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

An audit of oxygen prescribing in acute general medical wards

  • Comment

Mandy Howell, BSc (Hons), RGN, OND, FETC, DPSN, DMS, Asthma Dip, RespManDip.

Senior Clinical Nurse, General Internal Medicine, City Hospitals Sunderland NHS Trust, Sunderland Royal Hospital

Nurses are becoming increasingly involved in examining not only their own practice but that of other health-care professionals which may affect or influence nursing practice.


Nurses are becoming increasingly involved in examining not only their own practice but that of other health-care professionals which may affect or influence nursing practice.



Drug prescribing is an area that plays a fundamental role in nurses’ everyday practice, yet it is an area which is predominantly led by medical staff.



Oxygen is a drug that medical staff often fail to prescribe appropriately (Bell, 1995), while, in the author’s experience, nurses often fail to take appropriate action to ensure that a prescription for oxygen is written or updated before administration.



During a clinical governance meeting for respiratory medicine services at City Hospitals Sunderland NHS Trust, it was decided to audit the medical staff’s oxygen-prescribing practices across six acute medical wards under the topic of risk management.



Oxygen therapy
Two terms often used when discussing oxygen are hypoxeamia (insufficient oxygen in the blood) and hypoxia (decreased availability of oxygen in the tissues). Regardless of age, injury or illness, optimal cellular function depends upon adequate oxygenation and a balanced acid-base ratio. If left uncorrected, hypoxeamia results in hypoxia, leading to cellular injury or death (Horne and Derrico, 1999).



Oxygen is used in arterial hypoxaemia. A saturation of 97% means that 97% of the total amount of haemoglobin in the body is filled with oxygen molecules. A range of 95%-100% is generally considered normal (Carroll, 1997). Symptoms usually develop when the saturation falls to 90% or lower.



The binding of oxygen to haemoglobin is dependent on the local concentration of oxygen; however, the relationship is not precisely linear. This is usually referred to as the oxyhaemoglobin dissociation curve (Figure 1). The slope progresses steeply between 1.5 and 7kPa and then plateaus between 9 and 13.5kPa. This is important for oxygen therapy because it illustrates that haemoglobin is almost completely saturated at 9kPa and therefore further increases in the partial pressure of oxygen will cause only a slight rise in oxygen binding.



The main indications for oxygen therapy (Mallett and Bailey, 1996) are listed in Box 1.



Potential problems
Although the administration of oxygen may be a life-saving treatment, its use is not without risks and it should be regarded in the same way as any other therapeutic drug. Guidance on oxygen prescription is given in the British National Formulary (BMA/RPSGB, 2000), which states that the concentration to be given depends on the condition being treated. It also states that an inappropriate concentration may have serious or even lethal effects.



Foss (1990) states that as little oxygen as possible should be given, consistent with the patient’s clinical state, and for the shortest time required. This is particularly important for patients with chronic obstructive pulmonary disease (COPD), who have a reduced hypoxic drive with carbon dioxide retention. A study by Chien et al (2000) demonstrated that the administration of 100% oxygen to people acutely ill with asthma might adversely influence carbon dioxide elimination.



At too high a concentration, oxygen is toxic to the cells of the respiratory tract and prolonged use may cause damage to the alveoli (Fisher and Paykel Healthcare, 1995). Such concentrations also predispose patients to atelectasis and collapse of lung segments (Fisher and Paykel Healthcare, 1995). Oxygen therapy is particularly hazardous in patients with lung disease because, even in low concentrations, it may suppress the patient’s drive to breathe, leading to chronic respiratory failure. In diseases characterised by airway obstruction, such as COPD, in which there is a danger of carbon dioxide retention, great care needs to be taken in oxygen delivery.



Importance of prescribing procedures
The Royal College of Physicians have produced guidelines on oxygen therapy (1999).



Fisher and Paykel Healthcare, manufacturers of respiratory products (1995), advise doctors to prescribe the concentration of oxygen to be given and the method of administration to be used. The flow rate can be determined only once the specific administration device is known because identical flow rates may produce different concentrations in different devices.



Nurses have many responsibilities in caring for patients receiving oxygen therapy, including the choice of administration device and understanding of its function, supporting the patient with appropriate education and encouragement and ensuring that the correct prescription of oxygen is administered. Foss (1990) claims that nurses must also have an understanding of the purpose of oxygen therapy if it is to be given safely and effectively.



Methods of delivery
Patients receiving supplemental oxygen may have specific needs relating to its delivery. The method of delivery will depend on certain factors. These are listed in Box 2.



Oxygen can be delivered to the patient via a mask, a nasal cannula or a tracheostomy. High-flow masks deliver 100% oxygen, while fixed-rate masks deliver a fixed percentage of oxygen.



For those on oxygen for many hours of the day a mask may be claustrophobic and interfere with eating and talking. For these reasons a nasal cannula may be used as an alternative delivery device. Cannulae are better suited to low-flow rates, usually 2 litres/minute, as higher rates, such as 6 litres/minute, tend to make the mucous membrane of the nose sore. A comparison of delivery rates of various devices is given in Table 1.



Patients who are mouth-breathers should use masks, as the percentage of oxygen delivered via nasal cannulae may be inaccurate.



It is important that patients on oxygen therapy are monitored and prescriptions for oxygen altered accordingly.



Monitoring oxygen levels
Arterial blood gases are often regarded as the gold standard in monitoring oxygen levels and give other clinical information besides oxygen saturation values, such as the acid-base balance. They are, however, often taken intermittently, and results are not always available immediately.



The value of continuously monitoring patients’ arterial oxygen saturation is in assessing the need for and effect of supplemental oxygen therapy, which before the advent of pulse oximetry, was given on a ‘hit-and-miss’ basis (Keenan, 1995).



In the ward, pulse oximetry may be used to monitor the effects of changes in patients’ oxygen therapy (Lowton, 1999). A pulse oximeter measures peripheral oxygen saturation (SpO2), via a sensor, which is usually clipped on to the patient’s earlobe or finger. The main function of the device is to detect hypoxaemia in a patient before it can be detected by sight. Cyanosis, the visible sign of hypoxaemia, occurs at a saturation of about 75% in normally perfused patients. Pulse oximetry can be used intermittently or continuously to monitor the effect of oxygen therapy. However, there are no published studies which claim that patients are better off with continuous bedside pulse oximetry monitoring, nor are there any which indicate a reduced rate of illness or death resulting from the use of pulse oximeters. Indeed, studies have demonstrated that, while hypoxaemia is generally detected much earlier in patients monitored by pulse oximeters (Jubran and Tobin, 1990), technology is not infallible and may not reduce rates of undesirable outcomes.



In order for pulsatile flow to be detected there must be sufficient perfusion in the monitored area. If the patient has a weak or absent peripheral pulse, pulse oximeter readings will not be precise. Patients most at risk of low perfusional states are those with hypotension, hypovolaemia and hypothermia and those in cardiac arrest (Schnapp and Cohen, 1990). Non-arterial pulses can be detected if the probe is secured too tightly, creating venous pulsations in the finger. If a blood pressure cuff above the probe creates a tourniquet effect, this may stop the pulsatile flow altogether and no saturations will be recorded (Carroll, 1997)



It is also important to remember that pulse oximetry does not present a complete picture of oxygen transport. Information is not provided about haemoglobin concentration, cardiac output, efficiency of oxygen delivery to the tissues, the consumption or sufficiency of oxygenation, adequacy of ventilation (Jensen et al, 1998) or carbon dioxide retention (Jevon, 2000).



It is worthwhile noting that it is the trend of the patient’s response and not a single observation that provides the most valuable information on the patient’s condition (Sheppard, 2000).



Oxygen saturation measurements cannot reliably be used as a substitute for arterial oxygen tension measurements for the prescription of oxygen (Carlin et al, 1994), although studies have shown oximetry measurements to be accurate to within 2% of blood samples (Jones, 1995).



It is important to know the inspired oxygen concentration when analysing oxygen saturation readings. Pulse oximetry can give important information about the patient’s oxygen saturation level but, in order to interpret the oxygen saturation in relation to pulmonary function, the amount of inspired oxygen needs to be known.



As with any other drug, oxygen may have toxic effects, therefore nurses should consult regularly with medical staff to establish whether any change in oxygen therapy is appropriate (Woodrow, 1999).



Oxygen prescription audit: aims and method
During the monthly respiratory medicine clinical governance meeting, the inconsistencies in the practice of prescribing oxygen, which posed high risks for patients, were discussed and it was agreed that an audit would enable us to benchmark practice.



The overall aim of the audit was to measure practice within the trust. The specific objectives were to determine if:



- Oxygen was being prescribed



- Oxygen administration was being recorded if not correctly prescribed



- Appropriate delivery devices were being used



- Patients who were receiving oxygen treatment were being monitored by pulse oximetry or arterial blood gases.



A simple audit-recording tool was designed. It consisted of a range of questions that required completion following the examination of patients’ drug prescription sheets, medical case notes, nursing records, and the observation of patients receiving oxygen at the time of audit.



The audit was conducted across six acute medical wards consisting of two 27-bed medical assessment wards, a 45-bed respiratory ward, a 30-bed general medical ward and two 27-bed elderly care wards, and was completed on one day.



Results and discussion
Out of 177 in-patients, 45 (25%) were receiving oxygen therapy on the day of the audit. Of those, 38 (84%) were using it at the time of the audit.



Written instructions - Only seven (16%) of the patients receiving oxygen therapy that day had a prescription for oxygen written on the medication prescription sheet. Of these, four (57%) were patients in the respiratory ward, which accounted for 18 (40%) of the total number of patients on oxygen.



Of the 45 patients receiving oxygen therapy, 18 (40%) had written instructions in the medical case notes, and the nursing records of 38 (84%) recorded the administration of oxygen. The percentage and/or rate of oxygen and the means of administration were, however, not always specified. Thus doctors are relying on nurses’ knowledge to ensure the correct delivery device is used.



Devices and labelling - Although there is no statistical difference between the performance of nasal cannulae and facemasks (English, 1994), assuming performance is equal, it is presumed that patients receiving oxygen therapy in wards would prefer nasal cannulae. These cannulae do not interfere with eating and speaking and are therefore more likely to be left in situ than the facemask, which patients frequently fail to wear.



Twenty-six patients (58%) were receiving oxygen via a mask, 17 (38%) had nasal cannulae and two (4%) received oxygen via a tracheostomy.



Of the 26 masks in use, only six (23%) were labelled with the patient’s identity. None of the nasal cannulae tubing included any means of recording the patient’s identity. This was particularly relevant in areas that used a single-flow meter between two patients and in patients who also had separate nebuliser masks that were changed between therapies. The absence of clear patient identification of mask/cannulae could lead to a patient receiving the wrong delivery device and incorrect flow/percentage rate, especially if the prescription for oxygen is not readily available. It may also have infection control implications, with increased risk of respiratory infection if patients share delivery devices.



Documentation of monitoring results - Twenty-seven patients (60%) had pulse oximetry charts in situ, of which 18 (40%) showed evidence of oxygen saturation measurements being recorded at least every four hours. Of the 27 patients having regular pulse oximetry monitoring, 18 (67%) were from the respiratory ward and six (22%) from the adjacent general medical ward.



No pulse oximetry chart contained information for staff regarding the amount of oxygen the patient was to receive or whether or not this was to be continuous or on request only.



Arterial blood gases were recorded as having been taken in 31 patients (Figure 2). Arterial blood gases are taken to evaluate oxygenation, ventilation and acid base balance (Coombs, 2001). By offering more detailed information than pulse oximetry, blood gases can be used to monitor the effects of oxygen or artificial ventilation therapy (Higgins, 1996). Blood gases are thus a useful aid to ensure that the correct prescription of oxygen is being given to ensure maximum benefit to the patient.



The results of this audit confirmed what we feared, that medical staff were failing to prescribe oxygen and that nurses were administering oxygen despite not having an authorised prescription. A similar audit, conducted by Bell (1995), reported that junior doctors claimed that they were not taught about oxygen at medical school, hence they were unaware of the importance of a prescription for oxygen therapy. They also lacked education in oxygen devices and delivery, and assumed that these decisions could be left to a nurse, which was a similar excuse given by our medical staff.



Recommendations following the audit include examining the feasibility of a separate prescribing sheet for oxygen that clearly identifies rate, concentration and device. In the meantime, the prescription of oxygen on the current medication sheets is to be encouraged at all times.



Education sessions for both medical and nursing staff in the use and monitoring of oxygen therapy have also been arranged.



Nurses have many responsibilities when it comes to oxygen administration, not least to educate the patient in the benefits of oxygen therapy in order to improve adherence, but also to aid in the selection of appropriate devices and prevent any discomfort a device may cause.



The dangers of oxygen therapy cannot be emphasised enough, and the administration of oxygen without a valid prescription leaves nurses wide open to clinical risk and negligence claims should things go wrong. Sometimes, however, oxygen will need to be administered in an emergency and local policies should stipulate occasions when oxygen that has not been prescribed can be administered by nurses (Jevon and Ewens, 2001).



Nurses should remember that oxygen is a drug and therefore should be prescribed and administered accordingly. A clear, accurate and immediate record should be made when the oxygen is administered in accordance with local policy and UKCC guidance for drug administration (UKCC, 2000).





Bell, C. (1995)Is this what the doctor ordered? Professional Nurse 10: 6, 297-300.



British Medical Association/ Royal Pharmaceutical Society of Great Britain. (2000)Oxygen. In: British National Formulary No 39 (March) 157-158. London: BMA/ RPSGB.



Carlin, B.W., Clause, J.L., Ries, A.L. (1994)The effects of exercise testing on the prescription of oxygen therapy. Chest 106: 2, 361-365.



Carroll, P. (1997)Pulse oximetry at your finger tips. RN 60: 2, 22-27.



Chien, J.W., Ciufo, R., Novak, R. et al. (2000)Uncontrolled oxygen administration and respiratory failure in acute asthma. Chest 117: 3, 728-733.



Coombs, M. (2001)Making sense of arterial blood gases. Nursing Times 97: 27, 36-38.



English, I. (1994)Oxygen mask or nasal catheter? An analysis. Nursing Standard 8: 26, 27-31.



Fisher and Paykel Healthcare. (1995)Oxygen Therapy Made Simple. High Wycombe, Bucks: Fisher and Paykel Healthcare.



Foss, M.A. (1990)Oxygen therapy. Professional Nurse 5: 5, 180-190.



Higgins, C. (1996)Principles and practice of blood gas measurement. Nursing Times 92: 46, 45-47.



Horne, C., Derrico, D. (1999)Mastering ABGs. American Journal of Nursing 99: 8, 26-32.



Jensen, L.A., Onyskiw, J.E., Prasad, N.G.N. (1998)Meta-analysis of arterial oxygen saturation monitoring by pulse oximetry in adults. Heart and Lungs 27: 6, 387-408.



Jevon, P. (2000)Pulse oximetry: 1. Nursing Times 96: 26, 43-44.



Jevon, P., Ewens, B. (2001)Assessment of a breathless patient. Nursing Standard 15: 16, 48-53.



Jones, S.E. (1995)Getting the balance right: pulse oximetry and inspired oxygen concentration. Professional Nurse 10: 6, 368-373.



Jubran, A., Tobin, M.J. (1990)Reliability of pulse oximetry in titrating supplemental oxygen therapy in ventilator-dependent patients. Chest 97: 1420-1425.



Keenan, J. (1995)Pulse oximetry. Nursing Standard 9: 35, 55.



Lowton, K. (1999)Pulse oximeters for the detection of hypoxaemia. Professional Nurse 14: 5, 343-350.



Mallett, J., Bailey, C. (1996)Respiratory therapy (chapter 35). In: The Royal Marsden NHS Trust Manual of Clinical Nursing Procedures (4th edn). Oxford: Blackwell Science.



Respiratory Education Resource Centres. (2001)COPD Training Module. Liverpool: University Clinical Department, University Hospital Aintree.



Royal College of Physicians. (1999)Domiciliary Oxygen Therapy Services: Clinical guidelines and advice for prescribers. London: RCP.



Sheppard, M. (2000)Oxygen therapy: 1. Nursing Times 96: 29, 43-44.



Schnapp, L.M., Cohen, N.H. (1990)Pulse oximetry: uses and abuses. Chest 98: 1244-1250.



UKCC. (2000)Guidelines for the Administration of Medicines. London: UKCC.



Woodrow, P. (1999)Pulse oximetry. Nursing Standard 13: 42, 42-46.


  • 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.