Why Patients Miss Appointments: Lessons from 100,000 Healthcare Records
peris ndanu
In today’s fast-paced world, missed medical appointments can cost both patients and healthcare providers time, money, and even lives. Curious about the patterns behind no-shows, I analysed over 100,000 real-world appointment records using SQL and visualised the findings in Looker Studio.
The goal? Understand the behavioural and logistical reasons behind missed appointments and recommend actionable solutions.
What the Data Told Us
The dataset included key fields such as age, gender, appointment dates, SMS reminders, health conditions, and whether the patient showed up or not.
Before diving in, I engineered new features like:
Wait Time: Days between booking and appointment
Age Groups: Binned from toddlers to seniors
Weekday Breakdown: To spot day-of-week trends
This allowed for a more focused exploration of patterns.
Visualising the Problem
The interactive dashboard revealed compelling stories:
| Insights | Key Findings |
| Day Of Week | Most no-shows occurred on Monday, Tuesday, and Wednesday |
| Age Group | Young adults (15–30) had the highest no-show rates |
| SMS Reminder | Surprisingly, those who received SMS reminders no-showed more often |
| Wait Time | The longer the wait, the higher the no-show probability |
| Health Conditions | Conditions like diabetes or hypertension had no strong predictive impact |
My Favourite Insight
The counterintuitive finding was that SMS reminders didn’t reduce no-shows—in fact, the no-show rate was ~62.3% for those who received an SMS versus ~37.7% for those who didn’t.
This raises a question: Is the message format ineffective? Or is there a behavioural bias where people receiving reminders are already disinterested?
This discovery shows the power of digging deeper into assumptions using data.
Final Recommendations
Based on the analysis, I recommend:
Shorten wait times between booking and appointments
Test new SMS formats (e.g., conversational, emoji-rich, personalised)
Target young adults with behaviorally informed reminders or incentives
Add filters like day-of-week when overbooking to reduce loss
Invest in deeper behavioural research, not just medical profiling
Key Takeaways
Behavioural and logistical issues are stronger no-show predictors than health factors.
Longer delays between booking and appointment increase the likelihood of no-shows.
Reminder systems need to be evaluated, not just implemented.
Explore the Dashboard + Full Report
👉 View Looker Studio Dashboard
👉 Read Full Case Study on GitHub
Final Thoughts
This project helped me combine SQL analytics, data storytelling, and dashboard design to uncover real-world healthcare insights. Data isn’t just about numbers—it's about asking the right questions and communicating solutions.
Have you ever missed a doctor’s appointment? What could’ve helped? I’d love to hear your thoughts!
Subscribe to my newsletter
Read articles from peris ndanu directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by