Introduction

Therapy-associated errors such as medication errors are becoming increasingly common in public perception. Today consensus exists that medication errors pose a serious risk to the well-being of patients, which is underestimated both in terms of incidence and in terms of the impact of errors. Therefore, increasing efforts are being made to investigate their causes and effects. According to a prospective cohort study by Barker, errors occur in up to 20% of all medication-related processes, with the most frequent being drug administration at the wrong time (43%), no administration of a prescribed medication (30%), and wrong dose administration (17%) [1]. Healthcare professionals may make these errors but also patients who are self-administrating medications. Thereof, 7% of all medication errors have the potential to cause adverse drug events [1].

Up to now, there has been no consistent definition of medication errors in the scientific literature and consequently the incidence of medication errors varies from five to 967 per one thousand patient days [2–4]. The World Health Organization, for example, defines medication error as “a preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the healthcare professional, patient, or consumer” [5–7]. Conversely/Similarly, the British psychologist James Reason postulated a system-oriented error theory with the help of the "Swiss cheese model of system accidents" [8]. According to his theory, risks and errors only occur when several safety barriers, which can be both human and technical, fail simultaneously. This means that the faulty system barriers are by no means rigid, but change over time [9].

A prominent risk factor for the occurrence of medication errors could be a lack of work experience [10]. Nurses with little work experience as well as doctors at the beginning of their career are at an increased risk of making a medication error. A study examining the theoretical and practical pharmacological knowledge of undergraduate students from seventeen European countries revealed that 46% of pharmaceutical therapies established by undergraduate students were wrong and that 55% of prescriptions contained at least one error. The most frequent error was choosing the wrong active agent or prescribing a wrong, usually too high, dosage. Even students in the final year of their training did not feel confident enough to prescribe medication safely and correctly [11]. Moreover, 30% of first-year physicians rated their knowledge of clinical pharmacology and pharmacotherapy as insufficient, while only 8% evaluated their knowledge as good [12–14].

Although diabetes mellitus is not an independent risk factor for the occurrence of medication errors, patients with diabetes mellitus are at higher risk to medication errors than others [15] due to the fact that antihyperglycaemic drugs, insulin in particular, belong to the group of so-called high-alert medications [16–20]. Furthermore, despite diabetes mellitus being a complex and widespread disease of great concern not only to those affected but also to their environment, including healthcare professionals, to date there are hardly any data on medication errors from the patients’ point of view. Therefore, the aim of the study was to survey hospitalized patients with type 2 diabetes mellitus (T2D) in terms of their perception of error management, therapy-associated satisfaction, and diabetes-specific treatment aspects during their hospital stay.

Materials and methods

Study design and setting

This single-center survey was carried out from April 2017 to February 2019 at the ward of the Division of Endocrinology and Diabetology. The subjects included a hundred patients with T2D: of these, fifty patients were on oral antihyperglycaemic agents (OAH) and the other fifty were on preexisting insulin therapy (with or without OAH). The classification of the two groups was based on individual treatment regimen prior to inpatient admission independent of the glucose-lowering therapy the patients received during their hospital stay. The patients were randomly recruited and those who met all inclusion (≥ 18 years of age, type 2 diabetes mellitus with pre-existing insulin therapy or pre-existing therapy with oral antidiabetic drugs) and none of the exclusion (type 1 diabetes mellitus, gestational diabetes, other forms of diabetes mellitus, patient unable to provide informed consent) criteria were included in the survey. Furthermore, demographic data (age, sex), height, weight and health data (diabetes duration, diabetes-specific therapy, HbA1c value from the routine laboratory), were also recorded. The patients were asked to complete a 22-item paper-based questionnaire (Table 1) which was collected after completion by study staff. Data, which were collected on the paper-based questionnaire, were manually entered into a digital database. All database entries were independently double-checked for correctness by two study team members. To obtain usable answers, only patients fluent in German were invited to participate in the survey.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics.

https://doi.org/10.1371/journal.pone.0267570.t001

Results

Patients in the insulin group were older and had diabetes longer as compared to the OAH group. The proportion of women in the OAH group was 38% (n = 19) compared to 48% (n = 24) in the insulin group. The full set of baseline characteristics can be found in Table 1. Complete questionnaire results are shown in Tables 2–4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Beliefes on medication errors (group I = insulin; group O = oral antihyperglycaemic agent).

Differences were calculated between both groups (O/I), duration of diabetes (D), age (A), gender (G) and education level (E). Any significance is given in the column “p-value” (n.s. = not significant). Significant differences for duration of diabetes, age, gender and education level are indicated in the description below the table.

https://doi.org/10.1371/journal.pone.0267570.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Therapy associated satisfaction (group I = insulin; group O = oral antihyperglycaemic agent).

Differences were calculated between both groups (O/I), duration of diabetes (D), age (A), gender (G) and education level (E). Any significance is given in the column “p-value” (n.s. = not significant). Significant differences for duration of diabetes, age, gender and education level are indicated in the description below the table.

https://doi.org/10.1371/journal.pone.0267570.t003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Diabetes specific aspects (group I = insulin; group O = oral antihyperglycaemic agent).

Differences were calculated between both groups (O/I), duration of diabetes (D), age (A), gender (G) and education level (E). Any significance is given in the column “p-value” (n.s. = not significant). Significant differences for duration of diabetes, age, gender and education level are indicated in the description below the table.

https://doi.org/10.1371/journal.pone.0267570.t004

Discussion

This survey evaluated for the first time the perceptions of patients with T2D concerning medication errors, therapy associated satisfaction and diabetes-specific therapy concerns during hospitalization. So far, questionnaires on medication errors assessed only the attitude of healthcare providers towards error reporting while the patient perspective was not considered [22–24]. As negative outcome of medication errors directly affects patients it is of utmost importance to also assess patients’ perception on medication errors. To better address patients’ fears of medication errors in an inpatient setting where they are not in control of their medication, we need to understand their view on HCPs’ awareness, reporting and prevention of medical errors. Additionally, communication strategies to lay people on already established preventive error management can be adjusted to avoid knowledge gaps and fake news. Although patients with insulin-treated diabetes are classically regarded as high-risk with regard to the occurrence of medication errors, in this study no significant differences emerged with respect to medication errors when compared to patients on oral antihyperglycaemic agents alone [15–20]. In both groups, 25% of patients stated having experienced a medication error at least once. Most frequently, errors such as prescribing the wrong dosage or agent was mentioned. However, in line with the results of recent studies, only a small number of medication errors led to serious harm, while a large proportion of those had no lasting effect on patients’ wellbeing.

Of the cohort of patients with preexisting insulin therapy, only 26% assumed that medication errors that occurred were openly addressed. At the same time, slightly more than half of respondents in the OAH group (54%) believed that enough was done to avoid medication errors. Therefore, it can be assumed that there is still a chance to increase patients’ confidence in error prevention within the hospital. A survey of hospital pharmacists on medication error disclosure found, that the majority agreed that open communication is essential but differences in communication details and behavior when communicating to the patient (quote: “to give it a positive spin”) were described. Additionally, the response rate to this survey was low potentially having caused a positive selection bias overrepresenting pharmacists preferring open communication after occurrence of a medication error [25].

However, regardless of the preexisting, form of therapy the majority of surveyed patients, were not concerned about experiencing a medication error and felt that they were taken seriously enough when they expressed concerns about their medication. Furthermore, patients treated with OAH seemed to be more critical of the intra-hospital medication management than patients with insulin treated T2D, even if the differences were not statistically significant. Furthermore, patients undergoing insulin therapy were significantly more afraid of hypoglycaemic events than patients treated with OHAs. This is probably due to the fact that patients undergoing insulin therapy have already experienced hypoglycaemic episodes in the past. In addition, patients with longer diabetes duration feared hypoglycaemia more often than patients with a shorter duration of the disease. This finding may be explained by the fact that these patients were more often already receiving insulin therapy. Finally, women reported more frequently to having previously experienced hypoglycaemic episodes, which is in line with studies researching gender differences in hypoglycaemia in T2D [26]. This might also explain the greater fear of hypoglycaemia in females seen in our survey.

The fear of being subjected to a medication error has so far received little attention as a risk factor for non-adherence. In the present survey, nearly 25% of all patients—irrespective of the preexisting treatment type—indicated that they have refused to take medication at least once due to the fear of being subjected to a medication error. Chronic health conditions such as diabetes mellitus often require years of conscientious medicating to achieve the best possible therapeutic success. However, according to conservative estimates, more than 50% of all patients today do not take their medication as prescribed and thus jeopardize their own therapeutic success. The reasons for non-adherence found in the literature include inadequate health literacy, a challenging relationship between physicians and patients (especially in the sense of a paternalistic approach on the part of physicians), complex therapy regimes, communication problems, inadequate information on adverse drug events or limited access to healthcare [27]. A large majority of the surveyed patients saw stress as well as medical staff shortages as the root causes of medication errors. In addition, the lack of control mechanisms, multitasking, poor communication and similarities in drug name and design were mentioned as important medication errors triggers. Drugs with similar names [28, 29] or with similar appearance [30, 31] are a well-recognized risk factor for the occurrence of medication errors. It is estimated that between 7 and 20% of all medication errors occur due to similarly sounding name or similar physical shape of the drug [31].

In general, the vast majority of surveyed patients felt confident in their daily clinical routine and did not make any concrete suggestions for improvement. However, their desire to communicate more with the medical professionals as a measure to increase their subjective sense of safety in the treatment seemed to be a global concern. This finding is of great importance since the equally frequent request for more healthcare professionals is certainly less easy to accommodate than the desire for more time for medical consultation and education about disease and therapy. Overall, the survey revealed that, above all, improving the interpersonal relationship aspects (for example, appreciativeness, open-mindedness, etc.) helps to increase the patients’ personal sense of safety. Today, it is believed that good doctor-patient communication not only positively impacts the patient’s subjective feelings of wellbeing and their confidence, but may also improve compliance and in turn reduce the risk of medication errors [32]. Finally, the value of good communication is clearly underlined by the fact that most complaints about physicians are not based on their professional competence but on communication problems [33]. This is particularly important since it has been shown that patients who feel well looked after and adequately informed submit significantly fewer complaints and experience even less medication errors or complications, which is certainly a great advantage from a medical point of view [34].

Conclusions

The survey results clearly show that patients experienced at least one medication error during hospitalization, however, those errors only rarely led to patient harm. The survey also highlighted the value of an honest and respectful doctor-patient relationship regarding the perception of medication errors or complaints in general. In addition, the fact that the patients’ confidence in in-hospital medication management can still be improved, it also seems to be important for day-to-day clinical routine, since many patients assume that errors are more likely to be swept under the rug than openly addressed. Increasing patient education on in-hospital error management systems could eliminate these concerns and create a climate of trust that is essential for effective treatment.

Supporting information

Acknowledgments

The authors wish to express their gratitude to the patients at the Division of Endocrinology and Diabetology for their participation in the study.