NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Clinical Guideline Centre (UK). Nocturnal Enuresis: The Management of Bedwetting in Children and Young People. London: Royal College of Physicians (UK); 2010. (NICE Clinical Guidelines, No. 111.)

Cover of Nocturnal Enuresis

Nocturnal Enuresis: The Management of Bedwetting in Children and Young People.

Show details

12Enuresis Alarms in the management of bedwetting

12.1. Introduction

An enuresis alarm is a device that is activated by getting wet. According to Mowrer (1938) 98, the first enuresis alarms were bed-based, with the child sleeping on a pad or mat containing an electrical circuit. A bell would then ring as a result of the urine contacting the electrical circuit. There are several types of enuresis alarms available: pad-and-bell alarms where the sensor pad is positioned under a draw sheet beneath the child in the bed; body-worn alarms where the tiny sensor is attached to the child’s pants e.g. between 2 pairs of tightly fitting underpants and the alarm is worn on the pyjama top; and vibrating alarms.

Alarms achieve dryness over time by training the child to recognise the need to pass urine and to wake to go to the toilet or hold on. As such it would be expected that they would be most successful in children with arousal difficulties rather than children who are overproducing urine at night. Where possible therefore the evidence review highlights the population of children included in the study i.e. bedwetting with daytime symptoms, bedwetting only and mono-symptommatic nocturnal enuresis (see 2.1.4 and 2.2 for more detail).

Alarms were compared in the evidence review to pharmacological interventions singly and to each in combination with alarm: desmopressin (spray, tablets and melts), imipramine, amitriptyline, nortriptaline, oxybutinin, , long-acting tolterodine. Alarms were also compared to dry bed training with enuresis alarm, retention control training and star charts.

12.2. What is the clinical and cost effectiveness of enuresis alarms for children and young people under 19 years old who have bedwetting?

12.2.1. Evidence review

Study populations- children with bedwetting and possible daytime symptoms

12.2.1.1. Enuresis alarm compared to no treatment for children with bedwetting and possible daytime symptoms

Six randomised control trials evaluated enuresis alarm treatment compared to no treatment, or a waiting list group; these were: Baker (1969) 23, Bollard (1981) 93, Bollard (1982) 99, Houts (1986) 100, Jehu (1977) 101 and Moffatt (1987) 102

Table 12-1Enuresis alarm compared to no treatment - Clinical summary of findings

OutcomeAlarmNo treatmentRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights108/141 (76.6%)3/135 (2.2%)RR 16.9 (7.17 to 39.85)350 more per 1000 (from 136 more to 855 more)LOW
Mean number of wet nights per week at end of treatment (no SDs)1411-not pooledVERY LOW
Number of drop outs at end of trial4/34 (11.8%)0/31 (0%)RR 4.16 (0.5 to 34.6)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
12.2.1.2. Unsupervised enuresis alarm compared to supervised enuresis alarm for children with bedwetting and possible daytime symptoms

One randomised control trial Bollard (1981) 93 compared the supervision of enuresis alarm treatment for children with nocturnal enuresis, comparing an unsupervised enuresis alarm to a supervised enuresis alarm. The supervision was the parent or child (if old enough) contacting the author by telephone to report progress at a specific time, if contact was not made the author contacted the parent or child by telephone or letter.

Table 12-2Unsupervised enuresis alarm compared to supervised enuresis alarm - Clinical summary of findings

OutcomeUnsupervised alarmSupervised alarmRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights12/15 (80%)9/15 (60%)RR 1.33 (0.82 to 2.16)198 more per 1000 (from 108 fewer to 696 more)VERY LOW
Mean number of wet nights per week at end of treatment (no SDs)1515-not pooledVERY LOW
12.2.1.3. Enuresis alarm compared to imipramine

Two randomised control trials Fournier (1987) 81 and Kolvin (1972) 103 compared enuresis alarm to imipramine,. Fournier (1987) 81 gave 25 mg imipramine to children, Kolvin (1972) 103 did not state the dose of imipramine given to children.

Table 12-3Enuresis alarm compared to imipramine - Clinical summary of findings

OutcomeAlarmImipramineRelative risk (95% CI)Absolute effectQuality
Over 80% improvement in number of wet nights at the end of treatment17/32 (53.1%)16/35 (45.7%)RR 1.16 (0.71 to 1.89)73 more per 1000 (from 133 fewer to 407 more)VERY LOW
Mean number of wet nights per week at end of treatment (no SDs)4043-not pooledVERY LOW
Mean number of wet nights per week at follow-up (no SDs)3230-not pooledLOW
12.2.1.4. Enuresis alarm compared to amitriptyline

One randomised control trial Danquah (1975) 104 compared enuresis alarm to amitriptyline.

Table 12-4Increasing amitriptyline (Desmopressin)compared to enuresis alarm (placebo) - Clinical summary of findings

OutcomeDesmopressinPlaceboRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week after treatment (no SDs)1010-not pooledVERY LOW
12.2.1.5. Enuresis alarm compared to enuresis alarm with intranasal desmopressin

One randomised controlled trial Bradbury (1995) 105 compared enuresis alarms to enuresis alarms with 40mcg intranasal desmopressin.

Table 12-5Enuresis alarm compared to enuresis alarm with desmopressin - Clinical summary of findings

OutcomeAlarmAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 4 consecutive dry weeks16/27 (59.3%)27/33 (81.8%)RR 0.72 (0.51 to 1.03)229 fewer per 1000 (from 401 fewer to 25 more)LOW
Mean number of wet nights per week at end of treatment (no SDs)3536-not pooledLOW
Number of children relapsed at 6 months3/16 (18.8%)4/27 (14.8%)RR 1.27 (0.32 to 4.95)40 more per 1000 (from 101 fewer to 585 more)LOW
Number of drop outs at end of trial2/35 (5.7%)0/36 (0%)RR 5.14 (0.26 to 103.37)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
12.2.1.6. Enuresis alarm and placebo compared to enuresis alarm with desmopressin

One randomised controlled trial Sukhai (1989) 106 compared enuresis alarms and placebo to enuresis alarms with desmopressin.

Table 12-6Enuresis alarm and placebo compared to enuresis alarm and desmopressin - Clinical summary of findings

OutcomeAlarm and placeboAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week at the end of treatment2828-MD 1 (0.79 to 1.21)MODERATE
12.2.1.7. Enuresis alarm compared to enuresis alarm with imipramine

One randomised control trial Fournier (1987) 81, compared enuresis alarm alone to enuresis alarm with imipramine.

Table 12-7Enuresis alarm compared to enuresis alarm and imipramine - Clinical summary of findings

OutcomeAlarmAlarm and imipramineRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week at follow-up88-not pooledVERY LOW
12.2.1.8. Enuresis alarm compared to dry bed training with an enuresis alarm

One randomised controlled trial, Bennett (1985) 85 compared enuresis alarm treatment to dry bed training which included the use of an enuresis alarm. Bennett (1985) 85 reported dry bed training to include waking schedule, retention control training, positive practice and cleanliness training.

Table 12-8Enuresis alarm compared to dry bed training and alarm- Clinical summary of findings

OutcomeAlarmDBT and alarmRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights4/9 (44.4%)5/10 (50%)RR 0.89 (0.34 to 2.32)55 fewer per 1000 (from 330 fewer to 660 more)VERY LOW
Mean number of wet nights per week at the end of treatment910-MD −0.4 (−2.09 to 1.29)VERY LOW
Number of children who dropped out9/18 (50%)10/20 (50%)RR 1 (0.53 to 1.89)0 fewer per 1000 (from 235 fewer to 445 more)VERY LOW
12.2.1.9. Enuresis alarm compared to retention control training with an enuresis alarm

Three randomised control trials Fielding (1980) 107, Geffken (1986) 108 and Houts (1986) 100 compared enuresis alarm treatment to retention control treatment with an enuresis alarm.

Table 12-9Enuresis alarm compared to enuresis alarm and retention control training - Clinical summary of findings

OutcomeAlarmAlarm and retention control trainingRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights31/43 (72.1%)37/43 (86%)RR 0.84 (0.68 to 1.04)138 fewer per 1000 (from 275 fewer to 34 more)VERY LOW
Mean change of number of wet nights during treatment (no SDs)2020-not pooledVERY LOW
Mean change of number of wet nights during follow up (no SDs)2020-not pooledVERY LOW
Number of children who relapsed at 6 months5/12 (41.7%)9/19 (47.4%)RR 0.92 (0.42 to 2.02)38 fewer per 1000 (from 275 fewer to 483 more)VERY LOW
Number of children who relapsed at 12 months5/12 (41.7%)10/19 (52.6%)RR 0.82 (0.38 to 1.77)95 fewer per 1000 (from 326 fewer to 405 more)VERY LOW
Number of drop outs by end of trial3/15 (20%)2/15 (13.3%)RR 1.5 (0.29 to 7.73)67 more per 1000 (from 94 fewer to 895 more)VERY LOW
12.2.1.10. Enuresis alarm compared to enuresis alarm plus a star chart

Van Londen (1993) 88, a randomised controlled trial evaluated enuresis alarms compared to two types of star charts in combination with enuresis alarm treatment.

Table 12-10Enuresis alarm compared to enuresis alarm and star charts for correct behaviour - Clinical summary of findings

OutcomeAlarmAlarm and star chart for correct behaviourRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 dry consecutive nights26/36 (72.2%)37/38 (97.4%)RR 0.74 (0.6 to 0.91)253 fewer per 1000 (from 88 fewer to 390 fewer)VERY LOW
Number of relapses at 2.5 years13/26 (50%)10/37 (27%)RR 1.85 (0.96 to 3.56)230 more per 1000 (from 11 fewer to 691 more)VERY LOW

Table 12-11Enuresis alarm compared to enuresis alarm and star charts for dry night - Clinical summary of findings

OutcomeAlarmAlarm and star chart for dry nightRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 dry consecutive nights26/36 (72.2%)33/39 (84.6%)RR 0.85 (0.67 to 1.09)127 fewer per 1000 (from 279 fewer to 76 more)VERY LOW
Number of relapses at 2.5 years13/26 (50%)15/33 (45.5%)RR 1.1 (0.64 to 1.88)46 more per 1000 (from 164 fewer to 400 more)VERY LOW

Table 12-12Enuresis alarm and star chart for correct behaviour compared to enuresis alarm and star charts for dry night - Clinical summary of findings

OutcomeAlarm and star chart for correct behaviourAlarm and star chart for dry nightRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 dry consecutive nights33/39 (84.6%)37/38 (97.4%)RR 0.87 (0.75 to 1)127 fewer per 1000 (from 244 fewer to 0 more)LOW
Number of relapses at 2.5 years15/33 (45.5%)10/37 (27%)RR 1.68 (0.88 to 3.22)184 more per 1000 (from 32 fewer to 599 more)VERY LOW

Study populations- children with bedwetting only

12.2.1.11. Enuresis alarm compared to no treatment for children with bedwetting only

Four randomised control trials evaluated enuresis alarm treatment compared to no treatment, a waiting list group; Lynch (1984) 109, Nawaz (2002) 94, Wagner (1982) 110 and Wagner (1985) 111 for children with bedwetting.

Table 12-13Enuresis alarm compared to no treatment for children with bedwetting - Clinical summary of findings

OutcomeAlarmNo treatmentRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights28/55 (50.9%)3/55 (5.5%)RR 7.35 (2.56 to 21.11)349 more per 1000 (from 86 more to 1000 more)LOW
Mean number of wet nights per week at end of treatment3030-MD −2.78 (−4.42 to −1.14)VERY LOW
Number of children who relapsed at 6 months7/18 (38.9%)2/2 (100%)RR 0.54 (0.24 to 1.19)460 fewer per 1000 (from 760 fewer to 190 more)VERY LOW
Number of drop outs at end of trial2/20 (10%)2/20 (10%)RR 1 (0.16 to 6.42)0 fewer per 1000 (from 84 fewer to 542 more)VERY LOW
12.2.1.12. Pad and bell enuresis alarm compared to body worn enuresis alarm for children with bedwetting only

One randomised control trial Butler (1990) 112, compared effectiveness of two different enuresis alarms, one body worn enuresis alarm and a pad and bell enuresis alarm for children with bedwetting.

Table 12-14Pad and bell enuresis alarm compared to body worn enuresis alarm - Clinical summary of findings

OutcomePad and bell alarmBody worn alarmRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights14/20 (70%)14/20 (70%)RR 1 (0.67 to 1.5)0 fewer per 1000 (from 231 fewer to 350 more)MODERATE
Mean number of wet nights per week at end of treatment (no SDs)1718-not pooledMODERATE
Number of children who relapsed at 6 months4/14 (28.6%)3/14 (21.4%)RR 1.33 (0.36 to 4.9)71 more per 1000 (from 137 fewer to 835 more)MODERATE
Number of drop outs at end of trial3/20 (15%)2/20 (10%)RR 1.5 (0.28 to 8.04)50 more per 1000 (from 72 fewer to 704 more)MODERATE
12.2.1.13. Enuresis alarm compared to desmopressin for children with bedwetting only

Two randomised control trials Ng (2005) 113 and Wille (1986) 114 compared enuresis enuresis alarms to any form of desmopressin in children with bedwetting.

Table 12-15Enuresis alarm compared to desmopressin - Clinical summary of findings

OutcomeAlarmDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights8/35 (22.9%)16/38 (42.1%)RR 0.54 (0.27 to 1.11)194 fewer per 1000 (from 307 fewer to 46 more)LOW
Number of children who achieved 5 wet nights in 28 nights19/22 (86.4%)17/24 (70.8%)RR 1.22 (0.9 to 1.66)156 more per 1000 (from 71 fewer to 467 more)VERY LOW
Mean number of wet nights per week at end of treatment5060-MD −0.46 (−1.53 to 0.62)VERY LOW
Number of children who relapsed at 3 months1/27 (3.7%)19/33 (57.6%)RR 0.09 (0.02 to 0.45)524 fewer per 1000 (from 317 fewer to 564 fewer)LOW
Number of children who dropped out by the end of the trial8/57 (14%)2/62 (3.2%)RR 3.69 (0.95 to 14.34)86 more per 1000 (from 2 fewer to 427 more)VERY LOW
Adverse event - False alarm21/22 (95.5%)0/0 (0%)not poolednot pooledLOW
12.2.1.14. Enuresis alarm compared to imipramine for children with bedwetting only

One randomised controlled trial Wagner (1982) 110 compared enuresis alarm to imipramine (25 mg for children < 32 kg, 50 mg for children > 32k g) for children with bedwetting and was identified.

Table 12-16Enuresis alarm compared to imipramine for children with bedwetting - Clinical summary of findings

OutcomeAlarmImipramineRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights10/12 (83.3%)4/12 (33.3%)RR 2.5 (1.08 to 5.79)500 more per 1000 (from 27 more to 1000 more)VERY LOW
Mean number of wet nights per week at end of treatment (no SDs)1212-not pooledVERY LOW
Number of children who relapsed at 6 months5/10 (50%)4/4 (100%)RR 0.56 (0.29 to 1.07)440 fewer per 1000 (from 710 fewer to 70 more)VERY LOW
12.2.1.15. Enuresis alarm compared to enuresis alarm with desmopressin for children with bedwetting only

One randomised controlled trial Ng (2005) 113 compared enuresis alarms to enuresis alarms with desmopressin for children bedwetting only

Table 12-17Enuresis alarm compared to enuresis alarm and desmopressin for children with bedwetting- Clinical summary of findings

OutcomeAlarmAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights8/35 (22.9%)20/32 (62.5%)RR 0.37 (0.19 to 0.71)394 fewer per 1000 (from 181 fewer to 506 fewer)MODERATE
Mean number of wet nights per week at the end of treatment2829-MD 1.5 (0.43 to 2.57)LOW
Number of children who relapsed at 3 months0/8 (0%)7/20 (35%)RR 0.16 (0.01 to 2.44)294 fewer per 1000 (from 346 fewer to 504 more)LOW
Number of children who dropped out by the end of the trial7/35 (20%)3/32 (9.4%)RR 2.13 (0.6 to 7.56)106 more per 1000 (from 38 fewer to 617 more)LOW
12.2.1.16. Enuresis alarm compared to dry bed training with an enuresis alarm for children with bedwetting

One randomised controlled trial Nawaz (2002) 94 compared enuresis alarm treatment to dry bed training which included the use of an enuresis alarm for children with bedwetting. Nawaz (2002) 94 reported dry bed training to include waking schedule, retention control training, positive practice and cleanliness training.

Table 12-18Enuresis alarm compared to dry bed training for children with bedwetting - Clinical summary of findings

OutcomeAlarmDBTRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights3/12 (25%)8/12 (66.7%)RR 0.38 (0.13 to 1.08)414 fewer per 1000 (from 580 fewer to 53 more)LOW
Mean number of wet nights per week at the end of treatment1212-MD 2.42 (0.71 to 4.13)LOW
Number of children who relapsed at 6 months1/3 (33.3%)1/8 (12.5%)RR 2.67 (0.23 to 30.4)209 more per 1000 (from 96 fewer to 1000 more)LOW
12.2.1.17. Enuresis alarm compared to retention control training with an enuresis alarm for children with bedwetting

One randomised control trial Fielding (1980) 107 compared enuresis alarm treatment to retention control treatment which included an enuresis alarm for children with bedwetting.

Table 12-19Enuresis alarm compared to enuresis alarm and retention control training for children with bedwetting - Clinical summary of findings

OutcomeAlarmAlarm and retention control trainingRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights14/17 (82.4%)11/16 (68.8%)RR 1.2 (0.81 to 1.78)138 more per 1000 (from 131 fewer to 537 more)VERY LOW
Number of children who relapsed at 6 months5/14 (35.7%)3/11 (27.3%)RR 1.31 (0.4 to 4.32)85 more per 1000 (from 164 fewer to 906 more)VERY LOW
Number of children who relapsed at 12 months8/14 (57.1%)4/11 (36.4%)RR 1.57 (0.64 to 3.88)207 more per 1000 (from 131 fewer to 1000 more)VERY LOW

Study population- children with monosymptomatic nocturnal enuresis

12.2.1.18. Enuresis alarm compared to desmopressin for children with monosymptomatic nocturnal enuresis

Two randomised control trials Longstaffe (2000) 115 and Tuygun (2007) 116 compared enuresis alarms to desmopressin, Tuygun (2007) 116. Both studies considered children with monosymptomatic nocturnal enuresis.

Table 12-20Enuresis alarm compared to desmopressin for children with monosymptomatic nocturnal enuresis - Clinical summary of findings

OutcomeAlarmDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive or a 90% improvement in the number of dry nights55/96 (57.3%)54/109 (49.5%)RR 1.16 (0.89 to 1.5)79 more per 1000 (from 54 fewer to 248 more)VERY LOW
50%–90% reduction in number of wet nights at end of treatment9/35 (25.7%)15/49 (30.6%)RR 0.84 (0.42 to 1.7)49 fewer per 1000 (from 177 fewer to 214 more)VERY LOW
Mean number of wet nights per month at end of treatment3549-MD −7.29 (−11.27 to −3.31)VERY LOW
Number of children relapsed at 6 months10/35 (28.6%)27/49 (55.1%)RR 0.52 (0.29 to 0.93)264 fewer per 1000 (from 39 fewer to 391 fewer)VERY LOW
Number of children who dropped out of the trial8/61 (13.1%)5/60 (8.3%)RR 1.57 (0.55 to 4.54)47 more per 1000 (from 37 fewer to 294 more)LOW
12.2.1.19. Enuresis alarm compared to enuresis alarm with desmopressin for children with monosymptomatic nocturnal enuresis

One randomised controlled trial Ozden (2008) 117 compared enuresis alarms to enuresis alarms with desmopressin for children with monosymptomatic nocturnal enuresis and was identified in the update search.

Table 12-21Enuresis alarm compared to enuresis alarm with desmopressin for children with monosymptomatic nocturnal enuresis - Clinical summary of findings

OutcomeAlarmAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved at least 75% reduction in the number of wet nights7/22 (31.8%)6/30 (20%)RR 1.59 (0.62 to 4.08)118 more per 1000 (from 76 fewer to 616 more)VERY LOW
Mean number of wet nights per week at the end of treatment2230-MD 0.5 (0.19 to 0.81)VERY LOW
Number of children who dropped out by the end of the trial5/22 (22.7%)3/30 (10%)RR 2.27 (0.61 to 8.52)127 more per 1000 (from 39 fewer to 752 more)VERY LOW

Study populations- children with severe wetting

12.2.1.20. Enuresis alarm compared to no treatment for children with severe wetting

One randomised controlled trial Ronen (1992) 90 compared enuresis alarms to no treatment for children with severe wetting.

Table 12-22Enuresis alarm compared to no treatment for children with severe wetting - Clinical summary of findings

OutcomeAlarmNo treatmentRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights12/19 (63.2%)0/18 (0%)RR 23.75 (1.51 to 373.78)0 more per 1000 (from 0 more to 0 more)VERY LOW
Mean number of wet nights per 3 weeks at the end of treatment1918-MD −15.99 (−20.78 to − 11.2)LOW
Number of drop outs at end of trial4/19 (21.1%)2/18 (11.1%)RR 1.89 (0.39 to 9.11)99 more per 1000 (from 68 fewer to 900 more)VERY LOW
12.2.1.21. Enuresis alarm compared to enuresis alarm with intranasal desmopressin for children with severe wetting

One randomised controlled trial Bradbury (1995) 105 compared enuresis alarms to enuresis alarms with 40 mcg intranasal desmopressin for children with severe wetting.

Table 12-23Enuresis alarm compared to enuresis alarm and desmopressin for children with severe wetting - Clinical summary of findings

OutcomeAlarmAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 4 consecutive dry weeks6/19 (31.6%)14/21 (66.7%)RR 0.47 (0.23 to 0.98)354 fewer per 1000 (from 13 fewer to 514 fewer)LOW
Mean number of wet nights per week at end of treatment (no SDs)1921-not pooledLOW
Number of children relapsed at 6 months2/19 (10.5%)2/21 (9.5%)RR 1.11 (0.17 to 7.09)10 more per 1000 (from 79 fewer to 579 more)LOW
12.2.1.22. Enuresis alarm and placebo compared to enuresis alarm with desmopressin for children with severe wetting

One randomised controlled trial Leebeek (2001) 118 compared enuresis alarms and placebo to enuresis alarms with desmopressin for children with bedwetting.

Table 12-24Enuresis alarm and placebo compared to enuresis alarm and desmopressin for children with bedwetting - Clinical summary of findings

OutcomeAlarm and placeboAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who had greater than 90% improvement in the mean number of wet nights per week at the end of treatment18/38 (47.4%)15/43 (34.9%)RR 1.36 (0.8 to 2.3)126 more per 1000 (from 70 fewer to 454 more)LOW
Number of children who had a 90% improvement in the number of dry nights at 6 month follow up17/37 (45.9%)17/41 (41.5%)RR 1.11 (0.67 to 1.84)46 more per 1000 (from 137 fewer to 349 more)LOW
Mean number of wet nights per week at the end of treatment3943-not pooledLOW

Study population- children with family and behavioural difficulties

12.2.1.23. Enuresis alarm compared to enuresis alarm with intranasal desmopressin for children with family and behavioural problems

One randomised controlled trial Bradbury (1995) 105 compared enuresis alarms to enuresis alarms with 40 mcg intranasal desmopressin for children with family and behavioural problems.

Table 12-25Enuresis alarm compared to enuresis alarm and desmopressin for children with family and behavioural problems - Clinical summary of findings

OutcomeAlarmAlarm and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 4 consecutive dry weeks4/14 (28.6%)13/16 (81.3%)RR 0.35 (0.15 to 0.83)528 fewer per 1000 (from 138 fewer to 691 fewer)LOW
Mean number of wet nights per week at end of treatment (no SDs)1416-not pooledLOW
Number of children relapsed at 6 months2/14 (14.3%)2/16 (12.5%)RR 1.14 (0.18 to 7.08)17 more per 1000 (from 102 fewer to 760 more)LOW

Study population-children with hearing impairment

12.2.1.24. Light enuresis alarm for children with hearing impairment with nocturnal enuresis

One observational study, Baller (1970) 119 considered light enuresis alarms for children with hearing impairment with nocturnal enuresis. Children were treated with a pad and bell device with a light which had a cone shaped shade to shine the light directly at the child’s face. Children were given an explanation of the treatment by a consultant.

12.2.2. Network Meta-Analysis

Enuresis alarms were amongst the interventions included in a network meta-analyses of interventions used for nocturnal enuresis. The summary of results of this analysis is presented in chapter 24 and a detailed description of the analysis is presented in appendix F. If studies did not meet the inclusion criteria of the network meta-analysis protocol as stated in appendix F they were not included in the network meta-analysis.

12.2.3. Health economic evidence review

Given the lack of published evidence assessing the cost-effectiveness of different interventions, including enuresis alarms, used in the treatment of bedwetting, the GDG identified this area as high priority for original economic analysis. Therefore, a cost-utility analysis was undertaken where costs and quality-adjusted life-years (QALYs) were considered from a UK National Health Service and Personal Social Services perspective. The time horizon for the analysis was 13 years, modelling patients from the time they entered at age 7 years until they reached age 20.

A summary of the analysis is provided below. The full report is presented in appendix G.

Summary of results

The results of the probabilistic sensitivity analysis are summarised in table 12-26 in terms of mean total costs and mean total QALYs and mean net benefit for each treatment sequence, where each mean is the average of 20,000 simulated estimates. The option with the greatest mean net benefit is the most cost-effective at a specified threshold (for example, £20,000). The percentage of simulations where each strategy was the most cost-effective gives an indication of the strength of evidence in favour of that strategy being cost-effective.

Table 12-26. Basecase probabilistic sensitivity analysis results.

Table 12-26

Basecase probabilistic sensitivity analysis results.

The results of the incremental analysis in the probabilistic analysis, excluding dominated and extendedly dominated strategies, are presented in table 12-27.

Table 12-27. Incremental analysis of basecase probabilistic results with dominated and extendedly dominated sequences removed.

Table 12-27

Incremental analysis of basecase probabilistic results with dominated and extendedly dominated sequences removed.

The GDG considered that the differences between intervention sequences were relatively small and the probabilistic results indicated substantial uncertainty around the mean cost and benefit estimates. Small changes to the model inputs appears to result in substantial changes to the conclusions about modelled sequences’ relative and overall cost-effectiveness.

A series of sensitivity analyses were undertaken to test some of the assumptions feeding into the model and none of these affected the cost-effectiveness of the sequence alarm followed by combined alarm and desmopressin and then desmopressin alone compared to no treatment.

The economic analysis conducted and presented here represents the first undertaken to assess the cost-effectiveness of interventions used in the treatment of children with bedwetting. And although the analysis is directly applicable to decision making in the UK NHS, it has some potentially serious limitations, some of which may significantly impact the overall conclusions that can be drawn. The main limitations of the analysis are related to the fact that assumptions had to be made in the absence of evidence. Some of these key assumptions centre around:

  • treatment effectiveness being independent of age
  • health care resource use having been estimated by GDG
  • utility weights having been estimated by GDG

A full discussion of these can be found in appendix G.

12.2.4. Evidence statements

Studies which included children with bedwetting and possible daytime symptoms

Enuresis alarm compared to no treatment
Baker (1969) 23, Bollard (1981) 93, Bollard (1982) 99, Houts (1986) 100, Jehu (1977) 101, Moffatt (1987) 102
  • Six studies showed that more children achieved 14 consecutive dry nights with enuresis alarm treatment than with no treatment. Relative risk 16.9, 95% CI 7.17, 39.85. Children had an age range of 8.1 to 10.05 years and the length of treatment ranged between 10 to 20 weeks.
Bollard (1982) 99
  • One study showed that children treated with an enuresis alarm had 3.8 fewer wet nights in the final week of treatment compared to those who had no treatment. Children had a mean age of 8.6 to 9.7 years and the length of treatment was 20 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Houts (1986) 100, Jehu (1977) 101
  • Two studies showed there was no statistically significant difference in the number of children who dropped out of the trial when placed in the enuresis alarm treatment group compared to the no treatment group. Relative risk 4.16, 95% CI 0.5, 34.6. Children had a mean age of 8.35 to 10.05 years and the length of treatment was 12 to 18 weeks.
Unsupervised enuresis alarm compared to supervised enuresis alarm
Bollard (1981) 93
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights with a supervised enuresis alarm (weekly telephone contact with parent) than with an unsupervised enuresis alarm. Relative risk 1.33, 95% CI 0.82, 2.16. Children had a mean age of 9 years and 8 months and the length of treatment was 20 weeks.
  • One study reported that children treated with a supervised enuresis alarm had 0.4 fewer wet nights in the final week of treatment compared to those who treatment with an unsupervised enuresis alarm. Children had a mean age of 9 years and 8 months and the length of treatment was 20 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Enuresis alarm compared to other single treatments
Kolvin (1972) 103
  • One study showed there was no statistically significant difference in the number of children who had an 80% improvement in the number of dry when treated with imipramine compared to enuresis alarm treatment. Relative risk 1.16, 95% CI 0.71, 1.89. Children had a mean age of 9 years and 4 months and the length of treatment was 2 months. (Kolvin (1972)103 did not state the dose of imipramine given to children)
  • One study showed that children treated with an enuresis alarm had 1.1 fewer wet nights per week at follow up compared to those treated with imipramine. Children had a mean age of 9 years and 4 months and the length of treatment was 2 months. (Kolvin (1972) 103 did not state the dose of imipramine given to children). No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Fournier (1987) 81, Kolvin (1972) 103
  • Two studies evaluated the number of wet nights in the final week of treatment, one study showed no difference and one showed children treated with imipramine had 0.4 fewer wet nights than those treated with an enuresis alarm. Children in Kolvin (1972)103 had a mean age of 9 years and 4 months and the length of treatment was 2 months, children in Fournier (1987) (Fournier et al. 849–53) had a mean age of 8.5 years and the length of treatment was 6 weeks. Fournier (1987) (Fournier et al. 849–53) gave 25 mg imipramine to children, Kolvin (1972) (Kolvin et al. 715–26) did not state the dose of imipramine given to children. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Danquah (1975) 104
  • One study showed that children treated with an enuresis alarms had 0.8 fewer wet nights in the final week of treatment compared to those treated with amitriptyline. Children had a mean age of 10.4 years and the length of treatment was 7 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Enuresis alarm compared to enuresis alarm plus star charts
van Londen (1993) 88
  • One study showed that more children achieved 14 consecutive dry nights with enuresis alarm plus a star chart with rewards for correct behaviour (for waking up to the enuresis alarm within 3 mins, going to the toilet after, returning to bed and resetting the enuresis alarm) and returning a sticker if correct behaviour not demonstrated than with an enuresis alarm alone. Relative risk 0.74, 95% CI 0.6, 0.91. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 2.5 years in children treated with enuresis alarm plus a star chart with rewards for correct behaviour (for waking up to the enuresis alarm within 3 mins, going to the toilet after, returning to bed and resetting the enuresis alarm) and returning a sticker if correct behaviour not demonstrated than an enuresis alarm alone. Relative risk 1.85, 95% CI 0.96, 3.56. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights between treated with an enuresis alarm and children treated with an enuresis alarm plus a star chart with reward for a dry night and returning a sticker for a wet night. Relative risk 0.85, 95% CI 0.67, 1.09. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 2.5 years between children treated with an enuresis alarm and children treated with an enuresis alarm plus a star chart with reward for a dry night and returning a sticker for a wet night. Relative risk 1.1, 95% CI 0.64, 1.88. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
  • One study showed that more children achieved 14 consecutive dry nights with an enuresis alarm plus a star chart with rewards for correct behaviour (for waking up to the enuresis alarm within 3 months, going to the toilet after, returning to bed and resetting the enuresis alarm) and returning a sticker if correct behaviour was not demonstrated than with an enuresis alarm plus a star chart with reward for a dry night and returning a sticker for wet a night. Relative risk 0.87, 95% CI 0.75, 1. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 2.5 years between children treated with an enuresis alarm plus a star chart with rewards for correct behaviour (for waking up to the enuresis alarm within 3 months, going to the toilet after, returning to bed and resetting the enuresis alarm) and returning a sticker if correct behaviour not demonstrated and children treated with an enuresis alarm plus a star chart with reward for a dry night and returning a sticker for a wet night. Relative risk 1.68, 95% CI 0.88, 3.22. Children had a mean age of 8.6 years and the length of treatment was 20 weeks.
Enuresis alarm compared to enuresis alarm in combination with another treatment
Bradbury (1995) 105
  • One study showed there was no statistically significant difference in the number of children achieving 14 consecutive dry nights with enuresis alarm treatment than with 40 mcg intranasal desmopressin and enuresis alarm treatment. Relative risk 0.72, 95% CI 0.51, 1.03. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
  • One study showed that children treated with 40 mcg intranasal desmopressin and enuresis alarm had 1.3 fewer wet nights in the final week of treatment compared to those who had enuresis alarm alone treatment. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no statistically significant difference in relapse at 6 months in children when treated with an enuresis alarm compared to enuresis alarm and 40 mcg intranasal desmopressin. Relative risk 1.27, 95% CI 0.32, 4.95. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial when place in the 40 mcg intranasal desmopressin and enuresis alarm treatment group compared to the enuresis alarm alone treatment group. Relative risk 5.14, 95% CI 0.26, 103.37. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
Sukhai (1989) 106
  • One study showed children treated with an enuresis alarm and desmopressin had 1 fewer wet night per week at the end of treatment compared to children treated with an enuresis alarm and placebo. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable. Children had a mean age of 11 years and the length of treatment was 2 weeks.
Fournier (1987) 81
  • One study showed children treated with an imipramine and enuresis alarm had 1.5 fewer wet nights in the final week of treatment compared to those who had enuresis alarm alone treatment. Children had a mean age of 8.5 years and the length of treatment was 6 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Bennett (1985) 85
  • One study showed there was no statistically significant difference in the number of children that achieved 14 consecutive dry nights with an enuresis alarm alone than with dry bed training and enuresis alarm treatment. Relative risk 0.89, 95% CI 0.34, 2.32. Children had a mean age of 8.5 years and had 12 weeks of treatment.
  • One study showed there was no statistically significant difference in the mean number of wet nights per week at the end of treatment between children treated with an enuresis alarms and children treated with dry bed training and an enuresis alarm. Mean difference −0.4, 95% CI −2.09, 1.29.
  • One study showed there was no difference in the number of children who dropped out between children treated with an enuresis alarms and children treated with dry bed training and an enuresis alarm. Relative risk 1, 95% CI 0.53, 1.89.
Fielding (1980) 107, Geffken (1986) 108, Houts (1986) 100
  • Three studies (1 of which had 2 subgroups) showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights with an enuresis alarm alone treatment than with retention control training and an enuresis alarm treatment. Relative risk 0.84, 95% CI 0.68, 1.04. Children in Fielding (1980) had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks. Children in Geffken (1986) had an age range of 5 to 13 years and the length of treatment was 14 weeks; children in Houts (1986) (Houts, Peterson, and Whelan 462–69) had a mean age of 8.35 to 9.06 years and the length of treatment was 16 weeks.
Geffken (1986) 108
  • One study showed that children treated with retention control training and an enuresis alarm had 0.3 and 0.4 fewer wet nights in the final week of treatment compared to those who had an enuresis alarm alone. Children had an age range of 5 to 13 years and the length of treatment was 14 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study (which had 2 subgroups) showed that children treated with retention control training and an enuresis alarm had 1.5 and 0.4 fewer wet nights at follow up compared to those who had an enuresis alarm alone. Children had an age range of 5 to 13 years and the length of treatment was 14 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Fielding (1980) 107 Houts (1986) 100
  • Two studies showed there was no statistically significant difference in the number of children who relapsed at 6 months between the group treated with a retention control training and an enuresis alarm and those treated with an enuresis alarm alone. Relative risk 0.92, 95% CI 0.42, 2.02. Children in Fielding (1980) had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks. Children in Houts (1986) had a mean age of 8.35 to 9.06 years and the length of treatment was 16 weeks.
  • Two studies showed there was no statistically significant difference in the number of children who relapsed at 12 months between the group treated with a retention control training and an enuresis alarm and those treated with an enuresis alarm alone. Relative risk 0.82, 95% CI 0.38, 1.77. Children in Fielding (1980) had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks. Children in Houts (1986) had a mean age of 8.35 to 9.06 years and the length of treatment was 16 weeks.
Houts (1986) (Houts, Peterson, and Whelan 462–69)
  • One study showed there was no statistically significant difference in children who dropped out of the trial when placed in the enuresis alarm treatment group compared to the retention control training and enuresis alarm group. Relative risk 1.5, 95% CI 0.29, 7.73. Children had a mean age of 8.35 to 9.06 years and the length of treatment was 16 weeks.

Studies included children with bedwetting only

Enuresis alarm compared to no treatment
Lynch (1984) 109, Nawaz (2002) 94, Wagner (1982) 110, Wagner (1985) 111
  • Four studies showed that more children achieved 14 consecutive dry nights with an enuresis alarm treatment than with no treatment. Relative risk 7.35, 95% CI 2.56, 21.11. Children had a mean age of 7.9 to 9.93 years and the length of treatment was 10 to 16 weeks.
Lynch (1984) 109, Nawaz (2002) 94
  • Two studies showed that children treated with an enuresis alarm had fewer wet nights in the final week of treatment compared to those who had no treatment. Mean difference −2.78, 95% CI −4.42, −1.14. Children in Lynch (1984) had an age range of 5 to 12 years and length of treatment was 10 weeks; children in Nawaz (2002) had a mean age of 9.84 and 9.93 years and the length of treatment was 6 weeks.
Wagner (1982) 110, Wagner (1985) 111
  • Two studies showed there was no statistically significant difference in the number of children who relapsed between children treated with an enuresis alarm and children who had no treatment. Relative risk 0.54, 95% CI 0.24, 1.19. Children had a mean age of 7.9 years and length of treatment was 12 weeks.
Lynch (1984) 109
  • One study showed there was no difference in the number of children who dropped out of the trial when placed in the enuresis alarm treatment group compared to the no treatment group. Relative risk 1, 95% CI 0.16, 6.42. Children had an age range of 5 to 12 years and the length of treatment was 10 weeks.
Pad and bell enuresis alarm compared to body worn enuresis alarm
Butler (1990) 112
  • One study showed there was no difference in the number of children who achieved 14 consecutive dry nights between children treated with a body worn enuresis alarm and children treated with a pad and bell enuresis alarm. Relative risk 1, 95% CI 0.67, 1.5. Children had a mean age of 8.11 to 10.6 years and the length of treatment was 16 weeks.
  • One study showed children treated with a body worn enuresis alarm had 0.2 fewer wet nights than those treated with a pad and bell enuresis alarm. Children had a mean age of 8.11 to 10.6 years and the length of treatment was 16 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 6 months between the group treated with a body worn enuresis alarm and those treated with a pad and bell enuresis alarm. Relative risk 1.33, 95% CI 0.36, 4.90. Children had a mean age of 8.11 to 10.6 years and the length of treatment was 16 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial when placed in a group treated with a body worn enuresis alarm and those treated with a pad and bell enuresis alarm. Relative risk 1.50, 95% CI 0.28, 8.04. Children had a mean age of 8.11 to 10.6 years and the length of treatment was 16 weeks.
Enuresis alarm compared to single other treatment for children with bedwetting
Ng (2005) 113
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights between children treated with an enuresis alarm and children treated with desmopressin. Relative risk 0.54, 95% CI 0.27, 1.11. Children were aged over 6 years and the length of treatment was 3 months. Ng (2005) considered 0.2 mg tablet desmopressin.
Wille (1986) 114
  • One study showed there was no statistically significant difference in the number of children who achieved only 5 wet nights in 28 nights between children treated with an enuresis alarm and children treated with desmopressin. Relative risk 1.22, 95% CI 0.9, 1.66. Children were aged over 6 years and the length of treatment was 3 months. Wille (1986) considered 200 micro
  • One study showed that there was a 78% rate of false enuresis alarms during the trial. Children were aged over 6 years and the length of treatment was 3 months. Wille (1986) considered 200 micro grams intranasal desmopressin.
Ng (2005) 113, Wille (1986) 114
  • Two studies showed there was no statistically significant difference in the number of wet nights in the final week of treatment of those treated with an enuresis alarm compared to those treated with desmopressin. Mean difference −0.46, 95% CI −1.53, 0.62. Children were aged over 6 years and the length of treatment was 3 months. Ng (2005) considered 0.2 mg tablet desmopressin and Wille (1986) considered 200 micro grams intranasal desmopressin.
  • One study, Wille (1986), showed that children treated with desmopressin had a faster response compared to children treated with an enuresis alarm. Wille (1986) considered a response to be the number of dry nights.
  • One study, Ng (2005), showed that children treated with an enuresis alarm had a faster response compared to children treated with desmopressin. Ng (2005) considered a response to be a reduction in the number of wet nights.
  • Two studies showed after treatment children treated with an enuresis alarm had a continued higher response compared to children treated with desmopressin. Ng (2005) considered a response to be a reduction in the number of wet nights and Wille (1986) considered a response to be the number of dry nights. Children were aged over 6 years and treatment was for 3 months. Ng (2005) considered 0.2 mg tablet desmopressin and Wille (1986) considered 200 micro grams intranasal desmopressin.
  • Two studies showed children treated with desmopressin were more likely to relapse at 3 months compared to children treated with an enuresis alarms. Relative risk 0.09, 95% CI 0.02, 0.45. Children had a mean age of 9.5 years and the length of treatment was 12 weeks. Ng (2005) considered 0.2 mg tablet desmopressin.
  • Two studies showed there was no statistically significant difference in the number of children who dropped out of the trial when placed in the enuresis alarm treatment group compared to the desmopressin treatment group. Relative risk 3.69, 95% CI 0.95, 14.33. Children were aged over 6 years and the length of treatment was 3 months. Ng (2005) considered 0.2 mg tablet desmopressin and Wille (1986) considered 200 micro grams intranasal desmopressin.
Wagner (1982) 110
  • One study showed children treated with an enuresis alarm were more likely to achieve 14 consecutive dry nights compared to children treated with imipramine treatment. Relative risk 2.5, 95% CI 1.08, 5.79. Children had a mean age of 7.9 years and the length of treatment was 14 weeks. Wagner (1982) gave 25 mg imipramine for children < 32 kg, 50 mg imipramine for children > 32k g.
  • One study showed that children treated with an enuresis alarm had 2.17 fewer wet nights in the final week of treatment than those treated with imipramine. Children had a mean age of 7.9 years and the length of treatment was 14 weeks. Wagner (1982) gave 25 mg imipramine for children < 32 kg, 50 mg imipramine for children > 32k g. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no statistically significant difference in the number of children relapsing at 6 months when treated with an enuresis alarm compared to imipramine. Relative risk 0.56, 95% CI 0.29, 1.07. Children had a mean age of 7.9 years and the length of treatment was 14 weeks. Wagner (1982) gave 25 mg imipramine for children < 32 kg, 50 mg imipramine for children > 32k g.
Enuresis alarm compared to enuresis alarm in combination with other treatments
Ng (2005) 113
  • One study showed more children treated with an enuresis alarm and desmopressin achieved 14 consecutive dry nights compared to those treated with an enuresis alarm treatment. Relative risk 0.37, 95% CI 0.19, 0.71. Children had a mean age of 9.5 years and the length of treatment was 12 weeks. Ng (2005) considered 0.2 mg tablet desmopressin.
  • One study showed children treated with an enuresis alarm and desmopressin had fewer wet nights per week at the end of treatment compared to children treated with enuresis alarm alone. Mean difference 1.5, 95% CI 0.43, 2.57. Children had a mean age of 9.5 years and the length of treatment was 12 weeks. Ng (2005) considered 0.2 mg tablet desmopressin.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 3 months between children treated with an enuresis alarm and children treated with an enuresis alarm and desmopressin. Relative risk 0.16, 95% CI 0.01, 2.44. Children had a mean age of 9.5 years and the length of treatment was 12 weeks. Ng (2005) considered 0.2 mg tablet desmopressin.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial when place in the desmopressin and enuresis alarm treatment group compared to the enuresis alarm alone treatment group. Relative risk 2.13, 95% CI 0.6, 7.56 Children had a mean age of 9.5 years and the length of treatment was 12 weeks. Ng (2005) considered 0.2 mg tablet desmopressin.
Nawaz (2002) 94
  • One study showed there was no statistically significant difference in the number of children that achieved 14 consecutive dry nights with enuresis alarm alone than with dry bed training and enuresis alarm treatment. Relative risk 0.38, 95% CI 0.13, 1.08. Children had a mean age of 9.93 years and the length of treatment was 16 weeks.
  • One study showed children treated with dry bed training and an enuresis alarm had fewer wet nights per week at the end of treatment compared to children treated with an enuresis alarm. Mean difference 2.42, 95% CI 0.71, 4.13. Children had a mean age of 9.93 years and the length of treatment was 16 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 6 months in children treated with dry bed training with an enuresis alarm compared to an enuresis alarm alone. Relative risk 2.67, 95% CI 0.23, 30.4. Children had a mean age of 9.93 years and the length of treatment was 16 weeks.
Fielding (1980) 107
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights with an enuresis alarm alone treatment than with retention control training and an enuresis alarm treatment. Relative risk 1.2, 95% CI 0.81, 1.78. Children had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 6 months between the group treated with a retention control training and an enuresis alarm and those treated with an enuresis alarm alone. Relative risk 1.31, 95% CI 0.4, 4.32. Children had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 12 months between the groups treated with a retention control training and an enuresis alarm and those treated with an enuresis alarm alone. Relative risk 1.57, 95% CI 0.64, 3.88. Children had a mean age of 7.96 to 9.08 years and the length of treatment was 14 weeks.

Studies included children with monosymptomatic nocturnal enuresis

Enuresis alarm compared to desmopressin for children
Longstaffe (2000) 115, Tuygun (2007) 116
  • Two studies showed there was no statistically significant difference in the number of children achieved 14 consecutive dry nights with desmopressin than with enuresis alarm treatment. Relative risk 1.16, 95% CI 0.89, 1.5. Children were aged over 7 years and the length of treatment was 3 to 6 months. Longstaffe (2000) considered 200 micro grams intranasal desmopressin, and Tuygun (2007) considered 20 to 40 micro grams intranasal desmopressin or 0.2 to 0.4 mg tablet desmopressin.
Tuygun (2007) 116
  • One study showed there was no statistically significant difference in the number of children who had a 50 to 90% improvement in the number of dry nights when treated with desmopressin compared to enuresis alarm treatment. Relative risk 0.84, 95% CI 0.42, 1.7. Children had a median age of 8 years and the length of treatment was 3 months. Tuygun (2007) considered 20 to 40 micro grams intranasal desmopressin or 0.2 to 0.4 mg tablet desmopressin.
  • One study showed that children treated with an enuresis alarm had fewer wet nights in the final month of treatment compared to those in the desmopressin group. Mean difference −7.29, 95% CI −11.27, −3.31. Children had a median age of 8 years and the length of treatment was 3 months. Tuygun (2007) considered 20 to 40 micro grams intranasal desmopressin or 0.2 to 0.4 mg tablet desmopressin.
  • One study showed that fewer children relapse at 6 months when treated during 3 months with an enuresis alarm compared to desmopressin. Relative risk 0.52, 95% CI 0.29, 0.93. Children had a median age of 8 years and the length of treatment was 3 months. Tuygun (2007) considered 20 to 40 micro grams intranasal desmopressin or 0.2 to 0.4 mg tablet desmopressin.
Longstaffe (2000) 115
  • One study showed that giving children treatment for nocturnal enuresis improved their psychological scores in both treatment groups. Children were age over 7 years and the length of treatment was 6 months. Longstaffe (2000) considered 200 micro grams intranasal desmopressin.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial when placed in the enuresis alarm treatment group compared to the desmopressin treatment group. Relative risk 1.57, 95% CI 0.55, 4.54. Children were age over 7 years and the length of treatment was 6 months. Longstaffe (2000) considered 200 micro grams intranasal desmopressin.
Enuresis alarm compared to enuresis alarm with desmopressin
Ozden (2008) 117
  • One study showed there was no statistically significant difference in the number of children who achieved a greater than 75% reduction in the number of wet nights between the children treated with desmopressin and enuresis alarm and those who had enuresis alarm alone treatment. Relative risk 1.59, 95% CI 0.62, 4.08. Children had a mean age of 10.1 years and the length of treatment was 6 weeks.
  • One study showed children treated with an enuresis alarm and desmopressin had fewer wet nights per week at the end of treatment compared to children treated with an enuresis alarm. Mean difference 0.5, 95% CI 0.19, 0.81. Children had a mean age of 10.1 years and the length of treatment was 6 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial when placed in the desmopressin and enuresis alarm treatment group compared to the enuresis alarm alone treatment group. Relative risk 2.27, 95% CI 0.61, 8.52. Children had a mean age of 10.1 years and the length of treatment was 6 weeks.

Studies included children with severe wetting

Enuresis alarm compared to no treatment for children
Ronen (1992) (Ronen, Wozner, and Rahav 1–14)
  • One study showed that more children achieved 14 consecutive dry nights an enuresis alarm compared to children who had no treatment. Relative risk 23.75, 95% CI 1.51, 373.78. Children had a mean age of 10.5 (SD 2.28) years and the length of treatment was 3 weeks.
  • One study showed that children treated with an enuresis alarm had fewer wet nights per 3 weeks at the end of treatment compared to those who had no treatment. Mean difference −15.99, 95% CI −20.78, −11.2. Children had a mean age of 10.5 (SD 2.28) years and the length of treatment was 3 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out between children treated with an enuresis alarm and children who had no treatment. Relative risk 1.89, 95% CI 0.39, 9.11. Children had a mean age of 10.5 (SD 2.28) years and the length of treatment was 3 weeks.
Enuresis alarm compared to enuresis alarm with intranasal desmopressin
Bradbury (1995) 105
  • One study showed that more children achieved 14 consecutive dry nights with 40 mcg intranasal desmopressin and enuresis alarm treatment than with enuresis alarm treatment. Relative risk 0.47, 95% CI 0.23, 0.98. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
  • One study showed that children treated with 40 mcg intranasal desmopressin and an enuresis alarm had 2 fewer wet nights in the final week of treatment compared to those who had enuresis alarm alone treatment. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 6 months in children treated with 40 mcg intranasal desmopressin and enuresis alarm compared to an enuresis alarm alone. Relative risk 1.11, 95% CI 0.17, 7.09. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
Enuresis alarm and placebo compared to enuresis alarm with desmopressin
Leebeek (2001) 118
  • One study showed there was no statistically significant difference in the number of children who had a 90% improvement in the number of dry nights between children treated with an enuresis alarm and placebo and children treated with enuresis alarm and desmopressin. Relative risk 1.36, 95% CI 0.8, 2.3. Children had an age range of 6 to 14 years and the length of treatment was 6 weeks.
  • One study showed there was no statistically significant difference in the number of children who had a 90% improvement in the number of dry nights at 6 month followup between children treated with an enuresis alarm and placebo and children treated with an enuresis alarm and desmopressin. Relative risk 1.11, 95% CI 0.67, 1.84. Children had an age range of 6 to 14 years and the length of treatment was 6 weeks.
  • One study showed children treated with an enuresis alarm and placebo had 0.56 fewer wet nights per week compared to children treated with enuresis alarm and desmopressin. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable. Children had an age range of 6 to 14 years and the length of treatment was 6 weeks.

Studies included children with family and behavioural problems

Enuresis alarm compared to enuresis alarm with intranasal desmopressin
Bradbury (1995) 105
  • One study showed that more children achieved 14 consecutive dry nights with 40 mcg intranasal desmopressin and enuresis alarm treatment than with enuresis alarm treatment. Relative risk 0.35, 95% CI 0.15, 0.83. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.
  • One study showed that children treated with 40 mcg intranasal desmopressin and enuresis alarm had 4.5 fewer wet nights in the final week of treatment compared to those who had enuresis alarm alone treatment. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no statistically significant difference in the number of children who relapsed at 6 months in children treated with 40 mcg intranasal desmopressin and enuresis alarm compared to enuresis alarm alone. Relative risk 1.14, 95% CI 0.18, 7.08. Children had a mean age of 9.7 to 10 years and the length of treatment was 6 weeks.

Studies included for children with hearing impairment

Light enuresis alarm for children with hearing impairment
Baller (1970) 119
  • One observational study showed all children 21 treated with the light enuresis alarm gained complete dryness (10 consecutive dry nights) within 30 nights. Children had an age range of 7 to 16 years and had 30 nights of treatment.
  • One observational study showed 1 child relapsed but after 2 more treatments with the light enuresis alarm he gained dryness. Children had an age range of 7 to 16 years and had 30 nights of treatment.

NCGC network meta-analysis (see appendix F)

For children with bedwetting and possible daytime symptoms
  • The NCGC NMA showed there was a statistically significant difference in the number of children who achieved a full response between children treated with alarm and no treatment/placebo. Relative risk 8.601, 95% CI 7.294, 9.103. Children had an age range of 5 to 17 years and treatment for a minimum of 12 weeks.
For children with bedwetting only
  • The NCGC NMA showed there was a statistically significant difference in the number of children who achieved a full response between children treated with alarm and no treatment/placebo. Relative risk 8.601, 95% CI 7.294, 9.103. Children had an age range of 5 to 17 years and treatment for a minimum of 12 weeks.
  • The NCGC NMA showed there was a statistically significant difference in the number of children who experienced a recurrence of bedwetting at 6 months between children treated with alarm and no treatment/placebo. Relative risk 0.0364, 95% CI 0.005, 0.840. Children had an age range of 5 to 17 years and treatment for a minimum of 12 weeks.

For estimates of treatment effect relative to other active comparators, please see section 24.4 in chapter 24.

12.2.5. Health economic evidence statements

NCGC economic evaluation (see appendix G)

  • Alarms are a cost-effective initial intervention even if they need to be replaced at least once during a course of treatment. This evidence has potentially serious limitations and direct applicability.
  • An intervention sequence starting with an alarm (and followed by combined alarm and desmopressin and then by desmopressin alone) is cost-effective in the treatment of children with bedwetting starting at age 5 or 7 years. This evidence has potentially serious limitations and direct applicability.

12.2.6. Evidence to recommendations

Relative values of different outcomes

The GDG considered that sustained dryness was the outcome wished for by children and young people and their parents or carers. This was represented by the outcome of 14 consecutive dry nights to show initial success and indicate the effectiveness of the treatments being evaluated. The mean number of wet nights was also considered by the GDG in evaluating the effectiveness of treatments. Outcomes such as relapse and follow up rates were considered to evaluate sustained dryness.

Trade off between clinical benefit and harms

No evidence was identified of harms of alarm treatment.

Economic Considerations

Enuresis alarms were evaluated as part of original economic modelling undertaken for this guideline and were shown to be a likely cost-effective first line treatment option. The analysis showed that there was considerable uncertainty about which intervention was the most cost-effective first line option, and this was likely caused by the uncertainty around estimates of treatment effectiveness observed in the pairwise and network meta-analyses. The GDG considered that given the substantial uncertainty between interventions, it would be reasonable to recommend first line treatment with an alarm as it was consistently shown to be among less costly and still effective options.

As children and young people who have previously responded to an alarm are likely to respond to it again, it would be a good use of NHS resources to encourage children, young people and parents and carers to retain their alarm and reuse it before trying other options that have associated costs. The economic model assumed that prescribed alarms were given, not loaned, to patients and under this assumption, repeat use of alarms was considered cost-effective. Even if all alarms must be replaced at least once during treatment, they are still considered to be a cost-effective intervention.

Alarms are likely to be considered to be a cost-effective first line treatment regardless of age at initiation.

Quality of evidence (this includes clinical and economic)

The quality of evidence for the outcomes preferred by the GDG was generally low. The individual direct comparisons found in the evidence review were of underpowered studies with small sample sizes. Some studies did not give standard deviations and therefore mean difference and CI could not be calculated giving incomplete evidence.

The GDG considered that the available evidence on alarms compared to no treatment contained inadequate description of the study groups, mainly in terms of the patients’ age and the number of girls included in studies. One study compared supervised alarms to unsupervised alarms; the GDG considered that the type of supervision involved in the studies was not part of common clinical practice in England and Wales.

Other considerations

The GDG considered the direct evidence, the network meta-analysis and the health economic evidence in making their recommendations. They considered that the evidence from the direct comparisons indicated that alarms and desmopressin had similar effects on dryness (both complete dryness and reduced number of wet nights) when receiving treatment but children were more likely to have recurrence of bedwetting following use of desmopressin. In the study that examined monosymptomatic enuresis desmopressin had a faster response (described in Ng (2005) 113 as reduction in the number of wet nights, described in Wille (1986) 114 as the number of dry nights); however alarms had more continued success and children were less likely to experience a recurrence of bedwetting. The GDG considered that when children, young people and families and carers consult with healthcare professionals about bedwetting they are seeking a ‘cure’ from bedwetting. Pharmacological agents have an effect on bedwetting primarily while the child is taking them and do not affect the underlying pathophysiology; alarms develop a conditioned response of waking in response to full bladder which is more likely to continue after alarm treatment.

The evidence of continued success associated with alarms from the evidence review and from professional experience of the GDG was pivotal in the GDG decision to recommend alarms as the choice for first line treatment if the child, young person and family and carers could use alarm. The GDG considered that the first line use of an alarm as a non-pharmacological intervention in children and young people was also perceived as an advantage. The patient carer members specifically highlighted concern about using pharmacological interventions in young children if alternatives are available.

Children with special needs

There was no evidence that one type of alarm was better than another. The GDG considered that if different alarms were available children and families should be given choice. The evidence also indicated that alarms have been used successfully as treatment in children and young people with hearing problems and childrenand young people with behavioural problems. The GDG considered it important that these children and young people do not lose out on a potentially good treatment modaility and where possible, and with the needs of the child, young person and family and carers considered, alarms should be considered as treatment.

Children with and without daytime symptoms

The GDG were interested in whether there was any difference in response to alarm among population groups with different wetting patterns i.e. population who did or did not have daytme wetting. The evidence indicated that bedwetting in children and young people who also had daytime wetting did respond to an alarm and there was no difference in the number achieving 14 consecutive dry nights when alarm was compared with enuresis and desmopressin. The GDG considered it important that children with daytime wetting did have access to alarm treatment and made a recommendation to ensure that alarm is considered as an option for these children and young people. No specific evidence was found regarding treatment of secondary enuresis but the professional experience of the GDG was that these children and young people do respond similarly to children and young people with primary enuresis.

Children and young people who are very infrequent bedwetters will not wet often enough to have the conditioned responses by which an alarm works.

Assessment at 4 weeks

The GDG discussed the lack of evidence for when a child or young person should be assessed after starting treatment. From clinical experience the GDG discussed the benefits of following up early at 4 weeks or less to encourage the patient and report on progress with the treatment. The GDG made a consensus decision on assessment at 4 weeks after starting treatment. In younger children it may be advisable to stop at this stage as child may respond when older and proceeding with treatment for longer at this stage may engender negativity in the child, young person and family and carers about the alarm.

Continue alarm until minimum of 2 weeks uninterrupted dryness has been achieved

The GDG discussed the lack of evidence for how long the alarm should be used. The GDG discussed from clinical experience that to ensure continuing success it was important the child or young person continued to use the alarm until 14 consecutive dry nights was achieved to reduce the chance of experiencing a recurrence of bedwetting after treatment.

Review at 3 months

The GDG considered that it can take several weeks for an alarm to have an effect and it is important to inform and encourage the family or carers to use it for some weeks to get an effect. However a child not getting any response from an alarm should stop the alarm. The GDG considered it not appropriate to continue beyond 3 months if there was not continuing improvement. The use of an alarm can be difficult for a family and an assessment of motivation and ability to continue its use should be made.

Addition of reward systems

The evidence supported the addition of reward systems to alarms and this finding is consistent with psychological theory. The GDG considered that as part of the instructions in how to use the alarm it might be useful to suggest to the family and help them to consider how they will approach this.

Use of alarm in children between 5 and 7

While the GDG considered that children between 5 and 7 years may not require treatment those that do, and are appropriately motivated and mature enough to cope with an alarm, should not be denied use of an alarm by virtue of age alone.

Combination of alarm and pharmacological treatments

The direct evidence indicated that the combination of alarms with desmopressin were similarly effective to alarms alone in the number of wet nights at end of treatment and drop out rates for children with MNE. However relapse rates were inconclusive for children and young people with bedwetting and possible daytime symptoms. The GDG did not consider the evidence supportive of using combination treatment as first line, although the evidence indicated this may be better for children with severe wetting and may also be helpful for children and young people with behavioural difficulties. The evidence comparing alarms to imipramine (two small studies) had contradictory findings (for number of wet nights at the end of treatment). Alarms had fewer wet nights at follow up compared to imipramine. The addition of imipramine to an alarm was not supported by clinical evidence.

12.2.7. Recommendations (on offering and treatment)

12.2.7.1.

Offer an alarm as the first line treatment to children and young people whose bedwetting has not responded to advice on fluids, toileting or an appropriate reward system, unless:

  • an alarm is considered undesirable to the child or young person or their parents and carers or
  • an alarm is considered inappropriate, particularly if:
    -

    the child has very infrequent bedwetting (that is, less than 1–2 wet beds per week)

    -

    the parents or carers are having emotional difficulty coping with the burden of bedwetting

    -

    the parents or carers are expressing anger, negativity or blame towards the child.[1.8.1]

12.2.7.2.

Assess the response to an alarm by 4 weeks and continue with treatment if the child or young person is showing early signs of response10. Stop treatment only if there are no early signs of response.[1.8.2]

12.2.7.3.

Continue alarm treatment in children and young people with bedwetting who are showing signs of response until a minimum of 2 weeks’ uninterrupted dry nights have been achieved [1.8.3].

12.2.7.4.

Assess whether it is appropriate to continue with alarm treatment if complete dryness is not achieved after 3 months. Only continue with alarm treatment if the bedwetting is still improving and the child or young person and parents or carers are motivated to continue. [1.8.4]

12.2.7.5.

Do not exclude alarm treatment as an option for bedwetting in children and young people with:

12.2.7.6.

Consider an alternative type of alarm (for example, a vibrating alarm) for the treatment of bedwetting in children and young people who have a hearing impairment. [1.8.6]

12.2.7.7.

Consider an alarm for the treatment of bedwetting in children and young people with learning difficulties and/or physical disabilities. Tailor the type of alarm to each individual’s needs and abilities.[1.8.7]

12.2.7.8.

Consider an alarm for the treatment of bedwetting in children under 7 years, depending on their ability, maturity, motivation and understanding of the alarm.[1.8.8]

12.2.7.9.

Inform children and young people and their parents or carers about the benefits of alarms combined with reward systems. Advise them to use positive rewards for desired behaviour, such as waking up when the alarm goes off, going to the toilet after the alarm has gone off, returning to bed and resetting the alarm.[1.8.9]

12.2.7.10.

Encourage children and young people with bedwetting and their parents or carers to discuss and agree on their roles and responsibilities for using the alarm and the use of rewards [1.8.10]

12.2.8. Supporting recommendations -evidence to recommendations

Economic considerations

No economic evidence was identified to support these recommendations, however the GDG felt that informing children and their parents/carers of the aims of treatment, how to use the device and how and when they might observe results may encourage adherence to a treatment that may produce only a gradual response. Additionally, offering children and their parents/carers advice on what to do if they experience a recurrence following success may reduce re-presentation at clinics.

Quality of evidence (this includes clinical and economic)

No evidence was identified.

Other considerations

The GDG considered that while alarms may have a sustained effect on dryness an alarm requires considerable effort and perseverance from both child, young person and family, including siblings and extended family. The GDG considered that an important part of considering an alarm was assessing whether the child, young person and family have the necessary motivation, time and energy to use an alarm. Contectual factors such as a new baby in the house, might make an alarm a less attractive first line treatment. The GDG were particularly concerned that in situations where family members are already finding it difficult to cope with bedwetting and where parents or carers may be expressing anger to the child, the introduction of an alarm might result in a more punitive approach to the child.

The GDG considered it important that child, young person and parents or carers were properly informed about how an alarm works and that it may take some weeks for it to have an effect. The GDG also discussed from clinical experience the importance of recording the time child waked and how wet they were as this can be a sign of commitment and allows for positive feedback during follow up clinics.

The GDG considered that the use of an alarm can be difficult for a child, young person and parent or carers to master and that families may need considerable advice and support and access to expertise when starting to use an alarm. The GDG used their experience both as professional and patient members to develop recommendations around the likely information children, young people and parents and carers need when using an alarm. This covered how to use the alarm, what to expect, the signs of response and information about dealing with problems with the alarm and how to return it. Offering and agreeing appropriate support was considered by the GDG to be vital in helping families have confidence in their use of alarm.

12.2.9. Recommendations

12.2.9.1.

Explore and assess the ability of the family to cope with using an alarm for the treatment of bedwetting.[1.4.4]

12.2.9.2.

Consider whether it is appropriate to offer an alarm or drug treatment, depending on the age of the child and young person, the frequency of bedwetting and the motivation and needs of the child and young person and family.[1.4.5]

12.2.9.3.

Ensure that advice and support are available to children and young people and parents or carers who are given an alarm, and agree how these should be obtained. Be aware that they may need a considerable amount of help in learning how to use an alarm.[1.8.11]

12.2.9.4.

Inform the child or young person and their parents or carers that the aims of alarm treatment for bedwetting are to train the child and young person to:

  • recognise the need to pass urine
  • wake to go to the toilet or hold on
  • learn over time to hold on or to wake spontaneously and stop wetting the bed.[1.8.12]
12.2.9.5.

Inform the child or young person and their parents or carers that:

  • alarms have a high long term success rate
  • using an alarm can disrupt sleep
  • that parents or carers may need to help the child or young person to wake to the alarm
  • using an alarm requires sustained parental and child commitment, involvement and effort
  • they need to record their progress (for example, if and when the child or young person wakes and how wet the child and bed are)
  • alarms are not suitable for all children and young people and their families.[1.8.13]
12.2.9.6.

If offering an alarm for bedwetting in children, inform the child and young person and their parents or carers how to:

  • set and use the alarm
  • respond to the alarm when it goes off
  • maintain the alarm
  • deal with problems with the alarm, including who to contact when there is a problem
  • return the alarm when they no longer need it.[1.8.14]
12.2.9.7.

Inform the child and young person and their parents or carers that it may take a few weeks for the early signs of a response to the alarm to occur and that these may include:

  • smaller wet patches
  • waking to the alarm
  • the alarm going off later and fewer times per night
  • fewer wet nights.[1.8.15]
12.2.9.8.

Inform the child or young person and their parents or carers that dry nights may be a late sign of response to the alarm and may take weeks to achieve. [1.8.16]

12.2.9.9.

Inform the parents or carers that they can restart using the alarm immediately, without consulting a health professional, if the child or young person starts bedwetting again following response to alarm treatment[1.8.17]

Footnotes

10

Early signs of a response may include smaller wet patches, waking to the alarm, the alarm going off later and fewer times per night and fewer wet nights.

Copyright © 2010, National Clinical Guideline Centre.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of the National Clinical Guideline Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

Bookshelf ID: NBK62711

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...