Hard to Swallow: A Review of Interventions to Improve Swallowing Solid Medication

We performed a comprehensive search of the PubMed, Medline (OVID), CINAHL, Scopus, and Web of Science databases to identify relevant studies. The PubMed search was conducted using (((Adolescent[MeSH] OR Child[MeSH] OR Infant[MeSH] OR Pediatrics[MeSH] OR adoles* OR child* OR paed* OR pediatr* OR teen* OR youth)) AND (Administration, Oral[MeSH] OR Capsules[MeSH] OR Tablets[MeSH] OR capsule* OR “oral medication” OR pill* OR tablet*)) AND (Deglutition[MeSH] OR deglutition OR swallow*). Search details for each additional database are included in Supplemental Information. Identified articles were reviewed first by title and abstract to determine relevance to our study objective, and then by full text to ensure included articles met our inclusion criteria, as written in the following section. Citations were exported to and managed in Microsoft Excel.

Figure 1 summarizes the total articles identified by our search and describes our process for determining which articles were to be included in the final evaluation. Included articles were published in English from January 2014 through April 2022. We excluded articles published before January 2014 because they would have been included in a previous review. Studies must include patients aged 0 to 21 years of age; studies with a mix of pediatric and adult patients were excluded because the majority of the sample population were adults. Study participants had difficulty swallowing solid oral medications without a comorbid condition affecting swallowing ability, such as dysphagia or severe developmental disabilities. We also included all types of interventions, from behavioral approaches to novel medication formulations.

Our primary objective was to evaluate the quality of each study because most previous studies were significantly limited by study design and small sample size. Reviewers graded each study as “poor,” “fair,” and “good” based on 3 categories: sampling strategy, study design, and the strength of outcome measures. Sampling strategy was considered stronger if steps were taken to limit sampling bias (eg, participants selected from multiple sites, variation in ages) and with a larger sample size. Study design was evaluated based on how clearly the intervention protocol was described, likely increasing the consistency in intervention delivery. Finally, outcome measures were graded higher if there were multiple interventions per patient, and higher still if there were multiple interventions per patient spread over time. The secondary outcome investigated if the study intervention was ultimately considered successful in improving patients’ ability to swallow solid oral medications.

Five reviewers with experience in pediatric hospital medicine and 1 reviewer with a background in pediatric infectious disease assigned a 3-point grading scale to each category of each study. This method was identical to the methods used by the systematic review evaluating literature before 2014.¹²  Reviewers assigned 0 points for a category considered “poor,” 1 point for “fair,” and 2 points for “good.” The scores for each category from a single reviewer were averaged to determine a composite quality score for the study. Finally, reviewers’ composite scores were averaged to determine an overall quality rating. Values calculated between 0 and 1.0 were assigned a “poor” rating, values of 1.0 to <1.5 were assigned a “fair” rating, and values ≥1.5 were assigned a “good” quality rating, indicating a study with minimal methodological limitations.

Our search results identified 1402 records, of which 821 were duplicates; an additional 545 articles were excluded after title and abstract review for lack of relevance to the review objective; full text of 2 records were unable to be obtained; 34 articles were reviewed by full text; and, ultimately, 10 studies included in the final review. These 10 articles varied in study design (1 randomized control study, 2 nonrandomized crossover studies, 1 randomized crossover study, 3 cohort studies, 1 case series, and 2 quality improvement initiatives), sample size (a small case series of 4 patients to a randomized control trial of 372 children), and population ages (1 included infants as young as aged 6 months, the maximum age studied was 18 years). Types of interventions also varied from behavioral therapies to supplementary products like a flavored throat spray and an in situ tablet or capsule coating to minitablets, which is a novel medication formulation. Table 1 summarizes the study year, type, sample size and study population, main intervention, and study outcomes.

Overall, each study found its investigated intervention to help patients successfully swallow solid medications. However, each study varied greatly in the type of intervention and methods described. El Edelbi et al found that the use of a thin, flavored coating for tablets or capsules improved both palatability (85% of participants) and ability (87% of participants), and no difference was found in the improvement rate between older and younger age groups.¹⁴  Harman et al reported on four 4-year-old patients with cancer who all were able to learn how to swallow tablets or capsules after behavioral interventions. The interventions included shaping and modeling techniques in one to six 15-minute sessions delivered by psychologists or child life therapists.¹⁵  Jacobsen et al⁷  used a quality improvement initiative to screen and identify children on the general inpatient service who had difficulty swallowing solid medications. Thirty-four children were screened and 6 identified for intervention; ultimately, only 3 children participated in behavioral interventions, and all were successfully discharged from the hospital with solid medications. At follow-up in 5 months, 2 of the 3 children were still able to swallow tablets or capsules.⁷  In a study of 25 children, aged 6 to 17 years, Jagani et al studied the effects of 2 weeks of standard behavioral therapy followed by the combined use of a flavored throat spray with the previously learned behavioral therapies for 1 week.¹⁶  Patients’ experiences were recorded using a 6-point numerical score or a facial hedonic score and then converted into a Medicine Taking Difficulty Score. Authors reported the use of the flavored throat spray decreased the Medicine Taking Difficulty Score by 0.93 points (P = .002). One limitation of the study was that only 10 of 25 children completed it.¹⁶ 

The largest study was a randomized open-label, 3-way crossover study of 374, done by Klingmann et al.¹⁷  The participants were divided into 2 groups by age: 6- to 23-month-old and 2- to 5-year-old children. Each group was administered 2 different doses of placebo minitablets and assessed for both noninferiority and superiority for acceptability and swallowability of the minitablets versus a glucose syrup. In the younger age group, superiority of acceptability was demonstrated for doses of 25 minitablets (P < .017) and 100 minitablets (P < .0001) and swallowability (both doses P < .0001). In the older group, noninferiority of acceptability was demonstrated for 400 minitablets (P ≤ .0003), and noninferiority of swallowability for 100 minitablets (P < .01).¹⁸  Another study of minitablets, by Kluk et al,¹⁸  investigated the acceptability of progressively larger dose units of minitablets administered in fruit jelly. They reported 83% of patients successfully swallowed several doses; however, only 57% successfully swallowed the largest dose without mastication before swallowing.¹⁸  Another large randomized crossover trial with 280 children aged 1 to 5 years was performed by Munch et al¹⁹  assessing acceptability of a 2.5 × 6-mm oblong tablet with direct comparison with either 3 mL of glucose syrup or three 2-mm minitablets. All formulations were administered with the child’s choice of liquid. Noninferiority of the oblong tablet was demonstrated in both arms with an acceptability rate difference of 4.29% (95% confidence interval [CI], –3.00 to 11.57) in the tablet-to-syrup group and a rate difference of –2.88% (95% CI, –7.31 to 1.55) in the tablet-to-minitablets group. Limitations of this study were that the study was not blinded, potentially introducing bias, and the authors depended on subjective ratings of acceptability to determine noninferiority.¹⁹ 

Rashed et al used quality improvement methods to implement an inpatient “pill school” in a children’s hospital in the United Kingdom.²⁰  Thirty children aged 3 to 14 years were trained to swallow solid oral medications via behavioral interventions and practice with small candies. Following the intervention, 86% (n = 26) were successful and 24 children were discharged from the hospital with medications in solid formulations.²⁰  Study authors used patients who were prescribed liquid formulations as a surrogate for difficulty swallowing solid oral medications. Because no screen confirming difficulty swallowing solid formulations was performed, the intervention success rate may be falsely elevated. In 2019, Tse et al also used quality improvement methods in a multidisciplinary renal clinic to incorporate behavioral training to teach children how to swallow tablets. Over the course of 3 months, they screened 90 children and successfully converted 25 children from liquid to tablet medications, with an estimated cost savings of £46 588.²¹ 

Finally, Zombori et al performed a retrospective study of a cohort of patients with perinatally acquired HIV receiving care from 1 tertiary center from 1995 to 2017.²²  Patients who had difficulty adhering to antiretroviral therapy were identified and divided into 2 groups based on the primary intervention for improving medication adherence. The first group received multidisciplinary support in addition to percutaneous endoscopic gastrostomy (PEG) placement. This group of 15 patients (median age, 17 years) had acceptable adherence to treatment as evidenced by low viral copies (<50 copies/mL) and a median CD4 count of 940 cells/µL). At the end of the study period, 9 of the 15 patients had their PEG removed and were taking tablet formulations. The second group of 7 patients (median age, 10 years) were identified in the final 7 months of the study period. In addition to multidisciplinary support, this group used a proprietary lubricating throat spray to aid in swallowing tablet formulations. At the end of the study period, all reported improved ease of swallowing tablets and all achieved/maintained suppressed viral loads.²²  Despite a presumed preference for a less-invasive intervention such as a lubricating throat spray, direct comparison is difficult between the 2 groups; significant bias is introduced by inclusion of patients over the lengthy study period because antiretroviral therapy regimens from the 1990s would not be comparable to modern regimens.

Reviewers rated the studies performed by Kluk et al, Munch et al, and Rashed et al as the only studies to receive quality rating of “good” with minimal limitations.^18–20  Five studies were rated as “fair,” and the small case study by Harmen et al and the retrospective study by Zombori et al were considered as “poor,” with considerable limitations.^15,22  All quality ratings are outlined in Table 2. Additionally, all studies showed their intervention to be successful to varying degrees in improving pediatric patients’ ability to swallow solid oral medications.

Since the last review of this topic covering literature published through 2013, only 10 additional articles were identified that met our inclusion criteria. These studies had a wide variability in the methods used to address the problem of pediatric patients who cannot swallow tablets or capsules, and most studies identified were small with notable limitations in study design. Despite these limitations, each intervention demonstrated some benefit for certain subgroups of pediatric patients. The wide variability in interventions suggests that although continued efforts have been made, no 1 intervention has been identified that fully addresses medication swallowing ability. Even with the existence of several different interventions, parents continue to struggle to get their children to take their medication, and we believe that pediatric providers have not sufficiently addressed this issue when prescribing medications.

To date, pediatric providers have little guidance as to which intervention to use because only 20 articles have been identified since 1969 that have reported on available methods.^12,13 

Some, but not all, interventions are appropriate for certain age groups. Behavioral techniques or products to coat medications or the throat are likely not the best option for young toddlers or infants, but minitablets have been used safely in infants as young as aged 6 months.¹⁷  Specific formulations such as minitablets or small oblong tablets are often not available for many commonly prescribed medications; however, these formulations have significant potential for pediatric drug development because of the flexibility that minitablets or other small tablets offer for weight-based dosing, as well as increased ease in transport or shelf-stability compared with liquid formulations (including compounded). Glecaprevir/pibrentasvir, a combination antiviral regimen for hepatitis C virus recently approved for children as young as 3 years old, offers a great example of minitablets offering flexibility for weight-based dosing.²³  Novel products or formulations may be the most appealing of the discussed interventions because they may require the least amount of time or individual training to implement; however, there have been no studies comparing the cost-effectiveness in terms of time or money. Experienced pediatricians often have suggestions for methods parents can try (taking the medication in a food such as pudding, ice cream, or peanut butter) or involving a child life therapist to provide training. However, no standard of care exists to routinely identify patients with difficulty taking tablets or capsules and to intervene as appropriate, although previous work has demonstrated the successful implementation of a screening process.^7,24 

Current evidence suggests that pediatric patients would benefit from implementation of universal screening to identify who would benefit from interventions to increase the ability to swallow solid oral medications. This skill becomes increasingly vital as evidence mounts for early transition from intravenous medication to oral medications (eg, the 2021 Guideline on Diagnosis and Management of Acute Hematogenous Osteomyelitis, the use of oral antibiotics for postdischarge treatment of complicated pneumonia, treatment with oral antibiotics for children with complicated appendicitis).^2–4  Implementation of universal screening, followed by application of a local or national guideline recommending appropriate patient-centered interventions along with outcome reporting and trending, should become a national quality benchmark for pediatric hospitals as an additional way to show institutional commitment to high-value care.

Our findings need to be viewed in the light of certain limitations. Although our project missed 2 elements of a true systematic review (PROSPERO registration, use of a validated rating system), our methods do represent the spirit of a rigorous literature review and still offer value to the reader as a useful inventory of the current state of taking oral medications. We searched the PubMed, Medline (OVID), CINAHL, Scopus, and Web of Science databases, but did not include psychology-based databases such as APA PsycNet, potentially excluding other studies that used behavioral therapies as the primary intervention. We limited our search to articles since the 2014 publication of a previous systematic review,¹²  but this article excluded infants and adolescents as part of their search criteria. Therefore, we may have missed studies published on those populations before 2013. We excluded studies of children with comorbidities that would affect swallowing ability and studies describing interventions in adults, thus possibly missing interventions that may be helpful in children. Also, we limited articles to those with full text availability in English, and we did not restrict our search based on sample size, resulting in inclusion of a small case series. As a result, each study had patients of different ages and drawn from different settings. This may make it difficult to generalize and apply these findings broadly. Finally, to mitigate rater bias, multiple reviewers independently critiqued each study, and the raters were averaged to mitigate the effect of bias.

Our review of the literature indicates that additional work remains to be done in this area of difficulty swallowing oral medication, including more studies to identify the most efficacious, long-lasting, or cost-effective intervention. The consistent improvement seen over many studies also suggest that next steps should include implementation of universal screening, development of a guideline to suggest the most effective strategy for the patient based on age, and incorporation of these into national benchmarks for quality and safety.