Self-reported confidence and perceived training needs of surgical interns at a regional hospital in Ghana: a questionnaire survey

Background Due to disparities in their regional distribution of the surgical specialists, those who have finished “housemanship,” which is the equivalent of an internship, are serving as main surgical care providers in rural areas in Ghana. However, the quantitative volume of postgraduate surgical training experience and the level of self-reported confidence after formal training have not been investigated in detail in sub-Saharan Africa. Methods The quality-assessment data of the Department of surgery at a regional hospital in Ghana was obtained from the convenience samples of house officers (HOs) who had their surgical rotation before July 2019. A self-reported questionnaire with 5-point Likert-type scale and open-ended responses regarding the 35 topics listed as learning objectives by the Medical and Dental Council of Ghana were retrospectively reviewed to investigate the volume of surgical experience, self-reported confidence, and perceived training needs. Results Among 52 respondents, the median self-reported number of patients experienced for each condition was less than 11 cases. More than 40% of HOs reported that they had never experienced cases of liver tumor (n = 21, 40.4%), portal hypertension (n = 23, 44.2%), or cancer chemotherapy/cancer therapy (n = 26, 50.0%). The median self-confidence score was 3.69 (interquartile range, 3.04 ~ 4.08). More than 50% of HOs scored ≤2 points on the self-confidence scale of gastric cancer (n = 28, 53.8%), colorectal cancer (n = 31, 59.6%), liver tumors (n = 32, 61.5%), and cancer chemotherapy/cancer therapy (n = 38, 73.1%). The top 3 reasons for not feeling confident were the limited number of patients (n = 42, 80.8%), resources and infrastructure (n = 21, 40.4%), and amount of supervision (n = 18, 34.6%). Eighteen HOs (34.6%) rated their confidence in their surgical skills as ≤2 points. Of all respondents, 76.9% (n = 40) were satisfied with their surgical rotation and 84.6% (n = 44) perceived the surgical rotation as relevant to their future work. Improved basic surgical skills training (n = 27, 51.9%) and improved supervision (n = 18, 34.6%) were suggested as a means to improve surgical rotation. Conclusions Surgical rotation during housemanship (internship) should be improved in terms of cancer treatment, surgical skills, and supervision to improve the quality of training, which is closely related to the quality of surgical care in rural areas.


Background
Inadequate infrastructure, supplies, and human resources for essential surgical care have been observed in Low-and-Middle-Income Countries (LMICs) [1]. Regarding human resources, a severe shortage of surgical specialist workforce has been widely emphasized [2]. In the World Health Organization (WHO) African region, the density of surgeons, anesthesiologists, and obstetricians was 1.0 per 100,000 population [3], which is far behind the target of 20 to 40 per 100,000 population [4]. Inequitable regional distribution of doctors within the country has also been pointed out [5]. For example, in Ghana, although the overall doctor-to-population ratio is decreasing, 58.3% of non-specialist doctors and 68.0% of specialists are working in two major regions of the country, where 35.4% of the overall population resides [6].
Apart from short-term surgical training of nonspecialist physicians [7] or task-shifting to non-physician clinicians [8,9], training more surgical specialists has been a main stem concerning the surgical workforce. To increase the specialist surgical workforce, postgraduate surgical training in Africa has been established by the West African College of Surgeons (WACS) [10], the Ghana College of Physicians and Surgeons (GCPS) [10], and the College of Surgeons of East, Central, and Southern Africa (COSECSA) [11]. In Ghana, although formal residency training is well-developed, a low number of physicians is entering surgical residency and only a few qualified surgeons are being trained every year. In addition, it is rare for surgical specialists to work in rural areas [12]. As a consequence, medical officers who have completed the period of housemanship are the main surgical care providers in Ghanaian district hospitals [13].
A recent systematic review reported that 19 of the 34 LMICs had an internship or housemanship program [14]. Although one must complete this mandatory period of training to earn a full medical license, the quality of training during this period has not been prioritized and there is little baseline data related to it. This period of training should be considered seriously because the rates of enrollment in residency programs are low in LMICs and the majority of doctors who work in rural areas are those who have just finished their housemanship or internship [12,13]. However, it has been reported that medical officers experienced few supervised cases during their formal training, although they were responsible for performing major surgical procedures at district hospitals [12,13].
Therefore, improving the quality of the surgical education of house officers (HOs) during housemanship would be a realistic and effective approach to improve the quality of surgical care, especially in rural areas. In this study, the authors report the HOs' degree of experience and level of self-confidence in surgical conditions after surgical rotation to highlight the current status of HOs' surgical competency and provide baseline data to improve the quality of surgical training.

Overview of housemanship in Ghana
In Ghana, the two-year housemanship period consists of four six-month rotations between internal medicine, obstetrics and gynecology, pediatrics, and surgery, and takes place shortly after medical school graduation [15]. During housemanship, two of the four rotations are done at teaching hospitals or their equivalents, including regional hospitals [15]. Emphasizing the importance of surgical rotation, the Medical and Dental Council (MDC) of Ghana has provided 35 conditions or procedures that can be managed by HOs "to prepare the house officer for safe and independent practice either in the community or health facility" [16].

Study site
Located in the capital city Accra, the Greater Accra Regional Hospital (GARH) is one of the 10 regional hospitals in Ghana. The Department of surgery at GARH has 17 specialists (six general surgeons, four trauma and orthopedic surgeons, three urologists, two neurosurgeons, one pediatric surgeon, and one plastic and reconstructive surgeon). The annual volume of surgeries in the department is 1100 cases on average, 44% of cases being emergency operations. Among the specialties, the proportion of general surgery cases is the highest (43%), followed by trauma and orthopedic surgery (20%) and urology (13%). The department accommodates an average of 24 surgical HOs each term.

Data collection
From July to August 2019, a quality assessment of the Department of Surgery was conducted with a convenience sample of HOs who had their surgical rotations before July 2019 at GARH. HOs voluntarily responded to the electronic evaluation form, which was anonymized and did not collect any personal identifiable information, such as name, age, gender, or personal contacts. The duration of the respondents' surgical rotations, number of cases experienced, self-confidence regarding the 35 topics listed as learning objectives by the Medical and Dental Council of Ghana, and level of satisfaction with the surgical rotation were rated on a 5-point Likert-type scale. The details of the evaluation form are listed in Table 1. Concerning level of satisfaction and relevance, four or five points were considered satisfactory and relevant.

Data analysis
The quality assessment results were retrospectively reviewed to evaluate the surgical competency of the HOs during or after completing their surgical rotations and improve the quality of surgical training. Statistical analysis was conducted with STATA Version 15.1 (Stata Corp, Texas, USA). The aggregated scores for the number of cases experienced and self-confidence were derived from Likert-type scale scores in all 35 subdomains. The summary measures were presented in median and interquartile ranges (IQR) because the Construct validity of the level of satisfaction was demonstrated by positive correlation with the self-confidence score (Spearman's correlation coefficient 0.392, p = 0.001). Although competence and confidence do not have a linear correlation [17], it is important for the HOs to achieve some degree of preparedness at least before they are posted to district or sub-district level hospitals as independent doctors. It is widely known that multiple factors affect self-efficacy beliefs in surgery [18], and case volume or length of training has a controversial correlation [17,19]. However, the self-confidence of surgical trainees has been reported to have a positive correlation with level of satisfaction [20]. Therefore, the authors investigated the level of satisfaction as a measure of experience and self-confidence obtained during the surgical rotation.

Characteristics
The overall response rate was 95.9% (71/74). Seven of the 71 respondents who had parts of their surgical rotation at other facilities (five at teaching hospitals, two at district-level hospitals) were excluded to investigate the authentic training status of a regional hospital. Fifty-two of the remaining 64 respondents who had completed 5 to 6 months of surgical rotation in GARH were included for final analysis. Each HO has a different order of rotations between the specialties. Seven (13.5%) were in their first year of housemanship, 11 (21.2%) were in their second year, and 34 (65.4%) had already finished the period of housemanship.

Self-reported score for the number of cases experienced during surgical rotation
Although there is a risk of recall bias, the number of cases experienced was directly asked in the questionnaire because the majority of the HOs fill up their logbook at the end of the year with a higher risk of recall bias and inaccuracy. The median self-reported score for the number of cases experienced for each condition was 3.37 (IQR, 2.75~3.81). The median scores reported for each of the 35 conditions or procedures and the proportion of the HOs who had experienced five or fewer cases of them are shown in Table 2. The proportion of reported case scores for each topic is presented in Fig. 1. Eleven conditions or procedures were experienced in five or fewer cases in more than 50% of the HOs (shock, acute renal failure, chest injuries, peripheral vascular disease, typhoid, hand infections, colorectal cancer, gastric cancer, portal hypertension, liver tumors, and cancer chemotherapy/cancer therapy). More than 40% of the HOs reported that they had never experienced a case of liver tumor (n = 21, 40.4%), portal hypertension (n = 23, 44.2%), or cancer chemotherapy/cancer therapy (n = 26, 50.0%).
Liver tumors (n = 1) and portal hypertension (n = 1) were identified as items that could be removed from the checklist because of the limited number of cases. Surgical skills training with simulation or hands-on training in the operating theater (n = 1), splenectomy (n = 1), and cancer chemotherapy (n = 1) were identified as items that could be added to it.

Self-reported confidence score
The median self-reported confidence score was 3.69 (IQR, 3.04~4.08). The median confidence scores reported for each of the 35 conditions or procedures and the proportion of the HOs who rated their selfconfidence as 1 or 2 points are shown in Table 3. The proportions of reported confidence scores for each item are presented in Fig. 2. Four conditions scored 1 or 2 points on the self-confidence scale among more than 50% of HOs (gastric cancer, colorectal cancer, liver tumors, and cancer chemotherapy/cancer therapy). The self-reported confidence score was positively correlated with the self-reported score for the number of cases experienced (Spearman's correlation coefficient 0.601, p < 0.001). Those who rated their confidence scores at 4 or 5 had higher self-reported scores for the number of cases experienced (median 3.66, IQR 3.47~4.09 [confidence score ≥ 4, n = 17] vs. median 3.09, IQR 2.66~3.60 [confidence score < 4, n = 35]; p = 0.011).
Among the 11 items that the HOs had experienced five or fewer times during their surgical rotations, shock and typhoid had a relatively high proportion of HOs who rated their confidence scores at 4 or 5 (73.1 and 55.8%, respectively). Of the four items that elicited scores of 1 or 2 points in regard to self-confidence among more than 50% of HOs, more than 78% of HOs experienced five or fewer cases during surgical rotation (colorectal cancer n = 41, 78.8%; gastric cancer n = 41, 778.8%; cancer chemotherapy/cancer therapy n = 43, 82.7%; liver tumors n = 46, 88.5%). The reasons given for the lack of confidence towards these topics are listed in Table 4.

Level of satisfaction and relevance to future work
Of all respondents, 76.9% (n = 40) were satisfied with their surgical rotation and 84.6% (n = 44) perceived the surgical rotation as relevant to their future work. The respondents' level of satisfaction with the surgical rotation had a statistically significant correlation with the number of cases experienced, self-confidence, and perception of its relevance to their future work (Spearman's

Key lessons learned during the surgical rotation
The areas in which respondents felt most confidence after their surgical rotations were preoperative management, followed by postoperative management, knowledge of basic principles, and surgical skills (Fig. 3). In regard to surgical skills, 34.6% (n = 18) of the respondents rated their confidence level as 1 or 2 points. Their level of confidence regarding surgical skills was not correlated with the mean self-reported number of cases experienced (p = 0.108). Those who had more confidence in their surgical skills perceived a higher level of satisfaction after the surgical rotation (Spearman's correlation coefficient 0.375, p = 0.006).

Suggestions from the HOs
Of the respondents, 36 (69.2%) HOs responded to the open-ended portion of the survey with their suggestions for improving the quality of the surgical rotation (Table 5). Twenty-seven (51.9%) suggested the introduction of basic surgical skills training, while 14 of them emphasized the need for hands-on training. A desire for more supervision in daily practice was expressed by 18 respondents (34.6%).

Discussion
This study revealed that the HOs who had undergone training at the biggest regional hospital in Ghana experienced an average of six to 10 cases of each surgical condition or procedure during their 6-month rotations. The HOs had experienced more than six cases of benign conditions, including inguinoscrotal hernia, injury, appendicitis, acute abdomen, fracture management, intestinal obstruction, and peptic ulcer disease, with an average self-confidence score greater than 3.5. However, over 75% of the HOs had experienced five or fewer cases of cancer-related conditions (gastric cancer, colorectal cancer, liver tumors, and cancer chemotherapy/cancer therapy), with a self-confidence score lower than 2.5. As shown in the data, one of the main reasons for lack of confidence in treating the conditions was the limited number of patients with such conditions that the respondents had encountered. However, the learning objectives suggested by MDC are in accordance with cancer prevalence in the surgical field, which showed that breast cancer had the highest prevalence in Ghana, followed by liver, colorectal, and gastric cancer (71.3, 8.2, 7.9, and 3.8 cases per 100,000 population, respectively) [21]. The incidence of cancer in Africa is increasing rapidly because of changes in demographics and lifestyle as well as the urbanization of the population [22]. Due to the high prevalence of breast cancer, the HOs had experienced more than six cases of it, with a self-confidence score of 3. However, the regional hospital had inadequate case volumes for cancer treatment because the majority of cancer patients are referred to  teaching hospitals. Therefore, training for cancer treatment should not be overlooked, and solutions such as case-scenario training or rotation in oncology units at teaching hospitals should be considered to improve the quality of surgical training [23]. In addition, the current state of training at teaching hospitals should be investigated to decide where and how long the rotation should take place. The level of satisfaction with surgical rotation was higher among those who reported a higher level of perceived relevance of the surgical rotation to their future work. The HOs who perceived that the surgical rotation was more relevant to their future work may have participated more actively during the training, which may have resulted in a higher level of self-confidence and satisfaction. This observation highlights the importance of motivation, which is critical for successful learning [24]. However, external motivation should also be provided for effective training. A review of postgraduate medical education in sub-Saharan Africa suggested that unstructured programs, high workload and fatigue, and limited incentives for supervisors were challenges to conducting effective training [23]. In this study, 34.6% of the HOs rated their self-confidence in surgical skills as 1 or 2 points, and 51.9% of the respondents suggested a need for basic surgical skills training with hands-on experiences, reflecting their unmet needs in surgical skills training [23]. To improve preparedness among surgical interns, techniques like "shadowing" current interns in the United Kingdom and participating in specialtyfocused surgical skills courses called "boot camps" in the United States have been utilized [25]. As a consequence, a weekly basic surgical skills training session that includes training in instrument handling, basic suturing techniques, and knot-tying commenced in 2019 at the Department of Surgery at GARH. However, the surgical procedures that the HOs must perform at district hospitals are beyond the scope of basic surgical skills [13]. Building on basic training, structured procedural skills training with exposure to real cases in the theater or emergency room is urgently needed.
On the other hand, 34.6% of respondents focused on the unsatisfactory supervision in daily practice. "Physically and mentally remote teachers" and "theoretical, inconsistent, and irrelevant teaching, such as grand rounds or didactic education sessions" have been reported as factors related to poor clinical practice [26]. However, unlike in developed countries, where the density of the specialist workforce is high, the massive workload caused by the paucity of specialists in LMICs makes it difficult for supervisors to maintain a strong passion for medical education. Supervisors' competing responsibilities, including clinical practice and administrative work, and limited acknowledgment of such educational activities by colleagues or institutions, impede the prioritization of education [27]. Therefore, rather than relying on moral responsibility, systemic support, including peer recognition, career development, and various incentives, should be considered.
In addition, the MDC checklist included all of the "must do" procedures (laparotomy, treatment of open and closed fracture, and wound debridement) and parts of the "should do" (hernia repair, gastroscopy, cholecystectomy, intracranial hematoma evacuation, and mastectomy) and "can do" (chest injuries, retention of urine, and hematuria) procedures in the Lancet Commission on Global Surgery's list in the field of general surgery