Patient compliance with preoperative instructions prior to an elective procedure: A quality improvement audit

Prior to undergoing an elective procedure, patients are given instructions regarding which medications to withhold or continue, are told to fast from a certain time, and complete an informed consent process for their procedure. Failure to understand and comply with this information can have significant patient safety, health economic and medicolegal implications.

We aimed to determine the prevalence of incomplete compliance with medication and fasting instructions and incomplete knowledge of the procedure for which they had consented in our hospital.

A quality improvement audit was performed on patients undergoing an elective procedure at a major tertiary hospital in Sydney over a two-week period in July 2020 and over another two-week period in July 2021. The project met New South Wales Health guidelines for a quality improvement activity not requiring ethics review, ¹ which was confirmed by the South Eastern Sydney Local Health District Human Research Ethics Committee.

Adult patients (≥18 years of age) directly admitted to the day surgery unit on the day of their elective surgical or endoscopy procedure during the audit period and who were able to communicate in English without an interpreter were included. Prior to this, each patient had provided written informed consent for their procedure witnessed by a member of the procedural team at least at registrar level. This written consent process occurred at various times and places (i.e. public and private, inpatient or outpatient) prior to their procedure. Patients were asked routine clinical questions by the anaesthetics team to assess their compliance with medication advice, fasting instructions (with respect to both solids and clear fluids) and their knowledge of the procedure to which they had consented. Patient compliance/knowledge was categorised as either complete or incomplete for each of the three domains: medication instructions, fasting instructions and nature of procedure.

Compliance with medication instructions was categorised as complete if the patient had followed the instructions documented during their preoperative anaesthetic assessment. In the case of unclear documentation, it was assumed patients had been instructed according to our hospital policies. The inappropriate taking and/or the unnecessary withholding of medications warranted an ‘incomplete’ classification. Simple vitamin and mineral supplements were not included in this assessment.

Compliance with fasting instructions was categorised as complete if the patient had been fasting for at least as long as instructed by the day surgery unit or, if bowel preparation was required, their proceduralist’s prescribed bowel preparation protocol. Fasting for longer than required was not categorised as incomplete compliance, except for procedures not requiring fasting (e.g. flexible cystoscopy) in which case any fasting warranted being classified as incomplete compliance.

Knowledge of the procedure was categorised as complete if the patient expressed a basic description of their procedure as documented on their consent form; otherwise, their knowledge was classified as incomplete. A basic description was one that demonstrated awareness of the principal physical outcome of their procedure, which related to the actual procedure and not merely its indication, and which did not contain significant inaccuracies. For example, to be categorised as complete, a patient undergoing a ‘laparoscopic appendicectomy ± open’ would need to reference removal of their appendix; referencing the indication only (e.g. treatment of appendiceal abscess) would be not be sufficient. They would not be required to volunteer information regarding the surgical approach. If, however, they volunteered incorrect information, their understanding would be classified as incomplete. Although patients were interviewed by multiple members of the anaesthetics team (M.A., N.C., J.J., V.L., J.Y. and T.Z.), to maintain consistency, the same resident medical officer (M.A.) was readily available on site throughout the study period to determine whether patient compliance/knowledge was incomplete or complete; where the decision was unclear, this was escalated to the same consultant anaesthetist (M.D.).

These compliance/knowledge categorisations were de-identified and recorded alongside basic demographic information, including patient age, sex and admitting team. The type of preoperative anaesthetic assessment undergone by patients was also recorded. Elective patients were either seen by an anaesthetist in a pre-admission clinic (PAC), or bypassed PAC on the basis of triage by a pre-admission nurse. For a patient to be bypassed from PAC, all of the following criteria needed to be met: no major cardiac, respiratory or other significant comorbidities as determined by a pre-admission nurse in liaison with a consultant anaesthetist; planned for day surgery or one night admission only; body mass index <35 kg/m²; no previous problems with anaesthesia; the patient has no questions or concerns about the procedure or the anaesthetic proposed.

Patients attending PAC received verbal medication instructions, as well as written instructions on a standardised template, whereas bypassed patients received only verbal instructions. Patients attending PAC received instructions from an anaesthetist directly, whereas bypassed patients received instructions from a pre-admission nurse in liaison with a consultant anaesthetist. All patients were sent a text message with fasting instructions the day before their procedure and had their understanding of their procedure confirmed verbally whilst attending PAC or by phone if PAC was bypassed.

A total of 465 patients were assessed (mean age 57.9 (standard deviation 17.3) years; 48.4% female; 45.2% in 2020). Of these, 24.3% attended PAC, 68% were bypassed, and 7.7% did not undergo a pre-admission anaesthetic assessment (4.1% did not require a general or regional anaesthetic, 3.6% represented potential systems failure). The majority of patients were admitted under general surgery (33.8%), gastroenterology (29.2%) or urology (17.8%). Other disciplines contributing a significant number of patients included gynaecology, plastic surgery and vascular surgery. Overall patient compliance was incomplete for 24.5% of patients for medication instructions, 9.5% for fasting instructions and 16.1% for knowledge of their procedure. Among PAC patients, patient compliance was incomplete for 38.1% of patients for medication instructions, 8.8% for fasting instructions and 16.8% for knowledge of their procedure. Among PAC bypass patients, patient compliance was incomplete for 20.3% of patients for medication instructions, 8.9% for fasting instructions and 16.1% for knowledge of their procedure.

Compared to previous literature, this audit found a similar rate of noncompliance with medication instructions (24.5% versus 6.5% to 40%),^2–5 and a slightly higher rate of noncompliance with fasting instructions (9.5% versus 2% to 7.9%).^4–6

Our findings indicate that many patients scheduled for elective procedures in our hospital do not comply fully with medication advice or fasting instructions, and many have incomplete knowledge of the procedure to which they have consented. Given this was a quality improvement audit, we cannot comment on specific factors associated with noncompliance and poor understanding; for example, we did not to collect the socioeconomic or education status of patients, did not consider their English-speaking status fully and had no control or information on the surgical consent procedure. Our audit was also limited by our sample size and our use of a single assessor of patient compliance (M.A.).

Nevertheless, it is evident that simply providing information to patients—whether by anaesthetists in PAC, proceduralists through a written consent process, pre-admission nurses over the phone or fasting instructions via text message—does not ensure adequate compliance. Preoperatively, it is therefore incumbent upon clinicians to communicate more effectively with our patients the rationale and importance of compliance with preoperative instructions and to ensure this has been understood. Where the instructions are more complex or there exists doubt about a patient’s or carer’s ability to comply, liaison with the patient’s general practitioner and/or pharmacist could be considered; this may be particularly useful in bridging any communication gaps for patients from culturally and linguistically diverse communities. By presenting our findings to the anaesthetists, proceduralists and pre-admission nurses at our institution, we aimed to inspire culture change with respect to preoperative instructions and surgical consent. While our findings apply only to our institution, we believe our findings might be similar in other institutions and encourage other institutions to conduct similar audits.