Staged Rollout of a Patient Portal at CareLink Health
CareLink Health runs nine outpatient clinics and is launching a patient portal for appointment booking, record access, and bill payment. Marketing has already announced a six-month launch because a competitor launched a similar portal. The EHR vendor can expose APIs but has a poor track record on release quality. Clinical staff are skeptical: they fear patients will arrive with incomplete information and that extra work will land on already stretched nurses. The call centre has not been trained, and the security review is only partly complete. IT wants to use a waterfall plan with a single go-live date. The chief medical officer wants a pilot in two clinics first. The board has approved the budget but not the launch date. No single person owns the portal service.
1. Apply the VeriSM stages-Define, Produce, Provide, Respond-to this launch. What would you do differently from the single go-live plan?
Read the worked answer
Define: agree outcomes such as patients booking and managing appointments, viewing records, and paying bills; identify consumers including patients, carers, clinicians, and the call centre; and record constraints such as privacy, EHR API limits, and clinical safety. The marketing date should not define the service. Produce: build with a cross-functional team and select mesh practices including architecture, security, supplier management, change enablement, project management, and knowledge management. Run a two-clinic pilot rather than a single big-bang go-live. Provide: define the support model, service levels, call-centre training, clinical escalation, and patient communication. Respond: monitor incidents, adoption, complaints, and clinical queries, and feed lessons back into Define and Produce. The chief medical officer's pilot is a risk-reduction move, not a delay. The security review must complete before any patient data is exposed, and a service owner must be appointed.
2. Which management practices from the management mesh are most critical here, and why not simply apply every practice equally?
Read the worked answer
The mesh is selected according to context, not applied as a fixed checklist. Here the high-risk practices are information security, clinical safety and risk, supplier management because of the EHR vendor's release quality, change enablement, knowledge management, and incident management. Project management needs stage gates with clinical sign-off rather than a single date. Practices such as detailed capacity management or financial planning for a mature service may matter later but are not the first constraint. Selecting practices should follow the Define-stage outcomes and risks. Applying every practice equally would slow the pilot and obscure the real clinical and security risks. The service owner should document why each selected practice is included and review the mesh as the portal moves from pilot to production.
3. During the pilot, Respond shows low adoption and incomplete patient records. What is the appropriate response, and what should not be done?
Read the worked answer
Treat the signal as information, not failure. Investigate usability, clinician workflow, patient trust, communication, and data completeness before adding features. Check whether patients understand what the portal is for and whether clinicians can correct or annotate records safely. Review call-centre contacts to see where patients get stuck. Do not simply extend the pilot unchanged, blame patients for low digital uptake, or let marketing increase claims to drive adoption. Do not proceed to the remaining seven clinics until the root causes are understood and the risk is acceptable. The service owner should convene clinical, security, call-centre, and patient representatives, update the Define-stage assumptions, and adjust the Produce and Provide plans. Respond should feed the next iteration, not become a one-off report.