Please Scroll Down to See Forums Below 
George Stark
New member
This has always been a concern with BB'rs splitting tabs. I have always been on the not split side of the fence. I have tried in my own words to explain why this is a bad idea before. It seems my testing did little to sway any one over to my side. Well here is a study the may help every one understand Why I feel the way I do.
GS
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Posted 07/18/2002
Abstract and Introduction
Abstract
Objective. To divide 11 commonly split tablets and evaluate the resulting half-tablets for content uniformity.
Design. Pre-post comparison.
Setting. Laboratory. Interventions: A trained individual split tablets of 11 products using a single-edged razor blade and 3 products by hand alone.
Main Outcome Measures. The Uniformity of Dosage Units test published in the United States Pharmacopoeia 24 (USP), which applies to whole tablets, was adapted liberally to assess the dose uniformity of the resulting split tablets.
Results. Of the 11 razor-split products, 8 failed the liberal adaptation of the USP uniformity test. No visible tablet features (e.g., scoring) predisposed a product's split tablets to pass or fail the uniformity test. All three hand-split tablets failed the uniformity test and yielded worse results than did razor-split tablets.
Conclusion. The majority of the 11 drug products we tested, when assessed for their ability to be split into half-tablets of equal dose, failed a liberally interpreted USP uniformity test. The practice of dividing tablets to save costs or to improve a dosage regimen may not cause problems for patients using drugs with low toxicity and relatively flat dose-response relationships, but it is not recommended for patients using drugs with more substantial toxicity and steep dose-response efficacy curves.
Introduction
Pharmacists sometimes instruct patients to split tablets into halves or quarters in an effort to titrate doses or reduce costs. Some practitioner groups and insurance plans have promoted the splitting of double-strength tablets to reduce medication costs.[1]
Recommendations to split whole tablets into two halves rest on the assumption that the tablet will be split with sufficient accuracy. However, Stimpel et al.[2] showed that a majority of antihypertensive tablets did not break evenly, despite being scored. Sedrati et al.[3] identified several tablet products that, when split, resulted in half-tablets with doses outside an 85% to 115% range of the target half-dose. McDevitt et al.[4] evaluated the ability of healthy volunteers to split hydrochlorothiazide tablets. More than 40% of the resulting split tablets deviated by 10% or more from the target weight, and more than 10% of the split tablets deviated by at least 20%.
An important public health standard for prescription medications is the Uniformity of Dosage Units test in the United States Pharmacopeia 24 (USP).[5] This test requires relatively uniform drug content from tablet to tablet, essentially ensuring that the vast majority of individual tablets contain between 85% and 115% of the labeled dose while allowing for a 6% relative standard deviation (RSD) in drug content.
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Objectives
The objective of this study was to divide 11 commonly split tablets and evaluate whether the resulting half-tablets provided equal doses, based on a liberal adaptation of the USP Uniformity of Dosage Units test.
Methods
Selection of Tablet Products for Splitting
Eleven tablet products identified through anecdotal reports of which products are commonly split were selected and subjected to splitting:
Paxil 40 mg (lot 9258B13) (paroxetine -- GlaxoSmithKline).
Zestril 40 mg (lot CTB671) (lisinopril -- AstraZeneca).
Zoloft 100 mg (lot 9JP040A) (sertraline -- Pfizer).
Glucophage 850 mg (lot MAH93) (metformin -- Bristol-Myers Squibb).
Glyburide 5 mg (lot 109257A) (Novopharm).
HydroDIURIL 50 mg (lot D6811) (hydrochlorothiazide -- Merck).
Lipitor 20 mg (lot 09609V) and 40 mg (lot 0012108) (atorvastatin -- Parke-Davis; Pfizer).
Oretic 25 mg (lot 16053AF21) and 50 mg (lot 14884AF21) (hydrochlorothiazide -- Abbott).
Viagra 50 mg (lot 89R035A) (sildenafil -- Pfizer).
A single, trained individual divided tablets of each product using a single-edged razor blade. Each tablet was placed on a bench top, the individual held the razor blade, positioned the razor at the tablet surface, and pressed the razor into the tablet.
Additionally, three of the products' tablets were soft enough to split by hand alone: glyburide 5 mg, HydroDIURIL 50 mg, and Oretic 50 mg. These tablets were split by placing the tablet between each hand's forefinger and thumb and pushing the thumb through the tablet score.
Criteria for Split Tablet Performance: Uniformity Test
The Uniformity of Dosage Units test was adapted to assess the dose uniformity of resulting split tablets.[5] This test is intended to ensure the content uniformity of whole tablets and does not appear to apply to half-tablets. Briefly, the test calls for selection of 30 tablets from each batch of manufactured tablets. Of these 30 tablets, 10 are selected for analysis. Each tablet is weighed, and each must be within the 85% to 115% range of the target tablet weight (or dose content). Furthermore, the RSD must be less than 6%. In some situations, the remainder of the 30 tablets can be evaluated to allow for a second opportunity to pass this test.
In this study, half-tablets were evaluated for their uniformity through the following steps:
For each of the 11 products, 30 tablets were selected and weighed using a Mettler AE 100 analytical balance (Mettler Toledo, Inc., Columbus, Ohio). The mean weight per tablet was calculated. For each product, the acceptable 85% to 115% range for a perfectly split tablet was calculated based on this mean weight. A 75% to 125% range was also calculated. A Microsoft Excel 97 spreadsheet was used to perform all calculations.
Of the 30 tablets, 10 were individually weighed. Each tablet was split, resulting in 20 fragments. Each half-tablet was weighed. The percentage dose loss was calculated from the weight of the resulting two tablet halves by comparing their sum against the original whole tablet weight. This percentage dose loss due to the splitting process was simply recorded and was not used in the uniformity test.
From the 20 half-tablets, the number outside the 85% to 115% range was counted. The number of half-tablets outside the 75% to 125% range was also counted. The RSD of the half-tablet weights was calculated. If at most one half-tablet was outside the 85% to 115% range (but within the 75% to 125% range), and if the RSD was 10% or less, they were considered to have passed this uniformity test.
If two split tablets were outside the 85% to 115% range (but within the 75% to 125% range), or if the RSD was more than 10.0%, the additional 20 tablets were split. To pass our uniformity test, none of the additional 40 half-tablets could be outside the 85% to 115% range, and the RSD for all 60 half-tablets had to be 10% or less.
If three or more of the 20 half-tablets were outside the 85% to 115% range, the product was considered to have failed this uniformity test. Also, if any half-tablets were outside the 75% to 125% range, they failed this uniformity test.
Hence, like the Uniformity of Dosage Units test in USP, the uniformity test here for half-tablets could fail a batch of split tablets due to any or all of three characteristics: too many half-tablets outside the 85% to 115% range, too many half- tablets outside the 75% to 125% range, or too high an RSD. It should be noted that a more liberal maximum acceptable RSD of 10% was used here, rather than the 6% specified in USP. Also, tablet weight, rather than chemical assay of actual drug, was used in assessing whether drug content was equal between the split portions. These two aspects facilitated passage of the uniformity test for split tablets. Relative to the USP test, the use here of 10% RSD (rather than 6%) and tablet weight (rather than drug assay) provides for a more liberal test and facilitates half-tablets to pass this uniformity test.
Results
Table 1 and Table 2 summarize the performance of the 11 products in the uniformity test when they were split using a razor blade. Table 3 reports the performance of three products that were split by hand.
Of the 11 tablet products evaluated, only 3 (Paxil 40 mg, Zestril 40 mg, and Zoloft 100 mg) passed the uniformity test (Table 1). Splitting these products produced no half-tablets with weights outside the target range of 85% to 100%. Each product also exhibited RSD less than 10%; Paxil 40 mg provided RSD less than 6%, which is the maximum acceptance level for whole tablets in the USP uniformity test. As evidenced by very low loss of tablet weight after tablet splitting (loss of less than 2% of the dose of any single tablet of these three products), these tablets did not crumble upon splitting. These products shared no common visible physical tablet characteristic that could explain their uniformity when split. While Zoloft 100 mg perhaps fits the preconceived profile for favorable splitting (scored tablet, oval or oblong in shape, and flat facial surface), Zestril 40 mg is not scored and is spherical (i.e., is round in shape and has rounded facial surfaces).
Eight products yielded divided tablets that failed the uniformity test (Table 2). Except for Lipitor 40 mg, each product's split tablets failed the uniformity test after the evaluation of 20 split tablets because of too many half-tablets with weights beyond the 85% to 115% acceptability range. Except for HydroDIURIL 50 mg, each product's half-tablets exhibited RSD greater than 10%. Moreover, glyburide 5 mg, Lipitor 40 mg, and Oretic 25 mg each failed since some half-tablets had weights beyond the 75% to 125% range. For all products, the dose loss secondary to splitting was relatively low, less than 3% on average. The highest dose loss from any individual tablet was 14.4% (Oretic 25 mg).
Like the three products that passed our uniformity test, these eight products shared no common visible physical tablet trait that could explain their failure. Comparing Table 1 and Table 2, no tablet characteristic such as scoring, tablet shape, or tablet surface flatness was predictive in differentiating whole tablets that split well from whole tablets that split poorly.
The characteristics of half-tablets split by hand are listed in Table 3. Overall, hand splitting yielded worse results than razor splitting. Many tablet halves were beyond the 85% to 115% acceptance range. All products had RSDs of more than 10%. Hand splitting did, however, result in very low dose loss (less than 0.5% on average). This reduced crumbling is unfortunate since it visually indicates that "cleaner" splits, relative to razor splitting, are obtained, which indeed was the case. Unfortunately, hand splitting did not generate uniform tablet halves, and the half-tablets generated using this method performed worse than razor-split tablets in our uniformity test.
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Discussion
The old pharmacy adage "right drug, right patient, right dose" articulates a basic goal of pharmacy practice. From a drug product quality perspective, selecting the right dose of a commercially available product is generally as simple as reading the strength on a product label, allowing pharmacists to concentrate their efforts on more complex clinical questions rather than questions of product quality. Pharmacists and patients expect -- and rely on -- high public health standards with regard to the manufacture and availability of quality pharmaceuticals. One such standard is the Uniformity of Dosage Units test published in USP, which addresses the issue of the right dose from a drug product quality perspective. This test assures that drug lots contain relatively uniform drug content from tablet to tablet, in agreement with the product label.
Because of efforts to reduce costs or achieve lower doses that are not produced commercially, patients are sometimes instructed to split tablets. For these patients, one would hope that "right drug, right patient, right dose" still applies, including the desire to provide an accurate and precise dose. For these patients, a half-tablet is essentially the dosage form. Since public standards for whole tablets are not immediately applicable for split tablets, an unresolved question is what should be the quality standard for split tablets. One can argue that the issue of tablet splitting is a pharmacy practice issue outside the scope of public standards. On the other hand, one can argue that a public standard for half-tablet uniformity is needed, given the potential role of insurance plans in promoting tablet splitting.
Our results strongly suggest that split tablets generally fail to meet expectations for content uniformity. While we would agree that the safety and efficacy profiles of many tablet products guard against complications from even twofold variations in dosing, we recommend against the general practice of splitting tablets. Although no public standard for half-tablet content uniformity exists, the split tablets we tested generally fell far below our expectations for content uniformity. With the involvement of pharmacists, prescribers, and patients, splitting tablets can certainly be appropriate in individual cases. However, given the public's expectations for high product standards (as exemplified by the standards for whole tablet content uniformity) and the results of our study, a broad institutional promotion of dividing higher-dose tablets to reduce costs appears to be more of a compromise of public standards than a practice prerogative. Of course, the public may indeed elect to exchange cost savings for a lower public health standard, and this issue merits public discussion. However, pharmacists, prescribers, and patients should not be excluded from this discussion under the banner of pharmacy practice, medical practice, or managed care, given current expectations for receiving the right dose. With the apparent propensity of tablets to split into portions that fail the content uniformity standard for whole tablets, the general practice of splitting tablets is not recommended unless the public's expectations for content uniformity are lowered.
Limitations
Our study was subject to two principal limitations. First, only one individual performed tablet splitting. Different results might be obtained if tablets were split by several individuals, and future research might focus on what individual factors contribute to successful tablet splitting. However, the individual who split the tablets in our study was a 25-year-old pharmacy student who had been specially trained to split tablets efficiently. Further, she performed this task in the presence of two other pharmacy professionals and in a controlled laboratory environment. We suggest that this situation represents a favorable scenario for products to be successfully split, relative to what may happen when patients split tablets. We do not believe that the majority of studied products failed the content uniformity test because one individual split all the tablets.
Second, the tablets evaluated were selected based on anecdotal reports that they are commonly split in practice. These products were not selected as a means of determining which types of products can or cannot be split, or as a means of predicting the performance of all tablets during splitting. In this regard, caution is needed in extrapolating the results of this study to any other product or group of products.
Conclusion
The majority of the 11 drug products we tested, when assessed for their ability to be split into half-tablets of equal dose, failed a liberally interpreted USP uniformity test. No visible tablet features (e.g., scoring) indicated whether a product's split tablets would pass or fail the uniformity test. The practice of dividing tablets to save costs or to improve a dosage regimen may not cause problems for patients using drugs with low toxicity and relatively flat dose-response relationships, but it is not recommended for patients using drugs with more substantial toxicity and steep dose-response efficacy curves.
Acknowledgements
To Alfred Abramson, BS, for donating tablets for this study, and to Catherine Cooke, PharmD, for her helpful suggestions of commonly split tablets.
Disclaimer
The authors declare no conflicts of interest or financial interests in any product or service mentioned in this article, including grants, employment, gifts, stock holdings, or honoraria.
Reprint Address
James E. Polli, PhD, School of Pharmacy, University of Maryland, 20 North Pine Street, Baltimore, MD 21201. Fax: 410-706-0346. E-mail: [email protected].
GS
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Posted 07/18/2002
Abstract and Introduction
Abstract
Objective. To divide 11 commonly split tablets and evaluate the resulting half-tablets for content uniformity.
Design. Pre-post comparison.
Setting. Laboratory. Interventions: A trained individual split tablets of 11 products using a single-edged razor blade and 3 products by hand alone.
Main Outcome Measures. The Uniformity of Dosage Units test published in the United States Pharmacopoeia 24 (USP), which applies to whole tablets, was adapted liberally to assess the dose uniformity of the resulting split tablets.
Results. Of the 11 razor-split products, 8 failed the liberal adaptation of the USP uniformity test. No visible tablet features (e.g., scoring) predisposed a product's split tablets to pass or fail the uniformity test. All three hand-split tablets failed the uniformity test and yielded worse results than did razor-split tablets.
Conclusion. The majority of the 11 drug products we tested, when assessed for their ability to be split into half-tablets of equal dose, failed a liberally interpreted USP uniformity test. The practice of dividing tablets to save costs or to improve a dosage regimen may not cause problems for patients using drugs with low toxicity and relatively flat dose-response relationships, but it is not recommended for patients using drugs with more substantial toxicity and steep dose-response efficacy curves.
Introduction
Pharmacists sometimes instruct patients to split tablets into halves or quarters in an effort to titrate doses or reduce costs. Some practitioner groups and insurance plans have promoted the splitting of double-strength tablets to reduce medication costs.[1]
Recommendations to split whole tablets into two halves rest on the assumption that the tablet will be split with sufficient accuracy. However, Stimpel et al.[2] showed that a majority of antihypertensive tablets did not break evenly, despite being scored. Sedrati et al.[3] identified several tablet products that, when split, resulted in half-tablets with doses outside an 85% to 115% range of the target half-dose. McDevitt et al.[4] evaluated the ability of healthy volunteers to split hydrochlorothiazide tablets. More than 40% of the resulting split tablets deviated by 10% or more from the target weight, and more than 10% of the split tablets deviated by at least 20%.
An important public health standard for prescription medications is the Uniformity of Dosage Units test in the United States Pharmacopeia 24 (USP).[5] This test requires relatively uniform drug content from tablet to tablet, essentially ensuring that the vast majority of individual tablets contain between 85% and 115% of the labeled dose while allowing for a 6% relative standard deviation (RSD) in drug content.
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Objectives
The objective of this study was to divide 11 commonly split tablets and evaluate whether the resulting half-tablets provided equal doses, based on a liberal adaptation of the USP Uniformity of Dosage Units test.
Methods
Selection of Tablet Products for Splitting
Eleven tablet products identified through anecdotal reports of which products are commonly split were selected and subjected to splitting:
Paxil 40 mg (lot 9258B13) (paroxetine -- GlaxoSmithKline).
Zestril 40 mg (lot CTB671) (lisinopril -- AstraZeneca).
Zoloft 100 mg (lot 9JP040A) (sertraline -- Pfizer).
Glucophage 850 mg (lot MAH93) (metformin -- Bristol-Myers Squibb).
Glyburide 5 mg (lot 109257A) (Novopharm).
HydroDIURIL 50 mg (lot D6811) (hydrochlorothiazide -- Merck).
Lipitor 20 mg (lot 09609V) and 40 mg (lot 0012108) (atorvastatin -- Parke-Davis; Pfizer).
Oretic 25 mg (lot 16053AF21) and 50 mg (lot 14884AF21) (hydrochlorothiazide -- Abbott).
Viagra 50 mg (lot 89R035A) (sildenafil -- Pfizer).
A single, trained individual divided tablets of each product using a single-edged razor blade. Each tablet was placed on a bench top, the individual held the razor blade, positioned the razor at the tablet surface, and pressed the razor into the tablet.
Additionally, three of the products' tablets were soft enough to split by hand alone: glyburide 5 mg, HydroDIURIL 50 mg, and Oretic 50 mg. These tablets were split by placing the tablet between each hand's forefinger and thumb and pushing the thumb through the tablet score.
Criteria for Split Tablet Performance: Uniformity Test
The Uniformity of Dosage Units test was adapted to assess the dose uniformity of resulting split tablets.[5] This test is intended to ensure the content uniformity of whole tablets and does not appear to apply to half-tablets. Briefly, the test calls for selection of 30 tablets from each batch of manufactured tablets. Of these 30 tablets, 10 are selected for analysis. Each tablet is weighed, and each must be within the 85% to 115% range of the target tablet weight (or dose content). Furthermore, the RSD must be less than 6%. In some situations, the remainder of the 30 tablets can be evaluated to allow for a second opportunity to pass this test.
In this study, half-tablets were evaluated for their uniformity through the following steps:
For each of the 11 products, 30 tablets were selected and weighed using a Mettler AE 100 analytical balance (Mettler Toledo, Inc., Columbus, Ohio). The mean weight per tablet was calculated. For each product, the acceptable 85% to 115% range for a perfectly split tablet was calculated based on this mean weight. A 75% to 125% range was also calculated. A Microsoft Excel 97 spreadsheet was used to perform all calculations.
Of the 30 tablets, 10 were individually weighed. Each tablet was split, resulting in 20 fragments. Each half-tablet was weighed. The percentage dose loss was calculated from the weight of the resulting two tablet halves by comparing their sum against the original whole tablet weight. This percentage dose loss due to the splitting process was simply recorded and was not used in the uniformity test.
From the 20 half-tablets, the number outside the 85% to 115% range was counted. The number of half-tablets outside the 75% to 125% range was also counted. The RSD of the half-tablet weights was calculated. If at most one half-tablet was outside the 85% to 115% range (but within the 75% to 125% range), and if the RSD was 10% or less, they were considered to have passed this uniformity test.
If two split tablets were outside the 85% to 115% range (but within the 75% to 125% range), or if the RSD was more than 10.0%, the additional 20 tablets were split. To pass our uniformity test, none of the additional 40 half-tablets could be outside the 85% to 115% range, and the RSD for all 60 half-tablets had to be 10% or less.
If three or more of the 20 half-tablets were outside the 85% to 115% range, the product was considered to have failed this uniformity test. Also, if any half-tablets were outside the 75% to 125% range, they failed this uniformity test.
Hence, like the Uniformity of Dosage Units test in USP, the uniformity test here for half-tablets could fail a batch of split tablets due to any or all of three characteristics: too many half-tablets outside the 85% to 115% range, too many half- tablets outside the 75% to 125% range, or too high an RSD. It should be noted that a more liberal maximum acceptable RSD of 10% was used here, rather than the 6% specified in USP. Also, tablet weight, rather than chemical assay of actual drug, was used in assessing whether drug content was equal between the split portions. These two aspects facilitated passage of the uniformity test for split tablets. Relative to the USP test, the use here of 10% RSD (rather than 6%) and tablet weight (rather than drug assay) provides for a more liberal test and facilitates half-tablets to pass this uniformity test.
Results
Table 1 and Table 2 summarize the performance of the 11 products in the uniformity test when they were split using a razor blade. Table 3 reports the performance of three products that were split by hand.
Of the 11 tablet products evaluated, only 3 (Paxil 40 mg, Zestril 40 mg, and Zoloft 100 mg) passed the uniformity test (Table 1). Splitting these products produced no half-tablets with weights outside the target range of 85% to 100%. Each product also exhibited RSD less than 10%; Paxil 40 mg provided RSD less than 6%, which is the maximum acceptance level for whole tablets in the USP uniformity test. As evidenced by very low loss of tablet weight after tablet splitting (loss of less than 2% of the dose of any single tablet of these three products), these tablets did not crumble upon splitting. These products shared no common visible physical tablet characteristic that could explain their uniformity when split. While Zoloft 100 mg perhaps fits the preconceived profile for favorable splitting (scored tablet, oval or oblong in shape, and flat facial surface), Zestril 40 mg is not scored and is spherical (i.e., is round in shape and has rounded facial surfaces).
Eight products yielded divided tablets that failed the uniformity test (Table 2). Except for Lipitor 40 mg, each product's split tablets failed the uniformity test after the evaluation of 20 split tablets because of too many half-tablets with weights beyond the 85% to 115% acceptability range. Except for HydroDIURIL 50 mg, each product's half-tablets exhibited RSD greater than 10%. Moreover, glyburide 5 mg, Lipitor 40 mg, and Oretic 25 mg each failed since some half-tablets had weights beyond the 75% to 125% range. For all products, the dose loss secondary to splitting was relatively low, less than 3% on average. The highest dose loss from any individual tablet was 14.4% (Oretic 25 mg).
Like the three products that passed our uniformity test, these eight products shared no common visible physical tablet trait that could explain their failure. Comparing Table 1 and Table 2, no tablet characteristic such as scoring, tablet shape, or tablet surface flatness was predictive in differentiating whole tablets that split well from whole tablets that split poorly.
The characteristics of half-tablets split by hand are listed in Table 3. Overall, hand splitting yielded worse results than razor splitting. Many tablet halves were beyond the 85% to 115% acceptance range. All products had RSDs of more than 10%. Hand splitting did, however, result in very low dose loss (less than 0.5% on average). This reduced crumbling is unfortunate since it visually indicates that "cleaner" splits, relative to razor splitting, are obtained, which indeed was the case. Unfortunately, hand splitting did not generate uniform tablet halves, and the half-tablets generated using this method performed worse than razor-split tablets in our uniformity test.
Lack of Medication Dose Uniformity in Commonly Split Tablets
from Journal of the American Pharmaceutical Association
Discussion
The old pharmacy adage "right drug, right patient, right dose" articulates a basic goal of pharmacy practice. From a drug product quality perspective, selecting the right dose of a commercially available product is generally as simple as reading the strength on a product label, allowing pharmacists to concentrate their efforts on more complex clinical questions rather than questions of product quality. Pharmacists and patients expect -- and rely on -- high public health standards with regard to the manufacture and availability of quality pharmaceuticals. One such standard is the Uniformity of Dosage Units test published in USP, which addresses the issue of the right dose from a drug product quality perspective. This test assures that drug lots contain relatively uniform drug content from tablet to tablet, in agreement with the product label.
Because of efforts to reduce costs or achieve lower doses that are not produced commercially, patients are sometimes instructed to split tablets. For these patients, one would hope that "right drug, right patient, right dose" still applies, including the desire to provide an accurate and precise dose. For these patients, a half-tablet is essentially the dosage form. Since public standards for whole tablets are not immediately applicable for split tablets, an unresolved question is what should be the quality standard for split tablets. One can argue that the issue of tablet splitting is a pharmacy practice issue outside the scope of public standards. On the other hand, one can argue that a public standard for half-tablet uniformity is needed, given the potential role of insurance plans in promoting tablet splitting.
Our results strongly suggest that split tablets generally fail to meet expectations for content uniformity. While we would agree that the safety and efficacy profiles of many tablet products guard against complications from even twofold variations in dosing, we recommend against the general practice of splitting tablets. Although no public standard for half-tablet content uniformity exists, the split tablets we tested generally fell far below our expectations for content uniformity. With the involvement of pharmacists, prescribers, and patients, splitting tablets can certainly be appropriate in individual cases. However, given the public's expectations for high product standards (as exemplified by the standards for whole tablet content uniformity) and the results of our study, a broad institutional promotion of dividing higher-dose tablets to reduce costs appears to be more of a compromise of public standards than a practice prerogative. Of course, the public may indeed elect to exchange cost savings for a lower public health standard, and this issue merits public discussion. However, pharmacists, prescribers, and patients should not be excluded from this discussion under the banner of pharmacy practice, medical practice, or managed care, given current expectations for receiving the right dose. With the apparent propensity of tablets to split into portions that fail the content uniformity standard for whole tablets, the general practice of splitting tablets is not recommended unless the public's expectations for content uniformity are lowered.
Limitations
Our study was subject to two principal limitations. First, only one individual performed tablet splitting. Different results might be obtained if tablets were split by several individuals, and future research might focus on what individual factors contribute to successful tablet splitting. However, the individual who split the tablets in our study was a 25-year-old pharmacy student who had been specially trained to split tablets efficiently. Further, she performed this task in the presence of two other pharmacy professionals and in a controlled laboratory environment. We suggest that this situation represents a favorable scenario for products to be successfully split, relative to what may happen when patients split tablets. We do not believe that the majority of studied products failed the content uniformity test because one individual split all the tablets.
Second, the tablets evaluated were selected based on anecdotal reports that they are commonly split in practice. These products were not selected as a means of determining which types of products can or cannot be split, or as a means of predicting the performance of all tablets during splitting. In this regard, caution is needed in extrapolating the results of this study to any other product or group of products.
Conclusion
The majority of the 11 drug products we tested, when assessed for their ability to be split into half-tablets of equal dose, failed a liberally interpreted USP uniformity test. No visible tablet features (e.g., scoring) indicated whether a product's split tablets would pass or fail the uniformity test. The practice of dividing tablets to save costs or to improve a dosage regimen may not cause problems for patients using drugs with low toxicity and relatively flat dose-response relationships, but it is not recommended for patients using drugs with more substantial toxicity and steep dose-response efficacy curves.
Acknowledgements
To Alfred Abramson, BS, for donating tablets for this study, and to Catherine Cooke, PharmD, for her helpful suggestions of commonly split tablets.
Disclaimer
The authors declare no conflicts of interest or financial interests in any product or service mentioned in this article, including grants, employment, gifts, stock holdings, or honoraria.
Reprint Address
James E. Polli, PhD, School of Pharmacy, University of Maryland, 20 North Pine Street, Baltimore, MD 21201. Fax: 410-706-0346. E-mail: [email protected].
