Antibiotics are often unnecessary

“GPs are still prescribing antibiotics unnecessarily for coughs and colds”, the BBC and newspapers reported. The Daily Mail reported: “Many of the cases would clear up on their own” and that millions are being wasted on unnecessary treatments.

The BBC said that current guidelines advise GPs not to routinely prescribe antibiotics for patients with upper respiratory tract infections, such as coughs, colds and sinusitis, as well as sore throats and ear infection as these illnesses tend to be caused by a virus. Despite this, a study of the General Practice Research Database (GPRD) has demonstrated that antibiotics are still being given to more than 90% of patients with chesty coughs, 80% with ear infections, and 60% with sore throats.

The Daily Mail continues that the study’s researchers claim that there is no evidence that giving antibiotics prevents serious complications from developing, and highlights the problem that overuse of antibiotics can lead to the problem of drug resistance.

As mentioned by the BBC, current guidelines already state that GPs should exercise some restraint and not routinely prescribe antibiotics for minor infections. The issues of over-prescribing antibiotics, such as drug resistant infections and the minimal benefit that many patients get from antibiotics, are already well known to the medical profession.

While this study highlights the need to treat chest infections in the elderly with antibiotics to avoid pneumonia, many common infections are usually resolved by themselves, and GPs and the public should bear this in mind.

Where did the story come from?

The research was carried out by I. Petersen and colleagues of the Centre for Infectious Disease Epidemiology, Department of Primary Care and Population Sciences, University College London. The study was funded by the Department of Health. The study was published in the (peer-reviewed) British Medical Journal.

What kind of scientific study was this?

This was a retrospective cohort study where the researchers looked back over records in the GPRD to investigate the extent to which the prescription of antibiotics reduces the risk of serious complications following common upper respiratory tract infections (URTI) e.g. coughs, colds, sore throats, ear infections.

The researchers examined data contributed to the GPRD from 162 GP surgeries in the UK between July 1991 and June 2001. Their aim was to investigate whether patients diagnosed with common URTIs developed complications in the following month, and whether the prescription of antibiotics by their GP on the day of first presentation affected their risk of these complications.

The main complications the researchers looked for included; quinsy following tonsillitis (abscess and inflammation around the tonsils and surrounding tissues), mastoiditis following ear infection (a serious infection involving part of the bones in the skull), and pneumonia. The researchers also looked at whether giving antibiotics affected the chances of developing a chest infection.

The researchers searched the database for the codes that are assigned when a patient with URTI has their initial GP consultation. Statistical methods were then used to calculate the extent of the protective benefit that was obtained by using antibiotics and how many patients with URTI would need to be treated with antibiotics in order for just one to gain a benefit. The researchers took into account potential contributing factors such as age, sex, and social deprivation according to GP location.

What were the results of the study?

The researchers found that while the number of consultations for URTIs during the period studied was very high, the rate that complications developed was very low. Although prescribing antibiotics did reduce the risk of developing any of the complications, the actual number of patients that would need to be treated to prevent one tonsillitis patient from developing quinsy, one person with ear infection going on to develop mastoiditis, or to prevent pneumonia developing in the month after a URTI, was more than 4,000 in each case.

They found that 17 out of 1,000 untreated URTI patients presented with a chest infection in the following month, which was reduced to 11 out of 1,000 in patients who received antibiotics. Calculations revealed that in order to prevent one patient needing to consult a GP about a chest infection in the month following a URTI, 161 patients would also need to be treated.

The greatest benefit of antibiotics appeared to be for reducing the risk of pneumonia after a chest infection. The size of this risk increased with age: In patients over 65, 403 patients per 1,000 were at risk of pneumonia if a chest infection was untreated. This was reduced to 146 per 1,000 if treated with antibiotics. Only 39 patients over 65 years of age would need to be treated with antibiotics to prevent one case of pneumonia, compared to 119 patients between the ages of 16 and 64 years.

What interpretations did the researchers draw from these results?

The authors conclude that there is no justification in prescribing antibiotics for mild URTI, sore throats or ear infections. However, antibiotics do reduce the risk of pneumonia developing following a chest infection, particularly in the elderly.

What does the NHS Knowledge Service make of this study?

This research is a reliable analysis of the benefit obtained from prescribing antibiotics for common respiratory tract infections in terms of reducing the risk of developing complications. It highlights the well-known fact that many mild infections often only have a minimal benefits from antibiotics. It also indicates that elderly patients with chest infections may be at risk of pneumonia if they remain untreated.

Although this is a study of a large quantity of reliable data from the GPRD, there are still several points that need to be considered:

The main potential source of error is that this research relied upon the use of database codes to identify GP consultations and diagnoses. The codes applied will have been entered by individual GPs and may therefore, be a source of discrepancy. For example, the term “chest infection” is quite broad and could have included cases of pneumonia as well as milder chesty coughs or acute bronchitis.
It is also likely that the two groups of treated and untreated patients were not completely balanced and matched to each other and this may introduce error in the estimated risk of complications. For instance, the patients who have been treated with antibiotics are likely to include a higher proportion with more serious infections, or with other comorbid medical conditions, where the GP was more concerned that they would develop complications if left untreated. This means that we cannot observe the benefits of treating or not treating in more serious cases.
The researchers have attempted to account for the possible confounding effects of age, sex, social deprivation, and smoking. However, the effects of the patient having other medical conditions, a history of repeated infections, or poor recovery or hospitalisation following previous infections, cannot be considered by this research.
It is not possible to assess from this research whether prescribing antibiotics made any difference to the recovery time from illness, or whether their use is associated with any particular adverse side effects.

Current guidelines already state that GPs should exercise some restraint and not routinely prescribe antibiotics for minor infections. The issues of over-prescribing antibiotics, such as drug resistant infections and the minimal benefit that many infections get from antibiotics, are already well known to the medical profession.

Sir Muir Gray adds...

People love antibiotics and hate MRSA, but the two are intimately related like yin and yang.