Perception of Caring Among Patients and Nurses.

BACKGROUND: Empirical evidence supports the contention that implementing caring nurse behaviors results in improved patient experience; however, previous studies find differences between patient and nurse perceptions of caring. SIGNIFICANCE: Good patient experience is positively related to desired clinical and financial outcomes. Nurse caring is a critical component in the patient experience.
OBJECTIVE: The purposes of this project were to evaluate the congruency between nurse and patient perceptions of nurse caring in a long-term acute care hospital and to determine how much patient perception of nurse caring changes over time.
METHOD: The study employed mixed methods using a triangulation strategy in which quantitative data from patients and qualitative data from nurses were collected simultaneously and compared for interpretation.
RESULTS: Time affected patient perception of caring significantly. Patients and nurses disagreed about the extent to which nurses ask patients what they know about their illnesses, help them deal with bad feelings, and make them feel comfortable.
CONCLUSION: Patients and nurses do not always agree about the quality of caring behaviors, but exposure to nurses over time positively affects patient perception of nurse caring.

While there is inherent value to patients and their families that makes delivering a high-quality patient experience important, a positive patient experience is also associated with improved clinical outcomes, enhanced revenue, and the less tangible outcome of hospital reputation (1). Evidence supports the notion that a good patient experience is positively related to desired clinical outcomes including lower readmission and mortality rates (2,3). Additionally, patients’ experience of care, particularly communicating with their care providers, leads to improved adherence with care advice and treatment plans, especially among patients with chronic conditions (4,5). From the financial perspective, Medicare’s Hospital Value-Based Purchasing program directly rewards hospitals that have better patient experience scores, and there are indirect revenue enhancements for the hospitals related to lower medical malpractice risks and lower staff turnover (6). Deloitte investigated the association between patient experience and hospital financial performance, including operating and net profit margins and return on assets, and found that “hospitals with excellent HCAHPS patient ratings between 2008 and 2014 had a net margin of 4.7 percent, on average, as compared to just 1.8 percent for hospitals with low ratings.” (7, p1). Deloitte notes the correlation between profitability and patient experience for all hospital types (7).

In their 2013 Research Brief, National Research Corporation identifies a link between patient experience and reputation. National Research Corporation found this was especially true for hospitals with below-average scores for patient experience. National Research Corporation noted that what happens within a health-care facility at any given time may impact the reputation of the hospital in the near future even among individuals who have never had any direct health-care experience (1).

Because nursing care is of paramount importance to patients and families, studying the congruency between nurse and patient perceptions of caring behavior can help health-care facilities provide positive patient and family experiences. The Quality-Caring Model (QCM), developed by Joanne Duffy, contends that caring relationships between the nurse and the patient or family are the central focus of nursing’s work. Furthermore, Duffy asserts that caring relationships are cultivated and sustained over time (8). When a hospital system in north Texas created a new nursing professional practice model (PPM), Duffy’s QCM was integrated in it to help promote patient- and family-centered care experiences. This mixed-methods study compared patient perceptions of the caring they received from nurses, as defined by Duffy, to nurse interpretations of their own caring behaviors.

Problem

Patient-centered care (PCC) is an important goal for health care that is increasingly consumer-driven in the United States. In 2001, the Institute of Medicine defined PCC as “providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide all clinical decisions” (9). Donald Berwick later proposed a definition consistent with the contemporary patient experience movement: “The experience (to the extent the informed, individual patient desires it) of transparency, individualization, recognition, respect, dignity, and choice in all matters, without exception, related to one’s person, circumstances, and relationships in health care” (10).

Good quality patient–clinician relationships are essential to the application of PCC principles (11). Recognizing this fundamental belief, implementation of PPMs structured around caring relationships has swelled. Duffy’s QCM, which targets nursing care specifically, assumes when caring characterizes nurse–patient–family relationships, patient outcomes are better than when relationships are indifferent (12 –14). However, previous studies have also demonstrated differences between patient and nurse perceptions of the caring relationship (15). Thus, nursing has embraced PCC and the centrality of the nurse–patient relationship in theory for many years, but problems applying PCC in nursing practice may continue to challenge the profession (16).

Purpose

The patient-centered approach holds that patients know better than health-care providers if they are experiencing quality care, and patient ratings of the quality of care received are essential for improving patient outcomes. Therefore, judgments about the quality of caring behaviors manifested by nurses must be based on information obtained from patients, as well as nurses. With the foregoing in mind, we asked the following research questions: (1) Are perceptions of nurse caring congruent between nurses and patients in a long-term acute care hospital (LTACH) and (2) How much does patient perception of nurse caring change over time when patients are exposed to nurses over periods greater than the usual admission in an acute care hospital (5 days). The hypothesis advanced for the quantitative portion of the study was that patient perception of nurse caring changes positively over time. The objective of the qualitative portion of the study was to explore differences in patient and nurse perceptions of nurse caring.

Methods

Design

The study employed mixed methods using a triangulation strategy in which quantitative data from patients and qualitative data from nurses were collected simultaneously from patients and nurses and compared for interpretation (17). Weight was applied equally to quantitative and qualitative data. This design permitted analysis and interpretation from both patient and nurse perspectives. Data were collected during a single phase of the project and were analyzed independently using traditional methods.

Human Subjects’ Protections

This study was approved by the institutional review board governing research at the study hospital and complied with principles set forth in the Belmont Report (18) and Declaration of Helsinki (19) for protection of human research subjects. The investigators have no conflicts of interest to declare. Subjects provided written consent to participate in the study.

Sample and Setting

The setting for the study was a 10-bed LTACH in North Texas. The study hospital admits patients with complex conditions who have had an extended stay in an acute care hospital and need a minimum of 25 more days of acute medical management. This setting is a highly specialized environment in which nurses and patients interact over a longer period than typically occurs in acute care hospitals.

The convenience sample included 25 nonventilated, English- or Spanish-speaking adult patients without cognitive impairment who completed an objective measure of perceived nurse caring behaviors for the quantitative part of the study. The majority of patients were nonminority and covered by private insurance. Any registered nurse (RN) employed full-time by the study hospital was eligible to participate. Seven day-shift nurses of diverse ethnicities, races, and ages consented to participate in the qualitative part of the study and wrote 85 brief stories describing patient or family encounters.

Procedures

Patients and nurses were informed about the study by the investigators and provided written consent prior to participation. To assess the patient experience of nurse caring behaviors, each patient subject completed the Caring Assessment Tool (CAT-V) during the first 1 to 2 weeks after admission and again during the week of discharge.

Information about the Duffy model and its 8 “caring factors” was provided as part of a system-wide educational initiative for nurses when the PPM was deployed. This information was reviewed for all nurses in the target hospital prior to implementing procedures in this project. Participant RNs wrote reports about patient encounters between the nurses and patients enrolled in the study from the perspective of the caring factors and the new PPM. Participant nurses were also told that nurse caring from the patient perspective would be measured with the CAT-V and they were free to examine the items, but nurses were asked to classify their stories into the 8 “caring factors” classes. They were also free to relate their stories to specific CAT-V items. Participating nurses were blinded to subject responses to the CAT-V. Patient subjects were blinded to nurse stories.

Guided by Watson’s Theory of Human Caring (20,21), the CAT was developed in 1990 by Joanne Duffy and later revised to create the 27-item CAT-V used here. Evidence to support the validity and reliability of the CAT-V has been published (22,23). The 27-item tool is said to measure a single dimension (caring clinician–patient relationships) which includes characteristics (caring factors) listed in Table 1. The CAT-V is presented in a Likert-style format with a rating scale for items that ranges from never (low caring, 1) to always (high caring, 5). The sum of item scores can range from 27 to 135, with higher scores reflecting greater perceived caring globally.

Table 1.

Major Themes From Nurse Narratives and Related Patient Perception Items and Theoretical Domains.

Theme	Number of Theme Events	Related CAT-V Items	Related Theoretical “Caring Factors”a
Encouragement/reassurance	25	Help me deal with my bad feelingsEncourage my ability to go on with lifeSupport my sense of hopeHelp me see some good aspects of my situationHelp me to believe in myselfEncourage my ability to go on with lifeHelp me feel less worried	Encouraging manner Attentive reassurance
Respect/therapeutic presence	22	Seem interested in meSupport me with my beliefsShow respect for those things that have meaning to meRespect my need for privacyRespect meAllow me to choose the best time to talk about my concernsPay attention to me when I am talking	Human respectHealing environment
Problem-solving (tailoring care)	18	Know what is important to meHelp me figure out questions to ask other health-care professionalsAre concerned about how I view thingsAnticipate my needs	Appreciation of unique meaningsMutual problem-solving
Enhancing family access and understanding	13	Talk openly with my familyAre responsive to my family	Affiliation needs
Providing information	12	Help me explore alternative ways of dealing with my health problems	Mutual problem-solving
Physical care	5	Treat my body carefullyMake me feel as comfortable as possible	Attending to basic human needs
Eliciting information	3	Ask me how I think my health-care treatment is goingAsk me what I know about my illness	Appreciation of unique meanings
Humor	3	Help me deal with my bad feelings	Healing environmentEncouraging manner

Abbreviation: CAT, Caring Assessment Tool.

a From the Duffy Quality Caring Model introduced to nurses as part of a new professional practice model.

Data Analysis

For quantitative data, related-samples Wilcoxon signed ranks test was used to test the hypothesis that there was no significant difference between CAT-V scores early in hospitalization and at discharge. For qualitative data, content analysis was performed on 85 nurse narratives to count the frequencies of particular behaviors and identify recurring themes. Open coding was performed on the narratives by one member of the research team who had no contact with patients (Newcomb). This coding process associated themes from narratives with CAT-V items. Associations were inferred on the basis of words and concepts that referred to any activities in items on the CAT-V tool. Further coding and interpretation was done jointly by the authors.

Evidence from texts was sought to assess congruence between self-described nurse behavior and patient-described nurse behavior from the CAT-V tool. Although an 8-factor structure was originally described for the CAT-V (23), work in 2014 established the unidimensionality of the tool (11), thus statements in nurse texts were compared directly with related items in the CAT-V and no attempt was made to classify nurse texts into discreet “caring factor” categories until the end of the analysis. Each CAT-V item was an open code (theme).

Results

Reliability of the CAT-V was good. Cronbach’s alpha was .93 for the second survey and .89 for the first survey. Descriptive statistics for individual items on the CAT-V are shown in Table 2. Mean scores on the first administration of the CAT-V (early during admission) were significantly lower than mean scores on the CAT-V at discharge (Figure 1). Because CAT-V responses were skewed toward higher scores, the null hypothesis of no significant difference in total scores was tested using a Wilcoxon signed rank test and was rejected (P = .02). The effect of time on CAT-V scores was moderate (r = 0.33). Exposure to the cadre of nurses delivering care over the course of the admission accounted for about 11% of the variance (r 2 = 0.109) in the responses patients provided to the CAT-V questions.

Table 2.

Frequencies for CAT-V Items.

	Mean		Standard Deviation		Standardized Test Statistic	P Value	Effect Size (r)
	Time 1	Time 2	Time 1	Time 2
Since I have been a patient here, the nurse(s)…
Help me to believe in myself	4.4	4.3	0.81	0.74	−0.406	Non-sig	0.06
Make me feel as comfortable as possible	4.7	4.7	0.48	0.54	0.577	Non-sig	0.08
Support me with my beliefs	4.2	4.3	1.12	0.74	0.258	Non-sig	0.04
Pay attention to me when I am talking	4.6	4.7	0.57	0.56	0.577	Non-sig	0.08
Help me see some good aspects of my situation	4.4	4.2	0.86	0.91	−1.03	Non-sig	0.15
Help me feel less worried	4.4	4.2	0.70	0.78	−0.905	Non-sig	0.13
Anticipate my needs	4.1	4.3	0.83	0.80	1.387	Non-sig	0.20
Allow me to choose the best time to talk about my concerns	4.1	4.4	1.08	0.76	1.73	Non-sig	0.24
Are concerned about how I view things	4.1	4.3	0.87	0.89	1.435	Non-sig	0.20
Seem interested in me	4.6	4.8	0.76	0.66	2.0	.05	0.28
Respect me	4.7	4.8	0.62	0.50	1.89	Non-sig	0.27
Are responsive to my family	4.6	4.8	0.64	0.52	1.732	Non-sig	0.24
Acknowledge my inner feelings	4.18	4.16	0.988	0.85	.000	Non-sig	0
Help me understand how I am thinking about my illness	4.0	4.1	1.08	0.99	0.233	Non-sig	0.03
Help me explore alternative ways of dealing with my health problems	3.98	4.08	1.07	0.997	0.250	Non-sig	0.04
Ask me what I know about my illness	3.5	4.0	1.37	1.06	1.679	Non-sig	0.24
Help me figure out questions to ask other health-care professionals	3.5	4.0	1.17	0.99	2.177	.03	0.31
Support my sense of hope	4.2	4.3	1.09	0.945	0.000	Non-sig	0
Respect my need for privacy	4.3	4.7	0.98	0.63	2.456	.01	0.35
Ask me how I think my health-care treatment is going	3.8	4.2	1.12	0.97	2.392	.02	0.34
Treat my body carefully	4.5	4.7	0.65	0.54	2.236	.03	0.32
Help me with my special routine needs for sleep	4.3	4.7	0.89	0.46	2.251	.02	0.32
Encourage my ability to go on with life	4.1	4.3	1.24	0.95	0.885	Non-sig	0.12
Help me deal with my bad feelings	3.7	4.1	1.31	1.02	2.223	.03	0.31
Know what is important to me	4.1	4.3	1.19	0.94	0.546	Non-sig	0.08
Talk openly with my family	4.2	4.5	1.18	1.05	1.51	Non-sig	0.22
Show respect for those things that have meaning to me	4.3	4.7	1.1	0.69	2.124	.03	0.30

Abbreviations: CAT, Caring Assessment Tool; Non-sig, nonsignificant.

Figure 1.

Differences between mean CAT-V scores over time. CAT indicates Caring Assessment Tool.

The list of CAT-V items (each serving as a theme for preliminary coding) is shown in Table 2. After linking each statement in the texts to one of the CAT-V items, the codes were collapsed into broader themes, which roughly matched the 8 caring factors originally proposed by Duffy. The 3 most common descriptions of caring in nurse narratives were respect/presence, encouragement/reassurance, and mutual problem solving/tailoring care (Table 1). Examples included,The caring behaviors nurses described least were helping patients with basic human needs and providing a healing environment. For example,Congruence between nurses’ narrative evidence and CAT-V item scores was mixed (Table 3). Patient and nurse data were congruent in regard to nurses showing respect. Patients and nurses also agreed that nurses often failed to help patients figure out questions to ask other health-care professionals, although the score for this item on the CAT-V increased with time. In other areas, patients and nurses disagreed. Patients gave nurses low scores on helping patients deal with “bad feelings,” but nurses described themselves as providing strong care in this area when they classified their stories. Although nurses classified many stories into categories that might relate to helping patients with their “bad feelings,” no examples of helping patients with bad feelings, specifically, other than encouragement and reassurance, were found.

Table 3.

Congruence Between Patient and Nurse Perceptions.a

	Patient High CAT-V Scores	Patient Low CAT-V Scores
NurseStrong (high) narrative evidence of Caring	(High patient score/High nurse theme frequency)Item 11	(Low patient score/ High nurse theme frequency)Item 24
NurseWeak (low) narrative evidence of caring	(High patient score/Low nurse theme frequency)Item 2	(Low patient score/ Low nurse theme frequency)Item 16Item 17

aItem 2: Since I have been a patient here, the nurse/s make me feel as comfortable as possible. Item 11: Since I have been a patient here, the nurse/s respect me. Item 16: Since I have been a patient here, the nurse/s ask me what I know about my illness. Item 17: Since I have been a patient here, the nurse/s help me figure out questions to ask other health-care professionals. Item 24: Since I have been a patient here, the nurse/s help me deal with my bad feelings.

I listened to him explaining what happened to him over years. I was able to listen, reassure, and address some of his questions. The atmosphere gradually relaxed. (Reassurance/nursing presence)

Room temp, coffee mixture, meds before his bedtime: When you could get this just right he was a great [patient]. (Problem-solving/tailoring care)

We discussed together [nurse and patient] what we needed to get done on this day and we set up a planned time table. (Problem-solving/tailoring care)

I talked on and off all day with this patient’s wife. She is very stressed out with the patient’s condition and her financial situation. I gave her reassurance that it is OK to wait a few days before making big decisions. (Reassurance, nursing presence, family access and understanding)

I talked to her while giving her total care. (Nursing presence, physical care)

I turned and cleaned this patient every 2 hours. (Physical care)

“Privacy: close door at all times” sign on the door. He didn’t want anyone to see him, his Foley bag, wound vac canister, his bedside commode. (Respect, healing environment)

The term “bad feelings” was vague enough that patients and nurses weren’t sure how to interpret it. The most common question from patients regarding the CAT-V tool was the meaning of the term “bad feelings.” Patients usually wondered if this term referred to depression or suicidal thoughts. In most nurse narratives, bad feelings referred to unpleasant physical symptoms, such as dyspnea, rather than emotional distress. Nurses responded in most cases with reassurance or framed the bad feeling as a problem and tried to solve the problem.

Patients assigned low scores to nurses for asking about patients’ knowledge of their illnesses. What nurses described on their side was informing patients about their illnesses, not asking about the patient’s knowledge of their illnesses. Paradoxically, nurses rarely mentioned helping patients feel comfortable or attending to their basic physical needs, but patients gave high scores to nurses for these caring activities.

The most frequent theme in nurse narratives was encouragement, which was almost always confused or paired with reassurance. Hope appeared to be the most salient concept for nurses writing about reassurance and encouragement. Most nurse stories about encouragement/reassurance referred to communicating hope by informing patients about possible good outcomes. This was consistent with the high mean score on the CAT-V item, “nurses support my sense of hope.”

Discussion

Patient Perception of Care Improves Over Time

The hypothesis that patient perception of care improves over time was supported by the quantitative patient data. In this LTACH setting, each patient was exposed to the same nurses over many weeks. The improvement in CAT-V scores from admission to discharge suggests that as patients are exposed to the same cadre of nurses over time, perception of nurse caring improves, most likely through establishing trust as suggested by prior research (24). However, if building trust in nurses requires exposure to the same nurse multiple times over longer periods than a few days, earning patient trust may challenge nurses providing care in shorter stay facilities.

Strategies to build trust quickly have been suggested in business literature. Recommendations for building trust with consumers range from microactivities, such as crafting memorable first impressions, to overarching relationship-building practices, such as avoiding manipulation, being consistent, and engaging in real dialog (25). Coaching regarding evidence-based trust-building behaviors might benefit nurses who do not have long periods with patients to build relationships.

Patients and Nurses May Not Share Perceptions of Nursing Care Behaviors

Findings are consistent with prior evidence of incongruence between patient and nurse perceptions of care. Most research on quality of nursing care has been conducted from the nurse perspective (26).. The scant evidence that compares patient and nurse perceptions of nurse caring suggests that patients’ perceptions of care are not always congruent with nurses’ perceptions (27,28). In this study, the areas in which nurses and patients disagreed on the delivery of caring behaviors are shown in Table 3.

The finding that patients did not endorse the notion that nurses help with negative feelings while nurses thought they helped substantially in this regard may have been related to listening. Helping with negative feelings typically involves listening. The word “listen” occurred in 6% of the narratives. Terms most frequently used to describe conversing with patients included “talk to,” “explain,” “teach,” and “inform.” The quality of listening provided by nurses is impossible to ascertain, but if nurses listen passively without providing feedback to indicate patient concerns are heard and understood, the interaction may be dissatisfying to patients. Likewise, perceiving negative feelings as problems to be solved likely promotes the use of favored approaches, such as teaching and explaining, at the cost of more empathetic responses.

Another area of disagreement was helping with basic needs. Ironically nurses mentioned this caring behavior less often than any other, while patients scored nurses highly for it. Obviously, nurses recognize that meeting basic needs is nursing care because such activities are documented in the health record. The authors speculate that activities to meet basic needs has become so routine that nurses may not think of these care activities as manifesting caring. They may also think of these activities as within the purview of other health-care workers, such as unlicensed personnel.

Limitations

The sampling of nurse texts in the study ceased when saturation was attained. Because the number of nurses in the study hospital is small, saturation would be expected quickly and samples of nurses in additional LTACH facilities would improve the study. Although some methodologists argue that the notion of generalizability is not relevant to qualitative research (29), confidence in findings can only be enhanced with larger samples. Regarding the quantitative aspect of the study, the sample of patients was large enough to demonstrate the within-group effect of time, but the sample was too small to perform multivariable analysis with confidence.

In 28% of cases, patient ratings of caring behaviors delivered by nurses worsened over time. Because of small sample size, common themes that might associate with their disaffection were not identified. Further research specifically on factors that trouble patients about nursing care is important and should stretch beyond the limited responses to commercial patient satisfaction surveys that drive reimbursements for hospitals.

Conclusions

Nursing care in the study LTACH was guided by a PPM rooted in a patient-centered, quality care framework. Patients were generally positive about the caring behaviors their nurses demonstrated, but nurses are not always aware when their perceptions of caring behavior is not shared by patients. Nurse responses indicated that additional training on specific strategies that convey empathy or promote less directive patient teaching could be useful for enhancing the patient experience. Finally, time affects patient perception of care, and very short hospitalizations may present challenges to establishing caring relationships between patients and nurses.