Awareness and knowledge levels of patients regarding pre-analytical errors

Introduction

Around 70 % of medical decisions in healthcare rely on laboratory results. Mistakes in medical laboratories directly impact patient outcomes, leading to more frequent hospital visits, unnecessary interventions, and financial losses [1]. Laboratory medicine categorizes errors in medical laboratories into three stages: pre-analytical, analytical, and post-analytical. Pre-analytical errors occur during the initial stages of the process and have a significant impact on subsequent steps, contributing to approximately 70 % of all laboratory errors [2], 3].

Errors in the pre-analytical phase can be categorized into two main groups: controllable errors and uncontrollable errors. Uncontrollable variables encompass individual, physiological, and immutable factors, including biological aspects (age, gender, race), environmental factors (height, temperature, seasonal and diurnal variations, place of residence), the menstrual cycle, and medical conditions (high fever, shock, trauma, transfusion). These factors contribute to pre-analytical errors that are unavoidable. On the other hand, controllable variables can lead to pre-analytical errors that are preventable or can be minimized. These errors include the patient’s failure to adequately prepare for a blood test, physical activity, prolonged bed rest, circadian changes, travel, obesity, severe hunger/malnutrition, pregnancy, dietary habits, coffee consumption, smoking, alcohol intake, posture, and errors in blood sampling, such as mistakes during the drawing or transfer process to the laboratory [4]. Although the sample is subjected to numerous factors arising from the operational cycle within the hospital, a lack of patient awareness or failure to adhere to necessary pre-sampling protocols can lead to sample contamination or alteration even before the patient arrives at the hospital [3], [4], [5].

To minimize pre-analytical errors, it is essential to offer ongoing training to doctors, nurses, and staff in departments that send samples to the laboratory [3], [4], [5], [6]. Additionally, patients should be educated about the procedures being carried out and made aware of the factors that can influence test results. It is important to explain to patients that specific guidelines must be followed to ensure the accuracy of test results, and to clarify the reasons behind these guidelines. Emphasis should be placed on the critical importance of adhering to these rules to enable the correct interpretation of test results [7].

The “Laboratory Errors and Patient Safety” (WG-LEPS) Working Group of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) has introduced the Model of Quality Indicators (MQI) to classify and standardize errors throughout the total test process (TTP). To support this initiative, WG-LEPS organized and published consensus conferences aimed at developing effective quality assurance tools for medical laboratories. Researchers have identified acceptable ranges for the 25th, 50th, and 75th percentiles, as well as sigma values, for factors that can be controlled in the pre-analytical phase [8], 9]. However, the standardization of patient preparation for laboratory testing remains an area that requires further improvement [10].

The present study, designed for this purpose, aims to assess the knowledge and awareness of pre-analytical factors that can be controlled by individuals who visit our hospital for blood tests following a physical examination in the outpatient clinics.

Materials and methods

This questionnaire study was carried out in individuals over the age of 18 (excluding healthcare professionals) who referred to the blood sampling unit as outpatients to give a blood test between January and March 2020 and provided voluntary consent. A questionnaire of 20 questions was administered to the volunteers. The survey included 4 open-ended and 16 non-open questions with demographic information (Supplement 1). No questions were asked to interfere with personal data privacy. The Pamukkale University Faculty of Medicine Ethics Committee approved this study (approval number: E-60116787-020-567301, dated 05.09.2023), and all subjects gave their informed consent. The study was conducted in accordance with the Helsinki Declaration’s principles.

All statistical analyses were performed using SPSS 25.0 (IBM SPSS Statistics 25 software (Armonk, NY: IBM Corp.)). Continuous variables were defined by the mean ± standard deviation and categorical variables were defined by number and percent. The Chi-square test was used to compare categorical variables. p<0.05 was considered statistically significant.

Results

The study involved 328 individuals with the mean age 30 ± 14 years who were not healthcare professionals. The ratio of women and men was 55.79 % (n=183) and 44.20 % (n=145). The questionnaire is presented in Supplement 1.

In total, 3.35 % (n=11) were elementary school, 1.21 % (n=4) were secondary school, 8.84 % (n=29) were high school, 82.62 % (n=271) were university graduates, and 4.0 % (n=13) were post-graduates. Demographic information about individuals was given in Table 1.

In Table 2, we revealed the questions (6th, 8th, 9th, 11th, 13th, 14th, 16th and 19th) that show significance (p<0.05) when the participants are classified according to their educational status.

Participants mostly live in the city center (56.8 %, n=184). Also, 86.5 % (n=284) of the participants did not have a chronic disease and 14.6 % (n=48) had received healthcare at least 1 time in the last 1 month. When we classified the participants according to the last time giving a blood sample due to any complaint, 8th, 12th and 16th questions revealed statistical significance (p<0.05) (Table 3).

Regarding the question “Do you know there are rules to follow before giving blood?” 33.8 % (n=111) of volunteers answered no and 66.2 % (n=217) answered yes. Based on this information, we classified the participants who claimed to know the rules to follow before giving blood and analyzed the other questions according to this point (Table 4). In this table, 8th, 9th, 17th and 18th questions showed statistical significance.

In response to the question “Do you know how long participants should be fasting before giving blood?” 42.7 % (n=140) answered no and 57.3 % (n=188) answered yes.

In Table 5, we revealed the questions from 9th to 19th and their answers with their number and percentages.

Finally, regarding the statement “The questions you asked me showed me that there may be points I don’t know about blood sampling”, 11.0 % (n=36) of the participants disagreed and 89.0 % (n=292) agreed.

Discussion

When the relevant literature was reviewed, we found that most of the studies conducted on this subject evaluated the effect of pre-analytical factors on laboratory results. While there is a vast body of research on healthcare professionals’ training, laboratory workflow and laboratory inspection systems, there was limited data on the awareness of patients [3], 8], 9]. As is well-known, the Working Group for the Preanalytical Phase (WG-PRE) of the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) has held numerous meetings and developed guidelines and recommendations to identify, classify, and minimize errors in the pre-analytical phase [9]. However, information about the awareness and knowledge levels of individuals regarding the extra-analytical phases is still lacking.

The present study aimed to assess the awareness of patients who provided biological samples for laboratory analysis regarding controllable pre-analytical factors that can affect the results.

In our study, where there were more female participants, the highest number of participants were university graduates. Our analysis, which classified the participants according to their educational status, showed significant differences between groups in Questions 6, 8, 9, 11, 13, 14, 16, and 19 (Table 1). No significant difference was detected in terms of education in other questions. Alanizy and his colleagues used a survey containing 17 closed-ended questions, applied it to 346 participants and they found that participants with higher education levels and better jobs had a higher level of knowledge about medical errors and complications than other participants [11].

When the participants were classified according to their answers according to the last time they gave a biological sample, significant significance was detected in questions 8, 12 and 16 (Table 2). It was observed that as the frequency of participants coming to the health services decreased, the tendency to make mistakes before giving a biological sample also increased.

33.8 % of the 328 participants revealed did not know that there were rules to be followed before giving blood. Noticeably, 66.2 % of participants rated “Yes, I know” option. In the analysis made by evaluating the answers given by these people to questions 8 to 19 by separating those who answered yes to the question “Do you know there are rules to follow before giving blood?” statistical significance was detected between the answers given by individuals to questions 8, 9, 17 and 18 (Table 3). This classification we made showed that many of the things that the participants knew to be true were wrong behaviors.

We found that 42.7 % of the participants needed to be made aware of how long they needed to be fasting before giving blood. Steinmetz et al. conducted a study that compared the levels of 15 analytes studied in blood samples collected before and after a 700-kcal meal in 200 volunteers and found increased levels of AST, glucose, phosphorus and potassium after the meal [12]. Hazra and his colleagues conducted a questionnaire study for evaluating awareness, understanding, and compliance with fasting requirements with 98 patients and 187 controls. Participants who informed about fasting processes showed better compliance than the others, but the awareness about fasting still incomplete [13].

Studies report that a high-protein diet results in a four-fold increase in plasma urea concentration within four days, along with increased levels of cholesterol, phosphate, urate, and ammonia [12], 14]. Foods such as bananas, tomatoes and avocados are rich in serotonin and have been found to increase 5-HIAA excretion in urine [15].

11.9 % of our participants thought smoking 1 cigarette in the morning did not affect their test results. However, carboxyhemoglobin, catecholamine and cortisol levels increase with smoking. In addition, smoking leads to high levels of leukocyte and erythrocyte count and mean corpuscular volume [16], 17]. Smoking increases plasma catecholamine levels due to its stimulant effect on the adrenal medulla. Smoking 1–5 cigarettes daily increases plasma fatty acid, aldosterone and cortisol levels, and the associated glucose concentration also increases. In the long term, total cholesterol, triglycerides, LDL cholesterol, erythrocyte count and carboxyhemoglobin levels increase, and HDL cholesterol levels decrease. Smoking, which also influences immunity, causes a decrease in plasma immunoglobulin levels [16], [17], [18].

We found that 4.9 % of the people in our study thought that consuming alcohol the evening before giving blood did not affect test results. However, it has been found that alcohol has acute and chronic effects, such as smoking. Moderate alcohol intake can increase blood glucose concentration by 20–50 %. Chronic alcohol intake also increases the activity of many enzymes, such as GGT, AST and ALT, through enzyme induction. Coffee, cigarettes and alcohol should not be consumed during oral glucose tolerance testing due to their effects on glucose metabolism [19], 20].

Our study found that 79 % were unaware that heavy exercise and training affected test results. Sanchis-Gomar [21] and Lippi [22] revealed that individuals with elevated or decreased analytes in the blood without clinical symptoms should be asked whether they have recently done light or heavy exercise. They also recommended blood sampling at least 48 h after the physical activity [21].

We noticed that 50.9 % of the participants thought there was no difference between sitting and lying while giving blood. Lippi et al. conducted a study, with participants consisting of volunteer laboratory personnel, three sample sets were given sitting, lying, and standing. The results showed that position affected blood results [22].

In our study, 25.9 % of the participants said they thought there would be no difference between the blood tests given between morning, noon and evening hours. İhtiyar and his colleagues took separate blood samples from 17 healthy participants aged 18–50 at 9:00 am, 12:00 pm, 3:00 pm, 6:00 pm, and 12:00 pm. They found changes of up to 40–50 % in analytes such as Blood urea nitrogen, Total and Direct Bilirubin, and triglyceride between blood drawn in the morning and at other hours. They suggested that clinicians should consider this diurnal variation [23]. Another study conducted by Yucel and his colleagues, involving 12 healthy individuals, showed that the ESR test does not vary throughout the day and that it can be evaluated at any time if it is necessary [24].

In our study, 6.4 % of the participants thought that medication and treatments did not affect the test results. In the study conducted by Simundic and her study group with 200 outpatients who applied to hospitals from 18 different European countries, it was revealed that patients mainly used over the counter (OTC) drugs and dietary supplements, but did not share this with the healthcare professional who cared for them without giving a blood sample to the laboratory [25].

There are certain limitations to the present study. First, the participants who volunteered for the survey were generally young and had higher education levels, meaning our population predominantly consisted of young individuals. Additionally, since the educational level is usually higher among younger people in Turkey, the survey results reflected this trend, with 82.6 % of participants being university graduates. Another limitation was the lack of a standardized survey. Instead, we used a questionnaire that we created ourselves based on conventionally recognized controllable factors. No standardized guidelines or questionnaires were available at the time of the survey. The literature suggests further studies on the impact of pre-analytical errors on blood results using standardized questionnaires in larger populations.

Conclusions

This study elucidates the survey findings regarding the potential influence of participants’ knowledge and awareness on the accuracy of tests conducted on blood samples collected for routine analysis. Considering these results, comprehensive training should be extended to both healthcare professionals and patients. Moreover, the provision of an informational card to individuals before blood sampling for laboratory tests may contribute to a reduction in error rates.