SQL Reporting

Report
Geoff Kimber





Realize the importance of learning SQL
Recognize SQL platform differences
Identify relational database concepts
Apply SQL syntax to a real-life example
Review SQL references




Data organized in tree structure with parent
and child segments
Parent to child is a one to many relationship
Implies repeating information, generally in
child segments
Fileman
 Prescription file with refill multiple field




Data consists of records stored in tables
Each row is unique
Column values are of the same data type
Sequence of rows and columns is insignificant

Structured Query Language [SQL]
 Is an ANSI computer language that is used to
interact with many relational databases
▪ Oracle
▪ MS-SQL Server
▪ MySQL, PostgreSQL and others
 SQL is a set-based language
 SQL is structured like English
▪ SEQUEL (Simple English QUEry Language)

VA is truly data rich
 Each site has between 2-4 TB of data†
▪ non-imaging data

VA FileManager [FileMan]
 Only within VistA
 Limited functionality for non-programmers
 Runs in the production environment
▪ Workflow and resource concerns
† - Pham R. The VA Data Lifecycle (Internals, Data Flows, and Business Intelligence)

Data repositories (warehouses)
 Moves the analysis off-line
 Provides accessible, yet very secure, data
▪ ODBC, ADO and LINQ
 Greater functionality and access control
▪ The “official” data is secured within VistA
▪ Independent of VistA
▪ Microsoft Active Directory based access
▪ Maintain the front-line stance of CPRS/VistA

Various VA relational databases




Corporate Data Warehouse (CDW)
Regional Data Warehouse (RDW)
VISN Data Warehouse (VDW)
Potential for local data warehouses
▪ Extract raw data from VistA off-hours
▪ UPSERT the data into an SQL server automatically
▪ Users run their own reports and access is secure

Data warehouses = relational databases
 SQL = the language of relational databases

Relational databases are everywhere
 Learn a portable skill
 Take a systems approach to collaboration
 Build reusable code
 Work more efficiently


VA has an enterprise license
Microsoft SQL Server is functionally superior
to Microsoft Access
 Secure
 Backups
 Procedural code
 No database size limit (software)
 Efficiency

Two of the different SQL flavors
 Transact SQL (Microsoft SQL Server)
▪ Manage databases/services, procedural code
 Jet SQL (Microsoft Access)
▪ Information retrieval, program component

Each program has its own SQL flavor
 Core keywords and functions follow standards
 ANSI guidelines are “considered”

In SQL:
 Something is either part of a set or it is not
 Order is meaningless
 SQL Server supports procedural code
▪ VBA can be used in MS Access for procedures

Dimension tables

Fact tables



Contain attributes that describe records in
fact tables
Each record contains unique identifier
Generally more stable
 Drug name, provider name, patient name and
address
▪ Similar to Drug file (#50), New Person file (#200),
Patient file (#2)


Capture operational data
Contain multiple unique identifiers
 One identifier for each record
 One identifier for each data element in dimension
tables
 Similar to Fileman IEN

Usually contain numerical and date fields
 Prescription fills, Outpatient visits, lab test results
 Similar to prescription file (#52), V POV
(#9000010.07)

Dim.LocalDrug
 LocalDrugSID
 Sta3n (3 digit station number)
 LocalDrugNameWithDose
 VaClassification
 PricePerDispenseUnit
 NationalDrugSID
LocalDrug
SID
LocalDrugNameWithDose
PricePer
VaClassifi Dispense National LocalDrug
cation
Unit
DrugSID SID
735055
ATENOLOL 50MG TAB
CV100
0.0079
2223484 735055
735056
ATENOLOL 100MG TAB
CV100
0.0075
2221541 735056
739331
ATENOLOL 25MG TAB
CV100
0.0068
2223485 739331
741066
ATENOLOL 12.5MG (1/2 OF 25MG) TAB CV100
0.0042
2223485 741066
748289
STUDY: ATENOLOL TAB
NULL
-1
IN000
748289

Dim.NationalDrug
 NationalDrugSID
 DrugNameWithoutDose
 DosageForm
 Strength
NationalDrugSID
DrugNameWithoutDose
DosageForm Strength
2221541
ATENOLOL
TAB
100
2221542
ATENOLOL
INJ
500
2221543
ATENOLOL
TAB
50
2221544
ATENOLOL/CHLORTHALIDONE
TAB
NULL
2223485
ATENOLOL
TAB
25

sPatient.sPatient
 PatientSID
 PatientName
 PatientSSN
 AddressLine1
 City
 State
 ZipCode

RxOut.RxOutpat
 RxOutpatSID
 PatientSID
 ProviderSID
 LocalDrugSID
 IssueDate
 RxNumber

Rxout.RxoutpatFill
 RxoutpatFillSID
 RxOutpatSID
 FillType
 ReleaseDateTime
 Qty
 DaysSupply
 UnitPrice

Metadata
 Information about data
 Includes CDW to VistA field crosswalk
 http://vaww.cdw.r02.med.va.gov/metadata/Reports/Form
s/AllItems.aspx

Clauses perform set-based calculations
CLAUSE
SELECT
Definition
Defines what you will see. You can identify columns,
calculations, literal text, and even perform branching.
FROM
Defines the origin of the data. Tables can be joined.
WHERE
Defines data filters. Filters data before it is queried.
GROUP BY
HAVING
ORDER BY
Defines how data is grouped. Works hand in hand with
aggregate functions, i.e. how many in each group
Defines data filters. Filters data after the query is run. An
example: return only those items having a count > 5.
Defines the field(s) used to sort the data – ASC or DESC
Function
Definition
COUNT
Aggregate Function: Counts the number in a group.
SUM
Aggregate Function: Sums the numbers in a group
AVG
Aggregate Function: Averages numbers in a group
MIN and MAX
Aggregate Function: Finds the minimum and maximum
numeric values in a group.
UCASE and LCASE
Scalar Function: Returns text in uppercase or lowercase,
respectively
LEN
Scalar Function: Returns the length of a string
Microsoft Access supports the aggregate functions FIRST and LAST
SQL Server has a COUNT(DISTINCT...) aggregate function
Operator
AND
OR
NOT
=, >, <, >=, <=
Definition
Logical Operator: A and B are both true to equal TRUE
Logical Operator: A or B is true to equal TRUE
Logical Operator: Switches TRUE to FALSE and vice versa
Equality Operators: equals, greater than, etc.
BETWEEN
Equality Operator: TRUE if value is between the two criteria.
Values that equal the criteria are included.
Like
Equality Operator: TRUE if text contains the pattern defined.
LIKE allows for wildcard characters (see below)
+, -, *, /
+ or &
Mathematical Operators
Concatenation Operators: Join two text strings. (see below)
Access uses “*” as a wildcard and double quotes; SQL Server uses “%” and single quotes
Access uses the ampersand to concatenate “&”, SQL Server the plus sign “+”
Select LocalDrugSID
,NationalDrugSID
,LocalDrugNameWithDose
from vdw.dim.localdrug
Where LocalDrugNameWithDose like
'%atenolol%'


Case is not normally important, but helps
with readability
Spaces, tabs, and hard returns have limited
importance
 Important in text fields
 Not important between commands
 Help improve readability

Microsoft query analyzer color codes key
words for readability

SQL databases have many different data types
 Varchar
 Int
▪ Bigint
▪ Smallint
 Decimal
 Float
 Datetime
 Money

Select
 Main command used to retrieve data from SQL
tables
 Followed by list of fields to return
 ‘*’ (without quotes) returns all fields in table
▪ Normally, avoid using ‘*’ unless you need all the fields in
a table

From vdw.dim.localdrug
 Indicates source of data from which you are
selecting
 Vdw.dim.localdrug
▪ VDW is database name
▪ VDW => VISN Data Warehouse
▪ RDW => Regional Data Warehouse
▪ CDW => Corporate Data Warehouse
▪ Dim is schema
▪ Localdrug is table name

Where
 Allows you to restrict or filter the table contents
to just the results you want
 Extremely versatile

Like
 ‘%’ is a multicharacter wildcard
 ‘%atenolol%’ - contains ‘atenolol’
 ‘atenolol%’ – begins with ‘atenolol’
 ‘%atenolol’ – ends with ‘atenolol’


Case is not normally important, but helps
with readability
Spaces, tabs, and hard returns have limited
importance
 Important in text fields
 Not important between commands
 Help improve readability

Microsoft query analyzer color codes key
words for readability
Select LocalDrugSID
,NationalDrugSID
,LocalDrugNameWithDose
from vdw.dim.localdrug
Where localdrugnamewithdose like
'%atenolol%'
LocalDrugSID
NationalDrugSID
LocalDrugNameWithDose
735055
2223484
ATENOLOL 50MG TAB
735056
2221541
ATENOLOL 100MG TAB
739331
2223485
ATENOLOL 25MG TAB
741066
2223485
ATENOLOL 12.5MG (1/2 OF 25MG) TAB
1200004446
2223484
STUDY: ATENOLOL 50MG TAB
Select VA FileMan Option: PRINT File Entries
OUTPUT FROM WHAT FILE: DRUG//
SORT BY: GENERIC NAME// 'GENERIC
NAME["ATENOLOL"
WITHIN GENERIC NAME["ATENOLOL", SORT BY:
FIRST PRINT FIELD: NUMBER
THEN PRINT FIELD: PSNDF:NUMBER
THEN PRINT FIELD: GENERIC NAME
THEN PRINT FIELD:
DRUG LIST
PAGE 1
NUMBER NUMBER
GENERIC NAME
-----------------------------------------------------------------------3217 4331 ATENOLOL 25MG TAB
4068 4329 ATENOLOL 50MG TAB
4069 4328 ATENOLOL 100MG TAB
9751 4331 ATENOLOL 12.5MG (1/2 OF 25MG) TAB
12975 4329 STUDY: ATENOLOL 50MG TAB



Used to connect two or more tables together
using data elements that are shared by
individual tables
Similar to fileman jumps, with more power
Inner joins, outer joins and full joins

Produces a result that includes only the
records that are the same in both tables
Select a.localdrugnamewithdose
, b.strength
from vdw.dim.localdrug as a
inner join vdw.dim.nationaldrug b
on a.NationalDrugSID = b.NationalDrugSID
where LocalDrugNameWithDose like
‘%atenolol%’
localdrugsid
localdrugnamewithdose
strength
735055
ATENOLOL 50MG TAB
50
735056
ATENOLOL 100MG TAB
100
739331
ATENOLOL 25MG TAB
25
741066
ATENOLOL 12.5MG (1/2 OF 25MG) TAB
25

Results contain all records in one table and
only matching records in another table
 Left joins – return all results in table specified in
‘from’ statement and all matching tables from
‘joined’ table
 Right joins – return all results in table specified in
‘joined’ table ‘from’ statement and all matching
tables from ‘from’ statement
Select a.localdrugnamewithdose
, b.strength
from vdw.dim.localdrug as a
left outer join vdw.dim.nationaldrug b
on a.NationalDrugSID = b.NationalDrugSID
where LocalDrugNameWithDose like
‘%atenolol%’
localdrugsid localdrugnamewithdose
strength
735055
ATENOLOL 50MG TAB
50
735056
ATENOLOL 100MG TAB
100
739331
ATENOLOL 25MG TAB
25
741066
ATENOLOL 12.5MG (1/2 OF 25MG) TAB
25
1200004446 STUDY: ATENOLOL 50MG TAB
NULL

Indicates lack of data
 Does *not* mean zero
 Math operations with null values have null results
▪ 2*null = null
▪ 0 * null = null
 Text and logic operations on null values return null
 Several functions exist to deal specifically with
null values

Returns all values in both tables

Not particularly applicable in our setting

http://www.w3schools.com/sql/
 w3schools is an excellent learning resource

http://msdn.microsoft.com
 Microsoft Developer Network is the “go to” place
for all coding questions for Microsoft

SAMS Teach Yourself SQL in 10-Minutes
 Excellent starting point (uses older SQL style)

SAMS Teach Yourself Microsoft SQL Server T-SQL in
10-Minutes (good intro for Microsoft SQL Server)

\\vhacdwa10\BIPLTrainingContent\CDW_Data_101
 Recorded Live Meetings

http://vaww.cdw.r02.med.va.gov/Pages/
 CDW Home
 CDW Best practices
 MetaData
 Relationship Diagrams
 Additional Training

VA TMS
 Books 24x7











Forta, Ben. SAMS Teach Yourself Microsoft SQL Server T-SQL in 10-Minutes. Indianapolis, IN:
SAMS, 2008. Print.
Forta, Ben. SAMS Teach Yourself SQL in 10-Minutes –Third Edition. Indianapolis, IN: SAMS,
2004. Print.
Kline, Kevin, Hunt, Brand, and Kline, Daniel. SQL in a Nutshell –Third Edition. Cambridge,
MA: O’Reilly Media, 2008.
Molinaro, Anthony. SQL Cookbook. Cambridge, MA: O’Reilly Media, 2005.
Henderson, Ken. The guru's guide to Transact-SQL. Boston: Addison-Wesley, 2000. Print.
Roman Steven. Access Database Design & Programming –Third Edition. Cambridge, MA:
O’Reilly Media, 2002. 11-124. Print
Bluttman Ken, and Freeze Wayne S. Access Data Analysis Cookbook. Cambridge, MA:
O’Reilly Media, 2007. 1-106. Print
Getz, Ken, Litwin, Paul, and Baron, Andy. Access Cookbook Cambridge, MA: O’Reilly Media,
2002. 1-56. Print
Date CJ. SQL and Relational Theory. Cambridge, MA: O’Reilly Media, 2009.
http://www.w3schools.com/sql/
http://msdn.microsoft.com

similar documents