Assessment of medication use process in adult oncology unit of Tikur Anbesa Specialized Hospital

A cross-sectional study in Addis Ababa, Ethiopia

Abstract

Background: 

Chemotherapy errors affect several steps of medication-use process. A reduction of errors and toxicities, and an increase in error awareness, dose verification, proper documentation, and appropriateness of supportive care treatment are important. The present study was conducted to assess the medication-use process and dose-related errors in an outpatient adult oncology unit of Tikur Anbesa Specialized Hospital.

Methods: 

An institutional-based cross-sectional study was conducted in an oncology outpatient adult oncology unit of Tikur Anbesa Specialized Hospital between May 1 and June 30, 2012. A total of 212 patient records for 583 chemotherapy administrations were reviewed during the study period.

Results: 

The overall dose-related error rate was found to be 228 (39.1%). Specific rate of dose-related errors were (under dosing 58 (25.4%), overdosing 52 (22.8%), inaccurately reconstituted doses 106 (46.5%), and inappropriately adjusted doses 12 (5.3%). Dose labeling and documentation of appropriate time of administration for each chemotherapy preparation were not evident in the unit records. Only 3/14 dose verifications and 3 dose documentation processes were found. Supportive care treatment was not performed according to the recommended standards. Of the 23 equipment and supplies needed for chemotherapy preparation and administration, only 8 were available.

Conclusions: 

The findings of our study indicate that the medication-use process in an outpatient adult oncology unit was below expected standards in 2012. The hospital and the school of pharmacy should give emphasis on the establishment of pharmacist run oncology pharmacy for chemotherapy preparation. Policies and guidelines for chemotherapy mixing and administration service and standardization of the process workflow should be developed.

1 Introduction

Cancer was becoming a leading cause of death worldwide among the noncommunicable disease.^[1] There was 14.1 million new cancer cases and 8.2 million cancer deaths worldwide in 2012.^[2] The most common cancer in the world was lung cancer and the most common for women was breast cancer.^[2] In 2012 in Africa, there was 847,000 new cancer cases (6.0% of the world total) and 591,000 cancer deaths (7.2% of the world total) were estimated to have occurred. In males, cancer of prostate was the leading and in women breast cancer was the most common form of cancer.^[3]

Ethiopia has no national cancer registry in 2012. An earlier study showed that cancer of cervix was a leading cause of cancer deaths in the country, and accounted 39.5%, followed by breast cancer 27.8%.^[4] However, according to the annual cancer registration report of Addis Ababa, from November 2011 to October 2012 there were 2926 cancer cases and breast cancer (738 cases) was the number one diagnosis. A small proportion of the diagnoses of breast cancer were made in male patients (43 patients).^[5]

Medication use process is a multistep process in which a drug travels from pharmacy to the patient and consists of prescribing, transcribing and documenting, dispensing, administering, and monitoring.^[6] Medication error can occur at any steps of medication use process. Dose calculation, preparation, and administration of chemotherapy have to be strictly performed since they are error prone areas. Chemotherapy errors affect multiple steps of medication use process.^[7] Medication errors in chemotherapy occur frequently and have a high potential to cause considerable harm.^[8]

Little information is available about outpatient medication errors among patients with cancer to help guide the development of interventions to improve patient safety in low-income countries in 2012. The limited data that are available suggest that medication errors in outpatient cancer patients are potentially harmful.^[9] One observational study done in Iran indicated that medication errors occurred in administration of chemotherapy and an overall 32.5% error rate were detected.^[10] Studies in chemotherapies error rate were very limited in African countries in 2012. A study from Dar El Shefaa Hospital in Egypt Hospital indicated that Medication errors were higher during ordering/prescription stage (38.1%) followed by administration phase (20.9%).^[11] In Ethiopia, studies have been done on medication errors in general rather than specifically on cancer chemotherapy. One study in 2010, in the pediatric inpatient department of Jimma University Specialized Hospital, indicated a prevalence of medication prescribing errors to be (89.9%).^[12]

To reduce chemotherapy medication errors and toxicities, increasing awareness about the occurrence and consequences,^[13] verifying doses,^[14] proper documentation, and confirmation of appropriate supportive care are all important in medication use process. In addition to medication errors, contamination of personnel who handle cytotoxic has been proven via traces in urine.^[15] Hence, the availability of separate, clearly marked clean room with engineering controls (laminar flow hood) and personal protective equipment and other safety measures for a chemotherapy service is invaluable.^[16]

Tikur Anbesa Specialized Hospital (TASH) is one of referral and teaching hospital under Addis Ababa University, and it is the training center for undergraduate and postgraduate medical students, dentists, nurses, pharmacists, laboratory technicians, and others. TASH had 600 beds and it has radiotherapy center. TASH radiotherapy center was the only governmental hospital that provided cancer chemotherapy services for Ethiopia in 2012.

The patient waiting time for chemotherapy service was approximately 3 months. Chemotherapy was given in the unit by considering various patient factors in addition to clinical findings and the patient's socioeconomic status. The center had 19 beds. The center was staffed with 12 nurses, 1 internist with palliative care specialization, 3 oncologists, 2 medical physicists, and 3 monthly exchanged residents from the internal medicine and surgery department. Pharmacists were not involved in the unit, but they gave chemotherapy drugs and advised patient in the general pharmacy directorate.

There was no separate oncology pharmacy service during the study period. The radiotherapy center offered radiotherapy, chemotherapy, and palliative care services for cancer patients. Chemotherapy admixture and administration was given by the nurses in the unit. There was no separate room for chemotherapy admixture during the study period. The center provides chemotherapy services both for the inpatient and outpatients. Mostly in an outpatient adult oncology unit chemotherapy is given for patient with breast and colorectal cancer and the rest of cancer patient is treated in the inpatient services.

This study was the first assessment of chemotherapy dose-related errors at TASH. Unfortunately, at the time of this study, there was no documentation of the availability of equipment and supplies, dose verification, documentation, or appropriate supportive care. Therefore, the present study aimed to fill this gap in assessing the medication use process in an outpatient adult oncology unit of TASH.

2 Methods

This was an institutionally based cross-sectional study conducted in an outpatient adult oncology unit of TASH in Addis Ababa between May 1 and June 30, 2012. The source and study population constituted all medical records of adult outpatients’ taking chemotherapy in the outpatient oncology unit during the study period. The data were prospectively collected from patient clinical records of each administration visit over the 2-month study period. Clinical records with what actually done during the preparation and administration period for each of chemotherapy were collected. All patients’ medical record that show patients age who are above 18 years and who were fit to receive chemotherapy (i.e., WBC count, Hgb and platelets, and other laboratory value with normal range) were included. A data abstraction format was used to assess dose-related errors (Annex 1 and 2). Error recording were done after the collection of data by pharmacist and dose were recalculated by Mosteller formula at the analysis. The abstraction format were self-made by the authors, which was modified from patient history sheet of the department patient card. An observational checklist was used to confirm dose verification and documentation process and the availability of materials and equipment for chemotherapy service.

Structured observations were made using checklists based on standards from the British Oncology Pharmacy Association (BOPA) standard (Annex 3) and Elmhurst Memorial Hospital, Chemotherapy Policy (Annex 4) for evaluating dose verification, documentation, and assessing the availability of equipment and supplies for chemotherapy (Annex 5).^[17,18] To assess the appropriateness of supportive care for chemotherapy, the National Comprehensive Cancer Network (NCCN) guidelines 2011 version 1, 2012, were used.^[19] All clinical records were reviewed and body surface area and doses were recalculated using the Mosteller equation.^[20] 

All medication orders were reviewed and body surface area and medication doses were recalculated. Medication ordered dose ≥ or ≤5% over or under the recalculated dose was considered as a dosing error (over or underdose). Laboratory tests such as Creatinin clearance and bilirubin were used to check for dose modifications in organ dysfunction.

Data for dose-related errors included dose verification and documentation process were entered and analyzed using Statistical Package for the Social Sciences (SPSS Inc., Released 2007, SPSS for Windows, Version 16.0, Chicago, IL). For these assessments, descriptive statistics, that is, percentage frequencies, were computed to determine the rate. Each dose of chemotherapy was re-calculated using Mosteller BSA calculations. For documenting the availability of materials and equipment and the appropriateness of supportive care for chemotherapy, the data were manually tabulated.

Ethical clearance for the study was obtained from the ethics review committee of Addis Ababa University (AAU) School of Pharmacy. The study was also approved by the AAU radiotherapy center. The confidentiality of patient histories was protected since the pharmacists working in the TASH pharmacy collected the data and name of each patient card was not noted during data collection.

3 Results

For a total of 212 adult patients with cancer received chemotherapy during the study period. There were 583 chemotherapy drug administration's doxorubicin (Adriamycin), cyclophosphamide, epirubicin, fluoruracil, irinotecan, methotrexate, and leucovorin. The overall dose-related error rate was 228 (39.1%) during preparation and administration of chemotherapy, excluding errors such as unlabeled doses and wrong intervals between doses. The most commonly occurring error was incorrect reconstitution 106 (46.5%) and of these, 95 (89.6%) occurred during cyclophosphamide preparation and 11 (4.8%) during adriamycin preparation. The numbers of underdoses were 58 (25.4%) and overdoses were 52 (22.8%) (Table 1).

Table 1: Frequency of dose-related errors, in outpatient adult oncology unit of TASH, 2012 N = 228.

From the chemotherapy drugs administered, 58 were found to be underdoses, which accounted for 25.4% of the total medication errors (Table 1). Epirubicin was the most common underdose 33 (56.9%) followed by cyclophosphamide 12 (20.7%), adriamycin and fluorouracil each 6 (10.3%), and irinotecan 1 (1.7%) (Fig. 1).

Figure 1: Frequency of underdose, outpatient adult oncology unit, TASH, 2012. N = 58. TASH = Tikur Anbesa Specialized Hospital.

The rate of overdose was 22.8%. The most common overdoses were cyclophosphamide and fluorouracil 12 (23.1%) and 11 (21.2%), respectively. The lowest overdose frequency was Leucovorine 1 (1.9%) (Fig. 2).

Figure 2: Frequency of overdose, in outpatient adult oncology unit, of TASH, 2012. N = 52. TASH = Tikur Anbesa Specialized Hospital.

Dose adjustment was conducted by using the patient laboratory values for, renal and liver function tests, such as creatinin and bilirubin. In 96 (16.5%) chemotherapy administrations, doses were not adjusted because of the absence of patient laboratory data, whereas 12 (2.5%) errors were recorded as wrongly adjusted doses. The most common errors occurred, whereas calculating doses of methotrexate 4 (33.5%), cyclophosphamide, and adriamycin (Fig. 3). Only 3/14 dose-verification activities were practiced in the outpatient adult oncology unit of TASH for each patient (Table 2).

Figure 3: Frequency of wrong adjustment to individual patient, in outpatient adult oncology unit, of TASH, 2012. N = 12. TASH = Tikur Anbesa Specialized Hospital.

Table 2: Assessment of dose verification process, in outpatient adult oncology unit, of TASH, 2012.

Drug and dosages administered were recorded for 38 (17.9%) of the total 212 patients. Documentation of the sequence of drugs administered was observed in 38 (17.9%) patient records for receiving chemotherapy. Specific discharge instructions and appointment dates were recorded for only 163 (76.9%) of the patients (Table 3).

Table 3: Documentation of chemotherapy, in outpatient adult oncology unit, of TASH, 2012. N = 212.

The premeditations were given as a supportive care treatment and this included metoclopramide 10 mg and dexamethasone 8 mg given in bolus form for all patients receiving chemotherapy for all regimens during the study period.

Out of the total 23 equipment and supplies supposedly available, only 8 (34.78%) were actually available and accounted for a safe chemotherapy service (Table 4).

Table 4: Availability of equipment and supplies for chemotherapy at oncology unit, TASH, 2012.

4 Discussion

The present study provided insight into the assessment of the medication use process focusing on dose-related errors, appropriateness of supportive care treatment for chemotherapy administration, dose verification and documentation of chemotherapy, and availability of equipment and supplies for chemotherapy service in the adult oncology unit of TASH. The findings indicated that the overall dose-related error rate was 228 (39.1%). In 2012, there were no mechanisms for reporting sentinel events on the unit. Even though, studies on chemotherapy errors in low-income countries were scarce in 2012, our chemotherapy error rates are higher than those reported from other studies. According to Ford et al in the United States, the benchmark for chemotherapy error rates was 0.1% to 0.2% for medication administration errors, in which these numbers seem to be relatively independent of patient age and chemotherapy versus nonchemotherapy medications. However, specific to chemotherapy administration, 0.04% error rate was recorded, which is considerably lower than in this study.^[21]

Of course, those comparably low incidences of medication administration errors may have been due to self-reporting of nurses or due to preinstalled safety guidelines and obviously such systems are not functional in adult oncology unit of TASH. If we compare our findings with the benchmarks for error rates, the rate was much higher. Another study from the United States had a 4% error rate of chemotherapy administration with intercepted medication errors.^[22] Our error was also different from an observational study done in Iran, which indicated that an overall 32.5% error rate.^[10]

Our study indicated that wrong reconstitution 106 (46.5%) was the most common dose-related errors and cyclophosphamide 95 (89.6%) accounted for the highest share. Our observation indicated that drugs were prepared within the same room where patients receive their medications. Moreover, because many patients waited in line to receive such treatment, the administering nurses often appeared to be rushed and the chances of making errors were high as shown in our results. Even in 2012, the chemotherapy preparation and administration room did not have shaker machine or any other device to reconstitute powders with diluents, which resulted in particles remaining inside vials after reconstitution. During our study period, we observed that nurses sometimes received help from the patients with mixing the medications to save time. This, not only increases the chances of errors, but also is likely to result in unnecessary patient exposure to cytotoxic chemotherapy.

Other dose-related errors findings were overdose 52 (22.8%), underdose 58 (25.4%), and wrongly adjusted dose 12 (5.3%). Our study was very far from error rate recorded in Egypt which was underdose error 99 (6.4%) and overdose of 38 (2.4%).^[23] QUAPOs recommends double-checking doses prior to chemotherapy administration.^[16] The AAU radiotherapy center was critically understaffed and did not employ such practice. Therefore, a lack of double-checking chemotherapy is one of the likely reasons for such high number of dose-related errors.

During the study period in 2012, pharmacists were not involved in medication preparation and administration processes and their inclusion could have served as an additional safety net by participating in doses verification and recalculation prior to administration. One study in Cairo, Egypt, showed that the involvement of clinical pharmacist in an oncology unit, reduce the administration error from 15% to 0.9%.^[23] Updated protocols for each chemotherapy regimen were lacking and physicians use their own preference for dosing adjustments dependent on the patient's condition.

Even though data are scarce on specific types of chemotherapy errors, one US study showed that medication errors accounted for 7.1% from these; chemotherapy error range from 0.3 to 5.8 per 100 visit.^[8] Compared to our result, this difference may be attributable to chemotherapy safety systems in community hospitals abroad and their reporting systems that help to reduce chemotherapy errors. The TASH adult oncology unit did not have these systems, and therefore chemotherapy safety activities were not practiced.

This study also revealed that the dose verification and documentation for chemotherapy was not properly organized. Even though some activities were done, there is no preprinted protocol for dose verification and documentation. Thus, the nurse's lack of training in these practices resulted in obvious chemotherapy preparation and administration errors. Study in the US and Australia have confirmed that dose verification and documentation in chemotherapy are the major activities required to reduce chemotherapy errors.^[24]

Supportive care treatments for chemotherapy (dexamethasone 8 mg and metoclopramide 10 mg) were given before administering chemotherapy regimens for all patients. There were no postchemotherapy supportive care drugs provided for chemotherapy-induced nausea and vomiting (CINV) for any patient during the study period. This is due to the unavailability of these drugs in the hospital. However, according to the European Society for Medical Oncology (ESMO) clinical recommendation for antiemetic for chemotherapy, supportive care for CINV may differ for individual patients based on the dose and type of chemotherapy drugs.^[25]

In the TASH outpatient adult oncology unit the frequently administered regimens were FAC, FEC, AC, EC, MFC, FOLFIRI (irinotecan, flourouracil, and leucouvourine). For such regimens, the ESMO clinical recommendation recommends IV antiemetic drugs such as serotonin antagonists such as ondansetrone 8 mg and dexamethasone 8 mg as a prophylaxis.^[25] Most of the chemotherapy regimens used at TASH includes drugs with moderate emetogenic potential and oral antiemetic such as ondansetrone can be given as postsupportive care treatment for CINV depending on the patient's manifestation. This finding demonstrated that the supportive care for CINV was inadequate with the exception of prechemotherapy provided dexamethasone. Chan et al 2008 demonstrated the positive role that collaboration with an oncologic pharmacist can have an appropriate management of CINV in an oncology setting in Singapore. This model can be replicated in Ethiopia and other low- and middle-income countries (LMICs) to improve patient supportive care.^[26]

Apart from the medication use process, this study found that the availability of equipment and supplies was below standard practice. The chemotherapy preparation room was not separated and was unsterile. Since chemotherapy drugs are hazardous drugs, preparation should be done with in class II biological safety cabinet in a sterile room that ensures both the protection of the product and the protection of the drug handlers.^[10] There was no biosafety cabinet in the TASH adult oncology unit. Therefore, many patients receiving chemotherapy may be at risk for infection due to contamination of the prepared admixture. Personal protective equipment for the professionals working in chemotherapy preparation and administration was not adequately supplied from the pharmacy. This might have been due to the lack of proper drug supply management or poor communication between the hospital and manufacturers available in Ethiopia to avail such supplies. Available PPE was intermittently obtained from foreign organizations supporting oncology efforts in Ethiopia. PPE is recognized as an essential safety measure for handling hazardous medications, such as chemotherapy and is noted in ISOPP Standard of practice for chemotherapy.^[27] One study has shown that chemotherapy exposure may causes secondary malignancies such as leukemia and bladder cancer reported after a latency period of 1 to 10 years.^[28] In addition, workers experience other adverse health effects due to an association between exposure to antineoplastic drugs and adverse reproductive effects, including fetal loss^[29,30] congenital malformations,^[31] low birth weight and congenital abnormalities,^[32] and infertility.^[33] Accordingly, the use of PPE is invaluable to protect professionals handling cytotoxic drugs.

Other missing equipment necessary for chemotherapy preparation and packaging included a shaker, to avoid errors from incorrect reconstitution of chemotherapy and a label-printer, for labeling chemotherapy after preparation. There was also a lack of furniture, hands free telephone, medication tracking software, various medication storage racks, medication only refrigerators and adequate hazardous waste disposal. This finding was in line with a study in Iran regarding the lack of pharmacy equipment to ensure efficient and safe preparation and administration of medications.^[34] This situation indicates that the chemotherapy service in adult oncology unit is below expected standards partly due to a lack of advocacy and training on oncology services in Ethiopia and also a low emphasis on oncology pharmacy and patient and nurse safety by the hospital leaders and policy makers.

5 Conclusions

The findings of our study indicated that the medication use process in the outpatient adult oncology unit was below expected standards based on standard recommendations (ISOPP). Higher rates of dose-related errors, inappropriate supportive care treatment for chemotherapy and lack of dose verification and documentation were observed. In a 2008 international survey of pharmacists involved in oncologic services in 34 countries including LMIC, the authors found that compliance with international standards was not universal or 100% across all countries. Patient and healthcare staff safety must be paramount, even in a low-income country like Ethiopia. Oncology clinical pharmacists can be instrumental in advocating for needed change and prioritization of this fundamental human right.^[35]

It is important to note in the year 2012, several initiatives have been started: the establishment of TASH oncology pharmacy, separate room for chemotherapy admixing, and nursing practices and standards. The impact of the integration of clinical pharmacists in patient safety is well documented, even in LMIC.^[36] In addition systems for ensuring sustained and adequate supply of medications and personal protective equipment should be established in consultation with the relevant stakeholders such as the pharmaceutical fund and supply Agency of Ethiopia, local manufacturers, and importers. Besides, policies and guidelines for chemotherapy mixing and administration service and standardization of the process workflow should be developed.

Conflicts of interest statement

The authors declare that there are no conflicts of interest.