Predictors of therapeutic communication between nurses and hospitalized patients.

BACKGROUND: Therapeutic communication is a purposeful interaction between health professionals and patients that helps to achieve positive health outcomes. There is a pressing need for research examining factors influencing effective implementation of therapeutic communication in relation to patient-centered care and satisfaction.
OBJECTIVE: This study was aimed at determining the effective implementation of therapeutic communication and its predictors.
METHODS: Institution based cross-sectional study was conducted at the Jimma University Specialized Hospital from March 21 to April 9, 2016. One hundred ninety two patients were recruited using stratified sampling. A questionnaire was used to collect data. One-way ANOVA for mean difference by socio-demographic characteristics, simple and multivariable linear regressions were conducted.
RESULTS: The study revealed that 67(34.9%) of the patients rated high level of therapeutic communication. Significant predictors of therapeutic communication implementation were educational status (β = 5.87, P = 0.011), language difference (β = -6, P = 0.014), education difference (β = 5.21, P = 0.010) and perceived patient view score (β = 3.57, P˂0.001).
CONCLUSION: Therapeutic communication was poorly implemented. Education, language difference, education difference and perceived patient view scores were significant predictors of therapeutic communication.

Introduction

Nursing practice is related to the interrelationships of people. Hildegard E. Peplau's theory of interpersonal relations stated that the relationship has orientation, identification, exploitation, and resolution phases (Peplau, 1952). This interactive relationship is a powerful medicinal tool (Peplau, 1952; Hemsley et al., 2011). Therapeutic communication is a central element of the nurse-patient interaction, which helps to achieve positive health outcomes (Lima et al., 2012; Rezende et al., 2013). Younis et al. (2015). The importance of nurse-patient communication in the nursing profession has been stated since the 19th century (Fleischer et al., 2009). An essential nursing skill is providing care by showing concern and supporting the patient with a good word (Włoszczak-szubzda and Jarosz, 2013).

The nursing process is achieved through an interpersonal environment. Each interaction relied on the principle that promotes well-being and enhances satisfaction (Younis et al., 2015; Henson, 2007). According to the World Health Organization (WHO) report, communication serves an instrumental role that is at the heart of who we are as human beings (WHO, 2010). According to the Ethiopian Federal Ministry of Health reference manual for nurses and health care managers, nurses' communication should be accurate, timely and effective (Ministry of Health, 2011). Faye Glenn Abdellah has described that communication incorporates verbal and non-verbal aspects (Abdellah et al., 1960). Nonverbal communication is expressed through body motions, touch, facial expressions, reflexes, gestures, eye contact, postures, groaning, grunting, crying, cultural artifacts and appearances (Roberts and Bucksey, 2007).

Therapeutic communication is a purposeful interpersonal interaction. It allows an efficient exchange of information (UK Essays, 2015). According to Health as Expanding Consciousness theory, the relationship has a purpose of identifying meaningful patterns and facilitating client's decision-making (Newman, 1997).

Nurses spend 20%–30% of their time for providing direct care at medical-surgical care unit (Hendrich et al., 2008). Regardless of the employment setting, there is consensus within the nursing field that effective therapeutic communication is integral to good practice (Webster, 2013; Molla et al., 2014; Finke et al., 2008; Bridges et al., 2013). However, nurses working in the general wards often do not consider communication as a key component of nursing service delivery (Bridges et al., 2013; Cristhiane et al., 2013; Chapman, 2009). This has an impact on the patient's satisfaction (Mcgilton et al., 2012; Haugan, 2014; Hemsley et al., 2011; Lasiter, 2014). It also increases the length of hospital staythat accounts for 53% of hospitals' total waste (Agarwal et al., 2010).

Therapeutic communication is based on trust, respect, faith, hope, fulfillment of emotional, physical and spiritual needs (Pullen and Mathias, 2010, 2011; Travelbee, 1964). Virginia Henderson stated that nurses should act as a substitute for the patient, helper to the patient and a partner with the patient (Henderson, 1964). Furthermore, nurses should use clear, simple, and objective questions (Dewar and Nolan, 2013; Moser et al., 2010). According to Watson, a nurse-patient relationship implies congruence, empathy, non-possessiveness, warmth, and effective communication. Effective communication has cognitive, affective, and behavior response components (Watson, 1997).

Nurse characteristics, time, organizational values, and socio-demographic characteristics affected nurses’ ability to establish interaction (Madula, 2013; Ojwang et al., 2010; Berry, 2009; Sheldon et al., 2008; Anoosheh et al., 2009; Peters et al., 2013; Rasheed, 2015; Zamanzadeh et al., 2014; Chapman, 2009). Furthermore, specific patient characteristics, sensory impairment, personality, disability, and psychological barriers affect the therapeutic communication (Bakhtiari and Moshtagh, 2007; Anoosheh et al., 2009; Albagawi, 2014).

To improve patient satisfaction towards the nursing care, researchers must identify factors influencing the effective implementation of therapeutic communication between nurses and patients (Devi and Victoria, 2013; Balandin, 2007). Therefore, this study tries to quantify the relationship between therapeutic communication and its predictors using linear regression. The finding is also important in strengthening the concept incorporated in the Hildegard E. Peplau Theory of Interpersonal Relations. To the best of our knowledge, there is no documented evidence regarding the effective implementation of therapeutic communication and its predictors in Ethiopia. Identifying factors that influence the therapeutic communication might be supportive for the successful accomplishment of a policy aimed at creating compassionate, respectful and caring health professionals.

Materials and methods

Study setting

The study was conducted at the Jimma University Specialized Hospital (JUSH), which is found in Jimma town. It is the only teaching and referral hospital in the southwestern part of the country. It provides services for about 15 million people. Annually, it delivers service for 15,000 inpatient, 160,000 outpatient, 11,000 emergency and 4500 obstetrics cases.

Study period, design and population

The study was conducted from March 21–April 9, 2016. An institution based cross-sectional study design was used.

The source population was admitted patients and the study population was sampled patients who fulfilled the inclusion criteria. Patients who were at least 18 years old and hospitalized for at least three days were included in the study.

Sample size and sampling technique

The sample size was determined using single population proportion formula with the following assumptions: 50 % proportion, Z a/2 is the Z value at 95% Confidence level (1.96) and 0.05 margin of error (d). Since the source population was 344 (˂10,000), finite population correction formula was used. Adding 10% for the non-response rate, the final sample size was 200.

A stratified sampling technique was employed. The ward was considered as a stratum and samples were selected within each stratum by using simple random sampling method.

Study variables

Dependent variable

Effective implementation of therapeutic communication.

Independent variables

Age, sex, language, educational status, religion, emotional change, familiarity to the nurses’ responsibilities, previous hospitalization, disease-related change, nurses' willingness to share information, understanding patients' needs, intimacy, attitude towards therapeutic communication, taking consent before procedures, unfamiliar medical terms use, ward and rooms condition, visitors presence, caretakers presence, working time, communicating other health professionals.

Measures

Therapeutic communication

According to Peplaus’ theory, nursing is a therapeutic process that involves a therapeutic relationship between the nurse and patient. Therapeutic relationship involves a therapeutic communication (Peplau, 1997). There are three dimensions of therapeutic communication. These dimensions are Expressions group, Clarify group and Validation group techniques (Rezende et al., 2013; Lima et al., 2012). It was measured based on the five point scale in the response option, i.e. 1 = Never, 2 = Rarely 3 = Sometimes, 4 = often and 5 = Always. The total scores range from 18-90. Sum scores were used for calculating the overall therapeutic communication score. A tertiale analysis used to classify the level of therapeutic communication. The dimensions of therapeutic communications are defined as:

Expression group techniques: are methods that facilitate the expression of thoughts.

Clarify group techniques: are methods that enable clarifying what is expressed by the patient.

Validation group techniques: are methods that enable the establishment of a common meaning of what is expressed by the patient.

Language difference: It implies spoken language difference between the nurse and patient.

Education difference: It implies an educational status difference between the nurse and the patient.

Perceived patient view score: It was assessed through a question asking whether the nurses gave adequate description concerning the disease and procedures. The response included the two categories: (1) No (2) Yes.

Patient related factor: It refers to familiarity to the nurse's duties.

Data collection

Structured questionnaire

The structured questionnaire was adapted after a review of different literatures (Cristhiane et al., 2013; Webster, 2013; Anoosheh et al., 2009; Albagawi, 2014). The questionnaire sought information on respondents’ socio-demographic characteristics, perceived implementation of expression techniques (α = 0.732), perceived implementation of clarify techniques (α = 0.739), perceived implementation of validation techniques (α = 0.829), overall patient agreement on the effectiveness of therapeutic communication technique implementation (α = 0.704), patient agreement on patients, nurses and organization related factors (α = 0.829). The validity of the questionnaire was also considered. A valid questionnaire was adapted and opinion from the experts working in the Jimma University and nurses working in the Shenen Gibe Hospital was obtained. A pre-test and modification of vague concepts was done.

Three laboratory technologists working in JUSH, who were fluent speakers of the Afan Oromo and Amharic languages, were recruited as the data collector. One supervisor was supervising the data collection. Training was provided for the data collectors and the supervisor for two days. The training focused on the study objective, meaning of each question and interview techniques. In addition, the role of data collectors and supervisor was covered.

Data quality

The English version of the data-collection tool was translated to Afan oromo and Amharic language, then re-translated in the English version to evaluate its consistency. Pretesting of the data collection tools was conducted at the Shenen Gibe Hospital using, 5% of the total sample. Training and supervision were provided for the data collectors and the supervisor. Code was given on the questionnaires. Data collectors and supervisor checked the filled questionnaire for completeness every day. Computer frequencies and data sorting were used to check for missed variables, outliers or other errors during data entry.

Data processing and analysis

Data were checked for completeness, and then each completed questionnaire was assigned a unique code. Subsequently, the data was entered using EpiData Manager (V2.0.0.25) and EpiData Entry Client (V2. 0.7.22). The generated data were exported to SPSS version 20. The data were cleaned by visualizing, calculating frequencies and sorting. One-way ANOVA for mean difference by socio-demographic characteristics was done. Bivariate analyses between dependent and independent variables were performed using simple linear regression. Enter method was used to enter variables during the bivariate analysis. All explanatory variables that had association in simple linear regression analysis with p-value less than 0.25 was entered into multvariable linear regression model. Enter method was used to enter variables into the final model. Linearity was checked. Normality of the data was assessed using a normality plots with tests, Kolmogorov-Smirnov test and Shapiro-wilk test. Outliers were checked. Levene's Test for Equality of variance was used to check homogeneity of variance. Co-linearity between predictor variables were checked using Tolerance and variance inflation factor (VIF). A P-value less than 0.05 was taken as significant association. Results were presented in text, figure, and tables.

Ethical considerations

Ethical clearance was obtained from the institutional Review board of Jimma University (RPGC/40739/2076). Verbal informed consent was taken from respondents and the participants assured that their participation recorded anonymously.

Results

In the study, 192 admitted patients participated obtaining a response rate of 96%.

Socio-demographic characteristics

Majority (54.2%) were female and 113 (58.9%) were between the ages of 25 and 34. The mean age was 32.23 ± 6.94 with a range of 19–52. More than two-thirds (72.9%) were married (Table 1).

Table 1

Socio-demographic characteristics of patients.

Variables		Frequency	Percent
Sex	Female	104	54.2
	Male	88	45.8
Age	18–24	26	13.5
	25–34	113	58.9
	35–44	40	20.8
	≥45	13	6.8
Marital status	Married	140	72.9
	Not married	52	27.1
Religion	Muslim	113	58.9
	Orthodox	58	30.2
	Protestant	17	8.9
	Catholic	2	1.0
	Others	2	1.0
Education	No formal education	63	32.8
	Primary(grade 1–8)	64	33.3
	Secondary(grade 9–12)	53	27.6
	Post-secondary(12+)	12	6.3
Ethnicity	Oromo	120	62.5
	Amhara	31	16.1
	Dawro	18	9.4
	Keffa	12	6.2
	Tigrie	3	1.6
	Others	8	4.2
Occupation	Unemployed	66	34.4
	Private	45	23.4
	Farmer	43	22.4
	Government employed	25	13.0
	Others	13	6.8

Effective implementation of therapeutic communication techniques

Expression, clarification and validation group techniques had mean scores of 35.24(SD = 9.72), 12.75 (SD = 3.32) and 9.53(SD = 2.90), respectively (Table 2).

Table 2

Descriptive statistics of effective implementation of therapeutic communication techniques.

Techniques	Items	Range	Mean (SD)
Techniques of expression group	11	11–55	35.24(9.72)
Techniques of clarification group	4	4–20	12.75(3.32)
Techniques of validation group	3	3–15	9.53(2.90)

Patient agreement on effectiveness of therapeutic communication

Patients admitted in Gynecology ward rated higher therapeutic communication mean score (mean = 4.18, SD = 1.04) as compared with patients admitted in Surgical ward (mean = 3.55, SD = 1.24), Obstetrics ward (mean = 3.53, SD = 1.12), Medical ward (mean = 3.78, SD = 1.03) and Ophthalmology ward (mean = 3.24, SD = 1.49). In surgical ward, the highest ranking was participation in decision-making (mean = 3.65, SD = 1.26) and having discussion with the nurse on self-care behavior and self-reliance (mean = 3.66, SD = 1.25). In obstetrics ward, the highest ranking was having discussion with the nurse on self-care behavior and self-reliance (mean = 3.73, SD = 1.23). Adequate and clear description was the highest ranking in gynecology ward (mean = 4.30, SD = 0.97) and medical ward (mean = 4.00, SD = 1.05).In ophthalmology ward the highest ranking was having adequate time to express patients’ feeling and worries (mean = 3.90, SD = 1.02) (Table 3).

Table 3

Descriptive statistics of overall patient agreement on the implementation of therapeutic communication.

Item	Wards
	Surgical	Obstetrics	Gynecology	Medical	Ophthalmology
	Mean(SD)	Mean(SD)	Mean(SD)	Mean(SD)	Mean(SD)
Had adequate time to express my feeling and worries	3.48(1.12)	3.63(0.76)	4.00(1.09)	3.97(0.99)	3.90(1.02)
Had adequate and clear description concerning the disease and procedures	3.48(1.31)	3.27(1.20)	4.30(0.97)	4.00(1.05)	3.30(1.38)
Participated in decision making	3.65(1.26)	3.40(1.22)	4.22(0.95)	3.76(0.86)	2.50(1.47
Discussed me on self-care behavior	3.51(1.26)	3.73(1.23)	4.17(1.11)	3.49(1.15)	3.15(1.72)
The nurse allowed me to ask questions	3.66(1.25)	3.63(1.19)	4.22(1.08)	3.70(1.10)	3.35(1.89)
Average	3.55(1.24)	3.53(1.12)	4.18(1.04)	3.78(1.03)	3.24(1.49)

Therapeutic communication score of different categories

The therapeutic communication mean score differed significantly among the age groups and the four educational status groups (Table 4).

Table 4

ANOVA table showing the relationship between different categories and therapeutic communication mean score.

Variables		Therapeutic Communication
		N	Mean ± SD	F	P
Age	18–24	26	60.88 ± 12.16	2.763	0.043
	25–34	113	57.88 ± 13.31
	35–44	40	57.55 ± 15.54
	≥45	13	47.54 ± 16.87
Education	No formal education	63	61.71 ± 11.30	3.535	0.016
	Primary(1–8)	64	57.06 ± 13.64
	Secondary(9–12)	53	53.47 ± 15.91
	Post-secondary(12+)	12	55.75 ± 16.89
Religion	Orthodox	58	57.83 ± 11.25	0.293	0.882
	Muslim	113	57.09 ± 15.79
	Protestant	17	60.29 ± 11.85
	Catholic	2	54.50 ± 7.78
	Others	2	52.00 ± 16.97
Ethnicity	Oromo	120	58.34 ± 15.13	0.237	0.946
	Amhara	31	56.29 ± 8.89
	Tigrie	3	53.67 ± 20.03
	Dawro	18	56.11 ± 13.83
	Keffa	12	56.58 ± 16.03
	Others	8	55.88 ± 13.23
Occupation	Government employed	25	56.84 ± 14.39	0.409	0.802
	Private job	45	58.29 ± 11.99
	Farmer	43	59.09 ± 12.89
	Unemployed	66	55.95 ± 16.32
	Others	13	58.85 ± 13.10

Level of effective therapeutic communication

Mean score of effective implementation of therapeutic communication level was 57.52 ± 14.10. Sixty-five (33.9%), 60(31.3%) and 67(34.9%) of the respondents reported low, moderate and high level of therapeutic communication (Fig. 1).

Fig. 1

Level of therapeutic communication between nurses and admitted patients.

Predictors of effective implantation of therapeutic communication

Patients who had no formal education had on average 5.870 higher therapeutic communication as compared to those who attended primary education at p = 0.011. Language difference had a negative association with the therapeutic communication. Accordingly, patients who had reported language difference as a factor influencing the effective implementation of therapeutic communication had on average 6.002 lower therapeutic communication as compared to those who hadn't reported at p = 0.014. Patients who had reported educational difference as a factor influencing the effective implementation of therapeutic communication had on average 5.208 higher therapeutic communication as compared to those who hadn't reported educational difference as a factor influencing the effective implementation of therapeutic communication at p = 0.010. In addition, perceived patient view score had a positive association with the effective implementation of therapeutic communication (Table 5).

Table 5

Predictors of effective therapeutic communication implementation between nurses and admitted patients.

Variables	Unstandardized coefficients		Standardized coefficients	P	95% CI
	B	SD	Beta
(constant)	61.85	5.53		0.000	(50.943,72.754)
Age	-0.26	0.14	-0.13	0.060	(-0.521,0.010)
Education
No formal education	5.87	2.28	0.19	0.011	(1.378,10.361)
primary(1–8)*
Secondary(9–12)	-1.1	2.43	-0.03	0.651	(-5.887,3.685)
Post-secondary(12+)	1.84	4.14	0.03	0.656	(-6.324,10.015)
Language difference
Yes	-6	2.42	-0.16	0.014	(-10.786,-1.219)
No*
Education difference
Yes	5.21	2	0.18	0.010	(1.262,9.154)
No*
Perceived patient view score	3.57	0.95	0.25	<0.001	(1.698,5.448)
Patient related factor	0.14	0.29	0.03	0.634	(-0.440,0.720)

Adjusted R2 = 0.195, Maximum VIF = 1.411, *: Reference category.

Discussion

The study revealed 33.9% of the nurses had low level of therapeutic communication, This implies that the communication skills they acquired through education was not adequate. It could be related to the lack of recurrent training therapeutic communication techniques.

Ineffective communication is reported as a significant factor in medical errors and inadvertent patient harm (Devi and Victoria, 2013). Current health system is aimed at creating competent and responsible health professionals (Ministry of Health, 2011).With this low level of therapeutic communication, it is hard to deliver the expected and high quality care. Hence, educational curriculum development about therapeutic communication is needed in all specializations and practice settings.

The study showed that the patients who had no formal education had on average 5.870 higher therapeutic communications as compared to those patients who attended primary education. Patients’ communication with nurses is directly influenced by their educational status. Furthermore, patients who had no formal education have lowered capacity of obtaining and processing basic health information. The nurses might elaborate issues for those patients by considering their inability to understand the information easily (Jahromi and Ramezanli, 2014).

Patients who had reported educational difference as a factor influencing the therapeutic communication had on average 5.208 higher therapeutic communications as compared to those who had not reported educational difference as a factor influencing therapeutic communication. Patients who mentioned educational background difference as a factor influencing therapeutic communication might ask and interact effectively with the nurse.

Madeleine M. Leininger theory of culture care diversity and universality stated that nurses must meet the language demands of the patients (Leininger, 1985). In the study, patients who had reported language difference as a factor influencing the therapeutic communication had on average 6.002 lower therapeutic communications as compared to those who had not. This finding is consistent with other study results (Anoosheh et al., 2009; Fleischer et al., 2009; Bakhtiari et al., 2009), which indicated that patients who perceived language difference as a factor influencing the therapeutic communication faced a problem while communicating.

The client who is in pain or preoccupied with their condition might have difficulty of communicating effectively. In our study, 41.6% of the patients reported that therapeutic communication was affected by their emotions. This finding is consistent with results reported by Zamanzadeh et al. (2014), who stated 73.6% of the patients felt that depression, fear and anxiety affected the therapeutic communication. Elderly patients emphasize mainly emotional change during their interaction (Lima et al., 2012).

In this study, 42.2% of the patients agreed that the presence of visitors affected the therapeutic communication. This is higher than finding from a study conducted in Iran, on barriers of nurse-patient communication, in which 21% of the patients reported the presence of the patients’ visitors affected the nurse-patient communication (Bakhtiari et al., 2009). The discrepancy could be due to socio-cultural, socio-economic and study area difference. In addition, in our study area patient visitors were allowed to enter into the hospital despite the regular time of visits. This might increase the impact of the visitors on the therapeutic communication, as they are available during nursing care and procedure.

In this study, 26% of the patients agreed that sex difference affected the therapeutic communication, which is consistent with Bakhtiari et al. (2009) who reported that 34% of patients believed gender difference affected therapeutic communication. This might be related to the community culture and belief.

Regarding available time for care, 36% of admitted patients agreed that it affects the therapeutic communication. Whereas, a study conducted in Iran showed that 56.4% of the patients described that nurses’ lack of time influences the relationship (Zamanzadeh et al., 2014). The discrepancy might be due to a difference in the workload of the nurses in the two areas. Workload affects time available for therapeutic communication.

The finding from this study mainly reflects the situation in Jimma University Specialized Hospital. Therefore, the findings should be interpreted with caution. The result might be affected by social desirability bias. This in turn might overestimate the effective implementation therapeutic communication.

Conclusion

Therapeutic communication is one part of nursing assessment and care. To provide high-quality care, predictors of effective implementation of therapeutic communication should be recognized. In this study, effective implementation therapeutic communication and its predictors were identified. The study has clearly shown low level of therapeutic communication implementation. The main factors influencing the therapeutic communication related to the patient, nurse and the organization were disease-related change, use of an unfamiliar medical term and ward's condition, respectively. Educational status, language difference, education difference and perceived patient view scores were found as significant predictors of therapeutic communication between nurses and admitted patients.

Declarations

Author contribution statement

Robera Olana Fite: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Masresha Assefa, Asresash Demissie, Tefera Belachew: Analyzed and interpreted the data.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.