Medication Therapy in Ambulatory Medical Care National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, 1992 By Cheryl R. Nelson, M.S.P.H., and Deanne E. Knapp, Ph.D., Division of Health Care Statistics Introduction This report describes overall medication therapy in all ambulatory medical care settings, including office-based physicians and hospital outpatient and emergency departments. This report also examines differences among the three ambulatory health care settings as well as differences over time in office-based settings. The term medication therapy (previous reports used the term drug utilization) is defined as the prescribing or providing of a new or continued drug by a doctor of medicine or osteopathy or other health care practitioner in the course of a visit to a physician’s office, hospital outpatient department (OPD), or hospital emergency department (ED). It is not an indication of the patient’s compliance with the provider’s instructions. Medication therapy in this report will be described in terms of the frequency with which drugs are prescribed (drug mentions), the proportion of visits at which any medication was prescribed or provided (drug visits), and the average number of drugs mentioned per 100 patient visits (drug mention rate). The terms drug and medication are used interchangeably and are broadly defined to include any pharmaceutical agent the Abstract Objectives—This report describes medications provided or prescribed during ambulatory medical care visits in 1992. Total ambulatory care medication therapy combines data from office-based physicians, hospital outpatient departments (OPD’s), and hospital emergency departments (ED’s). Drug therapy is described along three dimensions: number of drugs provided or prescribed (drug mention), whether a visit had any drugs mentioned (drug visit), and average number of drugs mentioned per 100 visits (drug mention rate). Utilization in ambulatory care settings is compared in terms of patient, drug, provider, and visit characteristics. Methods—Annual use of medication therapy was determined using data collected in the 1992 National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS). NAMCS includes office visits to nonfederal physicians principally engaged in office practice. The target universe of NHAMCS includes visits to ED’s and OPD’s of non-Federal, short-stay, or general hospitals. Sample data were weighted to produce annual estimates. Drug mentions are defined as the number of drugs mentioned on the patient record form. Results—An estimated 1.1 billion medications were provided or prescribed at ambulatory care visits in 1992. The setting with the greatest percent of visits with medication therapy was the ED; OPD’s had the lowest percent with medications. Patients at the ED were provided more pain relief type drugs. The rate of drug mentions and percent of visits with medications were significantly higher in OPD clinics of general medicine and pediatrics compared with other types of OPD clinics. In office- based settings, physicians specializing in cardiovascular diseases were most likely to prescribe medications. Also, cardiovascular-renal type drugs accounted for the largest percent of office-based drug mentions. Visits with illness diagnoses are most likely to receive medication therapy. Trend data comparing 1980 to 1992 office-based mentions showed significant changes on several characteristics: single-ingredient drug status, physician specialty, and patient age. Conclusions—The profile of patients using office- and hospital-based ambulatory care settings are quite different as is the case-mix of conditions. These differences play an important role in medications utilized. The aging of the U.S. population from 1980 to 1992 appeared to have significant effects on several drug mention characteristics. Keywords: medication therapy c NAMCS c NHAMCS c drug mention Number 290 + August 8, 1997 From Vital and Health Statistics of the 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 CENTERS FOR DISEASE CONTROL AND PREVENTION
24
Embed
Medication Therapy in Ambulatory Medical Care - cdc.gov · PDF fileMedication Therapy in Ambulatory Medical Care ... outpatient and emergency departments. This report also examines
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Number 290 + August 8, 1997
From Vital and Health Statistics of the CENTERS FORDISEASECONTROLANDPREVENTION/National Center for Health Statistics
Medication Therapy in Ambulatory Medical CareNational Ambulatory Medical Care Survey and
National Hospital Ambulatory Medical Care Survey, 1992By Cheryl R. Nelson, M.S.P.H., and Deanne E. Knapp, Ph.D., Division of Health Care Statistics
AbstractObjectives—This report describes medications provided or prescribed during
ambulatory medical care visits in 1992. Total ambulatory care medication theracombines data from office-based physicians, hospital outpatient departments(OPD’s), and hospital emergency departments (ED’s). Drug therapy is describealong three dimensions: number of drugs provided or prescribed (drug mentionwhether a visit had any drugs mentioned (drug visit), and average number of dmentioned per 100 visits (drug mention rate). Utilization in ambulatory care setis compared in terms of patient, drug, provider, and visit characteristics.
Methods—Annual use of medication therapy was determined using data collectin the 1992 National Ambulatory Medical Care Survey (NAMCS) and the NationalHospital Ambulatory Medical Care Survey (NHAMCS). NAMCS includes office visitto nonfederal physicians principally engaged in office practice. The target universeNHAMCS includes visits to ED’s and OPD’s of non-Federal, short-stay, or generalhospitals. Sample data were weighted to produce annual estimates. Drug mentiondefined as the number of drugs mentioned on the patient record form.
Results—An estimated 1.1 billion medications were provided or prescribed atambulatory care visits in 1992. The setting with the greatest percent of visits withmedication therapy was the ED; OPD’s had the lowest percent with medications.Patients at the ED were provided more pain relief type drugs. The rate of drug meand percent of visits with medications were significantly higher in OPD clinics ofgeneral medicine and pediatrics compared with other types of OPD clinics. In officebased settings, physicians specializing in cardiovascular diseases were most likelyprescribe medications. Also, cardiovascular-renal type drugs accounted for the largpercent of office-based drug mentions. Visits with illness diagnoses are most likelyreceive medication therapy. Trend data comparing 1980 to 1992 office-based mentshowed significant changes on several characteristics: single-ingredient drug statusphysician specialty, and patient age.
Conclusions—The profile of patients using office- and hospital-based ambulatorcare settings are quite different as is the case-mix of conditions. These differencesan important role in medications utilized. The aging of the U.S. population from 1981992 appeared to have significant effects on several drug mention characteristics.
Keywords: medication therapyc NAMCS c NHAMCS c drug mention
U.S. DEPARTMENT OF HEALTH AND HUMAN SCenters for Disease Control and Prevention
National Center for Health Statistics
Introduction
This report describes overallmedication therapy in all ambulatorymedical care settings, includingoffice-based physicians and hospitaloutpatient and emergency departments.This report also examines differencesamong the three ambulatory health caresettings as well as differences over timein office-based settings. The termmedication therapy(previous reportsused the termdrug utilization) is definedas the prescribing or providing of a newor continued drug by a doctor ofmedicine or osteopathy or other healthcare practitioner in the course of a visitto a physician’s office, hospitaloutpatient department (OPD), or hospitalemergency department (ED). It is not anindication of the patient’s compliancewith the provider’s instructions.Medication therapy in this report will bedescribed in terms of the frequency withwhich drugs are prescribed (drugmentions), the proportion of visits atwhich any medication was prescribed orprovided (drug visits), and the averagenumber of drugs mentioned per 100patient visits (drug mention rate). Theterms drug and medication are usedinterchangeably and are broadly definedto include any pharmaceutical agent the
py
d),rugstings
ed
sof
s are
ntions
-toesttoions,
yplay0 to
ERVICES
CENTERS FOR DISEASE CONTROLAND PREVENTION
2 Advance Data No. 290 + August 8, 1997
health-care practitioner prescribes orprovides to the patient during a visit.
Previous reports have shown thatmedication therapy is the predominateform of therapy in ambulatory caresettings. Medications play a vital part inmaintaining and restoring health andpreventing disease, but what is theextent of use of medication therapy inambulatory care settings? In addition,are there differences in medicationtherapy or characteristics amongambulatory care settings?
Methods
The data presented in this report afrom the 1992 National AmbulatoryMedical Care Survey (NAMCS) and the1992 National Hospital AmbulatoryMedical Care Survey (NHAMCS).NAMCS and NHAMCS, year-longsample surveys of the Nation’s healthcare providers, are part of the NationalHealth Care Survey, which is conducteby the Centers for Disease Control andPrevention, National Center for HealthStatistics, Division of Health CareStatistics. This report does not includevisits for ambulatory surgery athospital-based or freestanding surgeryunits, which are covered by a thirdsurvey, the National Survey ofAmbulatory Surgery (NSAS). NAMCSsamples patient visits to the Nation’snon-Federal office-based physicians, anNHAMCS samples visits to emergencyand outpatient departments of non-Federal, short-stay, or general hospitalsA descriptions of the surveys includingstatistical design, sampling errors,nonsampling errors, adjustments fornonresponse, test of significance, anddefinition of terms can be found in theTechnical Notessection. A summaryof general findings from the 1992ambulatory care settings (1–3) andreports on medication therapy in officepractice since 1980 have beenpublished (4–7).
Medication data are based onentries in item 17 on the NAMCS andED NHAMCS Patient Record Forms(figures 1and3, respectively) and item16 on the OPD NHAMCS PatientRecord Form (figure 2). These items askthe responding health care practitionerto report the names of up to five specifi
re
d
d
.
c
drugs prescribed or provided in thecourse of the office, ED, or OPD visit(drugs prescribed through telephonecontact are excluded). Health-carepractitioners were asked to report bothnonprescription and prescription drugsand both new and continuedmedications.
The first survey year that healthcare providers reported the number annames of specific drugs prescribed topatients was 1980. The methodologyused to collect, classify, and processdrug information is reported elsewhere(8). Drug characteristics for medicationmentioned in the NAMCS andNHAMCS include new or continuedstatus (except for ED’s), therapeuticclass, generic or brand name, Federalcontrol schedule, and compositionstatus. These characteristics wereobtained for each medication mentionein the NAMCS and NHAMCS usingseveral sources including theAmericanDrug Index(9), Drug Topics Red Book(10), theNational Drug Code Directory(11), andFacts and Comparisons(12).
Results
In 1992, 64 percent of the estimate908.4 million visits made to thecombined ambulatory medical caresettings of physician offices, ED’s, andOPD’s in the United States wereclassified as drug visits, that is, visitsduring which one drug or more wasprescribed or provided to the patient.Visits to physician offices represented83.9 percent (762.0 million) of the totalambulatory care visits; visits to ED’saccounted for 9.9 percent (89.8 millionvisits); and 56.6 million visits to OPD’srepresented the remaining 6.2 percent.Figure 4shows that the distribution ofdrug mentions by type of setting wassimilar to the distribution of ambulatoryvisits by setting. Of the 1.1 billionmedications mentioned in ambulatorysettings the largest proportion of drugmentions was in physician offices(83.6 percent), followed by ED’s(10.6 percent) and OPD’s (5.7 percent)Because such a large percent of thedrugs mentioned are from physicianoffices, analyses on the combinedsettings will be heavily influenced bythe distribution within physician offices.
d
s
d
d
.
The tables show the data for thecombined settings. However, theanalyses of patient, drug, provider, andvisit characteristics focus on within andacross setting comparisons.
The percent of visits withmedication therapy varied according toambulatory setting (χ2=58.2,p<.001)(table 1). ED’s had the highest percentof drug visits (69.1) followed byphysician’s offices (63.8) and OPD’s(53.3). The rate of drug mentions pervisit was also associated with setting,with ED’s having a significantly higheraverage number of drug mentions pervisit than OPD’s.
Patient characteristics
Table 1displays the number andpercent distribution of drug mentions,percent of visits with drug mentions(drug visits), and the average number ofdrug mentions per 100 visits acrossambulatory care settings by patient age,sex, and race. Generally within eachambulatory care setting the percent ofvisits with medication therapy increasedwith patient’s age (χ2=20.2,p<.01).Similarly, figure 5shows that the drugmention rate increased from an averageof 93 drug mentions per 100 visits inthe 15–24 years old group to 165 drugmentions per 100 visits in the oldest agepatient group.
Although females accounted for agreater proportion of drug mentions, asthey do for ambulatory visits, comparedto males, there were no overall genderdifferences in the percent of visits withmedication therapy or drug mentionrates. Percent of visits with medicationtherapy varied by ambulatory medicalcare setting and patient’s race (χ2=11.5,p<.01). The OPD setting had the lowestpercent drug visits in all race groups.White patients had a higher percent ofdrug visits in ED’s as opposed tophysician offices; whereas the oppositewas true for black patients and patientsof ‘‘other races’’ (figure 6). In office-based settings the drug mention rate forblack patients was 18.2 percent higherthan for white patients. There were nosignificant differences in drug mentionrates by race in the ED or OPD settings.
.
-
-
Figure 1. National Ambulatory Medical Care Survey patient record
Advance Data No. 290 + August 8, 1997 3
Drug characteristics
Table 2describes some keycharacteristics of the medicationsprovided or prescribed for each of thethree ambulatory settings.
+ New or continued status—Almosthalf of the drug mentions weredescribed as continued medicationsA significantly higher percent of drugmentions at office visits were for newmedications compared with OPDvisits.
+ Composition status—About three-quarters of the drugs were for singleingredient medications. Hospital
settings had a significantly higherpercent of single-ingredient drugmentions than physician offices.
+ Control status—Noncontrolled drugsrepresented 85.7 percent of themedications used in drug therapy.About 7 percent of the medicationsprescribed or provided by health-carpractitioners were classified ascontrolled substances. Schedule IIdrugs are those with a high potentiafor abuse. Schedules III and IV arethose with some or low potential forabuse. Of the controlled drugs, thosthat were Schedule IV had asignificantly greater number of
e
l
e
mentions. Schedules II, III, and IVdrugs require prescriptions fordispensing. ED’s had a significantlyhigher percent of Schedule II and IIIdrug mentions and significantly lowerpercent of noncontrolled drugmentions than the other two settings.
In table 3, the estimated 1.1 billiondrug mentions are classified by theirprimary therapeutic effects.Antimicrobial agents and cardiovascularrenal drugs accounted for 30.3 percentof all drug mentions. ED’s had asignificantly higher percent thanphysician offices and OPD’s of drug
Figure 2. National Hospital Ambulatory Medical Care Survey outpatient department patient record
4 Advance Data No. 290 + August 8, 1997
Figure 3. National Hospital Ambulatory Medical Care Survey emergency department patient record
Advance Data No. 290 + August 8, 1997 5
n
Figure 4. Percent of visits and drug mentions by ambulatory medical care setting:United States, 1992
Figure 5. Average number of drug mentions for ambulatory medical care settings bypatient age: United States, 1992
6 Advance Data No. 290 + August 8, 1997
mentions for pain relief but a significantlylower percent for cardiovascular-renaldrugs and hormones. Within hospitalsettings, ED’s had a significantly higherpercent of antimicrobial agent drugmentions than OPD’s.Figure 7presentsthe comparison of drug mention rates pe100 visits for selected therapeutic classeacross ambulatory care settings illustratithe increased rate of drugs used for therelief of pain in the ED.
The data intables 4and5 showranked listings of the drugs mostfrequently prescribed or provided byambulatory care setting. It should benoted that estimates that differ in rankeorder may not be significantly different
rsg
d
from each other. Intable 4, the data arepresented by generic ingredients andprovide a more basic perspective ofmedication therapy in the ambulatorysetting. The two most commonlyutilized therapeutic categories intable 3were antimicrobials and cardiovascularrenal agents; which is also reflected intable 4. The most frequently usedgeneric substance was amoxicillin(4.6 percent), an antimicrobial. Eightother antimicrobials are in the top 50list, including erythromycin andcefaclor. Other drugs frequentlyprescribed or provided by ambulatorycare practitioners are decongestants(e.g., pseudoephedrine, phenylephrine
-
,
and phenylpropanolamine);bronchodilators (e.g., albuterol andtheophylline), and drugs used in treatingdiseases of the cardiovascular system(e.g., hydrochlorothiazide, furosemide,digoxin, enalapril, and diltiazem).
The three settings had 5 of the top 10generics in common (amoxicillin,acetaminophen, erythromycin, albuterol,and ibuprofen). Among the top 10generics, consistent withtable 3, thetherapeutic class table, ED’s had asignificantly higher percentage of painrelief drugs than the other settings. OPD’shad iron preparations specifically listedamong their top 10 (hematologics class),whereas the other two settings did not.
Table 5shows drugs by entry name,that is, the trade or generic nameentered on the patient’s prescription ormedical record. Two of the top entrynames are antimicrobials, amoxicillinand Amoxil (a brand of amoxicillin). Asexpected, emergency departments had asignificantly higher percent of pain reliefdrugs than the other two settings; half oftheir top 10 drugs were pain reliefagents. Outpatient departments hadferrous sulfate (an iron preparation)specifically listed among their top 10,whereas the other two settings did not.
Provider and visit characteristics
Table 6describes the relationshipbetween medication therapy andcharacteristics of office-based physiciansand OPD clinic types. Ninety-fourpercent of the office visits were tophysicians who identified themselves asmedical doctors (1), and 94.0 percent ofthe drug mentions were reported atvisits to doctors of medicine. Fifty-sixpercent of all office visits were tophysicians specializing in general andfamily practice, internal medicine,pediatrics, or cardiovascular diseases(1); together they accounted for69.5 percent of all drug mentions. Thesefour physician specialties, led bycardiovascular diseases, were also themost likely to prescribe or providemedications to their patients. Drugmentions, drug visits, and drug mentionrates were significantly higher in OPDclinics of general medicine comparedwith other types of clinics.
y
rery
y.
,
re
ed
he
.
ds.
Figure 6. Percent of visits with medication therapy by ambulatory setting and race:United States, 1992
Percentofvisits
Figure 7. Annual rate of drug mentions per 100 visits by ambulatory care setting andselected therapeutic classes: United States, 1992
Advance Data No. 290 + August 8, 1997 7
Table 7compares selected visitcharacteristics across settings withpercent of visits with medication therapand the average rate of drug mentionsper 100 visits. In all settings there weresignificantly higher percentages of drugvisits and drug mention rates for carerelated to illness rather than injury orsupplementary care. Supplementary cais defined as care not classified to injuor illness (e.g., general medicalexamination, routine prenatalexamination). The relative distributionof these selected visit characteristicsvary according to both ambulatory caresetting and extent of medication therapJust as illness visits are more likely tohave medication therapy in all settingsvisits where the patient was seenpreviously were more likely to havemedications prescribed.
Trend data for physicianoffice visits
NAMCS began in 1973 but did notcollect drug data until 1980 (6,7). TheNHAMCS began in 1992, socomparable data are not available prioto that time. The following describes thtrends in office-based physicianmedication therapy between 1980 and1992.
The percent of visits with drugmentions and the average number ofdrug mentions have essentially remainstatic. The percent visits with drugmentions was 63.1 percent in 1980 and63.8 percent in 1992. The averagenumber of drug mentions per visit was1.2 in both 1980 and 1992.
There was no significant changebetween 1980 and 1992 in percent ofdrug mentions for noncontrolled drugstatus. However, the percent of drugmentions for single-ingredient drugssignificantly increased from 1980 to1992.
As shown intable 8, 4 of the top10 ranked generic substances were tsame between 1980 and 1992. Thosefour generic substances werehydrochlorothiazide (cardiovascular-renal drug), aspirin (drug used for parelief), erythromycin (antimicrobialagent), and acetaminophen (drug usefor pain relief). However, 2 of the top10 generic substances in 1992
in
d
increased since 1980. Amoxicillinand acetaminophen utilizationincreased from 1.9 and 2.1 drugmentions per 100 visits in 1980 to 5.and 4.0 drug mentions per 100 visitsin 1992, respectively. Theantimicrobial agent, penicillin,
7
decreased from its 1980 rate of 3.0drug mentions per 100 visits to 1.1drug mentions per 100 visits in 1992It is noteworthy that, with theexception of the metabolic-nutrientclass in 1992, all of the top 10 rankeclasses were the same for both year
Figure 8. Rate of drug mentions per 100 visits by selected physician specialties:United States, 1980 and 1992
8 Advance Data No. 290 + August 8, 1997
Two significant changes by type ofoffice-based physician specialty occurrebetween 1980 and 1992. In 1980 (7),general and family practice physiciansentered 41.1 percent of all of the drugsmentioned on the Patient Record formswhereas in 1992, they entered34.1 percent—a decrease of17.0 percent. On the other hand,cardiovascular disease physiciansentered 1.5 percent of all drug mentionsin 1980 (7), but 4.4 percent of all drugmentions in 1992—an increase of193.3 percent. Percents of drug mentiondid not change significantly between1980 and 1992 for the other specialtiesIn 1992 the highest drug mention ratesin physician offices were in twotherapeutic categories, cardiovascular-renal drugs and antimicrobial agents. Itshould be noted that the percent ofgeneral and family practice doctors ofmedicine who were engaged primarilyin non-Federal, office-based patient careincreased by 22.7 percent between 198and 1992 (14). The respective changefor cardiovascular disease physicianswas 70.2 percent.
Figure 8shows the changes inaverage number of drug mentions per
d
,
s
.
0
100 visits by physician specialty. Drugmention rates for cardiovasculardisease and psychiatry specialtiessignificantly increased, whereas therate for dermatology significantlydecreased.
Discussion
The demographic characteristics ofpatients seeking health care fromoffice-based physicians versus hospitalhealth care providers were quitedifferent. In the office-based setting thepatient is usually a privately insured,white female whose principal reason fothe visit is a ‘‘general medicalexamination’’ (1). In the hospital settingalthough the percent of visits by whitepersons is higher than other racialgroups, the visit rate for black personshigher than in the office-based setting.The most frequently mentioned expectesource of payment is Medicaid inOPD’s or private/commercial insurancein ED’s (2,3). Like the office-basedsetting the principal reason for the visitin the OPD is a ‘‘progress visit’’ or‘‘general medical examination’’ (2). Inthe ED the top principal reasons forvisit are generally pain, such as stomac
r
,
is
d
h
and abdominal pain, chest pain, orheadache (3). Physician offices, OPD’s,and ED’s provide primary health careservices for populations of differentdemographic characteristics and degreesof health, and accordingly, thecharacteristics of medications utilizedmay also differ. The use of medicationtherapy in ambulatory medical care indifferent settings is directly related tothe case-mix of conditions found inthose settings. In all settings (table 7)visits with illness or injury diagnoseswere more likely to have medicationtherapy than were visits withsupplemental diagnoses such as routinemedical exams. Physician office andOPD visits for patients who hadpreviously been seen at the providersetting were also more likely to havemedication therapy, whereas theopposite was true for ED visits.Variation in provider characteristics suchas OPD clinic type and office-basedphysician specialty (table 6) also appearto be related to the use of medicationtherapy. Percents of patient visits withdrug mentions vary from a low of34.5 percent for general surgeryspeciality to a high of 85.8 percent forcardiovascular disease specialty.
In addition to the use of medicationtherapy in ambulatory medical care, thisreport examines differences in the rateof the number of drug prescriptions pervisit. The data indicate that ED’s had ahigher average number of drug mentionsper visit than the other two settings.This finding may be due to the fact thatoffice-based and OPD settings includedvisits in their top 20 principal diagnosesfor circumstances other than disease orinjury, for example, general medicalexamination. This was not true for theED setting for the top 20 principaldiagnoses (1–3). ED’s also utilizedmore pain relief drugs and drugs withhigh abuse potential. Within hospitalsettings, ED’s used more antimicrobialagents than OPD’s. However, it appearsthere were no differences between thethree settings with respect to half of theirtop 10 generic substances, that is,amoxicillin, acetaminophen, erythromycin,albuterol, and ibuprofen. The severity ofconditions in emergency departmentswould naturally result in a greater use of
Advance Data No. 290 + August 8, 1997 9
drugs to stabilize patients than in the othertwo settings.
The aging of the U.S. populationbetween 1980 and 1992 has had anoticeable effect on office-basedmedication therapy characteristics.Between 1980 and 1992 the residentpopulation 75 years and older increased38.6 percent (15). Compatible with thisincrease, the percent of drug mentionsaccounted for by those 75 years andolder increased significantly, by44.3 percent from 1980 (7) to 1992.Because the rate of cardiovasculardisease increases with age in adults (16the aging of the population probablycontributed to the significant increase inpercent of drug mentions bycardiovascular disease specialists from1980 (1.5 percent) (7) to 1992(4.4 percent) and perhaps to the increasin cardiovascular disease specialists aswell.
Between 1980 and 1992, there wereseveral noteworthy occurrences for thetop-two ranked therapeutic drugcategories, antimicrobial agents andcardiovascular-renal drugs. In theantimicrobial class, a number of new,later generation cephalosporins wereintroduced. Also, a new therapeuticallyimportant subclass of antimicrobialswas first marketed, thefluoroquinolones. These drugs are allsingle entity. Additionally, a numberof single-entity drugs belonging to anew subclass of the cardiovascular-renal category were first marketedduring this time span: calciumchannel-blocking drugs. These drugsare indicated for a variety ofcardiovascular problems (e.g.,arrhythmias, hypertension, andangina). Relevant to the reduction ofcardiovascular risks is a subclass oflipid-lowering drugs, the HMG-CoA(3-hydroxy-3-methylglutaryl-coenzyme A) reductase inhibitors;these single-ingredient drugs were firstmarketed during 1980 to 1992. All ofthese events would have contributed tothe significant increase in percent ofsingle-ingredient drug mentions from1980 to 1992.
),
e
References
1. Schappert SM. National AmbulatoryMedical Care Survey: 1992summary. Advance data from vitaland health statistics; no 253.Hyattsville, Maryland: NationalCenter for Health Statistics. 1994.
2. McCaig LF. National HospitalAmbulatory Medical Care Survey:1992 outpatient department summary.Advance data from vital and healthstatistics; no 248. Hyattsville,Maryland: National Center for HealthStatistics. 1994.
3. McCaig LF. National HospitalAmbulatory Medical Care Survey:1992 emergency departmentsummary. Advance data from vitaland health statistics; no 245.Hyattsville, Maryland: NationalCenter for Health Statistics. 1994.
4. Nelson CR. Drug utilization in officepractice, National AmbulatoryMedical Care Survey, 1990. Advancedata from vital and health statistics;no 232. Hyattsville, Maryland:National Center for Health Statistics.1993.
5. Koch H, Knapp DA. Highlights ofdrug utilization in office practice,National Ambulatory Medical CareSurvey, 1985. Advance data fromvital and health statistics; no 134.Hyattsville, Maryland: NationalCenter for Health Statistics. 1987.
6. Koch H. Drug utilization in officepractice: 1980. Advance data fromvital and health statistics; no 81.Hyattsville, Maryland: NationalCenter for Health Statistics. 1982.
7. Koch H. Drug utilization in office-based practice, a summary offindings: National AmbulatoryMedical Care Survey: United States,1980. National Center for HealthStatistics. Vital Health Stat 13(65).1982.
8. Koch H, Campbell WH. Thecollection and processing of druginformation: National AmbulatoryMedical Care Survey: United States,1980. National Center for HealthStatistics. Vital Health Stat 2(65).1982.
9. American Drug Index. Facts andcomparisons. St. Louis, MO.
10. Drug topics red book, pharmacy’sfundamental reference. Montvale, NJ:Medical Economics Company, Inc.
11. Food and Drug Administration.National drug code directory, 1985edition. Washington: Public HealthService. 1985.
12. Drug facts and comparisons,loose-leaf drug information service.St. Louis,MO: Facts andComparisons.
13. Public Health Service and HealthCare Financing Administration.International Classification ofDiseases, 9th Revision, clinicalmodification. Washington: PublicHealth Service. 1980.
14. National Center for Health Statistics.Health, United States, 1995.Hyattsville, Maryland: Public HealthService. table 99, p 221. 1996.
15. National Center for Health Statistics.Health, United States, 1995. table 1,p 79. Hyattsville, Maryland: PublicHealth Service. 1996.
16. National Center for Health Statistics.Health, United States, 1995.Hyattsville, Maryland: Public HealthService. table 53, p 162; table 69,p 181) 1996.
17. McCaig LF, McLemore T. Plan andoperation of the National AmbulatoryMedical Care Survey. NationalCenter for Health Statistics. VitalHealth Stat 10(184). 1992.
18. National Center for Health Statistics:National Ambulatory Medical CareSurvey: background andmethodology, United States. NationalCenter for Health Statistics. VitalHealth Stat 2(61). 1974.
19. Shah BV, Barneswell BG, Hunt PN,La Vange LM. SUDAAN user’smanual release 5.50. ResearchTriangle Park, North Carolina:Research Triangle Institute. 1991.
Table 1. Number and percent distribution of drug mentions, percent of visits with drug mentions, and drug mention rate with corresponding standard er rors for age, sex, and raceof patient by ambulatory care setting: United States, 1992
Age, sex and race
Number of drug mentions in thousands Percent distribution Percent of visits with drug mentionsAverage number of drugmentions per 100 visits
Table 1. Number and percent distribution of drug mentions, percent of visits with drug mentions, and drug mention rate with corresponding standard er rors for age, sex, and raceof patient by ambulatory care setting: United States, 1992—Con.
Age, sex and race
Standard error in thousands Standard error of percent distributionStandard error of percent visits
with drug mentionsStandard error of averagedrug mentions per 100 visits
Table 2. Number and percent distribution of drug mentions with corresponding standard errors for selected drug characteristics by ambulatory care s etting: United States, 1992
Drug characteristic
Number of drug mentions in thousand Percent distribution
. . . Category not applicable.1This variable was not included in the ED Patient Record.
NOTE: The estimated number of patient visits was 762,045,000 to physician offices, 56,605,000 to outpatient departments, and 89,796,000 visits to emergency room departments. There were a total of 908,446,000 patient visits to all ambulatory medicalcare settings in 1992.
12Advance
Data
No.290
+August8,1997
Table 3. Number and percent distribution of drug mentions and drug mention rate with corresponding standard errors for therapeutic category by ambul atory care setting:United States, 1992
Therapeutic category1
Number of drug mentions in thousands Percent distributionAverage number of
Table 3. Number and percent distribution of drug mentions and drug mention rate with corresponding standard errors for therapeutic category by ambul atory care setting:United States, 1992—Con.
Therapeutic category1
Standard error in thousands Standard error of percent distributionStandard error of averagedrug mentions per 100 visits
. . . Category not applicable.0.0 Quantity more than zero but less than 0.05.* Figure does not meet standard of reliability or precision (more than 30-percent relative standard error in numerator of percent or rate).1Therapeutic categories are based on the standard drug classifications used in the National Drug Code Directory, 1985 edition (11).
NOTE: The estimated number of patient visits was 762,045,000 to physician offices, 56,605,000 to outpatient departments, and 89,796,000 visits to emergency room departments. There were a total of 908,446,000 patient visits to all ambulatory medicalcare settings in 1992.
14Advance
Data
No.290
+August8,1997
Table 4. The top 50 and top 10 generic substances most frequently utilized by rank, generic name, and number and percent of drugmentions with standard errors and therapeutic classifications by ambulatory care setting: United States, 1992
Table 4. The top 50 and top 10 generic substances most frequently utilized by rank, generic name, and number and percent of drugmentions with standard errors and therapeutic classifications by ambulatory care setting: United States, 1992—Con.
. . . Category not applicable.1Frequency combines mentions of single-ingredient agents with combination-ingredient agents.2Percent is based on respective number of mentions for each setting.3Therapeutic categories are based on the standard drug classifications used in the National Drug Code Directory, 1985 edition (11).
NOTE: The estimated number of patient visits was 762,045,000 to physician offices, 56,605,000 to outpatient departments, and 89,796,000 visits to emergency room departments. There were atotal of 908,446,000 patient visits to all ambulatory medical care settings in 1992.
16 Advance Data No. 290 + August 8, 1997
Table 5. The top 50 and top 10 drugs most frequently mentioned by rank, entry name, number and percent of mentions, withcorresponding standard errors and therapeutic classifications by ambulatory care setting: United States, 1992
Table 5. The top 50 and top 10 drugs most frequently mentioned by rank, entry name, number and percent of mentions, withcorresponding standard errors and therapeutic classifications by ambulatory care setting: United States, 1992—Con.
. . . Category not applicable.1 The trade or generic name used by the health care provider on the prescription or other medical records The use of trade names is for identification only and does not imply endorsement by thePublic Health Service or the U.S. Department of Health and Human Services Because of its nonspecific nature, the entry ‘‘allergy relief or shots’’ is omitted.2Therapeutic categories are based on the standard drug classifications used in the National Drug Code Directory, 1985 edition (11).
NOTE: The estimated number of patient visits to physician offices was 762,045,000, 56,605,000 patient visits to outpatient departments, and 89,796,000 visits to emergency room departmentsThere was a total of 908,446,000 patient visits to all ambulatory medical care settings in 1992.
18 Advance Data No. 290 + August 8, 1997
Table 6. Number and percent distribution of drug mentions, percent of visits with drug mentions, and drug mention rate withcorresponding standard errors for provider characteristics by ambulatory care setting: United States, 1992
Table 7. Percent distribution of ambulatory care visits by selected visit characteristics and corresponding percent of visits withmedication therapy and drug mention rate, for each ambulatory setting: United States, 1992
Selected visit characteristics
Percent distribution of all visitsPercent of visits withmedication therapy
. . . Category not applicable.1Source of payment categories reclassified to allow only one source per visit according to the following hierarchy: Medicaid, Medicare, HMO, private/commercial, self-pay, and other (no charge,other government, or unspecified source).2Major diagnosis for visit determined using the ICD–9–CM principal diagnosis codes as follows; Illness=codes 001–799, Injury=codes 800–999, and Supplementary classification=codes V01–V99.Based on the International Classification of Diseases, 9th Revision, Clinical Modification (ICD–9–CM) (13).3Old patient status for emergency departments based on an injury or illness follow-up.4Physician seen based on ‘‘provider seen’’ item for hospital settings and the ‘‘duration’’ item in physician offices. Office-based visits where the patient did not see a physician were coded as havinga duration of 0 minutes.
NOTES: Based on the 1992 estimates of the number of patient visits to physician offices (762,045,000), outpatient departments (56,605,000), and emergency departments (89,796,000). HMO ishealth maintenance organization.
20 Advance Data No. 290 + August 8, 1997
Table 8. The top 10 generic substances most frequently utilized by rank, generic name, percent of mentions, and drug mention rate foroffice-based physician visits: United States, 1980, 1992
Rank Generic name
Percent Average of drug mention rate per 100 visitsTherapeuticclassification619801,2 19922,3 19804 19925
. . . Category not applicable.– Quantity zero.1Based on 679,593,000 drug mentions for 1980.2Total drug mentions for 1980 and 1992 combine mentions of single-ingredient agents with combination-ingredient agents.3Based on 922,584,000 drug mentions for 1992.4Based on 575,745,000 visits for 1980.5Based on 762,045,000 visits for 1992.6Therapeutic categories are based on the standard drug classifications used in the National Drug Code Directory, 1985 edition (11).7Common to top 10 for both years.
Advance Data No. 290 + August 8, 1997 21
d
re
22 Advance Data No. 290 + August 8, 1997
Technical notesThis report is based on data
collected during the period January 199through December 1992 in the NationaAmbulatory Medical Care Survey(NAMCS) and December 1991 throughDecember 1992 in the National HospitaAmbulatory Medical Care Survey(NHAMCS). The NAMCS is a nationalprobability sample survey of ambulatorycare visits to private office-basedphysicians. The NHAMCS is a nationalprobability sample survey of non-Federal, general, and short-stayhospitals. Both surveys are conducted bthe National Center for Health StatisticsCenters for Disease Control andPrevention. The NAMCS and NHAMCSsurvey designs and procedures arepresented briefly below. Detaileddescriptions of the plan and operationsof the NAMCS and NHAMCS havebeen published (17,18). Summaryreports and reports on special topicsusing NAMCS and NHAMCS data arepresented in Series 13 of the NCHSVital and Health Statistics series as welas in Advance Data from Vital andHealth Statistics reports. NAMCS andNHAMCS microdata are also availableon public-use tape and CD-ROM.
Statistical design
Scope of the survey
The target universe of the 1992NAMCS included office visits made inthe United States by ambulatory patiento nonfederally employed physicianswho were principally engaged in officepractice, but not in the specialties ofanesthesiology, pathology, or radiology.Physicians who are principally engagedin teaching, research, or administrationwere also excluded. The sampling framincluded physicians who were classifiedby the American Medical Association(AMA) or the American OsteopathicAssociation (AOA) as ‘‘office-based,patient care.’’ Visits to privatenonhospital-based clinics and healthmaintenance organizations were withinthe scope of the survey, but those thattook place in government-operatedfacilities and hospital-based outpatientdepartments were not.
2l
l
y,
l
ts
e
The target population of theNHAMCS includes visits in the UnitedStates to emergency departments (ED’and outpatient departments (OPD’s) ofnoninstitutional general and short-stayhospitals, exclusive of Federal, military,and Veterans Administration hospitals.Only OPD clinics under the supervisionof physicians were included within thescope of the survey. Clinics specializingin radiology, laboratory services,physical rehabilitation, or other ancillaryservices were excluded from the surveTelephone contacts were excluded inboth NAMCS and NHAMCS.
Characteristics of the physician’spractice (such as primary specialty andtype of practice) and the hospital (suchas ownership and expected number ofOPD and/or ED visits) were obtainedfrom the physician or hospitaladministrator during an inductioninterview. The U.S. Bureau of theCensus, Housing Surveys Branch, wasresponsible for data collection for bothsurveys. Data-processing operations anmedical coding were performed by theNational Center for Health Statistics,Health Care Surveys Section, ResearcTriangle Park, North Carolina.
Sample design
NAMCS uses a multistageprobability sample design involvingsamples of primary sampling units(PSU’s), physician practices withinPSU’s, and patient visits withinphysician practices. The PSU’s arecounties, groups of counties, countyequivalents (such as parishes orindependent cities), or towns andtownships (for some PSU’s in NewEngland). For 1992, a sample of 3,000nonfederal office-based physicians wasselected from master files maintained bthe AMA and AOA. Physicians werescreened at the time of the survey toensure that they were eligible for surveparticipation. Of those screened, 858physicians were ruled ineligible (out ofscope) due to being retired; beingemployed primarily in teaching,research, or administration; or otherreasons. The remaining 2,142 physiciawere in scope or eligible to participatein the survey. The physician response
s)
y.
d
h
y
y
ns
rate for the 1992 NAMCS was71.4 percent.
Sample physicians were asked tocomplete Patient Record forms (figure 1)for a systematic random sample of officevisits occurring during a randomlyassigned 1-week reporting period.Responding physicians completed34,606 Patient Record forms, on whichthey recorded 36,647 drug mentions.
NHAMCS utilizes a 4-stage surveyprobability sample design involvingsamples of PSU’s, hospitals withinPSU’s, outpatient clinics and emergencyservice areas (ESA’s) within hospitals,and patient visits within outpatientclinics and ESA’s. For 1992, a sampleof 524 non-Federal, short-stay, orgeneral hospitals was selected from theSMG Hospital Market Database. Of thisgroup, 474 hospitals were in scope, oreligible to participate in the survey. Thehospital response rate for the NHAMCSduring this period was 93 percent. Baseon the induction interview, 314 of thesample hospitals had OPD’s and 437 ofthe sample hospitals had ED’s. Hospitalstaff were asked to complete PatientRecord forms (figures 2and3) for asystematic random sample of patientvisits occurring during a randomlyassigned 4-week reporting period. Thenumber of Patient Record formscompleted for OPD’s was 35,114, onwhich they recorded 38,507 drugmentions. The number of Patient Recordforms completed for ED’s was 36,271,on which they recorded 45,844 drugmentions.
Sampling errors—The standarderror is primarily a measure of thesampling variability that occurs bychance when only a sample, rather thanan entire universe, is surveyed. Thestandard error also reflects part of themeasurement error but does not measuany systematic biases in the data. Thechances are 95 out of 100 that anestimate from the sample differs fromthe value that would be obtained from acomplete census by less than twice thestandard error. The standard errorspresented in tables of this report andused in tests of significance wereapproximated using SUDAAN software.SUDAAN computes standard errors byusing a first-order Taylor approximationof the deviation of estimates from their
e
Advance Data No. 290 + August 8, 1997 23
expected values. A description of thesoftware and the approach it uses hasbeen published (19). Standard errors foall estimates are presented in each tabThe relative standard error of anestimate can be calculated by dividingthe standard error by the estimate itselfThe result is then expressed as a perceof the estimate.
Nonsampling errors—Estimatesbased on the 1992 NAMCS andNHAMCS are subject to nonsampling awell as sampling errors. Nonsamplingerrors include reporting and processingerrors, as well as biases due tononresponse or incomplete response.Although the magnitude of thenonsampling errors cannot be computethese errors are kept to a minimum byprocedures built into the operation ofthe survey. To eliminate ambiguities andencourage uniform reporting, carefulattention was given to the phrasing ofquestions, terms, and definitions. Also,extensive pretesting of most data itemsand survey procedures was alsoperformed. Quality control proceduresand consistency and edit checks reduceerrors in data coding and processing.Because survey results are subject tosampling and nonsampling errors, thetotal error will be larger than the errorfrom sampling variability alone.
Adjustments for nonresponse
Offıce-based nonresponse—Estimates from NAMCS data wereadjusted to account for samplephysicians who were in scope but didnot participate in the study. Thisadjustment was calculated to minimizethe impact of response on final estimateby imputing to nonrespondingphysicians data from visits to similarphysicians. For this purpose, physicianswere judged similar if they had thesame specialty designation and practicein the same PSU.
Hospital nonresponse—Estimatesfrom NHAMCS data were adjusted toaccount for sample hospitals that werein scope but did not participate in thestudy. This adjustment was calculated tminimize the impact of nonresponse onfinal estimates by imputing tononresponding hospitals data from visitto similar hospitals. For this purpose,
rle.
.nt
s
d,
d
s
d
o
s
hospitals were judged similar if theywere in the same region, ownershipcontrol group, and metropolitanstatistical area control group.
ED and/or clinic nonresponse—Estimates from NHAMCS data wereadjusted to account for ED’s and samplclinics that were in scope but did notparticipate in the study. This adjustmenwas calculated to minimize the impactof nonresponse on final estimates byimputing to nonresponding ED’s orclinics data from visits to similar ED’sor clinics. For this purpose, emergencydepartments or clinics were judgedsimilar if they were in the same ED orclinic group.
Test of significance and rounding
In this report, the determination ofstatistical inference was based on thetwo-tailed t-test. The Bonferroniinequality was used to establish thecritical value for statistical significanceand differences (0.05 level ofsignificance over all analyses performedon estimates in a table). Terms relatingto differences such as ‘‘greater than’’ or‘‘less than’’ indicate that the differencesare statistically significant. A lack ofcomment regarding the differencebetween any two estimates does notmean that the difference was tested andfound to be not significant. Chi-squaretests were performed using SUDAANroutines that take into account thecomplex sampling design.
Estimates in this report are roundedto the nearest thousand. For this reasondetailed figures within tables do notalways add to totals. Rates and percenare calculated based on the original,unrounded figures and may not agreeprecisely with rates and percentscalculated from rounded data.
Definition of terms
Ambulatory patient—An ambulatorypatient is an individual seeking personahealth services who is not currentlyadmitted to any health care institutionon the premises. Patients are defined ain scope or out of scope as follows:
+ In scope—In NAMCS, patients seenby physicians or staff members intheir private offices. In NHAMCS,
e
t
,
ts
l
s
patients seen by hospital staff in anin-scope emergency service area orclinic.
+ Out of scope—Patients seen in anursing home or other extended careinstitution or at home. Patients whocontact and receive advice fromhealth care providers via telephone.Patients who come to the hospital oroffice only to leave a specimen, topick up insurance forms, to pick upmedication, or to pay a bill.
Clinic—A clinic is an administrativeunit within an organized outpatientdepartment of a hospital that providesambulatory medical care under thesupervision of a physician. Thisexcludes the ‘‘hospital as landlord’’arrangement in which the hospital onlyrents space to a physician group and isnot otherwise involved in the delivery ofservices. Clinics are grouped into thefollowing six specialty groups forpurposes of systematic sampling andnonresponse adjustment: generalmedicine, surgery, pediatrics, obstetrics/gynecology, substance abuse, and other.The following are examples of the typesof clinics excluded: ambulatory surgerycenters, chemotherapy, employee healthservice, renal dialysis, methadonemaintenance, and radiology.
Control status—Controlledmedications, because of their significantpotential for dependence or abuse andtheir possible diversion into illicitchannels are regulated under Federallaw by the Department of Justice, DrugEnforcement Agency. The ControlledSubstances Act of 1970 characterizeseach controlled drug into one of fiveschedules. Schedule I drugs, like heroinand LSD, have a high potential forabuse and no current accepted medicalusefulness for treatment in the UnitedStates. Schedule I drugs are outside thescope of this report. Each successiveschedule, II through V, reflects adecreasing degree of dependence andpotential for abuse.
Drug mention—A drug mention isthe physician’s or other health careprovider’s entry on the Patient Recordform of a pharmaceutical agent—by anyroute of administration—for prevention,diagnosis, or treatment. Generic as wellas brand-name drugs are included, as ar
24 Advance Data No. 290 + August 8, 1997
nonprescription and prescription drugs.Along with all new drugs, the physicianor other health care provider also recordcontinued medications if the patient wasspecifically instructed during the visit tocontinue the medication. Up to fivemedications may be reported per visit.
Emergency department—Anemergency department (ED) is ahospital facility staffed 24 hours a dayfor the provision of unscheduledoutpatient services to patients whoseconditions require immediate care. If anED provided emergency services indifferent areas of the hospital, then allthese areas were selected with certaininto the sample. Off-site ED’s open lesthan 24 hours a day are included ifstaffed by the hospital’s emergencydepartment.
Emergency service area—The areawithin the emergency department wheremergency services are provided. Thisincludes services provided under the‘‘hospital as landlord’’ arrangement in
Suggested citation
Nelson CR, Knapp DE. Medication therapy inambulatory medical care: National AmbulatoryMedical Care Survey and National HospitalAmbulatory Care Survey, 1992. Advance datafrom vital and health statistics; no 290.Hyattsville, Maryland: National Center forHealth Statistics. 1997.
DEPARTMENT OFHEALTH & HUMAN SERVICES
Centers for Disease Control and PreventionNational Center for Health Statistics6525 Belcrest RoadHyattsville, Maryland 20782
OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300
To receive this publication regularly, contactthe National Center for Health Statistics bycalling 301-436-8500E-mail: [email protected]: http://www.cdc.gov/nchswww/nchshome
DHHS Publication No. (PHS) 97-12507-0566 (8/97)
s
oftys
e
which the hospital rents space to aphysician group.
Offıce—An office is the spaceidentified by a physician as a locationfor her or his ambulatory practice.Offices customarily include consultation,examination, or treatment spaces thatpatients associate with the particularphysician.
Offıce-based physician—A physicianis a duly licensed doctor of medicine(M.D.) or doctor of osteopathy (D.O.)who is currently in office-based practiceand who spends time caring forambulatory patients. Excluded from theNAMCS are physicians who are hospitalbased; who specialize in anesthesiology,pathology, or radiology; who are federallyemployed; who treat only institutionalizedpatients; or who are employed full timeby an institution and spend no time seeinambulatory patients.
Outpatient department—Anoutpatient department is a hospitalfacility where nonurgent ambulatory
Trade name disclaimer
The use of trade names is for identificationonly and does not imply endorsement by thePublic Health Service, U.S. Department ofHealth and Human Services.
Copyright information
All material appearing in this report is in thepublic domain and may be reproduced orcopied without permission; citation as tosource, however, is appreciated.
.htm
g
medical care is provided under thesupervision of a physician.
Ownership—Hospitals aredesignated according to the primaryowner of the hospital based on the SMGHospital Market Database.
+ Government, non-Federal—Hospitalsoperated by a State or localgovernment.
+ Proprietary—Hospitals operated byindividuals, partnerships, orcorporations for profit.
+ Voluntary nonprofit—Hospitalsoperated by a religious or othernonprofit organization.
Visit—A visit is a direct personalexchange between an ambulatory patientand a physician or health care provider orstaff member working under thephysician’s supervision for the purpose ofseeking care and rendering personal healthservices. Excluded are visits wheremedical care was not provided, such asvisits made to drop off specimens, paybills, make appointments, and walkouts.