Bioequivalence and Bioavailability Forum

Dear drgopal!

❝ Can anyone tell me from where to take the reference ranges for the laboratory values in the diagnostics .

You should ask the clinical laboratory for their current ranges.

❝ Is there a standard reference book/site from where the values could be considered.

Google is your friend and comes up with 2Mio+ hits.

The normal range for any measured parameter is the 95% confidence interval gained from evaluation of a large sample of obviously healthy subjects.
Although you will find roughly similar ranges stated by different labs, these values may differ because of distinct

• analytical methods,
  
• reagents used,
  
• standards employed,...

It's not uncommon that labs change their reference ranges for some parameters twice a year...
If you really want to do statistics on clinical chemistry in a BE study (I haven't seen this anyhow) and have to struggle with changed normal ranges in the shift analysis, you may consider standardisation of normal ranges.
For a reference see:
Chapter 12.5 (Analysis of Laboratory Data), in:
S-C Chow and J-p Liu
Design and Analysis of Clinical Trials
John Wiley, New York, pp 560-571 (1998)

❝ Can we have a separate acceptable range on our own (+ 10-15% of the range) as our company's range beside the reference range?

This doesn't make sense, since the normal range is a 95% confidence interval of healthy subjects. +10-15% would include all values.
It's quite common that some values (of the >20 measured ones) in a healthy subject are outside the reference range. This does not necessarily mean that the subject is ill (actually we would expect this for 1/20 of subjects for any parameter). Therefore each value must be assessed in synopsis with related ones, the medical history of the subject, etc. by a competent investigator.
Example: one value for GGT, GOT, GPT outside the range shouldn't be a problem, but all of them should be of concern.
In all protocols I've seen in the last 25+ years the investigator rated every value outside the normal range as either 'ncs' (not clinically significant) or 'cs' (clinically significant), where only the latter lead to exlusion in screening or to a follow-up in the post-treatment assessment.

Dear drgopal,

❝ Can we have a separate acceptable range on our own (+/- 10-15% of the range) as our company's range beside the reference range?

In addition to Helmut's response: if you were to adopt such a system you would need to discuss the extended range on a parameter-by parameter basis, not just extend everything by 10 %. This would be OK for e.g. SGOT - 44 IU/l instead of 40 is no big deal, if the elevation is isolated. But take sodium, with normal ranges of 135-145 mmol/l. Extending the range to 122-159 would definitively not be acceptable ! And once again you can't interpret each parameter separately, you also have to look at the complete picture. And this is the investigator's job, not an SOP's.

Regards
Ohlbe