The validation of the Concept

To evaluate the concept, I prepared 23 patient cases composed of different LUT disorders — bedwetting, post-void dribble, stress incontinence, extraordinary daytime frequency, overactive bladder, small bladder — and I confronted the prescribed urotherapy plan with the one proposed by the BeDRY app. I named this automated, personalized, digital urotherapy tool “Coach” — it is the coach for the bladder training. To compare the Coach plan with the one prescribed by a specialist in a tertiary hospital (that would be me, from the time before I applied for the proof-of-concept grant), I asked two pediatric nephrologists with whom I often work and who I know see and treat children with bladder symptoms independently. My proof-of-concept goal was that the digital urotherapy plan would have 80% or more agreement with the doctor's prescription.

Both of them received a zip file with a list of anonymized cases. Each case was accompanied by the medical report from a tertiary outpatient clinic at a specific point in time when parents had recorded the diary for at least two days, and a doctor who is an expert in incontinence treatment had prescribed a urotherapy plan. With it, they also received the corresponding automated diary report, the calculated Your Bladder Score, the list of won and where-to-improve badges with a short description, and the Coach digital urotherapy plan. The hypothesis for this TRL4 concept was that, on models of lower urinary tract symptom cases, the digital remote tools were comparable to the doctor's plan. Evaluators received an online form with a list of questions to answer. One question of particular interest was: “Does the digital recommendation (COACH) agree with the doctor's urotherapy recommendation?” Evaluators concluded that in 9 out of 10 cases (91%) the digital plan agreed with the doctors' evaluation, and that in 2 cases out of these 9 (20%) the digital plan was superior.

This was an enormous relief, and proving that the concept worked brought great excitement. But are these tips safe? Could a remote tool give wrong or harmful advice? In all but one case, Coach produced no harmful or incorrect advice; the one case had a minor inconsistency (a boy got post-void dribble tips meant for a girl, which pointed to a bug in our algorithm — and was corrected), with no patient-safety impact. All of the generated plans were judged as practical for families to follow, and evaluators predicted that 94% of cases would show symptom improvement with three months of consistent app use (70% present improvement, 24% major improvement). From a clinical perspective, this feature required clinical evaluation before implementing in production, but it was a great starting point before investing more time and money in development.

Why did the digital plan do so well? It was built from fifteen years of outpatient work at a tertiary centre, where the difficult cases that did not respond to therapy at other hospitals were referred to us. The Coach is not generic advice; it is the prescription pattern of a clinic that has already seen the hardest version of each problem. I remember having exceptional motivation to help these families and children — I wanted to witness their complete treatment success. I read medical books, published reports, attended incontinence conferences, asked colleagues who had more experience at the time, and ran various smaller studies to try to explain anything I thought wasn't researched enough. I employed all the diagnostic tools at my disposal and was willing to go with up-to-date therapy plans if I thought they could result in better outcomes. I strongly believe that developing the BeDRY digital bladder diary with clear instructions, evaluation of diary data, and urotherapy prescription brings standardization into clinical practice — and that standardization is necessary for optimal healthcare. I knew that commitment and patience are crucial to achieving success. BeDRY's Coach urotherapy recommendations are the careful, comprehensive outcome of a long time spent in this field. And I acknowledge that this work is far from finished. Feedback from people who use the app will bring upgrades and eventually improve the app — that is inevitable if we plan to go into production.

During development I realized that 100% success in all patients with the digital tool is impossible. But if we bring an available, affordable, remote tool that will reach complete success in some cases, or partial success in more, contribute as a clinical-decision tool, and help medical experts become more proficient and successful in their own prescription, then it will be a great success. Another relief was that evaluators agreed that 100% of the time users would have no trouble following the prescribed digital urotherapy plan, and that in more than 40% of cases the prescribed plan would be enough for home-only treatment, relieving immediate pressure on the healthcare system.

One of the most interesting aspects of evaluating the concept was the results of Your Bladder Score, compared with how evaluators concluded what the general bladder-control status was in these cases. Both evaluators agreed: Your Bladder Score is too strict. It systematically gave lower scores than the doctors would have given for the same case. The algorithm was harder on the children than the clinicians were. Doctors and Your Bladder Score agreed only in half the number of cases, and Your Bladder Score was mostly lower, meaning it regarded bladder control — as calculated from the available bladder-diary data — as worse than what a clinician would say. To explain this, I did a sub-analysis to see where to improve. The biggest underestimate was for cases that won the “Night Star” badge (the dryness-at-night badge). Their Your Bladder Score values were ranked by evaluators about 17 points higher than the algorithm. This told me that for an experienced clinician, getting through the night dry matters more than my weighting acknowledged. On the other hand, I noticed that some badges with high weights were very hard to win, so I divided these badges into smaller ones to allow families to see incremental progress, even for hard-to-win badges.

After several years of development, a lot of working hours — sometimes waking up at 3 or 4 AM before regular work to get more focused time — and a lot of brainstorming, here is the app I am satisfied with. BeDRY is a digital bladder and stool diary that accurately captures different types of incontinence; it uses the validated Bristol Stool Scale to get reliable data about bowel habits; it has automated summaries with advanced metrics that can be shared with doctors; it has a user-friendly tool to evaluate the level of bladder control and track it over time; and finally, it provides personalized recommendations to improve bladder and bowel habits. This app is aimed at children, adolescents, and even adults with bladder symptoms — especially urinary incontinence — but it can also be used to maintain good bladder habits and track bowel movements.

None of this would have been possible without the great people I met and worked with along the way: Kristijan and his team, Ivana, Ana, Ivan, another Ana, Helena, Stjepan, Maja, Rea, and Hrvoje. My endless gratitude and love go to my wife, Martina, and her patience with my never-ending ideas. Our children, Toma, Mara, Mata, and Ruža, are my infinite source of inspiration; all my work is dedicated to them. Finally, dear reader, if you or your beloved ones have problems with bladder control, I built this app to help you. If you can get your bladder back on its feet again, then my work has not been in vain.