Ambulatory Surgery Presentation

1How to understand and use National Ambulatory Medical Care Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) data for clinical researchYuwei Zhu10-29-2004Dept of Biostatistics2OverviewI. Survey Background II. Survey Methodology III. Technical ConsiderationsIV. Getting the Data –Using Raw Data FilesV. ExampleVI. Data Analysis –SAS, STATA, SUDAANVII. Other Public Domain Data3Performed by:Centers for Disease Control and Prevention (CDC)National Center for Health Statistics, Division of Health Care Statistics, and National Health Care SurveyNAMCS and NHAMCS4National Ambulatory Medical Care Survey (NAMCS) HistorySurvey began in 1973 Annual data collection through 1981Conducted in 1985 Annual began again in 1989 5NAMCSClassified by the American Medical Association and the American Osteopathic Association as delivering “office-based, patient care” Healthcare providers within private, non–hospital-based clinics and health maintenance organizations (HMOs) are within the scope of the survey6NAMCSPatient visits made to the offices of non–federally employed physicians –Excluding:AnesthesiologyRadiologyPathology7In-Scope NAMCS locationsFreestanding clinicFederally qualified health centerNeighborhood and mental health centersNon-federal government clinicFamily planning clinicHMOFaculty practice planPrivate solo or group practice8Out-of-Scope NAMCS locationsHospital EDs and OPDsAmbulatory surgicenterInstitutional setting (schools, prisons)Industrial outpatient facilityFederal Government operated clinicLaser vision surgery9NAMCSNAMCS uses a multistage probability sample design to obtain –Primary sampling units (PSUs)–Physician practices within the PSUs–Patient visits within physician practices 10Sample design -NAMCS112 PSUs (counties)–Counties–Groups of counties–County equivalents (such as parishes or independent cities)–Towns–TownshipsNonfederally employed, office-based physicians stratified by specialty, 3,000 physiciansAbout 30 visits per doctor over a randomly selected 1-week period, 25,000 visits11National Hospital Ambulatory Medical Care Survey (NHAMCS) HistorySurvey began in 1992 Annual data collection12National sample of visits to the EDs and outpatient departments of noninstitutional general and short-stay hospitals in the United StatesExcluded hospitals:–Federal–Military–Veterans AdministrationNHAMCS13NHAMCSThis survey uses a 4-stage probability design with samples–geographically defined areas–hospitals within these areas–clinics within the hospital–patient visits within clinics.The first stage is similar to NAMCS14Sample design -NHAMCS112 PSUs (counties)Panel of 600 non-Federal, general or short stay hospitalsClinics (OPDs) and emergency service areas (EDs), 400 EDs and 250 OPDsAbout 200 visits per OPD, 100 per ED over random 4-week period,37,000 ED and 35,000 OPD visits15NHAMCS ScopeOPD was intended to be parallel to the NAMCS in the hospital settingGeneral medicine, surgery, pediatrics, ob/gyn, substance abuse, and “other” clinics are in-scopeAncillary services are out of scope16Data ItemsPatient characteristics –Age, sex, race, ethnicityVisit characteristics–Source of payment, continuity of care, reason for visit, diagnosis, treatmentProvider characteristics–Physician specialty, hospital ownership…Drug characteristics added in 1980–Class, composition, control status, etc.17Repeating fields (from text entries)Up to 3 fields each…–Reason for visit –Physician’s diagnosis–Cause of injuryDiagnostic services (6 fields)Surgical procedures (2 fields)Medications (6 fields)–Drug ingredients (5 fields)–Therapeutic class (3 fields –2002 on)18Coding Systems UsedReason for Visit Classification (NCHS)ICD-9-CM for diagnoses, causes of injury and proceduresDrug Classification System (NCHS)National Drug Code Directory19Drug Data in NAMCS/NHAMCSWhat is a “Drug Mention” ?Any of up to 6 medications that were ordered, supplied, administered, or continued during the visit. Respondents are asked to report trade names or generic names only (not dosage, administration, or regimen). 20Drug CharacteristicsGeneric Name (for single ingredient drugs)Prescription StatusComposition StatusControlled Substance StatusUp to 3 NDC Therapeutic Classes (4-digit)Up to 5 Ingredients (for multiple ingredient drugs)21Some User ConsiderationsNAMCS/NHAMCS sample visits, not patientsNo estimates of incidence or prevalenceNo state-level estimatesNot sampled by setting or by non-physician providersMay capture different types of care for solo vs. group practice physicians22Data usesUnderstand health care practiceExamine the quality of careTrack certain conditionsFind health disparitiesMeasure Healthy People 2010 objectivesServe as benchmark for states23Data usersOver 100 journal publications in last 2 yearsMedical associationsGovernment agenciesHealth services researchersUniversity and medical schoolsBroadcast and print media24Sample WeightEach NAMCS record contains a single weight, which we call Patient Visit WeightSame is true for OPD records and ED recordsThis weight is used for both visits and drug mentions25Reliability of EstimatesEstimates should be based on at least 30 sample records ANDEstimates with a relative standard error (standard error divided by the estimate) greater than 30 percent are considered unreliable by NCHS standardsBoth conditions should be met to obtain reliable estimates26How Good are the Estimates?Depends on what you are looking at. In general, OPD estimates tend to be somewhat less reliable than NAMCS and ED. Since 1999, Advance Data reports include standard errors in every table so it is easy to compute confidence intervals around the estimates.27Sampling ErrorNAMCS and NHAMCS are not simple random samplesClustering effects of visits within the physician’s practice, physician practices within PSUs, clinics within hospitalsMust use some method to calculate standard errors for frequencies, percents, and rates28Ways to Improve Reliability of EstimatesCombine NAMCS, ED and OPD data to produce ambulatory care visit estimatesCombine multiple years of dataAggregate categories of interest into broader groups.29NAMCS vs. NHAMCSConsider what types of settings are best for a particular analysis–Persons of color are more likely to visit OPD's and ED's than physician offices–Persons in some age groups make disproportionately larger shares of visits to ED's than offices and OPD's30File StructureDownload data and layout from websitehttp://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htmFlat ASCII files for each setting and yearNAMCS: 1973-2002NHAMCS: 1992-200231Trend considerationsVariables routinely rotate on and off surveyBe careful about trending diagnosis prior to 1979 because of ICDA (based on ICD-8)Even after 1980-be careful about changes in ICD-9-CMNumber of medications varies over years1980-81 –8 medications1985, 1989-94 –5 medications1995-2002 –6 medications2003+ –8 medicationsDiagnostic & therapeutic checkboxes varyUse spreadsheet for significance of trends 32Example Hypothesis --Educational Efforts Targeted at Judicious Antibiotic Use Will Reduce Prescription Rates in all Treatment Settings33Study DesignRetrospective collection of data from–NAMCS –NHAMCS1994-2000 study yearsAntibiotic prescribing patterns and diagnosesChildren <5 years of ageClinic type --PediatricPhysician type –Pediatrician or Family Medicine34Data StratificationRace –White, Black and otherTime period –94 & 95, 96 & 97, 98 & 00Antibiotics –Penicillin's, Cephalosporins, Erythromycin/lincosamide/macrolides,Tetracyclines, Chloramphenicol derivatives, Aminoglycosides, Sulfonamides and trimethoprim, Miscellaneous antibacterial agents, and Quinolone/derivativesDiagnoses --Otitis media, Sinusitis, Pharyngitis,Bronchitis,Upper respiratory tract infection (URI)35Overall Antibiotic Rates in Children <5 Based on Source of Care05001000150020001994199519961997199819992000YearsRates per 1000 childrenHospital-based ED Office-based 36Total CareYearsWhite BlackRate Ratio95% CI Visit rates per 1000 children aged <5 years1994-1995415031021.341.22, 1.47*1996-1997452943201.051.02, 1.08*1998-2000420443020.980.70, 1.34370%20%40%60%80%100%1994-19951996-19981999-20001994-19951996-19981999-2000Years% Distribution health care visit siteHospital-based ED Office-based White childrenBlack Children38Total CareYearsWhite BlackRate Ratio95% CI Antibiotic prescription rates per 1000 children aged <5 years1994-199514949981.501.48, 1.51*1996-1997142113201.080.96, 1.221998-2000111810741.040.86, 1.2439Total CareYearsWhite BlackRate Ratio95% CI Otitis media rates per 1000 children aged <5 years1994-19958165201.571.46, 1.69*1996-19977797391.061.04, 1.07*1998-20006306031.050.69, 1.5840ResultsDecline in antibiotic prescribing in children <5 years; most notable in office-based and emergency department settings Penicillin's were common antibiotics usedMost common diagnosis in all three settings was otitis mediaNatasha B. Halasa, Marie R. Griffin, Yuwei Zhu, and Kathryn M. Edwards. Difference in antibiotic prescribing patterns for children aged less than five years in the three major outpatient settings, Journal of Pediatrics. 2004; 144:200-20541Code to create design variables: survey years 2001 & earlier CPSUM=PSUM;CSTRATM = STRATM;IF CPSUM IN(1, 2, 3, 4) THEN DO;CPSUM = PROVIDER +100000;CSTRATM = (STRATM*100000) +(1000*(MOD(YEAR,100))) + (SUBFILE*100) + PROSTRAT;END;ELSE CSTRATM = (STRATM*100000);42proccrosstabdata=test1 design=WOR filetype=sas;Nest stratm psum subfile prostrat year provider dept su clinic/missunit;Totcnt poppsum _zero_ _zero_ _zero_ popprovm _zero_ popsum _zero_ popvism;Weight patwt;Tables sex*ager;run;SUDAAN version 8.0.2 example43proccrosstabdata=test1 filetype=sas;Nest stratm psum ;Weight patwt;Tables sex*ager;run;SUDAAN version 8.0.2 example44Use http://***/test1svyset [pweight=patwt], strata(cstratm) psu(cpsum)svytab sex agersvymean ageSTATA version 8. example45procsurveyfreqdata=test1;tables sex*ager;strata cstratm;cluster cpsum;weight patwt;run;SAS version 9.1 example46Some considerations: SUDAAN vs. SAS Proc SurveymeansSUDAANPROC Surveymeans•design variables=cstratm, cpsum (1-stage design)•design variables=cstratm, cpsum (1-stage design)•nest=cstratm, cpsum•strata cstratm •cluster cpsum•Sort by design variables•Sort not needed•Weight data: Patwt•Weight data: Patwt•Subgroup=identify categorical variables•Class=identify categorical variables•Tables=analysis variables•Var=analysis variables47If nothing else, remember…The Public Use Data File Documentation is YOUR FRIEND!Each booklet includes:–A description of the survey–Record format–Marginal data (summaries)–Various definitions–Reason for Visit classification codes–Medication & generic names–Therapeutic classes48Other Public Domain Data CDC WONDER --http://wonder.cdc.gov/National Center for Health Statistics --http://www.cdc.gov/nchs/National Health and Nutrition Examination Survey (NHANES) --http://www.cdc.gov/nchs/nhanes.htmNational Health Interview Survey (NHIS) --http://www.cdc.gov/nchs/nhis.htmNational Survey of Family Growth (NSFG) --http://www.cdc.gov/nchs/nsfg.htmCensus --http://www.census.gov/49Other Public Domain Data (cont.)Dept. of Health, TN http://hitspot.state.tn.us/hitspot/hit/main/SPOT/frames/SPOT/index.htm 50Thanks Natasha HalashaSusan Schappert -National Center for Health StatisticsLinda McCaig & David Woodwell -National Center for Health Statistics51Questions?