Currently, surgical training is often conducted using the traditional "apprentice model", where a trainee observes a qualified surgeon and learns from him/her, and then the surgeon supervises the trainee performing surgery on a patient. The investigators believe that this conventional model has substantial limitations and drawbacks, making surgical training less efficient and less safe.

The investigators will test the hypothesis that intense virtual reality (VR) simulation-based ophthalmic surgical training improves initial acquisition of competence in key stages of manual small incision cataract surgery (MSICS). To do this, the investigators are proposing a randomized multi-country study. This mixed-method study will combine qualitative and quantitative data collection.

Orbis International partnered with FundamentalVR to create a manual small incision cataract surgical (MSICS) simulator, using virtual reality software combined with existing gaming technology. The result is a VR simulator available at a fraction of the cost of products currently on the market. This VR simulator will be the subject of this study.

All training within the 'educational intervention' of this study will be performed using simulation. There is no testing or surgical training on patients.

Study Design:

Prospective, investigator-masked education-intervention randomized controlled study of intensive virtual reality (VR) simulation-based surgical education of ophthalmologists in China, Ethiopia, India, Mongolia, Bangladesh, UK and USA.

Construct validity study of assessment scores generated by the VR simulator for novices versus experts.

Qualitative study of face validity of VR simulator, and acceptability questionnaire survey of users.

Purposes of study:

To investigate the efficacy of intensive VR simulation-based surgical education using the Orbis-FVR simulator. To examine whether it improves competence, is acceptable and has validity. To assess the construct validity of the VR simulator's assessment capacity.

Design Summary This research program involves a prospective, investigator-masked education-intervention randomized controlled study of intensive virtual reality (VR) simulation-based surgical education of ophthalmologists in China, Ethiopia, India, Mongolia, Bangladesh, UK and USA. Participants will be randomized, and surgical competency will be assessed at baseline and 1-month between groups. Competency will be assessed at baseline, using artificial eyes. Both groups will perform three full MSICS cases using model eyes at baseline. At one-month post intervention, both groups will again perform three full MSICS surgeries using model eyes for grading by masked evaluators using an existing, validated rubric (main outcome).

The study will also assess the construct validity of assessment scores generated by the VR simulator for novices versus experts. Both novices and experts will perform three full VR cases using the simulator. The first score will be disregarded to allow for familiarization with the hardware, and the mean of the second two cases will be recorded.

There will also be a qualitative study of face validity of the VR simulator, and an acceptability questionnaire-based survey of users. An anonymous survey will be administered to experts and novices, using a Likert scale.

Study Setting This is a multi-center study. The investigators will enroll trainee ophthalmologists (doctors who have graduated from medical school and are currently undergoing ophthalmology specialist training) at 9 ophthalmology training institutions in in China, Ethiopia, India, Mongolia, Bangladesh, the UK and USA.

Collaborating institutions:

London School of Hygiene and Tropical Medicine, UK Chittagong Eye Infirmary and Training Complex, Chittagong, Bangladesh Addis Ababa University, department of Ophthalmology, Menelik II Hospital, Addis Ababa, Ethiopia University of Gondar, Gondar, Ethiopia Dr. Shroff's Charity Eye Hospital (SCEH), New Delhi, India H.V. Desai Eye Hospital, Pune, India The First Central Hospital,, Ulaanbaatar, Mongolia Shenyang He Eye Specialist Hospital, China. Emory Eye Center, Atlanta, GA USA

Study Duration The training will be conducted during 2021, and will last 6 months.

Study Participants Current trainees enrolled between January and December of 2021 in all 9 training institutions will be selected according to the inclusion and exclusion criteria.

Informed Consent Potential participant trainees will be informed of the training opportunity and the study. Heads of Department will be involved in the process. Trainee participants will be informed in detail about the nature of the education-intervention study, that the training offered provides no official qualification and will not be recorded in their annual clinical training evaluation. All surgeons participating will be free to leave the study at any time.

Withdrawal Criteria Trainee participants are free to leave the study at any time. If this is the case for any participant, no effort will be made to recover any costs incurred. Data collected up to the point of withdrawal of consent will have been pseudo-anonymized and securely stored and will still be held and included in data analysis

Baseline Data Resident Trainees

• Age
• Sex
• Year of training
• Training Institution
• Number of MSICS surgeries performed as primary surgeon
• Number of MSICS surgeries assisted in or performed in part
• Total number of other ophthalmic surgeries performed (disaggregated by type)

Attending Consultants (experts)

• Sex
• Training institution
• Number of MSICS surgeries performed as primary surgeon

Randomisation Sequence generation The randomization sequences will be computer generated and administered centrally by a statistician based at the London School of Hygiene and Tropical Medicine, who is independent of all other aspects of the trial. The investigators will use block randomization (block size 2 or 4), with a separate sequence for each of the ten recruitment sites, to ensure balance. The statistician will generate the code / sequence (as a block of 2 or 4).

Allocation Concealment The statistician will not have access to information about subsequent allocation, and the individual potential participants. The PI, co-investigators, and participants will have no prior access to the random sequence.

Randomization Implementation Trainees within the same institution, who have met the appropriate inclusion and exclusion (as detailed previously), will be eligible for randomization to the 'intervention' or 'control' arm. Each group of four trainee participants will be agreed by the Head of Training Program or Department.

For example (Blocks will be in groups of 4, 6 or 8 depending on the institution and the number of residents that meet the inclusion criteria):

A block of four potential participants is identified at Chittagong Eye Institute Training Center (CEITC), Bangladesh. Cards with the allocation or a block of four (two intervention and two control) will be printed and placed in sealed opaque envelopes. Physically, in Bangladesh, a block of four identical envelopes (e.g. block number 11) will be selected. Participants will be invited by the Head of Department to pick one of the four envelopes. In this example, CEITC randomization block 11 allocation might be:

IRO1101 Intervention IRO1102 Control IRO1103 Control IRO1104 Intervention

Training Intervention - Orbis-FVR Training Program

Pre-Learning (one month in advance)

• Surgical video review ***Stipulated number or total time spent?***

Pre-Course Assignment (one week in advance)

• Complete and record three full MSICS cases using the Philips MSICS model eye and following the steps of the OSSCAR (Appendix 4).
• Upload the 3 recorded simulated surgical procedures to the Cybersight platform

The Orbis-FVR Virtual Reality MSICS simulator has 19 steps in total. Currently, 10 steps are animated and 9 are interactive, as follows:

1. Patient Preparation - Animated Step
2. Conjunctival Peritomy - Animated Step
3. Paracentesis - Animated Step
4. Formation of the AC - Animated Step
5. Anterior Chamber Maintainer - Animated Step
6. Marking the Scleral Incision - Interactive Step
7. Scleral Incision - Interaction Step
8. Sclerocorneal tunnel - Interactive Step
9. AC Entry - Interactive Step
10. CCC - Interactive Step
11. Hydrodissection - Animated Step
12. Mobilization of the nucleus - Interactive Step
13. Releasing the nucleus - Animated Step
14. Nucleus removal - Interactive Step
15. Cortex removal - Animated Step
16. IOL insertion - Interactive Step
17. IOL Dialling and Implantation - Interactive Step
18. Inspection of the wound - Animated Step
19. Closure of the wound - Animated Step

Course Overview Participants must attend and successfully complete all sessions. All simulator tasks are to be completed in sequential order, in line with the course outline. A required score of 85 points for each interactive step must be reached three times in a row to advance to the next step in the course.

For each step within the simulation the following is recorded:

Step Time - The total time spent on a step including ALL attempts. Pass / Fail - The interpreted pass or fail result of the last attempt of the step.

Step score - Each step is scored between 0 and 100 points, based on user performance in completing the surgical objective, taking into account any adverse events.

Adverse events - For each step, if triggered, all of the adverse events are recorded.

Number of attempts - The number of attempts (including resets) of the step.

Course Content Each course will combine autonomous learning and live mentorship.

Autonomous learning will consist of a series of 15-minute video presentations, which include:

• Narrated demonstration video of specific VR step(s)
• PowerPoint presentations
• Instrument use demonstrations

Live mentorship will be delivered in two ways:

• Local mentor who provides in-person mentorship
• International mentor from technical advisory group provides live tele-mentorship

Content hosting All content will be hosted on the Orbis Cybersight Learning Management System platform. Data will be randomized and anonymized by the Cybersight Program manager, who will not be involved in any other aspect of the research initiative.

Course delivery

Each day:

Local mentor will facilitate introductions in each morning session. This will cover:

• An overview of the day's training plan
• Outlining the learning objectives for the day's training plan
• Introduction of the content covered in the day's training plan Local mentor will also facilitate discussion and answer questions on autonomous content viewed in the morning and midday sessions Local mentor will provide periodic check ins throughout the day's session International mentor will provide tele-mentorship during the afternoon sessions.

Trainees will do self-practice for the remainder of the afternoon session

The MSICS simulator has a built in surgical assessment per step. Each interactive step has a possible score between 0 and 100. A score of 85 indicates "passing." Each trainee will complete each interactive step on the VR simulator a minimum of 5 times with a reliability gate (3) and minimum score of 85. Reliability gate of 3 was chosen based on similar interventions using like VR cataract simulators. Each trainee will complete at least 5 full cases on the VR simulator with a reliability gate (3) and minimum score of 85

Attendance to be recorded as follows:

• Confirmation of three video uploads pre-intervention
• Simulator use and assessment logs
• Attendance records taken per daily session
• Confirmation of three video uploads post-intervention

Oversight and validation of the above will be done by the local mentor, as well as study personnel. The simulator itself will log that participants completed each step, with a detail of their performance, length of time spent, adverse events, score and final mark of pass/fail.

Risks and Limitations

There are no clinical risks within this study, as all the intervention training uses simulation. No patients are involved in any of the training. Patients are involved only as part of fully supervised, standardized, regulated and accredited post-graduate clinical and surgical training within the collaborating training institutions, all of which falls outside the study.

There are a number of broad risks in conducting this study.

• Trainees, meeting the inclusion criteria, not being available for enrolment (due to examinations, closure of training institutions, personal reasons).
• Connectivity issues undermining access to the VR software, orientation and training
• New technology: user comfortability with engaging in new technology
• Civil unrest (including national elections).
• Customs delays in the import of equipment or consumables.

There are expected limitations to this study:

• Some of the study locations might have opportunities for alternative cataract surgical training, including ECCE or phacoemulsification.

VR Training Efficacy

Inclusion Criteria:

• Trainee ophthalmologist in collaborating institution.
• Agree to undertake simulation procedure assessments
• Agree to undertake and complete the intensive VR simulation training course.
• Performed 0 MSICS as primary surgeon and assisted or part-performed less than 10 MSICS cases

Exclusion Criteria:

• Trainee who performed 1 or more MSICS cataract surgeries as primary surgeon and assisted or part-performed 10 or more

Construct Validity

Inclusion Criteria

• The first cohort is novice MSICS surgeons having performed 0 MSICS procedures as primary surgeon and assisted in less than 10.
• The second cohort is proficient MSICS surgeons having performed a minimum of 1000 MSICS procedures

Exclusion Criteria

• MSICS surgeons having performed 1 or more MSICS cataract surgeries as primary surgeon and assisted or part-performed 10 or more; or MSICS surgeons who have performed less than 1000 MSICS procedures as primary surgeon.

Video recordings of procedures will be allocated a random 7-digit number, and subsequently stored onto an encrypted computer, and a separate encrypted hard drive. This random number will be the only identifiable information available when the simulation/surgical procedure is assessed, thus masking the assessor to the participant's intervention/control arm.

Recordings will be converted to an MP4 format, and coded. At CyberSight, the recording will be renamed as a randomly generated seven-digit number (e.g. 6253815). The code sheet will be generated by an independent statistician and only be known to them and the CyberSight administrator. Once assessors are notified that the video is ready for marking, this random number will be the only identifiable information available when the simulation/surgical procedure is assessed, thus completely masking the assessor to the participant's intervention/control arm and personal identity.