THE RELATIONSHIP BETWEEN FAMILIES’ PERCEPTIONS AND NURSES’ PERCEPTIONS OF FAMILY NURSING PRACTICE

The prevalence of diabetes, a chronic illness, is expected to substantially rise over the next fifteen years (Whiting, Guariguata, Weil, & Shaw, 2011). One approach to ease the burden on the US health care system is the involvement and participation of family in care of the hospitalized adult. There is increasing evidence that involvement of family during exacerbations and hospitalizations increases client and family satisfaction during admissions and may also decrease length of stay and therefore cost (Powers & Rubenstein, 1999). The purposes of this study were to examine family members’ perceptions of family functioning, family health and the social support received from nurses when an older adult family member with diabetes is hospitalized. Also examined were nurses’ critical appraisals of their family nursing practice, as well as their experiences of the reciprocity and interaction in the nurse-family relationship. This study further explored the relationships between nurses’ critical appraisals of their family nursing practice and nurses’ experiences of the interaction and reciprocity in the nurse-family relationship with families’ perceptions of family function, family health and perceived social support from nurses. Finally, this study examined if nurses’ critical appraisal of their family nursing practice, and nurses’ experience of the interaction and reciprocity in the nurse-family relationship, differed across nursing units, and what the impact was on families’ perceptions of family function, family health and social support received. Wright and Leahey’s Calgary Family Intervention Model (CFIM) (1994) undergirded this descriptive study, which was conducted on four medical-surgical units in a community hospital. Sixty registered nurses and sixty family members of older adult patients participated. Two instruments were used to address the variables of interest in this study. Family member participants completed the Family Function, Family Health and Social Support Instrument (Astedt-Kurki, Tarkka, Paavilainen, & Lehti, 2002; Astedt-Kurki, Tarkka, Rikala, Lehti, & Paavilainen, 2009) as well as a demographic questionnaire. Registered Nurse participants completed demographics and the Family Nursing Practice Scale (Simpson & Tarrant, 2006). Significant variation was found across the four study units in how nurse participants reflected on their experiences with interaction and reciprocity in the nurse-family relationship. However, family member participants had no significant variation in their perceptions of family functioning, family health and social support received from nurses. This research informs practice by providing insight into nurses’ perceptions regarding the advantages and the disadvantages of working with families. Additionally, this study contributes evidence of what nurses are currently doing to include families in their nursing practice. More research is needed which focuses on collaboration and inclusion of families in care of their loved one.

. This is likely to continue to increase as members of the "baby boomer" generation move into older adulthood and reach an age where they require greater numbers of health care services. This increased demand for health services related to the ongoing management of complex chronic illness is projected to further strain the already overburdened and inefficient US health system, creating challenges in the ability of the system to provide adequate or cost-effective health care to the growing numbers of chronically ill older adults (Wagner, Austin, Davis, Hindmarsh, & Schaefer, 2001). Chronic illnesses account for nearly three-quarters of the Unites States' health care costs, with diabetes at the forefront, accounting for an estimated $174 billion in medical care costs annually (Gabbay, Bailit, Mauger, Wagner, & Siminerio, 2011). Further, exacerbation of chronic illness or deterioration secondary to chronic illnesses accounts for 90% of inpatient hospital expenses (Merrill & Elixhauser, 2005).
Those chronic illnesses presenting in middle-aged to older adults most commonly are hypertension, coronary artery disease, and diabetes mellitus. Chronic illnesses, most notably diabetes, are often marked by long periods of home management, interspersed with multiple hospitalizations either related to exacerbation of the disease or hospitalizations for other reasons that are complicated by the illness.
According to the Centers for Disease Control (CDC, 2011)

and the Agency for
Healthcare Research and Quality (AHRQ, 2000) diabetes is one of the leading causes of chronic illness in the general population, and medical expenditures are about 2.3 times higher for those with diabetes than those without. Currently, 7.8% of the US population has diabetes, and in 2007 alone, 1.6 million new cases were reported. The mortality rate for people with diabetes is twice as high (in any given age group) than for those without diabetes, and if current trends continue, 1 in 3 Americans will develop diabetes (CDC, 2011). Much of the expenditure associated with diabetes can be linked to inconsistencies in care, including ineffective management of glycosylated hemoglobin levels, blood pressure control and maintenance of cholesterol levels (Gabbay et al., 2011;Saydah, Fradkin, & Cowie, 2004). Patients diagnosed with chronic illnesses, especially those who are older or who have co-morbid conditions, are usually managed at home with assistance. These persons are the most likely to depend substantially on their family for assistance, especially at home (Institute for Family Centered Care, 2008).
Management of patients with chronic illnesses, such as diabetes, is rarely limited to only the individual with the disease, especially in aging adults who often have co-morbid conditions. In fact, it is estimated that family caregivers provide 75-80 percent of long-term care in the community (Levine, Halper, Peist, & Gould, 2010).
Family caregivers of the older adult with chronic illness have a significant impact on the day-to-day life of their loved ones. Assistance may include supervision of activities of daily living, meal preparation, financial management, medication management, as well as skills specific to each disease, such as blood sugar management for patients with diabetes. Over time, family caregivers develop competence in their care of chronic illness and in particular in understanding the specific constellation of symptoms and management strategies unique to the individual.
Individuals who most frequently require hospitalization include the very young, the very old and those discussed here, patients with chronic illness. Thus, throughout the trajectory of most chronic illnesses patients suffer exacerbations and must be hospitalized, an event that requires intricate care-planning and should include the family (Bauer, Fitzgerald, Haesler, & Manfrin, 2009). Upon admission to the hospital, however, the usual course of events is that the health care team takes over management of the patient's care with very little input from family, a process that neglects the family's expertise and knowledge of the patient and may inadvertently imply that the that the family is not competent in their care. Then, when the patient is discharged, family is expected to resume care with little or no ongoing preparation.
Several authors, however, have found that families who are involved during hospitalizations display increased satisfaction within the entire family system, while patients have demonstrated discernible improvement in their condition (Chesla, 1996;Rutledge et al., 2000a;Rutledge et al., 2000b). Furthermore, family members who are encouraged to participate in care-giving during hospitalizations report feeling less anxious and more importantly less out of control (Wright & Leahey, 2005). There is increasing evidence that families are able to improve both patient outcomes and patient satisfaction when given appropriate opportunities to be directly involved during inpatient stays (Bauer et al., 2009;Grimmer, Moss, & Falco, 2004). Family participation during exacerbations and subsequent hospitalizations may not only increase client and family satisfaction during admissions but may also decrease length of stay and therefore cost to the health care system (Pearson Hodges, 2009;Rosenbloom-Brunton et al., 2010) There is a critical need to provide effective health care for the chronically ill, while at the same time offering valuable partnerships with the family. Currently, however, there is a dearth of literature that examines the ways in which nurses might enhance family participation in the care of the hospitalized older adult. Further, while the importance of the nurse-family relationship has been explored in a variety of populations, there are no descriptions in the literature specific to the older adult hospitalized with diabetes. Therefore, the purpose of this study was to examine the relationships between the family and the nurse when an older adult with diabetes is hospitalized.

Theoretical Framework
There is some evidence available that suggests that family care and partnerships between nurses and family members during hospitalization may improve clinical outcomes for both the patient and the family, in part due to family member influence over client adherence with therapeutic regimens (Rutledge, et al., 2000a;Rutledge, et al., 2000b). This change in patient outcomes may in turn positively impact nurses' perspectives on the importance of improving family nursing in practice. There is, however, limited theoretical or empirical attention given to the complexity of the relationships between nurse, family and patient.
One theoretical approach that attempts to provide a lens through which to examine these relationships is Family Systems Nursing (FSN). The FSN approach provides for exploration of family strengths and evaluation of interactive family behaviors (Robinson, 1994;Wright & Leahey, 1990). FSN focuses on the whole family as the unit of care, allowing the nurse to simultaneously focus on the patient, the family and their illness. FSN was developed as a way of creating partnerships between families and nurses, and this fundamental approach is intended to have implications which can change the nature of relationships within family systems and between patients, families and nurses (Wright & Leahey, 1990).
This study was undergirded by the Calgary Family Intervention Model (CFIM) (Wright & Leahey, 1994), which is a model developed using the FSN approach. One of the main assumptions of the CFIM is that the family-nurse relationship is characterized by reciprocity, which is the nature of the mutual relationship that develops during interactions between the nurse, the individual and the family. CFIM conceptualizes an intersection between one of three domains of family functioning (cognitive, affective and behavioral) and a specific intervention offered by the nurse (Wright & Leahey, 2005). The cognitive domain of family functioning encompasses beliefs that a family may have about illness, and, if a change is needed within this domain, nurses may work with the family to change perceptions about health problems. Affective domain family functioning concerns intense family emotions that may be hindering a family's ability to problem-solve. In this case interventions that validate responses, encourage discussion, and promote listening are best suited.
Behavioral domain family functioning includes the ways in which family members interact with or behave towards one another, specifically when health problems arise.
When modifications in this domain of family function are indicated, interventions that offer family participation, respite or even rituals are suggested. The CFIM model suggests that interventions developed in collaboration between nurses and families may produce a change in any, or potentially all three, family functioning domains.
Some families, however, may have specific needs in one domain versus another and the nurse may be able to offer solutions that target specific family functioning domains.

Purpose of Research
This study was designed to provide descriptive data regarding the relationships between the family and nurses when an older adult with diabetes is hospitalized. There were four aims of this study. The first was to examine family members' descriptions of family functioning, family health and social support that the family reports receiving from nurses when an older-adult family member with diabetes is hospitalized. The second aim was to explore how nurses on four nursing units caring for hospitalized older adults with diabetes value their family nursing practice and the reciprocal nurse-family relationship. The third aim was to explore if there is an association between the perceptions of nurses working with family members of older adults with diabetes who have been admitted to an acute care facility and how the family describes family function, family health and perceived social support. Finally, this study examined if differences occur in nurses' appraisal of their family nursing practice across units and if so, were they related to families' perceptions of family functioning, family health, and perceived social support.
This study was designed to answer the following research questions: 1. How do families of older-adult, patients with diabetes describe their family functioning, family health and perceived social support from nurses during hospitalization?
2. How do nurses caring for older-adult, patients with diabetes, and their families, appraise their family nursing practice and how do they reflect on the nursefamily relationship?
3. What is the relationship between nurses' perceptions of their family nursing practice and families' perceptions of family function, family health and perceived social support from nurses?
4. Do nurses' appraisal of their family nursing practice differ across units and if so, are these related to families' perceptions of family functioning, family health and perceived social support?
As the incidence of chronic illness increases, it becomes more important to examine the relationship between nurses' family practice and how families evaluate their family functioning, family health and perceived social support when caring for a family member with diabetes. It was proposed that the results of this study would provide a description of the complex relationships between nurses' orientations to family practice and family members' perceptions of family health and functioning, as well as their descriptions of social support provided by nurses. These results will serve 8 as a foundation for the design of targeted family level interventions when older adults are hospitalized with diabetes. While there has been an increase in the nursing literature that addresses family participation, there is little empirical base that describes the effects the nurse-family relationships have on both nurses and family members when an older adult is hospitalized with diabetes. Existing studies that examine both the family and nurse are scarce, especially for the chronically ill adult populations (Fegran & Helseth, 2009). Thus, this study investigated if a relationship existed between how nurses perceived their family nursing practice and how families reported their family functioning, family health and the support they receive from nurses.

LITERATURE REVIEW
Family caregivers of the older adult with chronic illness have a significant influence on the day-to-day life of their loved ones and there is reason to believe that the nature of the relationship between nurses and family members may impact family functioning and older adult outcomes. This study draws on broad bodies of literature that examine chronic illness, the effect of a family member's chronic illness on the family, the nature of family caregiving, and the trajectory of chronic illness through hospitalization and home. This study also builds on previous work that has examined family nursing and its potential impact in the acute care setting.

Chronic Illness in Older Adults
Chronic illness accounts for over 90% of hospitalizations in the United States (Merrill & Elixhauser, 2005). In 2009, the Centers for Disease Control (CDC) reported that while only 12% of the United States population was age 65 and over, this group required 43% of total days of hospital care and comprised 38% of all in-patient discharges (Buie, Owings, DeFrances, & Golosinksiy, 2010). According to Coleman (2003), older adults who require hospitalization are very likely to require multiple post-hospital transfers due to the complex nature of their numerous chronic conditions. These older adults with chronic co-morbidities are also presumably under the care of multiple health care providers for various conditions (Wenger & Young, 2007), which places them at increased risk for potential complications including medication errors, inconsistencies in disease management, and lack of preparation for caregivers (Coleman, Smith, et al., 2004). Coleman, Min, Chomiak and Kramer (2004) further suggest that mismanaged post-hospital care can lead to costly consequences, including re-hospitalizations and even death.

Diabetes Mellitus
In 2010  between those that had heart failure and diabetes as compared to those with only heart failure or diabetes (Bogner, Miller, de Vries, Chhatre, & Jayadevappa, 2010).

Family Caregiving
While family care for adults with chronic illness is proposed to be very important, much of the published work related to family involvement in health care has revolved around the parents of pediatric clients (Rutledge et al., 2000a;Rutledge et al., 2000b). Family care in pediatrics came into focus largely due to the work of Shelton, Jeppson and Johnson (1987), who developed an approach to family-centered care (FCC) for children with chronic illness. This foundational work in FCC suggested that because families are ultimately the primary caregivers of the child, it is crucial that they are supported and allowed to participate in their child's health care (Shelton, Jeppson, & Johnson, 1987 Shelton, Jeppson and Johnson (1987), once these elements become valued by the health care team, central FCC strategies can then be implemented, which includes sharing of all information and establishing institutional policies to better support family participation in patient care. The authors also suggest that gaining understanding of each family's strengths and resources as well as their coping strategies is essential to creating an environment of respect. In light of this work, there has been widespread adoption of many of the recommended practices in family-centered care in acute care facilities including implementation of extended visiting hours and parent rooming-in.
The data regarding family nursing in a pediatric population suggests how important family involvement is during a child's hospitalization. For instance, the early work of Cleary et al. (1986) , demonstrated that children who have parents involved in their care during hospitalization cry less, are alone less when awake, had nearly 90% of their adult contact with family members and had greater social interactions than children whose parents were not present. Taylor and O'Connor (1989) reviewed 586 admissions to the National Children's Hospital in Washington D.C. over an eight month period and discovered that children who were admitted and accompanied by a "resident" adult had 31% shorter inpatient stays than those without a parent who resided with them. Benefits of family involvement include reduced anxiety and stress to children and parents, as well as decreased lengths of stay and less re-hospitalizations (Powers & Rubenstein, 1999).
Investigators of pediatric clients with chronic illness have recently begun to examine the experiences and possible effects that families have when their child requires hospitalization (Board & Ryan-Wenger, 2002;Mitchell et al., 2007;Sloper, 2000). In a study of distress in parents of children with cancer, Sloper (2000) collected data from parents of children with cancer at six months (time one) and eighteen months (time two) post-diagnosis. The researcher found that higher levels of distress were associated with lower levels of family cohesion, or perceptions of strong family relationships. Sloper also found that measures of family cohesion at time one were predictive of parental distress at time two. Similarly, Board and Ryan-Wenger (2002) examined long-term effect of pediatric intensive care hospitalization on families with young children. These authors found that parents of either ill or hospitalized children reported high stress symptoms and that they perceived their family as dysfunctional even after discharge from the hospital (Board & Ryan-Wenger, 2002). In a study of parents' perspectives and health care utilization in children with sickle cell disease, Mitchell et al. (2007) discovered that positive patient coping was related to positive family functioning and lower health care utilization.

Family Care Giving and Older Adults
Care giving for chronically ill older adults, with family members as the primary providers, has become increasingly common. According to Bass and Noelker (1987) eight out of ten functionally disabled older adults choose to live in a community setting and for many of these individuals, family members provide care. A family member, or members, may move into the role of caregiver due to a number of scenarios, including hospital discharge after an acute illness, transitioning back home after an elective surgical procedure, or assistance needed because of the impacts of chronic illness (Bass & Noelker, 1987). A survey of 1,480 family caregivers, conducted for the National Alliance for Caregiving (NAC) and AARP (Greenwald, Naiditch, & Weber-Raley, 2009), reported that 44% of caregivers provide care for a person who is over 75 years of age, and that the average age of adult care recipients in the community has risen from 66.5 years of age in 2004 to 69.3 in 2009, with over 51% of all care recipients being over age 75. These statistics will be compounded by the predicted 17% increase in the 60 and over age group in the US by 2030 (Kreidler, Campbell, Lanik, Gray, & Conrad, 1994).
Interestingly, while the data suggest that family caregiving for older adults is becoming more common, there is a limited amount of published research that has focused on families of older adults managing chronic illness. In fact, Naylor and Keating (2008) proposed that many of the studies that purport to have examined family care only collect data from the adult patient who is the receiver of care.
Consequently, families have not been extensively studied and there is limited evidence available regarding what may improve caregiver outcomes like burden, stress or depression (Mattila, Leino, Paavilainen, & Åstedt-Kurki, 2009;Robinson, 1994). Fisher (2006) described the "notable absence of studies that address adults with chronic disease and their families" (p. 375). Gavaghan and Carroll (2002) suggest that even though there is literature available that has addressed family member needs, nursing studies that might offer solutions are only beginning to emerge. There are deficits in our knowledge base regarding family systems impact on diseases such as diabetes, asthma/chronic obstructive pulmonary disease, obesity, and cardiovascular risk as they relate to adults (L. Fisher, 2006;Mattila et al., 2009). Again in this population, there are few studies that have examined the relationship between caregiving and outcomes for patients or families. (L. Fisher, 2006;L. Fisher & Weihs, 2000;Mattila et al., 2009;Wright & Leahey, 2005).
There is also limited knowledge of the processes through which family members choose to become caregivers to an older adult or are able to sustain a caregiver relationship. In a 2007 study, Piercy interviewed intergenerational family caregivers in order to identify the characteristics that are associated with a strong commitment to caring for older adults at home. Interview data were analyzed from two qualitative studies asking similar questions directed at adult children, children-inlaw, grandchildren or nieces who were providing care for an adult family member age 65 years or older over a period of at least three months. Those providers with a strong commitment to caregiving (defined as providing hands-on care for at least 6 months) were found to share some commonalities, the foremost being that all primary caregivers with strong commitments were women. According to Piercy, those caregivers with strong commitments also "offered compassionate care and went to considerable lengths to preserve the home care arrangement, or intended to do so if it was threatened" (2007, p. S383). Piercy identified four common themes expressed by those exhibiting a strong commitment to care, including a moral or religious basis for providing care, embracing/internalizing the caregiver identity, affection for the care recipient and the ability to provide compassionate care. Strongly committed caregivers also used common strategies to sustain their commitment, including accepting the situation and making adjustments, seeking support from other family members, making use of formal care services to supplement care or provide respite and conceptualizing caregiving as a growth experience. In contrast, those members of the sample with weaker commitments to care described ambivalent feelings and were unable to see caregiving as a growth experience that provided purpose or enhanced self-esteem. Those with weaker commitments also expressed concerns regarding the lifestyle restrictions imposed by caregiving and described strained relations within the family. The impact of caregiving on family functioning and family relationships described in this study are similar to that described in the literature regarding parent caregiving for chronically ill children. However, this study does not contain a specific assessment of family health or functioning that can be used in adult patients. Gallant, Spitze and Prohaska (2007) used focus group methodology to explore the positive and negative influences that family and friends had on the management of chronic illness. Participants who were enrolled were placed into 13 focus groups with a professional moderator who used a discussion guide. Groups were audio taped and analyzed independently by the study's co-investigators (Gallant, Spitze, & Prohaska, 2007). Results demonstrated that family and friends had positive influences related to dietary activities, physical activity and health care appointments. Family members, more than friends, were more likely to offer support for activities specifically related to disease-management, such as dietary activities, medication management and relationships with primary care providers (Gallant et al., 2007). In the case of older adults with diabetes, family members are often intimately involved in the day to day management activities described above, as well as recognition of symptoms and strategies for treatment of hyperglycemia or hypoglycemia, yet there are no reported studies that describe how the nature of the family's participation changes when an older adult is hospitalized.
Another study by Prohaska and Glasser (1996) (Prohaska & Glasser, 1996). The researchers conducted three interviews with older adults attending physician office visits. Initial interviews occurred with older patients in the waiting room prior to an office visit, then a second face-to-face interview one week after the office visit, and a third interview via phone discussion 10-12 weeks after the office visit. Most participants reported that having a family member, or friend, present during the appointment was an advantage. In fact, a majority reported that the companion assisted with physical help, aided in communication between physician and patient, assisted in patient understanding of treatments, and provided emotional support (Prohaska & Glasser, 1996). This study provides empirical evidence for the importance of having support from family and friends during interactions with the health care system. However, it does not examine the efforts of patients and families who cope with specific disease processes, such as diabetes, which require substantial care management, nor does it examine the nature of family participation during hospitalization, presumably when the older adult is more vulnerable due to acute illness.  Fisher, 2005). Armour, et. al (2005) point out that there continues to be a need for more well designed studies that are longitudinal and include education and involvement of family members and patients with diabetes. Interestingly these studies did not explore the impact of family caregivers other than spouses or partners for those with diabetes, even though the data suggest that adult children are frequently involved in the care of older adults with diabetes.

Transitions in Caregiving
The most challenging periods of caregiving in chronic illness care occur as patients move across care settings, either from the home into the acute care setting or from an inpatient setting back home. The literature regarding family needs during hospitalizations of adults has focused mainly on family member's psychological coping ability during critical care admissions to specialty care units. One example is a study by Auerbach et al. (2005) who investigated the needs of family members during an intensive care admission. The main focus of this investigation was to determine if family members believed that their needs were being met and how this related to emotional stress and psychopathological dysfunction in the family (Auerbach et al., 2005). The data demonstrated that family members had increased levels of stress during patient admission to an ICU setting and that family members expressed concern with unmet needs, including a lack of clear information regarding the patient's treatment plan and medical equipment being used.
Similarly, Kosco and Warren (2000) examined differences between nurses and family members perceptions of family needs during intensive care hospitalizations.
There were significant differences found between nurses and family member's perceptions in several areas. Families found it more important than nurses to have a specific staff member who could be called for updates. They also reported wanting to have reassurance that there was someone caring for the well-being of their relative if they could not visit, while nurses did not find this as important. Additionally, families described that their need to visit at any time was not being met, yet nurses believed they were meeting this need (Kosco & Warren, 2000). Eggenberger and Nelms (2007) also explored the family experience during a critical illness of a relative using a hermeneutic phenomenological design, which included interviews with family members. The findings indicated that family members suffered along with their critically ill relative and many reported a sense of vulnerability and distress due to troubled nurse-family relationships (Eggenberger & Nelms, 2007 & Tarkka, 2006;Tarkka, Paavilainen, Lehti, & Astedt-Kurki, 2003).
In a study designed to describe the perceptions that one has about their family's health, the family of patients with heart disease were asked to explore five components, including knowledge, ill-being, activity, well-being and values. For the purposes of this study, knowledge was defined as knowing when one is healthy in comparison to others, while ill-being referred to feelings of discomfort or pain. The component of activity was described as interests that promote health, such as exercise.
Well-being referred to effortless coping, freedom from pain and a carefree existence and values included items such as a sense of freedom, security, aesthetics and relationship to the natural environment. The researchers reported that overall, study participants reported good family health, and the best predictors of family health were family stability, effects of symptoms on daily life and family relationships (Astedt-Kurki et al., 2004). Paavilainen, et. al. (2006) further investigated how family members of cardiac patients described their perceptions of family functioning including family relationships, family stability, family resources and friendships. This study demonstrated that families of heart patients described high levels of family function, which was positively correlated with greater instrumental support from nurses.
In a phenomenological study, Tanner, Benner, Chesla and Gordon (1993) explored the unique expertise that family can offer during hospitalization. This study aimed to describe how nurses acquire the practice of "knowing the patient." Tanner and colleagues discussed their findings within two broad categories of how 130 intensive care unit nurses describe knowing their patient(s). The first type of "knowing the patient" is to appreciate the patient's patterns of responses, or to be able to recognize their patient(s) routine(s), coping mechanisms, physical ability and characteristics, and how they may respond to therapeutic treatments. The other way that nurses know their patient is to know the patient as a person, or on a personal level, which allows for advocacy and an immediate understanding of what is happening with the patient. Families can offer health care professionals the insightful ability to "know the patient" much earlier and may even be able to assist nurses to 'know' someone who is nonverbal or comatose (Tanner, Benner, Chesla, & Gordon, 1993). If allowed to, families can aide nurses in developing an understanding of the patient's everyday life, typical behaviors, and familiar patterns. This knowledge permits nurses to become aware of what may seem like insignificant warning signals of mounting distress (Tanner et al., 1993). This is especially important in patients who have diabetes because family members learn to become experts in this individual's experience of the disease. Family caregivers become skilled at recognizing symptoms of complications as wells as patient's responses to medications. However, often times during hospitalizations families feel that this expertise is overlooked or ignored.
During transitions in care, such as from the acute care environment back home, families may once again develop heightened levels of stress. Lough (1996) explored this hospital-to-home transition process in older adults diagnosed with congestive heart failure using a grounded theory approach. Semi-structured interviews were conducted with participants two weeks following discharge from an acute care facility.
The core variable identified in this study was conceptualized as 'a tentative situation' and three central processes illustrated it. Older adults in this study reported that central to their tentative situation were the ups and downs with managing a chronic illness.
The negative aspects of managing the disease, or downs, were related to uncertainty about new medication regimes as well as the daily impact of the disease on diet and prescribed activity levels. The positives, or ups of disease management, were focused on how the social support they received from family and friends allowed older adults to maintain their independence. When patients discussed the concept of caregiver issues, they reported their frustrations with having to be dependent on another for ADL and/or IADL assistance. Participants also expressed the stress they experienced if they, themselves, were also caregivers as this increased their anxiety over who would care for their loved one, as well as the participant at home. Many study participants reported quality of life challenges such as feelings of hopelessness, depression, or self-blame if the illness had worsened (Lough, 1996). Although this study examined transitional inconsistencies at the individual patient level, it is also relevant when working with families caring for older adults with diabetes, as they are equally prone to the negative impacts of chronic illness management.
At times family caregivers in the home work in conjunction with more formal caregivers. While intended to be supportive, this can also lead to frustration. Thus, Sims-Gould and Martin-Matthews (2010) investigated the experiences that caregivers of older family members had with in-home formal support services using in-depth semi-structured interviews with family caregivers. The authors utilized a conceptual model that they developed in a previous study to guide analysis and theme organization after interviews were read, re-read and coded independently by three While family members describe the benefits and satisfaction of providing care for a loved one, there is also an associated cost. One consequence that has received a great deal of attention in the literature is caregiver burden. The strain of caregiving has been associated with physical and emotional exhaustion, as well as stress and suffering (Andren & Elmstahl, 2008;Faison, Faria, & Frank, 1999;Glasdam, Timm, & Vittrup, 2010;Greenberger & Litwin, 2003;Kim & Schulz, 2008;Walsh, Estrada, & Hogan, 2004). Caregiver burden research has focused mainly on the individual caregiver.
However, because caring for older adults with chronic illness may involve the whole family unit, burden also has the potential to impact the entire family unit. Therefore it is essential to examine the effects that caregiving has on the family system and how nurses may offer better support.
Family caregivers in the home provide many types of assistance for patients, which range from relatively straightforward to highly complex skills. In a national survey of 1,002 informal caregivers, Donelan et al. (2002) found that 23% of respondents provided some type of health care assistance, such as ADLs, IADLs or more complex medical tasks. Fifty-four percent of caregivers who assisted with ADLs, such as feeding, bathing, toileting or lifting reported that they had received no formal training on how to safely perform these tasks. Fifty-four percent of caregivers also reported that their family member had been hospitalized during the year leading up to this survey and 74% stated that their family member had a chronic illness. This survey also described that 43% of caregivers who were surveyed were responsible for medical tasks, such as wound care, intravenous infusion pumps, home dialysis machinery, and medication administration. Most disquieting was the discovery that 37% of caregivers were responsible for administering more than five medications per day and 12% administered ten different medications per day. As the number of medications being administered by caregivers went up, so too did the number or caregivers reporting errors in administration. Many caregivers reported they received no instruction on how to properly administer medications (18%) or how to change dressings or use medical equipment (one-third) (Donelan et al., 2002). This study provides clear details of the complexities involved in family caregiving of chronically ill patients. The literature, however, provides relatively few descriptions of the specific nature of family caregiving for older adults with diabetes and even less about how this changes when older adult is hospitalized.
As one of the leading chronic illnesses with great potential for complications as well as ADL or IADL impairments, diabetes has a great potential for leading to high levels of caregiver burden. This was corroborated by a national US study that computed the amount of hours of weekly caregiving that caregivers undertake when caring for older adults with diabetes (Langa et al., 2002). This study examined the amount of time and cost associated in the informal care of 7,438 older adults and found that caregivers of family members who have diabetes spend greater amounts of time providing care than caregivers of family members without diabetes. The amount of weekly hours in caregiving was mainly associated with medication administration, with patients who require insulin administration averaging the highest weekly hours at 14.4 hours (p < 0.01) (Langa et al., 2002).
The amount of weekly hours in relation to the number and type of activities, especially medication administration, needed by care recipients imposes a significant burden on caregivers. One study highlights what types of caregiving, as well as characteristics of the caregiver, are associated with higher levels of burden (Faison et al., 1999). In a descriptive survey study of 88 family caregivers of chronically ill patients, Faison et al. (1999) report a significant correlation between increased ADL assistance needed by care recipients and caregiver burden. Those direct care activities with the highest correlations to caregiver burden were bathing (r = .215, p<.05), transfer (r = .255, p<.05) and continence care (r = .269, p<.05); and indirect care activities relating to high caregiver burden were meal preparation (r = .325, p<.01), medication assistance (r = .237, p<.05) and housework (r = .294, p<.05) (Faison et al., 1999). Andren and Elmstahl (2008) studied the relationships between caregiver burden, perceived health and sense of coherence in 130 family caregivers of people with dementia. In this study, 57% of respondents reported moderate levels of burnout, and the highest levels were associated with being a close relative of the person receiving care. Family caregivers were asked yes or no questions in relation to symptoms using the Nottingham Health Profile Scale, which measures energy, emotional reactions, social isolation, sleep, pain and physical mobility. The authors described strong correlations between high levels of burden, such as strain, isolation, disappointment and emotional involvement with perceived health and sense of coherence. Caregivers who reported higher levels of burden also reported a lower sense of overall health as well as a lower sense of coherence. Andren and Elmstahl (2008) found that a close relationship to the patient was correlated with higher burden for the caregiver. This study also highlighted the relationship between healthier caregiver coping strategies, such as considering problems as a challenge rather than a misfortune, with better-perceived health. This has great implications for how nurses can assist caregivers in positively reframing experiences through more effective coping strategies.
There are resources that nurses can use to better support families in attaining improved family functioning and overall family health. For instance, if families receive proper instruction, they may be better prepared to prevent future exacerbations that are so often associated with chronic illness. Levine, Halper, Peist and Gold (2010) point out that gaps in continuity of care, including lack of instruction, can occur because of poorly planned out transitions in care with family members. A breakdown in communication between acute care providers and home caregivers has been linked to high re-hospitalizations, as well as low satisfaction rates, and most concerning, adverse effects (Naylor & Keating, 2008 …intentional human interaction that involves one or more of the following elements: affect, which refers to appreciation, admiration respect or love, as well as creating a sense of security; affirmation which includes reinforcement, feedback, and influencing the individual's way of making decisions and finally concrete aid, such as objects or money, and spending time in helping someone (Tarkka et al., 2003, p.737) .
The concept of social support has been widely explored in the literature, and the definition used by Tarkka, et. al is consistent with House's (1944) foundational definition. House described four broad domains of support, including emotional support, instrumental support, informational support and appraisal support. The first category of social support is emotional support, which is providing empathy, love, caring, etc. Instrumental support constitutes those behaviors that directly help the person in need, such as paying bills or grocery shopping. Informational support includes those activities that provide a person with information that the person can use to better cope. The last domain is appraisal support, which is providing support that is significant to self-evaluation (House, 1944). Tarkka et al. (2003) found that families frequently describe receiving emotional support from nurses, however generally they were not satisfied with the amount of support they received. There were several predictors of increased satisfaction with social support including family structure, of trust was important in allowing for expressions of concerns and reflection of their situation. Caregivers also described the importance of feeling as though they were participating in decision-making and that that their contribution mattered (Levesque et al., 2010). Contrary to this, Grimmer, Moss and Falco (2004) reported that many caregivers described being unprepared and imposed upon for their role as caregiver of an elderly person in their qualitative study utilizing semi-structured interviews with family caregivers. The researchers in this study analyzed data from repeated interviews, over the course of six months, with twenty-four family caregivers of an elderly person who had been discharged home after an acute hospital admission (Grimmer et al., 2004). The researchers noted that many participants believed they had no choice in becoming caregivers and were given little information to help them in caring for the patient after discharge. Overall caregivers described that receiving more education prior to discharge regarding what to expect, how to manage medications, how to performs ADL's and where to seek assistance would have been beneficial (Grimmer et al., 2004). Often patients with diabetes are discharged home with changes to medication regimes, alterations in dietary needs and greater ADL assistance.
However, family caregivers often receive very limited discharge instructions.
Some studies have attempted to evaluate the causes of the disparity between what families feel they need to know and what information or training nurses feel families should receive. In a descriptive qualitative study by Yen et al. (2010), patients' and health care professionals' were asked to describe their perspectives related to problems and solutions that are associated with chronic illness. The researchers conducted ten focus groups and seven interviews (n=88) to explore how health care professionals viewed the three main themes that were revealed in a previous qualitative study of 54 patients and 14 caregivers related to chronic illness concerns. The main concerns discovered were related to economic hardship, managing co-morbid conditions and managing multiple competing demands (Yen et al., 2010).
Two main themes were identified, patient compliance and service fragmentation.
Patient compliance was described as the perspective that patients frequently act in ways that are not in accordance with what their health care providers see as normal or optimal. The theme of service fragmentation relates to the gaps in communication and philosophy that so often exists between, and within, health care agencies. Health care professionals most often used the term 'compliance' as being the actual issue for patients who described problems in managing co-morbid conditions, daily lives and/or economic burdens related to chronic illness. Patients stressed their feelings regarding fragmentation of services, health care system inefficiencies and the inability of individual agencies to collaborate and work together. In contrast, health care professionals faulted colleagues from other disciplines rather than addressing what their discipline could do differently (Yen et al., 2010). This study illustrates the inconsistencies in how families and nurses interact, and further demonstrates the importance of exploring the best ways to enhance nurse-family relationships.
Nurses have many different views about the best way they can work with families. Robinson (1994) has identified three main orientations impacting the way nurses collaborate with families, which she labeled as traditional, translational and non-traditional. Robinson proposed that these three orientations or approaches shape the way nurses practice family nursing. The traditional family orientation views the family as context rather than a unit of care (Robinson, 1994). The primary focus for health care providers who hold a traditional orientation in family intervention is the patient with the chronic condition. A nurse who holds a traditional orientation believes that family influences their loved one's experiences, therefore family treatments may only focus on the family as it assists them in caring for the client. Professional caregivers who hold traditional views of chronic illness care will often only seek out the family member most likely to give the desired response. If a family member resists a recommended intervention, then there must be something "wrong" with the family, and more offering of family education is often the recommended intervention (Robinson, 1994). Robinson (1994) posited that nurses with a transitional orientation concentrate on the family as a group, made up of multiple individuals, not as a system with interacting parts. There is a belief that there is a "right" way for families to experience and interact during chronic illness and that there is a "correct" approach. Transitional methods tend to be those that are more standardized and do not take into account the reciprocity which occurs in families (Robinson, 1994). Nurses who have a transitional orientation to family care may choose to intervene in ways that help family members individually cope with a chronic illness of a loved one but may not evaluate how the family unit impacts the progression and treatment of the illness. The practice of FCC is consistent with the transitional approach to family nursing, in which standard practices such as pre-scheduled family meetings are the main focus for all patients admitted to an institution.
Non-traditional orientations to family nursing have been identified as supporting a family's ability to change (Robinson, 1994). This orientation focuses on the family system as the client, as opposed to the more linear views of family, seen in the traditional and the transitional orientations. Nurses who intervene using a nontraditional approach to family care realize that there may not be one "correct" or "right" intervention. The main concentration in a non-traditional approach is on family interactions and relationships and the reciprocity between family functioning and chronic illness. Robinson (1994) proposes that families who are approached through a non-traditional family nursing orientation are able to reframe situations in a positive way, which enables better coping and problem-solving skills. As chronic illnesses of older adults become more and more of a family matter, nurses must become more adept at both holistic family assessments and the most beneficial family interventions for a particular family. Wright and Leahey (2005)  communication and better rapport with nurses, which seemed to improve overall family health and function (Legrow & Rossen, 2005).

Summary
With chronic illness in the older adult population, notably diabetes mellitus, accounting for such large percentages of heath care expenditures, the development of nursing knowledge related to family care needs is necessary for improvement of future health care delivery. The literature presented describes the increasing evidence that involvement of family during transitions in care for chronically ill family members may not only increase client and family satisfaction during admissions, it may also decrease length of stay and therefore cost to the health care system. Family caregivers, however, are faced with a multitude of tasks related to care for their chronically ill family member and may feel inadequately prepared for their role.
Although there has been an increase in family nursing research over the past several decades, there are few studies that explore family involvement in adult patients with a chronic illness. There are large gaps in the literature that examine how caring for a chronically ill older adult with diabetes impacts families, especially when the older adult is hospitalized or discharged back home. There are also few studies that have explored how nurses may better support families, and none that specifically investigate families of older adults with diabetes. Therefore there is a need to further investigate if a relationship exists between how nurses perceive their family nursing practice and how families caring for older adults with diabetes view their family functioning, family health and the social support they receive from nurses.

THEORETICAL FRAMEWORK
This study was framed by the Calgary Family Intervention Model, which is a model rooted in Family Systems Nursing. The Family Systems Nursing approach focuses on the whole family as the unit of care, allowing the nurse to simultaneously focus on the patient, the family and the patient's illness. Leahey (1990, 1994) point out that this is in contrast to the concept of family nursing, which focuses either just on the family or just on the individual patient. One example of family nursing is family centered care (FCC), which focuses predominantly on techniques such as nurses providing patient information for family and instituting better family visitation policies (White et al., 2002). FCC has been used primarily in pediatrics as a means of improving parent satisfaction, and although the ideal of FCC is promoted by many nurses, some authors are now reporting that it may not be as effective as was first thought (Berman, 1991;Shields, 2010).
Distinct from FCC is Family Systems Nursing, which establishes a partnership of mutual trust, regular communication and relationship building among nurses, patients and families, which equates to improved healthcare outcomes for patients (Leahey, Harper-Jaques, Stout, & Levac, 1995). By permitting such a multi-faceted interface, there is an opportunity for authentic and meaningful relationships to occur between the nurse and the family. This approach to family nursing is useful in a variety of client populations, including families who are managing the care of patients who have diabetes.
Family Systems Nursing (FSN) is first mentioned in the literature in 1990, although it was developed and used clinical slightly earlier (Wright & Leahey, 1990).

The creation of FSN resulted out of what Wright and Leahey believed was a need to
focus on the whole family as the unit of care. The foundational work by Wright and Leahey (Wright & Leahey, 1984) was truly the basis for the Calgary Family Assessment Model, while the Calgary Family Intervention Model was not added until 1994 (Wright & Leahey, 1994 The fundamental concept in Family Systems Nursing is the interaction between the family members, and the belief that the questions which best assess this interaction will focus on the relationships amongst all individuals that the patient considers significant. Family systems nursing was developed as a way of creating more structured partnerships between families and nurses, and the outcomes were intended to have implications for nursing practice, research and education (Wright & Leahey, 1990). They propose that the relationship between the health care professional and the family greatly affects the outcome of family nursing practices (Wright & Leahey, 1994). This is why examining the potential relationship between how families describe nursing support and how nurses describe their family nursing practice is so important.
The work of Family Systems Nursing is accomplished using two practice models, the Calgary Family Assessment Model CFAM and the Calgary Family Intervention Model (CFIM). A nurse who practices family systems nursing may incorporate one, or both, the CFAM and the CFIM. Many family systems nurses incorporate a family assessment, which may utilize CFAM, in which the bio-psychosocial structure of the family is evaluated, allowing for application of individualized family nursing practice as suggested by the CFIM (Svavardottir, 2008). The CFAM can be used as an organizing framework or as a template to guide nurses when working with families. Both of these practice models are especially useful in family research where the objective is to investigate both family dynamics and specific nursing practices that enhance family functioning. Calgary Family Intervention Model (CFIM) gives nurses, as well as other health care providers, the tools to generate change for a family that is managing the tribulations and exacerbations that can occur during the course of a chronic illness such as diabetes. Wright and Leahey (2005) assert that in order to accurately apply the underlying principles and meanings of the CFAM and CFIM models, it is important to understand their influences. Family systems nursing, as well as the CFAM and CFIM frameworks, have been developed from multiple philosophies and standpoints, such as: postmodernism, systems theory, cybernetics, communication theory, change theory, and biology of cognition. According to Wright and Leahey, CFAM and CFIM have been greatly influenced by a worldview know as the biology of cognition (Maturana & Varela, 1992). The biology of cognition suggests that there are two ways of viewing the world, objectivity and objectivity-in-parentheses (Wright & Leahey, 2005). In other words, we never really leave our own domain of activity, there is no independent object of study, thus there is nothing subjective either (Efran & Lukens, 1985). Maturana (1988) furthers this by stating that reality is not constructed, nor is it independent of individuals, it is instead drawn forth. Therefore, there are multiple views of reality, or a multiverse, which we bring forth through our interactions with the world and others through language.
The Calgary family intervention model has also been influenced by frameworks such as postmodernism, which inspired Wright and Leahey to infuse their practice models with multiple ways of understanding and experiencing the world.
These values are especially significant when working with families of patients with diabetes, as in this study, who experience and understand chronic illness in various manners (Wright & Leahey, 2005). Wright and Leahey (2005) believe that if a connection can be made between a particular area of family functioning and activities offered by the nurse, there will be improvements in the family's health as well as enhancements in the nurse-family relationship. There are three domains of family functioning (a) the cognitive; (b) the affective; (c) the behavioral. The CFIM attempts to promote, improve and/or sustain change in any one of the domains, or all three simultaneously (Wright & Leahey, 1994. After a family assessment has been conducted, the nurse (or other health care professional) can then come to a decision regarding the best approach for that family. Nurses may work with families to invoke change in any of the three domains.
However, modifications in the cognitive domain are believed to be the most compelling, as they impact family beliefs. Wright and Leahey (2005) additionally state that the family must be open to the ideas that are being offered. One potential lens for initially evaluating family systems nursing and the fit of CFIM is to describe if a relationship exists between how families perceive their family function and the support received from nurses with the perceptions of nurses regarding their family nursing practice.
A significant factor that determines whether or not change will occur is allowing the family to feel as though they have contributed to the type of support that is most needed. Wright and Leahey (1994) state that while health professionals can suggest what methods may best help the family, family members must be willing participants, which may vary according to their genetic make-up and history of family interactions. The ability for the nurse to incite change is also profoundly influenced by the relationship between the nurse and the family and the nurse's ability to invite the family to contemplate the family unit's health problems (Wright & Leahey, 1994).
This is why a study which aimed to examine associations between families' perceptions of social support received by nurses and nurses' perceptions of their practice and the nurse-family relationship is so important.
In their description of the CFIM, Wright and Leahey recommend the use of various techniques for nurses to carry out with families, many of which target a particular domain of family functioning. For example, if a family is having trouble coping with a health problem due to their perceptions, then the cognitive domain may be where the nurse will want to work at improving family function. The first technique suggested is to commend family and individual strengths, by observing patterns of behavior. There are a few considerations when offering commendations (a) compliments should not just be simple one-time occurrences; (b) they should be warranted; (c) they should occur within the first ten minutes of meeting with a family; (d) they should become part of a practice with each family. Also the offering of information and opinions is important and places strong emphasis on giving patient information. Wright and Leahey (2005) recommend the use of easily readable literature, community support and encouragement of the family's continued pursuit of information. The avoidance of judgment if a family does not make use of the information is also relevant in family nursing practice.
The aim of increasing or reducing intense emotions, which may be interfering with family functioning, is part of the affective domain. In order for a family to resolve issues within this domain, it is crucial to decrease feelings of isolation, therefore validating or normalizing emotional responses is advised. Encouraging the use of "storying" (Wright & Leahey, 1994) or the telling of illness narratives (Wright & Leahey, 2005) is also suggested, with an emphasis on narrating the experience of the illness, not just the "medical story." Additionally, nurses can be a catalyst for enhancing therapeutic communication among family members by drawing forth family support and allowing all family members to feel heard. This is quite the opposite of what often takes place when an older adult family member with chronic illness is hospitalized and the family's input, feelings and knowledge are discarded.
The behavioral domain requires that the nurse facilitate alternative interactions among family members by encouraging the use of specific behavioral tasks. The first behavior suggested is to encourage family members to be caregivers and offering caregiver support. What this task really encourages is a family member's participation in care giving, while at the same time offering needed information, advocacy and support. Associated with this is the act of encouraging respite, so as to avoid burden and burnout. Devising rituals, especially when daily routines have become disrupted due to illness can have a beneficial outcome on a family's behavioral interactions (Wright & Leahey, 2005) Wright and Leahey (2005) also describe a simple, yet highly influential skill of questioning as a key technique offered by the CFIM. There are two basic types of questioning for nurses to use with families: linear and circular. Linear questions are directed at eliciting information from families, which may be significant to health care providers. However they may not provide information about the family's comprehension of a problem. Circular questions alternatively explore explanations of problems in an effort to change a family's behavior regarding either illness or a family difficulty (Simpson, Yeung, Kwan, & Wah, 2006;Wright & Leahey, 1994. functioning (Wright & Leahey, 2005).

Theoretical Description
Wright and Leahey (2005)  Family Systems Nursing is "quite distinct from the positivist-based idea of two separate components, either family or nurse" (Wright & Leahey, 2005).
Family systems nurses view the family-nurse relationship as non-hierarchical (Wright & Leahey, 2005). Therefore, contributions to assessment and intervention are co-created by both family and nurses, where each person's involvement is needed and valued. Wright and Leahey (2005) also believe that nurses and families each have specialized expertise in maintaining health and managing health problems. Because families are the ones living with, near and alongside chronic illnesses on a daily basis, nurses must recognize and appreciate the knowledge which families provide.
Similarly, nurses also have expert knowledge based on their education and experience.
Nurses must realize that their expectations of illness management are not necessarily the only way, nor even the best approach for individual families.
A family systems nurse believes that nurses and families each bring strengths and resources to the family-nurse relationship (Wright & Leahey, 2005). Nurses who identify individual resources within each family including those unique to a family's culture, ethnicity, and spiritual or other beliefs are able to assist families in acknowledging their strengths or their weaknesses. Another critical element in providing family systems nursing is to recognize that feedback processes can occur simultaneously at several different relationship levels. Wright and Leahey (2005) state that nurses should invite the family to reflect on their evaluation of the encounter they had with the nurse, while also self-reflecting on how they believed they impacted the family's health and/or coping. Wright and Leahey (2005) additionally discuss guiding principles that a Family System's nurse should follow. Family System's nurses will conceptualize the interaction between an illness and the individual patient as well as having an understanding of the reciprocal influence that the patient (and family) has in maintaining, aggravating or ameliorating the illness. Therefore nurses must concentrate on the interconnections between illness, individual and family. Wright and Leahey (2005) also suggest that nurses should reflect on research that illustrates these interconnections and should assess interaction at all systems levels and across system levels with a focus on intervening at the family system level with the greatest leverage for change (Wright & Leahey, 1990).
There are several essential concepts that are critical in understanding the FSN, CFAM and CFIM. The first is the family system, which is distinct for each family unit and can be defined as 'the family is who they say they are' (Wright & Leahey, 2005).
In family systems nursing, the focus of care is seen as circular, so instead of 'either/or' it is 'both/and' (Wright & Leahey, 1990). Interaction is a second important concept to

Summary
Older adults are particularly susceptible to errors occurring during the numerous transitions in care that they so often experience. A Family Systems Nursing approach, particularly using the CFIM, allows for the inclusion of family in care during transitional periods for older adults. By encouraging such thorough interactions, it has been suggested that there will be improvements in the family-nurse relationships as well as family well-being. Wright's and Leahey's (1984Leahey's ( , 1990Leahey's ( , 1994Leahey's ( , 2005 approach recognizes the family as the individual unit of care, but some of the work in the literature around older adults suggests that while many members of the family are involved in care, they each have individual perceptions. This study used a family systems nursing approach, guided by the CFIM model to examine how the family describes family functioning, family health and perceived social support received from nurses and how this relates to nurses appraisal of their family nursing practice.

METHODOLOGY
While the literature related to family nursing supports the belief that families who have an older adult hospitalized with diabetes may experience an interruption in their family's normal health and function (Chesla, 1996(Chesla, , 2010Mattila et al., 2009;Rutledge et al., 2000a;Rutledge et al., 2000b), there are no comprehensive descriptions of these relationships in the literature. Similarly, while there is theoretical support for the belief that a nursing approach which values family participation may impact family function, family health and perceived social support (Wright & Leahey, 1994), the literature has not yet begun to explore these relationships as they relate specifically to the families of older adults hospitalized with diabetes.

Research Questions
This descriptive study was designed to answer the following research questions: 1

Sample Size
This study assessed two distinct populations on four medical-surgical units in one community hospital. The sample size of 60 was estimated using detectable difference for fixed samples. As Table 1 shows, to achieve 80% power a moderate to slightly larger than moderate effect size for two normally distributed variables (α =.05, two sided test) is determined. In order to detect a nonzero association when using a fixed sample, stronger associations (larger effect sizes or r) are needed to increase power. Table 2 illustrates correlation coefficients and effect sizes which remain in the moderate to slightly greater than moderate range for a sample size of 60 (n=60, r = 0.345, power=80%) (Cohen, 1988(Cohen, , 1992.

Measures
Two instruments were used to assess the variables of interest in this study. The and social support subscale (α = 0.97 and 0.98) demonstrating acceptable to excellent internal consistency reliability. The FAFHES instrument was also evaluated for construct validity, using factor analysis, as well as content validity, using expert reviewers. The FAFHES was administered to family members of patients with diabetes admitted to four medical-surgical units of a community hospital during the data collection period. Family member participants additionally completed a demographic questionnaire.
The Family Nursing Practice Scale (FNPS) (Appendix I) (Simpson & Tarrant, 2006; was administered to all Registered Nurses working on the same medical-surgical units as the family/patient participants at the same community hospital. The FNPS was used to assess two variables: 1) nurses' critical appraisal of their individual family nursing practice and 2) nurses' experiences of interaction and reciprocity in the nurse-family relationship (Simpson & Tarrant, 2006).

Nursing, including the Calgary Family Assessment Model and the Calgary Family
Interventional Model and has demonstrated reliability and validity in a preliminary psychometric analysis (Simpson & Tarrant, 2006). Simpson and Tarrant (2006) conducted a pilot study of the FNPS evaluating stability using test-retest reliability, as well as internal consistency which revealed acceptable internal consistency for each of the subscales (practice appraisal subscale: α =0.85 and nurse-family relationship subscale: α = 0.73). The instrument was also evaluated for face and content validity utilizing expert content review, and factor analysis was conducted to establish construct validity. The FNPS also includes three open-ended questions providing additional levels of rich detail in three domains, including: the advantages of involving families in nursing practice, the disadvantages of involving families in nursing practice and how nurses are currently including families in their nursing practice (Simpson & Tarrant, 2006). In addition, all Registered Nurse participants completed a demographic questionnaire.

Ethical Considerations
Institutional Review Board approval from both the hospital and the University Face-to-face discussion was also available if the nurses had any questions or concerns.
Return of the completed FNPS instrument, as well as the nursing demographic form (see Appendices H and I) implied consent. Completion of the surveys took approximately 10-15 minutes to complete the questionnaire.

Data Analysis
Descriptive statistics (means, standard deviations, frequencies, and percentages) were used to summarize demographic characteristics for both the nurse and family samples. Continuous variable distributions were further examined for violations of normality. Additionally, demographics and study variables were examined at the unit level, using Chi-Square analysis to determine if these variables differed by medical-surgical unit. The method of analysis chosen for each research question is shown in Table 3.
Analysis of variance was used to examine differences in the continuous variables (family nursing practice, practice appraisal, nurse-family relationship, family functioning, family health and perceived social support) by medical-surgical unit. Chi Squared tests were used to assess the associations of categorical variables (age, gender, education, nursing years, current unit years, marital status, relation to patient, living arrangements (i.e. with or apart from the identified patient), years with diabetes, and number of times hospitalized) by medical-surgical unit (Polit & Beck, 2008).
Accounting for the impact of the medical-surgical unit is very important since significant unit variations can adversely affect the ability to find statistically significant results in the variables of interest (Chen, 2012).
The relationships between nurses' critical appraisal of their family nursing practice and nurses' experiences of the interaction and reciprocity in the nurse-family relationship, with families' perceptions of family function, family health and perceived social support, were analyzed using linear regression analysis (Munro, 2005). To determine if nurses' critical appraisal of their family nursing practice and nurses' experiences of the interaction and reciprocity in the nurse-family relationship differed across units and if so, how this related to families' perceptions of family functioning, family health, and perceived social support, ANOVA with Bonferroni correction technique was conducted on the nursing data and the family data respectively (Munro, 2005). All analyses were conducted using SPSS Version 21.0 (SPSS IBM, New York, U.S.A). The significance level was set at .05 for all analyses.

Qualitative Data Analysis
The FNPS includes three open-ended questions posed to nurses. The data were analyzed using the approach to manifest content analysis described by Granehiem and Lundman (2004) where all open-ended textual data could be condensed and abstracted into meaningful codes and categories. All textual data were entered into a spreadsheet format. Each participant's responses were reduced into meaning units, or groups of statements that relate to the same central context (Graneheim & Lundman, 2004). These were then further abbreviated into condensed meaning units and then abstracted and labeled with codes. The codes were then compared for similarities and differences and were finally sorted into sub-categories and categories.
Trustworthiness. While there are widely recognized methods for assessing reliability and validity in quantitative analysis, there are also a number of approaches posed in qualitative research to ensure trustworthiness. One such method for achieving trustworthiness in qualitative research is credibility, which according to Lincoln and Guba (1985), make it more likely that the findings will address the intended focus and, therefore be considered believable. The analysis conducted on the qualitative data for this study attempted to achieve credibility by using a peer debriefing method as described by Lincoln and Guba (1985). This aids the researcher in reporting the participants' perspectives as accurately as possible. The author and a member of the dissertation committee discussed and examined the categories and sub-categories until overall agreement was reached on the coding system. Granaheim and Lundman (2004) suggest that this is done not just to make certain that data are labeled and organized exactly the same, but to ensure that multiple researchers and experts could reach agreement in the way that data is sorted and labeled.
The coded data were additionally assessed for dependability, or the reliability and stability of the data (Lincoln & Guba, 1985). Determining interrater reliability, or the amount of agreement between two independent raters coding the same data, assesses for dependability. The interrater reliability for this analysis was calculated with Cohen's Kappa and indicated high agreement (κ= 0.92 -0.96) (Downe-Wamboldt, 1992).

Limitations
There were limitations to the study. First, the use of a convenience sample poses challenges that create a great risk for sampling bias (Polit & Beck, 2008).
Nurses who chose to participate in this study may not have been an accurate representation of all nurses working on these four units, or of nurses in general.
Additionally, nurses may have floated to multiple nursing units, which may then affect families' and nurses' perceptions of family nursing practice on individual units. The use of a convenience sample was a reasonable approach, however, in this initial descriptive study.
The quantitative results generated from this study have limited generalizability, as the study was conducted on four medical-surgical nursing units in one community hospital. The results therefore may not be generalizable to nurses and families involved in family care at other acute care facilities, in particular those institutions that do not have Magnet status. This study also used a small sample size, therefore the results may not be generalizable to larger populations of nurses and families.
Finally, although the approach of surveying one family member rather than the entire family unit is applicable to this study, it is important to note that other family members may have different perspectives regarding family function, family health and perceived social support.

RESULTS
This study was designed to describe the relationships between the family and nurses when an older adult with diabetes was hospitalized. Additionally this study examined the impact these differences had on family perceptions of family function, family health and social support received.
A sample of 60 registered nurses and 60 family members of older adult patients admitted with diabetes to four medical-surgical units in a community hospital consented to participate in this study. with more than 20 years of in practice. Forty percent (n=24) of the current sample reported having less than five years in their current job assignment, 41.7% (n=25)

Characteristics of the Participants
reported having 5-10 years in their current job assignment, 10% (n=6) had 10-15 years in their current job assignment, 6.7% (n=4) had 15-20 years in their current job assignment and 1.7% (n=1) had greater than 20 years in their current job assignment.
Of the sixty nurses who participated in this study, 3.3% (n=2) reported their highest level of education was a nursing diploma, 18.3% (n=11) had an Associate's degree, 75% (n=45) had a Bachelor's degree and 3.3% (n=2) had a Master's degree.
Family participants. There were sixty family members who completed the Family Functioning, Family Health and Perceived Social Support Scale (FAFHES).
The participants were family members of patients admitted to the same four units where nursing data was gathered. Fifteen surveys were completed by family members from each unit. As described in Table 5, most family members who completed the survey instrument were over the age of 56 (58.3%, n=35), 25% (n=15) were between age 46 to 55, 6.7% (n=4) were between the ages of 36 to 45, 3.3% (n=2) were in the 26-35 age range and 6.7% (n=4) were in the 18-25 age range. There were thirty-eight female participants and twenty-two male participants. The majority of family member participants had a high school degree (35%, n=21), some college (28.3%, n=17) or a bachelor's degree (18.3%, n=11). The majority of participants were married (71.7%, n=43). Thirty-three percent (n=20) were the patient's spouse. Fifty percent (n=30) were either a son or daughter to the patient currently hospitalized. Many of the patients lived with the family member who was completing the survey (61.7%, n=37) and a little more than half (54.2%, n=32) had been hospitalized 1-5 times in the past year.

Family Perspectives
Family members perceptions' of their family functioning, family health and perceived social support were positive for participants who had an older adult family member hospitalized. year. Family correlation analysis results are illustrated in Table 6.
Family health. Family health is defined as the family's sense of knowledge, ill-being, well-being, activity, and values as it relates to the family unit (Harju et al., 2011). In this sample, family health was reported as high (M = 4.92, SD = .55). There were no significant correlations between family health and the family member participant's age, gender, education level, marital status, relationship to the patient, living arrangements (i.e. with or apart from the identified patient), or number of years with diabetes. Table 6 illustrates the significant, small to moderate, negative relationship between the number of times that the patient had been hospitalized in the past year and the reported family health (r 2 [59] = -.268, p = .040). This suggests that as the patient is hospitalized more frequently, family members report an overall decline in family health.
Perceived social support. Social support is an intentional human interaction that involves affect, affirmation and/or concrete aid offered by someone (Tarkka et al., 2003). Nurses display affect as they show appreciation, respect, and the ability to create a sense of security. Affirmation consists of reinforcement, feedback or having some influence on decision-making, while concrete aid may entail spending time helping patients and families or taking care of an issue (Tarkka et al., 2003). Family member participants in this study reported high levels of social support from nurses (M = 4.77, SD = .83). There were no significant correlations found between perceived social support and family member demographics, including age, gender, education level, marital status, relationship to the patient, living arrangements (i.e. with or apart from the identified patient), number of years with diabetes, or number of hospitalizations this year. Family correlation analysis results are illustrated in Table 6.

Nurses' Perceptions
The Practice Appraisal. Nurses' critical practice appraisal of their family nursing practice reflects nurses' assessment of their confidence, satisfaction, knowledge, skill and comfort when working with families (Simpson & Tarrant, 2006). For this sample, nurses' critical practice appraisal was reported as high (M = 1.86, SD = .58).
Spearman's correlation analysis was used to explore relationships between nursing demographic variables and nurses' critical practice appraisal. FNPS variable and demographic variable correlations are shown in Table 7. There were no significant relationships between nurses' critical practice appraisal and gender or education level.
There were, however, significant relationships between several other demographic variables and nurses' critical practice appraisal. There was a moderate and significant, negative relationship found between age and nurses' critical practice appraisal (r 2 [60] = -.450, p =.000). In this sample, the older the Registered Nurse, the more confident he/she was in family nursing practice.
There was also a moderate and significant, negative correlation between total years in nursing and nurses' critical appraisal of family nursing practice (r 2 [60] = -0.538, p =.000). For nurses in this study, the more years a nurse spent as a practicing nurse, the higher they critically appraised their family nursing practice. The total years that nurses have worked on their current unit also had moderate and significant negative correlation with nurses' critical appraisal of their family nursing practice (r 2 [60] = -.406, p =.001). This suggests that nurses with the most number of years on their current work unit critically appraised their family nursing practice higher than nurses' with fewer years working on their current unit. Registered Nurse correlation analysis results are illustrated in Table 7.
Reciprocity. The nurse-family relationship is a mutual relationship that develops during interactions between the nurse, the patient and the family (Simpson & Tarrant, 2006;. It is characterized by nurses' reflections on planning care, promoting family participation and reciprocity when working with families (Simpson & Tarrant, 2006). Appreciating reciprocity in the nurse-family relationship includes having the ability to assess family needs and issues, engaging and listening to the entire family unit, focusing on strengths and facilitating necessary change . Nurses in this study reported their experiences of the interaction and reciprocity in the nurse-family relationship as high (M = 1.78, SD = .47).
There were no significant relationships found between nurses' experiences of the interaction and reciprocity in the nurse-family relationship with either gender or education level. Age, however, was moderately, and significantly negatively correlated with nurses' experiences of the interaction and reciprocity in the nursefamily relationship (r 2 [60] = -.277, p = .032). These findings demonstrate that older nurses reported their experiences of the interaction and reciprocity in the nurse-family relationship higher than younger nurses in this study.
There was a significant, moderate and negative correlation between total years in nursing with nurses' experiences of the interaction and reciprocity in the nursefamily relationship (r 2 [60] = -0.342, p = .008). This indicates that nurses with more years in practice reported their experiences of the interaction and reciprocity in the nurse-family relationship higher than nurses with fewer numbers of years practicing.
Registered Nurse correlation analysis results are illustrated in Table 7.

Nurses' Perceptions of Family Nursing Practice
The hospitalizations. Some nurses described this as a lack of physical family members presence. Still other nurses addressed problems that they see in families related to heightened anxiety brought on by hospitalization of their loved one. In some cases, families may not be able to completely support their loved one because of a diminished ability to cope.
Family member issues. Some nurses described family member issues, which may include challenges within the family unit prior to admission or family member health problems. One nurse stated, "sometimes the family members have health issues of their own [and] have no coping mechanisms." Another nurse explained that difficulties in working with families might stem from "old family habits." No family presence. Nurses expressed concerns they have with some families being unavailable during their loved ones' hospitalizations. Some nurses described this as a lack of family member presence. Several nurses described that sometimes there is no family presence, or when patients are in the hospital, family are not available.
Nurses expressed this may be because family members are working or "they are nowhere to be found." While others were markedly different, expressing that family may not want to be included in care planning, "some families aren't willing to be involved or helpful." Situational anxiety. Still other nurses addressed problems that they saw in families related to heightened anxiety brought on by hospitalization of their loved one. Nurses also reported that some family members might be "…resistant to any changes in patient care…" and disagree with the nurse over best treatments for the patient.
Family conflict. Nurses described issues related to conflict that mainly related to differences in opinion which arise among family members. Within this category, there were two distinct sub-categories: family-family disagreements and familypatient disagreements. Assists the nurse. Several nurses expressed that including family assists the nurse because many families like to provide care to their loved one. Families may offer direct care, or they may offer psychosocial support in a way that nurses cannot.
One nurse reported that including family "helps staff and the patient, [as well as] helps with confused patients" and another stated that family "help with care." Common goals. Being able to share common goals with family was described by nurses as collaboration or shared agreement between family and the nurse regarding the patient's plan of care. One nurse expressed "it's great for the patient, [and] for the family to be on the same page," and another pointed out "some family members are realistic and very helpful and understand the goals we set for their loved ones." Help with planning. Several nurses expressed that families help with planning, specifically with discharge planning or with care planning during acute care admissions. For example one nurse stated, "[family] can help with discharge planning with patients" and another described that working with families allows for a "more realistic plan of care." Build family capacity to care. Some nurses believed involving families in care is a way to build family capacity to care for their loved one. This was described as continuity in care that keeps families and nurses informed about patients' conditions.
One nurse explained, "they learn more about how to care for the patient in terms of skin care and treatments the patient may need." Another nurse described that the work in partnership with families may increase family members desire to care for the patient, she stated, "by educating the family well -some people are more inclined to monitor their loved ones." Communicate / translate to patient. Some nurses described how families explain the plan of care and illness management in a very individual way, which the patient may understand better. They are able to communicate / translate to patient, which is illustrated by one nurse who stated, "often, family can explain the plan of care and discharge plan in a way the patient can understand better…" Strengthening nurse-family relationship. Many nurses communicated that the advantage of including family in care is that it can strengthen the relationship they develop with families. As families feel more involved, this creates a sense of trust and comfort level between the nurse, patient and family. Within this category there were two sub-categories labeled as: patient, family and nurse satisfaction and reducing family anxiety.
Patient, family and nurse satisfaction. Several nurses reported that when they involved families in care, it increased patient, family and nurse satisfaction. One nurse stated, "By involving the family you avoid family members feeling upset, angry… The more they feel involved, the more confidence and satisfaction they feel with the hospital." Another described it in terms of nurse satisfaction, "I get great satisfaction in keeping the family informed." Reduces family anxiety. Many nurses also described that including families reduces family anxiety. Involving families in the plan of care allows families to be more at ease, as well as, increases their trust in the nurse. For example, one nurse Provided education for family. When participants were asked how they had involved families in their nursing practice, many nurses described having provided education for the family. There were two sub-categories regarding the education nurses provide: teaching and providing information. Family member teaching was explained as more formal, sometimes mandated education, while providing information was described in more intimate terms and was on an ongoing, daily basis.
Teaching. Some nurses described the education they gave to families as Advocating for families' wishes. Several nurses referred to advocating for families' wishes in the past week by assisting families in staying with their admitted family member rather than just visiting. Some nurses saw that by contacting other health care providers and asking them to discuss the plan of care with the family, they were serving as a liaison for the family. One nurse explained how he/she was "making sure their wishes were passed onto the next nurse coming on." Another nurse described advocating for family by following through on a diagnostic suggestion from a family member, "[I] took daughter's suggestion of obtaining a UA [urinalysis]." Comfort from home. Some nurses discussed how they had created partnerships with families by encouraging families to bring in some comfort from home. This mainly involved dietary items, which the nurses believed would help improve the patient's nutrition, as well as allow the family to feel involved. One nurse stated "I had family bring in a patient's favorite food to a patient that had a very poor po [oral] intake." Another nurse described how family could be "encouraging [the] patient to increase participation in [activities of daily living] ADL's and increase intake at mealtimes. I encourage patient's families to bring in foods/ drinks the patient enjoys."

The Relationship Between Nurses' and Family Members Perceptions
Linear regression analysis evaluated the relationship between nurses' critical appraisal of their family nursing practice, and nurses' experience of the interaction and reciprocity in the nurse-family relationship, with families perceptions of their family function, family health and social support received existed. Linear regression analysis allowed each family's scores on all three subscales to be analyzed by including the corresponding nursing unit's aggregated scores as a covariate. There was no significant relationship (R 2 =.009, F(2, 57) =.263, p = .770) found between nurses' critical appraisal of their family nursing practice, and nurses' experience of the interaction and reciprocity in the nurse-family relationship with family functioning.
There was also no significant relationship found between nurses' critical appraisal of their family nursing practice, and nurses' experience of the interaction and reciprocity in the nurse-family relationship with family health (R 2 =.020, F(2, 57) = .593, p = .556). The last regression analysis, which assessed for a relationship between nurses' critical appraisal of their family nursing practice, and nurses' experience of the interaction and reciprocity in the nurse-family relationship with perceived social support (R 2 =.001, F(2,57) = .035, p = .966) was not significant.

Nurses' Perceptions Across Acute Care Units and Impact on Family Perceptions
As shown in Table 8, Analysis of variance (ANOVA) did not reveal significant differences in nurses' critical appraisal of their family nursing practice. There were significant (p < 0.05) differences discovered across the study units regarding nurses' There were no significant differences found across the units in family member perceptions, which is illustrated in Table 9. Therefore, there were no differences found between the nurses' critical appraisal of their family nursing practice, and nurses' experience of the interaction and reciprocity in the nurse-family relationship with families' perceptions of family functioning, family health and perceived social support across study units.

DISCUSSION
The cost of caring for patients with chronic illness accounts for approximately 75% of the United States current health care spending (Anderson, 2005;Hoffman, Rice, & Sung, 1996). This cost is estimated to increase as the number of persons over the age of 65 doubles in the next 30 years (Wolff, Starfield, & Anderson, 2002).
Considering that nearly 88% of Americans older than 65 have at least one chronic condition, it can be expected that health care spending will continue to rise, as will the need for specialized clinical care (CDC, 2009;Wolff et al., 2002) . Nearly 25% of those diagnosed with one chronic condition have activities of daily living (ADL) limitations, which often necessitates that families assume caregiving needs in the home (Anderson, 2005). There is, however, a notable absence of research studies that specifically address the impact of caring for chronically ill older adults on family structure and function.
Diabetes is a leading cause of chronic illness, which costs approximately $116 billion U.S. healthcare dollars annually (CDC, 2011). Many older patients managing chronic illnesses, such as diabetes, are taken care of at home by a family member.
Often, however, those with illnesses such as diabetes require hospital admissions due to exacerbations of the disease or for other co-morbid conditions which often accompany diabetes. Hospitalized older adult patients with chronic conditions are more likely to experience poor coordination in their care leading to contraindicated medical treatment (Wolff et al., 2002). Health care professionals working in pediatrics have long recognized the importance that families have in ensuring the best possible health outcomes for patients (Eichner & Johnson, 2012). There are, however, few studies that assess patient care coordination among older adult patients and their families. Therefore, this study aimed to examine the relationships between family and nurses when an older adult with diabetes is hospitalized.
The Calgary Family Intervention Model (CFIM) was used as the theoretical framework to guide this study. The CFIM is a model developed using a family systems nursing (FSN) approach. FSN provides a lens for examination of family strengths and evaluation of interactive family behaviors (Robinson, 1994;Wright & Leahey, 1990). FSN focuses on the whole family as a unit of care, allowing the nurse to simultaneously focus on the patient, the family and their illness. A Family Systems Nursing approach, particularly using the CFIM allows for the inclusion of family in care during transitional periods for older adults. The CFIM further advocates the importance of nurse-family reciprocity, which forms the foundation for mutual relationships that develop during interactions between the nurse, the patient and the family. By encouraging such purposeful interactions, this model proposes that there will be improvements in the family-nurse relationship as well as overall family wellbeing (Wright & Leahey, 1990).
The purpose of this study was to examine family perceptions of family functioning and family health during hospitalization, as well as family perceptions of social support received from nurses. This study also explored nurses' appraisal of their family nursing practice (FNP), as well as perceptions, or feelings, of the reciprocity involved in a nurse-family relationship. Also investigated was the relationship between the value that nurses place on a nurse-family relationship, and how families describe their family function, family health and social support. Finally, this study explored differences in nurses' appraisal of their FNP across the nursing units, as well as, any correlations between nurse appraisal of FNP and family function, family health and perceptions of social support.

Family Perspectives Related to Family Functioning, Family Health and Perceived
Social Support Family functioning. Leahey (1994, 2005) define family functioning as the ways that family members behave towards one another. Family members were asked to assess the ability of all members to plan events, contribute to household tasks, voice positive feelings, feel supported, and to convey perspectives about relationships within and outside of the family unit. It was not anticipated that family members caring for older adults with diabetes would describe their family functioning favorably.
The Calgary Family Intervention Model, which uses a family systems approach views the entire family as a unit of care (Wright & Leahey, 1994.
According to the tenets of the theory, then, it would be expected that chronic illness would negatively impact family functioning. The family as a whole unit, lives with, and alongside, the chronically ill family member thus making the illness a family experience, affecting the entire unit.
Family members in this study, however, reported their functioning as high.
These results were somewhat higher than family functioning findings in a similar study using the same instrument (Paavilainen et al., 2006), which assessed the family functioning of Finnish families of adults with heart disease. In Paavilainen, et al.'s (2006) study, family members of Finnish patients with heart disease were asked to describe their family functioning. These authors surveyed 161 participants, all of whom had a loved one diagnosed with heart disease.
The natures of these two diseases are very different. Home management of diabetes is likely to be considerably different from that of heart disease. Family members caring for an elderly person with diabetes describe being involved in meal planning, blood sugar monitoring and the administration of medications including injectable insulin. This is likely to have a substantial impact on family functioning.
However, in illnesses such as diabetes there are periods of quiescence that allow families to adjust the ways in which they function. Cardiac illnesses can be considerably different and often require families to continually adjust and re-adjust to acute exacerbations (Newby, 1996).
In a study of pediatric patients with sickle cell disease, Mitchell et al. (2007) found that there was a relationship between patients coping abilities with family functioning. Additionally, these researchers uncovered a relationship between negative patients coping abilities with higher rates of health care utilization (Mitchell et al., 2007). This reinforces the tenets of the CFIM, which proposes that it is important to note the differences in how families are functioning given various chronic illnesses.
According to Wright and Leahey (2005), this allows nurses to evaluate for the most appropriate nursing interventions, thus effecting the greatest change in families.
Nurses or families may identify a concern in either their cognitive, affective or behavioral domain of family functioning. Once this determination has been made, nurses can then decide on a nursing intervention that most fits the change needed in that domain of family functioning (Wright & Leahey, 2005).
Similar to the current study, many of the respondents in the Finnish study lived with the patient (83% in the Finnish study versus 61.7% in the current study) and most were married (92% in the Finnish study versus 71.7% in the current study). It is interesting to note that the ages of family member respondents were slightly different.
The Finnish study's participants had a mean age of 56.5 years. In the current study, family member respondents were slightly younger, primarily between the ages of 46-55. There were also distinctions in gender noted between the studies. The Finnish study's respondents were primarily female; while the current study gender was more evenly distributed between male and female respondents. The difference in ages and gender of family member respondents may account for higher reports of family functioning in the current study. Earlier studies have demonstrated that younger, female caregivers tend to report more distress than older, female caregivers (Ebbesen, Guyatt, McCartney, & Oldridge, 1990;Okkonen & Vanhanen, 2006). Conversely, Sloper (2000) found that male parents were more likely to feel more distressed than female parents, especially with repeated hospitalizations of pediatric cancer patients.
Another important distinction in the study conducted by Paavilainen et al. (2006) is that patients' ages ranged from 19-89 years, while in the current study all patients were required to be older than 65. There was also considerable variation in the length of time with diabetes in this study (15 had diabetes less than one year, 10 for 1-5 years, 10 for 6-10 years, 10 for 11-15 years, 4 for 16-20 years, 10 for 20 or more years and one patient was unsure of onset). Conversely, the Finnish study mainly included patients who have had heart disease for many years (Paavilainen et al., 2006).
Other authors have suggested that family adaptation can be more stressful in chronic illnesses such as cardiac disease, which are more acute in onset, are progressive and are more likely to be viewed as life threatening. These types of illnesses often challenge family members to constantly adapt and adjust to the illness (Newby, 1996), which may explain why respondents in this study reported better family functioning than in the Finnish study.
Exactly half of the family respondents in this study were children of the patient hospitalized, and another 33% were spouses. The Finnish study was different with only 21% of the participants being offspring of the patient and 77% spouses. The findings in the current study of families of older adults reinforces Leahey's (1994, 2005) views that the adult children often consider family functioning to be higher than do spouses of chronically ill patients.
Conversely, in a study of prostate cancer patients and their families (Harju et al., 2011), the researchers found that previous hospitalizations of the patient were associated with a better sense of well-being. According to Harju et al. (2011)  Interventions can be aimed to improve a family's cognitive domain by affecting changes in the way a family perceives its health problems. Treatment goals can also be aimed to reduce or increase emotion as needed to create change in the affective domain, which may better enable families to employ problem-solving techniques.
Finally, the behavioral domain of the family may have interventions directed at helping family members behave differently towards one another.
Perceived social support. Tarkka, Paavilainen, Lehti and Astedt-Kurki (2003) define social support as an intentional human interaction that involves affect, affirmation and/or concrete aid offered by someone. Families who participated in this study reported surprisingly high levels of social support from nurses, which differs considerably from the low social support scores described in a similar study of the families of Finnish cardiac patients using the FAFHES instrument (Tarkka et al., 2003). The identified patients in the Finnish study were adults, ranging in age from 19 to 89 years, while in the current study all patients were older adults, over the age of 65. This distinction in the age of patients is one possible explanation for the differences discovered in perceived social support found between these two studies. Tarkka et al. (2003) found that family members of older patients reported higher levels of social support received than family members of younger patients. It is probable that because the patients in the current study were older, as were the family member respondents in this study (58.3% were 56 or older), they had prior experience in caregiving for a loved one or had adjusted to the illness.
Although the length of illness was not significantly correlated with perceived social support in this study, as previously mentioned, family participants reported a wide range in length of time with diabetes. This may account for the higher levels of social support reported in this study when compared to a similar study by Tarkka et al. (2003) who found that family members of patients with advanced cardiac disease reported lower levels of social support.
According to Wright and Leahey (2005) nurses who practice family systems nursing will assist family members in discovering solutions to help alleviate suffering.
The high levels of perceived social support reported by respondents in this study suggests that nurses at this institution are currently employing many of the types of solutions suggested by the CFIM. This was to a certain degree unanticipated.
However, the rich descriptions offered by nurse participants in the qualitative data that was collected offers possible rationales into how families were integrated into care.

Reciprocity in the Nurse-Family Relationship
Practice appraisal. Nurses caring for older-adult patients with diabetes and their families critically appraised their family nursing practice at high levels. This indicates that respondents in this study were confident in their knowledge, skill and comfort in working with families. Although the nurses in the current study appraised their family nursing practice at high levels, it is important to note that often times perceptions and practice are inconsistent. Bruce and Ritchie (1997) found incongruences between nurse and family perceptions in an earlier study of 124 pediatric nurses caring for children and their families in an acute care setting. The nurses in this study reported that there was a lack of support and a greater need for skill development in their family nursing practice (Bruce & Ritchie, 1997). The current study demonstrates that older nurses, and those who had a greater number of years in practice as a nurse, rated their family nursing practice more highly than those who were younger and had fewer years in practice. Bruce and Ritchie similarly found that the age of the respondent influenced nurses' perceptions of their family practice.
However, years in nursing were not correlated with family nursing practice perceptions.
While there are limited studies reported in the literature regarding nurses' assessment of their family nursing practice with adults, some authors have explored nurses' perceptions of the emotional support they provide to families (Coco, Tossavainen, Jaaskelainen, & Turunen, 2013). In a study of 115 staff nurses caring for families and patients who had suffered traumatic brain injury (TBI), respondents reported that they regularly assisted family members' who were experiencing difficult emotions (Coco et al., 2013). Although Coco et al. did not specifically assess nurses' confidence, satisfaction, knowledge, skill and comfort in working with families, they did examine the level of competence nurses believed was needed to provide support to families.
The nurse respondents in the TBI study related that respecting and treating family members as individuals were basic competences that neurosurgical nurses should possess. In addition, there are more advanced skills, such as supporting family members ability to cope, which may need more training (Coco et al., 2013). A family systems nursing approach, such as that offered by Leahey in the CFIM (1994, 2005), which takes into account family needs and effective nursing interventions, may improve the skills needed for complex family nursing practice. with interaction and reciprocity in the nurse-family relationship higher than younger nurses. Takemoto et al. (2007) found that longer length of clinical experience influences a nurse's ability to practice enhanced family nursing. Wright and Leahey (1994) state that beginning clinicians are often overwhelmed with the intricacies of providing family nursing. The findings in the current study suggest that there is a role for veteran nurses to coach younger nurses using a model such as the CFIM to target family-specific interventions.

Reciprocity
Nursing challenges when involving family in assessment and care planning. Nurses inevitably influence the families they work with via their views, opinions, theories and thoughts (Wright & Leahey, 2005

Advantages of involving the family in assessment and care planning.
Nurses in the current study articulated that the main advantages of involving families in care planning are that it enhances patient care and strengthens the nurse-family relationship. A key reason that nurses in this study believed involving families would enhance patient care is that it facilitates their ability to know the patient. This is in accord with Wright and Leahey's belief (2005) that family members and nurses both bring strengths and resources to the nurse-family relationship that may go unnoticed by health care providers.
Tanner, Benner, Chesla and Gordon (1993) also discovered that nurses described how family presence enabled them to know the patient in a unique, individual way, allowing for more informed and skilled clinical judgments. Nurses in this study also described better outcomes for patients when families are involved.
Similarly,  found that nurses in their study believed patients are more cooperative and are more accepting of advice when their families participate.
Several nurses in the current study described how including families assists the nurse in the provision of care because family members help provide physical or psychosocial care to their loved one. Simpson and colleagues  likewise found that nurses described involving family as a time saver. According to nurses in the present study, as nurses and families work together they can share common goals. This can help with planning care or with discharge planning, which was also an advantage of family nursing described by nurses in this study.
Nurses in the current study described that including families builds family members' capacity to care for the patient. This allows for family members to better understand the patient and develop needed confidence and skills for care required after discharge . Nurses also described that families enhance patient care because they can communicate or translate the plan of care to the patient in a way the patient may better understand. Many families have their own way of communicating with each other, both verbally and non-verbally. Wright and Leahey (2005) suggest that the best way for nurses to ascertain how families communicate with one another is to assess for verbal and nonverbal forms of communication.
Consideration of all communication amongst family members as meaningful and purposeful enables for a better determination of possible interventions needed by families (Wright & Leahey, 2005).
Nurses in an earlier study by Simpson and colleagues  reported that family nursing promotes rapport between the patient, the family and the nurse. Many nurses in this study also indicated that an advantage to family nursing is that it strengthens the nurse-family relationship. Nurses in the current study reported that involving families increases patient, family and nurse satisfaction. Other nurses believed that including the family reduces their worry and anxiety. This is consistent with Simpson et al.'s (2006) study, which showed that when nurses involve families, they believed they better understood the reciprocity between the family, the illness and the nurse .
How nurses currently involve families in their nursing practice. Nurse respondents additionally described the ways in which they include families in their nursing practice. They discussed various methods used to include families in care, which ranged along a broad continuum from merely responding to a particular family need to more sophisticated strategies that actively encouraged family members to be part of the health care team.
Some nurses reported providing reassurance or education, while others actively engaged with families by creating partnerships. Much of the education that nurses reported providing was in the form of information sharing regarding the patient's plan of care or diagnostic tests. This is consistent with the earlier research of Astedt-Kurki et al. (2001), who found that nurses often associate family interactions primarily with information distribution. There were, however, some participants in this study who described more in-depth teaching practices they carry out with family members.
Nurses in this study also described more complex, indepth and bidirectional relationships that sometimes occur when they partner with families. Some nurses describe how they promote nurse-family relationships by actively encouraging family members to help them know the patient. This may be shaped around interviewing family members to find out patients' tendencies and preferences, or drawing family members into the care team. These complex techniques that nurses employ are supported and encouraged by models such as CFIM (Wright & Leahey, 1994.
The findings from this study reinforce the importance of how better descriptions of what nurses do to include families in care may help to inform best practices.

Relationship Between Family and Nurse Perceptions
This study explored the relationship between nurses' critical appraisal of their family nursing practice and their experiences of interaction and reciprocity in the nurse-family relationship with families' perceptions of family function, family health and perceived social support from nurses. This study found no relationship between nurses' critical practice appraisal and nurses' experiences of interaction and reciprocity in the nurse-family relationship with family functioning. There was also no relationship found between nurses' critical practice appraisal and nurses' experiences of interaction and reciprocity in the nurse-family relationship with family health.
Lastly, there was no relationship found between nurses' critical practice appraisal and nurses' experiences of interaction and reciprocity in the nurse-family relationship with perceived social support.
These findings suggest that, in this sample of nurses and families, there was no evidence that nurses' critical practice appraisal and nurses' experiences of interaction and reciprocity in the nurse-family relationship influences how families view their family functioning, family health or the social support received during hospitalization.
This is in contrast to the findings of Maxwell, Stuenkel and Saylor (2007) who assessed nurses' and family members' perceptions of needs being met in a critical care unit from one community hospital. These authors found significant differences in how families perceived their needs being met when compared to how nurses perceived meeting these needs. For example, nurses believed it was less important to talk to the family daily, inform the family how and why the patient was being treated and provide updates regarding the patient's condition. These inconsistencies may effect family perceptions, as well as family coping during hospitalizations (Maxwell, Stuenkel, & Saylor, 2007).
Finally, the current study tested to see if nurses' appraisal of their family nursing practice differed across units and if so, did these differences relate to families' perceptions of family functioning, family health, and perceived social support.
Findings suggest that there were significant differences in the way nurses perceived their family nursing practice between study units 1 and 3 in how nurses' experience interaction and reciprocity in the nurse-family relationship. Previous research has proposed that different nursing units within one organization may display considerable differences in the way a given unit's nurses practice, which supports the results found in this current study (Lauzon Clabo, 2008).
While there were differences in nurses' perceptions across the study units, there were no significant differences across the units in family member perceptions of nursing care and thus no relationship was found between nursing perceptions by unit and family perceptions by unit. This was not an expected finding, and may be associated with a threshold effect of investigating in one Magnet hospital. Hospitals that have been granted Magnet status are recognized for their quality patient care, nursing excellence and innovations in professional nursing practice. Thus, patients and families may feel that they receive high-quality nursing care at a Magnet institution regardless of which nursing unit they are admitted to.

Study Limitations
Several limitations need acknowledgment when interpreting the findings of this pilot study. The first concerns generalizability of the quantitative data, which is limited firstly because this study was conducted at one acute care institution, which may limit its comparability to other institutions. The sample was a non-probability convenience sample, which was small, further limiting the ability to generalize findings to all medical-surgical nurses or to other families caring for older adults with diabetes. Although nursing perceptions differed across units, family member perceptions across units were not found to be significantly different. This may relate to the small sample size, or it may also be explained by the phenomenon of having a loved one admitted to one Magnet institution, which is known for encouraging family practice and nursing education. Therefore, a possible explanation for these unexpected findings is that although nursing perceptions may differ across units in one institution, their actual family nursing practice is fairly consistent and well received by families.
Also important to note, is that the nursing participants were primarily female.
Previous work has suggested that male nurses tend to view families as less important and as less of a resource (Benzein, Johansson, Arestedt, & Saveman, 2008). It is, therefore, possible that if data had been collected from a larger more diverse sample, which included more male nurses, the responses may have differed. More research with larger, more representative samples is needed to further examine if gender differences do exist in nursing perceptions towards family nursing practice.
This study utilized two instruments, the FNPS and the FAFHES. Both of these instruments have reported validity and reliability, however, they have had relatively limited use. Further testing of this instrument with larger sample sizes is recommended. The FAFHES provides important quantitative data regarding family perspectives about their family health, family functioning and social support received from nurses, with no qualitative component. This researcher discovered that, on several occasions, family members had a desire to share much more data than a quantitative survey collection method allows. Future studies are planned that would allow family member participants to share their thoughts and feelings in a more indepth way.

Further Implications
The findings of this study have several implications, including theory testing, research, clinical practice and educational initiatives.

Theoretical.
A Family Systems Nursing approach, using the CFIM was used to guide this research study. The CFIM allows for an examination of family perceptions and nursing perceptions during care of the older adult with diabetes in the hospital. Family systems nursing and the CFIM were created so that whole families could be considered as the unit of care. One of CFIM's central tenets is the interaction between families and nurses. CFIM has been used to guide other research, but data has only been collected from either nurses or family members. Examining the possible relationship between how families describe nursing support, and how nurses describe their family nursing practice, is critical in understanding the interactions between nurses and families.
In the current study, the nurses' critical appraisal of their family nursing practice and how they perceived their experiences with interaction and reciprocity in the nurse-family relationship did not significantly impact how families perceived their family functioning, family health and social support received. It is important to note, however, that this study was conducted in one community hospital with Magnet status, which has high standards and expectations in nursing care delivery. This suggests that further application and testing of this model in family nursing research, are pertinent at diverse acute care settings.
Future Research. The current study findings taken in conjunction with the available literature suggest the need for future research studies. There is a need to develop well-planned research studies, which focus on collaboration, and participation of families in care of their chronically ill loved ones. Nurses are an integral part of this process and must be able to support and promote reciprocity in the nurse-family relationship. Effective nurse-family relationships encourage family involvement and have the potential to influence length of stay (Powers & Rubenstein, 1999). Further research may include a pre-test/post-test design, with implementation of a family-level intervention that is designed to improve family outcomes, as well as family nursing practice. A post-test may then evaluate if Family Systems Nursing was employed, and if so, did patients, families and/or nurses recognize more effective family nursing support. There are also potential implications for using and evaluating Family Systems Nursing in additional patient areas within various acute care settings. Multiple settings with larger sample sizes, which may have greater variability would permit for the use of stronger research designs, such as randomized control trials utilizing control and experimental groups.
Education and practice. This study offers implications for both nursing education and nursing practice. The first suggestion would be that schools of nursing and institutional administration provide greater learning opportunities, which promote the emphasis of family as a whole unit of care. Patients who suffer with chronic illness do not manage their conditions in isolation. Families of chronically ill patients, especially those with diabetes, provide tremendous amounts of care and should be encouraged to participate during hospitalizations as well as at home. Education about the value of including family in everyday nursing practice should be emphasized in nursing curriculums and continue into the practice setting. With utilization of a framework such as family systems nursing and a better understanding of family nursing practice, nurses will be much better prepared to assess and intervene when patients or families have an unmet need.    Note. **Correlation is significant at the .01 level (2-tailed); *Correlation is significant at the .05 level (2-tailed).

Reassurance
Provided support

Provided educa/on for family
Providing informa3on Teaching

Partnered with family
Ac3vely encouraged family to help me know the pa3ent Drawing into care team Advoca3ng for families' wishes Comfort from home