Over the past 100 years, recommended toilet training (TT) methods have oscillated between the two most common TT methods used in North America – rigid adult-directed programs and child-oriented ones (1). In 1962, Brazelton (2) developed the ‘child readiness’ approach, which focused on gradual training and is child-oriented. Current TT guidelines developed by the Canadian Paediatric Society and the American Academy of Pediatrics include a child-oriented approach, not starting before 18 to 24 months of age, and beginning when the child displays interest (3,4). The Foxx and Azrin (FA) (5) method emerged in 1971 as a parent-oriented method that emphasized structured behavioural end point training aimed at eliciting a specific chain of independent events by quickly teaching the component skills of TT. These two methods differ with respect to goal development, end points and emphasis on the child’s self-esteem. Other methods include variations of operant conditioning and assisted infant TT (6,7). The goal of operant conditioning is to establish habits and proper behaviours through positive reinforcement with rewards (8). Assisted infant TT emphasizes simultaneous training of bowel and bladder control by the parent learning the infant’s elimination signals (6). This can begin at two to three weeks of age (9). This method has been criticized as the ‘parent training’ method because the parents must be trained to recognize their infant’s elimination cues. A systematic review was performed to determine which method of TT was best for healthy children. MEDLINE, EMBASE, ERIC, PsycINFO and Cinahl databases were searched using combinations and variations of the following terms – toilet or potty; and behaviour, learn, train or condition. While there are six noncomparative studies examining TT in a single cohort of children, only four comparative studies were included. Matson and Ollendick (10) randomly assigned 10 children (20 to 26 months of age) into one of the following groups – the FA method with an experienced trainer or the FA method conducted by mothers using a book to guide the TT process (10). The goal was to toilet train the children within five daily 4 h sessions. Four of five children in the former group were successfully trained within five sessions, while only one of five in the latter group was completely successful (one obtained partial success and the remaining three failed). All mothers reported that their child exhibited emotional side effects of the training, such as avoidance behaviour or temper tantrums. Candelora (11) randomly assigned 71 healthy children between 18 and 35 months of age to FA’s method of TT in less than one day or the child-oriented method outlined in Spock’s Baby and Child Care book (12). When comparing pretraining and post-training results, and pretraining and follow-up results, the FA method was found to be superior to Spock’s method. Using the FA method, the number of accidents per child per day was reduced by 3.17 from pretraining to follow-up, whereas the child-oriented Spock method resulted in a reduction of 1.89 accidents per child per day. The number of successes per child per day was increased by 3.37 from pretreatment to follow-up when using the FA method and 1.91 for children randomized to the Spock method. Taubman et al (13) randomly assigned 406 children, 17 to 19 months of age, to one of two TT groups – one group was given instructions to praise defecation in the diaper before TT and avoid negative terms to describe defecation, and the other group received no such direction. All parents trained their children using the child-oriented approach and determined when TT would commence. There was no difference in the number of children who developed stool toileting refusal, but duration of the refusal was significantly longer in the control group (7.3 versus 5.1 months). No significant differences were found between the groups in the incidence of stool withholding or hiding during defecation. TT was achieved at a significantly earlier age among those in the intervention group (40 versus 43 months). Bakker et al (14) conducted a retrospective study that collected data on 4332 primary school age children and compared children who did (n=928) and did not (n=3404) develop abnormal bladder control symptoms, such as daytime and/or night-time wetting, and urinary tract infections. Significantly more children in the control group had parental prompting during TT than those in the symptom group (68% versus 62%). Parents of the symptom group used significantly more rewards and punished behaviour than parents in the control group (53% versus 46%). Parents in the control group were significantly more likely to encourage the child to try again later (83% versus 67%). Parents in the symptom group were significantly more likely to make the child wait until voiding, encourage the child to push or strain, make special noises and open a tap.