Are pediatrics taking the prescribed tablet dosage form? Practices of off-label tablet modification in pediatric wards: A prospective observational study

Abstract

Tablet manipulation (modification) is a practice where the licensed tablet product is customized into reduced fractions of doses or changed into dispersions using solvents because of suitable dose for paediatrics and neonates are frequently unavailable. As a result, unlicensed dosage form frequently used after manipulation, outside what is approved by drug regulatory authorities.

Objective

To assess the practice of off-label tablet manipulation in pediatric and neonatal wards at selected public hospitals in Ethiopia.

Methods

A prospective, direct observational approach was used to investigate the frequency, nature and appropriateness of tablet manipulations in neonatal and pediatrics patients of two public hospitals of Ethiopia from April 12, to June 30, 2021.

Results

During the study period a total of 303 tablet manipulations were observed. Two hundred nine (69%) of tablets were dispensed to pediatric patients to be taken after split into lower strengths. The remaining nighty four (31%) tablets were manipulated into dispersion using 0.9% normal saline as a main solvent. Interestingly, 48 (15.8%) of tablet manipulations into dispersions involved practically insoluble drugs whose manipulation may probably affect their bioavailability. In 12.5% (12/94) dispersion manipulations, large undissolved fractions were observed during administration through naso-gastric tubes. The most commonly manipulated tablets were drugs for central nervous system (n = 135, 44.6%) followed by cardiovascular drugs (n = 85, 28%).

Conclusions

and recommendations: The study revealed the off-label use of tablets for pediatrics age groups is very common in Ethiopia. Evidence-based guidelines for tablet manipulations should be practiced to improve the safety of paediatric drug use. As for policy implication, this study concurs with previous scientific recommendations that manufacturers introduce a wide range of dosage forms to reduce the need for manipulations.

Abbreviations

AMCP

American Academy of managing care pharmacy

API

Active Pharmaceutical Ingredient

BCS

Biopharmaceutical Classification System

DTC

Drug and Therapeutic Committee

EP

European Pharmacopoeia

SUPAC

Scale-up and Post-Approval Changes

SOPs

Standard Operating Procedures

US FDA

United States Food and Drug Administration

USP

United States Pharmacopoeia

WHO

World Health Organization

1. Background

Tablet manipulation is a practice where the licensed tablet product is customized into reduced fractions of doses or changed into dispersions using solvents. This is because of suitable dose for paediatrics and neonates are frequently unavailable [1]. As a result, unlicensed dosage form frequently used after manipulation, outside what is approved by drug regulatory authorities [2]. Tablet dosage forms may not needed at all in some situations, especially for neonates and paediatrics with the presence of naso-gastric tube, as they are unable to take whole tablets or capsules [3]. Consequently, the off-label manipulation of medicinal products to extract a part of the whole dosage form, to facilitate administration or to improve acceptability is a common practice in paediatric drug therapy setting where limited dosage regimens are available [[4], [5], [6]]. As a result, the practice remains an important public health issue in both developed and developing countries [2] and seems to increase, if responsible bodies do not give attention.

Manipulation goes further than splitting as it includes dispersion of commercially available tablet products [7,8]. The patients most affected by this practice are children. Many medicines given to children are dosage forms designed for adults. The magnitude of doses required throughout childhood can vary up to 100 folds [9]. Medicines are thus used by the physical alteration of the dosage form with the aim of achieving the required (usually smaller) dose for administration. Examples include splitting a tablet and administering a portion or further dilution at the aim of getting optimum pediatric dose. However, in order to investigate drug manipulation accurately, the processes that occur in practice need to be explored [10].

Although commonly acknowledged among professionals as a widespread practice, reports about tablet manipulations with the aim of achieving the required dose are limited. Manipulations may be time consuming, can be inaccurate, and have unknown effects on the stability and bioavailability of drugs [11,12]. Drug manipulations may also increase the risk of errors. Dose calculation errors are the most common medication errors in neonatal and pediatric practice. This includes calculation of the amount of liquid dispersion needed after a tablet is manipulated into dispersions or unequal split fragments [13]. An Australian survey of solid dosage form modifications found that, for children, on 60 (82%) of occasions, the inability to swallow, and in 14% occasions, because the correct dose was not commercially available, were reasons for modifications [14].

This study aimed at determining the overall tablet manipulation practices at paediatric wards of two hospitals in Ethiopia and providing ideas for potential interventions that would help improve these practices. Moreover, there have not been similar studies in Ethiopia; the findings of this study will serve as a foundation for further studies.

2. Methods

2.1. Study design and setting

A prospective observational study was conducted at pediatrics and neonatal wards of two referral public hospitals in Ethiopia. Observation was conducted in blocks of three weeks for each hospital (from April 12 to June 30, 2021). A trained pharmacist did the observation. During each three-week block, potential manipulations were identified prospectively via daily observation for 8 h. Since there were more than one pediatric wards in each hospital, nurses were requested to call the observer when they identified tablet manipulation for administration of medications. Where manipulations were identified, attempts were made to observe the manipulations occurring in practice. Observational findings and patients’ demographic data were documented in an observational tool. The observer was as unobtrusive as possible while collecting data to document the actual tablet manipulation practice.

In the absence of an evidence-base for this practice in the area, the sample frame was designed to sample maximum variation by observing as many different types of drug manipulations as possible. This led to targeted observations, or purposive sampling. A pediatric patient fulfilling the inclusion criteria were included during a study period.

Both hospitals have two pediatric wards (infectious and non-infectious disease ward and malnutrition ward).

2.2. Inclusion and exclusion criteria

Non-tablet manipulations were not included in the study. All patients less than 18 years and eligible to take their medication after manipulation were included.

2.3. Operational definitions

Tablet manipulation into dispersion: is defined as all processes related to the alteration of individually prescribed tablet dosage forms into a liquid using a solvent prior to administration.

Appropriate manipulation: is defined as the manufacturer supports crushing or splitting a tablet for getting an optimum dose/dosage form. These tablets should have score-line for splitting and/or have a label supporting manipulation.

Inappropriate manipulation: when splitting or crushing into dispersion is performed in unscored tablets, coated tablets or tablets having a label claim against manipulation.

Biopharmaceutics classification system (BCS): is defined as a system classifying a drug substance (API) based on its minimum aqueous solubility in the pH range of 1–7.5, dose and human fraction absorbed or intestinal membrane permeability into class I (soluble and permeable) class II (insoluble but permeable), class III (soluble but impermeable) and class IV (both insoluble and impermeable).

2.4. Data analysis

Duplicate single tablet manipulations with identical patients were excluded from the data analysis to avoid biases by patients with an extended length of hospital stay. If a single drug was administered more than once to the same patient, the manipulation was recorded, only once, to obtain maximum variation and avoid biases.

The prevalence and type of tablet manipulations were calculated in different pediatric age groups. Pediatrics age grouping was done according to WHO age classifications for pediatrics. Accordingly, the age stages defined for pediatrics is classified neonates (0–1 month), infants (1 month- 2 years), Children (2–12 years) and Adolescent (12–18 years) [15].

World Health Organization (WHO) Model list of Essential Medicines according to the biopharmaceutics classification system (BCS) and pharmacopoeial monographs were used to classify dispersed tablets in terms of their solubility characteristics.

For statistical analysis, SPSS Statistics Version 22 was used. Differences between the groups of pediatric patients with appropriate manipulation and inappropriate manipulation were calculated using the Mann-Whitney U test for independent samples. The correlation between the age of the pediatric patients and the type of manipulations performed was done using Pearson's bivariate regression at a significance level of α = 0.05 (two-tailed) for all tests.

3. Results

During a study period, a total of 303 tablet manipulations were observed in a total of 275 pediatric patients. Manipulations are either splitting (cutting) of tablets into fragments or converting tablets into dispersion using a solvent. Two hundred nine (69%) of tablets were dispensed to pediatric patients to be taken after split into lower fragments (Table 1). The remaining nighty four (31%) tablets were manipulated into dispersion before administration. Interestingly, 48 (15.8%) of tablet manipulations into dispersions involved practically insoluble drugs whose manipulation may probably affect their bioavailability. Among dispersed manipulations, in 12.5% (12/94) large undissolved fractions were observed during administration through naso-gastric tubes. The most commonly manipulated tablets were drugs for central nervous system (n = 135, 44.6%) followed by cardiovascular drugs (n = 85, 28%). In 87 dispersed manipulations, 0.9% normal saline was used as a solvent for dispersing the drugs before administration in pediatric and neonatal wards. The remaining seven dispersed manipulations were performed using distilled water. Types of tablet manipulation undertook has statistically significant association with pediatrics age groups (p = 0.03). This is to mean that in neonates tablets were manipulated into dispersions (in all 5 occasions) while in adolescent age group tablets were mostly manipulated into lower splits (40 out of 59 manipulations are splitting and the reaming 19 were manipulations into dispersions).

Table 1.

Descriptive analysis of patient age groups and frequency of tablet manipulations (Man.).

Patient age Group	No of patient with Man. into Dis.	No of Man. into Dispersion	No of patient with splitting	No of splitting	Total No of Man.
Neonates (0–1 month)	4 (1.5%)	5 (1.7%)	0	0	5 (1.7%)
Infants (1 month- 2 years	37 (13.5%)	42 (13.9%)	7 (2.5%)	9 (3%)	51 (16.8%)
Children (2–12 years)	28 (10.2%)	28 (9.2%)	145 (52.7%)	160 (5.8%)	188 (62%)
Adolescent (12–18 years)	17 (6.2%	19 (6.3%)	37 (13.5%)	40 (13.2%)	59 (19.5%)
Total	86 (31.3%)	94 (31%)	189 (68.7%)	209 (69%)	303 (100%)

During the observation days, 488 tablet dosage forms were dispensed to pediatric age groups in both ward units. The prevalence of tablet manipulation was therefore 62% (303/488) in pediatrics wards. The frequency of manipulation was highest in schoolchildren (2–12 years) with 62%, followed by adolescents (12–18 years) with 19.5% and infants (1 month −2 years) with 16.8% (P < 0.05, Table 1).

3.1. Manipulations of tablets into dispersion

Nurses, midwives or intern medical doctors undertook manipulations of tablets into dispersions. Narrow therapeutic window drugs like phenytoin and phenobarbital were commonly manipulated, especially in neonatal wards. Nearly 45% of tablet manipulations into dispersion were undertook in infant age groups. Fig. 1.

Fig. 1.

Recorded tablet manipulations into dispersion for different pediatrics age groups.

The active pharmaceutical ingredients frequently observed during manipulation into dispersion are shown in Fig. 2. Of these manipulated substances, 48 manipulations (51%) were practically insoluble or only very slightly soluble according to the European Pharmacopoeia [16] as shown in Table 2. Obviously, manipulation of such medicines into dispersion may affect their bioavailability. Two of the substances listed were enteric-coated tablets, valproic acid and chlorpromazine, whose delayed release functions are lost by manipulation of the tablets.

Fig. 2.

Recorded frequency of tablet dosage forms manipulation into dispersion.

Table 2.

Tablet dosage forms that were manipulated into dispersion in pediatric and neonatal wards and their solubility profile (BCS classification).

Drugs (tablet form)	Dose	Frequency of manipulations into dispersion	Solubility	BCS class
Phenytoin	50 mg	19	Low	II
Phenytoin	100 mg	12	Low	II
Phenobarbital Sodium	30 mg	15	High	I
Phenobarbital Sodium	100 mg	9	High	I
Carbamazepine	200 mg	5	Low	II
Spironolactone	25 mg	5	Low	II/IV*
Furosemide	40 mg	6	Low	IV
Levothyroxine	0.05 mg	1	High	III
Sodium valproate	200 mg	4	EC	EC
Zinc phosphate	20 mg	3	Low	NA
Prednisolone	5 mg	3	High	I
Warfarin	5 mg	1	High	I
Artemether + Lumefantrine	20 mg + 120 mg	2	Low	II/IV*
Albendazole	400 mg	4	Low	II/IV*
Diazepam	5 mg	3	High	I
Chlorpromazine	100 mg	2	EC	EC

EC = enteric coated tablet, NA = not well analyzed/insufficient data, * inconclusive data.

References for BCS classification [[16], [17], [18], [19], [20]].

The entire tablet was typically given crushed and dispersed in liquid or the intact tablet is soaked in the dispersing liquid without prior crushing. The reason for manipulation of some of the drugs, for example, furosemide was due to the lack of other dosage forms like injectables during the study period. There was a separate room for medication admixture and manipulation in the first hospital. However, manipulation were undertook at bedside in the other hospital. In all cases of observed manipulations into dispersion at the first hospital, the intact tablets were soaked in the solvent without crushing or splitting. The amount of solvent (normal saline solution or distilled water) used for manipulation depended only on the dose of the drugs, and not on their solubility. Whether the drug is highly soluble or not, the ratio of solvent to drug is 1 ml: 10 mg.

Perhaps, the amount of solvent used for manipulation in the second hospital was not dependent on the dose and type of tablet. Whether the drug is large dose or small dose, the amount of solvent used to disperse the drug was similar, 10 ml in all manipulations. So, when the drug is poorly soluble, undissolved crystals were clearly seen within the syringe. In two occasions, after the drug was dispersed, the dose directly administered into the naso-gastric tube without removing the needle of the syringe. This obviously is not a good manipulation practice as the insoluble crystals in the heterogeneous mixture may not pass through the narrow needle, and as a result, the patients may not have received the accurate titrated doses.

BCS class II, class IV drugs are poorly soluble and are difficult to manipulate into dispersions using normal saline, and water as it is generally difficult to get homogeneous solutions from such drugs. Accordingly, attempt was made to correlate the BCS classes of the manipulated drugs and the presence of undissolved fractions in the manipulated dispersions as shown in Table 2.

During the observation, a single manipulation was used for a single administration and the remaining fraction was discarded due to potential stability problems. In many cases of the dispersed manipulations, only fractions of the original dispersed dosage form were administered. In neonates in all manipulations, only portions of the dispersion were given. However, in many occasions the original dosage (full dose) was given after manipulation for adolescents (P < 0.05). (Fig. 3).

Fig. 3.

Manipulations and proportions of the dispersion given to different paediatric age stages.

In 12 (12.5%) manipulations, large undissolved fractions were seen during administration through the naso-gastric tubes (see Fig. 4). This will have undesired effect to deliver the optimum dose to patients because of the concentration of the drug varies in different depth of the container. Except phenobarbital sodium, all the drugs were BCS class II or IV, poorly soluble.

Fig. 4.

Proportion of undissolved fractions seen in naso-gastric tube while administering the dispersions.

3.2. Reasons and appropriateness of manipulation of tablets

Inappropriate dosage strength was the main reason for medication splitting. However, absence of suitable dosage form was the frequent reason for manipulation into dispersion. Reasons for manipulation has statistically significant association with pediatrics age groups (p = 0.01). In younger pediatrics, absence of suitable dosage form is the commonest reason for manipulation.

Some caregivers were split their medication by biting apart. Tablet dosage forms, which are not appropriate for splitting, were dispensed to patients to be taken after split. For example, enteric-coated products were dispensed to be split into lower strengths. Nurses or midwife professionals may not aware of the fact that these products are not candidates for splitting.

Overall, 63% of tablet medications were not suitable (unacceptable) candidate for manipulation (Table 3). Absence of score-line, coating and manufacturers’ restriction are considered to preclude from manipulating these tablet products.

Table 3.

Comparison of patients hospital stay when they are in appropriate and non-appropriate tablet manipulations (Mann-Whitney U test independent samples).

	Patient with Acceptable Manipulation					Patient with Unacceptable manipulation
	Mean	SD	Median	Min.	Max.	Mean	SD	Median	Min.	Max.	P-value
Age (years)	6.4	5.3	4.0	0.1	17.8	6.0	4.7	3.8	0.1	17.9	0.004
Weight (kg)	17.4	15.3	11.2	3	38	17.3	16.0	11.0	2.8	37	0.06
Duration of Hosp. stay in days	5.2	2.5	4.3	1.0	30	8.3	4.0	5.9	1.0	32	0.003

4. Discussion

Tablet manipulation is commonly practiced in health settings, but has rarely been explored especially in developing countries. Frequency of manipulations are expected to be higher in resource-limited settings because of less access to wide range of dosage forms. During manipulation, a large share of the patients have encountered errors. This and many previous studies have shown patients encountered problems like non-uniform splitting and taking over or under dose, non-adherence and forgetting splitting although counselled to take split fragments [21,22].

In this study, normal saline (0.9%) and sometimes, distilled water were used as a solvent for dissolution. The use of 0.9% normal saline solution as the main solvent for manipulation is considered problematic as ingestion of sodium chloride is restricted in some cardio-vascular diseases [23]. On the other hand, the presence of this electrolyte could decrease the solubility due possibly to salting out effect [24] and result in significant undissolved fractions especially when BCS class II/IV drugs are manipulated. It is understood that highly variable dosing may occur when insoluble drugs especially BCS class II/IV drugs are dispersed in water. For instance, in few occasions, partially soluble phenytoin were administered to get immediate response when patients were experiencing severe seizures and in need of an immediate treatment. In these cases, the patients probably did not receive the right doses as the partially soluble portion of the dose could result in inadequate pharmacological response.

The majority of medications intended for manipulation were dispensed for continuous use in the treatment of chronic diseases. According to a study in Germany, anti-epileptics were most commonly manipulated (24.9%), followed by vitamins (19.8%) and drugs for acid-related disorders (12.7%) and manipulations were common in infant age groups [25]. Similarly, in the current study, anti-epileptic drugs were commonly manipulated. Tablets used for the treatment of central nervous system were the most commonly manipulated products.

In two occasions, significant fractions of the undissolved tablets were seen in patients with enteral feeding tubes or nasogastric tubes and these fractions were irrigated by flashing normal saline into the feeding tubes. This practice is commendable as it ensures the drug to reach the gastrointestinal tract for absorption.

According to the FDA guidance for the industry ‘Waiver of in vivo bioavailability and bioequivalence studies for immediate-release solid oral dosage forms based on a biopharmaceutics classification system (BCS)’, a biowaiver can currently be requested only for solid, orally administered immediate-release products (>85% release in 30 min), containing drugs with a high solubility over the pH range from 1 to 7.5 (dose/solubility ratio of less than 250 ml and a high permeability fraction absorbed >90%) [26]. Based on this, BCS class II, class IV drugs are poorly soluble and are difficult to manipulate into dispersions using normal saline, and water as it is generally difficult to get homogeneous solutions from such drugs. Accordingly, in this study attempt was made to correlate the BCS classes of the manipulated drugs and the presence of undissolved fractions in the manipulated dispersions. More than half (51%) of the manipulated drugs in the current study were slightly soluble or insoluble (BCS class II, IV) in water and undissolved crystals were observed. This resulted in inconsistent doses. A study by Broadhurst et al., showed even dispersible tablets can yield inconsistent doses when withdrawn from different depths of a container after manipulation particularly when the drug is poorly soluble [27]. Another study done by Richey et al. showed that 44% of manipulated tablets were slightly soluble or totally insoluble in water [10].

The current study demonstrates that manipulations of dosage forms are integral activities to pediatric practice. However, it is not entirely clear who is responsible for the manipulations; different health professionals performed the manipulations including nurses, midwives and medical doctors. Sometimes, nurses make a professional judgment to meet a patient's preference when a dosage form is available. However, since there were no clear standard operating practices and guidelines on selecting candidate drugs for manipulation, and persons assigned to undertake manipulations, errors are expected to arise frequently. Manipulation of tablets into dispersions needs expertise as several steps must be followed before the actual administration, including interpreting medication labels, converting measurement units within a system or between systems, calculating medication doses, identifying the physicochemical properties and dosage form type of a given tablet and its solubility or its BCS class. All of these require expertise in the field like pharmacists or alternatively, the other health professionals will require additional on the job training. In addition, appropriate working areas equipped with appropriate materials and equipment are required to minimize errors encountered during manipulation. Unfortunately, most of the hospital staff including the hospital management, medical doctors, nurses and pharmacists did not give adequate attention to the manipulation practices. The problems encountered as well as the consequences on health outcomes of patients as a result of these poor manipulation practices are underestimated. Regarding training, none of the hospital staffs had ever taken on-the-job training on tablet manipulation.

4.1. Strengths and limitations

This was the first study investigating the type and nature of tablet manipulations into dispersion in pediatric and neonatal wards in Ethiopia. This study tried to see the scope and nature of wide range manipulations in different pediatric age groups. The study duration of six weeks was comparable and noticeably longer than other studies [10,28,29]. Thus, the results can be considered reliable and close to the reality for the described age groups. However, this study did not evaluate clinical outcomes following the administration of the manipulated medicines. The investigated dosage forms were only tablet dosage forms. If other dosage forms were included the frequency of manipulations might have been higher. Convenience sampling was employed and not every manipulation was observed. As a result, the problem could be underestimated.

5. Conclusions and recommendations

This study revealed that the tablet manipulation practices were common. To the best of our knowledge, no available guidelines regulate this practice in Ethiopia. Although manipulations are practiced commonly especially on drugs that are indicated for chronic ailments, the hospitals did not give adequate attention to this practice.

Adequate training to pharmacists and other healthcare professionals about tablet manipulation (both pre-service and in-service trainings) is needed. The government could also support the local pharmaceutical manufacturing sector so that the local industries will increase local production of a wide range of dosage forms, including pediatric formulations, to minimize the need for tablet manipulations.

On individual level, it is crucial to create a spirit of team work between pharmacists, prescribers and other healthcare team members to share all the required information for a tablet to be manipulated. The practice of tablet manipulation can lead to drug wastage and medication non-adherence. As the findings of this study showed, involvement of pharmacists in tablet manipulations in pediatric wards is minimal and it may increase the manipulation errors. Pharmacists are highly encouraged to educate healthcare providers and patients as to when tablet manipulation is an acceptable practice.

In situations where a patient is unable to get the required dose of the prescribed tablets, a different brand of the same product or a different dosage form like solution, injectable or suspensions could be more appropriate. When no alternatives are available, therapeutic substitution with an alternative that is available in an appropriate strength may sometimes be an option. The stability of the manipulated tablets should also be guaranteed. Further studies are needed to see the clinical consequences of the current tablet manipulation practices.

Ethical consideration

The study was performed in accordance with the ethical standards stipulated in the Declaration of Helsinki as revised in 2013. Ethical approval was obtained from the research and ethics review committee of the School of Pharmacy, University of Gondar. The Medical Directors of the selected hospitals were contacted and informed about the purpose of the study. The observer obtained verbal consent from patients before the commencement of the study. They were informed that the information provided would be kept confidential and analyzed in aggregate.

Funding source

The author's received no specific funding for this work.

Author contribution statement

Asmamaw Emagn Kasahun: Conceived and designed the experiments; Analyzed and interpreted the data; Wrote the paper.

Ashenafi Kibret Sendekie: Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Data availability statement

Data will be made available on request.

Declaration of competing interest

We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no financial support for this work that could have influenced its outcome. No funding was received for this work.