Recurrent cystitis: patients' needs, expectations and contribution to developing an information leaflet - a qualitative study.

Cystitis is an extremely frequent complaint, one out of two women developing an episode over her lifetime. 1 It is a benign condition generally treated with antibiotics prescribed by the primary care physician. Recurring cystitis (RC) is defined as the occurrence of at least four episodes of cystitis over a 12-month period, according to the French Language Society for Infectious Diseases (SPILF) 2 Although the prevalence rate of RC among the female population is not known, some studies on small cohorts of patients suggest it may be quite high. 3–5 Many women report the major impact of the condition on their daily life, as pain and urinary frequency can be invalidating, as well as on their sexual activity. 6 The significant psychological consequences, which are dominated by anxiety, have rarely been explored. 7 According to the French National Agency for Medicines and Health Products Safety, urinary tract infections (UTI) currently rank third among ambulatory antibiotic prescriptions in France. 8 Choice of antimicrobial agents and duration of treatment both appear inappropriate: fluoroquinolone and third generation cephalosporin prescriptions and treatment duration are excessive 9 and do not take either the epidemiology of antimicrobial-resistant bacteria nor the impact on the gut microbiota into account, and thus do not comply with recommendations. 2 10 In some countries, treatment is discussed with patients and alternatives to antimicrobial treatment are offered, which have not shown an increased rate of complications or recurrence compared with patients treated with antibiotics. 11–13 Such an approach should be made more broadly available to women with RC and with no risk of complications. The UK National Action Plan aims to prevent the need for antimicrobials and improve the publics’ infection prevention behaviours. Indeed, the primary objective of a recent qualitative study conducted in the UK was to explore patients’ needs on provision of self-care, which could reduce consultations and unnecessary antibiotic use. An information leaflet was developed to this end. 14 In France, the needs and expectations of patients suffering from RC have not been evaluated. The ReSO-InfectiO PACA EST includes a group of healthcare institutions, laboratories and health authorities in the Provence Alpes Côte d’Azur (PACA) region in South-Eastern France and aims to conduct research and harmonise the management of infectious diseases across the area. Infectious diseases physicians of the RéSO InfectiO PACA EST, coordinated by infectious diseases specialists at Nice University Hospital, consulting women referred for RC, conducted a survey of trigger factors, care pathways and management. 15 This showed that women wished for more autonomy and treatment options. To this end, a qualitative survey was conducted among women with RC to inform the contents of an information leaflet intended to improve patients’ knowledge and to help them manage their condition.

Qualitative semistructured interviews were undertaken by a single researcher trained in qualitative research methods, with a purposive sample of female patients with a definite diagnosis of RC, that is, at least three episodes of cystitis over 12 consecutive months, over 18 years of age, with no cognitive impairment. 16 Patients were recruited via their community-based general practitioner (GP), gynaecologist or urologist as well as via hospital-based physicians. Recruitment was subsequently extended to include patients attending medical laboratories (where information describing the study was delivered through leaflets posted in waiting rooms), pharmacists and via social networks. A snowball effect was produced as recruited patients had contacts with women with similar complaints. Recruitment continued until content saturation was achieved, as observed through immediate debriefing and ongoing data analysis. 17–19 Indeed, data saturation is considered to be reached when there is enough information to replicate the study, when the ability to obtain additional new information has been attained and when further coding is no longer feasible. 20 The interview guide ( online supplemental appendix 1 ) included a brief introduction, a qualitative section with seven neutral, open-ended questions that followed a guiding thread with the possibility of using topical probes if necessary, and a quantitative section with sociodemographic (age, educational level, socioprofessional category, area of residence) and medical details related to RC (attending physician, age at start of RC, main past or current medical conditions). The interview guide was initially tested on two patients and proved satisfactory. No further alteration was required. Recorded interviews were conducted and recorded according to patients’ availability and in any quiet location they chose by a single trained researcher (LB). The aims of the study and the interviewing procedure were explained, and patients provided written informed consent to participate. Each recorded interview was transcribed verbatim by a single trained secretary, with as many details as possible, both verbal and non-verbal. A debriefing procedure by the researcher (LB) took place immediately following each interview to record the overall impression and identify the main ideas put forward by the patients and was shared with coauthors (PT, VM). This allowed to identify the point when theoretical data saturation was reached after discussion and agreement from all researchers, that is, no new ideas arose, and discontinue recruitment. 18 19 Inductive thematic analysis 18 was performed using NVivo software. This included six stages: getting acquainted with the content of the interview (familiarisation), followed by initial coding where codes were approved by all researchers after discussion, identification of overarching themes, grouping of themes or categories of ideas, exploration of links and interaction between themes, description and supporting quotations, according to an iterative procedure. 18 20 On an ongoing basis, this data analysis procedure was discussed among all the coauthors at all the different stages to reach agreement and conducted in accordance with their comments. Each category was summarised in the results section and illustrated with relevant quotes ( online supplemental appendix 2 ). At the end of the results section, the implications derived from these results were illustrated in a table intended to be used as a basis for developing the information leaflet ( online supplemental appendix 3 ). Patients or members of the public were not directly involved in the design and planning of this study.

A breakdown of patient inclusion is illustrated in the flowchart shown in figure 1 Patient flowchart. Patient characteristics are shown in table 1 . Patient characteristics High school or less Some college or tech school Bachelor’s degree or higher Employed Retired Jobless Among the stated risk factors, many women mentioned insufficient hydration, sexual intercourse, pregnancy and parturition, withheld micturition, constipation, and stress. Among the 53 physicians contacted, 50 agreed to participate in recruiting patients (33 community-based practitioners: 31 GPs, 2 gynaecologists and 17 hospital-based practitioners: 9 accident and emergency physicians, 4 gynaecologists, 2 urologists and 2 infectious disease specialists). The interviews were conducted between 30 January 2018 and 3 April 2018. Their mean duration was 22 min (range 9–39 min). Sixteen interviews took place in a medical setting, six in patients’ home, three in patients’ workplace and one in a coffee shop. Among the 26 patients interviewed, 8 had been recruited via their physician (mainly GPs), 5 via leaflets, 7 via Facebook and 6 through snowball sampling. Results of patient interviews are detailed in table 2 . Summary of results and implications (patient quotes are listed in online supplemental appendix 2 ) FT, Fosfomycine Trometamol Unique Dose; RC, recurring cystitis.

This qualitative survey of women’s needs and expectations regarding recurring RC has revealed their need to understand and self-manage their condition. Such patient empowerment is indeed increasingly favoured insofar as women are aware of those situations, which might require a physician’s intervention. Avoiding unnecessary consultations and antibiotic prescriptions thanks to an information leaflet specifying both prevention and management is a major objective which has been advocated namely through a National Action Plan in the UK to ‘raise public awareness to encourage self-care and reduce expectations of antibiotics’. 21 Unlike the present study, the qualitative study conducted in England involved both patients and healthcare providers and was based on focus groups, rather than face-to-face interviews, with a comparable number of patients and a similar approach in seeking patients’ opinion on informing a leaflet. 14 This qualitative study in the form of individual interviews revealed the expectations and needs of patients suffering from RC, as well as their opinions and attitudes. The method provided the opportunity for an in-depth approach of the subject, thanks to the conversational character of the interviews, which were able to overcome any embarrassment these might have caused. The semistructured interview guide contained open questions that allowed to adjust the interview as it progressed. It was tested and found suitable after two pilot interviews. The various recruitment approaches resulted in a diverse and complementary theoretical sample, with a substantial number of respondents with varied characteristics, regarding age, number of years with the condition, age when it began, educational level. To reduce loss of information to a minimum, each interview was followed by immediate debriefing. Thematic analysis was optimal thanks to the use of N-Vivo software. Participants’ response was subject to their level of comprehension and motivation and the time they could allow for the interview. As in all qualitative face-to-face surveys, adjusting to the patient introduces an inevitable bias linked to the interaction between patient and interviewer. Physicians’ involvement was lower than expected as only 5 of all 31 contacted GPs recruited patients. This may be because physicians omitted or forgot to inform patients, or patients refused to participate. Several patients’ profession was related to healthcare, as information on the study and contact details were provided in healthcare facilities, and also due to snowball sampling. This may have translated into heightened health-related awareness of their condition, and thus stronger support for the proposed intervention. Lastly, qualitative thematic analysis and interpretation of results necessarily imply a degree of subjectivity on the part of the researcher. Coauthors were involved in every step of the analysis. Confusion regarding the difference between cystitis and urinary colonisation could explain patients’ inadequate strategies, such as resorting to antibiotics because of smelly urine. Few patients had any idea of the origin of their condition. Indeed, the physiopathology of recurrent cystitis remains obscure and multifactorial. In the case of relapsing RC, the possible presence of intracellular uropathogenic Escherichia coli within the bladder epithelium could interfere with an effective immune response and give rise to re-emergence of infection from this reservoir, 22 23 which may require reconsidering treatment approaches. Several patients mentioned hormonal causes. Oestrogen insufficiency has been described as a causative factor. 24 25 The SPILF suggests local oestrogen therapy after menopause if approved by the gynaecologist. 2 Oestrogen insufficiency results in decreased Lactobacillus vaginal colonisation and E. coli proliferation, and a study of local prophylactic treatment with probiotics shows encouraging results. 26 In the present study, none of the patients took topical oestrogens and only one took probiotics. Some patients mentioned hereditary factors, and a history of UTI in mothers has been noted by infectious disease specialists, suggesting possible genetic susceptibility to infection, 27 although this may also be related to behavioural factors within families. 28 29 The high frequency of RC on initiation of sexual activity is described in the literature. 24 However, stress, fatigue and apprehension of further episodes were also identified as risk factors, as well as bowel dysfunction, as identified in an ongoing prospective study conducted in our area. 30 Social isolation, sexual abstinence during episodes, but also reduced sexual activity at other times were mentioned and have also been described in an Italian study. 6 The major psychological impact, namely anxiety, resulting both from the RC episodes themselves (but which could also be a possible cause) and from the lack of adequate management, emphasises the need for a multidisciplinary approach, taking the stress factor into. 31 The variety of strategies to cope with RC point to a wish by many patients to avoid antibiotics and to self-manage their condition: herbal medicine, aromatherapy. The importance of increasing fluid intake was widely known, if not sufficiently applied. Although many patients resorted to various forms of cranberry preparations, these have not been evaluated in terms of effectiveness, while French recommendations advise a minimum daily dose of 36 mg proanthocyanidin. 32 33 Back-up antibiotic prescriptions, often mentioned in this survey, have been advocated, along with guidance as to their appropriate use. 34 While urine culture was performed too often for simple RC, dipstick tests were infrequent despite recommendations by the SPILF. 9 However, the cost of dipstick tests is not endorsed by the national health insurance, which limits their use. Another inappropriate approach was to treat urinary colonisation revealed by an unnecessary follow-up urine microbiology and culture after a clinically effective antibiotic course, which should not lead to further antimicrobial prescription. Although a wish to avoid antibiotics was expressed, these were used by many patients, especially Fosfomycine -trometamol (FT) due to the convenience of a single dose, but also inappropriate use of other compounds, whether self-administered or inadequately prescribed by a physician. Such variable coping measures when confronting initial signs of cystitis highlight the need for standardised approaches since the stated treatment strategies did not conform with recommendations: self-medication with fluoroquinolones, systematic urine culture, secondary adaptation of antibiotic treatment to susceptibility test results for uncomplicated RC; regular empirical antibiotic treatment for potentially complicated RC, with a single, thus suboptimal, FT dose; nitrofurantoïne as antibiotic prophylaxis, which is strictly contra-indicated according to French guidelines. 2 5 9 28 Many non-antibiotic options for RC have been explored. 35 Treatment strategies have been evolving in Scandinavian countries and in Germany, where painkillers/non-steroid anti-inflammatory drugs can be offered for treating cystitis with mild/moderate symptoms in a watch and wait approach although their efficacy remains controversial. 12 36 37 Phytotherapy has been shown to be effective. 38 39 French guidelines state that topical oestrogens can be beneficial to menopaused women, 40 while, according to a Spanish study, prevention with D-Mannose significantly decreased the frequency of UTI. 29 41 Immunotherapy using a vaccine based on a bacterial extract is currently being tested. 42 Lastly, among the various approaches aiming to alleviate pain and stress, hypnosis could prove useful: it has been shown to alleviate symptoms in irritable bowel syndrome and to reduce the need for analgesics during surgical procedures. 43 Preliminary results of its use in RC are promising (ongoing hypnocyst protocol by the same author, unpublished data). The discrepancy between symptom intensity and the reputedly benign character of RC stands out as a frustrating situation whereby the condition is not seriously considered, although patients’ distress has recently been acknowledged. 44 The lack of physicians’ and society’s understanding regarding RC contrasts with the major impact of the condition on patients’ activities, perception and degree of anxiety. Ignorance of the cause of RC, cost of antibiotic alternatives, lack of investigation or of conclusive results thereof led patients to adopt a fatalistic attitude. This was noted by Italian authors who concluded to the ‘cost of resignation’ related to physicians’ lack of involvement. 6 Certain patients even expressed surprise at being questioned regarding their opinions on their management of RC. Patients’ request for more autonomy has been addressed in France with a strategy put forward in 2014, targeting selected, educated women with no risk factor, subject to twio times yearly reassessment: this consists in self-treatment thanks to a delayed prescription, following a (non-reimbursed) dipstick test to confirm cystitis. 2 A recent qualitative survey conducted in the UK pointed to the need for addressing physicians’ knowledge and skill gaps on UTI in women under the age of 65 years, including non-pharmaceutical recommendations for self-care. 45 Lastly, the request for more thorough, multidisciplinary management relying on various strategies is not in line with recommendations put forward by the SPILF for uncomplicated RC in 2014 2 : in non-menopausal women with a normal pelvic and urethral clinical examination, no further investigation is systematically required. In other situations, for women at risk for complications, management should be decided by a multidisciplinary team. Few women (25%) had consulted a urologist or a gynaecologist, while infectious disease specialist advice was very rarely sought. Clarification of this trajectory should result in more standardised approaches and reduce patients’ anxiety. Based on the requests put forward by the interviewees, an information leaflet should be made available and include the following items to meet the needs and expectations of patients with RC: A reminder of the definitions. Description of the known causes and risk factors of RC supported by a diagram and information on the usefulness and timeliness of urinalysis, dipstick test and urine microscopy and culture. Procedure to be followed when first signs of cystitis appear. Role of antimicrobial treatment and preferred compounds. Summary of possible non-antibiotic treatments. Clear suggested strategy for initial investigations/specialist referral. Such an information leaflet can be considered as a means of patient empowerment, as recommended by the SPILF in 2014 and which has still not been put into practice. It can contribute to patient education, aiming to involve patients in their healthcare and quality of life, while reducing inappropriate antibiotic use. The effectiveness of information leaflets has been demonstrated in various contexts, namely regarding paediatric antibiotic prescriptions when coupled with GP online training. 46 A patient–clinician shared decision-making leaflet was developed in the UK that addressed the consultation barriers and promoted patient empowerment, with both leaflet and corresponding explanations delivered by the physician. 47 Although the use of such a leaflet will depend on the physician’s wish to grant patients more autonomy and on the time available for dialogue, this may prove beneficial for both in the long run. Patients would have a clear, handy and relevant resource, which would contribute to reduce their anxiety by addressing several of their queries and describe the procedure to follow in case of cystitis. For previously selected and educated patients provided with back-up of urine microbiology and culture and/or antibiotics, the leaflet would increase autonomy. A printed resource facilitates memorisation and assimilation. Clear and explicit definitions can eliminate any confusion and thus prevent inadequate treatment. A diagram can help understand the links between risk factors and hygiene and dietary measures to adopt. Improved understanding should lead to better adherence to the suggested management strategy. A summary of various non-antibiotic treatments, whether validated or not, can meet patients’ request, describe how each should be administered and allow patients to test their respective effectiveness. Lastly, its use would contribute to improve to harmonise the currently highly disparate management approaches reported by patients. Such an approach reflects most patients’ demands. The leaflet ( online supplemental appendix 3 ) could thus contribute to train physicians from various specialties in good clinical practice. This could even be complemented with a specific resource for physicians. Assessment of this resource on a wider scale is now necessary, by distributing it to the RésO GPs and their patients, to confirm its relevance and consequently offer it to all women suffering from RC.