Figures
- Figure 1–
Flow chart of included and excluded studies.
Tables
- Table 1– Characteristics of the 51 included studies
Characteristics Studies RCT 15 (29) Cohort study 36 (71) Evaluation of a pharmacological compound 15 (29) Patients median (IQR) 109 (78–259) Follow-up period months median (range) 12 (3–90) Data are presented as n (%), unless otherwise stated. RCT: randomised controlled trial; IQR: interquartile range.
- Table 2– Overview of exacerbation onset algorithms in the included studies#
Algorithm Studies n Criteria for defining onset Consecutive days DY SV SP CO WH ST CS FE CT FA DE NA Modified Anthonisen 32 Increase in ≥2 symptoms including 1 major ≥2+ Ma Ma Ma Mi Mi Mi Mi - - - - - Modified Anthonisen + fever 4 Increase in ≥2 symptoms including 1 major ≥2 Ma Ma Ma Mi Mi Mi Mi Mi - - - - Only major symptoms 4 Increase in ≥1 symptom ≥2§ + + + - - - - - - - - - Vijayasaratha 11 1 Graded symptom scale; increase in ≥1 point in 1 major symptom ≥2 Ma Ma Ma Mi - - Mi Mi Mi - - - Alvarez-Mon 22 1 Increase in ≥2 symptoms NS + + + + - - - - - - - - Ekberg-Jansson 23 1 ≥3 points; major symptom = 2 points, minor = 1 point ≥2 Mi Mi Mi Ma - - - - - - Mi - Van Schayk 24 1 Increase in ≥2 symptoms including 1 major NS Mi Mi Ma Ma Mi Mi Mi Mi Mi Mi Mi - Sund 25 1 Graded symptom score; increase of ≥1 point for 2 symptoms ≥2 + + + - - - - + + - - Banerjee 26 1 “Increase in symptoms” NS - - - - - - - - - - - - Calverley 27 1 Graded symptom score, 5 different criteria described¶ ≥3 + - - + - - - + - - - + Casaburi 28 1 Increase in ≥1 symptom ≥3 + - - + + - - - + - - - Dowson 29 1 Increase in ≥2 symptoms ≥1 + + + - - - - - - - - - Pela 30 1 Graded symptom score; increase in ≥3 points NS + + + + - - - + - - - - Dahl 31 1 Increase in ≥1 symptom ≥3 + + + + + - - - - - - - DY: dyspnoea; SV: sputum volume; SP: sputum purulence; CO: cough; WH: wheezing; ST: sore throat; CS: coryzal symptoms; FE: fever; CT: chest tightness; FA: fatigue; DE: difficulty with expectoration; NA: night-time awakenings; NS: not specified; Ma: major symptom; +: included in the algorithm but no distinction between major and minor symptoms; -: not included in the algorithm; Mi: minor symptom. #: n = 51. ¶: see online supplementary material. +: not specified in three studies. §: not specified in two studies.
- Table 3– Recovery time, severity classification, data registration, attempts to enhance validity of diary registration, blinded adjudication of exacerbations and handling of missing diary data#
Criteria Studies n (%) Recovery rules¶ Not reported 15 (29) Symptom score back to an individual baseline 28 (55) Symptom score back to a fixed value 1 (2) First day not meeting onset criteria 6 (12) All symptoms resolved 1 (2) Data registration Not reported 3 (6) Daily written diary cards 39 (76) Daily electronic diary cards 1 (2) Telephone consultations+ 4 (8) Clinic visits 4 (8) Blinded adjudication of exacerbations Not reported 48 (94) Yes 3 (6) Concurrence with event-based exacerbations Not reported 12 (24) Yes 39 (76) Severity classification Not reported 18 (35) By symptom count at onset§ 24 (47) By number of major symptoms 4 (8) By duration§ 5 (10) By healthcare utilisation 3 (6) Attempts to enhance validity and compliance of diary registrationƒ Not reported 12 (31) Run-in period## 2 (5) Telephone check## 4 (10) Clinic visit check 24 (62) Random home visits 1 (3) Handling of missing diary data Not reported 48 (94) Multiple imputation 2 (4) Retrospective interviews 1 (2) #: n = 51; ¶: rules to define recovery and/or identify new or recurrent events; +: performed monthly in all studies; §: three studies used a severity classification by symptom count and duration; ƒ: from 39 studies using daily diary cards; ##: two studies combined a run-in period with telephone checks.
- Table 4– Exacerbation rate, type, severity and concurrence with reported/treated events of exacerbation using four different algorithms
Symptom algorithm Exacerbations# n Incidence exacerbations·person-yr−1 (95% CI) Exacerbations n (% total) (% patients with an exacerbation) Clustering type n (% total) Recovery time days Anthonisen classificaton n (% total) Total symptom count n Concurrence with reported and treated events 0 1 >1 Initial Relapsed Recurrent I II III Reported events n (%) Treated events n (%) Missed admissions¶ n 1. 119 1.7 (1.4–2.1) 81 (59) 25 (18) (45) 31 (23) (55) 69 (58) 13 (11) 50 (42) 14 (7–22) 27 (23) 38 (32) 54 (45) 57 (27–94) 62 (52) 31 (26) 1 2. 119 1.7 (1.4 –2.1) 81 (59) 25 (18) (45) 31 (23) (55) 69 (58) 13 (11) 50 (42) 14 (7–22) 27 (23) 38 (32) 54 (45) 57 (27–94) 62 (52) 31 (26) 1 3. ≥1 132 1.9 (1.6–2.3) 73 (53) 30 (22) (47) 34 (25) (53) 75 (57) 14 (11) 57 (43) 13 (7–22) 27 (20) 39 (30) 66 (50) 51 (23–91) 62 (47) 31 (24) 1 4. ≥2 54 1.5 (0.6–1.0) 105 (77) 20 (15) (63) 12 (8) (37) 35 (65) 2 (4) 19 (35) 14 (8–21) 27 (50) 27 (50) 70 (42–125) 41 (76) 22 (41) 5 Data are presented as median (interquartile range), unless otherwise stated. The symptom alogrithms are: 1. Modified Anthonisen; 2. Modified Anthonisen+fever; 3. ≥1 major symptom; 4. ≥2 major symptoms. #: relapsed exacerbations excluded (original exacerbation and relapse are considered as the same event); ¶: hospital and emergency room visits missed.
Supplementary material
Files in this Data Supplement:
