Effect of a CKD Dashboard on Patient Activation and Shared Decision Making (NEFRODASH)

The goal of this clinical pre-post study is to evaluate the effect of using a CKD dashboard, which visualizes individual patient data in patients with Chronic Kidney Disease stage 3b-4, and is used to structure the conversation during patient-clinician health visits. The main question[s] it aims to answer are: • what is the effect of the dashboard on patient activation levels (or outcomes related to patient activation)? • What is the effect of the dashboard on the conversation during health visits, including Shared Decision Making (SDM), Motivational interviewing (MI) and conversation topics? Participants will receive three surveys and have two routine care follow up consultations with their clinician audio recorded. Researchers will compare the same patients before and after implementation as well as compare them with patients in another hospital where no dashboard is (yet) implemented to see if patient activation levels (and related outcomes) increase, and whether there is an increase in SDM, MI and a difference in conversation topics.

Research questions/objectives: In this study we aim to better understand how patient activation can be improved on the consultation level. We hypothesize PROMs, the level of SDM and visualization of information (through dashboarding) are important mediators in this process and we which to explore their intertwined relations. The Santeon collaborative is currently developing a consultation dashboard that targets these mediators and aims to increase patient activation. The dashboard visualizes treatment goals and corresponding PROMs and clinical data of the individual patient (n=1) and will be used during consultations by healthcare professionals (HCPs) in their interaction with CKD patients in stage 3-4. With the dashboard, PROMs will also be newly implemented. In this study we thrive to study whether the use of the dashboard relates to an increase of patient activation through the mentioned mediators. The primary objective is to study the effect of the CKD consultation dashboard on patient activation in CKD management Secondary objectives: Assess how decision making is performed in 'all types of' decisions of CKD consultations and explore what the effects are of the CKD consultation dashboard on the level of shared decision making in these decisions. Asses how the topic of the conversation may change during consultations when the CKD dashboard is used. Assess the effects of the CKD consultation dashboard on outcomes related to patient activation: the patients view on HCP-patient relationship, patient perceived efficacy in interactions with HCPs, patients self-reported therapy adherence and disease insight. Moreover the (long) term effect on clinical outcome: eGFR (kidney function). The final objective is to evaluate the satisfaction with information the dashboard provides. Study design To evaluate the effects of the CKD consultation dashboard an observational mixed methods pre-post design with time-control is chosen. Participants will be followed longitudinally over time with measurements before and after implementation of the dashboard intervention. To increase validity, interrupted time series elements will be included in the design. A hospital where no intervention will be implemented (in the timeframe of the study) will be used as a control group during the same time-frame of measurements. This will control for time-related external factors influencing the measurements and provide insight in possible learning-effects of repeated questionnaires. Pre and post measurements will start in Maasstad Hospital, and St. Antonius Hospital in Nieuwegein will function as a control-hospital. Since the dashboard is developed as a quality improvement incentive and the goal is to successfully implement it in the standard of care, the pre-post design is favored over a randomized controlled trial (RCT). Moreover, the pre-intervention measurements provide valuable knowledge about the current state of patient activation in CKD and the level of involvement in decision making in consultations (focusing on all decisions in consultations instead of the renal replacement decision). Procedure Pre and post measurements will be done in this study in the group receiving the intervention in Maasstad hospital and during the same timepoints of measurement in the control group in st. Antonius hospital. These (pre and post) measurements consist of: A pre- and post-questionnaire assessing patient activation (primary outcome), perceived efficacy in patient-HCP interactions, perception of the HCP-patient relation, disease insight, therapy adherence and satisfaction with consultation (and dashboard). Two audiotaped observations of patient-HCP consultations; one pre implementation and one post implementation. Video is preferred over audio since it can show how the dashboard is used during consultations. However, since audio is less intrusive this might be preferable. In addition, for video conference consultations or consultations by telephone audio is the only technical possible method to record. The first 5 taped consultations will therefore be assessed on whether audio can be sufficient to analyze when the dashboard is used during live consultations (we suspect that with verbal explanation of the dashboard it can be extracted when the dashboard is used in conversation). If this is not the case we can switch to video for the live consultations. Semi-structured interviews with patients and HCPs to assess usability of the intervention and gather more in-depth knowledge on how the intervention affects (intermediate) outcomes. Not all included patients in the study will be interviewed. Patients can state on their informed consent forms whether they are willing to be contacted for an interview. Variables related to kidney function are retrieved from electronic health records to assess improvement on these outcomes in the long term. Difference with usual care The dashboard is being developed as a quality improvement intervention and will be implemented independent of the execution of this study. The usual care is changing; the use of the dashboard will be the new usual care. This study will evaluate this changing usual care. The difference for patients in the intervention hospital is that when they visit their, HCPs provide information/feedback through the dashboard during the consultations. HCPs will use the dashboard to give feedback on the course of disease and to evaluate and discuss the PROMs and treatment goals. In contrast to patients with kidney failure (G5A3) with or without kidney replacement therapy, PROMs have not yet been implemented as standard of care for patients in the trajectory of prevention of CKD progression (stages G3b-G4). The extended implementation of PROMS to a wider group of nephrology patients is advised by Nefrovisie, the Dutch national organization supporting the quality system for the entire nephrological chain. Nationally used PROMs in nephrology are the SF-12 (Short Form Health Survey) or PROMIS and the DSI (Dialysis Symptom Index). Thus, usual care differs because the dashboard influences information-provision during consultations and earlier implementation of PROMs in CKD patients with stages G3b-G4. Screening/selection participants All patients of stage G3b-G4, visiting the HCP during our inclusion period are eligible for this study. Every nephrologist and specialized nurse (treating professionals) will screen the list of patients planned for outpatient visits in a beforehand given time frame. The treating professionals will contact elidgible patients by e-mail or telephone. They will inform about the study and ask permission for the research coordinators (research nurses) of the internal department of the hospital of the patients and the coordinating investigator to contact the patient (by phone) to give additional information about the study. When this permission is given, name, telephone number and e-mail is shared with the research nurses and the coordinating investigator. They will contact the patients by phone for a more detailed explanation of the study. When the patient is interested in participating, a letter is send with information about the study and informed consent forms. The letter will be send a minimum of one week before the consultation of the patients in order to give enough time to deliberate about participation. If interested in participating in the study, patients will be asked to bring these forms to the upcoming consultation with their HCP. If there are additional questions, these can be asked during the consultations or by contacting the research team through the contact form in the letter. A selected group of participants (both HCPs and patients) will be asked to participate in an interview. HCPs that have worked with the dashboard will be selected on availability. Patients will be selected randomly i.e. by approaching every fifth included patient for participation in the interview. Separate informed consent for the interview will be obtained. For tele consults the same screening and selection pathway will be followed. However when participants do not visit their HCP live they are asked to hand in their informed consent forms by returning envelope to the corresponding hospital instead of returning them in their next live consultation. Sample size calculation The sample size calculation is based on the primary outcome; Patient Activation Measure. With an expected effect size of 0.4, desired power (beta) of 0.80, level of significance (alfa) of 0.05 and a two sided test, the study would require a sample size of 52 participants. In earlier studies with the PAM a dropout rate of 25% is taken into account in calculating the sample size, resulting in a minimum of 65 participants that should receive the intervention (Schuit et al. 2019), (van der Hout et al. 2020). Thus we thrive to include 65 patients in Maasstad hospital (receiving implementation) and 65 in the control group in St. Antonius, in total including 130 patients. Data-inspection By using histograms and scatter plots assumptions of linearity and homoscedasticity and normality will be checked. Outliers will be identified using Cook's distance. We will record the percentages of drop-out and missing data at each point of measurement. The sample calculation is adjusted to possible missing data or drop-outs (accounting a drop-out rate of 25%). When there are many missing's imputation techniques might be executed. Analysis Pilot questionnaires A pilot will be done with the questionnaire to check understandability and duration. This pilot will be done with patients from the research-patient panel of the kidney patient federation (NVN) that volunteer to give feedback on researchers. These are not patients that will be included in the study. Statistical analysis Patient characteristics and data from the questionnaires will be presented for continuous variables as mean and standard deviation (normal distribution) or median and interquartile range (no normal distribution). Categorical variables will be presented as a number with percentage. The data from the questionnaires will be described with existing or self-made scoring metrics. Descriptive statistics will be assessed with paired samples t-test (or Wilcoxon test when data is not normally distributed) and chi-square tests for differences between measured scores pre and post implementation. Chi-square and independent samples t-test (or Mann-Whitney U test when data is not normally distributed) will be used to analyze differences between implementation group and the time control group. To compare changes in the primary and secondary outcomes between both groups and over time, absolute scores will be presented as well as appropriate tests to make comparisons (either linear mixed models or (non)parametric tests, depending on the distribution). Analysis audiotapes The tapes will not be transcribed, but directly scored using the 4SDM coding scheme. For this purpose all decisions that are mentioned in the consultation are scored to eventually formulate groups of decisions and relate this to the level of observed shared decision making according to the 4SDM coding scheme. The analysis will also include the usage of the dashboard; is it used during the consultation and how? It will be analyzed whether or not the dashboard is used in the steps of decision making of the 4SDM coding scheme. Additionally content analysis will be performed on discussed PROMs to gather information on which PROMs are discussed during consultations. Finally, the duration of the consultation will be measured in minutes. Analysis qualitative research (semi-structured interviews) All of the interviews with both patients and HCPs will be audiotaped and transcribed verbatim. To be able to gather new concepts in possible effects of the consultation dashboard, first open coding will be applied through inductively coding the transcripts. Thereafter through axial coding, codes will be placed in groups to form concepts. Data analysis will be done by two researchers independently. Software Atlast.TI for qualitative analysis SPSS 26 for statistical analysis Ethical considerations Patients are asked to fill in a questionnaire that will take approximately 1 hour to fill out. This questionnaire will be repeated 2 times; when the dashboard has been implemented and patients visited their HCP again (T1 post implementation) and one year after implementation (T1 post implementation). Additionally two consultations with their HCP will be recorded which will require no further action from the participating patient or HCP. The usual care of included patients does not differ from other nephrology patients treated in the hospital, since the dashboard will be implemented for all nephrology patients as a healthcare improvement initiative. The usage of the dashboard does not bring any risks since it is only used as a tool to visualize and summarize data to enhance the information exchange between patients and HCPs during consultations. It is not replacing or altering existing data from EHR or PROMS. HCPs participating in this study will be asked to record consultations and will be asked to participate in interviews. There are no additional risks. Both patients and HCPs will not receive compensation for participating. However, when patients incur parking or transportation costs due to the study, these will be reimbursed. Including patients The treating HCP will perform screening for eligible patients in an upcoming timeframe of outpatient planning. Eligible patients will be contacted by the treating HCP by e-mail or telephone to inform about the study and ask for permission to share name, e-mail and telephone to be able to contact the research nurses/coordinators of the local department or the coordinating investigator. When permission is given patient will be contacted by either the research nurses/coordinators of their hospital or the coordinating researcher. The study will be explained in more detail and when interested in participation a letter will be send to the patient. This letter contains: information regarding the study, informed consent forms regarding the questionnaires, usage of EHR data (coded), and recordings and contact sheets. Additionally there will be asked for permission to contact the patient (by the research group) in order to ask them for an interview. The letter will be sent maximum of 2 weeks and a minimum of 1 week before the patients visit their HCP. When the patients visit their HCP they are asked to bring the informed consent and the consultation can be recorded if the forms are filled in properly. In the case of teleconsultations eligible patients receive the same patient information and informed consent forms by post, but are asked to return the informed consent by returning envelope. To inform and approach HCPs for participating in this study, meetings of the HCPs will be attended by a researcher in which information is given regarding the study. HCPs will be asked to participate verbally, informed consent will also be verbal. If some HCPs were not able to attend such meetings they will be contacted personally by the researcher to further inform them and ask for consent. Patients will be informed through the treating HCP, telephone through the coordinating research nurses of the local hospital or the coordinating investigator and with an information letter. With this letter a contact sheet of the researchers is added so that they can be contacted if there are any questions. Patients can also ask their questions during their consultation with the HCP and if needed the HCP can direct the patients to the research team. Patients for which teleconsultation is planned will be informed with the same information letter and an additional call. HCP's will be informed thoroughly before the start of the study by presentations about the study and per e-mail. The information letter with informed consent forms will be send minimally 1 week before the patients have their consultation. Therefor there will be enough time for considering the participation of the study and asking questions to the research team. We choose not to send the letter more than 2 weeks before the consultation because extending this period of time might result in people forgetting about it (and not bringing their forms to the upcoming consultation). For teleconsultations, since the returning envelope will be used for returning informed consent forms; 2-3 weeks before the planned teleconsultation information on the study is given. Patients sign the informed consent forms at home before they hand them over during the following consultation. Datamanagement The patient questionnaires will be digitalized and stored in a secure digital environment within the St. Antonius Ziekenhuis server in a secure management system (Redcap) that is password protected. Signed informed consent forms will be stored in a secured area in the hospital of inclusion, accessible to the principle investigator and coordinating investigator (or his representative) for a duration of 15 years. After 15 years these documents will be discarded and deleted. The paper versions of the patient questionnaire will be destroyed immediately after they have been digitalized. Informed consent forms will be located in a secured place in either st. Antonius or Maasstad hospital depending on where they were obtained. In St. Antonius forms will be stored in a secured archive of which the key is only kept by the trial bureau of internal medicine. Only above mentioned researchers can ask for this key. After the study it will be stored on a central archive of the St. Antonius Hospital, also secured and only accessible for the above mentioned researchers. In Maasstad the forms will be stored in a secured closed cabinet in the office of the principle investigator. Every study participant will be asked for written informed consent on participation, and on the use and storage of their data. All signed informed consent forms will be stored in a secured area in the hospital of inclusion, accessible to the principle investigator and coordinating investigator (or his representative) for a duration of 15 years. After 15 years, these documents will be discarded and deleted.

Filled out PROMs before consultation and discussed the CKD consultation dashboard together with their clinician in the (already scheduled) routine care follow up consultation.

In the period of the study PROMs and the dashboard are not yet implemented for this group of patients.

Online dashboard that visualizes patients' individual PROM- and clinical data related to their kidney disease.

Patient activation relates to patients having the knowledge, skills and confidence to take an active role in managing their condition. A widely used and validated measure for these aspects of patient activation is the Patient Activation Measure. The PAM contains four levels of activation indicating an increased level of patient activation per level: 1) believing that your role is important; 2) having confidence and knowledge to take action; 3) taking action; 4) staying on course in disease management under stress. Total scores range from 0-100; higher scores indicate higher patient activation.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Patient activation relates to patients having the knowledge, skills and confidence to take an active role in managing their condition. A widely used and validated measure for these aspects of patient activation is the Patient Activation Measure. The PAM contains four levels of activation indicating an increased level of patient activation per level: 1) believing that your role is important; 2) having confidence and knowledge to take action; 3) taking action; 4) staying on course in disease management under stress. Total scores range from 0-100; higher scores indicate higher patient activation.

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

Coding audio-recordings of consultations on the level of SDM according to the four steps of SDM from the model of Stiggelbout et al.Total scores ranges 0-24. Higher=better.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Coding audio-recordings of consultations on the level of motivational interviewing according to the global relational and technical components of the Motivational Interviewing Treatment Integrity (MITI) instrument. Relational global score, scale 1-5, ≥3.5 is considered sufficient, >4 is considered good. Technical global score, scale 1-5, ≥3 is considered sufficient, >4 is considered good

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Patient perceived self-efficacy to interact with clinicians. Range total score, 10-50. Higher=better.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Patient perceived self-efficacy to interact with clinicians. Range total score, 10-50. Higher=better.

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Non-standardized questions related to disease insight, including assessing patients' memory regarding their current kidney function.

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Non-standardized questions related to disease insight, including assessing patients' memory regarding their current kidney function.

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

After more than 3 months and after one year, only for intervention arm after implementation of the dashboard

Change from baseline (pre-implementation dashboard) to the following consultation more than 3 months later (post-implementation dashboard)

Change from baseline (pre-implementation dashboard) to one year after implementation of the dashboard.

Inclusion Criteria: Patients will be eligible for participations if they meet all the following criteria: Patients with Chronic Kidney Disease in stages G3b until G4 due to all causes Patients are treated for chronic kidney disease in the participating hospitals ≥ 18 years of age Be able to understand Dutch language in speech and writing or have someone assist them in understanding Dutch language. Able to provide informed consent Exclusion Criteria: Patients with dementia or severe cognitive impairment are excluded from this study.

Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the Patient Activation Measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Serv Res. 2004 Aug;39(4 Pt 1):1005-26. doi: 10.1111/j.1475-6773.2004.00269.x.

Henselmans I, Smets EMA, de Haes JCJM, Dijkgraaf MGW, de Vos FY, van Laarhoven HWM. A randomized controlled trial of a skills training for oncologists and a communication aid for patients to stimulate shared decision making about palliative systemic treatment (CHOICE): study protocol. BMC Cancer. 2018 Jan 8;18(1):55. doi: 10.1186/s12885-017-3838-8.

Stiggelbout AM, Pieterse AH, De Haes JC. Shared decision making: Concepts, evidence, and practice. Patient Educ Couns. 2015 Oct;98(10):1172-9. doi: 10.1016/j.pec.2015.06.022. Epub 2015 Jul 15.

Chan AHY, Horne R, Hankins M, Chisari C. The Medication Adherence Report Scale: A measurement tool for eliciting patients' reports of nonadherence. Br J Clin Pharmacol. 2020 Jul;86(7):1281-1288. doi: 10.1111/bcp.14193. Epub 2020 May 18.

Greenhalgh J, Gooding K, Gibbons E, Dalkin S, Wright J, Valderas J, Black N. How do patient reported outcome measures (PROMs) support clinician-patient communication and patient care? A realist synthesis. J Patient Rep Outcomes. 2018 Sep 15;2:42. doi: 10.1186/s41687-018-0061-6. eCollection 2018 Dec.

Ryan BL, Brown JB, Tremblay PF, Stewart M. Measuring Patients' Perceptions of Health Care Encounters: Examining the Factor Structure of the Revised Patient Perception of Patient-Centeredness (PPPC-R) Questionnaire. J Patient Cent Res Rev. 2019 Jul 29;6(3):192-202. doi: 10.17294/2330-0698.1696. eCollection 2019 Summer.

Maly RC, Frank JC, Marshall GN, DiMatteo MR, Reuben DB. Perceived efficacy in patient-physician interactions (PEPPI): validation of an instrument in older persons. J Am Geriatr Soc. 1998 Jul;46(7):889-94. doi: 10.1111/j.1532-5415.1998.tb02725.x.

Hankins MC, Llewellyn CD. Is the Satisfaction with Cancer Information Profile (SCIP) valid for tailoring information for patients with head and neck cancer? BMC Cancer. 2008 Jun 6;8:164. doi: 10.1186/1471-2407-8-164.

Degner LF, Sloan JA, Venkatesh P. The Control Preferences Scale. Can J Nurs Res. 1997 Fall;29(3):21-43.

Moyers TB, Rowell LN, Manuel JK, Ernst D, Houck JM. The Motivational Interviewing Treatment Integrity Code (MITI 4): Rationale, Preliminary Reliability and Validity. J Subst Abuse Treat. 2016 Jun;65:36-42. doi: 10.1016/j.jsat.2016.01.001. Epub 2016 Jan 13.