Diabetes related knowledge among residents and nurses: a multicenter study in Karachi, Pakistan

Since the prevalence of diabetes has been rapidly rising, it has become one of the major public health problems. According to latest data released by an IDF, the worldwide number of people with diabetes has reached 366 million contributing to approximately 4.6 million of deaths [1]. In terms of country wise prevalence if current trends prevail, Pakistan would become tenth leading nation in the world by 2030 comprising of more than 11 million individuals suffering from this burgeoning epidemic [1, 2].

Not only the prevalence is high but Pakistanis also develop diabetes and associated complications at a relatively younger age [3] which are the most productive years of one’s life, posing a significant health burden at individual, household, community and health sector levels. This becomes a significant concern in a country like Pakistan where non-availability of national health and scarcity of a health insurance system, the patients have to bear the health cost expenses out of their own pocket. The overall mean economic cost (both direct & indirect) spent by individuals to visit their physicians for outpatient treatment of diabetes is around Rs.2, 070 per visit in Pakistan [4]. The situation becomes gloomier if the inpatient cost of diabetes care is included in the overall expenses. Therefore, the role of a health care professional becomes pivotal in providing safe, effective and evidence based care to these patients to achieve optimal clinical outcomes. Moreover, clinical benefits of adequate glycemic control have been consistently demonstrated in numerous clinical studies [5, 6]. Since trainee residents and nurses are involved in provision of primary medical care to the patients where diabetes as a primary diagnosis or is one of the comorbid conditions, it is essential that they should be well equipped in terms of their knowledge to deal with this epidemic effectively. There are several studies that have looked into the issue of diabetes related knowledge among health care professionals [7–9] and in Pakistan such work includes a study done by Shehra et al [10]. However, there has been no exploration of diabetes knowledge of nurses and trainees in Pakistan. Assessment of knowledge among resident trainees is also very important since majority of patients admitted in hospital especially non-medical admissions (surgery, orthopedics etc.) have underlying diabetes which could lead to adverse clinical outcomes if not managed efficiently.

Inadequate understanding and knowledge, related to the time action profile of both premixed and new insulin analogues could also give rise to unfavorable patient outcomes. In addition, health care providers also need to be fully equipped in terms of their knowledge to face medical emergencies related to diabetes since they can occur in any inpatient unit.

The purpose of this study was to assess the knowledge related to both inpatient and out-patient aspects of diabetes management among the trainees in different specialties (Internal Medicine, Family Medicine & Surgery) and nurses and to identify the major areas of deficiency requiring educational reinforcement.

This study is of paramount importance because to date, in our literature search we didn’t find any study evaluating diabetes related knowledge of both trainee residents and nurses in this region. Our goal was to recognize the deficiencies and then to incorporate the themes accordingly in future diabetes education programs. The ultimate aim is to improve the quality of diabetes care in order to reduce the complication rate and health burden of this epidemiologically important chronic disease.

This was a Cross sectional, multicenter study. A pre tested validated questionnaire (Appendix) with test- retests reliability of 0.71 consisting of 21 open ended questions related to diabetes awareness. This was acquired from a study done at Thomas Jefferson University Hospital, Philadelphia with the permission of primary author (Rubin et al) [11]. The questions were divided into categories of outpatient (items1-4), medications (items 5-10) and in-patient (items 11- 21).

Furthermore the questionnaire was assessed for content validity by our team of Endocrinologists and nurses with expertise in diabetes care. A 3 member panel (Endocrinologist, Endocrine trainee and one nurse with diabetes teaching background) outside the study team was set up. The questions were assessed qualitatively keeping in view the training system and curriculum of residents and nurses to avoid asking unrelated questions. The content was rated valid by the panel on clarity and relevance and no significant issues were identified.

Furthermore, a pilot of the study was performed in one of the institute initially in separate group consisting of nurses and trainees not included in the final study to assess the level of difficulty of the test.

The questionnaire was administered without any modification at 5 tertiary care university/teaching hospitals in Karachi, Pakistan through non probability purposive sampling method to internal medicine residents (IMR), family medicine residents (FMR), surgery residents (SR) and registered nurses (RN) in wards or at regular meetings after taking their informed consent verbally. Ethical approval was obtained from the Aga Khan University Ethics Review Committee.

Although there were variations in terms of training of residents and nurses among institutes most probably related to limited availability of financial and personnel resources.

The study coordinator visited the selected institutes during working hours and trainee resident and nurses were asked to fill in the forms under his supervision after explaining the purpose of the study in detail and with their consent. Mostly, the participants were approached right after their teaching sessions to catch the maximum number of trainees.

The total time allocated to fill the questionnaire was 15 minutes. The responses were kept anonymous.

Statistical analysis was performed using SPSS software version 16. Frequency and percentages were reported for qualitative variables like correct response of ADA guideline for HbA1c, medication etc. and mean with SD presented for quantitative variables (Specially mean percentage score).

The content analysis strategy was used for coding open ended answers. The reason for using this strategy was the fact that some of the answers were descriptive. Initially coding was formulated by the study team (consisting of Endocrinologist and diabetic nurse) after thoroughly reading all the answers and classified as correct (1) or wrong (0), however, about 5% of the missing or inadequate responses were left uncoded. The resulting coding was again later independently assessed by the study investigators and inter-rater agreement was assessed.

Unanswered answers were counted as wrong (score 0) with an assumption that participants most probably were not aware of the correct answer in view of the open ended nature of the questionnaire. The total score was acquired by adding respective points which were then expressed as mean percentage.

The study participants were characterized using the following variables: professional category (nurse/trainee residents) and years of training. Differences in proportion among specialties were assessed by Pearson Chi-square test however means differences were compared by Analysis of variance (ANOVA) and multiple comparisons within specialty were assessed by Bonferroni test. The survey had a good reliability coefficient (Cronbach α of 0.81). Statistical significance was considered to correspond to a P value <0.05.

Out of the 472 doctors and nurses who were administered the questionnaire 381(80%) responded. The response rate varied between different groups of resident trainees and nurses with rate of 90% from IMR, 83% from FMR, 75% from SR & 73% from RN.

The majority of IMR were in their first year of training (n = 59) followed by approximately equal number of participants from second to fourth years. In addition, the level of knowledge showed improvement from their first to second year of training (p = 0.03). Thereafter, their level of knowledge didn’t demonstrate any improvement. On the other hand in comparison to IMR, the level of knowledge of FMR & SR didn’t show improvement with level of experience at all (p = 0.46 & 0.80 respectively).

Some of the questions answered correctly and stratified by type of participant are outlined in table 1.

In the questions addressing out patient management of diabetes, sixty three percent of the participants were aware of the current American Diabetes Association (ADA) recommended goal for HbAIC (nonenzymatic binding of glucose with free amino group of globin chain), but only thirty four percent of the participants knew about the low density lipoprotein cholesterol (LDL) goal for diabetic patients without known ischemic heart disease. After stratifying the results by the participant’s specialty, FMR scored better compared with IMR, 74% vs. 50%, on this particular question (p = 0.018). Two thirds of both IMR and FMR were aware of the criteria for diagnosing diabetes, in sharp contrast to only sixteen percent of SR (p = <0.001). FMR outscored IMR on number of questions related to outpatient management of diabetes but it was not statistically significant (p = 0.12) (Figure 1). Regarding the knowledge about oral anti diabetic agents, 59% of the responders were aware of the fact that metformin (biguanide) is contraindicated in renal insufficiency. Assessment of knowledge about the most commonly used form of insulin in hospitalized patient revealed that less than half of the participants knew about the peak action of regular insulin. Similarly, only 38% of all participants were familiar with the duration of action of various insulin analogs with RN’s displaying the lowest score (2.0%). More than half of the responders were familiar with the exact components of humulin70/30 insulin. IMR scored highest on items pertaining to the knowledge related to oral anti diabetic agents and various insulin preparations compared to FMR but this differences was not statistically significant (p = 0.29). When it came to treatment of mild hypoglycemic episodes RN’s score was similar to that of IMR (71% & 73% respectively). On the other hand only seventeen percent RN’s were aware of the correct treatment for severe hypoglycemia compared to IMR score of 35%. Overall 46% of the participants recognized that continuous intravenous infusion is the ideal mode of treatment for type 1 diabetics undergoing major surgical procedure. No statistical difference was found regarding knowledge of in-patient management of diabetes between IMR and FMR (p = 0.175) (Figure 2). Although the mean percentage score of nurses related to in-patient management was 40% ± 20 considerably better than their out-patient knowledge of 16% ± 18 but still very low in comparison with other published studies. Furthermore, knowledge about insulin and oral anti-diabetic medications was also found to be suboptimal among RN and SR with scores of 27% ± 20 & 34% ± 24 respectively (Figure 3).

Management of Diabetes among other non-communicable diseases has recently drawn substantial attention due to its associated complications and socio-economic impact.

Lack of knowledge among health care providers has been found to be one of the major obstacles in the management of hyperglycemia [13, 14]. Among health care providers, our trainee residents and nurses possess a significant role in provision of care to these patients. Before developing a teaching material for educational activities for health professionals, a comprehensive assessment of health professional’s current knowledge must be done. The present study assessed whether our trainee residents and nurses are well-equipped in terms of their knowledge to manage patients with diabetes. Identification of the gaps in the knowledge would help us in highlighting these areas more specifically in future educational programs with an ultimate aim of improving diabetes care delivery in Pakistan.

According to this current study, the overall mean score of the participants was found to be quite low which points towards the inadequacy of diabetes training. The results were even lower to the study, from which the tool was acquired (61%) [11] and other study done in UK [14]. In contrast, the results were better than one of the recent study done in Switzerland (43 ± 22) [12] but the difference could be attributed to the use of a slightly advanced questionnaire. The results were also low in comparison to the previous study investigating the diabetes related knowledge among family physicians in Pakistan (62%) [10]. These results are distressingly low keeping in view the exceedingly high number of diabetics in our country. Since the burden of non-communicable diseases like diabetes not only depends on preventive measures but also on the optimal care provided to existing cases.

The expected increment of knowledge from first to second year of IMR truly reflects impact of good quality postgraduate training. But subsequent lack of improvement consistent with findings from another study [15] points towards the lack of knowledge among senior medicine house staff. Interestingly, one of the intervention studies revealed that even with education the knowledge plateaus at a senior level [16]. In comparison, FMR & SR didn’t show any significant augmentation in their knowledge with their level of experience pointing towards need to add more diabetic sessions and lectures in their curriculum. These findings demonstrate that educational intervention should focus all PGY levels with greater emphasis on reinforcement of the same knowledge through lectures and tutorials. Furthermore, the gap in diabetes related knowledge among SR found in our study and two other previous studies [11, 12] is a source of concern since diabetic patients are extremely prone to adverse outcome of surgery if not managed optimally. Moreover, majority of the patients admitted under surgical care have underlying diabetes and evidence suggest that good glycemic control reduces the rate of blood transfusion, the need for dialysis, critical illness polyneuropathy, blood stream infections and hospital stay [17]. One could strongly argue that frequent consultation with diabetologists could resolve this issue. However, reluctance of health care providers to follow the recommendations of endocrine service was one of the most frequent barriers acknowledged in a study to control hyperglycemia in hospitalized patients due to various misconceptions [14]. Hence, to prevail over this clinical inertia, residents including SR need to have basic knowledge concerning management of hyperglycemia. Insulin is identified as one of the highest risk medications, globally and with an advent of new insulin analogues it is observed that most of the trainee residents and nurses are not very well acquainted with these new insulin formulations, so the chances of errors are amplified. Our study clearly proved this and demonstrated that majority of the participants especially nurses were not aware of the time action profile of these new insulin analogues.

The most serious problem recognized in this study was inadequate knowledge of RN regarding management of diabetes. Previous studies have also revealed the deficiency in their knowledge with scores of 41-72% [9, 11, 12, 18, 19] but our results are quite inferior to the results of above mentioned studies. Nurses play a major role in the treatment of diabetic patients and should be well equipped to perform recommended standards of care. A Cochrane review revealed that trained nurses can indeed replace physicians in various aspects of diabetes management [20].

In summary, this study demonstrates considerable gaps in diabetes knowledge of all survey participants especially residents in surgery and nurses which indirectly reflects the quality of care provided to diabetic patients. This led us to conduct the sessions and seminars for residents and nurses to teach them the basic knowledge regarding the management of diabetes focusing mainly on in patient component. Subsequently, standardized protocols and algorithm (both hand written & computerized) have been designed for management of inpatient hyperglycemia based on latest published guidelines for guidance of our trainee residents and nurses. In addition, online education courses for physicians, trainee doctors and nurses is planned to be introduced to fill the gap in diabetes education. There is further need to assess the impact of these interventions on overall knowledge of our trainee residents and nurses in the form of post intervention survey in future.