Health Care Disparities, Social Determinants of Health, and Emotional Impacts in Patients with Ulcerative Colitis: Results from a Global Ulcerative Colitis Narrative Patient Survey

Abstract

Background

The Ulcerative Colitis (UC) Narrative global survey assessed aspects of living with UC. This analysis aimed to identify health care disparities, social determinants of health, and emotional impacts related to UC disease management, patient experience, and quality of life.

Methods

The survey was conducted by The Harris Poll from August 2017 to February 2018 among adults with UC. Responses from 1000 patients in the United States, Canada, Japan, France, and Finland were analyzed based on patient income, employment status, educational level, age, sex, and psychological comorbidities. Odds ratios (ORs) with significant P values (P < .05) from multivariate logistic regression models are reported.

Results

Low-income vs high-income patients were less likely to have participated in a peer mentoring (OR, 0.30) or UC education program (OR, 0.51). Patients not employed were less likely to report being in “good/excellent” health (OR, 0.58) than patients employed full time. Patients with low vs high educational levels were less likely to have reached out to patient associations/organizations (OR, 0.59). Patients aged younger than 50 years vs those aged 50 years and older were less likely to have visited an office within an inflammatory bowel disease center/clinic in the past 12 months (OR, 0.53). Males were less likely to be currently seeing their gastroenterologist than females (OR, 0.66). Patients with vs without depression were less likely to agree that UC had made them more resilient (OR, 0.51).

Conclusions

Substantial differences in disease management and health care experience were identified, based on categories pertaining to patient demographics and psychological comorbidities, which may help health care providers better understand and advance health equity to improve patient care.

Graphical Abstract

Graphical Abstract.

Key Messages.

• What is already known?

Social determinants of health contribute to disparities in disease burden and health outcomes.

• What is new here?

In a global survey of patients with ulcerative colitis (UC), substantial differences in disease management and health care experience were identified based on income, employment status, educational level, age, sex, and psychological comorbidities.

• How can this study help patient care?

These results provide insight into the experience of patients with UC and may help health care professionals better understand the perspective of their patients to advance health equity and address disparities that affect outcomes, patient care, and quality of life.

Introduction

Ulcerative colitis (UC) is a chronic inflammation of the colon ranging from proctitis to pancolitis.¹ Clinically, UC is characterized by intermittent symptoms, most commonly diarrhea and rectal bleeding, but can also include fecal urgency, incontinence, increased frequency of bowel movements, fatigue, abdominal pain, weight loss, and fever.¹ However, the burden of disease often goes beyond physical impairment, and patients also commonly report psychological distress and poor overall general health, even when in remission.^2,3 Depression and anxiety are also prevalent.^4,5 Together, this physical and mental burden can adversely affect the overall quality of life of patients with UC.³

Furthermore, studies have shown that medical care alone only accounts for an estimated 10% to 20% of modifiable factors that contribute to health outcomes, while the other 80% to 90% are shaped by nonmedical factors, including social determinants of health.⁶ Social determinants of health have been defined by the World Health Organization as the conditions in which people are born, grow, live, work, and age.⁷ They include socioeconomic factors, such as education, income, and wealth; geographic factors, such as access to healthful food and recreational areas; and exposure to crime.⁷ These socially determined variables have an important influence on health inequities. In patients with inflammatory bowel disease (IBD), socioeconomic-based disparities can lead to poor patient outcomes, including increased disease severity, diagnostic delays, poor appointment and medication adherence, or increased hospitalizations and death.^8,9

The UC Narrative is a global initiative that aims to identify how patients with UC are impacted by the disease, as well as barriers to optimal care, to help improve outcomes and enhance communication between patients and health care professionals (HCPs).^2,10 The original UC Narrative surveys, conducted among 2100 patients and 1254 physicians across 10 countries worldwide, examined the experiences of patients with UC and physicians, including day-to-day impact, disease management, goal setting, and communication.^2,10

The results of these surveys demonstrated a general satisfaction with patient-physician communication; however, differences were identified between patient and physician perceptions of the impact of UC symptoms¹⁰ and perceived delays in diagnosis of UC.² Areas of improvement were identified relating to access to information, materials, and support.¹⁰ Patients also described their most common fears and concerns and highlighted the mental exhaustion they experience due to UC; however, mental health issues were found to be infrequently discussed with their physicians.² Although these data provide evidence for improving disease management, and highlight the emotional and mental health burden of patients with UC at the global level, these factors are yet to be fully explored across social determinants of health in patients with UC.

Here, we report the results of a post hoc analysis performed on a subset of the original UC Narrative data among patients with UC from 5 of the 10 countries included in the original survey (the United States, Canada, Japan, France, and Finland). By stratifying survey responses by patient demographics and diagnosed psychological comorbidities, we aimed to identify health care disparities, social determinants of health, and emotional impacts related to UC disease management, patient experience, and overall quality of life.

Materials and Methods

Study Design and Patient Population

The UC Narrative is a collaborative global initiative directed by an advisory panel comprising adult patients living with UC, gastroenterologists, IBD nurses, a psychologist, and representatives of IBD patient advocacy organizations from multiple countries. The UC Narrative survey was conducted online and via telephone by The Harris Poll between August 2017 and February 2018 in 10 countries. Survey questions were designed to assess multiple aspects of UC and its management, and they reflect patient and physician experiences and perceptions of UC.^2,10 A separate UC Narrative COVID-19 survey was conducted between August 25 and December 13, 2021, to assess the effects of the COVID-19 pandemic on patients with UC from 5 core countries (the United States, Canada, Japan, France, and Finland).¹¹

This post hoc analysis of data from the UC Narrative survey included patients who resided in the United States (n = 301), Canada (n = 215), Japan (n = 210), France (n = 169), and Finland (n = 105), in line with the 5 countries included in the UC Narrative COVID-19 survey. The majority of respondents were recruited and selected from databases of individuals who agreed to participate in market research and provided informed consent, and a subset of the United States participants (n = 126) was recruited through public links shared by The Crohn’s and Colitis Foundation. Patients who were eligible to take part were those aged 18 years and older, who had received a diagnosis of UC confirmed by endoscopy and no prior colectomy, had been to a gastroenterologist or internist’s office within the past year, and had ever taken a UC prescription medication other than only aminosalicylates (5-ASAs).

The use of prescription medication was used as a proxy for UC disease severity. Patients were categorized as having “milder” disease if they had ever taken a prescription medication for UC (excluding those who had only taken 5-ASAs, who were defined as having “mild” UC, and who were excluded from this analysis), had taken corticosteroids for <4 of the past 12 months, and had never taken a biologic or immunosuppressant therapy. Patients with “moderate to severe” disease were defined as those who had taken corticosteroid treatment for ≥4 of the past 12 months or, alternatively, had ever taken a biologic or immunosuppressant for their UC. The proportion of patients with milder UC was capped at 20% of the total number of survey respondents, to allow focus on patients with moderate to severe UC, as it was assumed that these patients were more likely in need of support and resources, which the survey could help better identify.

Survey Outcomes

Patient-reported survey responses focused on (1) factors that they considered important to managing their UC (eg, being able to conduct daily activities, psychological impacts of the disease, impact on sex life and personal relationships, and ability to travel), (2) disease management (eg, HCP visits and use of resources and support groups, knowledge of UC, and path to diagnosis), and (3) overall quality of life (eg, missed events due to their UC, work and life changes, and UC impact on daily life and starting a family).

Disease activity was assessed through (1) the self-reporting of flares, defined as a time with a dramatic increase in symptoms compared with those typically experienced, and (2) remission, defined as UC disease control with few to no symptoms.^2,10

Survey items were analyzed based on patient demographics, including income level, employment status, educational level, age, self-identified sex, and psychological comorbidities. Stratification by income and education was country-specific and included low, middle, and high ranges (Table 1, Supplementary Data Content 1). Patients were categorized by their employment status: full-time employment, part-time employment, or not employed. Patients in the “not employed” group were retired, stay-at-home spouses or parents, students, not working due to disability or illness, or simply not employed (including patients who were not employed but looking for work, or not employed and not looking for work). Regarding comparisons by age, sex, and psychological comorbidities, patients aged younger than 50 years were compared with those aged 50 years and older; males were compared with females, and patients with a diagnosis of depression or anxiety were compared with patients without depression or anxiety, respectively. Diagnosis was self-reported via a survey question that asked patients if they had been told by a doctor that they had the condition (question: “Have you personally been told by a doctor that you have any of the following health conditions? Please select all that apply” (Patient Survey Questions, Supplementary Data Content 6).

Statistical Analysis

Survey responses were analyzed globally using a forward stepwise logistic regression analysis¹² performed on the unweighted results from the 5 countries mentioned previously. The independent variables included demographics (income [n = 831], employment status [n = 1000], educational level [n = 1000], age [n = 1000], sex [n = 1000]), and psychological comorbidities (depression and/or anxiety [n = 1000]). The dependent variables included questions on the importance of daily activities, events missed due to illness, work and life changes, overall health, other diagnosed conditions, visits to HCPs, access to health care supplemental resources (including support groups), medications taken for UC, years between symptom onset and official diagnosis, and ­knowledge of UC. To determine which variables to include, each regression used an entry P of 0.05 and a removal P of 0.1. Multicollinearity was checked, and none of the variation inflation factors were above 3, so no adjustment was needed. Odds ratios (ORs) and 95% confidence intervals (CIs) from multivariate logistic regression models measured the relative difference between demographic groups, adjusting for other confounders, including other demographic variables and diagnosed conditions. Unweighted data were used for the logistic regression due to a large margin of error around the ORs in the post-weighted results, which were a result of large weights in countries with higher populations. Subgroup ­analysis of the independent variables of income, education, sex, and psychological comorbidities (depression or anxiety) used descriptive statistics from the 5-country total, post-weighted to adjust for the relative population size of each country. A P of < .05 indicated statistical significance.

Ethical Considerations

This survey was noninterventional, not intended to provide clinical data to support treatment decisions, and not conducted as a clinical trial for any endpoints; therefore, ethics approval was not required. All patients provided their informed consent before completing the survey and were compensated on behalf of the sponsor by the investigators for their participation. In the United States, the research method and survey questionnaire were reviewed and received independent review board approval (Western Institutional Review Board protocol number 20171627).^2,10

Results

Patient Demographics, Disease Characteristics, and Psychological Comorbidities

This analysis included 1000 patients, of whom 301 patients were from the United States, 215 patients were from Canada, 210 patients were from Japan, 169 patients were from France, and 105 patients were from Finland. Patient demographics and psychological comorbidities are described in Table 1. Overall, 50% of patients identified as female. The mean age of respondents was 41.3 (standard deviation [SD], 12.6) years, and 75% of patients were aged younger than 50 years. The majority of patients (82%) had moderate to severe disease and had been diagnosed with UC for a mean of 9.0 (SD, 9.5) years, with 57% having ever received biologic therapy for their UC. Less than half of the patients (45%) described their overall health as “good/excellent.” Regarding demographic variables of interest, 27%, 36%, and 32% of patients had a low-, middle-, and high-income level, respectively (income was not requested in the original UC Narrative survey in France). Most patients (66%) were in full-time employment, 9% were employed part time, and 25% were not employed. A full breakdown of patients’ employment status, including details of those not employed, is found in the Tables of Supplementary Data Content 2. Patients with a low or middle income were less likely to be employed (part time or full time) than high-income patients (proportions in employment were 56%, 79%, and 87% for each income level, respectively). Similarly, patients with a high educational level were more likely to be employed than those with middle or low educational levels (proportions in employment were 85%, 58%, and 60% for patients with a high, middle, and low educational level, respectively). Males (82%) were more likely to be employed than females (68%; Table, Supplementary Data Content 3).

Table 1.

Patient demographics, baseline disease characteristics, and psychological comorbidities.^a

Characteristics	Patients (N = 1000)
Age (years), mean (SD)	41.3 (12.6)
<50, n (%)	753 (75)
≥50, n (%)	247 (25)
Female, n (%)	498 (50)
Age at UC diagnosis (years), mean (SD)	32.3 (12.4)
Time since UC diagnosis (years), mean (SD)	9.0 (9.5)
Time since UC diagnosis, n (%)
<1 year	49 (5)
1–4 years	415 (41)
5–9 years	198 (20)
≥10 years	338 (34)
Severity of UC,bn (%)
Milder UC	183 (18)
Moderate to severe UC	817 (82)
Currently in remissionc	676 (68)
Overall health, n (%)
Good/excellent	447 (45)
Poor/fair	553 (55)
Prescription medications ever taken for UC, n (%)
5-ASAs (mesalamine, mesalazine)	664 (66)
Corticosteroids	694 (69)
Biologic therapy	570 (57)
Anti-TNF biologic	538 (54)
Other biologicd	120 (16)
Immunosuppressant	452 (45)
Another prescription medication	204 (20)
Prescription medications currently taking for UC, n (%)
5-ASAs	434 (43)
Corticosteroids	225 (23)
Biologic therapy	458 (46)
Anti-TNF biologic	395 (39)
Other biologicd	84 (11)
Immunosuppressant	287 (29)
Another prescription medication	110 (11)
Not currently taking prescription medication	50 (5)
Impact at work, n (%)	836 (84)
Took time away from work due to their UC, n (%)	624 (62)
Income level,en (%)
Lowf	238 (27)
Middleg	317 (36)
Highh	284 (32)
Employment status, n (%)
Not employedi	250 (25)
Part-time employment	90 (9)
Full-time employment	661 (66)
Educational level, n (%)
Lowj	111 (11)
Middlek	261 (26)
Highl	628 (63)
Diagnosis of depression,an (%)	147 (15)
Diagnosis of anxiety,an (%)	210 (21)

Abbreviations: 5-ASAs, aminosalicylates; BEP; Baccalauréate Professionnel; CAD, Canadian dollars; CAP, Certificate d'Aptitude Professionelle; CEGEP, Collège d’enseignement général et professionnel; n, weighted number of patients with each characteristic; N, number of patients included in the analysis; SD, standard deviation; TNF, tumor necrosis factor; UC, ulcerative colitis; USD, United States dollars.

^aDiagnosed condition; Question: “Have you personally been told by a doctor that you have any of the following health conditions? Please select all that apply” (Patient Survey Questions, Supplementary Data Content 6).

^bPatients with a “milder” form of the disease were defined as those who had ever taken a prescription medication for their UC (excluding those who had only taken 5-ASAs, who were defined as having “mild” UC, and who were excluded from this analysis), had taken corticosteroids for <4 of the past 12 months, and had never taken biologic or immunosuppressant therapy. Patients with “moderate to severe” disease were defined as those who had taken corticosteroid treatment for ≥4 of the past 12 months, or had ever taken a biologic or immunosuppressant for their UC.

^cUC disease control with few to no symptoms.

^d N = 756; Japan was excluded from the analysis because this information was not requested in the original UC Narrative survey in Japan.

^e N = 881; France was excluded from the analysis since income was not requested in the original UC Narrative survey in France.

^fUnited States and Canada: <$50 000 USD, CAD, respectively; Japan: <¥4 000 000; Finland: <€30 000.

^gUnited States and Canada: $50 000–99 999 USD, CAD, respectively; Japan: ¥4 000 000–6 999 999;

Finland: €30 000–49 999.

^hUnited States and Canada: ≥$100 000 USD, CAD, respectively; Japan: ≥¥7 000 000; Finland: ≥€50 000.

ⁱIncludes patients who were not employed but looking for work, not employed and not looking for work, retired, stay-at-home spouses or parents, students, or not working due to disability or illness.

^jUnited States and Japan: high school degree or less; France: high school diploma or less; Canada: graduated from secondary school or less; Finland: matriculation examinations or less.

^kUnited States: job-specific training, some college but no degree, associate degree; Canada: trade-certificate/diploma, certificate or diploma from community college/institution/CEGEP, teaching certificate from provincial department of education, completed some university study but no degree, university certificate or diploma below bachelor level; Japan: junior college degree; France: CAP/BEP (vocational training certificate/technical education certificate), 2-year college degree/associate’s degree; Finland: vocational and professional qualifications or polytechnic degree.

^lUnited States: college degree or more; Canada: bachelor or first professional degree or more; Japan: bachelors or university degree; France: 3-year college degree/bachelor’s degree or more; Finland: university degree.

In total, 26% of patients had a current diagnosis of depression (15%) or anxiety (21%). Low-income patients were more likely than high-income patients to report having either of these conditions; 23% and 11% of low- and high-income patients, respectively, were diagnosed with depression; 28% and 16% of low- and high-income patients, respectively, were diagnosed with anxiety. More females reported having depression or anxiety (19% and 31%, respectively) compared with males (10% and 11%, respectively; Table, Supplementary Data Content 3).

Sex differences were also present in prescription medications ever taken for UC. More females than males had ever taken corticosteroids (females, 81%; males, 58%), 5-ASAs (females, 79%; males, 54%), or immunosuppressants (females, 56%; males, 35%), whereas the proportions of patients who had ever taken a biologic therapy were similar (female, 54%; male, 60%; Table, Supplementary Data Content 4).

Disparities in Patients with Ulcerative Colitis

Income level

Findings stratified by income level are presented in Figure 1. Compared with high-income patients, low-income patients were 70% (OR, 0.30; 95% CI, 0.14-0.65) and 49% (OR, 0.51; 95% CI, 0.32-0.82) less likely to have participated in a peer mentoring or UC education program, respectively, and 48% less likely (OR, 0.52; 95% CI, 0.32-0.84) to have reached out to patient associations or organizations. Low-income patients were also 41% less likely (OR, 0.59; 95% CI, 0.36-0.98) than high-income patients to say that they had been referred to patient associations or organizations by their gastroenterologist/internist.

Figure 1.

Disparities related to income level in patients with UC. ORs (adjusted for confounders) and relative differences (defined as OR minus 1) in patient response between low-income and high-income patients and middle-income and high-income patients. France was excluded from the analysis since income was not requested in the original UC Narrative survey in France. ^aUnited States and Canada: <$50 000 USD, CAD, respectively; Japan: <¥4 000 000; Finland: <€30 000. ^bUnited States and Canada: ≥$100 000 USD, CAD, respectively; Japan: ≥¥7 000 000; Finland: ≥€50 000. ^cUnited States and Canada: $50 000–99 999 USD, CAD, respectively; Japan: ¥4 000 000–6 999 999; Finland: €30 000–49 999. ^dAmong employed patients. ^*P < .05; ^**P < .01 for low- or middle- vs high-income patients. Abbreviations: CAD, Canadian Dollars; CI, confidence interval; OR, odds ratio; UC, ulcerative colitis; USD, United States Dollars.

Compared with high-income patients, middle-income patients were 56% less likely (OR, 0.44; 95% CI, 0.28-0.71) to have been currently visiting a gastroenterologist, and 37% less likely (OR, 0.63; 95% CI, 0.44-0.91) to say that minimizing or avoiding side effects from medications was important for managing their UC. Middle-income patients were 64% more likely (OR, 1.64; 95% CI, 1.13-2.40) to say that they were too tired because of their UC to excel in the workplace (among employed patients) compared with high-income patients.

Employment status

Findings stratified by employment status are presented in Figure 2A. Compared with patients in full-time employment, those patients who were not employed were 89% less likely (OR, 0.11; 95% CI, 0.03-0.50) to have stopped treatment to start a family, 53% less likely (OR, 0.47; 95% CI, 0.30-0.74) to be at least somewhat satisfied with their medication, and 42% less likely (OR, 0.58; 95% CI, 0.39-0.86) to say that they were in “good/excellent” health. Also, patients employed part time were 57% less likely (OR, 0.43; 95% CI, 0.20-0.94) than patients employed full time to say that preventing the need for injections or infusions was important to managing their UC.

Figure 2.

Disparities related to (A) employment status and (B) educational level. ORs (adjusted for confounders) and relative differences (defined as OR minus 1) in patient response. ^aIncludes patients who were not employed but looking for work, not employed and not looking for work, retired, stay-at-home spouses or parents, students, or not working due to disability or illness. ^bNot requested in the original UC Narrative survey in the United States. ^cAmong those taking prescription medication. ^dUnited States and Japan: high school degree or less; France: high school diploma or less; Canada: graduated from secondary school or less; Finland: matriculation examinations or less. ^eUnited States: college degree or more; Canada: bachelor or first professional degree or more; Japan: bachelors or university degree; France: 3-year college degree/bachelor’s degree or more; Finland: university degree. ^*P < .05; ^**P < .01 for not employed or part-time employed vs full-time employed (panel A) or low vs high educational level (panel B). Abbreviations: CI, confidence interval; OR, odds ratio; UC, ulcerative colitis.

Educational level

Findings stratified by educational level are presented in Figure 2B. Compared with patients with a high educational level, patients with a low educational level were 57% less likely (OR, 0.43; 95% CI, 0.27-0.68) to say that having less impact on their sex life and personal relationships was important to managing their UC, and 56% less likely (OR, 0.44; 95% CI, 0.29-0.66) to say that being able to travel (for work or pleasure) was important to UC management. Patients with a low educational level were also 41% less likely (OR, 0.59; 95% CI, 0.36-0.95) than those with a high educational level to have reached out to a patient association or organization and 75% more likely (OR, 1.75; 95% CI, 1.19-2.59) to say that they had missed travel plans due to their UC.

Age

Findings stratified by age are presented in Figure 3. Compared with patients aged 50 years and older, those aged younger than 50 years were 81% less likely (OR, 0.19; 95% CI, 0.10-0.37) to not be currently taking prescription medications for UC, 47% less likely (OR, 0.53; 95% CI, 0.36-0.79) to have visited an office located in an IBD center or clinic in the past 12 months, and, among employed adults, 89% more likely (OR, 1.89; 95% CI, 1.18-3.03) to agree that their employer was very understanding of their condition. When considering factors important to managing their UC, patients aged younger than 50 years vs those aged 50 years and older were 53% less likely (OR, 0.47; 95% CI, 0.33-0.68) to say that reducing the need for prescription medication was important. Conversely, patients aged younger than 50 years were 73% more likely (OR, 1.73; 95% CI, 1.13-2.63) to say that having less impact on their sex life and personal relationships was important to managing their UC.

Figure 3.

Disparities in patients aged younger than 50 years vs those aged 50 years and older. ORs (adjusted for confounders) and relative differences (defined as OR minus 1) in patient response. ^aAmong employed patients. ^bAmong those taking prescription medication. ^*P < .05; ^**P < .01 for patients aged younger than 50 years vs those aged 50 years and older. Abbreviations: CI, confidence interval; IBD, inflammatory bowel disease; OR, odds ratio; UC, ulcerative colitis.

Sex

Findings stratified by sex are presented in Figure 4. Compared with females, males were 40% less likely (OR, 0.60; 95% CI, 0.42-0.84) to have visited an office within an IBD center or clinic in the past 12 months, and 34% less likely (OR, 0.66; 95% CI, 0.44-0.98) to be currently seeing their gastroenterologist. Males were also 74% (OR, 1.74; 95% CI, 1.26-2.43) and 73% (OR, 1.73; 95% CI, 1.23-2.42) more likely to have been hospitalized or have had 2 or more flares in the past 12 months, respectively, compared with females (Figure 4A).

Figure 4.

Disparities in patient-reported (A) health care utilization and outcomes and (B) knowledge of UC and disease management in male vs female patients with UC. ORs (adjusted for confounders) and relative differences (defined as OR minus 1) in patient response. ^aTrue or false questions relating to UC disease and treatment knowledge (described in full in Survey Questions, Supplementary Data Content 6). ^bAmong patients who were employed. ^*P < .05; ^**P < .01 for male vs female patients. Abbreviations: CI, confidence interval; IBD, inflammatory bowel disease; OR, odds ratio; UC, ulcerative colitis.

Compared with females, males were 56% more likely (OR, 1.56; 95% CI, 1.09-2.22) to agree that they felt comfortable discussing health issues in their workplace (among employed patients; Figure 4B). When considering factors important to managing their UC, compared with females, males were 61% less likely (OR, 0.39; 95% CI, 0.29-0.54) to say that reducing fatigue, 53% less likely (OR, 0.47; 95% CI, 0.34-0.66) to say that being able to manage the psychological impacts of the disease, 41% less likely (OR, 0.59; 95% CI, 0.43-0.80) to say that being able to control pain, and 31% less likely (OR, 0.69; 95% CI, 0.50-0.96) to say that having less impact on familial or social relationships were important to managing their disease (Figure 4B).

Emotional and mental health impacts

Depression

Findings stratified by a diagnosis of depression are presented in Figure 5A. Patients with vs without depression were 59% less likely (OR, 0.41; 95% CI, 0.26-0.65) to say that they were in “good/excellent” health and 49% less likely (OR, 0.51; 95% CI, 0.34-0.77) to agree that UC has made them more resilient. Patients with vs without depression were also 87% more likely (OR, 1.87; 95% CI, 1.20-2.92) to agree that UC controlled their life, rather than them controlling the disease, and 79% more likely (OR, 1.79; 95% CI, 1.20-2.69) to have taken time off from their job due to their disease/symptoms (ie, not due to disability).

Figure 5.

Factors impacted by psychological comorbidities in patients^a with vs without self-reported (A) depression and (B) anxiety. ORs (adjusted for confounders) and relative differences (defined as OR minus 1) in patient response. ^aAmong patients who were employed. ^bAmong those taking prescription medication. ^cAmong students. ^*P < .05; **P < .01 for patients with vs without depression (panel A) or anxiety (panel B). Abbreviations: CI, confidence interval; OR, odds ratio; UC, ulcerative colitis.

Regarding factors considered important to disease management, employed patients diagnosed with depression were 93% more likely (OR, 1.93; 95% CI, 1.20-3.09) than those without depression to say that working more often/consistently was important to managing their UC. Patients with vs without depression were more likely to say that reducing their fatigue (92%; OR, 1.92; 95% CI, 1.25-2.94), and having less impact on their sex life and personal relationships (65%; OR, 1.65; 95% CI, 1.08-2.53) was important to managing their UC.

Anxiety

Findings stratified by a diagnosis of anxiety are presented in Figure 5B. Patients with anxiety were 47% less likely (OR, 0.53; 95% CI, 0.33-0.85) to have been at least somewhat satisfied with their medication than those without anxiety. Patients with vs without anxiety were 88% more likely (OR, 1.88; 95% CI, 1.08-3.26) to have left their job due to their UC, and students with anxiety were 79% more likely (OR, 1.79; 95% CI, 1.06-3.04) to have made special arrangements/accommodations at school due to their UC compared with students without anxiety.

Regarding factors considered important to managing their UC, patients with vs without anxiety were 84% more likely (OR, 1.84; 95% CI, 1.22-2.78) to say that being able to control pain was important to managing their UC, and 83% more likely (OR, 1.83; 95% CI, 1.15-2.89) to say that the ability to conduct daily activities was important. Patients with anxiety were also more likely than patients without anxiety to say that having less impact on familial and social relationships (61%; OR, 1.61; 95% CI, 1.08-2.39) was important to managing their UC.

Disparities in Time to Diagnosis of Ulcerative Colitis

Numerical differences were observed in the average time between the onset of first symptoms and diagnosis of UC based on income level, educational level, sex, and psychological comorbidities (Table, Supplementary Data Content 5). The time between first symptoms and diagnosis for low- and middle-income patients was slightly longer (mean, 2.7 years and 2.5 years, respectively) compared with high-income patients (mean, 1.8 years). Patients with a low educational level were diagnosed on average 3.3 years after their first symptom, whereas patients with a middle educational level were diagnosed sooner (mean, 1.4 years). There was a numerically longer delay in diagnosis of UC in females compared with males (mean, 2.4 years and 1.8 years, respectively). The mean time to diagnosis for patients without depression or anxiety was 1.8 years compared with 3.0 years for patients with either depression or anxiety.

Disparities in Disease Management and Emotional Impacts

Subgroup analyses identified numerical differences in disease management and the emotional impacts of UC (Figure 6). More patients with a low income or low educational level described their overall health as “poor/fair” compared with those of a high income or high educational level (low- vs high-income level, 67% vs 44%, respectively; low vs high educational level, 77% vs 47%, respectively). A higher proportion of low-income patients (44%) reported that their most recent UC flares lasted for 1 month or more compared with middle-income patients (28%). More patients with depression (77%) or anxiety (64%) also described their overall health as “poor/fair” compared with those with neither condition (51%), and almost half of patients with anxiety or depression (43%) did not consider themselves to be in remission compared with just 28% of those without either condition.

Figure 6.

Disparities in disease management and outcomes^a by (A) income level, (B) educational level, (C) sex, and (D) self-reported psychological comorbidities. ^aDescriptive statistics showing proportion of patients. ^bFrance was excluded from the analysis since income was not requested in the original UC Narrative in France. ^cSelf-identified. Abbreviations: IBD, inflammatory bowel disease; UC, ulcerative colitis.

Regarding sex differences, as reported previously, a lower proportion of males than females were currently seeing their gastroenterologist (males, 68%; females, 89%). However, 42% of females reported that their most recent flare lasted for 1 month or more compared with 23% of males. Subgroup analyses also revealed that more males than females had visited an office located in an IBD center or clinic in the past 12 months (57% vs 36%). More males than females agreed that if their treatment made them feel good enough, they did not see a need to consider other treatment options even if they might make them feel even better (males, 65%; females, 48%). Also, more males than females worried that they would be perceived as a difficult patient if they asked too many questions and that it would affect the quality of their care (males, 53%; females, 36%), or were hesitant to tell their gastroenterologist/internist if they were not taking their medication exactly as prescribed (males, 50%; females, 33%).

Discussion

This post hoc analysis from the global UC Narrative patient survey found significant disparities in the day-to-day impact of UC, disease management, and health care utilization based on patient demographics, social determinants of health, and psychological comorbidities. These results support previous findings that social determinants of health contribute to disparities in health outcomes in patients with IBD,^8,9 while identifying populations at risk of lack of support and emphasizing population differences in disease management.

Overall, we found that low-income patients were less likely than high-income patients to have relied on, or have been referred to, patient associations/organizations or education/peer mentoring programs to help manage their disease. Patients with a low educational level were also less likely than patients with a high educational level to have reached out to patient associations/organizations. These results suggest that low-income patients and those with a low educational level may be particularly vulnerable to lack of information and additional support. It has previously been reported that patients with UC wished for more information and support at diagnosis¹⁰ and that there was a need for more education about disease risks and therapy options.² However, over 50% of physicians did not discuss patient advocacy organizations with their patients.¹⁰ Furthermore, improving access to education has been linked to improved health through various mechanisms, including health literacy.¹³ These results suggest that HCPs may benefit from attentiveness to identifiers of socioeconomic status, such as low income or low education in their patients, and point them towards additional support resources, which could help close the gap.

Compared with patients employed full time, those who were not employed were less likely to have described their current overall health as “good/excellent,” to be at least somewhat satisfied with their medication, or to have stopped treatment to start a family, suggesting that these patients may experience a higher mental and physical disease burden. This is supported by lower socioeconomic status being associated with worse disease outcomes¹⁴ and higher rates of psychiatric disorders.⁵

Patients aged younger than 50 years vs those aged 50 years and older were more likely to have prioritized disease impacts on their sex life and personal relationships as a factor important to managing their UC, suggesting that these patients may also experience a high psychosocial burden of the disease.⁵ Furthermore, patients aged younger than 50 years were found to be less likely to have visited an office within an IBD center or clinic, and less likely to have reported a need to reduce prescription medications as being important to managing their UC compared with patients aged 50 years and older. Compared with younger patients, older patients have an increased risk of infection and malignancies when treated with immunosuppressants^15,16 and are more likely to experience comorbidities that require additional medications.^16,17 Therefore, reducing the amount of prescription medications in these patients could help improve their overall well-being.

Compared with female patients, male patients were less likely to be currently seeing a gastroenterologist or have visited an office located within an IBD center or clinic in the past 12 months and were more likely to have experienced hospitalization or 2 or more flares in the past 12 months. Although no consistent sex differences have been reported to date in patients with UC,^18-20 our findings suggest that there may be variation in male and female behaviors regarding disease management and outcomes. Interestingly, subgroup analyses revealed that a higher proportion of males had negative perceptions around communications with their HCPs, which may contribute to less interaction with disease specialists, ultimately impacting health outcomes. This may also explain why males said they were more likely to settle for a treatment that made them feel good enough, rather than considering other treatment options, even if they might make them feel better. Further research could be done to evaluate why males are less likely to reach out to specialists, even if they have more flares and hospitalizations.

Overall, males were also less likely to have experienced certain negative emotional impacts of UC including familial/social, psychological, and quality of life factors compared with females. Although quality of life and the ability to control pain have been identified previously as being central to UC management,² we found these factors to be less of a priority for male patients vs female patients.

Patients with vs without depression or anxiety were more likely to have said that they experienced negative work impacts, along with more pronounced psychosocial impacts of UC, and subgroup analyses revealed that a high proportion of patients with vs without depression or anxiety described their health negatively. Taken together, these findings indicate that these patients seem to suffer from a particularly high disease burden and therefore may require more care and attention. This is supported by a population-based study where the incidence rate ratio of psychiatric comorbidity was significantly increased in patients with IBD compared with controls.⁵ Moreover, it has been found that patients with IBD who also have depression or anxiety are at an increased risk of disease relapse, and these comorbidities are also risk factors for poorer quality of life,^21-23 while anxiety has also been linked to lower adherence to medication.²⁴ Our findings also demonstrate a delay in diagnosis for patients with depression or anxiety, which may be due to patients’ reluctance to see physicians due to their depression or anxiety, concerns about diagnosis, or potential negative outcomes from the appointment. Delays in diagnosis could also be due to diagnostic confusion caused by overlapping symptoms, which can be prevalent in patients with psychological comorbidities (eg, chronic diarrhea). Furthermore, previous results suggest that mental health concerns in patients with UC are not routinely addressed² and that almost 50% of patients did not feel comfortable discussing emotional concerns with their physician.¹⁰ More specific attention to mental and emotional impacts of UC could help to improve patient care and quality of life.

Limitations of this analysis include a reliance on accurate recall by patients and their understanding of the questions surveyed, and that recruitment for the survey was based on self-reported diagnosis of UC, with no clinical disease activity assessment. This analysis also included data from 5 countries compared with the 10 countries included in the original survey.^2,10 Furthermore, though these results have been reported as global outcomes, the findings may not be applicable to all countries due to regional differences in health care systems, variable access to care, and cultural differences. In addition, patients with mild disease (defined as those who had ever been treated with 5-ASAs only) were excluded from the analysis in order to focus on those patients with moderate to severe disease who would more likely benefit from the findings of this research. Consequently, it is difficult to generalize these findings to the entire UC patient population. In addition, a lack of data in some categories resulted in wide confidence intervals, which may have impacted the reliability of the results. Also, some patient categories included a wide range of patients who may have differing viewpoints, which may have influenced their responses. For example, those categorized as “not employed” included those not employed who were both looking for work and not looking for work, retired participants, students, or those not employed due to disability; it is likely that retired patients may have responded differently than students or patients with a disability.

In addition, other than capturing self-identified sex, there was no further analysis of patients by, for example, sexual orientation. It is possible that heterosexual patients may have responded differently than patients identifying as part of the lesbian, gay, bisexual, transgender, queer+ community; however, further stratification of participants by sexual orientation and/or identity was not possible due to low sample sizes and/or a lack of data. Furthermore, though there are known racial and ethnic disparities in IBD health care,^8,9,25 insufficient data regarding race and ethnicity were captured in the original patient survey; therefore, subgroup analysis by race/ethnicity was not possible. Lastly, it is important to note that subgroup analyses do not take into account other potentially confounding variables, such as age, psychological comorbidities, or employment status, whereas the regression analysis isolates the effect from other confounding factors.

In conclusion, this global survey of patients with UC identified substantial disparities in patient-reported assessments of disease management and health care experience, based on factors such as income level, employment status, educational level, age, sex, and psychological comorbidities. These results expose the importance of addressing social determinants of health, by providing insight into the experience of patients with UC. This may help HCPs to better understand the perspective of their patients, to advance health equity and address disparities that affect outcomes, patient care, and quality of life.

Author Contributions

All authors contributed to the analysis and interpretation of study data, critically reviewed the manuscript, and approved the final submission. M.C.D., L.P.B., K.Y., A.S., J.C.C., L.J.R., M.S., S.G., E.P.J., A.M., and R.P. designed the study.

Funding

This study was sponsored by Pfizer.

Conflicts of Interest

F.D.O. has received consulting fees from Janssen and Pfizer Inc. M.C.D. has received consultancy fees from AbbVie, Arena Pharmaceuticals, Bristol Myers Squibb, Celgene, Eli Lilly, Galapagos, Genentech, Gilead Sciences, Janssen, Pfizer Inc, Prometheus Laboratories, Takeda, Trellus Health, and UCB. L.P.B. has received personal fees from AbbVie, Amgen, Biogaran, Biogen, Boehringer Ingelheim, Celgene, Celltrion, Eli Lilly, Ferring, Forward Pharma, Genentech, H.A.C. Pharma, Hospira/Pfizer, Index Pharmaceuticals, Janssen, Lycera, Merck, Mitsubishi, Norgine, Samsung Bioepis, Sandoz, Takeda, Theravance, Tillotts Pharma, and Vifor. K.Y. was an employee of Sidekick Health, Kopavogur, Iceland, at the time of the analysis and is currently an employee and shareholder of Pfizer Inc; has received lecture fees from IBD ja muut suolistosairaudet Ry (patient association) and Pfizer Inc; and has received consultancy fees from Pfizer Inc, Sandoz, Sidekick Health, Takeda, and Tillotts Pharma. A.S. has received consultancy fees from Pfizer Inc for work unrelated to this analysis. J.C.C., L.J.R., and S.G. are employees and shareholders of Pfizer Inc. M.S. is a postdoctoral fellow employed by Rutgers University, Ernest Mario School of Pharmacy, which received funding from Pfizer in connection with the development of this manuscript. E.P.J. and A.M. are employees of The Harris Poll, which was a paid contractor to Pfizer Inc to conduct the initial research and related statistical analysis. R.P. has received advisory and/or consultancy fees from Abbott, AbbVie, Alimentiv, Amgen, Arena, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Celltrion, Cosmos Pharmaceuticals, Eisai, Elan, Eli Lilly, Ferring, Galapagos, Genentech, Gilead Sciences, GlaxoSmithKline, Janssen, Merck, Mylan, Oppilan Pharma, Pandion Therapeutics, Pfizer Inc, Progenity, Protagonist, Roche, Sandoz, Satisfai Health, Schering-Plough, Shire, Sublimity Therapeutics, Takeda, Theravance, and UCB.

Data Availability

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.