JCO Interviews Dr. Homer W. Phillips on Computers in Orthodontic Practice, Part 1

DR. GOTTLIEB Should an orthodontist have a computer?

DR. PHILLIPS You don't start with needing a computer. First you have to say, "Here are five tasks in my office. What is the most efficient, accurate, and economical way to accomplish them?" If it turns out that the computer is the best way to perform some of those tasks, then an orthodontist should consider a computer. Just to have the idea that it's a good thing to have a computer these days and to say, "I want a complete system, I want to automate my office", I think is a totally unrealistic, impractical approach. That is not the way to go about it at all, in my estimation.

The computer can help you accomplish tasks more efficiently. There's no question about that. However, it isn't all benefit. There are complexities involved that have to be considered, and I don't think that most orthodontic practices are in a position to utilize a computer well. For the computer to really be of help, rather than just be another way of doing routine office procedures, you have to have a very high degree of organization in your paperwork and business systems. You have to use a systems approach, rather than a haphazard approach, and most offices simply aren't organized that way. A large part of the management details of many practices is in the heads of the orthodontist and staff. It's not on paper or in a manual. To computerize these things, you must first have an organizational flow sheet describing exactly how you do each task. A computer does the routine tasks so quickly that, if they're not organized correctly, the office can quickly get into a lot of trouble.

DR. GOTTLIEB Can you give us a for instance?

DR. PHILLIPS Well, let's take appointing. A lot of orthodontists think they would like to have a computerized appointment book and schedule appointments like scheduling seats on an airplane. This can be done on a computer. However, it's one of the things that perhaps can better be done as it is being done right now--with paper and pencil.

DR. GOTTLIEB Why is that?

DR. PHILLIPS The computer can only show you so much on the screen at one time, and, no matter how fast you can hit keys to flip to another section, it's not as fast as flipping pages in the appointment book. It is also faster to add or delete appointments with paper and pencil than it is with a keyboard. It's one of those things that is only better in theory. There are many things that you can do on a computer that paper and pencil will still beat.

DR. GOTTLIEB Could you give us another example?

DR. PHILLIPS Oh, payroll: you can have the computer figure the amount of the check for each employee, and actually print it, but these tasks may take more time with the computer. For instance, if you have six employees and pay them twice a month, that's 12 checks a month. Let's say you want the computer to do this for you. Well, the employee's names and information such as the number of dependents have to be put into the computer, much as they have to be put on a paper form without the computer, and when it comes time to print the checks, you have to put the roll of computer checks into the printer--after first removing whatever computer forms were in it--and then print out those checks. Unless you have a sizeable payroll of 50 or 100 employees, I feel that it's not efficient to do it with the computer, compared to writing the checks in a checkbook.

DR. GOTTLIEB Not yet anyway. What uses should one consider first in starting with a computer?

DR. PHILLIPS In the business area I think the simplest thing to start with would be a recall system, which is important in building and maintaining a practice. This is something the computer can do very efficiently, and it's a procedure that I call a nonsensitive one--nonsensitive in the sense that it's not apt to cause bad public relations with your practice families if something goes wrong. The computer can very easily send out your recall messages every month. In my own practice, we have one computer turned on all day every working day with our recall program in it, among other things. Any patient who does not get a regular appointment--within six weeks--goes into our recall system. The patient's name and address are put into the computer one time. The appointment secretary can get their name on the screen and key in when they're to be seen next in about one minute. That is generally faster than entering the same information manually, when you consider pulling and refiling cards. Every month when we send out recall notices, we ask the computer for all the recalls we're to see that month, and it generates press-on mailing labels for those patients. The computer will allow you to find things so much better. If you don't remember the brand name of an item or the company name, you can find things by other approaches, perhaps by remembering the city where the item was ordered or the month the order was placed.

DR. GOTTLIEB Where do management functions come in?

DR. PHILLIPS If you are interested in various management reports concerning your practice--perhaps reports that you are receiving now from an outside source such as your CPA or practice management consultant--if this sort of thing is important to you in running your office--the computer can generate these reports from the same information that you send to an outside computer service, which then sends the reports back to you. The reports can be obtained faster and at less cost using your own in-house computer. To get into the financial area, I would start with an accounts payable program. If anything gets messed up, it's an outgoing check, and that's not likely to cause as big a problem as if you send a patient a wrong bill. That can cause people to become dissatisfied with your practice. The last things I would computerize would be accounts receivable and appointing, and these are probably the two things that most orthodontists think they want to do first with a computer. I think it's a mistake to start with something that is essential to daily operation until someone in the office really knows how to operate the equipment and understands the programs thoroughly. You shouldn't put yourself in the position of having to deal with operator errors compounding equipment or program malfunctions.

DR. GOTTLIEB When you talk about management reports, are you speaking only of financial statements, or more than that?

DR. PHILLIPS I would say more than that. You're not in a position to generate all the management programs that you might be interested in until your financial transactions are on line. Management reports include financial statements, but, in addition, a total breakdown of your practice in terms of referral sources, patient starts and stops--everything dealing with the statistics of your practice. A computer can provide that type of information in a very efficient manner. But I feel that these are not the important tasks to consider in starting with a computer. I think those are down the line a bit.

DR. GOTTLIEB You feel that the nonsensitive programs are the place to start.

DR. PHILLIPS Yes, and then approach things such as accounts receivable, if you want to send out statements using the computer, but I don't think you should do that until one or more people in the office are thoroughly familiar with the operation of the equipment, and you're sure that it's reliable and you're not having any trouble with it. The very last thing I think you should consider is an appointing system, because if you cannot find the appointment time or the whole system is down it causes ill will.

DR. GOTTLIEB Are those you mentioned the major benefits of a computer or just a few examples of possible applications?

DR. PHILLIPS I really think that the major benefit of a computer in an orthodontic practice is none of the above. I think the major thing that it can do for you is to establish a data base for your practice. If you are interested in how many people live at such and such a zip code, you can find that. If you want to find out how many orthognathic cases of a certain type you did last year, you can find that. This is the sort of thing that I feel is a major benefit, but it is not what people ordinarily think of when they think of computerizing their practice. I think most people are thinking of sending out bills and looking at financial statements.

DR. GOTTLIEB How about letter writing?

DR. PHILLIPS Letter writing, yes, that is a major benefit without a doubt, and I would give that a high priority. Any practice that sends out much printed information during the course of treatment can benefit from the word-processing capability of computers. In my practice, we don't use our dictating system any more. If I want to write a letter, I simply type it out myself on the computer. I can write, edit, and print out a letter with no typing errors, and get it into the mail faster than I could by dictating and revising drafts in the past.

DR. GOTTLIEB I don't know that people realize that word processors are superior to any typewriter that was ever born.

DR. PHILLIPS Word processors are by far the most efficient way to put words on paper that has ever been devised.

DR. GOTTLIEB Of course, much like the old mag card, the computer can combine the name, address, salutation, and the body of the letter, and compose a personalized form letter.

DR. PHILLIPS Yes, and that saves a world of time. But it goes beyond that. It actually helps you

think. Just as rapidly as you can think of something to say, you see it in front of you, and it's so easy to change words and word sequences.

DR. GOTTLIEB There's no question that what you say is true. We know it better than anybody else, I guess, turning out a journal, but what can a computer do beyond that in an orthodontic office?

DR. PHILLIPS In my practice, we find it useful in the diagnostic area. For several years now, we have used the computer for cephalometric analysis and for diagnosis in general. It never forgets to ask every question or to answer every question, it doesn't get tired, it doesn't leave anything out, and it forces a regime on your diagnostic procedures. It makes you look at certain things on every case, and it does measurements more quickly and accurately than a manual system. On occasion, when my computer is down or I don't have access to it, I really feel a little bit put upon if I have to trace a headplate using a pencil and a piece of tracing paper.

DR. GOTTLIEB That means you're putting tracings on the computer?

[show_img]746-jco-img-0.jpg[/show_img] Computer generated cephalometric tracing displayed on terminal screen.

[show_img]746-jco-img-1.jpg[/show_img] Digitizing the computer tracing.

Caliper feeding tooth sizes into computer.

DR. PHILLIPS Yes, we computerize every headplate tracing.

DR. GOTTLIEB It would be interesting to describe how that process works.

DR. PHILLIPS It's really a simple process and not new technology, but the orthodontic application is fairly new. Either a frontal or a lateral headfilm is placed on what amounts to an electronic viewbox, called a digitizer, and a computer program will request that certain anatomical points be indicated on the film. These points, such as orbitale, are asked for on the screen. You don't have to remember any sequence of points. As the points are requested, you touch an electronic pointer at that point on the film, and measurements are recorded from the x and y axes. It generally requires between two and three minutes of the operator's time. Following that, the programs we use allow one to call up a number of different analyses. It will convert the stored x/y coordinate measurements into the familiar linear and angular measurements of whatever analysis you want to look at. It will show you a tracing on the screen, print it out in 1-to-1 scale, and give you a printout of all the numbers. In addition to that, my group designed an electronic caliper that plugs into the system. With it, we can do such things as a Bolton analysis or arch discrepancy measurements from the model. We routinely delegate the caliper measurements to staff members. The computer does the actual reading, displays the number, stores it, and prints it out for you. I prefer to do the headplate digitizing myself, but that procedure can also be delegated.

DR. GOTTLIEB It certainly doesn't take a lot of your time.

DR. PHILLIPS That's right. Usually the patients' names are entered into the computer and maybe a half dozen headplates will be left on my desk. When I get to them, I can enter the points on all of them in just a few minutes. At some later time, staff members will print out the information that I request.

DR. GOTTLIEB Do you use your computer program at a case presentation? Do you call up a diagram on the screen and use it to explain what's going on?

DR. PHILLIPS I do it two ways. I either actually have the parents at the computer and show them data and tracings right on the screen, or I have printouts of those things from the computer, in which case I give them to the family.

DR. GOTTLIEB Do you find that they are particularly impressed with the fact that you are using a computer?

DR. PHILLIPS Oh, yes. When I started doing consultations this way, I was concerned because I had the incorrect notion that people would feel I was taking shortcuts, not doing enough work myself, that the machinery was doing everything. As it turned out, that wasn't the case at all. It really was interpreted by the parents as a case of third-party validation of what we were doing. If the computer said Johnny had a 10mm overjet, there was no question about it.

DR. GOTTLIEB The computer is an authority figure, whether you're sending bills or doing a case presentation.

DR. PHILLIPS That's true. It has a very positive effect of making the diagnosis acceptable.

DR. GOTTLIEB Do you get with-and without-treatment comparisons?

DR. PHILLIPS We do. We have a growth-without-treatment program based on the Ricketts modular growth increments. It will give you growth without treatment, and we also have an interactive visual treatment objective program that allows you to simulate treatment. The computer will ask such questions as "Is this patient going to grow, and if so, for how many years?" and "Where are you going to put the lower incisors?" and "What interincisal angle would you like?" and "Are you going to extract, and if so, how many teeth and which ones?" If it is an extraction case, the program goes back and checks the tooth file, which contains the stored sizes of that particular patient's teeth, which were entered into the computer with the electronic caliper. If you indicate, "It's a first bicuspid lower arch extraction case", the program goes to the tooth file and finds out the exact size of those two first bicuspids to a hundredth of a millimeter, and that is how much it takes out. Then it says, "Now you have an excess space of 7.3mm. Are you going to move the molars, or what are you going to do?" In other words, you can simulate treatment in many different ways. Certainly this is something that a staff member can do. You can say, "I'd like you to do a VTO for Susan, put the lower incisors 2mm ahead of the APo line, extract two upper first bicuspids, set the interincisal angle at 130 degrees and let me see what it looks like". If you like it, fine; if you don't, you can try some other plan.

DR. GOTTLIEB It seems to me that's the fascination of computers for orthodontists. The other things are the business end of orthodontics and can be done efficiently, but the romantic, fascinating end of orthodontics and computers is in diagnosis and treatment planning. Don't you feel that is justification enough for orthodontists to be interested in computers?

DR. PHILLIPS Well, it is for me, but there's no doubt I have a personal bias in this direction. A computer is such a beautiful tool for helping to make a diagnosis that if I did only one thing in the office with computers, that's what I would be doing.

DR. GOTTLIEB How about recording treatment visit information?

DR. PHILLIPS There are a number of ways to do that, but just because you can do it doesn't mean it's efficient to do it. If you're going to record everything that happens at every appointment, the data entry method must be fast and easy to use. This can be done by having a terminal and something

not a great deal different from the supermarket bar code reader near every chair. Every patient in your practice can have an identifying code. Then you can have, let us say, 100 procedures programmed so that you just touch line 17 and the indicated procedure will be entered. It really gets down to a matter of time and money. It gets complex in terms of programming and, in most instances, it isn't practical. You have to have several terminals for such a system to be efficient, and the complexities of a multiple-terminal system can get to the point where the tail is wagging the dog. It may get more complex than it's worth. But it can be done. There are offices that approach this to one degree or another, but I really think that the present state of the art is such that, for most practices, it is not practical to enter information appointment by appointment. I don't recommend it at this time. When voice recognition systems are versatile enough and reasonable in cost, then I will consider entering treatment data at the chair.

DR. GOTTLIEB How does an office approach the problem of investigating the computer field and deciding what suits their practice and what they want to invest in?

DR. PHILLIPS Well, one basic approach is to go to a computer store or even to a supermarket and pick up a number of computer magazines. There are over 150 publications on personal computers being published now. At first you won't understand many of the articles, but you'll learn from the ads, and eventually you'll get so that you're pretty conversant with what's being discussed. Then you can talk to personnel at computer stores and learn more.

DR. GOTTLIEB Are there some publications you could recommend that would be better than others to start with?

DR. PHILLIPS Yes, I would recommend one called Personal Computing that is general in nature. In the dental field, the Dental Computer Newsletter (published by Andent, Inc.,1000 North Avenue, Waukegan, IL 60085) is full of factual information each month. The editor, Dr. E.J. Neiburger, is a dentist who probably has more hands-on experience with computers than any other dentist in the country.

DR. GOTTLIEB Is Personal Computing on the newsstands?

DR. PHILLIPS Yes, Personal Computing is showing up on newsstands quite a bit now. You can also find it at any computer store. Another useful publication is called InfoWorld. It's a weekly newspaper that has a wealth of current information in it. Popular Computing is another one that has some pretty general information. With those alone you can start picking up a good bit of information. I would highly recommend against buying a computer on the basis of some of the ads. They are useful to get an introduction to the field. Then, if you really get serious about it, you need to start looking for programs that you're interested in before you buy any machinery. Even before that, you've got to decide what you want to do. You don't just buy a program for the fun of it and then make a job to fit the program. If you've got a job you want done, look for a program that does it to your satisfaction.

DR. GOTTLIEB Any buying decision starts with an establishment of a need first, then the search, and then the purchase.

DR. PHILLIPS Yes, that's right. And let's say you have established a need. If it's an accounting job that you want done, you have to start searching for accounting packages. And let's say you've done this and you've narrowed it down to four of them that sound as if they will do what you want done. The step beyond that is still not to buy the package, if at all possible, until you see it demonstrated, and, best of all, see it in use in someone's office. If you do find an office where it is in use, the person you should talk to is the person running the program, not the doctor (unless he is the one actually using it). To really find out what the program does and its problems and advantages compared to doing it manually, you need to talk to a person who has done the work both ways. It would be well for your proposed computer operator to be present and see the system firsthand.

DR. GOTTLIEB Where do you search for the programs?

DR. PHILLIPS There are a number of ways to locate programs. There are knowledgeable people in some computer stores who can help you with a search, and virtually every computer store has catalogs of programs available that you can look through. Sofsearch International Inc. (Box 5276, San Antonio, Texas 78201) is a company whose only business is locating software programs. Virtually every orthodontic meeting has one or more computer vendors, and you can see what they have. This coming January, the AAO is sponsoring a meeting dealing strictly with computers and computing in orthodontics, and I think this would be a worthwhile meeting for anyone who is thinking about putting a computer in his or her practice. Best of all, find someone who is doing what you think you want to do, and actually see the system in use in their office. Make arrangements to go see it with your bookkeeper or whoever is going to be doing the work, and plan to spend a day or two and get some in-depth understanding of what they're doing and what the advantages and problems are.

DR. GOTTLIEB And this would apply to each one of the programs?

DR. PHILLIPS Each one you are interested in.

DR. GOTTLIEB Maybe we ought to say what a program is.

DR. PHILLIPS A program in computer jargon is a list of instructions that have been magnetically stored in some fashion, which the computer uses to cause a certain thing to happen. If it's an accounting program, the instructions to the computer allow you to store things like names and addresses and financial information, and also to manipulate that information to give various reports. These instructions to the computer are ordinarily invisible to the user and are written in a language that a non-programmer would not understand. One set of instructions will make your computer do word processing. Another set of instructions will make it a recall machine. Still another set of instructions will make it into a diagnostic machine.

DR. GOTTLIEB These are prepackaged programs? An orthodontist would have to conform to the program, or find one that is exactly what he wants or so close that it will satisfy him?

DR. PHILLIPS That's true. These are prepackaged or off-the-shelf programs. In some cases they may be modified by the user; however, in most cases that's not practical. You simply use the program as you buy it. You should not expect to find a program that will do exactly what you have

in mind initially. Probably, at best, it will produce the information you want, but not in exactly the form that you had expected. A well-chosen, prepackaged program should provide 75-80% of your requirements; beyond that, custom software is needed.

DR. GOTTLIEB A lot of people might find that it was an improvement on what they were doing.

DR. PHILLIPS Oh, without a doubt. Most of us are not terribly efficient, and successful programs have been pretty well organized and thought out or they wouldn't work on the computer at all. Even poorly written programs are often more efficient than manual methods, simply because of the tremendous speed at which the programs are run.

DR. GOTTLIEB Is there any reason today that an orthodontist should have custom-made software?

DR. PHILLIPS Yes, there is. If the orthodontist has very specific needs, he will almost certainly have to have someone write a customized program from scratch or modify an existing program. It is extremely unlikely that you will be able to find off-the-shelf programs that do everything you want.

DR. GOTTLIEB Is that still true with all the companies that are offering software programs?

DR. PHILLIPS The potential number of sales to orthodontists is small. Even worldwide, there are not many of us. Most programs are written to satisfy large user groups, where sales of hundreds of thousands of programs can be anticipated. I think, in most cases, if a program is not available off the shelf that comes sufficiently close to what a person wants so it can be used "as is", probably it is not practical for him or her to have a computer. Sophisticated custom programming gets to be rather expensive, and, unless it is for a very large group practice, I don't think in most cases you can justify the cost of custom programming. I think you should be able to find off-the-shelf programs that do what you want to do within reason.

DR. GOTTLIEB Would the availability of software for orthodontists be greater with the major brands of equipment?

DR. PHILLIPS I would say so, simply because a software developer wants to develop software for machines that a lot of people own. They are not going to spend $100,000 or even $10,000 on programming for a machine that has a very low sales volume. They are going to do exactly the opposite, and spend their time and effort programming for machines that are big sellers. They can expect to sell more of their programs that way, which is a reason for buying the major brand machines. At some point you're going to want programs that may not be orthodontically related, and you will have a bigger selection of off-the-shelf programs if you've bought one of the major machines. It's very frustrating to go to a computer store to buy a program, or to see some program advertised in a magazine, and find out that it will not run on your machine.

DR. GOTTLIEB Is it a problem to deal with small companies such as we may see in our field?

DR. PHILLIPS A very small company may have good programs, but may not stay in business. On the other hand, a big, well established company may stay in business, but their programs may not be what you want, so this is indeed a problem. However, the problem is decreasing every year, because there are more programs available and they are more directly suited to orthodontics. But you have

to remember that there aren't very many orthodontists in the United States and a lot of dentists, and the name of the game in making money in the software business is volume. If a vendor has a market of 125,000 people versus maybe 7,000 or 8,000, you can understand why they put their efforts toward writing programs for general dentistry instead of for orthodontics. So in the past there really haven't been many good programs specifically written for orthodontics. This is still a problem. Although there are more orthodontic programs available every year, there are still relatively few. Most of them are modifications of medical programs or general dentistry programs.

DR. GOTTLIEB What are your priorities for an orthodontist setting out on this search?

DR. PHILLIPS It would be software first, but if the software I wanted was not available on one of the major hardware systems I would probably not buy that software. The only exception to that would be if it was from a really big software company or a big management company, because I'm very wary of buying a program that runs on what I would call an off-brand computer that might not be available very long. It takes a lot of time to enter practice data into any computer system, and if you spend three to six months of someone's time entering data and the hardware is no longer supported by the manufacturer, then it's been a lot of time wasted. So I would search for software first, but as part of that search be very aware of what equipment it ran on. I think that's very important.

DR. GOTTLIEB A great many orthodontists approach this problem with the idea that they are going to have custom software, and I don't think we can overestimate the problems that they can run into in trying to get custom software.

DR. PHILLIPS That is true. I don't think they realize the cost involved in custom software, or understand the difference between writing a program and testing it. I don't believe that a program of any magnitude has ever been written that didn't have some sort of problems in it, things that needed to be changed, so-called "computer bugs". Because of the literally hundreds of millions of possible combinations of data, it is not humanly possible to test every conceivable combination when you write a program. So unless a program has been tested and in use for awhile, it will have bugs or errors in it that will either stop it completely or give incorrect answers to questions. So, in most cases, you are better off with an off-the-shelf program that has had thousands of copies in use for years.

DR. GOTTLIEB Could we give some guidelines to someone who insists that he has to have custom software? For example, should he pay a premium to get a guaranteed system within a certain amount of time, and not have an open-ended contract?

DR. PHILLIPS That's one possibility. The ideal situation is to find a programmer who can demonstrate having designed a program similar to what you want that is in successful operation somewhere.

DR. GOTTLIEB Where do you find such a person?

DR. PHILLIPS Unless you're lucky enough to know an orthodontist who has a program that you want, I think your best bet is to go to a software company that specializes in designing software.

You'll find them in most major cities. I know of too many instances where an individual has gone to two or three programmers, spent a lot of time and money, and still did not get the program he wanted.

DR. GOTTLIEB So you would rather find someone who already has something close to what you want and modify the existing program?

DR. PHILLIPS I would certainly recommend that. If you're talking about something in the business or data management area, I would want to find a programmer who had actually written such programs, and in most cases the economical way is to modify an existing program. It's pretty rare that you'd have such a unique need that a program would have to be written from scratch. It isn't like something to operate a satellite. Business programs are basically the same, and you should be able to find a programmer familiar with that type of program. You would have to let him know the specific things that you want to incorporate into that program.

DR. GOTTLIEB Apart from a hobby interest, do you think there's any need for an orthodontist to know programming, either to alter the program or to write programs?

DR. PHILLIPS It's not necessary in my estimation. You could make a similar case for knowing enough about plumbing or electricity to handle all of those needs in your office. I think it's not a good use of an orthodontist's time, unless it's something that he or she wants to do to learn as a new skill or as a hobby. Programming is very time-consuming, and I don't think most orthodontists have the time to do a good job of it. You can hire skilled programmers, and spend your time on things that you enjoy more or that are more productive for you.

DR. GOTTLIEB If an orthodontist did want to get into custom programming, where would he find the programmer who could accomplish the job in a reasonable amount of time and at a reasonable cost?

DR. PHILLIPS If you don't have personal knowledge of such an individual, computer stores are probably your best bet, but to go to such a place cold and ask them to recommend someone to you is somewhat like going to the yellow pages for a dentist. My experience has been that, although there are lots of programmers available, they may or may not be able to do what you want done, either ever or at a reasonable price. Most software houses that do only programming, in contrast to the individual programmer who perhaps does it part-time, will not even give you a set dollar figure to do a certain job. They will do it on an hourly basis, and there is a good reason for that. It is very difficult for a non-technical computer person such as an orthodontist to tell a programmer exactly what he wants. The programmer may be able to do what the orthodontist wants, but once that is done the orthodontist will say, "Yes, but what I really wanted was this", and there is add-on after add-on. The people that write software are quite aware that in most cases the customer is not able to verbalize exactly what he or she wants. So usually they will take the job on an hourly basis, and if you keep saying that you want more things, they will be glad to do it for you. If you are very specific and know exactly what you want and can spell it out very accurately, it is not going to cost you very much. But if you go in with a vague idea that you want your bookkeeping automated, you are in for trouble.

DR. GOTTLIEB There is something about programming that is not as simple as it would appear to someone who does not know programming.

DR. PHILLIPS That is true. If you have a certain form or format for a report in mind, chances are there are programming reasons for not doing it that way. In nearly every case, the programmer can give you exactly what you had in mind, but he will be able to generate the information in a slightly different form much more easily from his standpoint. So, if you can live with some changes from your original concept, chances are the job will be accomplished much faster and cheaper. A programmer makes flow sheets that break down any procedure into a number of sequential parts, with more steps than you can imagine in most cases. You don't just go from what you want directly to the answer. That is what your brain seems to do, but for the computer to do it, the task must be broken down into many steps, and each one of those steps has to be programmed. It's a whole different language and a different way of thinking.

DR. GOTTLIEB The computer language the program is written in can also be one of the complications, can it not?

DR. PHILLIPS Oh, yes. Some languages will handle numbers better, and some will make a program operate or run much faster than others.

DR. GOTTLIEB Let's talk about language a minute. An orthodontist will hear about Basic, Pascal, Cobol, Fortran, and probably others. It's a non-standard field, just as the world has a lot of languages, and you can get into just as much trouble, I think, with program languages as you could in any foreign country. Is there a language that is more compatible with what orthodontists are doing than others?

DR. PHILLIPS I would have to answer "yes and no" to that. Actually, from the operator's standpoint, other than differences in operating speed, it doesn't matter too much what language a program is written in. Assuming the program does what you want it to do and operates as fast as you feel it needs to operate, then it's OK unless you need to modify it. If the program needs to be modified and is written in what is called machine language, then you are in big trouble. For all practical purposes, it is not modifiable by the operator. Technically, it is modifiable, but in terms of time required to do it, modification costs too much. You would be better off just to get another program.

DR. GOTTLIEB "Machine language"--is that the language internal to the system?

DR. PHILLIPS That is the language that the microprocessor chip understands or utilizes to manipulate data. It uses zeros and ones in various combinations. Because it is the language that the machine itself uses, it does not have to be translated by the computer before processing the data, so a program runs very quickly. Anything else, so-called high level languages such as Pascal or Basic, are more like English with As, Bs, and Cs, but they have to be interpreted by the machine and converted to machine language before processing by the computer. Programs written in high level languages take a little longer to run, but when operating at a few million calculations per second, you can understand that when we talk about slow or fast it is different from talking about slow or fast in almost anything other than computer terms.

DR. GOTTLIEB But you can't talk to every machine in every language, can you?

DR. PHILLIPS No, you can't. So the main thing to watch for is that the program you are interested in is written in a language that will run on the machine that you have; or, if you are looking for an off-the-shelf program, you buy the program first and then get the machine that will run it.

DR. GOTTLIEB Subsequently, you would have to get programs that are written in the language the machine can handle.

DR. PHILLIPS That is right. There are applications programs and operating programs. An applications program might be an accounting program or a file management program. Operating system programs instruct the computer how to run the applications programs, and these are often specific to a particular computer. In other words, the operating system for the Apple computer is not the same system that runs the IBM computer. Just because you have a program that is written in Basic doesn't mean that it will run on any machine that will handle Basic.

DR. GOTTLIEB Even though they both call themselves Basic, it may not be the same Basic. There might be subtle differences that don't permit you to run without knowing what those differences are.

DR. PHILLIPS That is absolutely true. Just because you have a word-processing system written in a particular language does not mean that it will run on any other machine than the one it's specifically written for. A so-called Apple Basic that runs on an Apple machine is not the same as IBM Basic. There are certainly similarities, but there don't have to be many differences. One difference may keep it from running on the other machine.

DR. GOTTLIEB Should the orthodontist know how to operate the system?

DR. PHILLIPS Here again, I think that's not totally necessary. I have an interest in the technology, so I make it a point to know how to operate the system, but what is important is that someone in the office be very knowledgeable in the operation of the system--not in the theory behind it, but in the operation. I think that the main contribution of the orthodontist is to show that he or she wants the system to function. There is a certain amount of frustration involved in getting any computer system going, and I think that the staff members involved with it have to know that the orthodontist is vitally interested that this succeed--that he or she is really interested in the project, knows that there are going to be problems, is not going to get totally upset if something goes wrong, and will help rather than fuss about it. I think this real strong support is what is needed more than actually being able to operate the equipment.

DR. GOTTLIEB But if you only had one person in the office who could operate the equipment, then you'd better know how.

DR. PHILLIPS Yes, I would say so, and fortunately within the last year things have gotten so much less complicated for the user that it's not a major task to learn to operate the equipment these days. You can make a how-to-do-it book in your own words, starting with how to plug it in, where to turn it on, and so on. The operation of the equipment itself is very simple, and recently the

programs have also gotten much easier to use. It's really not the formidable task that it was a few years ago, so anyone who is capable of operating other business machines in the office can certainly learn to operate the computer.

DR. GOTTLIEB If you were to lose the person or persons who operate the computer, what kind of shape are you in then? Is it simple to find people in the community who can operate it or can be trained rapidly?

DR. PHILLIPS How big a problem that is depends on the programs you are using. If you have a very small staff, it would be well for the orthodontist to know how to operate the equipment. At the very least he could fill in in an emergency and perhaps be the one who trains the new person. A computer can help you maintain continuity in the office if you do lose key staff members, but you have to have more than one person who knows how to operate the system or you are pretty vulnerable. So, whether it's the orthodontist and one other person or two staff members, I feel you need at least two knowledgeable people to keep the minimum functions going without any interruption.

DR. GOTTLIEB Is it simple enough now so that in the situation we're describing you could train people in-house with a procedure manual and with one knowledgeable person?

DR. PHILLIPS I would say so. If you bring a new person in to run bookkeeping programs, what they really have to know that you can't teach quickly is bookkeeping. If they know that, then it is easy to teach them how to do it on the computer.

DR. GOTTLIEB Do you think that a computer is threatening to staff members?

DR. PHILLIPS Yes. Although some staff members are extremely eager to get started on a computer and adapt to it very quickly, there are definitely others who are threatened by it. And this again is where the orthodontist has to show a lot of support, to ease someone into it who is somewhat reluctant to start using it.

DR. GOTTLIEB Actually, I think the computer is a stimulant for a staff. You might hire a higher achiever than you would normally expect to hire in an orthodontic office and challenge her with a computer--advance to the next plateau, so to speak.

DR. PHILLIPS I would agree with that. I have experience with a number of staff members who look upon using computers as an opportunity to learn new skills, and they are eager to do it. There is always another level of complexity, so they're not going to get bored with it very quickly. And it can help them to do their job more efficiently, which has to be satisfying.

DR. GOTTLIEB Can the orthodontist expect a computer to replace any staff?

DR. PHILLIPS No, I would say not. It makes staff members more efficient, and it can, in many cases, allow the same size staff to accomplish more. In that sense, it may make it possible for you to delay hiring additional staff people, but it's not likely to replace anyone. In my experience, it simply allows those there to do their jobs more effectively and perhaps faster. One of the advantages of having a computer is that it forces an office to become better organized than it has ever been before

and, once that is accomplished, it also makes the transition from one employee to another much easier. Knowledge of office procedures does not leave with the employee; it is still with you if it's been programmed into the computer.

DR. GOTTLIEB Does an orthodontist have to think in terms of saving dollars in salaries in order to consider a computer to be cost-effective?

DR. PHILLIPS Not in my view. To me the advantage is not strictly in saving dollars, although that is a possibility. The advantage is in being able to do things more accurately in some cases, and in other cases to do things that you aren't doing and can't do without the computer. My concern is not strictly about the dollars per hour that the system costs to operate; it's more a desire to do things that are what I consider state of the art, much as orthodontists no longer sit down and assemble bands and brackets when they can buy preformed bands and brackets or bondable brackets. I feel that there are emotional or psychological advantages to be gained from knowing that you're doing a good job and doing things as accurately as they can be done. Any time a machine can take over a good part of a task from a person, I would rather have the machine do the things it can do very well and have the people in the office do things that only they can do well, such as relating to other people, making patients comfortable, making people happy, and thinking.

DR. GOTTLIEB That's why I have to think that one day they'll iron out the problem of entering treatment visit information, because pulling and refiling records is not human work either, and computers can do it so quickly and efficiently.

DR. PHILLIPS One of the major advantages of using the computer for that job is that no one has to search for the record in the file cabinet and later possibly refile it incorrectly. It's always safely stored at the same place and can't be misfiled.

DR. GOTTLIEB An orthodontist could send his data outside the office to a big, mainframe computer, or he could keep the whole operation in the office on a small computer. Which way do you see him going?

DR. PHILLIPS Back in the late '50s and '60s, mainframes were large, expensive machines costing millions of dollars, obviously out of the financial reach of any private practitioner, and very complicated to operate. So they were not available to an orthodontist. The next step was the development of what are called minicomputers, which were generally in the range of one hundred thousand to a few hundred thousand dollars in cost. This made them more available to smaller businesses, but not down to the level of an orthodontic practice. More recently, the very small computers, called microcomputers or personal computers, were developed. Initially, anyone using a computer was using a so-called mainframe. In the '70s, what was called distributive processing came along, which meant that you could have a terminal in your office on which you could type information and send it to a mainframe for storing or processing. The information might be sent back and forth by mail or transmitted over the phone lines. At that time, if you were going to use a computer in as small a business as an orthodontic practice, that was the way to do it. The step beyond that was the development of the small, inexpensive, so-called stand-alone system that is completely self-contained in your office. That's the state of the art now and I think by far the most

practical system in use today. I think the stand-alone system will continue in use for many years and will be the predominant one. However, I think that mainframes will again become accessible to the individual orthodontist using a small computer in his office as a terminal. Transmission of data over telephone lines or by satellite will allow the orthodontist to gain access to information stored in a few centers around the country or perhaps around the world. For instance, one might ask for all the information on mandibular advancement. I think someday there will be diagnostic data banks to which we will be able to transmit information concerning a patient and have the best minds in the world look at it and send back treatment suggestions for that patient. I think that is not in any way farfetched. Even now, there are national medical-dental data services that you can reach with your microcomputer by telephone, and receive abstracts or complete journal articles on virtually any subject.

DR. GOTTLIEB What's the distinction between a minicomputer and a microcomputer?

DR. PHILLIPS At one time there were very large differences among the mainframe, mini, and micro, in the amount of information that they could store, in their speed of operation, and in what they were able to do with the information, based on the complexity of their operating systems. There was also a physical difference in size and a dollar difference in both initial and operating cost. At the present time, that difference still holds between mainframes and anything else, but minicomputers and microcomputers are almost the same. In fact, the minicomputer is going the way of the large V8 sedan, and being replaced by the equivalent of the 4-cylinder compact. Today's microcomputer is the minicomputer of 10 years ago. They can store as much information as the mini. The speed is slightly slower, but from a practical standpoint it's not significant. The physical size is smaller. The cost is certainly much less. Perhaps the main difference between most microcomputers and minicomputers at the present time is in what is called networking, or connecting a number of different computers together to operate as one large system. Even that is not a clear distinction, because it is entirely possible to connect a large number of microcomputers together, but it may be more trouble than it's worth.

DR. GOTTLIEB How small or inexpensive a computer could an orthodontist think about?

DR. PHILLIPS Well, we have to be specific about what we want the computer to do. You don't need a $30,000 system to do most everything that an orthodontist would want to do. I would say to do the variety of things that most orthodontists would want to do, by the time you put in the printers, the computer itself, the storage devices, all the so-called peripheral pieces of equipment that you need to go with the computer itself, plus purchasing adequate programs, you're very likely going to invest $10,000-20,000 for most systems. If you want a much more elaborate system, involving multiple offices, then you can easily be up in the $50,000-plus range. This brings up another point. I think an orthodontist would be well advised to think of more than one machine. In the past, that would have meant a minicomputer with two or more terminals. In the present, it can very well mean two or more discrete, stand-alone microcomputers, which is not as extravagant or ridiculous as it may seem at first. Because of the small physical size of the machinery and the relatively small price, you can afford to have multiple computers located where they are needed. If something goes wrong with one of them, you have another that you can use, even though it may not

be in the handiest location in the office.

DR. GOTTLIEB Are you speaking of two simpler, less expensive systems to do two separate jobs, such as diagnostics on one and business functions on the other, which would total the same approximate cost as a large machine?

DR. PHILLIPS Yes, that's exactly what I'm talking about. There's a lot to be said for simplicity. Not only does less complexity mean less electronic interaction, but not using one computer for everything means that each computer is used a little bit less. Since there is no time sharing, there's a lot less chance of one program adversely affecting another in some completely unpredictable way.

DR. GOTTLIEB A large office would have to make sure that the machine it was using for the business functions was adequate for now and for the foreseeable future, and they might wind up with a computer just for that function.

DR. PHILLIPS That' true, but my experience has been that there is more chance that a computer salesman will push to sell you more computer than you will ever need than that you will be sold a computer that is not upgradeable to the extent that you may eventually need.

DR. GOTTLIEB Are you saying that an orthodontist could go to a computer store, and, assuming he made the right choice of everything, buy two entire systems--meaning a CPU, a CRT, and a printer--one he might use for diagnostics and one he might use for business, and pay say $8,000-10,000 for each system?

DR. PHILLIPS I'd say he could have two functional systems that in each case will do what he wants to do for less than $20,000 total, including the software.

DR. GOTTLIEB So you don't really have to worry about residual value, particularly since you have taken an investment credit and have depreciated the equipment over the years.

DR. PHILLIPS That's entirely true. The cost factor is just not the major one any more. Computers are now within a price range at which you can get your value out of them in a fairly short time. It's not something that you have to be able to justify for 20 years in order to make it a reasonable purchase.

Continued in December 1983 issue