Concept: Dose Intensity
Last Updated: 20041201
Total DDDs
(e.g., amoxicillin
250 mg) =

Dosage form strength (250mg) X Quantity of amoxicillin 250 mg dispensed ___________________________________________________________ WHO assigned DDD for amoxicillin (1 g or 1000 mg) 
DDDs are assigned per ATC 5th level by the WHO Collaborating Centre for Drug Statistics Methodology. See the
WHO Collaborating Centre for Drug Statistics Methodology Web Site at: http://www.whocc.no/ddd/definition_and_general_considera/
for more information on DDDs.
For some types of drugs, DDDs have not been assigned by WHO either because it is difficult to find appropriate DDDs (e.g. dermatologicals or drugs combined in the same dosage form) or because no requests for DDD have been made to the WHO Centre.
Calculating DDD Rates
Four different types of rates can be obtained for the number of DDDs: per day, per year, per resident, or per user. Rates can also be age and sex adjusted using either the Manitoba population or the group of user proportions for the year of analysis and a direct method of standardization.
 The numerator, which can be summed for an entire year of usage, yielding the number of DDDs dispensed in the population, is:
quantity / (DDD/strength) = # of DDDs, where
 quantity = the quantity dispensed
 DDD = the level assigned by the WHO
 strength = the strength of the drugs in the same unit as the DDD.
 The denominator is the number of residents or users respectively.
Example: ( (Metge et al. (1999) )
Consider the DDD rate calculation for 600 mg tablets of potassium chloride. The DDD for potassium chloride is 3.0 g per day, so the denominator in the above formula is 3.0 g/600 mg = 5. That is, 5 tablets are required to make up 1 DDD. For example, if 3,000,000 tablets were dispensed to the population in a year, then the # of DDDs would be 3,000,000 tablets/5 tablets per DDD = 600,000 DDDs per year.
(Note: Analyses for this study were limited to solid dosage forms (tablets and capsules) due to inconsistency in how quantities are measured; (e.g.: a 200 mL bottle of 250 mg/5 mL amoxicillin liquid could be described as: 1 (bottle), 200 mL, 100 g or even 10,000 mg.). This example also has been somewhat simplified, since rates were age and sex adjusted, which required calculating the DDDs at the claim level).
 Rates per residents per year (assuming a population of 1 million residents)
# of DDDs per year per 1000 residents for potassium chloride
= 600,000 DDDs per year / 1,000,000 residents * 1000
= 600 DDDs per 1000 residents per year
This could be interpreted as an estimate of how many days of treatment every resident, if placed on the drug, would have consumed in a year.
 Rates per residents per day
# of DDDs per day per 1000 residents
= 600,000 DDDs per year / 365 days per year / 1,000,000 residents * 1000
= 1.6 DDDs per 1000 residents per day
This provides a rough estimate of the proportion of the population treated daily with a specific drug.
 Rates per user per year (assuming actual number of users of potassium chloride of 10,000)
# of DDDs per year per user
= 600,000 DDDs per year / 10,000 users
= 60 DDDs per user per year
This is interpreted as the average duration of treatment (number of days per year) for every user of the drug/drug class.
 Rates per user per day  to determine whether the DDD is close to the average daily maintenance dose for the drug's main indication (as determined by WHO).
# of DDDs per day per user
= 600,000 DDDs per year / 365 days per year / 10,000 users
= 0.16 DDDs per user per day
We expect this measure to be close to 1 for drugs used for chronic diseases (i.e., for drugs for which the DDD is likely to be prescribed for every user for every day of the year). To the extent that users are not dispensed 1 DDD per day of a group of drugs depends on the objective of the research.Using DDDs Clinically
A clinical measure can also be calculated to help interpret the DDDs/user/day value which assumes dispensation to users over an entire year. (Note that age and sex adjustment, if needed, are done prior to the calculation).
This can be done in two steps:
1) Calculate an intermediate rate of the number of days supplied per user by summing the number of days supplied (DAYSUPP) recorded on each prescription claim (over a year), and dividing by the number of users. For example, if the sum of the number of days supplied of potassium chloride for the year was 1,210,000 days (assuming 10000 users again), then the average number of days supplied per user would be:1,210,000 days / 10,000 users
= 121 days supplied per user (per year)2) Calculate a clinical measure of how the drug (potassium chloride in this case) is actually being used by dividing the rate of the # of DDDs per user by the rate of the number of days supply per user:
# of DDDs per user per year / avg. days supplied per user
= 60 DDDs per user per year / 121 days supplied per user per year
= 0.49 DDDs / day suppliedDDDs used this way should equal 1 if the drug being examined is primarily used for its main indication at its recommended dose. Values greater than 1 or less than 1 could indicate that the drug is being used primarily for a different indication at a different dose, or that the drug is being over or underdosed for its main indication.
If age and sex adjustment are not needed:# of DDDs per year / # of days supplied per year
= 600,000 DDDs per year / 1,210,000 days supplied per year
= 0.49 DDDs per day suppliedThis method avoids the rates altogether, and the DDDs can still be collapsed across different drugs. For example, the different groups of drugs used to treat hypertension could be described using this clinical measure and interpretation of DDD.
Example: PDD rate calculation for 600 mg tablets of potassium chloride.2) Rate 2: (average number of days supplied per user): number of days supplied (DAYSUPP) summed for all claims/number of users .Numerator (in grams)  assuming 3,000,000 tablets dispensed to the population in a year:3,000,000 tablets * 0.6 g per tabletDenominator  assuming 10,000 users:
= 1,800,000 g1,800,000 g per year / 10,000 users
= 180 g per user per year
Example: assuming 1,210,000 days supplied of potassium chloride for the year:3) Ratio: Clinical measure of PDD  divide the first rate by the second rate1,210,000 days / 10,000 users
= 121 days supplied per user, on average, (per year)
Example: 180 g per user per year / 121 days supplied per user per yearA comparison of a drug's PDD with its DDD gives one an insight into actual use of the drug when compared to its most common and recommended use. The DDD for potassium chloride, for example, is 3 g.= PDD of 1.49 g per user per day
total quantity * strength / total days suppliedNote:
= 3,000,000 tablets * 0.6 g per tablet / 1,210,000 days supplied
= 1.49 g per day supplied per user