Background Indwelling urinary catheters (IUCs) placed in acute care are a leading cause of healthcare-associated urinary tract infection. Despite initiatives to minimise the placement of IUCs, levels of inappropriate use are still considered unacceptable. IUC practice is difficult to change, and factors influencing clinicians’ decisions need to be better understood.
Objective To explore why clinicians decide to place IUCs in acute medical care.
Methods We conducted a qualitative study in the emergency department and acute medical wards of a 1200+ bed hospital, undertaking 30 retrospective think aloud and 20 semistructured interviews with nurses and physicians who made the decision to place an IUC. A purposive sample and thematic analysis were used.
Results Opinions on when an IUC was warranted varied considerably. Inconsistency in decision-making was caused by differing beliefs on when an IUC was appropriate for each clinical indication. Numerous patient and non-patient factors, including clinical setting, resources, patient age and gender and staff workload, also impacted on each decision. Assessing when the benefit of an IUC outweighed the risk could be problematic due to conflicting goals.
Conclusions These findings help to explain why clinicians sometimes deviate from IUC best practice guidance and resist interventions to modify practice. In order to engage nurses and physicians in change, interventions to reduce IUC use should acknowledge and respond to the complexity and lack of clarity often faced by clinicians making the decision to place an IUC. However, it is equally important that inconsistencies in IUC-related beliefs are recognised, investigated and, where appropriate, challenged.
- Decision making
- Patient safety
- Qualitative research
- Quality improvement
- Infection control
Statistics from Altmetric.com
Despite being a leading cause of healthcare-associated urinary tract infection (UTI),1 indwelling urinary catheters (IUCs) remain overused in acute care.2 ,3 Recent point prevalence surveys in Europe,1 the USA4 and Australia5 indicate that 17–25% of patients in acute care have an IUC placed. Numerous attempts have been made to reduce use with some success, but the levels of IUCs considered to be inappropriate remain high. Recent studies have reported levels of inappropriate catheterisation to be 24%,6 46.4%2 and 62%7 of the total number of catheterisations in acute care settings.
The cornerstone of most interventions to reduce IUC use is a list of indications providing guidance on agreed upon reasons for placing an IUC. Given the current lack of empirical evidence on the appropriateness of IUC use, such lists are generally derived from guidelines based on expert consensus.8 ,9 Lists vary,10 but commonly accepted indications in acute care include (1) relieving acute urinary retention, (2) measuring urine output in critically ill patients, (3) perioperatively for some procedures, (4) assisting healing for open sacral or perineal wounds in patients with urinary incontinence (UI) and (5) improving comfort at the end of life.11
Interventions generally involve one or more methods to promote adherence to restricted IUC use, for example, introducing a programme of education, improving provision of resources to enable use of IUC alternatives and introducing a ‘catheter champion’.12–15 The highest profile initiative came from Michigan in the USA where the incidence of catheter-associated UTI was reduced by 25% by introducing a range of measures, referred to as the Bladder Bundle.2 The Bladder Bundle is one of the few initiatives that has employed a theoretical framework to support change, using a model based, in part, on the ‘4 Es approach’: engage, educate, execute and evaluate.16
Clinicians’ resistance to changing IUC practice in acute care has been acknowledged, but there have been remarkably few attempts to understand what influences decision-making. Some attempts have been made to understand the limitations of the Bladder Bundle.17 ,18 The key barriers to implementation were found to be difficulties with clinician engagement, patient and carer influence and clinical practices in emergency departments (EDs).17 Clinicians’ perceptions of risk were influential, notably that use of IUCs and IUC-associated risks were not always linked in clinicians’ minds. Competing priorities and organisational pressures led clinicians to find ways to work around IUC restrictions.18 These studies raised the need for pre-implementation assessment of practice before initiatives are introduced18 and to better understand clinicians’ concerns.17
Two studies have undertaken focus groups with specific clinical groups to consider the use of IUCs within EDs. One found that ED nurses did not consistently follow IUC guidelines and reported a lack of clear protocols.19 In the other, ED clinicians reported IUCs are overused and poor communication and understanding of practice is a barrier to change.20 Another study conducted semistructured interviews to explore clinician beliefs when placing a catheter with acute stroke patients and found practice varied considerably, unwritten rules dominated and clinicians perceived that clearer guidance was required.21
No published studies have examined clinicians’ decision-making for placing an IUC with individual patients or that compared clinical areas within an institution. Understanding clinicians’ decision-making is fundamental to designing effective initiatives to change practice.22 Consequently, the aim of this study was to explore why clinicians in acute medical care make individual decisions to place an IUC.
Study design and data collection
A qualitative study was undertaken to analyse the factors that influence clinicians in acute medical care to make the decision that an IUC is needed. The study took place within the English National Health Service (NHS). Data were collected from the ED, medical assessment unit, cardiology wards and older people's acute medicine wards in a large NHS hospital in England. A hospital where no recent quality improvement programmes aimed at reducing IUC use had taken place was chosen in order to explore pre-initiative decision-making. The units were selected to provide a diverse range of decision-making contexts. Two methods of data collection were used: retrospective think aloud (RTA) interviews and semistructured interviews.
During data collection periods, one researcher (CM) visited the participating units five times per week (including day and night) to ask if, within the last 4 h, the decision to place an IUC had been made for a patient over the age of 18, who had not had an IUC in the previous 24 h. Clinicians who had made the decision were invited to participate in an RTA interview within 4 h of the decision being made. These clinicians were asked to take part in a semistructured interview within 3 weeks. In the RTA interview, clinicians were asked to describe how they decided that a specific patient needed an IUC. Follow-up questions were used to gain further detail, for example, alternative options and patient involvement in the decision. Using an interview guide developed for the study and reviewed by clinical experts, semistructured interviews examined common reasons for making the decision to place an IUC, experiences of making the decision and what influenced decisions.
A purposive approach to sampling was taken to achieve a variety of indications and clinical circumstances. The sampling strategy was determined by the exploratory nature of the study. Although it was not feasible to collect data for all possible clinical circumstances or from the full range of clinicians in each department, the sample provided sufficient data to meet the study aim, delivering a detailed picture of decision-making.
We took a thematic approach to data analysis using constant comparison techniques.23 We were informed but not restricted by sensitising concepts resulting from the research question. The data were transcribed verbatim, coded, analysed and interpreted to provide a thematic summary of the phenomenon. The transcripts were reviewed line-by-line to identify sections of text to apply codes using a constant comparison approach. Material was iteratively rearranged as themes and subthemes were developed. Variations were highlighted, with divergent or contradictory cases explored further. Data from the two types of interviews were labelled to ensure that the interview source was identifiable and were analysed jointly to add depth of understanding. Analysis was overseen by all authors and consensus achieved on the development of themes.
In total, 34 clinicians were approached and 30 took part in an RTA interview, 8 nurses and 22 physicians (see table 1). In this context, registrars are physicians under specialty training, consultant physicians are practising a specialty and consultant nurses are senior, autonomous, specialist practitioners. Of these, 20 also took part in a semistructured interview. Four clinicians declined to take part in the RTA interviews, all giving time pressure as the reason for non-participation. Ten clinicians undertook an RTA interview, but not a semistructured interview due to workload/time pressure (five), absence from the work area (four) or change in role (one). Of the 30 decisions made, 18 related to male and 12 to female patients. All patients were aged over 60, except two women, one in her 20s and one in her 50s. Table 2 shows the breakdown of where decisions were made by the indication for IUC use given in the RTA interviews: relieving acute urinary retention (n=11), monitoring urine output (n=10), protecting skin (n=2) or a combination of two or more (n=7).
Differences found in the data provided by the two types of interview were central to a deeper understanding of clinicians’ decision-making. These differences are highlighted where relevant.
Five key decision-influencing themes were identified.
Indications for IUC use
This section describes beliefs on three commonly accepted indications, followed by clinicians’ descriptions of combining two or more weak indications together or with other goals to provide a plausible rationale for IUC placement.
Monitoring urine output
Participants presented individual decisions to place an IUC to monitor output as unequivocal clinical choices, with no doubt for the need of the device. In the semistructured interviews, many clinicians reported that using an IUC for this purpose was often a knee-jerk or default decision. A consultant physician in ED admitted “I really don't think about the decision at all". It was found that junior physicians learn to consider placing IUCs in the ED for urine output monitoring as part of their routine patient assessment. One staff nurse in ED expressed frustration, stating,
If a patient's difficult to catheterise they sometimes say oh don't bother we don't really need it now and you think either she does or she doesn't. (ED staff nurse)
One physician drew attention to the comfort provided to clinicians by numbers on an hourly urine output chart, but questioned the need to monitor so closely. However, alternatives to measuring output were not considered reliable or feasible, with one consultant asking, “What happens if that [urinary] sheath does come off and that precious urine output is lost?” (ED consultant physician). Summarising her views on this matter, another clinician noted, “Sometimes, it's just easier to stick a tube in” (medical assessment unit registrar).
Relieving acute urinary retention
Beliefs on the amount of urine in the bladder that warranted the use of an IUC varied from 200 to 500 mL (as measured by a bladder scanner) and most clinicians did not require symptoms of retention to be present. Many physicians (both junior and senior) expressed a desire for greater clarification for the appropriate threshold for IUC use. The key reason that clinicians were ‘very twitched to get a catheter in’ (older people's medical wards consultant physician) was that “It is a disaster if it's missed” (ED consultant physician). Horror stories of the consequences of missing urinary retention were imprinted on the minds of many clinicians, with one physician recalling two separate situations where elderly patients had been admitted with abdominal pain, but no one had asked if they had passed urine, stating, “Nobody was listening to them and they were both in huge amounts of retention” (medical assessment unit registrar). Another physician recalled a patient who had needed bladder surgery following missed urinary retention.
Protecting skin integrity
This indication was associated with polarised opinions. A minority of clinicians stated that if the patient was incontinent of urine an IUC might be appropriate to protect skin integrity, even when there was no existing skin damage, as described below:
It's not a great for someone who's wet with potential pressure areas and lying in their own wee, so, yes it can be argued. (ED consultant physician)
At the other end of the spectrum, some clinicians, predominantly in the older people's medical wards, believed using an IUC for skin integrity reasons was rarely justified.
The language used was noteworthy; none of the interviewees referred to moisture lesions, incontinence-associated dermatitis or other similar description. Commonly IUCs were described as protecting ‘pressure areas’.
Combined indications and goals
During RTA interviews, participants revealed that in 7 of the 30 decisions commonly accepted indications, such as monitoring urine output or protecting skin, were either combined together or with goals not considered by IUC guidelines11 to provide acceptable indications (eg, preventing UI, assisting with patient comfort and dignity or providing a restful night's sleep). The following quote from a registrar illustrates how clinicians combined seemingly weak, commonly accepted indications with other goals to make a plausible decision:
One it helps us manage her inputs and outputs and two, it would be for management of her pressure areas because we wouldn't be able to control her urinary continence and keep her comfortable. (medical assessment unit registrar)
A clue to the unstated use of combined indications was clinicians’ views on when IUCs would be removed. During the RTA interviews, clinicians were asked when they anticipated that the IUC would come out. The usual response was along the lines of “when they've improved". These vague reasons did not correspond with the rationale for catheter insertion.
Resource availability and workload
The availability of incontinence management products varied between clinical areas, directly impacting on IUC decision-making. The ED had a limited range of continence management products available, providing context to comments made in the ED about patients benefiting from IUCs to manage UI for comfort and dignity. One ED staff nurse commented, “We don't get full pads down here. We did get them once but we used too many of them, so that's a bit of a problem”. Bladder scanners were in also short supply and, in some wards, kept under lock and key due to their scarcity.
The indication most frequently linked to workload was the management of UI or what was regularly referred to as ‘nursing care’ or ‘nursing issues’. Physicians tended to acknowledge that IUCs were used to reduce workload for nursing staff, accepting this as an unfortunate but pragmatic option. However, the nurses interviewed maintained that IUCs were not necessary for this purpose, with one commenting, “from my point of view, it's laziness” (ED staff nurse).
By contrast, some physicians acknowledged the role their own workload could play. One junior physician reported catheterising a patient to find a residual volume of only 100 mL in the bladder, but leaving the IUC in situ as he felt there was a risk of the patient going into urinary retention in the near future, commenting, “You tend to leave it in, because you're on-call and you've popped something in, to take it out again seems fairly nonsensical, because it can only add to your work load later”.
Clinicians reported that IUCs were used to avoid a range of potential negative incidents. A key contingent event that clinicians were trying to avoid was UI or toileting problems. Participants reported that this was a factor in several of the combined decisions. Many clinicians admitted that IUCs provide a pragmatic option to manage UI in some circumstances, often linked with patient comfort, as illustrated below:
If you're transferring someone that's critically ill to CT, I would be more likely to catheterise someone than not simply because you've got a patient who is potentially going to wet themselves in the CT scanner and that's going to cause problems for the CT scanners, it's going to cause problems for the patient, you don't want them lying in their own urine. (ED consultant physician)
Notably, the patients who clinicians felt would benefit from having an IUC placed to manage their urine output were usually not incontinent of urine. Instead, they were acutely ill with decreased mobility and had lost the ability to be independent with their toileting needs. IUCs were being placed to help them adapt to a, hopefully, temporary situation.
In the majority of instances, clinicians were trying to avoid a negative event for a patient, for example, to avoid the potential of the patient's skin breaking down due to poor care or to avoid retention being missed. However, for a sizeable minority of examples, the adverse consequences would be for the clinician. One example was the fear expressed by junior physicians that they would be criticised by senior colleagues in ED and medical assessment unit if they did not place an IUC to monitor urine output. They would, “err on the side of caution” (ED junior physician) if they were unsure whether it was needed. One registrar explained that, “It's easier to justify putting one in than not putting one in” (medical assessment unit registrar). One senior physician confirmed that he would be unlikely to criticise a junior colleague for being overzealous placing IUCs.
Compared with the wards, clinicians in the ED (and to a lesser extent, the medical assessment unit) were more likely to make the decision to place an IUC at a lower threshold. Junior and mid-level physicians acknowledged this difference, attributing it to time pressure, clinical uncertainty and reliance on routine practice in the ED.
Several clinicians stated that they believed that the decision to place an IUC in ED was often part of a bundle of care, for example, for the management of sepsis. Although junior and mid-level physicians recognised the higher level of IUC use in the ED, senior physicians did not, sometimes commenting on overuse elsewhere, demonstrated by the comment, “Also I think at the back of my mind is the likelihood is that they're going to get catheterised in the near future any way when they hit the wards for a management reason” (ED consultant physician).
Patient age and gender
All clinicians agreed that patient age and gender influenced decisions to place an IUC (“There's a massive perception that catheters are for old ladies”, ED junior physician). It was widely acknowledged that older patients were often assumed to be frail or immobile, increasing the likelihood of receiving an IUC. By comparison, many clinicians reported an aversion to catheterising young patients, with one junior physician stating,
People don't like putting catheters in younger people even if they're really, really sick. It was awkward for him and it was awkward for me. I was like why, I've never been awkward putting a catheter in, but I'd never put one in a 26 year old guy before. (ED junior physician)
Men were less likely to get an IUC partly because there are more alternatives to managing UI such as urinary sheaths and it is easier for men to pass urine when less mobile, but also because in the clinical areas in this study it is usually a physician who has the job of placing the device. Some nurses believed that physicians would easily agree to women being catheterised, but would try to avoid male catheterisations, typified by this comment on the subject, “If you're being particularly cantankerous then you'll think up ten reasons why they don't need one” (ED junior physician).
The findings of this study have shed light on why clinicians sometimes deviate from IUC best practice guidance and resist interventions to change practice. With the influence of a multitude of beliefs and contextual factors, clinicians’ decisions to place an IUC in an individual patient can be complex and the choice does not always fit comfortably into the category of appropriate or inappropriate. Instead, clinicians combine cues from multiple domains to reach a plausible decision.
Fundamental to this decision are clinicians’ beliefs about the indications commonly considered appropriate. Three of these indications (monitoring output, relieving acute urinary retention and protecting skin) were used in this study and beliefs on when an IUC was warranted varied considerably. It is likely that the lack of empirical evidence to support the indications8 ,9 contributes to this level of variation.
It is evident that providing clinicians with a more prescriptive list of indications might reduce catheter use in some cases, for example, in the case of acute urinary retention, providing clinicians with guidance about the quantity of urine that should be in a bladder before catheterising. However, findings also indicated that clinicians drew from multiple cues to conclude that a patient would benefit from an IUC and would combine weak indications together or with other goals to reach a plausible justification for placing the device. In these circumstances, clinicians have goals unrelated to any of the commonly accepted indications, helping to explain why attempts to minimise inappropriate use by restricting indications have had some success, but levels of non-adherence remained high.2 ,7 ,24 Goals and cues influencing clinicians are highlighted in the other four themes.
Successfully avoiding or finding alternatives to IUCs can require additional resources such as bladder scanners25 or urinary sheaths (also known as condom catheters),26 and the lack of availability of these products influenced clinicians’ decision-making. More ambiguous was whether clinicians believed that IUCs can be time-saving devices and whether workload considerations impacted on the decision-making. Previous studies have focused on the impact on nursing workload, with mixed findings.18 ,20 ,21 A survey27 following the implementation of an initiative to reduce IUC use asked nurses how they would rate the effect of the ‘No Foley’ programme in terms of the difficulty of their job and 45% said it made it worse, 25% better and 30% gave no feedback or said it had no effect, indicating that IUCs have at least some impact on workload. In this study, many clinicians acknowledged the use of IUCs to manage workload as a clinical reality, but claimed it did not impact on their own decision-making, with the exception of junior physicians who were frank about their occasional use of IUCs to help manage their workload.
An important goal of IUC use was the avoidance of contingent events for both patients and clinicians. Previous studies have found that clinicians use IUCs to manage risks, principally to avoid falls18 ,21 or inaccurate measurement of urine output.18 This study has shown that the perceived protection provided by IUCs can be viewed more broadly as shields against diverse potential harms. In particular, in some circumstances, both nurses and physicians sought to remove from the patient the burden of manage their own bladder emptying and anxiety over UI, issues known to threaten the dignity and comfort of patients in hospital.28 It was evident that the avoidance of embarrassment, discomfort and loss of dignity for patients was an influential factor that should be considered when devising interventions to reduce IUC use. However, the link between the management of UI and workload, particularly for nurses, cannot be ignored, and it is difficult to assess the extent to which clinicians were also avoiding contingent events for themselves.
Beliefs on indications, avoiding contingencies and managing workload varied with the clinical environment and the patient. It has previously been observed that the overuse of IUCs frequently occurs in EDs.15 ,20 ,24 In this study, the issues of time restrictions and clinical uncertainty in the ED were identified to promote IUC placement decisions. This study also supports previous findings18 that beliefs about IUC overuse elsewhere in the hospital encourage IUC use and that the usual rules should not apply in the ED. It is known that older women are at greater risk of catheterisation.15 All clinicians in this study agreed that age was a factor in decision-making, but an interesting point was the lengths clinicians would go to avoid catheterising a young patient. Similarly, physicians were found to go to some lengths to avoid catheterising men.
The inconsistency between different patient groups and clinical areas, combined with the variation in interpretation of indications and IUC-related goals, suggests that there is room for reducing IUC use. To achieve maximum potential, IUC-reducing initiatives need to explore, understand and address clinicians’ decision-making processes for each indication and other goals. Identifying clinicians’ challengeable beliefs could enhance the efficacy of interventions. Examples of such beliefs reported by senior physicians in the ED in this study include:
if there is the potential for UI, then an IUC should be placed;
UI cannot be effectively managed in the ED without an IUC;
IUCs should be used routinely in some circumstances;
older people are more likely to benefit from an IUC;
‘pressure areas’ can be protected using IUCs;
junior clinicians are not under pressure to place IUCs.
Interventions that do not acknowledge and respond to the complexity of the decision to place an IUC are likely to continue failing to engage clinicians and this has been identified as a key limiting factor when attempting to changing practice.29 ,30 Using the example of the ‘4 E’ model, the addition of a fifth E, ‘explore’, before engage, educate, execute and evaluate would improve understanding of clinicians’ motivations, goals and priorities.
This is the first published study to present the results from RTA interviews on individual IUC decisions, to combine these findings with broader, semistructured interviews and to examine decisions made in an ED in contrast to those made in hospital wards.
A key strength of this study lies in combining the two interview methods allowing new insight into the realities of IUC decision-making by providing the opportunity to compare clinicians’ decision-making for individual patients with their views on practice more generally. Each interview method unearthed detail that was not found by the other, for example, the use of combined indications was revealed by the RTA interviews and not the semistructured. The study has limitations. First, it was undertaken in one hospital and findings might not be generalisable to other settings. In particular, it could be expected that findings from a hospital where a recent IUC reduction initiative had taken place would vary considerably. Second, although clinicians appeared frank and open in their responses, the data are self-reported and liable to social desirability bias.
This work has important clinical and research implications. Globally, considerable resources have been put into reducing IUC use. The success of these attempts has been limited, and future initiatives should consider the factors highlighted in this paper to identify the roles that IUCs play in the care of individual patients and establish how their use could be influenced to improve the quality of care. In particular, in order to engage clinicians in change, attempts to modify practice must be underpinned by an understanding of the often equivocal and complex nature of the decision and the multiple patient-centred and non-patient-centred factors at play. However, equally, it is important that inconsistent IUC-related beliefs and variation in interpretation of indications should be identified, explored and, where necessary, challenged.
Contributors All authors contributed to the development of the study design, gaining ethical approval, analysing results and writing this paper. CM undertook the data collection and transcription.
Funding This paper is independent research by the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care (CLAHRC) Funding Scheme.
Disclaimer The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.
Competing interests None declared.
Ethics approval NHS Research Ethics Committee (Wales).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.