• Anon Writer

Elimination Communication Survey

Research paper by Núria Estrada-Zambrano

Toilet training is usually a challenge for parents all over the world, and difficulties associated with toilet training are a great concern and can cause much stress in families. However there hasn’t been much research regarding this topic in the medical literature, and guidelines about how to achieve this milestone have changed over the years, from rigid habit training approaches that even included coercive methods to the so-called child-centred approach described by Brazelton(1) in 1962 that has greatly influenced the current guidelines(2). In Brazelton’s study, parents of 1170 children were instructed to start toilet training when their child was 18 months or older, introducing a potty chair, showing the child what it was for and encouraging him to use it, and of course praising his or her success. The average age for day continence was 28.5 months, and 33.3 months including night-time. Urine and bowel continence were usually achieved simultaneously.

An earlier observational study3 followed a cohort of 992 children born in Baltimore (USA) in 1952. This study, prior to the widespread commercialization of disposable diapers in the late 50’s, aimed to determine the age of bladder control during the day and at night. It must be stressed that children were categorized as having become dry after a month of having one accident at most, therefore using a strict definition of dryness. Although the authors don’t specify the age when training was started or the method parents used, day-time dryness was reached just before 2 years of age in 50% of children and night-time dryness just over 2 years of age, showing earlier achievement of this milestone than Brazelton had observed.

There are also descriptions in the literature of cultures in which mothers learn to identify their baby’s need to void or defecate, provide an appropriate position in a certain location and cue the baby with a certain sound so that the child doesn’t wet or soil him or herself. Dryness is achieved as early as 6 months old (4 5).

It has been shown that toilet training changes the physiological functions involved in bladder response (increasing the bladder capacity, improving the coordination between the bladder and its sphincter and emptying the bladder completely), and that from birth bladder emptying is not an automatic function without the influence of the brain. In a study comparing healthy Vietnamese (where toilet training is started in infancy and diapers are infrequently used) and Swedish children(6), the authors observed that early potty training of Vietnamese children led to complete bladder emptying by 9 months of age, while Swedish children did not show complete emptying until the age of 36 months (when they were toilet trained). By the age of 24 months, 98% of Vietnamese children had achieved bladder control independent of support from an adult, while by the age of 36 months, only 55% of Swedish children had reached this milestone.

In 2000, a survey in Belgium(7) was conducted to assess the differences in toilet training practices in the previous 60 years. Parents who responded to the questionnaire were grouped according to their generation (those born in 1920-40, those born in 1940-1960, and those born in 1960-1980). In the eldest generation group, toilet training was usually started before 18 months of age (and 50% claimed to have started during the first year), disposable diapers were infrequent and most parents used prompting as a training method. Among the youngest generation, only 22% had started before 18 months, 98% used disposable diapers and only 8% used prompting as a training method. The duration of training was not different in the three groups, with an average duration of 6 months. 71% and 61% of children in the eldest generation group were dry and clean by 18 months of age during the day and at night, respectively, while these percentages dropped to 17% and 8% in the youngest generation of parents.

As shown in this last study, a gradual delay in the age of toilet training worldwide has consistently been observed over the past century, with recent data indicating that the current average start age is 25.5 months for girls and 30.5 months for boys(8), while there is no reason to assume that children’s abilities, capacities or physiological development have changed over the last hundred years. It has been suggested that later initiation of training is related to the belief that early training does not lead to earlier completion of training and the availability of disposable diapers(9).

Unfortunately, this delay in training has been linked to higher rates of stool toileting refusal, defined as the refusal to defecate in the toilet or potty for at least one month while voiding urine consistently on the potty, with 50% of children trained between 42 and 48 months being considered stool toileting refusers and rising up to 73% in those being trained after 4 years of age, while only 11% of children exhibited stool toileting refusal before 2 years(10). It is noteworthy that stool toileting refusal can lead to stool withholding, severe constipation and, if not resolved, primary encopresis(11). Moreover, a review of urinary continence has shown that a later age of training is consistently associated with a higher risk of urinary incontinence, and it has therefore been suggested that the modern trend for late toilet training may have caused an increase in enuresis(12), but no controlled studies on early versus late toilet training exist to evaluate this hypothesis. A large anonymous survey with over 4000 respondents evaluated factors associated with voiding problems such as incontinence, urinary tract infections and urgency, and found a strong association between these problems and the age of onset of toilet training(13). They concluded that bladder-sphincter coordination is not purely a maturational process and can be influenced by toilet training, and suggested that the best time to start this training should be before 18 months of age.

Prior to the present survey and of great relevance is a study conducted by the team of Dr. Rugolotto(14) aimed at examining the practice of toilet training started during the first year of life. An anonymous questionnaire comprising 44 questions was developed in English and translated into Chinese, Dutch, German and Italian. It was distributed in several countries and 286 responses were obtained. The demographic characteristics of the participants were similar to those found in the present survey. The mean age of daytime dryness (with a maximum of 2 accidents per week) and bowel control (excluding episodes of diarrhoea) was 17.4 and 15 months respectively. At night, the mean age for dryness (with a maximum of one accident per week) was 13.5 months (with self-awakening for voiding) and 17.7 months while sleeping through the night. Disposable diaper use and a later start (first vs. second semester of life) were found to significantly delay completion age for both bladder and bowel control.

Elimination Communication

Elimination Communication (EC), also called natural hygiene or assisted infant toilet training, is the practice in which parents or caregivers use timing, signals or cues to tend to the child’s need to void or defecate(15). There is no doubt that communication between infants and caregivers starts right after birth, with signals such as crying, moving, fussiness, and facial expressions indicating hunger, pain or fatigue, and that are intuitively recognized by the adult. Early toilet training can help the caregiver better understand and meet the child’s needs, facilitate the transition to independent toilet completion and achieve continence at an earlier age.

In order to find out about the current and common EC practices and results, an anonymous and voluntary online survey was designed and distributed through Facebook EC communities, Reddit and EC mailing lists. The survey was written in English and included 44 items regarding demographic characteristics of the participants, motivation for EC, aspects about their current practice and results, difficulties experienced and global satisfaction. It was aimed at those currently practicing EC or who had recently finished doing so (in the previous year).

During the time the online survey was open (from 28/05/2020 until 12/06/2020), 1229 responses were received, but for the purpose of the study, only those who claimed to have started elimination communication practices before the first birthday of the child were included, with a total of 1160 responses. Even though the questionnaire didn’t have any compulsory questions, most participants answered all the questions, with usually no more than 5 missing values per answer.


Most respondents were living in the United States of America (492), followed by the UK (167), Canada (97) and Australia (66), with Norway (57) as the first non-English speaking country. However, responses were received from all over the globe:

  • North America 593 (USA 492, Canada 97, Mexico 4)

  • South America 4 (Chile 2, Argentina 1, Brazil 1)

  • Europe 423 (UK 167, Norway 57, The Netherlands 37, Germany 36, Spain 20, Belgium 15, Denmark 14, France 11, Switzerland 11, Slovenia 9, Ireland 8, Sweden 6, Finland 5, Portugal 5, Austria 3, Estonia 3, Italy 2, Latvia 2, Romania 2, Bulgaria 1, Czech Republic 1, Hungary 1, Iceland 1, Lithuania 1, Luxembourg 1, Malta 1, Poland 1, Serbia 1, Ukraine 1)

  • Australinea 78 (Australia 66, New Zealand 12)

  • Asia 31 (Singapore 13, Israel 6, India 3, China 2, Turkey 2, UAE 2, Indonesia 1, Japan 1, Malaysia 1)

  • Africa 23 (South Africa 19, Egypt 1, Ethiopia 1, Kenya 1, Zambia 1)

  • Not specified 8

Most of the respondents were female (1143, 98,5%). The participants’ mean age was 32.57 years (SD 4.78), with a range from 19 to 61. The oldest participant claimed to be the grandmother of the child with whom she was practicing EC.

Most of the participants had completed university or college studies (90.2%), while 9.5% had completed high school or secondary education as their highest education level. Regarding current occupations, 32.8% were working for a company, 16.3% were self-employed, 25.2% were on maternity or paternity leave, 21% were unemployed and 2.5% were students, with a remaining 2.2% in other situations (on sick leave, disabled or didn’t answer this question).

General Information Regarding EC

The majority of respondents were living with the other parent of the child (95.1%), while the rest (in descending order) were living with their own parents or other relatives, single parents, sharing house, living with a partner who was not the father of the child, divorced or in other family situations. Most had only one child (67.2%), 24.3% had two, 8.2% had more than two children and the rest didn’t answer this question. Regarding the people who had more than one child, the majority (61.8%) had previous experience with EC, while the rest hadn’t tried it with their other children.

Regarding the involvement of the partner, 78.5% of the respondents claimed that their partner was just as motivated to do EC as them, even though over half of these stated that they participated less because they spent less time with the baby due to work or other duties. However 17.9% said that their partner was more reluctant to do EC or didn’t usually participate. 1.4% claimed that their partner played a more active role in EC than themselves, and 2.2% didn’t live with a partner.

The majority of the participants (53.4%) claimed that none of their acquaintances practiced or had practiced EC, 25.2% stated that they knew only one person or family that practiced EC, 19.4% knew 2 to 5 people/families and 1.9% knew more than 5 families. Furthermore, most participants (62.6%) stated that none of their acquaintances with babies had followed their example and had started practicing EC, while 32.4% claimed that one or two families had followed their example, and 5% reported more than two families had started practicing EC after them.

Most participants used more than one source of information about EC. 65% used online information (podcasts, websites…), 40.5% participated in online communities (Facebook, WhatsApp groups or similar), 27.9% used books, and 27.7% learned from friends or relatives, while only 1.4% claimed their pediatrician, pediatric nurse or other healthcare provider had been a source of information for EC.

Participants of the survey were asked to state which was the first and most important reason to practice EC for them, and also the second and third reason (8 options were offered from a list, but they could also describe alternative reasons). Listed below are the most important reasons stated, in descending order, with the percentage of respondents choosing that answer (second column). In the third column there is the percentage of respondents who chose this option as one of the three main reasons.

Among the other reasons, participants said that EC felt more attuned with their parenting style, that it was “the natural way”, that it was important for them to tend to an important need of the child (the need to be clean) and help the child avoid wetting or soiling him or herself, their curiosity to see if the method would work, that they wanted to see how capable the baby was, that it was challenging and rewarding, to help their baby be more aware of their bodily functions, to empower the child and encourage their independence, to avoid inconvenient diaper leaks, to prevent or heal diaper rash, to travel more lightly, and also after observing the child’s preference for EC or feeling that the baby was clearly demanding EC.

Characteristics of EC Practice

The following questions were about the specific way the participants were practicing EC and their results. The current mean age of the baby was 12.1 months (SD 7.9) with a range from 0 to 48 months. They had started EC on average before the baby was 3 months old (mean 2.8, SD 2.9), ranging from birth to 11 months (participants who said they had started EC after the baby had reached his or her first anniversary were excluded from the study). The average duration of EC until completing the survey was 9.3 months (SD 7.9), with a range from 0 to 48 months.

During the day, half of the respondents practiced full-time EC (49.3%) and the other half practiced part-time EC (50.6%), while only one person answered they didn’t usually do EC during the day. Regarding the type of backup generally used most participants used cloth diapers (50.4%), while the rest used disposable diapers (18.2%), training pants (10.2%) or no backup at all (naked bottom or regular underwear, 17.7%), while the rest used an equal combination of more than one category or didn’t answer this question (3.5%).

At night, most participants claimed they didn’t usually practice EC (60.7%), while some practiced part-time EC (24.7%) and the rest full-time EC (14.6%). Regarding the type of backup generally used at night, most participants used disposable diapers (53.1%), while the rest used cloth diapers (38.1%), training pants (1.7%) or no backup at all (5.4%), and the remaining 1.6% used an equal combination of more than one category or didn’t answer this question.

The place most frequently used for elimination was a baby potty (53.8%), followed by a seat reducer (19.5%), holding the baby over the sink/floor/other (13.3%), a top hat potty (10.4%), a regular toilet (1 %) or other methods (2.1%).

While away from home 32% claimed to keep practicing EC always or almost always, 19.7% often, 26% sometimes and 22.3% rarely or never.

Participants were asked when they usually took their baby to the potty. They could choose more than one answer. Regarding urine, 93.1% did so at certain times (after naps, after eating, when the baby woke up..), 60.5% used baby cues (fussing, pushing away…), 36.3% did so when their baby intentionally showed them their need (with a word, sound or signal, or going to or pointing at the potty), 31.2% used scheduled timing (i.e. every 30 minutes), and 13.2% claimed to have no tricks at all and took their child randomly to the potty. These methods were somewhat different regarding stool movements: 81.9% used baby cues (fussing, pushing away, red face…), 63% did so at certain times (after naps, after eating, when the baby woke up, 40.9% when their baby intentionally showed them their need (with a word, sound or signal, or going to or pointing at the potty), 9.7% used scheduled timing (i.e. every hour), and 5.3% claimed to have no tricks at all and took their child randomly to the potty.

Effectiveness of EC

To evaluate how successful EC practices were, and the relationship with other variables, the participants were asked to rate how continent their baby was at the time of the survey during the day and also at night (only those who stated that they usually practiced EC at night, full-time or part-time), regarding urine and bowel movements, on a 10 point-scale. For the statistical analysis, the pool of participants was divided into those who were still in training, and those who considered that their child had reached continence in each of the four categories (urine and stool, day and night).

Statistical analyses were conducted to study the influence of the start age (groups were formed by those who started in the first vs. the second semester of life, and also those who started in the first vs. second trimester of life), type of EC (full-time vs. part-time) and type of backup used. For this purpose, participants who had practiced EC for less than 3 months were excluded from the analysis, considering that the time of training was too short to influence the results.

Since none of the variables registered had a normal distribution (as assessed by the Kolmogorov-Smirnov test), non-parametric tests were used (Spearman’s rhos, Man-Whitney or Kruskal-Wallis tests). The statistical analyses were carried out using the statistical software PASW Statistics 18.0.

In all four types of continence full-time EC showed greater effectiveness (p=0.008 in stool at night and p<0.000 in the other categories) than part-time EC. However this may be explained in different ways: better effectiveness might be a consequence of full-time, rather than part-time, EC; participants experiencing more success in EC might tend to switch from part-time to full-time EC; or other variables might be influencing the two, such as the current age of the child. In fact, in this study, the babies engaged in full-time daytime EC were 13.5 months old on average, while those practicing part-time EC were 10.5 months old, while at night the difference was even greater (15.4 vs. 10.7 months old for full-time and part-time EC) and this difference was statistically significant in both situations (p<0.000).

In general, the effectiveness of bowel and bladder control was related to the type of backup used, although this was not always significant. In the same way as with part or full-time EC, this relationship doesn’t imply a causal effect, and it seems realistic that the type of backup used changes according to improvements in EC, starting with diapers (cloth or disposable), continuing towards training pants and finishing with regular underwear. Age could also play an important role: the mean age of the babies wearing normal underwear or training pants was significantly higher than the age of those using diapers (p<0.000), while the age of those using disposable vs. cloth diapers was not significantly different, and this was true both for daytime and nighttime backup. Therefore, in the effectiveness analysis, the efficacy of bowel and bladder control was compared between those using cloth and disposable diapers in each of the four categories.

Regarding daily stool continence (N=439), those who started EC in the first semester had higher effectiveness rates than the later start group (p=0,011). There was no difference between those using cloth or disposable diapers. Daily urine continence (N=626) was not affected by start age, but those using cloth diapers were more successful than those using disposables (p=0.012).

Nighttime stool continence effectiveness (N=121) and night urine continence (N=237) didn’t show any significant difference between those who started in the first vs. the second semester of life, or whether cloth or disposable diapers were used. However, as a limitation of this survey, the start age was only registered in global terms (“How old was the baby when you started EC?”), and it seems logical that many people might have started with daytime EC and then, some months later, continued with nighttime EC. Therefore, the variable “start age” regarding nighttime continence may not be reliable. There were no significant differences in any of the four categories when the start age in the first trimester of life was compared to those who started in the second trimester.

Age of Urine and Stool Continence

The respondents were asked if their baby had finished their toilet training. Out of the total of 1160 respondents, around 40% claimed that their baby was fully trained regarding passing stools during the day (459) and at night (470). Fewer respondents claimed their babies were continent regarding urine during the day (252) and at night (167). The most common age for controlling stools during the day and at night was 6 months, while the most frequent age of urinary continence was 18 months during the day and 20 months at night. The average age (“Media”) and standard deviation (“Desviación típica”) are shown in the graphics, and the percentiles 25, 50 and 75 (age by which 25%, 50% and 75% of babies have reached this milestone) are shown in the table below.

Regarding the ability of the baby to intentionally communicate elimination needs, 38.1% claimed their baby had already started signaling the need to void or defecate, at least a significant number of times, not just sporadically. Among these, the most common age was 12 months; while 25%, 50% and 75% of babies had started signaling by the ages of 9, 13 and 17 respectively.

The correlation between the age when they had reached continence and started signaling and the age when they had started EC was analyzed. There was a significant correlation (p<0.000) regarding stool continence (for both day and night), while this was not the case for urine continence, suggesting that the start age (by our definition between 0 and 12 months only) is important regarding stool continence but not so regarding urine continence. The age when the babies started signaling their toilet needs intentionally was also significantly (p=0.004) correlated with the start age.

Potty Strikes, EC Pauses and Abandonment

50.1% of participants stated that their baby had gone through at least one “potty strike”, refusing to use the potty or experiencing a significant decline in success that could not be explained otherwise. Most of these participants (57.3%) claimed to have experienced only one strike, 23.6% had experienced two or three strikes, and 10.3% more than three, while 8.8% didn’t specify the number of strikes. While 49.9% had not experienced any strikes, this percentage could be influenced by the ages of the babies whose parents answered the survey. When only children older than 14 months were selected, the percentage of those who had experienced potty strikes increased up to 74%.

The graphic below shows the age at which babies experienced a potty strike, with 12 months old being the most common age.

Participants were asked if they had completely stopped practicing EC during a certain period of time (at least a week) and then resumed EC. Most participants said no (71.4%), while 14.6% said they had stopped due to practical reasons (i.e. travel, moving house…) and 13.3% due to frustration or lack of success.

Only 28 participants (2.4%) claimed they had definitively abandoned EC. Their reasons were predominantly due to lack of success and potty refusal, but participants also mentioned stress and domestic or external workload, lack of support, depression or anxiety.

Global Burden and Satisfaction

1155 participants rated how tiring and/or time-consuming they felt EC was on a scale of 1 (not tiring or time-consuming at all) to 10 (enormously tiring or time-consuming). Most people rated 3, with a mean of 3.9.

On a similar 10-point scale the same amount of participants rated how satisfied they were with EC globally, ranging from 1 (not satisfied at all) to 10 (totally satisfied). Most participants (39.6%) claimed they were totally satisfied, with a mean of 8.5.


The survey has several limitations that need to be taken into account when interpreting the results. It was based on retrospective information and parental reports, therefore being prone to memory or subjective biases. Even though answers were received from all over the world, the majority of respondents were North American, and this might hinder the generalization of the results observed. The gender of the babies was not registered, while gender differences have been described in the literature regarding toilet training completion, and thus this effect could not be analyzed. There was no clear definition of continence facilitated to the respondents, contributing to differences in the interpretation of the concept. The age of starting training was only asked globally, without differentiating between daytime and nighttime practice. Finally, regarding the effectiveness analysis, variables that played an important role (such as the type of backup or whether EC was practiced full-time or part-time) were only noted once. No history of these variables was recorded, making it impossible to analyze the influence of recent previous practices in the current effectiveness.


Although EC is a common practice in many developing countries, it is still a minority practice in industrialized societies. However, the number of responses received, over 1200 in less than three weeks, and the variety of countries of residence of the participants show there is a current interest in Elimination Communication practices around the world.

The survey was mostly answered by North American mothers with a high level of education, employed or self-employed, living with their partner, who was also motivated by EC, and one child. According to the results of the survey, EC is a very isolated practice, with most people having no acquaintances around them involved in EC and using online resources as the main source of information.

The main motivation to practice EC was feeling that it was more respectful for the baby, but participants also wanted to finish potty training at an earlier age, to clean fewer diapers or save money on diapers and to contribute to protect the environment.

The average start age was just under 3 months, and the average duration of EC of the respondents of the survey was just over 9 months. The mean current age of the babies was one year. During the day, half of the participants practiced full-time EC and the rest part-time EC, and most used cloth diapers as a backup. At night most participants didn’t usually practice EC and used disposable diapers. The potty was the most common instrument for elimination, and while away from home about half usually continued practicing EC. Most participants said that they used baby cues (fussing, pushing away…) to predict elimination and also offered the potty at certain times (after naps, after eating, when the baby wakes up...).

By 13 months of age half of the children had already started intentionally signaling their need to eliminate. 50% of the children had achieved stool continence by 6 months (at night) and 10 months (during the day), and had attained bladder control (both during the day and at night) by 19 months of age.

Among those still in training, the effectiveness of EC was greater for those who practiced full-time EC, and also for those who used regular underwear or training pants, although no causal relation can be determined, and this could also be influenced by other variables, such as the age of the baby. Daily stool continence was related to the start age, with greater effectiveness in those who started earlier, and daily bladder control was related to the type of diaper used (those using cloth diapers experienced more success than those using disposables).

Among those who had already achieved continence, those who had started at an earlier age reached this milestone earlier in the case of stools, but not urine. Also, the children of those who started EC earlier started intentionally signaling their elimination needs earlier.

Half of the children had already experienced one or more potty strikes, with 12 months old being the most common age. Most parents were always consistent with EC, and very few abandoned the practice. Participants stated that EC was not too tiring or time-consuming and were highly satisfied.

Considering the results of this survey and the review of previous research, it seems highly advisable to inform parents or caregivers of the possibility, the method and the expected results of early toilet training, so that parents can make an informed choice. It seems clear that infants are prepared for toilet training at earlier ages than mainstream medicine advocates, that an earlier start does lead to an earlier completion and that, despite the effort involved, there are potential health benefits for infant toilet training.


This study would not have been possible without the generous contribution of over one thousand respondents (parents of young babies who kindly took their time to answer the survey). I am thankful to Rebecca Mottram and Andrea Olson for helping me distribute the survey through their mailing lists. I am grateful to Laurie Boucke for her careful revision of the manuscript and her very valuable suggestions and corrections.

1. Berry Brazelton, T. A Child-Oriented Approach to Toilet Training. www.aappublications.org/news.

2. Toilet learning: Anticipatory guidance with a child-oriented approach. Paediatr. Child Health 5, 333–344 (2000).

3. Oppel, W. C., Harper, P. A. & Rider, R. V. The age of attaining bladder control. Pediatrics 42, 614–626 (1968).

4. deVries, M. W. & deVries, M. R. Cultural Relativity of Toilet Training Readiness: A Perspective From East Africa. www.aappublications.org/news (1977).

5. Richards, C. G. M. Ready, steady, hiss. Archives of Disease in Childhood vol. 66 172 (1991).

6. Duong, T. H., Jansson, U.-B., Holmdahl, G., Sillén, U. & Hellström, A.-L. Urinary bladder control during the first 3 years of life in healthy children in Vietnam--a comparison study with Swedish children. J. Pediatr. Urol. 9, 700–706 (2013).

7. Barker, E. & Wyndaele, J. J. Changes in the toilet training of children during the last 60 years: the cause of an increase in lower urinary tract dysfunction? vol. 86 (2000).

8. Schum, T. R. et al. Factors associated with toilet training in the 1990s. Ambul. Pediatr. 1, 79–86 (2001).

9. Berk, L. B. & Friman, P. C. Epidemiologic aspects of toilet training. Clin. Pediatr. (Phila). 29, 278–282 (1990).

10. Taubman, B. Toilet Training and Toileting Refusal for Stool Only: A Prospective Study. PEDIATRICS vol. 99 www.aappublications.org/news (1997).

11. Beaudry-Bellefeuille, I., Booth, D. & Lane, S. J. Defecation-Specific Behavior in Children with Functional Defecation Issues: A Systematic Review. Perm. J. 21, 1–8 (2017).

12. Wu, H. Y. Achieving urinary continence in children. Nature Reviews Urology vol. 7 371–377 (2010).

13. Bakker, E., Van Gool, J. D., Sprundel, M. Van, Van Der Auwera, C. & Wyndaele, J. J. Results of a questionnaire evaluating the effects of different methods of toilet training on achieving bladder control. BJU Int. 90, 456–461 (2002).

14. Rugolotto, S., Sun, M., Boucke, L., Calò, D. G. & Tatò, L. Toilet training started during the first year of life: a report on elimination signals, stool toileting refusal and completion age. Minerva Pediatr. 60, 27–35 (2008).

15. Boucke, L. Infant Potty Training: A Gentle and Primeval Method Adapted to Modern Living. White-Boucke (2000).

  • Black Facebook Icon