Background Blister packaging (mechanical repackaging of drugs in individual patient rations) in our setting involves a noteworthy number of partial tablets as they are common in patients´ medication regimens. Partial tablets are inserted manually, in a personnel and time consuming manner, via a tray adapter into the blister machine (Proud Model, Baxter). Dispensing them through cassettes is not recommended by the machine manufacturer and not implemented in the software settings due to difficulties with handling asymmetric parts by cassette rotors, danger of grinding and potentially higher dust formation.

Purpose Our aim was (a) to increase productivity of blister packaging by implementing cassettes for the most frequently repackaged partial tablets: Trittico (trazodon) 150 mg (one-third), Dominal (prothipendyl) 80 mg (half), furosemid 1A 40 mg (half), Concor (bisoprolol) 5 mg (half) and Lasix (furosemide) 40 mg (half), with a total monthly repackaged volume of about 9200 tablets and (b) to validate this change showing consistent high quality of production.

Material and methods We ordered cassettes for these 5 partial tablets from Baxter and programmed a workaround for the software limitation.

A trial order was generated to test:

1. the software adaptation and the interface with the prescription software and;

2. correct blister filling with partial tablets (ie, functionality of the cassettes).

We compared production time and visible dust formation before and after the change.

Alterations in error rates due to blister misfillings were assessed from in-process controls and systematically examined customer complaints.

Results Partial tablets in the trial order matched the prescription and were correctly repackaged.

Visual inspection of the machine showed no increase in dust formation after implementation of the new cassettes.

The average monthly repackaging time for approximately 78 000 blisters (175 000 tablets) could be reduced from 78 to 60 h. Blister production accelerated from 1000 to 1300 bags/h.

Inaccurate blister fillings detected and corrected in internal visual blister controls increased from 0.11% to 0.21%. Misfillings reported by customers remained unchanged (on average 2/month).

Conclusion Cassettes for partial tablets present a major improvement in our blister setting. Increased but still extremely low blister misfillings were compensated by our final controls. Therefore, consistent quality of the end product as well as higher efficiency and no increase in dust formation were established.

No conflict of interest.