Raw data documentation

 

Here you will find answers to the following questions:

  • When should the raw data be documented?
  • How much raw data should be documented?
  • What is the procedure, what must be taken into consideration?
  • What requirements are to be made on the documentation?

(See chapter 15.C.4 Testing procedures and test protocol)

14.I.1 Principles

Documentation plays a central role in quality control. Supported by the BARR verdict (see chapter 14.H.1.1 The BARR Laboratories case) the FDA primarily assumes for the case of incomplete and fragmentary documentation that the appropriate testings were not carried out or, much more seriously, that the submitter is attempting to commit fraud. Opinions with regard to what makes for good documentation during an inspection continue to be divided in discussions. However, all agree on one thing: only a careful, complete and unfragmented documentation can provide the written evidence that the analyses were actually carried out. This concerns all areas that are somehow involved in the chain from raw materials through to the drug product.

Previously, the predominant opinion was that the people of central Europe, thanks to their good education system and consequent high level of training as such, would be certain to produce the evidence that their analyses were carried out correctly. An obvious counter-argument to this is that people will always make mistakes and therefore, a documentation system needs to be introduced. This system would simply guarantee that human inadequacies are kept under control.

One basic principle of good manufacturing and analytical documentation is the traceability of the data. In the lengthy documentation procedure which culminates in the final result, e.g. in a assay value of 99.2% of a capsule which ultimately leads to the release decision, no grey areas may appear. All the stages, starting from ordering the materials, processing, sampling and release analyses through to packaging and finally dispatch, must be carefully and comprehensively recorded.

The results that apply for the raw data documentation (see also chapter 15.C.4 Testing procedures and test protocol) in quality control, are given below. A solution for a documentation system is also presented.

There are various requirements that a documentation system must fulfil. The FDA still gives preference to pre-numbered, bound laboratory notebooks (pre-numbered notebooks). However, other systems with far greater flexibility yet which are still secure when handled are also possible. The example given (see figure 12) shows one alternative.

Several principles can be derived for the documentation (see chapter 15 Documentation):

Figure 11 General principles for documentation  

General principles for documentation

  • It must be neatly presented, using authentic ink (blue ballpoint pen or ink). Pencils are not permitted.
  • Entries may not be highlighted with a highlighter pen or covered up with Tippex.
  • Any corrections to incorrect entries must be crossed through with a line so that the original entry is still legible. The correct entry must then be added. This should then be authenticated with a date and signature/initials and justified if so required.
  • Each analyst shall be responsible for his own results. The checks on the analytical documents shall be the duty of the manager, who must be able to delegate the task to reliable and experienced persons. The persons authorised to carry out the checks shall be specified in writing in a list.
  • The documentation of the steps carried out and the observations made must be recorded immediately afterwards. All entries must be made on a continuous basis and in chronological order (raw data, no "subsequent comments"). Empty fields must be crossed at.
  • Information on slips of paper (e.g. before entering them in the official documentation) are not permitted.
  • Printouts of balances, for example, must be identified with analysis numbers and fixed into place in the documentation, and then signed and dated on the edges. Thermo paper should be copied first and the copy then appropriately labelled.
  • Chromatograms or comprehensive documents must be labelled as inserts and form a component of the documentation.
  • If errors are found when checking the documentation they must be discussed with the analysts and then corrected. The corrections will be dated by the analyst and by the person commissioned to carry out the checks, signed and, where necessary, justified.
  • It must be clear at all times who has done what, when, how and why. If, for example, raw data was entered by another person, it should be entered on the raw data who carried out the analysis and at what point it was carried out.
  • All observations must be carefully recorded.
  • Even if raw data will not be considered in the result, it must nonetheless still be documented. The reason why the data will not be considered in the results must be provided.
  • The documentation of the raw data must be checked by an authorised second person and this check must be documented with the date and signature of this second person.

14.I.2 Single sheet documentation system

In response to the dominant opinion, a significant deviation from the FDA instructions can be made in that pre-numbered notebooks must not necessarily be used for the documentation. A documentation system based on single sheets (single sheet system) can be successfully used. This system consists of the following elements:

  • Data sheet
  • Cover sheet
  • Index sheet

Together, these sheets form the documentation system.

A particular advantage of this documentation system consists in the fact that not only routine analyses can be tidily documented, but also special assignments. Ways of developing and improving the methods should be the major consideration here and also the analyses as part of the validation studies (process, cleaning, tubing, and many other factors).

14.I.2.1 Cover sheet

On the cover sheet (no reconciliation of the evidence of use is necessary) all the data relevant to the sample identification must be indicated. The pages used in the data sheet must be correctly entered in the field provided. All inserts (analytical point outs, chromatograms) should have the number of pages indicated on them and if necessary, the place where the electronic file is archieved. At the foot of the page, the analyst must enter the start and end point of the analyses and sign it (date/signature). A second person shall document the checks in accordance with the checklist (see figure 14) by signing in the field provided.

Figure 12 Cover sheet

Link to 14.I-2.jpg

14.I.2.2 Data sheet

All raw data forming a part of the analyses shall be recorded in a data sheet. The top right-hand corner of each sheet shall bear a clearly printed number in different colour (e.g. red) in one place only. These data sheets can be obtained in 50 piece packs. Writing must be on one side only. Empty noteblocks must be returned to the place of issue. This is because a check list is carried out (evidence of use) in relation to the notebooks that were submitted, thus reconciling the use of the data sheets.

As mentioned above, all entries must be entered on a continuous basis and in a chronological sequence. The work steps and observation must be documented with immediate effect. The records must be tidy and complete, so that experienced third parties are also bale at all times to understand and follow the results achieved. The general documentation principles must be taken into consideration here (see figure 11).

Figure 13 Data sheet 

Link to 14.I-3.jpg

Figure 14 Documentation checklist  

Checklist for checking documentation

  • Can key information regarding the sample be found on the cover sheet (product/project number, batch number/sample) and are all the tests also specified with title/code?
  • Is the top line on each cover sheet (key sample information) completed has each test begun with a title/code/version (e.g. FT-IR / A 4.2 / 01-Apr-1999)?
  • Has all raw data been comprehensively recorded in the data sheets? Have original printouts (e.g. initial weights) been affixed? Have these been individually validated on the edges and have the sample names and apparatus numbers been added? Are there several affixed printouts which do not overlap?
  • Was the correct testing procedure applied and was it adhered to? Were the evaluations and calculations correct? Are any possible modifications to or deviations from the testing procedure documented and justified? If no official testing procedure exists, is the process that is chosen described in such a way that an experienced third party can understand it?
  • Is the preparation of the samples and standards documented in detail, particularly the initial weights, the expiry dates of materials used (identification numbers) and all dilution steps?
  • Have supplementary documents (calculation sheets, etc.) been affixed or enclosed? Has all apparatus that was used been entered together with the identification number? Have the columns that were used been documented (column logbook)? Have reagents, solvents and other chemicals been identified with receipt number and expiration date? Do specially produced reagents contain a reference to the manufacturing documentation (reagents logbook)? (See chapter 14.B Reagents)
  • Has the sequence of the analysis (e.g. for HPLC) been documented on a computer printout or by hand? Is it clear which sample was injected when? Are the method parameters specified? Have the spectra and chromatograms been printed out?
  • Have System Checks or System Suitability Checks been carried out, documented and successfully passed?
  • Are the results summarised in a table which enables specifications to be compared and evaluated?

All data sheets are a component of the documentation and must be reconciled. None of the sheets may be thrown away. Data sheets which can no longer be used must be marked invalid and be enclosed in the documentation.Each data sheet must be signed (date/approval) at the foot of the page by the analyst once it is complete. The check carried out by the second person in accordance with the check list (see figure 14) must be appropriately documented on each page (date/approval).

If data sheets are not completely filled when completed, the sections where nothing was entered must be crossed out, dated and approved. Alternatively, the term "End of documentation" or a note to that effect can be entered after the end of the text. This will ensure that no subsequent additional entries can be made.

The use of pre-printed forms is permitted. These should be printed directly onto or affixed to blank sheets or attached in the form of enclosures. These pre-printed forms have the added advantage that nothing is forgotten in the documentation and that the documentation has a more uniform layout. This then also facilitates the checks if the same parameters are always in the same place no major discrepancies are apparent between one analyst and another. Finally, a structured documentation also requires less time because the only thing left to note is the variables.

The detailed documentation instructions and the contents for checks must be set out in an SOP. It is particularly important to stipulate the contents signed for by the second person. Is the checking carried out for plausibility or is the aim to make the entire documentation easy to follow in its accuracy and stepwise structure, in other words, checking the calculations? A thorough verification must be carried out from the GMP point of view. A checklist can be very helpful for this purpose. In addition to the general principles previously referred to, the check list shall also contain the points referred to in figure 14.

Note: In contrast to the production area, the initial weighing in the laboratory is not checked directly by a second person present at the time. This second person is much more likely to check after completion of documentation whether the initial weighing complied with the test procedure and was used correctly in the calculations.

All completed data sheets form the analytical documentation together with the cover sheet and the enclosures. This set may no longer be separated but must be stapled together, for example.

14.I.2.3 Index sheet

The Index sheets (see figure 15) are fixed components of the block referred to above. All the data sheets used and the purpose of their use (description of analysis, e.g. analysis number) are entered in the index sheets. This guarantees the complete traceability of the data sheets. The index sheets are used for the accountability of the data sheets used and must therefore be continuously traced. If all the data sheets from a set have been used, the empty set must be returned to the place of issue. The completeness of the entries shall be checked by a second person and confirmed with a signature, e.g. from the department that issued the sheets (QA) when receiving them back.

The administration of the data sheet sets is the duty of a central department (e.g. QA). This department shall carry out checks on the sets issued and shall also check the entries on the index sheets for completeness.

Figure 15 Index sheet

Link to 14.I-5.jpg

Summary

Documentation must be carried out continuously and in a chronological sequence (no "belated entries").

The documentation should be brief and concise but contain all essential steps right through to the final result, so that the result can be understood by an experienced third party at all times.

A single sheet system can also be used as a documentation system, provided that the traceability is guaranteed.

All raw data must be checked by an authorised second person; these checks must be documented with a date and signature.

The single sheet system is a very versatile one. The use of pre-printed forms facilitates documentation and checking considerably.