- Email: [email protected]
Revolving drug funds: Conducting business in the public sector
Sot. Sci. Med. Vol. 22. No. 3. pp. 335-343, 1986 Britain. All rights reserved
Printed in Great
Copyright C
REVOLVING DRUG FUNDS: CONDUCTING BUSINESS IN THE PUBLIC
0277-9536.86S3.00+ 0.00 1986Pergamon Press Ltd
SECTOR*
PETER N. CROSS. MAGGIE A. HUFF, JONATHAN D. QUICK and JAMES A. BATES Management Sciences for Health, Drug Management Program. 165 Allandale Road, Boston. MA 02130. U.S.A. Abstract-Pharmaceuticals are essential for preventive and therapeutic health services. Unfortunately, significant demand, limited funds and high prices contribute to frequent shortages of drugs in many public health programs. One method for financing pharmaceutical supplies has been the establishment of revolving drug funds (RDFs) in which, after an initial capital investment, drug supplies are replenished with monies collected from the sale of drugs. All too often however, the funds actually recovered are insufficient to replenish supplies and the fund is soon depleted. In this paper we consider the potential benefits and common pitfalls of revolving drug funds and then focus on the central role of financial planning in establishing drug sales programs. Experiences from a variety of countries suggest several causes for the failure of some RDFs, including: (I) under-estimation of capitalization costs, (2) prices set below true replacement cost, frequent failure to collect payment, (3) delays in cash flow which make funds unavailable for replenishment of drug stocks, (4) rapid program expansion for which additional capital funds are not available, (5) losses due to theft and deterioration of drugs, (6) unanticipated price increases due to inflation or changes in parity rates and (7) foreign exchange purchase restrictions. Common to many of these problems is the lack of a businesslike orientation to RDFs and, in particular, lack of careful financial planning and management. Financial planning for an RDF includes four analytical tasks: (I) (2) (3) (4)
assessment of the potential market, estimation of the costs of an RDF, establishment of the cost-recovery objectives, definition of the role of subsidies and surcharges.
Revolving drug funds offer great potential for supplying drugs in many parts of the Third World, but they must be carefully planned and the organizers should bear in mind the pitfalls which commonly undermine these programs.
Pharmaceuticals-including vaccines, vitamins, antibiotics, oral rehydration solutions and other products-are essential for preventive and therapeutic health services. In addition to their direct health impact, the effectiveness of pharmaceuticals against many common diseases serves to establish the credibility of health workers that they need to promote long-term health improvements through environmental and nutritional changes. The importance and popularity of pharmaceuticals has led many governments in developing countries to espouse a policy of providing free drugs. Unfortunately, few countries have the resources to fully implement such a policy. Significant demand, limited funds and high prices contribute to frequent shortages. Thus the cost of ‘free’ drugs becomes a scarcity of drugs and disillusionment. Health workers are dispirited by trying to provide services without the resources they have been trained to use; the people are frustrated by receiving so much less than they *An earlier version of this paper was presented at the National Council for International Health Conference, Washington, D.C., June 1982. The authors would like to express their appreciation to Day1 Donaldson -and David Dunlop from the Boston University Health Management program and to Catherine Overholt, Margaret Hume, Ronald O’Connor and John Turnbull from Management Sciences for Health for their careful review and constructive critiques of this paper.
were promised. At the same time however, even the rural poor have been found willing to spend significant amounts of their limited cash resources on commercially available, but often highly priced, brand name drugs. One response to the challenge of financing drug supply has been the establishment of revolving drug funds (RDFs) in which, after an initial capital investment, drug supplies are replenished using monies collected from the sale of drugs. Revolving drug funds, which are one type of drug sales program or cost-recovery scheme, attempt to mobilize financial resources based on a demonstrated willingness of people to pay for health services. Unfortunately, the establishment of such funds frequently has been fraught with difficulty. All too often the monies actually recovered are insufficient to replenish the original drug stocks and the fund is soon depleted or ‘decapitalized’. Many factors contribute to the failure of revolving drug funds, one of the most important of which appears to be a resistance to thinking of the fund in business terms and, in particular, a lack of careful economic and financial analysis in planning the fund. Unlike a public sector entity that receives a budget allocation from the government and where management’s principal financial concern is simply staying within that budget, a revolving drug fund must sell a product and, through sales, generate sufficient revenues to meet its cost recovery objectives. In many respects, manage-
335
PETERN. IN
THEORY-EXPENSES
CROSS et a[.
FOR EACH
CYCLE
EQUAL
RECEIPTS
s,,,,iersf==== / o”:J,‘cc~~““a’ /
Procuremenf unit
/
Government medical srores
\
Heal~hi~l~~~~~
Fig. I. Revolving drug funds in theory. From: Management Sciences for Health: Manrzging Drug Supply: the Selection, Procurement. Distribution and ure of Pharmaceuticals in Primary Health Care. p. 23 I. Management Sciences for Health. Boston. Mass.. 1981.
merit’s concerns should reflect those found in private sector businesses-if the RDF is to survive and thrive. In this paper. the potential benefits and common pitfalls of revolving drug funds are considered. Subsequently the paper addresses the central role of financial planning in establishing viable drug sales programs. The potential market, the capital and operating costs of a revolving drug fund, the fund’s cost-recovery objectives, and the role of subsidies and surcharges must all be considered. POTENTIAL REVOLVING
BEXEFITS OF DRUG FUNDS
Revolving drug funds are attractive because they are theoretically self-financing after a one-time capital investment by the community, the govemment, or an outside donor. As illustrated in Fig. 1, the supply of drugs can be continued indefinitely without further government budget allocations as long as the receipts from sales (or in some cases local health budgets) are sufficient to cover replacement costs. In comparison to public free-drug programs, revolving drug funds make use of an untapped financial resource: patients’ payment for drugs. The result of this can be a significant increase in the availability of pharmaceuticals and a corresponding increase in public participation in public health care services [I, 21. In addition, an RDF provides a set of incentives to the population to use pharmaceuticals more cost-effectively. For example, in a free-drug system it is not unusual for relatively healthy villagers going into town for other reasons to stop at a health center with minor complaints and with the aim of obtaining some drug supplies. The cost-recovery objectives of RDFs also force improvements in management information and accounting systems, in inventory control, and in other supply management activities. In comparison to commercial drug stores, publicly sponsored RDFs have several potential advantages. Through the selection of the most cost-effective therapeutic alternatives and competitive bulk procurement by generic name, cash and foreign exchange outlays can be used more efficiently. The same level
of funding can then achieve substantially greater health impact or government funds can be freed for other public health services. Finally, since an RDF can be structured to use both the Ministry of Health infrastructure and financing, it can serve remote markets with limited commercial appeal.
COhlMON
PITFALLS OF REVOLVING DRUG FUNDS
Revolving drug funds or modifications of revolving drug funds have been established in Peru [3]. Guatemala [4], India [4], Bolivia [5], Haiti [6], Senegal [7], Niger [7], Afghanistan [I]. Mali [S]. Indonesia [4], Thailand [9] and elsewhere [4, lo]. Establishment and/or maintenance of many of these programs has been fraught with difficulties. In Montero, Bolivia, for example, a regional revolving drug fund began selling drugs in August 1976, and expanded rapidly. By May 1979, drug supplies at the central level were depleted, there were insufficient funds to procure additional supplies, and the drug sales program was at a standstill [5]. Similarly, a plan in Gosses, Senegal. to have the vaccination program pay for itself failed when villagers proved unwilling to pay for the entire series of vaccinations and revenues proved insufficient to replenish health workers’ supplies [7l. To these two cases could be added other examples of revolving drug funds which have faced difficulties. All too often, RDFs fail to generate sufficient revenues to replenish supplies and, in effect. soon cease to ‘revolve’. In practice. the expenses for each cycle, including losses, can quickly exceed receipts (Fig. 2). Specific causes for loss of available funds or supplies include: rapid program expansion for which additional capital funds are not made available; under-estimation of the capitalization costs of the supply system; unanticipated losses of drugs through theft or deterioration; high operating costs which exceed budgeted amounts;
Revolving IN PRACTICE
- EXPENSES
FOR
EACH
drug funds CYCLE,
337
INCLUDING
LOSSES,
EXCEED
RECEIPTS:
drugs
Key:
w
_ _--
-
-
-c
Flow
of druq
Flow
of
funds
supphes
Loss
of
avpilpble
funds
or drugs
Fig. 2. Revolving drug funds in practice. From: Munagemenr Sciencesfor Health: Managing Drug SuppI!: the Seleciion. Procuremenr. Distribution and use of Pharmaceuricals in Primary Health Care. p. Xl. Management Sciences for Health, Boston. Mass., 1981.
prices set too low for intended level of cost recovery; failure to collect payment for drugs with no subsidy system; delays in collection of subisidies and other payments from government agencies; funds tied up in national banking system or Ministry accounting mechanisms; unanticipated price increases from inflation or changes in parity rates; foreign exchange limitations which restrict foreign purchases for resupply. Many of these problems are avoidable through careful financial planning and sound fund management. The impact of unavoidable problems often can be minimized by anticipating these problems through the planning process.
FINANCIAL
PLANNING
For planning and management purposes it is important that government officials and other individuals involved in establishing a revolving drug fund view the fund more as a commercial operation than a public service. Programming and administrative concepts and procedures familiar to policy-makers and administrators in the public sector will prove insufficient to create a viable revolving fund. Although the aim of an RDF is to provide a needed public service, rather than to turn a profit, its planning and implementation should be approached with a strong business perspective. Financial planning for a revolving drug fund includes four analytical tasks: (1) assessement of the potential market, (2) estimation of the costs of an RDF, (3) establishment of cost-recovery objectives
and (4) definition surcharges. Potential
of the role of subisidies
and
market
The potential market for a drug sales program is determined by the household or consumer demand for pharmaceuticals from public as well as commercial suppliers. The market potential for an RDF is influenced by households’ ability and willingness to pay, the extent of public pharmaceutical supply capability, the prescribing practices of health workers, the availability and accessibility of alternative commercial pharmaceutical outlets, and the prices of commercially available pharmaceutical products. Each of these factors must be assessed during the planning for an RDF. Consumers’ ability and willingness to pay for health care has been established by surveys of the so-called ‘rural poor’ in several developing countries. These surveys have demonstrated that people already spend considerable amounts of their limited cash resources on health care. In Afghanistan, families were spending an average of USS15.80 per year or 6.8% of the annual household income on health care services. Almost one-half of this amount went to drugs purchased from local pharmacies or ‘dokans’ [l I]. In a survey in Yemen Arab Republic, families were found to be spending about USS20.00 per person per year on treatments, including drugs [12]. In Honduras, a 1983 Ministry of Health household survey [13] found that expenditures associated with family illness represented the third largest expense category after food and clothing. Though most Ministry of Health services are provided without fee, 93% of the households interviewed expressed willingness to pay a small fee for services and medications. A ‘more reliable supply of medicines’ was the most commonly recommended area for improvement of
338
PETERN.
\linistry services. Finally. experiences with revolving drug funds in Afghanistan [I]. Haiti [6]. Tanzania [4] and Thailand [9] have demonstrated public willingness to pay for drugs. Where RDFs do exist. however. mechanisms are often devised at the community level to provide for patients who are clearly unable to pay for drugs. In contrast to consumer items such as food and clothing. the demand for drugs is generated not only by individual preferences. but also by the health care providers prescribing the drugs that the patients consume. When a revolving fund is introduced. public health care providers are usually concerned about the selection of drugs for the program. the ability of their patients to buy the drugs, and the quality of drugs bought in bulk by generic name. Thus. involvement of physicians and other health workers in drug selection IS an important aspect of developing drug sales programs. An analysis of the potential market must also consider competition from retail pharmacies as well as local sources of traditional medicine paid for in cash or kind. Revolving drug funds have generally been established in areas where commercial distribution is limited. In these instances. prices can be based on cost-recovery objectives (discussed below) and local ability to pay, without regard to competitive prices. Where subsidized RDFs have come into competition with the private sector, there has been predictable private sector opposition. In Cameroon, for example, the private pharmacists’ association was able to have restrictive legislation enacted when customers flocked to subsidized community-operated pharmacies [3]. Thus, to maximize the market opportunity for an RDF, its planners must identify geographical areas in which reasonable private sector alterantives are not already in existence and strike a balance between the willingness of consumers to pay for drugs and the costs of providing drugs. Costs of an RDF Depending on an RDF’s cost-recovery objectives (considered below) an RDF may or may not seek to recover all of its casts. Nev-ertheless, it is necessary for planners to consider all of the cost categories associated with starting and maintaining a revolving drug fund: drug costs, operating costs and capitalization costs. Drug costs are simply the quantity of drugs purchased multiplied by their unit prices and acquisition costs. The unit prices will be reduced by competitive bulk purchase of drugs by their generic name, and by increasing sales volume. For planning purposes, however. drug prices can be estimated from recent prices listed by international procurement services such as UNICEF in Copenhagen. The Supply of Equipment to Charity Hospitals Overseas (ECHO) in Great Britain, or the International Dispensary Association in the Netherlands. Related drug acquisition costs must be added to the basic unit price: freight. insurance, port charges, government duties (including foreign import duties) and quality assurance expenses. These costs, which can amount to more than 20% of basic unit prices,
CROSS er a/. should be carefullv estimated and assessed for cost reduction possibiliiies. Projection of drug consumption is necessary to determine purchase quantities for a new RDF. This is a crucial. but difficult aspect of estimating RDF costs. There are at least three available estimation techniques which can be used: consumption-based. service-based or population-based [j]. Consumption-based estimates use historical data on drug consumption to forecast future demand. A variety of forecasting formulas can be used, depending on the anticipated rate of program growth [IJ]. Unfortunately. reliable historical drug consumption data are frequently unavailable in public health programs. When consumption data are available, they are often misleading because frequent past shortages result in under-estimation of the real demand, or estimation of demand at a zero money price. Service-based projections require estimates of the number of health workers or health facilities using the fund and the frequency of specific health problems encountered by each. In addition, treatment norms must be developed, listing the types and quantities of drugs needed to treat each health problem. The service information is then multiplied by the treatment norms to calculate total requirements for each drug. Service-based estimates are useful when information on morbidity and mortality is organized by the geographic area to be served by an RDF. Population-based estimates combine epidemiological information on a country’s most important health problems with the same treatment norms mentioned above. Such estimates are useful when only aggregate morbidity and mortality data are available. Service-based and population-based estimates are quite similar in that they may provide the only rational framework for guiding estimates of drug requirements when varying amounts of reliable data are available. Planners should recognize that population-based estimates tend to over-estimate demand unless sufficient attention is paid to identifying a realistic target population. Once quantity estimates have been obtained they can be combined with unit cost information to obtain an overall estimate of drug procurement costs. Operating costs include salaries. facility costs, utilities, office supplies, packaging supplies, telephone and other communication costs, vehicle maintenance and other transportation costs, and allowances for theft, spoilage and inflation. Operating costs can be divided between fixed and variable costs. Administrative salaries, facility costs, utilities and communication costs are usually relatively fixed. Salaries associated with warehouse management and packaging operations, transportation costs and supplies will vary depending on the volume of drugs being handled. Since the bulk of the operating costs will be fixed, an increase in the volume of drugs will reduce the ratio of operating costs to direct costs, thus reducing the total average cost per unit sold. Most operating costs can usually be obtained from existing government accounting information. However, unless drug supply activities are organized under one central service. the total operating costs
Revolving drug funds
Month
PIpelIne
0123456789 r I I 57
I
for
proposed
Haitian IO
I
I
8
I
Foreign
I
DurchaSe
I
50 %'ua'
---‘$
I
I
I,,
drug
sales
12 13 14 ,,,,,,,,,,
11
339
15
program 16
17
18
19
20
21
22
21 1
%
Central warehouse safety srock ICourse Of tnerfx3y
pocllapmq1
Fig. 3. Sample pipeline for a revolving drug fund. From: &fanager.lenrSciences/& He&h: Managing Drug Supply: the Selecrion, Procurement,
Distribution and use of Phmmaceuticals
in Primary Health Care.
Management Sciences for Health, Boston Mass., 1981. may be spread across several departments and budget categories. The necessity of collecting this disparate information will depend in part on the RDF’s cost-recovery objectives. Capifalixtion costs. or development costs, include the cost of designing and planning the system, construction and renovation of office and warehouse space, purchase of vehicles and equipment, andmost importantly-capital funds for initially stocking the system (‘working capital’) [15]: Capital requirements for construction, renovation, new vehicles or other parts of the logistics system will depend largely on the existing infrastructure. These requirements will, therefore, vary considerably from country to country. However, a common challenge which faces all RDFs is that of estimating the amount of working capital required to start the fund. To analyze working capital requirements, it is useful to think of the drug supply system as a ‘pipeline’, as illustrated in Fig. 3. To assure a continuous supply of drugs at health facilities, the pipeline must be filled; once filled, consumption must be matched by purchases at the central level. The pipeline includes not only the flow of drugs to the health facilities, but also the flow of funds back to the central procurement unit. Without the return of
Drugs b
(A)
funds, new procurements cannot be made. supply becomes erratic, and the system soon fails. Two factors influence the amount of capital required to fill the pipeline: the diameter of the pipeline and the length of the pipeline. The diameter is determined by the volume of sales per month (or per year) which reflects rate of drug consumption. The length of the pipeline represents the amount of time between the first commitment of funds for drugs until the funds collected from the sale of those drugs are again available for buying replacement stocks. The length is determined by procurement practices, supplier lead time, the distribution network. stocking policies, cash flow arrangements and related factors. The working capital required is simply the product of the diameter of the pipeline expressed in volume of sales per month and the length of the pipeline expressed in months. Figure 3 illustrates the pipeline model used in planning a revolving drug fund. In this example, the ‘length’ of the pipeline is based on a series of assumptions and estimates regarding the length of time funds are tied up in the foreign purchase pipeline; the amount of stock required at the central, regional and community level; and the time required for drug sales revenues to flow back to the central level. The length
(8)
--25%-
I Origin01 cost - recovery base before subsidies
s ._ E
base
drug subsidy 6 ._ p loo%-L1OOV. x 25%)
100 *I.
ii
S ._ G 3 % a
reduced
Cosr - recovery base further
:
2 cf
a"
I
J Fig. 4. Impact
of multiple
subsidies
on cost recovery
base of RDF.
30
PETER 5. CROSS rr
of the pipeline varies somewhat. depending on whether the source of supply is local or foreign and whether the sale is to a central agency or the community pharmacies (C.P.’ in Fig. 4). Based on these differences. the pipeline varies from 12 to over 22 months. It is sometimes difficult to convince publichealth decision-makers that the pipeline is as long as 12-24 months. As a result, the length of the pipeline is frequently under-estimated, and an initial phase of adequate supply and enthusiasm is soon followed by shortages and disenchantment. The 32 month pipeline illustrated in Fig. 3 would not be unusual for a primary health program in a country which relies heavily on imported drugs and lacks highly efficient supply management and transportation facilities. In planning the supply system and program expansion, it is important to bear in mind that Increases in the number of health facilities or the population covered increases the diameter of the pipeline. Increases or decreases in lead time, the number of levels in the distribution network, or stock-levels will change the length of the pipeline. Declines in the valuation of the local currency will increase the value of funds tied up in the foreign purchase pipeline; this, combined with difficulties in obtaining foreign exchange, should result in adjustments to procurement policies that reduce foreign purchases to a bare minimum by substituting local products or encouraging local production where no substitutes exist. Thus, program expansion as well as program design and management can influence the amount of working capital required. It is essential, therefore, to review estimates of working capital periodically throughout the planning and implementation of a revolving drug fund as modifications are made in original demand estimates or program plans. Cost-recocery
objectices
Unlike a commercial pharmaceutical distributor who must recover all of his expenses plus some profit, a publicly sponsored revolving drug fund can be designed with the objective of recovering any of the following combinations of costs: (I) all costs, including repayment of the capital investment; (2) drug costs and operating costs only; (3) drug costs only, with the Ministry of Health continuing to support the operating expenses: (4) partial cost of drugs; (5) any of the above combinations of costs plus a surplus for supporting community health workers, health centers, or other parts of the health system. The first strategy, that of full cost-recovery, is relatively uncommon for RDFs. Whether supported through bilateral or international aid programs or through government allocations, revolving drug funds are usually not required to repay the initial capital investment. The opportunity to make a onetime capital investment to create a self-sustaining drug supply program is attractive to many donors and repayment of the capital is rarely intended or desired. If the fund is capitalized through private investment (equity financing), then the RDF would need to pay a reasonable return on the investment. An RDF could also be capitalized through a govern-
al.
ment loan. in which case payments for interest and principal would need to be covered from RDF revenues. In general. however. repayment of the capital investment is not an objective of RDFs. A second strategy rvould be for the RDF to seek recovery of all operating and drug costs. The Agence d’Approvisionnement des Pharmacies Communitaires (AGAPCO) in Haiti established this objective. AGAPCO is a two-tiered RDF. in which the central and regional levels of the system seek to cover their own operating and drug costs with a mark-up from the original price of the drugs. while the ‘community pharmacies’ have a second mark-up to cover their own operating expenses. The ‘community pharmacies’ (more accurately ‘community dispensaries’ since they are not staffed by fully qualified pharmacists) are separate entities established with the assistance of AGXPCO. As of this writing. many of the more established community pharmacies have attained the objective of recovering operating as well as drug costs. The central and regional levels have not attained self-sufficiency, but. as the volume of drug sales increases. they are moving closer to that objective [6]. The third strategy would be for the RDF to seek recovery of the direct drug costs only. The government of Dominica is considering establishing an RDF with this cost-recovery objective [ 161; operating costs associated with the drug supply system would continue to be funded by the Ministry of Health’s (MOH) budget. The advantage to this approach is that it provides the revenue necessary to assure regular drug supply, while maintaining a relatively low cost to the patient. This approach also avoids the administrative, accounting, and sometimes legislative effort necessary to identify all operating costs and put them under one budget. The fourth strategy of collecting only a portion of the costs associated with the drugs would mean that the IMinistry of Health would subsidize the sales program’s direct drug costs: the RDF would continue to revolve only with a regular injection of drug procurement funds from the MOH budget. This strategy might be particularly appealing where there is strong public or political resistance to a public sector drug sales program. It may also be an effective intermediate strategy to reduce the financial burden on the government by gradually shifting toward greater cost-recovery. On the other hand, it makes RDFs vulnerable to untimely government budget cuts. Finally, within any of the cost-recovery objectives above, the RDF may establish a strategy that seeks to recover costs plus a surplus [l7]. Such a surplus might be required for future capital investment in vehicles or facilities. for financing working capital if the RDF has expanding accounts receivable and inventory requirements as sales volume expands, or for financing other health-related programs. In Afghanistan [I] and Niger [7], for example, health workers were expected to mark-up the price of drugs enough to maintain their supplies and at the same time pay themselves a small salary. Economic analysis can serve to identify the costs of an RDF and the financial implications of each of the cost-recovery options. However. the selection of a
Revolving drug funds
particular cost-recovery policy for a specific program is largely a matter of local political, economic and health priorities. Role of‘ subsidies and surcharges Once the cost-recovery objectives have been established. a pricing policy must be established. If effective demand was not influenced by price and there were no political or social issues related to pricing policy, this could be relatively straightforward: total expected operating expenses (and principal repayments of capital in the case of the first cost-recovery strategy) could be allocated to the total expected direct costs of the drugs. The percentage relationship between direct drug costs and indirect operating expenses (and principal repayments) would represent the percentage mark-up required to cover all costs. Obviously this pricing strategy is very sensitive to the estimated and actual volume of sales. For this and other reasons, it is probably unrealistic for an RDF to expect to attain its cost-recovery objective during the first few years of the program. In practice, establishing a pricing policy is rarely as simple as stated above. Political, social, health care and patient preference issues frequently enter into the price-setting process. Popular subsidy targets considered by policy-makers include essential drugs, drugs for mothers and children, drugs for the elderly or medically indigent, proven preventive measures (e.g. immunization), and high mortality diseases with costly but effective treatments such as tuberculosis. Subsidization of certain portions of the population or certain categories of pharmaceutical products requires charging higher prices for the remaining population or drugs (‘surcharges’), if full costrecovery is the objective. To demonstrate the interplay of subsidies and surcharges, the following hypothetical case of an RDF with a cost-recovery objective of recovering only direct drug costs is given. Consider the hypothetical cost components of a single package of paracetamol (acetaminophen or Tylenol &,a common fever and pain medication), for which the original purchase price is USSI .OO.Freight, insurance, demurrage, bank fees, customs storage and course-oftherapy pre-packaging add an additional 38% to the price. If inflation is estimated at 9% per year or 0.75% per month, and the ‘pipeline’ is I8 months, then an additional SO.19 (0.75% x 18 x fl.38) will be required to cover the cost of inflation, bringing the average cost per package to 51.59. If losses due to spoilage, expiration, and theft are 20%, there will only be 80 packages available for sale. To recover the replacement costs of 100 packages the remaining 80 must be sold for Sl.99 each ([IOOjSO]x $1.59) or a 99% mark-up. If all patients are charged for all drugs, then the cost-recovery base is 100% [Fig. 4(A)] and the paracetamol would sell for %I.99 per package. On the other hand, some drugs may be considered vital to public health and, therefore, may be provided without charge. If 25% of the drugs fall into this category, then the cost-recovery base is reduced 25% [Fig. 4(B)]. Paracetamol remains with the non-exempt drugs and to recover all drug costs, including those for exempt drugs, the paracetamol should sell for
341
52.65 ([ 100/75] x 5 I .99) or a 165% mark-up. Finally, the Ministry of Health may choose to exempt the medically indigent. those under age 5, and those over age 65. If these groups amount to 30% of the population, then the cost-recovery base is further reduced, with the combined drug and population exemptions leaving only 53% of the original cost-recovery base [Fig. 4(C)]. Again, if sales revenues are to cover all drug costs, the mark-up on non-exempt sales would result in the $1.00 packet of paracetamol selling for 53.75 ([100/53] x Sl.99) a mark-up of 275%. Although a mark-up of nearly 300% may seem substantial, with bulk procurement of selected generic drugs, the selling price may still be below private sector alternatives. It is apparent that exempting a number of vital drugs from the sales program as well as a significant portion of the population can drastically shrink the revenue base. An alternative to financing exempt sales from surcharges or mark-ups on non-exempt sales would be to subsidize exempt sales from national, regional or local health budgets. Clearly, this dilutes the cost-recovery impact of a revolving drug fund, but it represents a compromise approach which may make a revolving drug fund more politically acceptable. Another alternative to percentage mark-ups would be charging the same flat fee for all drugs. Although this approach enormously simplifies accounting, it raises a series of issues related to its impact on consumption patterns and the price-subsidy analysis. Regardless of the ultimate policy pricing decisions, the type of analysis presented in Fig. 4 is an essential element in assessing the impact of subsidies and surcharges on the cost-recovery base of the RDF. On -going financial planning and management
A critical determinant of an RDF’s viability may be its ability to develop, early in its creation, a management information and accounting system capable of providing the cost, revenue, and consumption data necessary to adjust and improve the planning estimates described in the previous pages. An efficient and reasonably comprehensive system which includes the following elements is fundamental to monitoring and controlling an RDF. A cost accounting mechanism should collect indirect operating costs by expense type and measure these against the original planning estimates or budget. Variable costs associated with the volume of drugs can also be collected, controlled and monitored. Direct costs associated with individual drugs or groupings of drugs can be collected and measured against revenues so that pricing policies, especially related to subsidies and surcharges, can be reconsidered as trends are revealed. The system should record sales quantities and revenues by drug or groupings of drugs and by type of health facility (health center, hospital, etc.). This information can be used to assess the impact of pricing policies and to more accurately predict future drug requirements. Regular trial balances and income statements should provide periodic status reports on drug inventories and financial reserves. Such reports are vital to assure that revenues are achieving or at least
313
PETER N.
approaching cost-recovery
objectives. that cash flow
is sufficient to procure new stock. and that inventory
is sufficient to fill expected demand. Finally, a double-entry control system will establish ‘audit trails’ of transactions necessary to assure that liquid assets (cash. negotiable checks and drugs) are safeguarded and that leakages of either drugs or funds can be traced. FUND
MAN.-\CE.VEhT
Although the major emphasis of this paper is on the financial planning and economic analysis necessary for establishing a revolving drug fund. the critical role of effective management in successfully implementing an RDF should not be overlooked. The necessity of an efficient and reasonably comprehensive management information and accounting system has already been discussed. Personnel or human resources management is another key management area. Operating a revolving drug fund requires far more expertise than simply distributing free drugs. As a result, it is sometimes difficult to find spe:alists who have the technical and business management skills necessary to implement and operate a revolving drug fund. For example, one difficulty which may be encountered in setting up the accounting system is that government managers and accountants are often unfamiliar with classic income statement and balance sheet accounting conventionally used in business [18]. Recruitment of qualified individuals (which may include I looking outside public service), clear definition of job responsibilities, adequate training and preparation for specific job assignments and regular supervision are all essential to achieve good staff performance. A functioning distribution network should be considered a prerequisite to beginning a revolving drug fund, since arrangements for delivery and storage have a major impact on the cost and reliability of drug supply [5, 191. Furthermore, significant disruptions in delivery can lead users of the fund to seek supplies elsewhere and, thereby, lead to a drain in capital resources. Where sufficient public distribution channels were not available, public revolving drug funds in Peru [3] and Thailand [4,9] have relied on private sector distribution. Prepackaging of all drugs in standard quantities sufficient for a usual course of therapy substantially simplifies accounting procedures, allows reliable dispensing of drugs with minimally trained staff, and is often a substantial improvement over the haphazard dispensing from bulk containers in many shops and dispansaries. An attractive design for the labels on these ‘course-of-therapy’ packets can serve a promotional role as well as provide the unit price. Finally, implementation of RDFs should include social marketing activities directed at both health care providers and patients [20,21]. Without the acceptance and support of preunderstanding, scribers, dispensers and consumers, the development and growth of an RDF will be problematic. CO?iCLLSlON Revolving drug funds offer an appealing and po-
CROSSer al. tentially successful means of supplying basic pharmaceutical needs for many parts of the Third World. The concept seems quite simple. but in practice these funds have proven to be substantially more complicated to plan and implement than systems which simply give drugs away. The technical problems involved in planning revolving drug funds are sub-
stantial, but analytic approaches to these problems are available. The greatest difficulty may be in convincing Ministry of Health decision-makers and other key individuals that a businesslike approach is necessary for the fund to succeed. To avoid some of the common pitfalls in revolving drug funds careful and systematic financial planning is essential. Such planning should consider market factors affecting the demand for and supply of pharmaceuticals, the capitalization costs of starting an RDF, the expected operating costs. the cost-recovery objectives and the role of subsidies and surcharges. Although experiences to date have provided considerable insight into the potential benefits and common pitfalls of revolving drug funds. there remain many issues for further analysis and research. Among these issues are the following: (I) clarification of the circumstances in which RDFs offer distinct advantages over commercial sales alternatives; (2) definition of pricing policies which optimally balance generation of sales revenue and public health goals of expanding population coverage; (3) determination of the extent to which the supply pipeline can be shortened and working capital requirements reduced by improved supply management techniques; (4) exploration of the role of pricing policy and social marketing in promoting more appropriate and cost-effective prescribing and patient use of drugs; (5) determination of the extent to which economies of scale achieved through bulk procurement of generically-named drugs can contribute to achieving cost-recovery objectives; and (6) definition of marketing techniques for health care providers and patients which are both effective and appropriate to a public health entity. The analyses and discussions presented in this paper are based on experiences from several continents. However, there is no simple. invincible approach to planning and implementing a revolving drug fund. Health decision-makers should take the ideas presented here as suggestions which can be used and adapted to the planning requirements of individual programs, By bearing in mind the difficulties which have faced other revolving drug funds, planners should increase the likelihood that their work will survive and thrive.
REFERESCES
R. (Ed.) Impiemenring a ?4arional Rural Healrh Swem: Managemenr Experiences from Afghanisran. Lexington Press. Boston, 1980. Quick J. er al. Who will pay? Financing drug supplies O’Connor
through revolving funds. Presented at the Conference on Essential Drugs for Primar) Health Care, Harvard School of Public Health. Boston, %lass.. 1984. Pan American Health Organization. Peruvian basic drugs program (CD 23/29). PAHO. Washington, D.C., 1975.
Revolving 4. Stinson W. Community financing of primary health care. Primary Health Care Issues Series 1, No. -1. 1982. 5. Management Sciences for Health. .Cfanaging Drug Supplr: The Selection. Procurement. Distribution. and tiie. df Pharmaceuticals in Primary Health Care. Xlanagement Sciences for Health, Boston, 1981. 6. AGAPCO. Analysis of Drug Sales. Management Sciences for Health. Port-au-Prince, 1984 (unpublished report). 7. Over A. M. Five primary care projects in the Sahel and the issue of recurrent costs. Harvard institute for International Development. Cambridge, Mass., 1980 (mimeograph). 8. Brinkerhoff D. W. Projet de Sante rurale: a history and analysis of nineteen months of operation. Harvard Institute of International Development. Cambridge, Mass., 1980 (mimeograph). 9. The Lampang Health Development Project. T’nailand’s fresh approach to rural primary health care. Project to help practitioners help the rural poor. Case Study No. 8. ICED, Essex, Conn.. 1979. IO. American Public Health Association. Revolving funds: trials and errors in the search for self-sufficiency. Salubritas 45, 198 I. II. Management Sciences for Health. A field survey of and resources in rural health needs, practices. Afghanistan: conclusions and recommendations. Management Sciences for Health. Boston, Mass.. 1975.
drug
funds
343
12. Management Sciences for Health. Tihama primary health care project baseline community survey. IManagement Sciences for Health. Boston. IMass.. 1981. 13. Hartman A. F.. Foreit K. G. and Garcia Y. The costs of illness in Honduras. Management Sciences for Health, Tegucigalpa, Honduras, 1983 (unpublished report). 14. Wheelwright S. C. and Makridakis S. Forecasring .Lfethods for .Lfanagement. 3rd edition. U’iIey. New York, 1980. IS. Sishtia V. S. P. Working Capital Managemmt in Public Enterprises. Kumarian Press. West Hartford. Conn.. 1982. 16. Huff M. A. Pricing strategy for the revolving drug fund in Dominica. hlanagement Sciences for Health. Boston, Mass.. 1983 (unpublished paper). 17. Young D. W. “Nonprofits”‘ need surplus ioo. Hart. Busin. Rev. 121131. Januarv-February, 1982. 18. Anthony R. N’. Making sense of nonbusiness accounting. Hara. Busin. Reu. 83-93, May-June, 1950. 19. Attwood P. R. Planning a Distribution S~srmr. Gover Press, London. 197 I. 20. Population Information Program. Social marketing: does it work? Pop. Rep. Series I. No. 21. January. 1980 (reprinted January, 1982). ?I. Hume M. L. Planning for public acceptance. Manage1983 ment Sciences for Health. Boston, &lass., (unpublished paper).
Printed in Great
Copyright C
REVOLVING DRUG FUNDS: CONDUCTING BUSINESS IN THE PUBLIC
0277-9536.86S3.00+ 0.00 1986Pergamon Press Ltd
SECTOR*
PETER N. CROSS. MAGGIE A. HUFF, JONATHAN D. QUICK and JAMES A. BATES Management Sciences for Health, Drug Management Program. 165 Allandale Road, Boston. MA 02130. U.S.A. Abstract-Pharmaceuticals are essential for preventive and therapeutic health services. Unfortunately, significant demand, limited funds and high prices contribute to frequent shortages of drugs in many public health programs. One method for financing pharmaceutical supplies has been the establishment of revolving drug funds (RDFs) in which, after an initial capital investment, drug supplies are replenished with monies collected from the sale of drugs. All too often however, the funds actually recovered are insufficient to replenish supplies and the fund is soon depleted. In this paper we consider the potential benefits and common pitfalls of revolving drug funds and then focus on the central role of financial planning in establishing drug sales programs. Experiences from a variety of countries suggest several causes for the failure of some RDFs, including: (I) under-estimation of capitalization costs, (2) prices set below true replacement cost, frequent failure to collect payment, (3) delays in cash flow which make funds unavailable for replenishment of drug stocks, (4) rapid program expansion for which additional capital funds are not available, (5) losses due to theft and deterioration of drugs, (6) unanticipated price increases due to inflation or changes in parity rates and (7) foreign exchange purchase restrictions. Common to many of these problems is the lack of a businesslike orientation to RDFs and, in particular, lack of careful financial planning and management. Financial planning for an RDF includes four analytical tasks: (I) (2) (3) (4)
assessment of the potential market, estimation of the costs of an RDF, establishment of the cost-recovery objectives, definition of the role of subsidies and surcharges.
Revolving drug funds offer great potential for supplying drugs in many parts of the Third World, but they must be carefully planned and the organizers should bear in mind the pitfalls which commonly undermine these programs.
Pharmaceuticals-including vaccines, vitamins, antibiotics, oral rehydration solutions and other products-are essential for preventive and therapeutic health services. In addition to their direct health impact, the effectiveness of pharmaceuticals against many common diseases serves to establish the credibility of health workers that they need to promote long-term health improvements through environmental and nutritional changes. The importance and popularity of pharmaceuticals has led many governments in developing countries to espouse a policy of providing free drugs. Unfortunately, few countries have the resources to fully implement such a policy. Significant demand, limited funds and high prices contribute to frequent shortages. Thus the cost of ‘free’ drugs becomes a scarcity of drugs and disillusionment. Health workers are dispirited by trying to provide services without the resources they have been trained to use; the people are frustrated by receiving so much less than they *An earlier version of this paper was presented at the National Council for International Health Conference, Washington, D.C., June 1982. The authors would like to express their appreciation to Day1 Donaldson -and David Dunlop from the Boston University Health Management program and to Catherine Overholt, Margaret Hume, Ronald O’Connor and John Turnbull from Management Sciences for Health for their careful review and constructive critiques of this paper.
were promised. At the same time however, even the rural poor have been found willing to spend significant amounts of their limited cash resources on commercially available, but often highly priced, brand name drugs. One response to the challenge of financing drug supply has been the establishment of revolving drug funds (RDFs) in which, after an initial capital investment, drug supplies are replenished using monies collected from the sale of drugs. Revolving drug funds, which are one type of drug sales program or cost-recovery scheme, attempt to mobilize financial resources based on a demonstrated willingness of people to pay for health services. Unfortunately, the establishment of such funds frequently has been fraught with difficulty. All too often the monies actually recovered are insufficient to replenish the original drug stocks and the fund is soon depleted or ‘decapitalized’. Many factors contribute to the failure of revolving drug funds, one of the most important of which appears to be a resistance to thinking of the fund in business terms and, in particular, a lack of careful economic and financial analysis in planning the fund. Unlike a public sector entity that receives a budget allocation from the government and where management’s principal financial concern is simply staying within that budget, a revolving drug fund must sell a product and, through sales, generate sufficient revenues to meet its cost recovery objectives. In many respects, manage-
335
PETERN. IN
THEORY-EXPENSES
CROSS et a[.
FOR EACH
CYCLE
EQUAL
RECEIPTS
s,,,,iersf==== / o”:J,‘cc~~““a’ /
Procuremenf unit
/
Government medical srores
\
Heal~hi~l~~~~~
Fig. I. Revolving drug funds in theory. From: Management Sciences for Health: Manrzging Drug Supply: the Selection, Procurement. Distribution and ure of Pharmaceuticals in Primary Health Care. p. 23 I. Management Sciences for Health. Boston. Mass.. 1981.
merit’s concerns should reflect those found in private sector businesses-if the RDF is to survive and thrive. In this paper. the potential benefits and common pitfalls of revolving drug funds are considered. Subsequently the paper addresses the central role of financial planning in establishing viable drug sales programs. The potential market, the capital and operating costs of a revolving drug fund, the fund’s cost-recovery objectives, and the role of subsidies and surcharges must all be considered. POTENTIAL REVOLVING
BEXEFITS OF DRUG FUNDS
Revolving drug funds are attractive because they are theoretically self-financing after a one-time capital investment by the community, the govemment, or an outside donor. As illustrated in Fig. 1, the supply of drugs can be continued indefinitely without further government budget allocations as long as the receipts from sales (or in some cases local health budgets) are sufficient to cover replacement costs. In comparison to public free-drug programs, revolving drug funds make use of an untapped financial resource: patients’ payment for drugs. The result of this can be a significant increase in the availability of pharmaceuticals and a corresponding increase in public participation in public health care services [I, 21. In addition, an RDF provides a set of incentives to the population to use pharmaceuticals more cost-effectively. For example, in a free-drug system it is not unusual for relatively healthy villagers going into town for other reasons to stop at a health center with minor complaints and with the aim of obtaining some drug supplies. The cost-recovery objectives of RDFs also force improvements in management information and accounting systems, in inventory control, and in other supply management activities. In comparison to commercial drug stores, publicly sponsored RDFs have several potential advantages. Through the selection of the most cost-effective therapeutic alternatives and competitive bulk procurement by generic name, cash and foreign exchange outlays can be used more efficiently. The same level
of funding can then achieve substantially greater health impact or government funds can be freed for other public health services. Finally, since an RDF can be structured to use both the Ministry of Health infrastructure and financing, it can serve remote markets with limited commercial appeal.
COhlMON
PITFALLS OF REVOLVING DRUG FUNDS
Revolving drug funds or modifications of revolving drug funds have been established in Peru [3]. Guatemala [4], India [4], Bolivia [5], Haiti [6], Senegal [7], Niger [7], Afghanistan [I]. Mali [S]. Indonesia [4], Thailand [9] and elsewhere [4, lo]. Establishment and/or maintenance of many of these programs has been fraught with difficulties. In Montero, Bolivia, for example, a regional revolving drug fund began selling drugs in August 1976, and expanded rapidly. By May 1979, drug supplies at the central level were depleted, there were insufficient funds to procure additional supplies, and the drug sales program was at a standstill [5]. Similarly, a plan in Gosses, Senegal. to have the vaccination program pay for itself failed when villagers proved unwilling to pay for the entire series of vaccinations and revenues proved insufficient to replenish health workers’ supplies [7l. To these two cases could be added other examples of revolving drug funds which have faced difficulties. All too often, RDFs fail to generate sufficient revenues to replenish supplies and, in effect. soon cease to ‘revolve’. In practice. the expenses for each cycle, including losses, can quickly exceed receipts (Fig. 2). Specific causes for loss of available funds or supplies include: rapid program expansion for which additional capital funds are not made available; under-estimation of the capitalization costs of the supply system; unanticipated losses of drugs through theft or deterioration; high operating costs which exceed budgeted amounts;
Revolving IN PRACTICE
- EXPENSES
FOR
EACH
drug funds CYCLE,
337
INCLUDING
LOSSES,
EXCEED
RECEIPTS:
drugs
Key:
w
_ _--
-
-
-c
Flow
of druq
Flow
of
funds
supphes
Loss
of
avpilpble
funds
or drugs
Fig. 2. Revolving drug funds in practice. From: Munagemenr Sciencesfor Health: Managing Drug SuppI!: the Seleciion. Procuremenr. Distribution and use of Pharmaceuricals in Primary Health Care. p. Xl. Management Sciences for Health, Boston. Mass., 1981.
prices set too low for intended level of cost recovery; failure to collect payment for drugs with no subsidy system; delays in collection of subisidies and other payments from government agencies; funds tied up in national banking system or Ministry accounting mechanisms; unanticipated price increases from inflation or changes in parity rates; foreign exchange limitations which restrict foreign purchases for resupply. Many of these problems are avoidable through careful financial planning and sound fund management. The impact of unavoidable problems often can be minimized by anticipating these problems through the planning process.
FINANCIAL
PLANNING
For planning and management purposes it is important that government officials and other individuals involved in establishing a revolving drug fund view the fund more as a commercial operation than a public service. Programming and administrative concepts and procedures familiar to policy-makers and administrators in the public sector will prove insufficient to create a viable revolving fund. Although the aim of an RDF is to provide a needed public service, rather than to turn a profit, its planning and implementation should be approached with a strong business perspective. Financial planning for a revolving drug fund includes four analytical tasks: (1) assessement of the potential market, (2) estimation of the costs of an RDF, (3) establishment of cost-recovery objectives
and (4) definition surcharges. Potential
of the role of subisidies
and
market
The potential market for a drug sales program is determined by the household or consumer demand for pharmaceuticals from public as well as commercial suppliers. The market potential for an RDF is influenced by households’ ability and willingness to pay, the extent of public pharmaceutical supply capability, the prescribing practices of health workers, the availability and accessibility of alternative commercial pharmaceutical outlets, and the prices of commercially available pharmaceutical products. Each of these factors must be assessed during the planning for an RDF. Consumers’ ability and willingness to pay for health care has been established by surveys of the so-called ‘rural poor’ in several developing countries. These surveys have demonstrated that people already spend considerable amounts of their limited cash resources on health care. In Afghanistan, families were spending an average of USS15.80 per year or 6.8% of the annual household income on health care services. Almost one-half of this amount went to drugs purchased from local pharmacies or ‘dokans’ [l I]. In a survey in Yemen Arab Republic, families were found to be spending about USS20.00 per person per year on treatments, including drugs [12]. In Honduras, a 1983 Ministry of Health household survey [13] found that expenditures associated with family illness represented the third largest expense category after food and clothing. Though most Ministry of Health services are provided without fee, 93% of the households interviewed expressed willingness to pay a small fee for services and medications. A ‘more reliable supply of medicines’ was the most commonly recommended area for improvement of
338
PETERN.
\linistry services. Finally. experiences with revolving drug funds in Afghanistan [I]. Haiti [6]. Tanzania [4] and Thailand [9] have demonstrated public willingness to pay for drugs. Where RDFs do exist. however. mechanisms are often devised at the community level to provide for patients who are clearly unable to pay for drugs. In contrast to consumer items such as food and clothing. the demand for drugs is generated not only by individual preferences. but also by the health care providers prescribing the drugs that the patients consume. When a revolving fund is introduced. public health care providers are usually concerned about the selection of drugs for the program. the ability of their patients to buy the drugs, and the quality of drugs bought in bulk by generic name. Thus. involvement of physicians and other health workers in drug selection IS an important aspect of developing drug sales programs. An analysis of the potential market must also consider competition from retail pharmacies as well as local sources of traditional medicine paid for in cash or kind. Revolving drug funds have generally been established in areas where commercial distribution is limited. In these instances. prices can be based on cost-recovery objectives (discussed below) and local ability to pay, without regard to competitive prices. Where subsidized RDFs have come into competition with the private sector, there has been predictable private sector opposition. In Cameroon, for example, the private pharmacists’ association was able to have restrictive legislation enacted when customers flocked to subsidized community-operated pharmacies [3]. Thus, to maximize the market opportunity for an RDF, its planners must identify geographical areas in which reasonable private sector alterantives are not already in existence and strike a balance between the willingness of consumers to pay for drugs and the costs of providing drugs. Costs of an RDF Depending on an RDF’s cost-recovery objectives (considered below) an RDF may or may not seek to recover all of its casts. Nev-ertheless, it is necessary for planners to consider all of the cost categories associated with starting and maintaining a revolving drug fund: drug costs, operating costs and capitalization costs. Drug costs are simply the quantity of drugs purchased multiplied by their unit prices and acquisition costs. The unit prices will be reduced by competitive bulk purchase of drugs by their generic name, and by increasing sales volume. For planning purposes, however. drug prices can be estimated from recent prices listed by international procurement services such as UNICEF in Copenhagen. The Supply of Equipment to Charity Hospitals Overseas (ECHO) in Great Britain, or the International Dispensary Association in the Netherlands. Related drug acquisition costs must be added to the basic unit price: freight. insurance, port charges, government duties (including foreign import duties) and quality assurance expenses. These costs, which can amount to more than 20% of basic unit prices,
CROSS er a/. should be carefullv estimated and assessed for cost reduction possibiliiies. Projection of drug consumption is necessary to determine purchase quantities for a new RDF. This is a crucial. but difficult aspect of estimating RDF costs. There are at least three available estimation techniques which can be used: consumption-based. service-based or population-based [j]. Consumption-based estimates use historical data on drug consumption to forecast future demand. A variety of forecasting formulas can be used, depending on the anticipated rate of program growth [IJ]. Unfortunately. reliable historical drug consumption data are frequently unavailable in public health programs. When consumption data are available, they are often misleading because frequent past shortages result in under-estimation of the real demand, or estimation of demand at a zero money price. Service-based projections require estimates of the number of health workers or health facilities using the fund and the frequency of specific health problems encountered by each. In addition, treatment norms must be developed, listing the types and quantities of drugs needed to treat each health problem. The service information is then multiplied by the treatment norms to calculate total requirements for each drug. Service-based estimates are useful when information on morbidity and mortality is organized by the geographic area to be served by an RDF. Population-based estimates combine epidemiological information on a country’s most important health problems with the same treatment norms mentioned above. Such estimates are useful when only aggregate morbidity and mortality data are available. Service-based and population-based estimates are quite similar in that they may provide the only rational framework for guiding estimates of drug requirements when varying amounts of reliable data are available. Planners should recognize that population-based estimates tend to over-estimate demand unless sufficient attention is paid to identifying a realistic target population. Once quantity estimates have been obtained they can be combined with unit cost information to obtain an overall estimate of drug procurement costs. Operating costs include salaries. facility costs, utilities, office supplies, packaging supplies, telephone and other communication costs, vehicle maintenance and other transportation costs, and allowances for theft, spoilage and inflation. Operating costs can be divided between fixed and variable costs. Administrative salaries, facility costs, utilities and communication costs are usually relatively fixed. Salaries associated with warehouse management and packaging operations, transportation costs and supplies will vary depending on the volume of drugs being handled. Since the bulk of the operating costs will be fixed, an increase in the volume of drugs will reduce the ratio of operating costs to direct costs, thus reducing the total average cost per unit sold. Most operating costs can usually be obtained from existing government accounting information. However, unless drug supply activities are organized under one central service. the total operating costs
Revolving drug funds
Month
PIpelIne
0123456789 r I I 57
I
for
proposed
Haitian IO
I
I
8
I
Foreign
I
DurchaSe
I
50 %'ua'
---‘$
I
I
I,,
drug
sales
12 13 14 ,,,,,,,,,,
11
339
15
program 16
17
18
19
20
21
22
21 1
%
Central warehouse safety srock ICourse Of tnerfx3y
pocllapmq1
Fig. 3. Sample pipeline for a revolving drug fund. From: &fanager.lenrSciences/& He&h: Managing Drug Supply: the Selecrion, Procurement,
Distribution and use of Phmmaceuticals
in Primary Health Care.
Management Sciences for Health, Boston Mass., 1981. may be spread across several departments and budget categories. The necessity of collecting this disparate information will depend in part on the RDF’s cost-recovery objectives. Capifalixtion costs. or development costs, include the cost of designing and planning the system, construction and renovation of office and warehouse space, purchase of vehicles and equipment, andmost importantly-capital funds for initially stocking the system (‘working capital’) [15]: Capital requirements for construction, renovation, new vehicles or other parts of the logistics system will depend largely on the existing infrastructure. These requirements will, therefore, vary considerably from country to country. However, a common challenge which faces all RDFs is that of estimating the amount of working capital required to start the fund. To analyze working capital requirements, it is useful to think of the drug supply system as a ‘pipeline’, as illustrated in Fig. 3. To assure a continuous supply of drugs at health facilities, the pipeline must be filled; once filled, consumption must be matched by purchases at the central level. The pipeline includes not only the flow of drugs to the health facilities, but also the flow of funds back to the central procurement unit. Without the return of
Drugs b
(A)
funds, new procurements cannot be made. supply becomes erratic, and the system soon fails. Two factors influence the amount of capital required to fill the pipeline: the diameter of the pipeline and the length of the pipeline. The diameter is determined by the volume of sales per month (or per year) which reflects rate of drug consumption. The length of the pipeline represents the amount of time between the first commitment of funds for drugs until the funds collected from the sale of those drugs are again available for buying replacement stocks. The length is determined by procurement practices, supplier lead time, the distribution network. stocking policies, cash flow arrangements and related factors. The working capital required is simply the product of the diameter of the pipeline expressed in volume of sales per month and the length of the pipeline expressed in months. Figure 3 illustrates the pipeline model used in planning a revolving drug fund. In this example, the ‘length’ of the pipeline is based on a series of assumptions and estimates regarding the length of time funds are tied up in the foreign purchase pipeline; the amount of stock required at the central, regional and community level; and the time required for drug sales revenues to flow back to the central level. The length
(8)
--25%-
I Origin01 cost - recovery base before subsidies
s ._ E
base
drug subsidy 6 ._ p loo%-L1OOV. x 25%)
100 *I.
ii
S ._ G 3 % a
reduced
Cosr - recovery base further
:
2 cf
a"
I
J Fig. 4. Impact
of multiple
subsidies
on cost recovery
base of RDF.
30
PETER 5. CROSS rr
of the pipeline varies somewhat. depending on whether the source of supply is local or foreign and whether the sale is to a central agency or the community pharmacies (C.P.’ in Fig. 4). Based on these differences. the pipeline varies from 12 to over 22 months. It is sometimes difficult to convince publichealth decision-makers that the pipeline is as long as 12-24 months. As a result, the length of the pipeline is frequently under-estimated, and an initial phase of adequate supply and enthusiasm is soon followed by shortages and disenchantment. The 32 month pipeline illustrated in Fig. 3 would not be unusual for a primary health program in a country which relies heavily on imported drugs and lacks highly efficient supply management and transportation facilities. In planning the supply system and program expansion, it is important to bear in mind that Increases in the number of health facilities or the population covered increases the diameter of the pipeline. Increases or decreases in lead time, the number of levels in the distribution network, or stock-levels will change the length of the pipeline. Declines in the valuation of the local currency will increase the value of funds tied up in the foreign purchase pipeline; this, combined with difficulties in obtaining foreign exchange, should result in adjustments to procurement policies that reduce foreign purchases to a bare minimum by substituting local products or encouraging local production where no substitutes exist. Thus, program expansion as well as program design and management can influence the amount of working capital required. It is essential, therefore, to review estimates of working capital periodically throughout the planning and implementation of a revolving drug fund as modifications are made in original demand estimates or program plans. Cost-recocery
objectices
Unlike a commercial pharmaceutical distributor who must recover all of his expenses plus some profit, a publicly sponsored revolving drug fund can be designed with the objective of recovering any of the following combinations of costs: (I) all costs, including repayment of the capital investment; (2) drug costs and operating costs only; (3) drug costs only, with the Ministry of Health continuing to support the operating expenses: (4) partial cost of drugs; (5) any of the above combinations of costs plus a surplus for supporting community health workers, health centers, or other parts of the health system. The first strategy, that of full cost-recovery, is relatively uncommon for RDFs. Whether supported through bilateral or international aid programs or through government allocations, revolving drug funds are usually not required to repay the initial capital investment. The opportunity to make a onetime capital investment to create a self-sustaining drug supply program is attractive to many donors and repayment of the capital is rarely intended or desired. If the fund is capitalized through private investment (equity financing), then the RDF would need to pay a reasonable return on the investment. An RDF could also be capitalized through a govern-
al.
ment loan. in which case payments for interest and principal would need to be covered from RDF revenues. In general. however. repayment of the capital investment is not an objective of RDFs. A second strategy rvould be for the RDF to seek recovery of all operating and drug costs. The Agence d’Approvisionnement des Pharmacies Communitaires (AGAPCO) in Haiti established this objective. AGAPCO is a two-tiered RDF. in which the central and regional levels of the system seek to cover their own operating and drug costs with a mark-up from the original price of the drugs. while the ‘community pharmacies’ have a second mark-up to cover their own operating expenses. The ‘community pharmacies’ (more accurately ‘community dispensaries’ since they are not staffed by fully qualified pharmacists) are separate entities established with the assistance of AGXPCO. As of this writing. many of the more established community pharmacies have attained the objective of recovering operating as well as drug costs. The central and regional levels have not attained self-sufficiency, but. as the volume of drug sales increases. they are moving closer to that objective [6]. The third strategy would be for the RDF to seek recovery of the direct drug costs only. The government of Dominica is considering establishing an RDF with this cost-recovery objective [ 161; operating costs associated with the drug supply system would continue to be funded by the Ministry of Health’s (MOH) budget. The advantage to this approach is that it provides the revenue necessary to assure regular drug supply, while maintaining a relatively low cost to the patient. This approach also avoids the administrative, accounting, and sometimes legislative effort necessary to identify all operating costs and put them under one budget. The fourth strategy of collecting only a portion of the costs associated with the drugs would mean that the IMinistry of Health would subsidize the sales program’s direct drug costs: the RDF would continue to revolve only with a regular injection of drug procurement funds from the MOH budget. This strategy might be particularly appealing where there is strong public or political resistance to a public sector drug sales program. It may also be an effective intermediate strategy to reduce the financial burden on the government by gradually shifting toward greater cost-recovery. On the other hand, it makes RDFs vulnerable to untimely government budget cuts. Finally, within any of the cost-recovery objectives above, the RDF may establish a strategy that seeks to recover costs plus a surplus [l7]. Such a surplus might be required for future capital investment in vehicles or facilities. for financing working capital if the RDF has expanding accounts receivable and inventory requirements as sales volume expands, or for financing other health-related programs. In Afghanistan [I] and Niger [7], for example, health workers were expected to mark-up the price of drugs enough to maintain their supplies and at the same time pay themselves a small salary. Economic analysis can serve to identify the costs of an RDF and the financial implications of each of the cost-recovery options. However. the selection of a
Revolving drug funds
particular cost-recovery policy for a specific program is largely a matter of local political, economic and health priorities. Role of‘ subsidies and surcharges Once the cost-recovery objectives have been established. a pricing policy must be established. If effective demand was not influenced by price and there were no political or social issues related to pricing policy, this could be relatively straightforward: total expected operating expenses (and principal repayments of capital in the case of the first cost-recovery strategy) could be allocated to the total expected direct costs of the drugs. The percentage relationship between direct drug costs and indirect operating expenses (and principal repayments) would represent the percentage mark-up required to cover all costs. Obviously this pricing strategy is very sensitive to the estimated and actual volume of sales. For this and other reasons, it is probably unrealistic for an RDF to expect to attain its cost-recovery objective during the first few years of the program. In practice, establishing a pricing policy is rarely as simple as stated above. Political, social, health care and patient preference issues frequently enter into the price-setting process. Popular subsidy targets considered by policy-makers include essential drugs, drugs for mothers and children, drugs for the elderly or medically indigent, proven preventive measures (e.g. immunization), and high mortality diseases with costly but effective treatments such as tuberculosis. Subsidization of certain portions of the population or certain categories of pharmaceutical products requires charging higher prices for the remaining population or drugs (‘surcharges’), if full costrecovery is the objective. To demonstrate the interplay of subsidies and surcharges, the following hypothetical case of an RDF with a cost-recovery objective of recovering only direct drug costs is given. Consider the hypothetical cost components of a single package of paracetamol (acetaminophen or Tylenol &,a common fever and pain medication), for which the original purchase price is USSI .OO.Freight, insurance, demurrage, bank fees, customs storage and course-oftherapy pre-packaging add an additional 38% to the price. If inflation is estimated at 9% per year or 0.75% per month, and the ‘pipeline’ is I8 months, then an additional SO.19 (0.75% x 18 x fl.38) will be required to cover the cost of inflation, bringing the average cost per package to 51.59. If losses due to spoilage, expiration, and theft are 20%, there will only be 80 packages available for sale. To recover the replacement costs of 100 packages the remaining 80 must be sold for Sl.99 each ([IOOjSO]x $1.59) or a 99% mark-up. If all patients are charged for all drugs, then the cost-recovery base is 100% [Fig. 4(A)] and the paracetamol would sell for %I.99 per package. On the other hand, some drugs may be considered vital to public health and, therefore, may be provided without charge. If 25% of the drugs fall into this category, then the cost-recovery base is reduced 25% [Fig. 4(B)]. Paracetamol remains with the non-exempt drugs and to recover all drug costs, including those for exempt drugs, the paracetamol should sell for
341
52.65 ([ 100/75] x 5 I .99) or a 165% mark-up. Finally, the Ministry of Health may choose to exempt the medically indigent. those under age 5, and those over age 65. If these groups amount to 30% of the population, then the cost-recovery base is further reduced, with the combined drug and population exemptions leaving only 53% of the original cost-recovery base [Fig. 4(C)]. Again, if sales revenues are to cover all drug costs, the mark-up on non-exempt sales would result in the $1.00 packet of paracetamol selling for 53.75 ([100/53] x Sl.99) a mark-up of 275%. Although a mark-up of nearly 300% may seem substantial, with bulk procurement of selected generic drugs, the selling price may still be below private sector alternatives. It is apparent that exempting a number of vital drugs from the sales program as well as a significant portion of the population can drastically shrink the revenue base. An alternative to financing exempt sales from surcharges or mark-ups on non-exempt sales would be to subsidize exempt sales from national, regional or local health budgets. Clearly, this dilutes the cost-recovery impact of a revolving drug fund, but it represents a compromise approach which may make a revolving drug fund more politically acceptable. Another alternative to percentage mark-ups would be charging the same flat fee for all drugs. Although this approach enormously simplifies accounting, it raises a series of issues related to its impact on consumption patterns and the price-subsidy analysis. Regardless of the ultimate policy pricing decisions, the type of analysis presented in Fig. 4 is an essential element in assessing the impact of subsidies and surcharges on the cost-recovery base of the RDF. On -going financial planning and management
A critical determinant of an RDF’s viability may be its ability to develop, early in its creation, a management information and accounting system capable of providing the cost, revenue, and consumption data necessary to adjust and improve the planning estimates described in the previous pages. An efficient and reasonably comprehensive system which includes the following elements is fundamental to monitoring and controlling an RDF. A cost accounting mechanism should collect indirect operating costs by expense type and measure these against the original planning estimates or budget. Variable costs associated with the volume of drugs can also be collected, controlled and monitored. Direct costs associated with individual drugs or groupings of drugs can be collected and measured against revenues so that pricing policies, especially related to subsidies and surcharges, can be reconsidered as trends are revealed. The system should record sales quantities and revenues by drug or groupings of drugs and by type of health facility (health center, hospital, etc.). This information can be used to assess the impact of pricing policies and to more accurately predict future drug requirements. Regular trial balances and income statements should provide periodic status reports on drug inventories and financial reserves. Such reports are vital to assure that revenues are achieving or at least
313
PETER N.
approaching cost-recovery
objectives. that cash flow
is sufficient to procure new stock. and that inventory
is sufficient to fill expected demand. Finally, a double-entry control system will establish ‘audit trails’ of transactions necessary to assure that liquid assets (cash. negotiable checks and drugs) are safeguarded and that leakages of either drugs or funds can be traced. FUND
MAN.-\CE.VEhT
Although the major emphasis of this paper is on the financial planning and economic analysis necessary for establishing a revolving drug fund. the critical role of effective management in successfully implementing an RDF should not be overlooked. The necessity of an efficient and reasonably comprehensive management information and accounting system has already been discussed. Personnel or human resources management is another key management area. Operating a revolving drug fund requires far more expertise than simply distributing free drugs. As a result, it is sometimes difficult to find spe:alists who have the technical and business management skills necessary to implement and operate a revolving drug fund. For example, one difficulty which may be encountered in setting up the accounting system is that government managers and accountants are often unfamiliar with classic income statement and balance sheet accounting conventionally used in business [18]. Recruitment of qualified individuals (which may include I looking outside public service), clear definition of job responsibilities, adequate training and preparation for specific job assignments and regular supervision are all essential to achieve good staff performance. A functioning distribution network should be considered a prerequisite to beginning a revolving drug fund, since arrangements for delivery and storage have a major impact on the cost and reliability of drug supply [5, 191. Furthermore, significant disruptions in delivery can lead users of the fund to seek supplies elsewhere and, thereby, lead to a drain in capital resources. Where sufficient public distribution channels were not available, public revolving drug funds in Peru [3] and Thailand [4,9] have relied on private sector distribution. Prepackaging of all drugs in standard quantities sufficient for a usual course of therapy substantially simplifies accounting procedures, allows reliable dispensing of drugs with minimally trained staff, and is often a substantial improvement over the haphazard dispensing from bulk containers in many shops and dispansaries. An attractive design for the labels on these ‘course-of-therapy’ packets can serve a promotional role as well as provide the unit price. Finally, implementation of RDFs should include social marketing activities directed at both health care providers and patients [20,21]. Without the acceptance and support of preunderstanding, scribers, dispensers and consumers, the development and growth of an RDF will be problematic. CO?iCLLSlON Revolving drug funds offer an appealing and po-
CROSSer al. tentially successful means of supplying basic pharmaceutical needs for many parts of the Third World. The concept seems quite simple. but in practice these funds have proven to be substantially more complicated to plan and implement than systems which simply give drugs away. The technical problems involved in planning revolving drug funds are sub-
stantial, but analytic approaches to these problems are available. The greatest difficulty may be in convincing Ministry of Health decision-makers and other key individuals that a businesslike approach is necessary for the fund to succeed. To avoid some of the common pitfalls in revolving drug funds careful and systematic financial planning is essential. Such planning should consider market factors affecting the demand for and supply of pharmaceuticals, the capitalization costs of starting an RDF, the expected operating costs. the cost-recovery objectives and the role of subsidies and surcharges. Although experiences to date have provided considerable insight into the potential benefits and common pitfalls of revolving drug funds. there remain many issues for further analysis and research. Among these issues are the following: (I) clarification of the circumstances in which RDFs offer distinct advantages over commercial sales alternatives; (2) definition of pricing policies which optimally balance generation of sales revenue and public health goals of expanding population coverage; (3) determination of the extent to which the supply pipeline can be shortened and working capital requirements reduced by improved supply management techniques; (4) exploration of the role of pricing policy and social marketing in promoting more appropriate and cost-effective prescribing and patient use of drugs; (5) determination of the extent to which economies of scale achieved through bulk procurement of generically-named drugs can contribute to achieving cost-recovery objectives; and (6) definition of marketing techniques for health care providers and patients which are both effective and appropriate to a public health entity. The analyses and discussions presented in this paper are based on experiences from several continents. However, there is no simple. invincible approach to planning and implementing a revolving drug fund. Health decision-makers should take the ideas presented here as suggestions which can be used and adapted to the planning requirements of individual programs, By bearing in mind the difficulties which have faced other revolving drug funds, planners should increase the likelihood that their work will survive and thrive.
REFERESCES
R. (Ed.) Impiemenring a ?4arional Rural Healrh Swem: Managemenr Experiences from Afghanisran. Lexington Press. Boston, 1980. Quick J. er al. Who will pay? Financing drug supplies O’Connor
through revolving funds. Presented at the Conference on Essential Drugs for Primar) Health Care, Harvard School of Public Health. Boston, %lass.. 1984. Pan American Health Organization. Peruvian basic drugs program (CD 23/29). PAHO. Washington, D.C., 1975.
Revolving 4. Stinson W. Community financing of primary health care. Primary Health Care Issues Series 1, No. -1. 1982. 5. Management Sciences for Health. .Cfanaging Drug Supplr: The Selection. Procurement. Distribution. and tiie. df Pharmaceuticals in Primary Health Care. Xlanagement Sciences for Health, Boston, 1981. 6. AGAPCO. Analysis of Drug Sales. Management Sciences for Health. Port-au-Prince, 1984 (unpublished report). 7. Over A. M. Five primary care projects in the Sahel and the issue of recurrent costs. Harvard institute for International Development. Cambridge, Mass., 1980 (mimeograph). 8. Brinkerhoff D. W. Projet de Sante rurale: a history and analysis of nineteen months of operation. Harvard Institute of International Development. Cambridge, Mass., 1980 (mimeograph). 9. The Lampang Health Development Project. T’nailand’s fresh approach to rural primary health care. Project to help practitioners help the rural poor. Case Study No. 8. ICED, Essex, Conn.. 1979. IO. American Public Health Association. Revolving funds: trials and errors in the search for self-sufficiency. Salubritas 45, 198 I. II. Management Sciences for Health. A field survey of and resources in rural health needs, practices. Afghanistan: conclusions and recommendations. Management Sciences for Health. Boston, Mass.. 1975.
drug
funds
343
12. Management Sciences for Health. Tihama primary health care project baseline community survey. IManagement Sciences for Health. Boston. IMass.. 1981. 13. Hartman A. F.. Foreit K. G. and Garcia Y. The costs of illness in Honduras. Management Sciences for Health, Tegucigalpa, Honduras, 1983 (unpublished report). 14. Wheelwright S. C. and Makridakis S. Forecasring .Lfethods for .Lfanagement. 3rd edition. U’iIey. New York, 1980. IS. Sishtia V. S. P. Working Capital Managemmt in Public Enterprises. Kumarian Press. West Hartford. Conn.. 1982. 16. Huff M. A. Pricing strategy for the revolving drug fund in Dominica. hlanagement Sciences for Health. Boston, Mass.. 1983 (unpublished paper). 17. Young D. W. “Nonprofits”‘ need surplus ioo. Hart. Busin. Rev. 121131. Januarv-February, 1982. 18. Anthony R. N’. Making sense of nonbusiness accounting. Hara. Busin. Reu. 83-93, May-June, 1950. 19. Attwood P. R. Planning a Distribution S~srmr. Gover Press, London. 197 I. 20. Population Information Program. Social marketing: does it work? Pop. Rep. Series I. No. 21. January. 1980 (reprinted January, 1982). ?I. Hume M. L. Planning for public acceptance. Manage1983 ment Sciences for Health. Boston, &lass., (unpublished paper).