Introduction

Patient experience and satisfaction are inextricably linked to the perceived relationship a patient has with their physician [3, 15, 17, 19, 23, 24, 27, 40]. Patient satisfaction has been associated with patients reporting good communication with their physician [24]. The quality of the physician-patient relationship is of great importance in medical oncology, in which breaking bad news, discussing priorities, and conversations about prognosis and death are frequent topics [2, 39].

Studies of oncology patients have found patients experience significant levels of psychological distress [18, 28] which has been linked to decreased survival, suboptimal quality of life, and increased healthcare utilization [32]. Oncology patients have cited their relationship with their physician as highly impactful in their psychological adjustment to cancer [16, 18] and care experience [13], and a source of support in assisting with distressing events [30]. In particular, oncology patients are looking for communication with their physician that allows them to feel guided, build trust, and sustain hope [1]. In fact, research has shown that patients with higher levels of hopelessness, distress, and maladaptive coping reported feeling that their physicians were disengaged and less supportive [29]. The patient-physician relationship also affected perception of pain intensity [12], understanding of information, satisfaction with medical care, degree of hopefulness, and ensuing psychological adjustment [2]. Unfortunately, research has found that oncologists are frequently unaware of patients’ personal preferences, life priorities, and quality of life [28]. Furthermore, oncologists are often unaware of the full range of patient concerns [1] and are unable to meet patient information and emotional needs [18]. These unmet patient needs can make the creation of a positive patient-physician relationship more difficult.

An inpatient admission to the oncology unit is often a stressful time for patients [14]. Most admissions are for uncontrolled symptoms and may signal progression of disease or declining performance status [38] such that further anti-cancer treatments could be limited. Furthermore, patients are not likely to be cared for by their primary outpatient oncologist during a hospital admission. Inpatient attendings may not have met the patient before the admission and, though they may be oncologists who are adept at managing the medical issues requiring admission, the relationship may be established for the first time during the admission.

The goal of this study was to explore oncology patients’ perceptions of their inpatient hospital experience, focusing on quality of care. Using a qualitative and quantitative approach, oncology patients were interviewed to assess experience with physician communication, best practices, and recommendations for improvement. Additionally, physician daily rounds were observed and specific communication styles were studied.

Methods

A qualitative and quantitative study design was implemented to explore patients’ perceptions of their hospital experience, focusing on quality of care.

Sample

Patients admitted to the inpatient medical oncology and hematology units of Smilow Cancer Center at Yale New Haven Hospital were considered for eligibility. Eligibility criteria included adult oncology patients admitted during December 2016–July 2017. Of the 191 patients eligible, 90 declined and 43 were excluded, resulting in a final sample of 58 patients. Exclusion criteria included the following: non-English speaking, medically incapacitated, or unavailable > 2 occasions. Yale University IRB granted an exemption for this project as it met criteria for Quality Initiative Research. Consent to be interviewed was obtained from each patient.

Data collection

Interviews

The interview guide was modified with permission from Banka and colleagues to meet the needs of our research question and to reflect the oncology setting [3] (Supplemental A). Fourteen questions were added, five questions were modified, and five questions removed. During interviews, patients were asked to assign a grade for the quality of a physician’s communication skills according to their perceptions, A being always/perfectly, B as mostly, C as sometimes/occasionally, and D as never. Letter grade definitions were described to each patient during the interview. Validation of the adapted guide was not performed. Interviews were conducted in person using the adapted interview guide. The interview guide consisted of a 26-question script that included open-ended questions and 4-point Likert scale questions that assessed providers’ performance in a variety of patient communication domains. The interviewer was trained in performing qualitative interviews by one of the authors with extensive qualitative research experience (JPH). All interviews were recorded. Interviews averaged from 5.5 to 29.5 min with all questions addressed.

Observations

One hundred five physician-to-patient interactions were observed by one of the authors (AOA) during morning rounds over a three-month period. Behaviors observed included the frequency with which the physician introduced him or herself and clarified roles of the team and how often the physician sat at the eye level of the patient. Observations of patient-physician interaction were done before patient interviews and were collected using an abbreviated form of the REDE checklist [42]. Data obtained were used to compare to patients’ assessment of physician actions in the interview guide. Of the final 58 patients, 83% were observed directly by the author during morning rounds.

Analysis

Coding

Interviews were conducted by one of the authors (AOA) and analyzed by two coders (AOA and AK). Each transcript was reviewed using the constant comparative method. Each transcript was independently coded; then, consensus coding was performed in which two authors examined each transcript, identified themes, and reviewed discrepancies until agreement was reached. After several rounds of consensus coding, a preliminary coding structure was created and used for remaining coding. The final coding scheme was applied to all transcripts and reviewed by the second author. A third reviewer (JPH) with extensive qualitative research experience independently reviewed and coded transcripts to confirm accuracy of the final coding scheme. Content thematic analysis was used to identify themes and observe for systemic deficiencies.

Data organization

Quantitative data were created from the qualitative data. Frequency counts on polar questions (yes or no) and 4-point Likert scale questions were calculated. Frequency counts of the subthemes that emerged from the data served to triangulate with qualitative data.

Results

The majority of patients (53%) were over 60 years old and 52% were male; lung and gastrointestinal cancer were the most common diagnoses. Table 1 depicts patient demographic information.

Table 1 Patient demographics
Full size table

Thematic saturation, defined as the point in data coding when no new codes emerged [37], was reached after two-thirds of the interviews were completed but a decision was made to code the complete sample. Two major themes emerged from the qualitative data consensus analysis: hospital experience (Table 2) and physician communication skills (Table 3). Each major theme had multiple subthemes. All patients made comments in at least one major theme. Representative quotes for each subtheme are depicted in Tables 2 and 3. Within each table, each row depicts a subtheme, the frequency, percent, and representative quotes of the subtheme by column.

Table 2 Hospital experience
Full size table
Table 3 Physician communication skills
Full size table

Major theme 1: Hospital experience

Experience was defined as all components of the patient’s hospital stay separate from the patient’s interaction with physicians. Twelve subthemes emerged as defining the hospital experience: (1) attended to, (2) facility/staff, (3) nurses, (4) long wait time, (5) doctors, (6) coordination of care, (7) unnecessary medical procedures, (8) medications, (9) night awakenings, (10) pain, (11) not getting better, and (12) decreased mobility. All quotes for each subtheme are depicted in Table 2. Of the 58 patients interviewed, 49 elaborated with specific statements.

Within attended to, patients described varying amounts and quality of attentiveness received, with 35% of patients describing high-level attentiveness from physicians while 14% described less than desired level of attentiveness from physicians.

Facility and staff were emphasized as contributing factors to hospital experience by 35% of patients. Twenty-nine percent of patients provided high ratings of quality of Yale’s facility and staff while 6% of patients described that they were dissatisfied with their care from Yale and its staff.

Nurses were frequently mentioned as specific influencers of patients’ hospital experience with 33% of patients describing them as consistent sources of comfort and care.

Long wait time for medications or staff assistance was described as a dissatisfactory component of the hospital experience by 29% of patients.

Doctors were described by 20% of patients as a source of “wonderful” care. Sixty-seven percent of comments about doctors were linked with nurses. In contrast, 52% of comments about nurses were made independent of doctors.

Coordination of care was frequently reported by patients to be a relevant component to their hospital experience. Twelve percent described confusing coordination of care while 6% reported pleasure at the quality of coordination of care provided.

Thirty percent of patients described unnecessary medical procedures, medications, night awakenings, pain, not getting better, or decreased mobility to be sources of frustration and interruption within their hospital experience. See Table 2 for percent frequency of subthemes.

In general, patients commented that nurses, staff, and physicians made them feel cared for. Importantly, they emphasized the desire for being attended to and having organized, high-quality care in their hospital experience. Patients described the following subthemes exclusively in a negative context: long wait time, medical procedures, night awakenings, pain, not getting better, and decreased mobility, essentially noting discomfort with the administrative and psychosocial aspects of being an inpatient.

Major theme 2: Physician communication skills

Physician communication skills were defined as verbal and non-verbal components of exchange between patient and physician. Six subthemes emerged: (1) involving the patient and/or family in the care process, (2) method of information sharing, (3) lack of coordination of care, (4) use of medical jargon, (5) attending to patient needs, and (6) lack of patient’s perspective. All quotes for each subtheme are depicted in Table 3. Of the 58 patients interviewed, 40 elaborated with specific statements.

Within involving patient and/or family in the care process, 28% of patients reported feeling adequately involved in their care while 13% of patients noted frustration with either their lack of involvement and over-involvement of family or their family’s lack of involvement.

Within method of information sharing, 18% of patients reported feeling overwhelmed with the amount of material shared and preferred information that was provided in succinct briefings. Others noted the need for more information and repetition of information.

Lack of coordination of care was reported by 15% of patients to be a source of frustration due to changing doctors without notice, lack of clarity about who is in charge, and overall lack of communication amongst provider teams.

Ten percent of patients reported the use of medical jargon during communication with physicians, with all finding its use to be a language barrier due to their inability to “understand everything.”

Patients noted attending to their needs to be an important quality in communication with physicians with 8% reporting feeling adequately tended to.

Patients reported that physician’s lack of the patient’s perspective was a crucial deterrent to quality communication. Specifically, 8% of patients noted that physician awareness and understanding of their perspective as it related to team dynamics and inpatient flow was frustratingly lacking.

In general, patients commented that effective engagement of the patient in the care process and attending to and understanding patient needs were important and necessary qualities in physician-patient communication while lack of such and disorganized coordination of care defined suboptimal physician-patient communication. Interestingly, the majority of patient comments on communication focused on aspects they found to be suboptimal.

Hospital experience and physician communication skills

Importantly, common subthemes of patient engagement, attending to patient needs, and coordination of care were described as determining aspects in overall patient satisfaction and quality analysis within both hospital experience and quality communication with physicians.

Quantitative and frequency data

Purpose

The purpose of quantitative data from observations and Likert scale questions were to observe how well patients’ categorical assessment corresponded to observed actions during rounds or their qualitative assessment of care provided.

Observations and Likert scale-like physician action item questions

During morning rounds, physician teams were noted to introduce themselves and clarify individual roles in 56% of interactions. Similarly, 54% of the time the primary physician was observed to place him or herself at the patient’s eye level during discussion.

Likert scale-like grades on quality of physician actions are shown in Table 4. C and D responses were grouped for simplicity of reporting. Across all questions, physicians received A (always/perfectly) grades < 80% of the time. Physicians had their lowest grade (graded C/D most frequently) on how well they communicate about treatment (15% C/D, somewhat/never). In all other questions, physicians received always/perfectly grades 72–79% of the time, with the highest score in the area of clarity of terminology (79%). Physicians received always/perfectly grade of 76% in sensitivity to physical and emotional needs. Seventy-five percent of patients gave always/perfectly grades in questions about educating the patient and family about their care. In the overall question on quality of patient experience, patients gave always/perfectly grade 72% of the time.

Table 4 Physician action items Likert scale-like questions
Full size table

Amongst patients interviewed, they noted that the physician team always introduced themselves and described individual roles 72% of the time and that the primary physician placed him or herself at their eye level 43% of the time. Introducing team members and sitting at patient eye level are observations of attentiveness, a subtheme noted by patients to be of importance in determining satisfaction with hospital experience and physician communication. This discrepancy between observation and patient grades reveals that patients may significantly overestimate how often physician teams introduce themselves but are relatively accurate in recalling how often physicians sat with them. This observed inconsistency further supports the importance of attentiveness to patient satisfaction identified in the qualitative data.

Discussion

This study found that feeling attended to is important in shaping the patient experience. Specifically, having persistent attentiveness from physicians, a non-interrupted hospital environment, and organized, high-quality coordinated care defined a satisfactory hospital experience while suboptimal quality in any of these areas defined a dissatisfactory experience. Within physician communication, we found that having patient-centered care, which includes effective engagement of the patient, attending to patient needs, and understanding the patient perspective, defines quality communication while suboptimal quality in any of these areas defined dissatisfactory communication. All subthemes were noted less than 50% of the time, indicating that important areas of dissatisfaction were identified at this level and should be addressed. All subthemes described positively in hospital experience included either the concept of attentiveness or mention of select groups (i.e., nurses, doctors, and staff) who provided attentive care, thereby emphasizing that the most important aspect to creating a satisfactory hospital experience is engagement of the care team. Similarly, within the theme of physician communication skills, physician attentiveness to patient needs was highlighted as a contributing factor to high-quality communication. Along with attentiveness, nurses were found to be highly ranked in creating a satisfactory patient experience. Other studies have also found that nurses have been highly ranked as contributing to a satisfactory patient experience [27, 41].

Subthemes described as dissatisfactory, such as lack of coordination of care, involvement of the patient and/or family in the care process, method of information sharing, and medical jargon, were again related to physician inattentiveness. Research has shown that oncology patients want emotional support from physicians [19, 32], their families involved in their care [13], accessible information sharing, and to participate in the decision-making during the treatment process [6, 7, 20, 21, 26]. The discrepancy between patient communication needs and physician actions is not necessarily due to intentional neglect but possibly due to patient reliance on “ambiguous and indirect hints or cues” [32] to express emotion which may be difficult for providers to interpret. In addition, there is the common phenomenon of miscommunication in oncology in which information shared by the provider is believed to be complete or adequately explained but is in fact misinterpreted by patients [26].

Our findings emphasize the need of physician attentiveness that can appropriately recognize and respond to individual patient needs. It is important to note that patient satisfaction scales often show significant positive skew in responses [6, 7]. As a result, areas of dissatisfaction identified may be even more prevalent than reported here. Research has shown that there are many reasons for heterogeneity in information preferences, such as cross-cultural variation [9, 39], differences in processing large amounts of information [17], and variation in preferences of the patient and caregiver [28]. As such, oncologists must acknowledge this individual need if they want positive patient-physician relationships. Communication skills training, based on patient preferences, has been demonstrated to enhance the patient experience [19]. As a result, further research geared to creating a training program to create an environment that allows and encourages attentive individualized care can be designed to address our results. As patients also noted medical student communication style as shaping their perception of physician communication, medical students should be included in such communication training.

Patients defined wait time, medical procedures, medications, night awakenings, pain, not getting better, and decreased mobility as frustrating aspects of their hospital experience. These subthemes can be grouped together under the umbrella term of hospital environment, which include the administrative and psychosocial aspects of being an inpatient. The overall negative view of the hospital environment is unsurprising as research has found that administrative aspects such as noise and quality of pain control were important features that influenced the patient experience [27].

Amongst oncologists, a wide array of communication skills has been strongly linked to patient satisfaction in the literature. In particular, some communication skills have been shown to interfere with the creation of patient-centered care. Those communication skills included “broadcasting” in which physicians share information in long uninterrupted monologs [10], minimal use of questions to assess psychosocial health of patients [18], exclusive use of close-ended questions, and avoidance strategies [35]. The variety of suboptimal communication skills likely reflects the fact that up until the late 1990s, oncologists self-reported insufficient communication skills training [16]. As a result, intervention efforts to improve physician communication with patients have become more widespread and include interventions such as perspective training, integrated case conference, and intensive psychosocial and communication skills training (CST) [4, 5, 19, 33, 34]. Notably, CST programs have been repeatedly shown to improve communication skills of oncology clinicians [35] in addition to their attitudes and beliefs towards addressing certain topics such as psychosocial issues [25]. The positive results of such interventions show that increasing physician awareness of patient needs can significantly improve patient satisfaction [4, 5, 19, 33, 34].

Our results align with recommendations from existing empirically validated and widely used communication skills training programs such as VitalTalk and Comskil. VitalTalk offers online communication tools, including training courses and checklist, in addition to in-person physician workshops to assist physicians in improving their communication skills. Comskil programs aim to train physicians and other healthcare professionals in how to effectively communicate with patients with cancer via program modules typically taught in small-group workshops in a format that emphasizes role plays with simulated patients [22]. VitalTalk’s communication tools [11] used for discussing prognosis, serious illness, and goals of care [40], and the Comskil Model [8] collectively recommend incorporation of patient preference when information sharing, minimizing jargon, and responding to patient displayed emotion. All of these were areas reported by our patients to be lacking and provide opportunity to enhance training aimed at improving the patient experience.

Quantitative data asking patients to assess how well physicians performed, resulted in an average of 70% A (perfectly) grades overall except in communication about treatment where only 59% of patients gave a perfect grade. This finding supports the qualitative interviews, suggesting communication was lacking as part of the patient experience. In general, we found the grades assigned corresponded with the qualitative analysis.

Research has shown the value of sitting in improving the patient-physician relationship [31, 36]. We observed half of the time physicians sat (54%) and that the majority of patients remembered those efforts (43%). As a result, effort needs to be made to encourage and facilitate physician sitting during patient interactions. Furthermore, observations revealed that patients focused on aspects that are meant to build a close patient-physician relationship: team introductions and sitting. Interestingly, patients even overestimated the frequency of these behaviors, further supporting the importance of those aspects to them.

Agreement in themes from the qualitative-quantitative approach shows effectiveness of methods in exploring patients’ perceptions on quality of care. The study intends to inform clinical and operational practices physicians can incorporate into their patient relationships. These data are being incorporated into the Academy of Communication in Healthcare Relationship-Centered Communication Workshop for MDs and RNs at YNHH.

Limitations identified in the study include single-institution study, inability to include non-English speaking patients, narrow patient demographics, such as having a predominantly older age patient population, variable interview length, and non-response bias all of which decrease generalizability of results. Interestingly, in shorter interviews, patients still provided expanded qualitative descriptions of their perception of their care; however, they provided more immediate descriptions (i.e., less pauses) and used more succinct descriptors.

Conclusions

Our study demonstrated that physician attentiveness or lack thereof is a defining aspect of the quality of the patient experience. It is an important quality in communication and patients perceived, in this study, that their needs were not being fully addressed. Based on these results, the following are practices that physicians can incorporate into daily operations to improve patient satisfaction: (1) partner with nursing colleagues to work as a team to develop effective communication skills in patient attentiveness, (2) sit at eye level of patient, (3) decrease use of medical jargon, and (4) utilize reiterative inquiry about patient preferences for method of information sharing and desired family involvement.