Digital Gaming for Persons With Traumatic Brain Injury

This is a feasibility study using a pragmatic, randomized controlled trial with three arms.

The study will be conducted within the Turku University Hospital, Division of Clinical Neurosciences (Operational Division of Clinical Neurosciences, Department of Rehabilitation and Brain Trauma in collaboration with former neurology outpatient clinic and rehabilitation units of the university hospital). The Division is specialized in demanding neurological and psychiatric inpatient rehabilitation, inter-professional neurological education, rehabilitative examinations, and the inpatient and outpatient care of traumatic brain injuries. In each year, about 500 patients are admitted in the Division services units.

The recruitment will be focused on patients who have been discharged from the Turku University Hospital, Division of Clinical Neurosciences at least one year before the starting day of the recruitment process. Based on the previous literature, using games in longer interventions will be facilitated if training is provided in the persons' home, due to easier access and reduced impact on school or work activities (Burbea et al. 2011).

The randomisation and patient allocation will be fully centralized; we will use a central randomization service by the University of Turku. The randomization will be targeted on patients who have been hospitalized in Turku University Hospital and who are registered in the Hospital electronic medical records. The study will be individually randomized. Allocation will be computer generated. The participants will be randomly assigned in numbers via computerized assignment developed by the independent Trial Statistician and implemented by the Research Assistant who are trained for patient randomization and data collection.

Participants and the investigators enrolling the patient recruitment and randomization will not foresee assignment. Allocation will be masked to participants in two intervention groups; participants will be aware about the gaming intervention, but not which one is the experimental and which is the comparator. Investigators will not be masked to group allocation, while data analysts (outcome assessor) will be kept blinded to the allocation. As far as we are aware, there will be no contact between participants in different groups; participants are living in wide geographic area in the catchment area of Southwest Hospital District.

To capture a representative sample of patients, the Turku University Hospital electronic medical records will be used. The medical records will be accessed by the authority of the Chief medical doctor of the Hospital and screened by the Research Assistant. All patients with TBI diagnosis will be screened to find out patients who fulfill the eligibility criteria for the study participation.

Those patients who fulfill the inclusion criteria will be contacted with the telephone call by the Research Assistant. A preliminary description of the study will be offered and the inclusion criteria for the study will be described to them. The participants are able to answer whether they fulfill the specific inclusion criteria and their preliminary interest toward the study will be asked.

Those who fulfill the inclusion criteria and are willing to participate in the study, will receive postal information of the study, two items of informed consent forms, questionnaires to be filled at home (baseline data), and a short description of eight entertainment games to be used (if allocated into the control group). Further, possible participants are aware that after 1-2 weeks they will receive another call from the Research Assistant. This will ensure that possible participants have enough time to read all material, ask questions, and make their decision whether to participate in the study or not (and what type of game they would like to play if allocated into the control group).

During the second telephone call, those who are willing to participate in the study will get information about practical arrangement of the study and the place of the data collection. They are also informed that all travel costs will be covered and that the time for face-to-face meeting may take totally 1-1.5 hours. Participants are also aware that they are free to withdraw from the study at any time. At the end of the telephone call, a specific time for the face-to-face meeting with the Research Assistant will be arranged.

During the face-to-face meeting at the University of Turku, it will be ensured that participants will fulfill the inclusion criteria. If a patient fulfills the criteria, a signed written informed consent will be returned/obtained. Should a person consent, their pre-filled baseline data will be gathered. The Research Assistant will receive this information by email, text message or telephone call. Cognitive measurement will be also conducted by the psychologist. After that eligible patients will be informed to Trial Manager who allocates patients in one of the three arms of the trial based on computer generated randomization list (intervention vs. control group vs. "As-usual"). Recruitment will continue until the required data has been received.

Criteria for the feasibility:

• Willingness to participate 80 % (refusal rate 20 %)
• We will expect that more than 75% of the prescheduled measurements would be performed (adherence)
• Less than 25 % of participants would drop out due to any reason (adherence)
• The acceptability of the game use would be at least 95% (in intervention and control groups) (adherence)
• Usability evaluation for the gaming system would be at least 80% (1 item; dichotomous scale yes/no) (usability)
• More than 80% will be satisfied with the games (1 item; dichotomous scale, yes/no) (satisfaction)
• We will expect that 60% are willing to use the games later as part of their recovery process (use in the future)

The calculations for sample size needed in each group are preliminary estimations to guide our data collection. This has been performed with two outcome variables used in this study: a) TMT version A (Lezak et al. 2004) and b) depression (PHQ-9, Kroenke et al. 2001).

1. Let's assume that the average level of the TMT version A score is about 71 and the mean change in the scores during the follow-up is 30. Let's assume that the standard deviation of the TMT scores is 53. This difference could be expected to be significant (with power 85% on significant level 0.05) if the sample size in each group is 30 subjects.
2. Let's assume that the average level of the PHQ-9 score is about 10 and the mean change in the scores during the follow-up is 3. Let's assume that the standard deviation of PHQ-9 scores is 5. This difference could be expected to be significant (with power 85% on significant level 0.05) if the sample size in each group is 27 subjects. Conclusion: Based on these preliminary power calculations, sample size to be used in this study (30/a group) will be quite sufficient to detect significant within-group changes between baseline and follow-up in the outcome measures. However, this means that the attrition rate of the study should be near 0 %. The outcomes of this feasibility study can be used as a guide to estimate a sample size for our future study.

The cumulative monitoring data during the 8-week period regarding gaming for the total number of received parameters (frequency, timing, time) will be calculated. Information available for each patient will also be calculated (participation/refusal rate, measurement instrument filled, drop-outs, acceptability, usability, satisfaction, willingness) and compared to the highest possible representative numbers (100 % of participation rate; measurement instrument filled, drop-out rate 0 %, acceptability 100 %, usability 100%, satisfaction 100%, willingness to use games in the future 100 %).

For the secondary outcomes, all the participants will be analysed at the baseline, after 8 weeks when the intervention has been finalised, and 2 months, on all the scales as described earlier. Paired t-tests or Chi square tests will be performed to detect significant differences before and following treatment in one group. Significant differences between groups will be evaluated applying the unpaired Student's t-test using the conventional 95 % level of confidence. In all tests, p-values <0.05 were considered significant.