Richards J
Norwich Public Health Laboratory, Bowthorpe Road, Norwich NR2 3TX.
Commun Dis Public Health. 2002 Sep;5(3):226-9.
Awareness of the rise in antimicrobial resistance has led to recommendations for better prescribing guidelines, based on accurate laboratory sensitivity data. However, concerns have been raised of possible biases in sampling leading to skewed resistance rates results, and so a range of different enhanced surveillance models have been proposed. This study compared the differences between routine results from unselected primary care urine samples with results from samples from spotter practices. Five hundred and eighty-eight out of 2,288 urine samples received from three spotter practices were culture positive. Of the 510 isolates tested against a panel of commonly prescribed antibiotics, levels of resistance to ampicillin were 32.5%, and to trimethoprim 18.0% (p < 0.05). By comparison, of 40,026 urine specimens submitted by other practices, 9,382 were culture positive and the percentage antibiotic resistant was higher--36.3% to ampicillin and 22.6% to trimethoprim. Routinely submitted laboratory samples may overestimate the true levels of antibiotic resistance in the community. A spotter practice-model is an effective and viable way of collecting unbiased data.
对抗菌素耐药性上升的认识促使人们根据准确的实验室敏感性数据，提出了更好的处方指南建议。然而，有人担心采样过程中可能存在偏差，导致耐药率结果出现偏差，因此提出了一系列不同的强化监测模型。本研究比较了未经选择的初级保健尿液样本的常规结果与监测点样本的结果之间的差异。从三个监测点收集的2288份尿液样本中，有588份培养呈阳性。在针对一组常用抗生素进行测试的510株分离菌中，对氨苄西林的耐药率为32.5%，对甲氧苄啶的耐药率为18.0%（p<0.05）。相比之下，其他医疗机构提交的40026份尿液标本中，9382份培养呈阳性，抗生素耐药百分比更高——对氨苄西林为36.3%，对甲氧苄啶为22.6%。常规提交的实验室样本可能高估了社区中抗生素耐药性的真实水平。监测点样本模型是收集无偏差数据的一种有效且可行的方法。
