Indicators for Monitoring National Drug Policies
(1999; 250 pages) [French] Voir le document au format PDF
Table des matières
Afficher le documentACKNOWLEDGEMENTS
Ouvrir ce répertoire et afficher son contenuCHAPTER I: Introduction
Ouvrir ce répertoire et afficher son contenuCHAPTER II: Development of the manual
Ouvrir ce répertoire et afficher son contenuCHAPTER III: Model lists of indicators
Fermer ce répertoireCHAPTER IV: Methodology for indicator calculation
Afficher le documentOrganizing the data collection
Afficher le documentCollecting data
Afficher le documentAnalysis and reporting
Afficher le documentConducting surveys
Afficher le documentCalculating the value of a basket of drugs
Ouvrir ce répertoire et afficher son contenuCHAPTER V: Detailed presentation of indicators
Ouvrir ce répertoire et afficher son contenuANNEX 1: Data collection forms
Afficher le documentANNEX 2: Glossary
Afficher le documentANNEX 3: Table of random numbers
Afficher le documentBACK COVER
 

Calculating the value of a basket of drugs

Selection of a basket of drugs

Various methods can be used to constitute a basket of drugs for monitoring trends in prices. However, for a basket of drugs to be representative of the national drug consumption, a number of principles should be respected in the choice and the number of the drugs.

Choice of drugs

The drugs to be included in the basket should be selected from among the essential drugs which are most used in the country. This selection should be carefully made according to the following steps:

• the drugs to be included in the basket should be from the national list of essential drugs (EDL). Essential drugs are those that satisfy the health care needs of the majority of the population and should be available at all times in adequate quantity and in the appropriate dosage form. Most countries have a national list of essential drugs; the other countries can use the model list of WHO;

• drugs to be included in the basket should also be selected on the basis of their general usefulness and their wide availability internationally. The advantage of using the same selected list (see Table 2 for a model of a basket of drugs) is that comparison could be made among different countries;

• some drugs (for example, acetylsalicylic acid and chloroquine) are used with different dosages. For each of the drugs, countries should select those which are most commonly used and on the basis of local availability and need;

• criteria for selecting drugs for the basket should include the concept of affordability for the majority of the population. For example, praziquantel, although useful in many developing countries and especially many of the poorest ones, should not be included in the selection as it is an expensive drug which cannot be afforded by many people.

Number of drugs

The number of drugs to be included in the basket cannot be defined at global level as it depends on many variables which are country-specific (level of precision required, type of consumption, pricing policy, etc.). The ideal would be to follow the price of as many products as possible and eventually of all the products. Experience shows that such a decision implies expensive, lengthy procedures for selecting the basket and for monitoring the price trends. However, for calculating the indicators in the manual, it is recommended that the number of drugs in the basket should be limited, on the condition that these drugs are (i) on the national list of essential drugs, (ii) among the most important therapeutically and (iii) among the most used. If these principles are respected, a basket of 10 essential drugs can be considered sufficient to calculate indicators PR21, PR22, PR30, PR32 and OT4 in a developing country with limited resources. Such a basket is also suitable for building a price index to monitor price trends. An example of such a basket of drugs is given in Table 2.

Table 2: Example of a basket of drugs

acetylsalicylic acid

tablet, 300 mg

aluminium hydroxide

tablet, 500 mg

amoxicillin

tablet, 250 mg

chloroquine

tablet, 150 mg (as phosphate or sulfate)

ferrous salt

tablet, equivalent to 60 mg iron

mebendazole chewable

tablet, 100 mg

metronidazole

tablet, 500 mg

paracetamol

tablet, 500 mg

phenoxymethylpenicillin

tablet, 250 mg (as potassium salt)

sulfamethoxazole + trimethoprim

tablet, 400 mg + 80 mg

Calculate the value of the selected basket

Different methods exist to calculate the value of a basket of drugs for a given year. For use as a basis for a monitoring system on price trends, the price of each of the drugs of the basket should be weighted by its level of consumption. The method proposed in the manual is simple to implement and scientifically acceptable. It is based on principles used everywhere to calculate price indexes.

The main problem with this method is the need to know with some precision the drug consumption for the year used as reference year (Y0), as this consumption will be used as a basis for calculating the price index for the following years. In many developing countries, such data on drug consumption are not reliable. In these situations, the value of the basket can be based on a reasonable estimation of the consumption of each of the selected drugs. This reference basket will then be used every year until the quality of the data collection improves. A clear increase or decrease in the consumption of certain drugs of the basket, and/or therapeutic changes, will necessitate its modification.

How to calculate the value of the basket of drugs for the year of reference

- For a given year (for instance 1997), take the first 20 causes of consultations and the prevalence of each cause (p1, p2,..., p20, i.e. pi). This can be obtained nationally if a health information system exists or from a representative sample of health facilities. The sample proposed for collecting data for certain indicators can be used for this purpose.

- For each of the causes, the standard treatment (or the most common treatment) used in the country should be noted (name of the drug under INN, standard dosage, length of the treatment).

- Select 10 essential drugs among those most used for the treatment of the 20 first causes of consultations. For each drug, calculate an estimated consumption by cause of consultation (Q1, Q2,..., Q20, i.e. Qi) by multiplying the number of units (tablets, vials, etc.) included in the standard treatment (a1, a2,..., a20, i.e. ai) by the prevalence (pi); then add all the quantities obtained for each drug:

Qi = Σai x pi = a1 x p1 + a2 x p2 + ... a20 x p20

- Calculate the average price for the given year of each of the drugs of the basket (P1, P2,... P10, i.e. Pi) from a survey in a representative sample of drug outlets (the sample proposed for collecting data for certain indicators can be used for this purpose). According to the indicator, the calculation will be done for different prices: CIF/ex-factory price, international reference price or retail price.

- The value of the reference consumption for each drug of the basket is obtained by multiplying the consumption Qi by the price Pi. For the year Y0, the value of the basket will be:

V0 = ΣPi x Qi = P1 x Q1 + P2 x Q2 + ... P10 x Q10

This figure will be designated arbitrarily as the basis 100 of the value of the consumption calculated for the year of reference Y0.

How to calculate a drug price index

A drug price index can be defined as the value of a basket of drugs calculated for a year Yx relative to the value of the same basket in the year of reference Y0. The formula for the price index will be:

This means that the price index will give a precise representation of the price trend every year, 100 being the value of reference of the consumption calculated for the year Y0.

CALCULATING THE VALUE OF A BASKET OF DRUGS

Cause of consultations
Name of drugs

No. 1

No. 2

No. 3

No. 4

No. 5

No. 6

No. 7

No. ...

No. 16

No. 17

No. 18

No. 19

No. 20

Quantity of drugs consumed in millions of units
Qi = Σai x pi

Average unit price for the year
Pix

Consumption value of each drug for the year
Vx = ΣQi x Pix

1. ...

a1

a2

a3

a4

a5

a6

a7

a...

a16

a17

a18

a19

a20

Q1

P1

Q1 xP1

2. ...

b1

b2











b20

Q2

P2

Q2 xP2

3. ...

c1

c2











c20

Q3

P3

Q3 xP3

4. ...

d1

d2











d20

Q4

P4

Q4 xP4

5. ...

e1

e2











e20

Q5

P5

Q5 xP5

6. ...

f1

f2











f20

Q6

P6

Q6 xP6

7. ...

g1

g2











g20

Q7

P7

Q7 xP7

8. ...

h1

h2











h20

Q8

P8

Q8 xP8

9. ...

i1

i2











i20

Q9

P9

Q9 xP9

10. ...

j1

j2











j20

Q10

P10

Q10 xP10

Prevalence

p1

p2











p20

-

-

ΣQixPi

 

vers la section précédente vers la section suivante
 

Dernière mise à jour: le 24 avril 2012