Computers have become an indispensable part of every clinic. Though it may not be economically viable to install an 'Electronic Medical Records'[1] or EMR system in a small clinic, computers are used in some form or other for patient data management.

An EMR is software that allows you to create, store, organize, edit and retrieve patient records on a computer. But it is more than just the electronic equivalent of paper. Advanced EMRs also allow you to automate many time-consuming, paper-driven office tasks. Some are even equipped with 'Computerized Physician Order Entry'2 tool. EMR software has many advantages which are beyond the scope of this article.

Most of the clinics use custom made software to record patient data, but this does not include all the features of a typical EMR. This is generally used for storage of demographic and personal data, storage and retrieval of patient ID and appointment scheduling. Even these simple software packages require a database backend which in the majority of cases is MS Access though some use more advanced RDBMS (Relative Data Base Management Systems)3 like Oracle or MS-SQL server.

The data stored in these database systems is usually used only for income auditing, billing and other similar tasks. However this data could provide a wealth of information which is important from the practice monitoring and marketing perspective, though this is often overlooked in small practises and clinics.

Big hospitals use various kinds of data to monitor performance. However small clinics use only traditional financial statements. Going beyond figures like net revenues and costs and using meaningful financial reporting can offer a deeper and ultimately more effective analysis.
Medical practice is traditionally considered as a venerable profession outside the realm of a business environment. However the trend is slowly changing and more and more practitioners are considering their practice as a business and their patients as customers. This is especially true for certain specialties like cosmetology and dentistry and they have started realising that the principles of management applies to practice management as well. Hence it is important to think what else needs to be monitored apart from total income in a small medical establishment.

A few of the indices which may be important in practice management are mentioned below.

1. New Patient flow: This can be expressed as a percentage of previous month or a certain time period as it is important only when considered with respect to the previous month. It is not relevant for a new clinic. However for an old clinic it indicates the efficiency of the marketing management.

2. Old patient rate (OPR): This can be expressed as a percentage of total number of patients. OPR indicates the efficiency of the practitioner to retain the patients. However for a more sensitive measure of patient retention, the number of old patients returning after the standard follow-up period needs to be considered.

3. Average collection per new patient: This index may increase the income for a short period. However it is important to retain it within acceptable limits for each specialty as the longterm effects are not favourable.

4. Average collection per patient: This is a very sensitive measure of patient loyalty.

5. Revenue growth as a percentage of previous year.

6. Descriptive statistics of geographic location, socio-economic status, demographic profile and referral from other practitioners and clinics: This is very important for strategic planning.

It is important to understand that most, if not all, the information needed to calculate the above indices is essentially data that the hospital already has available. It is a matter of taking the various datasets and merging them into useful information upon which decisions can be formed.

Databases are fundamentally tools to allow people to organize and manipulate large amounts of data using the power of the computer to quickly translate and deliver that information in a humanly readable format.

Databases fundamentally organize the data into hierarchies. The building blocks for creating databases starting from the top level are the data structure or schema. Each data structure is made up of a series of records, and each record has a set of predefined fields.

Databases use a series of Tables to store the data. A table simply refers to a two dimensional representation of your data using columns and rows. Each column in the table is given a unique name which would be something like first_name, last_name, email.

To reduce the redundancy of information, some tables will have a primary key which can be used to link to other tables. It will become a foreign key for the linked table. For example if a patient comes for multiple visits we may have to enter the patient data each time if there is only one table. However if the patient data is stored in a separate table with a unique number, only that number needs to be entered to the 'visits' table so that data redundancy is minimised. This unique number is called as the primary key for the first table and foreign key for the 'visits' table.

If you give data from your patient management system to database professionals for analysis then you have to deal with certain crucial ethical issues like data privacy. The data often includes the personal details of the patient, diagnosis and investigation reports which have to be treated as confidential. Hence it is imperative for healthcare practitioners to learn basic data warehousing techniques to remove sensitive data before outsourcing it to professionals for more detailed analysis.

Data warehousing is the process of consolidating data to a central store so as to make analysis of data easier. The practitioner who owns the data has to remove the personal information in the database before giving it for analysis. Usually all the personal information will be grouped together in a single table. Though in some cases this database table can be safely dropped or removed. This may make the database unstable because the primary key is often stored in this table. This table can also contain some useful information like age, sex, region, occupation and socio-economic status. Hence it is better to remove individual fields like first_name, last_name, phone number, address, email etc.

Data warehouse also implies that the data is manipulated and consolidated in a separate location, different from operational data used for day-to-day activities. For a small clinic, it essentially means that you have to take a copy of the database and work on that so that original database is not damaged. Though I have used the term 'Data warehousing' here, it is a much broader concept than just backup and deleting unnecessary fields.

Microsoft® Access is a powerful program to create and manage your databases. It has many built in features to assist you in constructing and viewing your information. A detailed description of MS Access is beyond the scope of this article. Deleting a field is usually done in the table design view as depicted in Fig 1. It involves the following steps.

1. Click on the Tables tab on the Access main screen
2. Highlight the name of the table to be modified and click on the Design button.
3. Make the necessary changes.
4. Save the table by pulling down the File menu and choosing Save.
5. Close the table by pulling down the File menu and choosing Close.

A database query is usually expressed in SQL or structured query language. The indices I mentioned earlier can be easily calculated from the database by using simple SQL. In MS Access you can easily create a query in Design View as depicted in Fig 2 to filter the information in your table. You establish a set of criteria when you create a query.
However there is other information which can be derived from the database apart from the calculation of above indices and descriptive statistics of demographic data like age, sex and location.

Data mining has been defined as "The nontrivial extraction of implicit, previously unknown, and potentially useful information from data" and "The science of extracting useful information from large data sets or databases". Although it is usually used in relation to analysis of data, data mining, like artificial intelligence, is an umbrella term and is used with varied meaning in a wide range of contexts. It is usually associated with a business or other organization's need to identify trends. Data mining techniques can be used for patient data also. It will help us find patterns which may not be evident at the first look like the referral pattern between specialists in a clinic. Data mining converts data into knowledge.

As with any complex business, clinics can create all kinds of statistics. But if not used wisely, the only thing all these data will add up to is a bunch of numbers. Organizations that get the best results from this data are those that get all levels of leadership involved in analyzing the information.

Strategy formulation frequently includes a SWOT analysis, or an assessment of internal strengths and weaknesses (SW) combined with environmental opportunities and threats (OT).

However, SWOT analysis may sometimes fail to reflect the true picture. Senior managers act only on perceived strengths, weaknesses, opportunities, and threats as filtered through their own lenses and those of middle managers, physicians, nurses, and others in the organization. These filtered perspectives prevent senior management from seeing the real SWOT, thereby creating a potential for suboptimal strategic decisions. Patient data will help in having a more realistic SWOT analysis.

Using the patient data we can monitor the volumes and the practice indices for all departments to help management understand where the hospital is growing or declining during the year. The analysis will also help in understanding the impact of various factors like changes in payer contracts, coding and reimbursement regulations. It can even unearth a few not so obvious factors like practice patterns of physicians. This includes the pattern of sending investigations and other departmental referrals. Patients may initially succumb to physician demands for cutting-edge technology when less expensive technology would do an acceptable job without compromising quality. This might increase the 'average collection per new patients' but makes patient retention difficult. Clinics need to think carefully about the mix of supportive services they provide to patients, their cost, and whether a less expensive approach could achieve the same results.

Another important aspect which needs to be considered is whether a clinic can provide every service or whether they should specialise in certain key services. Modern healthcare organizations should consider which services and programs they wish to emphasize, for which kinds of patients, and in which localities and then eliminate programs and other activities that do not fit that focus. Analysis of existing patient data will provide key insight into this.

Strategic trade-offs should also be evaluated in terms of competitive scope and pricing policy both of which can be defined from the existing patient data. The goal of organizations with a low-price strategy is to deliver a product of acceptable quality at a price below that of its competitors. By contrast, organizations with a high-price strategy select small market segments and offer services that are widely acknowledged as superior. It is important to assess based on previous data, which segment of the population the clinic is catering to before introducing these services.

Detailed analysis of patient data will also provide useful information for other managerial responsibilities like controlling and staffing. For example it will help in procuring the right supplies at the right time and in the right quantities so that there will not be any wastage or shortage. It will also help in identifying the bottle necks in patient flow and adequately staff such segments.