Saturday, January 30, 2010

Signing off

The time has come for me to sign off.  As you can see in the title, "I will be implementing an Electronic Medical Record in my practice, starting in March 2006. This blog is a diary of what happened". 

The EMR has now been implemented; my practice has been redesigned to meet goals for patient access (wait times for appointments are now routinely same day or next available clinic day; time sitting in the waiting room is <1/2 hour, we use email with patients), quality (routine measurement and monitoring, regular team meetings), and efficiency.  We work as an interdisciplinary team now; these are not just "buzz words", we actually are doing it.

All of us in this primary care team have traveled far along the road to better care for our patients in the past four years, and the EMR has been a key part of this redesign.  We will not stop, but I do feel that a large part of the work has now been done.  The key issue remaining is that those of us using EMRs continue to function as electronic islands in a sea of paper and systemic inefficiency.  We cannot change this from our practices; such a change will take leadership and vision from the people managing our health care system.

As for me, I am finishing my Masters of Science; my thesis "Effect of EMR implementation on preventive services" will be completed this year, and I intend to publish it.  I will continue to work three days a week at my office and two days a week on research projects; I think EMRs and quality of care are an important subject, and that is what my research will focus on.


Thank you  for bearing with me as I navigated the twists and turns of an EMR implementation in this small community based family practice.  It has certainly been challenging at times, but the outcome is more than worth it.

Michelle



Monday, January 04, 2010

Not yet good enough

Here we are, at the start of a new year.  This seems to be a good time to take stock of things.  It has now been almost four years that my group started using an EMR.  My office is much further along for some things than I thought we would be by now, but also much behind for other things.

We have done well in terms of getting rid of paper inside the office.  We have no paper charts at all, and no filing cabinets for patient data.  All patient data is stored directly into the EMR, whether entered directly by someone in the office, or scanned in.  All members of this practice (physicians, Allied Health Professionals, staff) use the EMR.  Almost all tasks and patient-related communication are entered in the system.

We are efficient.  As of next Monday, I am on Open Access; my patients have all received a brochure outlining what this means (this was mailed in October).  We have eliminated delays and waiting times to see me are now essentially 0 or 1 day. 

We are also effective.  We measure our quality monthly; 89% of diabetics in this practice have had an eye exam in the past two years, and 91% have had a foot exam in the past year.  We have consistent alerts and reminders for  overdue services, and are always looking for ways to improve quality.

Through QIIP, we were provided with a spreadsheet outlining how many visits were expected for a patient in each age group, for a family physician on capitation or fee for service.  I had 3709 encounters for a year; expected for my practice is 4301, a difference of 592 encounters (13.8% less than expected).

While this does not tell me what exactly is contributing to the difference, some of the effect may be due to:
  • "max-packing" visits (doing everything that needs to be done in a single visit)
  • increasing time between repeat visits if appropriate
  • phone management
  • use of email with patients
  • Allied Health Professionals and team-based care; task distribution
  • working to top of scope for all team members 
The monthly number of visits appears to have decreased since September, from about 290 per month to 220 per month.  The number of "no shows" has decreased from 20 to 10 per month, and was down to 3 in December.  It may be fair to expect the difference between expected visits and booked visits to be larger by next year.

And yet, it does not seem to be good enough; we continue to suffer from systemic inefficiencies.  The number of proprietary referral forms has not diminished; there is no electronic prescribing in my neck of the woods; email communication between doctors and patients continue to be poorly supported by our health and privacy organizations; and worst of all, the onslaught of non-electronic incoming patient data has not abated one bit since we started.  Not one of the paper based reports that we started scanning four years ago have been switched over to electronic format (hospital, Diagnostic Imaging, specialist reports, non electronic laboratories); all these continue to be reported on paper.

The new funding for EMRs will make a big difference in adoption; I think the majority of family physicians are now considering switching to EMR.  The funding is also available to specialists, and I think they will switch as well.  Perhaps this will provide enough "push" for the system to integrate at last; labs that do not transmit reports electronically may well see a significant hit to their business, as physicians actively drive patients away from those facilities. 

We are pushing ahead with practice redesign.  On the agenda for this year is Group Medical Visits, where several patients with a similar condition (for example, diabetes) are seen together; our clinical pharmacist is in charge of arranging this.  My secretaries are now routinely collecting email addresses from all patients.  I have configured Outlook Express on every computer in the practice with my office email (drgreiveroffice@rogers.com), outgoing only.  If the secretary has trouble reaching a patient by phone, she sends an email asking them to call back, or sends the date of the appointment and asks for a phone call to confirm.  One day we'll have online booking for patients.  Incoming email to the office address gets redirected to our office manager, and she then takes action or forwards to the physician if appropriate.  I have been told by eHealth Ontario that they are looking at the possibility of giving patients access to the secure OneMail; as well, they are considering adding pharmacies to OneMail--that would be good, perhaps the pharmacist could email me if they have a question, instead of faxing or calling.  Perhaps I could even send prescriptions via secure email instead of fax. 

I think we have made progress in re-engineering how we look after patients in this office, but I'm ambitious.  I don't think it is good enough yet. 

Michelle

Saturday, December 05, 2009

Upgrading the office hardware

My tablet is now 3 1/2 years old; that's old for a computer!  It is still working well, but I expect it to have an increasing number of problems over the next few months.  It is time to get a new machine.

The other computers I initially bought are all desktops, and are functioning well; I bought pretty solid Dell business machines.  These two computers are used at the front, and are limited to business functions only.  One computer has the scanner and fax machine attached to it, and our IT manager added some extra memory a few months ago.  I have a second scanner attached to a newer computer in the staff room, and a second fax machine in the staff room in case of failure--we have had to learn to plan for redundancy over the years.  It looks like we can get a number of extra years out of the two computers at the front, although they will eventually fail and have to be replaced.

As we grew, I added more machines.  I tried to keep the new PCs the same (all Dells, all Windows XP), so that maintenance would be easier (same look and feel, same software).  There are now six desktops, four Tablets, three laptops and a Netbook in the office.  We have two scanners, two fax machines, and four labelers.  We have twelve printers; nobody has to walk very far to get a printed document.  Everyone has ready access to computers.  All this is for a (paperless) three physician office.

I talked to our IT manager before buying the new Tablet.  I have to decide what operating system to buy (XP vs Windows 7), and which machine.  I use the Tablet every single day; I decided to upgrade to as much RAM (4 MB) and to as fast a processor as they had available.  I added three years of on-site warranty. My IT manager recommended that I downgrade to Windows XP, so that the machine would be fully compatible with the rest of my office network, and I have done this.

I will have to start thinking about what to do to upgrade the entire office, as our machines age.  This is becoming a more complex issue due to the number of machines we have.  My colleagues in our Family Health Organization will have the same issue, as we all bought our hardware at the same time.  We should probably  put this on the agenda at one of our IT committee meetings.

As you can see, we have more IT help and organization than when we first started.  We have an IT committee, composed of two representatives from each of the two Family Health Organizations on our server and our IT manager.  Our IT manager oversees the functioning and daily maintenance of our common server, and recommends server upgrades as needed.  Because all FHO offices have similar hardware and software, our IT manager is able to provide some support for in-office issues over the phone (as he is now very familiar with what is in the offices), and we are experiencing far fewer hardware failures than in the first two years. 

We are also starting to have a bit more IT structure for our Family Health Team (which is composed of six different Family Health Organizations).  The FHT has an IT manager to oversee issues for the administrative office, as well as for the new FHT Allied Health clinical offices; this office was used for our H1N1 vaccination clinic, and is now slated to open later this month. It is two doors down from my office.  There are a lot of computers there as well.

Last week, my eHealth Ontario internet connection was failing:  it started to run more slowly, and would intermittently disconnect.  The problem was isolated to the SOFA (Small Office Firewall Appliance), which is the router supplied by eHO.  The SOFAs for all the practices here are now in the phone room of our floor, and the FHT's IT manager looks after problems for that equipment.  He disconnected the malfunctioning SOFA and plugged my office into another office's SOFA as a temporary measure (worked well), and then notified eHO.  They shipped another router by the next day, and we were back in business.  Frankly, I like this managed approach much better than the old panic attacks at my office.  Support makes a big difference; working as part of a larger group of practices helps; having an organization behind you is good.

We are starting to enlist the help of our patients.  I received an email from a patient asking if we had a "preferred lab" in her area of the city.  We are making a concerted effort to direct patients away from non-electronic labs: we have handouts on lab locations, and we call the patient if we get a lab result on paper (they get asked to avoid that facility in the future).  A patient with a complex health condition said that she takes the printed CPP I give her to all the different specialists, because it helps them keep track of her overall health; she is very picky about making sure that I update the CPP properly, and we do this together.  She can see that most of the specialists do not use computerized records.  Another patient was furious about the fact that the hospital never sent me anything, despite the fact that she gave them directions to forward the information to her family doctor.  She is going to complain to her Member of Parliament about the facility's lack of abilitiy to send her data electronically.  Maybe we can partner with our patients to push for interconnectivity; after all, it is their health and their quality of care that is affected by the fact that their data does not follow them.  Disconnected care is bad care.

Michelle

Tuesday, November 10, 2009

H1N1 vaccination clinic and EMR

Our H1N1 vaccination clinics are now running at full capacity, five afternoons a week. All Family Health Team physicians have volunteered to staff the clinics. Family Health Team RNs have been withdrawn from their regular duties in our offices, and are staffing the vaccination clinics.

We are running this at a single location for all 53 physicians in the FHT, and this was all organized in a few days. We are currently vaccinating high risk patients in our FHT. The clinic has one administrator, three nurses and two physicians. We vaccinate about 360 people in 3 ½ hours.

The clinic is two doors down from my office, and I see the line up in the corridor when I poke my head out the door. The waiting time is about 40 minutes, and nobody has to wait outside in the cold. Patients getting vaccinated are in a different location than those coming in to see their family physician for the flu, so this approach limits viral transmission. We have had very good feedback on our clinic from patients getting vaccinated.

The vaccination team holds a debriefing session at the end of each clinic. They review their processes, identify bottlenecks, and quickly implement changes.

We use the EMR to record data, because otherwise we would have to keep enormous amounts of personal information on paper (the consent forms, the name of each patient getting the vaccine, their ages, gender, and risk category). The FHT, as an organization, does not have a mandate to hold personal information, nor do we have storage facilities for this at our central FHT office. In Ontario, all patient data has to be kept for a minimum of ten years; for children, until they are 28 years of age.

The first issue that we ran into is that we do not have an integrated EMR for all FHT physicians. There are two EMR software applications being used; the larger one, which my group uses, is split into three databases (or enterprises), one for 14 physicians, one for 9 physicians and one for 20 physicians. The physical location of the offices does not matter for data entry, as it is all done remotely—we log in to a server located off site. The other EMR, serving 10 physicians, uses software on servers located in individual offices, so there are actually a larger number of individual databases for fewer offices. You have to have a different log on for each database, and you have to establish remote access.

In a FHT-wide program, especially where speed is key, you cannot have multiple different log-ins. What I mean by that, is you can’t have Mr Smith, who is in Dr A’s practice, having demographic data and H1N1 immunization entered in Dr A’s database, then Mrs Jones, in Dr B’s practice, comes, and the clerk has to enter demographic data by logging into a different database etc. It doesn’t work because it is too complicated, is too much additional work, and slows things down too much.

We decided that all data entry was going to be done in the 20 physician database. Here is the process:


1. the patient checks in, the clerk swipes the health card
2. if the patient is already in the current database, the system recognizes this and automatically registers the patient. If not, then the swiped card automatically adds data to the EMR (name, date of birth, health card number, gender), and the patient is entered as new
3. patients are sequentially added by the software to the schedule, so that the clerk can see who is next in line when calling patients in.
4. The clerk gives the patients the screening form, and the patient waits until called in to see the nurse.
5. When checking in, the nurse reviews the data; an encounter is opened, which contains the screening questions as checkboxes, all defaulted to no (most patients are all “no”). If there is a “yes”, the nurse changes that, and notifies the physician
6. The patient signs the consent electronically on a signature pad, and this is saved to the EMR.
7. The physician sees the patient and gives him or her the H1N1 vaccine. Lot numbers and dosages are pre-set as defaults. The dosage is changed if this is a pediatric injection.
8. The encounter, screen, consent, signature, and vaccination are all electronically saved in the EMR.
9. The patient goes to wait for 15 minutes in our post vaccination room, which has chairs and a television. They are given a paper to inform their family physician of the vaccination. If there are any reactions, these can be entered in the correct field of the EMR, so that the data can be extracted later.

The first day that we had the clinic, the waiting time was longer. The rate-limiting step was the screening; the initial EMR screening template used drop down lists, so all information had to be individually entered, which slowed things down. Once it was changed to checkboxes defaulted to “no” for all screening questions, this was much faster. The change was implemented at the end of the very first clinic.

One early challenge for us was not setting up and managing the clinic, it was communicating with Public Health. They had a great deal of difficulty letting us know if we would receive vaccines so that we could run our clinics and take some pressure off their own, overwhelmed vaccination program—and a few days later we heard that a private clinic downtown had been sent vaccines for their “executive physicals”. I understand that this will be investigated.

The vaccine supply chain at Public Health runs well for our regular supplies; it appears to have been initially overwhelmed by this large scale H1N1 program. I cannot tell where the internal problem was; I hope that processes similar to our daily debrief and rapid improvement cycles have been implemented. The problem at the present time, as I understand it, is vaccine shortages due to manufacturing problems.

We do have ongoing issues with our local Public health unit; as an example, we do not have an efficient method of reporting issues to them (usually this is by fax or by phone). For those of us on EMR, the issue is even worse, as they insist on paper-based proprietary lab reqs and will not accept anything generated out of an EMR; I cannot track HIV tests in my practice to ensure that I have received the result. I think the Public health unit does an excellent job in many areas, such as their “safe dining” restaurant program; however, they do less well when communicating and collaborating with family physicians.

The EMR company has programmed software to extract the data that public health needs in an excel format; we should be able to extract information such as age, gender, vaccine lot number, adverse reactions so we can report for our large clinic. This is a much better way to report than duplicating the information on a piece of paper for each patient. Perhaps we’ll eventually have a secure electronic method for forwarding these data; wouldn’t it be nice to be able to automatically send data on child vaccination to public health.

I am pretty proud of this Family Health Team for doing such a great job of organizing and managing this project. I think there are a couple of early conclusions that I am coming to:

1. This couldn’t have been done without a good organizational structure. Primary care has traditionally been composed of individual offices, often working in isolation; we are now moving towards more organized groups (Family Health Groups, Organizations and Networks, Family Health Teams)
2. EMRs are part of this new organizational structure. Newly forming groups should consider having one common database for all offices if they want to run projects in common in the future. This can be done with many of the EMR applications being sold
3. Local Public health units need to consider primary care as an integral partner. The experience with SARS shows that these new diseases often show up in our offices first. Talk to us, work with us, we can help you.
4. The IT infrastructure at Public Health may need to be revisited. They receive large amounts of data, and they need robust systems to analyze and manage the information. During SARS, information was tracked via yellow sticky notes on the wall; I hope it is different now. Upgrade Public Health lab to electronic reporting standards. Work on having EMR data from family practices reported electronically to the local Unit.

I think this H1N1 vaccination clinic is a beautiful example of a large, geographically dispersed primary care team banding together to provide needed services, quickly and efficiently. The EMR in this case is an important facilitator, but the key aspect was leadership and group cohesiveness. Family Health Teams are a fairly recent initiative from our provincial Ministry of Health, and our experience shows the value and benefit of this approach. I hope the FHT program will be expanded and offered to more of my colleagues in family practice.

Michelle

Friday, October 30, 2009

H1N1 influenza

There are increasing numbers of people with H1N1 influenza reported in the past two weeks. This feels a bit like SARS did in 2003 (my hospital was at the epicentre of the second Toronto outbreak). However, we are better prepared now, and I have electronic tools, both EMR and web based.

Public Health is running H1N1 vaccination clinics for the public. Their initial mail out about vaccine requests to family physicians almost sounded like they did not want us to order the vaccine: only order in lots in multiples of 500 doses, reconstitute vaccines with adjuvant and use the ten reconstituted doses within 24 hours. Send a report on the number of vaccines, number of males, numbers of females, ages, high risk conditions to Public Health every Monday by noon (their faxes are going to be very busy on Mondays, we may not even get through). There were also additional conditions for ordering this vaccine. I was told by a colleague who is a Medical Officer of Health that he was advising family physicians not to order the vaccine, due to the complexity of managing it.

I can understand that the requirements for this vaccine are different from the usual flu shot, and some of the problems are due to vaccine packaging, but the result is extremely long line ups for the Public Health clinics in Toronto, which are currently overwhelmed.

My Family Health Team quickly organized a common vaccination clinic for the 53 physicians and 60,000 patients registered in our practices. Our FHT executire director is pulling nurses out of offices, and most of our physicians have volunteered to man the clinic. This was all arranged by our executive team over email. All patients are being directed to the FHT clinic for H1N1 vaccinations, and we will not be running clinics in individual offices. I have added information on our clinics to our website.

We are getting more patient phone calls at the office about what to do; I post announcements on our practice website, http://drgreiver.com , and my front staff is re-directing inquiries there. This is helping us to keep the phone lines reasonably free, so that patients who do need to get through on the phone can get through. I am also getting a much larger volume of emails from patients than usual, and I am able to redirect these to our website as well, so this is currently manageable. Volume of visits to the website has more than doubled, but volume of visits to the office is holding steady so far.

The EMR company has added new vaccination profiles for the H1N1 (both adjuvanted and unadjuvanted) so that we can quickly enter the data. My colleagues in Nova Scotia have reported that their Public Health authority can obtain weekly reports containing all the data needed (without patient names), directly exported from the EMR, under the authority of the NS Provincial Chief Medical Officer of Health (through the Nova Scotia Public Health Act). What a fine example of collaboration between Ministry of Health, Public Health, EMR company and physicians! I wish we had this in Ontario; it really represents a good example of the reporting power of the EMR, especially under strained conditions, but it needs to have a health care system that is less fragmented than what we currently have in Ontario. One physician said that we should be using a similar system to report routine vaccinations to Public Health, something which is currently fully paper based.

Because of the size of our FHT, we have administrative support to enter data directly in the EMR when patients come to one of our H1N1 clinics, and we will be able to generate data similar to Nova Scotia's. Hopefully Public Health will accept our reports as long as the needed data is present, without requiring it be entered in their proprietary excel format (or worse, on paper).

Just prior to the outbreak, I volunteered to program the Public Health guideline on the diagnosis and management of H1N1 influenza as an EMR template. We had enough time for several colleagues to test the template and offer feedback; it was modified to make it as useful in practice as possible, and I then added it to all three EMR enterprises for our FHT. We now have a common management tool, with clear information on when to prescribe Tamiflu available to physicians electronically, and it is currently being used. Perhaps in the future we will be able to export this to Public Health in an anonymized fashion as well.


OntarioMD announced this week that about $280 million will be made available to subsidize EMRs for the majority of Ontario physicians. The subsidy is $29,800 for those using an ASP model, and $27,100 for those using a server located in their own offices. These amounts are very similar to what physicians in my group received, and will cover about 70% of the cost of EMR. I think we will now see large scale computerization in Ontario; this step is done in conjunction with eHealth Ontario, and this is the right approach for them to take. The auditor general said that there was no traffic on the super-highway that eHO built; having data generated in the majority of physician offices will put traffic there. Once we are computerized, we quickly learn to dislike non electronic data generated on the outside. However, there have been too few of us on EMR to generate the kind of pressure needed to get other organizations to connect to us for incoming data and stop using paper-based processes for outgoing data; I think this is about to change. I expect to see a lot of changes in the next few years; I think a lot of early adopters will also be kept very busy helping our colleagues implement and use their new EMRs effectively.

I was asked by a perceptive journalist if I still expected to be writing this blog over three years after EMR implementation started in my practice. I have to admit that I did not; I thought things would be settled and going smoothly by about 18 months--how wrong I was. It now looks like I am going to continue this journal; thank you for bearing with me through these interesting times.

Michelle

Sunday, September 27, 2009

Open Access



We are continuing to implement new things in my office. Here you see an initiative to decrease the number of days that patients have to wait until they get an appointment. We count the "Third Next Available" appointment every Tuesday, and average it for each month. By Third Next Available, I mean that we look at my schedule, and see when the third next open slot is. This is the best way of measuring access to the physician. We went from 23 days to 11 days to 5 days, and last week, we had our first zero: you called that day, you got an appointment that day.

My practice Team is part of QIIP, the Quality Improvement and Innovation Partnership. We are using multiple Plan-Do-Study-Act cycles to change the way we work. It has been hard; after all, routine is comfortable and change is not. However, we are making progress on many fronts.

Part of what you are seeing up there is the result of multiple changes to improve efficiency; we are now trying to do everything when the patient comes in. If you are in for a cold, and the EMR shows that you are overdue for a foot exam for your diabetes, there is an alert in the chart and you get offered a foot exam. If you come in for a rash, and you are due for a Blood pressure check, you also get a blood pressure check. We use the phone a lot; my RN checks in by phone to follow up with patients who are depressed--we use the EMR for messaging about this. We distribute tasks: our Team Clinical Pharmacist now sees several of my diabetics for their routine follow-ups, and monitors all their parameters. I am always on site and available when she is seeing them, but it gives me the time to deal with more complicated problems. My Team is testing a much more complex EMR-based flowsheet (the Chronic Disease Management flowsheet) that incorporates templates that are linked to the CDM flowsheet, along with automatic time-based alerts--and other people in our Family Health Team are interested in doing this as well. In my practice, we have agreed on common ways for all of us to enter data so that it can be extracted to monitor our quality.

I don't just want to cut waiting times for our patients, I also want to maintain or even improve our quality. However, if there is no access there is no quality, so you have to work on both at the same time. I often hear about the effect of EMR on efficiency, and I think what people mean when they talk about this is how fast charting for encounters is, or how fast you can access information when seeing a patient, or how long a consultation takes. I think there is another dimension to efficiency, and that is how well the EMR helps you manage your practice as a whole, and how well it helps you function as a team. If these things improve, then you can see your patients sooner--efficiency gains translated as improved patient access.

Interestingly, the move to the new office has not translated (so far) into a significant decrease in the number of patients I look after. My current roster is 1296, down from 1306. What I have to decide now is whether to open my practice up to new patients again; if I do that without planning, then I think access will worsen and waiting times to see me will increase. My staff are tracking Demand; this is how many appointments are generated each day, whether because I ask the patient to book a follow up (internal demand), or because the patient calls to make an appointment (external demand). We also recently measured how many appointments I have available, by counting how many days I am in the office and how many patients I see each day; this is Supply. Having all this data will help my Team to decide what to do in terms of opening up the practice to new patients. Clearly, if we become more efficient, then we increase Supply; the EMR is an important factor there.

My Team has decided to go to Open Access in January. Several other Teams participating in QIIP have already make the leap, and we have used materials developed in Dr Peterkin's practice (Mapleton Family Health Team) to begin to let our patients know about the change. Open Access means reducing wait times: you call and you get an appointment same day or next clinic day. I have posted initial information about Open Access in this practice here, and Dr Peterkin's poster is now on the Community bulletin board in my waiting room with a note that we will be starting a similar initiative on January 11th 2010.

Michelle

Tuesday, September 08, 2009

Internet speed and quality of care

I have a fast connection to the Internet in my new office. The reason for this is that we have several practices on our floor, with a lot of physicians, staff and Allied Health Providers. eHealth Ontario has provided us with a bigger connection, because we are now a medium sized business rather than a single small office.

At my previous office, I had a “regular” internet connection, similar to what most of us have at home. There was some security overlay on this connection, through eHO’s central circuits, which reduced speed. Security is important (these are medical records we are talking about), but there is an effect on speed. As well, the security overlay failed at times, which affected multiple practices across Ontario. My access speed at home, via VPN, was always much greater than at the office. The single office internet access line was shared between all of us at my office, and the net speed per PC varied, depending on how many people were using our computers: number of physicians present, front staff, residents, medical students, “special projects” (such as our preventive services audits). As well, we found fluctuations during the day: it was slower when everyone logged on together at 9 am, or when there was heavy system internet use, such as 4 pm (probably kids coming home from school and logging in).

Our server at the hospital also had an older access line, and an older firewall. This limited speed of access at times, even if Internet speed was good. The lines coming into the hospital were unstable at times, and we had several outages.

eHealth Ontario has been working collaboratively with our IT manager to solve these issues. It does not seem like a simple problem to me; there is a cost/benefit ratio to providing faster lines, and I know that funding is not unlimited. There are many factors influencing speed. However, slow or unstable lines do have an impact on patient care.

The three months before my move were more difficult than usual, with more frequent slowdowns and outages; when the system slows, we call our IT manager, and he tells us whether to switch to our backup internet line, which we buy privately. Our IT manager is allowed to handle this for our group, as it works better than having each office call eHO individually. Switching to backup means that all office computers access the EMR via VPN (Virtual Private Network); we only have a limited number of VPN passwords, so the number of computers in use is restricted. Some networked printers don’t work. VPN at the office is unstable, and logs us off periodically. Using the backup internet line does not work well for more than a few hours, but does allow us to continue using the system. Once our IT manager tells us that the eHO lines are working again, then we switch back; we have to log off, stop the VPN, wait for 10 minutes for the eHO Internet connection to come back, change some computer settings, and then reboot all the computers. Because of all the work, it is impractical to do this while seeing patients; we stay on backup and switch at lunch or after last patient is seen in the afternoon.

During the slowdowns, I often could not look at my flowsheets, and I did not print information for my patients to the same extent that I normally do—no labs, no CPPs, fewer handouts. Printing just took too long and was too frustrating. I limited the amount of information entered in the chart to essentials, and less information was coded, because that took too long. I finished charts at home. If it was really too slow, I wrote prescriptions on a paper prescription pad. Sometimes patients would ask me about results, and I just couldn’t look them up; one of my patients commented that I had problems for two of her last three visits. In addition to being unable to look at and input information, I was very distracted and upset, which made it more difficult to give my patients the care they deserve.

eHO has now upgraded our hospital line, as well as the firewall at the hospital server. Access from home is noticeably faster. However, the biggest difference is access from the office, via the new lines— we are no longer on “normal slow”: my secretaries are not frustrated with information coming in at eyedropper speed, and I can look at records without thinking about what I can or can’t do. This just feels right.

A colleague in Markham on another EMR system told me that they had two sites: one with a local server, and a second with remote access to a hospital-based server. The second site eventually switched to local because of the same issues we had (access line speed and stability). His comment: “an absolute requirement is fast, stable access to servers”. We just don’t tolerate slow access speeds while using EMR.

Despite the difficulties, I still believe that ASP (one large server for multiple small practices, remotely managed) is ultimately the way to go. We could not have done our preventive services project, our diabetes reminders and common flowsheets, or the data quality improvement summer projects if we had individual, isolated servers in each office. I believe that Quality Improvement initiatives should start and be tested within individual practices, then be spread to the group if successful. Spreading QI is much easier if you have a common server—you can have similar data entry for several practices from a single location once you all agree on what to do. However, a prerequisite for this is IT stability and speed.

We have been talking with eHO, and I think there is a good understanding and appreciation of the importance of this issue with respect to the quality of care we provide to patients. I am now seeing sure signs of progress at the front line. eHO is upgrading the Small Office Firewall Appliance (SOFA) in our offices to a more modern firewall and router system in a month, as part of a provincial initiative. The move to my new office involved a complex IT installation; however, we were up and running from day 1, due to collaborative efforts between our IT manager, our FHT’s IT manager, and eHO’s staff. My husband’s large business moved last year, and he commented that one of the most difficult aspects of the move was making sure that the IT transfer was seamless. eHO has been criticized heavily in the press for their consultants’ billing practices; it is harder to talk about what goes well, and I can say that the job was done right in this instance.

In my practice, we are now posting monthly graphs for quality in our staff room’s bulletin board (for example, percentage of diabetics reaching targets for blood pressure, cholesterol, blood sugar control); you can’t improve what you don’t measure. Perhaps we should think about having reports of system access uptimes and access speeds for practices using EMRs posted online. I think that this may give a more genuine indication of progress at eHO.

Michelle