Table D-9Outcomes and results of included studies (observational studies)

Author-YearToilet Training Method(s)Study DesignPrimary OutcomeOther Outcomes or Effect ModifiersResultsConclusions
Healthy children
Bakker 2002Other: Increased prompting v. less promptingRetrospective cohortResidual lower urinary tract symptoms (LUTS) at ≥10 yrComparisons based on sex, family structure, number of bedwetting-relatives, school performance, self-management, daily hygiene, age TT begun, parent attitudesN=4332Data show significant differences between children with lasting problems with bladder control and those without.
928/4332 (21%) symptomatic at 10 to 14 yr. Symptom group v. control: significant difference in:
1.

Proportion female (62.1% v. 45.2%)

2.

Fewer from stable first marriage (82% v. 86%)

3.

Bedwetting relatives (37% v. 25%)

4.

Poorer school performance (12% v. 8%)

5.

Less ability to manage homework and appointments independently (73% v. 68%)

6.

Less capacity to manage daily hygiene (37% v. 31%)

7.

Began TT at older age (22% v. 32% began TT before 18 mo)

8.

Parents in symptom group used less prompting, were more liberal, rewarded and punished more, and exerted more pressure if attempt to void failed

Beginning TT >18 mo and using certain methods to provoke voiding if attempt failed increased the risk of LUTS.
Brazelton 1962Child-orientedRetrospective cohortAge, day, and nighttime bladder continenceComparisons based on sexN=1170Day and nighttime training effected later in first child than subsequent siblings. Boys took longer to be night trained. A child-orientated program helps prevent residual symptoms.
Day training achieved 1–2 mo later and nighttime 1–7 mo longer in 1st child
 54.7% began TT at 24 mo
 12.3% achieved bowel training first
8.2% bladder, and 79.5 % simultaneous training
80.7% daytime trained by average 28.5 mo and 80.3% night trained by 3 yr.Average age to complete all training was 33.3 mo; males took 2.3 mo longer to complete
1.4% had residual problems >5 yr
Healthy Children
Butler 1976Azrin and FoxxProspective cohortBowel and bladder accidentsComparisons based on sex and ageN=49Males and all children >25 mo. trained faster. 20% stopped wetting at night. Some children reacted negatively to positive practice-sessions following an accident, and some parents found it difficult not to prompt at signs of self-initiation.
End 1st intensive training session
 <25 mos: 25/34 (74%) success
 >25 mos: 14/15 (93%) success
Mean intensive training time: all =4.5 hr (range 1.25–10)
 <25 mo.= 5.15 hr
 >25 mo.= 4.08 hr
2 wk post training: mean # accidents
 bladder ↓ 6.03 to 0.60/day (p<0.001)
 bowel ↓ 1.23 to 0.07/day
 3 failures: 2 male, 1 female
8 wk post training: mean # accidents
 bladder = 0.58/day
 bowel = 0.11/day
 2 failures: 1 male, 1 female
Foxx 1973Azrin and FoxxProspective cohortSelf toileting without remindersBowel and bladder accidents, training timeN=34 (results at 1–4 mos)Virtually all healthy children >20 mo. were daytime TT in a few hours and 26–36 mo.old train faster. Nighttime dryness often resulted as well.
33/34 (97%) successChildren reacted favorably to trainer and early tantrums were short-lived.
 Mean no. accidents/day: pre-training / 1 day post- training / 4 mo. post training
Bladder: ~6 / ~0.5 (↓of 97%) / ~0
Mean training time: 3.9 hr (range 0.5–14)
 26–36 mo. = 2.25 hr
 20–25 mo. = 5 hr
1/34 regressed at 3 mo. followup
Kaffman 1972Child-orientedRetrospective cohortEnuresisComparisons by age and sexN=1376Up to 6–7 yr, kibbutz raised children had higher incidence of enuresis and by > 10 yr, it was lower than non-kibbutz raised children. Regressive enuresis was rare.
TT begun at 15–26 mo.
Proportion enuretic:
 4–15 yr: 13.9%; M/F = 14.9/12.8%
 4–7 yr: 133/508 (26%); M/F = 13/13%
 7–15 yr: 59/868 (7%) M/F = 5/2%
Healthy and Mentally Handicapped Children
Mahoney 1971Operant conditioningProspective cohortLevel of performance from 1–10 where Level 10 = independent toiletingNumber of training hoursHealthy: n=3; MC n=5Complete toileting behavior includes complex chain of behaviors and other behaviors should be taught prior to elimination.
Number of trials(results after 29 hrs of training)
Healthy: 3/3 (100%) attained Level 10 over average of 29 hr and 118 trials.
MC: 4/5 (80%) attained Level 10 over average 29 hrs and 262 trials
1 failure
Taubman 1997Child-orientatedProspective cohortStool toileting refusal (STR)Associated factorsN=482Two behaviors associated with STR may need intervention: withholding causing constipation and unsuccessful training by 42 mo.
≤ 3 yr: 292/482 (61%) trained
4 yr: 471/482 (98%) trained
22% experienced ≥ 1 mo of STR, 73% resolved without intervention
STR significantly associated with presence of younger siblings, parental inability to set limits, and later age (>42 mo.) to complete TT
Overall: 48% males and 30% females trained by 3 yr.(p=0.0004)
48% began TT < 24 mo.and 32% not trained until >3 yr
52% begun TT > 24 mo.and 46% not trained until > 3yr
No association between age of TT with mother's work status, attending daycare, behavior scores, or presence of siblings
STR group
 Age STR began: 73% between 24–36 mos
 Siblings: 33% had younger and 17% had older
 Behavior scores: ≤ 2 = 32% with STR, > 2 = 20% with STR
Physically Handicapped Children
Forsythe 1970Other: progressive intervention if unsuccessful bowel controlProspective cohortIndependent bowel controlN=47A combination of regular toileting, initial enemas, and Senokot was the most satisfactory to bowel train children with spina bifida. Dulcodos tablets were almost as effective in those >6 yr.
Bowel control obtained:Manual evacuation and repeated enemas were unsatisfactory due to social reasons and creating dependence on others.
1. Regular toileting (RT)
 8/47 (17%) maintained >2 yr
 8 others relapsed at 8–15 mo.
2. Enemas + RT
 7/39 (18%) maintained >2 yr
 2 others relapsed at 6–8 mo.
1 & 2 combined
 15/47 (32%) success
 10/47 (21%) temporary improvement
 22/47 (47%) unchanged
3. Enemas + RT + suppositories
 Glycerin: 0/25 (0%) success
 Dulcolax: 3/25 (12%)
 Stopped dulcolax at 6 mo.
 Maintained for 2 yr.
 18/25 (72%) relapsed 3 mo. after enemas stopped
 Micralax micro-enemas
 2/22 (8%) success
 Stopped tx at 6 mo.
 Maintained >2 yr
20/25 (80%) unchanged
4. Enemas + RT + purgatives
 Dulcodos: 8/16 (50%)
 Maintained >9 mo.
 Senokot syrup: 15 /15 (100%)
 Maintained >6 mo.
King 1994Other: patient and family education and regular reflex-triggered bowel evacuationRetrospective cohortBowel continenceComparisons based on age, compliance, and reflexes presentN=35Continence more likely achieved if ≤ 6 yr, were compliant, and if had intact bulbocavernosus or anocutaneous reflex or both.
Bowel continence ↑ from 0 to 54.3%
24/35 (68.6%) were compliant; 19/24 (79%) of these became continent
Reflexes present
 Anocutaneous reflex: 8/8 (100%) with reflex became continent
 10/25 (40%) without reflex did not
 Bulbocavernosus reflex: 13/19 (68%) with reflex became continent
 5/14 (36%) without reflex did not
Age
 ≤ 6 yr: 11/17 (65) became continent
 >6 yr: 8/18 (44%) became continent
Sullivan-Bolyai 1984Other:Retrospective cohortDependent and independent toiletingComparisons based on sex, IQ, and training methodN=525Individuals with higher-level lesions may experience delay in achieving successful skill.
bowel trainingSocially acceptable and unacceptable toiletingHigh lumbar/thoracic lesions: 80% became socially acceptable by 16–17 yr; 50% were dependentThose with ileal diversions gained independence at a later age and experienced ammonia odor, persistent infection, recurrent obstruction, and renal stones. Penile collectors caused odor and penile ulcers.
1.

untimed random collection with diaper or insert

2.

infrequent enema

3.

small rapid low level enema

4.

suppositories

5.

timed evacuation ± digital stimulation

Low lumbar/sacral lesions: 80% became socially acceptable by 10–11yr 50% were dependentProblems with CIC and timed voiding included leakage with Valsalva maneuver. A generalset of toilet training predictor guidelines can assist parents and children, avoid frustration, discouragement and psychological problems. Close clinic followup is required.
bladder trainingAll levels: 50% became socially acceptable between 7–9 yr 70% were dependent
1.

diaper/pant insert + periodic cleansing.

2.

penile collector

3.

urinary diversion

4.

timed emptying ± medication

5.

clean intermittent catheterization (CIC) ± medication

No differences based on age, sex, or time TT begun
IQ <69: 1/30 successful
Bowels:
<3 yr: n=41
Socially acceptable/dependent: 39/41 (95%)
 7/41 (17%) timed evacuation
 14/41 (35%) Bisacodyl suppository
 3/41 (7%) expansion enema
 15/41 (41%) diapers
Socially unacceptable/dependent: 2/41 (5%)
> 4 yr: n=184
Socially acceptable/independent: 84/184 (46%)
 55/184 (30%) timed evacuation
 24/184 (13%) bisacodyl suppository
 3/184 (2%) small expansion enemas
 1/184 (.005%) diaper/pant insert
 1/184 (.005%) infrequent enema
Socially acceptable/dependent: 57/184 (31%)
 18/184 (10%) timed evacuation
 (3/18 (17%) with digital stimulation)
 24/184 (13%) bisacodyl suppository
 12/184 (7%) expansion enemas
 2/184 (1%) diaper/pant insert
 1/184 (0.005%) infrequent enema
Socially unacceptable/dependent: 23/184 (13%)
Socially unacceptable/independent: 19/184 (10%)
Bladder:
< 6yr: n=57
Socially acceptable/dependent: 45/57 (79%)
 3/57 (5%) timed
 10/57 (18%) CIC
 32/57 (56%) diaper/pant insert
Socially acceptable/independent: 1/57(2%)
 1/57 (2%) CIC
Socially unacceptable/dependent: 11/57 (19%)
> 6 yr: n=158
Socially acceptable/dependent: 52/158 (33%)
 23/158 (15%) heat diversion
 4/158 (3%) timed evacuation
 12/158 (8%) CIC
 2/158 (0.01%) diaper/pant insert
 11/158 (7%) penile collectors
Socially acceptable/independent: 55/158 (35%):
 27/158 (17%) heat diversion
 8/158 (5%) timed evacuation
 10/158 (6%) CIC
 1/158 (0.01%) diaper/pant insert
 9/158 (6%) penile collectors
Socially unacceptable/dependent: 36/158 (23%)
Socially unacceptable/independent: 51/158 (32%)
van Kuyk, 2000aOperant conditioningRetrospective cohortBowel continence (Templeton and Wingspread scores)Comparisons based on ageN=16Multidisciplinary intervention effective in treatment of constipation and incontinence in children with Hirschsprung's disease. The children improved in all aspects.
Constipation Defecation behavior14/16 (88%) achieved good continence 12/16 (75%) were clean
8/12 (67%) recovered from constipation
Templeton score: ↓ from 2.7 ± 0.48 to 1.1 ± 0.34 (p=0.00)
Wingspread score: ↓ from 3.5 ± 0.52 to 1.3 ± 0.60 (p=0.00)
Constipation: ↓ from 1.8 ± 0.45 to 1.3 ± 0.45 (p=0.01)
Defecation behavior: ↓ from 2.9 ± 0.34 to 1.1 ± 0.34 (p=0.00)
No difference based on age
van Kuyk 2000bOperant conditioningRetrospective cohortFeces in toiletComparisons based on age and high v. lower anal atresiaN=43The intervention was effective and there were no differences based on age. Both somatic and behavioral factors affect persistence of defecation problems, therefore treatment should include behavioral modification techniques.
Number of days without soiling17/43 (40%) achieved good continence
Templeton score Wingspread score Constipation score21/43 (49%) achieved fair continence
Parental judgment incontinence scales51% were clean
40% only staining
10/18 (55%) recovered from constipation
Templeton score: ↓ from 2.7 ± 0.45 to 1.6 ± 0.59 (p=0.00)
Wingspread score: ↓ from 3.4 ± 0.85 to 2.2 ± 0.80 (p=0.00)
Constipation: ↓ from 1.5 ± 0.51 to 1.2 ± 0.41 (p=0.01)
Defecation behavior: ↓ from 2.8 ± 0.39 to 1.4 ± 0.55 (p=0.00)
Straining technique: ↓ from 2.5 ± 0.67 to 1.2 ± 0.43 (p=0.00)
No difference based on age
Mentally Handicapped Children
Ando 1977Operant conditioningProspective cohortSelf-initiated eliminationN=5One should not expect the same dramatic results in TT autistic children that have been shown in the profoundly retarded. Suggest a long baseline record of elimination and a long treatment period plus individual study to determine positive and negative reinforcers.
4/5 (80%) achieved improved self-initiated elimination (SIE):
 1 improved SIE by 50% in 11 mos
 1 improved SIE by 60% in 3 mos
 1 improved SIE by 32% in 3 mos
 1 improved SIE by 18% in 12 mos
 1 made no progress
None completely eliminated inappropriate urination
Better result in those with some receptive language skills
Colwell 1973Operant conditioningProspective cohortToileting skills under verbal control (top score 18)Dressing skillsN=47The majority made significant gains in toileting, dressing, and feeding skills and also improved mental age score.
Feeding skills(results achieved in ~ 7.1 mos)
Mental ageMean toileting score ↑ from 6.0 ± 4.7 to 10.0 ± 4.7 (p<0.001)
33/47 (70%) made gains
8/47 (17%) made no gains
3/47 (6%) regressed from baseline
Connolly 1976Operant conditioningProspective cohortWetting incidentsN=9Positive use of operant conditioning can help toilet train the severely mentally handicapped.
Soiling incidents(results at 7 wk followup)
Successful toileting following accident free periodWetting accidents ↓ 14% and soiling accidents ↓ 25%
2/9 (22%) were successfully trained
Successful toileting following accident free period periods continued to decrease
Didden 2001Azrin and Foxx (modified)Prospective cohortAccidents and correct toileting/day and at 2.5 yr.post training followupTime spent trainingN=6Azrin and Foxx TT significantly increased correct prompted toileting that was sustained at followup. It somewhat decreased accidents.
(results at 2.5 yr.followup)
Mean incorrect toileting/day ↓ from 1.65 ± 1.76 to 0.12 ± 0.29 at followup (p=0.07)
Mean correct toileting/day ↑ from 0.80 ± 0.95 to 3.1 ± 1.57 (p=0.02)
Mean TT time was 17.2 days (range 12–25) and 108 ± 31 hrs
Giles 1966Operant conditioningProspective cohortConsistent self-initiated (SI) bowel and bladder elimination in the toiletN=5Operant conditioning can be an effective means of establishing self-care behavior in institutionalized retardates. Reinforcement must be tailored individually.
5/5 (100%) success
Individual results:
 1 achieved consistent SI bowel & urine elimination at 3 wks; night soiling ended at 5 wks
 1 achieved consistent SI bowel & urine elimination at 7 wks
 2 achieved consistent SI bowel & urine elimination at 8 wks
 1 achieved consistent ‘other’ initiated bowel elimination at 3 wks; urine at 8 wks with some SI
Holverstott-Cockrell 2002Azrin and FoxxProspective cohortBowel and bladder successes and accidentsIntervention acceptabilityN=10; 9 completedPostprogram, there was a significant increase in successes and fewer accidents across all classrooms (4 children continued to have accidents). The intervention was highly acceptable but not carried out consistently by teachers (did not like dry pants checks and positive reinforcement for being dry) and, thus, possibly decreased effectiveness of the intervention. Only a few parents returned data indicating low parental participation
Study takes place in 4 special education preschool classrooms in the same school district. Children attended 1/2 days.Parent participation(results at 6 wk)
Classroom A: n=4
Trend of successes ↑ from 0.6 (0–1) to 4.0 (0–8)/day, accidents ↓ from 2.2 (1–3) to 1.2 (0–3)/day
Classroom B: n=2
Trend of successes remained stable (from 2.4 (1–4) to 4.3 (1–8)/day), accidents ↓ from 1.1 (0–3) to 0.7 (0–3)/day)
Classroom C: n=2
Trend of successes remained stable (from 2.2 (1–3) to 3.2 (0–6)/day), accidents ↓ 1.2 (0–2) to 0.3 (0–2)/day)
Classroom D: n=2
Trend of successes ↑ from 0.07 (0–1) to 2.1 (0–4)/day, accidents ↓ from 1.5 (0–2) to 0.6 (0–2)/day
Kimbrell 1967Operant conditioning v. conventional groupProspective cohortVineland social maturity scale (VSMS) scores (social age, social quotient)Comparisons based on age Laundry useN=40Improved scores on VSMS for social age and social quotient. Soiling decreased and laundry use was cut in half.
Frequency of soilingChange in social age and social maturation(results at 7 mos)
VSMS scores (post test)
Social age: OC: ↑ 0.42 v. CG: ↑ 0.10 (p<0.05)
Social quotient: OC ↑ 3.30 v. CG ↓ -0.15 (p=ns)
Toilet Training: OC ↑ 4.10 v. CG ↑ 0.30 (p<0.001)
No significant developmental gains
No differences based on age
Experimental group laundry use cut in half
Lancioni 1980Azrin and Foxx (modified)Prospective cohortIndependent toileting for urinationPartially-independent, incomplete, and assisted toiletingN=9Comparisons based on age, sex, sensory condition, and degree of self-stimulation were not significant. Suggest punishment may or may not have played a useful role. Some achieved bowel control as well.
Urinary accidents(results at 44 day followup)
Comparisons based on age, sex, sensory condition, and degree of self stimulation9/9 (100%) achieved daytime independence and accidents ↓ to zero
Partial and incomplete toileting not present
1/9 (11%) continued to have occasional accidents
Lancioni 1981Other: Intervention A: 25 potties in training setting v. Intervention B: no potties displayedProspective cohortIndependent toiletingPartially-independent, incomplete, and assisted toileting Urinary accidentsN=5Intervention A: All increased independent toileting and decreased accidents while continuing normal programs unaltered. The immediate presence of potties may be crucial for developing independence.
Punishment used(results at 60 day followup)
Intervention A: 5/5 (100%) achieved independent toileting that continued at 60-day followup
Mean successes:
6.8 actions/day
2 had no accidents
3 had the odd accident
Intervention B was not effective: when switched to Intervention A with potties improvement began
Lancioni 1981Other: Intervention A: 25 potties in training setting v. Intervention B: no potties displayedProspective cohortIndependent toiletingPartially-independent, incomplete, and assisted toiletingN=4Intervention A: All increased independent toileting and decreased accidents; results continued at followup. Suggest punishment useful in those who have history of accidents, but not necessary in those who toilet when prompted or assisted.
No punishment usedUrinary accidents(results at 60 day followup)
Intervention A: 4/4 (100%) achieved independent toileting that continued at 60-day followup
Mean successes:
6.5 actions/day
1 accident occurred
Intervention B was not effective: when switched to Intervention A with potties improvement began
Spencer 1973Operant conditioningProspective cohortIndex of bowel controlN=38Operant conditioning can improve toileting behaviors in the profoundly retarded.
(results at 6 wk)
Accidents ↓ by 17%
Spontaneous toileting ↑ 9%
Those who had greater initial bowel control did not improve, those totally incontinent showed considerable progress
Smith 1977Azrin and FoxxRetrospective cohortWetting accidents/wkN=8Significant drop in wetting accidents. Those <20 yr.trained faster; those with SA 2–2.5 yr.progressed faster than those with SA 1.5–2 yr.
(results at 10 wk)
Wetting accidents:
Younger, low social age (SA):
2/5 (40%) averaged 1 accident/wk
Younger, high SA: 3/3 (100%) zero accidents
Tierney 1973Operant conditioning v. controlControlled clinical trialReduced incontinenceSoiled linenN=36Operant conditioning led to improvement. A significant reduction in use of nappies, laundry and staff time to manage incontinence was attributed to operant conditioning. Also noticed general functional improvement among the operant conditioning group.
Successful continence of urine and feces on 16-level scale where 16=independentStaff workload(results on 18 experimental patients at 30 day followup)
Number in nappiesOC: 14/18 (78%) improved and were removed from diapers
 7/18 (39%) achieved level 4
 5/18 (28%) achieved level 3
 2/18 (11%) achieved level 2
 6/18 (33%) achieved nocturnal continence
 4/18 (22%) showed no improvement
Behavior relating to ‘sitting’ levels (5–8) more easily achieved than behavior relating dressing (9–12) and ‘going to the toilet’ (13–16) levels
Control: showed minimal improvement
Van Wagenen 1969Other: auditory signal followed by rapid forward moving series of training eventsProspective cohortCriterion levels 1–6 where 6= self-initiated urination with no promptsN=9This procedure successfully trained the profoundly retarded subjects and the training transferred to other environments.
(results at 19.5 hr to 22 days)
9/9 (100%) achieved level 6

CCT indicates controlled clinical trial; GTRP: group training regular potting; IIRP: intensive/individual regular potting; IITP: intensive/individual timed potting; mos: months; NS: not significant; OC: operant conditioning; RCT: randomized controlled trial; SCH: scheduled; STR: stool toileting refusal; TTLD: toilet training in less than a day

From: Appendix D: Evidence Tables

Cover of The Effectiveness of Different Methods of Toilet Training for Bowel and Bladder Control
The Effectiveness of Different Methods of Toilet Training for Bowel and Bladder Control.
Evidence Reports/Technology Assessments, No. 147.
Kiddoo D, Klassen TP, Lang ME, et al.

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