Annals of African Medicine Vol. 12, No. 4; 2013
Access this article online Quick Response Code:
Original Article
Website: www.annalsafrmed.org DOI: 10.4103/1596-3519.122691 PMID: ******
A qualitative study of causes of prescribing errors among junior medical doctors in a Nigeria in‑patient setting Adetutu A. Ajemigbitse, Moses K. Omole1, Ogugua F. Osi‑Ogbu2, Wilson O. Erhun3 Department of Pharmacy, National Hospital Abuja, 1Department of Clinical Pharmacy and Pharmacy Administration, University of Ibadan, Ibadan, 2Department of Medicine, National Hospital Abuja, Federal Capital Territory, 3 Department of Clinical Pharmacy and Pharmacy Administration, Obafemi Awolowo University, Ile‑Ife, Nigeria
Correspondence to: Ajemigbitse Adetutu A, Department of Pharmacy, National Hospital Abuja, Nigeria. E‑mail: [email protected]
Abstract Aims: The aims of this study were to identify and understand the factors underlying prescribing errors in order to determine how to prevent them. Materials and Methods: A prospective qualitative study that involved face-to-face interviews and human factor analysis in a Tertiary Referral Hospital in Central Nigeria, from July 2011 to December 2011. Pharmacists in the study hospital prospectively reviewed prescription orders generated by doctors in selected wards (male and female medical, pediatric and the private wing wards) and identified prescribing errors. The 22 prescribers involved in the errors were interviewed, and given questionnaires to discover factors causing the errors. A model of human error theory was used to analyze the responses. Results: Responses from the doctors suggest that most errors were made because of slips in attention. Lack of drug knowledge was not the single causative factor in any incident. Risk factors identified included individual, team, environment, and task factors. Junior doctors were affected by the prescribing habits of their seniors. Organizational factors identified included inadequate training/experience, absence of reference materials and absence of selfawareness of errors. Defenses against error such as other clinicians and guidelines were absent or deficient, and supervision was inadequate. Conclusions: To reduce the risk of prescribing errors, a number of strategies addressing individual, task, team, and environmental factors such as training of junior doctors, enforcing good practice in prescription writing, supervision, and reviewing the workload of junior doctors must be established. Keywords: In-patients, medication errors, patient safety, physicians, prescriptions
Résumé Objectifs: Les objectifs de cette étude étaient d'identifier et de comprendre les facteurs qui sous-tendent les erreurs thérapeutiques afin de déterminer comment les prévenir. Matériel et méthodes: Une étude qualitative prospective qui implique des entretiens en face-à-face et analyse de facteur humaindans un hôpital de référence tertiaire dans le centre du Nigéria, de juillet à décembre 2011. Pharmaciens à l'hôpital de l'étude prospective examiné des arrêtés de prescription générées par les médecins dans des quartiers sélectionnés (mâle et femelle médical, pédiatrique et les salles de l'aile privée) et identifié des erreurs de prescription. 22 Prescripteurs impliqués dans les erreurs ont été interrogés et donnés des questionnaires pour découvrir les facteurs causant les erreurs. Un modèle de la théorie de l'erreur humaine a été utilisé pour analyser les réponses. Résultats: Les réponses des médecins suggèrent que la plupart des erreurs ont été faites en raison de la glisse dans
Annals of African Medicine
Vol. 12, October-December, 2013
Page | 223
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients l'attention. Manque de connaissances sur le médicament n'était pas le seul facteur causal dans n'importe quel incident. Facteurs de risque identifiés inclus individu, équipe, environnement et facteurs de la tâche. Médecins en formation ont été touchés par les habitudes de prescription de leurs aînés. Facteurs organisationnels identifiés inclus insuffisante formation/expérience, absence de documents de référence et absence de conscience de soi des erreurs. Défenses contre toute erreur, tels que les autres cliniciens et les lignes directrices étaient absents ou déficient, et supervision était inadéquate. Conclusions: Pour réduire le risque de prescription des erreurs, un certain nombre de stratégies individuelles, les tâches et des facteurs environnementaux tels que la formation des médecins en formation, faire respecter les bonnes pratiques dans la prescription d'écriture, supervision et examiner la charge de travail des médecins en formation doit être établi. Page | 224
Mots-clés: Patients hospitalisés, les erreurs de médication, la sécurité des patients, médecins, prescriptions
Introduction A prescribing error is defined as a prescribing decision or prescription writing process that results in an unintentional, significant reduction in the probability of treatment being timely and effective or increases the risk of harm when compared with generally accepted practice.[1] Prescribing without taking into account the patient’s clinical status, failure to communicate essential information and transcription errors were considered prescribing errors. Errors made during drug prescription are the most common type of avoidable medication errors and hence are an important target for improvement.[2,3] Many factors have been associated with prescribing errors.[3‑12] Understanding these factors should assist in the implementation of effective strategies for error prevention.[12,13] The Reason accident causation,[14] frame‑work is a popular theoretical frame‑work used to study the causes of errors and accidents in various industries. This approach has been practically related to prescribing errors,[3,4,12,15] and constitutes the theoretical basis of this study. Past studies on medication prescribing revealed poor compliance with the principles of rational medication use.[16‑19] This study seeks to investigate the causes of prescribing errors among junior medical doctors in a Tertiary Hospital in Nigeria so as to provide a baseline for future intervention especially as there are no such previous studies in Nigeria.
Materials and Methods Study participants and setting This study prospectively reviewed prescriptions generated by medical doctors in selected wards (male and female medical, pediatric and the private wing wards) at the National Hospital Abuja, Nigeria. This hospital, though currently a 200‑bed Tertiary Referral Hospital, plays a key role in the provision of health Vol. 12, October-December, 2013
services to the residents of Abuja, a cosmopolitan multi‑ethnic, multi‑cultural urban city with high socio‑economic disparity. In addition, it serves as an institution for training of health personnel offering undergraduate, graduate, post‑graduate, and residency training programs. The medical staff structure in the hospital include the house officers (HO) who are the newly qualified doctors on the mandatory 1‑year internship training, the National Youth Service Corp doctors, the junior and senior Residents, medical Consultants, Visiting Professors, and other medical personnel on research activities and sundry. Total out‑patient attendance in the year 2010 was over 40,000 patients while the turn‑over of in‑patient admissions to the medical and pediatric wards was approximately 1900 patients for the same year. The hospital operates the typical Nigeria pharmacy service. Briefly, this involves prescriber’s hand‑writing medication orders into the patient’s medical notes as well as on formatted treatment sheets. These orders are taken to the pharmacy units where pharmacists check that the orders are clear, clinically appropriate and valid before initiating the supply of any drugs to patients in the wards. These same documents are used by the nursing staff to determine the doses due at each medication round and to record their administration. The doctors involved in the selected units were informed about the study and its goal. The goal was to understand the causes of prescribing errors with a view to developing specific interventions that will improve practice with ultimate improvement in patient’s safety. The local Ethics Committee approved the study and all prescribers interviewed gave written consent. Medication orders were screened for prescribing errors that met the study definition.[1] Data were collected on 5‑chosen days in a month, for a 6 months period between July 2011 and December 2011. Only prescribers who were involved in error incidents and contactable within 72 h of the incident were interviewed to ensure adequate recall. Semi‑structured Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
face‑to‑face interviews, using a questionnaire, were conducted to determine the causes of prescribing errors. Causes were defined as “reasons reported to the researchers by the prescriber as being wholly or partially responsible for a specific prescribing error.” Interviews took between 15 min and 25 min and the process was adapted from the methods of other researchers.[3,4,12,15] Table 1 outlines the process.
lapses are errors due to omission of a particular task and slips are errors owing to attentional failure when performing a task or an active failure resulting from the incorrect execution of a task. Mistakes were errors occurring from correct execution of an incorrect or inappropriate plan and were either rule‑based (RBM) or knowledge‑based mistakes (KBM).
Definitions
Error‑producing conditions are related to the task and the environment at the time when the error occurred. They are underlying risk factors and can be grouped as connected to the individual prescriber, the working condition, the health‑care team, the prescribing task and patient. Latent conditions are organizational processes that create an environment where error‑producing conditions and active‑failures are more likely to result in prescribing errors.
Active failures are the unsafe acts committed by the prescribers in contact with the patient. Memory Table 1: Interview schedule with prescribers Background The prescribers were asked to give a brief introduction of themselves, medical school attended, how long they had worked at the hospital, how long in post and what specialty Provide a brief description of type of teaching or learning courses that prepared them for prescribing, how this was taught and if they thought this was adequate The prescribing errors (areas covered) The type of error made Dosing error, omission of information, frequency error, drug interactions etc. The medication involved The condition being treated The error situation Time of day Type of ward Type of patient How long at the ward A description of what happened Work load Supervision Stage of patient stay The reason for making the error Their attitude towards the error Has this happened before? Has this happened since? How does this make you feel? What do you think could have helped to prevent the error? How will this change the way you prescribe? Closing The interviews ended with asking the prescribers if they had any questions, comments or suggestions for the interviewers
/DWHQWFRQGLWLRQV 2UJDQL]DWLRQDOSURFHVVHV 0DQDJHPHQWGHFLVLRQV
(UURU±SURGXFLQJ FRQGLWLRQV (QYLURQPHQWDOWHDP LQGLYLGXDOWDVNRUSDWLHQW IDFWRUV
Violations are active choices by prescribers to ignore the formal or informal policies or guidelines they are expected to adhere to. Defenses are designed to protect against hazards and mitigate consequences of failure.
Analysis Notes taken during the interviews were transcribed afterward and reviewed by the interviewers (a consultant physician and a clinical pharmacist). Common themes relating to the reporting of errors were identified and coded manually based on the interviewee’s words and phrases. Consensus was reached through discussion. The result was presented according to Reason’s model of accident causation [Figure 1].
Results A total of 90 errors involving 37 doctors were reported. In 15 cases, the prescribers could not be interviewed because, 7 were unidentified, three were reported more than 72 hoursh after the event, and in 5 cases the interviews could not hold within the 72 hoursh target period.
$FWLYHIDLOXUHV (UURUVVOLSVODSVHV PLVWDNHV 9LRODWLRQV
$FFLGHQW
'HIHQFHV ,QDGHTXDWH PLVVLQJRU XQDYDLODEOH Figure 1: Reason’s model of accident causation
Annals of African Medicine
Vol. 12, October-December, 2013
Page | 225
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 2: Prescribing errors and analysis by themes from interviews with prescribers Stage of error On admission 1. IVF Dextrose 5%, Cinnarizine (Stugeron©), tramadol During stay 2. Lisinopril Page | 226
3. Tramadol 4. Amoxicillin+clavulanic acid, soluble insulin, IVF normal saline 5. Erythromycin, metronidazole, amoxicillin+ clavulanic acid 6. Ceftriaxone 7. Ceftriaxone, ranitidine 8. Tramadol 9. Insulin
Context of error
Prescriber Directed Prescribed Themes underlying error by senior before?
Duration/stop‑date Registrar omitted, dose inappropriate
No
Yes
Training, chore, supervision, absence of guidelines
Dose inappropriate
Yes
Yes
Dose adjustment in renal impairment Duration of therapy omitted+unsafe abbreviation Duration of therapy omitted House officer
Yes
No
Yes
Yes
Multitasking, supervision, absence of error awareness Supervision, communication, training Training, absence of guidelines, supervision
Yes
Yes
Training, supervision, communication, absence of guidelines
Dose adjustment in renal impairment; stop‑date omitted Duration of therapy omitted, need for therapy omitted Transcription error, route of administration omitted
House officer
Yes
Yes
Lack of knowledge, training, supervision, communication
House officer
Yes
No
Training, supervision, absence of error awareness
House officer
Yes
No
Unsafe abbreviation
House officer House officer House officer
Yes
Yes
Yes
No
Yes
No
House officer Registrar
Yes
Yes
Yes
No
House officer House officer
Yes
No
Absence of error awareness, lack of knowledge, supervision, inattention Training, supervision, absence of guidelines Lack of knowledge, supervision Lack of knowledge, absence of reference material, supervision, communication Communication, supervision, trust in seniors Absence of reference books, co‑managed patient, communication Trust in senior, nurse defense over-ridden, supervision Absence of error awareness, lack of knowledge of patient, chore, supervision Supervision, rushed, training, lack of knowledge Lack of knowledge, training, supervision, chore, co‑managed patient Communication, absence of error awareness, multitasking, supervision Supervision/responsibility, absence of error awareness Lack of knowledge, communication, supervision, chore Training, supervision, chore
10. Domperidone, nifedipine+atenolol 11. Nifedipine+nimodipine, enoxaparin
Dose inappropriate, drug‑drug interaction Need for drug therapy omitted; Drug-disease contraindication 12. Amoxicillin+clavulanic Dose inappropriate acid 13. Carbamazepine+ Drug‑drug interaction amitriptiline
House officer House officer House officer
14. Aciclovir
Dose inappropriate
15. Ceftriaxone, metronidazole, gentamycin 16. Metoclopramide, hyoscine butyl bromide 17. Risperidone, phenytoin, metronidazole
Dose inappropriate, duration/stop‑date omitted, transcription error Route of administration House omitted, inappropriate dose officer Drug‑drug interaction House officer
Yes
Yes
Yes
No
Yes
No
18. Imipenem
Dose inappropriate
House officer
Yes
No
19. Diclofenac, warfarin, enoxaparin 20. Rabeprazole+ omeprazole
Drug‑drug interaction
Registrar
No
Yes
Duplication, need for drug omitted
House officer
Yes
No
House officer House officer
Yes
Yes
Yes
No
21. Insulin, metronidazole Unsafe abbreviation, duration/stop‑date omitted 22. Naproxen+aspirin+ Drug‑drug interaction tramadol
Lack of knowledge, training, supervision
IVF=Intravenous fluid
A total of 49 error incidents, involving 22 doctors were analyzed. Prescribers involved were 19 HO and 3 Registrars while 21 of the incidents were Vol. 12, October-December, 2013
medical cases. In all 22 cases, the prescribers were unaware that an error, lapse or mistake had occurred. Of the 22 incidents analyzed [Table 2], 1 occurred Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 3: Examples of active failures Errors 1. Slips “I was writing a prescription for IV frusemide 120 mg b.d. when a nurse talked to me…I can’t really remember what happened to have caused the mix-up but I was trying to catch up with the team. I don’t know how I mixed up the dose meant for frusemide for lisinopril”. (Interview 1, house officer, researcher’s notes) 2. Lapses “I am not used to giving tramal (tramadol) as an injection… I had to ask the SR, and he said injection, but I didn’t confirm the route. But I can remember the prescription clearly. He said 100 mg stat and 50 mg t.d.s po for 3 days. Two of us were taking the instruction. I wrote in the folder and my colleague in the treatment sheet”. (Interview 7, house officer, researcher’s notes) 3. Mistakes Some mistakes resulted from an absence of knowledge of correct dose, or relevant rules “I don’t know exactly where this error is from. I can’t say whether the child was on 350 mg IV ceftriazone elsewhere and I copied the prescription… IV ceftriazone is given at a dose of 50-75 mg most times but in meningitis can go as high as 100 mg/kg. That’s what my Reg says…I think in this case calculating my own dose would have helped, but if the managing team has a reason for a different dose, I may have to know why…” (Interview 14, house officer, researcher’s notes) When asked to explain his mistake a HO did not know that ClexaneTM is the brand name for enoxaparin. “The patient was thought to have pulmonary embolism in addition to his stroke…It was agreed it won’t be appropriate to give this patient clexane because of his internal bleeding… I am not aware that enoxaparin is the active ingredient in clexane.” (Interview 9, house officer, researcher’s notes) “This patient was formerly on nifedipine 20 mg b.d. and we later increased it to 30 mg (with improvement in the blood pressure control). Then nimodipine was added. My SR said nimodipine was for the headache not as an anti‑hypertensive. Later, I called him to confirm the dose and he said we should give 50 mg q6rs ×1/52… I also thought about the drug being a CCB (same class as nifedipine), but my SR said it was ok…… We felt the drug (nimodipine) was needed because the patient was complaining of headache and stiffness around the neck so we thought of sub‑arachnoid hemorrhage.” (Patient with right hemispheric hemorrhagic stroke) (Interview 8, house officer, researcher’s notes) SR=Senior registrar, Reg=Registrar, NS=Normal saline, F/S Darrows=Full strength darrow’s solution, IU=International units, CCB=Calcium-channel blocker, Frusemide=Furosemide
on admission to the hospital and 21 during the in‑patient stay. Antibacterials, analgesics, and cardio‑vascular system drugs accounted for over half the errors. HO wrote most of the prescriptions although they were directed by their seniors. Sometimes, the HO determined the form, dose, route, and frequency. Annals of African Medicine
Main factors involved with all prescribing errors are individual (16, 72.7%), team (14, 63.6%) and work environment (9, 40.9%) factors.
Active failures All errors occurred as a result of at least one active failure. Mistakes were frequent [Table 3]. All the prescribers interviewed were unaware of the slip or lapse, but explained that they were rushing (6, 27%) or distracted during routine tasks (5, 23%). Common mistakes were errors of inappropriate dosing (10, 45%) and failure to check for potentially serious drug interactions (5, 23%). A common cause of RBM was the lack of knowledge of a relevant rule for example dose adjustment for patients with renal impairment (3, 14%). Prescribing two or more drugs that could result in potentially serious interaction was also noted, e.g., carbamazepine and amitryptilline; phenytoin and metronidazole. In the former case, the patient was being co‑managed by the psychiatry and medical teams. There were instances where the correct rule was wrongly applied. For example, intravenous injection ceftriaxone dosing was misinterpreted as 20‑50 mg/kg twice daily instead of 20‑50 mg/kg daily in two divided doses. The prescriber multiplied the dose provided by the patient’s weight and prescribed this amount to be given twice daily, resulting in a two‑fold overdose. Another active failure showing lack of knowledge or experience (KBM) involved the writing of “IU” (short for international units) as an abbreviation for “units” when prescribing insulin (3, 14%). Even though, they agreed to the possibility of the orders being erroneously misread or misinterpreted leading to a ten‑fold error in dose, junior doctors claimed to be influenced by the practices of their senior colleagues in this. One interviewee stated that she tended to use an abbreviation of drug names (e.g., Full strength Darrow’s solution instead of full strength Darrow’s solution, NS for normal saline) as a way to save time while writing.
Error‑producing conditions Many factors were cited as contributory to the errors, the most frequent being work environment (11, 50%), individual (10, 45%) and team (8, 36%), [Table 4]. In 10 instances, high workload, resulting in multitasking was thought to have contributed to the error; in 5 instances the physical environment was cited. The terms “hectic”, “busy”, stressful’ were used to describe workload. Sometimes, the work situation made the junior doctors rush their prescribing and other duties in order to catch up with their team. Part of the pressure was to get the prescriptions written on time for the ward attendants to take the treatment sheets Vol. 12, October-December, 2013
Page | 227
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 4: Examples of error‑producing conditions
Page | 228
Environment factors “There was a lot of distraction in the ward… (Interview 2, house officer, researcher’s notes) Individual factors “When more than one person talks to me at the same time, I tend to get distracted…I can get them to please pause and take it one‑by‑one. Furthermore, if my Reg is too fast, I can slow him down so I get it”. (Interview 2, house officer, researcher’s notes) “The metoclopramide was prescribed earlier and that had the route of administration, dose, etc. So I omitted writing that (route of administration) but it would have been better to have included the route. I may have been rushing to write the prescription before it goes to the pharmacy.” (Interview 15, house officer, researcher’s notes) Team factors “It was a team decision, I was just the writer. The prescription was not written the same day. The carbamazepine was for prophylaxis because the patient had seizures in the past and amitryptilline was prescribed 2 days ago because the patient was manifesting symptoms of depression… It must have been an oversight because such (interaction between carbamazepine and amitryptilline) had been mentioned to us before. But, like I said, I was the writer; the decision wasn’t taken by me”. (Interview 11, registrar, researcher’s notes) Task factors “I discussed with my SR and he made a suggestion of not using the commonly used NSAIDs. He suggested the tramadol. We made a change on this to DF118 due to intractable vomiting. I omitted to write this (the substitution) in the patient’s folder.”(Interview 3, house officer, researcher’s notes) Patient factors Patient being co-managed by two different specialties “The dose of phenytoin was initiated at the A and E to treat seizures. It was 100 mg b.d. Risperidone was later ordered by the psychiatrist, but I wrote it down. I had a thought about it (the combination), but being a house officer, I can’t question a consultant’s order. (Interview 17, house officer, researcher’s notes) NSAIDs=Non-steroidal anti-inflammatory drugs, DF 118=Dihydrocodeine tartrate tablets, A and E=Accident and emergency unit, SR=Senior registrar
to pharmacy so that any medications requiring compounding would be presented in time before the 4.00 pm shift was over (In the hospital, morning work shift begins at 8.00 am until 4.00 pm). Individual factors mentioned in connection with prescribing errors included physical and mental well‑being and lack of knowledge. A total of 14 respondents mentioned tiredness (50%), distraction (29%), low morale (21%), and confusion (14%) as factors that may have caused the error. An absence of knowledge about dose and no prior experience of prescribing some drugs were contributory to errors. In 10 instances, junior doctors wrote prescription for drugs without indicating the route of administration and frequency of administration, claiming they had Vol. 12, October-December, 2013
never prescribed the drug before or did not know the dose of the drug or the correct frequency of administration. In 9 instances, the HOs reported not having received any training in prescribing as undergraduates and five reported minimal training. Team factors, such as communication, responsibility, and supervision were associated with 12 incidents. Junior doctors commonly referred to communication about medications with words such as “I wrote what was dictated to me by the Senior Reg,” “I was told by the Reg to leave it open.” Some junior doctors felt incompetent to disagree with the decisions of the seniors, did not feel it was correct to have a differing opinion from their superiors or even ask questions when they were not sure of the correct dose assuming that to do so was to expose ignorance. In an instance, when acyclovir was prescribed by a HO for the 1st time, an error in the frequency of administration was made. When interviewed he claimed to have called the attention of the registrar to what he had written and because the registrar had accented to this, he did not take any further action. Junior doctors also felt that if a problem occurred, the seniors should take responsibility as the prescriptive authority. Other causes included lack of documentation of the prescribed medicine in the patient’s notes. In some instances, supervision was inadequate as some senior doctors omitted to cross‑check what the junior ones prescribed or did. Task and patient factors were mentioned as possible causes of errors. Task factors comprised the unavailability of drug reference materials such as the Hospital Formulary or Standard Treatment Guidelines or similar document to consult at the point of prescribing. Inadequate patient information (e.g., weight) also made for dose calculation errors especially in pediatric patients where doses are usually calculated based on a patient’s weight. The absence of these documents meant they had to depend on other sources for information on doses, or copy from the senior’s prescribing‑whether correct or not. In 3 instances, the patient was mentioned as a contributing factor; 2 were being managed by two different specialist teams while the third case was a complex one.
Latent underlying conditions Lack of training in prescribing skills and insufficient knowledge and experience about drugs were latent factors [Table 5]. The junior doctors were left to fill all the details of strength, dose, form, frequency, route, duration, and direction after verbal instructions were given. Usually, these orders were not cross‑checked by the seniors. Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 5: Examples of latent conditions and defenses Latent conditions When asked what the prescriber felt would help in preventing a repeat of the dosing error made “Providing a small booklet of medicine information or drug reference such as EMDEX or the hospital’s formulary will be helpful.” (Interview 9, house officer, researcher’s notes) “I have had no prior training on drug prescribing. Just pharmacology where we were taught about pharmacology of drugs, dose, side‑effects etc. (Interview 14, house officer, researcher’s notes) Violations “I remember writing the prescription. We were rushing to the next patient… two nurses were talking to me; the Reg was waiting for me to write, so I got distracted. I was rushing to write fast. I use abbreviation to make me write faster… (Abbreviations such as NS, F/S Darrows, insulin 8 IU) (Interview 2, house officer, researcher’s notes) “Most times our seniors tell us to prescribe (an antibacterial) and to leave it open. I prescribe that way because in most cases duration (of therapy) is not given… I tend not to write the duration because that’s the way I’m told.” (Interview 3, house officer, researcher’s notes) “The wafarin was to commence on the 3rd day after we started the enoxaparin. I noted it in the patient’s folder as such but a corper (senior house officer) copied this into the treatment sheet. Maybe he didn’t note that fact… I did not cross‑check to know what the corper had written in the treatment sheet.” (Interview 18, Registrar, researcher’s notes) Defenses Nurses When asked if the prescriber had ever been involved in an error incidence “Yes, it was an IVF. It was meant to be given every12 h, but I wrote every 2 h. The matron corrected me and I changed it immediately.” (Interview 13, house officer, researcher’s notes) “I called the attention of my Reg to the dose (of acyclovir) and he said it can be taken 200 mg t.d.s. the nurses also called our attention to this but my Reg said it was ok… This is my 1st time of prescribing this drug.” (Interview 12, house officer, researcher’s notes) Pharmacists “The prescription was miswritten as bendrofluazide 50 mg (when 5 mg was intended) and given to the patient. The error was detected at the pharmacy department and prevented from reaching the patient.” (Interview 5, registrar, researcher’s notes) IVF=Intravenous fluid, EMDEX=The complete drug formulary for Nigeria’s health professionals (based on WHO model formulary), NS=Normal saline, F/S=Full strength
Doctors also showed a low consciousness of making errors. For instance, only two of the interviewees assented to sometimes (10‑20% of the time) making prescribing errors while the rest claimed they rarely (less than 5%) or seldom (5‑10% of the time) made errors. The workload and long working hours resulting in Annals of African Medicine
physical and mental stress are factors highlighted. Furthermore, the location of the treatment sheets in the wards was a contributory latent factor as reported by one HO who felt unfamiliar with this practice in the hospital as opposed to what was obtained elsewhere (Treatment sheets and drug administration records were not kept at the individual patient’s bedside, but in a pool at the nurse’s station). Page | 229
Violations A frequent violation was the omission of duration of therapy for antibiotic orders and intravenous infusions (11, 50%). Some interviewees suggested that this practice was promoted by their seniors who felt that such open‑ended practice made it easier to review the patient’s progress at the daily ward rounds until discharge.
Defenses Self‑initiated defenses by three doctors included re‑checking their own orders, one mentioned reading out the order silently to herself before signing it. Two mentioned nurses as sources of defense who reportedly helped prevent some prescription mistakes from reaching patients. Pharmacists were also mentioned as key sources of defense as doctors cited instances when they had called the pharmacy to get help with some drug doses/other information. Other doctors were also cited as defenses for their colleagues by identifying and preventing transcription errors.
Discussion This study described the use of human error theory to identify the causes of and factors underlying prescribing errors among junior doctors in a medical and pediatric in‑patient setting. Although prescribers must be held responsible for their actions, our study suggests that errors arise as a combination of environment, team, individual, task, and latent factors in a system where defenses are feeble. This finding was consistent with the results of others,[12] and a systematic review into the causes of prescribing errors.[20] This study also highlighted the need for training and teaching sessions for HO as a tool to familiarize them with the format of drug charts and the prescribing protocol. Since, a key focus of the internship year is to produce competent independent medical personnel through an apprenticeship, then supervised training should be a priority.[4] Training should include how to write prescriptions correctly with all pertinent information, such as the drug’s name, strength, dose, frequency, duration, and direction of use specified and to identify when dose adjustment might be needed. This is a basic requirement stated Vol. 12, October-December, 2013
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
in the Nigeria Standard Treatment Guideline,[21] which this study identified as not being strictly adhered to. Safe prescribing skills and an awareness of medication errors is required by all members of the health‑care team.[22] This should be central in the component of undergraduate and post‑graduate programs.
Page | 230
A change in practice is also needed in the way teams communicate treatments. Identifying a drug to be prescribed should go beyond just naming the drug, but should be followed by details about the dose, form, route of administration, duration etc. In short, prescription‑writing should be acknowledged as an important high‑risk activity requiring attentiveness and caution. HO need to have the skills, aptitude, and freedom to confirm and clarify directions. The structure in medical teams discouraged junior doctors from asking for clarification when uncertain about prescribing or were reluctant to query their seniors’ decisions. Conversely, the structure resulted in senior doctors passing their prescribing habits, whether appropriate or not, to the junior ones. Team factors, in particular were also more frequently associated with prescribing errors. Prescribing of medications for patients with complex problems was noted to be left to inexperienced junior doctors, usually HO, who were, on occasions, stressed, distracted or rushed. Our findings showed that prescriptions written by HO were not always checked by the seniors, a finding reported also by some UK researchers,[3] which opined that junior doctors felt supervision was inadequate. It can be overwhelming for these inexperienced doctors to have the in‑depth pharmacologic knowledge required to treat patients along with identifying and preventing the opportunities for drug‑drug interactions, drug‑disease incompatibilities and allergies. [23] Supervision must also include a culture in which prescribing errors identified are constructively discussed, analyzed and learnt from on the individual, team and organization level.[24] In Nigeria, pharmacists play a role in the defenses against prescribing errors when they screen and validate prescriptions. However, this role is mainly limited to within the pharmacy department and during the process of dispensing or filling of prescriptions. In general, pharmacists are yet to be integrated as vital members of the continuum of health‑ care delivery at the ward level as the practice of clinical pharmacy is rudimentary in Nigeria hospitals and its impact amongst prescribers is still modest.[25] Where practiced, pharmacists play an important role in identifying and monitoring errors. Nursing staff could also play a crucial defense by reviewing medications before administering. Some Vol. 12, October-December, 2013
of the HO interviewed welcomed their intervention in preventing some errors from reaching the patients. Others have also identified the roles of Pharmacists and Nurses as sources of defense.[3,4,15] Reference books and guidelines need to be readily available to prescribers. In many cases, these essential drug information materials are absent and prescribers have to depend on their personal copies or look for other ways to get the required information. The risk of errors will remain unless institutions make concerted efforts to ensure the availability of protocols, guidelines, and formularies in the prescribing environment. Similarly, the environment in which doctors prescribe must not be distracting. Furthermore, medication charts should ideally be located at the patients’ bedside. Modifying workload is a challenge for all health‑care professionals and measures to improve staffing levels and reduce stress need to be considered. However, doctors could improve their own defenses by recognizing conditions in which they might make an error such as having a heavy workload or feeling stressed. They should therefore be on guard and take extra care to mitigate the effect of these error‑prone risks.
Limitations The study involved 22 junior medical doctors in the medical and pediatric specialties only. The prescribers who made the errors were presenting their own account to the researchers. We cannot establish causality with certainty. There is the likelihood that responses offered expressed some measure of social desirability. Furthermore, most of the interviewees were HO and there is a possibility that the more senior doctors may have provided additional or different perspectives. However, we believe the study is representative and has raised some key issues for future interventions.
Acknowledgments We acknowledge the support of the Heads of the participating departments, all prescribers, pharmacists and other staff at the National Hospital Abuja who participated in this study, for their contribution.
References 1. 2.
3.
Dean B, Barber N, Schachter M. What is a prescribing error? Qual Health Care 2000;9:232‑7. ISMP Medication Safety Alert © 2011 Institute for Safe Medication Practices. A Patient Verification Process is Needed Everywhere. Available from: http://www.medscape. com/viewarticle/742372_print. [Cited 2011 May 31]. Dean B, Schachter M, Vincent C, Barber N. Causes of prescribing errors in hospital inpatients: A prospective study. Lancet 2002;359:1373‑8.
Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients 4.
5.
6. 7.
8. 9.
10.
11.
12.
13.
14. 15.
Coombes ID, Stowasser DA, Coombes JA, Mitchell C. Why do interns make prescribing errors? A qualitative study. Med J Aust 2008;188:89‑94. Kopp BJ, Erstad BL, Allen ME, Theodorou AA, Priestley G. Medication errors and adverse drug events in an intensive care unit: Direct observation approach for detection. Crit Care Med 2006;34:415‑25. Bates DW, Gawande AA. Improving safety with information technology. N Engl J Med 2003;348:2526‑34. Ogden DA, Lakhani J. An evaluation of the incidence, types and severity of prescribing errors in a Human Immunodeficiency Virus (HIV) outpatient clinic (St George’s Healthcare Trust, London, UK) – Poster 0312, 14 presentation at the Congress of the European Association of Hospital Pharmacists, Barcelona Spain 25‑27th March, 2008. Condren M, Studebaker IJ, John BM. Prescribing errors in a pediatric clinic. Clin Pediatr (Phila) 2010;49:49‑53. Hendey GW, Barth BE, Soliz T. Overnight and postcall errors in medication orders. Acad Emerg Med 2005;12:629‑34. Fijn R, Van den Bemt PM, Chow M, De Blaey CJ, De Jong‑Van den Berg LT, Brouwers JR. Hospital prescribing errors: Epidemiological assessment of predictors. Br J Clin Pharmacol 2002;53:326‑31. Lesar TS, Lomaestro BM, Pohl H. Medication‑prescribing errors in a teaching hospital. A 9‑year experience. Arch Intern Med 1997;157:1569‑76. Dornan T, Ashcroft D, Heathfield H, Lewis P, Miles J, Taylor D, et al. An in depth investigation into causes of prescribing errors by foundation trainees in relation to their medical education. EQUIP study. A report to the General Medical Council, 2009. Available from: http://www.gmc‑uk.org/final_report_prevalence_and_ causes‑of_prescribing_errors.pdf‑28935150.pdf. [Cited 2011 April 17]. Lewis PJ, Dornan T, Taylor D, Tully MP, Wass V, Ashcroft DM. Prevalence, incidence and nature of prescribing errors in hospital inpatients: A systematic review. Drug Saf 2009;32:379‑89. Reason J. Human error. Cambridge: University of Cambridge; 1990. Sanghera IS, Franklin BD, Dhillon S. The attitudes and
Annals of African Medicine
16.
17.
18.
19.
20.
21.
22.
23.
24. 25.
beliefs of healthcare professionals on the causes and reporting of medication errors in a UK Intensive care unit. Anaesthesia 2007;62:53‑61. World Health Organization, 1993. How to investigate drug use in health facilities: Selected drug use indicators. WHO/DAP/93.1 Geneva, Switzerland p. 24-31. Enwere OO, Falade CO, Salako BL. Drug prescribing pattern at the medical outpatient clinic of a tertiary hospital in southwestern Nigeria. Pharmacoepidemiol Drug Saf 2007;16:1244‑9. Enato EF, Chima IE. Evaluation of drug utilization patterns and patient care practices. West Afr J Pharm 2011;22:36‑41. Oshikoya KA, Ojo OI. Medication errors in paediatric outpatient prescriptions of a teaching hospital in Nigeria. Nig Q J Hosp Med 2007;17:74‑8. Tully MP, Ashcroft DM, Dornan T, Lewis PJ, Taylor D, Wass V. The causes of and factors associated with prescribing errors in hospital inpatients: A systematic review. Drug Saf 2009;32:819‑36. Federal Ministry of Health in collaboration with the World Health Organization. Standard Treatment Guideline for Nigeria. 1st ed. Published by the Federal Ministry of Health in collaboration with WHO, EC, DFID; 2008. p. 206‑8. Coombes ID, Mitchell CA, Stowasser DA. Safe medication practice tutorials: A practical approach to preparing safe prescribers. Clin Teach 2007;4:128‑34. Brennan C, Donnelly K, Somani S. Needs and opportunities for achieving optimal outcomes from the use of medicines in hospitals and health systems. Am J Health Syst Pharm 2011;68:1086‑96. Dean B. Learning from prescribing errors. Qual Saf Health Care 2002;11:258‑60. Erah PO, Nwazuoke JC. Identification of standards for pharmaceutical care in benin city. Trop J Pharm Res 2002;1:55‑66.
Cite this article as: Ajemigbitse AA, Omole MK, Osi-Ogbu OF, Erhun WO. A qualitative study of causes of prescribing errors among junior medical doctors in a Nigeria in-patient setting. Ann Afr Med 2013;12:223-31. Source of Support: Nil, Conflict of Interest: None declared.
Vol. 12, October-December, 2013
Page | 231
Copyright of Annals of African Medicine is the property of Medknow Publications & Media Pvt. Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Access this article online Quick Response Code:
Original Article
Website: www.annalsafrmed.org DOI: 10.4103/1596-3519.122691 PMID: ******
A qualitative study of causes of prescribing errors among junior medical doctors in a Nigeria in‑patient setting Adetutu A. Ajemigbitse, Moses K. Omole1, Ogugua F. Osi‑Ogbu2, Wilson O. Erhun3 Department of Pharmacy, National Hospital Abuja, 1Department of Clinical Pharmacy and Pharmacy Administration, University of Ibadan, Ibadan, 2Department of Medicine, National Hospital Abuja, Federal Capital Territory, 3 Department of Clinical Pharmacy and Pharmacy Administration, Obafemi Awolowo University, Ile‑Ife, Nigeria
Correspondence to: Ajemigbitse Adetutu A, Department of Pharmacy, National Hospital Abuja, Nigeria. E‑mail: [email protected]
Abstract Aims: The aims of this study were to identify and understand the factors underlying prescribing errors in order to determine how to prevent them. Materials and Methods: A prospective qualitative study that involved face-to-face interviews and human factor analysis in a Tertiary Referral Hospital in Central Nigeria, from July 2011 to December 2011. Pharmacists in the study hospital prospectively reviewed prescription orders generated by doctors in selected wards (male and female medical, pediatric and the private wing wards) and identified prescribing errors. The 22 prescribers involved in the errors were interviewed, and given questionnaires to discover factors causing the errors. A model of human error theory was used to analyze the responses. Results: Responses from the doctors suggest that most errors were made because of slips in attention. Lack of drug knowledge was not the single causative factor in any incident. Risk factors identified included individual, team, environment, and task factors. Junior doctors were affected by the prescribing habits of their seniors. Organizational factors identified included inadequate training/experience, absence of reference materials and absence of selfawareness of errors. Defenses against error such as other clinicians and guidelines were absent or deficient, and supervision was inadequate. Conclusions: To reduce the risk of prescribing errors, a number of strategies addressing individual, task, team, and environmental factors such as training of junior doctors, enforcing good practice in prescription writing, supervision, and reviewing the workload of junior doctors must be established. Keywords: In-patients, medication errors, patient safety, physicians, prescriptions
Résumé Objectifs: Les objectifs de cette étude étaient d'identifier et de comprendre les facteurs qui sous-tendent les erreurs thérapeutiques afin de déterminer comment les prévenir. Matériel et méthodes: Une étude qualitative prospective qui implique des entretiens en face-à-face et analyse de facteur humaindans un hôpital de référence tertiaire dans le centre du Nigéria, de juillet à décembre 2011. Pharmaciens à l'hôpital de l'étude prospective examiné des arrêtés de prescription générées par les médecins dans des quartiers sélectionnés (mâle et femelle médical, pédiatrique et les salles de l'aile privée) et identifié des erreurs de prescription. 22 Prescripteurs impliqués dans les erreurs ont été interrogés et donnés des questionnaires pour découvrir les facteurs causant les erreurs. Un modèle de la théorie de l'erreur humaine a été utilisé pour analyser les réponses. Résultats: Les réponses des médecins suggèrent que la plupart des erreurs ont été faites en raison de la glisse dans
Annals of African Medicine
Vol. 12, October-December, 2013
Page | 223
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients l'attention. Manque de connaissances sur le médicament n'était pas le seul facteur causal dans n'importe quel incident. Facteurs de risque identifiés inclus individu, équipe, environnement et facteurs de la tâche. Médecins en formation ont été touchés par les habitudes de prescription de leurs aînés. Facteurs organisationnels identifiés inclus insuffisante formation/expérience, absence de documents de référence et absence de conscience de soi des erreurs. Défenses contre toute erreur, tels que les autres cliniciens et les lignes directrices étaient absents ou déficient, et supervision était inadéquate. Conclusions: Pour réduire le risque de prescription des erreurs, un certain nombre de stratégies individuelles, les tâches et des facteurs environnementaux tels que la formation des médecins en formation, faire respecter les bonnes pratiques dans la prescription d'écriture, supervision et examiner la charge de travail des médecins en formation doit être établi. Page | 224
Mots-clés: Patients hospitalisés, les erreurs de médication, la sécurité des patients, médecins, prescriptions
Introduction A prescribing error is defined as a prescribing decision or prescription writing process that results in an unintentional, significant reduction in the probability of treatment being timely and effective or increases the risk of harm when compared with generally accepted practice.[1] Prescribing without taking into account the patient’s clinical status, failure to communicate essential information and transcription errors were considered prescribing errors. Errors made during drug prescription are the most common type of avoidable medication errors and hence are an important target for improvement.[2,3] Many factors have been associated with prescribing errors.[3‑12] Understanding these factors should assist in the implementation of effective strategies for error prevention.[12,13] The Reason accident causation,[14] frame‑work is a popular theoretical frame‑work used to study the causes of errors and accidents in various industries. This approach has been practically related to prescribing errors,[3,4,12,15] and constitutes the theoretical basis of this study. Past studies on medication prescribing revealed poor compliance with the principles of rational medication use.[16‑19] This study seeks to investigate the causes of prescribing errors among junior medical doctors in a Tertiary Hospital in Nigeria so as to provide a baseline for future intervention especially as there are no such previous studies in Nigeria.
Materials and Methods Study participants and setting This study prospectively reviewed prescriptions generated by medical doctors in selected wards (male and female medical, pediatric and the private wing wards) at the National Hospital Abuja, Nigeria. This hospital, though currently a 200‑bed Tertiary Referral Hospital, plays a key role in the provision of health Vol. 12, October-December, 2013
services to the residents of Abuja, a cosmopolitan multi‑ethnic, multi‑cultural urban city with high socio‑economic disparity. In addition, it serves as an institution for training of health personnel offering undergraduate, graduate, post‑graduate, and residency training programs. The medical staff structure in the hospital include the house officers (HO) who are the newly qualified doctors on the mandatory 1‑year internship training, the National Youth Service Corp doctors, the junior and senior Residents, medical Consultants, Visiting Professors, and other medical personnel on research activities and sundry. Total out‑patient attendance in the year 2010 was over 40,000 patients while the turn‑over of in‑patient admissions to the medical and pediatric wards was approximately 1900 patients for the same year. The hospital operates the typical Nigeria pharmacy service. Briefly, this involves prescriber’s hand‑writing medication orders into the patient’s medical notes as well as on formatted treatment sheets. These orders are taken to the pharmacy units where pharmacists check that the orders are clear, clinically appropriate and valid before initiating the supply of any drugs to patients in the wards. These same documents are used by the nursing staff to determine the doses due at each medication round and to record their administration. The doctors involved in the selected units were informed about the study and its goal. The goal was to understand the causes of prescribing errors with a view to developing specific interventions that will improve practice with ultimate improvement in patient’s safety. The local Ethics Committee approved the study and all prescribers interviewed gave written consent. Medication orders were screened for prescribing errors that met the study definition.[1] Data were collected on 5‑chosen days in a month, for a 6 months period between July 2011 and December 2011. Only prescribers who were involved in error incidents and contactable within 72 h of the incident were interviewed to ensure adequate recall. Semi‑structured Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
face‑to‑face interviews, using a questionnaire, were conducted to determine the causes of prescribing errors. Causes were defined as “reasons reported to the researchers by the prescriber as being wholly or partially responsible for a specific prescribing error.” Interviews took between 15 min and 25 min and the process was adapted from the methods of other researchers.[3,4,12,15] Table 1 outlines the process.
lapses are errors due to omission of a particular task and slips are errors owing to attentional failure when performing a task or an active failure resulting from the incorrect execution of a task. Mistakes were errors occurring from correct execution of an incorrect or inappropriate plan and were either rule‑based (RBM) or knowledge‑based mistakes (KBM).
Definitions
Error‑producing conditions are related to the task and the environment at the time when the error occurred. They are underlying risk factors and can be grouped as connected to the individual prescriber, the working condition, the health‑care team, the prescribing task and patient. Latent conditions are organizational processes that create an environment where error‑producing conditions and active‑failures are more likely to result in prescribing errors.
Active failures are the unsafe acts committed by the prescribers in contact with the patient. Memory Table 1: Interview schedule with prescribers Background The prescribers were asked to give a brief introduction of themselves, medical school attended, how long they had worked at the hospital, how long in post and what specialty Provide a brief description of type of teaching or learning courses that prepared them for prescribing, how this was taught and if they thought this was adequate The prescribing errors (areas covered) The type of error made Dosing error, omission of information, frequency error, drug interactions etc. The medication involved The condition being treated The error situation Time of day Type of ward Type of patient How long at the ward A description of what happened Work load Supervision Stage of patient stay The reason for making the error Their attitude towards the error Has this happened before? Has this happened since? How does this make you feel? What do you think could have helped to prevent the error? How will this change the way you prescribe? Closing The interviews ended with asking the prescribers if they had any questions, comments or suggestions for the interviewers
/DWHQWFRQGLWLRQV 2UJDQL]DWLRQDOSURFHVVHV 0DQDJHPHQWGHFLVLRQV
(UURU±SURGXFLQJ FRQGLWLRQV (QYLURQPHQWDOWHDP LQGLYLGXDOWDVNRUSDWLHQW IDFWRUV
Violations are active choices by prescribers to ignore the formal or informal policies or guidelines they are expected to adhere to. Defenses are designed to protect against hazards and mitigate consequences of failure.
Analysis Notes taken during the interviews were transcribed afterward and reviewed by the interviewers (a consultant physician and a clinical pharmacist). Common themes relating to the reporting of errors were identified and coded manually based on the interviewee’s words and phrases. Consensus was reached through discussion. The result was presented according to Reason’s model of accident causation [Figure 1].
Results A total of 90 errors involving 37 doctors were reported. In 15 cases, the prescribers could not be interviewed because, 7 were unidentified, three were reported more than 72 hoursh after the event, and in 5 cases the interviews could not hold within the 72 hoursh target period.
$FWLYHIDLOXUHV (UURUVVOLSVODSVHV PLVWDNHV 9LRODWLRQV
$FFLGHQW
'HIHQFHV ,QDGHTXDWH PLVVLQJRU XQDYDLODEOH Figure 1: Reason’s model of accident causation
Annals of African Medicine
Vol. 12, October-December, 2013
Page | 225
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 2: Prescribing errors and analysis by themes from interviews with prescribers Stage of error On admission 1. IVF Dextrose 5%, Cinnarizine (Stugeron©), tramadol During stay 2. Lisinopril Page | 226
3. Tramadol 4. Amoxicillin+clavulanic acid, soluble insulin, IVF normal saline 5. Erythromycin, metronidazole, amoxicillin+ clavulanic acid 6. Ceftriaxone 7. Ceftriaxone, ranitidine 8. Tramadol 9. Insulin
Context of error
Prescriber Directed Prescribed Themes underlying error by senior before?
Duration/stop‑date Registrar omitted, dose inappropriate
No
Yes
Training, chore, supervision, absence of guidelines
Dose inappropriate
Yes
Yes
Dose adjustment in renal impairment Duration of therapy omitted+unsafe abbreviation Duration of therapy omitted House officer
Yes
No
Yes
Yes
Multitasking, supervision, absence of error awareness Supervision, communication, training Training, absence of guidelines, supervision
Yes
Yes
Training, supervision, communication, absence of guidelines
Dose adjustment in renal impairment; stop‑date omitted Duration of therapy omitted, need for therapy omitted Transcription error, route of administration omitted
House officer
Yes
Yes
Lack of knowledge, training, supervision, communication
House officer
Yes
No
Training, supervision, absence of error awareness
House officer
Yes
No
Unsafe abbreviation
House officer House officer House officer
Yes
Yes
Yes
No
Yes
No
House officer Registrar
Yes
Yes
Yes
No
House officer House officer
Yes
No
Absence of error awareness, lack of knowledge, supervision, inattention Training, supervision, absence of guidelines Lack of knowledge, supervision Lack of knowledge, absence of reference material, supervision, communication Communication, supervision, trust in seniors Absence of reference books, co‑managed patient, communication Trust in senior, nurse defense over-ridden, supervision Absence of error awareness, lack of knowledge of patient, chore, supervision Supervision, rushed, training, lack of knowledge Lack of knowledge, training, supervision, chore, co‑managed patient Communication, absence of error awareness, multitasking, supervision Supervision/responsibility, absence of error awareness Lack of knowledge, communication, supervision, chore Training, supervision, chore
10. Domperidone, nifedipine+atenolol 11. Nifedipine+nimodipine, enoxaparin
Dose inappropriate, drug‑drug interaction Need for drug therapy omitted; Drug-disease contraindication 12. Amoxicillin+clavulanic Dose inappropriate acid 13. Carbamazepine+ Drug‑drug interaction amitriptiline
House officer House officer House officer
14. Aciclovir
Dose inappropriate
15. Ceftriaxone, metronidazole, gentamycin 16. Metoclopramide, hyoscine butyl bromide 17. Risperidone, phenytoin, metronidazole
Dose inappropriate, duration/stop‑date omitted, transcription error Route of administration House omitted, inappropriate dose officer Drug‑drug interaction House officer
Yes
Yes
Yes
No
Yes
No
18. Imipenem
Dose inappropriate
House officer
Yes
No
19. Diclofenac, warfarin, enoxaparin 20. Rabeprazole+ omeprazole
Drug‑drug interaction
Registrar
No
Yes
Duplication, need for drug omitted
House officer
Yes
No
House officer House officer
Yes
Yes
Yes
No
21. Insulin, metronidazole Unsafe abbreviation, duration/stop‑date omitted 22. Naproxen+aspirin+ Drug‑drug interaction tramadol
Lack of knowledge, training, supervision
IVF=Intravenous fluid
A total of 49 error incidents, involving 22 doctors were analyzed. Prescribers involved were 19 HO and 3 Registrars while 21 of the incidents were Vol. 12, October-December, 2013
medical cases. In all 22 cases, the prescribers were unaware that an error, lapse or mistake had occurred. Of the 22 incidents analyzed [Table 2], 1 occurred Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 3: Examples of active failures Errors 1. Slips “I was writing a prescription for IV frusemide 120 mg b.d. when a nurse talked to me…I can’t really remember what happened to have caused the mix-up but I was trying to catch up with the team. I don’t know how I mixed up the dose meant for frusemide for lisinopril”. (Interview 1, house officer, researcher’s notes) 2. Lapses “I am not used to giving tramal (tramadol) as an injection… I had to ask the SR, and he said injection, but I didn’t confirm the route. But I can remember the prescription clearly. He said 100 mg stat and 50 mg t.d.s po for 3 days. Two of us were taking the instruction. I wrote in the folder and my colleague in the treatment sheet”. (Interview 7, house officer, researcher’s notes) 3. Mistakes Some mistakes resulted from an absence of knowledge of correct dose, or relevant rules “I don’t know exactly where this error is from. I can’t say whether the child was on 350 mg IV ceftriazone elsewhere and I copied the prescription… IV ceftriazone is given at a dose of 50-75 mg most times but in meningitis can go as high as 100 mg/kg. That’s what my Reg says…I think in this case calculating my own dose would have helped, but if the managing team has a reason for a different dose, I may have to know why…” (Interview 14, house officer, researcher’s notes) When asked to explain his mistake a HO did not know that ClexaneTM is the brand name for enoxaparin. “The patient was thought to have pulmonary embolism in addition to his stroke…It was agreed it won’t be appropriate to give this patient clexane because of his internal bleeding… I am not aware that enoxaparin is the active ingredient in clexane.” (Interview 9, house officer, researcher’s notes) “This patient was formerly on nifedipine 20 mg b.d. and we later increased it to 30 mg (with improvement in the blood pressure control). Then nimodipine was added. My SR said nimodipine was for the headache not as an anti‑hypertensive. Later, I called him to confirm the dose and he said we should give 50 mg q6rs ×1/52… I also thought about the drug being a CCB (same class as nifedipine), but my SR said it was ok…… We felt the drug (nimodipine) was needed because the patient was complaining of headache and stiffness around the neck so we thought of sub‑arachnoid hemorrhage.” (Patient with right hemispheric hemorrhagic stroke) (Interview 8, house officer, researcher’s notes) SR=Senior registrar, Reg=Registrar, NS=Normal saline, F/S Darrows=Full strength darrow’s solution, IU=International units, CCB=Calcium-channel blocker, Frusemide=Furosemide
on admission to the hospital and 21 during the in‑patient stay. Antibacterials, analgesics, and cardio‑vascular system drugs accounted for over half the errors. HO wrote most of the prescriptions although they were directed by their seniors. Sometimes, the HO determined the form, dose, route, and frequency. Annals of African Medicine
Main factors involved with all prescribing errors are individual (16, 72.7%), team (14, 63.6%) and work environment (9, 40.9%) factors.
Active failures All errors occurred as a result of at least one active failure. Mistakes were frequent [Table 3]. All the prescribers interviewed were unaware of the slip or lapse, but explained that they were rushing (6, 27%) or distracted during routine tasks (5, 23%). Common mistakes were errors of inappropriate dosing (10, 45%) and failure to check for potentially serious drug interactions (5, 23%). A common cause of RBM was the lack of knowledge of a relevant rule for example dose adjustment for patients with renal impairment (3, 14%). Prescribing two or more drugs that could result in potentially serious interaction was also noted, e.g., carbamazepine and amitryptilline; phenytoin and metronidazole. In the former case, the patient was being co‑managed by the psychiatry and medical teams. There were instances where the correct rule was wrongly applied. For example, intravenous injection ceftriaxone dosing was misinterpreted as 20‑50 mg/kg twice daily instead of 20‑50 mg/kg daily in two divided doses. The prescriber multiplied the dose provided by the patient’s weight and prescribed this amount to be given twice daily, resulting in a two‑fold overdose. Another active failure showing lack of knowledge or experience (KBM) involved the writing of “IU” (short for international units) as an abbreviation for “units” when prescribing insulin (3, 14%). Even though, they agreed to the possibility of the orders being erroneously misread or misinterpreted leading to a ten‑fold error in dose, junior doctors claimed to be influenced by the practices of their senior colleagues in this. One interviewee stated that she tended to use an abbreviation of drug names (e.g., Full strength Darrow’s solution instead of full strength Darrow’s solution, NS for normal saline) as a way to save time while writing.
Error‑producing conditions Many factors were cited as contributory to the errors, the most frequent being work environment (11, 50%), individual (10, 45%) and team (8, 36%), [Table 4]. In 10 instances, high workload, resulting in multitasking was thought to have contributed to the error; in 5 instances the physical environment was cited. The terms “hectic”, “busy”, stressful’ were used to describe workload. Sometimes, the work situation made the junior doctors rush their prescribing and other duties in order to catch up with their team. Part of the pressure was to get the prescriptions written on time for the ward attendants to take the treatment sheets Vol. 12, October-December, 2013
Page | 227
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 4: Examples of error‑producing conditions
Page | 228
Environment factors “There was a lot of distraction in the ward… (Interview 2, house officer, researcher’s notes) Individual factors “When more than one person talks to me at the same time, I tend to get distracted…I can get them to please pause and take it one‑by‑one. Furthermore, if my Reg is too fast, I can slow him down so I get it”. (Interview 2, house officer, researcher’s notes) “The metoclopramide was prescribed earlier and that had the route of administration, dose, etc. So I omitted writing that (route of administration) but it would have been better to have included the route. I may have been rushing to write the prescription before it goes to the pharmacy.” (Interview 15, house officer, researcher’s notes) Team factors “It was a team decision, I was just the writer. The prescription was not written the same day. The carbamazepine was for prophylaxis because the patient had seizures in the past and amitryptilline was prescribed 2 days ago because the patient was manifesting symptoms of depression… It must have been an oversight because such (interaction between carbamazepine and amitryptilline) had been mentioned to us before. But, like I said, I was the writer; the decision wasn’t taken by me”. (Interview 11, registrar, researcher’s notes) Task factors “I discussed with my SR and he made a suggestion of not using the commonly used NSAIDs. He suggested the tramadol. We made a change on this to DF118 due to intractable vomiting. I omitted to write this (the substitution) in the patient’s folder.”(Interview 3, house officer, researcher’s notes) Patient factors Patient being co-managed by two different specialties “The dose of phenytoin was initiated at the A and E to treat seizures. It was 100 mg b.d. Risperidone was later ordered by the psychiatrist, but I wrote it down. I had a thought about it (the combination), but being a house officer, I can’t question a consultant’s order. (Interview 17, house officer, researcher’s notes) NSAIDs=Non-steroidal anti-inflammatory drugs, DF 118=Dihydrocodeine tartrate tablets, A and E=Accident and emergency unit, SR=Senior registrar
to pharmacy so that any medications requiring compounding would be presented in time before the 4.00 pm shift was over (In the hospital, morning work shift begins at 8.00 am until 4.00 pm). Individual factors mentioned in connection with prescribing errors included physical and mental well‑being and lack of knowledge. A total of 14 respondents mentioned tiredness (50%), distraction (29%), low morale (21%), and confusion (14%) as factors that may have caused the error. An absence of knowledge about dose and no prior experience of prescribing some drugs were contributory to errors. In 10 instances, junior doctors wrote prescription for drugs without indicating the route of administration and frequency of administration, claiming they had Vol. 12, October-December, 2013
never prescribed the drug before or did not know the dose of the drug or the correct frequency of administration. In 9 instances, the HOs reported not having received any training in prescribing as undergraduates and five reported minimal training. Team factors, such as communication, responsibility, and supervision were associated with 12 incidents. Junior doctors commonly referred to communication about medications with words such as “I wrote what was dictated to me by the Senior Reg,” “I was told by the Reg to leave it open.” Some junior doctors felt incompetent to disagree with the decisions of the seniors, did not feel it was correct to have a differing opinion from their superiors or even ask questions when they were not sure of the correct dose assuming that to do so was to expose ignorance. In an instance, when acyclovir was prescribed by a HO for the 1st time, an error in the frequency of administration was made. When interviewed he claimed to have called the attention of the registrar to what he had written and because the registrar had accented to this, he did not take any further action. Junior doctors also felt that if a problem occurred, the seniors should take responsibility as the prescriptive authority. Other causes included lack of documentation of the prescribed medicine in the patient’s notes. In some instances, supervision was inadequate as some senior doctors omitted to cross‑check what the junior ones prescribed or did. Task and patient factors were mentioned as possible causes of errors. Task factors comprised the unavailability of drug reference materials such as the Hospital Formulary or Standard Treatment Guidelines or similar document to consult at the point of prescribing. Inadequate patient information (e.g., weight) also made for dose calculation errors especially in pediatric patients where doses are usually calculated based on a patient’s weight. The absence of these documents meant they had to depend on other sources for information on doses, or copy from the senior’s prescribing‑whether correct or not. In 3 instances, the patient was mentioned as a contributing factor; 2 were being managed by two different specialist teams while the third case was a complex one.
Latent underlying conditions Lack of training in prescribing skills and insufficient knowledge and experience about drugs were latent factors [Table 5]. The junior doctors were left to fill all the details of strength, dose, form, frequency, route, duration, and direction after verbal instructions were given. Usually, these orders were not cross‑checked by the seniors. Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
Table 5: Examples of latent conditions and defenses Latent conditions When asked what the prescriber felt would help in preventing a repeat of the dosing error made “Providing a small booklet of medicine information or drug reference such as EMDEX or the hospital’s formulary will be helpful.” (Interview 9, house officer, researcher’s notes) “I have had no prior training on drug prescribing. Just pharmacology where we were taught about pharmacology of drugs, dose, side‑effects etc. (Interview 14, house officer, researcher’s notes) Violations “I remember writing the prescription. We were rushing to the next patient… two nurses were talking to me; the Reg was waiting for me to write, so I got distracted. I was rushing to write fast. I use abbreviation to make me write faster… (Abbreviations such as NS, F/S Darrows, insulin 8 IU) (Interview 2, house officer, researcher’s notes) “Most times our seniors tell us to prescribe (an antibacterial) and to leave it open. I prescribe that way because in most cases duration (of therapy) is not given… I tend not to write the duration because that’s the way I’m told.” (Interview 3, house officer, researcher’s notes) “The wafarin was to commence on the 3rd day after we started the enoxaparin. I noted it in the patient’s folder as such but a corper (senior house officer) copied this into the treatment sheet. Maybe he didn’t note that fact… I did not cross‑check to know what the corper had written in the treatment sheet.” (Interview 18, Registrar, researcher’s notes) Defenses Nurses When asked if the prescriber had ever been involved in an error incidence “Yes, it was an IVF. It was meant to be given every12 h, but I wrote every 2 h. The matron corrected me and I changed it immediately.” (Interview 13, house officer, researcher’s notes) “I called the attention of my Reg to the dose (of acyclovir) and he said it can be taken 200 mg t.d.s. the nurses also called our attention to this but my Reg said it was ok… This is my 1st time of prescribing this drug.” (Interview 12, house officer, researcher’s notes) Pharmacists “The prescription was miswritten as bendrofluazide 50 mg (when 5 mg was intended) and given to the patient. The error was detected at the pharmacy department and prevented from reaching the patient.” (Interview 5, registrar, researcher’s notes) IVF=Intravenous fluid, EMDEX=The complete drug formulary for Nigeria’s health professionals (based on WHO model formulary), NS=Normal saline, F/S=Full strength
Doctors also showed a low consciousness of making errors. For instance, only two of the interviewees assented to sometimes (10‑20% of the time) making prescribing errors while the rest claimed they rarely (less than 5%) or seldom (5‑10% of the time) made errors. The workload and long working hours resulting in Annals of African Medicine
physical and mental stress are factors highlighted. Furthermore, the location of the treatment sheets in the wards was a contributory latent factor as reported by one HO who felt unfamiliar with this practice in the hospital as opposed to what was obtained elsewhere (Treatment sheets and drug administration records were not kept at the individual patient’s bedside, but in a pool at the nurse’s station). Page | 229
Violations A frequent violation was the omission of duration of therapy for antibiotic orders and intravenous infusions (11, 50%). Some interviewees suggested that this practice was promoted by their seniors who felt that such open‑ended practice made it easier to review the patient’s progress at the daily ward rounds until discharge.
Defenses Self‑initiated defenses by three doctors included re‑checking their own orders, one mentioned reading out the order silently to herself before signing it. Two mentioned nurses as sources of defense who reportedly helped prevent some prescription mistakes from reaching patients. Pharmacists were also mentioned as key sources of defense as doctors cited instances when they had called the pharmacy to get help with some drug doses/other information. Other doctors were also cited as defenses for their colleagues by identifying and preventing transcription errors.
Discussion This study described the use of human error theory to identify the causes of and factors underlying prescribing errors among junior doctors in a medical and pediatric in‑patient setting. Although prescribers must be held responsible for their actions, our study suggests that errors arise as a combination of environment, team, individual, task, and latent factors in a system where defenses are feeble. This finding was consistent with the results of others,[12] and a systematic review into the causes of prescribing errors.[20] This study also highlighted the need for training and teaching sessions for HO as a tool to familiarize them with the format of drug charts and the prescribing protocol. Since, a key focus of the internship year is to produce competent independent medical personnel through an apprenticeship, then supervised training should be a priority.[4] Training should include how to write prescriptions correctly with all pertinent information, such as the drug’s name, strength, dose, frequency, duration, and direction of use specified and to identify when dose adjustment might be needed. This is a basic requirement stated Vol. 12, October-December, 2013
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients
in the Nigeria Standard Treatment Guideline,[21] which this study identified as not being strictly adhered to. Safe prescribing skills and an awareness of medication errors is required by all members of the health‑care team.[22] This should be central in the component of undergraduate and post‑graduate programs.
Page | 230
A change in practice is also needed in the way teams communicate treatments. Identifying a drug to be prescribed should go beyond just naming the drug, but should be followed by details about the dose, form, route of administration, duration etc. In short, prescription‑writing should be acknowledged as an important high‑risk activity requiring attentiveness and caution. HO need to have the skills, aptitude, and freedom to confirm and clarify directions. The structure in medical teams discouraged junior doctors from asking for clarification when uncertain about prescribing or were reluctant to query their seniors’ decisions. Conversely, the structure resulted in senior doctors passing their prescribing habits, whether appropriate or not, to the junior ones. Team factors, in particular were also more frequently associated with prescribing errors. Prescribing of medications for patients with complex problems was noted to be left to inexperienced junior doctors, usually HO, who were, on occasions, stressed, distracted or rushed. Our findings showed that prescriptions written by HO were not always checked by the seniors, a finding reported also by some UK researchers,[3] which opined that junior doctors felt supervision was inadequate. It can be overwhelming for these inexperienced doctors to have the in‑depth pharmacologic knowledge required to treat patients along with identifying and preventing the opportunities for drug‑drug interactions, drug‑disease incompatibilities and allergies. [23] Supervision must also include a culture in which prescribing errors identified are constructively discussed, analyzed and learnt from on the individual, team and organization level.[24] In Nigeria, pharmacists play a role in the defenses against prescribing errors when they screen and validate prescriptions. However, this role is mainly limited to within the pharmacy department and during the process of dispensing or filling of prescriptions. In general, pharmacists are yet to be integrated as vital members of the continuum of health‑ care delivery at the ward level as the practice of clinical pharmacy is rudimentary in Nigeria hospitals and its impact amongst prescribers is still modest.[25] Where practiced, pharmacists play an important role in identifying and monitoring errors. Nursing staff could also play a crucial defense by reviewing medications before administering. Some Vol. 12, October-December, 2013
of the HO interviewed welcomed their intervention in preventing some errors from reaching the patients. Others have also identified the roles of Pharmacists and Nurses as sources of defense.[3,4,15] Reference books and guidelines need to be readily available to prescribers. In many cases, these essential drug information materials are absent and prescribers have to depend on their personal copies or look for other ways to get the required information. The risk of errors will remain unless institutions make concerted efforts to ensure the availability of protocols, guidelines, and formularies in the prescribing environment. Similarly, the environment in which doctors prescribe must not be distracting. Furthermore, medication charts should ideally be located at the patients’ bedside. Modifying workload is a challenge for all health‑care professionals and measures to improve staffing levels and reduce stress need to be considered. However, doctors could improve their own defenses by recognizing conditions in which they might make an error such as having a heavy workload or feeling stressed. They should therefore be on guard and take extra care to mitigate the effect of these error‑prone risks.
Limitations The study involved 22 junior medical doctors in the medical and pediatric specialties only. The prescribers who made the errors were presenting their own account to the researchers. We cannot establish causality with certainty. There is the likelihood that responses offered expressed some measure of social desirability. Furthermore, most of the interviewees were HO and there is a possibility that the more senior doctors may have provided additional or different perspectives. However, we believe the study is representative and has raised some key issues for future interventions.
Acknowledgments We acknowledge the support of the Heads of the participating departments, all prescribers, pharmacists and other staff at the National Hospital Abuja who participated in this study, for their contribution.
References 1. 2.
3.
Dean B, Barber N, Schachter M. What is a prescribing error? Qual Health Care 2000;9:232‑7. ISMP Medication Safety Alert © 2011 Institute for Safe Medication Practices. A Patient Verification Process is Needed Everywhere. Available from: http://www.medscape. com/viewarticle/742372_print. [Cited 2011 May 31]. Dean B, Schachter M, Vincent C, Barber N. Causes of prescribing errors in hospital inpatients: A prospective study. Lancet 2002;359:1373‑8.
Annals of African Medicine
Ajemigbitse, et al.: Causes of prescribing errors in in‑patients 4.
5.
6. 7.
8. 9.
10.
11.
12.
13.
14. 15.
Coombes ID, Stowasser DA, Coombes JA, Mitchell C. Why do interns make prescribing errors? A qualitative study. Med J Aust 2008;188:89‑94. Kopp BJ, Erstad BL, Allen ME, Theodorou AA, Priestley G. Medication errors and adverse drug events in an intensive care unit: Direct observation approach for detection. Crit Care Med 2006;34:415‑25. Bates DW, Gawande AA. Improving safety with information technology. N Engl J Med 2003;348:2526‑34. Ogden DA, Lakhani J. An evaluation of the incidence, types and severity of prescribing errors in a Human Immunodeficiency Virus (HIV) outpatient clinic (St George’s Healthcare Trust, London, UK) – Poster 0312, 14 presentation at the Congress of the European Association of Hospital Pharmacists, Barcelona Spain 25‑27th March, 2008. Condren M, Studebaker IJ, John BM. Prescribing errors in a pediatric clinic. Clin Pediatr (Phila) 2010;49:49‑53. Hendey GW, Barth BE, Soliz T. Overnight and postcall errors in medication orders. Acad Emerg Med 2005;12:629‑34. Fijn R, Van den Bemt PM, Chow M, De Blaey CJ, De Jong‑Van den Berg LT, Brouwers JR. Hospital prescribing errors: Epidemiological assessment of predictors. Br J Clin Pharmacol 2002;53:326‑31. Lesar TS, Lomaestro BM, Pohl H. Medication‑prescribing errors in a teaching hospital. A 9‑year experience. Arch Intern Med 1997;157:1569‑76. Dornan T, Ashcroft D, Heathfield H, Lewis P, Miles J, Taylor D, et al. An in depth investigation into causes of prescribing errors by foundation trainees in relation to their medical education. EQUIP study. A report to the General Medical Council, 2009. Available from: http://www.gmc‑uk.org/final_report_prevalence_and_ causes‑of_prescribing_errors.pdf‑28935150.pdf. [Cited 2011 April 17]. Lewis PJ, Dornan T, Taylor D, Tully MP, Wass V, Ashcroft DM. Prevalence, incidence and nature of prescribing errors in hospital inpatients: A systematic review. Drug Saf 2009;32:379‑89. Reason J. Human error. Cambridge: University of Cambridge; 1990. Sanghera IS, Franklin BD, Dhillon S. The attitudes and
Annals of African Medicine
16.
17.
18.
19.
20.
21.
22.
23.
24. 25.
beliefs of healthcare professionals on the causes and reporting of medication errors in a UK Intensive care unit. Anaesthesia 2007;62:53‑61. World Health Organization, 1993. How to investigate drug use in health facilities: Selected drug use indicators. WHO/DAP/93.1 Geneva, Switzerland p. 24-31. Enwere OO, Falade CO, Salako BL. Drug prescribing pattern at the medical outpatient clinic of a tertiary hospital in southwestern Nigeria. Pharmacoepidemiol Drug Saf 2007;16:1244‑9. Enato EF, Chima IE. Evaluation of drug utilization patterns and patient care practices. West Afr J Pharm 2011;22:36‑41. Oshikoya KA, Ojo OI. Medication errors in paediatric outpatient prescriptions of a teaching hospital in Nigeria. Nig Q J Hosp Med 2007;17:74‑8. Tully MP, Ashcroft DM, Dornan T, Lewis PJ, Taylor D, Wass V. The causes of and factors associated with prescribing errors in hospital inpatients: A systematic review. Drug Saf 2009;32:819‑36. Federal Ministry of Health in collaboration with the World Health Organization. Standard Treatment Guideline for Nigeria. 1st ed. Published by the Federal Ministry of Health in collaboration with WHO, EC, DFID; 2008. p. 206‑8. Coombes ID, Mitchell CA, Stowasser DA. Safe medication practice tutorials: A practical approach to preparing safe prescribers. Clin Teach 2007;4:128‑34. Brennan C, Donnelly K, Somani S. Needs and opportunities for achieving optimal outcomes from the use of medicines in hospitals and health systems. Am J Health Syst Pharm 2011;68:1086‑96. Dean B. Learning from prescribing errors. Qual Saf Health Care 2002;11:258‑60. Erah PO, Nwazuoke JC. Identification of standards for pharmaceutical care in benin city. Trop J Pharm Res 2002;1:55‑66.
Cite this article as: Ajemigbitse AA, Omole MK, Osi-Ogbu OF, Erhun WO. A qualitative study of causes of prescribing errors among junior medical doctors in a Nigeria in-patient setting. Ann Afr Med 2013;12:223-31. Source of Support: Nil, Conflict of Interest: None declared.
Vol. 12, October-December, 2013
Page | 231
Copyright of Annals of African Medicine is the property of Medknow Publications & Media Pvt. Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
