Treatment of acute cough/lower respiratory tract infection by antibiotic class and associated outcomes: a 13 European country observational study in primary care
Table of Contents
Abstract
aim acute accent cough/lower respiratory tract infection ( LRTI ) be one of the common reason for consult and antibiotic order. there be theoretical argue why treatment with particular antibiotic class may aid convalescence more than others, merely empirical, pragmatic evidence exist miss. We investigate whether treatment with adenine finical antibiotic class ( amoxicillin ) be more powerfully consociate with symptom score resolution and time to patient coverage convalescence than each of eight other antibiotic class oregon nobelium antibiotic treatment for acute cough/LRTI. method acting
clinician recorded history, examination find, symptom severity and antibiotic treatment for 3402 patient indium angstrom thirteen state prospective experimental study of adult confront inch fourteen primary care research network with acute cough/LRTI. 2714 patient completed adenine symptom score daily for up to twenty-eight day and read the day along which they felt recover. adenine three-level autoregressive move average model ( 1,1 ) model investigate log casual symptom score to analyze symptom settlement. adenine two-level survival model analyze time to report recovery. clinical presentation be control for use clinician-recorded symptom, phlegm color, temperature, age, co-morbidities, smoke condition and duration of illness anterior to consultation.
result compare with amoxicillin, no antibiotic class ( and nobelium antibiotic treatment ) embody consociate with clinically relevant improved symptom resolution ( all coefficient inch the range −0.02 to 0.01 and wholly p value capital than 0.12 ). no antibiotic class ( and no antibiotic discussion ) be consociate with firm time to recovery than amoxicillin. conclusion treatment aside antibiotic course constitute not consociate with symptom resolution operating room time to recovery in adult confront to basal care with acute cough/LRTI .
Introduction
controversy wall which class of antibiotic should embody choose once clinician have make vitamin a decisiveness to order antibiotic for acute cough/LRTI in elementary care. while test and experimental cogitation generally indicate that there be little oregon no benefit from antibiotic treatment of acuate cough in primary care, one antibiotic order persist common exercise for this circumstance. two The european respiratory club ( emergency room ) and the european society of clinical microbiology and infectious disease ( ESCMID ) guidepost for the management of pornographic turn down respiratory nerve pathway infection ( LRTIs ) three promote vitamin a conservative approach to antibiotic treatment for differently well people indium the residential district and recommend amoxicillin oregon tetracycline american samoa first-line agent for certain patient : those with distrust oregon definite pneumonia ; those with choose exacerbation of chronic clogging pneumonic disease ( affected role with three of the following symptom : increase dyspnea ; increase phlegm volume ; and increase phlegm purulence ) ; those aged over seventy-five old age with fever ; those with cardiac bankruptcy ; those with insulin-dependent diabetes mellitus ; and those with serious neurological disorder. however, these guidepost argue that there cost insufficient robust empiric tell for their recommendation approximately antibiotic choice, which be based on adept consensus. The uracil center for disease control and prevention treatment guideline for acute cough illness recommend against empiric antibiotic for acuate bronchitis. four ampere recent taxonomic revue conclude that there be insufficient tell to recommend one antibiotic class over another for the discussion of community-acquired pneumonia. five approximately general medical practitioner order modern, broad-spectrum agent early because they think these agent will give the affected role the best gamble of rapid cure and prevent hospital admission. six besides, the function of agent such a fluoroquinolones for acute cough/LRTI be sometimes justify on the basis of difference indium regional bacterial resistance rate. six The anti-inflammatory place of macrolides are mention a ampere reason to choose these agent. seven We consequently analyze datum from the thirteen state seemliness ( genomics to battle resistance against antibiotic in Community-acquired LRTI in europe ), eight associate in nursing experimental study of the presentation, management and consequence of acute cough/LRTI in primary wish to determine whether any antibiotic class be consort with clinically meaningful remainder indium symptom resolution. We design to do this aside investigate whether discussion with any individual antibiotic class ( oregon no antibiotic treatment ) be associate with either remainder indiana daily symptom score oregon time until patient report recovery when compare with treatment with amoxicillin, equally this cost the most normally commend and order antibiotic for acute accent cough/LRTI in europe. two
Patients and methods
Patients
eligible patient be senesce eighteen year and all over, consult with associate in nursing illness where associate in nursing acute operating room worse cough be the chief oregon dominant symptom, oregon own ampere clinical presentation that suggest associate in nursing LRTI, with ampere duration of up to and include twenty-eight day. enter general practitioner ( global positioning system ) in fourteen basal caution research net ( based inch antwerpen, helsinki, Rotenburg, utrecht, Balatonfured, milan, Tromsø, lodz, bratislava, barcelona, Mataro, Jonkoping, cardiff and Southampton ) indiana thirteen country [ belgium, finland, germany, netherlands, hungary, italy, norway, poland, slovakia, spain ( two network ), sweden and the united kingdom ( wale and england ) ] constitute ask to recruit back-to-back eligible patient from october to november 2006 and late january to march 2007. ethical motive review committee in each area approve the sketch.
Data
clinician record aspect of the affected role ’ history, symptom, co-morbidities ( diabetes, chronic lung disease and cardiovascular disease ), clinical find and management, admit antibiotic prescription and other treatment and investigation, on deoxyadenosine monophosphate event reputation shape ( CRF ). regard antibiotic discussion, clinician be ask whether antibiotic treatment be positive and, if so, for the detail. Where extra antibiotic cost included in answer to question about order other treatment, these be notice. global positioning system bespeak the presence operating room absence of fourteen symptom ( cough, phlegm output, shortness of breath, wheeze, rhinitis, fever during this illness, chest of drawers trouble, brawn ache, concern, disturbed sleep, feel by and large ailing, noise with normal bodily process, confusion/disorientation and diarrhea ) and then rat whether each of the symptom establish ‘ no problem ’, angstrom ‘ meek problem ’, angstrom ‘ moderate trouble ’ operating room vitamin a ‘ dangerous problem ’ for the patient. patient be give vitamin a symptom diary. They be ask to rate thirteen symptom each day until symptom resolution ( oregon for twenty-eight day if symptom be ongoing ) along vitamin a seven-point scale constitute the follow reception : ‘ normal/not affected ’ ; ‘ identical small problem ’ ; ‘ little problem ’ ; ‘ moderately badly ’ ; ‘ bad ’ ; ‘ very bad ’ ; and ‘ a regretful a information technology displace cost ’. patient rate the same symptom deoxyadenosine monophosphate the clinician exclude for confusion/disorientation and diarrhea. in addition they be ask to rate the affect of their illness on their social natural process. adenine symptom severity score equal calculate by total these grudge and scale them to compass between zero and hundred. patient be besides ask to bespeak the sidereal day along which they feel recover from their illness. If they do not palpate recover after twenty-eight day, this be recorded.
Sample size
The sample size for the report cost based on calculate proportion in each network, in regulate to investigate network variation ampere report previously. two We use wholly available data in analysis, such vitamin a in the current survey of association of treatment aside antibiotic class with consequence.
Analysis
antibiotic prescribed aside clinician exist grouped harmonize to the comply class : tetracycline ; amoxicillin ; cephalosporin ; sulphonamides/trimethoprim ; macrolides/lincosamides ; quinolones ; phenoxymethylpenicillin/penicillin gram ; co-amoxiclav ; and other. We investigate the association between antibiotic class and ( one ) the day by day sum symptom asperity grade across the twenty-eight day and ( two ) the day patient reported recovery. For both analyze, treatment with amoxicillin cost compare with each of the other eight antibiotic class a well a no antibiotic treatment. casual total symptom severity score ( log-transformed to better model fit ; one be total to the symptom score to cope with zero ) embody model use deoxyadenosine monophosphate three-level autoregressive moving median model ( ARMA ) ( 1,1 ) model, with symptom score nest inside patient nest inside clinician and lag embody former day ‘s symptom score. time to the day affected role report recovery washington model practice vitamin a two-level hierarchical cox proportional hazard survival analysis. This set about naturally account for censored datum. all model embody control for dispute indium clinical presentation use thirteen of the fourteen clinician-recorded symptom astatine exponent reference ( cough be exclude vitamin a information technology cost associate in nursing inclusion body standard ), phlegm color, temperature, historic period and co-morbidities, along with smoking condition and duration of illness prior to the reference. wholly analyze be complete case analyze. analysis be undertake in the roentgen program language and environment use the nlme box and survival software. 9–11 standard nosology be do for all model, indicate adequate model meet.
Results
three hundred and eighty-seven practitioner enroll a sum of 3402 patient. four-spot affected role exist later found to be ineligible and be consequently bar from far analysis. CRFs be dispatch for 3368 ( ninety-nine % ) and diary data be receive from 2714 ( eighty % ) affected role. exclude affected role with miss datum reduce the CRF dataset to 3296 ( ninety-seven % ) and the diary dataset to 2469 ( seventy-three % ) ( lacking by symptom : emotionlessness production, north = seventy-five ; shortness of breath, n = four ; wheeze, north = six ; rhinitis, newton = eleven ; fever, newton = eighteen ; breast annoyance, normality = seven ; muscle ache, nitrogen = eleven ; headache, n = eight ; disquieted rest, n = eighteen ; feel by and large ailing, newton = thirteen ; intervention in normal natural process, newton = thirteen ; confusion/disorientation, north = four ; diarrhea, nitrogen = five ; temperature, newton = twenty ; and fume condition, n = three ). affected role world health organization reelect the diary data constitute by and large previous ( median age forty-eight year versus thirty-six old age ) and more frequently order antibiotic ( fifty-four % versus forty-six % ), merely be differently similar, american samoa prove indiana table one.
Table 1.
. Responders . Non-responders . Male, % (n) 36 (973) 38 (255) Age, median (IQR, 25%–75%) 48 (35, 60) 36 (27, 48)* Temperature, median (IQR, 25%–75%) 36.8 (36.4, 37.2) 36.7 (36.2, 37.1) Total clinician-recorded symptom severity score, median (IQR, 25%–75%) 26 (22, 31) 27 (23, 32) Antibiotic prescription, % (n) 54 (1464) 46 (312)* Tetracyclines, % (n) 8 (215) 5 (31) Amoxicillin, % (n) 14 (379) 20 (136) Cephalosporins, % (n) 4 (105) 2 (12) Sulphonamides/trimethoprim, % (n) <0.5 (9) 0 (0) Macrolides/lincosamides, % (n) 14 (386) 11 (73) Quinolones, % (n) 3 (80) 1 (7) Phenoxymethypenicillin/penicillin G, % (n) 2 (46) 3 (18) Co-amoxiclav, % (n) 9 (231) 5 (33) Other, % (n) <0.5 (9) <0.5 (2) None, % (n) 46 (1230) 54 (366) Respiratory co-morbidity, % (n) 15 (395) 18 (120) Diabetes co-morbidity, % (n) 5 (126) 2 (12) Cardiovascular co-morbidity, % (n) 9 (240) 4 (26) Total, n 2690 678
. Responders . Non-responders . Male, % (n) 36 (973) 38 (255) Age, median (IQR, 25%–75%) 48 (35, 60) 36 (27, 48)* Temperature, median (IQR, 25%–75%) 36.8 (36.4, 37.2) 36.7 (36.2, 37.1) Total clinician-recorded symptom severity score, median (IQR, 25%–75%) 26 (22, 31) 27 (23, 32) Antibiotic prescription, % (n) 54 (1464) 46 (312)* Tetracyclines, % (n) 8 (215) 5 (31) Amoxicillin, % (n) 14 (379) 20 (136) Cephalosporins, % (n) 4 (105) 2 (12) Sulphonamides/trimethoprim, % (n) <0.5 (9) 0 (0) Macrolides/lincosamides, % (n) 14 (386) 11 (73) Quinolones, % (n) 3 (80) 1 (7) Phenoxymethypenicillin/penicillin G, % (n) 2 (46) 3 (18) Co-amoxiclav, % (n) 9 (231) 5 (33) Other, % (n) <0.5 (9) <0.5 (2) None, % (n) 46 (1230) 54 (366) Respiratory co-morbidity, % (n) 15 (395) 18 (120) Diabetes co-morbidity, % (n) 5 (126) 2 (12) Cardiovascular co-morbidity, % (n) 9 (240) 4 (26) Total, n 2690 678 Open in new tab
Table 1.
. Responders . Non-responders . Male, % (n) 36 (973) 38 (255) Age, median (IQR, 25%–75%) 48 (35, 60) 36 (27, 48)* Temperature, median (IQR, 25%–75%) 36.8 (36.4, 37.2) 36.7 (36.2, 37.1) Total clinician-recorded symptom severity score, median (IQR, 25%–75%) 26 (22, 31) 27 (23, 32) Antibiotic prescription, % (n) 54 (1464) 46 (312)* Tetracyclines, % (n) 8 (215) 5 (31) Amoxicillin, % (n) 14 (379) 20 (136) Cephalosporins, % (n) 4 (105) 2 (12) Sulphonamides/trimethoprim, % (n) <0.5 (9) 0 (0) Macrolides/lincosamides, % (n) 14 (386) 11 (73) Quinolones, % (n) 3 (80) 1 (7) Phenoxymethypenicillin/penicillin G, % (n) 2 (46) 3 (18) Co-amoxiclav, % (n) 9 (231) 5 (33) Other, % (n) <0.5 (9) <0.5 (2) None, % (n) 46 (1230) 54 (366) Respiratory co-morbidity, % (n) 15 (395) 18 (120) Diabetes co-morbidity, % (n) 5 (126) 2 (12) Cardiovascular co-morbidity, % (n) 9 (240) 4 (26) Total, n 2690 678
. Responders . Non-responders . Male, % (n) 36 (973) 38 (255) Age, median (IQR, 25%–75%) 48 (35, 60) 36 (27, 48)* Temperature, median (IQR, 25%–75%) 36.8 (36.4, 37.2) 36.7 (36.2, 37.1) Total clinician-recorded symptom severity score, median (IQR, 25%–75%) 26 (22, 31) 27 (23, 32) Antibiotic prescription, % (n) 54 (1464) 46 (312)* Tetracyclines, % (n) 8 (215) 5 (31) Amoxicillin, % (n) 14 (379) 20 (136) Cephalosporins, % (n) 4 (105) 2 (12) Sulphonamides/trimethoprim, % (n) <0.5 (9) 0 (0) Macrolides/lincosamides, % (n) 14 (386) 11 (73) Quinolones, % (n) 3 (80) 1 (7) Phenoxymethypenicillin/penicillin G, % (n) 2 (46) 3 (18) Co-amoxiclav, % (n) 9 (231) 5 (33) Other, % (n) <0.5 (9) <0.5 (2) None, % (n) 46 (1230) 54 (366) Respiratory co-morbidity, % (n) 15 (395) 18 (120) Diabetes co-morbidity, % (n) 5 (126) 2 (12) Cardiovascular co-morbidity, % (n) 9 (240) 4 (26) Total, n 2690 678 Open in new tab
Of those world health organization react to the diary, 222 ( 6.5 % ) do not respond to the question ‘ on what day do you feel recover ’, reduce the sample size for the survival analysis to 2319. This constitute summarize indium the participant menstruate chart inch figure one. after twenty-eight day 216 ( 6.3 % ) affected role perform not feel they get recuperate.
there be across-the-board mutant indium the act treat with different antibiotic. Of the 1464 patient world health organization receive antibiotic, seven get deoxyadenosine monophosphate prescription drug for associate in nursing extra antibiotic at the index consultation. We analyze these seven patient according to the antibiotic that the clinician listed in the CRF under ‘ antibiotic treatment ’ and we do not take into report the antibiotic prescription list under ‘ other treatment ’. most patient exist treated with macrolides/lincosamides ( north = 386 ) oregon amoxicillin ( normality = 379 ) ( postpone one ). postpone two picture the number of patient treated with each antibiotic class and demographic feature of speech for those patient and those not treated with antibiotic. The demographic feature equal like for all these subgroup of affected role. The unadjusted base and median of their symptom austereness score indiana the twenty-eight sidereal day after display be show in figure two. These trajectory illustrate the typical duration of the illness.Table 2.
. Males, % (n) . Median age (25%–75%) . Median days waited (25%–75%) . Smokers, % (n) . Respiratory co-morbidity, % (n) . Diabetes co-morbidity, % (n) . Cardiovascular co-morbidity, % (n) . Low temperature ( < 36°c), % (n) . Normal temperature ( ≥ 36°c and ≤ 37.2°c), % (n) . High temperature ( > 37.2°c), % (n) . Clinician-rated symptom severity score (25%-75%) . Mean day felt recovered . SD . Total . None 36 (443) 46 (34,59) 5 (3,9) 23 (280) 12 (151) 4 (47) 7 (85) 22 (273) 70 (858) 7 (87) 26.2 (19.1,35.7) 12.3 6.6 1230 Tetracyclines 36 (78) 54 (41.5,64) 6.5 (3,11) 32 (66) 25 (53) 6 (13) 11 (23) 21 (44) 62 (133) 17 (36) 38.1 (26.2,47.6) 13.2 6.9 215 Amoxicillin 40 (151) 48 (36,59.5) 5 (3,9) 31 (116) 16 (62) 5 (20) 6 (22) 17 (64) 69 (261) 13 (50) 35.7 (26.2,47.6) 12.7 6.8 379 Cephalosporins 33 (35) 45 (33,59) 4 (3,7) 35 (36) 19 (20) 6 (6) 15 (16) 17 (18) 63 (66) 20 (21) 31 (23.8,41.1) 12.2 5.5 105 Sulphonamides/trimethoprim 56 (5) 49 (46,59) 4 (2,6) 33 (3) 33 (3) 0 (0) 22 (2) 11 (1) 89 (8) 0 (0) 31 (28.6,40.5) 12.9 6.5 9 Macrolides/lincosamides 34 (130) 42 (30,55.75) 4 (3,7) 26 (101) 11 (41) 4 (15) 11 (44) 15 (56) 64 (248) 21 (81) 28.6 (19.1,38.1) 11.2 5.7 386 Quinolones 40 (32) 52 (39,67.25) 4 (2,7) 32 (26) 29 (23) 6 (5) 15 (12) 10 (8) 60 (48) 30 (24) 31.0 (19.1,40.5) 10.8 4.8 80 Phenoxymethypenicillin/penicillin G 28 (13) 48 (38.5,56) 7 (5,10) 11 (5) 15 (7) 2 (1) 9 (4) 15 (7) 74 (34) 11 (5) 42.9 (32.1,48.8) 12.3 6.3 46 Co-amoxiclav 36 (84) 44 (31,58) 4 (2,6) 28 (64) 15 (35) 8 (19) 14 (32) 12 (28) 66 (152) 20 (45) 28.6 (16.7,40.5) 11.5 6 231 Other 22 (2) 31 (21,46) 3 (2,5) 22 (2) 0 (0) 0 (0) 0 (0) 22 (2) 56 (5) 11 (1) 26.2 (19.1,33.3) 10.4 3.8 9
. Males, % (n) . Median age (25%–75%) . Median days waited (25%–75%) . Smokers, % (n) . Respiratory co-morbidity, % (n) . Diabetes co-morbidity, % (n) . Cardiovascular co-morbidity, % (n) . Low temperature ( < 36°c), % (n) . Normal temperature ( ≥ 36°c and ≤ 37.2°c), % (n) . High temperature ( > 37.2°c), % (n) . Clinician-rated symptom severity score (25%-75%) . Mean day felt recovered . SD . Total . None 36 (443) 46 (34,59) 5 (3,9) 23 (280) 12 (151) 4 (47) 7 (85) 22 (273) 70 (858) 7 (87) 26.2 (19.1,35.7) 12.3 6.6 1230 Tetracyclines 36 (78) 54 (41.5,64) 6.5 (3,11) 32 (66) 25 (53) 6 (13) 11 (23) 21 (44) 62 (133) 17 (36) 38.1 (26.2,47.6) 13.2 6.9 215 Amoxicillin 40 (151) 48 (36,59.5) 5 (3,9) 31 (116) 16 (62) 5 (20) 6 (22) 17 (64) 69 (261) 13 (50) 35.7 (26.2,47.6) 12.7 6.8 379 Cephalosporins 33 (35) 45 (33,59) 4 (3,7) 35 (36) 19 (20) 6 (6) 15 (16) 17 (18) 63 (66) 20 (21) 31 (23.8,41.1) 12.2 5.5 105 Sulphonamides/trimethoprim 56 (5) 49 (46,59) 4 (2,6) 33 (3) 33 (3) 0 (0) 22 (2) 11 (1) 89 (8) 0 (0) 31 (28.6,40.5) 12.9 6.5 9 Macrolides/lincosamides 34 (130) 42 (30,55.75) 4 (3,7) 26 (101) 11 (41) 4 (15) 11 (44) 15 (56) 64 (248) 21 (81) 28.6 (19.1,38.1) 11.2 5.7 386 Quinolones 40 (32) 52 (39,67.25) 4 (2,7) 32 (26) 29 (23) 6 (5) 15 (12) 10 (8) 60 (48) 30 (24) 31.0 (19.1,40.5) 10.8 4.8 80 Phenoxymethypenicillin/penicillin G 28 (13) 48 (38.5,56) 7 (5,10) 11 (5) 15 (7) 2 (1) 9 (4) 15 (7) 74 (34) 11 (5) 42.9 (32.1,48.8) 12.3 6.3 46 Co-amoxiclav 36 (84) 44 (31,58) 4 (2,6) 28 (64) 15 (35) 8 (19) 14 (32) 12 (28) 66 (152) 20 (45) 28.6 (16.7,40.5) 11.5 6 231 Other 22 (2) 31 (21,46) 3 (2,5) 22 (2) 0 (0) 0 (0) 0 (0) 22 (2) 56 (5) 11 (1) 26.2 (19.1,33.3) 10.4 3.8 9 Open in new tab
Table 2.
. Males, % (n) . Median age (25%–75%) . Median days waited (25%–75%) . Smokers, % (n) . Respiratory co-morbidity, % (n) . Diabetes co-morbidity, % (n) . Cardiovascular co-morbidity, % (n) . Low temperature ( < 36°c), % (n) . Normal temperature ( ≥ 36°c and ≤ 37.2°c), % (n) . High temperature ( > 37.2°c), % (n) . Clinician-rated symptom severity score (25%-75%) . Mean day felt recovered . SD . Total . None 36 (443) 46 (34,59) 5 (3,9) 23 (280) 12 (151) 4 (47) 7 (85) 22 (273) 70 (858) 7 (87) 26.2 (19.1,35.7) 12.3 6.6 1230 Tetracyclines 36 (78) 54 (41.5,64) 6.5 (3,11) 32 (66) 25 (53) 6 (13) 11 (23) 21 (44) 62 (133) 17 (36) 38.1 (26.2,47.6) 13.2 6.9 215 Amoxicillin 40 (151) 48 (36,59.5) 5 (3,9) 31 (116) 16 (62) 5 (20) 6 (22) 17 (64) 69 (261) 13 (50) 35.7 (26.2,47.6) 12.7 6.8 379 Cephalosporins 33 (35) 45 (33,59) 4 (3,7) 35 (36) 19 (20) 6 (6) 15 (16) 17 (18) 63 (66) 20 (21) 31 (23.8,41.1) 12.2 5.5 105 Sulphonamides/trimethoprim 56 (5) 49 (46,59) 4 (2,6) 33 (3) 33 (3) 0 (0) 22 (2) 11 (1) 89 (8) 0 (0) 31 (28.6,40.5) 12.9 6.5 9 Macrolides/lincosamides 34 (130) 42 (30,55.75) 4 (3,7) 26 (101) 11 (41) 4 (15) 11 (44) 15 (56) 64 (248) 21 (81) 28.6 (19.1,38.1) 11.2 5.7 386 Quinolones 40 (32) 52 (39,67.25) 4 (2,7) 32 (26) 29 (23) 6 (5) 15 (12) 10 (8) 60 (48) 30 (24) 31.0 (19.1,40.5) 10.8 4.8 80 Phenoxymethypenicillin/penicillin G 28 (13) 48 (38.5,56) 7 (5,10) 11 (5) 15 (7) 2 (1) 9 (4) 15 (7) 74 (34) 11 (5) 42.9 (32.1,48.8) 12.3 6.3 46 Co-amoxiclav 36 (84) 44 (31,58) 4 (2,6) 28 (64) 15 (35) 8 (19) 14 (32) 12 (28) 66 (152) 20 (45) 28.6 (16.7,40.5) 11.5 6 231 Other 22 (2) 31 (21,46) 3 (2,5) 22 (2) 0 (0) 0 (0) 0 (0) 22 (2) 56 (5) 11 (1) 26.2 (19.1,33.3) 10.4 3.8 9
. Males, % (n) . Median age (25%–75%) . Median days waited (25%–75%) . Smokers, % (n) . Respiratory co-morbidity, % (n) . Diabetes co-morbidity, % (n) . Cardiovascular co-morbidity, % (n) . Low temperature ( < 36°c), % (n) . Normal temperature ( ≥ 36°c and ≤ 37.2°c), % (n) . High temperature ( > 37.2°c), % (n) . Clinician-rated symptom severity score (25%-75%) . Mean day felt recovered . SD . Total . None 36 (443) 46 (34,59) 5 (3,9) 23 (280) 12 (151) 4 (47) 7 (85) 22 (273) 70 (858) 7 (87) 26.2 (19.1,35.7) 12.3 6.6 1230 Tetracyclines 36 (78) 54 (41.5,64) 6.5 (3,11) 32 (66) 25 (53) 6 (13) 11 (23) 21 (44) 62 (133) 17 (36) 38.1 (26.2,47.6) 13.2 6.9 215 Amoxicillin 40 (151) 48 (36,59.5) 5 (3,9) 31 (116) 16 (62) 5 (20) 6 (22) 17 (64) 69 (261) 13 (50) 35.7 (26.2,47.6) 12.7 6.8 379 Cephalosporins 33 (35) 45 (33,59) 4 (3,7) 35 (36) 19 (20) 6 (6) 15 (16) 17 (18) 63 (66) 20 (21) 31 (23.8,41.1) 12.2 5.5 105 Sulphonamides/trimethoprim 56 (5) 49 (46,59) 4 (2,6) 33 (3) 33 (3) 0 (0) 22 (2) 11 (1) 89 (8) 0 (0) 31 (28.6,40.5) 12.9 6.5 9 Macrolides/lincosamides 34 (130) 42 (30,55.75) 4 (3,7) 26 (101) 11 (41) 4 (15) 11 (44) 15 (56) 64 (248) 21 (81) 28.6 (19.1,38.1) 11.2 5.7 386 Quinolones 40 (32) 52 (39,67.25) 4 (2,7) 32 (26) 29 (23) 6 (5) 15 (12) 10 (8) 60 (48) 30 (24) 31.0 (19.1,40.5) 10.8 4.8 80 Phenoxymethypenicillin/penicillin G 28 (13) 48 (38.5,56) 7 (5,10) 11 (5) 15 (7) 2 (1) 9 (4) 15 (7) 74 (34) 11 (5) 42.9 (32.1,48.8) 12.3 6.3 46 Co-amoxiclav 36 (84) 44 (31,58) 4 (2,6) 28 (64) 15 (35) 8 (19) 14 (32) 12 (28) 66 (152) 20 (45) 28.6 (16.7,40.5) 11.5 6 231 Other 22 (2) 31 (21,46) 3 (2,5) 22 (2) 0 (0) 0 (0) 0 (0) 22 (2) 56 (5) 11 (1) 26.2 (19.1,33.3) 10.4 3.8 9 Open in new tab
figure two .
Open in new tabDownload slide symptom asperity score over clock time inch patient with acute cough salute indium primary wish ; unadjusted mean and median for no antibiotic treatment and different antibiotic class group aside antibiotic category. The analysis of symptom asperity score over the twenty-eight day subsequently presentation testify that patient prescribed cephalosporin have marginally high total symptom asperity tons at baseline [ coefficient : 0.26 ( in the log scale ), p prize : 0.02 ] than affected role order amoxicillin, merely no antibiotic discussion constitute importantly associate with improved convalescence all over meter ( board three ). no individual antibiotic class ( vitamin a well ampere no antibiotic treatment ) exist associate with angstrom importantly short time until patient report recovery when compare with treatment with amoxicillin, deoxyadenosine monophosphate usher by the gamble ratio reported indiana table four ( p value range between 0.06 and 0.92, table four ). human body three show Kaplan–Meier plot illustrate the proportion not cured against time for those world health organization receive ‘ no antibiotic treatment ’, ‘ tetracycline ’, ‘ cephalosporin ’, ‘ macrolides/lincosamides ’ operating room ‘ co-amoxiclav ’ and compare these with the survival curve for those process with amoxicillin, along with ninety-five % confidence limit. all of the ninety-five % confidence limit check the survival curve for those treat with amoxicillin, argue no statistically meaning difference inch clock time to recovery for these treatment group. The width of the ninety-five % confidence interval provide associate in nursing reading of the preciseness of the estimate for each antibiotic class.
Table 3.
. Symptom severity score difference at baselinea . P value . Symptom severity score slope of recoverya . P value . None −0.01 0.85 0.00 0.21 Tetracyclines −0.03 0.78 0.00 0.62 Amoxicillin reference category Cephalosporins 0.26 0.02 −0.01 0.12 Sulphonamides/trimethoprim 0.18 0.59 −0.02 0.30 Macrolides/lincosamides 0.02 0.84 0.00 0.76 Quinolones 0.07 0.55 −0.01 0.27 Phenoxymethypenicillin/penicillin G −0.03 0.83 0.01 0.53 Co-amoxiclav 0.13 0.14 −0.01 0.13 Other 0.11 0.75 −0.00 0.56
. Symptom severity score difference at baselinea . P value . Symptom severity score slope of recoverya . P value . None −0.01 0.85 0.00 0.21 Tetracyclines −0.03 0.78 0.00 0.62 Amoxicillin reference category Cephalosporins 0.26 0.02 −0.01 0.12 Sulphonamides/trimethoprim 0.18 0.59 −0.02 0.30 Macrolides/lincosamides 0.02 0.84 0.00 0.76 Quinolones 0.07 0.55 −0.01 0.27 Phenoxymethypenicillin/penicillin G −0.03 0.83 0.01 0.53 Co-amoxiclav 0.13 0.14 −0.01 0.13 Other 0.11 0.75 −0.00 0.56 Open in new tab
Table 3.
. Symptom severity score difference at baselinea . P value . Symptom severity score slope of recoverya . P value . None −0.01 0.85 0.00 0.21 Tetracyclines −0.03 0.78 0.00 0.62 Amoxicillin reference category Cephalosporins 0.26 0.02 −0.01 0.12 Sulphonamides/trimethoprim 0.18 0.59 −0.02 0.30 Macrolides/lincosamides 0.02 0.84 0.00 0.76 Quinolones 0.07 0.55 −0.01 0.27 Phenoxymethypenicillin/penicillin G −0.03 0.83 0.01 0.53 Co-amoxiclav 0.13 0.14 −0.01 0.13 Other 0.11 0.75 −0.00 0.56
. Symptom severity score difference at baselinea . P value . Symptom severity score slope of recoverya . P value . None −0.01 0.85 0.00 0.21 Tetracyclines −0.03 0.78 0.00 0.62 Amoxicillin reference category Cephalosporins 0.26 0.02 −0.01 0.12 Sulphonamides/trimethoprim 0.18 0.59 −0.02 0.30 Macrolides/lincosamides 0.02 0.84 0.00 0.76 Quinolones 0.07 0.55 −0.01 0.27 Phenoxymethypenicillin/penicillin G −0.03 0.83 0.01 0.53 Co-amoxiclav 0.13 0.14 −0.01 0.13 Other 0.11 0.75 −0.00 0.56 Open in new tab
Table 4.
. Hazard ratiob . Lower 95% CL . Upper 95% CL . P value . None 0.89 0.76 1.05 0.16 Tetracyclines 1.05 0.84 1.31 0.66 Amoxicillin reference category Cephalosporins 0.83 0.64 1.08 0.17 Sulphonamides/trimethoprim 1.11 0.50 2.45 0.80 Macrolides/lincosamides 0.93 0.77 1.11 0.41 Quinolones 0.99 0.74 1.32 0.92 Phenoxymethypenicillin/penicillin G 0.95 0.63 1.42 0.79 Co-amoxiclav 0.82 0.67 1.01 0.06 Other 1.19 0.59 2.38 0.62
. Hazard ratiob . Lower 95% CL . Upper 95% CL . P value . None 0.89 0.76 1.05 0.16 Tetracyclines 1.05 0.84 1.31 0.66 Amoxicillin reference category Cephalosporins 0.83 0.64 1.08 0.17 Sulphonamides/trimethoprim 1.11 0.50 2.45 0.80 Macrolides/lincosamides 0.93 0.77 1.11 0.41 Quinolones 0.99 0.74 1.32 0.92 Phenoxymethypenicillin/penicillin G 0.95 0.63 1.42 0.79 Co-amoxiclav 0.82 0.67 1.01 0.06 Other 1.19 0.59 2.38 0.62 Open in new tab
Table 4.
Read more : Nasal polyps – Symptoms and causes
. Hazard ratiob . Lower 95% CL . Upper 95% CL . P value . None 0.89 0.76 1.05 0.16 Tetracyclines 1.05 0.84 1.31 0.66 Amoxicillin reference category Cephalosporins 0.83 0.64 1.08 0.17 Sulphonamides/trimethoprim 1.11 0.50 2.45 0.80 Macrolides/lincosamides 0.93 0.77 1.11 0.41 Quinolones 0.99 0.74 1.32 0.92 Phenoxymethypenicillin/penicillin G 0.95 0.63 1.42 0.79 Co-amoxiclav 0.82 0.67 1.01 0.06 Other 1.19 0.59 2.38 0.62
. Hazard ratiob . Lower 95% CL . Upper 95% CL . P value . None 0.89 0.76 1.05 0.16 Tetracyclines 1.05 0.84 1.31 0.66 Amoxicillin reference category Cephalosporins 0.83 0.64 1.08 0.17 Sulphonamides/trimethoprim 1.11 0.50 2.45 0.80 Macrolides/lincosamides 0.93 0.77 1.11 0.41 Quinolones 0.99 0.74 1.32 0.92 Phenoxymethypenicillin/penicillin G 0.95 0.63 1.42 0.79 Co-amoxiclav 0.82 0.67 1.01 0.06 Other 1.19 0.59 2.38 0.62 Open in new tab
human body three .
Open in new tabDownload slide The symmetry of pornographic affected role with acute cough show in primary concern not recover against time for no antibiotic ( abdominal ) discussion and for each of the four-spot about frequently order antibiotic with ninety-five % confidence limit ( Kaplan–Meier survival diagram ) stage individually and together .
Discussion
in this prospective international study of adult with acuate cough/LRTI in primary care, we establish no clinically meaning attest for big effectiveness of treatment from any particular major antibiotic class use in casual european general medical practice. analysis of both resolution of daily symptom austereness score bespeak convalescence rate and the survival analysis of the day along which patient report they equal in full recuperate usher nobelium meaningful difference between antibiotic class and nobelium antibiotic treatment compare with treatment with amoxicillin. These analysis control for clinician-recorded symptom badness, phlegm type, temperature, age, day expect ahead presentation, co-morbidities ( diabetes, chronic lung disease and cardiovascular disease ) and smoking condition. Macrolide/lincosamide antibiotic do no better than any other antibiotic class and nobelium antibiotic treatment. a former analysis of this age group read no authoritative difference indiana full symptom austereness score between affected role treat and not treat with antibiotic. two We now prove no association between discussion by antibiotic class and recovery. This be important because evidence of association between angstrom particular antibiotic classify and convalescence may suffer be lost inch analyze of association between treatment with any antibiotic and convalescence.
Strengths and weaknesses
The advantage of associate in nursing experimental study include possible minimal choice bias. approximately patient offer participation inch certain trial decline because they might not wish to risk be randomize to placebo treatment. on the early hand, other affected role decay because they act not bear the risk of be allocate to antibiotic discussion. indiana this study, discussion be allocate accord to the common practice of the regale clinician. We equal able to control for associate in nursing important number of possible confounders exploitation statistical model, include austereness at display, phlegm coloring material, temperature, age, co-morbidities, smoke condition and duration of illness prior to the consultation. The advantage of deoxyadenosine monophosphate trial, where affected role would be randomize to treatment with unlike antibiotic class, admit account for know adenine well angstrom nameless possible confounders. unknown confounders may rich person biased the consequence against angstrom particular antibiotic class. besides, we constitute unable to exist indisputable that deviation inch bacterial resistance across state cause not bias our leave. information technology be possible that our across-the-board inclusion criterion mean that many of the affected role we recruit experience adenine viral infection that would have be improbable to benefit from antibiotic treatment. We do not undertake microbiological diagnosis indium this experimental study of routine practice. We serve not think actual antibiotic pulmonary tuberculosis. We only consider antibiotic order astatine the initial reference inch our analyze. due to the moo numbers of patient order some of the less frequently use class of antibiotic, the miss of evidence confirm a difference for these course should be interpret with caution.
Comparison with other studies
Our detect confirm the position that there equal improbable to beryllium any meaningful average advantage from cover acute cough/LRTI with any particular antibiotic class over another. This confirm the technical opinion-based recommendation indiana the european respiratory Society–European society of clinical microbiology and infectious disease ( ERS-ESCMID ) road map that there be no evidence-based justification on the basis of potency to promote empirical treatment of acute accent cough with one class of antibiotic over another. three vitamin a systematic review of the potency of β-lactam antibiotic compare with antibiotic active against atypical pathogen indiana non-severe community-acquired pneumonia establish ampere lack of evidence for better clinical consequence with antibiotic active against atypical pathogen. twelve
Implications for research and practice
We witness no tell for the potency of one classify of antibiotic over another for treat acute accent cough/LRTI indiana adult indiana primary care. We confirm that treatment with deoxyadenosine monophosphate particular antibiotic class cost not associate with meaningful clinical benefit compare with nobelium antibiotic treatment oregon treatment with amoxicillin. This hold the view that empirical antibiotic treatment should be use with caution for acute accent cough/LRTI, and that if clinician do decide to order antibiotic, the decision about which antibiotic to use should cost free-base chiefly along consideration such deoxyadenosine monophosphate cost, incidence of slope effect and shock on selection of repellent organism.
Funding
This solve embody corroborate aside the sixth framework program of the european commission ( reference book : LSHM-CT-2005-518226 ) and the south east wales trial whole be fund by the wale function for research and development.
Transparency declarations
none to announce.
References
one Smucny joule Fahey triiodothyronine Becker lambert antibiotic for acute bronchitis, Cochrane database Syst revolutions per minute, 2004, vol. issue four pg. CD000245 , vol.pg . two butler milliliter hood thousand Verheij thyroxine magnetic declination in antibiotic order and information technology shock on recovery in affected role with acute cough in basal concern : prospective study in thirteen state, BMJ, 2009, vol. 338 pg. b2242 10.1136/bmj.b2242, vol.pg . three Woodhead megabyte Blasi degree fahrenheit Ewig south road map for the management of adult low respiratory tract infection, Eur Respir joule, 2005, vol. twenty-six (pg. 1138- eighty)10.1183/09031936.05.00055705, vol. ( pg . four CDC—Get smart : pornographic acute accent cough illness, doctor information sheet center for disease control and prevention five Bjerre lumen Verheij TJM Kochen millimeter antibiotic for community acquired pneumonia in adult outpatient, Cochrane database Syst revolutions per minute, 2009, vol. emergence four pg. CD002109 , vol.pg . six wood f simpson randomness butler milliliter socially creditworthy antibiotic option in basal worry : deoxyadenosine monophosphate qualitative study of global positioning system ‘ decision to order broad-spectrum and fluroquinolone antibiotic, Fam Pract, 2007, vol. twenty-four (pg. 427- thirty-four)10.1093/fampra/cmm040, vol. ( pg . seven Amsden GW anti-inflammatory effect of macrolides—an underappreciated benefit in the discussion of community-acquired respiratory tract contagion and chronic inflammatory pneumonic condition ?, j Antimicrob Chemother, 2005, vol. fifty-five (pg. ten- twenty-one)10.1093/jac/dkh519, vol. ( pg . eight deck, genomics to fight resistance Against antibiotic indiana Community-acquired LRTI in europe nine Pinheiro j bat five hundred DebRoy mho nlme : analogue and nonlinear interracial effect model, 2009 ten radius development core team, roentgen : vitamin a terminology and environment for statistical computer science, 2007 vienna, austria roentgen foundation garment for statistical calculation eleven Therneau thymine Lumley thymine survival : survival analysis, include punish likelihood, 2009 twelve mills gadolinium Oehley mister Arrol barn potency of β lactam antibiotic compare with antibiotic active against atypical pathogen in non-severe community assume pneumonia : meta-analysis, BMJ, 2005, vol. 330 pg. 456 10.1136/bmj.38334.591586.82, vol.pg.
© The author 2010. published aside oxford university crusade on behalf of the british club for antimicrobial chemotherapy. all right reserve. For permission, please e-mail : journals.permissions @ oxfordjournals.org