Gastrointestinal Symptom Science and Assessment
Margaret M. Heitkemper and Wendy A. Henderson
Symptom science focuses on developing personalized strategies to treat and prevent the adverse symptoms of illness across diverse populations and settings (National Institute of Nursing Research [NINR], 2016). At the same time, symptom science is about understanding mechanisms that account for symptoms so that appropriate therapies can be devised. Nurse scientists are particularly focused on linking patient reports of symptoms with real-time capture of physiological/biological variables that reflect biology and pathophysiology. Equally important is their ability to translate and disseminate the findings into evidence-based practice. In this chapter, we discuss irritable bowel syndrome (IBS) as a prototype of a chronic and life-interfering condition that is associated with intermittent but at times daily symptoms.
IBS is one of the most prevalent gastrointestinal (GI) disorders in the world and its diagnosis is based on patients’ symptoms along with exclusion of other diseases. A meta-analysis of 80 studies with more than 260,000 participants reveals that more than 11% of the world’s population (Lovell & Ford, 2012) meets the Rome III symptom-based criteria for a diagnosis of IBS. IBS is associated with substantial health care costs, with estimates ranging from $1,562 to $7,537 per year (Nellesen, Yee, Chawla, Lewis, & Carson, 2013). Work productivity of those with IBS is lower than that of other chronically ill populations such as those who have rheumatoid arthritis, migraine headache, and asthma (Taylor, Kosinski, Reilly, & Lindner, 2014). In numerous studies investigators have shown that IBS is associated with reduced quality of life (Monnikes, 2011). IBS is more common in women than men in Western cultures. The most recently revised version of the symptom-based IBS classification is the Rome IV criteria: recurrent abdominal pain on average at least 1 day a week in the past 3 months associated with two or more of the following: related to defecation; associated with a change in a frequency of stool; associated with a change in form (consistency) of stool; and symptoms must have started at least 6 months ago (Palsson et al., 2016). The diagnostic criteria include abdominal pain and alterations (diarrhea, constipation, or mixed diarrhea/constipation) with no discernible pathophysiological cause (e.g., infection, gross inflammation). There is currently no accepted biological marker specific to IBS. Therefore, clinical trials of behavioral therapy and drug therapy rely on self-report of symptoms and symptom intensity.
The biopsychosocial model of IBS first proposed by Drossman (Tanaka, Kanazawa, Fukudo, & Drossman, 2011) is a framework that encompasses the growing list of biological and psychological factors that contribute to the etiology of IBS, some of which are amenable to drug and/or behavioral therapy. In order to better manage symptoms, persons living with IBS need to both self-identify symptom triggers and self-manage to the degree that symptom burden is reduced and quality of life is enhanced. As noted in Figure 10.1, the biopsychosocial model includes pathophysiological factors which, if found to be relevant to symptom experiences, may be important in the design and testing of therapeutic interventions to improve patient-related outcomes.
ALIGNMENT WITH NURSING SCIENCE AND WOMEN’S HEALTH RESEARCH
Prior to the first published criteria for IBS diagnosis known as the Manning criteria in 1989, IBS was generally viewed as a “psychosomatic disorder.” The symptoms of abdominal pain and discomfort, diarrhea, and constipation were attributed to personality characteristics such as hysteria and catastrophizing tendencies, depression, and anxiety. Because there was no validated biomarker, the symptoms were often discounted and patients were left seeking medical advice from a series of providers (e.g., gastroenterologists, obstetrics/gynecologists, psychiatrists) with little to guide the management of their symptoms. Such gaps in the literature led to inadequate patient information and in some cases, a bias against those patients, particularly women, reporting symptoms in the absence of overt pathology. At that same time, pharmacologic approaches included agents that were directed at one specific symptom such as constipation, which did little to relieve other intestinal and extra-intestinal symptoms of IBS such as abdominal pain. Relatively recent qualitative studies continue to reveal that patients with IBS often report a sense of frustration, isolation, and dissatisfaction with the medical system and the need for more information (Drossman et al., 2009; Halpert, Rybin, & Doros, 2010).
FIGURE 10.1 A biopsychosocial model of IBS.
ANS, autonomic nervous system; CNS, central nervous system; ENS, enteric nervous system; Hx, history; IBS, irritable bowel syndrome.
Source: Adapted from Tanaka, Kanazawa, Fukudo, and Drossman (2011).
An early national survey (Heitkemper, Carter, Ameen, Olden, & Cheng, 2002) confirmed the gender gap in the diagnosis and health care–seeking behavior associated with IBS. At that time several researchers noted the overlap of a diagnosis of IBS with other chronic pain disorders such as fibromyalgia, headache with aura, interstitial cystitis, temporomandibular joint disorder, and pelvic pain, all of which were more likely to be diagnosed in women. Abdominal surgical procedures such as cholecystectomy and hysterectomy rates were higher in women with IBS relative to other patient groups. Likewise, other comorbid conditions such as insomnia, fatigue, and psychological distress (anxiety, depression) were more frequently reported by women with IBS relative to men with IBS and non-IBS female patients. Building on this published work, our team of nurse scientists, a psychologist, a gastroenterologist, and a psychiatrist began a series of descriptive studies in which women with a medical diagnosis of IBS were studied using both prospective and retrospective measures of symptoms and health-related quality of life.
These initial studies focused solely on women participants for two primary reasons. First, more women seek health care services for symptoms and they are most likely to be diagnosed with IBS relative to men. Second, in the early 1990s there was an impetus across NIH to include more women in clinical trials, as well as to study conditions that predominantly affect women, so as to reduce gaps in our understanding of sex and gender differences (Kirschstein, 1991). Although fewer in number, men are diagnosed with IBS, most frequently IBS-diarrhea. Women with IBS tend to have more reports of constipation.
Our work in the field of functional gastrointestinal disorders (FGDs) began with the question: Why do women report more GI symptoms? Our hypothesis was that women who experienced/reported more stress (e.g., life events, daily stress) resulting in physiological activation of the hypothalamic–pituitary–adrenal axis and the autonomic nervous system (ANS) would experience more GI symptoms. In our initial model, both of these physiological stress response systems work downstream to influence the GI tract either through increases or decreases in smooth muscle contractions and/or intestinal secretion.
The first project, titled A Nursing Study of Gut Function in Menstruating Women, was funded by the Division of Nursing in 1984. The basic premise of this proposal, as well as of subsequent descriptive, mechanistic studies, was that stress and GI symptoms are linked. We postulated that understanding this relationship was important to informing the design of behavioral interventions to reduce GI symptom distress. In this initial study, the focus was on women with and without dysmenorrhea (menstrual cramping pain) as a model for determining first if menstrual cycle phase differences in GI symptoms existed and if symptoms were related to ovarian hormone levels. Both well-validated retrospective (e.g., Symptom Checklist-90, Menstrual Distress Questionnaire) and prospective (i.e., Women’s Health Diary [WHD]) for three menstrual cycles) tools were used to characterize the types and severity of symptoms, including GI, menstrual cramping pain, and mood states. The WHD was developed by Woods for the longitudinal Seattle Midlife Women’s Study (Mitchell & Woods, 1996) and includes 26 symptoms with items related to sleep, fatigue, pain, and mood. Measures of gastric motility and serum levels of estrogen and progesterone were made in the two groups of women. Urine luteinizing hormone (LH) surge testing at ovulation was used to confirm menstrual cycle phase.
The results of this initial study demonstrated that many women both with and without dysmenorrhea report GI symptoms including diarrhea, constipation, stomach pain, and nausea. For both nondysmenorrheic and dysmenorrheic women, these symptoms are amplified during the late luteal and early menses phases of the menstrual cycle. However, the menses-related symptom amplification was greater in those with dysmenorrhea. We also found that menstrual cycle phase influences stool number/consistency (e.g., looser stools at menses), which is likely due to natural cyclic changes in the ovarian hormones estrogen and progesterone. As hypothesized, there were group differences, with the dysmenorrheic group reporting more moderate to severe pain and bowel symptoms than nondysmenorrheic women. Thus, menstrual cycle phase is a confounder when attempting to make comparisons of symptom frequency/intensity between IBS and non-IBS premenopausal women, as well as men-versus-women comparisons. In addition, menstrual cycle phase is an important consideration when conducting physiological testing in premenopausal women.
Additional secondary data analyses of this and later descriptive studies were used to describe the influence of oral contraceptives and perimenopausal status on GI, mood, and somatic symptoms in healthy control and women with IBS (Heitkemper, Cain, Jarrett, et al., 2003). For example, women in the perimenopausal period report an increase in the number of GI, somatic, and psychological distress symptoms. Given the fluctuations in ovarian hormones as well as life stressors during this stage, the cause of the increase in symptoms is likely multifactorial. Women with IBS who also report a history of dysmenorrhea and perimenstrual dysphoria disorder report greater symptoms and cyclic variation than women with IBS alone. This effect is not modified by the use of oral contraceptives (Heitkemper, Cain, Jarrett, et al., 2003). An outcome of the aforementioned early descriptive work is the necessity of gathering a comprehensive medical history that includes gynecologic evaluation when studying women with IBS.
Concurrent with the prior clinical studies were ongoing animal experiments to study the effects of estrogen (estradiol; E2) and progesterone on GI smooth muscle and intestinal transit under both resting (basal condition) and stress-related peptide (thyrotropin releasing hormone [TRH]) stimulation. Significant findings reported (Heitkemper & Bond, 1995) included that ovariectomized and E2-treated rats exhibited a blunted response to central (i.e., intracerebroventricular) administration of TRH when compared to sham-ovariectomized females and untreated male rats. Whether this is a direct effect of E2 on the smooth muscle because estrogen receptors are present in the gut or due to E2 modulation of vagal input subsequent to TRH administration is unknown.
The notion that gonadal hormone levels can modulate the response to stress-related hormone administered centrally raises the question of whether sex-related differences in stress responsivity are solely present in adult animals when higher levels of gonadal hormones are present. However, recently investigators demonstrated sex differences in visceral pain sensitivity related to altered stress paradigms administered early in neonatal development in a rat model (Prusator & Greenwood-Van Meerveld, 2016). Together these experiments, along with other animal data, support the hypothesis that ovarian hormones may contribute to the documented phenotypic differences in men and women with IBS.
Irritable Bowel Syndrome
In 1990, we began our first study specifically focused on women with IBS. The comparison groups included: women with medically diagnosed IBS; IBS-non-patients (IBS-NP), women who report current symptoms compatible with IBS (i.e., they met Rome I criteria but denied health care–seeking for their GI symptoms); and healthy control women. Data were collected on: (a) daily reports of GI symptoms and stool patterns, (b) daily and reported life time stress, psychological distress, (c) physiological arousal, and (d) diet. The IBS-NP group was chosen as the comparison group because while a significant portion of the U.S. adult population reports daily or monthly symptoms consistent with a diagnosis of IBS, many do not seek health care. Whether this is due to differences in symptom intensity/frequency, the presence of psychological distress, or the degree to which symptoms interfere with daily activity and work or quality of life was not known. An interdisciplinary team including experts in gastroenterology, psychiatry, psychology, bioengineering, and nutrition were brought together to design and conduct the study.
Some of the key findings from these early studies are that daily self-reported stress levels are higher in the IBS compared to the control women and that daily stress predicts next-day GI symptoms, especially pain (Jarrett et al., 1998; Levy, Cain, Jarrett, & Heitkemper, 1997). This confirms what many individuals say anecdotally when asked what they believe triggers their symptoms. Also, in this study it was found that more women in the IBS group (66%) as compared with controls (34%) have a history of symptoms compatible with a Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM-IIIR; DSM criteria available at that time) diagnosis (e.g., depression, anxiety disorder) and higher scores on a retrospective psychological distress tool (Heitkemper, Jarrett, Cain, et al., 1995). Indeed, the major difference between those seeking health care services and those not was found in the psychological distress variables, with the IBS group reporting more distress than non-health-care seekers. Ideally, a longitudinal study may reveal whether the non-health-care seekers with time ultimately sought health care services for their IBS. Additional variables such as lack of social supports in the environment or individual characteristics such as low resilience may also contribute to health care–seeking behavior. However, these factors are often not included in clinical (primary and tertiary care) and community studies of IBS.
In a separate follow-up study, we examined the role of psychological distress, both current and lifetime, with respect to GI symptoms. A history of psychological distress was measured with the Composite International Diagnostic Interview (CIDI) that determines if their symptoms meet the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV; version at the time of the study) criteria for a mood or anxiety disorder. Women with severe IBS symptoms were more likely to report a history of posttraumatic stress disorder (PTSD, 30%), social phobia public (SPP, 35%), and somatization (12%) relative to women with mild to moderate IBS symptoms (PTSD, 0%; SPP, 7%; somatization, 0%). Early adverse events including abuse in childhood, especially sexual abuse, are more commonly reported in women with IBS when compared to non-IBS groups. We assessed childhood trauma, including physical and emotional neglect and abuse, with the Childhood Trauma Questionnaire (Heitkamper et al., 2001). More women with severe IBS were likely to report a history of emotional and physical abuse, physical neglect, and sexual abuse as children compared to healthy controls and IBS individuals with mild to moderate symptoms (Heitkemper, Cain, Burr, Jun, & Jarrett, 2011). These findings were among the early studies showing the relationship of early childhood adverse events and symptom severity in IBS. Again, such results are important when considering multicomponent interventions for IBS. These data also reflect the need for a multidisciplinary approach (diet, exercise, stress reduction) to the management of IBS, rather than a focus solely on bowel pattern. In addition, it is clear that early negative childhood experiences may shape the trajectory of future pain-related conditions such as IBS. Hence, there is a need to study sociocultural and environmental factors that predispose to adult health problems. Understanding these as well as times of transitions (e.g., puberty) may be key to management of adult IBS.
The importance of stress exposure and environment and the development of IBS in adulthood is perhaps best exemplified in recent studies of U.S. military personnel. In a review of 314 new-onset cases of IBS from a cohort of more than 41,000 active duty participants, Riddle et al. (2016) described that both increased life stressors and reports of anxiety and depression increased the risk of developing IBS. They found that risk was greatest in those with these features combined with a history of infectious gastroenteritis (Riddle et al., 2016). In another study of 337 women veterans seen in primary care for PTSD, depression, and IBS (prevalence 33%), White et al. found that while depression and PTSD were more common in the women with IBS, this did not fully explain the association between trauma and increased risk of IBS (White et al., 2010). Given the prevalence of IBS in active duty service people and veterans, understanding the linkages among stress, environment (infectious exposure), and gender is an important direction for future work.
ALIGNMENT WITH SYMPTOM SCIENCE
Symptom science focuses on developing personalized strategies to treat and prevent the adverse symptoms of illness across diverse populations and settings. At the same time, symptom science is about understanding mechanisms that account for symptoms so that appropriate therapies may be devised. For example, urine and salivary proteomic measures can serve as potential biomarkers reflective of tissue damage produced by inflammation resulting in pain, nausea, and fatigue. While omics approaches provide information on potential pathophysiological mechanisms at play, they can also serve as outcome measures (e.g., diet changes and microbiome) depending on study design. As such, biological measures provide the opportunity to explicate mechanisms linking comorbid symptoms, such as disturbed sleep, with abdominal pain, and the opportunity to predict and/or monitor responses to nonpharmacological therapies. Explorations of biological alterations may help explain why individuals with the same clinical diagnosis (e.g., IBS) have different symptoms, symptom severity, symptom triggers, or degree of life interference due to symptoms.
One challenge in IBS symptom science research is the wide array of measures and approaches used to capture common symptoms such as poor sleep and pain. To address the general issue of measurement, the NINR center directors led an initiative to address common data elements (CDEs) for symptom science research (Corwin et al., 2014; Redeker et al., 2015). This work built on prior and ongoing NIH initiatives including PROMIS and NeuroQoL. The University of Washington School of Nursing incorporated symptom CDEs for our pilot projects funded through the Center for Research on Management of Sleep Disturbances (P30 NR 011400, 2009–2014) and our Center for Innovation in Sleep Self-Management (P30 NR016585, 2016–2021), as well as several R series grants. For example, we use PROMIS-sleep–pain interference, and Global Health tools in all studies of adults with a disorder or disease for which these symptoms are either examined as descriptors or as outcome measures. Recently, the GI Health PROMIS measures (Spiegel et al., 2014) became available and will be incorporated into future studies of FGDs. At a recent National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)-sponsored conference on FGD research, a common theme expressed by scientists was the need for a data repository that would link centers of excellence focused on FGDs. This could be much like the NIDDK Multidisciplinary Approach to the Study of Chronic Pelvic Pain (MAPP) project that focuses on urological chronic pelvic pain (www.mappnetwork.org).
As noted earlier, in many studies, daily symptom recordings provide not only a snapshot of a day’s experience, but also patterns within weeks, months, or years. Though not perfect, daily recordings attend to one of the important limitations of one-time retrospective measures: response bias, which is the tendency for participants to report their “worst” symptom intensity. Initially, paper-and-pencil diary versions were employed, whereas more recently, electronic diaries have been introduced to capture symptom experiences. Colleagues from medicine, engineering, and computer sciences assisted to test and employ an eDiary symptom log to capture real-time data elements and at the same time allow for symptom pattern recognition, with the goal of using this as a self-management tool. The use of these technologies both for data collection and as tools to enhance self-management enables collaborative industry–academic partnerships.
Abdominal pain is the most disruptive and distressing symptom reported by individuals with IBS. GI pain is unlike somatic pain in that it is often diffuse and challenging to discover its site of origin. Henderson and colleagues, in 2009, identified the need to develop a more sophisticated tool to help patients and providers to communicate regarding the location of GI symptoms. Together with an interdisciplinary team, they developed an electronic assessment tool, the Gastrointestinal Pain Pointer (GIPP). The GIPP employs a graphical interface, shown in Figure 10.2, where patients identify personal characteristics including gender, body type (normal or overweight) and record the location and intensity of their pain. The graphical interface also includes a list of descriptors from which to select the one that matches best the pain experience, including intensity, location, and qualitative components. At the same time, the provider records the patient’s heart rate and blood pressure. In a sample of 93 IBS and non-IBS participants, they found the GIPP to be a practical method of assessing GI symptoms such as bloating and simultaneously entering the data into an electronic medical record (Henderson et al., 2015). The tool is a novel measure of GI symptom intensity that enhances clinicians’ ability to better quantify, in real time, patient-related GI outcomes for both clinical care and research (Henderson et al., 2015).
Diet is another area of importance when considering factors that contribute to GI symptoms. As such, dietary assessments are increasingly viewed as important to the management of persons with chronic GI symptoms. Many persons with IBS can self-identify a food trigger that elicits or exacerbates their symptoms. Investigators have identified fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs) as triggers for symptoms such as abdominal pain and discomfort, intestinal gas/bloating, and diarrhea for many persons with IBS (Nanayakkara, Skidmore, O’Brien, Wilkinson, & Gearry, 2016). FODMAPs are short-chain carbohydrates (oligosaccharides, monosaccharides) and generate alcohols that are poorly absorbed in the small intestine. The elimination or restriction of FODMAPs from the diet is difficult to achieve and the long-term impact of such restrictions is unknown. More research in this area is clearly warranted.
FIGURE 10.2 The Gastrointestinal Pain Pointer.
The assessment of dietary intake associated with GI symptoms, as well as other chronic, episodic illnesses, is challenging. Recall tools such as the Food Frequency Questionnaire are dependent upon memory and do not provide real-time information about specific food triggers. Working with medical and computer science colleagues, a novel prospective computer-based application (app) for logging GI symptoms over a 2-week period was tested. The investigators found that 73% of IBS participants were able to identify a meal nutrient trigger of their GI symptoms. In another survey, gastroenterology patients were willing to use health-related apps (Zia, Le, Munson, Heitkemper, & Demiris, 2015).
As the necessity of rigorous approaches to symptom assessment moves to the forefront, it is also important to incorporate analytic strategies to consider the reality that a single symptom rarely exists in isolation from other comorbid symptoms. For example, abdominal pain, sleep disturbance, and fatigue are highly correlated and comorbid in individuals with IBS. Building on the work of Miaskowski et al. (Miaskowski et al., 2006; Miaskowski & Aouizerat, 2007), an approach of examining symptom clusters not only as an outcome or a means of identifying subgroups with a heterogeneous condition such as IBS is needed.
OVERLAPPING CONDITIONS: EMPHASIS ON SLEEP
Studies by us and others indicate that patients with IBS report other non-GI symptoms, of which insomnia, frequent awakenings, and fatigue are among the most common. A limited number of studies with small samples using polysom-nography (PSG) measures suggest that objective sleep may be different in IBS patients as compared to nonsymptomatic individuals (Tu, Heitkemper, Jarrett, & Buchanan, 2016). In addition, neuroendocrine factors, ANS imbalances, and psychological distress are known to influence the quality of sleep and are associated with GI symptom distress in women with IBS (Jarrett et al., 1998; Levy et al., 1997). Investigators suggest, based on limited PSG data, that there is a sleep disturbance associated with IBS and that this may lead to the exacerbation of symptoms in a significant subgroup of women. Many studies are not conclusive regarding objective sleep disturbances and IBS, due to methodological weaknesses including the use of a one-night sleep monitoring protocol; failure to attend to important cofac-tors such as age, gender (including ovarian hormone status or menstrual cycle phase), and psychological comorbidities (e.g., depression); and small sample size (Tu et al., 2016). A “first-night” effect is clearly evident when PSG sleep studies are performed with women with IBS. The phenomenon of sleep misperception (i.e., marked differences between self-report and objective indicators in the sleep laboratory) may be present.
Although our original IBS studies did not specifically include sleep variables, the WHD included three self-report variables: difficulty in getting to sleep, awakening during the night, and early morning awakening. Women with IBS self-report problems with sleep and these reports are related to report of GI symptoms and psychological distress in almost half of the women with IBS (Jarrett, Heitkemper, Cain, Burr, & Hertig, 2000). Based on these initial observations, we began to incorporate experts in nursing science and sleep physiology onto our research team. With their collaboration we designed two laboratory sleep studies. In these studies, we also stratified the IBS sample by symptom intensity (using items from Bowel Disease Questionnaire, e.g., abdominal pain, diarrhea, constipation) and found that women with the IBS-severe had more sleep complaints than those with IBS-mild–moderate symptoms.
The protocols for the sleep studies involved the recruitment of women with and without IBS between the ages of 20 and 45. Objective indices of sleep were obtained during a standard 3-night PSG study in a school of nursing sleep laboratory, as well as 7-day actigraphy recording. The first night in the sleep laboratory was considered “adaptation,” while the second was considered “baseline.” The third night involved the insertion of an intravenous line and serial blood drawings for the measurement of catecholamines, adrenocorticotropic hormone, and cortisol. In the second sleep study, a stressor (public speaking) was added to the study protocol. We also measured heart rate with a Holter monitor for the assessment of heart rate variability. Both PSG and actigraphy-determined sleep efficiency the night before are also related to pain and discomfort the next day (Buchanan et al., 2014; Burr, Jarrett, Cain, Jun, & Heitkemper, 2009; Heitkemper, Cain, Deechakawan, et al. 2012; Jarrett et al., 2009). Women with severe IBS-diarrhea consistently showed elevations in markers of vagal tone during the night, while those with IBS-constipation demonstrated reduced vagal tone. Evidence of neuroendocrine dysregulation, in particular elevations in serum cortisol, on the third night was also found. Together, these findings support the notion that there are physiological markers that distinguish IBS from healthy controls and that even within IBS there are differences in these markers based on bowel pattern predominance and symptom severity. Whether these physiological observations contribute to the clinical phenotype of IBS or are the outcome of a chronic condition remains to be determined. It is likely that there is a subgroup of women with IBS who experience disturbed sleep (e.g., prolonged sleep onset latency) and for whom strategies such as sleep hygiene and/or cognitive behavioral therapy-insomnia (CBT-I) may be beneficial for IBS symptom management. Clinically, evidence supports the inclusion of subjective sleep assessment in the comprehensive assessment of persons with IBS.
The observation that poor sleep predicts increased GI symptoms the next day has particular relevance for those who do shiftwork. Researchers conducting studies of nurses in both an American and a Korean medical center found the prevalence of IBS as well GI symptoms to be higher in those doing shiftwork. In the U.S. report, Nojkov, Rubenstein, Chey, and Hoogerwerf (2010) hypothesized that the circadian rhythm disruptions contribute to the pathogenesis of IBS (Nojkov et al., 2010). The development of chronic-pain-related conditions such as IBS is a relevant inclusion in the national dialogue about the impact of sleep disturbances on chronic comorbid symptoms.
SYMPTOM SCIENCE AND INTERVENTION RESEARCH
With our interdisciplinary team of investigators and clinicians, an advanced practice registered nurse (APRN)-delivered comprehensive self-management (CSM) program was designed and tested (Heitkemper, Jarrett, Levy, et al., 2004; Jarrett et al., 2009, 2016; Zia, Barney, Cain, Jarrett, & Heitkemper, 2016). Males and females were recruited to these studies; however, as men seek health care services in fewer numbers, our samples were composed of approximately 20% men. The major findings of these studies are that regardless of approach (in-person or telephone CSM delivery) used, the CSM intervention is effective for up to 12 months in reducing symptoms (as defined by a composite GI symptom score that included pain measures and diarrhea/constipation) and improving quality of life. These effects were found regardless of predominant bowel type in approximately 55% to 60% of those who participated. Qualitative interviews with the participants at the end of the study showed that at the 12-month follow-up, participants continued to employ the strategies such as avoiding diet triggers and performing relaxation response that they learned during the intervention (Zia et al., 2016).
Baseline data from these intervention trials were used for additional secondary analyses. The results of these secondary data analyses have relevance to future behavioral intervention trials and clinical practice. The first is that alcohol and dietary intake patterns are related to GI symptoms, and thus the focus on how best to tailor diet (i.e., exclusion versus restrictive) is an important and timely discussion (Reding, Cain, Jarrett, Eugenio, & Heitkemper, 2013). Second, the CSM intervention also improves sexual quality of life in women (Eugenio, Jun, Cain, Jarrett, & Heitkemper, 2012). This is an important finding in that women with IBS often report reduced satisfaction with sexual activity. Third, although not a primary outcome of the trials, the CSM intervention resulted in reduced self-reported daily stress at 3, 6, and 12 months postrandomization.
Despite these overall group differences, not all patients improved following the CSM program. One goal of the last intervention trial was not only to determine benefit but also to determine if baseline physiological factors (e.g., biomarkers) could predict who is most likely to benefit. Understanding who is most likely to benefit would also enlighten us as to the pathophysiology or etiology of functional GI disorders to identify early on who is most likely to benefit. We have found that IBS patients with higher vagal tone and lower sympathovagal balance at baseline experienced greater symptom reduction following the CSM intervention (Jarrett et al., 2016). It is interesting to note that lower vagal activity and reduced heart rate variability have been linked to a number of risk factors for cognitive impairment and lower cognitive flexibility (Kim et al., 2006; Ylitalo, Airaksinen, Sellin, & Huikuri, 1999). Specifically, in IBS, increased catecholamines may contribute to disease pathophysiology through their roles in pain perception and stress reactivity, particularly as part of the ANS branch of the stress response. There are data to indicate that patients with IBS have imbalances in the ANS, with some but not all patients exhibiting increases in sympathovagal balance as reflected in reduced vagal activity and increased circulating levels of norepinephrine. Whether this sympathetic dominance is a reflection of a genetic predisposition for the degradation of catecholamines or the result of chronic stress and/or environmental exposure is not known (Liu et al., 2014).
Omics and Symptom Assessment in IBS
In 2001, Levy and colleagues conducted an analysis based on more than 10,000 respondents representing 6,060 twin pairs (Levy et al., 2001). The concordance for IBS was significantly greater in monozygotic than in dizygotic twins, leading researchers to consider the potential role that genetics plays in IBS development. In 2014, a large, multinational genome-wide association study was published demonstrating that the genes that encode for KDELR2 (endoplasmic reticulum protein retention receptor) and GRID2IP (glutamate receptor, ionotropic, delta 2 [Grid2] interacting protein) were associated with a risk for IBS (Ek et al., 2015). Building on what is known about the interaction of psychological distress with symptoms, our collaborative laboratories have pursued a targeted genetic approach.
Prior to these genetic studies, clinical investigators reported that patients with IBS-diarrhea had higher levels of the urinary metabolite of serotonin (Bearcroft, Perrett, & Farthing, 1998). Following this, Caspi et al., in a Turkish study, reported that individuals who had a genetic polymorphism in the gene that regulates the production of the serotonin reuptake transporter (5-HTTLPR) protein were more vulnerable to life stress (Caspi et al., 2003). Subsequent to this, a number of investigators sought to examine whether the SERT gene polymorphisms and later the presence of rare alleles in the gene were associated with IBS (Zhang et al., 2014). Early reports with relatively small samples indicated that SERT polymorphisms were not associated with a diagnosis of IBS, but some differences in psychological variables were noted. Kohen examined SERT variants in both IBS and healthy control women by use of a combined dataset from the CSM intervention trials with related descriptive/mechanistic studies. While no IBS versus control differences were found in the distribution of alleles, a greater number of women with IBS had rare variants as compared to controls. A larger sample is needed to understand the clinical impact of these rare variants with respect to phenotype and response to treatment.
One of the genes with potential contribution to abdominal pain but also the responsiveness to the CSM intervention is Catechol-O-methyltransferase (COMT). This gene codes for an enzyme that is involved in the degradation of the neurotransmitters norepinephrine and dopamine. It was known that COMT plays important roles in neurocognitive processes like pain and emotional processing. The Val158Met polymorphism (rs4680) of the COMT gene leads to a substitution of valine (Val) by methionine (Met). The Met allele has less enzymatic activity than the Val allele (Lachman et al., 1996). There is an association between the Met/Met carrier status and cognitive performance and psychiatric conditions as well as pain processing and sensitivity (Gatt, Burton, Williams, & Schofield, 2015; Scheggia, Sannino, Scattoni, & Papaleo, 2012; Tammimaki & Mannisto, 2012).
The literature on COMT and IBS is inconsistent with one study in Chinese patients (Wang, Wu, Qiao, & Zhang, 2014) finding a higher preponderance of the Val/Met and a Swedish study (Karling et al., 2011) finding that the Val/Val was more common in IBS. Of interest to our team was the study by Hall and colleagues, who studied 104 U.S. patients with IBS and found that Met/Met homozygotes had the strongest response to a placebo intervention (i.e., the greatest symptom reduction), thus pointing to a possible role of COMT Val158Met in moderating the cognitively focused intervention response (Hall et al., 2012).
To determine if COMT genotype is associated with response to a behavioral intervention, we used data from two previously reported randomized clinical trials of CSM versus usual care (Jarrett et al., 2009, 2016). In keeping with prior observations of population-specific genotype effects, the analysis was restricted to self-identified European American participants only. Based on the prior literature, daily reports of abdominal pain, depression, anxiety, and feeling stressed were used as the primary outcome measures for analyses. Baseline abdominal pain among IBS patients did not differ by COMT genotype. However, study participants who carried at least one Val allele (i.e., either the Val/Met or Val/Val genotype) derived greater benefit, in terms of psychological distress relief, from the CSM program compared to participants with the Met/Met genotype. This finding is consistent with what had been described in other non-IBS populations, particularly those with psychiatric conditions, with respect to COMT genotype and response to behavioral therapies. The COMT genotype by CSM interactions on psychological distress may reflect the way in which COMT genotype shapes signal transmission through dopamine D1 and D2 receptors (Witte & Floel, 2012). COMT polymorphisms may confer or participate in cognitive flexibility, which is proposed to be a key factor of response to CBT, due to greater activation of dopamine D2 type receptors (Carroll et al., 2015; Witte & Floel, 2012). Whether this greater cognitive flexibility conveyed by the Val allele is associated with an improved learning capacity required to incorporate CSM strategies of self-management remains to be determined (Carroll et al., 2015). If so, this may have relevance in terms of self-management for other patient groups.
In another study of both men and women with chronic abdominal pain who had either high or low levels of perceived stress, the expression of genes involved in metabolic stress was examined. Henderson et al. found that Interleukin-1 alpha gene (IL1A) was upregulated in females with high stress versus females with low stress. Similarly, IL1A was upregulated in participants with high stress and chronic abdominal pain versus those with low stress and chronic abdominal pain. These findings suggested that the mechanism behind stress-related changes in GI symptoms is pro-inflammatory in nature (Peace et al., 2012). Further, these investigators found that CCL-16 gene expression was upregulated by 7.46-fold in IBS patients when compared with controls. More specifically, CCL-16 was overexpressed by more than 130-fold in IBS-constipation patients when compared with both controls and IBS-diarrhea patients. For some IBS patients, subclinical inflammation, epithelial dysfunction, and dysregulation of inflammatory cells (Del Valle-Pinero et al., 2011) may be important pathophysiological factors.
It is likely that there are multiple genes and/or epigenetic influences related to lifetime exposures, resilience, and personal resources, which influence patient outcomes. Our combined findings contribute to a growing literature exploring the role of genetic factors in individualized self-management and symptom outcomes. Inclusion of omics (e.g., genetics) in future clinical trials may help identify for whom interventions work and under what conditions (Grey, Schulman-Green, Knafl, & Reynolds, 2015; Henly, 2016).
Proteomics and Metabolomics
A biomarker discovery approach that may hold promise for nurse scientists is the examination of the proteome (Voss et al., 2011). Proteomics refers to the global analysis of cellular proteins with mass spectrometry (MS)-based techniques, image analysis, reverse-phase protein array, amino acid sequencing, and/or bioinformatics to identify and quantify proteins. Proteomics is focused on cellular or secreted proteins in terms of both their structure and the functional interaction among proteins. Advances in high-throughput proteomics exponentially increased the potential for new clinical biomarkers, particularly using noninvasive sources such as urine and saliva. To date, the majority of studies are focused on cancer detection and tissue injury. However, these approaches are being employed in symptom science research (Voss et al., 2011). Initially, a shotgun proteomics using urine or other body fluids was used to compare patient and healthy control group differences. This broad hypothesis-generating approach allows an initial identification of proteins for further validation studies.
Another biomarker approach that utilizes MS technology is metabolomics. Metabolomics is a multitargeted analysis of low molecular mass (< 1000D) of endogenous and exogenous metabolites and their pathways. This approach allows for the systematic study of the unique chemical fingerprints (amino acids, glucose) that are the result of cellular processes. Metabolites are both the intermediates and the products of metabolism. Complementing genomics and proteomics, metabolomics offers investigators the advantage of linking environmental (diet, stress) factors to cellular responses. Metabolites are found in all body fluids. The Human Metabolome Database (www.hmdb.ca) contains a list of those metabolites that are the result of cellular activities. Metabolomic approaches are increasingly used in the study of diabetes, dyslipidemia, cancer, chronic neurological disorders, pharmaceutics, and nutrition. The utility of metabolomics in symptom science is a new area of inquiry. Given the relationship of diet and stress to IBS symptoms, it is conceivable that metabolomics fingerprints may help distinguish those most likely to benefit from dietary restrictions or cognitively focused therapies.
There is growing appreciation of how the environment within us—that is, our microbiome—can affect health. Our physiological health is influenced by the large number of microbes that exist on or in the skin, mouth, respiratory tract, vagina, and urinary and GI tracts (Van Ness, 2016). There is recent evidence of nutritional and gender effects associated with the microbiome in early life (Cong et al., 2016). Of particular relevance to the development of IBS is the interaction of diet, intestinal bacteria, and the host response. When alterations within the microbiome occur due to disease or lifestyle changes, the balance of microbes is disrupted (dysbiosis), leading to local inflammation, immune response, and metabolic changes. Depending on genetic risk, as well as contextual factors, such responses may ultimately result in chronic inflammation. The Human Microbiome Project (HMP) has contributed to an explosive amount of interest in the gut microbiome and its role in chronic illnesses, including lung disease, diabetes, and psychiatric conditions (Taylor, Wesselingh, & Rogers, 2016).
Our team works collaboratively with pediatric gastroenterologists to test the impact of fiber therapy on stool microbiome and symptoms in children with chronic abdominal pain (Shulman et al., 2016). We observed no differences in stool microbiome composition between children treated for 6 weeks with psyllium compared to placebo, despite reductions in the number of pain episodes. It may be that more sustained efforts are needed to see substantial differences in gut microbial communities.
In an exploratory study, differences in the oral microbiome of IBS patients and healthy controls were examined, to see whether the oral microbiome related to GI symptom severity. The oral microbiome of 38 participants was characterized using PhyloChip microarrays. The severity of GI symptoms was assessed by orally administering a GI test solution. Participants self-reported their induced GI symptoms. Abdominal pain severity was highest in IBS patients and was significantly correlated to the abundance of 60 operational taxonomic units (OTUs), four genera, five families, and four orders of bacteria (r2 > 0.4, p < .001). Of note is that the patients with IBS who were also overweight reported the greatest symptom severity and exhibited differences in microbiome diversity characteristics. Abdominal pain severity was highly correlated to the abundance of many taxa, suggesting the potential utility of the oral microbiome in GI symptom phenotyping (Fourie et al., 2016).
Omics Toolbox for Symptom Assessments
Taken together, these new “omic” measures may be viewed as part of the “toolbox” for nurse scientists pursuing the biobehavioral basis for symptoms in a condition such as IBS. These tools, as well as the integration of data elements (i.e., systems biology), allow us to uncover the heterogeneity of each individual and open up future possibilities of designing therapies that are personalized or tailored. Nurse scientists, through their multilevel focus on the individual, family, and community, can interweave the sociocultural, behavioral, and biological influences on ability and disability. The toolbox as used by current and future nurse scientists has the potential to utilize the power of omics to address significant health care problems within the context of the person’s life.
Despite the impressive growth in technologies, there are challenges in the interpretation, utilization, and translation of omic findings to practice. A number of potential pitfalls associated with varying analytic approaches in testing contribute to false positives, false negatives, and conflicting results (Rehm, Hynes, & Funke, 2016). How and when these approaches are used to enhance patient self-management remains to be determined and is therefore an area much in need of research.
HEALTH POLICY IMPLICATIONS
In the 2015 State of the Union address, President Obama introduced the launching of the Precision Medicine (a.ka. health) Initiative. It is a bold plan to improve health and treat disease (www.whitehouse.gov/precision-medicine). He stated:
The future of precision medicine will enable health care providers to tailor treatment and prevention strategies to people’s unique characteristics, including their genome sequence, microbiome composition, health history, lifestyle, and diet. To get there, we need to incorporate many different types of data, from metabolomics (the chemicals in the body at a certain point in time), the microbiome (the collection of microorganisms in or on the body), and self-report symptom and behavioral data collected by both the health care providers and the patients themselves.