41. Drug packaging and labeling
S L I D E
Drug packaging and labeling: Coronary Drug Project (CDP)18 Number of drug regimens: 6 (1 placebo; 5 test treatments)
Route of administration: Oral (capsule; opaque gelatin shell; size #1)
Concentration/Capsule ESG1: Estrogen (Premarin) 0.28 mg ESG2: Estrogen (Premarin) 0.56 mg CPIB: Clofibrate (CPIB) 0.20 g DT-4: Dextrothyroxine (Choloxin) 0.67 mg
NICA: Nicotinic acid 0.33 g
PLBO: Lactose placebo 0.33 g
Formulary: USPHS pharmacy at Perry Point, Maryland Drug distribution to study clinics: From Perry Point pharmacy
Dispensing: In bottles of 100 capsules from study clinics or local hospital pharmacy Means of identification for purposes of assignment: Bottle number (1–30)
R E L AT E D E N T R I E S
Masking principles (page 111), Masking, censoring, and shielding specifications (page 113), Drug masking procedure (page 115), Masking safeguards (page 123)
D E F I N I T I O N S
batch number n- A number displayed on a product, on its container, or on the label serving to identify the production run from which it came; also lot number.
blister packn- A card or strip of bubble-like blisters containing unit doses of a medication;
blisters backed by a foil-like paper designed to break when pressed; used in trials to aid persons in following a time schedule for treatment (by arranging blisters to correspond to times or days on which medications are to be taken) or to facilitate the assessment of treatment adherence; may be on strips contained in a dispenser designed to electronically record times when the strip is pulled from the container.
bin number drug system n- A system in double-masked randomized trials in which treatment assignment is indicated by bin number (see Coronary Drug Project for example involving 30 bins18); typically a system in which more than one person receives medication from the same bin; system easier to implement and manage than med Id number system and typically more medication conserving; downside relates to potential for collateral unmasking if a bin number is unmasked.
lot numbern- The number used to identify units of a lot; also batch number.
med Id number drug systemn- A drug dispensing system in which assigned treatment is indicated by a unique number, e.g., a system in which patient Id number corresponds to med Id; dispensing system more difficult and expensive to implement and manage
118 V. M A S K I N G
and less medication conserving than the bin number dispensing system but immune to collateral unmasking because of unique numbering scheme, e.g., as discussed in relation to ADAPT.51
N A R R AT I V E
The devil is in the details when it comes to packaging and labeling drugs for masked administration. One obvious clue in the masking is lot and run numbers. Study drugs and matching placebos will have different lot and run numbers, whether packaged by the same or different pharmacies. The best that can be done is to imbed the numbers in sequences of numbers to make identification by those numbers difficult.
The package label will also display the product expiration date. Operationally, this means that all product, including placebo, must have the same expiration date.
Interstate shipment of drugs requires labeling as to content. One approach to complying with the regulation in placebo-controlled trials is to label the contents as being either the test drug or lactose (matching placebo).
For drugs dispensed to study subjects in pill form, the simplest, and usually least expensive, method of dispensing is in bottles. Investigators have to decide whether to dispense with childproof caps; prudent for drugs going into homes with small children but not necessarily for studies in the elderly with arthritic, hands.
An alternative to bottles is blister packs. This form of packaging can be useful when patients are required to take two kinds of pills or where the number of pills to be taken differ, depending on time of day or day of week. Disadvantages are those related to the cost and reduced utility. Blister cards are not as easy to pack for transport or store as bottles and may be more difficult to use, especially if the pills are small and the blisters are hard to break.
Medication has to be labeled for randomization. Basically, there are two options:
bin numbering or med numbering.
With med numbering, each patient receives uniquely labeled medication. The usual approach with such numbering is to supply clinics with medication labeled with patient Id numbers or with numbers linked to patient Id numbers.
The virtues of unique labeling lies in its robustness in regard to preservation of masking. Breaking the code for a particular patient affects only that assignment, revealing nothing about any other assignment. The disadvantages are in the costs and added logistical complexities in unique labeling. Product for a person has to be supplied assuming perfect compliance over the period of treatment. Drug not used, for whatever the reason, is wasted. The amount wasted can be large if compliance is low or if a person drops out early in the course of treatment.
Med numbering is reasonable in short-term trials when all drug needed for a person can be supplied on enrollment. It is problematic where patients have to be resupplied with drug over the course of the trial.
The alternative to unique labeling is bin numbering. In its simplest form, in a trial involving just one test drug and a matching placebo, there would be two bins, Bin 1 and Bin 2, of drug at a clinic: one containing the test drug and the other containing the placebo. Persons, on enrollment, would be randomized to receive drug from Bin 1 or Bin 2.
4 1 . D r u g P a c k a g i n g a n d L a b e l i n g 119
The shortcoming with such a simple system is in its lack of robustness in regard to treatment masking. Any unmasking, purposeful or accidental at a clinic, would unmask all patients in that clinic. The obvious fix is to increase the number of bins. For example, instead of two bins one could have 8 bins, with four bins containing drug A and the other 4 containing drug B. The robustness of the system to unmasking increases with the number of bins, but so does cost.
The bin system was used in the Coronary Drug Project.18,60It involved 30 bins per clinic, with each bin containing bottles of placebos or one of five different test medications. Unmasking a particular bin at a clinic had the effect of unmasking all other patients receiving medication from that bin. The amount of unmasking caused by such unmasking was a function of the number enrolled at a clinic, e.g., 4 to 6 patients at a clinic with an enrollment of around 150.
The obvious advantage of the bin system is in reduced wastage of drug and in simplifying resupply. Clinics are supplied on a per bin basis rather than on a per patient basis. Bins in short supply can be supplied accordingly. Similarly, clinics in multicenter trials running low on medication in a particular bin before resupply can be temporarily supplied from another clinic (assuming a common bin system). That possibility does not exist with the unique labeling scheme.
Also of concern in masked drug trials is the ability to unmask in case of emergency.
The two most common modes of providing such capability are by access to a 24-hour 1-800 number or by use of ‘‘scratch-off’’ or ‘‘open in case of emergency’’ labels. Usually, drugs are shipped to clinics with that label attached with instructions that it be removed before dispensing. The tear-off labels are retained at clinics to be opened in case of emergency.