Understanding and using NAMCS and NHAMCS

Ppt on data tool and basic programming techniques

1. 1
Understanding and Using
NAMCS and NHAMCS Data
Data Tools and Basic Programming Data Tools and Basic Programming
TechniquesTechniques
Donald Cherry
Ambulatory and Hospital Care Statistics Branch
Division of Health Care Statistics
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Centers for Disease Control and Prevention
National Center for Health Statistics
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Centers for Disease Control and Prevention
National Center for Health Statistics

2. 2
Overview

Some important features of NAMCS & NHAMCSSome important features of NAMCS & NHAMCS

File structureFile structure

SETSSETS

Exercises using SAS Proc Surveyfreq/Proc Surveymeans, Exercises using SAS Proc Surveyfreq/Proc Surveymeans,
SUDAAN, STATASUDAAN, STATA

Downloading data & creating a SAS datasetDownloading data & creating a SAS dataset

Simple frequencies with/without standard errors Simple frequencies with/without standard errors

Creating a new variable-AsthmaCreating a new variable-Asthma

Visit rates for asthma-male/femaleVisit rates for asthma-male/female

Total number of digestive write-in Total number of digestive write-in
proceduresprocedures

Time spent with physicianTime spent with physician

Considerations Considerations

SummarySummary

3. 3
NAMCS and NHAMCS

National Ambulatory Medical Care Survey National Ambulatory Medical Care Survey
(NAMCS)(NAMCS)

Visits to nonfederal, office-based physicians Visits to nonfederal, office-based physicians

CHC’s sampled beginning in 2006CHC’s sampled beginning in 2006

National Hospital Ambulatory Medical Care National Hospital Ambulatory Medical Care
Survey (NHAMCS)Survey (NHAMCS)

Visits to hospital outpatient and emergency Visits to hospital outpatient and emergency
departmentsdepartments

4. 4
NAMCS Sample Design

Three stage designThree stage design

112 PSUs112 PSUs

Physician practices within PSUsPhysician practices within PSUs

Patient visits within practicesPatient visits within practices

One-week reporting periodOne-week reporting period

About 30 visits per doctor are typically sampledAbout 30 visits per doctor are typically sampled
For 2006—3,350 doctors sampledFor 2006—3,350 doctors sampled

104 CHC’s sampled & physician visits included in 104 CHC’s sampled & physician visits included in
samplesample

Total visits 29,392Total visits 29,392

5. 5
Scope of the NAMCS
Basic unit of sampling is the physician-Basic unit of sampling is the physician-
patient visit patient visit
In scope visits:In scope visits:

Must occur in physician’s officeMust occur in physician’s office

Must be for medical purposesMust be for medical purposes

Administrative visits not sampledAdministrative visits not sampled

House calls, emails, phone calls not House calls, emails, phone calls not
sampledsampled

6. 6
Scope of the NAMCS (cont.)

Physicians must be:Physicians must be:

Classified by AMA or AOA as primarily Classified by AMA or AOA as primarily
engaged in office-based patient careengaged in office-based patient care

nonfederally employednonfederally employed

not in anesthesiology, radiology, or not in anesthesiology, radiology, or
pathology pathology

64 percent unweighted response rate in 64 percent unweighted response rate in
20062006

CHC’s are Federally Qualified or “look CHC’s are Federally Qualified or “look
alike”alike”

7. 7
NHAMCS Sample Design

Multistage probability designMultistage probability design

First stage sample of 112 PSUsFirst stage sample of 112 PSUs

Hospitals within PSUsHospitals within PSUs

Clinics within OPDs, Emergency Service Clinics within OPDs, Emergency Service
Area (ESA) within EDsArea (ESA) within EDs
Patient visits within clinics, ESAsPatient visits within clinics, ESAs

4-week reporting period4-week reporting period

382 hospitals sampled in 2006; 35,849 ED 382 hospitals sampled in 2006; 35,849 ED
visits and 35,105 OPD visitsvisits and 35,105 OPD visits

8. 8
Scope of the NHAMCS

Basic unit of sampling is patient visitBasic unit of sampling is patient visit

Emergency and outpatient departments Emergency and outpatient departments
of noninstitutional general and short-of noninstitutional general and short-
stay hospitalsstay hospitals
Not Federal, military, or Veterans Not Federal, military, or Veterans
Administration facilitiesAdministration facilities
Located in 50 states and D.C.Located in 50 states and D.C.

9. 9
Sample Weight

Each NAMCS record contains a Each NAMCS record contains a
single weight, which we call Patient single weight, which we call Patient
Visit WeightVisit Weight

Same is true for OPD records and ED Same is true for OPD records and ED
recordsrecords

This weight is used for both visits and This weight is used for both visits and
drug/procedure mentionsdrug/procedure mentions

10. 10
Data Items
Patient characteristics Patient characteristics

Age, sex, race, ethnicityAge, sex, race, ethnicity

Visit characteristicsVisit characteristics

Source of payment, continuity of care, reason Source of payment, continuity of care, reason
for visit, diagnosis, treatmentfor visit, diagnosis, treatment

Provider characteristicsProvider characteristics

Physician specialty, hospital ownership…Physician specialty, hospital ownership…
MULTUM drug characteristics added in 2006MULTUM drug characteristics added in 2006

11. 11
Coding Systems Used

Reason for Visit Classification (NCHS)Reason for Visit Classification (NCHS)

ICD-9-CM for diagnoses, causes of ICD-9-CM for diagnoses, causes of
injury and proceduresinjury and procedures

Drug Classification System-MULTUMDrug Classification System-MULTUM

12. 12
File Structure
Download data and layout from Download data and layout from
websitewebsite
http://www.cdc.gov/nchs/about/major/ahcd/http://www.cdc.gov/nchs/about/major/ahcd/
ahcd1.htmahcd1.htm
Flat ASCII files for each setting and Flat ASCII files for each setting and
year:year:
NAMCS: 1973-2006NAMCS: 1973-2006
NHAMCS: 1992-2006NHAMCS: 1992-2006
STATA files on Web:STATA files on Web:
NAMCS: 2003-2005NAMCS: 2003-2005
NHAMCS: 2003-2005NHAMCS: 2003-2005

13. 13
Creating a usable STATA dataset
Two options:Two options:
1)Use the self-extracting file in STATA
folder to open a complete dataset
for the 2003-2005 NAMCS,
NHAMCS-ED, & NHAMCS-OPD
2)Use the DO file (*.do) and the
dictionary file (*.dct) along with the
flat data file (*.exe) to create a
dataset
3)3)StatTransferStatTransfer

14. 14
Organizational structure-NAMCS data
Provider
provider info
practice info
geographic info
Visit
patient & visit info
treatment & outcome info
medications
Medications 1-8 Primary reason for Visit
MULTUM Categories
Primary diagnosis
Write-in scope procedure 1
Other Reason for Visit
Other Reason for Visit
Other diagnosis
Other diagnosis
Write-in scope procedure 2
Other test/service 1
Other test/service 2
Surgical procedure 1
Surgical procedure 2
Non-surgical procedure 1
Non-surgical procedure 2

15. 15
SETS-Statistical Export and
Tabulation System

16. 16
Hands-on Exercises
STATA Users
Double-click: My Computer\Double-click: My Computer\
Local Disk C:\DUC_08Local Disk C:\DUC_08
Open STATAOpen STATA
In the command window In the command window
type:type:

Set Set mem 1000mmem 1000m

Set Set matsize 5000matsize 5000
Under the “File” icon-double-Under the “File” icon-double-
click click namcs05.dtanamcs05.dta
Under “New Do File Editor”-Under “New Do File Editor”-
double-click: STATA double-click: STATA
exercises.doexercises.do
SAS/SUDAAN Users
Double-click: My Computer\
Local Disk C:\DUC_08
Double-click: Final Exercises

17. 17
Visit rate estimates
PhycodePhycode SexSex PatwtPatwt (Patwt/(Patwt/
Pop)*100Pop)*100
SexwtSexwt
14011401 11 100100 (100/800)*100(100/800)*100 12.512.5
18201820 11 300300 (300/800)*100(300/800)*100 37.537.5
10011001 11 5050 (50/800)*100(50/800)*100 6.256.25
500500 11 120120 (120/800)*100(120/800)*100 1515
71.25 visits per 71.25 visits per
100 persons100 persons
Female population=800 New variableCalculation*
*Note: Rate=est/pop=Σ patwt/pop=1/pop*Σ patwt.
Sample
size=4
Visits=570

18. 18
RecordRecordProc1Proc1Proc2Proc2Proc3Proc3Proc4Proc4Proc5Proc5Proc6Proc6Proc7Proc7Proc8Proc8TotprocTotproc
11 191119110000000000000000 00000000 00000000000000000000000000000000 11
22 218221822186218600000000 00000000 00000000000000000000000000000000 22
33 549054900000000000000000 00000000 00000000000000000000000000000000 11
44 000000000000000000000000 00000000 00000000000000000000000000000000 00
55 819281920000000000000000 00000000 00000000000000000000000082008200 22
Note: 0000=No procedure recorded.
Calculating Total Number of Write-in Procedures

19. 19
Data Considerations

20. 20
NAMCS vs. NHAMCS

Consider what types of settings are best Consider what types of settings are best
for a particular analysisfor a particular analysis

Persons of color are more likely to Persons of color are more likely to
visit OPD’s and ED’s than physician visit OPD’s and ED’s than physician
officesoffices

Persons in some age groups make Persons in some age groups make
disproportionately larger shares of disproportionately larger shares of
visits to ED’s than offices and OPD’svisits to ED’s than offices and OPD’s

21. 21
Procedures
ProgramProgram Categorical Categorical
VariablesVariables
Continuous Continuous
VariablesVariables
SASSAS PROC PROC
SURVEYFREQSURVEYFREQ
PROC PROC
SURVEYMEANSSURVEYMEANS
STATASTATA SVY: TABSVY: TAB SVY: MEANSVY: MEAN
SUDAANSUDAAN PROC CROSSTABPROC CROSSTAB PROC PROC
DESCRIPTDESCRIPT

22. 22
How Good are the Estimates?

Depends … In general, OPD estimates Depends … In general, OPD estimates
tend to be somewhat less reliable than tend to be somewhat less reliable than
NAMCS and ED. NAMCS and ED.

Since 1999, our Advance Data reports Since 1999, our Advance Data reports
include standard errors in every table so include standard errors in every table so
it is easy to compute confidence it is easy to compute confidence
intervals around the estimates.intervals around the estimates.

23. 23
RSE improves incrementally with
the number of years combined

RSE = SE/RSE = SE/xx

RSE for percent of visits by persons RSE for percent of visits by persons
less than 21 years of age with less than 21 years of age with
diabetesdiabetes

1999 RSE = .08/.18 = .44 (44%)1999 RSE = .08/.18 = .44 (44%)

1998 & 1999 RSE = .06/.18 = .33 (33%)1998 & 1999 RSE = .06/.18 = .33 (33%)

1998, 1999, & 2000 RSE = .05/.21 = .24 1998, 1999, & 2000 RSE = .05/.21 = .24
(24%)(24%)

24. 24
Some User Considerations

NAMCS/NHAMCS sample visits, not NAMCS/NHAMCS sample visits, not
patientspatients

No estimates of incidence or No estimates of incidence or
prevalenceprevalence

No state-level estimatesNo state-level estimates

May capture different types of care for May capture different types of care for
solo vs. group practice physicianssolo vs. group practice physicians

Data is only as good as what is Data is only as good as what is
documented in the medical recorddocumented in the medical record

25. 25
If nothing else, remember…The
Public Use Data File
Documentation is YOUR FRIEND!

Each booklet includes:Each booklet includes:

A description of the surveyA description of the survey

Record formatRecord format

Marginal data (summaries)Marginal data (summaries)

Various definitionsVarious definitions

Reason for Visit classification codesReason for Visit classification codes

Medication & generic namesMedication & generic names

Therapeutic classesTherapeutic classes