|Year : 2022 | Volume
| Issue : 5 | Page : 199-204
Discrepancy between admission diagnosis in emergency and final diagnosis in ward and its correlation with length of hospital stay and mortality
Umer Un Nabi1, Asma Rafi2, Muzaffar Maqbool2, Parvaiz Ahmad Shah2
1 Department of Medicine, Al-Falah School of Medical Science and Research Centre, Faridabad, Haryana, India
2 Department of Medicine, Government Medical College, Srinagar, Jammu and Kashmir, India
|Date of Submission||11-May-2022|
|Date of Acceptance||30-Aug-2022|
|Date of Web Publication||31-Oct-2022|
Department of Medicine, Government Medical College, Karan Nagar, Srinagar - 190 010, Jammu and Kashmir
Source of Support: None, Conflict of Interest: None
Background: Inception of emergency medicine as a specialty in India is on a rise. Data on the accuracy of diagnosis made in emergency rooms in India is scarce and with varied results with especially no such data available from our part of the country.
Aim: The aim of this study was to evaluate the discrepancy between admission diagnosis in emergency and final diagnosis in ward and its correlation with length of hospital stay and outcome. This study was an observational prospective study.
Materials and Methods: Patients were categorised as per the International Classification of Diseases – Version 10. A total of 2000 patients were enrolled in the study. The patients were followed from admission to discharge. Data were categorised into two major groups – 'Concordant' diagnosis and 'Discordant' diagnosis. Univariate analysis was performed using SPSS version 20.0.
Results: Five hundred and fifty-three (27.65%) patients had a final diagnosis in ward discordant from the initial diagnosis in the medical emergency room. The frequency of discrepancy was highest for the genitourinary system (39%) and relatively low for neoplasm (16%). The average length of hospital stay in the concordant group of patients was 5.15 days, whereas it was 7.05 days in the discordant group (P = 0.003). There was a statistically significant increase in percentage mortality in patients when initial and final diagnoses did not match (P = 0.0005).
Conclusions: A diagnostic discrepancy of 27.65% occurred between admission diagnosis in the medical emergency room and final diagnosis in ward. The diagnostic discrepancy resulted in a significant increase in the length of hospital stay and mortality.
Keywords: Concordant diagnosis, diagnostic discrepancy, discordant diagnosis, length of hospital stay, mortality
|How to cite this article:|
Nabi UU, Rafi A, Maqbool M, Shah PA. Discrepancy between admission diagnosis in emergency and final diagnosis in ward and its correlation with length of hospital stay and mortality. Curr Med Res Pract 2022;12:199-204
|How to cite this URL:|
Nabi UU, Rafi A, Maqbool M, Shah PA. Discrepancy between admission diagnosis in emergency and final diagnosis in ward and its correlation with length of hospital stay and mortality. Curr Med Res Pract [serial online] 2022 [cited 2023 Feb 2];12:199-204. Available from: http://www.cmrpjournal.org/text.asp?2022/12/5/199/359939
Diagnosis is one of the most important tasks performed by health-care providers. Errors in diagnosis can lead from wrong to delayed treatment. There are more than 8000 diseases according to the MESH system, and uncertainty is an inherent element at each step of the diagnostic process. Even countries with advanced care than ours, the percentage of diagnostic errors has been kept at 1 in 10. Among the diagnostic errors, the most common were incorrect diagnosis and missed or delayed diagnosis. The results of studies on diagnostic discrepancies are very variable, ranging from 3% to 52% in some Indian studies.
Medical error has been defined variously, as an unintended act (either omission or commission), one that does not achieve its intended outcome, the use of a wrong plan to achieve an aim or deviation from the process of care that may or may not cause harm to the patient. Despite being an established reason for increased mortality in hospitals, no specific code has been assigned to this entity, even in the International Classification of Diseases – Version 10 (ICD-10). Hussain et al. found that 15% of diagnostic errors, which included delayed and wrong diagnoses resulted in severe harm or death. Various factors can cause a discrepancy between the admission and discharge diagnosis, including inherently difficult to diagnose, poor pre-hospital diagnostics, multiple underlying diseases, leading to confusion regarding the principal decompensating disease, errors from doctors and coding errors.,,,
Multiple studies show the variation in the results for diagnostic accuracies, ranging from 43.3% to 93.5%.,, Diagnostic accuracy is vital in an emergency as it affects prognosis, has financial and legal implications and indirectly reduces revisit to the emergency department, thereby reducing the burden on the health sector.,,, However, this is among the understudied topics in emergency medicine. There is a scarcity of data regarding this subject in our part of the world, especially with emergency medicine evolving as a specialty in India and emergencies being manned by training physicians and internists. Therefore, this study aims to determine the discrepancy between admission diagnosis in the medical emergency room and final diagnosis in ward among different diagnostic groups as given in ICD-10 and the correlation between the rate of discrepancy and length of hospital stay and mortality.
| Materials and Methods|| |
The present observational study was conducted in a 700-bedded tertiary care hospital in Kashmir Division of Union territory of Jammu and Kashmir, North India, with a population of nearly 70 lakhs in the division. This hospital is one of the oldest and the primary referral centres for the whole division, with an average admission of 100 patients daily in medicine alone, primarily through the emergency room managed by internists and training physicians with no dedicated emergency medicine department currently. The ethical approval was obtained from the Ethical Committee of Government Medical College Srinagar with approval number of 237/ETH/GMC/ICMR dated 19th October, 2018. The study was conducted over nearly 1½ years, from October 2018 to February 2020. Patients were included in the study after obtaining written signed consent from the patients/next of kin after explaining the nature of the study in local and simple language. Patients were categorised into various groups as per ICD-10. Patients admitted through the medical emergency room of the medicine department with age >18 years were enrolled for the study. Patients received from other health-care centres after being treated as in-patients, patients who expired in the emergency room, terminally ill and palliative patients, out-of-hospital cardiac arrest patients, patients referred to the medical emergency room from any other specialty of the hospital for continuation of medical care, undiagnosed patients, left against medical advice and discharge on request patients and patients who refused to be part of the study were excluded from the study. Every day four patients were selected using the method of random tables out of all the patients admitted to a particular medical unit in the past 24 h. A total of 2000 patients were enrolled in the study. The patients were followed from admission to discharge.
Statistical analysis was performed using SPSS version 20.0 (SPSS version 20.0, a statistical software developed by IBM, Chicago IL). Categorical variables were summarised as frequencies and percentages. Continuous variables were summarised as mean and standard deviation. Scale variables with non-normal distribution were summarised as median and interquartile ranges. The relationship between two categorical variables was analysed using the Chi-square test. The Mann–Whitney test was used to compare the distribution of scale variables (not normally distributed) across the categories of a dichotomous variable. P < 0.05 was considered statistically significant.
| Results|| |
A total of 2000 patients were studied. Out of 2000, 1050 (52.5%) were females and 950 (47.5%) were males. The age of the studied patients varied from as young as 19 to 101 years [Table 1]. Five hundred and thirty-four (27%) patients were initially identified as having disorders of the circulatory system, 433 (22%) were initially identified as having disorders of the respiratory system and 257 (13%) were initially identified as having disorders of the nervous system [Table 1]. Five hundred and fifty-three (27.65%) patients had their final diagnosis in ward discordant from their initial diagnosis in the medical emergency room, whereas in 1443 (72.35%) patients, final diagnosis in ward was concordant to initial diagnosis in the emergency room [Figure 1]. Among the diseases of the circulatory system, 130 (24.2%) out of 536 patients had discordant diagnosis. The most common diagnosis was heart failure among disorders of the circulatory system. Out of 236 patients with heart failure, discordant diagnosis was seen in 66 (28%) patients. The highest discrepancy was seen in myocarditis (62.5%) of patients and the lowest for stroke (11.3%). Among respiratory disorders, 144 (33.3%) out of 433 patients had a discordant diagnosis. The most common respiratory diagnosis was pneumonia. Out of 212 patients with pneumonia, the discordant diagnosis was seen in 77 (36%) patients. The highest discrepancy was seen with ARDS (47%) of patients and the lowest for COPD (27%). Among nervous system disorders, 59 (23%) out of 257 patients had discordant diagnosis. The most common neurological diagnosis was seizures. Out of 80 patients with seizures, the discordant diagnosis was seen in 14 (17.8%) patients. The highest discrepancy was seen in Guillain–Barré syndrome in 11 (46%) of 24 patients and the lowest for encephalitis in 7 (17%) of 41 patients. Among diseases of the digestive system, 49 (30%) out of 161 patients had discordant diagnosis. The most common gastrointestinal (GI) diagnosis was Upper GI (UGI) bleed. Out of 55 patients with UGI bleeding, discordant diagnosis was seen in 15 (27%) patients. The highest discrepancy was seen in acute on chronic liver failure in 4 (44%) of nine patients and the lowest for acute gastroenteritis in 2 (9.5%) of 21 patients. Among endocrine disorders, 29 (19%) out of 147 patients had discordant diagnosis. The most common diagnosis was DKA/HHS. Out of 85 patients of DKA/HHS, discordant diagnosis was seen in 10 (12%) patients. The highest discrepancy was seen in hypopituitarism, in 5 (35%) out of 14 patients. Out of 164 patients who received a diagnosis of one of the certain infectious and parasitic diseases, 60 (36%) patients had a discordant diagnosis. The frequency of discrepancy between initial emergency diagnosis and final diagnosis in ward was highest for the genitourinary system. Thirty-five (39%) out of 90 patients had a discordant diagnosis. The highest discrepancy was seen with acute pyelonephritis (45%) patients. It was relatively low for neoplasm. Eight (16%) out of a total of 50 diagnosed patients with neoplasm had discordant diagnosis. Sixteen (21%) out of 76 patients admitted with poisoning had a diagnostic discrepancy. There was no discordant diagnosis in patients with corrosive poisoning, whereas the highest discrepancy was observed in rodenticide poisoning (30%). A total of 86 patients received an initial diagnosis of one of the diseases of musculoskeletal and connective tissue. Discrepancy occurred in 23 (27%) patients [Figure 2].
|Figure 1: Discrepancy between admission diagnosis in the emergency room and final diagnosis in ward|
Click here to view
The average length of hospital stay was significantly higher in patients with discordant diagnosis. In the concordant group of patients, the average length of hospital stay was 5.15 days, whereas it was 7.05 days in the discordant group (P = 0.003). Out of the studied 2000 patients, 184 patients died and total average mortality of 9.2%. Among 553 patients with discordant diagnosis, 71 patients died (12.8%), whereas out of 1447 patients with concordant diagnosis, 113 died (7.8%). There was a statistically significant increase in percentage mortality in patients when initial and final diagnosis did not match with χ2 (2, n = 2000) = 12.11, P = 0.0005 [Table 1].
| Discussion|| |
Accurate diagnosis in emergency is pivotal in effective management and influences outcome. Despite advanced health-care systems and organised emergency medicine facilities in developed countries, errors in diagnosis in emergency departments featured as the most important reasons for claims where patients expired. Various studies conducted on diagnostic discrepancies between emergency diagnosis and discharge diagnosis have shown variable results ranging from 3% to 52% discrepancy in diagnosis. In low- and lower-middle-income countries, emergency medicine departments are yet to evolve to acceptable standards of care; studying the importance of this diagnostic discrepancy is pivotal and may provide an impetus in improving the level of patient care in emergency.
This was a single-centre, prospective longitudinal study, in which 2000 patients admitted through the emergency room were enrolled and performed over 18 months. Our patients belonged to a wide range of age groups from <20 to >80 years. An overall discrepancy of 27.65% was observed between emergency room diagnosis and final diagnosis in ward. Our study results are in close agreement with Goh and Low. In their study, 23% of patients in the general medicine group had diagnostic discrepancy. A similar diagnostic discrepancy was also found in a study by Leske et al. who found discrepancy in 26.8% of all admissions. In our study, one of the largest study groups was of pneumonia patients. We studied a total of 212 emergency room (ER)-diagnosed pneumonia patients. There was a discrepancy of 36% in the subgroup, which is in agreement with a study carried out by Atamna et al., who found a discrepancy rate of 29% between pneumonia diagnoses in ED with internal medicine ward diagnoses. Similar results were found in a multicentre analysis of ED diagnosis of pneumonia by Chandra et al. with a discrepancy of 27.37%. One of the largest studies carried out with regard to the discrepancies between emergency department and hospital discharge diagnosis was carried out by Eames et al. on a total of 6381 patients. About 47.2% of patients had concordant diagnosis, whereas 52.8% had discordant diagnosis. The background literature on discrepancy has a range of results that varies across regions, in various studies, from as low as 6.5% in certain studies. The reasons for discrepancy could be multiple, including the heterogeneity of the diseases studied, methodology used and availability of the diagnostic tools available at the emergency room and the nature of the emergency staff.
In our study, the results of discrepancy varied across organ systems involved. It was the highest for genitourinary diseases (39%), followed by infectious diseases (36%). It was relatively low for neoplasm (16%), endocrine disorders (19%) and poisoning (21%). Our study results are in partial agreement with a similar study carried out by Mihailovic et al., who found the highest discrepancy in the genitourinary system group during their study period of 2006–2013. The disagreement was lowest for neoplasms (2.2%), certain infectious disease groups (4.8%) and diseases of the blood and blood-forming organs (7.5%). Our results are in agreement with the above study with respect to discrepancy in the diagnosis of neurological disorders (23.8% vs. 23%) and cardiovascular disorders (18.55 vs. 24.2%). However, our results are not in agreement with respect to other organ systems. Our results are in partial agreement with a study carried out by Avelino-Silva and Steinman, who showed a wide range of discrepancies across organ systems, with the highest discrepancy rates for endocrine, metabolic and nutritional diseases (26%) and 20% for diseases of genitourinary, musculoskeletal and connective tissue and blood/blood-forming organs. The overall baseline discrepancy of 12.5% differed from our study, although the relative discrepancy among various organ systems with respect to each other matches our study.
The average length of hospital stay of studied patients was significantly higher in patients in the discordant group, with an average of 1.90 days increase in hospital stay between discordant and concordant groups (P < 0.001). Our observation is in agreement with the study carried out by McNutt et al. The authors carried out a similar study on 2390 patients. The patients in the admission discharge group had an average of 4.2 days of hospital stay versus 3.9 days for the non-discrepant group. There was an average increase of 0.3 days between the two groups. Although the actual numbers do not match with our study but qualitatively an increase in the length of hospital stay was observed. Similar results were found in a study carried out by Hautz et al., diagnostic discrepancy was associated with a longer, hospital stay (mean 10. 29 days vs. 6.90 days; P = 0.002). A study carried out by Avelino-Silva and Steinman found that the length of hospital stay was greater in patients with diagnostic discrepancy. It was a median of 5 days (interquartile range = 3–7 days) for patients with substantial diagnostic discrepancy and a median of 4 days (interquartile range = 2–5 days) for patients without substantial diagnostic discrepancy.
In our study, the mortality was significantly higher in patients with discordant emergency diagnosis with a mortality of 12.8% in comparison to the mortality of 7.8% in the concordant group (P = 0.0005). The overall mortality in studied patients was 9.2% (n = 182). Similar results were found in a study carried out by Avelino-Silva and Steinman, who observed almost doubling of mortality rates with respect to substantial diagnostic discrepancy (26% vs. 12%; P < 0.001).
In this study, the presence of various comorbidities in study patients was not studied. We could not elucidate the factors responsible for a diagnostic discrepancy in nearly one-fourth of patients admitted through the emergency room. Detailed insight into discordant group diagnosis and studying in detail subcodes of ICD-10 warrant further studies elaborating on those factors which will help in improving diagnostic accuracy in emergency. It would be worthwhile to see whether diagnostic discrepancy would be less under similar circumstances if emergency is manned by trained emergency physicians.
| Conclusions|| |
A diagnostic discrepancy of 27.65% occurred between admission diagnosis in the medical emergency room and final diagnosis in ward. There was a significant increase in the length of hospital stay due to diagnostic discrepancy. Diagnostic discrepancy resulted in a significant increase in mortality. We conclude that a better application and exercise of clinical skills such as history, physical examination and correlation is required to reduce the diagnostic discrepancy in the emergency department and decrease consequences.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Singh H, Graber ML. Improving diagnosis in health care – The next imperative for patient safety. N Engl J Med 2015;373:2493-5.
Graber ML. The incidence of diagnostic error in medicine. BMJ Qual Saf 2013;22 Suppl 2:ii21-7.
Chatterjee S, Ray K, Das AK. Gap analysis between provisional diagnosis and final diagnosis in government and private teaching hospitals: A record-linked comparative study. J Family Med Prim Care 2016;5:637-40.
] [Full text]
Leape LL. Error in medicine. JAMA 1994;272:1851-7.
Reason J. Human Error. Cambridge: Cambridge University Press; 1990.
Reason J. Understanding adverse events: Human factors. Qual Health Care 1995;4:80-9.
Hussain F, Cooper A, Carson-Stevens A, Donaldson L, Hibbert P, Hughes T, et al.
Diagnostic error in the emergency department: Learning from national patient safety incident report analysis. BMC Emerg Med 2019;19:77.
Foot C, Naylor C, Imison C. The Quality of GP Diagnosis and Referral. Indeed UK: The King's Fund; 2010.
Zwaan L, Singh H. The challenges in defining and measuring diagnostic error. Diagnosis (Berl) 2015;2:97-103.
Ilgen JS, Humbert AJ, Kuhn G, Hansen ML, Norman GR, Eva KW, et al
. Assessing diagnostic reasoning: A consensus statement summarizing theory, practice, and future needs. Acad Emerg Med 2012;19:1454-61.
Hautz WE, Kämmer JE, Hautz SC, Sauter TC, Zwaan L, Exadaktylos AK, et al.
Diagnostic error increases mortality and length of hospital stay in patients presenting through the emergency room. Scand J Trauma Resusc Emerg Med 2019;27:54.
Lim GH, Seow E, Koh G, Tan D, Wong HP. Study on the discrepancies between the admitting diagnosis from the emergency department and the discharge diagnosis. Hong Kong J Emerg Med 2002;9:78-82.
Amiri H, Shams Vahdati S, Ghodrati N, Irandoust A, Sadeghi H, Ahmadpour H. Comparison of presumptive diagnoses in the emergency department and the final diagnoses in the wards. Turk J Emerg Med 2010;10:164-8.
Goh SH, Low BY. Accident & emergency department diagnosis – How accurate are we? Singapore Med J 1996;37:24-30.
Chattopadhyay A, Ghosh R, Das T, Chakroborty A, Paul S, Lahiri SK. Gap analysis between provisional diagnosis on admission and final diagnosis during discharge – A comparative study. IOSR J Dent Med Sci 2013;8:28-31.
Lo SM, Choi KT, Wong EM, Lee LL, Yeung RS, Chan JT, et al.
Effectiveness of emergency medicine wards in reducing length of stay and overcrowding in emergency departments. Int Emerg Nurs 2014;22:116-20.
Chiuv-Lung W, Wang FT, Chaing YC. Unplanned emergency department revisits within 24 hours in a referral hospital. J Emerg Crit Care Med 2008;19:146-54.
Eames J, Eisenman A, Schuster RJ. Disagreement between emergency department admission diagnosis and hospital discharge diagnosis: Mortality and morbidity. Diagnosis (Berl) 2016;3:23-30.
International Classification of Diseases. Ver. 2010-(ICD-10 Brochure). 10th
revision, Volume 2 instruction manual 2010 edition.
Leske MC, Sorensen AA, Zimmer JG. Discrepancies between admission and discharge diagnoses in a university hospital. Med Care 1978;16:740-8.
Atamna A, Shiber S, Yassin M, Drescher MJ, Bishara J. The accuracy of a diagnosis of pneumonia in the emergency department. Int J Infect Dis 2019;89:62-5.
Chandra A, Nicks B, Maniago E, Nouh A, Limkakeng A. A multicenter analysis of the ED diagnosis of pneumonia. Am J Emerg Med 2010;28:862-5.
Mihailovic N, Vasiljevic D, Milicic V, Luketina Sunjka M, Radovanovi S, Milicic B, et al
. Discrepancy between admission and discharge diagnoses in Central Serbia: Analysis by the groups of international classification of diseases, 10th
revision. Iran J Public Health 2020;49:2348-55.
Avelino-Silva TJ, Steinman MA. Diagnostic discrepancies between emergency department admissions and hospital discharges among older adults: Secondary analysis on a population-based survey. Sao Paulo Med J 2020;138:359-67.
McNutt R, Johnson T, Kane J, Ackerman M, Odwazny R, Bardhan J. Cost and quality implications of discrepancies between admitting and discharge diagnoses. Qual Manag Health Care 2012;21:220-7.
[Figure 1], [Figure 2]