Hypothesis / aims of study One of the challenges in health care today is providing affordable care for those in need, and identifying a reasonable level of care for care-seekers. Many women with urinary incontinence might be reluctant to seek care for various reasons. Recent reviews propose lifestyle advice, pelvic floor muscle training (PFMT) and, in some cases, behavioural changes as first-line treatment for urgency (UUI) and mixed (MUI) urinary incontinence in women (1). Treatment via a smartphone app containing lifestyle advice and PFMT has been shown to be effective for, and appreciated by, women with stress urinary incontinence (SUI) (2). A smartphone app could also be a way to make treatment available to more women with UUI and MUI. The traditional recommendation of an extensive examination, on the other hand, has been described as a potential barrier to offering diagnosis and treatment to women with those conditions (3). The results of one study support the use of an algorithm combined with dipstick urinalysis for diagnosing women with urgency-predominant incontinence suitable for pharmacological treatment (3). The first aim of this study was to develop and use an extensive algorithm intended for women with UUI or MUI, to identify those with symptoms that would motivate a physical examination within usual care. The algorithm was intended for women interested in treatment via a smartphone app. To our knowledge, this is the first attempt to identify this target group in this way. The second aim was to estimate the proportion of the people interested that might be suitable for smartphone app treatment, based on the algorithm. Study design, materials and methods This report is part of a larger RCT study aimed at evaluating smartphone app treatment for women with UUI and MUI. As part of the preparations for the RCT study, a team of experienced general practitioners (GP), a Specialist Continence nurse, a urologist and a urogynecologist together developed an algorithm with questions regarding symptoms for which an examination would be judged important within usual care. The team included both researchers and clinicians. The RCT study was approved by a regional ethics board and registered in the Clinical Trials register. Recruitment was carried out via conventional methods (press releases, information to midwives, advertisements in media) as well as via Facebook advertisements. The advertisements directed interested people to the homepage of the research project, where additional information about the study and a link to a web-based screening questionnaire was provided. The screening questionnaire included questions on inclusion criteria and some background information, before presenting the questions related to the algorithm (figure 1). People who did not meet the inclusion criteria (woman, ≥18 years, ≥2 leakages/week, ≥12 month symptom duration, urgency or mixed urinary incontinence), or those who were pregnant or used another PFMT app or antimuscarinic drugs could not proceed further with the questionnaire. If a respondent gave a positive answer on any symptom from the algorithm, she was excluded and recommended to contact her normal health care provider for further assessment. Any respondent who passed the screening questionnaire in full was asked to provide her email address and thereafter received an informed consent form and a bladder diary to complete. Once these were returned, the respondent received another questionnaire and was thereafter contacted via telephone by a Specialist Continence nurse or GP. The purpose of this telephone interview was to give the diagnosis and to verify the answers to the algorithm questions. Results Following a year of meetings and discussions in the research team, a final algorithm was decided via consensus, based on previous literature on the subject as well as clinical experience. The symptoms and conditions included in the algorithm were painful urges; pyelonephritis; three or more urinary tract infections (UTI) in the last 12 months; dysuria (burning upon urination); visible haematuria; non-investigated bladder emptying difficulties; metrorrhagia; cancer in the pelvic area, bladder or bowels; decreased mobility or sensibility in the legs or pelvic area; previous stroke; neurological disease and diabetes (figure 1). The algorithm was used in the web-based screening questionnaire as described above. Out of 765 women with UUI or MUI with ≥2 leakages/week and ≥12 month duration, 523 were identified as eligible to be offered e-health treatment after exclusions. The 238 women who were excluded for symptoms in the algorithm were automatically advised to contact their normal health care provider for further assessment (figure 1). A further four women left the questionnaire before completion of all questions and were therefore not included. Of the 523 eligible women, 142 women chose to complete all the successive steps and were interviewed via telephone. In the interviews, nine women presented algorithm-related symptoms. In five cases, those symptoms were neurological (i.e. a diffuse sense of numbness in regions of the lower limbs), one woman, aged 51, also had painful urges. Another woman, aged 45, had painful urges as her only symptom. One woman, aged 64, reported having recurring visible haematuria and dysuria three months prior to the interview and had earlier been examined with cystoscopy. Another woman, aged 70, had current dysuria and was being treated for a UTI. One woman had metrorrhagia and was being investigated in usual care. All of these cases were discussed with an experienced GP and/or urogynecologist and were excluded and redirected to their normal health care provider as an extra precaution. Interpretation of results It is possible to develop an algorithm as described above via consensus within a team of experienced clinicians and researchers. Approximately two-thirds of women with UUI or MUI with ≥2 leakages/week and ≥12 month duration who are interested in an e-health intervention might be suited to this kind of treatment. An algorithm such as the one described here might be one way to identify suitable women and redirect those who should contact usual care for an assessment of specific symptoms. However, we do not know whether the respondents who were redirected to usual care had already been examined for these symptoms and/or had relevant underlying pathology. Nonetheless, our view was that the occurrence of any of the other symptoms should motivate precaution, and was a reason for the patient to contact their normal health care provider. Concluding message An algorithm such as the one described here might both help the patient (or health care personnel) to choose a reasonable level of care, and possibly also identify women who had not previously considered seeking care for certain symptoms. In the long term an algorithm might help lessen the burden of ordinary health care providers by directing interested and eligible women to suitable e-health options. We are currently evaluating the efficacy of an app treatment for women with UUI/MUI, both in the short and long term. The results will include information from registers regarding diagnosis and care for relevant conditions.