Tuesday, November 11, 2014

3 Reasons Why Predictive Patient Scheduling is the Key to Primary Care Efficiency

In the 30 years I've been in healthcare there has been little innovation to optimize provider schedules and patient flow.  Patients are still frustrated by long waits and provider organizations are grappling with capacity constraint issues and rising operating costs. The most significant and often overlooked cause of patient flow inefficiency is the prevailing practice for scheduling appointments.  Visit durations are not based on data of actual time needed with the patient, and existing scheduling systems enable schedulers to enter their own interpretation of the visit reason.  This practice results in inaccurate scheduling, and causes uneven flow throughout the day.  Below are three reasons why predictive patient scheduling is the key to improving primary care efficiency.   

The Problem – unruly scheduling practice
Practice management systems and on-line scheduling services are designed to match a patient with an open appointment slot.  They are not designed to ensure that the right amount of time is given for each appointment.   They enable only broadly defined categories of appointment durations (e.g., allocating 30 minutes for new patients, and 15 minutes for established patients).  Most clinics establish written guidelines to further define visit categories and help schedulers determine the amount of time for each visit. 

There are several problems with written scheduling guidelines.  First, they are often incomplete.  Limited to one page per clinic or provider, they call out only a fraction of the reasons for primary care visits. Second, the visit durations are based on a “best guess” rather than an actual measure of provider time spent in each visit.   Third, it’s a manual process.  Without good data, updating guidelines is time-consuming and often the subject of much debate among providers. 

In practice, the guidelines are often not followed because they are difficult to access, out of date, or usurped by (unwritten) provider preferences.  Existing systems enable any scheduler to enter any visit reason and any duration, yielding significant variation.  On-line scheduling services are more rigid, requiring patients to select from a short generalized list of visit reasons.  Visits are often assigned the same or the longest appointment duration, requiring clinic staff to adjust the duration after the fact.  Regardless of how appointments are scheduled, today’s scheduling practices are unruly, and inevitably result in variation in provider schedules and inefficiency in patient flow.  

Optimizing flow – the case for predictive patient scheduling
Not many industries can afford to operate their core business with such a random process.   In today’s environment - with rising costs, provider shortages, and consumerization – it’s no longer affordable for provider organizations either.  To succeed, they need to optimize patient flow.  In much the same way as a shipping company packs a container to optimize space, providers need accurate scheduling to optimize patient flow.  This starts with predictive patient scheduling, a measured understanding of visit durations based on the actual time providers spend with patients (including time for documentation).  This data-driven rules-based scheduling approach is established by: 
  1. measuring provider time for each visit reason,
  2. analyzing the data patterns, and  
  3. incorporating evidenced-based rules into the scheduling process.  
Here’s an example:  The average primary care visit for an established diabetic patient takes 15 minutes.  With deeper analysis, the data shows a bi-modal distribution pattern – stable diabetic visits average 10 minutes, and unstable diabetic visits average 20 minutes.  Whether the patient schedules their appointment on the phone or on line, their status (stable vs. unstable) is ascertained at the time the appointment is scheduled.

Predictive patient scheduling helps:  1) improve patient experience, 2) reduce capacity constraints, and 3) reduce costs.  Here’s how:

Improve patient experience
Scheduling the right amount of time for each patient (e.g., 10 or 20 minutes vs. an average of 15 minutes in the diabetes example above) reduces the peaks and troughs that occur when back-to-back patients are scheduled with too little or too much time.  For example, if 5 consecutive patients need 5 more minutes than scheduled, the care team will be 25 minutes late starting their sixth visit.  Scheduling accurately streamlines daily flow, reduces wait time and improves the patient experience.

Reduce capacity constraints
With unpredictable flow, providers often buffer appointment durations to run on time.  Scheduling visit durations accurately reduces this “hidden access” and enables schedulers to see more appointment opportunities.  By reducing hidden access just 0.3 visits per hour (e.g., moving from 2.5 to 2.8 patients/ hour), for example, a full time primary care physician can see 2 more patients per day (e.g., moving from 20 to 22 patients/day).  For a health system with 150-physicians, the cumulative effect is significant.  It would open 60,000 more visits per year and help alleviate capacity constraints felt by most health systems today.

Reduce costs
Scheduling accurately and optimizing flow, reduces operating costs.  With a more predictable flow, there is less pressure to staff for peaks and less need to pay for overtime.  Reducing hidden access and improving throughput reduces the cost per visit.  In the example above, 60,000 more visits accommodated by existing providers is the equivalent of adding 16 providers – without the added cost! 

In the absence of intelligent visit data, individual providers or clinic sites solve their scheduling and flow problems individually.  This duplicative process is not only expensive, but also results in variable resource use (e.g., number of rooms or staff per provider) and variable outcomes (e.g., access, productivity, patient experience) across the network.  With comparable visit data, health systems can identify best practices and implement standard work to reduce costly variation.

The prevailing practice for scheduling appointments, whether via phone or on line, produces uneven flow, constrained access and higher costs.  Predictive patient scheduling is the key to optimizing patient flow and the mainspring to primary care efficiency.  It improves the patient and provider experience, reduce capacity constraints and reduce the cost to operate.

Tuesday, March 19, 2013

CadenceMD is joining 9 Mile Labs

CadenceMD is excited to be part of the first cohort for 9 Mile Labs, a new startup incubator/accelerator in Seattle, located at Surf Incubator. We're at Surf already, so we're just moving down the hall.

9 Mile Labs is a great fit for CadenceMD because they're focusing entirely on B2B companies, and the partners (Tom Casey, Kevin Croy, Enrique Godreau III, Sanjay Puri and Sandy Sharma) and mentors will be helping companies refine their business models, fine tune their go to market strategies and generally help the companies accelerate.

We'll be joining an extremely diverse group of eight other B2B startups: AeroMetric Wireless, AMP Tablet Solutions, Appuri, Comr.se, GIVINGtrax, MeritShare, Ombitron, and SpinRiot. We've met some of them already and we can't wait to get started working with them.

Wednesday, January 9, 2013

Improving Doctors’ Schedules for Better Throughput and Access

Improving scheduling can have a major impact on patient flow efficiencies, throughput, access, and patient experience. Physicians and healthcare executives, burdened by myriad new healthcare requirements, are feeling increased pressure to improve operational effectiveness and should not overlook this opportunity.

To understand the size of this opportunity, consider a few examples for a primary care doctor in a traditional practice:
  1. Leaving one hour of appointments per week unfilled costs $10,000+ per year in missed revenue.
  2. Running an average of 30 minutes late per day costs $3,000+ per year in staffing costs.
  3. Losing just one patient costs $1600+ per year in lost revenue, and 87% of patients say that wait time is a significant factor affecting their overall visit experience.
That is a lot of money to leave on the table, and the stakes have never been higher. Achieving core operational efficiencies and sustainable improvements in workflow is key to patient and physician retention, and clinic profitability.

Two common scheduling mistakes:

Focusing only on productivity and access:  Physician organizations are focused – and should be – on improving throughput and access. However, if this is done without understanding the root causes of delays and inefficiencies in the patient flow process, it can backfire. For example, when asked to add another patient, physicians rarely add time. Instead, they typically double-book appointments. This causes them to run further behind, which, in turn, leads to unhappy patients, greater physician and staff stress, and longer days (with more overtime pay). Exhausted by an unsustainable pace, the reaction is often to reduce – rather than increase – the amount of template time available. The net result is actually reduced productivity and access.

Scheduling appointments with insufficient data:  Physician schedules are typically set up in 10, 15 or 20-minute increments, but visits are scheduled with insufficient information to know the actual time needed for the appointment. Many clinics establish increments for short and long appointment “types” (often based simply on the average or mode). Other clinics give all patients the same time (e.g., 20 minutes) regardless of the reason for the visit. In either case, when appointments are set without full knowledge of the specific reason for each visit and actual provider time needed, the results are the same: a large number of “scheduling outliers” (appointments scheduled with either too little or too much time). These outliers (often >70% of the appointments) create an uneven flow during the day. Consider, for example, if five patients in a row require five more minutes than scheduled. The doctor is already 25 minutes behind half way through their morning! Conversely, if patients need less time than scheduled, doctors start to double book to fill perceived extra time. Both scenarios create inefficiencies, and fuel the ongoing tension between providers and schedulers (receptionists or call centers). These so-called “scheduling errors” are not the fault of the scheduling staff, but the scheduling process itself.

Two surefire ways to improve scheduling and operational efficiency:

Measure and analyze patient flow:  There is a surprising lack of understanding of the patient flow process. Understanding starts with measuring the visit cycle time in more detail, including the time it takes to perform each step in the visit process (e.g., check in, rooming, provider exam, etc.) and the time the patient waits between each step. The data then needs to be analyzed for issues and anomalies, and delays and bottlenecks. For example, if patients arrive an average of 4 minutes early and it takes 9 minutes to complete check-in and rooming, the doctor is 5 minutes behind schedule by the time they enter the exam room. Once the problem is understood, adjustments can be made. Alternatively, suppose it takes 9 minutes on average to room patients for one doctor and 3 minutes for another. What are the reasons for the variation? Is it that the first assistant is inefficient or that the second one is not performing all the rooming tasks, leaving them for the provider to do? Or is it something else? The analysis and ensuing discussion will help eliminate delays, optimize each team member’s role, and streamline the visit process.

Schedule visits according to the expected exam duration:  By measuring patient flow, you now have the actual provider face-to-face time with the patient, and the ability to analyze actual vs. scheduled exam duration by reason for visit. Detailed analysis of this data will typically reveal two things. First, the actual duration of exams frequently doesn’t match the scheduled 15 or 20 minute appointments. For example, the average diabetes visit might be scheduled for 15 minutes, but the providers’ time in the exam room is typically around 10 minutes for a stable diabetic patient, and around 20 minutes for an unstable diabetic patient. Second, the providers’ actual exam duration by reason for visit is generally consistent. Choosing a scheduling increment that enables appointments to be scheduled around the bimodal distribution, and scheduling based on the specific reason for the visit and expected exam duration, will result in fewer scheduling outliers and a more even flow throughout the day. This requires that schedulers have the right tools and education to appropriately schedule, but with greater clarity on the amount of time actually needed, this becomes more manageable.

Running on time is achievable. When doctors run on time, they are more efficient – and patients, doctors, and staff have a better experience. Understanding and optimizing physician scheduling and patient flow is the key to sustainable improvements in throughput, access and satisfaction.

Bonnie Cech, CEO and Founder

This post is cross-posted from Washington Healthcare News, February 2013.
You can download this article as a PDF.

Welcome to CadenceMD

Welcome to the CadenceMD blog, where we'll discuss scheduling and patient flow in physician practices, including tips, metrics, and best practices. Our goal is to make it easier for providers to run on time and improve productivity and throughput by sharing what we've learned and encouraging others to do the same. This is part of the mission of CadenceMD, providing innovative technology solutions that enable physician organizations to improve operational efficiency and patient experience.

Why the name CadenceMD? Good things happen when doctors run on time – patients are happier, it's less stressful for doctors and staff, it's easier to manage, and efficiency/productivity goes up! Running on time starts with a regular  sequence of flow – "cadence." When providers have good cadence, they have balance, consistency, and efficiency.

So whether you are on the front line (provider, care team member, call center, or reception staff), supporting staff (IT, finance, or lean/six sigma consultant), or management (healthcare executive or practice manager), we hope you'll benefit from – and join in – the discussion.