Means restriction for the prevention of suicide by jumping

May combine two claims.

In the UK, after domestic gas supplies changed from toxic coal gas to non-toxic natural gas, the numbers of carbon monoxide poisoning suicides fell, and despite a slight increase in other methods, the net effect was a large reduction in suicide deaths.

“With the change to natural gas, the numbers of these deaths fell.Despite a slight increase in other methods, the net effect was a large reduction in suicide deaths (Kreitman 1976).”

• cites Kreitman 1976 sourceSupported

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

Study withdrawal was not reported in any of the included studies.

“Study withdrawal was not reported in any of the included studies.”

May combine two claims.

While the results of individual studies were all in the same direction of effect, there was considerable heterogeneity between studies as evidenced by I 2 of 88.4%.

“While the results of individual studies were all in the same direction of effect, there was considerable heterogeneity between studies as evidenced by I 2 of 88.4%.”

May combine two claims.

Only one study assessing the effectiveness of restricting road access was eligible for inclusion (Skegg 2009), therefore we did not undertake a meta-analysis. This study reported a reduction in the number of suicides at the site of road closure (incident rate difference 1.3 deaths per year, 95% CI 0.60 to 2.00).

“Only one study assessing the effectiveness of restricting road access was eligible for inclusion (Skegg 2009), therefore we did not undertake a meta-analysis.This study reported a reduction in the number of suicides at the site of road closure (incident rate difference 1.3 deaths per year, 95% CI 0.60 to 2.00).”

May combine two claims.

Pre-intervention periods ranged from four years to 22 years (median 10 years), while the postintervention periods ranged from 0.4 years to 22 years (median 5 years).

“Pre-intervention periods ranged from four years to 22 years (median 10 years), while the postintervention periods ranged from 0.4 years to 22 years (median 5 years).”

May combine two claims.

A meta-analysis of two studies found a reduction in the number of suicides at sites where safety nets were installed, but this effect was not statistically significant (IRR = 0.09, 95% CI 0.01 to 1.30; P = 0.07).

“A meta-analysis of two studies found a reduction in the number of suicides at sites where safety nets were installed, but this effect was not statistically significant (IRR = 0.09, 95% CI 0.01 to 1.30; P = 0.07) (Figure 6).”

May combine two claims.

There was minimal heterogeneity between studies (I 2 = 29.3%) but this comparison included only two studies.

“There was minimal heterogeneity between studies (I 2 = 29.3%)but this comparison included only two studies.”

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

Twelve studies were included in the meta-analysis overall.

“We included 12 studies in the meta-analysis overall.”

May combine two claims.

In Bennewith 2007, in the three years prior to installation of bridge barriers, there were 25 fatal and 92 non-fatal incidents, while in the seven years after installation there were 24 fatal and 280 non-fatal incidents at the bridge.

“In the three years prior to installation of bridge barriers, there were 25 fatal and 92 non-fatal incidents, while in the seven years a er installation there were 24 fatal and 280 non-fatal incidents at the bridge.”

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

The total pre-and postintervention periods were 134.5 years and 92.4 years, respectively.

“The total pre-and postintervention periods were 134.5 years and 92.4 years, respectively (Table 1).”

May combine two claims.

In this study, the intervention was not primarily aimed at suicide prevention, but rather a coincidence of the need to restrict accessibility to the site due to construction work.

“In this study, the intervention was not primarily aimed at suicide prevention, but rather a coincidence of the need to restrict accessibility to the site due to construction work.”

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

It was proposed that the barrier would generate considerable cost savings because other suicide prevention activities might not be required.

“Furthermore, it was proposed that the barrier would generate considerable costsavings because other suicide prevention activities -such as suicide surveillance, negotiations, and recovery attempts by bridge patrols and other relevant authorities -might not be required.”

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

Thirteen of the included studies reported number of suicides per year in both the pre-and postintervention periods.

“Thirteen of the included studies reported number of suicides per year in both the pre-and postintervention periods.”

The quote only partly states this claim.

Only one study reported cost-effectiveness data (Whitmer 2013).

“Only one study reported cost-effectiveness data.”

• Also stated in: § Cost-effectiveness of interventions · plain

  “This study evaluated the cost-effectiveness of a proposed barrier on a bridge (Whitmer 2013).”

May combine two claims.

Bennewith 2007 reported that during the 10-year study observation period, there was a total of 421 'fatal' and 'non-fatal' incidents. There were 117 incidents (39 per year) recorded in the three years prior to installation of bridge barriers and 304 incidents (43 per year) in the seven years after installation.

“The study reported that during the 10year study observation period, there was a total of 421 'fatal' and 'non-fatal' incidents where someone jumped or appeared to be at risk of jumping from the bridge of interest.”

• Also stated in: § f. Safety nets · plain

  “There were 117 incidents (39 per year) recorded in the three years prior to installation of bridge barriers and 304 incidents (43 per year) in the seven years a er installation.”

May combine two claims.

Data on suicide attempts were limited and none of the studies used self-harm as an outcome.

“Data on suicide attempts were limited and none of the studies used self-harm as an outcome.”

The quote only partly states this claim.

There was considerable heterogeneity between studies (I 2 = 84.1%) for physical barriers.

“There was considerable heterogeneity between studies (I 2 = 84.1%).”

May combine two claims.

The cost-effectiveness study suggested that approximately 286 lives would be saved over a 20-year period at an average cost per life of approximately USD 180,419.

“This study suggested that approximately 286 lives would be saved over a 20-year period at an average cost per life of approximately USD 180,419.”

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

Of the 14 included studies, 13 were before-and-after studies.

“Of the 14 included studies, 13 were before-and-a er studies.”

May combine two claims.

In the 13 included observational studies, a total of 742.3 suicides (5.5 suicides per year) occurred during the pre-intervention period (134.5 study years), while 70.6 suicides (0.8 suicides per year) occurred during the postintervention period (92.4 study years) -a 91% reduction in suicides.

“In the 13 included observational studies, a total of 742.3 suicides (5.5 suicides per year) occurred during the pre-intervention period (134.5 study years), while 70.6 suicides (0.8 suicides per year) occurred during the postintervention period (92.4 study years) -a 91% reduction in suicides (Table 1).”

May combine two claims.

Skegg 2009 found that in the 10-year period before access to the headland was restricted, there were two suicide attempts at the site, and in the two years after restriction of access, there were no attempts.

“In the 10-year period before access to the headland was restricted, there were two suicide attempts at the site.In the two years a er restriction of access, there were no attempts.”

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

Given the study design of all the included studies, study withdrawal rates were not reported in any of them.

“Given the study design of all the included studies, study withdrawal rates were not reported in any of them.”

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

Of the 4 safety net interventions in Hemmer 2017, 2 were complete measures.

“Of the 4 safety net interventions, 2 were complete measures.”

May combine two claims.

Safety nets showed an effect in favour of the intervention; however, this effect was not significant.

“Safety nets showed an effect in favour of the intervention; however, this effect was not significant.”

May combine two claims.

All four reviews found relatively strong evidence of the effectiveness of means restriction in reducing suicide numbers, albeit that these studies were observational in nature so do not strictly meet the level in the hierarchy of evidence of RCTs.

“All four reviews found relatively strong evidence of the effectiveness of means restriction in reducing suicide numbers, albeit that these studies were observational in nature so do not strictly meet the level in the hierarchy of evidence of RCTs.”

Classified as probably not a substantial claim.

Data on suicide method substitution was limited in this review.

“Data on suicide method substitution was limited in this review.”

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

Cost-effectiveness of interventions was not assessed in any of the aforementioned reviews.

“Cost-effectiveness of interventions was not assessed in any of the aforementioned reviews.”

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

Of the 11 barrier interventions in Hemmer 2017, 5 were complete measures, i.e. secured the entire jump area.

“Of the 11 barrier interventions, 5 were complete measures, i.e. secured the entire jump area.”

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

Bennewith 2007 reported 'fatal' and 'non-fatal' events at a bridge. The non-fatal events were a combination of suicide attempts and incidents of people who were thought to be about to make attempts. The study findings suggested an increase in non-fatal incidents after the installation of bridge barriers. The three-year pre-installation incidents were 92, and the seven-year postinstallation were 92

“Bennewith 2007 reported 'fatal' and 'non-fatal' events at a bridge.The non-fatal events were a combination of suicide attempts and incidents of people who were thought to be about to make attempts.The study findings suggested an increase in non-fatal incidents a er the installation of bridge barriers.The three-year pre-installation incidents were 92, and the seven-year postinstallation incidents”

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

All 13 included observational studies reported data on the number of suicides pre-and postintervention.

“All 13 included observational studies reported data on the number of suicides pre-and postintervention.”

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

However, we identified three systematic reviews (all conducted by members of the same team) that assessed interventions to reduce suicides at high-frequency locations -including jumping sites (Cox 2013;Pirkis 2013;Pirkis 2015), one evidence review by the National Institute for Health and Care Excellence (NICE) on reducing access to means (NICE 2018), and two narrative reviews on suicide by jumping

“However, we identified three systematic reviews (all conducted by members of the same team) that assessed interventions to reduce suicides at high-frequency locations -including jumping sites (Cox 2013;Pirkis 2013;Pirkis 2015), one evidence review by the National Institute for Health and Care Excellence (NICE) on reducing access to means (NICE 2018), and two narrative reviews on suicide by jumping”

May combine two claims.

Similar to the NICE evidence review, our review identified a study that made an assessment of the extent of structural interventions (Hemmer 2017). Barriers and safety nets that secured the entire jump area (complete measures) were found to be more effective at preventing suicides than structural interventions that secured only a part of the jump area (incomplete measures).

“Similar to the NICE evidence review, our review identified a study that made an assessment of the extent of structural interventions (Hemmer 2017).Barriers and safety nets that secured the entire jump area (complete measures) were found to be more effective at preventing suicides than structural interventions that secured only a part of the jump area (incomplete measures).”

May combine two claims.

Some of the included studies included displacement to other jumping sites as an outcome, with the majority showing minimal displacement.

“Some of our included studies included displacement to other jumping sites as an outcome -with the majority showing minimal displacement.”

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

The systematic and evidence reviews differed from ours in three main ways: (a) the means of suicide were not restricted to jumping from a height, but also included other means such as carbon monoxide poisoning, collision with a train, and use of a firearm; (b) the interventions were not restricted to means restriction only but also included other interventions, such as those solely aimed at

“The systematic and evidence reviews differed from ours in three main ways: (a) the means of suicide were not restricted to jumping from a height, but also included other means such as carbon monoxide poisoning, collision with a train, and use of a firearm; (b) the interventions were not restricted to means restriction only but also included other interventions, such as those solely aimed at”

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

Similar to our review, there was evidence of heterogeneity across studies for the primary outcome of suicide or attempted suicide in both Pirkis' reviews and the NICE evidence review.

“Similar to our review, there was evidence of heterogeneity across studies for the primary outcome of suicide or attempted suicide in both Pirkis' reviews and the NICE evidence review.”

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

Of the 12 eligible included studies, most evaluated interventions at bridges or bridge-like structures.

“Of the 12 eligible included studies (two each from Switzerland, USA, Canada, New Zealand and Australia, and one from the UK), most evaluated interventions at bridges or bridge-like structures.”

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

We considered investigating the effects of geographical variation and structure type. However, with the available data we had, we did not believe these subgroup analyses would be useful.

“We considered investigating the effects of geographical variation and structure type.However, with the available data we had, we did not believe these subgroup analyses would be useful.”

May combine two claims.

We attempted to explore the heterogeneity by conducting a sensitivity analysis. This analysis did not substantially alter the direction of effect or the precision of the effect estimates, however heterogeneity still persisted.

“We attempted to explore the heterogeneity by conducting a sensitivity analysis.This analysis did not substantially alter the direction of effect or the precision of the effect estimates, however heterogeneity still persisted.”

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

One study compared the effectiveness of different suicide prevention interventions at various frequently-used jumping locations.

“One study compared the effectiveness of different suicide prevention interventions at various frequently-used jumping locations.”

May combine two claims.

This study suggested that approximately 286 lives would be saved over a 20-year period at an average cost per life of approximately USD 180,419.

“This study suggested that approximately 286 lives would be saved over a 20-year period at an average cost per life of approximately USD 180,419.”

May combine two claims.

Planned subgroup analyses were not feasible. Most studies included in this review evaluated interventions aimed at the general population, and did not specify comorbidities or previous history of suicidal behaviour of individuals.

“Planned subgroup analyses were not feasible.Most studies included in this review evaluated interventions aimed at the general population, and did not specify comorbidities or previous history of suicidal behaviour of individuals.”

May combine two claims.

Analyses performed both with (7 studies; IRR = 0.07, 95% CI 0.02 to 0.24; P < 0.001) and without this study (6 studies; IRR = 0.05, 95% CI 0.01 to 0.19; P < 0.001) did not substantially alter the direction of effect or the precision of the effect estimates.

“Analyses performed both with (7 studies; IRR = 0.07, 95% CI 0.02 to 0.24; P < 0.001) and without this study (6 studies; IRR = 0.05, 95% CI 0.01 to 0.19; P < 0.001) did not substantially alter the direction of effect or the precision of the effect estimates.”

May combine two claims.

Ten studies assessed jumping means restriction interventions delivered in isolation: barriers (n = 6), blocking off road access to headlands/ocean cliffs (n = 2), safety net (n = 1) and a guard rail (n = 1).

“Ten studies assessed jumping means restriction interventions delivered in isolation: barriers (n = 6), blocking off road access to headlands/ ocean cliffs (n = 2), safety net (n = 1) and a guard rail (n = 1).”

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

One of the strengths of this review is that, in addition to establishing the effectiveness of means restriction interventions, we also sought to investigate their cost-effectiveness. However, we found only one economic study that matched our inclusion criteria. Further economic studies are required.

“One of the strengths of this review is that, in addition to establishing the effectiveness of means restriction interventions, we also sought to investigate their cost-effectiveness.However, we found only one economic study that matched our inclusion criteria.Further economic studies are required.”

May combine two claims.

The search identified 13 observational studies evaluating the effectiveness of various means restriction interventions for the prevention of suicide by jumping and one study evaluating the cost-effectiveness of a proposed barrier on a bridge.

“The search identified 13 observational studies evaluating the effectiveness of various means restriction interventions for the prevention of suicide by jumping and one study evaluating the cost-effectiveness of a proposed barrier on a bridge.”

May combine two claims.

Skegg 2009 found that no suicide attempts were made in the period following restriction of road access to a headland, compared to two attempts that were made in the period before restriction of access. However, the pre-and postobservation periods were 10 and two years, respectively.

“Skegg 2009 found that no suicide attempts were made in the period following restriction of road access to a headland, compared to two attempts that were made in the period before restriction of access.However, the pre -and postobservation periods were 10 and two years, respectively.”

May combine two claims.

In both cases of road closure, the intervention was not primarily aimed at suicide prevention, but rather a coincidence of the need to restrict accessibility to the sites due to construction work and a Foot and Mouth outbreak, respectively.

“In both cases of road closure, the intervention was not primarily aimed at suicide prevention, but rather a coincidence of the need to restrict accessibility to the sites due to construction work and a Foot and Mouth outbreak, respectively.”

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

Three studies included details on the height of barriers/fences.

“Three studies included details on the height of barriers/fences.”

May combine two claims.

Jumping from a height is an uncommon but lethal means of suicide.

“Jumping from a height is an uncommon but lethal means of suicide.”

• Also stated in: § Means of suicide · plain

  “Jumping from a height, though an uncommon means of suicide, can be lethal (Elnour 2008).”

• cites Elnour 2008 sourceNot checked — no full text of the source

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

Three review authors independently assessed risk of bias for each included study.

“Three review authors (CO, UK, SW) independently assessed risk of bias for each included study.”

May combine two claims.

In these studies, publicity (of the intervention/jump site) by media reports could have had a potentially confounding effect, and this was not adjusted for in the analyses.

“In these studies, publicity (of the intervention/jump site) by media reports could have had a potentially confounding effect.This was not adjusted for in the analyses.”

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

We did not have access to study protocols for the studies included in this review, and as such it is difficult to assess this domain. We therefore gave a rating of 'unclear risk' for this domain to the 13 included observational studies.

“We did not have access to study protocols for the studies included in this review, and as such it is difficult to assess this domain.We therefore gave a rating of 'unclear risk' for this domain to the 13 included observational studies.”

May combine two claims.

Due to the nature of this topic, none of the included studies employed a randomised controlled design as these would be difficult to implement.

“Due to the nature of this topic, none of the included studies employed a randomised controlled design as these would be difficult to implement.”

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

For the cost-effectiveness analysis study, we made a judgement of 'yes' for study design; form of evaluation; benefit measurement and valuation; costing; adjustments for timing of costs and benefits; and presentation of results.

“For the evaluation of the cost-effectiveness analysis study (Table 15), we made a judgement of 'yes' for the following domains: study design; form of evaluation; benefit measurement and valuation; costing; adjustments for timing of costs and benefits; and presentation of results.”

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

Two review authors independently assessed all citations from the searches and identified relevant titles and abstracts.

“Two review authors independently assessed all citations from the searches and identified relevant titles and abstracts.”

May combine two claims.

We judged the remaining study to be at moderate risk of bias because electronic death records used in the study between 1960 and 1974 did not meet current data quality standards (Pelletier 2007), thereby leading to a potential underestimation of suicides during this time period.

“We judged the remaining study to be at moderate risk of bias because electronic death records used in the study between 1960 and 1974 did not meet current data quality standards (Pelletier 2007), thereby leading to a potential underestimation of suicides during this time period.”

May combine two claims.

We judged all 13 included observational studies to be at low risk of bias in this domain as there were no deviations from the intended interventions.

“We judged all 13 included observational studies to be at low risk of bias in this domain as there were no deviations from the intended interventions.”

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

Small study effects and potential publication bias were planned to be assessed using a funnel plot if a meta-analysis included results of at least 10 studies.

“We planned to assess small study effects and potential publication bias using a funnel plot if a meta-analysis included results of at least 10 studies.”

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

We made a judgment of 'not clear' for effectiveness data, modelling, and allowance for uncertainty.

“We made a judgment of 'not clear' for effectiveness data, modelling, and allowance for uncertainty.”

May combine two claims.

Evidence from research on near lethal suicide attempts supports the idea that, at least for a proportion of people, these acts may be an impulsive response and probably would not have occurred if the means had not been readily available.

“Evidence from research on near lethal suicide attempts supports the idea that, at least for a proportion of people, these acts may be an impulsive response and probably would not have occurred if the means had not been readily available (Hawton 2005).”

• cites Hawton 2005 sourceNot checked — no full text of the source

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

The quality of the evidence included in this review was summarised using the GRADE approach.

“We summarised the quality of the evidence included in this review using the GRADE approach.”

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

Imputation of missing data was not conducted because of the potential for introduction of bias.

“We did not conduct imputation of missing data because of the potential for introduction of bias.”

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

The only domain with a judgement of 'no' was selection of alternatives, as this was not described.

“The only domain with a judgement of 'no' was selection of alternatives, as this was not described.”

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

We conducted sensitivity analyses for the primary outcomes of suicide and attempted suicide. Where appropriate, studies that were methodologically different from other studies and those contributing to the marked heterogeneity in the analyses for the primary outcome were removed.

“We conducted sensitivity analyses for the primary outcomes of suicide and attempted suicide.Where appropriate, studies that were methodologically different from other studies and those contributing to the marked heterogeneity in the analyses for the primary outcome were removed.”

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

Lockley 2014 is at low risk of bias in classification of interventions because intervention status is well defined.

“Intervention status is well defined”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

In agreement with Pirkis, we believe that such heterogeneity is expected in this type of research due to the inherent differences within the studies, the jump sites, the types of interventions, and the differing demographics involved.

“In agreement with Pirkis, we believe that such heterogeneity is expected in this type of research due to the inherent differences within the studies, the jump sites, the types of interventions, and the differing demographics involved.”

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

This is in keeping with the argument regarding the inevitability of statistical heterogeneity in meta-analyses as a consequence of clinical and methodological diversity (Higgins 2011).

“This is in keeping with the argument regarding the inevitability of statistical heterogeneity in meta-analyses as a consequence of clinical and methodological diversity (Higgins 2011).”

• cites Higgins 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

Hemmer 2017 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

Hemmer 2017 is at unclear risk of bias in selection of reported result because no study protocols were accessible.

“No access to study protocols for this study therefore it is difficult to assess this domain”

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

The electronic literature search did not include grey literature searching. It is possible that additional eligible studies may have been missed.

“The electronic literature search did not include grey literature searching.It is possible that additional eligible studies may have been missed.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Whilst possible, it is unlikely that we missed any relevant studies.

“Whilst possible, it is unlikely that we missed any relevant studies.”

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

In the course of this review, we conducted an extensive literature search using a wide range of search terms and databases, searched the reference lists of included studies and relevant systematic reviews, and contacted subject experts for information on unpublished or ongoing studies.

“In the course of this review, we conducted an extensive literature search using a wide range of search terms and databases, searched the reference lists of included studies and relevant systematic reviews, and contacted subject experts for information on unpublished or ongoing studies.”

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

Lockley 2014 is at unclear risk of bias in selection of reported result because no study protocols were accessible.

“No access to study protocols for this study therefore it is difficult to assess this domain”

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

Since we did not judge any studies to be at high risk of bias across one or more domains, we did not feel a sensitivity analysis was appropriate for this reason.

“Since we did not judge any studies to be at high risk of bias across one or more domains, we did not feel a sensitivity analysis was appropriate for this reason.”

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

None of the aforementioned reviews assessed attempted suicides at frequently-used jumping sites. Exclusion of this outcome in a study or review could underestimate the effect of an intervention and would therefore not provide a complete picture of the scale of the problem or prevention efforts.

“None of the aforementioned reviews assessed attempted suicides at frequently-used jumping sites.Exclusion of this outcome in a study or review could underestimate the effect of an intervention and would therefore not provide a complete picture of the scale of the problem or prevention efforts.”

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

Lockley 2014 is at moderate risk of bias due to confounding because the study did not account for potential confounding from media publicity that could have promoted the jump site.

“The study did not take into account the possibility of potential confounding effects as a result of media publicity which could have promoted the jump site”

• AssumptionRests on an assumption: Media publicity could have promoted the jump site, and thus could confound studies of means restriction at jump sites.

May combine two claims.

The findings from the meta-analyses assessing the effectiveness of all jumping means restrictions interventions; means restriction interventions delivered in isolation; means restriction interventions delivered in combination with other interventions; and barriers, showed significant reductions in the number of suicides at sites where the interventions were installed.

“The findings from the meta-analyses assessing the effectiveness of all jumping means restrictions interventions; means restriction interventions delivered in isolation; means restriction interventions delivered in combination with other interventions; and barriers, showed significant reductions in the number of suicides at sites where the interventions were installed.”

Classified as probably not a substantial claim.

It was not feasible to undertake any of the planned subgroup analyses in this review.

“It was not feasible to undertake any of our planned subgroup analyses in this review.”

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

Law 2014 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

We conducted a sensitivity analysis for the analysis of studies assessing the effectiveness of barriers. We removed one study from this analysis because we judged it to be methodologically different from the other studies (Hemmer 2017).

“We therefore conducted a sensitivity analysis for the analysis of studies assessing the effectiveness of barriers (Figure 7).We removed one study from this analysis because we judged it to be methodologically different from the other studies (Hemmer 2017).”

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

We did not conduct sensitivity analyses for interventions combining multiple intervention types due to variations in these interventions.

“We did not conduct sensitivity analyses for interventions combining multiple intervention types due to variations in these interventions.”

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

Law 2014 is at moderate risk of bias due to confounding because the study did not account for potential confounding from media publicity that could have promoted the jump site.

“The study did not take into account the possibility of potential confounding effects as a result of media publicity which could have promoted the jump site”

• AssumptionRests on an assumption: Media publicity could have promoted the jump site, and thus could confound studies of means restriction at jump sites.

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

Furthermore, data on suicide method substitution was limited in all the reviews. The paucity of literature here means that we have to rely on evidence (which needs further exploration) that prevention in the acute suicidal episode reduces longer-term suicide risk.

“Furthermore, data on suicide method substitution was limited in all the reviews.The paucity of literature here means that we have to rely on evidence (which needs further exploration) that prevention in the acute suicidal episode reduces longer-term suicide risk.”

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

Lockley 2014 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

Hemmer 2017 is at low risk of bias in classification of interventions because intervention status is well defined.

“Intervention status is well defined”

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

Law 2014 is at unclear risk of bias in selection of reported result because no study protocols were accessible.

“No access to study protocols for this study therefore it is difficult to assess this domain”

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

Where applicable, we contacted study authors for missing outcome data. At least two review authors were involved in study selection, data extraction and quality assessment of included studies.

“Where applicable, we contacted study authors for missing outcome data.At least two review authors were involved in study selection, data extraction and quality assessment of included studies.”

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

Lockley 2014 is at low risk of bias due to missing data because data were reasonably complete.

“Data were reasonably complete”

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

Lockley 2014 is at low risk of bias in measurement of outcomes because outcome assessment methods were comparable across intervention periods.

“The methods of outcome assessment were comparable across intervention periods”

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

Hemmer 2017 is at low risk of bias due to missing data because data were reasonably complete.

“Data were reasonably complete”

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

Law 2014 is at low risk of bias due to missing data because data were reasonably complete.

“Data were reasonably complete”

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

Hemmer 2017 is at moderate risk of bias due to confounding because the study did not account for potential confounding from media publicity that could have promoted the jump site.

“The study did not take into account the possibility of potential confounding effects as a result of media publicity which could have promoted the jump site”

• AssumptionRests on an assumption: Media publicity could have promoted the jump site, and thus could confound studies of means restriction at jump sites.

May combine two claims.

All of our included studies were conducted in high-income countries, and as such, our study findings may not be applicable to low-or middle-income countries, where financial and/or operational constraints may present challenges.

“All of our included studies were conducted in highincome countries, and as such, our study findings may not be applicable to low-or middle-income countries, where financial and/or operational constraints may present challenges.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

It should be noted that due to the nature of this topic, it would be highly impractical to undertake a randomised controlled trial (RCT) to evaluate the effectiveness of interventions. Observational studies, being more prone to bias, will therefore receive lower quality ratings.

“It should be noted that due to the nature of this topic, it would be highly impractical to undertake a randomised controlled trial (RCT) to evaluate the effectiveness of interventions.Observational studies, being more prone to bias, will therefore receive lower quality ratings.”

May combine two claims.

We assessed the quality of the evidence included in this review as low using the GRADE criteria. This low rating was mainly due to methodological limitations of the included studies and considerable heterogeneity between studies.

“We assessed the quality of the evidence included in this review as low using the GRADE criteria.This low rating was mainly due to methodological limitations of the included studies and considerable heterogeneity between studies.”

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

Law 2014 is at low risk of bias in classification of interventions because intervention status is well defined.

“Intervention status is well defined”

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

Sinyor 2010/Sinyor 2017 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

Reisch 2005 is at moderate risk of bias due to confounding because the study did not account for potential confounding from media publicity that could have promoted the jump site.

“The study did not take into account the possibility of potential confounding effects as a result of media publicity which could have promoted the jump site”

• AssumptionRests on an assumption: Media publicity could have promoted the jump site, and thus could confound studies of means restriction at jump sites.

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

Whitmer 2013's economic evaluation's allowance for uncertainty was not clear.

“Allowance for uncertainty Not clear”

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

Reisch 2005 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

Reisch 2005 is at low risk of bias in classification of interventions because intervention status is well defined.

“Intervention status is well defined”

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

Whitmer 2013's economic evaluation adjusted for the timing of costs and benefits.

“Adjustments for timing of costs and benefits Yes”

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

Sinyor 2010/Sinyor 2017 is at unclear risk of bias in selection of reported result because no study protocols were accessible.

“No access to study protocols for this study therefore it is difficult to assess this domain”

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

Reisch 2005 is at unclear risk of bias in selection of reported result because no study protocols were accessible.

“No access to study protocols for this study therefore it is difficult to assess this domain”

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

Whitmer 2013's economic evaluation's modelling was not clear.

“Modelling Not clear”

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

Sinyor 2010/Sinyor 2017 is at low risk of bias in classification of interventions because intervention status is well defined.

“Intervention status is well defined”

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

Pelletier 2007 is at low risk of bias in measurement of outcomes because outcome assessment methods were comparable across intervention periods.

“The methods of outcome assessment were comparable across intervention periods”

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

Sinyor 2010/Sinyor 2017 is at moderate risk of bias due to confounding because the study did not account for potential confounding from media publicity that could have promoted the jump site.

“The study did not take into account the possibility of potential confounding effects as a result of media publicity which could have promoted the jump site”

• AssumptionRests on an assumption: Media publicity could have promoted the jump site, and thus could confound studies of means restriction at jump sites.

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

Whitmer 2013's economic evaluation adequately costed the intervention.

“Costing Yes”

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

Whitmer 2013's economic evaluation presented its results adequately.

“Presentation of results Yes”

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

Sinyor 2010/Sinyor 2017 is at low risk of bias in measurement of outcomes because outcome assessment methods were comparable across intervention periods.

“The methods of outcome assessment were comparable across intervention periods”

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

Reisch 2005 is at low risk of bias due to missing data because data were reasonably complete.

“Data were reasonably complete”

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

Reisch 2005 is at low risk of bias in measurement of outcomes because outcome assessment methods were comparable across intervention periods.

“The methods of outcome assessment were comparable across intervention periods”

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

Pelletier 2007 is at moderate risk of bias due to missing data because data from 1968 were not available.

“Data from 1968 were not available”

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

Sinyor 2010/Sinyor 2017 is at low risk of bias due to missing data because data were reasonably complete.

“Data were reasonably complete”

May combine two claims.

Pelletier 2007 is at low risk of bias in selection of participants/data because the study assessed interventions aimed at the general population, so all eligible individuals were included.

“The study assessed interventions that were aimed at the general population, therefore all individuals who would have been eligible to take part in the study were included”

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

Whitmer 2013's economic evaluation adequately measured and valued benefits.

“Benefit measurement and valuation Yes”