Plan-Do-Study-Act cycles of change during implementation
| Problem | Plan | Do | Study | Act |
| Primary care confusion on how to link patients with correct behavioural health provider as evidenced by inappropriate referrals. | Reorganise ordering from type of provider (PhD, LICSW, MD) to type of service (diagnostic evaluation, medication management, therapy). | Rollout of new electronic ordering. | Frequency of inappropriate referrals went down overall except for those from a few individual providers. | One-on-one assistance on the new ordering system offered to those slow to adapt to new system. |
| Include all referrals from primary care or only those with significant symptoms warranting this intervention? | Initially accepted all referrals regardless of symptom severity to help develop the programme. | Data entry on initial 57 patients to explore outcomes regardless of initial severity on GAD-7. | Review of initial 57 patient outcomes demonstrated difficulty in justifying treatment on subthreshold (GAD-7 <10) patients. | Full course of CALM offered only to those with GAD-7 score of 10 or more. Very brief1–3 session developed and offered to others. |
| Therapist documentation oriented around billing requirements and not around tracking of outcomes. | Registry developed and therapists given task of data entry into registry during or just after therapy visits | Trained therapy staff on web-based registry. | Significant missing data as therapists reported time constraints and lack of clarity on reasons for this aspect of the model. | Simplified documentation and created routine review of data in case reviews with psychologist, which improved both data entry completeness and allowed for case-based teaching. |
| Reference RCT CALM model also included an algorithm for medication management of anxiety. | Offer medication review for patients with anxiety as a part of parallel adult care coordination available in primary care settings. | Algorithm created and reviewed, option for medication review advertised. | Referral rates by primary care very low for routine review of medications. | Abandoned routine algorithm for anxiety medications in favour of individual consultation as needed. |
CALM, Coordinated Anxiety Learning and Management; GAD-7, Generalized Anxiety Disorder 7 Item scale; LICSW, licensed social worker; RCT, randomised controlled trial.