• Early initiation of highly active antiretroviral treatment Two cohort studies added. One prospective study found mortality was increased the longer HAART initiation was delayed after starting antituberculosis treatment, and found decreased mortality in people with bacteriologically confirmed tuberculosis, who started HAART within the first 120 days of antituberculosis treatment versus people who started HAART later than this. Another study (retrospective from 1996 to 2000; prospective from 2000 to 2004) found better survival in people who started HAART within 2 months of starting antituberculous treatment versus people who started HAART more than 3 months after antituberculous treatment.Categorisation unchanged (unknown effectiveness).

• Alginate dressing One systematic review added (search date 2008), which found no RCTs that met Clinical Evidence inclusion criteria. Categorisation unchanged (Unknown effectiveness).
• Chlorhexidine-impregnated paraffin gauze dressing One systematic review added (search date 2008), which did not pool data, and found no additional RCTs to those previously reported in this Clinical Evidence review. Categorisation unchanged (Unknown effectiveness).
• Foam dressing One systematic review added (search date 2008),which found no RCTs. Categorisation unchanged (Unknown effectiveness).
• Hydrocolloid dressing One systematic review added (search date 2008), which did not pool data and found no additional RCTs to those previously reported in this Clinical Evidence review. Categorisation unchanged (Unknown effectiveness).
• Hydrogel dressing One systematic review added (search date 2008), which found no RCTs that met the inclusion criteria for this Clinical Evidence review. One additional RCT added, which found that a liposome hydrogel including polyvinylpyrrolidone iodine reduced time to wound closure compared with silver sulfadiazine cream; however, the RCT found similar results for clinician assessments of inflammation and healing, and similar results for participant assessments of pain and itching, with both the polyvinylpyrrolidone iodine hydrogel and silver sulfadiazine cream. Categorisation unchanged (Unknown effectiveness).
• Polyurethane film One systematic review added (search date 2008), which did not pool data and found no additional RCTs to those previously reported in this Clinical Evidence review. Categorisation unchanged (Unknown effectiveness).
• Silicone-coated nylon dressing One systematic review added (search date 2008), which did not pool data and found no additional RCTs to those previously reported in this Clinical Evidence review. Categorisation unchanged (Unknown effectiveness).
• Silver sulfadiazine cream One systematic review added (2008), which did not pool data and found no additional RCTs to those previously reported in this Clinical Evidence review. Categorisation unchanged (Unknown effectiveness).

• Tonsillectomy versus no surgery in adults One systematic review added identifying one RCT that met Clinical Evidence inclusion criteria. The RCT found that tonsillectomy reduced the frequency and duration of sore throat at 5 to 6 months, but had short follow-up. Categorisation unchanged (Trade-off between benefits and harms).

New evidence; conclusion confirmed for:

• Tonsillectomy versus no surgery in children One systematic review updated (4 RCTs; 564 children), which found that tonsillectomy reduced the number of episodes of sore throat, and duration of sore throat compared with standard care at 1 to 3 years. Categorisation unchanged (Trade-off between benefits and harms).

• Statins versus placebo or no treatment One systematic review and two subsequent RCTs added. The review found that statins reduced all-cause mortality and cardiovascular mortality compared with placebo or no treatment. The first subsequent RCT found that simvastatin decreased all-cause mortality, major coronary events, and the composite outcome of death from coronary disease and non-fatal MI compared with placebo. The second subsequent RCT found no significant difference between pravastatin and placebo or no treatment in rates of progression to end-stage renal disease (ESRD), composite outcome of ESRD or 50% decrease in glomerular filtration rate (GFR), or the composite outcome of ESRD or 25% decrease in GFR. Categorisation changed from Unknown effectiveness to Likely to be beneficial.

New evidence; conclusion confirmed for:

• ACE inhibitors One RCT added comparing trandolapril versus placebo with a median follow-up of 4.8 years. It found that trandolapril reduced all-cause mortality compared with placebo in people with a glomerular filtration rate (GFR) of less than 60 mL/minute/1.73 m². Categorisation unchanged (Likely to be beneficial).
• ACE inhibitors plus angiotensin II receptor antagonist versus either drug alone One RCT added comparing ACE inhibitors plus angiotensin II receptor antagonists versus ACE inhibitors alone. The RCT found that ACE inhibitors plus candesartan significantly increased the level of serum creatinine at 3 years and significantly decreased urinary protein excretion at 1, 2, and 3 years compared with ACE inhibitors alone. Categorisation unchanged (Likely to be beneficial).
• Angiotensin II receptor antagonists Three RCTs added, one comparing angiotensin II receptor antagonists versus placebo,and two comparing angiotensin II receptor antagonists versus ACE inhibitors. The first RCT found that valsartan did not decrease progression to end-stage renal disease (ESRD) or rates of doubling in serum creatinine level, but it did decrease proteinuria and slowed the mean rate of decline of glomerular filtration rate (GFR) compared with placebo at 2 years. The second RCT found that angiotensin II receptor antagonists did not decrease the rate of progression to ESRD compared with ACE inhibitors. The third RCT found no significant difference between angiotensin II receptor antagonists and ACE inhibitors in the composite outcome of doubling of the serum creatinine, ESRD, or mortality. Categorisation unchanged (Unknown effectiveness).
• Targeted lowering of albuminuria/proteinuria One RCT added comparing up-titrated benazepril or losartan versus non-titrated benazepril or losartan. The RCT found that up-titrated doses of both benazepril and losartan reduced the risk of the composite outcome of doubling of serum creatinine, end-stage renal disease (ESRD), or death compared with the non-titrated doses. Categorisation unchanged (Unknown effectiveness).
• Lowering blood pressure below usual targets Two additional outcomes added for the African American Study of Renal Disease (AASK) trial in two papers. The first paper reported that the RCT found no significant difference between low and usual target blood pressure in cardiovascular mortality, cardiovascular composite outcome of cardiovascular death or admission to hospital for CVD, or all cardiovascular events, including multiple events in the same person. The second paper reported that the RCT found no significant difference in mean change in composite physical health or the composite mental health outcomes of the Short Form 36 between groups over the 4 years of the study. Categorisation unchanged (likely to be beneficial)

• Single- or multiple-session brief intervention in hazardous or harmful drinkers in primary care One systematic review and one subsequent RCT added comparing single- or multiple-session brief interventions with either control or extended interventions. Categorisation unchanged (Beneficial).
• Brief intervention (single- or multiple-session) in emergency departments One systematic review and one subsequent RCT added comparing single-session brief intervention (BI) with control. The review found no significant difference between single-session BI and control in quantity of or frequency of alcohol consumption, but found that single-session BI significantly decreased the number of alcohol-related injuries compared with control at 12 months. The subsequent RCT found that single-session BI reduced alcohol consumption at 12 months compared with baseline. Categorisation unchanged (Likely to be beneficial).

• H pylori eradication treatments (quadruple regimens compared with triple regimens as second-line treatment) New option for which we found three RCTs. The RCTs found that quadruple regimens as second-line therapies were more effective than triple regimens at eradicating H pylori. Most triple regimens given did not contain a nitroimidazole. Categorised as Likely to be beneficial.
• H pylori eradication treatments (sequential regimens compared with triple regimens as first-line treatment) New option for which we found one systematic review. The review found sequential therapy was more effective at increasing H pylori eradication rates compared with proton pump inhibitor triple regimens. Categorised as Likely to be beneficial.

New evidence; conclusions changed for:

• H pylori eradication treatments (quadruple regimens compared with triple regimens as first-line treatment) One systematic review results updated and published as a letter to the editor.The review found no significant difference in eradication rates between quadruple regimens and triple regimens. Categorisation changed from Likely to be beneficial to Unlikely to be beneficial with the rationale that quadruple regimens are no more effective than triple regimens. Adding a fourth drug to initial eradication treatment confers no additional benefit.
• H pylori eradication treatments (duration of H pylori eradication as first-line treatment) Three RCTs added, which found no significant difference in H pylori eradication rates between 1-week and 2-week triple regimens, although in all three RCTs eradication rates were higher with 2-week regimens. Categorisation unchanged: both Likely to be beneficial, but the 2-week triple regimen is more effective than the 1-week triple regimen.

New evidence; conclusion confirmed for:

• H pylori eradication (in uninvestigated dyspepsia) Three RCTs added. The first RCT found that eradication treatment was more effective at improving dyspepsia symptoms than placebo in long-term proton pump inhibitor users.Two RCTs found similar dyspepsia scores with test and treat, prompt endoscopy, and empirical eradication treatment. Categorisation unchanged (beneficial).
• H pylori eradication treatments (different triple regimens compared with each other) Four systematic reviews and 12 RCTs added. One systematic review and three RCTs (2 RCTs reported in 1 publication) found that nitroimidazole-based triple regimens and amoxicillin-based triple regimens were equally effective at eradicating H pylori. Two RCTs found no significant difference in H pylori eradication rates between higher and lower dose clarithromycin within nitroimidazole-based triple regimens. Three systematic reviews and three subsequent RCTs found no significant difference in H pylori eradication rates between different proton pump inhibitor-based triple regimens. Four RCTs found lower eradication rates in people infected with strains of H pylori resistant to antibiotics included in the eradication regimen compared with people infected with sensitive strains. It is unclear whether any one triple regimen is more effective than another. Categorisation unchanged (Unknown effectiveness).

• Photodynamic treatment One systematic review added. It found that photodynamic treatment was more effective at wart clearance than placebo or cryotherapy, and that neutral red photodynamic treatment and proflavine photodynamic treatment have similar effects. Categorisation unchanged (Likely to be beneficial).
• Intralesional bleomycin Data from one systematic review updated, and one RCT added. The RCT found that intralesional bleomycin increased wart clearance compared with placebo. Categorisation unchanged (unknown effectiveness).
• Duct tape occlusion One systematic review and two RCTs (3 publications) added. The systematic review found no significant difference in wart clearance between duct tape occlusion and cryotherapy. The two RCTs found no significant difference in wart clearance between duct tape occlusion and placebo. One of the RCTs also found no significant difference in wart recurrence between duct tape occlusion and placebo. Categorisation unchanged (Unknown effectiveness).
• Pulsed dye laser One RCT added, which found no significant difference in wart clearance between pulsed dye laser and placebo. Categorisation unchanged (Unknown effectiveness)

• Rotavirus vaccines New option for which we found one systematic reviewand eight subsequent RCTs; one RCT was reported in two papers. The systematic review found that live-attenuated bovine rotavirus vaccine and human-attenuated rotavirus vaccine were safe, and significantly reduced the number of episodes and severe episodes of gastroenteritis caused by rotavirus, and admissions to hospital with gastroenteritis caused by rotavirus, compared with placebo. The eight subsequent RCTs found similar results. Categorised as Beneficial.
• Ondansetron New option for which we found one systematic review. The review did not pool data because of clinical heterogeneity among the RCTs and a paucity of data. Three RCTs identified by the review found that ondansetron reduced episodes of vomiting compared with placebo. The RCTs found that ondansetron was associated with an increased risk of diarrhoea compared with placebo, but the RCTs were small in size and, taken together, do not provide sufficient evidence to draw a definitive conclusion regarding adverse effects of ondansetron. Categorised as Likely to be beneficial.

New evidence; conclusions changed for:

• Loperamide One systematic review added, which found that loperamide significantly reduced the mean duration of diarrhoea compared with placebo. However, the incidence of adverse effects was significantly higher in children treated with loperamide compared with placebo; all serious adverse effects occurred in children under 3 years of age. Categorisation changed from Likely to be beneficial to Trade-off between benefits and harms.

• Universal vaccination of infants Two observational studies added, which found that universal vaccination of infants reduced the proportion of hepatitis B carriers compared with no vaccination. Categorised as Beneficial.
• Selective vaccination of high-risk individuals No new evidence on effects of selective vaccination of high-risk individuals. One retrospective cohort study added to harms found no evidence of a link between autoimmune thyroid disease and hepatitis B vaccine. Categorised as Likely to be beneficial.
• Universal vaccination of infants No new evidence on the effects of universal vaccination of infants added. One retrospective cohort study added to harms found no evidence of a link between acute leukaemia and hepatitis B vaccination. Categorised as Likely to be beneficial.

• Oral antiviral agents to prevent recurrence One systematic review added (search date 2008),which did not pool data. It found three RCTs and one pooled analysis of two further RCTs that were already reported in this Clinical Evidence review. No new data added from the new review. Categorisation unchanged (Likely to be beneficial).
• Topical antiviral agents to prevent recurrence One systematic review added (search date 2008), which identified two RCTs. The review did not pool data, and the results of the RCT were reported from the original papers. Benefits and harms section enhanced. Categorisation unchanged (Unknown effectiveness).
• Sunscreen One systematic review added (search date 2008)identifying one small crossover RCT already reported in this Clinical Evidence review. No new data added from the new review. Categorisation unchanged (Likely to be beneficial).
• Oral antiviral agents for treating recurrent attacks One systematic review added (search date 2008), which did not pool data. It identified four RCTs previously reported in this Clinical Evidence review, and one additional RCT comparing famciclovir versus placebo not previously reported in this Clinical Evidence review. Results from this RCT added from the original report of the RCT. Categorisation unchanged (Likely to be beneficial).
• Topical antiviral agents for treating recurrent attacks One systematic review added (search date 2008),which did not pool data and identified 12 RCTs already reported in this Clinical Evidence review. No new data from the systematic review added.Categorisation unchanged (Unknown effectiveness).
• Topical anaesthetic agents for treating recurrent attacks One systematic review added (search date 2008), which found no RCTs of sufficient quality. No data added from the new review. Categorisation unchanged (Unknown effectiveness).
• Zinc oxide cream for treating recurrent attacks One systematic review added (search date 2008),which identified one RCT previously reported in this Clinical Evidence review. No new data added from the review.Categorisation unchanged (Unknown effectiveness).

• Glasses One cohort study added found that visual acuity was better at 1 year in over two-thirds of children who wore glasses. Categorisation unchanged (Likely to be beneficial, based on consensus).
• Occlusion (patching) Three systematic reviews added identified the same RCTs as previously found by Clinical Evidence. One further RCT identified found no significant difference in visual acuity at 18 weeks between 12 and 6 hours' prescribed occlusion, but added to the evidence for a dose response when actual hours of received occlusion can be measured. Categorisation unchanged; occlusion in combination with penalisation and near-vision tasks categorised as Beneficial compared with glasses alone.
• Penalisation One RCT added found that penalisation using atropine improved visual acuity at 6 months compared with optical penalisation. Categorisation unchanged (Likely to be beneficial).

• Anticoagulants in addition to antiplatelet treatments: Two systematic reviews added comparing intensive warfarin plus aspirin versus aspirin alone. Both reviews found that warfarin plus aspirin decreased the risk of non-fatal MI and stroke, but increased the risk of major bleeding compared with aspirin alone in people with acute coronary syndrome. The evidence supports the recategorisation from Likely to be ineffective or harmful to Trade-off between benefits and harms.
• Fish oil consumption (from oily fish or capsules): Two systematic reviews added comparing advice to increase fish oil consumption or fish oil consumption versus control. The reviews found that increased fish oil consumption decreased all-cause mortality when studies with heterogeneity were removed from the analysis. Categorisation changed from Unknown effectiveness to Likely to be beneficial.

New evidence; conclusion confirmed for:

• Combination antiplatelets: Two RCTs added comparing clopidogrel plus aspirin versus aspirin alone. The first RCT found a reduction in the combined risk of death, MI, or stoke at 1 year with clopidogrel plus aspirin compared with aspirin alone with extended follow-up of patients after PCI. The second RCT found no significant difference between groups for the primary efficacy end point of cardiovascular death, MI, or stroke at a median of 28 months' follow-up. Categorisation unchanged (Beneficial)
• Beta-blockers: One RCT and one systematic review added. The RCT compared carvedilol with metoprolol in people with MI and found no significant difference between groups in all-cause mortality. The review compared metoprolol versus placebo after CABG. The review found no significant difference between groups in repeat revascularisation, unstable angina, non-fatal MI, or mortality. Categorisation unchanged (Beneficial).
• ACE inhibitors: Three systematic reviews and two subsequent RCTs added comparing ACE inhibitors versus placebo in people with normal left ventricular function and no heart failure. All of the trials found that ACE inhibitors reduced the risk of mortality compared with placebo. One RCT added comparing ACE inhibitors versus placebo in people with left ventricular dysfunction. The RCT found no significant difference in mortality between groups, but may have had insufficient power to detect clinically meaningful differences in outcome. Categorisation unchanged (Beneficial).
• Angiotensin receptor blockers plus ACE inhibitors: One RCT added comparing valsartan plus captopril versus captopril alone in people with left ventricular dysfunction over a mean 24.7 months' follow-up. The RCT found no significant difference between groups for all-cause mortality or for cardiovascular mortality. We still don't know whether angiotensin receptor blockers plus ACE inhibitors are more effective than ACE inhibitors alone. Categorisation unchanged (Unknown effectiveness).
• Statins: Four systematic reviews and one RCT added comparing statin therapy versus control, higher-intensity statin therapy versus lower-intensity statin therapy, or statin therapy in different groups of people.Two systematic reviews found that statin therapy reduced the risk of all-cause mortality compared with placebo. A third systematic review found that statin therapy reduced the risk of MI compared with placebo. The fourth systematic review found that higher-intensity statin therapy reduced the risk of coronary death or MI compared with lower-intensity statin therapy, and the RCT found that older people benefit more from statin therapy compared with younger people, as they are at higher risk of events. Statin therapy has been associated with rare but serious adverse effects, rhabdomyolysis, creatine kinase increase to at least 10 times the upper limit of normal, and myalgia. Categorisation unchanged (Beneficial).
• Blood pressure reduction : One RCT added comparing amlodipine or enalapril versus placebo, and amlodipine versus enalapril. The RCT found that amlodipine reduced cardiovascular events compared with placebo, but there was no significant difference between enalapril and placebo for this outcome. The RCT also found that amlodipine reduced the risk of admission to hospital for angina compared with enalapril, but found no significant difference in all-cause mortality, cardiovascular mortality, cardiovascular events, non-fatal MI, stroke or TIA, coronary revascularisation, admission to hospital for heart failure, or new-onset peripheral artery disease. Categorisation unchanged (Beneficial).
• Antioxidant vitamin combinations: One RCT added comparing beta-carotene, vitamin C, and vitamin E alone or in combination with placebo. The RCT found no significant difference for any combination of the antioxidants compared with placebo for the primary composite outcome of all-cause mortality, cardiovascular mortality, MI, stroke, or revascularisation. However, the RCT found that vitamin C plus vitamin E significantly reduced the risk of stroke compared with placebo. Categorisation unchanged (Unlikely to be beneficial).
• Beta-carotene: One RCT added comparing beta-carotene, vitamin C, and vitamin E alone or in combination with placebo. The RCT found that beta-carotene did not reduce the risk of the primary composite outcome of all-cause mortality, cardiovascular mortality, MI, stroke, or revascularisation, or the secondary individual outcomes of CVD-related mortality, MI, stroke, or need for revascularisation compared with placebo. Categorisation unchanged (Likely to be ineffective or harmful).
• Vitamin C: One RCT added comparing vitamin C, vitamin E, or beta-carotene alone or in combination with placebo. The RCT found that vitamin C did not reduce the primary composite outcome of all-cause mortality, cardiovascular mortality, MI, stroke, or revascularisation, or secondary individual outcomes compared with placebo. Categorisation unchanged (Unlikely to be beneficial).
• Vitamin E: One RCT added comparing vitamin E, vitamin C, or beta-carotene alone or in combination with placebo. The RCT found that vitamin E did not reduce the primary composite outcome of all-cause mortality, cardiovascular mortality, MI, stroke, or revascularisation, or secondary individual outcomes compared with placebo. However, stronger results were found when the results were re-analysed factoring compliance. Categorisation unchanged (Likely to be ineffective or harmful).
• Cardiac rehabilitation including exercise: One systematic review and two subsequent RCTs added. The systematic review and one subsequent RCT compared cardiac rehabilitation with or without exercise versus usual care and cardiac rehabilitation programme of health education versus usual care. Both found that cardiac rehabilitation reduced the risk of all-cause mortality at longer-term follow-up. The other subsequent RCT compared two types of cardiac rehabilitation: physical training plus information for risk-factor management versus physical training, and information for risk-factor management supplemented with relaxation therapy versus psycho-education sessions. The RCT found no difference between groups in quality of life. Categorisation unchanged (Beneficial).
• Advice to eat less fat: One systematic review added comparing reduced intake of saturated fats with normal diet. The review found no significant difference in all-cause mortality between groups. Categorisation unchanged (Unknown effectiveness).
• Psychological treatment: One systematic review and one subsequent RCT added comparing psychological interventions, stress management, and behavioural interventions versus usual care. The review found that psychological interventions plus stress management, and stress management alone reduced the risk of non-fatal MI. However, there was heterogeneity between studies. The RCT found no difference between behavioural interventions and usual care for quality-of-life outcomes. Categorisation unchanged (Likely to be beneficial).
• CABG versus medical treatment alone: One RCT added. It found no significant difference between CABG and medical treatment in mortality or MI-free survival after 5 years, but found that CABG reduced the need for additional revascularisation and led to greater elimination of angina compared with medical treatment alone. Categorisation unchanged (Beneficial).
• PTCA versus medical treatment alone: One systematic review and four RCTs added comparing PTCA with medical treatment. The systematic review, a long-term follow-up of an RCT included in the review, and two subsequent RCTs found no difference in rates of mortality for PTCA compared with medical treatment. One subsequent RCT found that PTCA reduced the risk of mortality compared with medical treatment in people with an MI within the last 3 months and objective evidence of ischaemia. Categorisation unchanged (Likely to be beneficial).
• CABG versus PTCA: Seven systematic reviews, four subsequent RCTs, and one additional RCT added comparing CABG with PTCA with or without stents. The reviews and RCTs found that, in people with coronary artery disease, CABG reduced the risk of major coronary events compared with PTCA, particularly when repeat revascularisation was included in a combined outcome. Categorisation unchanged (Likely to be beneficial)
• PTCA with stents versus PTCA alone: Three systematic reviews and one follow-up RCT added comparing PTCA with stents versus PTCA. Most new evidence found that stents reduced the risk of major adverse cardiac events compared with PTCA, supporting the categorisation of Beneficial.

• Behavioural therapy in children and adolescents New option added, for which we found no data satisfying Clinical Evidence inclusion criteria. Categorised as Unknown effectiveness.
• Cognitive therapy or CBT in children and adolescents New option added, for which one systematic review and two RCTs were added. The systematic review found that CBT improved OCD symptoms versus waiting list control or placebo. However, it found no significant difference between CBT and SRIs. One RCT found no significant difference between intensive once a weekday CBT and once-weekly CBT in improvement of OCD symptoms. Categorised as Beneficial.
• SRIs in children and adolescents New option added, for which two systematic reviews were added. They found improvements in OCD symptoms with fluoxetine, paroxetine, fluvoxamine, sertraline, and clomipramine versus placebo in children and adolescents. Categorised as Trade-off between benefits and harms.
• Behavioural therapy or cognitive therapy plus SRIs in children and adolescents New option added, for which one systematic review was added. It found that CBT plus an SRI improved OCD symptoms versus placebo or an SRI alone, but not versus CBT alone. Categorised as Trade-off between benefits and harms.
• Optimum duration of maintenance treatment with SRIs in children and adolescents New option added, for which we found no data satisfying Clinical Evidence inclusion criteria. Categorised as Unknown effectiveness.

New evidence; conclusions changed for:

• Behavioural therapy in adults Two systematic reviews added. One review found that behavioural therapy significantly improved OCD symptoms over 6–16 weeks of treatment compared with waiting list control. One review identified one previously included RCT, satisfying Clinical Evidence inclusion criteria. No new data added. Categorisation changed from Beneficial to Likely to be beneficial.
• Cognitive therapy or CBT in adults One systematic review added comparing cognitive therapy or CBT versus waiting list control. It found no significant difference between cognitive therapy after 9‒16 weeks' treatment and waiting list control in symptoms of OCD. It found that CBT improved symptoms of OCD after 6‒20 weeks' treatment versus waiting list control. Categorisation changed from Beneficial to Likely to be beneficial.

New evidence; conclusion confirmed for:

• SRIs in adults One RCT added comparing sertraline versus cognitive-behavioural group therapy found no significant difference between groups in reduction in Yale–Brown Obsessive Compulsive Scale score or quality of life after 12 weeks. Categorisation unchanged (Beneficial).
• Addition of antipsychotics to SRIs in adults One systematic review added, which identified nine RCTs, four of which were already included in this Clinical Evidence review. It found that adding antipsychotics versus adding placebo to an SRI increased the proportion of treatment responders (adults with a reduction of 35% or more in the Yale–Brown Obsessive Compulsive Scale score). Categorisation unchanged (Likely to be beneficial).

New option(s) added for:

• Prompting mechanisms New option. Ten systematic reviews identified (search date 1996;search date 2000;search date 2002; search date 2003; search date 2004 ), which did not pool data. The reviews identified seven RCTs of sufficient quality and we found one subsequent RCT.Five RCTs compared various prompting mechanisms versus usual care; two RCTs compared a prompting mechanism plus usual care versus unit-of-use packaging plus usual care, unit-of-use packaging plus prompting mechanism plus usual care, or usual care alone; and the remaining RCT compared a prompting mechanism plus patient health education versus usual care. Prompting mechanisms categorised as Likely to be beneficial.
• Simplified dosing New option. Ten systematic reviews identified (search date 1996; search date 2000; search date 2002; search date 2003; search date 2004 ), which did not pool data. The reviews identified five RCTs of sufficient quality, and we found no subsequent RCTs. Three RCTs compared once-daily versus twice-daily medication, one RCT compared twice-daily versus three-times daily medication, and the remaining RCT compared twice-daily versus four-times daily medication. Simplified dosing categorised as Likely to be beneficial.
• Prescriber education New option. We found eight systematic reviews (search date 1996;search date 2000;search date 2002; search date 2003; search date 2004 ), which found no RCTs of sufficient quality. We found no subsequent RCTs. Prescriber education categorised as Unknown effectiveness.
• Reminder packaging (calendar [blister] packs; multi-dose pill boxes) New option. We identified ten systematic reviews (search date 1996; search date 2000;search date 2002; search date 2003; search date 2004 ), which did not pool data. The reviews identified three RCTs. We found no subsequent RCTs. One RCT compared a calendar (blister) pack versus usual care, while two RCTs compared unit-of-use packaging plus usual care versus a prompting mechanism plus usual care, unit-of-use packaging plus prompting mechanism plus usual care, or usual care alone. Reminder packaging categorised as Unknown effectiveness.
• Patient health education New option. We identified nine systematic reviews (search date 1996; search date 2000;search date 2002; search date 2003; search date 2004 ), which did not pool data. The reviews identified nine RCTs of sufficient quality published in 10 reports, and we found two subsequent RCTs. Ten RCTs, including one extended follow-up report of an RCTcompared various patient health education interventions versus usual care; and one RCT compared patient health education plus a prompting mechanism versus usual care.Patient health education categorised as Unlikely to be beneficial.

• Laser trabeculoplasty plus topical medical treatment (in people with primary open-angle glaucoma, ocular hypertension, or both) One systematic review added. It identified one RCT already included in this Clinical Evidence review, so no new data added. Categorisation unchanged (Likely to be beneficial).
• Topical medical treatment (in people with primary open-angle glaucoma, ocular hypertension, or both) Two systematic reviews added. One review found reduced visual field loss progression with topical medical treatment (beta-blockers, dorzolamide, or unspecified) versus placebo or no treatment at 2 to 3 years' follow-up. The other review found no significant difference in visual field loss progression with initial medical treatment versus initial laser trabeculoplasty in people with newly diagnosed glaucoma at 2 years' follow-up.Categorisation unchanged (Likely to be beneficial).
• Laser trabeculoplasty (in people with primary open-angle glaucoma, ocular hypertension, or both) One systematic review added.It found no significant difference in visual field loss progression with initial laser trabeculoplasty versus initial medical treatment in people with newly diagnosed glaucoma at 2 years' follow-up. Categorisation unchanged (Unknown effectiveness).

• Corticosteroid injection during vitrectomy surgery New option added for which we identified one RCT. The RCT found no significant difference between adjunctive triamcinolone acetonide and no adjunctive treatment in retinal re-attachment rate or visual acuity at 6 months. Categorised as Unknown effectiveness.
• Daunorubicin infusion during vitrectomy surgery New option added for which we found two RCTs. Both RCTs found no significant difference between adding daunorubicin and not adding daunorubicin in rate of retinal re-attachment and in visual acuity. One RCT found that a smaller proportion of people in the daunorubicin group required further vitreoretinal surgery within 1 year after initial surgery compared with the group not receiving daunorubicin.Categorised as Unknown effectiveness.

New evidence; conclusions changed for:

• Scleral buckling versus primary vitrectomy Condition restructured and three RCTs added. One large RCT analysed results separately for pseudophakic plus aphakic rhegmatogenous retinal detachment (RRD) and for phakic RRD. In pseudophakic and aphakic RRD, the RCT found a lower primary anatomical success rate at 1 year with scleral buckling compared with vitrectomy, but found no significant difference between groups in visual acuity. However, the RCT reported that additional scleral buckling was used (carried out at surgeon's discretion) at the time of primary surgery in over half of the vitrectomy group. In people with phakic RRD, the subgroup analysis and two other RCTs found similar rates of retinal re-attachment for scleral buckling and vitrectomy. Conflicting results were reported for improvement in visual acuity in people with phakic RRD. Although there is consensus that both techniques are effective in the treatment of RRD, there is insufficient evidence to assess the effects of scleral buckling versus those of primary vitrectomy. Categorisation changed from Likely to be beneficial to Unknown effectiveness.

• Buprenorphine for stabilisation: One systematic review that searched for systematic reviews and RCTs added. The systematic review found no new evidence that met our reporting criteria. The systematic review did not perform a meta-analysis for buprenorphine versus placebo but reported that the reviews found buprenorphine was more effective at retaining people in treatment (with higher doses being more effective than lower doses) and at reducing opiate use compared with placebo or no drug treatment. Categorisation of Buprenorphine versus placebo unchanged (Beneficial). The systematic review carried out a meta-analysis for buprenorphine versus methadone and found that flexible dosing regimens of buprenorphine were less effective at increasing retention in treatment compared with flexible dosing regimens of methadone. However, previous data reporting similar rates of opioid use with both buprenorphine and methadone were unchanged so still insufficient data to assess which drug is more effective. Categorisation of Buprenorphine versus methadone unchanged (Unknown effectiveness; both beneficial and seem as effective as each other).
• Methadone for stabilisation: Two systematic reviews added. One systematic review searched for systematic reviews and RCTs and found no new evidence that met our reporting criteria. The systematic review did not perform a meta-analysis for methadone versus placebo. It reported that the reviews found methadone was more effective at increasing retention in treatment and at reducing opioid misuse as assessed by self-reported heroin compared with placebo or no drug treatment. It also found higher doses of methadone were more effective than lower doses at increasing the proportion of people retained in treatment, and at reducing heroin abstinence rates (self-reported heroin use and urine-confirmed opioid abstinence). Categorisation of Methadone versus placebo unchanged (Beneficial). The systematic review carried out a meta-analysis for methadone versus buprenorphine and found that flexible dosing regimens of methadone were more effective at increasing retention in treatment compared with flexible dosing regimens of buprenorphine. The other systematic review found that methadone increased the proportion of people retained in treatment compared with placebo, control, and buprenorphine. However, previous data reporting similar rates of opioid use with both methadone and buprenorphine unchanged so still insufficient data to assess which drug is more effective. Categorisation of Buprenorphine versus methadone unchanged (Unknown effectiveness; both beneficial and seem as effective as each other).

• Anticonvulsant drugs New option added, for which we found no systematic review or RCTs. Categorised as Unknown effectiveness.
• Spinal manipulation (chiropractic and osteopathic treatment) Two systematic reviews and one further RCT added to benefits and harms. One RCT identified by one review found greater reduction in headache severity after treatment with palpatory examination plus osteopathic manipulation versus palpatory examination alone or versus instruction to rest in people with muscle tension-type headache. One RCT identified by both reviews found a reduction in headache intensity with amitriptyline versus spinal manipulation after 6 weeks’ treatment, but no significant reduction in headache frequency between groups. However, there was a significant reduction in headache intensity and frequency with spinal manipulation compared with amitriptyline at 4 weeks after treatment discontinuation. One further RCT found improved headache frequency with osteopathic treatment plus progressive muscular relaxation exercises versus progressive muscular relaxation exercises alone, but no significant difference between groups in headache intensity after 6–7 weeks’ follow-up. Two systematic reviews of observational studies added to the harms section found reported cases of serious adverse effects, including arterial dissection causing stroke, other stroke syndromes, and cerebellar and spinal cord injuries with the use of spinal manipulation. Categorisation as Likely to be ineffective or harmful.

New evidence; conclusions changed for:

• Noradrenergic and specific serotonergic antidepressants One RCT added found no significant difference in headache frequency, duration, or intensity with low-dose mirtazapine plus ibuprofen or mirtazapine alone versus ibuprofen alone or placebo. Categorisation changed (from Beneficial to Likely to be beneficial).

New evidence; conclusion confirmed for:

• Botulinum toxin Two RCTs added, comparing botulinum toxin versus placebo. One RCT found an increase in headache-free days with botulinum toxin 150 U (but not with lower doses) versus placebo after 60 days. It found no significant difference in headache duration or average headache severity with botulinum toxin versus placebo after 60 days. The other RCT found a reduction in headache intensity, but no significant difference in headache frequency with botulinum toxin versus placebo at an average of 178 days’ follow-up. Categorisation unchanged (Likely to be ineffective or harmful).
• Analgesics One RCT added found no significant difference in headache frequency and duration with ibuprofen alone, or plus low-dose mirtazapine, versus low-dose mirtazapine alone or placebo. It found an increase in headache intensity with ibuprofen alone compared with placebo. Categorisation unchanged (Likely to be ineffective or harmful).
• Acupuncture One RCT added, comparing acupuncture versus minimum acupuncture versus no intervention. It found improvements in headache frequency, duration, and headache score (assessing headache intensity) with acupuncture compared with no intervention. It found no significant difference in any of these outcomes with acupuncture compared with minimum acupuncture. Categorisation unchanged (Unknown effectiveness).

• Topical allylamines (naftifine, terbinafine): One already reported systematic review updated and two RCTs added. The updated review confirmed previous conclusions, and found naftifine and terbinafine were more effective at curing athlete's foot (defined as negative results on microscopy and no growth of dermatophytes in culture) compared with placebo, but not significantly different compared with each other. It found that the results for cure rates between allylamines and azoles were inconclusive. The RCTs found that topical allylamines were more effective than placebo at curing fungal skin infections. Categorisation unchanged (Beneficial).
• Topical azoles: One already reported systematic review updated, which confirmed previous conclusions. The updated review found that azole preparations were more effective at 4 to 6 weeks at curing athlete's foot (defined as negative results on microscopy and no growth of dermatophytes in culture) compared with placebo. It found that the results for cure rates between allylamines and azoles were inconclusive. The review found insufficient evidence of a difference in cure rates between different azoles at 2–6 weeks. It also found no significant difference in cure rates between clotrimazole and ciclopirox olamine. Categorisation unchanged (Beneficial).
• Topical ciclopirox olamine: One already reported systematic review updated, which confirmed previous conclusions. It found that ciclopirox olamine was more effective at curing athlete's foot (defined as negative results on microscopy and no growth of dermatophytes in culture) at 2 and 4 weeks compared with placebo, but found no significant difference between clotrimazole and ciclopirox olamine. Categorisation unchanged (Beneficial).

• Counselling people to increase physical activity versus no counselling: effects on level of physical activity One systematic review and three RCTs added to the existing reporting of three systematic reviews and 12 RCTs. The added systematic review did not pool data in our group of interest, and the three RCTs were therefore reported separately. Two of the added RCTs found that counselling increased physical activity, whereas the third found no significant difference among groups in physical activity. Categorisation unchanged (Likely to be beneficial).

• Intra-articular injections (of hyaluronan or corticosteroid) New option for which we found no systematic review or RCTs in people with osteoarthritis of the hip. Categorised as Unknown effectiveness.
• Hip resurfacing New option for which we found no systematic review or RCTs in people with osteoarthritis of the hip. Categorised as Unknown effectiveness.
• Arthroscopic debridement New option for which we found no systematic review or RCTs in people with osteoarthritis of the hip. Categorised as Unknown effectiveness.

New evidence; conclusion confirmed for:

• Acupuncture One systematic review added, which found no new evidence on the effects of acupuncture in people with osteoarthritis of the hip. One RCT added, which found that at 3 months, compared with no acupuncture, needle acupuncture improved WOMAC scores for pain, stiffness and function, and SF-36 physical component. However, the RCT found no significant difference between groups in the mental component of the SF-36 questionnaire. Categorisation unchanged (Unknown effectiveness).

• ABVD plus radiotherapy versus radiotherapy alone (stage I or II non-bulky Hodgkin's lymphoma) New option added for which we found one RCT. The RCT found that two cycles of ABVD plus radiotherapy improved freedom from treatment failure (composite outcome), and reduced risk of relapse at 7 years. However, the RCT found no significant difference between treatments in overall survival at 7 years. Categorised as Likely to be beneficial by consensus (ABVD plus radiotherapy is considered the international gold standard for early-stage Hodgkin's lymphoma).

New evidence; conclusion confirmed for:

• ABVD (stage II [bulky], III, or IV Hodgkin's lymphoma) One RCT added found that ABVD may be more effective than Stanford V at reducing disease recurrence (various outcomes) at 5 years, and as effective as MOPPEBVCAD. The RCT found similar rates for overall survival at 5 years for ABVD, Stanford V, and MOPPEBVCAD. Categorisation unchanged (Beneficial).
• ABVPP plus radiotherapy versus ABVPP alone One long-term follow-up studyadded found lower rates of survival at 10 years with combination treatment compared with ABVPP alone, which supported findings from 5-year follow-up. Categorisation unchanged (Unlikely to be beneficial).
• MOPP/ABV plus radiotherapy versus MOPP/ABV alone One long-term follow-up study added found similar rates of survival at 10 years with combination treatment compared with MOPP/ABV alone, which supported findings from 5-year follow-up. Categorisation unchanged (Unlikely to be beneficial).

• Advice to promote room sharing (without bed sharing): We found no direct evidence on the effects of room sharing (without bed sharing) on the prevention of SIDS. Categorised as Unknown effectiveness.

New evidence; conclusion confirmed for:

• Advice to avoid prone sleeping : One retrospective cohort study added assessing the change in incidence of SIDS in the 20 years since advice to avoid prone position sleeping campaign. The study found that factors contributing to SIDS have changed during this time period. Two observational studies added to the harms section assessing the effects of supine sleeping position on deformational plagiocephaly. The first observational study documented a temporal relationship between advice to avoid prone sleeping and an increase in the incidence of occipital plagiocephaly without synostosis. However, the incidence of other forms of plagiocephaly with synostosis did not change. The second observational study found that supine sleeping position was not a significant determinant of deformational plagiocephaly, but that sleeping with the head turned to the same side is a significant factor.Categorisation unchanged (Beneficial).
• Advice to avoid tobacco-smoke exposure: One retrospective cohort study added assessing the effect of smoking on SIDS. The study found that smoke-exposed infants were significantly more likely to die of SIDS. Categorisation unchanged (Likely to be beneficial).