Colon Cancer
Conditions
Keywords
quality improvement, Organization and Administration, Costs and cost analysis, Patient satisfaction, Health plan implementation
Brief summary
In this research study, investigators use colonoscopy as a case example to evaluate a predictive overbooking model derived using patient-level predictors of absenteeism. The no-show overbooking intervention employs a logistic regression model that uses patient data to predict the odds of no-showing with 80% accuracy. These projected no-show appointments will be overbooked by clerks for patients who agree to join a fast track short-call line. By rapidly processing endoscopy patients and moving them out of traditional slots, investigators predict more scheduling slots would become available for patients awaiting colonoscopy.
Detailed description
Patient no-shows are especially common in VA gastrointestinal (GI) endoscopy units, where both open-access endoscopy scheduling and patient dislike of procedures contribute to high absenteeism. In this proposal, investigators use endoscopy as a case example to evaluate a predictive overbooking model derived using patient-level predictors of absenteeism. The no-show overbooking intervention employs a logistic regression model that uses patient data to predict the odds of no-showing with 80% accuracy. These projected no-show appointments will be overbooked by clerks for patients who agree to join a fast track short-call line. However, patients scheduled for upper endoscopies in the fast track assume a small risk of service denial on the day of their overbooking in case of inaccurate predictions. If this occurs, the patient is guaranteed service in the next available position and is assured of having a shorter wait time. Patients scheduled for colonoscopies will never be turned down but may experience delays in the waiting room the day of their fast track appointment. By rapidly processing endoscopy patients and moving them out of traditional slots, investigators predict more scheduling slots would become available for patients awaiting colonoscopy. Investigators propose to conduct a prospective, 24-month, interrupted time series (ITS) trial in the WLAVA (West Los Angeles Veterans Administration) GI clinic endoscopy unit. During intervention periods, investigators will activate the no-show predictive overbooking strategy described above. Investigators will compare outcomes between scheduling strategies, including differences in percent utilization of capacity (primary outcome), number of Veterans served, mean patient lag time between scheduling and procedure, number of unexpected service denials (bumps) from no-show predictive overbooking, and direct costs of care. Investigators will analyze differences using both traditional univariate and multivariate approaches, and using autoregressive integrated moving average (ARIMA) analyses to adjust for auto-correlations in ITS data.
Interventions
OTHERPredictive no-show overbooking
During intervention period, every Veteran scheduled for an endoscopy will be offered fast-track offer, which gives them a chance to get their endoscopy procedure done earlier than usual scheduling by overbooking their appointment in a predictive no-show slot.
Sponsors
VA Office of Research and Development
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Patients who are scheduled for upper endoscopy and agree to the terms of fast track offer.
Exclusion criteria
* If a patient expresses concern about service denial, confusion about the bargain, or refuses to participate, the investigators will schedule these patients routinely.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Percentage of GI Clinic Capacity Filled | After 12 months of running study in clinic | Investigators' primary objective will be to evaluate the impact of no-show predictive overbooking on percentage of the GI endoscopy clinic that are filled on a given day. Days where at least one Fast-tracked patient attended an appointment were compared to days where only Control patients attended appointments. Percentage of GI Clinic Capacity is calculated as the number of appointments completed divided by number of appointment spots available on a given day. This percentage was compared between Fast-tracked days and Control days, using data from 1672 patients. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Scheduling-to-procedure Lag Time | After 12 months of running study in clinic | The investigators will calculate the mean daily lag time for all colonoscopy and upper endoscopies performed per day |
| Daily Service Denials (Bumps) | After 12 months of running study in clinic | The investigators will compare the number of patients bumped per day between scheduling approaches |
| Advanced Adenoma Detection/Cecal Intubation Rates | After 20 months of running study in clinic | The investigators will compare daily advanced adenomatous polyp detection and daily cecal intubation rates between groups. |
| Length of Workday | After 12 months of running study in clinic | Length of Workday in hours (comparing days with Fast-Tracked Appointments to Control days without) |
| Cost Comparisons | After 12 months of running study in clinic | For cost comparisons, the investigators will aggregate total provider overtime costs for colonoscopies performed. Cost is reported per day. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Fast-tracked Patients who volunteer to enroll in fast-track line, which gives them an opportunity to overbook their appointment for endoscopy earlier in a predictive no-show slots.
Predictive no-show overbooking: During intervention period, every Veteran scheduled for an upper endoscopy will be offered fast-track offer, which gives them a chance to get their endoscopy procedure done earlier than usual scheduling by overbooking their appointment in a predictive no-show slot. | 180 |
| Control Patients who are scheduled routinely | 4,855 |
| Total | 5,035 |
Baseline characteristics
| Characteristic | Fast-tracked | Control | Total |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 68 Participants | 2043 Participants | 2111 Participants |
| Age, Categorical Between 18 and 65 years | 112 Participants | 2812 Participants | 2924 Participants |
| Age, Continuous | 59.3 years STANDARD_DEVIATION 13.6 | 62.2 years STANDARD_DEVIATION 10 | 62.1 years STANDARD_DEVIATION 10.1 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 17 Participants | 565 Participants | 582 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 163 Participants | 4290 Participants | 4453 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 2 Participants | 35 Participants | 37 Participants |
| Race (NIH/OMB) Asian | 5 Participants | 109 Participants | 114 Participants |
| Race (NIH/OMB) Black or African American | 70 Participants | 1584 Participants | 1654 Participants |
| Race (NIH/OMB) More than one race | 2 Participants | 41 Participants | 43 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 2 Participants | 28 Participants | 30 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 16 Participants | 644 Participants | 660 Participants |
| Race (NIH/OMB) White | 83 Participants | 2414 Participants | 2497 Participants |
| Region of Enrollment United States | 180 participants | 4855 participants | 5035 participants |
| Sex: Female, Male Female | 12 Participants | 206 Participants | 218 Participants |
| Sex: Female, Male Male | 168 Participants | 4649 Participants | 4817 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 180 | 0 / 4,855 |
| serious Total, serious adverse events | 0 / 180 | 0 / 4,855 |
Outcome results
Primary
Percentage of GI Clinic Capacity Filled
Investigators' primary objective will be to evaluate the impact of no-show predictive overbooking on percentage of the GI endoscopy clinic that are filled on a given day. Days where at least one Fast-tracked patient attended an appointment were compared to days where only Control patients attended appointments. Percentage of GI Clinic Capacity is calculated as the number of appointments completed divided by number of appointment spots available on a given day. This percentage was compared between Fast-tracked days and Control days, using data from 1672 patients.
Time frame: After 12 months of running study in clinic
Population: These 1672 patients are a subset of all patients who were enrolled in this study. We dropped data on a first wave of participants (69 Fast-tracked and 3294 Controls) because of problems incorporating recruitment strategy into clinic. However, data from these individuals was used to build predictive model.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Fast-tracked | Percentage of GI Clinic Capacity Filled | 99.6 percentage of clinic capacity filled | Standard Deviation 19.4 |
| Control | Percentage of GI Clinic Capacity Filled | 86.4 percentage of clinic capacity filled | Standard Deviation 21.6 |
Comparison: We counted days containing at least one Fast-tracked participant as a Fast-tracked day for the sake of analysis at the clinic level. Our null hypothesis is that these days will not differ in terms of percentage of clinic capacity filled. We are powered to detect a 0.5 Standard Deviation (SD) difference in clinic capacity (Type I error rate = 5%; Power = 81%), with a equal ratio of Experimental and Control days.p-value: <0.0001t-test, 2 sided
Secondary
Advanced Adenoma Detection/Cecal Intubation Rates
The investigators will compare daily advanced adenomatous polyp detection and daily cecal intubation rates between groups.
Time frame: After 20 months of running study in clinic
Population: We only collected data on polyp detection for the first half of our Fast-tracked participants and all Controls seen over the same time period (4897 in total).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Fast-tracked | Advanced Adenoma Detection/Cecal Intubation Rates | 2.35 Number of Polyps Detected per patient | Standard Deviation 1.58 |
| Control | Advanced Adenoma Detection/Cecal Intubation Rates | 2.88 Number of Polyps Detected per patient | Standard Deviation 2.56 |
Comparison: We calculated this measure on the patient level, but were hampered by a large sampling ratio (more than 50:1). Our null hypothesis is that the number of polyps detected will not differ between groups. We are powered to detect a 0.25 SD difference in polyp count per patient (Type I error rate = 5%; Power = 80%).p-value: 0.05t-test, 2 sided
Secondary
Cost Comparisons
For cost comparisons, the investigators will aggregate total provider overtime costs for colonoscopies performed. Cost is reported per day.
Time frame: After 12 months of running study in clinic
Population: These 1672 patients are a subset of all patients who were enrolled in this study. We dropped data on a first wave of participants (69 Fast-tracked and 3294 Controls) because of problems incorporating recruitment strategy into clinic. However, data from these individuals was used to build predictive model.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Fast-tracked | Cost Comparisons | 63.01 dollars | Standard Deviation 107.01 |
| Control | Cost Comparisons | 36.88 dollars | Standard Deviation 101.73 |
Comparison: We counted days containing at least one Fast-tracked participant as a Fast-tracked day for the sake of analysis at the clinic level. Our null hypothesis is that these days will not differ in terms of cost per day. We are powered to detect a 0.5 SD difference in clinic capacity (Type I error rate = 5%; Power = 81%), with a equal ratio of Experimental and Control days.p-value: 0.11t-test, 2 sided
Secondary
Daily Service Denials (Bumps)
The investigators will compare the number of patients bumped per day between scheduling approaches
Time frame: After 12 months of running study in clinic
Population: These 1672 patients are a subset of all patients who were enrolled in this study. We dropped data on a first wave of participants (69 Fast-tracked and 3294 Controls) because of problems incorporating recruitment strategy into clinic. However, data from these individuals was used to build predictive model.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Fast-tracked | Daily Service Denials (Bumps) | 0 participants |
| Control | Daily Service Denials (Bumps) | 29 participants |
Comparison: We did not explicitly power this study to detect differences in the number of service bumps, but we can compare them using Fisher's Exact Testp-value: 0.49Fisher Exact
Secondary
Length of Workday
Length of Workday in hours (comparing days with Fast-Tracked Appointments to Control days without)
Time frame: After 12 months of running study in clinic
Population: These 1672 patients are a subset of all patients who were enrolled in this study. We dropped data on a first wave of participants (69 Fast-tracked and 3294 Controls) because of problems incorporating recruitment strategy into clinic. However, data from these individuals was used to build predictive model.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Fast-tracked | Length of Workday | 8.31 hours | Standard Deviation 1.32 |
| Control | Length of Workday | 7.84 hours | Standard Deviation 1.32 |
Comparison: We counted days containing at least one Fast-tracked participant as a Fast-tracked day for the sake of analysis at the clinic level. Our null hypothesis is that these days will not differ in terms of length of workday. We are powered to detect approximately 0.5 SD difference in workday length in hours (Type I error rate = 5%; Power = 81%), with a equal ratio of Experimental and Control days.p-value: 0.024t-test, 2 sided
Secondary
Scheduling-to-procedure Lag Time
The investigators will calculate the mean daily lag time for all colonoscopy and upper endoscopies performed per day
Time frame: After 12 months of running study in clinic
Population: These 1672 patients are a subset of all patients who were enrolled in this study. We dropped data on a first wave of participants (69 Fast-tracked and 3294 Controls) because of problems incorporating recruitment strategy into clinic. However, data from these individuals was used to build predictive model.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Fast-tracked | Scheduling-to-procedure Lag Time | 8.3 days | Standard Deviation 24.3 |
| Control | Scheduling-to-procedure Lag Time | 10.8 days | Standard Deviation 24 |
Comparison: We calculated this measure on the patient level. Our null hypothesis is that the lag time between scheduling and appointment, measured in days, will not differ between groups. We are powered to detect a 0.3 SD difference in lag time (Type I error rate = 5%; Power = 84%), accounting for the large sampling ratio (approximately 14:1).p-value: 0.29t-test, 2 sided