Home telemonitoring and remote feedback between clinic visits for asthma

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The mean percentage of male participants was 45.6%, with individual studies ranging from 23.5% to 74.0%.

“The mean percentage male indicated a relatively even split of males and females (45.6% male), although the percentage male in individual studies ranged from 23.5% to 74.0%.”

May combine two claims.

Adult or adolescent participants were recruited in 12 trials, children only in five, and both groups in one.

“Twelve studies recruited adults or adults and adolescents (Bateman 2000; Cingi 2015; Donald 2008; Finkelstein 2005; Kokubu 1999; Kokubu 2000; Liu 2011; Ostojic 2005; Prabhakaran 2009; Ryan 2012; van der Meer 2009; Young 2012), five studies recruited only children (Deschildre 2012; Guendelman 2002; Jan 2007; Voorend-van Bergen 2015; Xu 2010) and one study recruited both adults and children (Willems”

May combine two claims.

One adult study reported the number of people classed as well controlled on the ACT; the effect favoured telemonitoring, but the confidence interval included the possibility of no difference (OR 2.46, 95% CI 0.94 to 6.41).

“The effect favoured home telemonitoring and feedback, but the CI included the possibility that the effect may be the same as that for usual monitoring (OR 2.46, 95% CI 0.94 to 6.41).”

• cites Cingi 2015Not checked yet

  Not checked in the free run.

May combine two claims.

No included study recorded serious adverse events separately from asthma exacerbations, and none reported whether any participant died.

“None of the studies recorded serious adverse events separately from asthma exacerbations, and none reported whether anyone died during the study.”

May combine two claims.

Two adult studies using the ACQ produced very different effects, making the pooled result difficult to interpret (MD -0.24, 95% CI -0.72 to 0.24; 478 participants; two studies; I2=91%).

“These studies showed very different effects, which made the pooled result difficult to interpret (mean difference (MD) -0.24, 95% CI -0.72 to 0.24; 478 participants; two studies; I 2 = 91%).”

• cites Ryan 2012Not checked yet

  Not checked in the free run.

• cites van der Meer 2009Not checked yet

  Not checked in the free run.

May combine two claims.

Most trials enrolled participants whose asthma was mild to moderate and persistent, with follow-up lasting three to 12 months.

“Studies generally recruited people with mild to moderate persistent asthma and followed them for between three and 12 months.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Weighted mean participant age across studies was 31.6 years (range 7 to 52.8).

“The overall weighted mean of population ages was 31.6 years (range, seven to 52.8).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Included trials had 16 to 288 participants, with a median of 120.

“The number of participants in each study ranged from 16 to 288, and the median number was 120.”

May combine two claims.

Validated asthma control measures were used in only a minority of studies, and the four studies that used them could not be combined in one analysis.

“Most studies did not use validated measures of asthma control, and the four that did could not be pooled in a single analysis (Cingi 2015; Ryan 2012; van der Meer 2009; Voorend-van Bergen 2015).”

• cites Cingi 2015Not checked yet

  Not checked in the free run.

• cites Ryan 2012Not checked yet

  Not checked in the free run.

• cites van der Meer 2009Not checked yet

  Not checked in the free run.

May combine two claims. The quote only partly states this claim.

The review included 18 randomised trials with a total of 2268 participants.

“We found 18 studies including 2268 participants: 12 in adults, 5 in children and one in individuals from both age groups.”

• Also stated in: § Included studies · plain

  “Eighteen studies, including a total of 2268 participants, met the inclusion criteria for this review”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The mean age was 10.4 years in child populations and 41.6 years in adult studies.

“The mean age of child populations was 10.4 years (range, seven to 12.1) and the mean age in adult studies was 41.6 (range, 24.7 to 52.8).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Four trials measured asthma control using three distinct instruments.

“Asthm a control was reported in 3 dif f erent ways across 4 studies”

May combine two claims.

The test for subgroup differences for hospital admission was statistically significant, suggesting possible benefit for adults in reducing exacerbations requiring hospital admission (OR 0.24, 95% CI 0.06 to 0.94).

“However, the test for subgroup differences was statistically significant, suggesting possible benefit for adults in reducing the number of exacerbations requiring hospital admission (OR 0.24, 95% CI 0.06 to 0.94).”

May combine two claims.

One study had very unbalanced dropout (12% vs 57%) and showed effects different from the other two studies.

“Very unbalanced dropout in 1 study showing dif f erent ef f ects f rom the other 2 (12% vs 57%)”

May combine two claims.

Removing Deschildre 2012, Kokubu 2000, and Xu 2010 from the hospitalisation outcome gave a similar magnitude but more imprecise overall effect (OR 0.58, 95% CI 0.13 to 2.56).

“When all three studies (Deschildre 2012; Kokubu 2000; Xu 2010) were removed from the exacerbation requiring hospitalisation outcome, the overall effect showed a similar magnitude but became more imprecise (OR 0.56, 95% CI 0.21 to 1.49, to OR 0.58, 95% CI 0.13 to 2.56).”

• cites Deschildre 2012Not checked yet

  Not checked in the free run.

• cites Kokubu 2000Not checked yet

  Not checked in the free run.

• cites Xu 2010Not checked yet

  Not checked in the free run.

May combine two claims.

The pooled effect for unscheduled healthcare visits was difficult to interpret because of variation between study results (OR 0.99, 95% CI 0.37 to 2.62; 430 participants; three studies; I2=73%), and telemonitoring did not clearly increase or decrease the number of people making such visits.

“Variation between study results made the pooled effect difficult to interpret (OR 0.99, 95% CI 0.37 to 2.62; 430 participants; three studies; I 2 = 73%), but telemonitoring did not lead to a clear increase or decrease in the number of people making unscheduled healthcare visits.”

May combine two claims.

Deschildre 2012 reported mean unscheduled visits per participant; the telemonitoring group had a mean of 5.24 (SD 3.62) versus 4.43 (SD 4.13) in usual care.

“The study showed a slightly higher rate of unscheduled visits in the home telemonitoring group (mean 5.24, SD 3.62) compared with the usual care group (mean 4.43, SD 4.13).”

• cites Deschildre 2012Not checked yet

  Not checked in the free run.

May combine two claims.

Some adult studies recruited people with severe or uncontrolled asthma, but most did not specify, and there was a mix within and across analyses.

“Some of the adult studies recruited people with severe or uncontrolled asthma (Deschildre 2012; Kokubu 1999; Kokubu 2000; Prabhakaran 2009; Ryan 2012), but most did not specify, and we found a mix both within and across analyses.”

• cites Deschildre 2012Not checked yet

  Not checked in the free run.

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

• cites Kokubu 2000Not checked yet

  Not checked in the free run.

May combine two claims.

Too few studies and too much variation in the interventions prevented meaningful conclusions about which technologies may offer the greatest benefit.

“With the exception of hospital admissions, adult and child studies showed similar findings, and too few studies with too much variation in the interventions used prevented any meaningful conclusions about which types of technology may offer the greatest benefit.”

May combine two claims.

Voorend-van Bergen 2015 reported lung function parameters as z-scores; absolute final scores could not be used because of a significant baseline imbalance between groups.

“Additionally, Voorend-van Bergen 2015 reported several lung function parameters as z-scores in the paper; we could not use the absolute final scores, as we observed a significant baseline imbalance between groups.”

• cites Voorend-van Bergen 2015Not checked yet

  Not checked in the free run.

May combine two claims.

van der Meer 2009 reported FEV1 as litres change from baseline; the telemonitoring group had a mean increase of 240 mL (SD 810) and the control group a mean decrease of 10 mL (SD 752).

“This study showed a 240 mL mean increase from baseline in the home telemonitoring group (SD 810 mL) and a 10 mL mean decrease in the control group (SD 752 mL).”

• cites van der Meer 2009Not checked yet

  Not checked in the free run.

The quote only partly states this claim.

For unscheduled healthcare visits, the pooled odds ratio comparing home telemonitoring with feedback to usual monitoring was 0.99 (95% CI 0.37 to 2.62).

“Total (95% CI) 231 199 100.0 % 0.99 [ 0.37, 2.62 ]”

The quote only partly states this claim.

For exacerbations requiring hospital admission, the pooled odds ratio in adults was 0.24 (95% CI 0.06 to 0.94).

“Subtotal (95% CI) 309 312 52.6 % 0.24 [ 0.06, 0.94 ]”

The quote only partly states this claim.

For exacerbations requiring oral corticosteroids, the pooled odds ratio comparing home telemonitoring with feedback to usual monitoring was 0.93 (95% CI 0.60 to 1.44).

“Total (95% CI) 233 233 100.0 % 0.93 [ 0.60, 1.44 ]”

The quote only partly states this claim.

For exacerbations requiring hospital admission, the test for subgroup differences between children and adults was significant (p = 0.04).

“Test for subgroup differences: Chi 2 = 4.13, df = 1 (P = 0.04), I 2 =76%”

Probably not a substantial claim: may not pass S3 (load-bearing).

For the outcome ACT > 19 (well controlled), the odds ratio in the Cingi 2015 study was 2.46 (95% CI 0.94 to 6.41).

“Cingi 2015 (1) 29/60 8/29 2.46 [ 0.94, 6.41 ]”

The quote only partly states this claim.

For asthma control measured by ACQ, the pooled mean difference comparing home telemonitoring with feedback to usual monitoring was -0.24 (95% CI -0.72 to 0.24).

“Total (95% CI) 240 238 100.0 % -0.24 [ -0.72, 0.24 ]”

The quote only partly states this claim.

For exacerbations requiring oral corticosteroids, the test for subgroup differences between types of technology was not significant (p = 1.00).

“Test for subgroup differences: Chi 2 = 0.00, df = 2 (P = 1.00),I 2 =0.0%”

The quote only partly states this claim.

For exacerbations requiring oral corticosteroids, the pooled odds ratio for smartphone app technology was 0.93 (95% CI 0.52 to 1.65).

“Subtotal (95% CI) 140 141 55.9 % 0.93 [ 0.52, 1.65 ]”

The quote only partly states this claim.

For exacerbations requiring hospital admission, the pooled odds ratio for SMS technology was 0.05 (95% CI 0.00 to 0.66).

“Subtotal (95% CI) 8 8 8.9 % 0.05 [ 0.00, 0.66 ]”

Probably not a substantial claim: may not pass S3 (load-bearing).

GRADE was used to rate the evidence for every outcome.

“We rated evidence for all outcomes using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation Working Group) approach.”

May combine two claims.

Prior asthma reviews found wide differences in telehealth delivery and comparators, which restricted their conclusions.

“existing reviews have noted a large degree of variation in the way telehealth is delivered in studies and to whom and with what it is compared (Jaana 2009; McLean 2010), and have been limited for this reason in the conclusions that could be drawn.”

• cites Jaana 2009Not checked yet

  Not checked in the free run.

• cites McLean 2010 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Confidence in the emergency department visit effect was low because the confidence intervals were wide and important heterogeneity was present within and across subgroups.

“We had low confidence in the effect because confidence intervals were too wide to show whether one strategy is likely to be better than another, and because we noted important heterogeneity both within and across subgroups.”

Classified as probably not a substantial claim. May combine two claims.

It is widely accepted that self-monitoring and routine professional checkups help track symptoms and support preventer inhaler adherence, particularly in people at high risk of severe attacks.

“The importance of self monitoring and regular checkups with a healthcare professional to monitor symptoms and encourage adherence to preventer inhalers is now well accepted (Gibson 2002; NRAD 2014), especially for those at high risk of severe asthma attacks.”

• cites Gibson 2002Not checked yet

  Not checked in the free run.

• cites NRAD 2014Not checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

An overview of reviews cautioned against assuming telehealth benefits and suggested the largest gains may occur in those at greatest risk of serious outcomes.

“A recent overview of systematic reviews suggested that these benefits should not be assumed, and that people at highest risk of serious health outcomes are likely to show the biggest gains (Mclean 2013).”

• cites Mclean 2013 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Confidence in the oral corticosteroid exacerbation estimate was low because only four studies contributed, suggesting possible publication bias, and the confidence intervals were too wide to determine whether one monitoring strategy is better.

“We had low confidence in the estimate because only four studies could be included in the analysis (Deschildre 2012; Donald 2008; Ryan 2012; Xu 2010), indicating possible publication bias.In addition, confidence intervals were too wide to reveal whether one monitoring strategy is likely to be better than another.”

• cites Deschildre 2012Not checked yet

  Not checked in the free run.

• cites Donald 2008Not checked yet

  Not checked in the free run.

• cites Ryan 2012Not checked yet

  Not checked in the free run.

May combine two claims.

It remains unproven that adding telemonitoring improves symptom control or lowers oral steroid requirements relative to usual asthma care.

“Studies have not yet proven that additional telemonitoring strategies lead to better symptom control or reduced need for oral steroids over usual asthma care”

May combine two claims.

Child and adult studies were pooled, and important heterogeneity was noted (I2 = 54%, P = 0.06).

“Child and adult studies were pooled, and im portant heterogeneity was noted (I 2 = 54%, P value = 0.06) (-1 inconsistency)”

May combine two claims.

High dropout put the two most influential trials at high risk of bias; Cingi 2015 lost 57% of controls versus 12% of intervention participants.

“Two studies carrying m ost of the weight were judged to be at high risk of bias owing to high dropout; in particular, Cingi 2015 had 57% dropout in the control group com pared with 12% in the intervention group (-1 risk of bias)”

May combine two claims.

The Cingi trial's effect ran opposite to the other two, creating substantial heterogeneity (I2 = 73%, P = 0.03).

“Cingi showed an ef f ect in the opposite direction to that noted in the other two studies, which introduced im portant heterogeneity (I 2 = 73%, P value = 0.03) (-1 inconsistency)”

Probably not a substantial claim: may not pass S3 (load-bearing).

Two reviewers separately judged each study's risk of bias with Cochrane Handbook criteria over seven domains.

“Two review authors (KMK and CJC) independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).”

• cites Higgins 2011Not checked yet

  Not checked in the free run.

May combine two claims.

The modest quality-of-life gains carry risk-of-bias concerns since the trials lacked blinding.

“Investigators noted small benefits for quality of life, but these are subject to risk of bias, as the studies were unblinded.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Screening and data extraction were done by two reviewers working separately, with consensus used to settle disagreements.

“Two review authors screened the search and independently extracted risk of bias and numerical data, resolving disagreements by consensus.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

A published protocol guided the review, and any departures from it were documented.

“We conducted the review according to the published protocol and reported deviations from it in the Differences between protocol and review section of the systematic review.”

May combine two claims.

Blinding domains were usually judged high risk, potentially influencing subjective rating scales.

“Studies were generally at high risk of bias f or the blinding dom ains, which m ay have af f ected results on subjective rating scales (-1 risk of bias)”

Probably not a substantial claim: may not pass S3 (load-bearing).

Broader reviews are difficult to compare with this review because they compile evidence for interventions assessed in different ways.

“These broader reviews are difficult to compare with our own, as they compile evidence for interventions that often need to be assessed in quite different ways.”

May combine two claims.

For oral steroid exacerbations, studies with high attrition and selection bias carried less than one-fifth of the overall weight, so this was not judged serious enough to downgrade.

“For one of the primary outcomesexacerbations requiring oral steroids -a couple of the contributing studies had high attrition and uncertainty with selection bias, but they carried less than a fifth of the overall weight, so we did not judge this as serious enough to warrant a downgrade.”

May combine two claims.

Jaana 2009 reviewed monitoring interventions more similar to this review's scope, but included people with other respiratory illnesses and focused on patient attitudes and receptiveness rather than effectiveness and safety.

“Jaana 2009 conducted a review of monitoring interventions more similar to what we set out to assess in this review; however, that review included people with other respiratory illnesses and outcomes focused on patient attitudes and receptiveness rather than effectiveness and safety.”

• cites Jaana 2009Not checked yet

  Not checked in the free run.

May combine two claims.

Most high risk of bias judgements involved blinding domains, which reduced confidence only for subjective outcomes such as asthma control and quality of life.

“When considering the effects of risk of bias on evidence quality, we noted that most of the high risk of bias judgements involved the blinding domains, which reduced our confidence only for subjective outcomes (asthma control and quality of life).”

May combine two claims.

High or unbalanced dropout affected only the lung function and unscheduled healthcare visits analyses, where high-risk studies carried substantial weight.

“We also noted some issues with high or unbalanced dropout in some studies, although we considered this to have affected only the lung function and unscheduled healthcare visits analyses, for which high-risk studies carried a lot of weight.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The review recorded deviations from the published protocol and explained why planned actions were not possible or meaningful.

“We recorded any deviations from the published protocol and explained why we believed it was not possible or meaningful to do what was planned.”

May combine two claims.

GRADE confidence in the evidence ranged from moderate to very low, and no analysis was rated high quality.

“Our confidence in the evidence based on GRADE (Grades of Recommendation, Assessment, Development and Evaluation Working Group) quality ratings ranged from moderate to very low, with none of the analyses thought to be of high quality.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Communication technologies for interventions with another primary focus, such as education, compliance, or inhaler technique, were not considered, and this separation may complicate decisions for decision makers.

“Moreover, communication technologies for interventions with another primary focus such as education, compliance or inhaler technique have not been considered, and this splitting may cause difficulty for decision makers in a field that is as varied and dynamic as the technologies themselves (Rada 2015).”

• cites Rada 2015Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Jaana 2009 emphasized variations in study approaches and an absence of robust study designs and formal evaluations, which is a common problem for syntheses in this rapidly evolving area.

“The review authors emphasized the”

• cites Jaana 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

A degree of subjectivity in applying study eligibility criteria was unavoidable.

“However, a degree of subjectivity in the application of study eligibility criteria was unavoidable.”

May combine two claims.

The small number of studies reporting prespecified important outcomes, especially the lack of explicit adverse event reporting, is a limitation.

“A related issue is the small number of studies reporting what we considered to be the most important outcomes (i.e.those prespecified in our review protocol), in particular, lack of explicit reporting of adverse events, which may be related to the frequency of these events expected in different populations.”

May combine two claims.

Inconsistency between study results was a problem in several analyses and could not be explained in most cases by planned subgroup analyses for age or technology type.

“Inconsistency between study results was a problem in several of the analyses (asthma control, exacerbations requiring emergency department (ED) visits and hospital admissions, quality of life and unscheduled healthcare visits) and could not be explained in most cases by planned subgroup analyses for age or type of technology.”

May combine two claims.

Zhao 2014 included six studies and found that asthma function scores were not improved by telemedicine interventions.

“Similarly, Zhao 2014, which included six studies, noted that asthma function scores were not improved by 'telemedicine' interventions.”

• cites Zhao 2014 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

The review did not downgrade for indirectness because inclusion criteria were implemented as planned, and the included populations, interventions, comparisons, and outcomes were considered to reflect the review question.

“We did not downgrade the quality of evidence for any outcome for its indirectness to the study question, because we were careful to implement the study inclusion criteria as planned.Therefore, we considered the populations, interventions, comparison groups and outcomes of the studies included in the review to reflect well the question we set out to answer.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

McLean 2010 reviewed all telehealth interventions for asthma regardless of components or comparators and found 21 studies.

“McLean 2010 looked at all 'telehealth' interventions for asthma regardless of their constituent parts or comparators used, and found 21 studies.”

• cites McLean 2010 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Imprecision was perhaps the most limiting factor in the evidence, due to the relatively small number of participants contributing to most analyses.

“Imprecision was perhaps the most limiting factor in this body of evidence, which made results difficult to interpret.This was due to the relatively small number of participants whose data contributed to most of the analyses.”

Probably not a substantial claim: may not pass S3 (load-bearing).

McLean 2010 concluded that telehealth interventions are unlikely to benefit patients with relatively mild asthma.

“McLean 2010 noted the clinical heterogeneity and concluded that telehealth interventions are not likely to be of benefit for patients with relatively mild asthma.”

• cites McLean 2010 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

It cannot be determined whether people with regular exacerbations would benefit most from extra monitoring or are more likely to be harmed by weakened responsibility for their own health.

“This has implications for services that seek to implement monitoring strategies such as these, because we cannot be sure whether people who have regular exacerbations would derive the greatest benefit from extra monitoring, or in fact whether they are more likely to be harmed by the weakened responsibility patients feel in terms of their own health.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The review contacted study authors to clarify interventions and methods and to request unpublished data to reduce publication bias, and translated non-English papers in duplicate.

“Once the list of included studies had been decided, we contacted study authors to clarify anything about the interventions and study methods that was uncertain, and to ask for additional unpublished data to reduce publication bias in the analyses.We translated non-English language papers in duplicate with a structured data extraction form based on the one used for all other studies.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The review sifted independently to reduce bias and collated additional references from trial registries and related works to make the included studies as complete as possible.

“When devising the original short list, we sifted independently to reduce bias, and we collated a large number of additional references from trial registries and related works to make the list of included studies as complete as possible.”

May combine two claims.

Publication bias was suspected only for the oral steroid exacerbation analysis, because fewer studies reported it than similar emergency department or hospital analyses, suggesting oral steroid courses might have been recorded and not reported.

“We suspected that publication bias might have affected only the exacerbations requiring oral steroids analysis, as the number of studies reporting this outcome was smaller than the number included in similar analyses of exacerbations requiring ED visits or hospital admissions, suggesting that oral steroid courses might have been recorded and not reported.”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of reporting bias is low.

“Selective reporting (reporting bias) Low risk All outcomes were reported but could not be included in the meta-analysis, as nonparametric tests were used (which was appropriate for the study data)”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Donald 2008, the risk of detection bias for all outcomes is low.

“Blinding of outcome assessment (detection bias) All outcomes Low risk”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Bateman 2000, the risk of other bias is low.

“Other bias Low risk None noted”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Cingi 2015, the risk of performance bias for subjective outcomes is high.

“Blinding of participants and personnel (performance bias) Subjective outcomes High risk”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Bateman 2000, the risk of attrition bias for all outcomes is unclear.

“Incomplete outcome data (attrition bias) All outcomes Unclear risk Dropout not reported”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Finkelstein 2005, the risk of performance bias for objective outcomes is low.

“Blinding of participants and personnel (performance bias) Objective outcomes Low risk It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Cingi 2015, the risk of selection bias from allocation concealment is low.

“Allocation concealment (selection bias) Low risk”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the same study, the risk of other bias was low, with none noted.

“None noted”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the same study, the risk of bias from incomplete outcome data was unclear, because dropout was not reported and only interim data for the first 50 participants were available.

“Dropout not reported, only interim data for first 50 participants recruited”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

In Cingi 2015, the risk of detection bias for all outcomes is high.

“Blinding of outcome assessment (detection bias) All outcomes High risk This trial was described as double-blind, but the main outcome was a rating scale completed by participants who, by the nature of the intervention, could have worked out if they were in the intervention or control group”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Guendelman 2002 trial, the risk of other bias was low, with none noted.

“None noted”

• cites Guendelman 2002Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

In Donald 2008, the risk of attrition bias for all outcomes is high.

“Incomplete outcome data (attrition bias) All outcomes High risk 71 participants were randomised; 44 replies were received at 6 months and 49 at 12 months.No description of how data were modified or imputed for those not contributing to the analyses”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Cingi 2015, the risk of reporting bias is low.

“Selective reporting (reporting bias) Low risk Named outcomes were reported, but some could not be included in meta-analysis with the other studies owing to the statistical methods used (z-scores and non-parametric tests)”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Donald 2008, the risk of performance bias for objective outcomes is low.

“Blinding of participants and personnel (performance bias) Objective outcomes Low risk It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Jan 2007 trial, the risk of other bias was low, with none noted.

“None noted”

• cites Jan 2007Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Donald 2008, the risk of other bias is low.

“Other bias Low risk None noted”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Jan 2007 trial, the risk of bias from selective reporting was low, because although satisfaction questionnaire data were not shown, this was not one of the review's named outcomes.

“Satisfaction questionnaires data not shown, but this was not one of the review's named outcomes”

• cites Jan 2007Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

In Cingi 2015, the risk of other bias is low.

“Other bias Low risk None noted”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of performance bias for subjective outcomes is high.

“Blinding of participants and personnel (performance bias) Subjective outcomes High risk Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

May combine two claims.

For the Jan 2007 trial, the risk of bias from incomplete outcome data was low, because of 164 randomised, 82/88 (93.2%) in the intervention arm and 71/76 (93.4%) in the control arm completed the study, representing relatively low and balanced attrition.

“Of 164 randomised, 82/88 in the intervention arm (93.2%) and 71/76 in the control arm (93.4%) completed the study, representing relatively low and balanced attrition”

• cites Jan 2007Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Jan 2007 trial, the risk of bias from blinding of participants and personnel for subjective outcomes was high, because subjective measures like quality of life and symptom scales may have been subject to performance bias.

“Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

• cites Jan 2007Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Jan 2007 trial, the risk of bias for allocation concealment was low because a sealed envelope was used.

“Sealed envelope”

• cites Jan 2007Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

In Bateman 2000, the risk of selection bias from allocation concealment is unclear.

“Allocation concealment (selection bias) Unclear risk No information regarding allocation concealment”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Donald 2008, the risk of selection bias from random sequence generation is unclear.

“Random sequence generation (selection bias) Unclear risk Described as randomised, but method used to generate the random sequence not described”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Finkelstein 2005, the risk of selection bias from random sequence generation is unclear.

“Random sequence generation (selection bias) Unclear risk Described as randomised, but method used to generate the random sequence not described”

May combine two claims.

For the Jan 2007 trial, the risk of bias from blinding of participants and personnel for objective outcomes was low, because although allocation could not be hidden, this is unlikely to bias objective outcomes such as exacerbation counts.

“It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

• cites Jan 2007Not checked yet

  Not checked in the free run.

May combine two claims.

For the Kokubu 1999 trial, the risk of bias from blinding of participants and personnel for objective outcomes was low, because although allocation could not be hidden, this is unlikely to bias objective outcomes such as exacerbation counts.

“It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

In Bateman 2000, the risk of selection bias from random sequence generation is unclear.

“Random sequence generation (selection bias) Unclear risk Described as randomised, but method used to generate the random sequence not described”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of attrition bias for all outcomes is high.

“Incomplete outcome data (attrition bias) All outcomes High risk Attrition was much higher in the telemonitoring group (40%) than in the usual care group (20%), and only those dropping out after 120 days were included in the final analysis, representing 88% of the randomised sample”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of other bias is low.

“Other bias Low risk None noted”

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Kokubu 1999 trial, the risk of bias for random sequence generation was unclear because the study was described as randomised but details of how the sequence was generated were not described.

“Described as randomised, but details of how the sequence was generated not described”

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Donald 2008, the risk of performance bias for subjective outcomes is high.

“Blinding of participants and personnel (performance bias) Subjective outcomes High risk Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

May combine two claims.

For the same study, the risk of bias from selective reporting was high, because only an abstract was available with minimal methodological information, two conference abstracts from 2005 reported data for the first 50 participants to complete 4-month follow-up, and no data were available for the full 240 enrolled or for the full 12 months.

“Only an abstract was available with minimal information about methods.Two conference abstracts from 2005 report data for the first 50 participants to complete 4month follow-up.No data are available for the full population of 240 enrolled in the study, and none for the full 12 months of the trial”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of detection bias for all outcomes is high.

“Blinding of outcome assessment (detection bias) All outcomes High risk Investigators were not blind and appear to be those taking measurements”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Cingi 2015, the risk of performance bias for objective outcomes is low.

“Blinding of participants and personnel (performance bias) Objective outcomes Low risk”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Donald 2008, the risk of selection bias from allocation concealment is unclear.

“Allocation concealment (selection bias) Unclear risk No information regarding allocation concealment”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Cingi 2015, the risk of selection bias from random sequence generation is low.

“Random sequence generation (selection bias) Low risk”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Cingi 2015, the risk of attrition bias for all outcomes is high.

“Incomplete outcome data (attrition bias) All outcomes High risk”

May combine two claims.

For the Guendelman 2002 trial, the risk of bias from incomplete outcome data was low, because baseline characteristics of completers and non-completers did not differ, and completion rates were 62/66 in the intervention group and 60/68 in the control group.

“Baseline characteristics of children who did and did not complete the trial did not differ.62/66 participants in the intervention group and 60/68 participants randomised to the control group completed 12 weeks.”

• cites Guendelman 2002Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

In Deschildre 2012, the risk of selection bias from random sequence generation is unclear.

“Random sequence generation (selection bias) Unclear risk Random allocation to HM or CT group was performed at inclusion, resulting in groups of 6 participants at each investigation centre”

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Guendelman 2002 trial, the risk of bias for allocation concealment was low because the nurse opened a sealed envelope containing the treatment assignment after the baseline interview.

“Following baseline interview the nurse opened a sealed envelope containing the treatment assignment”

• cites Guendelman 2002Not checked yet

  Not checked in the free run.

May combine two claims.

For the Guendelman 2002 trial, the risk of bias from selective reporting was low, because all primary and secondary outcomes were reported.

“All primary and secondary outcomes are reported”

• cites Guendelman 2002Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Deschildre 2012, the risk of selection bias from allocation concealment is unclear.

“Allocation concealment (selection bias) Unclear risk No information regarding allocation concealment”

May combine two claims.

For the same study, the risk of bias from blinding of outcome assessment was high, because although blinding would have been possible, no information indicates it was done.

“It would have been possible to blind outcome assessors, but no information suggests this was done”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Guendelman 2002 trial, the risk of bias from blinding of participants and personnel for objective outcomes was low, because although allocation could not be hidden, this is unlikely to bias objective outcomes such as exacerbation counts.

“It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

• cites Guendelman 2002Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

In Donald 2008, the risk of reporting bias is low.

“Selective reporting (reporting bias) Low risk Outcomes named in the methods were well reported in the results section, although this could not be verified with a protocol”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Finkelstein 2005, the risk of selection bias from allocation concealment is unclear.

“Allocation concealment (selection bias) Unclear risk No information regarding allocation concealment”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the study whose risk of bias table appears first in this fragment, the risk of bias from blinding of participants and personnel for subjective outcomes was high, because subjective measures such as quality of life and symptom scales may have been influenced by performance bias.

“Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

Probably not a substantial claim: may not pass S3 (load-bearing).

In Deschildre 2012, the risk of performance bias for objective outcomes is low.

“Blinding of participants and personnel (performance bias) Objective outcomes Low risk It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

In Finkelstein 2005, the risk of performance bias for subjective outcomes is high.

“Blinding of participants and personnel (performance bias) Subjective outcomes High risk Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

For the Prabhakaran 2009 trial, the risk of other bias was low, with none noted.

“None noted”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

May combine two claims.

For the Liu 2011 trial, the risk of bias from incomplete outcome data was high because dropout was quite high and balanced between groups (28% and 23%), and outcomes were reported for the 43 and 46 participants completing the 6-month follow-up, not for the 60 and 60 randomised.

“Dropout was quite high and balanced between groups (28% and 23%).Outcomes are reported for the 43 and 46 participants completing the 6-month follow-up, not for the 60 and 60 randomised”

• cites Liu 2011 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

For the Prabhakaran 2009 trial, the risk of bias from blinding of outcome assessment was high because although blinding would have been possible, no information suggests it was done.

“It would have been possible to blind outcome assessors, but no information suggests this was done”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Kokubu 2000 trial, the risk of bias for random sequence generation was unclear because the study was described as randomised but details of how the sequence was generated were not described.

“Described as randomised, but details of how the sequence was generated not described”

• cites Kokubu 2000Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Kokubu 2000 trial, the risk of other bias was low, with none noted.

“None noted”

• cites Kokubu 2000Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Ostojic 2005 trial, the risk of bias for allocation concealment was unclear because no details about allocation concealment were provided.

“No details about allocation concealment”

• cites Ostojic 2005 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Prabhakaran 2009 trial, the risk of bias from blinding of participants and personnel for objective outcomes was low, because although allocation could not be hidden, this is unlikely to bias objective outcomes such as exacerbation counts.

“It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

For the Liu 2011 trial, the risk of bias from blinding of participants and personnel for subjective outcomes was high, because subjective measures like quality of life and symptom scales may have been subject to performance bias.

“Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

• cites Liu 2011 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Liu 2011 trial, the risk of bias from blinding of participants and personnel for objective outcomes was low, because although allocation could not be hidden, this is unlikely to bias objective outcomes such as exacerbation counts.

“It would not have been possible to hide allocation from participants and personnel, but it is unlikely that this would have introduced bias for objective outcomes (e.g.number of people having exacerbations)”

• cites Liu 2011 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Prabhakaran 2009 trial, the risk of bias for allocation concealment was unclear because it was unclear whether it was possible to predict allocation from the slips used.

“Unclear whether it was possible to predict allocation from the slips used”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

For the Prabhakaran 2009 trial, the risk of bias from blinding of participants and personnel for subjective outcomes was high, because subjective measures like quality of life and symptom scales may have been subject to performance bias.

“Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Prabhakaran 2009 trial, the risk of bias from selective reporting was low because although no study protocol was found, named outcomes were well reported.

“No study protocol found but named outcomes well reported”

• cites Prabhakaran 2009Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Kokubu 1999 trial, the risk of other bias was low, with no other bias noted.

“No other bias noted”

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Kokubu 1999 trial, the risk of bias from incomplete outcome data was unclear because no information was available.

“No information available”

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

May combine two claims.

For the Liu 2011 trial, the risk of bias from selective reporting was low because although there was no mention of trial registration, outcomes stated in the methods were well reported.

“No mention of trial registration.Outcomes stated in the methods were well reported”

• cites Liu 2011 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Liu 2011 trial, the risk of bias for random sequence generation was unclear because the study was described as randomised but the method used to generate the random sequence was not described.

“Described as randomised, but method used to generate the random sequence not described”

• cites Liu 2011 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Kokubu 1999 trial, the risk of bias from selective reporting was unclear because it could not be determined from the translation.

“Could not be determined from the translation”

• cites Kokubu 1999Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

For the Ostojic 2005 trial, the risk of other bias was low, with none noted.

“None noted”

• cites Ostojic 2005 sourceNot checked yet

  Not checked in the free run.

May combine two claims.

For the Kokubu 2000 trial, the risk of bias from blinding of participants and personnel for subjective outcomes was high, because subjective measures like quality of life and symptom scales may have been subject to performance bias.

“Subjective outcomes such as quality of life and symptom scales filled in by participants or personnel may have been subject to performance bias”

• cites Kokubu 2000Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

For the Ostojic 2005 trial, the risk of bias for random sequence generation was low because patients were randomised by computer.

“Patients were randomised by computer”

• cites Ostojic 2005 sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For the Ostojic 2005 trial, the risk of bias from incomplete outcome data was low because there was no evidence of incomplete outcome data and no study withdrawals.

“No evidence of incomplete outcome data; no study withdrawals”

• cites Ostojic 2005 sourceNot checked yet

  Not checked in the free run.

May combine two claims.