Skip to main content
Saudi Gastroenterology Association logo

Saudi Gastroenterology Association · Open course

Listen

A short spoken briefing for each module, read aloud by your browser’s built-in voice. The full text is written out below each one, so you can read instead.

  1. 1

    Health systems through a GI lens

    The WHO building blocks, UHC, and how an IBD or hepatitis service depends on all of them

    Read the script

    Module 1. Health systems through a GI lens. The WHO building blocks, UHC, and how an IBD or hepatitis service depends on all of them.

    Six building blocks, four goals. The WHO framework describes a health system by six building blocks: service delivery, health workforce, health information systems, access to essential medical products and technologies, financing, and leadership and governance. Together they pursue four goals: improved health, responsiveness to what people expect, financial protection and efficiency. The value of the framework is that it makes you look beyond the clinic. A service is a bundle of people, information, supplies, money and rules, and each strand has to hold. A weakness in one block caps what the others can achieve, which is why single-programme fixes, such as adding a drug to a formulary, so often disappoint. For a clinician-leader the practical use is a checklist: before backing any proposal, ask which blocks it touches and which it quietly assumes will look after themselves.

    Universal health coverage. UHC has three dimensions, often drawn as a cube: who is covered (population), which services are covered (benefit package), and how much of the cost is covered (financial protection). The aim is to fill the cube without letting any face collapse. Moving along one dimension often tests the others. Expanding the benefit package without more money or capacity lengthens waiting times or raises out-of-pocket payments. Extending coverage to more people without new services creates entitlement on paper only. Policy choices about which services to include are, in practice, choices about trade-offs between these dimensions, and they should be made openly with explicit criteria rather than by whoever argues loudest. Monitoring UHC means tracking all three, plus whether services reach people in practice.

    Roles in the Saudi system, simplified. Under Vision 2030 the Health Sector Transformation Program aims to separate roles that were once combined: regulator, payer and provider. The Ministry of Health sets policy and regulates, health clusters organise care for defined populations, and a holding company structure supports the public provider network. Separate bodies handle drug and device regulation (SFDA), workforce standards (SCFHS), insurance regulation (CCHI) and accreditation (CBAHI). Public health functions sit with the Public Health Authority. Separation matters because a body that both pays for and provides care has little reason to question its own spending, while separate roles allow purchasing to be strategic and quality to be inspected independently. For a GI leader, the practical lesson is to know which body owns each lever you want to pull. Mandates evolve, so confirm current mandates before citing roles.

    Diagnosing the weakest link, step by step. Naming the six blocks is easy; using them is a skill. A building-block diagnosis starts from a concrete service problem, such as a 14-week wait for a first IBD visit, and asks for evidence on each block rather than opinions. Score each block with a few simple measures, such as vacancies, stock-outs, share of patients with a recorded plan, payment per visit and named pathway owner. Then find the block that most limits the others. Often it is the one nobody owns, which is why information and governance are so commonly weakest. Sequence the fixes so that dependent blocks are not left behind: a register comes before a performance target, and a nurse role before a nurse-led clinic. State what will be measured and when you will check again, so the diagnosis becomes a plan and not a report.

    Goals in practice: access, quality and equity. The four goals are easy to state and hard to measure. Improved health needs outcome data. Responsiveness covers dignity, waiting and choice. Financial protection needs household data. Efficiency asks whether the same resources could achieve more. Two goals that deserve extra attention in a GI service are access and quality, because they can move in opposite directions. Opening many endoscopy lists improves access but may dilute quality if staffing and preparation standards slip. Raising quality standards without adding capacity improves quality for those who get a slot and lengthens the queue for the rest. Equity cuts across both: ask who is at the back of the queue, by region, income, sex and disability. A health system leader looks at all four goals together and asks which patients benefit and which are left behind by each proposal.

    A worked example. An IBD service can have good biologics supply (medical products), yet long delays because the workforce lacks IBD nurses (workforce), no registry to see who is lost to follow-up (information) and no payment for multidisciplinary clinics (financing). The weakest block, not the strongest, sets the performance.

  2. 2

    The policy process, actors and power

    Agenda setting, the policy triangle, stakeholder mapping, and why good evidence is not enough

    Read the script

    Module 2. The policy process, actors and power. Agenda setting, the policy triangle, stakeholder mapping, and why good evidence is not enough.

    The policy cycle, and its limits. A useful simplification: agenda setting, policy formulation, adoption, implementation, then evaluation, which feeds back to the agenda. Real policy is messier, with stages overlapping and loops back, but the cycle is a good checklist of what might be missing. Agenda setting asks why this problem, why now, and who is pushing. Formulation turns an idea into a costed option with a clear owner. Adoption is a decision, but not yet a change in care. Many policies fail not at formulation but at implementation, where nobody planned who would do what, with what money. Evaluation is the stage most often skipped, yet it is the only way the system learns. A clinician-leader who understands the cycle can ask at each stage what evidence and which allies are needed next, and can notice early when a stage has been skipped.

    Context, content, process, actors. The Walt–Gilson policy triangle asks four questions. Context: what social, economic and political conditions shape the policy? Content: what exactly is proposed? Process: how does an issue get on the agenda and move through decision stages? Actors: who has a stake and what power do they hold? Analysing only content, as technical experts tend to do, misses most of why policies succeed or fail. The triangle is a diagnostic. Use it before you propose something, to anticipate resistance, and afterwards, to understand why something stalled. Actors sit at the centre because individuals and groups shape context, content and process alike, and each of them acts on interests, beliefs and relationships as well as on evidence. Good analysis names specific people and roles, not just institutions, and notes who can block a decision as well as who can make it.

    Stakeholder mapping. Map stakeholders on two axes, interest and influence. High-interest, high-influence actors need close engagement and a seat at the table. High-influence but low-interest actors, often finance or legal officials, must be kept satisfied because they can quietly stop a proposal. High-interest, low-influence groups, such as many patients and junior staff, deserve to be kept informed and listened to, partly for fairness and partly because they can become powerful through coalitions. Low-interest, low-influence actors need only monitoring. A map is a living document: positions shift, so revisit it. Also record what each stakeholder wants and fears, not only where they sit. Engagement then means addressing those interests directly, for example by bringing cost estimates to finance and workload data to primary care, rather than sending everyone the same briefing.

    Policy windows and multiple streams. Kingdon’s multiple-streams idea says a policy window opens when three streams meet: a problem becomes salient, a workable solution exists, and the political climate favours action. The streams run separately and most of the time they do not meet. Problems rise with a focusing event or a worsening indicator. Solutions circulate among experts, who refine them for years. The political stream moves with elections, leadership changes and budget cycles. When all three align, a short window opens, and the people ready with a costed, tested proposal can move it through. Prepared advocates keep solutions ready for when the window opens, and watch for events they can honestly link to the problem. The lesson is not to wait for luck but to do the slow work on the solution stream so that a proposal is available, credible and aligned with a decision-maker’s own goals.

    Why adopted policies stall in implementation. Adoption is a signature, not a change in care. Between the signature and the patient lie funding lines, hiring, training, procurement, information systems and local leaders who must prioritise the new work among many demands. Each handoff loses momentum, so delivery is usually a fraction of what was announced. Implementation planning therefore needs the same rigour as policy design. Name an owner with authority, estimate capacity honestly, phase delivery so early sites succeed visibly, and build monitoring in from the start. Expect resistance, because a policy that changes workload or income for any group will meet pushback, and plan to meet it with data, training and clear communication. Finally, agree in advance what would count as failure, so the programme can be corrected instead of defended.

    A worked example. A national policy to add FIT-based colorectal cancer screening to the primary care package: agenda (rising CRC incidence at younger ages), formulation (age bands, test, pathway), adoption (cabinet or ministerial approval), implementation (invitations, labs, colonoscopy capacity), evaluation (uptake, positivity, stage at diagnosis).

  3. 3

    Evidence, data and indicators for policy

    Choosing indicators, reading care cascades, and being honest about uncertainty

    Read the script

    Module 3. Evidence, data and indicators for policy. Choosing indicators, reading care cascades, and being honest about uncertainty.

    Structure, process, outcome. Donabedian’s framework separates what a system has (structure: staff, equipment, beds), what it does (process: wait times, guideline adherence) and what results (outcome: survival, complications, stage at diagnosis). Good dashboards use all three and keep a few well-defined indicators rather than dozens. Process indicators move first and are easier to act on; outcome indicators matter most but lag, and often need adjustment for case mix before services can be compared fairly. The framework also protects against a common trap: measuring only structure, which is easy to count, and assuming quality follows. A service with ample rooms can still have poor detection rates. When choosing indicators, ask which decision each one would inform. An indicator nobody can act on is decoration, however accurate it is.

    Defining an indicator properly. A usable indicator has five parts: a clear name, a numerator, a denominator, a data source and a target with a time frame. Without them, two units will report the same indicator in different ways and any comparison is noise. Decide who is counted, for example every adult with a positive FIT in the quarter, and what counts as success, for example a completed colonoscopy within 30 days. Decide how often to report and who owns the number. Then test the definition on a small sample from real records before launching it widely, because ambiguities surface only when staff start filling in the data. Keep the set small and stable: changing definitions each year breaks trends. Finally, publish the definition alongside the results so readers can judge what the number means and what it leaves out.

    The care cascade. A cascade lists the steps patients must pass through, with numbers at each: eligible, invited, tested, positive, linked to care, treated, cured. The biggest drop-off is where action pays most. Cascades are the standard tool for programmes such as hepatitis C elimination, where WHO global targets for 2030 include diagnosing 90% of infected people and treating 80% of those eligible. Read a cascade in two ways. In absolute numbers it shows how many people fall out. In conditional percentages, each step as a share of the previous one, it shows which step is weakest. The second view is the one that guides action. Because denominators change at each step, always say what each percentage is a percentage of.

    Hepatitis C elimination against WHO targets. Elimination programmes use cascades against explicit targets. WHO global targets for 2030 include diagnosing 90% of infected people and treating 80% of those eligible. Estimating the infected population is the hardest step, since it relies on surveys and models, and every percentage depends on it. A programme can look successful on paper while a large undiagnosed group remains. Linkage to care, meaning that a person with a positive test is seen by someone who can prescribe, is the step most often neglected. Cure must be confirmed by a test after treatment, otherwise the cascade ends with an assumption. Policy levers differ by step: testing and case finding for diagnosis, simplified pathways and prescribing in primary care for linkage, and drug procurement and pricing for treatment. Matching the lever to the leak is the central skill.

    Uncertainty and what policymakers can use. Policymakers need clear messages but not false certainty. Give a best estimate with a range, say how sure you are and why, and say what would change the conclusion. Distinguish what the evidence shows (an effect), what is modelled (a projection) and what is judged (a value). Data limitations should be stated early, with a plan to improve them, rather than hidden. Lead with the decision-relevant message, then the range, then the caveats. Use plain words for certainty, such as “high”, “moderate” or “low”, and define them once. Avoid both extremes: false precision that invites distrust when numbers change, and vague hedging that gives the reader nothing to use. A short, honest statement of what is known, what is assumed and what would change the answer is more persuasive over time than a confident number that later proves wrong.

    A worked example. Colonoscopy: structure is the number of endoscopists and rooms. Process includes waiting time for a positive FIT, caecal intubation rate, adenoma detection rate and bowel-preparation adequacy. Outcomes include interval cancer rate and stage at diagnosis.

  4. 4

    Health financing, insurance and purchasing

    Raising, pooling and spending money, with payment methods for endoscopy and biologics

    Read the script

    Module 4. Health financing, insurance and purchasing. Raising, pooling and spending money, with payment methods for endoscopy and biologics.

    Collect, pool, purchase. Health financing has three functions. Revenue collection raises funds from taxes, insurance contributions or direct payments. Pooling combines funds so risk is shared across many people. Purchasing uses the pool to buy services, ideally strategically, linking payment to need, quality and value. Financial protection improves as the share of spending from pooled sources rises and out-of-pocket payments fall. The three functions are separable: a system can raise money fairly but pool it in many small funds that cannot share risk, or pool well but purchase passively by paying whatever is billed. For a GI leader, this matters because a drug or a procedure is only reachable if all three work. Ask which function is weakest before arguing for more money, because more money in a poorly pooled or passively purchased system buys less than expected.

    Risk pooling in numbers. Pooling works because costly events are rare and random for any one person but predictable across many. Imagine a fund of 1,000 members in which, on average, 10 in a year need a biologic costing SAR 60,000. The expected annual cost is SAR 600,000, which equals SAR 600 per member. A single household facing the full SAR 60,000 may be pushed into debt; as a premium, the same cost is small. The calculation assumes a large enough group, a mix of healthy and sick members, and compulsory or near-universal participation. If only the sick join, the average cost per member rises, premiums rise, healthier people leave and the pool fails, which is called adverse selection. Policy responses include mandatory enrolment, subsidies for the poor, risk adjustment between funds, and larger or merged pools. Calculate expected cost per member, then ask what could make it unstable.

    How providers are paid shapes what they do. Fee-for-service rewards volume and can lead to over-use. Capitation pays per enrolled person and rewards prevention but can encourage under-provision. Diagnosis-related groups pay per case type and reward efficiency, with a risk of early discharge or selecting simpler cases. Bundled and outcome-linked payments pay for an episode or result. No method is perfect; the best systems blend them and monitor for gaming. Each method moves financial risk: fee-for-service leaves the payer to carry volume risk, capitation transfers it to the provider, and bundles split it. Providers respond rationally to what is paid, so the design question is which behaviour you want and what could go wrong. Pair every method with indicators that would reveal the typical side effect, for example adenoma detection alongside colonoscopy volume.

    Strategic purchasing, out-of-pocket spending and equity. Strategic purchasing asks what to buy, from whom, at what price, and for whom, using evidence including HTA. Out-of-pocket spending is the weakest form of financing for protection and equity. A household is commonly considered to face catastrophic spending when out-of-pocket health payments exceed 10% of its budget, one of the thresholds used in SDG indicator 3.8.2. In the private sector, mandatory cooperative insurance, regulated by CCHI, covers employees; benefit design and co-payments decide what remains out of pocket. Equity requires looking beyond averages: a system with a low average out-of-pocket share can still push poorer households into catastrophic spending, particularly for chronic conditions such as IBD or cirrhosis. Include indirect costs such as transport and lost income when assessing burden, since they often decide whether treatment is completed.

    From price lists to purchasing for value. Passive purchasing pays whatever is billed from a price list. Strategic purchasing links the contract to the value it buys. It starts by defining the benefit package with evidence, using HTA to decide which technologies are worth funding and for which patients. It then chooses a payment method and price, writes quality terms into the contract, monitors with data and adjusts. For a payer, the evidence trail is crucial: if a high-cost drug is funded without defined eligibility, volume and spending will grow without a clear link to outcomes. For a GI service, contracts can specify minimum quality standards, such as adenoma detection or complication reporting, and link part of payment to meeting them. Purchasing also needs capacity: data, contract management and a willingness to renegotiate. Without these, strategic purchasing is only a slogan.

    A worked example. A patient with Crohn’s disease needing biologic therapy faces a cost that no household can predict. Pooled funding (tax or insurance) turns an unpredictable catastrophe into a predictable premium. Without pooling, treatment depends on ability to pay.

  5. 5

    Regulation, governance and data

    Who sets the rules for products, people, facilities and data, with endoscopy as the running example

    Read the script

    Module 5. Regulation, governance and data. Who sets the rules for products, people, facilities and data, with endoscopy as the running example.

    Regulatory functions and who performs them. Health regulation covers products (drugs and devices), people (workforce standards), places (facility licensing and accreditation), payers (insurance rules) and data (privacy and sharing). The reason to separate them is practical: each asks a different question and needs different expertise. Is the device safe? Is the clinician competent? Is the unit fit to run procedures? Is the benefit covered? Is the patient’s information protected? In Saudi Arabia these functions are assigned to different bodies, such as SFDA for drugs and devices, SCFHS for health professionals, CCHI for cooperative insurance, and CBAHI for accreditation, while data use is governed by the Personal Data Protection Law (PDPL) and health-sector rules. For a clinician leader the working method is simple: list every function a proposal touches, name the body that holds each lever, and approach them in a sensible order. The exact boundaries evolve, so confirm current mandates for any formal use.

    Instruments: licensing, standards, incentives, information. Regulators have a menu of instruments, and the skill lies in matching one to the problem. Licensing and credentialing set minimum entry requirements. Standards and accreditation set expected processes. Incentives, such as payment tied to quality, reward desired behaviour. Information, such as public reporting, lets patients and purchasers compare. A useful rule of thumb is proportionality: the greater and more certain the harm, the firmer the instrument. A safety hazard may need a mandatory standard with audit; a performance gap among competent clinicians may respond to feedback and benchmarking. Heavy tools also carry cost: inspection, paperwork and the risk of box-ticking. Light tools carry a different risk, which is that nothing changes. Whichever instrument is chosen, define who enforces it, how compliance is measured and what happens when it is not met. A rule that nobody audits is only advice.

    Transparency and conflicts of interest. Decision bodies earn trust through declared interests, recorded reasons and published outcomes. A conflict of interest is any financial or other interest that could influence, or appear to influence, judgement. It is not misconduct to have one; the failure is to hide it. Good practice has three parts. First, declare: everyone completes a structured form before the meeting and updates it when circumstances change. Second, judge: the chair decides what the interest means for this agenda item, using the size of the interest and its closeness to the decision. Third, record: the minutes state who declared what and how it was handled. Handling ranges from noting the interest, to allowing a member to speak but not vote, to asking them to leave the room. Banning every expert would leave panels without knowledge, so the aim is management, not purity. Publishing reasons and decisions allows outsiders to check that the process was fair.

    Registry governance and data protection. Registries turn routine care into knowledge, but they hold sensitive information about identifiable people, so they need governance before the first record is entered. A defined purpose comes first, because purpose limits what may be collected and who may see it. Next come a named data controller, the information given to patients, and a consent or other lawful basis under the data protection law. Security covers access control, encryption and an audit trail of who viewed what. Rules for access and sharing say who may request data, how requests are reviewed and what is released, usually in de-identified form. Finally, a plan for retention and deletion stops data being kept indefinitely. Patients and clinicians should sit on the steering group, and an annual public report shows the registry is used for improvement. Without these elements a useful registry becomes a liability, and clinicians will rightly decline to contribute.

    Accreditation, audit and the quality loop. Standards matter only if they change what happens at the bedside. The quality loop connects four steps: set a standard, measure against it, feed results back and act on gaps. For an endoscopy service the standards might cover reprocessing, consent, sedation monitoring, bowel-preparation quality and detection rates. Accreditation provides an external check on structure and process, while internal audit provides a continuous view of outcomes. The difference in tone matters. Audit that is confidential and educational invites honest data; audit that is used for ranking and blame invites gaming. A mature programme starts with confidential feedback and a benchmark, publishes aggregated results, and reserves sanctions for unsafe practice that persists after support. Leaders also need to protect time for audit, because a loop that depends on goodwill alone will stop during the first busy quarter. Report the spread, not just the average, because variation between clinicians is often the most useful signal.

    A worked example. A new capsule endoscopy system needs SFDA device clearance; the operators need credentialing recognised by the professional body; the hospital unit must meet licensing and accreditation standards; and image storage must satisfy data protection rules.

  6. 6

    Strategic planning for GI services

    From environmental scan to capacity, wait times, workforce and a measurable plan

    Read the script

    Module 6. Strategic planning for GI services. From environmental scan to capacity, wait times, workforce and a measurable plan.

    Scan, goals and indicators. A plan starts by understanding the context: population need, current capacity, workforce, finance, partners and policy direction. This environmental scan answers the questions “where are we?” and “what is pushing on us?” before anyone proposes a solution. Goals follow, written so they can be checked: what will change, by how much, by when, who owns it and how it will be measured. Vague goals (“improve access”) cannot be monitored, and unowned goals drift. Each goal needs one or two indicators and a baseline, because without a starting figure an improvement cannot be shown. Prefer a small number of goals that the team can remember. Three to five well-chosen indicators, reviewed regularly, beat a dashboard of thirty that nobody reads. Distinguish outcome indicators (waiting time, complications) from process indicators (sessions delivered, referrals triaged), and include at least one quality measure so that speed is never bought at the cost of safety.

    Queues, capacity and the backlog. A waiting list behaves like a queue. If referrals arrive faster than the service can treat them, the backlog grows every week, however hard staff work. The average wait is roughly the backlog divided by weekly capacity. To reduce waits, a service must have a positive surplus (capacity minus demand) or must cut demand. A one-off extra effort without more capacity clears a backlog once, but it returns. This arithmetic tells a leader what kind of problem they face. A backlog with balanced demand and capacity is a temporary problem that a one-off push can fix. A backlog with a recurring deficit is structural, and only added capacity or reduced demand will help. Also watch the trend in demand: screening programmes, ageing and new referral pathways can all raise referrals, so a plan that balances today may fail next year. Model the numbers before choosing the response.

    Levers and trade-offs. Waiting-list policy has three families of levers: add capacity (sessions, staff, rooms, extended hours), reduce inappropriate demand (triage using FIT or calprotectin, appropriate-use criteria, primary care protocols) and improve flow (pre-assessment, bowel-preparation support, reducing no-shows). Each has costs and risks. Triage lowers demand but needs primary-care capability; added sessions cost money and need staff who take time to recruit; no-show reduction works only if patients are reachable. Rarely does one lever suffice, so plans usually combine them. The skill is to estimate the effect and the cost of each, compare them on the same scale, and check for side effects. Extra volume can lower quality if teams are stretched, and tighter triage can delay diagnosis if primary care cannot follow results up. State the assumptions, test them in a pilot, and track a quality measure alongside the waiting time.

    From sessions to completed procedures. Capacity is often counted in slots, but patients experience completed procedures. The gap between the two is where services quietly lose output. A scheduled session has a fixed number of slots. Some patients do not attend. Some arrive with inadequate bowel preparation or unsuitable for sedation, and the procedure is cancelled or abandoned. Others need more time than a standard slot. Planning with slots alone overstates capacity and leads to a persistent shortfall. A better approach converts slots into expected completed procedures by applying observed rates for no-shows, failed preparation and overruns taken from local audit. This also shows where small improvements pay off: reducing no-shows or improving preparation increases completed procedures at almost no extra staffing cost. Staffing follows from the same arithmetic. Divide weekly demand by slots per session to estimate sessions, then check that rooms, nurses, recovery space and reprocessing capacity can support them. The scarcest of these sets the true limit.

    Monitoring, review points and decision rules. A plan that is never reviewed is a wish. Build review points into the plan from the start, with the indicators to be examined, the person responsible and the decision that follows. A decision rule states in advance what the team will do if the numbers move one way or the other. For example: if the backlog is still rising at three months, escalate and add capacity; if the median wait is below target for two successive months, consider releasing temporary sessions. Deciding rules before seeing the data protects against wishful interpretation. Monitoring should combine speed and quality: waiting time, no-show rate and backlog alongside detection rate and complications. Report the trend as well as the latest value, because a single month can mislead. When the plan is not working, ask whether the cause is the assumptions, the delivery or the context, and change the right element. Finally, record what was learned, so the next plan starts from evidence rather than memory.

    A worked example. Goal: reduce median wait for routine colonoscopy from 18 weeks to 8 weeks within 12 months, owned by the cluster endoscopy lead, measured monthly from the booking system. Baseline: 18 weeks (current audit).

  7. 7

    Policy communication and advocacy

    Writing a policy brief, making the economic case, and engaging decision-makers

    Read the script

    Module 7. Policy communication and advocacy. Writing a policy brief, making the economic case, and engaging decision-makers.

    The policy brief. A policy brief is short (two to four pages), standalone, and written for a busy reader who may give it five minutes. A typical structure follows the WHO EMRO template: title and key messages; the problem and why it matters now; the options with their pros and cons; a recommended option; implementation considerations; and a one-line ask. Lead with the conclusion, use plain words, avoid jargon, and put detail in annexes. The test of a good brief is whether a reader who stops after the first paragraph still knows what you want them to decide. Show the options fairly, including doing nothing, because decision-makers distrust papers that present only one answer. Put numbers in context, with a comparison or a unit the reader can picture. Name who would do what, by when, and at what cost. Finally, have someone outside your field read the draft, and rewrite any sentence they cannot summarise.

    Making the economic case. Decision-makers ask what it costs and what they get. A benefit–cost ratio (BCR) compares the present value of benefits with the present value of costs. For prevention and screening, benefits arrive years after costs, so the lag matters. Be explicit about the lag, discount rate and what is counted. A BCR above 1 means benefits exceed costs, but a ratio near 1 with a long lag is a hard sell. Report assumptions and a range, and say who gains: savings to one budget holder do not help another. Also say what is left out, such as life-years gained, because the unmeasured benefits may be the main reason to act. Test the result by moving the lag and the discount rate and showing how the ratio changes. A finance officer is more persuaded by a candid range than by a single figure that looks too good. If the BCR is below 1, say so, and argue on health and equity grounds rather than hiding the arithmetic.

    Audience and channel. Different audiences need different messages, and the same facts can fail if they are framed for the wrong reader. A minister needs a decision and a headline number; a finance officer needs cost and risk; clinicians need clinical credibility; the public needs clear action. Choose the channel to match: brief and meeting for decision-makers; press release and media interviews for the public; social media for reach, with accuracy first. The messenger matters as much as the message. A clinician who treats the patients, a patient who tells the story, or a respected official may carry more weight than an organisation’s letterhead. Before writing, ask three questions: who decides, what do they already believe, and what would make saying yes easy? Then tailor the first sentence, the evidence you lead with and the call to action. Keep core facts identical across audiences so that messages do not contradict each other when they are compared.

    Crisis communication. In a health crisis, such as a suspected outbreak, trust is built or lost in the first hours. The principles are to be first, be accurate, be consistent, be credible and show empathy, and above all to tell people what they should do. Say what is known, what is not known, what is being done, and when the next update will come. Acknowledge uncertainty openly; people forgive “we do not know yet” far more readily than a retraction. Use one coordinated set of facts so that every spokesperson says the same thing, and nominate who speaks for which topic. Correct mistakes quickly and visibly. Plan before the crisis: prepare holding statements, contact lists, approval routes and a rumour-monitoring routine. Do not wait for perfect information, because silence is filled by rumour. Finally, review afterwards: what was said, when, and what people did as a result, so the next response is faster and clearer.

    Stakeholders: mapping and engaging. Advocacy succeeds when the right people support a proposal, so it pays to map them. For each stakeholder, estimate how much influence they have over the decision and how supportive they currently are. This gives four groups. Influential supporters are your champions: equip them with clear messages and ask them to speak. Influential people who are not yet supportive are the priority: find out their concerns, and answer them with evidence and adjustments. Supporters with little influence can build a visible coalition and provide stories. Those with little influence and little support need only light-touch updates. Engage early, before positions harden, and listen as much as you speak. Seek common ground, such as shared goals for patients, and be honest about trade-offs. Keep a simple log of contacts and commitments, and thank people publicly when they help. Over time, a stable network of allies is more valuable than any single brief, because it lets a proposal survive changes of personnel.

    A worked example. Brief on hepatitis C elimination. Key message: “Treat everyone diagnosed within 12 weeks.” Problem: half of those diagnosed are untreated. Options: status quo, treatment hubs in clusters, primary care prescribing. Recommendation: hubs plus training. Ask: approve pilot funding for three clusters.

  8. 8

    Equity, ethics and implementation

    Measuring inequity, choosing under scarcity, and making policy work on the ground

    Read the script

    Module 8. Equity, ethics and implementation. Measuring inequity, choosing under scarcity, and making policy work on the ground.

    Measuring equity. Inequity is a difference in health or access that is avoidable and unfair. To manage it, measure it: compare an indicator across groups (region, income, sex, nationality) using the rate difference (absolute gap) and the rate ratio (relative gap). Report both. A small relative gap in a common outcome can be a large absolute burden, while a large relative gap in a rare outcome can be a small absolute burden. National averages can hide these gaps completely, so always break indicators down by the groups that matter. Choose the reference group deliberately, usually the best-off or the largest group, and say which you used. Interpretation then needs judgement: ask whether the difference is avoidable and whether it is fair. Differences in age structure should be adjusted for, but differences driven by distance or cost are the policy targets. Track the same measures over time so you can show whether a policy narrows or widens the gap.

    Choosing under scarcity: screening and priorities. Wilson and Jungner’s 1968 principles remain the standard checklist for screening: an important health problem, an accepted treatment, facilities for diagnosis and treatment, a recognisable early stage, a suitable test acceptable to the population, an understood natural history, an agreed policy on whom to treat, cost balanced against the benefit, and continuing case-finding. The criteria make the question concrete: can the system deliver everything that follows a positive test? Priority-setting under scarcity should be explicit, using criteria such as burden, effectiveness, cost, equity and feasibility, and should be open to challenge. Scoring options against shared criteria does not remove judgement, but it exposes it, so that people can disagree about weights rather than about hidden assumptions. Always consider harms alongside benefits: false positives, overdiagnosis, complications of follow-up procedures and the opportunity cost of what is displaced.

    The screening cascade: where people are lost. A screening programme is not a single test but a chain: people are invited, some take the test, some test positive, and some of those complete the follow-up examination. At every step a share of people is lost, and the losses are usually larger in groups that are already disadvantaged. The final yield, meaning cancers found early, depends on the product of the step rates, so a modest loss at each stage compounds. Mapping the cascade turns a vague concern about low impact into specific targets: if remote areas lose half of positive tests before colonoscopy, improving transport or tele-triage for that step may do more than extra invitations. Collect data on each step by group, not just a total, and look for the largest drop. Remember that a screening test without reliable follow-up can cause harm, because people are told they may have cancer and then not examined.

    Implementation and learning. Policies work when people can do what the policy asks. Implementation frameworks such as CFIR list domains to examine: the intervention itself, the outer setting, the inner setting, the people involved and the process. Typical barriers are workload, unclear roles, lack of training, weak data systems and misaligned incentives. Match strategies to barriers (training, protocols, reminders, feedback, payment changes) and monitor early, with a clear decision rule for adapting or stopping. Begin with a diagnosis: ask the staff and patients who face the barrier, because their account often differs from a manager’s assumption. Choose a small number of strategies that clearly address each barrier, rather than a long list that nobody can deliver. Pilot before scaling, and track both outcomes and the fidelity of delivery, since a policy that was never delivered cannot be judged to have failed.

    Monitoring, adaptation and stopping rules. Every pilot should end in a decision. Before it begins, agree the indicators, the review date and the thresholds that trigger each of three outcomes: scale up, redesign or stop. Scaling follows when targets are met and no new harms appear. Redesign follows when a gap persists but the idea remains sound, for example because a barrier was not addressed. Stopping follows when there is harm or no meaningful benefit, and it should be treated as a legitimate result, not a failure. Pre-agreed thresholds protect against two errors: continuing a failing policy because of sunk cost, and abandoning a promising one after a single bad month. Use equity-specific thresholds too, such as remote uptake and completion, so that an average improvement does not hide a widening gap. Share the results openly, including what did not work, because other clusters can then avoid repeating the same mistakes.

    A worked example. FIT screening uptake is 52% in urban areas and 31% in remote areas. Rate difference 21 percentage points; rate ratio 1.68. If remote eligible adults number 200,000, closing the gap would add 42,000 people screened.

Structure and topics adapted from the PHPSA academy (Public Healthcare Policy Society of Saudi Arabia). SGA authored the GI examples and figures here; PHPSA has not reviewed them. Descriptions of Saudi institutions are simplified and should be checked against current mandates. PHPSA

Educational content only. Figures in examples, calculators and simulators are illustrative — they are not Saudi price or cost data and do not represent any coverage, pricing or official policy decision.

Share this