of the costs involved for each; 4. to provide a basis for cost-benefit, cost- effectiveness, and efficiency analysis; 5. to assist donor agencies in understanding the trend of costs and expenditure in the health complex system. A thana is the smallest governmental adminis- trative unit in Bangladesh, with an approximate ' Assistant Professor, Department of Community Medicine, National Institute of Preventive and Social Medicine, Mohakhali, Dhaka 12, Bangladesh. a Costs refer to the total value of the resources used to produce certain services or goods, and not just to hard cash. b Expenditure is the amount of cash spent to provide certain services; it may be more or less than the costs involved. population of 200 000-300 000. There are 472 thanas in the country, and it is planned to have a health complex in each of the 356 rural thanas by 1985. At present, 312 of these are operating. Each thana health complex has several subcentres, and it is hoped that by 1985 there will be one subcentre in each union (a union has an approximate population of 15 000- 25 000). These subcentres will be known as family welfare centres and there will be 12-20 subcentres attached to each thana health complex. They will pro- vide outpatient consultancy by medical assistants and public health nurses, maternal and child health (MCH) care, family planning consultancy, and a dispensary. Each health complex has a pathological laboratory, a theatre for emergency surgery, inpatient facilities with 31 beds, outpatient clinic with 5 medical officers, maternal and child health and family planning ser- vices, and home visiting service. All these services are controlled by an administrative section. This study is the first of its kind in Bangladesh. The health complex studied is a prototype, and since each health complex has the same annual budget and all cover approximately equal populations, the results 4303 -477 A. M. ZAKIR HUSSAIN complexes in Bangladesh. MATERIALS AND METHODS The study examined two types of costs: (1) capital items (buildings, vehicles, furniture, and equipment), and (2) recurrent costs (food for the patients, petrol and oil for the vehicles, maintenance and repair costs, salaries and allowances of the staff, medicines, stationery, and general administrative expenses). Records were examined, including work state- ments, account books, inventories, log books of the vehicles, price lists for medicines and equipment, attendance registers, and stock cards of medical supplies. Data collection Each subcentre was visited in order to assess the cost of items belonging to each site and the type of activities carried out there. Records were examined to ascertain the capital and recurrent costs at each site of the health complex, and the market value of all furni- ture was estimated. Staff of the health complex were interviewed to assess expenditure on salaries and allowances and sample studies were carried out to assess work statements, and types of activities per- formed for which there was no reliable record. Classification of sites and activities The activities and associated sites of the health complex were divided into three groups: (a) general services, including maintenance, water and sanita- tion, security, and administration, (b) intermediate services, such as laboratory facilities, and (c) final services (sites) covering wards, the outpatient depart- ment, the emergency department, the home-visiting service, the maternal and child health service, tuberculosis control, and the subcentres. Each activity is carried out mainly on one site, although it may also be partly associated with other sites, e.g., tuberculosis control activities are carried out mainly in the outpatient department but patients may also be admitted to the wards. All the general and intermediate activities are in support of the final services. Description of the activities by site Emergency: activities consist of consultation, dis- pensing, dressing of wounds, and minor operations. Laboratory: simple pathological examinations are carried out, e.g., tests of sputum for acid-fast bacilli, microscopy of stools and urine, blood cell counts, 478 erythrocyte sedimentation rate, examination of blood slides for malaria parasites, haemoglobin estimation, and urine tests for sugar and albumin. These are all support activities. Ward: these services were reckoned in terms of patient-days. Outpatient department: consultations, dispensing of medicines, and injections. Domiciliary services: home visits during which blood slides are collected and immunizations are carried out. MCH services: consultations, immunization, and dispensing. Tuberculosis control services: consultations and dispensing of medicines. Subcentres: consultations and dispensing of medi- cines. Administration: supports the intermediate and final services, and consists of registration and issuing of notices, salaries, reports, etc. Basic services: for support in other ways, e.g., pro- viding water supplies (deep tubewell with petrol- driven pump), sanitation (septic tank), maintenance (plumber/electrician) and security (night guards). Costing Depreciation of capital assets was estimated in order to calculate annual capital costs. A 10% interest rate was assumed on the capital items to account for the interest paid against foreign loans and for the opportunity cost, i.e., the revenue foregone by the government in financing a non-profitable organiz- ation. The capital depreciation and interest are shown in this paper as the cost of the capital items for 1979. Costs were accounted ultimately against the com- ponents of activities so as to determine the unit.cost for each activity. Costs per patient were accounted on the basis of the number of new patients treated in each site. Costs were accounted in US dollars; the official exchange rate at the time of this study was 15 Bangla- deshi taka to the dollar. Cost apportionment Costs were apportioned in three stages. In the first stage all direct expenditure, such as salaries and supplies, was calculated for each site. Capital costs were distributed among the sites, using the following criteria: - costs of service buildings, according to the area utilized for each activity; COST ANALYSIS OF HEALTH CENTRE IN BANGLADESH - costs of staff lodgings, according to the area occupied by the resident staff working in each site; - cost of equipment and furniture, according to the site/activity in which they were employed; - vehicle costs, according to the mileage attri- butable to each site/activity. In the second stage, general service costs were divided among the intermediate and final services. The costs of basic services such as water, sanitation, maintenance, and security were allocated according to the building space occupied by each service site. The administrative costs were distributed according to the amount of paper work produced for each site. In the third stage, the cost of the laboratory services were distributed according to the number of examinations requested by each service. Also, a part of the costs of the outpatient department were allo- cated to MCH, tuberculosis control, and the wards, according to the services that had been provided for each one of these activities. In this way, the distribution of costs was carried to such an extent that no further subdivision was possible. The costs obtained in the third stage are thus the total costs for each final site and its specific activity. Costs incurred on the training of staff were not accounted in this study, since adequate data were not available. Costs associated with visits of supervisory staff from the district or national level were not included as the visits were infrequent. Unit costs for each type of service were obtained by dividing the total annual costs incurred on the associated site by the number of components of the activity carried out there. Costs per patient were obtained by dividing the total annual costs incurred on a site by the total number of new patients treated in the year. Costs per inhabitant were obtained by dividing the total costs incurred in the thana health complex by the total population of the thana in 1979. RESULTS Since 1964, US$ 227 993 have been spent in devel- oping the capital assets of the Dhamrai thana health complex (Table 1); of this amount, US$ 13 288 came from UNICEF in the form of equipment and vehicles, while US$ 6693 came from public donations which were used to help establish and equip some sub- centres. These sums represent, respectively, 5.8%o and 2.907o of the assets of the health complex. The capital costs for 1979 were US$ 36 382 which covered depreciation and a 10%o interest on the invest- ments. Of this, 84.6% was incurred on the buildings, Table 1. Capital assets of Dhamrai 1979 health complex, Asset Value (US$) Source Buildings Health complex 196 562 Subcentres 6 693 Public donation Subtotal 203 255 Vehicles Bus 6 000 Motorcycle 753 Subtotal 6 753 UNICEF Furniture 667 Equipment 10 783 National agencies 6 535 UNICEF Subtotal 17 318 Total 227 993 while the cost of the furniture was nominal, since most of the furniture had little or no market value at the time of this study (Table 2). The recurrent costs for 1979 were US$ 59 556, i.e., 62%o of the overall costs. This indirectly indicates the labour intensiveness of the health complex. In fact, 68.27o of the recurrent costs were incurred on the salaries and allowances of the staff. The next highest amount was spent on medicines, which comprised 16.4% of the recurrent costs for 1979. Of the total annual cost of US$ 96 000 incurred for the health complex, 35.8% was spent on buildings, 42.3% on salaries and allowances, and 10.2qo on medicines. The distribution of costs among the sites after the first stage of cost apportionment is shown in Table 3. The highest proportion was spent on the domiciliary service, the wards, and the administration. Table 4 shows the distribution of costs after the second and third stages of cost apportionment. At the second stage, the administrative costs were redistrib- uted to the appropriate sites; at the third stage, the laboratory costs were redistributed and certain final activities were allocated to the proper sites, e.g., all MCH activities, wherever they were carried out, are shown under MCH. At the final stage of the cost analysis, the total cost of the MCH service was second only to the domiciliary service, whereas at the first stage, MCH occupied sixth place. The final estimate of unit costs shows that one patient-day in Dhamrai cost US$ 4.88 in 1979; one consultation cost US$ 0.16 in the outpatient depart- ment, but US$ 1.16 in the emergency unit, US$ 0.09 479 A. M. ZAKIR HUSSAIN Table 2. Distribution of capital and recurrent costs for Dhamrai health complex, 1979 Capital costs Recurrent costs Total costs Item US$ % US$ % US$ % Buildings 30 784 84.6 3 564 6.0 34 348 35.8 Vehicles 2 126 5.8 1 544 2.6 3 670 3.8 Furniture 128 0.4 176 0.3 304 0.3 Equipment 3 344 9.2 0 0 3 344 3.5 Medicine 9 787 16.4 9 787 10.2 Food and laundry 2 100 3.5 2 100 2.2 Salary 38 837 65.2 38 837 40.5 Contingencies 1 757 3.0 1 757 1.8 Staff allowances 1 791 3.0 1 791 1.9 Total 36 382 100 59 556 100.0 95 938 100.0 Table 3. Total and unit costs, in all sites of the Dhamrai health complex, after the first stage of cost distribution' Unit of No. of Total cost Unit cost Site activity activities (US$) (US$) Wards Patient-day 4 355 16 314 3.75 Outpatient Consultations, 94 686 11 559 0.12 department dispensing Emergency service Consultations, 2 576 2 680 1.04 dispensing, dressings, or minor operations Laboratory Tests 4 836 2 778 0.57 Domiciliary Home visits 428 904 23 420 0.05 service Tuberculosis Consultations, 1 320 2 248 1.70 control dispensing Maternal and Consultations 21 587 7 507 0.35 child health dispensing, and service immunization Subcentres Consultations, 149 004 11 653 0.08 dispensing Administrationb Paper work 2 003 14 239 7.10 a These costs do not include those incurred for night guards and a plumber, septic tank and deep tubewell installation. b Administrative costs include the cost of the health complex store, and costs incurred on the vehicles when used for non-specific purposes. in the subcentres, and around US$ 0.04 for MCH services. Home visits, excluding MCH activities, cost US$ 0.06 each, and consultations for tuberculosis control cost US$ 2.65 each. Table 5 shows the cost per patient for each type of service. Thus, each hospitalization cost US$ 24.41, each tuberculosis patient cost US$ 31.75, and each new patient seen in the outpatient department cost US$ 0.37. It is possible that the average cost per hospital admission might be reduced if the occupancy of beds (38%o in 1979) were to increase. Finally, Table 6 shows that the annual cost of the health complex activities per head of population in the thana came to US$ 0.40. Of this, 90Gb was paid by the Government of Bangladesh, 7.5%o by UNICEF, and 2.5% by public donations. The per capita cost of US$ 0.40 should be seen in the light of the use of services by the population. In 1979, there were 1.1 consultations and 1.8 home visits per capita, 3.7 hospital admissions per 1000 popu- lation, and 0.5 tuberculosis control consultation per 1000 population. 480 COST ANALYSIS OF HEALTH CENTRE IN BANGLADESH Table 4. Total and unit costs, in all sites of the Dhamrai health complex, after the second and third stages of cost apportionment Second stage of apportionment Third stage of apportionment Site No. of Total cost Unit cost No. of Total cost Unit cost activities (US$) (US$) activities (US$) (US$) Wards 4 355 20 224 4.64 3 955 19 286 4.88 Outpatient 94 686 13 826 0.15 82 501 12 974 0.16 department Emergency 2 576 2 981 1.16 2 334 2 701 1.16 service Laboratory 4 836 3 500 0.72 Domiciliary 428 904 30 360 0.07 428 904 25 560 0.06 service Tuberculosis 1 320 3 248 2.46 1 320 3 493 2.65 control MCH service 21 587 8 588 0.04 35 878 20 939 0.58 Subcentres 149 004 13 213 0.09 124 188 10 985 0.09 Table 5. Cost per patient of the final activities of the Dhamrai health complex, in 1979 Cost per Total cost No. of patient Service (US$) patients (US$) Wards 19 286 790 24.41 Outpatient 12 974 34 705" 0.37 department Emergency 2 701 946 2.86 service Domiciliary 25 560 53 354b 0.48 service Tuberculosis 3 493 1lo0 31.75 control MCH service 20 939 35 878 0.58 Subcentres 10 985 51 024b 0.22 aNew patients only. b This figure covers only patients who received immunization, or who gave a blood sample. Other domiciliary services were not accounted for, since they are considered to be negligible in terms of the time required. Table 6. Per capita costs of the Dhamrai health complex and source of finance, 19790 Source of finance Total Govern- UNICEF Public ment donation Total cost 86 355 8 284 1 300 95 939 Cost per 0.36 0.03 0.01 0.40 capita a Total population of Dhamrai thona in 1979 was 237 109. DISCUSSION Literature on cost analysis of health complexes and services is scarce and is often concerned with specific problems rather than overall considerations (1-5). However, several workers have provided useful guidelines for a general cost-benefit analysis of health services (6-8). The present paper has outlined the costs involved in providing services in a primary health care centre in rural Bangladesh. Government efforts to provide health services to the rural population started in 1960. By 1970, there were 140 rural health centres (RHC), each with 6 beds, including 2 for MCH services (9). After independence in 1972, a scheme was drawn up to establish one health complex in each of the rural thanas of Bangladesh by 1985, with 3 subcentres attached to each complex; this was later revised in favour of establishing one subcentre in each union. Some of the complexes were to be converted from the existing rural health centres, of which there were then 150. At present, 312 health complexes are working, with 31 beds in each, including 6 forMCH care. These health complexes offer comprehensive health ser- vices, as well as referral to better equipped facilities. The budgetary provision for each thana health complex (including 3 subcentres) in 1974-75 was approximately US$ 356 000. Of this, US$ 201 230 were earmarked for extension of each existing rural health centre (9). The Dhamrai health facilities were built in 1964, as an RHC with 3 subcentres, at a cost of approximately US$ 41 000. In our study we found that only US$ 162 000 were spent on extending this to a health complex in 1974-75. Presumably the 481 A. M. ZAKIR HUSSAIN budgeted figure was an average one for the whole period covered by the first 5-year plan (1974-78). Naturally, towards the end of the period, costs would have been higher than during the earlier years, because of inflation. For example, in 1974-75, it was estimated that the establishment of a subcentre would cost approximately US$ 8000, while in 1981-82 the actual cost was US$ 26 500. In fact, the estimated national Annual Development Programme (ADP) budget for the health sector, which was formulated in 1976-77, had to be revised less than 12 months later from US$ 21 million to US$ 25 million. In 1978, the total budget was approximately US$ 29 million. In the same way, the recurrent budget increased from US$ 13 million to US$ 18 million within two years. Of this total, US$ 16 million were to be spent on direct services to the patients and the remainder (11%) was for administration and training (10). In our study, it was found that about 14%o of the total recurrent costs were incurred on administration. Of the US$ 16 million allocated for direct services, about 8% and 5% were to be spent on tuberculosis control and MCH activities, respectively, in 1976-77. In our study, the respective proportions were 307 and 7%o, if the general costs are considered. This difference is possibly because of differences in costing methods or the variety of institutes included in the national study. The ADP budget of 1976-77 envisaged US$ 29 904 for each of the 356 thana health complexes, including a foreign exchange component of US$ 5618 (1). In the present study, the corresponding sum was found to be US$ 31 255 (including the cost of equipment), which was obtained from the Government. The budgeted recurrent costs and expenditures for each health complex in 1976-77 were US$ 5165, US$ 1736, US$ 927, and US$ 862 for salaries, allowances, contingencies, and medicine/equipment, respectively (1). However, in Dhamrai, the actual costs in 1979 were found to be US$ 38 837, US$ 1791, US$ 1757, respectively, for salaries, allowances, and contin- gencies, and US$ 5204 for medicines with no expendi- ture on equipment. The reasons for these huge discrepancies are not known but may be related to differences in the programme budgets for 1976 and 1979. As has already been stated, this study is the first of its kind in Bangladesh, so it may be worth while to consider its potential use within the national health information system. This study has shown that the unit costs of services provided in a thana health com- plex can be estimated. The method of cost analysis used is easily reproducible, at least up to the first stage of allocation (10). Procedures for further distribution of the costs may be formulated. While the distri- bution of administrative costs may be carried out in several ways, it is believed that the method followed in this paper is easily replicable. It is hoped that this study will provide a basis for further work of a similar nature. The results obtained in studies of this sort will provide useful information on the cost of various services in an organization. This type of cost information should provide a sound background for the formulation of performance budgets. This paper has also indicated the utilization pattern and volume of different services provided in a health complex. The results show that, in Bangladesh, the cost per unit of activity depends mainly on the inten- sity of use of the resources. It seems that this sort of cost information can pro- vide a good basis for efficiency analysis. Compari- sons between complexes can thus be made on a sound basis. This method of cost analysis can also be used in analysing activities in the private health sector. Estimates of private expenditure on health should be linked with results such as those presented here, so that a complete picture of costs incurred on a particular service can be obtained. According to various sources (9, 11), Government expenditure on health services in Bangladesh is about US$ 0.54 per capita per year. This figure was arrived at by dividing the total national health budget by the population of the country. However, the greater share of the national health budget is allocated to institutions in urban areas, e.g., of the total ADP budget in 1976-77, only 39% was reserved for thana health complexes, the only form of health care available in rural areas, while the remainder was spent on hospitals and teaching institutes (9). Only 17% of the recurrent budget for that year went to the thana complexes. ACKNOWLEDGEMENTS The author is grateful to the staff of UNICEF, Bangladesh, Construction and Buildings Division VI of the Public Works Department of the Government of Bangladesh, the Central Medical Store, Bangladesh, Office of the Civil Surgeon, Dhaka, and the thana health complex, Dhamrai, for their unreserved cooperation. Thanks are due to Professor M. Ali, Director of the National Institute of Preventive and Social Medicine and to Professor C. McCord, Professor of Cardiology, for their advice during the study. The author also acknowledges the contribution of Mr Rahim and Mr Delwar of the National Institute of Preventive and Social Medicine, who typed the manuscript. 482 COST ANALYSIS OF HEALTH CENTRE IN BANGLADESH 483 RESUME ANALYSE DES COUTS D'UN CENTRE DE SOINS DE SANTE PRIMAIRES AU BANGLADESH Ce rapport decrit la premiere analyse des cofts d'un centre de soins de sante primaires au Bangladesh. L'objectif etait de fournir des renseignements sur les couits des diverses activites de soins de sante exercees dans un complexe sani- taire de thana (la plus petite unite administrative) en fonc- tion du nombre de malades et de l'intensite de l'utilisation des services. On a estime les couits globaux en examinant des dossiers et des documents et en interrogeant le personnel. La venti- lation des couts a ete effectuee en trois etapes. Au cours de la premiere, la totalite des depenses directes a et calculee pour chaque activite, par exemple salles de malades hospitalises, consultation externe, services de tuberculose, etc. Au cours de la deuxieme etape, les couts des services generaux, couvrant l'eau, l'assainissement, la securite et l'adminis- tration, ont et divises entre les differents services. Dans la troisieme etape, les couits des services de laboratoire ont et ventiles entre les services selon le nombre d'examens effec- tues pour chacun. Les depenses d'investissement pour 1979 ont ete de US$ 36 382, dont 84,61o destines aux batiments. Les de- penses de fonctionnement pour 1979 ont et de US$ 59 556, c'est-a-dire 62% des cofits totaux, refletant l'intensite du travail du complexe sanitaire. La ventilation des couits selon les diverses activites de soins de sante a e d6taill6e. On a observe qu'en general, le couDt par unite d'activit6 d6pendait principalement de l'intensite d'utilisation des ressources, par exemple les depenses unitaires encourues dans la consul- tation externe, les services de sante maternelle et infantile et les sous-centres etaient relativement basses en raison du taux eleve d'utilisation des services. Cette etude a demontre qu'il etait possible d'estimer les couits unitaires dans un complexe sanitaire de thana. On espere que les methodes utilisees ici serviront de base a de nouveaux travaux de meme nature. REFERENCES 1. HALL, T. L. Chile health manpower study: methods and problems. Internationaljournal of health services, 1: 166-184 (1971). 2. Functional analysis of health needs and services. Baltimore, Johns Hopkins Press, 1976. 3. MORLEY, D. Economic background of child care in developing countries. In: Paediatric priorities in the developing world. London, Butterworths and Co., 1973. 4. JOLLY, R. ET AL. The economy of a district hospital. In: King, M., ed., Medical care in developing countries, Nairobi, Oxford University Press, 1966. 5. WAGNER, R. Care of elderly: an exercise in cost-benefit analysis. London, 1972 (Commissioned by Essex County Council). 6. CULYER, A. J. & WRIGHT, K. G., ed. Economic aspects of health services. London, Martin Robertson, 1978. 7. WILLIAMS, A. H. The cost-benefit approach. British medical bulletin, 30: 252-256 (1974). 8. MISHAM, E. J. Cost-benefit analysis. London, George Allen and Unwin Ltd., 1975. 9. COUNTRY HEALTH PROGRAMMING, BANGLADESH. The information document. Dhaka, Bangladesh, Ministry of Health and Population Control, 1977. 10. Bangladesh health profile 1977. Dhaka, Health Infor- mation Unit, Health Division, 1977. 11. Health. Washington, DC, World Bank, 1975 (Sector Policy Paper).