Improving prescribing using a rule-based prescribing system C Anton 1, PG Nightingale 2, D Adu 3, G...
1
Improving prescribing using a rule-based prescribing system C Anton 1 , PG Nightingale 2 , D Adu 3 , G Lipkin 3 , RE Ferner 1 1. West Midlands Centre for Adverse Drug Reaction Reporting, City Hospital, Dudley Road, Birmingham B18 7QH 2. Wolfson Computer Laboratory, Queen Elizabeth Medical Centre, Edgbaston, Birmingham B15 2TH 3. Renal Unit, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH Introduction Difficulties can arise at any part of the prescription process from the moment the prescriber makes a choice of drug treatment to the time the patient receives that treatment. Medication errors are very common, however they are defined, 1 and in many instances avoidable. 2 Illegible prescriptions are one cause of avoidable medication errors, and electronic prescription systems are increasingly being introduced in order to remove this danger. 3 They can also act as 'expert' systems, and so prevent other drug errors, for example, from drug interactions; and they can enforce local prescribing rules. Computer-based prescribing systems have been shown to reduce drug errors. 4 Nature of the problem We used the data from one expert computer prescribing system in use in the Renal Unit at the Queen Elizabeth Hospital in Birmingham to test the following hypotheses: 5 • an intervention by an expert computer prescribing system improves the prescribers' future prescribing and so doctors will learn over time to avoid errors; • more senior doctors are more likely to disregard warnings; • warning messages following a prescription are less common for more senior doctors, and more common as the workload of the renal unit (measured by the number of patients) increases. We collected data between 1 st July and 31 st August 2001 at the 64 bedded Renal Unit of the Queen Elizabeth Hospital, Birmingham. During this time there were 5,995 individual prescriptions filed by 42 doctors and these generated 51,612 records of prescriptions and administrations in the system. Prescription process on the Unit Numbers and types of doctors and other users who used the system during July and August 2001 Results There was an association between numbers of warning messages and the grade and also with the amount of experience they had had in using it. The data are shown in the Tables Comparison between the warning messages per prescription generated by experienced doctors who stopped using the system in July and those inexperienced ones who started using the system in August Warning message rates calculated as the number of warning messages generated per prescription for 6 doctors who started in the unit in August 2001 compared with 3-4 weeks later Discussion Doctors who were experienced in the use of the system and who finished working on the unit in July were much less likely to generate a warning message when compared with the six doctors who were new to the system in August. However, the new doctors rapidly improved their prescribing, as judged by the number of warning messages per prescription. We have not examined in this study the number of non- intercepted medication errors that continued to occur, but we have examined the way in which the doctor’s prescribing behaviour was modified by interaction with an expert computerized prescribing system. Since the computer generates warning messages when errors are made in prescribing, a reduction in the References 1. Ferner RE, Aronson JK. Errors in prescribing, preparing, and giving medicines: definition, classification, and prevention. In: Aronson JK, ed. Side effects of drugs annual. 22nd ed. Amsterdam: Elsevier, 1999: xxiii-xvi. 2. Dean,B; Schnachter M; Vincent C; Barber N. Causes of prescribing errors in hospital inpatients: a prospective study. Lancet 2002; 359: 1373-8. 3. Nightingale PG, Adu D, Richards NT, Peters M. Implementation of rules based computerised bedside prescribing and administration: intervention study. BMJ 2000; 320: 750-3. 4. Bates DW, Leape LL, Cullen DJ, Laird N, Petersen LA, Teich JM, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA 1998; 280: 1311-6. 5. Anton C, Nightingale PG, Adu D, Lipkin G, Ferner RE. Improving prescribing using a rule-based prescribing system. Qual Saf Health Care 2004 (in press). G rade (n) N o of scripts TotalN o of warning m essages N o of warning m essages disregarded (%) warning m essages/ perscription N o of interaction or renalfailure warning m essages Consultant(9) 382 697 372 (53) 1.82 403 (79) Registrar(13) 2436 2388 1323 (55) 0.98 1030 (92) Seniorhouse officer(6) 1057 1157 537 (46) 1.09 490 (86) Pre-registration house officer(14) 1643 1917 960 (50) 1.17 801(80) Non-doctor (61) 477 433 242 (56) 0.91 D octor 1st7 days use of system 24th -30th August A 1.27 0.01 B 1.48 0.41 C 0.41 0.14 D 1.38 0.17 E 0.4 0.23 F 0.36 0.13 G roup N o of prescriptions N o of warning m essages N o (% )of warning m essages disregarded Warning m essages/ prescription July cohort (3 SH O s,5 PRH O s) 1515 1417 676 (48) 0.94 A ugustcohort (2 SH O s,4 PRH O s) 1170 1643 817 (50) 1.4 51 6 12 co n cernin g p rescrip tio n s a n d ad m in istratio n s reco rd s 15 8 53 d isp layed m essages 4 121 In fo rm atio n al m essages 5 14 0 System m essages 6 15 9 (+ 4 33* ) W arn in g m essages Disregarded in our analysis A cce p t w a rn in g and abandon p rescrip tion 29 6 7 (+ 19 1*) D isreg ard w a rn in g and continue with p rescrip tio n 319 2 (+ 242*) * numbers in parentheses refer to messages for nurses Nurse or doctor logs onto system to prescribe or administer drugs Message is displayed in accordance with the system’s rules Doctor or nurse considers the message displayed Doctor (or nurse) decides to heed or override any warning message We imported these data into a Microsoft Access database and analysed them to test the hypotheses in the following ways: • we compared a cohort of doctors who were experienced in the use of the system and who finished working on the unit in July 2001 with a cohort who started on the unit in August 2001 and had no prior experience of the unit. • we compared the rate of warning messages and the proportion that were "disregarded" for the August cohort during their first week on the unit with their fourth week at the end of August; • we examined the rate of warning messages and the proportion that were "disregarded" by grade of doctor for the entire sample over the two months; • we examined the rate of warning messages and the proportion that were "disregarded" by the number of patients on the unit as a surrogate of workload; and • we looked at the most commonly occurring warning messages and interaction warnings. number of warning messages equates to improved — that is, safer — prescribing. We do not know whether improved prescribing behaviour persists after doctors move to areas where there is no computerized prescribing. This is a potentially important question to answer. We conclude that clinical staff rapidly adapt to computer prescribing, and their prescribing behaviour is modified to reduce the number of warning messages of serious danger displayed by the system. Provided the rules governing warning messages are carefully constructed, the alignment of doctors' prescribing practice with the rules should improve patient safety.
-
Upload
emily-mclaughlin -
Category
Documents
-
view
213 -
download
1
Transcript of Improving prescribing using a rule-based prescribing system C Anton 1, PG Nightingale 2, D Adu 3, G...
Improving prescribing using a rule-based prescribing system
C Anton1, PG Nightingale2, D Adu3, G Lipkin3, RE Ferner1
1. West Midlands Centre for Adverse Drug Reaction Reporting, City Hospital, Dudley Road, Birmingham B18 7QH2. Wolfson Computer Laboratory, Queen Elizabeth Medical Centre, Edgbaston, Birmingham B15 2TH3. Renal Unit, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH
Introduction
Difficulties can arise at any part of the prescription process from the moment the prescriber makes a choice of drug treatment to the time the patient receives that treatment. Medication errors are very common, however they are defined,1 and in many instances avoidable.2 Illegible prescriptions are one cause of avoidable medication errors, and electronic prescription systems are increasingly being introduced in order to remove this danger.3 They can also act as 'expert' systems, and so prevent other drug errors, for example, from drug interactions; and they can enforce local prescribing rules. Computer-based prescribing systems have been shown to reduce drug errors.4
Nature of the problemWe used the data from one expert computer prescribing system in use in the Renal Unit at the Queen Elizabeth Hospital in Birmingham to test the following hypotheses:5
• an intervention by an expert computer prescribing system improves the prescribers' future prescribing and so doctors will learn over time to avoid errors;
• more senior doctors are more likely to disregard warnings;
• warning messages following a prescription are less common for more senior doctors, and more common as the workload of the renal unit (measured by the number of patients) increases.
We collected data between 1st July and 31st August 2001 at the 64 bedded Renal Unit of the Queen Elizabeth Hospital, Birmingham. During this time there were 5,995 individual prescriptions filed by 42 doctors and these generated 51,612 records of prescriptions and administrations in the system.
Prescription process on the Unit
Numbers and types of doctors and other users who used the system during July and August 2001
ResultsThere was an association between numbers of warning messages and the grade and also with the amount of experience they had had in using it. The data are shown in the Tables
Comparison between the warning messages per prescription generated by experienced doctors who stopped using the system in July and those inexperienced ones who started using the system in August
Warning message rates calculated as the number of warning messages generated per prescription for 6 doctors who started in the unit in August 2001 compared with 3-4 weeks later
DiscussionDoctors who were experienced in the use of the system and who finished working on the unit in July were much less likely to generate a warning message when compared with the six doctors who were new to the system in August. However, the new doctors rapidly improved their prescribing, as judged by the number of warning messages per prescription. We have not examined in this study the number of non-intercepted medication errors that continued to occur, but we have examined the way in which the doctor’s prescribing behaviour was modified by interaction with an expert computerized prescribing system. Since the computer generates warning messages when errors are made in prescribing, a reduction in the References1. Ferner RE, Aronson JK. Errors in prescribing, preparing, and giving medicines: definition, classification, and
prevention. In: Aronson JK, ed. Side effects of drugs annual. 22nd ed. Amsterdam: Elsevier, 1999: xxiii-xvi.
2. Dean,B; Schnachter M; Vincent C; Barber N. Causes of prescribing errors in hospital inpatients: a prospective study. Lancet 2002; 359: 1373-8.
3. Nightingale PG, Adu D, Richards NT, Peters M. Implementation of rules based computerised bedside prescribing and administration: intervention study. BMJ 2000; 320: 750-3.
4. Bates DW, Leape LL, Cullen DJ, Laird N, Petersen LA, Teich JM, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA 1998; 280: 1311-6.
5. Anton C, Nightingale PG, Adu D, Lipkin G, Ferner RE. Improving prescribing using a rule-based prescribing system. Qual Saf Health Care 2004 (in press).
Grade (n)No of scripts
Total No of warning messages
No of warning messages
disregarded (%)
warning messages/
perscription
No of interaction or renal failure
warning messages
Consultant (9) 382 697 372 (53) 1.82 403 (79)
Registrar (13) 2436 2388 1323 (55) 0.98 1030 (92)Senior house officer (6) 1057 1157 537 (46) 1.09 490 (86)Pre-registration house officer (14) 1643 1917 960 (50) 1.17 801 (80)
Non-doctor (61) 477 433 242 (56) 0.91
Doctor 1st 7 days use of system
24th - 30th August
A 1.27 0.01
B 1.48 0.41
C 0.41 0.14
D 1.38 0.17
E 0.4 0.23
F 0.36 0.13
GroupNo of
prescriptions
No of warning messages
No (%) of warning messages
disregarded
Warning messages/
prescription
July cohort(3 SHOs, 5 PRHOs)
1515 1417 676 (48) 0.94
August cohort(2 SHOs, 4 PRHOs)
1170 1643 817 (50) 1.4
51 612 concerning prescriptions and administrations
records
15 853 d isplayedm essages
4121 Info rmational messages
5140System messages
6159 (+433* )W arning messages
Disregarded in our analysis
A ccept w arning and abandon prescription
2967 (+191* )
Disregard w arning and continue w ith prescription
3192 (+242* )
* numbers in parentheses refer to messages fo r nurses
Nurse o r docto r logs o nto system to prescribe o r administer d rugs
M essage is d isp layed in accordance w ith the system’s ru les
Do cto r o r nurse co nsiders the message d isplayed
Do cto r (o r nurse) decides to heed or o verride any w arning message
We imported these data into a Microsoft Access database and analysed them to test the hypotheses in the following ways:
• we compared a cohort of doctors who were experienced in the use of the system and who finished working on the unit in July 2001 with a cohort who started on the unit in August 2001 and had no prior experience of the unit.
• we compared the rate of warning messages and the proportion that were "disregarded" for the August cohort during their first week on the unit with their fourth week at the end of August;
• we examined the rate of warning messages and the proportion that were "disregarded" by grade of doctor for the entire sample over the two months;
• we examined the rate of warning messages and the proportion that were "disregarded" by the number of patients on the unit as a surrogate of workload; and
• we looked at the most commonly occurring warning messages and interaction warnings.
number of warning messages equates to improved — that is, safer — prescribing.
We do not know whether improved prescribing behaviour persists after doctors move to areas where there is no computerized prescribing. This is a potentially important question to answer. We conclude that clinical staff rapidly adapt to computer prescribing, and their prescribing behaviour is modified to reduce the number of warning messages of serious danger displayed by the system. Provided the rules governing warning messages are carefully constructed, the alignment of doctors' prescribing practice with the rules should improve patient safety.
![Time Temperature Effects in the Nonintegrable SU(3) Lipkin ...€¦ · context of an SU(3) version of the Lipkin Meshkov Glick1 model [20, 21]. This model was proposed originally](https://static.fdocuments.in/doc/165x107/6083bfa75fd590791160f5d8/time-temperature-effects-in-the-nonintegrable-su3-lipkin-context-of-an-su3.jpg)
