A MediMobile Case Study
A five physician nephrology practice implemented MediMobile for charge capture. During this case study, only three CPT codes were used to analyze costs. Using these three CPT codes, this nephrology group increased their annual revenue by $37,473 per physician with an ROI of 1.7 months.
Looking back in history we can find a time when the health care system in the United States was the envy of the free world. We had the finest doctors using the latest technology and practicing in hospitals or offices that supported high quality and affordable care. Insurance was only needed for catastrophic issues since patients could pay for typical medical expenses and malpractice lawsuits were unheard of.
Today’s health care field has become much more complex. Medical advancements have been astonishing, the number of specialties has grown dramatically and the complexity of care has increased as well. Government and HMO oversight has increased to unprecedented levels in an effort to protect patients, increase the quality of patient care and drive down costs. Litigation has placed an air of caution around even the most common patient encounters.
A Merritt, Hawkins & Associates survey of physicians 50 years old and older indicated that only 54 percent of physicians indicated that the quality of health care in the U.S. has “generally improved” over the last 20 years. The same survey found that only 50 percent of physicians would choose medicine as a career if they were starting out today. While managed care has been cited as the number one issue many practitioners face today they also state that the added complexity of making a living is becoming as a major concern. In fact, a rising number of physicians are experiencing a reduction in income based on the overall challenges of practicing in this dynamic environment.
As Pay-for-Performance programs become the norm terms like efficiency, effectiveness, patient satisfaction and ease of doing business will become an integral part of managing a practice. These terms have been the “watch words” of most businesses around the world as they institute change to improve customer service and to manage the financial side of the business. It is time for the health care industry to take a page from general business management and gain the performance improvements that come from optimizing practice management.
Alas you say, “I became a doctor to care for patients and I didn’t expect to spend more time running the business”. Well you are in the majority and to take that thought one step further, when asked 60 percent of the doctors respond that their greatest professional satisfaction is in the relationships that they build with their patients. You exist for your patients so we suggest looking at improving your practice as a method for improving their experience with you. They should be able to expect a comfortable office with efficient service, accurate billing and reasonable availability when they have an urgent question. Providing outstanding health case enhanced with these services and still turning a profit is a challenge for anyone.
Opportunities for Performance Improvement
There are several areas within a typical practice that provide an opportunity for optimization. Research indicates that the most prevalent areas for improvement include; patient service, admissions, medical records, coding, billing, collections and risk/liability, to name a few. We decided to focus on the coding, billing and collections functions since they represent a significant opportunity for managing cash flow with a high potential for solid financial returns.
We needed to understand how big the problem was prior to searching for a solution so we agreed to complete an optimization study. If the study results justified we would first pilot a solution and then complete a full optimization project.
There are six steps included in a typical optimization project; situation analysis, technical investigation, solution formation, financial analysis & decision, implementation, measurement & monitoring. Our sample case to describe this project was a Nephrology Practice that employs five doctors that travel between two hospitals and two dialysis centers. We looked at the past performance of three doctors to benchmark the processes and procedures. We also analyzed the tools used during rounds and for the billing and collections processes. We piloted a mobile software tool with a single doctor and measured the results. Based on those results we rolled the product out to two additional doctors. After their implementation we measured the performance improvement along several dimensions and the financial results were astonishing.
Let’s meet the doctors that have made this practice successful. Doctor M is an early adopter for just about everything but particularly for technology. He will try anything and frequently has the latest cell phone or portable organizer. Doctor H could be called “old school”, he has been in practice for decades and is pretty set in his ways. He carries the reputation of being a sloppy record keeper and frequently leaves his patient files and tracking forms in the trunk of his car. Doctor R is our technophobe. Every time he gets a new piece of software is seems to have problems so he prefers manual methods.
He wants to keep his old cell phone since it finally works well for him after six years. Doctor B is driven, moves quickly and efficiently from patient to patient. Her patients love her and she leads the office in patient encounters. With that success in place she will only use technology if it’s proven to make a difference. And finally, Doctor V is very organized and methodical. She is clinical with processes & record keeping and so her billing and collections track record is fantastic. She is more than willing to make changes but her potential for improvement is much smaller than the other nephrologists.
We began by working with the office manager and CFO for the practice. She acted as our host as we laid out the requirements for a pilot and began the study. We decided to collect a baseline of data on doctor M for 30 days and then have him use the new tool to track patients and keep records for an additional 30 days. We also collected statistical data on doctor H and doctor R to use for comparisons. We could then use the averaged data to run a return on investment calculation and build a financial justification for further investment.
The first step for the project was a situation analysis which consisted of performing an audit of the existing systems and processes to determine the current performance around critical processes and tools. Our interviews of key personnel indicated that there were issues in the following areas:
- Down-coding too frequently when code selection was not clear
- Reduced charges occurred due to coding errors
- Lost charges resulted from misplaced forms
- Lower collection rates were observed when records were incorrect
- Insurance denials increased due to late submissions
- High denied charge write-off rates seemed to track insurance denials
We learned that the current process for these hospitalists required that they capture the patient information, a diagnosis and recommended code on a 4×5 paper form. The forms were collected throughout the day and returned to the office for processing at the next available opportunity. That might be the same day or several days later depending on the patient load and doctors’ schedule. When the forms are delivered some are coded incorrectly or a higher level code is used which reduces the revenue associated with the actual level of care. Frequently the forms are lost, misplaced or illegible and the staff would have to spend time with the doctor to get the forms ready for the billing process. By the time the specialists notice that a lost form is missing, and it can be found or reproduced, and then processed through the insurance company, the total time allowed for processing has frequently been exceeded. When this does occur the insurance company will usually refuse payment.
The coding specialists assigned to collect the information walked through the issues from diagnosis to completed records ready for billing. They estimated the actual lost time related to each of the activities and then we calculated the differences in revenue generated for the optimum situation versus the current situation. The table below illustrates the impact to Doctor M’s available time based on these issues. Of course, loss of billable time affects total annual revenue generated by Doctor M and since his income is tied to billable revenue these issues direct affect his paycheck.
|Avg. min/event||Avg. event/yr||Total (min/yr)|
|Sheets not delivered||15||17||255|
|Time lost to correct||45||188||8460|
The impact to the practice from errors and delays shows up in lost revenue. Some claims are uncollectible due to errors generated in the manual coding process that are not corrected prior to billing. Rejected claims occur when the mandated time for processing has been exceeded and the insurance company denies payment. And finally, after an insurance company denies payment the total bill is frequently beyond the reach of many patients and this increases the potential for additional write-offs. The impact on total revenue is captured in the table below.
|Issue||Percent of Total Revenue|
|Uncollectible (based on errors)|| 0.5%|
|Insurance payment denied (time)|| 1.4%|
|Write-off after denial (time)|| 19.5|
In an effort to quantify the improvement in revenue from solving these problems we analyzed a subset of procedures before and after implementing a solution. We then collected the aggregated revenue impact from all of the improve targets. Given the rough numbers we could begin defining the scale of solution that the practice could afford. But before addressing the total impact let’s take a look at how technology can be applied to solve these problems.
Since most of the patient encounters take place outside of the office we looked for a mobile product that would allow each doctor to capture patient information, record the diagnosis, provide a coding selection, and confirm that the record is ready for billing. Once the confirmation is complete we needed a method to communicate the record to the office staff for review and billing submission to the insurance company. If possible we also wanted to send the billing information to the insurance companies and patients in electronic form. Completing all of these steps electronically would eliminate a large number of manual steps, each with its own risk of errors.
Two of the three doctors, doctor M and doctor H wanted to minimize the number of devices that they carried so a solution that would run on their mobile phone was attractive. Doctor R felt that he would prefer to have a separate device from his phone so that critical calls would not interrupt the data entry process. In observing doctor M’s early trials doctor R felt that moving from data entry to answer a call might cause him to shut down the software and lose data. Therefore the solution should be able to run on a Personal Digital Assistant as well as other mobile devices.
There are several of these types of products on the market. On the surface they seem similar but as usual “the devils in the details”. We quickly narrowed the selection by careful analysis of ease-of-use and typical work flow patterns. The defining factor was complete and reliable mobility.
Most of the software tools on the market use a host computer located in the office and managed by office staff. That presented a problem since most of the staff did not have Information Technology backgrounds or the time to manage the hardware and software. Many of the possible solutions allow some level of mobility but require the user to set the device in a cradle for data synchronization. That would be fine except this practice is so distributed that doctors may not return to the office but once a week so we needed a product that could synchronize the data without a physical connection.
The software functions needed for this application were pretty straight forward so all of the packages were close in most categories. Ease-of-use and some level of customization were required to get all of the doctors from zero to productive use of the tool in a very short time. Support after installation was another factor so we needed to take care in selecting the supplier such that they had these capabilities at a reasonable cost.
We found few suppliers that could meet all of these requirements and only one that could capture the information on a mobile phone synchronize it with the office system without the use of a docking cradle and provide central service for system and data reliability. The selected product is a web-based service so there is no office installation that requires local expertise and on-going technical support. The product is called MediMobile from a company by the same name.
MediMobile has very easy to use screens that let you sort by patient or doctor. It handles scheduling for rounds and allows the transfer of patients between doctors. It provides a customized set of codes so that you don’t have to page through codes that don’t apply to your specialty. It even has text recognition so a description can be automatically completed by the software after typing 3-4 initial characters. It connects to a web application that is used by the office staff and captures the information in a backup database. The on-line synch sends data in smaller packages so it doesn’t tie up your phone for lengthy data transfers. Any time the phone can get cell coverage it will assure the data is synchronized with the central servers.
Another reason that we selected this particular package of hardware and software was that we could get the pilot up and running in a matter of minutes. The first step was to purchase a phone from the list of Windows-based and Palm-based mobile devices. Next we used the phone to log in to the MediMobile web site to load the software. It took a bit longer to select the customized list of nephrologists’ codes but that was automatically loaded over the internet and we were ready to run that same afternoon. After the learning curve from the first phone we found that we could easily get each additional phone loaded and running in under an hour each.
It did not take doctor M long to become proficient with the phone and the software. He was ready to adopt the new tools based only on their ease of use and data security. During the pilot doctor M actually dropped his phone in a swimming pool one evening! He had not been in the office for two days and was very concerned that he had lost 25 patients’ records. He had to replace the phone but all of the data had automatically synch’ed with the web site within minutes of finishing his rounds and none of the data was lost.
After 30 days using the MediMobile software we captured the statistics on doctor M’s patients. What we found was a clear improvement in many of the problem areas. To put the improvements in context we needed a financial model that compared the past state with what we believed we could achieve in a future state.
Financial Analysis & Decision
We first look at the total costs for implementing this type of solution. The costs associated with the project were as follows:
|Project Investment Costs||$1,979|
|– MediMobile System||$ 1,260|
|– Custom Code List||$175|
|– Pocket PC/Phone Hardware||$350|
|*The contingency was included to cover any miscellaneous costs which may not have been predicted at the beginning of the project.|
We assumed that the investment would be evaluated over a five year period. The phone would likely be replaced every two years due to damage or through upgrading to the latest version so an additional cost of $750 was added maintain a conservative case.
The benefits side of the financial analysis starts with an understanding of the total number of hours available for a doctor to use for patient care. That begins with all of the standard work hours per year. Add an additional 20 days per year to cover the weekend call duty that rotates from doctor to doctor. If you then subtract out vacation and time for training or conference attendance and you are left with approximately 1755 hours available for patient care. At an average visit duration over the mix of different required visits you might expect a typical doctor to see approximately 2956 patients per year. Our actual results found a range from 2901 to 4265 patients per year with well over 65% of the visits being for the same procedures.
Rather than try to address all the possible reasons for the variation in patients seen we maintained our focus on the procedural factors that could be consistently improved without negatively impacting patient care. The first area for improvement was the providing the correct coding for the diagnosis that was performed.
We looked at the actual coding splits that took place using the old paper system for coding and billing. We found the 99232 code was used much more frequently than would have been expected. The explanation from the doctors was that when there was any uncertainty it was safer to “down-code” than to code incorrectly at a higher code. Given the nature of government oversight and litigation this is the more cautious route to take even though it reduces the level of revenue generated.
We also found that if paperwork was finalized at the end of rounds it was more common to use the more conservative coding. Since several patients are seen in succession it sets up a situation that is more conducive to errors. From a liability perspective it made sense to have a tool that could populate the forms in a few clicks and therefore facilitate full documentation after every patient encounter.
This table outlines the current state of coding for the three most common codes. The rates are shown for these procedures as well as the actual splits between codes based on the paper system. Since each code is linked to the billing revenue it is clear that coding more visits at 99233 would increase revenue. So the next question we needed to address was, “what are the correct coding splits if the tracking issues are excluded from the equation?”
|TYPICAL Coding & Billing:||Doctor M|
|Typical total billing (per year)||$121,209.19|
The second table was based on the use of coding guidance, better accuracy and more complete coding once the software was in place. The difference in splits indicates the difference between an actual manual system and the implemented solution. Using the actual number of patients and the typical procedure charges we calculated that doctor M had the potential to generate $18,333 in additional billing for a full year.
|OPTIMUM Coding & Billing:||Doctor M|
|Maximum total billing (per year)||$139,542.53|
|Increased Revenue per Encounter||$18,333.34|
The number of coding forms that were lost or misplaced ranged between 5-9% of the total patients seen per year. As the patient load increased the number of visits back to the office was reduced and the number of lost forms increased. It took an average of 45 minutes for a doctor and coding specialist to reconstruct the forms from the hospitals records. Over the course of an entire year the total revenue impact of the lost time was $12,558.
|Lost Forms||Minutes per||Doctor M|
|Time lost to correct||45||188||patients/yr||$12,558.83|
|Increased Revenue per year||$12,558.83|
Three of the problems that had been cited with the paper system were illegible coding, incorrect coding, and paper tracking sheets not delivered to the coding specialists. If you calculate the number of lost minutes used to correct these problems and the number of times that the problem occurred then you can calculate the additional revenue that a doctor could generate if they were to use that time on rounds. After some discussion we agreed that some of these errors would still occur with a more automated system so we reduced the potential savings by 10%.
|Minutes per||Doctor M|
|Sheets not delivered||15||13||times/yr||$435.22|
|Percent that can be improved using MediMobile software||90%|
|Increased Revenue per year||$3,550|
The final three areas of improvement were related directly to the billing process and collections where required. As error rates came down the time required to process the billing records was also reduced. The number of insurance denials decreased and the collections rates improved. As a consequence the total write-off value reduces as well. The total calculated savings from improved billing and collections was $3,032 per year.
In summary, the total costs for the recommended solution were $2,679 and the solution generated an annual savings of $37,473.
|The estimated annual costs were:|
|First year project investment…||$1,979|
|Third year phone replacement…||$350|
|Fifth year phone replacement…||$350|
|The total estimated costs over five years are…||$2,679|
|The estimated annual savings were:|
|Increased revenue per encounter…||$18,333|
|Increased revenue from lost forms…||$12,558|
|Increased revenue from coding improvements…||$3,550|
|Increased revenue from improved billing & collections…||$3,032|
|The total estimated annual savings are…||$37,473|
Comparing this level of benefit to the costs we find a Payback Period of 1.7 months! Since most projects of this type are approved with a 1 year payback we found this investment to be outstanding. Looking at a five year period for the financial analysis the internal rate of return is also off the charts and clearly indicates that the return on this project is exceptional.
One thing that we learned during the pilot was that we should have allowed a few hours for familiarity training on the new phone and software. As you may recall doctor M is very comfortable trying new technologies. In fact, he enjoys the challenge and could easily teach himself. That was not the case with doctor H. We put the new phone in his hands early knowing that he would take a bit more time to get comfortable and he had some difficulty finding functions that he needed and his frustrations grew before he even had the medical software loaded. Had we provide a short hands-on training session he would have had a more positive initial experience. Now that he has learned where to find what he needs he is using the phone and software very effectively.
We also would recommend planning a phase in process so that all of the doctors are not moving though the learning curve at the same time. The early adopters can then act as internal “experts” to assist their peers as each learns how to make the tools most effective for them.
We would also recommend planning for data backup from the central server to protect against any lose at the host. Ultimately the practice is responsible for the records and having a periodic backup of the data will provide one more level of security.
Measurement & Monitoring
Finally, any system that is put in place takes effort to learn and use. If the projected results are to be sustained a program for measurement, confirmation and continuous improvement will be required. This has proven over time to be the most difficult step of the process but the using the necessary discipline can result in financial improvements year after year.
The results for the practice covered the items outlined below. As we move forward to expand the optimization process we expect to find other areas for improvement. If you would like to get further information on how to get these kind of results please contact the author.
- Increased patient encounters
- Down-coding to reduced revenue level
- Increased charge capture
- Improved collection rates
- Decreased denials (insurance time limits)
- Decreased denied charge write-off’s
MediMobile is the leading technology provider of hand held billing solutions. MediMobile, Inc. is focused on point of billing solutions and patient management systems by reducing costs, streamlining processes allowing physicians more time to focus on healthcare. MediMobile, Inc. is a Texas based Private Corporation. Request a Free Demo for more information: www.medimobile.com.