A randomized controlled trial on the effects of decision aids for choosing discharge destinations of older stroke patients

In Japanese medical practice, older stroke survivors are bombarded with information regarding their discharge locations, increasing their decision-making difficulties. This study used a randomized controlled trial to evaluate the influence of using decision aids (DAs) matching the values of older stroke patients and their families on the internal conflict and participation in discharge destination decisions. Participants were randomly allocated to intervention and control groups. The intervention lasted for two months, from admission to discharge, and a survey was conducted on both occasions. DAs were provided to the intervention group, and brochures to the control group. The decisional conflict scale (DCS) and the control preference scale (CPS) were the primary and secondary endpoints, respectively. An unpaired t -test and z-test analyzed inter-group differences in DCS, and CPS, respectively. This trial was registered with the University Hospital Medical Information Network (UMIN Registration No.: UMIN00032623), certified as a test registration institution by the World Health Organization. Ninety-nine participants completed a full analysis set, which revealed that the intervention group had significantly more participants who had already decided on their discharge destination while they were admitted to the hospital. These were "the same place as before admission" in a significant number of cases. No significant inter-group differences were found in the DCS and CPS scores. DAs were effective at reducing uncertainty and controlling the decline in participation rates, especially in participants living alone who were unable to decide their discharge destination, and at clarifying the values of those aged 75 and older. The DA made it possible to increase available choices and explain the disadvantages regarding various locations of discharge destinations, allowing fewer internal conflicts in the decision-making process. Going forward, there is a need to further our understanding of methods of offering DA, the ideal duration of these interventions, and the identification of beneficiaries.


Abstract
In Japanese medical practice, older stroke survivors are bombarded with information 26 regarding their discharge locations, increasing their decision-making difficulties. This study used 27 a randomized controlled trial to evaluate the influence of using decision aids (DAs) matching the 28 values of older stroke patients and their families on the internal conflict and participation in 29 discharge destination decisions. 30 Participants were randomly allocated to intervention and control groups. The 31 intervention lasted for two months, from admission to discharge, and a survey was conducted on 32 both occasions. DAs were provided to the intervention group, and brochures to the control group. 33 The decisional conflict scale (DCS) and the control preference scale (CPS) were the primary and 34 secondary endpoints, respectively. An unpaired t-test and z-test analyzed inter-group differences 35 in DCS, and CPS, respectively. This trial was registered with the University Hospital Medical 36 Information Network (UMIN Registration No.: UMIN00032623), certified as a test registration 37 institution by the World Health Organization. 38 Ninety-nine participants completed a full analysis set, which revealed that the 39 intervention group had significantly more participants who had already decided on their discharge Introduction 51 Medical care for strokes has advanced, and its mortality rates have drastically declined. 52 However, age-related morbidity and recurrence rates of strokes remain high, with strokes being 53 the second most common condition, only after dementia, leading to patients requiring long-term 54 nursing care [1]. As a result, the roles of care personnel and the need for recovery rehabilitation 55 enabling patients to live independently have intensified. However, individuals who have suffered 56 a stroke experience such drastic changes in their lives that an internal conflict arises between their 57 past self (that is hard to let go of) and their shattered self-image [2]. Due to these reasons, it is 9 149 computer at a 1:1 ratio, and (b) an allocation table of patients according to their condition's 150 severity designed by the ward's head nurse. The severity of illness was determined by the lowest 151 total score of a daily living function assessment and the Functional Independence Measure (FIM). 152 According to the facility criteria prescribed by the government, the severely ill are those who have 153 a daily living function assessment of 10 points or more, or a total FIM score of 55 points or less. 154 Until the allocation to the groups was completed, the order of allocation was concealed from 155 Research Assistant A, the ward's head nurse, the patients, their families, and multidisciplinary 156 professionals (as part of the "allocation concealment mechanism").
The flow of selecting participants 159 From October 2018 to May 2020, we invited 135 individuals who had met the eligibility 160 criteria to take part in the trial. After excluding those who had declined to take part (n = 28), we 161 randomly allocated 107 individuals to the intervention or the CGs. Further, eight individuals were 162 excluded with whom, in the course of follow-up, no questionnaire survey could be carried out. 163 Finally, a total of 99 people, comprising 51 in the IG and 48 in the CG, constituted the full analysis 164 set who were to undergo analysis (Fig 1).
. CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint Fig 1. Flow diagram of the CONSORT study. The figure describes the stages of the CONSORT 166 study, beginning with the enrollment of the participants, followed by their allocation into 167 intervention and CGs, their follow-up, and finally the analysis of the two groups.

169
Although we planned to enroll 122 patients, we were not allowed to enter the hospital is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint to that of the DA. The participants were explained that the brochure had been provided for their 212 reference while deciding their discharge destinations and that it contained the same methods of 213 usage and points of special note as those provided to the IG. 214 We held two meetings with the multidisciplinary professionals to explain the purpose, 215 significance, and method of research. We explained that we could not reveal the contents of the 216 DA or the brochure, or the allocation of patients between the two groups. We also informed them 217 that they may respond to the queries of the older stroke patients and their families but should 218 avoid providing instructions regarding the content of DA or about utilizing the tool. Furthermore, . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint Likert scale. All the DCS items are totaled, divided by 16, and multiplied by 25 to arrive at the 238 total score. The total score is converted into a score ranging from 0 to 100 points, with a high 239 score indicating a high decision-making internal conflict level. A score below 25 points indicates 240 implementation of decision-making, and a score of 37.5 points or higher indicates a delay in 241 decision-making and a feeling of uncertainty about its implementation [27].

242
The secondary endpoint pertained to participation in decision-making and was evaluated 243 using one control preference scale (CPS) item. The CPS was developed by Strull et al. [28] and 244 modified by Degner et al. [29]. Its reliability has been confirmed (Coombs' criterion of 50%).

245
The reliability of the Japanese edition of CPS has also been confirmed. The test-retest reliability were also calculated, using a 10-point Visual Analog Scale.

252
Besides these, we asked for the following details regarding the participants' attributes, 253 on admission: sex, age, disease name, family makeup, the desired and the ultimate discharge is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint destination, educational background, work history, duration of hospitalization, the status of 255 readiness for decision-making, and the person(s) with whom a decision had been made. Method of analysis 266 After checking the input data independently by two Research Assistants B, the primary 267 investigator, who was not blind to the allocation process, handled the data. To retain the random 268 allocation, we made all randomized data the targets of analysis following interventions that had 269 been initially allocated (intention-to-treat). All the participants' characteristics at the baseline 270 underwent descriptive statistical testing, a t-test, a χ 2 -test, and a Mann-Whitney's test. The 271 internal conflict over decision-making, which is the primary endpoint, was subjected to an . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint Table 1

300
Of the 16 participants who were discharged to a place that was different from before, 6 (37.5%) 301 were living alone. As to where the participants wanted to decide their discharge destination, the 302 largest number of the participants wanted to do so "With their family," followed by "With family 303 and healthcare professionals." 304 We found that the IG contained significantly more participants who had already decided 305 their discharge destination than the CG (p < 0.05) ( Table 1). It was also found that significantly . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022.   319 c) Educational background: Regarding "Graduated from schools below high school" and "Graduated 320 from schools above high school," "Graduated from schools below high school" includes graduation 321 from elementary school in the old educational system, graduation from a girls' school in the old 322 educational system, and graduation from middle school. "Graduated from schools higher than high . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint school" includes graduation from high school, vocational college, junior college, university and 324 graduate school.
325 d) Status of readiness for decision-making: "Have already decided where to go to after discharge" and "Have not decided where to go to after discharge" includes "Cannot even begin to think about a 327 discharge destination," "Not even considering discharge destinations other than the one desired," "Also 328 considering discharge destinations other than the one desired," and "On the verge of deciding where 329 to go to after discharge." . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint performed for "With whom the subjects wanted to decide their discharge destination"; and a z-test was performed for testing inter-group 337 differences. 338 a) "With whom the subjects want to decide their discharge destination" was studied with three items: "By themselves," "With family," and "With 339 family and healthcare professionals." Primary endpoint 341 In terms of conflicts over decision-making (measured by the DCS), both the intervention 342 and CGs had intense internal conflict over "Support," "Sufficient explanation of information," 343 and "Clarification of values." The intense state of internal conflict continued even after the 344 hospital discharge. On the contrary, the level of internal conflict over "Effective decision-making" 345 was the lowest, during admission and discharge. No significant inter-group differences were seen 346 in terms of the extent of change in DCS scores between admission and discharge (Table 3). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  In terms of participation in decision-making, as measured by the CPS, both the intervention and CGs gave the highest scores for "Active roles." However, no significant 394 differences were seen between the groups (Table 7). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022.

426
available to them. Schkade and Kahneman [34] showed the tendency to use only a part of the 427 information that may be used and underestimated information to which they do not direct their is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint were based solely on self-reporting by older stroke patients. Hence, we feel that they have not 481 been able to appropriately grasp whether or not they had actually participated in decision-making.

482
As cultural characteristics of decision-making among the Japanese, Kawai  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint 514 and standardize the findings and data and to further understand the period and method of offering 515 DA as well as the selection and content of target individuals. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint systematic development process for patient decision aids. BMC Med Inform Decis Mak. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint listen to me": A cross-sectional study of experiences of seniors and their caregivers . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 3, 2022. ; https://doi.org/10.1101/2022.07.28.22277637 doi: medRxiv preprint