Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Research report

Medication errors in patients with dysphagia

  • Comment

A study examined whether medication errors were more common in acute hospital patients with dysphagia than those without, and the type and cause of such errors.


Background The administration of medicines to patients with dysphagia is complex and prone to errors.

Aim To determine the medicine administration error (MAE) rate for those with dysphagia in acute hospitals.

Method Data was collected on the preparation and administration of oral medicines to patients with and without dysphagia, including those with enteral feeding tubes, using direct observation of 65 nurse-led medicine administration rounds at four acute general hospitals.

Results Excluding time errors, the normalised frequency of MAEs for patients with dysphagia was 21.1% compared with 5.9% for those without the condition.

Conclusion The increased MAE rate in patients with dysphagia means extra care is needed when prescribing, dispensing and administering medicines to this group.

Citation: Kelly J et al (2012) Medication errors in patients with dysphagia. Nursing Times [online]; 108: 21, 12-14.

Authors: Jennifer Kelly is lead tissue viability nurse, Queen Elizabeth Hospital King’s Lynn Foundation Trust; David Wright is professor of pharmacy practice, and John Wood is lecturer, both at the School of Pharmacy, University of East Anglia.

Older people comprise 20% of the population and take 50% of prescribed medicines (Rajaei-Dehkordi and McPherson, 1997). If they suffer from swallowing difficulties (dysphagia), they may have difficulty taking medication (Kelly et al, 2010; 2009). A study in a psychiatric hospital showed that medicine administration errors (MAEs) were more common in patients with dysphagia than those without (37.3% versus 20.1%; p<0.0001) (Haw et al, 2007). This finding and the concern of the National Patient Safety Agency (2009) prompted our study to determine the MAE rate for this group in acute hospitals and whether it was affected by the care setting.


The study used a convenience sample in one ward for older people and one stroke ward at each of four hospitals in the east of England between March and June 2008.

Using a standardised pro forma, a nurse observed eight drug rounds on each of the eight wards, giving a total of 64 drug rounds split equally between early morning and lunchtime. Information was collected on the oral medicines administered including dose, formulation, preparation and administration. Tablet crushing, capsule opening, tablet dispersion, adding to food and consistency/texture of liquid medications (including where thickeners were added) were recorded. Patients with and without dysphagia, including those with enteral tubes, were observed.

The observer intervened in incidents where there was potential for patient harm; these were recorded as errors. The errors were categorised using an adapted version of the American Society of Hospital Pharmacists (1993) classification.

The data set was analysed using a combination of formal statistical modelling and summary statistics. Most of the latter focused on the MAE rate, which was calculated from the number of administrations with one or more errors divided by the total opportunities for error (the sum of the number of observed drug administrations plus the number of omitted drug administrations), calculated as a percentage.


Including the pilot, 65 drug rounds were observed, involving 62 nurses. All nurses (mainly in bands 5 and 6) took part; three were observed twice. The results of a questionnaire they completed have been published previously (Kelly and Wright, 2009).

Sixteen drug rounds were observed in each hospital plus the stroke unit pilot (Table 1); 33 rounds took place in the early morning and 32 at lunchtime, split equally between weekdays and weekends. Some 2,129 drug administrations to 625 patients were observed, of which 817 (38.4%) were deemed MAEs (Table 2). The most common error was giving medicine over an hour early or, more usually, over an hour late.

Of the 625 patients, 214 (34.2%) had dysphagia, and accounted for 679 (31.9%) of drug administrations; 170 patients without dysphagia and 133 with the condition experienced at least one MAE; and 36 of the 50 patients with enteral tubes experienced at least one error. There were 313 MAEs to patients with dysphagia and 504 to those without (Table 2). After normalising, the frequency of MAEs for patients with dysphagia was 46.1% (95% confidence interval 42.3 to 49.8) compared with 34.8% (95% CI 32.3 to 37.3) for those without. Excluding time errors, the MAE rate for patients with dysphagia was 21.1% compared with 5.9% for those without.

Description of errors

Table 2 shows time errors to be more common in the non-dysphagic group. However, this is largely because other errors are less frequent. There is also a masking effect because each medicine was only recorded as having one error, and time errors were recorded only if there was no other error. For example, where an incorrect formulation was given late, the error was recorded as “wrong formulation”.

The main cause of MAEs for patients with dysphagia was wrong formulation and incorrect preparation of medicines. The wrong formulation category included eight incidences of patients inappropriately chewing medicines such as modified-release dipyridamole (six patients chewed this, just fewer than 1% of patients observed). Tablets were crushed in 24 cases, all without the prescriber’s written authorisation; here, a liquid formulation could have been dispensed in seven cases and, in 19 cases, the tablets could have been more appropriately dispersed in water (White and Bradnam, 2006).

The 65 preparation errors were: two cases of incorrect tablet breaking; one case of measuring 2.5ml morphine oral solution with a measuring cup rather than a syringe; two cases of using a luer lock syringe to administer enteral drugs contrary to NPSA guidelines (2007); 34 cases of enteral tubes not being flushed before or after administration; and 26 cases of mixing medicines when crushing and dispersing them.

Statistical modelling found no evidence of an association between patients with dysphagia and time errors, so time errors were ignored in subsequent analyses (Table 3).

No statistically significant difference was found between lunchtime and breakfast rounds, different days of the week, or between the stroke and care-of-the-elderly wards in terms of the risk of MAEs for patients with dysphagia.


This study found that MAEs were higher in patients with dysphagia in four acute hospitals, largely due to errors of drug formulation or preparation. It also identified an unrecognised group of patients who, although not diagnosed with dysphagia, chew their medicines and are therefore at risk of adverse effects. These patients need to be identified and a risk assessment of the medicines being chewed performed.

To facilitate medicine administration to patients with dysphagia, nurses crushed medicines even though licensed alternatives were available, or failed to disperse them when this was recommended (White and Bradnam, 2006). Guidelines suggest dispersal as an alternative to crushing, but both approaches can render the administration outside the medication’s licence and may cause adverse effects. This may also mean patients experience any unpleasant taste medications may have, which could affect medicine-taking behaviour post discharge (Shukla et al, 2009).

Only independent and supplementary prescribers are allowed to sanction unlicensed medicine use under the Medicines Act 1968. Administering an unlicensed medicine passes the liability for harm from the manufacturer to the prescriber, supplier and administrator, depending on who has sanctioned the action and who is aware of the medicine being administered in an unlicensed fashion (Griffith, 2005).

Although no slow-release preparations were identified as being inappropriately crushed, medicines with enteric coats, such as lansoprazole, and those with small therapeutic windows, such as digoxin, were identified as being crushed before administration. Digoxin tablets have a bioavailability of 0.7 (Lisalo, 1977) and crushing may increase this to 1, causing an almost 50% increase in dosage absorbed; crushing without monitoring could therefore reduce patient safety (Stubbs et al, 2007).

Preparation errors accounted for patients with enteral tubes having a greater risk of MAEs than those with dysphagia without tubes. This was due to the additional complexity of administration via this route (van den Bemt et al, 2007). Feeds need to be turned on and off and the tubes flushed before and after medicines are given, using an enteral syringe (British Pharmaceutical Nutrition Group, undated; NPSA, 2007). Where this is not undertaken, medicines can interact with each other, the feed or the tube, which can consequently block the tube (White and Bradnam, 2006).

Another preparation error was mixing different tablets before or after crushing them to speed up administration. This may illustrate nurses’ limited knowledge of dosage formulation and the considerable care taken by drug manufacturers to select excipients within a single formulation that do not affect the active ingredient.

Lack of knowledge is frequently given as a reason for MAEs (Tissot et al, 1999). This is supported by a qualitative study exploring the issues for nurses of altering medication dose forms before administration to care home residents. This found nurses felt conflict between the need to make sure residents received medication and uncertainty on how to do this safely (Barnes et al, 2006). A survey of 71 doctors and 104 nurses on the suitability of prescribing and administering nine medicines via enteral tubes found that both nurses and doctors had limited awareness of the suitability of medications for this route (Lonergan et al, 2009).

Stroke units, in line with standard five of the National Service Framework for Older People (Department of Health, 2001), have been set up, employing specialist stroke teams. It might be expected that management of medicine administration in relation to dysphagia would be of a higher standard here than on wards for the care of older people. Although organisational differences were noted between the two unit types in this study, they did not appear to affect the MAE rate.

Study limitations

This study is the first to concentrate on the MAE rate in patients with dysphagia. Although attempts were made to reduce observer bias through the use of structured observation, it can never be eliminated and this could have affected the numbers and types of errors recorded.

Additionally, caution is needed in interpreting the “negative” inferences from the study, where these are based on relatively low replication. In particular, the comparison of error rates between types of ward, while based on observation of a lot of administrations, is limited by the replication at “ward level” to just four stroke units against four care-of-the-elderly wards.

The study took place in hospitals in East Anglia so the findings may not apply to hospitals elsewhere or community settings.


This study found strong evidence of a marked rise in the risk of MAEs for patients with dysphagia, with a further increase if they have an enteral tube. Health professionals therefore need to take extra care when prescribing, dispensing and administering medicines to these patients.

  • This article is a summary of: Kelly J et al (2011) Medicine administration errors in patients with dysphagia in secondary care: a multi-centre observational study. Journal of Advanced Nursing; 67: 12, 2615-2627.

Key points

  • Older people comprise only 20% of the population but take half of prescribed medicines
  • Medicine administration errors (MAEs) are more common in patients with dysphagia than those without
  • MAEs are largely due to errors of drug formulation or preparation
  • The risk of MAEs is higher for patients with dysphagia who have enteral tubes compared with those without tubes
  • Health professionals need to take extra care when prescribing, dispensing and administering medicines to this patient group
  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.

Related Jobs