WF-03 — Course Specification and Industry Co-Development

This is the clearest case in the whole pilot of real activity that goes uncounted. The OBEF approved partner-type list explicitly includes healthcare institutions and professional associations. IoL sits inside an academic health system, and a substantial share of its teaching is already designed and delivered with clinical colleagues from Dubai Health entities and with professional bodies. Almost none of it counts, because counting requires a named contributor, a timetable entry and a recorded contact-hour share, and those are administrative acts rather than educational ones.

1. Purpose

To ensure that every IoL course has an approved, current specification in which the contribution of external partners to its design and its delivery is named, quantified in contact hours, scheduled and evidenced at the time it happens, so that teaching genuinely co-created with the health system is recognised as such by the Ministry, and so that a course claimed as jointly developed or jointly delivered can be defended line by line under an evidence request.

2. Scope

Scope statement. This process manages the specification of an IoL course from the decision to create or revise it, through the recording of partner contribution to design and delivery, to the annual classification and aggregation of joint-industry credit hours for submission.

Applies to. Every course carrying a Course Ref Number in the PGDipHPE and MScHPE study plans, including practicum and work-based components that appear in the study plan as courses. Covers both the development side (who shaped the specification, the assessment and the learning objectives) and the delivery side (who taught, when, for how many hours, and under whose name in the syllabus).

Does not apply to. Creating or removing a course from the study plan, which is WF-01. Learning-outcome drafting and mapping, which is WF-02. Assessment instrument design and moderation, which is WF-09. The partnership agreement itself, which is WF-24. Placement episode registration, which is WF-13; this workflow only receives the contact-hour record WF-13 produces.

Applicable requirements. OBEF guide v11.5 KPI 3.3 (qualification criteria for developed and delivered in partnership, and the exclusion rule); KPI 2.1 rubric; CAA course requirements; HEDB Data Dictionary 2026 definitions for Courses.xlsx and Course Faculty.xlsx; MBRU data protection obligations for external contributor contact data.

The eight criteria that govern everything below. KPI 3.3 recognises two routes, and each has four criteria, all of which must be met.

Developed in partnership, all four required:

1. At least 20% of course content by contact hours developed by industry partners, and never fewer than 10 contact hours in absolute terms.
2. Documented formal input: signed minutes, feedback forms, participation in producing the course specification, the assessment or the learning objectives, or recorded correspondence incorporated into the design.
3. Input occurred during the course-planning phase, not after delivery.
4. Contributors are named individuals or named companies.

Delivered in partnership, all four required:

1. At least 20% of contact hours taught directly by external industry partners, with a minimum of 10 hours.
2. Sessions linked to learning outcomes and/or assessment.
3. Partners named in the syllabus and effectively responsible for leading, hosting or delivering.
4. Delivery is on the formal timetable or documented in the LMS. Ad-hoc or optional activities such as open question and answer sessions explicitly do not count.

3. Trigger, boundary and endpoint

4. SIPOC

5. Accountability

Process owner. IoL Director of Programmes, or the role able to require a course lead to name a contributor in the syllabus and to timetable a partner-led session.

Escalation.

6. Process steps

1. Open or reopen the course specification. Record Course code, Course Ref Number, Credit hours, Contact Hours, Course type, Course level, the academic period and the owning Program Code. [CAPTURE] the course inventory record. Serves 3.3 numerator and denominator. Course Ref Number must be unique per offering, so a course taught twice in a year is not double counted.

2. Identify intended partner contribution during planning. Decide, before the course is designed, who from outside MBRU will shape it and who will teach on it. This is the step that makes the difference between a course that qualifies and an identical course that does not, because criterion 3 on the development side requires that input occurred during the course-planning phase.

3. Confirm the partner type. Route to WF-24. The approved partner-type list includes healthcare institutions and professional associations, which is the clause that matters here: clinical colleagues employed by Dubai Health entities, and officers of professional bodies, are partners of an approved type. Confirm and record the type; do not assume it. [CAPTURE] partner identity and type.

4. Record design input as it happens. Signed minutes of a design meeting, a completed feedback form on a draft specification, a recorded email chain whose content was incorporated, or documented participation in writing the learning objectives or the assessment. [CAPTURE] the dated design-input artefact and the share of content by contact hours it shaped. Serves 3.3 developed-in-partnership criteria 2 and 3. Exception route. Input received after delivery has begun does not meet criterion 3. Record it as feedback for the next revision, and do not count it this cycle.

5. Name the contributors in the syllabus. Named individuals or named companies, published in the syllabus the learners see. [CAPTURE] contributor names and organisations. Serves 3.3 development criterion 4 and delivery criterion 3. Decision point. A contributor who cannot or will not be named cannot support a claim. Resolve the naming before delivery, not after.

6. Timetable the partner-led sessions, or document them in the LMS. [CAPTURE] the timetable or LMS record with dates and durations. Serves 3.3 delivery criterion 4. This is where most genuine co-delivery is lost. A clinical expert who teaches a scheduled, assessed, syllabus-named session qualifies. The same expert delivering the same content as an invitation-only optional discussion does not, because the guide explicitly excludes ad-hoc and optional activities such as open question and answer sessions.

7. Link partner-led sessions to learning outcomes and assessment. Take the CLO references from WF-02. [CAPTURE] the session-to-CLO link. Serves 3.3 delivery criterion 2 and contributes to 2.1.

8. Deliver, then record actual contact hours. Actual, not planned. [CAPTURE] delivered contact hours per contributor, and Number of Industry-Led Sessions. Serves 3.3 delivery criterion 1.

9. Run both threshold tests. Every claim is tested twice.

   • Share test: Percentage of Co-developed course or Percentage of Co-delivered course is at or above 20.
   • Absolute floor test: Contact Hours × Percentage ≥ 10. Both must pass. Running only the share test is the most likely arithmetic failure in this KPI. Twenty per cent of a 40-hour course is 8 hours, which clears the share test and fails the 10-hour floor. A course must have at least 50 contact hours before a bare 20% share clears the floor. [CAPTURE] the test result per course, pass or fail, with the numbers.

10. Set the flags and percentages. [CAPTURE] Co-developed course, Percentage of Co-developed course, Co-delivered course, Percentage of Co-delivered course. A flag set without both tests passing is a misstatement, not an optimism.

11. Record the industry faculty. [CAPTURE] Faculty type, Name of the industry, Email of the industrial faculty, Percentage of teaching workload, Percentage of course design workload. Serves 3.3 and the Appendix B evidence request for industry partner contact data.

12. Assemble the evidence pack per claimed course. Design-input artefacts with dates, the syllabus naming the contributor, the timetable or LMS extract, the contact-hour record, and the partnership document from WF-24. A claim without this pack should not be made.

13. Apply the exclusion rule. Remove general education, mathematics and basic sciences courses from both the numerator and the denominator, using a version-controlled classification of Course type and Course level. [CAPTURE] the exclusion list version applied. Applying it to the denominator only, or to neither, changes the result substantially. [IoL to confirm whether either programme contains any course that falls within the excluded categories. For a postgraduate health professions education programme the expectation is that few or none do, which would make the exclusion arithmetically immaterial but still mandatory to document.]

14. Aggregate by credit hours, not course count. The KPI is a percentage of credit hours. [CAPTURE] the annual aggregate with its three-year rolling context.

7. OBEF data generated

Capture rule. Four things cannot be reconstructed after the course has run, and every one of them is a criterion the KPI actually tests: the date and content of the design input (criterion 3 is about when it happened), the contributor's name as published in the syllabus (a name added retrospectively is not what learners saw), the timetable or LMS entry (a session cannot be scheduled in the past), and the actual delivered contact hours. A course can be genuinely co-delivered and permanently unclaimable because none of these four was written down at the time.

Reproducibility test. A second analyst can reproduce KPI 3.3 from Courses.xlsx joined to Course Faculty.xlsx, filtered by the exclusion list and aggregated by credit hours, provided the flags and percentages are populated and correct. Today, on the assumption stated below, they are not, and the analyst would produce a value close to zero for programmes that in substance have substantial partner co-delivery. That is the whole problem this workflow exists to fix.

Assumption, stated as an assumption. This document assumes that Co-developed course, Co-delivered course and their percentage columns are currently unpopulated or default for IoL courses, and that clinical contributors are not systematically named in syllabi with recorded contact hours. [IoL to confirm.] If some courses are already recorded this way, the workflow's task is extension rather than construction.

8. Max-score design

The anchor structure is worth reading carefully. The gap between the 75-point anchor (35%) and the 100-point anchor (100%) is enormous, while the gap between 50 points (20%) and 75 points (35%) is 15 percentage points. The efficient target is 35 to 50%, not 100%. Pushing from 35% to 60% buys far less than moving from 15% to 35%, and chasing 100% would require nearly every course to be co-delivered, which is neither achievable nor educationally sensible.

What this workflow must do to reach the target.

1. Name contributors in the syllabus. This single administrative act converts existing teaching into countable teaching, and it costs a line of text.
2. Timetable partner-led sessions. Move genuine partner teaching from optional and ad-hoc formats, which the guide explicitly excludes, onto the formal timetable or into the LMS, which it accepts.
3. Record actual contact hours per contributor, so the share can be computed rather than estimated.
4. Run both threshold tests on every claim, and record the numbers, not just the verdict.
5. Capture design input at the planning phase with a date, since criterion 3 is a timing test and no later artefact can satisfy it.
6. Aggregate by credit hours and apply the exclusion identically on both sides.

[REDESIGN] actions.

A word on R5, because it is easy to over-read. The clause accepts courses taught by faculty with a current documented professional affiliation with an industrial or economic sector in the same field, provided the course meets the minimum requirements. It does not convert all clinically qualified faculty into industry partners automatically. The affiliation must be current, documented, and in the same field, and the 20% and 10-hour tests still apply. Treat it as a real but bounded route, evidence it through WF-08, and expect it to be scrutinised.

9. Indirect strategy where data cannot be collected

Category 4 applies, and this is the clearest case of it anywhere in the pilot: the activity exists but is invisible. IoL teaches inside an academic health system. Clinical colleagues from Dubai Health entities contribute to course design and teach on courses. Professional associations shape content. Both are approved partner types under the OBEF list. The activity is real, current and substantial. It counts for nothing, because counting requires four administrative artefacts that nobody has been asked to produce: a name in the syllabus, a timetable entry, a recorded contact-hour share, and a dated planning-phase artefact.

The response to Category 4 is capture, not tactics. Nothing in this section proposes a manoeuvre. Every action in section 8 is an instruction to write down something that is already true. That is the entire strategy, and it is why the modelled gain from this workflow is described in the Score Maximisation Plan as registration rather than improvement.

Why this KPI is under-reported specifically, and not by accident. The four delivery criteria are all administrative, and the administrative acts sit with different people. The course lead names the contributor, timetabling schedules the session, the contributor delivers, and the data steward sets a flag in a spreadsheet months later with no way to check any of it. The workflow's job is to move all four acts to the point of occurrence and to give them one owner per course.

A note on the temptation this KPI creates. Because the anchors reward reaching 35% and because the share is self-computed, KPI 3.3 offers more room for optimistic classification than most Track A indicators. Two specific temptations are worth naming. The first is counting an open question and answer session or an optional guest talk, which the guide excludes by name. The second is recording planned rather than actual contact hours, which passes every internal check and fails the moment the timetable is compared with the claim. Appendix B permits the Ministry to demand partnership documents and industry partner contact data, and a named contributor can be contacted.

A genuine ambiguity in the guide, flagged rather than resolved. The developed-in-partnership criterion sets a threshold of "at least 20% of course content by contact hours developed by industry partners", but it does not define how the contact-hour share of development is measured. Contact hours are a delivery unit. The natural reading is the proportion of the course's contact hours whose content the partner shaped, but a second reading, that it means the partner's own hours spent developing, is available and gives a different number. The defensible position is the first reading, applied consistently, with the basis stated in the Mechanism of collecting the data field for 3.3. Record the choice, hold it constant across the three years of the rolling average, and put the question to MoHESR.

A second ambiguity worth flagging. The guide permits work placements that are a formal part of the study plan to be counted as delivered in partnership, provided the course is genuinely delivered inside and by the employing organisations. It does not say whether the placement's contact hours enter the KPI 3.3 denominator as a course in their own right. If they do, a large placement both raises the numerator and enlarges the denominator, and the net effect depends on the rest of the study plan. Compute it both ways before claiming R4, state the basis, and hold it constant.

Boundary check. This workflow must never:

• count an ad-hoc, optional or open question and answer session, which the guide excludes by name;
• record planned contact hours as actual hours in order to clear the 20% share or the 10-hour floor;
• set Co-developed course or Co-delivered course where either threshold test fails, or where only the share test was run;
• claim a contributor who is not named in the syllabus the learners actually received;
• create a design-input artefact after delivery and present it as planning-phase input, which fails criterion 3 on its face;
• apply the general education, mathematics and basic sciences exclusion to the denominator only;
• aggregate by course count when the KPI is defined on credit hours;
• claim a partner type that is not on the approved list, or assume a partner type without confirming it through WF-24.

10. Service standards

11. Records and evidence

Appendix B readiness. IoL could most likely produce partnership agreements within 15 working days. It could probably not produce, for any specific claimed course, the dated planning-phase input artefact, the timetable extract and the contact-hour record as a matched set, because on the stated assumption these are not currently assembled per course. That is the exact evidence a KPI 3.3 request would ask for, and it is the reason a claim should not be made without the pack at step 12.

12. Risks and controls

13. Performance measures

The measure worth watching is the last outcome row. Near-misses are the characteristic failure of this KPI: a course with 15% partner delivery is doing almost exactly the same thing as a course with 21% and scores nothing for it. Identifying them annually and redesigning a handful is more productive than any general exhortation to involve partners more.

14. Change control