Non-antibiotic treatment of acute Urinary Tract Infection: qualitative study in primary care

word count: 225/250


Introduction
As antibiotic resistance continues to rise (1), there has been growing interest in non-antibiotic treatments to manage common bacterial infections, such as acute uncomplicated urinary tract infection (auUTI). AuUTIs are commonly managed in General Practice (2), and are almost always treated with immediate antibiotics (3). However, many UTIs are self-limiting, and there is potential to avoid antibiotics.
Trials of Non-steroidal anti-inflammatory drugs (NSAIDs) (4-7) and herbal treatments (8,9) for auUTI treatment have typically resulted in reduced antibiotic consumption, but worse symptom control. Many trials of cranberry extract for UTI prevention have been conducted, with promising results (10)(11)(12)(13). However, many have suffered methodological problems, such as high participant drop-out attributed to difficulty drinking large volumes of cranberry juice over extended periods (11). A systematic review found limited evidence for or against using cranberry extract to treat auUTIs (14). Despite this, up to 27% of women report consuming cranberry products for auUTI treatment (15) and around 17% use cystitis sachets to help manage auUTIs (15), despite an absence of randomised trial evidence (16).
In addition to establishing the efficacy of non-antibiotic treatments through clinical trials, it is critical to understand women's thoughts on and experiences of using them for symptoms of auUTI. Such qualitative exploration provides an understanding of whether and how women might engage with such treatments if they were shown to be clinically effective and introduced into routine clinical practice. Previous qualitative research has established that a delayed antibiotic strategy may be acceptable to some women (17,18). However, limited studies have focused on exploring the acceptability of non-antibiotic treatments as part of a delayed antibiotic strategy for auUTI symptom management (19).
The aim of this study was to explore women's views on treating auUTIs with non-antibiotic treatments.

Context and Recruitment
Interviews were embedded within the Cranberry for Urinary Tract Infection (CUTI) feasibility trial, an open-label, randomised feasibility trial of the use of cranberry extract in treating symptoms of auUTI in primary care (14). In the CUTI trial, auUTI patients presenting to participating General Practices in Oxfordshire were randomly assigned to: 1) Immediate antibiotic prescription 2) Immediate antibiotic prescription and immediate cranberry capsules 3) Immediate cranberry capsules and a delayed antibiotic prescription in case symptoms worsened, or did not improve within 3-5 days CUTI trial methods and results have been published in full elsewhere (20).
Semi-structured interviews (21) were conducted with: a sample of CUTI trial participants; and, non-CUTI trial patients who had experienced at least one, acute uncomplicated UTI in the preceding 12 months. Non-CUTI trial UTI patients were identified through an electronic search of women ≥ 18 years with an auUTI in the past year conducted at a General Practice in Oxfordshire, outwith the CUTI trial. This practice was chosen to facilitate maximum variation sampling: it was in an area of higher deprivation and with more ethnic diversity compared with CUTI trial practices. Fully informed, written consent was obtained from each participant prior to being interviewed.

Participants
We aimed to conduct 20 to 30 interviews with women ≥18 years who had experienced an auUTI in the preceding 12 months (22), with the final number determined by data saturation (23). We excluded immunosuppressed women, women with underlying urological abnormalities, and women receiving palliative care; such women are more likely to experience complicated UTIs and to require immediate antibiotics. We hoped to employ a purposive maximum variation sampling strategy (24) with regard to age, ethnicity, and whether or not women were CUTI trial participants.

Data collection
We used a narrative, semi-structured interview guide (21) to explore participants' experiences (see supplementary file). The topic guide was developed by OAG through review of literature on non-antibiotic treatments UTIs, in consultation with AMB, and was reviewed by patient and public involvement (PPI) contributors. UTI patients were encouraged to tell their story about their most recent UTI, from when they first suspected they had a UTI, to the end of the illness episode. Additional questions elicited further details on help-seeking behaviour, selfcare strategies, thoughts on non-antibiotic treatments, and experience of taking part in the 6 CUTI trial, where relevant. All interviews were conducted by OAG, audio recorded and professionally transcribed verbatim.

Philosophical approach
An interpretivist approach was taken (25), recognising that a person's beliefs about UTI and non-antibiotic treatments is dependent on their prior experiences, their context, and interactions that they have had. These beliefs are also changeable in light of new experiences, contexts and interactions (e.g. moving to a different country with a different healthcare system).

Data analysis
We conducted thematic analysis (26), with analysis and data collection performed concurrently. Thematic analysis involves constructing and analysing patterns (or themes) within data (26). Transcripts were read and audio-recordings listened to several times to aid familiarisation (by OAG), allowing immersion in the data. NVivo 12 software was used to organise the data and facilitate coding. OAG grouped codes relating to similar phenomena into categories, and subsequently generated themes and sub-themes to describe the data through an iterative process, and in discussion with AMB and STC. Once the thematic structure was finalised, theme labels were refined to comprehensively describe the data within, and supporting quotes chosen to illustrate the themes and sub-themes.

Patient and Public Involvement (PPI)
Four PPI contributors were involved with the CUTI trial and interview studies from the outset. They reviewed all public facing documents (e.g. participant information leaflet). The developing analysis was shared with PPI contributors, to seek their thoughts on the findings, which were incorporated into the analyses.

Results
Twenty seven interviews were conducted with CUTI trial participants (n=14) and non-trial UTI patients (n=13) between August 2019 and January 2020. One interview with a non-trial UTI patient was not analysed as she met an exclusion criterion (immunosuppressed). Interviews ranged from 28 -72 minutes (mean 54 minutes). Participant characteristics are described in Table 1.
The three themes presented here demonstrate women's thoughts on and experiences of UTI management, including: self-care; treatments (immediate/delayed antibiotics and nonantibiotic options, within and outwith the CUTI trial); and help-seeking behaviour.

Theme 1: Treatments, cures and symptom control
Women often spoke of finding a "cure" for their UTI, and for many women antibiotics represented a cure: most reported finding antibiotics an effective treatment for UTIs, which they perceived worked quickly. Interviewees often expressed a tension between finding antibiotics effective, whilst wanting to avoid taking them if possible. A recurring sentiment expressed was fear of becoming "immune" to antibiotics if they took them frequently.

You don't sort of want to, you know, take too many of them [antibiotics] and then like the next episode you have, you cannot sort of treat it well, well enough, you know. ((Non-trial participant (NT) 8)
For other women, typically women who had experienced recurrent UTIs or protracted UTI symptoms, a cure instead implied a permanent end to their UTIs, which went above and beyond their perceived capabilities of an antibiotic.

OAG: …when you then contacted your GP practice…what were you sort of hoping for?
NT6: I think I was…hoping for a cure but knowing that perhaps I would only get antibiotics.
Over-the-counter (OTC) remedies, such as cystitis sachets and cranberry products, were generally seen as more "natural" and more easily accessible than antibiotics, but less reliably effective. Some women used them as holding measures to provide symptom relief when it wasn't possible to get a General Practice appointment. However, other women used them as a UTI treatment in their own right, particularly with milder symptoms, in the earlier stages of their illness, and if they didn't have an important upcoming engagement (e.g. going on holiday).

They're [cystitis sachets] quite effective if you, if you get it early but once an infection takes hold, they're, they're not very, in my, from my experience, not very effective.(NT12)
Women frequently reported that they would increase their fluid intake as an early part of their UTI management. Similarly to OTC remedies, increasing fluid intake was viewed by some as a holding measure. However, whilst women usually discontinued OTC measures upon starting antibiotic treatment, women often continued increased fluid intake alongside taking antibiotic treatment, perceiving it as a treatment adjunct and a way to 'flush out' the infection. Some women felt that a mild UTI could even be treated through increased fluid intake alone.

If it's [the UTI] really mild… you can flush it out with water…. (Trial participant (T) 10)
Cranberry juice was commonly reported in this context. Many women perceived that cranberry juice might have specific therapeutic properties over and above other fluids.

I've heard cranberry mentioned so much over the past years, going back from my own GP who is saying, you know, "Drink, drink cranberry juice and whatever." So it's always been in the loop…(T7)
Few women were aware of, or had tried, cranberry in capsule/tablet form, prior to taking part in the CUTI trial/interview study. Women usually felt that cranberry capsules/tablets would be preferable to consuming cranberry juice, due to concerns about the taste and sugar content of juice formulations. Those women who had used cranberry tablets (outside of the CUTI trial) usually reported using them as a means of preventing UTI, rather than as a way of managing an acute UTI.

I've been taking the cranberry tablets just sort of daily (NT13)
Most women did not naturally link analgesia (e.g. paracetamol/ibuprofen) with treating a UTI. Women who reported taking analgesia saw it as a means of alleviating certain symptoms (e.g. abdominal pain), but not as a way of treating their UTI.
It [taking analgesia] wasn't something I really thought or associated really… I was thinking really more about actually making what was causing it better rather than actually taking the painkillers to mask it (T3)

Theme 2: Functional and formulaic -UTI consultations in General Practice and the role of the healthcare practitioner
Women typically contacted their General Practice when they perceived that their symptoms were severe and/or inconveniencing. Physical evidence, namely haematuria, was seen as confirmation that symptoms warranted attention and a legitimate reason for seeking medical attention.

If I see blood in my urine that's like the sign I need to ring a GP…it's not just in my mind… something is going on. (NT8)
Upon contacting their General Practice, women hoped to be seen quickly by a healthcare practitioner and expected an immediate antibiotic prescription; consultations appeared to be set up to meet this expectation.

I rang the doctor's appointment and asked them if I could come and expecting them to maybe just give me some antibiotics… because before when I've gone in, I've always had antibiotics (T1)
Women described quick and focused consultations, and sometimes sensed that healthcare practitioners seemed to be following an algorithm, for which the outcome was usually an antibiotic prescription.

They will just say, infection yes or no; antibiotics, yes or no. (NT13)
They checked the urine and they said, "You have a UTI; we'll give you antibiotics."(NT7) Outside of the CUTI trial, discussions about non-antibiotic treatments with healthcare practitioners were unusual. Women described that healthcare professionals tended to express negative sentiments (stating that non-antibiotic options did not work) or neutral feelings (stating that non-antibiotic options were unlikely to help, but unlikely to do harm).

Theme 3: Changing the treatment paradigm
Women tended not to report that a delayed antibiotic strategy had been used in the management of their UTIs, outside of the CUTI trial. Some women considered that the delayed antibiotic approach provided a welcome opportunity to avoid consuming antibiotics for an acute UTI. However, women weighed this potential benefit against other factors, such as the severity of their symptoms and the timing of their presentation to General Practice. A few women, typically those who had previously experienced an upper UTI, also factored in the potential risk of developing a complicated UTI if antibiotic treatment was delayed.

I might be a bit grumpy about it. I think it would depend how awful I felt…(NT3)
Many women felt that a delay of 3-5 days (as suggested in the CUTI trial) was too long; a shorter delay of 2-3 days was generally considered more acceptable. Women expressed that having contacted General Practice it was important for them to receive something by way of treatment in the interim.

I'm not sure why you'd just delay antibiotics without doing anything because the whole reason for going to your doctor is that you've reached a decision, a big decision that you want to go the doctor to get it sorted out and the fact that the doctor says we'll just wait another couple of days, you might already have waited three, four, five days before you made that decision to go to the doctor… (T14)
Women in the CUTI trial were randomly assigned to one of three groups. Group one of the trial (immediate antibiotics alone) aligned with what most women had originally hoped for upon contacting General Practice, and was therefore the preference for some women. This was particularly the case for women who were experiencing perceived severe symptoms, or who had already delayed seeking medical attention.
Group two (immediate antibiotics and immediate cranberry capsules) was seen by some as the 'best of both worlds,' allowing them to experience the benefit of taking an antibiotic along with any potential additional beneficial effect of cranberry.

Probably the safest one would be the immediate antibiotics and the cranberry capsules because you get both basically, so that's like a, you know, a full force attack…(NT8)
Women assigned to this group compared their UTI experience within the trial with their previous UTI experiences that had been managed with antibiotics alone, and often reflected that they perceived an additive benefit to taking cranberry alongside antibiotics.

Between the two [antibiotics and cranberry capsules], it cleared it up very quickly…it was really, really helpful.. (T8) I took the cranberry tablets alongside [antibiotics] and I thought the symptoms had gone but within a couple of days, I felt they were coming back again. [I]Took some cranberry tablets…I took another two days' worth…I think that probably if I hadn't had the cranberry tablets I would have had to go back to the GP for more antibiotics.(T5)
However, other women primarily saw the utility of cranberry as a way of avoiding antibiotics. Group three (immediate cranberry with delayed antibiotics) was therefore the preferred group for women expressing this view. These women were keen to establish whether cranberry extract would help them personally, and felt that combining cranberry with antibiotics would make this more difficult to ascertain. Women felt reassured that they would receive back-up antibiotics, which one woman described as a "parachute," giving confidence to try a new intervention.
Whilst some women in group three of the trial were able to avoid taking antibiotics (27), of the women who were interviewed in this group, all ended up taking their delayed antibiotic prescription. However, some women interviewed in this group suggested that cranberry had some effect, albeit not as potent as antibiotics, and seemed to prevent symptom deterioration.

Summary
Acute UTI symptoms can be severe and disruptive to women's lives. Antibiotics were usually perceived as a reliable treatment, or indeed a cure. However, respondents in this study were also aware of potential harms associated with antibiotics, such as becoming 'immune' to their effects. Non-antibiotic measures were variably used as UTI treatments, treatment adjuncts, holding measures, or for symptom relief. Non-antibiotic measures were perceived as more natural, but less potent than antibiotics, with better results and greater acceptability if used earlier in the illness course, with milder symptoms, and when patients had no important upcoming engagements.
Women were willing to consider a delayed antibiotic approach, but this option was not usually offered in General Practice consultations. The perceived binary choice between antibiotics or no antibiotics did not appear to leave much room for wider discussions. The acceptability of the delayed antibiotic approach was improved by offering an interim nonantibiotic treatment with perceived therapeutic potential.

Strengths and Limitations
Interviews were conducted with UTI patients within and outwith the CUTI trial. We conducted interviews with non-CUTI trial participants to increase the chance of capturing the views and experiences of women who might be less amenable to trying non-antibiotic treatments. However, it is possible that women who were interested in non-antibiotic treatments for UTI were more likely to respond positively to the interview invite.
Interviews were conducted with women of a range of ages and from a range of backgrounds. We attempted to diversify the sample by recruiting non-CUTI trial participants from a General Practice with more ethnic diversity compared with the CUTI trial practices. However, despite our best efforts, there was limited ethnic diversity in our sample. The results may therefore be less reflective of the experiences of women with UTIs from non-Caucasian backgrounds. Furthermore, electronic searches may have missed some potentially eligible patients.
All interviews were conducted by the Chief Investigator (CI) of the CUTI trial (OAG), which might have influenced the participants' responses. To minimise the potential for this, participants were reminded before the interviews started that they were the experts in their experiences and that the interviewer was keen to hear all views.

13
Finally, we did not receive approval from ethics to interview women who declined to take part in the CUTI trial. Interviewing CUTI trial decliners may have provided additional, useful insights.

Comparison with existing literature
The findings of the present interview study suggest that women would consider delaying antibiotics in certain situations, such as with earlier and milder UTI symptoms. This is in keeping with previous research that explored GP and patient experiences of delayed prescribing for UTIs (18). The authors concluded that the decision to delay antibiotics should depend on whether self-management strategies have already been tried before seeing the GP, and symptom severity.
Leydon et al (17) also explored women's thoughts on and experiences of delayed antibiotics for UTIs, and found that women would be prepared to delay antibiotics if their symptoms were not severe. The authors also suggest that General Practitioners should be mindful that women may not always want immediate antibiotics. The present interview study builds on these findings, and provides additional unique insights that suggest that the acceptability of a delayed antibiotic strategy for UTI may be increased by receiving a non-antibiotic alternative treatment in the interim.
Interviews with women taking part in a trial of immediate antibiotics versus immediate ibuprofen for auUTI in Germany showed they perceived that it was safe to take part because UTI is "not a serious condition." (19). Indeed, four of the five trial decliners interviewed had refused trial participation on the basis that they wished to avoid immediate antibiotics. This contrasts with the findings of the present study, in which women reported significant, disruptive symptoms associated with UTI. Furthermore, the commonest reason for women declining to take part in the CUTI feasibility trial was to avoid being assigned to the delayed antibiotics group (27). This difference may represent cultural differences in the way that UTIs and antibiotics are perceived. Of note, outpatient antimicrobial prescription in Germany is lower than in the UK (28).
A number of studies have evaluated the use of Ibuprofen as an alternative treatment for acute UTIs (4,5,7). However, women interviewed in the present study did not perceive analgesia as a UTI treatment, but rather a means of alleviating pain. This is in keeping with a questionnaire study of women with UTI, which found that most women take antibiotics because they want to 'combat bacteria.' (29) The need to 'combat bacteria' may also be prevalent amongst healthcare practitioners. In an interview study exploring General Practitioners' experience of delayed antibiotic prescribing for a UTI, General Practitioners felt that a firm UTI diagnosis warranted antibiotics (18). They might consider a delayed antibiotic approach for equivocal UTI symptoms, but were more used to using the delayed antibiotic approach in the context of acute respiratory tract infections (18). Urine samples sent for culture are the commonest specimens sent to microbiological laboratories (30). Healthcare practitioners routinely receiving urine culture results that show bacterial growth may serve to reinforce immediate antibiotic prescribing behaviour. As throat swabs/sputum cultures are not routinely sent to laboratories for respiratory infections in primary care (31, 32), the same positive reinforcement of antibiotic prescribing behaviour may not be present for acute respiratory tract illnesses.

Implications for research and practice
Women with auUTIs are amenable to trying certain non-antibiotic treatments with advice to delay antibiotics for a short period, in some situations. NICE guidance suggests that a delayed antibiotic prescription can be considered for women with auUTIs, taking into consideration various factors such as symptom severity and patient preference (33). The present interview study, in keeping with previous studies (17,18), suggests that this does not happen often. There is therefore scope for clinicians to have more discussions with women about a delayed antibiotic approach, particularly if women have presented earlier in their illness, have milder symptoms and do not have pressing upcoming engagements. Such consultations should be sensitively conducted, recognising that the patient is the expert in their own symptoms, and is best placed to make an appropriate decision.
Clinicians should also bear in mind that if a patient rejects a delayed antibiotic approach on one occasion, this does not necessarily mean that they will reject it in future; often, the decision is dependent on their experience of the UTI episode in question, rather than being a fixed preference. These discussions may lengthen a consultation, but they may have the potential to reduce antibiotic consumption, and also to empower women to take control of their acute UTI management. Whilst a delayed antibiotic approach will not prevent all women from consuming antibiotics, given that auUTIs are common and are almost always managed with an immediate antibiotic prescription (3), there is scope to meaningfully reduce antibiotic consumption for auUTIs through taking this approach.
A delayed antibiotic prescription may be more acceptable to women if they receive an interim, non-antibiotic treatment that they perceive to be potentially effective. The results of the CUTI feasibility trial are suggestive of possible, preliminary evidence of an effect of cranberry extract on reducing antibiotic consumption for acute UTI (27). Adequately powered clinical trials are needed to definitively establish whether non-antibiotic treatments like cranberry and cystitis sachets are safe and effective treatments, and to better define when delayed antibiotics are suitable. Ideally, these trials should include a qualitative evaluation to better understand whether/when/how women might find non-antibiotic treatments acceptable when integrated into auUTI management. Such trials and interviews should incorporate the views of people from diverse ethnicities, for example, by engaging community bridging researchers who speak a variety of languages (34).

Conclusion
Under certain conditions, women reported that they are willing to try a delayed antibiotic approach, made more acceptable by concurrently offering an alternative immediate treatment in the interim. There is scope for more discussions about this between women presenting with symptoms attributable to auUTI and healthcare practitioners in General Practice. Furthermore, there is an urgent need to establish the efficacy of non-antibiotic treatment options, such as cranberry extract and cystitis sachets, in the context of auUTIs.

Funding:
This work was supported by the NIHR School for Primary Care Research (Grant reference number: SPCR-2014-10043) and the Wellcome Trust (Grant reference number: 203921/Z/16/Z). For the purpose of Open Access, the author has applied a CC BY public copyright license to any author accepted manuscript version arising from this submission. STC received funding from the National Institute for Health Research (NIHR) Health Protection Research Unit (HPRU) in Healthcare Associated Infections and Antimicrobial Resistance at the University of Oxford in partnership with Public Health England. CJH receives funding support from the NIHR School of Primary Care Research, and the NIHR BRC Oxford and is funded by the World Health Organization for a series of Living rapid reviews on the modes of transmission of SARS-CoV-2 reference WHO registration No2020/1077093. The funders had no role in study design, manuscript submission, or collection, management, analysis or interpretation of study data. The views are those of the authors and not necessarily those of the NIHR or Department of Health and Social Care, nor