Mean kinetic temperature (MKT): how it is calculated and when it can be used

Why MKT sits above the arithmetic mean

Degradation does not follow the clock evenly. Chemical reactions speed up steeply as temperature rises, so for a product kept at 25 °C, an hour at 30 °C adds more degradation than an hour at 20 °C takes away. An arithmetic average treats the two hours as cancelling out. MKT does not, which is why it always lands at or above the average when temperatures vary.

How MKT is calculated

USP <1079.2> gives the equation, for temperatures taken at even intervals:

1. 1

   Convert each reading to kelvin

   Add 273.15 to each temperature in °C.

2. 2

   Turn each reading into a weight

   Raise e to the power −ΔH / RTᵢ. With USP's default activation energy, that is e to the power −10,000 / Tᵢ, a number that grows steeply with temperature.

3. 3

   Average the weights

   Add them up and divide by n, the number of readings.

4. 4

   Turn the average back into a temperature

   Divide 10,000 K by minus the natural logarithm of the average, then subtract 273.15 to return to °C.

In a spreadsheet, with the readings in °C in cells A2 to A25, that is one formula: =10000/-LN(AVERAGE(EXP(-10000/(A2:A25+273.15))))-273.15. At the default activation energy, a day at 40 °C carries about five times the weight of a day at 25 °C.

A worked example

Take a product stored at controlled room temperature, where USP's limit on mean kinetic temperature is 25 °C, calculated over 30 days when the time in your possession is not known. It spends 29 days at exactly 25 °C and then one day at 40 °C. Its arithmetic mean over the 30 days is 25.50 °C. Its mean kinetic temperature is 26.11 °C, above the limit, because the hot day counts five times over.

The same hot day after 29 days at 20 °C gives an MKT of 22.01 °C, comfortably within the limit. A product that has been kept cool has room to absorb a spike; one that has been sitting at the top of its range has none.

Calculated with the USP <1079.2> equation and the default activation energy, from evenly spaced readings. The interval itself does not change the result: a day logged every 15 minutes at a constant 40 °C carries the same weight as one daily reading of 40 °C.

The last two rows show why refrigerated products get so little room. Four hours at 15 °C keep the day's MKT under 8 °C. Four hours at 25 °C take it to almost 13 °C, and that excursion is outside what USP allows for refrigerated products in any case.

When USP allows it

USP <1079.2> sets out how MKT is used to evaluate excursions, with a window for each storage condition. The window matters: MKT over a long enough period can make almost any spike disappear, which is the misuse USP is guarding against.

For storage, MKT is accepted practice. For transport, USP says it can be considered for excursions in transit, provided the calculation is justified by showing that the product's stability-limiting characteristic (the property that would go out of specification first, such as potency) follows zero-order or first-order kinetics over the temperatures it met: that is, it degrades at a steady rate, or at a rate in proportion to how much is left. The revision dated 1 August 2025 added that MKT can be calculated over the whole shipment, from origin to destination, and added the climatic zone IVb row for hot and very humid regions.

When it does not apply

USP and the UK's Medicines and Healthcare products Regulatory Agency (MHRA) are both specific about the limits:

Not to justify a system out of control

USP: MKT may not be used to justify a storage or transport system with repeated excursions. The MHRA notes that wholesalers have proposed MKT as an alternative to adequate temperature control, and discourages its use without robust supporting information.

Not over an inflated window

USP names using 52 weeks of temperature data as the most significant misuse.

Not to undo an excursion

lowering the warehouse temperature afterwards brings the number down but not the damage, because degradation at the higher temperature is not reversible.

Not on its own

USP: MKT alone is not enough to assess the impact of an excursion. The duration, the peak, and the stability data are needed too.

Not for every product

since the 2025 revision, USP says MKT may not be appropriate for products subject to phase change, such as suppositories, suspensions, emulsions, and creams, and that biologics are a category for which it may not be suitable. Those are evaluated with their own stability and stress studies.

Not without a value from the authorization holder

for wholesalers, the MHRA says MKT is only appropriate where the marketing authorization holder provides an acceptable MKT value for the specific product, and the temperature record is consistent and complete from the moment the product leaves the manufacturer.

MKT and the excursion decision

MKT answers one question in the assessment of an excursion: how much thermal stress did the product take, expressed as a single temperature. The decision still needs the rest of the record and the product's stability data. Our guide to temperature excursions covers the full sequence, from reporting to disposition, and our guide to good distribution practice covers the wider rules for storage and transport.

Where Controlant fits

MKT is only as good as the record it is calculated from. The MHRA has pointed out that it fails where a complete chain of temperature recording cannot be tied to a specific consignment. Controlant's devices record temperature for every shipment, allocated to that shipment, from start to delivery, and Zero-Touch Release evaluates each product in the shipment against its own stability profile: its temperature criteria and its remaining stability budget.