"I'm calling for hope, comfort and maybe some advice..." : Interaction and caller satisfaction in telenursing

Abstract: Background: The provision of nursing over the telephone is an appreciated asset in many parts of the world. Interaction between telenurse and caller is important for caller satisfaction, and satisfaction in turn is positively related to outcomes such as adherence, engagement in self-care and well-being. Despite the long history of telehealth services, research about interaction in telenursing is still in a developmental stage.Aim: The overall aim of the thesis was to develop a questionnaire that enables valid and reliable exploration of telenurse–caller interaction, and caller satisfaction with calls, and to explore interaction and its importance for caller satisfaction.Methods: The thesis applied a deductive approach based on a theoretical nursing model; the Interaction Model of Client Health Behavior (IMCHB). Interaction in the IMCHB, is divided into four components: health information, affective support, decisional control, and professional-technical competencies. In the first study (I), the Telenursing Interaction and Satisfaction Questionnaire (TISQ) was developed. Variables were identified through a literature review, structured according to the IMCHB, and worded as questionnaire items. Content validity was evaluated with input from cognitive interviews with callers, and professionals using the Content Validity Index (CVI). A consecutive sample of 1,400 callers received the TISQ by post, and test-retest reliability of single items was evaluated. In Study two (II), 25 items from the TISQ focusing on interaction between callers and telenurses were selected to form the Telenursing Interaction and Satisfaction Scale (TISS), a scale for the measurement of caller satisfaction with interaction in four subscales according to the IMCHB. Psychometric properties of the TISS were evaluated with a focus on data quality, factor structure, convergent validity, and reliability. In Study three (III), caller satisfaction with interaction (i.e., TISS scores) and their associations to overall satisfaction with calls were explored using descriptive statistics, repeated measures ANOVA and ordinal logistic regression models. In Study four (IV), 30 transcripts of authentic calls were analysed using content analysis with directed approach. A coding scheme based on the four components of interaction in the IMCHB was developed and applied for deductive categorisation of data. Manifest content of conversations (n = 37,447 words) was presented quantitatively and qualitatively. In addition, data was coded and analysed in relation to phases in the conversation process for telenursing: opening, listening, analysing, motivating, and ending.Results: The 60 item TISQ showed good content validity in the telenursing setting (I), with a Scale-CVI of 0.91. Test–retest reliability of single items was moderate to good (I). TISS-data deviated significantly from a normal distribution, but all response options were endorsed (II). The confirmatory factor analysis confirmed the four-factor structure of the TISS, and factor correlations were high (rs = 0.88 to 0.96) (II). A higher order model showed marginally deteriorated model fit values (II). Ordinal alpha, scale reliability, and test–retest reliability were satisfactory for all scales, and convergent validity was satisfactory (II). Callers expressed the highest level of satisfaction with affective support, while their satisfaction with decisional control was comparatively lower (III). Callers’ satisfaction with interaction was positively associated with overall satisfaction with calls, even when considering experienced waiting time, main result of the call, fulfilment of expectations, age, gender, self-rated health status and time of call (III). A total of 97% of manifest words in conversations were identified as nursing interaction based on the IMCHB (IV). Health information was the primary focus, particularly during the listening and analysing phases (IV). Affective support was the least prominent component, and callers’ reactions to advice were seldom discussed (IV). The opening phase was the shortest, and analysing phase the most word consuming (IV).Conclusions: The TISQ, including the TISS, provide a valid and reliable questionnaire for the study of telenurse-caller interaction, and caller satisfaction. Caller satisfaction with the interaction can be measured using the four subscales in the TISS. In cases where there is a problem with multicollinearity, a total scale score from the TISS can be utilized. All four components of interaction in the IMCHB play a crucial role in determining overall caller satisfaction with calls. Therefore, the development of interaction in telenursing need to consider all four components. The exchange of health information presents challenges for both telenurses and callers, and affective support can be provided to callers without verbalizing emotions. Among the four components, decisional control may have the largest potential for improvement in the studied setting. This can be achieved by acknowledging callers’ reactions to advice.