An analysis of computer-assisted pre-screening prior to elective surgery.
In our institutions, the initial assessment by an anaesthetist for an elective surgical patient can take place either in a separate outpatient department (OPD) consultation some weeks prior to surgery, or on the day of surgery (DOS), dependent upon surgical complexity and available information regarding patient health from sources such as referring doctor, surgeon and patient-completed health questionnaire. This is a common preoperative model in many hospitals but variation can exist in the timing and extent of this first assessment depending upon hospital capacity and patient casemix. There are clear cost and convenience differences between the two pathways, and potentially a large penalty for choosing the wrong pathway for a given patient. However, having higher quality patient information well in advance of surgery could further refine decision-making and safely allow some patients who are currently sent for OPD assessment to be appropriately first seen on the DOS.
In practice, much of patient assessment relates to patient history, and tools for remote patient assessment and triage have previously been shown to be effective (5,6). We are currently exploring the concept of obtaining patient histories by telephone, using a detailed computer-assisted self-assessment questionnaire conducted by non-clinical personnel. We have nominally termed this process Computer Health Assessment by Telephone (CHAT). This study was conducted in parallel with our current OPD preoperative practice to formally evaluate the quality of the specific elements obtained by remote pre-assessment and whether the information provided by CHAT was of sufficient quality to further optimise the triage of patients currently attending OPD and correctly assign them to the appropriate pathways as outlined above.
MATERIALS AND METHODS
This study was performed using patients from the Royal Adelaide and Royal Perth Hospitals, both large public tertiary referral centres. The Research Ethics Committee of the Royal Adelaide Hospital considered this to be a management improvement or quality assurance exercise which could be undertaken without committee approval. Ethical approval for the study was granted by the Research Ethics Committee of Royal Perth Hospital (EC 2009/129). This work was conducted within the confines of the standard preoperative pathway of both institutions. Other than receiving a phone call prior to their scheduled outpatient clinic visit, no change in patient management or treatment occurred as a result of inclusion in this study and all decision-making was based on outpatient clinic findings. Patients were informed of the process and provided with the option to not participate.
Patients in the study were aged between 18 and 75 years, scheduled for elective non-cardiac surgery with a minimum of one night postoperative hospital stay and booked into the preoperative OPD for assessment by an anaesthetist. These patients have more complex conditions and/or surgery than patients who have already been streamed to Day Surgery and DOS assessment, and hence provided exposure to a broad range of medical conditions. There was no formal patient sample size calculation, the number of participants being as large as possible within the dictates of the practicality and expense of conducting and coordinating phone pre-assessment and scoring sessions.
The initial CHAT question set was derived from a range of existing hospital written questionnaires with specific relevance to anaesthesia and perioperative medicine. Based on pilot study experiences in 151 patients, a final question set was determined and formally evaluated in this study. The computer-guided questionnaire contains a maximum of 313 separate questions with branching logic whereby primary questions may lead to follow-up questions to clarify a medical issue. For example, a positive response to a history of angina will lead to questions on frequency, precipitants, treatments etc. Hence, the total number of questions asked depends upon the patient's medical history. The questionnaire and charting system uses custom-written software with selectable check-boxes and limited free-text entries to record patient responses.
Approximately one to two weeks prior to attending OPD, patients were contacted by telephone and underwent CHAT conducted by non-clinical staff. Interviewers were instructed to adhere to the computer scripted question and response options throughout. At the conclusion of the interview a summary of the CHAT was automatically produced, the layout mimicking existing preoperative assessment forms at each institution to assist clinicians to navigate and utilise the responses in their assessment in OPD.
Patients were contacted using either cold calling (with a maximum of three attempts) or at times pre-arranged by OPD booking staff. Patients with pre-arranged interview times were instructed to be prepared to answer questions related to their health and have details regarding current medications, height and weight ready for the interviewer. Following CHAT, patients were seen in OPD for conventional assessment by an anaesthetist. To trial intended clinical practice, the anaesthetist was provided with the CHAT summary to assist them in their assessment. They were also asked to manually amend the CHAT summary, to identify any errors made by the CHAT process.
Qualitative panel review
The anaesthetist's OPD assessment and the amended CHAT summary were reviewed at a later date by panels of three to six members drawn from a pool of 50 experienced anaesthetists from eight public hospitals in South Australia and Western Australia. All reviewers were familiar with the perioperative process and routinely performed preoperative anaesthetic assessments on these types of patients. All patient and staff identifiers were removed prior to review. Summaries from both the CHAT and OPD assessment were examined together by each panel, and each summary evaluated in the following manner.
Quality of data collection
Medical conditions or treatments considered of relevance to anaesthesia or surgery were classified under the term "medical indicators" and counted for each patient. Examples include heart disease, gastro-oesophageal reflux and vulnerable dentition. For each medical indicator in each patient, the panel provided a score of the accuracy and quality of information obtained from both assessments as follows: A=dealt with adequately; B=minor error or omission, which would not impact on patient health or delay surgery: could be successfully resolved by clinician phone follow-up or on the day of surgery; and C=major errors or omissions, which could compromise patient safety or result in surgery cancellation if not identified prior to surgery.
Streaming to care pathways
With the information provided by the CHAT summary alone, the panel was asked to provide an opinion on whether the patient could have safely bypassed the OPD assessment and been first seen by the anaesthetist on the DOS. This opinion was also separately sought from information provided by both the CHAT and OPD summaries, to detect situations where CHAT could have misled clinicians.
OPD assessment by anaesthetist represents the best available reference against which to compare the performance of a preoperative health assessment (7) and the overall quality of the CHAT assessment was compared to OPD assessment using Wilcoxon signed-rank test after converting A, B, C scores to a numeric value.
To determine the potential accuracy of a patient triage system based upon clinician review of CHAT summaries, the sensitivity and specificity of patient suitability for DOS assessment was calculated by comparing panel recommendations for DOS assessment based upon CHAT alone, to DOS recommendations based upon review of the OPD summary.
Data were analysed using the R-project statistical package (R Foundation for Statistical Computing, Vienna, Austria) and GraphPad Prism (GraphPad Software, San Diego, CA, USA). A P value of <0.05 was considered significant.
A total of 514 patients completed CHAT (291 cold-called; 223 pre-arranged) and OPD assessments.
Mean patient age was 49.8 years (SD 14.4). Surgical severity grading based upon National Institute of Clinical Excellence guidelines (8) indicates that 306 (59.8%) of patients were scheduled for major procedures (Table 1).
An average of 58.7 (SD 13.2) questions were asked per CHAT, with an average interview duration of 13.8 minutes (SD 4.7; range 6.0 to 35.4). There were 177 (34%) patients from non-metropolitan areas located at least 50 km and up to 2600 km from the site of their outpatient clinic appointment, a pattern of referrals common to both tertiary centres.
Failure to contact patients to conduct CHAT occurred in 31.1% of cold-called patients but in only 6.5% of patients with pre-arranged interview times. Once contact had been established, CHAT was abandoned in 3.8% of pre-arranged and 5.4% of cold-called patients. Reasons for abandoning CHAT observed with both contact methods were: changing or cancelling surgery dates (2.0%), communication issues (1.3%) or patients found it inconvenient to complete CHAT at the time (1.3%).
A mean of 6 (SD 2.6; range 0 to 17) significant medical indicators per patient were noted. The number of medical indicators increased with both age and American Society of Anesthesiologists (ASA) physical status classification. Many patients had serious and multiple co-morbidities, with 107 (21%) patients classified as ASA physical status III or greater. ASA classification was not recorded by the OPD anaesthetist in 81 (15.8%) patients. The most commonly recorded significant medical indicators are shown in Table 2.
Qualitative panel review
A total of 2852 individual medical indicators in 514 patients were reviewed by specialist panels. Panel review considered that 2315 (81.2%) of these medical indicators were described adequately (A scores), 434 (15.2%) had minor errors or omissions (B scores) and 103 (3.6%) medical indicators with significant implications for patient health were described inadequately (C scores) by CHAT. There was no significant difference in the quality of medical indicator descriptions between CHAT and OPD assessment (Figure 1).
In general, CHAT performance was reduced in areas which relied upon physical examination, such as airway assessment, dentition descriptions and identifying previously undiagnosed cardiac abnormalities such as cardiac murmurs. Imperfect description of important patient medications was a feature of CHAT assessment in the cold-called patient cohort, with 3.8% of medication summaries containing a major error (C score), but less common in pre-arranged interviews with only 0.45% of medications summaries containing a major error. The most common errors or omissions from CHAT are shown in Table 3.
[FIGURE 1 OMITTED]
Streaming to care pathways
Decisions on patient streaming were available for 512 (99.6%) of the 514 patients. The review panels considered that with the information provided by CHAT, 398 (60%) patients could have safely been evaluated on the DOS, thus potentially avoiding the need to separately attend OPD for assessment by an anaesthetist. The sensitivity of the CHAT summary to correctly determine which patients were suitable for DOS assessment was 98% (95% confidence interval 96 to 99%). Panel review determined that 7 (1.4%) patients would have been wrongly designated as being suitable for DOS assessment if the decision for streaming were based upon CHAT alone, yielding a specificity of 97% (95% confidence interval 92 to 98%) (Table 4). The errors in these cases were caused by CHAT failing to record or fully describe prednisolone usage; previously undetected cardiac murmur discovered in OPD; cognitive issues; significant cardiac disease X 2; family history of anaesthetic morbidity; patient anxiety. The panel's opinion on requirement for OPD was strongly, but not exclusively, associated with patient health complexity and the number of noted medical indicators, with 85% of ASA I, but only 32% of ASA III patients, deemed suitable for DOS assessment (Figure 2). Initial sentinel indicators of the need for OPD assessment include significant cardiac history and history of previous anaesthetic adverse events such as awareness or regional block complications. The 10 medical indicators most frequently identified in patients requiring OPD assessment are shown in Table 5.
In general, the CHAT process performed well in identifying and displaying medical issues which require attention prior to surgery, suggesting it may suit the purposes intended in its development. From panel discussions it was clear that the decision tree structures of electronic rather than paper-based questionnaires allowed positive answers to be elucidated to a greater degree to ascertain their significance, and that a legible, well laid out display of significant responses made navigation of summaries easy for anaesthetists. Further, the structured and consistent coverage of patient assessment, in a manner not unlike preflight checklists, provides confidence in the process and the quality of preoperative patient information gathered. The use of such a standardised approach is increasingly important in areas such as formal patient handover (9).
Not surprisingly, complex health issues and serious co-morbidities were often less well detailed by the CHAT process than through face-to-face assessment. However, in 98.6% of cases, CHAT recognised these issues sufficiently well to identify patients who required OPD work-up prior to surgery. Overall, this appears to compare favourably in terms of quality with approaches used in other systems. For example, a nurse-based clinic developed to assess patient readiness for surgery has shown a sensitivity and specificity of 83 and 87% respectively, with 1.3% of patients incorrectly classified as ready for surgery (7). In other settings, the use of a nurse-based pre-assessment system has resulted in cancellation rates on the day of surgery due to inadequate medical assessment of less than 1% (10,11). While effective, these systems require training of nurses, add to the significant nursing workforce shortages and still require patient attendance at a clinic. Online pre-assessment has been shown to allow 35.6% of cases to be streamed to DOS assessment by an anaesthetist as part of a preoperative system6. The current study, perhaps because of the detail obtained through the extensive question set, provided a potentially higher DOS assessment rate (60%), despite not including the non-complex patients undergoing minor surgery already streamed to DOS assessment. Addition of that cohort to this model would further increase the percentage of patients suitable for DOS assessment.
Although it is apparent that errors also occurred with formal OPD assessment, it is valuable to consider the indicators imperfectly addressed by CHAT, to develop strategies to minimise their impact. While difficult airway descriptions featured as common CHAT errors, the need for physical examination prior to DOS can often be inferred from a range of other physical and medical indicators identified during CHAT. The presence of other co-morbidities (such as obesity, obstructive sleep apnoea or the nature of surgery) frequently indicated a need to attend OPD in these patients. Further, panel discussions suggested that, even following OPD assessment, the procedural anaesthetist is still required to formally examine the airway to develop an airway management strategy in these patients and disparity often exists between the OPD and procedural anaesthetists' definition of a potentially difficult airway. The increasing availability of sophisticated airway equipment, such as videolaryngoscopes, may reduce the risk of cancelled surgery due to a potentially difficult airway but, until tested in situ, the full impact of patients presenting for day of surgery assessment with an unanticipated airway problem is unknown.
Unsuspected cardiac murmurs are an issue for many forms of screening and could lead to deferral of surgery in a small percentage of cases if history alone is used as a triage tool. Mechanisms to address unexpected cardiac murmurs potentially include good clinical training, appreciation of the clinical signs of significant murmurs (12,13), good data from referring primary care providers and the availability of bedside diagnostic tools such as portable transthoracic ultrasound.
Scheduled interviews, in a manner not dissimilar to the UK National Health Service based 'At your convenience' telephone preoperative assessment system (14), appeared to reduce the incidence of medication summary errors. They could also provide opportunities for patients to gather other relevant information (medical referrals, test results, etc.) and/or for involvement of carers to assist in communication problems. Integration with other information sources such as general practitioner summaries and electronic patient records could also improve the quality of pre-assessment. Prior to conducting the interview the CHAT could be seeded with data from these sources to improve the patient health summary and refine questioning. Ongoing quality review and improved question sets can also address systematic errors and simple omissions, such as the presence of vulnerable teeth.
There are a number of potential limitations to this study, and sources of bias, which limit the formal analysis of the data. The fact that some patients could not be contacted is a potential source of selection bias. These patients may have a different spectrum of disease and capacity to answer questions, although there were no apparent large differences in data quality between cold-called patients versus those with scheduled calls. Anaesthetists conducting OPD assessment had access to CHAT summaries, in order to mimic intended clinical use, which influenced their own assessments and is likely to have advantaged the quality of OPD assessment. Panel members were not blinded to the two forms of assessment, potentially influencing their opinions. However, the use of multiple panels, with the inclusion of a large cross section of experienced anaesthetists, was intended to obtain opinions from a large body of anaesthetists in two states. The value of the CHAT summary to the procedural anaesthetist is critical and while panel members were accustomed to interpreting OPD assessments performed by others for the benefit of the procedural anaesthetist, the opinion of the actual procedural anaesthetist was not sought and the degree to which errors or omissions may affect service is not known. An extension to this work is planned to examine any reduction in workload in those patients still requiring OPD assessment, the actual cost savings of such a triage tool and the value and impact on day of surgery services.
This study has begun to identify patient and surgical factors which determine clinicians' decisions to stream patients to OPD versus DOS assessment. As shown in Figure 2, simple decisions based solely on factors such as ASA classification will not be particularly accurate. Predictors, identified by phone, could be used as a guide to streaming to DOS assessment, and, if sufficiently sensitive, specific and practical, the load reduction on outpatient clinics would be substantial. Formal identification of predictors which influence clinicians' decisions on matters such as referral to DOS assessment, need for high dependency or intensive care unit availability and ordering of preoperative investigations is the subject of a separate analysis currently underway.
How CHAT is deployed and the potential impact and utility of this type of pre-screening tool will of course depend on the model of care currently in place and the circumstances of the patient population. The example of patient triage presented in this study examined the value of pre-screening in a single element of the preoperative process assuming relatively simplistic streaming to either OPD or DOS. In reality, any model of care is likely to be unique to an institution and there will be variation in pre-screening's adoption, execution and integration within existing elements of the process. For instance, in some hospitals, nurses or clerical staff may be a logical choice to conduct CHAT and even perform simple first-pass triage decisions. Healthy patients undergoing minor procedures or patients with serious co-morbidity or major surgery could be readily identified and streamed appropriately by such staff using predetermined guidelines and aided by specialist instruction. The procedural anaesthetist could then further refine the remaining patient list, externally ordering and reviewing any required preoperative patient investigations and then making decisions on the need for OPD assessment prior to the day of surgery.
The increasing availability of internet access for patients also suggests that a model incorporating online guided self-assessment may be a practical option. Commercial online systems are increasingly available (e.g. SurgiPrep, SourcePlus Passport (15)) and some patients even express a preference for such methods over traditional outpatient clinics (16). Validation of quality and accuracy however, particularly for patients with significant health issues or those undergoing complex surgery, is still required. While telephone contact was a part of the current study, the structure of the tool in practice meant the phone operator simply acted as a surrogate for the patient, reading the questions, ticking boxes, or adding free text. In practice, both options might be available, although the quality of data obtained online, without the input of a phone operator, may be different. This warrants exploration if such delivery platforms are to be employed.
Assessment by other clinicians, such as nurses or pharmacists, for example, is often a preoperative requirement. Our observation during this study found considerable overlap in the information required from these disciplines suggesting value in adding allied health questions to such a pre-screening process. Preoperative instructions and ordering preoperative investigations are also often necessary components of preparation for surgery. When distance or travel are issues for patients (as for one-third of patients in the current study), remote management of these aspects is also potentially feasible.
There are many elements and stake-holders within the perioperative process and no single model of care is likely to fit all institutions and changes to any single element of the process need to be viewed in context with other concurrent patient services. However, the exploration of new models of care provides the opportunity to examine current practice, new technologies, the role of outpatient clinics together with issues such as patient consent and education. Remote assessment tools such as CHAT can provide quality patient health information useful for clinical decision-making but its role within each institution's existing model of care needs to be carefully considered.
Input from anaesthetists from the following hospitals is gratefully acknowledged: South Australia--Royal Adelaide Hospital, Queen Elizabeth Hospital, Women's and Children's Hospital and Flinders Medical Centre; Western Australia--Royal Perth and Fremantle Hospitals. In particular, Drs James Ellwood, Andrew Hardy, Simon Macklin, Manith Kar, Rob Laing, Matthew Newman and Kym Osborn were instrumental in evaluating the developing question set. Allison Martinez and Joanne Petito conducted phone interviews and organised the collection and collation of Royal Adelaide Hospital patient data. Shauna Fatovich, Susan March and Felicity Vavra collated patient information from the Royal Perth Hospital. Statistical support was provided by Professor John Ludbrook, Biomedical Statistical Consulting Service, Melbourne, Victoria. Assistance in preparation of the manuscript from Dr Michael James, Royal Adelaide Hospital, is gratefully acknowledged.
This work has been supported by funding from Medibank Private, Medtel Australia, South Australian Health and also supported by a Research Translation Project grant from the State Health Research Advisory Committee of the Health Department of Western Australia. The topic, direction, content, analyses and issues identified in the article are solely the work of the authors with no intellectual involvement either directly or indirectly by any of these bodies.
The University of Adelaide and South Australian Health has an agreement with Medtel Australia for potential return on sales of their perioperative information system.
The CHAT software platform used in this study was written exclusively by the first author as a non-proprietary clinical research tool and is not designed for the delivery of clinical care outside of a research setting. There is no agreement with any institution or company regarding the future use of any part of this tool.
(1.) Kluger MT, Tham EJ, Coleman NA, Runciman WB, Bullock MF. Inadequate pre-operative evaluation and preparation: a review of 197 reports from the Australian incident monitoring study. Anaesthesia 2000; 55:1173-1178.
(2.) Gibbs N, ed. Safety of Anaesthesia. A review of anaesthesia related mortality reporting in Australia and New Zealand 20032005. Australian and New Zealand College of Anaesthetists, 2009.
(3.) Ferschl MB, Tung A, Sweitzer BJ, Huo D, Glick DB. Preoperative clinic visits reduce operating room cancellations and delays. Anesthesiology 2005; 103:855-859.
(4.) Correll DJ, Bader AM, Hull MW, Hsu C, Tsen LC, Hepner DL. Value of preoperative clinic visits in identifying issues with potential impact on operating room efficiency. Anesthesiology 2006; 105:1254-1259; discussion 6A.
(5.) Bunn F, Byrne G, Kendall S. The effects of telephone consultation and triage on healthcare use and patient satisfaction: a systematic review. Br J Gen Pract 2005; 55:956-961.
(6.) Parker BM, Tetzlaff JE, Litaker DL, Maurer WG. Redefining the preoperative evaluation process and the role of the anesthesiologist. J Clin Anesth 2000; 12:350-356.
(7.) van Klei WA, Hennis PJ, Moen J, Kalkman CJ, Moons KGM. The accuracy of trained nurses in pre-operative health assessment: results of the OPEN study. Anaesthesia 2004; 59:971978.
(8.) NCCAC. Preoperative Tests, The Use of Routine Preoperative Tests for Elective Surgery--Evidence, Methods and Guidance. London, NICE 2003.
(9.) Horn J, Bell MD, Moss E. Handover of responsibility for the anaesthetised patient--opinion and practice. Anaesthesia 2004; 59:658-663.
(10.) Reed M, Wright S, Armitage F. Nurse-led general surgical preoperative assessment clinic. J R Coll Surg Edinb 1997; 42:310313.
(11.) Rai MR, Pandit JJ. Day of surgery cancellations after nurse-led pre-assessment in an elective surgical centre: the first 2 years. Anaesthesia 2003; 58:692-699.
(12.) Shub C. Echocardiography or auscultation? How to evaluate systolic murmurs. Can Fam Physician 2003; 49:163-167.
(13.) Das P, Pocock C, Chambers J. The patient with a systolic murmur: severe aortic stenosis may be missed during cardiovascular examination. QJM 2000; 93:685-688.
(14.) Digner M. At your convenience: preoperative assessment by telephone. J Perioper Pract 2007; 17:294-298, 300-301.
(15.) Patients go online for preoperative screening. Software provides more complete preoperative information from patients prior to surgery dates and easy access for operating room staff. Health Manag Technol 2009; 30:24-25.
(16.) Hepner DL, Bader AM, Hurwitz S, Gustafson M, Tsen LC. Patient satisfaction with preoperative assessment in a preoperative assessment testing clinic. Anesth Analg 2004; 98:10991105.
C. GRANT *, G.L. LUDBROOK ([dagger]), E.J. O'LOUGHLIN ([double dagger]), T.B. CORCORAN ([section])
Department of Acute Care Medicine, University of Adelaide, Adelaide, South Australia, Australia
* MMedSc, Senior Research Associate.
([dagger]) MB, BS, PhD, FANZCA Professor of Anaesthesia, University of Adelaide and Royal Adelaide Hospital.
([double dagger]) MB, BS, FANZCA, Specialist Anaesthetist, Department of Anaesthesia and Pain Medicine, Royal Perth Hospital and Clinical Associate Professor, School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia.
([section]) MB, BCh, BAO, MRCPI, FCARCSI, MD, FCICM, Specialist Anaesthetist, Department of Anaesthesia and Pain Medicine, Royal Perth Hospital and Clinical Associate Professor, School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia.
Address for correspondence: Mr C. Grant, Department of Acute Care Medicine, University of Adelaide, Adelaide, SA 5005. Email: cliff.grant@ adelaide.edu.au
Accepted for publication on January 3, 2012.
Table 1 Patient surgical categories and surgical severity grade Surgical severity Patients Grade 1 (minor) 15 (2.9) Grade 2 (intermediate) 191 (37.3) Grade 3 (major) 225 (44.0) Grade 4 (major+) 81 (15.8) Numbers shown are absolute patient numbers (%). Surgical severity grades based upon National Institute of Clinical Excellence Guidelines. Examples are; Grade 1=excision of lesion of skin, drainage of breast abscess; Grade 2=primary repair of inguinal hernia, excision of varicose veins of leg, tonsillectomy; Grade 3=total abdominal hysterectomy, endoscopic resection of prostate, lumbar dissectomy, thyroidectomy; Grade 4=total joint replacement, lung operations, colonic resection, radical neck dissection, neurosurgery. Table 2 Incidence of the most frequent medical indicators recorded in patients Comorbidity Patients GORD 200 (38.8) Hypertension 172 (33.4) Asthma/COAD 129 (25) Adverse drug reaction 123 (23.9) BMI >35 kg/[m.sup.2] 116 (22.5) PONV history 101 (19.6) Diabetes 80 (15.5) Difficult airway 52 (10.1) Previous GA complications 52 (10.1) Obstructive sleep apnoea 49 (9.5) IHD/significant cardiac history 45 (8.7) Numbers shown are absolute values (%). GORD=gastrooesophageal reflux disease, COAD=chronic obstructive airway disease, BMI=body mass index, PONV=postoperative nausea and vomiting, GA=general anaesthesia, IHD=ischaemic heart disease. Table 3 Most common errors and omissions (C scores) in preoperative assessments provided by computer health assessment by telephone Category n Description/examples Airway 13 Mallampatti grade 4 (x1), grade 3 (x2); limited mouth opening not noted (x1) Medications 12 Significant omissions (e.g. prednisolone, oxycodone) Dental 7 E.g. vulnerable teeth or dental caps not noted GORD 6 Mild gastric reflux (x3); severe (x1) Adverse drug 5 E.g. bronchospasm (x1), anaphylaxis to reaction contrast media (x1) Arrhythmia 4 False positive descriptions by patient BMI 4 Incorrect height and weight supplied by patient during interview GORD=gastro-oesophageal reflux disease, BMI=body mass index. Table 4 Association of the review panel classification 'suitable for day of surgery assessment' and 'requirespreoperative outpatient assessment' based upon patient data provided by outpatient assessment vs data from telephone interview CHAT data OPD data DOS suitable OPD required Total DOS suitable 304 (59.4) 7 (1.4) 311 (60.7) OPD required 5 (1.0) 196 (38.3) 201 (39.3) Total 309 (60.4) 203 (39.7) 512(100) Numbers shown are patient numbers and (%). Sensitivity 98% (304/309), 95% CI 96-99%; Specificity 97% (196/203), 95% CI 92-98%; Positive predictive value=98% (304/311), 95% CI 96-99%; Negative predictive value=98% (196/201), 95% CI 95-99%. CHAT=computer health assessment by telephone, DOS=day of surgery, OPD=outpatient department, CI=confidence interval. Table 5 Ten medical indicators most frequently associated with the need for preoperative outpatient clinic assessment. Values show the percentage and absolute number of patients with this comorbidity requiring outpatient clinic assessment Patients requiring OPD assessment Diminished exercise tolerance * 100% (24) Cardiac implants 100% (8) Gastric banding 100% (9) Angina 87% (15) Previous regional block complication 83% (6) COAD 76% (25) Hepatitis 75% (12) OSA 72% (49) Orthopnoea 71% (17) DVT 70% (20) Cardiac murmur 70% (10) OPD=outpatient department, COAD=chronic obstructive airway disease, OSA=obstructive sleep apnoea, DVT=deep vein thrombosis. * Including inability to climb one flight of stairs without shortness of breath or chest pain. Figure 2: Number of patients suitable for either day of surgery assessment (DOS=grey) or requiring outpatient preoperative assessment (OPD=white) stratified by American Society of Anaesthetists (ASA) physical status classification score. Numbers shown are absolute patient numbers. DOS=day of surgery, OPD=outpatient department. DOS OPD unkown 47 34 ASA IV 7 ASA II 32 69 ASA II 167 82 ASA I 64 11 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Author:||Grant, C.; Ludbrook, G.L.; O'Loughlin, E.J.; Corcoran, T.B.|
|Publication:||Anaesthesia and Intensive Care|
|Date:||Mar 1, 2012|
|Previous Article:||Lignocaine plasma levels following topical gel application in laparoscopic and hysteroscopic procedures.|
|Next Article:||Ketofol (mixture of ketamine and propofol) administration in electroconvulsive therapy.|