25th August 2010
A systematic review and meta-analysis found that individuals prescribed an antibiotic for a respiratory or urinary infection were around twice as likely to develop resistance to that antibiotic compared with those who hadn’t taken one. This risk was greatest in the first month after taking the antibiotic, but persisted for up to a year.
Level of evidence:
Level 2 (limited quality patient-oriented evidence) according to the SORT criteria.
Health professionals should ensure they prescribe antibiotics appropriately and in line with NICE or local guidance based on HPA advice. Prescribing for viral or mild, self-limiting infections such as coughs and colds is unlikely to improve the course of the illness, puts patients at risk of unnecessary adverse reactions (e.g. vomiting, diarrhoea, rash, fungal infection) and encourages further consultations. Clinicians should select the patients who are at higher risk of complications and those who are more likely to have a bacterial cause for their symptoms. In certain scenarios providing reassurance that the symptoms will resolve without antibiotic treatment and the use of watchful waiting or a delayed prescription may be preferable. The NPC Patient decision aid relating to upper respiratory tract infections (URTIs) may be helpful in some consultations. Clinicians in primary care should work with their patients to consider the implications of this study when discussing the benefits and risks of taking antibiotics.
What is the background to this?
As discussed in a MeReC Bulletin on the management of common infections in primary care, the widespread use of antibiotics is associated with the emergence of resistant bacteria, many of which are multi-drug resistant. Furthermore, as very few new antibiotics are entering practice, antibiotic resistance is now a major threat to public health.
Most antibiotic prescribing is in primary care. There are concerns that some common infections are becoming increasingly difficult to treat and that illnesses due to antibiotic resistant bacteria may take longer to resolve. The findings of this study therefore support the case for primary care clinicians and patients to see antibiotic resistance as a reason to refrain from inappropriate antibiotic use. A lack of patient awareness may be contributing to the problems caused by antibiotic resistance, and it may appear that there is a perception amongst some patients and clinicians that antibiotic resistance is only of minimal risk.
Previous studies that have examined the link between antibiotic prescribing in primary care and bacterial resistance have mostly been at population level. To date, only a limited number of good quality studies have reported on the relationship between prescribing and prevalence of resistance for at individual-patient level. Therefore, the aim of this study was to examine the effect of antimicrobial use on the emergence of antibiotic resistance in individual patients in primary care. The authors aimed to quantify the strength and duration of any association as well as identifying which antibiotics were most and least likely to cause resistance.
What does this study claim?
This review included 24 studies; 22 involved patients with symptomatic infection and two involved healthy volunteers; 19 were observational studies (of which two were prospective) and five were RCTs. In five studies looking at the urinary tract (14,348 participants), individuals were between two and three times more likely to test positive for resistant organisms within three months of antibiotic treatment (pooled odds ratio [OR] for resistance was 2.48, 95% confidence interval [CI] 2.06 to 2.98). This was shown to reduce over time (P<0.001), although a residual association was still present at twelve months.
In seven studies of respiratory tract bacteria (2,605 participants), a non-statistically significant result was found, with a pooled OR for resistance of 1.48 (95%CI 0.95 to 2.32) within three months. However, a statistically significant association between antibiotics and resistance to respiratory tract bacteria was found within one month (pooled OR 2.10, 95%CI1.04 to 4.23), and within two months (pooled OR 2.37, 95%CI 1.42 to 3.95). No significant association between resistance of respiratory tract bacteria was seen with time (P=0.91).
Studies reporting the quantity of antibiotic prescribed found that longer duration and multiple courses were associated with higher rates of resistance. Studies comparing the potential for different antibiotics to induce resistance showed no consistent effects, possibly due to the limited number of studies available.
This study found that antibiotics prescribed to an individual in primary care were consistently found to be associated with resistance of urinary and respiratory bacteria to those antibiotics in that individual. Despite this association, the clinical significance of the findings of this study remains unclear. However, health professionals should still follow appropriate NICE guidance (for example on antibiotic prescribing for self-limiting respiratory tract infections) to ensure appropriate targeting of antibiotics to those patients most likely to benefit.
Another implication of the study was that in some instances the antibiotics prescribed in primary care may impact on bacterial resistance in individual patients for up to 12 months. Furthermore, the study’s findings support the Standing Medical Advisory Committee report recommendations that the fewest number of antibiotic courses should be prescribed for the shortest period possible.
Finally, this study serves to highlight the need to use first-line antibiotics whenever possible, by following appropriate NICE and local guidance based on HPA advice. This study supports the on-going need for a shift in clinicians’ and patients’ attitudes to the use of antibiotics in primary care by providing supporting evidence linking the emergence of resistance to antimicrobial use.
What are the limitations of this study?
The authors found some evidence of positive publication bias in the urinary bacteria studies investigating resistance in E coli but were unable to assess publication bias for the respiratory flora, as there were too few studies. Another source of bias could include other confounding factors, such as the association between primary care prescribing and any recent hospital admissions, where antibiotics could have been used. However, the studies that attempted to adjust for potential confounders such as age, sex, comorbidities, catheter use, and smoking status rarely demonstrated substantial difference between crude and adjusted results. A further limitation was that, the existing body of evidence is mainly reliant on observational or case-controlled studies, and there is a lack of data from clinical trials. These limitations highlight that further research is needed to assess the link between antibiotics prescribed in primary care and more serious infections that require secondary care treatment, as well as to further clarify the effects of interactions between antibiotic dose, duration, and adherence on resistance.
Systematic review and meta-analysis of 24 studies conducted in any country. The studies consisted of five RCTs and 19 observational studies, two of which were prospective, and 17 retrospective controlled observational or case control studies. The studies were included if they investigated relations between primary care prescribed antibiotics and antimicrobial resistance in bacteria sampled from any body site; analysed at the level of the individual. All studies were based in countries where antibiotics are available by prescription only
The 24 studies investigated effects in 15,505 adults and 12,103 children. Twenty-two studies sampled bacteria from patients with symptomatic infection: urinary tract infections (seven studies); respiratory tract infections (seven); otitis media (two); chronic obstructive pulmonary disease (one); methicillin resistant Staphylococcus aureus (MRSA) infection (four); and trachoma in children (one). Two studies examined asymptomatic healthy adult volunteers
Intervention and comparison
Studies presented a wide range of antibiotic exposure analyses including those for: macrolides (eight studies); penicillins (five); sulphonamides and trimethoprim (six); cephalosporins (six); tetracyclines (two); quinolones (two); nalidixic acid (one); metronidazole (one); nitrofurantoin (one); and “any antibiotic” (seven). The antibiotics were given between two and 104 weeks before measurement of antibiotic resistance.
Adherence to antibiotic regimen was not assessed in most of the studies as they were retrospective and researchers were only able to measure prescribing from patients’ records or questionnaires, though in one of the RCTs adherence was measured by recording medicine bottle weights at various time points throughout the study.
Outcomes and results
In five studies of urinary tract bacteria (14,348 participants), individuals were between two and three times more likely to test positive for resistant organisms within three months of antibiotic treatment (pooled odds ratio [OR] for resistance was 2.48, 95% confidence interval [CI] 2.06 to 2.98). This was shown to reduce over time (P<0.001), although a residual association was still present at twelve months.
A similar result was found in seven studies of respiratory tract bacteria (2605 participants), with a pooled OR for resistance of 1.48 (95%CI 0.95 to 2.32) within three months, which did not reach conventional levels of statistical significance. However, some association between antibiotics and resistance to respiratory tract bacteria was found within one month (pooled OR 2.10, 95%CI1.04 to 4.23), and within two months (pooled OR 2.37, 95%CI 1.42 to 3.95). No significant association between resistance of respiratory tract bacteria was seen with time (P=0.91).
Studies reporting the quantity of antibiotic prescribed found that longer duration and multiple courses were associated with higher rates of resistance. Studies comparing the potential for different antibiotics to induce resistance showed no consistent effects, possibly due to the limited number of studies available. The prospective Malhotra-Kumar study was the only one that reported changes in resistance over a long period; pooled ORs fell from 12.2 (95% CI 6.8 to 22.1) at 1 week to 6.1 (95% CI 2.8 to 13.4) at 1 month, 3.6 (95% CI 2.2 to 6.0) at 2 months, and 2.2 (95% CI 1.3 to 3.6) at 6 months.
University of Bristol in collaboration with the University of Oxford which both received a proportion of their funding from the Department of Health’s NIHR School for Primary Care Research.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you