HLTH 6882 Week 10 Improvement Measurement Plan Example

Measurement Plan: Outcome, Process and Balancing Measures for the Booking Change at Prairie Bend

Student Name

Walden University

HLTH 6882: Improving Healthcare Performance Through Social Entrepreneurship

Instructor Name

Month Day, Year

Measurement Plan: Outcome, Process and Balancing Measures for the Booking Change at Prairie Bend

Introduction

This plan sets out how Prairie Bend Community Hospital will know whether the booking change worked, before the change begins across the visiting specialist clinic. It names one outcome measure, three process measures and two balancing measures. Each is defined so that two analysts working separately would reach the same figure. The outcome measure uses exactly the definition of a lost slot set in the gap analysis, so results after the change can be compared with the 12-month baseline. Every measure goes on a run chart, a display well suited to learning from variation over time in health care processes (Perla et al., 2011). All figures describe a composite hospital and are illustrative.

Outcome Measure

The outcome measure is the kept rate: the share of booked specialist slots that end in a completed visit, reported per session day.

Table 1

Definition of the Outcome Measure

Process Measures

Three process measures show whether the change is being carried out as designed. Each is linked to one component.

Table 2

Definitions of the Process Measures

Why These Process Measures

Should the kept rate stay flat, these three figures point to the part of the change that broke down. If short-notice offers are rarely made within a day, the list is not being used, and refilled cancellations cannot be expected. If reminder reach is low, the reminder is not reaching patients, perhaps because phone numbers are out of date. The median days from referral to visit shows whether the window is shortening the wait overall, which matters because advanced access work treats the delay between request and visit as a central measure of access (Murray & Berwick, 2003). A Cochrane review found reminders improved attendance only when they reached patients (Gurol-Urganci et al., 2013), so reach is the relevant process measure rather than reminders sent.

Balancing Measures

A change that fixes one problem can create another, so two balancing measures watch the places most likely to suffer.

Registration wait time on clinic days is the median minutes from a patient's arrival at registration to completed check-in on specialist session days, for all patients registering that day, recorded by the registration lead from check-in time stamps. It watches for the effect of moving the scheduler off registration coverage on session days.

Sessions running past contracted end is the share of session days on which the last patient was seen more than 30 minutes after the contracted session end time, recorded by the outpatient manager from the visit log. It watches for overbooking, since a full schedule refilled from the short-notice list may run long and strain the specialist group's goodwill.

A third possible harm, urgent referrals waiting on the holding list, is not a measure but a stop rule: any instance is reviewed the same day, as set out in the pilot design.

Run Charts

Figure 1 shows the template for the outcome measure, using the 12 baseline months from the gap analysis. The baseline median of 71.5% is the center line. When the change starts, the chart will add an annotation at the pilot start on orthopedic sessions and a second annotation when the change extends to all specialties, so that any shift can be read against the date each part began.

Figure 1

Run Chart Template: Kept Rate by Month, Baseline Year (Illustrative)

What Counts as a Real Shift

The plan uses standard run chart rules to tell a real change from ordinary variation (Perla et al., 2011). Four signals are used, following Perla and colleagues. A shift is a stretch of six or more session results all landing on one side of the center line; results sitting exactly on the line neither count nor break the stretch. A trend is five or more results that keep moving the same way, every one above, or every one below, its predecessor. The count of runs, meaning unbroken sequences on one side of the line, is compared with the range expected by chance for that many results. An astronomical point, a result plainly out of line with every other, is investigated before anyone draws a conclusion from it.

The baseline year itself shows why these rules matter. Months 4 to 6 were well above the median, at 87%, 91% and 89%, but three points do not make a shift, and nothing in the clinic's records explains them; they may have reflected a quiet season with shorter waits. A post-change result will count as a real improvement only if the kept rate shows a shift above the baseline median, sustained for at least six session days for orthopedics in the pilot and six months for the clinic as a whole. A single good month will not be treated as success, and a single poor one will not be treated as failure.

Reporting

The outpatient manager will report the outcome, process and balancing measures to the director of outpatient services after each orthopedic session during the pilot and monthly after the change extends. After adoption, the hospital quality committee's monthly report will carry the kept rate, alongside the existing measures it reviews.

References

Gurol-Urganci, I., de Jongh, T., Vodopivec-Jamsek, V., Atun, R., & Car, J. (2013). Mobile phone messaging reminders for attendance at healthcare appointments. Cochrane Database of Systematic Reviews, (12), Article CD007458. https://doi.org/10.1002/14651858.CD007458.pub3

Murray, M., & Berwick, D. M. (2003). Advanced access: Reducing waiting and delays in primary care. JAMA, 289(8), 1035-1040. https://doi.org/10.1001/jama.289.8.1035

Perla, R. J., Provost, L. P., & Murray, S. K. (2011). The run chart: A simple analytical tool for learning from variation in healthcare processes. BMJ Quality & Safety, 20(1), 46-51. https://doi.org/10.1136/bmjqs.2009.037895