Conservative management is considered the first-line treatment option for lower urinary tract symptoms (LUTS), such as urgency, polyuria, enuresis and urinary incontinence (UI) (1). Guidelines on conservative management usually include lifestyle interventions, such as caffeine reduction (1). The mild diuretic effect of caffeine could provide a physiological basis to support this recommendation (2). However, systematic reviews on which are based these guidelines still hold very limited evidence supporting caffeine reduction benefits. As these systematic reviews relied only on randomized controlled trials (RCTs), a more inclusive review would allow for a wider understanding of current evidence. To this end, we conducted a scoping review to map out the existing literature on caffeine intake and LUTS in adults.

Literature search:
In this critical systematic scoping review, we searched the following electronic databases for all studies available until June 2018: MEDLINE, Embase, CINAHL, Cochrane Central Register of Controlled Trials, PsycINFO, LILACS, LiSSa, Web of Science, Joanna Briggs Institute Evidence-Based Practice Database. We used the search terms “urgency”, “polyuria”, “enuresis”, “urinary incontinence” and related terms, combined with “caffeine” and related terms through Boolean logic. In addition, we manually searched the reference lists of all relevant articles, as well as unpublished studies using ProQuest, MedNar, OpenGrey, Google Scholar and our research center’s local grey literature catalogue. 

Study selection:
We included intervention studies on the effects of caffeine reduction in adults with LUTS and observational studies on the association between caffeine intake and LUTS in adults. No study was excluded based on language, publication year or risk of bias. Studies were included through a two-step process (titles/abstracts, full-text articles) with two independent reviewers for each step. Discrepancies were resolved by consensus or involvement of a third-party.

Data extraction:
Data from each study were extracted using a modified version of the Cochrane Incontinence group data extraction form, specific to study type (intervention/observational). Articles were screened by study characteristic (author, year, title, journal), design, sample, setting, participant flow, intervention or exposure to caffeine, and reported outcomes. Data extraction was conducted independently by two separate reviewers. Discrepancies were resolved by consensus or involvement of a third-party.

Risk of bias assessment:
All studies were assessed for risk of bias by two independent reviewers using the Mixed Methods Appraisal Tool (MMAT), which offers criteria specific to study design (3). Discrepancies were resolved by consensus or involvement of a third-party.

The literature search identified 1851 records. Manual searches identified an additional 10 records. Of these, 15 records on 14 distinct intervention studies and 12 records on 11 distinct observational studies were included.

Intervention studies on the effects of caffeine reduction:
Among the 14 intervention studies, eight were RCTs, four were quasi-experimental studies with a pre-post design and two were quasi-experimental studies with an interrupted time series design. Study populations were diverse with different LUTS: urgency/polyuria (3/14, 21%), enuresis (1/14, 7%), unspecified UI (6/14, 43%) or multiple symptoms (4/14, 29%). Measures of caffeine intake were heterogeneous with 6/14 (43%) not reporting specific measurement methods. There was also heterogeneity among the LUTS outcome reporting, with 3/14 (21%) using unspecified tools and 2/14 (14%) using non-validated tools, such as staff reports. Eight studies reported the effects of caffeine reduction alone on LUTS while six reported the effects of multimodal interventions including caffeine reduction (Table 1). Overall, no study on caffeine reduction was deemed at “low risk” of bias in all five MMAT categories and 6/8 (75%) had at least one item deemed at “high risk” of bias. For multimodal interventions, 1/6 (17%) studies was deemed at “low risk” of bias in all five MMAT categories and 4/6 (67%) had at least one item deemed at “high risk” of bias.

For urgency symptoms, all studies (n=2) reported a positive effect of caffeine reduction. Yet among these two studies, one did not report any measure of caffeine intake. Both were deemed at “high risk” of bias due to incomplete outcome data. For enuresis frequency, all studies (n=2) reported a positive effect of caffeine reduction. Yet both studies had a small sample size (n=30, n=18) and one study did not report any measure of caffeine intake. Additionally, one study was deemed at “high risk” of bias due to incomplete outcome data and the other was deemed at “high risk” of bias due to both confounding bias and uncontrolled exposure. A combination of positive and negative results or no effect was found in studies evaluating the impact of caffeine reduction on voiding frequency, UI frequency, UI volume, symptom severity and quality of life.

Multimodal interventions included other treatment modalities, such as pelvic muscle exercises, biofeedback, education or counselling (i.e. bowel habits or weight) in addition to caffeine reduction. Although, 5/6 (83%) studies reported favorable results, only 1/6 (17%) studies included secondary analyses associating treatment effectiveness with caffeine reduction. Despite a clear positive effect of the multimodal intervention on UI frequency, no association between caffeine intake and the UI frequency was found.

Observational studies:
Six of the 11 observational studies were cross-sectional; other designs were qualitative (n=1), case-control (n=1), longitudinal (n=1), case report (n=1) and survey (n=1). Study populations were diverse with different LUTS: urgency/polyuria (3/11, 27%), enuresis (1/11, 9%), unspecified UI (4/11, 36%), stress UI only (1/11, 9%) or multiple conditions (2/11, 18%). Reported measures of caffeine intake were heterogeneous, and 3/11 (27%) studies did not report on their measurement methods. There was also heterogeneity among the LUTS outcome reporting, with 3/11 (27%) using unspecified tools. Eight studies reported on the association between caffeine intake and LUTS while three studies reported the effect of caffeine intake on the success of LUTS treatment (i.e. pharmacologic treatment, pelvic floor muscle training) (Table 2). Overall, 3/8 (38%) studies on the association between caffeine intake and LUTS were deemed at “low risk” of bias in all five MMAT and 1/8 (13%) had at least one item deemed at “high risk” of bias. No study on the effect of caffeine intake on the success of LUTS treatment was deemed at “low risk” of bias in all five MMAT categories and 2/3 (66%) had at least one item deemed at “high risk” of bias.

All studies (n=1) reported a positive association between enuresis frequency and caffeine intake. However, this was a case study with only one participant. No outcome measurement methods or statistics were reported. Mixed results or no significant associations were observed between caffeine intake and urgency, voiding frequency, UI frequency, UI volume and symptoms severity. 

Finally, studies on the effect of caffeine intake on the success of LUTS treatment reported mixed results.

In our critical systematic scoping review, we included 14 intervention studies. Four (29%) studies reported a positive effect of caffeine reduction on LUTS, and one study out of 11 observational studies (9%) reported a positive association between caffeine intake and LUTS. Our results highlighted high heterogeneity in studied populations, caffeine measures and reported outcomes. Furthermore, there was unknown or high risk of bias in most identified studies.

Future studies should target specific populations and confirm related UI symptoms during the recruitment process with valid and reliable tools. Valid and reliable measures of caffeine intake and LUTS outcome measures should also be included. Finally, the use of sound study design and reporting guidelines would ensure lower risk of bias in future studies.

Smith A, Bevan D, Douglas HR, James D. Management of urinary incontinence in women: summary of updated NICE guidance. Bmj. 2013;347:f5170.
Creighton SM, Stanton S. Caffeine: does it affect your bladder? BJU International. 1990;66(6):613-4.
Hong QN, Fàbregues S, Bartlett G, Boardman F, Cargo M, Dagenais P, et al. The Mixed Methods Appraisal Tool (MMAT) version 2018 for information professionals and researchers. Education for Information. 2018(Preprint):1-7.