Research in Gerontological Nursing

Empirical Research 

Exploring the Applicability of Interdependence Theory to CNA-Resident Relationships

Louis J. Medvene, PhD; Carissa K. Coleman, PhD


This study explored the quality of certified nurse assistant (CNA)-resident relationships. Of interest was the extent to which interdependence theory could be used to code CNAs’ responses to questions about their relationships with residents and factors that promoted or limited these relationships. Interdependence theory defines closeness in terms of outcome interdependence and provides an account of how trust and commitment can develop. Seventeen socially skilled CNAs from nine long-term care facilities participated in structured personal interviews. All of the CNAs said they had developed close relationships with some residents. CNAs’ answers to interview questions indicated they were thinking “relationally” about their interactions with residents. Many CNAs made reference to commitment and pro-relationship behaviors that promoted relationships, and their responses supported the applicability of interdependence theory. Implications for training are that CNAs should be encouraged to think relationally, as well as dispositionally, and that boundary issues need to be explored.


This study explored the quality of certified nurse assistant (CNA)-resident relationships. Of interest was the extent to which interdependence theory could be used to code CNAs’ responses to questions about their relationships with residents and factors that promoted or limited these relationships. Interdependence theory defines closeness in terms of outcome interdependence and provides an account of how trust and commitment can develop. Seventeen socially skilled CNAs from nine long-term care facilities participated in structured personal interviews. All of the CNAs said they had developed close relationships with some residents. CNAs’ answers to interview questions indicated they were thinking “relationally” about their interactions with residents. Many CNAs made reference to commitment and pro-relationship behaviors that promoted relationships, and their responses supported the applicability of interdependence theory. Implications for training are that CNAs should be encouraged to think relationally, as well as dispositionally, and that boundary issues need to be explored.

Interest in the nature of certified nursing assistant (CNA)-resident interactions in long-term care (LTC) settings is growing, and the goal of humanizing the experience of living in LTC settings is one force driving this interest. Many LTC facilities are implementing person-centered care practices to make the experience more homelike and less institutional. Person-centered care includes knowing the resident, promoting the resident’s personhood, and developing nurturing relationships (Crandall, White, Schuldheis, & Talerico, 2007). However, the literature on CNA-resident interactions suggests that achieving these goals will not be easy. Reports on the quality of CNA-resident interactions varies, sometimes noted as impersonal and uncaring and other times as being close (Foner, 1994; Pietrukowicz & Johnson, 1991). Identified barriers to the development of close relationships include role conflicts between institutional demands for standardized task completion versus allowing for personalized interactions (Hullett, McMillan, & Rogan, 2000), cultural differences between residents and CNAs (Berdes & Eckert, 2007), and emotional withdrawal in anticipation of residents’ death (Campbell, 2003).

Several authors have reported that CNAs sometimes use the term family to describe their relationships with residents (Berdes & Eckert, 2007; Sumaya-Smith, 1995). In addition, several studies show that CNAs’ relationships with residents can be a source of job satisfaction (Mittal, Rosen, & Leana, 2009). From the residents’ perspective, Bowers, Fibich, and Jacobson (2001) reported that some residents define quality of care in terms of relationships with CNAs that are characterized by friendship and reciprocity. McGilton and Boscart (2007) reported that residents experience close relationships with CNAs and nurses.

This small-scale, qualitative study evaluated the extent to which interdependence theory—a generalized theory of relationships—could be applied to CNA-resident relationships. Interdependence theory is a social psychological theory of relationships that defines closeness in terms of the extent to which there is outcome interdependence between individuals (Rusbult, Kumashiro, Coolsen, & Kirchner, 2004). It is concerned with the ways in which partners cause one another to experience good versus poor outcomes. Interdependence theory involves an account of how pro-relationship behaviors (i.e., behaviors that are helpful to one’s partner) lead to the development of commitment, dependence, and trust within relationships. The current study explored the applicability of interdependence theory to CNA-resident interactions through structured interviews with CNAs who were identified by administrators as socially skilled. In this article, the term relationship is used to describe CNA-resident exchanges that involve some degree of closeness. The term interaction is used to characterize CNA-resident exchanges that may involve minimal closeness.

Person-Centered Caregiving: The Current Context

Current efforts to promote person-centered caregiving are taking place in an atmosphere in which LTC facilities are implementing policies based on residents’ individual preferences rather than on institutional needs for standardized practices (e.g., flexibility around mealtimes). New models of LTC facilities include architectural changes involving the creation of small-scale settings such as The Eden Alternative® and “neighborhoods” in larger facilities. These culture change efforts are being promoted by a wide variety of organizations, including the grassroots Pioneer Network, federal agencies such as the Centers for Medicare & Medicaid Services, and philanthropic organizations such as the Robert Wood Johnson Foundation (White, Newton-Curtis, & Lyons, 2008).

Yet, as White et al. (2008) noted, a cohesive definition or even agreed-on terms related to these concepts have yet to emerge. In a major contribution to the research literature on person-centered caregiving, White et al. (2008) developed a 64-item measure of person-directed care that identified five distinct concepts based on an exploratory factor analysis: personhood, knowing the person, comfort care, autonomy, and support relations. Three of these concepts are associated with relationships, broadly defined: (a) personhood, which emphasizes that each person is unique, has inherent value, and is worthy of respect and honor; (b) knowing the person, referring to the importance of knowing each resident’s life story, cultural experiences, personality, and needs and preferences; and (c) support relationships, which refers to the idea that each person lives and functions within a web of relationships and that person-centered environments strive to reduce social isolation and promote friendships. Each of these concepts of person-centeredness is relevant to the focus of the current study on the relational aspects of caregiving work.

Literature Review

The research literature includes studies of direct care workers and their clients in a variety of settings such as nursing homes (Berdes & Eckert, 2007; Foner, 1994), home care (Karner, 1998), and assisted living facilities (Ball, Lepore, Perkins, Hollingsworth, & Sweatman, 2009). This literature includes reports of the absence of relationships as well as the presence of close relationships. Little is known about the relative frequency of close positive relationships, or the ways in which they develop. However, a common theme across settings is that many workers experience family-like relationships with some of their clients (Ball et al., 2009; Berdes & Eckert, 2007; Karner, 1998). In her small-scale study of CNAs, Sumaya-Smith (1995) reported that more than 90% experienced a surrogate family bond with at least one resident, and more than 30% experienced such a bond with 10 to 50 residents. Stone (2000) and others (Bowers, Esmond, & Jacobson, 2000; Karner, 1998) have noted that care providers frequently use the word love to describe their relationships with residents and that they view the affective quality of their care relationships as an important determinant of residents’ quality of care and quality of life.

Similarly, Berdes and Eckert (2007) reported that caregivers who were African American or members of immigrant groups used the metaphor of family care to describe the high standard of care they worked to achieve. These caregivers extended their boundary of family to include the Caucasian residents for whom they cared. Female caregivers also referred to female residents as friends and girlfriends.

Additionally, in a study of care providers, residents, and their family members, McGilton and Boscart (2007) reported that caregivers felt close to residents when they believed they knew the residents’ preferences (i.e., what they liked) and their personality (i.e., how they were likely to behave). A second basis for connecting with residents was reciprocity. Reciprocity was “described as an engagement, a partnership of both the resident and the care provider in the relationship: ‘We do this together, it’s a give-and-take relationship’” (McGilton & Boscart, 2007, p. 2152). Both residents and family members identified providers’ lack of trustworthiness and lack of interest in the resident as a person as barriers to relationship development.

The bulk of this literature concerns the experiences of caregivers in nursing homes. However, research by Ball et al. (2009) indicated that a similar variety of relationship experiences occur in assisted living facilities. Currently unexamined is the extent to which the nature of CNA-resident relationships differ in assisted living facilities versus nursing homes.

In summary, these studies and the related literature indicate that close relationships do exist between residents and care workers. A useful next step would be to develop a more detailed description of such relationships, including the kinds of experiences that lead CNAs to think of their relationships with residents as being like those with family or friends. Of interest is learning more about the extent to which commitment, trust, and pro-relationship behaviors take place in CNA-resident interactions. These are relationship processes that interdependence researchers have demonstrated to be characteristic of a wide range of relationships, including friendships and family relationships, as well as romantic relationships (Rusbult et al., 2004).

Interdependence Theory

Interdependence theory is a social psychological theory that has been applied primarily to romantic relationships between adults who are presumably equals in terms of power. However, due to the basic similarities shared by all relationships, as described in the theory, it can also be applied to relationships characterized by nonmutual dependence, where one person is in a relatively more powerful and less dependent position (e.g., helping relationships, parent-child relationships) (Rusbult et al., 2004). Interdependence theory includes two important aspects to understanding the relations between individuals: the structure and the process of interdependence. The structure of interdependence is best conceptualized as an outcome matrix involving two individuals, two possible behavior strategies, and the consequences of those choices as represented by the outcomes of each person. One well-known type of interdependence structure in relationships characterized by mutual dependence is the “prisoner’s dilemma” (Deutsch, 1973). This matrix is constructed such that each partner has two strategic options: to cooperate or compete. The best possible outcome for both partners occurs when each decides to cooperate.

This study applies interdependence theory to CNA-resident relationships that involve nonmutual dependence. Relationships that involve nonmutual dependence can be thought of as a prototype for caregiving relationships, both paid and voluntary. Interdependence theory is well suited to explore the nature of interdependence within such relationships because it poses several dilemmas central to caregiving relationships. Behaviorally, individuals in the more dependent role choose how much they are willing to expose their vulnerability and express their needs; alternatively, they can be defensive and resist help. Behaviorally, individuals in the less dependent role choose the extent to which they will be responsive to the other; alternatively, they can be indifferent or exploitative (Rusbult et al., 2004). Within caregiving relationships, it would be expected that the best possible outcomes would occur when the more dependent individuals expose their vulnerability and expresses their needs, and the less dependent person is responsive. Of interest is how such patterns of interdependence are established.

By virtue of arrangements within LTC facilities, residents and CNAs can be thought of as being involved in interdependent relationships. The CNA is dependent on cooperation from the resident to accomplish the required care tasks. Likewise, the resident is dependent on the CNA to provide needed care. Of interest is how CNAs and residents negotiate their interdependence. From the perspective of interdependence theory, residents’ willingness to express their needs and CNAs’ responsiveness is reciprocally determined. A CNA’s expression of interest in a resident’s preferences may increase the likelihood that the resident will disclose his or her needs. Conversely, a resident’s grace in forgiving a CNA’s mistake may increase the CNA’s responsiveness to the resident.

The process of interdependence is determined by each individual’s needs, thoughts, motives, and behaviors in relation to the other (Rusbult et al., 2004). The theory assumes that individuals start out being primarily self-interested and oriented to short-term outcomes. However, as a result of a series of interactions in which partners may choose cooperative behavioral strategies, they may begin a process that Rusbult et al. (2004) labeled “mutual cyclic growth” (p. 154). Mutual cyclic growth is a pattern found in romantic relationships involving a reciprocal, long-term association between pro-relationship behaviors, trust, dependence, and commitment. This study assumes that mutual cyclic growth can also occur in relationships characterized by nonmutual dependence.

Of interest in this study was whether CNAs talked about their interactions with residents in ways that indicated they thought of themselves as being involved in relationships with residents; more specifically, whether they talked about their interactions with residents in ways that could be coded as involving pro-relationship behaviors, trust, dependence, and commitment (i.e., the elements of mutual cyclic growth). Within interdependence theory, a possible outcome of mutual cyclic growth is cognitive interdependence, and this was of interest as well. Cognitive interdependence is a term used to characterize relationships in which partners come to describe themselves and the other as us or as a unit (e.g., family) (Agnew & Etcheverry, 2006).

Study Purpose

The purpose of this study was to explore the applicability of these concepts of interdependence theory to CNA-resident interactions and to explore the extent to which the development of closeness in CNA-resident relationships follows a pattern of mutual cyclic growth. This research was guided by the question: When close relationships exist between CNAs and residents, how applicable are the concepts of interdependence theory to CNAs’ descriptions of these relationships and of the types of factors that facilitate or limit the development of close CNA-resident relationships?


Research Design

An exploratory, qualitative design was used to determine the extent to which concepts related to interdependence theory applied to CNAs’ experiences of relationships with residents and to the factors that facilitated or limited the development of these relationships. Seventeen 90-minute, semi-structured personal interviews were conducted with CNAs working in a LTC facility. A purposive sample was selected whereby an administrator from each participating LTC facility identified one or two socially skilled CNAs within their facility who then participated in a face-to-face interview. Administrators were asked to select CNAs who were skilled at establishing rapport with residents, families, and staff. The rationale for selecting socially skilled CNAs was that such CNAs likely had good communication skills in relating to residents and constituted credible sources of experiential expertise that might be taught to CNAs with less social skill.

Sample and Setting

Nine LTC facilities were selected based on convenience to represent a diversity of caregiving arrangements: Five were proprietary and four were nonprofit. The for-profit facilities were diverse in size and managerial practice, and each was part of a larger chain. One proprietary facility operated as a Green House® model (Weiner & Ronch, 2003) and was a residential home that had been converted to a group home. This facility provided care for six residents diagnosed with Alzheimer’s disease. The nonprofit facilities also varied in size and managerial practice. Several were associated with religious organizations (i.e., Catholic, Mennonite, and Methodist), and all were members of a larger chain. Two of the proprietary facilities had both assisted living and nursing home components, as did two of the nonprofit facilities. The assisted living components of both the proprietary and nonprofit facilities ranged in size from 35 to 89 licensed beds (mean = 51), and the nursing home components, with the exception of the Green House model, ranged from 42 to 90 licensed beds (mean = 79).

To increase diversity, CNAs from both assisted living facilities (n = 13) and nursing homes (n = 4) were included in the study. These CNAs were identified and recruited separately at each of the nine participating facilities by one administrator at each facility. The criterion for inclusion in the study was the administrator’s judgment that the CNA was skilled in establishing rapport with residents, family, and staff.

Ethical Considerations and Procedures

After approval from the University’s Institutional Review Committee, administrators at each facility selected socially skilled CNAs at their facility. Selected CNAs were then given a letter describing the study and were asked to contact the research team for information and to schedule an interview. All of the selected participants agreed to be interviewed.

Measures and Data Collection

This exploratory qualitative study focused on CNAs’ answers to a series of questions and probes about their relationships with residents. The interviews were framed by an introduction to relationships: “Sometimes CNAs have closer relationships with some residents than others. This is something beyond working with residents and helping them with their different needs—hygiene, dressing, et cetera.” The following four probes were then introduced:

  • Are there residents whom you feel especially close to?
  • Please describe the kinds of things that led to this closeness (with appropriate probes as warranted).
  • Are there some residents whom you don’t feel especially close to?
  • What kinds of things limit the closeness you feel toward this resident?

These probes about relationships were part of a longer interview concerning CNA-resident relationships. These 90-minute audiorecorded personal interviews took place with each participating CNA. The interviews were arranged during the CNAs’ workdays with the help of administrators at each facility. The lead author (L.J.M.) trained and supervised two psychology doctoral students who conducted the 17 interviews. All interviews were confidential, and participants were given the option of editing portions they did not want included. None of the participants edited their interviews.

Data Analysis and Ensuring Rigor

Structured content analysis was used to determine how the concepts of interdependence theory fit CNAs’ descriptions of their interactions with residents or whether new categories would emerge. Specifically, directed content analysis was used, a content analytic technique that Hsieh and Shannon (2005) identified as appropriate when the goal is to validate and extend an already existing set of categories derived from previous research.

The interviews were transcribed, coded, and judged as to whether CNAs’ responses to the interview questions could be captured by Rusbult et al.’s (2004) concepts. The concepts developed by Rusbult et al. (2004) in articulating the idea of mutual cyclic growth constituted most of the a priori categories: pro-relationship behaviors that lead to trust, increased dependence, and commitment. The category of cognitive interdependence, a frequent consequence of commitment, was also included. General categories of pro-relationship behaviors include accommodation (i.e., “turning the other cheek”) and willingness to sacrifice (i.e., forgoing short-term benefit for the long-term welfare of the relationship). Additionally, giving positive regard by expressing approval of the resident or spending time with the resident were coded as pro-relationship behaviors on the basis of previous research by the first author (Medvene & Lann-Wolcott, 2010). With regard to trust and dependence, expressions of trust and/or comfort with dependence from the perspective of either the CNA or the resident were coded under these categories. Commitment refers to both the development of a long-term orientation and to expressions of psychological attachment (Rusbult et al., 2004); CNAs’ comments related to these ideas were coded under the category of commitment. In addition, CNAs’ references to their relationships with residents in terms of we, family, or friendship were coded as examples of cognitive interdependence on the basis of prior research (Agnew & Etcheverry, 2006).

Hsieh and Shannon (2005) suggested creating an audit trail to overcome any bias that might lead researchers to confirm the existing set of categories, rather than discovering new ones. The following audit process was used to establish the trusthworthiness of the findings. Each audiorecorded interview was transcribed. The second author and an undergraduate student coded the first several interviews and explored the applicability of the major concepts of interdependence theory. The coders were surprised by the number of times CNAs’ descriptions of their interactions with residents referred to their interdependence with residents. To capture this content, the coders created a macrocategory titled Acknowledgment of Interdependence. This category was defined as acknowledging two people are in a relationship and that each person’s behavior influences both the positive and negative outcomes of the other. Acknowledgement of Interdependence included four subcategories that referred to slightly different aspects of interdependence: Relationship Awareness, Partner Perspective Taking, Cognitive Interdependence, and Organizational Interdependence. These subcategories were named based on well-established concepts from the social-psychological research on close relationships.

The final set of coding categories included nine other macrocategories: Commitment, Barriers, CNAs’ Pro-Relationship Behaviors, Trust, Residents’ Pro-Relationship Behaviors, Extended Temporal Structure, Empowerment, Residents’ Dependence, and Miscellaneous (Table). The coders created these categories in the process of coding the first several interviews. The coders reached a saturation point after coding the fourth interview and determined that no new categories needed to be added to code the remaining 13 transcripts.

Coding Categories Related to CNAs’ InterdependenceCoding Categories Related to CNAs’ Interdependence

Table: Coding Categories Related to CNAs’ Interdependence

The second stage of the coding process involved unitizing the section of the transcripts that concerned CNAs’ interactions with residents. Once these categories and subcategories were developed, criteria for unitizing the interviews were created by the second author. The criteria for unitizing was based on portions of the transcript that seemed to fall logically into a distinctive category. A new unit was created each time the coding category changed or when a single episode of a codeable behavior emerged. A total of 524 units were created throughout the 17 interviews. Both coders coded all 17 interviews. On the basis of the coding of Interviews 5 through 17, the Cohen’s kappa coefficient was 0.77 (Cohen, 1960), which indicated a high rate of intercoder reliability.



The sample of CNAs included 16 women and 1 man. The CNAs ranged in age from 22 to 67 (mean = 41.5, SD = 12.15 years). The average number of residents with whom CNAs worked in a typical week was 32. CNAs typically worked 8-hour shifts during one of the three blocks of time. Fifty-nine percent of the CNAs worked the first shift, 35% worked the second shift, and 6% worked the third shift. The mean number of years participants had worked at the facility was 7 (SD = 6.5; range = 1 to 20). Participants’ level of education varied; 3 participants had less than a high school education, 9 were high school graduates, and 5 had some college.

Closeness and CNA-Resident Interactions

All 17 CNAs reported that they had close relationships with many residents, and all reported multiple examples of specific people and/or instances in which they had close relationships. For example:

I have quite a few [close relationships] over the years. It is hard not to get attached or close, especially when you have a resident that doesn’t have family close. I mean, they really depended on you, not just for emotional support, but you know, I’ve stopped on the way to work to pick up hearing aid batteries, or ran to the store for snacks or different things. So…it really does go above and beyond, but it’s, they become so much like your family, that it’s not really a job, you do it because…essentially, they are your friend or your family. I don’t see how there would be a possible way that you could work with someone that much and not become close to them.

Applicability of Interdependence Theory

The coding categories and their definitions are presented in the Table. There were 10 macrocategories, four of which included at least one subcategory. The relative frequency of each macrocategory was measured in two ways: the percentage of CNAs who used this category and the percentage of the total units accounted for by the category. An example of each category is included in the Table.

Acknowledgement of Interdependence. All of the CNAs engaged in Acknowledgement of Interdependence, which was the most frequent category, accounting for 44% of the total units. By definition, this category involved recognizing that two people are in a relationship and that each person’s behavior influences the other. This macrocategory had four subcategories: Relationship Awareness, Partner Perspective Taking, Cognitive Interdependence, and Organizational Interdependence.

The quote in the Table that exemplifies the subcategory of Relationship Awareness is rich and complex. Of interest is the self-awareness the CNA is demonstrating in talking about her feeling that a resident was “[getting] on [her] nerves.” She then described choosing a strategy of being “polite” to this resident. The quote describes the CNA’s awareness of relational dynamics: If she is merely polite, the resident will, in turn, be less likely to be “needy,” which will ultimately result in a somewhat distant relationship. This kind of relational thinking likely guided the CNA’s behavioral skills in caring for residents.

Cognitive Interdependence was a second important subcategory of Acknowledgement of Interdependence. This subcategory is defined as “a habit of thinking that supports pro-relationship motivation and behavior” (Agnew & Etcheverry, 2006, p. 282). All of the CNAs, in one quote or another, referred to their feelings of being like “family” or being “friends” with a resident.

Commitment. All of the CNAs interviewed had at least one unit that was coded as Commitment. This macrocategory accounted for 15% of the total units. Clearly, CNAs developed feelings of attachment and a long-term orientation toward many of the residents with whom they were working.

Barriers. While all of the CNAs had close relationships with some residents, almost all of them (94%) also talked about barriers that limited their ability to form close relationships. The example quoted in the Table cites time limitations. Additional factors included residents’ independence or their lack of interest in relationships. In other cases, CNAs described how residents’ communication difficulties were a factor.

Especially striking were CNAs’ comments that their training was a limiting factor and had not adequately prepared them to relate to the residents. One CNA commented:

That was one of the biggest things in my CNA class was “don’t be personal!” Do not talk about your family life, do not talk about your personal life, and do not talk about your faith, your religion. Do not talk about any of this stuff. You are not their friend! You are their caregiver. I mean, I came in here thinking I couldn’t talk to them about anything! And then I finally realized that’s not me. So, I’m gonna do it the way I think it needs to be done, and I’m still here, so I must be doing it okay.

CNAs’ Pro-Relationship Behaviors. Nearly 90% of the CNAs described pro-relationship behaviors they enacted to help residents. The example in the Table involved listening (i.e., a form of giving positive regard). Other examples included asking about a resident’s welfare or showing interest in the resident. The following quote demonstrates a CNA’s strategic use of communication skills to build relationships. During a follow-up question about an CNA’s close relationship, the interviewer asked, “What kinds of things do you think led to that closeness?” The CNA answered:

Being open to visit, and recognizing their needs, showing care and concern about their welfare, recognizing when they’re comfortable or uncomfortable, and saying “Can I help you?,” or “Are you hurting right now?” And being receptive and then you notice then too that some people’s pain levels are different than others. Some can do more for themselves and work over their pain better and others, ya know, their pain level, you can see that it’s harder and that they need a little extra assistance because it’s a little harder for them to cope with.

Especially striking about this quote is the CNA’s conscious use of her communication skills as a tool to build relationships with residents. This quote also demonstrates the CNA’s highly individualized approach to helping—noticing differences between residents in their tolerance of pain.

Trust. Eighty-eight percent of the CNAs made comments that related to trust. Many CNAs described how it is easier to work with a resident if the resident trusts the CNA. For example, one CNA said, “They feel more comfortable with you, they have more trust in you.”

Residents’ Pro-Relationship Behaviors. CNAs described specific behaviors performed by residents that seemed to involve some kind of reciprocation for the help CNAs had given them. Seventy-nine percent of the CNAs described residents’ pro-relationship behaviors; this category accounted for 3% of the total units. Many CNAs described residents asking them if they “had a good weekend” or commenting that residents would be forgiving if the CNA made a mistake or forgot to do something.

Extended Temporal Structure. This macrocategory refers to the length of the relationship or its growth over time. Fifty-three percent of the CNAs made reference to extended temporal structure, and this category accounted for 3% of the total units.

Empowerment. This macrocategory refers to allowing residents to have more power in the decision-making process. Giving Choices was a subcategory. Twenty-nine percent of the CNAs made reference to empowerment, and this category accounted for 2% of the total units.

Residents’ Dependence. This macrocategory was exemplified by statements regarding the degree to which an individual relies on a partner, in that his or her outcomes are influenced by the partner’s actions. Twenty-three percent of the CNAs made reference to dependence, and this category accounted for 1% of the total units.

Comparison Between CNAs in Assisted Living And Nursing Home Settings

Because the sample included CNAs from assisted living facilities, as well as nursing homes, the coding of CNAs’ comments in both settings was compared. The percentages of CNAs whose responses were coded into each of the nine macrocategories were calculated. With the exception of Residents’ Pro-Relationship Behaviors, roughly equivalent percentages of the CNAs in both settings made comments that were coded into each macrocategory. In the case of Resident Pro-Relationship Behaviors, 62% of CNAs working in assisted living settings made comments coded into this category, versus 100% of the CNAs working in nursing home settings. In the cases of the remaining eight macrocategories, differences between the percentages of CNAs whose comments were coded into each category varied by 15% or less.


The research question addressed by this study concerned the applicability of the concepts of interdependence theory to CNAs’ descriptions of their relationships with residents and the factors that facilitated or limited these relationships. The fact that all of the CNAs reported developing close relationships with at least some residents made this a viable data set to answer this question. The categories that proved useful in coding CNAs’ descriptions of their relationships were largely drawn from interdependence theory. Following the logic of directed content analysis, the goal of which is to validate and extend an already existing set of categories, the coding here constitutes evidence supporting the validity of applying the concepts from interdependence theory to CNA-resident relationships. For example, all of the CNAs referred to commitment, and more than 80% referred to the concepts of pro-relationship behaviors on their own part, as well as on the part of the residents, as well as trust. A note of caution is appropriate, however, as the use of directed content analysis may bias the results toward confirmation. This is discussed below under Study Limitations.

Especially applicable were concepts related to cognitive interdependence (i.e., CNAs’ talking about the residents as closely linked to themselves, such as a family member or a friend). Emerging from the process of coding CNAs’ descriptions of their relationships with residents was the relational nature of their talk. All of the CNAs spoke with an awareness that the outcomes of their interactions with residents were determined by both their own and the residents’ behaviors. The CNAs demonstrated an awareness of interaction patterns whereby their own behaviors could predictably lead to distancing (i.e., if they were merely polite) or to engagement (i.e., if they showed personal interest). This relationship mindedness is likely a cognitive skill associated with the social skills that distinguished the CNAs in this study from their peers. More generally, all of the elements of mutual cyclic growth were represented in the coding categories. In addition, there was some evidence that CNAs talked about the sequential processes explicit in mutual cyclic growth. Explored in the discussion below are the implications of applying interdependence theory to CNA-resident relationships, both for theorizing about the distinctive nature of these relationships, as well as for training CNAs and nurses.

Developing Relationships with Residents

Applied to CNA-resident relationships, the model of mutual cyclic growth can be thought of in terms of a continuum. For CNAs and residents, mutual cyclic growth likely involves CNAs’ pro-relationship behaviors leading to residents’ increased comfort with dependence and trust, thus leading to commitment and to residents’ reciprocal pro-relationship behaviors benefitting CNAs. In some CNA-resident relationships, relatively little trust, comfort with interdependence, or commitment develops. In other relationships, greater trust and interdependence may be generated. High levels of commitment and cognitive interdependence may be developed in relatively few. From the perspective of interdependence theory, a reasonable goal in CNA-resident relationships might be to develop moderate amounts of trust and comfort with interdependence. Unlike romantic relationships, long-term commitment may not be widely desired or realistic.

Person-centered caregiving, as noted in the introduction, includes a relational component, especially with regard to caregiver-resident interactions. Interdependence theory offers a picture of what such caregiving relationships might be like and how they might be created. From the caregivers’ perspective, person-centered caregiving relationships would seem to require caregivers’ acknowledgement of the interdependence of their interactions with residents. It would also seem to require that caregivers have a large repertoire of pro-relationship behaviors, especially communication skills, such as giving positive regard (Medvene & Lann-Wolcott, 2010). Communication skills are common content for training CNAs. The findings from the current study indicate that communication skills can be taught as important tools for building cooperative relationships with residents.

Boundary Issues in CNA-Resident Relationships

From the perspective of interdependence theory, which has been primarily applied to romantic relationships, CNA-resident relationships are distinctive in terms of the boundary issues they raise. As the findings in this and other studies point out, CNAs’ interactions with residents become more complicated when they also begin to see themselves as “family” or “friends” of the residents. This additional level of complexity in CNA-resident interactions can create boundary problems (i.e., the expectation that family members should “always be there for each other” conflicts with the circumscribed hours and job descriptions of CNAs). Challenging ethical, legal, and moral issues are raised when CNAs relate to residents in the context of multiple roles. The American Association of Homes and Services for the Aging (AAHSA, 2006) has already held one audio conference about such problems. For example, boundary violations arise when residents want to give CNAs gifts or when CNAs want to spend time with residents—either as friends or as private nurses—in roles that go beyond their employment as CNAs in a LTC facility. Viewing such boundary issues from the perspective of interdependence theory may help clarify the conflicting norms and expectations.

Implications for Training

Inservice training programs for CNAs should encourage CNAs to think about their interactions with residents in relational, as well as dispositional, terms. Since the philosophy of person-centered caregiving encourages CNAs to develop close relationships with residents, training programs should normalize the appropriateness of such relationships. One CNA in the current study said she was taught: “Don’t get personal…. You are not their friend, you are their caregiver.” However, if inservice training programs normalize close relationships, such programs will also need to give CNAs guidance in dealing with boundary issues, personal relationships, and multiple roles. CNAs should be encouraged to develop a repertoire of behavioral approaches to residents and to think strategically about interaction patterns that can lead to engagement as well as distancing. For example, training programs should give CNAs a vocabulary for talking about boundary issues and could normalize and validate CNAs’ feelings of closeness, affection, and family-like relationships. At the same time, training programs could also provide clear guidance about the need to refrain from acting inappropriately on these feelings. Additionally, CNAs will vary in their comfort and ability to manage the ambiguity of being both a caregiver and a friend (Medvene, Grosch, & Swink, 2006), should such opportunities present themselves. Training programs will need to help CNAs develop relational styles with which they are comfortable.

Developing training programs that deal with boundary issues would be innovative. The current research literature includes reports of inservice training programs that increased CNAs’ social conversation with residents, therapeutic communication, and development of more cooperative relationships around caregiving tasks (Burgio et al., 2001; Levy-Storms, 2008; Sloane et al., 2004; Williams, 2006; Williams, Ilten, & Bower, 2005). McGilton et al. (2003) developed an intervention that increased the relational aspects of nursing staff’s interactions with nursing home residents but did not deal with boundary issues. Currently, many facilities deal with boundary issues through policies that limit gift-giving and allow or prohibit staff from joint employment where they are working both as facility staff and as employees of a resident (e.g., as a caregiver) (AAHSA, 2006)

Study Limitations and Future Research

The current study was small in scale and exploratory; because of this and other reasons, the generalizability of the results may be very limited. There is a need for research involving larger samples that are varied in terms of the functional level of residents and in terms of the size and program models of LTC facilities. Future studies should include larger numbers of CNAs from both nursing homes and assisted living facilities to determine the extent to which differences in settings influence CNA-resident relationships. Nursing home residents are more dependent than assisted living residents in terms of instrumental activities of daily living. Likewise, practices in some nursing homes may encourage dependency, while practices in some assisted living facilities encourage residents’ independence (Ball et al., 2004).

Additionally, the sample in the current study was selected by administrators who were asked to pick socially skilled CNAs. This likely biased the sample toward finding positive relationships. Future studies should involve random selection so that CNAs with varying levels of social skills would be included, and a wider variety of relationships would be sampled. Other factors that influence CNA-resident relationships such as gender and culture need to be taken into account in future research.

One of the challenges to naturalistic inquiry that directed content analysis presents is that it may involve a bias to find evidence that is supportive of the given theory (Hsieh & Shannon, 2005). While past interdependence research provided clear a priori definitions of most of the coding categories used in this study, it is possible that coders in this study missed contextual factors that may have led to different results. Future studies could include observational data of CNAs’ interactions with residents and could be supplemented by CNAs’ comments about their thoughts of what was going on during their interactions with residents. These data could confirm the relevance of interdependence theory.


Consistent with previous research, the current study provided evidence that CNAs develop close relationships with residents. The findings also demonstrate that interdependence theory is helpful in understanding the nature of close relationships that can develop between CNAs and residents. Future training programs for both CNAs and nurses could include content based on interdependence theory, which normalizes close relationships and provides guidelines for maintaining role boundaries.


  • Agnew, C.R. & Etcheverry, P.E. (2006). Cognitive interdependence: Considering self-in-relationship. In Vohs, K.D. & Finkel, E.J. (Eds.), Self and relationships: Connecting intrapersonal and interpersonal processes (pp. 274–293). New York: Guilford.
  • American Association of Homes and Services for the Aging. (2006, December14). Relationships between consumers and staff: The HR and legal implications [Audio conference]. Washington, DC: Author.
  • Ball, M.M., Lepore, M.L., Perkins, M.M., Hollingsworth, C. & Sweatman, M. (2009). “They are the reason I come to work”: The meaning of resident-staff relationships in assisted living. Journal of Aging Studies, 23, 37–47. doi:10.1016/j.jaging.2007.09.006 [CrossRef]
  • Ball, M.M., Perkins, M.M., Whittington, F.J., Hollingsworth, C., King, S.V. & Combs, B.L. (2004). Independence in assisted living. Journal of Aging Studies, 18, 467–483. doi:10.1016/j.jaging.2004.06.002 [CrossRef]
  • Berdes, C. & Eckert, J.M. (2007). The language of caring: Nurse’s aides’ use of family metaphors conveys affective care. The Gerontologist, 47, 340–349. doi:10.1093/geront/47.3.340 [CrossRef]
  • Bowers, B.J., Esmond, S. & Jacobson, N. (2000). The relationship between staffing and quality in long-term care facilities: Exploring the views of nurse aides. Journal of Nursing Care Quality, 14(4), 55–64. doi:10.1097/00001786-200007000-00009 [CrossRef]
  • Bowers, B.J., Fibich, B. & Jacobson, N. (2001). Care-as-service, care-as-relating, care-as-comfort: Understanding nursing home residents’ definitions of quality. The Gerontologist, 41, 539–545. doi:10.1093/geront/41.4.539 [CrossRef]
  • Burgio, L.D., Allen-Burge, R., Roth, D.L., Bourgeois, M.S., Dijkstra, K., Gerstle, J. & Bankester, L.,… (2001). Come talk with me: Improving communication between nursing assistants and nursing home residents during care routines. The Gerontologist, 41, 449–460. doi:10.1093/geront/41.4.449 [CrossRef]
  • Campbell, S.L. (2003). Empowering nursing staff and residents in long-term care. Geriatric Nursing, 24, 170–175. doi:10.1067/mgn.2003.40 [CrossRef]
  • Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37–46. doi:10.1177/001316446002000104 [CrossRef]
  • Crandall, L.G., White, D.L., Schuldheis, S. & Talerico, K.A. (2007). Initiating person-centered care practices in long-term care facilities. Journal of Gerontological Nursing, 33(11), 47–56.
  • Deutsch, M. (1973). The resolution of conflict: Constructive and destructive processes. New Haven, CT: Yale University Press.
  • Foner, N. (1994). Nursing home aides: Saints or monsters?The Gerontologist, 34, 245–250. doi:10.1093/geront/34.2.245 [CrossRef]
  • Hsieh, H.-F. & Shannon, S.E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15, 1277–1288. doi:10.1177/1049732305276687 [CrossRef]
  • Hullett, C.R., McMillan, J.J. & Rogan, R.G. (2000). Caregivers’ predispositions and perceived organizational expectations for the provision of social support to nursing home residents. Health Communication, 12, 277–299. doi:10.1207/S15327027HC1203_04 [CrossRef]
  • Karner, T.X. (1998). Professional caring: Homecare workers as fictive kin. Journal of Aging Studies, 12, 69–82. doi:10.1016/S0890-4065(98)90021-4 [CrossRef]
  • Levy-Storms, L. (2008). Therapeutic communication training in long-term care institutions: Recommendations for future research. Patient Education and Counseling, 73, 8–21. doi:10.1016/j.pec.2008.05.026 [CrossRef]
  • McGilton, K.S. & Boscart, V.M. (2007). Close care provider-resident relationships in long-term care environments. Journal of Clinical Nursing, 16, 2149–2157. doi:10.1111/j.1365-2702.2006.01636.x [CrossRef]
  • McGilton, K.S., O’Brien-Pallas, L.L., Darlington, G., Evan, M., Wynn, F. & Pringle, D.M. (2003). Effects of a relationship-enhancing program of care on outcomes. Journal of Nursing Scholarship, 35, 151–156. doi:10.1111/j.1547-5069.2003.00151.x [CrossRef]
  • Medvene, L., Grosch, K. & Swink, N. (2006). Interpersonal complexity: A cognitive component of person-centered care. The Gerontologist, 46, 220–226. doi:10.1093/geront/46.2.220 [CrossRef]
  • Medvene, L.J. & Lann-Wolcott, H. (2010). An exploratory study of nurse aides’ communication behaviours: Giving ‘positive regard’ as a strategy. International Journal of Older People Nursing, 5, 41–50. doi:10.1111/j.1748-3743.2009.00183.x [CrossRef]
  • Mittal, V., Rosen, J. & Leana, C. (2009). A dual-driver model of retention and turnover in the direct care workforce. The Gerontologist, 49, 623–634. doi:10.1093/geront/gnp054 [CrossRef]
  • Pietrukowicz, M.E. & Johnson, M.M.S. (1991). Using life histories to individualize nursing home staff attitudes toward residents. The Gerontologist, 31, 102–106. doi:10.1093/geront/31.1.102 [CrossRef]
  • Rusbult, C.E., Kumashiro, M., Coolsen, M.K. & Kirchner, J.L. (2004). Interdependence, closeness, and relationships. In Mashek, D.J. & Aron, A. (Eds.), Handbook of closeness and intimacy (pp. 137–161). Mahway, NJ: Erlbaum.
  • Sloane, P.D., Hoeffer, B., Mitchell, C.M., McKenzie, D.A., Barrick, A.L., Rader, J. & Koch, G.G.,… (2004). Effect of person-centered showering and the towel bath on bathing-associated aggression, agitation, and discomfort in nursing home residents with dementia: A randomized, controlled trial. Journal of the American Geriatrics Society, 52, 1795–1804. doi:10.1111/j.1532-5415.2004.52501.x [CrossRef]
  • Stone, D. (2000). Caring by the book. In Meyer, M.H. (Ed), Care work: Gender, labor, and the welfare state (pp. 89–111). New York: Routledge.
  • Sumaya-Smith, I. (1995). Caregiver/resident relationships: Surrogate family bonds and surrogate grieving in a skilled nursing facility. Journal of Advanced Nursing, 21, 447–451. doi:10.1111/j.1365-2648.1995.tb02726.x [CrossRef]
  • Weiner, A.S. & Ronch, J.L. (Eds.). (2003). Culture change in long-term care. New York: The Haworth Press.
  • White, D.L., Newton-Curtis, L. & Lyons, K.S. (2008). Development and initial testing of a measure of person-directed care. The Gerontologist, 48(Suppl. 1), 114–123. doi:10.1093/geront/48.Supplement_1.114 [CrossRef]
  • Williams, K.N. (2006). Improving outcomes of nursing home interactions. Research in Nursing & Health, 29, 121–133. doi:10.1002/nur.20117 [CrossRef]
  • Williams, K.N., Ilten, T.B. & Bower, H. (2005). Meeting communication needs: Topics of talk in the nursing home. Journal of Psychosocial Nursing and Mental Health Services, 43(7), 39–45.

Coding Categories Related to CNAs’ Interdependence

Category/Subcategory % CNAs % Units Description/Examples
Acknowledgement of interdependence 100 44 Acknowledging there are two people in a relationship and that each person’s behavior influences the other.
Relationship awareness Showing awareness of a relationship by thinking or talking about the interaction patterns, comparisons, or contrasts between the CNA and the resident. EXAMPLE: “I had one resident that if you’re nice to her, then she just will come onto you so much, and I know she’s a needy person. So I try to be polite to her, but yet I don’t think I could ever be as close to her as I could some other residents, just because, I guess because I feel sorry for her but yet sometimes she gets on your nerves.”
Partner perspective taking Thinking about the situation from the resident’s viewpoint, respecting the resident’s preferences, or showing empathy for the resident. EXAMPLE: “It’s not like you’re going to Boeing and working on the same machine every single day. Every day is a new day, I mean they have good days, bad days, and you need to approach them with that, you know, ‘So this is a bad day for you, maybe you don’t want to do anything and that’s fine.’”
Cognitive interdependence Referring to the resident as family, as a friend, using we or our, or thinking about the residents and staff as one. EXAMPLE: “They become so much like your family, that it’s not really a job, you do it because, essentially, they are your friend, family.”
Organizational interdependence Acknowledging interdependence on an organizational level. The CNA acknowledges the link between organizational structure and relationship building. EXAMPLE: “It’s not just the aides, the nurses also get attached. It’s like a little community. We have independent living, assisted living, and then the nursing home, so it is a community all in itself.”
Commitment 100 15 Having a long-term orientation toward the relationship with the resident, which includes intent to persist and feelings of psychological attachment.
Closeness Acknowledging or referring to a sense of closeness or comfort between CNAs and the resident. EXAMPLE: “I would have to say that the more I do for them, the more I’m going to get to know them. I think that’s why I get so close to them because they feel comfortable with me and I feel comfortable with them.”
Barriers 94 11 Specific barriers to relationship building. EXAMPLE: “A lot of times, people tend to think, ‘I’m so busy. I don’t have time to do this.’ You don’t have to go in and spend 20 minutes. You can go in and spend a few minutes with them. A few minutes of your time makes a big difference in their lives.”
CNAs’ pro-relationship behaviors 88 13 Specific behaviors performed by the CNA that lend themselves to building or maintaining a mutually positive relationship.
Positive regard Acknowledging the resident, treating the resident with respect, and showing interest in, approval of, and/or liking of the resident. EXAMPLE: “You stop in and to say ‘Hey, I was here, you know, I’m not working, how was your day?’ And they’ll look at you and say, ‘Well, life is good.’ And, they’ll start talking, talking, talking.”
Positive attitude Creating a positive, upbeat environment. EXAMPLE: “If you got both [conversation and humor] these people just eat it up…. It makes it go a lot better, I mean, you don’t want to be straight-faced all the time, it just ain’t gonna work.”
Self-disclosure Telling residents about themselves or sharing themselves with the resident. EXAMPLE: “Mostly it’s open to everybody with me, that’s why I have a relationship with them really good, you tell them about your life, it’s just you need to be open, you don’t have to tell them your financial thing, but it’s up to you, just like you’ve been on vacation, England, Scotland, Canada, and wherever, and they get interested too, they say, ‘Oh, I’ve been there too,’ so it’s really nice”.
Accommodation Reacting in a constructive way when the resident’s behavior is destructive toward the CNA. EXAMPLE: “The ones that are real combative, the ones that are what I call ‘ugly,’ their disease has made them ‘ugly.’ They’re very hard to get close to, but you still care for ‘em, because you know that they probably weren’t like this.”
Retaining dignity with conversation Relying on light conversation to help the resident feel more relaxed while completing an unpleasant task with the resident. EXAMPLE: “And there again dignity, you know, if people are incontinent, if they had to be fed, you know, make them feel good about it, you now, do something else to take that attention away from what their needs really are, make them feel good about something that they can do…while you’re doing everything else, talk about something else.”
Advocacy Advocating on behalf of the resident with management. EXAMPLE: “Well I say, ‘I can’t answer that question for you right now, but I’ll go find out how I can help you or how you can receive help.’ So sometimes you can’t give them the full answer, you have to say, ‘I’ll look into it,’ or I’ll direct them to this person or that person that can help them with their needs.”
Nonjudgmental attitude Creating an open, accepting environment. EXAMPLE: “But you learn over the years, you learn in your work that you have to accept everybody and that everybody is different.”
Willingness to sacrifice Forgoing immediate self-interest to promote the well-being of the resident. EXAMPLE: “If I see one that’s really sick or they having a problem eating, I don’t have a problem coming in on my day off, ya know, around lunchtime or dinnertime or whatever. If they’ll eat for me, I’m gonna make the sacrifice and come. Ya know, ‘cause it’s important”.
Gratitude Feelings of gratefulness or thankfulness on the part of either the CNA or resident toward the other. EXAMPLE: “So there’s that bond to be able to appreciate each other’s lives along the way.”
Trust 88 6 Expecting that a given partner can be relied on to behave in a benevolent manner and be responsive to one’s needs. Examples of trust would be the returning to work nearly every day, consistency in their responsiveness and caregiving abilities, looking forward to seeing the other, the mention of privacy or confidentiality, or the outright acknowledgement of trust between the CNA and the resident. EXAMPLE: “If you approach them with confidence, like you know what you’re doing and they’re going to be safe…then they’re going to do what you need them to do because they trust you, it’s trust like with any relationship.”
Residents’ pro-relationship behaviors 79 3 Specific behaviors performed by the resident that lend themselves to building or maintaining a mutually positive relationship. EXAMPLE: “It’s a happy time to greet one another and catch on how you have been and then they’ll share what’s been happening in their lives. And they will say, ‘What have you been doing?’ or ‘Did you have a good weekend?’”
Extended temporal structure 53 3 Referencing time and/or growth. EXAMPLE: “It grows over time like any relationship, you’re not going to develop this in 2 or 3 days, you know, but as time goes on you become very close to people”.
Empowerment 29 2 Allowing the resident to have more power in the decision-making process, emphasizing more independence rather than dependence for the resident, using a cooperative approach, or attempting to equalize power by placing the resident in charge of the interaction.
Giving choices Giving residents choices such as the clothing they would like to wear or food they would like to eat. EXAMPLE: “I think it’s very important that you ask, you do not say, ‘Okay, let’s go to the table, I’m going to take you out of the chair and take you to the table, I’m going to take you to the bathroom, it’s time to change your clothes.’ You approach them: ‘May I?,’ or ‘Is it alright if we go back to the bathroom?,’ or ‘It’s almost lunchtime, may I take you to the table?’ You are going to get a lot more out of them if you ask.”
Residents’ dependence 23 1 The degree to which an individual relies on a partner, in that his or her outcomes are influenced by the partner’s actions. EXAMPLES: “And a lot of times we’re the only family those residents have”; “I know that the residents are dependent on me to meet their needs”.
Miscellaneous 53 2 The unit does not fit into any of the categories.

Dr. Medvene is Professor of Psychology, Department of Psychology, Wichita State University, Wichita, and Dr. Coleman is Postdoctoral Fellow, University of Kansas School of Nursing, Kansas City, Kansas.

The authors disclose that they have no significant financial interests in any product or class of products discussed directly or indirectly in this activity. This project was funded by a grant from the Wichita State University LINK project and by matching funding from Larksfield Place. The authors thank the following long-term care facilities for their support and cooperation with this project: Andover Court Assisted Living, Comfort Care Homes, Friendly Acres, Halstead Place, Kansas Masonic Home, Larksfield Place, The Vine at Derby, Park West Plaza, and Riverside Village.

Address correspondence to Louis J. Medvene, PhD, Professor of Psychology, Department of Psychology, Wichita State University, Box 34, 1845 Fairmount, Wichita, KS 67260-0034; e-mail:

Received: March 05, 2010
Accepted: November 12, 2010
Posted Online: April 13, 2011


Sign up to receive

Journal E-contents