Patient decision‐making and regret in pilonidal sinus surgery: a mixed‐methods study

Little is known about optimal management strategies for pilonidal sinus disease (PSD). We conducted a mixed‐methods study to understand why patients make, and sometimes regret, treatment decisions.


INTRODUC TI ON
Pilonidal sinus disease (PSD) affects 26 in 100 000 people, predominantly men of working age [1]. Obstruction and rupture of hair follicles in the natal cleft lead to abscess and sinus formation, progressing through insertion of hairs into the sinuses [2,3]. Patients present to emergency services with a painful abscess between the buttocks or electively with a chronic cycle of pain and discharge [4].
The proliferation of small, single-centre studies, using different classification systems and outcome assessments, means that there is no clear consensus on what constitutes optimal treatment-one that most general surgeons could perform, which results in rapid healing and minimal complications [5,6]. If surgical treatments are perceived as ineffective, individuals with simple sinuses may be asked to wait for their condition to deteriorate or offered medical management, before surgical intervention [7][8][9][10]. On emergency presentation, a pilonidal abscess requires hospitalization for its incision and drainage, following which one in five individuals re-present with recurrent symptoms [11]. Treatment of chronic PSD typically involves excision of affected skin and fat and management of the wound, if it is not closed using fibrin glue [12]. The most common excisional approaches leave sizeable open wounds [13,14]. The need for prolonged wound care over a period of months impacts education, work, intimacy, social life; recurrent infection, fear of wound deterioration and pain can profoundly affect quality of life [15,16].
Treatments may be advocated without evidence of clinical or cost effectiveness [17] while inadequate preparation for postoperative wound care, uncontrolled pain and delayed wound healing affect patient well-being and activities of living [18,19].
For these reasons, in 2018 the UK National Institute for Health Research commissioned the Pilonidal Sinus Treatment-Studying the Options (PITSTOP) study. The centrepiece of this work was a large observational cohort intended to understand the effectiveness of common excision and closure techniques used in UK practice [20,21]. To enhance our understanding of patient decision-making and responses to treatment we undertook a nested mixed-methods sub-study, triangulating findings from semi-structured interviews with patient-reported experience measures (PREMs). The objectives of the sub-study, reported in this paper, were an overview of patient views and experiences, including (1) how treatment decisions are made and why they might be regretted and (2) attributes of interventions they would rather avoid and which outcomes they most value.

ME THODS Design
Case study methodology is a mixed-methods research design used for in-depth examination of complex phenomena in real-world settings [22]. The study was a multiple case design: it compared more than one data type between and within more than one person [23].
The case study was nested in an observational cohort, with two embedded units of analysis-longitudinal semi-structured interviews at baseline and 6 months, and quantitative cohort data at the same time points.

Case selection
All participants with symptomatic PSD, referred for elective surgical treatment and participating in the PITSTOP cohort, were invited to take part in a semi-structured telephone interview at baseline (either before or immediately after surgery) and 6 months after surgery. We sampled for maximum variation based on the Wysocki classification (an indicator of disease severity) the method of surgical excision and closure [20]. Initial contact was made by telephone, which was followed by an email and an information sheet.

Survey instruments (PREMs)
At baseline, participants completed the CollaboRATE three-question PREM of shared decision-making (SDM), scored 0 (indicating poor SDM) to 9 (indicating good SDM) [24]. Six months after surgery, we collected pain, length of time to healing, post-surgery complications and a decision regret (DR) PREM, using a five-point scale scored 0 (low DR) to 100 (high DR), measuring healthcare decision regret [25].

Interviews
Semi-structured telephone interviews were conducted between June 2019 and September 2020 by EB and ES. A minimum of 20 interviews was considered adequate to understand common perceptions and experiences of treatment choices, thereby achieving thematic saturation [26,27]. Interviews were recorded on encrypted digital recorders and transcribed. Baseline interviews adapted key 'choice' (e.g., 'did you let the surgeon choose your treatment?') and 'options' (e.g., 'did the surgeon talk you through the risks and benefits?') questions from the coping in deliberation (CODE) framework (Table 1) [28]. At 6 months, the interview guide asked CODE questions related to decision 'consolidation' (e.g., 'was this the right decision?'). Throughout, probing questions covered dimensions of Sekhon's acceptability framework [29] as well as intervention attributes, to inform a discrete choice experiment (to be published separately).

Analysis of interviews
ES, TC, EB and DH used the National Centre for Social Research 'Framework' analysis approach which, unlike some other analytical approaches, allows for coding of a priori and de novo themes [30].
After familiarizing ourselves with the transcripts, we independently coded a sample of transcripts using NVivo (QSR International) version 11 before conferring. Integration of qualitative and quantitative data occurred during analysis and interpretation, to understand (1) how disease characteristics and surgeon preferences interacted with patient values in treatment choices and (2) how participants appraised treatments given particular outcomes. We used joint display tables to look for convergences and divergences between cohort data (disease features/treatment choices/outcomes) with experiences, views and values [31,32]. We invited a patient expert to provide feedback on a lay summary of triangulated results from interviewees.

Health threat
Newly diagnosed participants, unfamiliar with PSD, expressed confusion about its cause and prognosis. They discussed soreness, inflammation, discharge and odour, disrupting employment, exercise habits, social life and intimacy. They reported negatively impacted well-being and mental health. Whilst newly diagnosed participants assumed it was an acute, curable condition, those with recurrent disease understood that recurrence was common.
I was told initially, 'Oh that could be it, and then it might go away'… but once you get it once, that's it: it's coming back… If I was a bit more aware of that I would have probably started to look into the surgeries quicker.  as not taking PSD seriously and failing to escalate their case to secondary care. One presented with recurrent disease, which was disputed by the GP. …just gave me some antibiotics… it just kept getting more painful and worse… I went back three times… then she put me on sort of the path to go back to TA B L E 1 Description of constructs from the coping in deliberation (CODE) and acceptability frameworks

Choice and options
Where participants were given a choice, preferences were con-  (Table 3, participant   15). Some reported using significant others, friends, relatives or the internet to support decision-making, with some deriving a sense of control (self-efficacy; intervention coherence) from researching the condition and treatment options.

Key outcomes at the time of decision-making
Around the time of surgery, not every participant would specify a single most important outcome (Table 3). However, the following outcomes were highlighted during decision-making: avoiding recurrence (n = 8), return to normal activities (n = 6) and/or the elimination of symptoms (n = 7). Six participants were not aware of procedural risks; others expressed awareness of risks presented by anaesthesia (n = 2), infection or bleeding (n = 4), the wound not healing (n = 5) and recurrence (n = 8).

Consolidation
After surgery, most participants were fearful of aggravating the wound and/or delaying healing. They made physical adaptations (altered sitting and reclining positions) and behavioural adaptations (reducing duration and type of exercise), which negatively affected their well-being.
It has made me reticent to engage in some activities… exercise and things like that… through the pain and discomfort, and also the chance of sort of popping the cyst….
(1: no previous pilonidal disease) Participants visited the GP or were visited by the district nurse for wound care, daily or weekly. Because of its location, many had difficulty attending to the wound themselves: they used mirrors or were reliant on others for daily examination and wound management (including cleaning, dressing and packing the wound), often resenting this loss of independence (self-efficacy) and expressing embarrassment.
On the other hand, the emotional support provided by participants' social networks was important in alleviating emotional distress.
I think the worst part of it is that you always have to rely on someone else to do, like, a dressing for you… you can't drive cos you can't sit down… you basically you can't do anything.

Key outcomes at the time of follow-up
Six months after surgery, people recalled hoping surgery would address pain (n = 3), recurrence (n = 5), wound healing (n = 1), the smell

Notes:
The key outcome is the primary desired outcome for each participant expressed at the baseline interview. CollaboRATE score is the mean collaboRATE score regarding shared decision-making of treatment recorded at baseline; high score, more shared decision-making, low score, less shared decision-making (0-9); the table reports collaboRATE scores low-high. The sample quotes are taken from interview at baseline.
Abbreviation: ADL, activities of daily living.  Notes: Pain/post-surgery complications were recorded at the 6-month follow-up. Score is the decision regret score regarding treatment decision recorded at 6-month follow-up: high score, high decision regret; low score, low decision regret (0-100); the table orders participants in low-high decision regret scores. The sample quotes were taken from the 6-month follow-up interview. Abbreviation: LTFU, lost to follow-up. (n = 1), the inconvenience (n = 1) and impaired ability to perform activities of daily living (n = 1). In five cases (Table 4, In a closed question-one in which the range of potential answers was limited-participants felt that the single most important outcome was the wound healing in the expected time (n = 3), avoidance of recurrence (n = 4) and return to activities of daily living (n = 1). Three participants who had undergone PSD surgery for the first time found their treatment to be effective and, when asked, did not think they would have done anything differently (Table 4, participants 5, 10 and 11).
Wound healing took a median of 51 (8-112) days (Table 4). One participant reported that the wound had not healed 6 months after surgery ( Table 4,   Qualitative and quantitative data from the other (18) was discordant.
On the decision regret PREM, the participant was ambivalent as to whether the decision was right, whether she regretted the decision and whether the choice did her harm. The transcript was more positive but demonstrated acute anxiety about possible recurrence.

DISCUSS ION AND CON CLUS IONS
This study explored how patients make, and sometimes regret, PSD This study does not attempt null hypothesis significance testing using quantitative data, and is reasonably large by qualitative research standards [27], the largest yet conducted in people with pilonidal sinus [7,19]. The sample is satisfactory for the application of existing theory to empirical data [33]. While the participants' surgical treatment is representative of current UK practice, emerging minimally invasive techniques increasingly common in other health systems are not represented; their acceptability should be the subject of future research.
As post-surgical wound healing can take over 6 months and recurrence of PSD may take place over many years, our study is limited, and the attitudes of participants may be affected by its short follow-up period. Like other contemporary studies [34,35] the COVID-19 pandemic limited our ability to follow up participants.
Even where remote data collection was possible, asking people to engage with research activities was difficult, given the burden placed on them by the pandemic [36]. Attrition rates are poorly reported in PSD studies but thought to be high, due to the young, mobile, mainly male population [5,37] a widely reported challenge [38,39]. Attrition may also result from reluctance to express negative thoughts [40] or loss of interest in research after wound healing [41].
Our mixed-methods approach identified divergences and in- Surgeons [45,46] and patients [47] may have reasons for avoiding shared decision-making, and our findings complicate the common assumption that shared decision-making leads to increased decisional satisfaction [48]. Systematic reviews in other contexts suggest that unmet information needs are common and distressing [49][50][51][52]. There are growing concerns that self-report measures of shared decision-making may not capture the quality of the interaction or the multi-staged nature of the process [53,54]. PREMs may be compromised by social desirability or acquiescence bias [55][56][57][58] and open-ended questions may reveal significant problems from patients who report high levels of satisfaction on survey instruments [59,60]. Triangulation of research methods is useful to identify such problems [61,62].
Clinical teams should ensure patients are properly informed about available surgical techniques and manage their expectations about aftercare and the uncertainties which surround clinical outcomes. Surgeons may not actively engage in wound care discussions because that is seen as the responsibility of primary care services [17]. But information gaps reduce patients' ability to self-manage and teams should ensure that patients receive verbal and written information, tailored to their needs, at the right time [63]. Surgical consultations may concentrate more on interventions than outcomes, but asking patients about their expectations gives surgical teams the opportunity to address false optimism [64][65][66]. In other settings, patient expectations predict satisfaction and functional outcomes following surgery [67].
Awareness-raising among primary and secondary care is needed to avoid delays in treatment where PSD is poorly recognized. Where pilonidal surgery is seen as unglamorous [68] or surgeons only specialize in one technique [69] patients with recurrent disease should be referred rapidly onward to genuine specialists. Both shared decision-making and the consent process itself are compromised if patients are poorly informed about their condition, available treatments and the probability of various outcomes [45]. This is challenging when there are many available treatments supported by variable evidence [5]. There are around 20 systematic reviews and meta-analyses on surgical techniques alone, and around 15 more on medical, wound care and other topics. An overview of these reviews should be an urgent research priority to adequately inform shared decision-making and the development of decision support tools.
Until then, the review by Stauffer and colleagues remains one of the most comprehensive overviews focusing on time-to-recurrence with different surgical techniques [70]. Finally, discharge planning should begin at pre-assessment visits, involving the patient, day surgery nurses and district nurses [71][72][73]. Postoperative wound care is enhanced by continuity of care from a limited number of communitybased health professionals [74].
In conclusion, giving people with pilonidal sinus better information about the burden of wound care and the risks of recurrence associated with different surgical approaches may improve decision-making and minimize regret. An overview of systematic reviews is needed to inform decision support tools. Surgical teams should fully communicate uncertainties about treatment effects, as well as the timescale, tasks and psychosocial issues associated with recovery.

INFORMED CONS ENT
Informed consent was obtained from all individual participants included in the study. This paper is submitted as an original article.

ACK N OWLED G EM ENTS
We would like to thank all the participants who we interviewed.

CO N FLI C T O F I NTE R E S T S
The authors declare no conflict of interest.

E TH I C A L A PPROVA L
The study received approval from East of England-Cambridge South Research Ethics Committee (REC reference 18/EE/0370).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.