0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

The Impact of Prescribing Safety Alerts for Elderly Persons in an Electronic Medical Record:  An Interrupted Time Series Evaluation FREE

David H. Smith, RPh, PhD; Nancy Perrin, PhD; Adrianne Feldstein, MS, MD; Xiuhai Yang, MS; Daniel Kuang, MS; Steven R. Simon, MD, MPH; Dean F. Sittig, PhD; Richard Platt, MS, MD; Stephen B. Soumerai, ScD
[+] Author Affiliations

Author Affiliations: Kaiser Permanente Center for Health Research (Drs Smith, Perrin, Feldstein, and Sittig and Messrs Yang and Kuang) and Oregon Health and Sciences University (Dr Perrin), Portland; and Harvard Medical School and Harvard Pilgrim Health Care, Boston, Mass (Drs Simon, Platt, and Soumerai).


Arch Intern Med. 2006;166(10):1098-1104. doi:10.1001/archinte.166.10.1098.
Text Size: A A A
Published online

Background  Considerable effort and attention have focused on medication safety in elderly persons; one approach that has been understudied in the outpatient environment is the use of computerized provider order entry with clinical decision support. The objective of this study was to examine the effects of computerized provider order entry with clinical decision support in reducing the use of potentially contraindicated agents in elderly persons.

Methods  With data from a 39-month period of a natural experiment, we evaluated changes in medication dispensing using interrupted time series analysis to estimate changes, controlling for prealert prescribing trends. The setting was a large health maintenance organization in the Pacific Northwest. All adult enrollees of the health plan participated. The intervention was computerized alerts cautioning against using certain medications in elderly persons. The main outcome measure was dispensing per 10 000 members per month.

Results  Following the implementation of the drug-specific alerts, a large and persistent reduction (5.1 prescriptions per 10 000, P=.004), a 22% relative decrease from the month before alert implementation, in the exposure of elderly patients to nonpreferred medications was observed. We found no evidence of a decrease in use of nonpreferred agents for nonelderly patients. The reduction seen in use of nonpreferred agents for elderly persons was driven primarily by decreases in dispensing for tertiary tricyclic agents.

Conclusions  We found that alerts in an outpatient electronic medical record aimed at decreasing prescribing of medication use in elderly persons may be an effective method of reducing prescribing of contraindicated medications. The effect of the alerts on patient outcomes is less certain and deserves further investigation.

Figures in this Article

Considerable effort and attention have focused on the development and dissemination of evidence-based guidelines to reduce medication errors. Unfortunately, these efforts have not yet translated into wanted changes in physician behavior.13 Even when the physician is aware of a guideline, time and mental processing constraints at clinical decision making hamper its use.4 Guideline implementation strategies, including alerts, reminders, and other clinical decision support,5 are more likely to be effective because they deliver specific advice at the time and place of a consultation. Such strategies have been associated with reduced medication prescribing errors, improved preventive and chronic disease care, and improved physician and patient satisfaction in controlled trials, predominantly in academic settings or inpatient environments.612 However, these studies are difficult to generalize to community-based care, where most prescribing occurs. In addition, many previous analyses have had short follow-up times, limiting inferences regarding erosion of effect over time.13,14

Systems analysis suggests that computerized provider order entry (CPOE) with clinical decision support systems can reduce medication errors.15 A recent systematic review16 supports the effectiveness of CPOE with clinical decision support in reducing the frequency of medication errors in inpatients. However, there is little research on the effectiveness of electronic medical record (EMR) alerts for promoting medication safety in the outpatient environment. This topic is important because about half of medication errors occur during the prescribing phase,17,18 and most prescribing occurs in the outpatient setting.

The goal of this study was to evaluate the impact of decision support at provider medication order entry, which sought to reduce the use of 2 classes of medications generally contraindicated in elderly persons. We took advantage of a natural experiment and 39 months of data at a large health maintenance organization (HMO), using a strong quasi-experimental design and interrupted time series analysis, to examine the effects of CPOE with clinical decision support in reducing the use of potentially contraindicated agents in elderly persons. To our knowledge, this is the first study examining the effects of medication safety alerts (CPOE with clinical decision support) in a population-based primary care setting.

STUDY DESIGN

By using data from a 39-month period, we evaluated changes in the use of medications generally contraindicated in elderly persons, associated with the implementation of a CPOE alert cautioning against using these 2 classes of medications in elderly persons. We used interrupted time series analysis to estimate changes in the use of these medications, controlling for prealert trends.

SETTING AND INTERVENTION

The setting for the study was a group-model HMO in the Pacific Northwest caring for about 450 000 people, with demographic characteristics closely mirroring those of the surrounding metropolitan area. The Research Subjects Protection Office approved the study. The HMO uses an EMR (EpicCare; Epic Systems Corporation, Verona, Wis) for all outpatient encounters, including the ordering of all medications. From September to October 2000, decision support was implemented, which alerted clinicians to preferred alternative medications when they ordered certain nonpreferred agents that carry potential contraindication in elderly persons.19 These nonpreferred agents were certain long-acting benzodiazepines (diazepam, flurazepam, triazolam, and chlordiazepoxide) and tertiary amine tricyclic antidepressants (TCAs) (imipramine, amitriptyline, and doxepin); the shorter-acting and less-sedating alternative benzodiazepines (oxazepam and temazepam), secondary amine TCAs (nortriptyline and desipramine), or other medications (buspirone, trazodone, and paroxetine) were suggested as alternatives. Paroxetine was subject to changes in formulary status during the period of study, leading to large fluctuations in use that were unrelated to the alerts, so we did not include it in any of our analyses.

The alerts (Figure 1) made it possible to change the order by accepting the alternative medication suggested. These alerts were drug specific, meaning that they were presented to the clinician after ordering one of the nonpreferred medications, regardless of patient characteristics, such as age or comorbidities. As Figure 1 illustrates, the alert was constructed so that the warning regarding falls and fractures was prominent. If desired, the user could scroll down the message pane to see the rest of the text, which contained further information on specific preferred alternative agents.

Place holder to copy figure label and caption
Figure 1.

Example screen shot of an alert for diazepam, a nonpreferred benzodiazepine.

Graphic Jump Location
STUDY SAMPLE

We collected electronic data on monthly outpatient medication dispensings from prescriptions written by primary care providers between October 1, 1999, and December 31, 2002; this period included 12 months before the alerts were implemented and 27 months after implementation. We included data from all 15 primary care clinics. Providers were family practitioners and internal medicine clinicians, including physicians (n = 152), nurse practitioners (n = 25), and physician assistants (n = 32) (the numbers are for the 12 months preceding the intervention). All members of the HMO during the observation period were included in the study.

DATA SOURCES AND MEASURES

We were interested in examining the effect of the alerts on initial medication prescribing, because it is during prescribing that alerts are presented to the clinician. Our data source was the prescription-dispensing database. The dispensing database includes prescription refills, but refills are not subject to the alert. Prescriptions are given a unique identification number, and this identification number is associated with the entire stream of prescription refills, making it impossible to directly differentiate between initial dispensings and refills. To minimize inclusion of refills, we included dispensings only when (1) the patient had not been exposed to the medication in the previous 6 months and (2) the prescription was the first in a stream of refills for a given prescription identification number.

The alerts in this analysis were presented when any nonpreferred drug was prescribed, regardless of age or clinical characteristics of the patient. It was left to the prescriber to determine the clinical appropriateness of the advisory. Because of this, we were interested in whether there was any spillover effect in nonelderly persons, so we examined results for patients in the elderly (aged ≥65 years) and nonelderly (aged <65 years) groups.

STATISTICAL ANALYSIS

Outcomes of interest (number of dispensings of nonpreferred and preferred drugs per 10 000 population) were aggregated into 39 monthly intervals. We used interrupted time series regression models (segmented regression) to estimate changes in outcomes that occurred after the alert implementation, controlling for preintervention trends and other autocorrelation (including seasonal effects).2023 The models included a constant, a baseline slope term to control for secular trends (eg, first-, second-, and higher-order autocorrelation), and terms estimating changes in the level and slope of outcome rates. Analyses were conducted using SAS statistical software (Proc Autoreg; SAS Institute Inc, Cary, NC).

At baseline, patients with a dispensing for the nonpreferred agents were older (nonelderly patients, 44.9 vs 31.4 years; and elderly patients, 75.1 vs 74.7 years) and more likely to be female (nonelderly patients, 72.1% vs 51.7%; and elderly patients, 69.4% vs 56.2%) compared with the general HMO population. Data on the use of nonpreferred and preferred medications at baseline are given in the Table. Controlling for baseline trends, a sudden reduction in the rate of initial dispensing of nonpreferred agents among elderly persons was observed when the alerts were placed (Figure 2A), from 21.9 to 16.8 per 10 000 (P<.01). This decrease of 5.1 prescriptions per 10 000 represented a relative decrease of 22% from the month before alert implementation and was sustained over the entire 2-year postalert observation period. For nonelderly persons, no significant (P=.48) changes were observed in the use of nonpreferred agents.

Place holder to copy figure label and caption
Figure 2.

Time series of nonpreferred (A) and preferred (B) agent dispensings per 10 000 by age. Fitted trend lines show predicted values from the segmented regression analysis.

Graphic Jump Location
Table Graphic Jump LocationTable. Use of Nonpreferred and Preferred Medications at Baseline*

The dispensing rate of preferred agents did not change significantly (P=.66) for elderly persons, but did continue an upward trend throughout the observation period (Figure 2B). Controlling for preintervention trends, the use of preferred agents increased suddenly by 4.0 dispensings per 10 000 (P<.01) for nonelderly persons, representing a 20% increase from the month before alert implementation. An upward trend continued over the observation period.

Controlling for baseline trend, the rate of use of nonpreferred TCAs for elderly persons decreased abruptly by 4.5 dispensings per 10 000, a 35% decrease from the month before the alert, with a further continued decline of 0.4 prescription per 10 000 per month (P<.01) observed over the entire 39-month study period. The rate of initial dispensing of the nonpreferred TCAs for elderly persons had been increasing at a rate of 0.3 prescription per 10 000 per month (P=.02) before the alerts were placed—this trend was reversed after alert implementation. The rate of overall dispensing for nonpreferred TCAs did not change significantly (P=.22) before or after alert implementation for nonelderly persons.

The medications with the largest observed changes in dispensing rates were the TCA amitriptyline (nonpreferred) and its metabolite, the preferred TCA, nortriptyline. Nortriptyline was specifically suggested in the alert as the preferred alternative to amitriptyline. The rate of dispensings for the nonpreferred TCA (amitriptyline) was higher than for the preferred TCA (nortriptyline) for both age groups at every time point before the alert (Figure 3). This trend was suddenly reversed after the alerts. Controlling for baseline trends, there was an abrupt and sustained 34% decrease from 9.8 (1 month before alert implementation) to 6.4 dispensings per 10 000 (P<.01) in the use of the TCA amitriptyline for elderly persons, with a further slight decline (−0.3 prescription per 10 000 per month, P=.004) during the follow-up period of observation. Use of the TCA amitriptyline also decreased for nonelderly persons, but not significantly (P = .22) (Figure 3A). Use of the preferred TCA nortriptyline increased suddenly in both age groups, by 3.5 prescriptions per 10 000 for elderly persons (P<.01) and by 1.9 prescriptions per 10 000 for nonelderly persons (P<.01) (Figure 3B).

Place holder to copy figure label and caption
Figure 3.

Time series of the dispensings per 10 000 of the tricyclic antidepressants amitriptyline (A) and nortriptyline (B) by age. Fitted trend lines show predicted values from the segmented regression analysis.

Graphic Jump Location

Although the monthly rate of initial prescribing for nonpreferred benzodiazepines did not change, the monthly initial prescribing rate of flurazepam per 10 000 did decrease immediately following the alert for elderly persons, by 0.3 dispensing per 10 000 (P<.10), with a continued decrease of 0.04 dispensing per 10 000 per month (P = .02), which persisted to the end of the study period; no changes were observed for nonelderly persons. No significant (P = .36) changes were noted in the rate of use of preferred benzodiazepines or “other” preferred agents (buspirone and trazodone) in either age group—the small sample sizes did not produce stable effects.

Following the implementation of drug-specific alerts in an outpatient setting aimed at decreasing the use of medications that are often contraindicated in elderly persons, we found an abrupt, large, and persistent reduction in the exposure of elderly patients to nonpreferred medications. Among elderly persons, we found that the use of nonpreferred agents decreased immediately by 5.1 prescriptions per 10 000 (P<.01), a 22% relative decrease from the month before alert implementation. This reduction was sustained over the entire 2-year postalert period of observation, and was driven primarily by decreases in dispensings for nonpreferred TCAs. The reduction in the use of nonpreferred TCAs was especially striking given that their rate of dispensing was increasing significantly before the intervention, then continued significantly downward at every time point after the intervention. We found no evidence of a decrease in the use of nonpreferred agents for nonelderly patients.

Among all the medications studied, the TCAs amitriptyline and nortriptyline had the largest changes in dispensing rates. Before the alert, the nonpreferred TCA amitriptyline was prescribed more frequently at every time point than the preferred TCA nortriptyline, but after the alert, this finding was exactly reversed. One reason may be that there is a clear therapeutic interchange—amitriptyline is metabolized to nortriptyline—perhaps making the exchange more amenable for clinician and patient acceptance. In addition, for other nonpreferred medications, specifically the long half-life benzodiazepines, there is conflicting opinion on whether they should be avoided in elderly persons.24

We found that while the use of nonpreferred medications decreased for elderly persons, there was no offsetting increase in the use of preferred medications. This finding suggests that the alerts had the effect of decreasing overall exposure of older patients to these types of drugs, nonpreferred and preferred. While this may have been clinically appropriate for some patients, it is possible that others were denied appropriate medication. Alternatively, clinicians may have switched to other medications that we did not include in our analysis.

This research suggests there may be some unexpected consequences from the alerts, namely, that the rates of use of preferred medications increased for people in younger age groups. One reason for this finding may be the high fraction of prescribing encounters with nonelderly patients for whom the alerts were generated. Also, the alert functioned such that it was easy to choose the recommended alternative medication, and the clinicians' “reward” for choosing the alternative was a completed prescription automatically prepopulated (with dose, directions for use, and refills) that was ready for electronic signature. Anecdotes from clinicians suggest that they would start a prescription with the nonpreferred medication to access the prepopulated prescription.

This study includes alerts with particular characteristics, potentially limiting generalizability. First, even though the premise of the safety message contained in these alerts was related to patient age, because of the functionality constraints of the specific EMR in use at the time of the study, ordering of specific medications, regardless of patient age, triggered the alerts. At the HMO, about 25% of all prescriptions for the nonpreferred drugs are for people older than 65 years. An alert with more patient specificity (ie, only present for elderly patients) would reduce provider burden; whether such an alert would be differentially effective is unknown.

A remaining unanswered question is the relationship of the intervention to patient outcomes. The alerts may have decreased the use of nonpreferred medications, but that decrease does not necessarily translate into improved outcomes (eg, decreased rate of daytime sedation and fewer falls). In addition, there is some controversy surrounding the clinical appropriateness of recommendations regarding some of the medications targeted by the alerts in this analysis. Specifically, some research24 has suggested that long half-life benzodiazepines are less likely than short half-life benzodiazepines to be associated with hip fractures. This literature may have had some effect on the effectiveness of the alerts if, for example, clinicians chose to ignore the alerts because of their awareness of the controversy. However, these medications remain on recent lists of medications to avoid in elderly persons.19 A recent study25 suggests that elderly nursing home patients who are exposed to potentially inappropriate agents (including those in our study) experience a higher risk of hospitalization and death. That study examined a broader array of agents and a more frail population, but does suggest that decreasing exposure to these agents is beneficial.

The alerts were not entirely successful in eliminating the use of these medications in elderly persons. Some use of these drugs in elderly persons is probably clinically appropriate, however, so we would not expect to see the use of the drugs decline to 0. For example, it might be reasonable for a patient who has successfully used diazepam for relief of anxiety in the past to use it again at a low dosage for the short term, particularly if the patient has not responded well to an alternative preferred agent, such as oxazepam or buspirone.

Compared with studies16 conducted in the inpatient setting of CPOE with clinical decision support, our results suggest a similar direction of effect for the outpatient environment (ie, toward decreasing contraindicated prescribing). Comparing the results of these studies is difficult because of differing clinical topics, study designs, and outcome measures. However, the reduction of 22% in potentially inappropriate prescribing of nonpreferred agents in those older than 65 years that we observed is consistent with findings from studies16 of inpatient CPOE with clinical decision support.

One small Canadian study26 in the outpatient setting that examined the effectiveness of clinical decision support (but without CPOE) found a reduction of 13% (relative rate, 0.77; 95% confidence interval, 0.59-1.00) in inappropriate prescribing of long-acting benzodiazepines and active metabolite TCAs. Our results reinforce that finding in a larger population with a mature EMR.

There are alternative explanations for our findings that should be considered. Specifically, the issue of avoiding certain drugs in elderly persons has gained press in recent years,19,27 raising awareness in the clinical community. However, given our strong quasi-experimental methods, and the sudden decrease we observed contemporaneous with the alert initiation, this is an unlikely explanation for the observed changes in medication use. In addition, we are unaware of any ongoing cointervention that could explain these results. We did investigate this possibility thoroughly by directly inquiring with appropriate personnel in pharmacy and other relevant departments at the HMO.

In conclusion, we found that alerts in an outpatient EMR aimed at decreasing prescribing of medication use in elderly persons may be an effective method of reducing prescribing of contraindicated medications. Additional work on the impact of more patient-specific clinical decision support will be useful to determine whether an intervention method with less clinician burden can have similar effects in the outpatient setting. More important, the effect of the alerts on patient outcomes is less certain and deserves further investigation.

Correspondence: David H. Smith, RPh, PhD, Kaiser Permanente Center for Health Research, 3800 N Interstate Ave, Portland, OR 97227 (David.H.Smith@kpchr.org).

Accepted for Publication: December 27, 2005.

Financial Disclosure: None.

Funding/Support: This study was supported by grant 1 U18 HS11843 from the Agency for Healthcare Research and Quality, Rockville, Md.

Role of the Sponsor: The funding body had no role in data extraction and analyses, in the writing of the manuscript, or in the decision to submit the manuscript for publication.

Acknowledgment: We thank the many individuals who assisted with this project, particularly those on the Regional Formulary and Therapeutics Committee and in Information Technology.

Grimshaw  JMRussell  IT Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations. Lancet 1993;3421317- 1322
PubMed Link to Article
Grimshaw  JMThomas  REMacLennan  G  et al.  Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol Assess 2004;8iii- iv, 1-72
PubMed
Gilbody  SWhitty  PGrimshaw  JThomas  R Educational and organizational interventions to improve the management of depression in primary care: a systematic review. JAMA 2003;2893145- 3151
PubMed Link to Article
McBride  PEPacala  JTDean  JPlane  MB Primary care residents and the management of hypercholesterolemia. Am J Prev Med 1990;671- 76
PubMed
Shiffman  RNLiaw  YBrandt  CACorb  GJ Computer-based guideline implementation systems: a systematic review of functionality and effectiveness. J Am Med Inform Assoc 1999;6104- 114
PubMed Link to Article
Balas  EABoren  SAGriffing  G Computerized management of diabetes: a synthesis of controlled trials. Proc AMIA Symp November1998;29- 5299
PubMed
Balas  EAWeingarten  SGarb  CTBlumenthal  DBoren  SABrown  GD Improving preventive care by prompting physicians. Arch Intern Med 2000;160301- 308
PubMed Link to Article
Elson  RBConnelly  DP Computerized patient records in primary care: their role in mediating guideline-driven physician behavior change. Arch Fam Med 1995;4698- 705
PubMed Link to Article
Rind  DMSafran  CPhillips  RS  et al.  Effect of computer-based alerts on the treatment and outcomes of hospitalized patients. Arch Intern Med 1994;1541511- 1517
PubMed Link to Article
Jha  AKKuperman  GJTeich  JM  et al.  Identifying adverse drug events: development of a computer-based monitor and comparison with chart review and stimulated voluntary report. J Am Med Inform Assoc 1998;5305- 314
PubMed Link to Article
Classen  DCPestotnik  SLEvans  RSBurke  JP Computerized surveillance of adverse drug events in hospital patients. JAMA 1991;2662847- 2851
PubMed Link to Article
Bates  DW Using information technology to reduce rates of medication errors in hospitals. BMJ 2000;320788- 791
PubMed Link to Article
Demakis  JGBeauchamp  CCull  WL  et al.  Improving residents' compliance with standards of ambulatory care: results from the VA Cooperative Study on Computerized Reminders. JAMA 2000;2841411- 1416
PubMed Link to Article
Abookire  SATeich  JMSandige  H  et al.  Improving allergy alerting in a computerized physician order entry system. Proc AMIA Symp November2000;2- 6
PubMed
Leape  LLBates  DWCullen  DJ  et al. ADE Prevention Study Group, Systems analysis of adverse drug events. JAMA 1995;27435- 43
PubMed Link to Article
Kaushal  RShojania  KGBates  DW Effects of computerized physician order entry and clinical decision support systems on medication safety: a systematic review. Arch Intern Med 2003;1631409- 1416
PubMed Link to Article
Bates  DWCullen  DJLaird  N  et al. ADE Prevention Study Group, Incidence of adverse drug events and potential adverse drug events: implications for prevention. JAMA 1995;27429- 34
PubMed Link to Article
Kaushal  RBates  DWLandrigan  C  et al.  Medication errors and adverse drug events in pediatric inpatients. JAMA 2001;2852114- 2120
PubMed Link to Article
Fick  DMCooper  JWWade  WEWaller  JLMaclean  JRBeers  MH Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Arch Intern Med 2003;1632716- 2724
PubMed Link to Article
Wagner  AKSoumerai  SBZhang  FRoss-Degnan  D Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 2002;27299- 309
PubMed Link to Article
Soumerai  SBMcLaughlin  TJRoss-Degnan  DCasteris  CSBollini  P Effects of a limit on Medicaid drug-reimbursement benefits on the use of psychotropic agents and acute mental health services by patients with schizophrenia. N Engl J Med 1994;331650- 655
PubMed Link to Article
Soumerai  SBRoss-Degnan  DGortmaker  SAvorn  J Withdrawing payment for nonscientific drug therapy: intended and unexpected effects of a large-scale natural experiment. JAMA 1990;263831- 839
PubMed Link to Article
Gillings  DMakuc  DSiegel  E Analysis of interrupted time series mortality trends: an example to evaluate regionalized perinatal care. Am J Public Health 1981;7138- 46
PubMed Link to Article
Wagner  AKZhang  FSoumerai  SB  et al.  Benzodiazepine use and hip fractures in the elderly: who is at greatest risk? Arch Intern Med 2004;1641567- 1572
PubMed Link to Article
Lau  DTKasper  JDPotter  DELyles  ABennett  RG Hospitalization and death associated with potentially inappropriate medication prescriptions among elderly nursing home residents. Arch Intern Med 2005;16568- 74
PubMed Link to Article
Tamblyn  RHuang  APerreault  R  et al.  The medical office of the 21st century (MOXXI): effectiveness of computerized decision-making support in reducing inappropriate prescribing in primary care. CMAJ 2003;169549- 556
PubMed
Beers  MH Explicit criteria for determining potentially inappropriate medication use by the elderly: an update. Arch Intern Med 1997;1571531- 1536
PubMed Link to Article

Figures

Place holder to copy figure label and caption
Figure 1.

Example screen shot of an alert for diazepam, a nonpreferred benzodiazepine.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.

Time series of nonpreferred (A) and preferred (B) agent dispensings per 10 000 by age. Fitted trend lines show predicted values from the segmented regression analysis.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 3.

Time series of the dispensings per 10 000 of the tricyclic antidepressants amitriptyline (A) and nortriptyline (B) by age. Fitted trend lines show predicted values from the segmented regression analysis.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable. Use of Nonpreferred and Preferred Medications at Baseline*

References

Grimshaw  JMRussell  IT Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations. Lancet 1993;3421317- 1322
PubMed Link to Article
Grimshaw  JMThomas  REMacLennan  G  et al.  Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol Assess 2004;8iii- iv, 1-72
PubMed
Gilbody  SWhitty  PGrimshaw  JThomas  R Educational and organizational interventions to improve the management of depression in primary care: a systematic review. JAMA 2003;2893145- 3151
PubMed Link to Article
McBride  PEPacala  JTDean  JPlane  MB Primary care residents and the management of hypercholesterolemia. Am J Prev Med 1990;671- 76
PubMed
Shiffman  RNLiaw  YBrandt  CACorb  GJ Computer-based guideline implementation systems: a systematic review of functionality and effectiveness. J Am Med Inform Assoc 1999;6104- 114
PubMed Link to Article
Balas  EABoren  SAGriffing  G Computerized management of diabetes: a synthesis of controlled trials. Proc AMIA Symp November1998;29- 5299
PubMed
Balas  EAWeingarten  SGarb  CTBlumenthal  DBoren  SABrown  GD Improving preventive care by prompting physicians. Arch Intern Med 2000;160301- 308
PubMed Link to Article
Elson  RBConnelly  DP Computerized patient records in primary care: their role in mediating guideline-driven physician behavior change. Arch Fam Med 1995;4698- 705
PubMed Link to Article
Rind  DMSafran  CPhillips  RS  et al.  Effect of computer-based alerts on the treatment and outcomes of hospitalized patients. Arch Intern Med 1994;1541511- 1517
PubMed Link to Article
Jha  AKKuperman  GJTeich  JM  et al.  Identifying adverse drug events: development of a computer-based monitor and comparison with chart review and stimulated voluntary report. J Am Med Inform Assoc 1998;5305- 314
PubMed Link to Article
Classen  DCPestotnik  SLEvans  RSBurke  JP Computerized surveillance of adverse drug events in hospital patients. JAMA 1991;2662847- 2851
PubMed Link to Article
Bates  DW Using information technology to reduce rates of medication errors in hospitals. BMJ 2000;320788- 791
PubMed Link to Article
Demakis  JGBeauchamp  CCull  WL  et al.  Improving residents' compliance with standards of ambulatory care: results from the VA Cooperative Study on Computerized Reminders. JAMA 2000;2841411- 1416
PubMed Link to Article
Abookire  SATeich  JMSandige  H  et al.  Improving allergy alerting in a computerized physician order entry system. Proc AMIA Symp November2000;2- 6
PubMed
Leape  LLBates  DWCullen  DJ  et al. ADE Prevention Study Group, Systems analysis of adverse drug events. JAMA 1995;27435- 43
PubMed Link to Article
Kaushal  RShojania  KGBates  DW Effects of computerized physician order entry and clinical decision support systems on medication safety: a systematic review. Arch Intern Med 2003;1631409- 1416
PubMed Link to Article
Bates  DWCullen  DJLaird  N  et al. ADE Prevention Study Group, Incidence of adverse drug events and potential adverse drug events: implications for prevention. JAMA 1995;27429- 34
PubMed Link to Article
Kaushal  RBates  DWLandrigan  C  et al.  Medication errors and adverse drug events in pediatric inpatients. JAMA 2001;2852114- 2120
PubMed Link to Article
Fick  DMCooper  JWWade  WEWaller  JLMaclean  JRBeers  MH Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Arch Intern Med 2003;1632716- 2724
PubMed Link to Article
Wagner  AKSoumerai  SBZhang  FRoss-Degnan  D Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 2002;27299- 309
PubMed Link to Article
Soumerai  SBMcLaughlin  TJRoss-Degnan  DCasteris  CSBollini  P Effects of a limit on Medicaid drug-reimbursement benefits on the use of psychotropic agents and acute mental health services by patients with schizophrenia. N Engl J Med 1994;331650- 655
PubMed Link to Article
Soumerai  SBRoss-Degnan  DGortmaker  SAvorn  J Withdrawing payment for nonscientific drug therapy: intended and unexpected effects of a large-scale natural experiment. JAMA 1990;263831- 839
PubMed Link to Article
Gillings  DMakuc  DSiegel  E Analysis of interrupted time series mortality trends: an example to evaluate regionalized perinatal care. Am J Public Health 1981;7138- 46
PubMed Link to Article
Wagner  AKZhang  FSoumerai  SB  et al.  Benzodiazepine use and hip fractures in the elderly: who is at greatest risk? Arch Intern Med 2004;1641567- 1572
PubMed Link to Article
Lau  DTKasper  JDPotter  DELyles  ABennett  RG Hospitalization and death associated with potentially inappropriate medication prescriptions among elderly nursing home residents. Arch Intern Med 2005;16568- 74
PubMed Link to Article
Tamblyn  RHuang  APerreault  R  et al.  The medical office of the 21st century (MOXXI): effectiveness of computerized decision-making support in reducing inappropriate prescribing in primary care. CMAJ 2003;169549- 556
PubMed
Beers  MH Explicit criteria for determining potentially inappropriate medication use by the elderly: an update. Arch Intern Med 1997;1571531- 1536
PubMed Link to Article

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 56

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles