PERSPECTIVE Version 4.0
Research type
Research Study
Full title
A questionnaire survey to explore the patient perspective of functional bowel disorders (irritable bowel syndrome / constipation / diarrhoea / faecal incontinence / evacuatory dysfunction): effects on lifestyle, experience of transanal irrigation and effects of the COVID-19 pandemic.
IRAS ID
296856
Contact name
Yan Yiannakou
Contact email
Sponsor organisation
County Durham and Darlington NHS Trust
Duration of Study in the UK
3 years, 0 months, 0 days
Research summary
Research Summary:
What was PERSPECTIVE?
PERSPECTIVE stands for Patient pERSPective of functional bowel disorders: Effects on lifestyle, experienCe of Transanal Irrigation and effects of the COVID-19 pandEmic.
It was a research study that asked people living with functional bowel disorders (FBDs) to share their experiences through a detailed questionnaire. FBDs are a group of gut conditions — including irritable bowel syndrome (IBS), functional constipation, and functional diarrhoea — where patients experience persistent, often debilitating symptoms without there being a clear structural or biochemical cause that doctors can point to on a scan or blood test.
The study was run by the Durham Bowel Dysfunction Service at University Hospital North Durham, and it drew on two existing groups of patients: people registered on the ContactMe-IBS national registry, and patients on the service's own transanal irrigation (TAI) database. Questionnaires were sent to 4,739 people in late 2021, and 1,175–1,176 responded — a response rate of around 25%. The average age of respondents was 52 years, and the large majority (around 85%) were female, which is consistent with what is known about who these conditions affect most.
The questionnaire covered a wide range of topics: what symptoms people had and how severe they were, how their condition affected daily life and work, their mental health and wellbeing, how they accessed and experienced healthcare, how COVID-19 had affected them, and how effective various treatments had been — including TAI.
The study produced two published papers. One focused specifically on the findings about TAI. The other took all the data and analysed it using advanced computer methods — machine learning and a type of network analysis — to look at the bigger picture of what shapes a patient's experience of living with a functional bowel disorder.
________________________________________
Paper 1: Transanal Irrigation for the Management of Functional Bowel Disorders What is transanal irrigation?
Transanal irrigation (TAI) — sometimes called rectal irrigation — is a treatment in which a controlled amount of water is introduced into the bowel via the rectum using a specialist kit. The idea is to flush out the lower bowel on the patient's own schedule, which can help restore a degree of control over bowel emptying. It is a specialist treatment, usually introduced by a nurse or doctor with expertise in bowel dysfunction.
What did the study find?
Of the 1,175 or so questionnaire respondents, 183 (about 20%) had experience of using TAI. Of those, 110 (60%) were still using it at the time of the survey; 73 (40%) had tried it but stopped.
A 60% success rate. The fact that 60% of those who started TAI were still using it is itself a meaningful finding. TAI is time-consuming, requires motivation and a degree of planning, and needs the patient to fit it into their daily routine. It is reasonable to assume that people only keep using it if it is genuinely helping them. This 60% continuation rate — used as a surrogate for effectiveness — is consistent with findings from other studies of TAI.
Long-term benefits. Many patients had been irrigating for a very long time. More than half (53%) had been using TAI for three years or more, and a significant proportion had been using it for five to twelve or more years. This strongly suggests that, for many people, TAI continues to work over the long term rather than just providing short-term relief.
Early response predicts long-term success. Effectiveness at three months was strongly correlated with effectiveness at six months (a strong positive correlation, R=0.77), and there was a clear trend for this to continue through to twelve months. In practical terms, this means that if TAI is helping someone at three months, it is very likely to still be helping them at twelve months. Guidelines already recommend reassessing TAI at three months and referring on if it is not working — these findings support that approach and may give patients reassurance that if they are doing well early on, they are likely to keep doing well.
TAI for diarrhoea and mixed symptoms — a new finding. Most previous research into TAI has focused on constipation. One of the most significant findings from PERSPECTIVE was that TAI also appears to benefit people with diarrhoea-type symptoms and mixed or alternating bowel habits. Among those using TAI for diarrhoea symptoms, three quarters (74%) were still irrigating — the highest continuation rate of any symptom group. This is an area that has received very little research attention to date, and the findings suggest it warrants serious investigation.
People wait a very long time before starting. Almost half of current TAI users (44%) had been living with their symptoms for ten years or more before starting TAI. The study suggests this long delay is largely due to the time it takes for patients to reach specialist services — typically ten to fifteen years after symptoms begin. This may reflect patients not seeking help early, complex journeys through the healthcare system, or TAI being regarded as a "last resort" rather than an option to consider earlier. Raising awareness among healthcare professionals about the potential benefits of TAI across all types of FBD could help get patients started sooner.
Most patients irrigate at a frequency that suits them. Irrigation guidelines suggest that once a good routine is established, patients can reduce from daily to alternate-day irrigation. The study found that people naturally adapt their frequency to what works for their symptoms and lifestyle — ranging from every day to as infrequently as once a month or less. A substantial proportion (35%) were long-term, infrequent irrigators, suggesting that even a less intensive regimen can be effective over time.
The most common reason for stopping was that it did not improve symptoms. Of those who stopped, the most common reason was that TAI simply did not help their bowel symptoms. Other reasons included increased abdominal pain, difficulty with the equipment, and (occasionally) cost — particularly prescription costs. Some patients had stopped and then restarted, perhaps when no alternative treatment proved effective.
Gender and diagnosis. The vast majority of TAI users were female (91%), reflecting the preponderance of women among people with FBDs generally. The most common diagnoses among TAI users were functional constipation and mixed IBS. Male irrigators were more likely to be using TAI for diarrhoea-type symptoms. The numbers of male respondents were too small to draw firm statistical conclusions about gender differences in effectiveness.
________________________________________
Paper 2: Deep Phenotyping of Patient Lived Experience in Functional Bowel Disorders Using Machine Learning What does "deep phenotyping" mean?
A phenotype is simply a description of what a person or condition actually looks and feels like in practice. "Deep phenotyping" means going well beyond the usual clinical label — such as "IBS-C" or "functional diarrhoea" — to build a richer, more detailed picture of how a condition affects every aspect of someone's life. The "machine learning" part refers to using computer algorithms to find patterns across large amounts of complex data that would be impossible to spot with conventional statistical methods.
What did the study do?
The researchers took the full dataset from the PERSPECTIVE questionnaire — all 1,175 patients and 59 different pieces of information about each one — and applied two types of sophisticated analysis.
The first was a set of machine learning models (specifically, a type called XGBoost). For each of the 59 patient factors — things like personal health rating, severity of anxiety and depression, employment status, frequency of GP attendance, and effectiveness of various treatments — the computer was asked: how well can we predict this one thing from all the other information we have about the patient? And which pieces of information are the most useful for making that prediction?
The second was a Bayesian network analysis — a way of mapping out all the connections between different patient factors and understanding which ones are most central and influential, and which ones are merely connected to lots of others without actually driving much.
What did the models find?
Healthcare use and life impact were the most predictable things. The model predicting how often a patient attended for bowel-related healthcare was the strongest performer, explaining 71% of the variation between patients. The impact of symptoms on daily activities, on work, and on activities of daily living were also predicted well. This tells us that the factors shaping healthcare use — how much a condition encroaches on a person's working and daily life, their mental health, and whether they are already known to the service — are relatively consistent and identifiable from routine information.
Treatment response was much harder to predict — but there was a striking pattern. The models predicting how effective individual treatments were (laxatives, diet changes, pelvic floor exercises, suppositories, irrigation, probiotics, and so on) were more variable in their performance. However, a very important finding emerged: the best predictor of whether a patient would respond to any given treatment was whether they had responded to other treatments. In other words, there appears to be a group of patients who tend to respond to multiple different interventions, and another group who are refractory — that is, who do not respond well to anything they try. It is this refractory group who typically end up in specialist care, and the study suggests that understanding what makes a patient refractory should be a major focus for future research.
What matters most for how patients feel?
This is perhaps the most clinically important finding of the whole study. Doctors and nurses working in this field often assume that the most important driver of how unwell a patient feels is the severity of their gut symptoms or the specific diagnosis they carry. The machine learning models told a different story.
Diagnosis was rarely selected as a useful predictor. Across all the modelling tasks — predicting health ratings, quality of life, pain, life impact, and treatment response — the patient's diagnostic label (IBS-C, IBS-D, functional constipation, and so on) was almost never chosen by the computer as one of the top five most useful pieces of information. This is striking, because it is the diagnostic label that most clinical guidelines, treatment pathways, and prescription decisions are built around.
What did matter? The factors that most powerfully predicted how patients felt and how their lives were affected were:
• The impact of symptoms on daily life — whether they could manage usual activities, care for themselves, get around, and sleep.
• Mental wellbeing — levels of anxiety and depression were among the strongest predictors of how patients rated their health and quality of life.
• Employment status — whether someone was in work was one of the most powerful predictors of overall life impact, and vice versa: the impact of bowel symptoms on daily activities was itself the strongest predictor of unemployment.
• Age — being younger when living with an FBD was associated with a worse perception of health, a finding that had not been highlighted before.
None of the top five predictors of a patient's personal health rating specifically interrogated gut symptoms. This is a radical finding — it suggests that when a patient tells their doctor how unwell they feel, the most important things shaping that answer are not the frequency of their diarrhoea or the severity of their constipation, but what those symptoms are doing to their life, their mind, and their ability to work.
Mental health and healthcare access were closely linked. The model predicting the impact of COVID-19 on mental health in this group found that the strongest predictor was frequency of attendance for bowel symptoms — in other words, those whose mental health suffered most during the pandemic were those who had previously been attending most regularly and were therefore most affected by reduced access to healthcare. This underlines the importance of maintaining and widening access to specialist gut services, not just for bowel reasons but for mental wellbeing too.
What did the network analysis show?
The network analysis produced a "map" of how all the different patient factors are connected to one another. It showed that the patient factors cluster into four broad communities: the diagnostic or nosological domain (the labels we give to the conditions), life impact, treatment effects, and symptoms.
The most provocative finding from the network analysis was this: diagnostic labels function as "hubs" in the network — they are connected to many other things — but they are not "authorities." In network terms, a hub links to many other nodes, but an authority is something that genuinely shapes or influences others. The network analysis showed that while the diagnostic category of a patient connects to many other aspects of their health and wellbeing, it has low authority — meaning it does not actually drive those other aspects in a meaningful way.
By contrast, the communities of life impact and treatment effectiveness had high authority and high influence across the network. These are the things that most powerfully shape the rest of a patient's experience.
Put simply: a patient's diagnosis tells us something about how to name their condition, but it tells us comparatively little about what their life is actually like, or what treatments will help them.
What does this mean in practice?
The authors make several practical suggestions flowing from these findings.
For clinical consultations, the study suggests that asking about employment status should be routine — it is a powerful, simple indicator of general life impact that any clinician can elicit in a few seconds. Mental health should be assessed as a matter of course, not as an afterthought. And the focus of the consultation should be less on refining the diagnostic label and more on understanding how the condition is affecting the whole person.
For healthcare planning, the findings that healthcare attendance can be predicted from a relatively small number of factors — previous contact with the service, employment status, mental health, and life impact — has potential applications in planning services and identifying patients who may need more intensive support before they reach crisis point.
For clinical trial design, the finding that treatment responders tend to cluster together — and that non-responders tend to be refractory to everything — has important implications. Future trials could try to stratify patients not just by diagnosis but by their predicted likelihood of responding, using information about previous treatment responses, employment, and life impact. This could make trials more efficient and help identify the refractory subgroup for targeted investigation.
For clinical guidelines and drug regulation, the authors suggest that professional bodies should reconsider the heavy reliance on diagnostic categories for treatment algorithms, and that regulators should consider whether approving treatments for "functional constipation" or "IBS-D" as discrete diagnostic groups is the most meaningful way to evaluate whether an intervention helps real patients.
________________________________________
Overall Conclusions
The PERSPECTIVE study produced two complementary sets of findings.
On transanal irrigation, it demonstrated that TAI has a meaningful success rate of around 60%, with benefits that persist over the long term. It provided the first substantial evidence that TAI may benefit patients with diarrhoea and mixed bowel symptoms, not just constipation. Early effectiveness at three months is a reliable guide to long-term effectiveness. And the study highlighted that too many patients wait far too long — often a decade — before accessing this treatment, suggesting that earlier introduction should be considered.
On the broader lived experience of FBDs, it challenged something fundamental about how these patients are currently cared for. The expectation that the diagnostic label is the most important guide to treatment and to understanding how a patient feels is not supported by the data. What actually shapes a patient's experience — their quality of life, their mental wellbeing, their ability to work, how often they need healthcare — is far more closely tied to life impact, mental health, and employment than to whether they have IBS-C versus functional constipation versus IBS-M. Patients who respond to treatment respond to multiple treatments; those who do not, tend not to respond to any.
The study's central message to healthcare professionals is perhaps best captured in a quote cited by the authors from the physician William Osler, written over 125 years ago: "Care more particularly for the individual patient than for the especial features of the disease." PERSPECTIVE provides rigorous modern data to support exactly that approach.
________________________________________
This lay summary covers findings from two publications arising from the PERSPECTIVE study:
1. Henderson M, et al. (2022) Transanal Irrigation for the Management of Functional Bowel Disorders: An Observational Study. Int J Nurs Health Care Res 5: 1360.
2. Ruffle JK, et al. (2025) Deep phenotyping of patient lived experience in functional bowel disorders using machine learning. Scientific Reports 15: 35349.Functional bowel disorders are common gastrointestinal (digestive) conditions. They include constipation (stools become harder and less often), diarrhoea (stools become looser and more frequent) and irritable bowel syndrome (IBS), which is accompanied by abdominal pain and bloating. Also evacuatory dysfunction and faecal incontinence.
Bowel conditions remain a taboo subject and a hidden disability. Our patients describe a life structured around toilets, a sense of ‘loss of control’, time off work because of symptoms and lives dominated by “laxative days” at home.
Clinical experience suggests that all types of functional bowel disorders impact on quality of life and ability to work. This study will examine the impact of these disorders.A small number of people with functional bowel disorders use a treatment called transanal irrigation to manage their symptoms. It can be an effective treatment, but not for everyone and sometimes patients stop using it. We want to learn more about how effective this treatment is in the long term and why patients stop using it.
These findings may positively influence clinical practice through managing patients’ expectations about effectiveness of treatment and also encourage patients to seek help sooner if treatment becomes less effective.
Also, the impact of the pandemic on functional bowel disorders will be explored. Anecdotally, some patients have benefited from being at home more during lock-down. Easier access to toilet facilities has reduced anxiety. Home working has allowed more flexibility with routines. Increased awareness of factors affecting bowel habits can help patients develop better coping strategies.
Potential participants will be recruited by email, from ContactMe-IBS – a dedicated National IBS registry and our service database of transanal irrigation users. Participation will involve completion of a single on line questionnaire.
REC name
South West - Cornwall & Plymouth Research Ethics Committee
REC reference
21/SW/0086
Date of REC Opinion
9 Aug 2021
REC opinion
Further Information Favourable Opinion