By Dr. Cole Stanley
On a grey mid-afternoon in British Columbia, a family doctor (we’ll call her Dr. J, a composite of many clinicians we’ve worked with) sat staring at a blank Quality Improvement (QI) form.
It was the kind that many physicians are encouraged to do regularly and may feel a professional obligation to complete: a request to “pick a project,” preferably something measurable, ideally something meaningful, though the instructions rarely said how to find either.
Dr. J cares deeply about her patients. She cares about better chronic disease care, better prevention, fewer crises. But in truth, like many clinicians, she was guessing. What should she work on? What mattered most? Would her effort even make a noticeable impact?
Here’s something we don’t always name: every practice is itself a small system. It has workflows, handoffs, data flows, patterns of care that repeat hundreds of times a year. But as clinicians, we’re trained to see one patient at a time, not to look at these systems and improve them. Dr. J was being asked to improve a system she’d never been taught to navigate, using data she’d never been trained to interpret, under conditions that often ensured she would be working alone.
And she felt it.
A Missing Piece in Primary Care’s Foundation
It’s a familiar story in primary care: improvement done in isolation, driven more by obligation than by inspiration. In an era of rising complexity and mounting administrative burden, clinicians are often left trying to fix system-level problems with individual-level tools.
What’s less often named is that this isolation isn’t a personal failing, it’s a structural gap. Unlike hospital-based medicine, where interdisciplinary rounds, morbidity and mortality reviews, and unit-level meetings are woven into the fabric of the work, community-based primary care has no equivalent infrastructure for collective learning and improvement. There is no built-in space where family physicians routinely come together to examine their practice data, compare approaches, and improve as a group.
This isn’t a nice-to-have. It is a missing foundation. And without it, clinicians are asked to carry system-level improvement with individual-level effort, a recipe for burnout, frustration, and stagnation.
Evidence supports what many of us intuitively know: collaborative approaches to quality improvement can be quite effective. A systematic review of 64 quality improvement collaboratives found that 83% reported improvements in their primary outcome measures (Wells et al., BMJ Quality & Safety, 2018). The mechanism isn’t mysterious: peer learning reduces cognitive load, increases accountability, sparks ideas you wouldn’t generate alone, and creates the psychological safety needed to look honestly at your own data.
A recent pilot in BC suggests there may be a practical path forward, one that turns improvement into a collective act rather than a solitary chore.
The Old Way: Improvement by Guesswork
Before the two pilot groups began, many physicians described their annual QI cycle the same way: pick something you can finish quickly, something not too risky or embarrassing, something you can manage without a team, without funded time, and without much support.
To be clear, not all individual QI work is low value. Many clinicians do thoughtful, meaningful improvement with excellent support from organizations like the Health Data Coalition (HDC) and the Practice Support Program (PSP), offerings of the Family Practice Services Committee (FPSC), that consistently elevate the experience of QI work.
The problem isn’t the quality of that support. It’s that too often QI still happens side-of-desk, time-limited, and in isolation, asking clinicians to carry improvement alone, even when solid coaching is available.
Meanwhile, the real needs (timely diabetes care, heart failure follow-up, multimorbidity, safe prescribing, chronic pain) are bigger than any one clinician could hold alone. These are practice-level system problems, not individual knowledge gaps. A physician might know the latest diabetes guidelines perfectly, but without workflow changes, team coordination, and data to track whether things are actually improving, that knowledge sits unapplied. This is the know-do gap.
The consequence is predictable: QI rarely touches the problems that need it most.
This is the world Dr. J operated in: improvement driven by intuition, effort, and hope, but without a clear signal about what actually matters to her patients or her practice as a system.
The New Way: A Small Group with a Shared Purpose
Last year, HDC piloted a different model of improvement: Small Group Learning Sessions (SGLS), a structured, fully supported, low-burden way for clinicians to improve together. This isn’t a one-off data party or a single reflect-and-discuss session. SGLS brings clinicians together over time to align on a shared aim, build trust, and work through iterative improvement cycles on a topic they’ve chosen based on their own data.
Two groups were formed, each made up of experienced primary care physicians from different practices. The groups met virtually every three months for 90 minutes, facilitated by myself and co-supported by HDC’s Cathy McGuinness and Aigul Musambetova and PSP Coach Suzanne Beyrodt-Blyt.
From the first meeting, the tone was different.
Instead of arriving with preselected topics, the groups began with something medicine often rushes past: trust. Participants co-created a group charter, agreeing to principles like psychological safety, confidentiality, generosity of listening, and patient-centred goals. Only after establishing this foundation did they look at their data, collectively, openly, with curiosity.
For the first time, improvement didn’t start with picking a project. It started with asking: What do we, as a small community of learning, see in our data that suggests where we could have the most impact?
From Data to Clarity: Choosing the Right Topic
When the groups examined their HDC Discover measures (HDC currently offers over 400 clinical measures spanning chronic disease, prevention, mental health, substance use, and more), a different kind of conversation emerged. Instead of “What project is easiest?”, physicians asked:
- What patterns surprise us in our data?
- Where is there variation within our group?
- What signals suggest opportunity, or urgency?
Both groups considered several clinical areas before settling on their topics. The conversations ranged across diabetes, heart failure, COPD, mental health, and preventive care. One group chose diabetes. The other chose heart failure. Not because they were simple, but because the data made the needs unmistakably clear. A separate question, “What work feels meaningful enough to sustain?”, helped filter the list once options were on the table.
“I’m used to designing and doing QI projects on my own. Working with colleagues kept the work on track and sparked ideas I never would’ve come up with alone.“
Dr. Angela Jennings,
Family Physician
Over time, topics didn’t stay static. Heart failure work, for example, evolved from basic case-finding to deeper questions about disease staging, lifestyle interventions, and how to sustain gains, illustrating how quality planning naturally flows into improvement and then into quality control (the ongoing monitoring of important outcomes to catch problems early, distinct from external quality assurance or auditing).
A Light Lift with a Large Impact
What made the SGLS pilots different wasn’t only the structure, it was how light the lift felt compared to the return. The SGLS model doesn’t replace strong individual coaching; it amplifies it by pairing skilled facilitation with peer connection, shared data, and funded time.
Why this Model Matters Now
Primary care is cracking under the weight of complexity: chronic disease rates are rising, psychosocial needs intensifying, administrative burden is growing, and downstream crises regularly overwhelm our hospitals.
Many of these pressures (patient deterioration, preventable hospitalizations, delayed diagnosis) are shaped upstream in primary care. HDC Discover provides a shared system of measurement for important clinical topics, backed by an organization that promotes frontline data use for learning and improvement (and helps clinicians avoid the many pitfalls of healthcare data, which we’ve explored in our Circles of Healthcare Data Hell series). SGLS gives clinicians the psychological safety and peer support to consider something different from the status quo.
Importantly, the work remains locally driven. Topics emerge from clinicians’ own data and patient populations, allowing groups to align around what matters in their communities, while still pointing toward shared provincial priorities like chronic disease management and prevention.
When data becomes a shared mirror rather than a judgment, improvement becomes lighter, faster, and far more human.
From Clinic to Community: What Scaling Could Look Like
The SGLS pilots brought together individual clinicians from different practices into small communities of learning. But the model carries a natural next step: what if groups like these were organized around a community’s health needs?
Imagine practitioners within a Primary Care Network or Division of Family Practice coming together not just to improve individually, but to collectively address the health priorities of the population they serve. A community with high rates of heart failure could form an SGLS group around that theme. A region seeing rising mental health burden could align its peer learning accordingly. The data infrastructure already exists in HDC Discover to support community-level views; what’s been missing is the structured peer learning process to act on it.
This isn’t about top-down direction. The power of the model lies in preserving clinician and community autonomy: the group decides what matters, informed by their own data and local context, while aligning naturally with broader system priorities like the Ministry’s chronic disease and prevention objectives. The question shifts from “What project should I pick?” to “What can we improve together that matters to our patients and our community?”
A Better Way Forward
Participation in the pilots was supported with sessional funding, an acknowledgment that meaningful quality improvement activities require funded time. This support created the conditions for intrinsic motivation to surface; it wasn’t the primary driver, but it removed the barrier of asking clinicians to add unpaid work to an already overwhelming schedule.
Looking ahead, the SGLS model doesn’t depend on new funding streams. Similar structures could be sustained through existing CME allocations or Division-level social or community investment funds (as some Divisions of Family Practice already use). The key insight is that the model is low-cost relative to its return: a few hours of funded peer time, supported by existing PSP coaching and HDC data infrastructure, can unlock improvement that years of solo QI could not.
For Dr. J, the transformation was simple but profound: improvement shifted from guessing to knowing, from isolation to connection, from administrative burden to professional meaning.
The SGLS model offers BC something primary care has long needed: a replicable, scalable, virtual, low-burden way to move together from quality planning, to quality improvement, to quality control.
It didn’t fix the system. But it illuminated what’s possible when clinicians don’t have to fix it alone.
And Dr. J hasn’t stared at a blank QI form since.
Interested in learning more about SGLS or starting a group in your community? Contact HDC at info@hdcbc.ca or visit hdcbc.ca.
* Panel management refers to the systematic approach of tracking and managing care for an entire patient panel (not just the patients who walk through the door), including outreach, recalls, and population-level screening.
**The Doctors Technology Office (DTO) provides BC physicians with technology support, including EMR optimization and digital health tools. Both are available through the FPSC ecosystem.
Reference
Wells S, Tamir O, Gray J, Naidoo D, Beez M, Wickham D. Are quality improvement collaboratives effective? A systematic review. BMJ Quality & Safety. 2018;27(3):226–240.