Making use of Data to Reduce No-shows in Healthcare


No where is a no-show as costly as it is in healthcare. Expensive resources sit idle. Most providers come up with innovative ways to fill the empty spots (calling people on a waiting list, overbooking in the first-place). Overbooking does alleviate the no-show problem but it creates other problems: long wait times and potentially lower quality of care (reduced time with the clinician). Thus, preventative measures are an absolute must when tackling the no-show problem.

Preventative measures should include reminding patients of their appointment. Not only is this the most commonly practiced preventative measure, it is also the most successful.. Some providers employ this technique to the fullest, texting, emailing, and phone calling patients, sometimes repeatedly. Although some patients dislike excessive use of the technique, even counting it as an intrusion, most are obliged to make their appointment.

Online appointment scheduling software that is available to patients can also alleviate the problem. As patient engagement increases, so does commitment. Online schedulers can be configured to synchronize with the patient’s personal calendar, and here also, automatic reminders can be setup.

Can the no-show problem also be tackled at the root? WHY are patients missing their appointments? This question might not be answerable on a case by case basis, but even if it could, the variability of the causes would diminish the value of the information. One proposal is that providers analyze patient data to identify trends which could reveal patterns and causes of no-shows. This involves gathering data from the patients when they first contact the provider to make the appointment and then throughout their entire experience with the provider. Statistical and quantitative analysis can then be done on the data to surface information that can aid in preventing no-shows.

Gathering incident data can include:

  • Data that qualifies and quantifies the urgency of the patient’s health problem.
  • Day of the week, time of the month, time of day of the scheduled appointment.
  • Day and time data for the patient’s initial contact to make the appointment.
  • Demographics data such as age, income, and distance to provider should be included.

Correlation to no-shows can be studied and by means of regressions and other techniques, general causes can begin to surface. For example, it might be found that 18-22 year-olds miss appointments the most when those appointments are between 8am and 10am. Also, patients who live farther from the provider might be found to miss more appointments than those living near. Armed with this type of data, providers can schedule around these factors (even if they are statistically significant, they are still not causes but only possible factors), and possibly reduce no-shows. Certain age-groups can be scheduled at certain times of the day, etc.

For my second example of patients who live farther missing more appointments, I am not proposing that providers drop these patients. The other preventative measures listed at the beginning of this post can be stressed to strengthen the patient’s commitment. In addition, overbooking can even be employed for population sub-groups most likely to no-show versus general overbooking. Thus, making use of data and analysis can alleviate no-shows as well as contribute to better use of other preventative measures.

Reducing No-shows may also be achieved by removing barriers for the patients such as reducing time to appointments and simplifying the appointment scheduling process. As long as the collecting of data does not present a barrier to the patient, more data is deemed beneficial. The data can be analyzed to identify possible operational and even clinical improvements.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>