Feedback Loops
A patient completes six sessions of physiotherapy, reports feeling significantly better, and then simply... disappears. No cancellation, no explanation, just silence. Three months later, they're back, in more pain than ever, wondering why their progress didn't hold. This pattern isn't a mystery; it's a predictable consequence of one of the most well-documented failures in human decision-making: people behave differently when they can't see the consequences of their choices unfolding in real time.
The Science Behind Feedback Loops
Feedback loops are one of the foundational mechanisms through which humans learn, adjust, and make better decisions. At their core, they work like this: when people receive clear, timely, and relevant information about the results of their behaviour, they are far more likely to course-correct before a small problem becomes a large one. Richard Thaler and Cass Sunstein explored this concept in depth in their landmark 2008 book *Nudge*, arguing that well-designed feedback is one of the most powerful and cost-effective 'nudges' available to any organisation trying to improve human outcomes. The principle doesn't rely on lectures, warnings, or willpower, it simply makes consequences visible.
The psychology behind feedback loops draws on several established cognitive mechanisms. The first is the 'out of sight, out of mind' problem: humans are notoriously poor at tracking slow-moving, invisible changes. A patient whose mobility gradually deteriorates over eight weeks after stopping treatment rarely perceives the decline as clearly as they perceived the original improvement. Pain returns incrementally, range of motion reduces by degrees, none of it triggers an alarm because there is no alarm. Feedback loops install that alarm. The second mechanism is what behavioural economists call 'present bias', our tendency to dramatically overweight what we feel right now relative to future consequences. A patient feeling 70% better after six sessions genuinely believes they'll manage on their own. Without feedback showing them the statistical trajectory of patients who stop at that exact stage, they have no rational basis to question that belief.
Thaler and Sunstein draw on a range of evidence to demonstrate how feedback transforms behaviour across health, energy use, and financial decisions. Their most instructive examples involve situations where the gap between action and consequence is invisible or delayed, precisely the situation in allied health. A patient doesn't feel the moment their disc rehab stalls. They don't experience the physiological point at which neural pathway reinforcement begins to reverse. They experience only the subjective sensation of 'feeling fine,' which is a deeply unreliable guide to clinical progress. Structured feedback closes the gap between what patients feel and what is actually happening inside their bodies.
Critically, feedback loops work not just by informing, but by reframing. When a patient sees their own data, mobility scores, pain indices, functional milestones, presented in a clear trajectory, the decision to stop treatment is no longer abstract. It becomes a specific, personalised choice with a visible cost attached. Research in behavioural science consistently shows that people respond far more strongly to concrete, individualised information than to generic advice. Telling a patient 'most people benefit from completing their full treatment plan' produces minimal behaviour change. Showing that specific patient 'your mobility improved 40% over six sessions, and patients with your profile who stop now typically lose around 60% of those gains within eight weeks' is an entirely different cognitive experience.
The Research
One of the most compelling real-world demonstrations of feedback loops cited in *Nudge* involves household energy consumption studies conducted by Opower (now Oracle Utilities), which partnered with utility companies across the United States to send households personalised energy reports. These reports didn't simply show people their energy usage, they compared it against similar neighbouring households and provided a clear feedback signal: you are using more (or less) than your neighbours. The results, studied by economists Hunt Allcott and Judd Kessler among others, showed that households receiving these personalised feedback reports reduced their energy consumption by an average of around 2% compared to control groups. While 2% sounds modest, across millions of households the effect was equivalent to taking hundreds of thousands of cars off the road, and crucially, it was achieved with no financial incentive, no new technology, and no change in pricing. The sole intervention was making consequences visible and personally relevant. The parallel to allied health is direct: patients who can see their own progress data, benchmarked against clinical expectations for their condition, are in a fundamentally different decision-making position than patients who simply stop receiving care and receive no information about what that means for their recovery trajectory.
How to Apply This in Your Practice
The most effective way to apply feedback loops in your allied health practice is to build a structured progress reporting system that activates at clinically meaningful milestones, not just at the end of treatment, but throughout it, and critically, at the moment a patient goes quiet. Start by identifying the two or three measurable outcomes your practitioners already track for common presentations: range of motion for shoulder rehab, pain scale scores for lower back conditions, pressure threshold scores for podiatry patients with plantar fasciitis. These don't need to be exotic metrics, they need to be consistent, recorded at each session, and retrievable.
Once your data collection is systematic, the feedback message itself can be straightforward and deeply effective. Consider a re-engagement SMS or email that reads something like: 'Hi Sarah, we noticed you haven't been in for a few weeks. Looking at your progress notes, your shoulder flexion improved from 95° to 145° over your six sessions with us, that's a 53% improvement. Clinical research shows patients who pause treatment at this stage without a maintenance plan typically lose a significant portion of that gain within 6-8 weeks as the surrounding musculature returns to compensatory patterns. We'd love to help you protect what you've worked for. Reply here or book online.' Notice what this message does: it is specific (her numbers, not generic averages), it names a real consequence, and it frames the re-booking not as a sales pitch but as loss protection, which is psychologically far more motivating, drawing on the well-established principle of loss aversion.
At a workflow level, this requires three practical components. First, your practice management software or a connected platform needs to flag patients who have missed two or more appointments without a future booking, these are your lapsed patients, and the window for effective feedback-based re-engagement is typically within the first four to six weeks of lapsing, before the habit of not attending becomes entrenched. Second, your practitioners should document progress metrics in a format that can be automatically retrieved and inserted into communications. Third, you need message templates that are condition-specific, because a feedback message for a post-surgical knee rehab patient reads very differently from one for a patient managing chronic neck tension through regular osteopathy.
For practices implementing this without enterprise-level software, a simpler version works well: at the end of each session, the practitioner verbally summarises progress to the patient ('Your dorsiflexion has improved 15% since we started, you're tracking ahead of where most patients are at this point') and this summary is captured in the session notes. When the patient lapses, a trained receptionist or automated system references those notes in the outreach message. The behavioural science doesn't require sophisticated technology, it requires consistency in making progress visible and consequences concrete. Even a well-crafted postcard with a patient's key milestone number written on it outperforms a generic 'we miss you' message, because it transforms an administrative reminder into a meaningful, personalised feedback signal.
Get one behavioral science principle per week
Applied to patient retention. Backed by research. No fluff.
Seeing It in Action
Marcus, a 41-year-old project manager, was referred to a chiropractic clinic after months of chronic lower back pain aggravated by long hours at a standing desk. Over seven sessions across ten weeks, his practitioner documented consistent improvements: his Oswestry Disability Index score dropped from 42% (moderate disability) to 18% (minimal disability), and his reported pain on a numerical rating scale moved from a 7 to a 2. Feeling dramatically better, Marcus cancelled his eighth appointment and didn't rebook. From his perspective, the problem was solved.
Five weeks later, Marcus received an automated message through the clinic's patient engagement platform. It referenced his actual scores, specifically that his disability index had dropped 57% over his treatment, and included a brief, plainly worded note explaining that patients who reach minimal disability scores and then discontinue without a maintenance or discharge plan show a statistically elevated risk of symptom recurrence within three months, often returning at a higher disability score than their discharge point. The message wasn't alarmist; it was informative. It closed with a single link to book a 20-minute progress review at a reduced rate.
Marcus booked within 48 hours. At the review appointment, his practitioner identified two postural patterns that had already begun to revert and updated his home exercise programme. Marcus subsequently committed to a monthly maintenance schedule. Twelve months later, he had not experienced a significant relapse. The feedback loop didn't guilt him or pressure him, it simply gave him information he didn't have: that his subjective experience of being 'fixed' was not the same thing as his clinical discharge status, and that the gap between those two things had a predictable, avoidable consequence.
Your Action Plan
- 1Audit your current session documentation to identify two or three measurable progress metrics your practitioners already record for your most common presentations, these become the foundation of your feedback messages.
- 2Set up an automated flag in your practice management system to identify patients who have not attended or rebooked within 21 days of their last appointment, creating a reliable trigger for feedback-based outreach.
- 3Write condition-specific feedback message templates that include a placeholder for the patient's actual progress data, a plain-language explanation of what typically happens when treatment stops at that stage, and a single, low-friction call to action such as a direct booking link.
- 4Train your front desk and clinical team to verbalise progress feedback at every session, practitioners should close each appointment with a brief, specific summary of measurable improvement so that progress feels real and owned by the patient before they ever receive a written follow-up.
- 5Review the re-engagement rate of lapsed patients who receive feedback-based messages against those who receive generic recall messages every quarter, and refine your message copy based on which specific data points and framings produce the highest rebooking rates.
Key Takeaway
Patients don't abandon their recovery out of indifference, they abandon it because no one has made the cost of stopping as visible and personal as the relief of feeling better, and fixing that is a design problem your practice can solve.
Related Principles
Default Effects: How Auto-Scheduling Boosts Rebooking Rates
Nudge · Richard H. Thaler & Cass R. Sunstein
People overwhelmingly stick with whatever option is presented as the default. When no active choice is required, inertia wins.
Status Quo Bias: Why Patients Stick with Routines (and How to Use It)
Nudge · Richard H. Thaler & Cass R. Sunstein
People prefer the current state of affairs and resist change, even when change would benefit them. Disrupting their routine requires effort they instinctively a
System 1 vs. System 2 Processing: Design Messages for Fast, Intuitive Patient Decisions
Thinking, Fast and Slow · Daniel Kahneman
Most daily decisions are made by the fast, automatic System 1. People don't deliberate about routine health appointments, they either have an automatic habit
Loss Aversion: Frame What Patients Lose to Drive Rebooking
Thinking, Fast and Slow · Daniel Kahneman
People feel the pain of losing something roughly twice as strongly as the pleasure of gaining the equivalent. Losses loom larger than gains.
