Incontinence is a critical issue in the lives of many individuals worldwide. Urinary incontinence (UI) affects at least 15-30% of adult women, and one-fourth to one-third of elders (Executive summary, 2003). Internationally, authors (e.g., Kim, 2001;Milne, 2000;Rigby, 2001;Shaw, 2001) have documented that incontinence is a common problem and that there is a lack of care-seeking for incontinence by those affected. Compared to individuals without UI, individuals with UI have reported social isolation, lower purpose in life, less positive relations with others, higher negative moods including depression, and poorer perceived health (Chiverton, Wells, Brink, & Meyer, 1996; Heidrich & Wells, 2004;Kim, 2001;Paterson, 2000).

As underreported, underdiagnosed, and undertreated conditions, UI and fecal incontinence (FI) also are understudied (Executive Summary, 2003; Shaw, 2001). Past researchers have focused primarily on describing biophysiologic factors and outcomes regarding continence, using a "one size fits all" approach. Most researchers have not examined either a breadth of variables or the specificity of interventions. Little attention has been paid to differences in interventions for continence customized for different population groups (Shaw;Tomlinson et al., 1999) or to differences in outcomes by physiologic, psychological, or sociocultural variables (Executive summary). Yet knowledge about, willingness to report, and experiences with incontinence may differ by clinical condition, age, gender, race, and ethnicity (Ruff, van Rijswijk, & Okoli, 2002; Shaw).

In this presentation, our aims are to: (a) briefly share a conceptual definition of "patient-centered;" (b) discuss kinds of patient-centered interventions (PCIs;Table 1); (c) offer examples of tailored interventions in particular; (d) apply ideas of tailored interventions to incontinence; and (e) suggest ideas for future research regarding incontinence. We hope to stimulate research in incontinence that will increase the applicability of research findings to nursing practice and encourage new discoveries to improve the lives of those with incontinence.

TABLE 1.

Patient-Centered Interventions

Patient-centered interventions refer to interventions in which: (a) a patient is assessed on selected characteristics; (b) the approach is selected to address characteristics of the patient's experiences; (c) the process is responsive to the patient's characteristics (e.g., type of incontinence, medications, perceptions of cause or treatment of incontinence, social situation, or goals); or (d) any combination of these. Related concepts include consumer-centered, client-centered, and women-centered interventions (Lutz & Bowers, 2000;Writing Group, 1997). To clarify, patient does not refer only to an individual, but also can be a social group or aggregate. Patient-centered interventions commonly studied in the past decade involve explaining, guiding, counseling, or building skills among participants to adopt health-related behaviors such as eating well, getting regular mammograms, and engaging in physical activity. Although these interventions have been studied well and have been valued highly by patients (Ryan & Lauver, 2002;Skinner, Campbell, Rimer, Curry, & Prochaska, 1999), they are not the only type of PCI. For example, PCIs for continence have included not only explaining about the particular condition and counseling about management approaches, but also teaching participants skills such as scheduled voiding and pelvic floor muscle exercise (PFME;Borrie, Bawden, Speechley, & Kloseck, 2002;Kim, 2001;Shaw, 2001;Tomlinson et al., 1999).

The four terms referring to PCIs used most often in the literature are personalized, targeted, tailored, and individualized. Although these terms have been used somewhat interchangeably and inconsistently, these kinds of interventions differ in complexity (Ryan & Lauver, 2002). The terms targeted, tailored, and individualized may be viewed loosely along a continuum that reflects greater patient-centeredness.

Personalized Interventions

A personalized intervention is an intervention in which either the name of a person or another of their characteristics is used. An example of a personalized intervention would be the use of names in letters addressed to patients or prospective clients (e.g., "Mr. Jones, we are expecting you on Monday for your followup visit in our clinic."). However, using a name or single characteristic does not address characteristics that are relevant to health. Some researchers believe that this type of intervention is not truly patient-centered (Lauver et al., 2002).

Targeted Interventions

A targeted intervention is one designed to match a limited number of characteristics shared by a group of people, such as sociodemographic characteristics (e.g., gender, age, or ethnic group), type of clinical condition (e.g., overactive bladder or FI and obesity), or behavioral characteristics (e.g., exercise or smoking). Targeted interventions often have focused on health promotion and screening messages and have used mass media approaches for delivery (Huff & Kline, 1999). For example, the National Association for Continence is working on informational materials to target people with multiple sclerosis, FI, and those in assisted-living facilities (N. Muller, personal correspondence, November 5, 2003).

As we seek to understand more fully the effectiveness of interventions for continence, we will want to study subgroups defined by gender, ethnicity, and class more carefully (Gray, 2003). We can assess whether or not our interventions targeted to selected variables (e.g., type of incontinence or gender) are more effective in promoting continence than interventions which are not. Also, we can assess whether or not the typically studied variables are the most appropriate variables to address regarding incontinence. We need to think critically about whether or not gender is the most relevant variable to incontinence or whether other variables such as parity, hysterectomy, obesity, occupation, and level of smoking should be examined more fully (Gray).

Tailored Interventions

In comparison to a targeted intervention, a tailored intervention has more complexity (de Vries & Brug, 1999;Kreuter, Strecher, & Glassman, 1999;Lauver et al., 2002). Tailored interventions are customized based upon a set of individual, rather than group, characteristics. Participants are assessed on multiple and individual characteristics, each of which could have a large number of possible values. Examples of such characteristics include medication use and dose, perceptions of self-efficacy of increasing continence (Kim, 2001), perceptions of outcome efficacy of incontinence intervention (Shaw, 2001), or prior experience with behavioral strategies (e.g., PFME or scheduled voiding). This can result in a larger number of possible alternative strategies than for a targeted intervention. Based on the anticipated situations determined by the combination of variables assessed, the researchers would establish specific protocols to be followed for each anticipated situation. This would result in a finite number of possible interventions.

Standardized, health-related, informational interventions have been compared to tailored, health-related, informational interventions. Although tailored informational messages designed to increase health behaviors are only one example of tailored interventions, they are the type of tailored interventions that have been studied most often. Tailored messages have been more effective in terms of process (e.g., they have been read, remembered, and discussed more often) than standard information. Tailored informational and behavioral skills interventions have been more efficacious in promoting health behaviors (e.g., mammography, diet, or physical activity) than standard messages in about half of studies (de Vries & Brug, 1999;Ryan & Lauver, 2002;Skinner et al., 1999).

Because tailored interventions are an important and evolving area of research, but have not been applied systematically and explicitly to incontinence, testing tailored interventions promises to aid in delineating effective interventions. Some researchers have concluded that we need to tailor interventions for UI, using simpler approaches (e.g., bladder education, self-monitoring, or bladder training) before initiating more complex ones (e.g., PFME or biofeedback;Dougherty et al., 2002;Dowd, Kolcaba, & Steiner, 2003). Whereas in the past researchers targeted interventions for UI mostly by gender because of physiological differences (i.e., long and rigid vs. short and distensible urethras), future researchers could refine tailored interventions on psychological and sociocultural variables. Wells (2003) has argued that we need to identify patient characteristics that respond to particular interventions so we can tailor our interventions to these characteristics. Krause, Wells, Hughes, Brink, and Mayer (2003) have suggested that we tailor interventions for UI on variables such as perceived severity of symptoms, interference in life, degree of shame, and perceived helpfulness of treatment.

More specifically, Gross (2003) has stated that research on incontinence could be facilitated by studying tailored interventions among persons after stroke. These studies could be tailored on type of UI as well as abilities and deficits after stroke and build on previously effective interventions for UI among persons with similar characteristics. Findings from such studies would enable nurses to treat UI after stroke more effectively.

Example of a Theory-Based Intervention Related to Continence

A tailored intervention could be conducted to promote care-seeking for incontinence symptoms (Kincade, Peckous, & Busby-Whitehead, 2001;Shaw, 2001). From a theory, the variables of physiologic arousal, affect, beliefs, habits, and external barriers about a behavior have been proposed to influence that behavior, such as care-seeking (Lauver, 1992;Triandis, 1980). Prior to participant contact, a research team could design a menu of potential responses and strategies to expected participants' comments and needs. This could include (a) bio-physiologic factors related to incontinence (e.g., type of incontinence, frequency of episodes, volume of urine, and medications taken); (b) beliefs (e.g., about the cause of incontinence, about outcome efficacy of proposed treatments, or about ability to address incontinence;Kim, 2001;Shaw, 2001); (c) feelings, such as embarrassment or distress over incontinence; (d) habits (e.g., voiding patterns, past use of self-care measures, and skill level of self-care measures for incontinence (Milne, 2000); and (e) external barriers regarding receipt of continence services (e.g., financial affordability, geographic accessibility, or personal acceptability of services). Upon initial contact, the team would assess all participants on the same dimensions: biophysiologic factors, beliefs, feelings, habits, and external barriers regarding use of continence services. Following predetermined protocols, nurses could tailor their intervention to participants' characteristics. Individuals' interventions would likely differ in order, depth, and breadth. Interventions could differ on the particulars of either acceptance or alteration of biophysiologic factors, clarification of beliefs and myths, acknowledgment and coping with feelings, reflection on prior habits, encouragement to adopt new habits, clarification of accurate behavioral efforts (e.g., for scheduled voiding or PFMEs), and provision of supportive information about affordability and accessibility of care services (Lauver, 1992, 1994;Triandis, 1980).

Individualized Interventions

Individualized interventions are highly customized to particular individuals and their situations. For such interventions, core concepts are identified and general guidelines are outlined before the intervention is delivered. However, the exact content of individualized messages is not determined before interactions occur between researchers and participants. Rather, an individualized intervention develops collaboratively and as a consequence of such interactions. Individualized interventions may be customized to the extent that no two persons would receive the same message content. The number of such interventions may be infinite. The levels of specificity and complexity of individualized interventions are much higher than are those for other kinds of PCIs. For example, Brooten and colleagues (1986) and Naylor and colleagues (1999) have compared the effects of advanced practice nurses' individualized interventions in home care to usual care; the individualized interventions have had greater clinical improvements at less cost. They deliver an individualized intervention based on the individual assessments made and guided by the protocol, similar to that done in expert practice. This is an important distinction to make because for most of our history in nursing research, our interventions have been tightly controlled and have not allowed for individual differences in delivery (Lauver et al., 2002).

Past research involving interventions for continence has included individualized components, yet these components have not been labeled consistently as patient-centered or categorized as individualized. Tomlinson and colleagues (1999) performed assessments of biophysiologic factors (e.g., caffeine use, fluid intake, and involuntary urine loss) and women's needs, goals, and circumstances. The latter assessments guided the strategies incorporated for individual women (i.e., self-monitoring, including decreasing caffeine, increasing fluid intake, or decreasing voiding interval; bladder training; or PFME). To the extent that these approaches had not been delivered according to predetermined protocol, we would categorize this as an individualized intervention.

Kim (2001) offered a Continence Efficacy Intervention Program with individualized components. This program included continence guidelines and initial training in PFMEs for all participants. However, experimental group participants also received a theory-based intervention designed to increase self-efficacy regarding continence. This included not only written and audiotaped information, a calendar, and a schedule guideline; but also phone counseling; encouragement; and reassessment, refinement, and review of self-care strategies as needed. The latter dimensions are characteristics of an individualized intervention.

In the past, most PCIs have been customized on theory or key concepts determined by researchers (e.g., participants' present beliefs or clinical conditions). If interventions were individualized based on participants' preferences, then they would be shaped by participants themselves. Highly individualized interventions may be more effective in promoting recommended actions for continence because they could elicit individuals' own motivations (Stuifbergen, Becker, Timmerman, & Kullberg, 2003). We believe that promising research in the area of PCIs and incontinence would involve individualization based on participants' preferences (e.g., in approach, timing, or nature of interventions), future desired state, or both. Some researchers in UI have incorporated elicitation of participants' goals as part of a multidimensional behavioral intervention. One such intervention was associated with marked decreases in severity of urine loss when compared to a control condition (Dougherty et al., 2002).

Although individualized interventions are promising and exciting, they may not always be practical in terms of time and cost needed for individualized assessments and follow-up. Costs for individualized, targeted, and tailored interventions need to be compared. Perhaps population-based interventions targeted to patients with particular diagnoses, or group-based interventions tailored on type of incontinence and beliefs, may be less costly in terms of time and finances than individualized interventions.

Future Directions

Research that involves testing the efficacy of PCIs in improving healthcare outcomes in general and incontinence in particular is relatively new. Patient-centered interventions may differ in type of content or preferences, degree of mutual decision-making between researchers and participants, number of assessments, number of customized messages, and method of message delivery (in person, printed, or computer-based). The effects of PCIs have not been demonstrated consistently across such dimensions (Lauver et al., 2002).

More specific descriptive studies are needed in future research on incontinence. Using similar definitions of continence across studies of prevalence, we can learn more about prevalence rates by subgroups such as gender, race, and ethnicity. We also need to better understand the role of potential explanatory variables such as biophysical factors, beliefs, sociocultural practices, or economic access to healthcare in influencing experiences of incontinence (Gray, 2003). We need to have strategies for recruiting and retaining participants from diverse populations in order to develop knowledge bases that are applicable to these populations. With findings from such studies, we could design customized PCIs.

In regard to ethnicity and race, there are inconsistencies in the literature about the prevalence of incontinence and associated symptoms (Duong & Korn, 2001;Ruff et al., 2002). Perhaps ethnicity and race do not fully capture the factors critical to elimination or incontinence. When insufficiently precise measures of critical factors are used, there is a high probability of false overgeneralizations of a particular group and inconsistencies of data across groups. By more carefully describing and understanding the particular characteristics associated with ethnicity and race, we can customize PCIs for subgroups in the most effective manner. We need to understand more about biopsychosocial factors (e.g., perceptions of severity, distress, quality of life, and strategies used) in order to address incontinence across diverse groups more effectively (Shaw, 2001;Tomlinson et al., 1999). Dreher and Macnaughton (2002) proposed that to the extent that we customize our approach to special populations, we are being patient-centered and thus more culturally sensitive.

The critical dimensions on which to customize our interventions need to be delineated and tested more specifically (Suhonen, Valimaki, Leino-Kilpi, & Katajisto, 2004). Future researchers could delineate whether or not critical dimensions such as preferences, goals, and behavioral strategies for continence are similar or different by cultural groups. They can distinguish critical dimensions selected by participants and researchers. Furthermore, they can test whether or not interventions designed to improve outcomes selected by researchers or outcomes selected by participants are similarly effective, and whether the choice of clinical or behavioral outcomes in an intervention influences participants' subsequent status.

Different types of interventions for UI have been associated with comparably good outcomes (Wells, Brink, Diokno, Wolfe, & Gillis, 1991;Wyman, 2003;Wyman, Fantl, McClish, & Bump, 1998). Perhaps examining other variables would illuminate the conditions under which particular interventions are more or less effective. We need to examine whether or not selected variables (e.g., gender, race, and culture) moderate the effects of PCIs on incontinence; interventions may be more or less effective for men or women, or for Whites or African Americans. Other moderating variables could include psychosocial variables such as beliefs (e.g., perceived severity of incontinence or self-efficacy in dealing with incontinence), feelings (e.g., shame), moods, and social interactions.

Researchers can disentangle the effects of content and process-the what versus the how of interventions. Researchers can examine how selected dimensions of interventions influence outcomes; that is, what mediates the effects of interventions (e.g., perhaps self-efficacy about steps to deal with incontinence mediates the relationship between an intervention and outcome;Kim, 2001). We are challenged to explain why different interventions (i.e., bladder training, PFME instruction with biofeedback, or both) can improve UI similarly. Perhaps there is a key mediating variable. The intervention might provide structure desired by symptomatic individuals or valuing participants' self-worth or demonstrating empathy might elicit adoption of behavioral strategies for continence (Sampselle, 2003;Wyman, 2003;Wyman et al., 1998).

In summary, we propose that developing and testing PCIs regarding incontinence promises to advance knowledge about more effective interventions, conditions under which they are more or less effective, and how they are effective. With such knowledge, we can increase the applicability of nursing research to practice. In addition, we will have a stronger foundation upon which to build our nursing practice and from which to derive our approaches for individuals with incontinence in order to promote their continence and overall health status.

References