The Recurrent Urinary Tract Infection Symptom Scale

Recurrent urinary tract infection (rUTI), is defined by the European Association of Urology (EAU)⁴ and the American Urological Association (AUA)⁵ as experiencing a minimum of two episodes of UTI within six months, or three episodes within a year.

Recent research estimates that over 400 million people experience UTI each year worldwide,⁶ over 90% of whom are female.⁷ Up to 50% of females who experience a UTI are likely to experience a recurrent infection within a year.⁷

The RUTISS was developed for self-completion by adults (aged 18 years and above) living with rUTI. Readability analysis established that this UTI assessment measure is suitable for completion by people with a reader's age of 12 years old and above (7th US grade, equivalent to UK Key Stage 3/Year 8).

Section B, the global rating of change scale, utilises an 11-point Likert scale. A central response (zero) represents “no change”, with the left side ranging from –1 to –5 (“very much worse”) and the right side ranging from +1 to +5 (“very much better”).

Section C, aimed at exploring the severity of key rUTI symptoms and pain and/or discomfort, employs an 11-point numerical rating scale. The scale ranges from 0 – 10, with 0 representing “not present”, 1 “very mild”, and 10 “extremely severe”.

The RUTISS provides a comprehensive assessment of key rUTI symptom and pain factors from the patient perspective, encompassing a UTI symptom scale and UTI pain scale. Each section explores a distinct domain, with Section C also assessing four subdomains of the UTI symptom experience and UTI assessment:

A. Frequency of UTI symptoms

B. Global change in UTI symptom severity

C. Severity of key rUTI symptoms and pain and/or discomfort

• • • Urinary symptoms (3 items)
    • Urinary presentation (3 items)
    • UTI pain and discomfort (2 items)
    • Bodily sensations (3 items)

These domains perform robustly as statistically separate unidimensional scales and may be administered separately if required.

Both Sections B and C of the RUTISS assess patient-reported symptoms over the past 24 hours. This recall period was adopted due to the potential for rapid fluctuation and variability in UTI symptoms over time.

Respondents typically complete the RUTISS within 3-5 minutes.

The RUTISS was developed to facilitate simple and straightforward assessment for UTI and collection of patient-reported outcomes in both research and clinical settings. As a PROM, this tool is intended as a self-report measure for patients to complete without clinician or proxy involvement, and the language has been validated with a diverse patient sample (see development methodology).

Respondents will initially read an introductory paragraph aimed at defining key terms used in the RUTISS, including what is meant by ‘recurrent urinary tract infection’. All three sections of this UTI questionnaire are introduced with instructions, outlining their objectives and guiding them on key information to consider whilst responding (e.g., the recall period). Simple checkboxes are used for Likert and numerical scale response options. A closing statement is provided at the end to signal that there are no further questions.

Upon completion of a licence for the RUTISS, administrators will be provided with scoring and administration instructions and, where applicable, a concept elaboration document to assist in translation development. A sample copy of the RUTISS is available to view here.

Currently, the RUTISS can be administered either via paper and pen completion, or electronically as an ePROM (please get in touch to find out more).

A straightforward scoring system was established, with the flexibility to prepare scores from individual domains as well as overall RUTISS severity score (based on Section C). A full scoring guide and calculation tool are available upon completion of a licence.

Section B is a single item global rating of change scale, exploring change in symptom severity. Scores range between –5 (very much worse), to 0 (no change), to +5 (very much better). Positive scores indicate symptom improvement over the past 24 hours, whilst negative scores indicate relative worsening or deterioration in symptoms.

For Section C, individual domain scores may be calculated by adding together the scored responses within a single domain (e.g., urinary symptoms, urinary presentation, etc.). An overall RUTISS severity score may be computed from the sum of all Section C responses, divided by 11 and then multiplied by 10. A score ranging between 0 and 100 will be generated, with greater scores indicating greater overall UTI symptom and pain severity. The minimal clinically important difference (MCID) for the overall RUTISS severity score is 6.5 points.⁸

The RUTISS may be used independently to assess rUTI symptoms, or alongside additional generic and/or condition-specific measures of the rUTI patient experience.

The Recurrent UTI Impact Questionnaire (RUTIIQ), also developed by PARED Insights, is a validated PROM which evaluates the psychosocial impact of living with rUTI. Comprised of 18 items, this UTI assessment tool explores key domains of quality of life that are uniquely impacted by this urological condition, comprising:

• Personal wellbeing, including mental health and mood
• Social wellbeing, including isolation and social anxiety
• Work and activity interference, including productivity and withdrawal from activities
• Satisfaction with rUTI healthcare, including confidence and feeling listened to
• Sexual wellbeing (optional), including avoidance and impact on relationships

In order to capture a full assessment for UTI, exploring both rUTI symptoms and quality of life experiences from the patient perspective, it is recommended that the RUTISS and the RUTIIQ are administered together. It is possible to request a licence for both measures at the same time.

A Delphi consensus method was utilised to systematically gain quantitative and qualitative feedback on the RUTISS from 15 expert rUTI clinicians.¹² Clinicians were based in the UK, USA, and Canada, and demonstrated diverse expertise including from both general practice and specialist backgrounds (e.g., in urology and urogynaecology).

Two rounds of feedback facilitated consensus building, calculation of content validity indices, and identification of potential areas for refinement. Item refinements were implemented based on clinician feedback prior to Stage 2.

One-to-one cognitive interviews were conducted with a heterogeneous sample of 28 people living with rUTI, aged between 18 and 82 years old (approximately 47 years old on average).¹² Participants were English-speaking and based in 10 countries, predominantly the UK and USA.

A cognitive debriefing methodology, including ‘think aloud’ techniques, was applied to generate qualitative feedback on the RUTISS from the patient perspective. Interviews facilitated evaluation of the relevance and clarity of each item and questionnaire instruction, and identification of potential areas for refinement.

Interviews were conducted in two phases to allow iterative refinements, with an opportunity to validate revisions made after the first phase during the second phase.

A pilot evaluation of the preliminary RUTISS was conducted online, facilitating exploratory psychometric analysis and item refinements.¹² A total of 240 people living with rUTI, aged between 18 and 84 years old (45 years old on average), completed the RUTISS and existing measures of related constructs (UTI pain scales and UTI symptom scales). Questionnaires were completed twice, separated by 24 hours, to facilitate test-retest analysis. Participants were English-speaking and resided in 24 countries in total.

Exploratory factor analysis determined the latent factor structure (structural validity) of the RUTISS. Item reduction was implemented to improve performance, resulting in a preliminary 28-item RUTISS. Four factors were identified: ‘urinary symptoms’, ‘urinary presentation’, ‘UTI pain and discomfort’, and ‘bodily sensations’. These factors represented a strong fit for the data and collectively account for 75.4% of the total variance in scores.

Internal consistency of the RUTISS domains was excellent (Cronbach’s alpha coefficients, ɑ, were 0.87 and above), and test-retest reliability was strong (intraclass correlation coefficients, ICC, were 0.73 and above). Construct validity, assessed via comparison with existing related measures of urinary symptoms and pain, was similarly strong (Spearman’s rho coefficients, ⍴, were 0.60 and above).

This stage collected further RUTISS data from a new sample, of 389 people living with rUTI in 37 countries.¹³ Participants were aged between 18 and 87 years old, with an average age of approximately 45.

This new data facilitated confirmatory psychometric analysis, including bifactor modelling and item response theory analysis. From this, final item reduction was implemented which resulted in an optimised 15-item recurrent UTI questionnaire (see sample copy here). The final RUTISS comprises a 3-item symptom frequency section, a single item global rating of change scale, and an 11-item general ‘rUTI symptom and pain severity’ subscale with four subdomains. These subdomains assess ‘urinary symptoms’, ‘urinary presentation’, ‘UTI pain and discomfort’, and ‘bodily sensations’. Please see the key characteristics of the RUTISS above.

The latest stage of RUTISS validation involved the assessment of the measure’s responsiveness to change following antibiotic treatment, and determination of its minimal clinically important difference (MCID).8 Longitudinal data from a sample of 108 people living with rUTI were collected before, during, and after antibiotic treatment.

In accordance with COSMIN recommendations, responsiveness was evaluated using receiver operating characteristic (ROC) curve analysis, effect size determination, and comparison with external anchors, including the UTI Symptom Assessment (UTISA) and Patient Global Impression of Change (PGIC). The MCID was established through triangulation of anchor-based methods and distribution-based estimates, including effect sizes, standardised response means, and standard error of measurement.

This rigorous analysis confirmed that the RUTISS is sensitive to clinically meaningful patient-perceived changes in symptom severity, with an established MCID of 6.5 points for the overall RUTISS severity score. Sensitivity analyses supported the robustness of these findings across different analytical approaches. These results provide clinicians and researchers with essential interpretation guidelines for monitoring treatment response and evaluating intervention effectiveness in recurrent UTI management.

PROMs can facilitate:

• Standardised assessment and quantification of patient-reported outcomes
• Patient monitoring
• Shared decision-making
• Patient-centred and value-based care
• Improved patient-clinician communication
• Identification of target symptoms and other health outcomes that may require intervention and/or additional support
• Identification of change (improvement or deterioration) in response to treatment and interventions
• Assessment of new and existing health interventions (e.g., in clinical trials)
• Evaluation of healthcare quality and patient satisfaction
• Quality improvement
• Electronic data collection and longitudinal reporting

Effective PROMs should be developed and validated using rigorous procedures, such as those outlined by the US Food and Drug Administration (FDA)⁹ and the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) initiative.^10,11

To ensure the resultant measures are truly patient-centred and reflect the health outcomes which are both relevant and important for those responding, a high level of patient and public involvement and engagement (PPIE) should be maintained throughout all stages of development and validation.^11,15

Effective development methodologies typically include both qualitative and quantitative assessment of new PROMs, including cognitive debriefing techniques with patients to identify areas for improvement from the patient perspective. Refinements are made iteratively throughout development to continually optimise the measure’s psychometric performance before finalisation.

Robust statistical testing should be implemented to validate PROMs,^10,11 including assessment of: