Essential Drugs Monitor No. 025-026 (1998)
(1998; 36 pages) [French] [Russian] [Spanish] Voir le document au format PDF
Table des matières
Fermer ce répertoireEditorial. Managing Drug Supply
Afficher le documentHealth reform and drug financing
Afficher le documentHow “topping-up” improved drug management at a small clinic in Ghana
Afficher le documentGood drugs at low cost: Thailand’s provincial collective bargaining system for drug procurement
Afficher le documentDrug supply choices: what works best?
Afficher le document“Contracting-out” drug procurement and distribution: experience with a primary distributor system in South Africa
Afficher le documentImproving the supply, quality control and access to essential drugs in Guatemala
Ouvrir ce répertoire et afficher son contenuNational Drug Policy
Ouvrir ce répertoire et afficher son contenuResearch
Ouvrir ce répertoire et afficher son contenuTraining
Ouvrir ce répertoire et afficher son contenuNewsdesk
Ouvrir ce répertoire et afficher son contenuDrug Information
Afficher le documentMeetings & Courses
Afficher le documentNetscan
Afficher le documentLetters to the Editor
Afficher le documentPublished Lately
Ouvrir ce répertoire et afficher son contenuRational Use

Drug supply choices: what works best?

PERSPECTIVES on the role of government in health vary from a social welfare approach (the state should provide all health and other social services except where it is unable to do so) to a self-help or market economy approach (the private market should provide most health services). In the past many governments (in both developing and developed countries) have subscribed to the social welfare approach, particularly when health technology was limited in scope and cost and was therefore affordable. Increasing complexity and technology has taken the cost of health care provision beyond the reach of most developing country governments, and there is a growing emphasis on the responsibility of the individual to provide for their own health care.

In fulfilling the goals of a national drug policy government has a central role in ensuring that drugs distributed through the public and private sectors are of acceptable quality, safe and effective. Also government has a responsibility to promote the rational use of drugs. In addition, it is necessary to actively promote drug availability (geographic access) and affordability (economic access) if a large share of low income and remote populations depend on private sector drug supply.

Among the decisions which governments have to face in the pharmaceutical sector the most complex and the most costly often concern the financing and supply of drugs for government health services. In some countries public sector drug supply is well financed and administratively efficient. In other countries the drug supply system is unreliable and shortages are common. Such systems suffer from inadequate funding, outdated procedures, inefficiency or a mixture of these and other problems.

In a situation of diminishing resources one response is to maximise them by increasing efficiency. Can private sector mechanisms be used to improve public sector efficiency and, thereby, improve access to essential drugs through government health services? It is important to remember that one element which stimulates efficiency in the private sector is the existence of competition.

Various strategies have been tried in order to provide access to pharmaceuticals and, in particular, to essential drugs. At least five alternatives exist for supplying drugs to governmental and nongovernmental health services.

Effective supply of essential drugs includes rational selection, quality assurance, good procurement and storage practices, and timely distribution

Photo: Sister Patricia McCusker


Central medical stores (CMS): conventional drug supply system, in which drugs are procured and distributed by a centralised government unit;

Autonomous supply agency: a centralised supply system in which the management responsibility is devolved to an autonomous or semi-autonomous Board;

Direct delivery system: a decentralised, non CMS approach in which drugs are delivered directly by suppliers to districts and major facilities. The government drug procurement office tenders to establish the supplier and price for each item, but the government does not store and distribute drugs from a central location;

Primary distributor system: another non CMS system, in which the government drug procurement office establishes a contract with a single primary distributor, as well as separate contracts with drug suppliers. The primary distributor is contracted to manage drug distribution by receiving from the suppliers, storing, and distributing all drugs to districts and major facilities;

Fully private supply: selection, in some countries, practices, drugs are provided by private pharmacies in or near government health facilities.

It is possible to identify some advantages and disadvantages for each of the above systems and to make some theoretical comparisons, but true comparisons of cost-effectiveness have not been made. In part this is because other issues have made such comparisons very complex. The introduction of policies on user charges, decentralisation, contracting-out and privatisation all have an impact on the drug supply system.

The Central Medical Stores system

The historical approach to public sector drug supply is the Central Medical Stores (CMS) approach, in which drugs are financed, procured and distributed by the government, which is the owner, funder and manager of the entire supply system. Selection, procurement and distribution are all handled by the central government, often by a unit within the ministry of health. Financing is usually from central treasury allocations and/or donors, though a CMS can function as a revolving drug fund.

This is a demanding approach in terms of human resources, physical infrastructure, management systems and communication systems, requiring the state to manage and fund every aspect of the system. It has been a logical approach in situations where the vast majority of items were imported through one channel; the demand was predictable; finance and administration were highly centralised; and a developed and professional private sector did not exist. Traditional CMS structures continue to be used in a number of countries including Ghana, Oman and Zimbabwe. However the distribution and financing mechanisms in use vary.

Using the traditional CMS system the availability of drugs in the public sector has deteriorated in many (but not all) countries as the nature of medical practice has changed and real financial resources have diminished. At the same time the demand for, and cost of, health service provision has increased.

The causes are many. The drug supply environment has changed drastically from a range of chemicals and galenicals to a multiplicity of manufactured finished products. CMSs have experienced problems with financial management, quantification of requirements, management of tenders, warehouse management, transport and security of drugs. These problems have also been exacerbated by political or administrative influences and weak staff discipline procedures in an atmosphere of diminishing incentives. Where public financing is adequate then the public sector CMS system can be effective in providing a supply of drugs as has been demonstrated in the past in many countries and is still seen in some.

The inability of the CMS system to achieve its purposes in the current economic and commercial climate in many countries indicates a need to consider alternative mechanisms for the procurement and distribution of drugs.

Autonomous supply agency system

The problems which have been experienced with CMS systems have led some governments to establish systems which place the responsibility for bulk procurement, quality assurance, storage, distribution and financial management in the hands of an autonomous or semi-autonomous supply agency. This model has been tried in several countries, particularly in Africa and Latin America.

Autonomous supply agencies are often constituted as parastatals, either under the ministry of health or as independent organizations, with a board of directors including representation from other (than health) government ministries. Their primary and priority client is government health services and they may or may not operate on a non profit basis. Examples of such supply agencies have been functioning in Benin, Haiti, Sudan, Tanzania, Uganda and Zambia.

A commonly seen Ministry of Health organigram is as follows: where the board is autonomous in running the agency but reports to the minister of health, who may be involved in the appointment of the chairman of the board and/or the executive officer.

The goals of establishing an autonomous supply agency are to achieve the efficiency and flexibility associated with private management and private sector employment conditions. At the same time sufficient public sector supervision is maintained to ensure that the services provide a range of essential drugs, at reasonable prices, with adequate control of quality.

The basic concept is that, under the right conditions, a well-constituted management board or board of directors will have the freedom to appoint qualified senior managers, who will in turn ensure an efficient, accountable supply agency.

Supply agencies may be established in the context of a public sector revolving drug fund, where fees are used to purchase drugs on a cash-and-carry basis as in Benin, or in a system where government institutions purchase drugs with centrally allocated treasury funds as in Uganda, or where budgets are still centrally controlled as in Tanzania. Whatever the financing mechanism, autonomous agencies can only function if there is a market for their products and if the client(s) has funds to purchase the products.

Graphic: Zimbabwe Essential Drugs Action Programme

Autonomous agencies are likely to improve drug supply only if structured to overcome the constraints of the CMS approach. Experience to date, though limited, suggests that features which should be sought in establishing autonomous supply agencies, include:

supervision by an independent management board;
professional pharmaceutical supply managers;
good personnel management;
adequate financing;
public accountability and sound financial management;
focus on essential drugs (rather than “profitable” alternatives);
focus on quality assurance, both in terms of products and of services provided.

The intention is that an autonomous supply agency will achieve greater value for money and improved drug availability through more efficient management. Three important characteristics that are needed to promote efficiency are flexibility, incentive and competition. The existence of competition will encourage efficiency, but in the majority of situations the monopoly of the CMS continues to apply to the autonomous agency, with no pressure to improve the quality of services and products or aim for the lowest prices.

A whole series of difficulties may occur, for example, if senior managers are political appointees rather than professional managers appointed by an independent management board. Similarly if the government retains the authority to require distribution of drugs without charge or on a credit basis (without ensuring payment). Again, if special interests outside the agency influence drug procurement, or if the agency is required to retain staff members regardless of their ability or performance, these factors will be counter-productive.

Countries considering an autonomous supply agency should recognise that this approach will not solve problems related to lack of funding for drugs.

Direct delivery system

In general, CMS and autonomous supply services involve bulk procurement into, and distribution from, a central warehouse. The costs and logistical problems associated with central storage and distribution are substantial. An alternative is the direct delivery system.

In this non CMS model, a government procurement office tenders to establish prices and suppliers for each essential drug, but drugs are then delivered directly by the suppliers to individual regional stores, district stores or major health facilities. Variations of direct delivery contracts have been implemented in many countries, including Chile, Colombia, the Eastern Caribbean, Indonesia, Mexico, South Africa, Thailand, UK and Peru. In Indonesia annual allocations for drugs are made on a per capita basis to each district. Using their budget and the Ministry’s current price list for essential drugs, each district determines its own drug order.

Contracts for direct delivery may specify fixed quantities with scheduled deliveries (generally the approach in Indonesia) or estimated quantity tenders with orders placed by the local store or health facilities as needed. Financing arrangements can be a sensitive issue. Debts can quickly accumulate if drug supplies are not balanced against available funds. This means maintaining separate accounts for each supply point (if funding is from central allocations) or ensuring that all supplies are paid for at the time of delivery.

Like most procurement systems, direct delivery contracts require a sole-source commitment, that is, for the tender drugs the local warehouses and facilities will order from the supplier who holds the tender contract. The local purchasers are free to order drugs that were not on the tender from any supplier.

Direct delivery supply agreements depend on and encourage further development of an effective private sector distribution system. In principle, they reduce storage and transport requirements and risks for the government by specifying in the procurement contracts that drugs are to be delivered directly to district stores and major health facilities. The government then has only to store drugs at the regional or district level and deliver them to health centres and peripheral health units.

Direct delivery contracts preserve the benefits of centralised selection (the essential drugs list), bulk procurement (suppliers offer favourable prices to get all the business for the products they are awarded), and centralised quality control (only reputable suppliers are invited to tender). Hospitals and/or districts benefit from being able to manage their own funds and determine the exact quantities needed. Also, the problems of security, central storage and transport are shifted from the ministry to the private suppliers.

With a direct delivery system, however, district level and facility level drug management responsibilities are much greater, since the quantities and quality of drug deliveries must be confirmed. Success will depend on the ability and willingness of staff to manage the increased responsibilities. Finally, direct delivery in itself cannot solve problems of inadequate financing.

Primary distributor system

The primary distributor system is a variation of direct delivery in which the public procurement agency tenders and establishes two types of contracts. The public procurement agency contracts with any number of suppliers to establish the source and price for each drug. But the drugs are not delivered by the suppliers directly to health facilities; instead, a separate contract is negotiated (through tender if feasible) with a single private sector distributor, the primary distributor. Two provinces in South Africa use the primary distributor system for delivery of drugs and medical supplies to hospitals. Contracts for the procurement of drugs are negotiated nationally using competitive tenders.

The suppliers deliver tender drugs to the primary distributor, who is responsible for maintaining sufficient stocks to fill orders from regional warehouses, district stores and/or health facilities. Primary distributors may maintain their own vehicle fleet or subcontract transportation.

Like other direct delivery contracts, this system depends on sole-source commitment for the essential drugs list, though districts and health facilities may be allowed to purchase non-tender drugs from any source. The system also requires the same level of good information and monitoring.

The primary distributor is paid a fee or commission for storage and delivery services. The primary distributor may appear to add an extra middleman and extra costs, but the expectation is that the cost of the primary distributor will be less than the cost to the government of running the warehouse and distribution system itself. Competitive bidding for the primary distributor contract is important to achieve this efficiency.

Fully private supply system

Finally, national policy, insufficient financing or management problems have led some countries to avoid responsibility for providing hospitals and health centres with even essential drugs. Where this is the case, retail pharmacies become the source of supply, especially in urban environments. Often the pharmacies are located very close to the hospital, and may be located inside the hospital. Such pharmacies may be part of a parastatal pharmaceutical enterprise (Sudan), or they might be an institutional enterprise (Ghana), or they may be independent enterprises. In some countries, patients receiving health care through the public sector are left to buy virtually all drugs on their own from the private sector. This situation usually results from complete lack of government funds, rather than as part of an official drug management plan.

Sometimes Central Medical Stores leave a lot to be desired, as can be seen here in a country in Sub-Saharan Africa

Photo: WHO/O. Brasseur

As with revolving drug funds operated by the government, the greatest concern with fully private supply is equity of access for the poor, children, patients with communicable diseases and other vulnerable groups.


These health reform strategies are among many being tried in developed and developing countries to improve the efficiency and outcomes of health care delivery. They have often been initiated in response to the situation in developed countries and are being proposed as potential answers to the problems faced in the delivery of health care in developing countries. These strategies have significant implications for drug supply systems in developing countries which may not have been factors in their implementation in developed countries.

Decentralisation either as delegation or devolution is intended to improve the responsiveness, quality and efficiency of health services. Decentralisation aims to achieve these benefits through greater local involvement, more direct public accountability, increased flexibility to adjust to local circumstances, more rapid and more accurate communication, and quicker adaptation to changing conditions.

Problems that have occurred when attempts have been made to decentralise drug management functions include the following:

Lack of capacity: drug management responsibility may be decentralised without ensuring there are sufficient local staff and management capacity to sustain such services.

Lack of financial resources: responsibility is sometimes decentralised for all drug supply without providing an adequate budget. In this case decentralisation simply becomes abandonment of responsibility.

Increased corruption: because of the money involved, interference for personal gain is common in drug supply systems. While decentralisation is meant to improve accountability, it makes it easier for local officials or other special interests to profit fraudulently.

Increased cost: decentralisation of procurement usually means smaller order quantities. It can result in higher prices for essential drugs, although this problem can be overcome with central contracts coupled with decentralised ordering.

Contracting-out or resourcing in health care has most commonly been tried for non clinical services such as equipment maintenance, laundry and food services. Contracting out services is common within the private sector. Often companies find that outside contractors who specialise in specific services such as managing staff cafeterias or repairing computers can provide those services at lower cost and higher quality than providing the service in house. The philosophy prevails in industrialised economies that if you can find someone else to do a specific task then pay them to do it rather than committing the capital resources to do it yourself; for example, distribution is usually contracted out by large supermarket firms.

Primary distributor systems, transport contracts, port-clearing services and related approaches to private sector involvement require contracting for services in contrast to contracting for products (drugs, for example). However to contract out activities requires the skills of writing, negotiating and monitoring contracts. Contracting functions most effectively where there is competition as with any tendering process. Contracting-out also demands a commitment to pay the contractors according to the terms of the contract.

Privatisation in health is properly defined as transfer of ownership from the public to the private sector. But the term is also applied, less precisely, to contracting government services to the private sector (as with direct delivery contracts) or introducing private sector features into the public sector (as with government owned but semi autonomous supply agencies). The full privatisation of drug supply would have implications for equitable access to drugs in an environment where profit will become the motive for supply.


If a different drug supply system is not chosen as a result of a careful analysis of the underlying causes for the weaknesses of the existing system in a particular country, the change may not produce the desired outcome. Systems chosen because they function in the climate of a “successful” market economy may not prove to be the solution to the problems faced in the supply of drugs in the context of a developing country.

To weigh the pros and cons of one drug supply system against another cannot properly be done from a global perspective without detailed study. Each country brings unique political, economic and geographical factors to the equation. It will never be possible to state that one particular system is “the best”. However some basic factors will point in the direction of certain systems, for example, the existence of an effective private sector is necessary for either direct delivery or prime distributor systems to function.

The Action Programme on Essential Drugs has initiated a multicountry study to examine in depth at country level the outcomes of the drug supply systems in use. This study will look at the advantages and disadvantages of the systems, and the factors which influence the level of success in correcting problems and meeting needs.

This article has been adapted from Drug Supply Strategies, Chapter 6 of Managing Drug Supply (see reference below).

Some experiences with competitive mechanisms for public drug supply

Autonomous supply agency in Tamil Nadu, India

The Tamil Nadu Medical Services Corporation (TNMSC) was created in 1995 to contain drug costs and reduce shortages by purchasing and supplying drugs to government health care facilities. TNMSC is set up as a government company, with a Board chaired by the Secretary of Health, which is accountable to the Minister of Health. The TNMSC created a list of 267 essential drugs from the previous state drug list of 900 items.

Drugs are procured through tender and delivered directly to district level stores. Quality assurance procedures are in place, including sampling of products from manufacturers and district stores. Testing is contracted to reputable private laboratories through tenders. TNMSC adds a 5% charge to fund its own operations. Each facility is given a budget target and issued a “pass book” in which to record the value of drugs it has received. Through prompt payment and other administrative efficiencies, TNMSC has considerably reduced drug costs, while maintaining reliable supply.

Combined supply strategies in Zimbabwe

The Zimbabwe Essential Drugs Action Programme (ZEDAP) uses different systems for different categories of drugs. High-usage drugs on the essential drugs list are procured, stocked and distributed in bulk through the central medical stores. For high-cost, slow-moving specialist items, direct delivery contracts are used. For most specialist items an annual tender is conducted to fix the price for the year. Drugs are then ordered as needed by the roughly two dozen national hospitals and NGO hospitals which require them. Orders are delivered directly to the hospitals. Finally, for cancer agents and some other highly specialised drugs, no contract exists. Instead, drugs are purchased by the Ministry’s pharmaceutical division by individual order, with permission from the Secretary of Health.

Source: Bennett S, Muraleedharan VR. Personal communication on Tamil Nadu Medical Services Corporation. 1997.

MSH/WHO/DAP. Managing Drug Supply, 2nd ed. Hartford, CT: Kumarian Press; 1997.

vers la section précédente
vers la section suivante
Le Portail d'information - Médicaments essentiels et produits de santé a été conçu et est maintenu par l'ONG Human Info. Dernière mise à jour: le 29 octobre 2018