Christopher C Butler

Amoxicillin for acute lower-respiratory-tract infection in primary care when pneumonia is not suspected: a 12-country, randomised, placebo-controlled trial

BACKGROUNDnLower-respiratory-tract infection is one of the most common acute illnesses managed in primary care. Few placebo-controlled studies of antibiotics have been done, and overall effectiveness (particularly in subgroups such as older people) is debated. We aimed to compare the benefits and harms of amoxicillin for acute lower-respiratory-tract infection with those of placebo both overall and in patients aged 60 years or older.nnnMETHODSnPatients older than 18 years with acute lower-respiratory-tract infections (cough of ≤28 days duration) in whom pneumonia was not suspected were randomly assigned (1:1) to either amoxicillin (1 g three times daily for 7 days) or placebo by computer-generated random numbers. Our primary outcome was duration of symptoms rated moderately bad or worse. Secondary outcomes were symptom severity in days 2-4 and new or worsening symptoms. Investigators and patients were masked to treatment allocation. This trial is registered with EudraCT (2007-001586-15), UKCRN Portfolio (ID 4175), ISRCTN (52261229), and FWO (G.0274.08N).nnnFINDINGSn1038 patients were assigned to the amoxicillin group and 1023 to the placebo group. Neither duration of symptoms rated moderately bad or worse (hazard ratio 1.06, 95% CI 0.96-1.18; p=0.229) nor mean symptom severity (1.69 with placebo vs 1.62 with amoxicillin; difference -0.07 [95% CI -0.15 to 0.007]; p=0.074) differed significantly between groups. New or worsening symptoms were significantly less common in the amoxicillin group than in the placebo group (162 [15.9%] of 1021 patients vs 194 [19.3%] of 1006; p=0.043; number needed to treat 30). Cases of nausea, rash, or diarrhoea were significantly more common in the amoxicillin group than in the placebo group (number needed to harm 21, 95% CI 11-174; p=0.025), and one case of anaphylaxis was noted with amoxicillin. Two patients in the placebo group and one in the amoxicillin group needed to be admitted to hospital; no study-related deaths were noted. We noted no evidence of selective benefit in patients aged 60 years or older (n=595).nnnINTERPRETATIONnWhen pneumonia is not suspected clinically, amoxicillin provides little benefit for acute lower-respiratory-tract infection in primary care both overall and in patients aged 60 years or more, and causes slight harms.nnnFUNDINGnEuropean Commission Framework Programme 6, UK National Institute for Health Research, Barcelona Ciberde Enfermedades Respiratorias, and Research Foundation Flanders.

Probiotics for Antibiotic-Associated Diarrhoea (PAAD): a prospective observational study of antibiotic-associated diarrhoea (including Clostridium difficile-associated diarrhoea) in care homes

BACKGROUNDnAntibiotic prescribing rates in care homes are higher than in the general population. Antibiotics disrupt the normal gut flora, sometimes causing antibiotic-associated diarrhoea (AAD). Clostridium difficile (Hall and OToole 1935) Prévot 1938 is the most commonly identified cause of AAD. Little is known either about the frequency or type of antibiotics prescribed in care homes or about the incidence and aetiology of AAD in this setting.nnnOBJECTIVESnThe Probiotics for Antibiotic-Associated Diarrhoea (PAAD) study was designed as a two-stage study. PAAD stage 1 aimed to (1) prospectively describe antibiotic prescribing in care homes; (2) determine the incidence of C. difficile carriage and AAD (including C. difficile-associated diarrhoea); and (3) to consider implementation challenges and establish the basis for a sample size estimation for a randomised controlled trial (RCT) of probiotic administration with antibiotics to prevent AAD in care homes. If justified by PAAD stage 1, the RCT would be implemented in PAAD stage 2. However, as a result of new evidence regarding the clinical effectiveness of probiotics on the incidence of AAD, a decision was taken not to proceed with PAAD stage 2.nnnDESIGNnPAAD stage 1 was a prospective observational cohort study in care homes in South Wales with up to 12 months follow-up for each resident.nnnSETTINGnRecruited care homes had management and owners agreement to participate and three or more staff willing to take responsibility for implementing the study.nnnPARTICIPANTSnEleven care homes were recruited, but one withdrew before any residents were recruited. A total of 279 care home residents were recruited to the observational study and 19 withdrew, 16 (84%) because of moving to a non-participating care home.nnnMAIN OUTCOME MEASURESnThe primary outcomes were the rate of antibiotic prescribing, incidence of AAD, defined as three or more loose stools (type 5-7 on the Bristol Stool Chart) in a 24-hour period, and C. difficile carriage confirmed on stool culture.nnnRESULTSnStool samples were obtained at study entry from 81% of participating residents. Over half of the samples contained antibiotic-resistant isolates, with Enterobacteriaceae resistant to ciprofloxacin in 47%. Residents were prescribed an average of 2.16 antibiotic prescriptions per year [95% confidence interval (CI) 1.90 to 2.46]. Antibiotics were less likely to be prescribed to residents from dual-registered homes. The incidence of AAD was 0.57 (95% CI 0.41 to 0.81) episodes per year among those residents who were prescribed antibiotics. AAD was more likely in residents who were prescribed co-amoxiclav than other antibiotics and in those residents who routinely used incontinence pads. AAD was less common in residents from residential homes.nnnCONCLUSIONSnCare home residents, particularly in nursing homes, are frequently prescribed antibiotics and often experience AAD. Antibiotic resistance, including ciprofloxacin resistance, is common in Enterobacteriaceae isolated from the stool of care home residents. Co-amoxiclav is associated with greater risk of AAD than other commonly prescribed antibiotics.nnnTRIAL REGISTRATIONnCurrent Controlled Trials ISRCTN 7954844.nnnFUNDINGnThis project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 18, No. 63. See the NIHR Journals Library website for further project information.

Antibiotic prescribing and associated diarrhoea: a prospective cohort study of care home residents

BACKGROUNDnthe risk factors for and frequency of antibiotic prescription and antibiotic-associated diarrhoea (AAD) among care home residents are unknown.nnnAIMnto prospectively study frequency and risks for antibiotic prescribing and AAD for care home residents.nnnDESIGN AND SETTINGna 12-month prospective cohort study in care homes across South Wales.nnnMETHODnantibiotic prescriptions and the development of AAD were recorded on case report forms. We defined AAD as three or more loose stools in a 24-h period occurring within 8 weeks of exposure to an antibiotic.nnnRESULTSnwe recruited 279 residents from 10 care homes. The incidence of antibiotic prescriptions was 2.16 prescriptions per resident year (95% CI: 1.90-2.46). Antibiotics were less likely to be prescribed to residents from dual-registered homes (OR compared with nursing homes: 0.38, 95% CI: 0.18-0.79). For those who were prescribed antibiotics, the incidence of AAD was 0.57 episodes per resident year (95% CI: 0.41-0.81 episodes). AAD was more likely in residents who were prescribed co-amoxiclav (hazards ratio, HR = 2.08, 95% confidence interval, CI: 1.18-3.66) or routinely used incontinence pads (HR = 2.54, 95% CI: 1.26-5.13) and less likely in residents from residential homes (HR compared with nursing homes: 0.14, 95% CI: 0.06-0.32).nnnCONCLUSIONnresidents of care homes, particularly of nursing homes, are frequently prescribed antibiotics and often experience diarrhoea following such prescriptions. Co-amoxiclav is associated with greater risk of AAD.

Eligibility for interventions, co-occurrence and risk factors for unhealthy behaviours in patients consulting for routine primary care: results from the Pre-Empt study

BackgroundSmoking, excessive drinking, lack of exercise and a poor diet remain key causes of premature morbidity and mortality globally, yet it is not clear what proportion of patients attending for routine primary care are eligible for interventions about these behaviours, the extent to which they co-occur within individuals, and which individuals are at greatest risk for multiple unhealthy behaviours. The aim of the trial was to examine ‘intervention eligibility’ and co-occurrence of the ‘big four’ risky health behaviours – lack of exercise, smoking, an unhealthy diet and excessive drinking – in a primary care population.MethodsData were collected from adult patients consulting routinely in general practice across South Wales as part of the Pre-Empt study; a cluster randomised controlled trial.After giving consent, participants completed screening instruments, which included the following to assess eligibility for an intervention based on set thresholds: AUDIT-C (for alcohol), HSI (for smoking), IPAQ (for exercise) and a subset of DINE (for diet). The intervention following screening was based on which combination of risky behaviours the patient had. Descriptive statistics, χ2 tests for association and ordinal regressions were undertaken.ResultsTwo thousand sixty seven patients were screened: mean age of 48.6xa0years, 61.9xa0% female and 42.8xa0% in a managerial or professional occupation. In terms of numbers of risky behaviours screened eligible for, two was the most common (43.6xa0%), with diet and exercise (27.2xa0%) being the most common combination. Insufficient exercise was the most common single risky behaviour (12.0xa0%). 21.8xa0% of patients would have been eligible for an intervention for three behaviours and 5.9xa0% for all four behaviours. Just 4.5xa0% of patients did not identify any risky behaviours. Women, older age groups and those in managerial or professional occupations were more likely to exhibit all four risky behaviours.ConclusionVery few patients consulting for routine primary care screen ineligible for interventions about common unhealthy behaviours, and most engage in more than one of the major common unhealthy behaviours. Clinicians should be particularly alert to opportunities to engaging younger, non professional men and those with multi-morbidity about risky health behaviour.Trial registrationISRCTN22495456