Table 1

PDSA improvement cycles

PDSA cycle	Plan/prediction	Do	Study	Act	Time required (months)
Baseline			OM1in: 7.7 days; OM1out: 33.2 days ☒ OM2: 11% intubation failures ☒ PM1n: 6 TOE slots per month PM1w: 13% slots wasted ☒ PM2: 2.92 mg of midazolam (median 3 mg)
1	Institute a clinical scientist led service, extending role (CI A) and reduce sedation (CI D). This will improve OM1 (reduced waiting times due to more capacity, PM1n) and OM2 (greater intubation success rate).	Clinical cardiac scientist as lead operator 2nd cardiac scientist/physiologist in support (image acquisition) Weekly TOE lists (capacity 6 → 8 TOEs per month) Unused slots → TTE cardiac nurse delivers reduced sedation (midazolam).	OM1in: 3.9 days; OM1out: 15.0 days ☒ but successful reduction OM2: 0% intubation failures ☑ PM1n: 8 TOE slots per month ☑ PM1w: 29% slots wasted ☒ Unsuccessful—worse! (but can use for TTE) PM2: 1.66 mg of midazolam (median 1.5 mg) ☑	Do not recalculate the SPC limits for OM1in; recalculate for OM1out. Worthwhile improvement, make permanent change: Capacity=8 TOEs per month; Cardiologist 2 days a month released; Continue lower sedative. Inappropriate bookings: institute referral review	3
2	Institute review of referrals prior to booking and give priority to inpatients (CI B). This will further improve OM1 (by reducing slots wasted, PM1w, by reducing inappropriate bookings).	Cardiac nurse prints referrals. Reviewed by clinical scientist Appropriate cases booked, prioritising inpatients	OM1in: 2.1 days; OM1out: 8.3 days ☑ OM2: 0% intubation failures ☑ PM1w: 13% slots wasted ☒ (but used for TTE)	Recalculate SPC for OM1in, (at PDSA1). Provisional recalculation for OM1out at PDSA2. Worthwhile improvement, make permanent change. Continuing problem with end of week referrals—unable to review—refine in future.	2
3	Reassign roles to avoid need for a support (2nd) clinical scientist (CI C1). Can be done without reducing performance Frees resource for other departmental services.	Clinical scientist (lead operator) and cardiac nurse share image acquisition. Do not require a support clinical scientist.	No change to performance metrics (success!)	Release 2nd clinical scientist from here on. Technical success, but care management is difficult with challenging patients: refine	2
4	Add nursing assistant role to help with manipulation of patient’s head and ensure safety is maintained (CI C2). Easier management, especially of challenging patients.	Add nursing assistant role.	Easier patient management—successful	Worthwhile improvement, additional role justified, retain as permanent change.	3
5	Add ‘clinical scientist’ as TOE operator on patient information leaflet and institute patient satisfaction form to assess experience and gather patient suggestions (CI E). Patients will be reassured.	Discuss with stakeholders. Change the information leaflet. Include a patient satisfaction form with appointment letter.	No change to performance metrics (as expected) Patient feedback confirms new process is not upsetting patients ☑	Worthwhile information, retain as permanent change. Easy way to seek patient ideas for areas of improvements.	2

☒, target not achieved; ☑, target achieved; CI, change idea; IP, inpatient; OM, outcome metric; OP, outpatient; PDSA, plan-do-study-act; PM, process metric; SPC, statistical process control; TOE, transoesophageal echocardiography; TTE, transthoracic echocardiography.