Clinical diagnosis by junior doctors-How confident and accurate are they ?

Background: Accurate diagnosis is necessary for effective treatment. Over the past few years it has been noticed that the junior doctors are not confident enough to make a diagnosis after initial history and examination. Aim of this study is to determine the ability of junior doctors to document a clinical diagnosis and accuracy of the diagnosis. Method: This single centre study included case records of patients admitted to Professorial Medical Unit (PMU) and Emergency Treatment Unit (ETU) at Colombo South Teaching Hospital (CSTH). The junior doctors‘ on admission diagnosis is compared with the medical consultants‘ diagnosis. Only case records of patients belonging to four common specialty domains were studied. Results: In the PMU out of 200 case records no diagnosis or symptom as a diagnosis was documented in 99(49.5%) cases and a diagnosis was documented in 68(34.0%) case records of which 53(77.9%) diagnoses were concordant with the medical consultants‘ diagnosis. When case records of patients admitted to ETU considered, no diagnosis or symptom as a diagnosis was documented in 56(56.0%) case records and a diagnosis was documented in only 21(21.0%) case records of which 15(71.4%) diagnoses were concordant with the medical consultants‘ diagnosis. Documentation of correct diagnosis improved with the grade of the doctor, from intern medical officer (IMO) to medical registrar (MR) in both study settings and also with the order of clerking. Conclusion: Recording of symptom based diagnosis or no diagnosis remains high among most of the junior doctors in all specialty domains and at all grades.


Background
The difference between a doctor and other health care personnel is the ability of the doctor to work as a clinical scientist who is able to apply the principles and procedures of medicine to diagnose and treat patients (The Medical Schools Council, 2008).Ability of diagnosis varies among individual doctors, and greatly depends on the knowledge and experience of the doctor.
While easy cases can be diagnosed by identifying the patterns of presentation, difficult diagnostic problems can be solved by generating a limited number of hypotheses early in the diagnostic process.Those hypotheses can be used to guide subsequent investigations (Elstein, 2002).Comprehensive clinical history and examination is essential to identify patterns of presentation and to formulate reasonable hypotheses.The doctor should be able to make a reasonable diagnosis by critically analysing the information that they gather from the initial history and examination.
The diagnostic ability and the accuracy of the diagnosis can be influenced by several factors.The ability of doctor to communicate effectively with the patient is one of the important factors (Leavitt & Leavitt, 1970).Published guidance has stressed the importance of the doctors to show their performance and competence when treating patients and also the importance of identifying non-verbal communication during the medical consultation (General Medical Council, 2009).
There have been many advances in the diagnostic techniques for most diseases.However, taking a detailed and accurate history and performing a proper clinical examination still remains important as it is adequate to lead to a diagnosis in more than 75% of patients (Peterson et al., 1992;Hampton et al., 1975).It has been observed that many junior doctors are not documenting a diagnosis or a differential diagnosis following clerking of acutely ill patients.A single centre audit done in UK has shown that the ability to make a diagnosis or a differential diagnosis remains suboptimal regardless of the specialty even among relatively experienced junior doctors (Bhandari, 2009).Current work place conditions and the training structures may have contributed to this.
Studies regarding diagnostic ability of junior doctors in Sri Lanka have not been published before.The aim of this study is to see whether the junior doctors are able to identify the correct system involved, ability to complete their clinical assessment by documenting a reasonable diagnosis, accuracy of their diagnosis when compared with the final diagnosis and the diagnostic ability in relation to grade of the doctor and order of clerking.

Methods
This retrospective single centre study was carried out in a Teaching Hospital in Sri Lanka.We analysed case records of patients admitted to the Professorial Medical Unit (PMU) and the Emergency Treatment Unit (ETU) at the Colombo South Teaching Hospital (CSTH) for two consecutive months.In both these settings patients were seen by different groups of doctors, including intern medical officer (IMO), medical officer emergency treatment unit (MO-ETU), senior house officer (SHO), medical registrar (MR) and medical consultant.All groups of doctors except consultants were identified as junior doctors.Junior doctors were graded according to seniority and post graduate qualifications.Thus, MO-ETU/SHO was at a higher grade than IMO due to seniority and MR was graded higher than MO-ETU and SHO due to postgraduate qualifications.Data were obtained from the case records of patients available at the record room at CSTH.The details of clinical diagnosis documented by junior doctors who clerked the patient initially were obtained together with the final documented diagnosis.
Our study methodology is similar to the single centre audit conducted in UK to assess the diagnostic activity of junior doctors (Bhandari, 2009).
The four commonest specialty domains of clinical cases were identified based on hospital statistics (Cardiovascular, Gastroenterology /hepatology, Respiratory and Neurology).The case records under the above specialty domains were identified according to the symptomatology related to the relevant system.For example, the patients presenting with exertional chest pain and autonomic symptoms were grouped in the cardiovascular domain and patients with pleuritic chest pain and productive cough were grouped in the respiratory domain.Case records of all elective admissions, deaths and those who did not have symptoms related to selected specialty domains were excluded from the study.For convenience, only case records of patients with a hospital stay of less than five days were selected.
According to CSTH admission policy, patients admitted to PMU will be first seen by IMO, SHO or MR in the ward and all patients admitted to ETU will be first seen by the MO-ETU.If further opinion is needed, MO-ETU will refer the patient to the casualty medical ward MR or if further opinion is not needed the patient is directly transferred to the casualty medical ward after initial management and once the patient is stable.Thus, IMO and MR will be seeing the ETU patients after MO-ETU as a second clerking.
Data collection was carried out until 50 case records for each specialty domain in the PMU and 25 case records for each specialty domain in the ETU were completed.Descriptive statistical methods (tables and charts) were used to describe data.Cross tabulations was used to find out the association between the categorical variables and Chi squared test was used to assess the significance.P value <0.05 was considered significant.All analysis was done using SPSS statistical package 11.

Results
Case records of two hundred patients admitted to PMU and one hundred patients admitted to ETU were analyzed.There were 181 males and 119 females aged 15-99 years with a mean of 68.5 years.

Confidence and accuracy of junior doctors' diagnosis in PMU
Of 200 case records of patients at the PMU, correct specialty domain was identified in 167(83.5%).No diagnosis or a symptom as a diagnosis (e.g.haematemesis, chest pain) was given for 99(49.5%)case records.A diagnosis was documented in only 68(34.0%)case records of which 53(77.9%) were accurate and concordant with the medical consultants' diagnosis.Out of total number of cases analyzed, the junior doctors were able to document a correct diagnosis only in 26.5% of case records (Table 1).Documentation of correct system and correct diagnosis according to the different grades of the doctor (IMO and MR) are summarized in Figure 1.Details of documentation by the SHO were excluded since there were only 7 case records of patients which carried their diagnosis.
No significant association was found between ability of identification of the correct system and grade of the doctor (P>0.05).When junior doctors' ability to identify the correct system was considered (Table 2), significant association was found, with the proportion of the cases correctly documented by the MR being significantly higher than that of the IMO (P<0.001).

Correct diagnosis
*For SHO there were only seven cases seen and therefore the data were not interpretable  2).Note: Numbers do not add up to 100 since wrong system was identified in six cases and are not included for the above analysis  (Shojania, 2003).

Figure 3: Percentage of identifying correct specialty system and making a correct diagnosis according to order of clerking in Emergency Treatment Unit
This study is mainly aimed at finding the confidence and accuracy of documentation of a correct diagnosis among casualty admissions.If the junior doctor clearly documents the most likely diagnosis or differential diagnosis after the clinical assessment, the consultant can use this information to establish the correct diagnosis and start necessary treatment without delay.Diagnostic standards can be improved by observing and questioning others who make diagnoses and also establishing more continuity of care of patients for junior doctors and allowing them to follow patients through to final discharge (Bhandari, 2009).

Figure 1 :
Figure 1: Percentage of doctors of different grades and their ability to identify correct system and making a correct diagnosis in the Professorial Medical Unit

Table 2 : Relationship between ability of documentation diagnosis and grade of doctor in PMU
Numbers do not add up to 200 since wrong system was identified in thirty three cases and six cases were seen by the SHO and are not included for the above analysis.

Table 3 : Relationship between ability of documentation of correct diagnosis and grade of second clerking doctor in ETU
Variable educational standards and teaching structures of different universities can influence the doctor's performance.Group of Shojania, K.G., Burton, E.C., McDonald, K.M. & Goldman, L. (2003) Changes in rates of autopsydetected diagnostic errors over time: a systematic review, Journal of the American Medical Association 289, pp.2849-2856.The Chief Medical Officers of England, The Medical Schools Council (2008) The consensus statement on the role of the doctor, Available from: http://www.medschools.ac.uk/AboutUs/Projects/D ocuments/Role%20of%20Doctor%20Consensus %20Statement.pdf[Accessed 16 th Oct 2011].World Health Organization (2007) 11 Health questions about 11 SEAR Countries, India: WHO Regional Office for South-East Asia, Available from: http://www.searo.who.int/en/section1243. htm [Accessed on 16 Oct 2011].