Background

Inappropriate opioid prescribing is an internationally recognized threat to population health and a pressing challenge for healthcare services (Kiang et al. 2020; Curtis et al. 2019; Degenhardt et al. 2019). The North American ‘opioid crisis’ continues, with an ongoing rise in opioid-related mortality, initially due to prescription opioids and more recently illicit heroin and fentanyl use (Berterame et al. 2016). Despite increased awareness of risks and opioid abuse, prescription opioid use remains historically high in both North America and Europe (Jani et al. 2020; Schieber et al. 2019; Verhamme and Bohnen 2019; Lancet 2022). Inappropriate prescribing following surgery is increasingly recognized as a contributor to the problem. Opioids are effective analgesics for managing acute pain following surgical trauma (Small and Laycock 2020) and were increasingly used in longer and higher doses following the publication of guidelines on post-operative pain management (Ballantyne et al. 2016). However, opioids also have significant adverse effects, including sedation, constipation, nausea, and confusion. Long-term use can lead to tolerance, dependence, hyperalgesia, addiction and increased mortality (Colvin et al. 2019). There is increasing attention being paid to the role of postoperative opioids in slowing recovery from surgery and contributing to long-term opioid use (Glare et al. 2019; Levy et al. 2021; Daliya et al. 2021). Enhanced recovery after surgery (ERAS) programmes often include the provision of multi-modal analgesia to promote faster recovery with fewer complications. Whilst short-term opioid use provides effective relief from acute pain following surgery, it is increasingly recognized that the perioperative period is a time when longer-term opioid usage may begin. Effective opioid stewardship in the perioperative period is therefore of critical importance.

Improving opioid stewardship in the peri- and post-operative management of patients with bowel cancer has the potential to improve recovery, lead to faster discharge, improve outcomes and most importantly, prevent patient harm. A requirement for an effective opioid stewardship program is the ability to measure the appropriateness of opioid use.

Health care quality indicators are a type of performance measure (Stelfox and Straus 2013) that evaluate aspects of quality of care, without which the monitoring of healthcare quality is impossible (Mainz 2003; Arah et al. 2006). Quality indicators are used to measure the variability in the quality of care, identify potential areas for improvement and can be used to feedback on performance to healthcare teams to change clinical practice. They should be relevant, actionable, reliable, show room for improvement and data collection should be feasible (Ivers et al. 2012; Kelley and Hurst 2006; Fabian and Geppert 2011). Donabedian’s framework (Donabedian 1988) describes quality as a function of three domains: structure, process and outcome. The structure is defined by the attributes of the setting in which care is provided, process by the input of the practitioners working in that system and outcome by the change in health status of the patient.

No quality indicators for perioperative opioid use are currently described in the National Institute for Health and Care Excellence Standards and Indicators library (Standards and Indicators | NICE (accessed 22nd March 2022). A rapid review was performed to identify quality indicators for perioperative opioid stewardship for patients undergoing abdominal surgery for bowel cancer. This is a form of knowledge synthesis that streamlines the process of conducting a traditional systematic review to produce evidence in a rapid resource-efficient manner (Hamel et al. 2021) and has been chosen to allow timely evidence synthesis to inform decision-making (Haby et al. 2016).

The objective of this rapid review was to identify and extract potential quality indicators from the best available evidence on perioperative opioid use in patients undergoing major abdominal surgery for bowel cancer. This approach to the development of actionable quality indicators has been described and applied effectively in other clinical settings (Kallen et al. 2018).

Methods

Cochrane rapid review methods were followed (Garrity et al. 2021). A systematic literature search of Medline was performed and included all articles available to April 2021. Systematic reviews and primary studies were sought. The types of participants were not restricted and could be individuals, organizations or systems. Search terms are shown in Table 1 and include terms and truncations for quality indicators, opioids, surgery (with potential limitation to colorectal cancer surgery) and development. The search was limited to studies of adult subjects and studies published in English. A manual search was conducted of the reference lists of the selected papers. Searches were conducted between the 1st and the 25th August 2021 and supplementary searches of reference lists were conducted in December 2021. The National Quality Measures Clearinghouse project (2022) was also reviewed for relevant content.

Table 1 MEDLINE search strategy
Full size table

The initial search identified 588 abstracts. These results were imported into Rayyan (http://rayyan.qcri.org/) (Ouzzani et al. 2016) a free web tool used to facilitate the screening and selection of studies for systematic and scoping reviews.

Three members of the project team (MA, KP, DY) screened all titles and abstracts for inclusion. Duplicate studies, case reports, editorials, non-English language studies, quality improvement not concerning opioid use, abdominal or colorectal surgery, or performance measures and quality improvement in specific subgroups of patients were excluded. Where a decision on inclusion could not be reached, two further team members (CT, SH) reviewed the titles and abstracts. All studies included at this stage underwent full-text review, undertaken by two members of the project team (CT, SH). MA, KP, and DY accessed the full-text articles of all included papers, which were uploaded and accessed using Rayyan. The extraction of quality indicators from the full texts was undertaken by CT and SH. All those included were papers from which quality indicators could be extracted. A quality indicator extraction tool was developed in advance of data extraction, with potential indicators categorized to the stage of perioperative care they relate to (Supplementary materials 1) to enable reproducible results. Finally, a full list of potential indicators was composed, in which indicators were rephrased where needed and duplicate indicators removed. Reporting has been guided by the PRISMA extension for scoping reviews (PRISMA-ScR) (Extension and for Scoping Reviews (PRISMA-ScR): Checklist and Explanation | The EQUATOR Network (equator-network.org) (accessed 22nd March 2022).

Results

Five hundred eighty-eight publications were identified by the literature search. Three duplicates and a further 425 abstracts were removed as they did not meet inclusion criteria. Eighty-three of these papers were included for full-text review, of which 53 were excluded because they reported on another outcome or population, or because the paper did not include quality indicators. Thirty papers were included, the references of which were reviewed. A further 35 full-text papers were reviewed, of which 17 were included. The study selection flowchart (Fig. 1) details this process. In total, 47 papers were identified from which quality indicators were extracted. Review of the quality indicators clearinghouse did not yield results.

Fig. 1
figure 1

Study selection flowchart

Full size image

The characteristics (study design and numbers of participants) of the included papers are shown in Supplementary Materials 2.

One hundred twenty-eight quality indicators from 47 papers were extracted, with some papers describing several indicators. See Supplementary materials 1 for full details of all raw extracted quality indicators. Duplicates were removed, leading to the identification of 24 discrete indicators. Table 2 shows the numbers of discrete quality indicators identified at each stage of the patient journey.

Table 2 Numbers of discrete quality indicators identified at each stage of the patient journey
Full size table

Instruments for collecting data on quality indicators, and structural, process, and outcome indicators were collated. These are grouped according to stage in the perioperative journey and are shown in Table 3.

Table 3 Structural, process, and outcome quality indicators across the patient perioperative journey
Full size table

The quality indicators identified which could be grouped into five topics: patient education, staff education, preoperative patient optimization, patient and procedure-specific prescribing and deprescribing and opioid-related adverse drug events (ORADEs) and are shown in Table 4. Full details of the quality indicator topics are shown in Supplementary materials 3.

Table 4 Full list of proposed quality indicators
Full size table

Discussion

Opioids are highly effective analgesics but can cause harm and there is now increasing concern about their perioperative use. A number of contributing problems have been identified. Opioid tolerance preoperatively is a risk factor for poorer outcome (Cron et al. 2017; Gan et al. 2020; Gan et al. 2015; Kessler et al. 2013). When opioids are used by either opioid-naïve or opioid tolerant patients, they are put at risk of opioid-related adverse drug events (ORADEs) (Macintyre et al. 2014; Minkowitz et al. 2014; Keller et al. 2019; Kessler et al. 2013; Oderda et al. 2013), and opioid use is associated with postoperative complications and increased length of stay (Cron et al. 2017; Gan et al. 2020; Gan et al. 2015). Additionally, the perioperative period has been identified as a period of risk for the development of chronic opioid use (Lee et al. 2017; Brummett et al. 2017; Clarke et al. 2014; Macintyre et al. 2014; Roughead et al. 2019; Srivastava et al. 2021). At discharge, opioid prescriptions in excess of requirements are widely reported (Neuman et al. 2019; Bromberg et al. 2021; Hill et al. 2017; Hill et al. 2018; Pruitt et al. 2020; Bartels et al. 2016; Agarwal et al. 2021; Howard et al. 2019; Meyer et al. 2021) and opioids initially used for short-term pain relief can become part of repeat prescriptions following hospital discharge. Poor practice around safe storage and disposal of opioids following discharge contributes to increased opioid in the community with the potential for opioid diversion (Fujii et al. 2018; Hill et al. 2017; Bartels et al. 2016). These factors contribute to the development of persistent postoperative opioid use (PPOU) with the increased potential for ORADEs in the community following discharge.

Effective opioid stewardship is therefore an important part of the provision of opioids in the perioperative period, and a need to improve has been identified (Srivastava et al. 2021). Quality indicators are used to monitor and improve quality in healthcare (Stelfox and Straus 2013; Mainz 2003; Fabian and Geppert 2011; Donabedian 1988; Rademakers et al. 2011). Good quality indicators are based on the best available evidence, should be highly specific and sensitive, with the integration of best clinical evidence, clinical expertise, and patient values. No quality indicators for perioperative opioid stewardship currently exist. The review of the supporting evidence base is required to enable the development of a practical set of reliable quality indicators (Stelfox and Straus 2013).

Extracted quality indicators

Our review identified indicators relating to five key topics during the perioperative patient journey. These five topics are patient education, staff education, preoperative patient optimization, patient and procedure-specific prescribing and deprescribing and opioid-related adverse drug events. All five topics include structure, process, and outcome quality indicators (Table 4).

Definitions and comparisons

Varying definitions used in the literature have emerged from this review and consideration of these when discussing quality indicators is useful. Persistent perioperative opioid use is frequently described as the ongoing use of opioids at 90–180 days postoperatively (Fields et al. 2019; Lee et al. 2017; Clarke et al. 2014; Roughead et al. 2019; Pullman et al. 2021). Opioid tolerance is variably described as being present if a patient has used opioids for more than 7 days in the 60 days prior to surgery, any opioid use in 12 months prior to surgery or any opioid on the admission medication list (Fields et al. 2019; Brat et al. 2018; Cron et al. 2017; Gan et al. 2020; Hilliard et al. 2018; Truong et al. 2019). Milligrams of morphine equivalents (MME) or oral morphine equivalents (OME) are the most widely used methods to describe and compare opioid use. When reviewing postoperative patients in the community, the postoperative prescription can be considered to have been used if the prescribed opioids are dispensed between 2 and 7 days following discharge (Roughead et al. 2019). The detection of opioid misuse or PPOU after discharge is defined as at least one of the ICD-9 diagnosis code of opioid dependence, abuse or overdose (Brat et al. 2018). When reviewing the time to opioid cessation, a suggested definition is a period without an opioid prescription equivalent to three times the estimated supply duration in preoperatively opioid naïve patients (Roughead et al. 2019).

Data collection tools

Instruments to collect data for quality indicators are also reported although none have been specifically developed for postoperative opioid use. The Opioid Risk Tool (ORT), Screener for Opioid Assessment and Patients with Pain (SOAPP), and Brief Risk Interview (BRI) have been proposed for use in perioperative practice when screening patients preoperatively for risk of PPOU (Macintyre et al. 2014). The frequency, severity, and distress caused by opioid-related side effects can be scored as 0 to 60 on the Perioperative Opioid-related Symptom Distress scale and has been reported as a tool to assess ORADEs (Lee et al. 2010).

Addressing the gaps

More process quality indicators than structure or outcome quality indicators are described in the literature. However, the factors which are reported to make the greatest difference to a patient’s assessment of healthcare quality are process-related and process quality indicators are especially useful to consider when quality improvement is desired (Rademakers et al. 2011). Fewer quality indicators concern the intraoperative and immediate recovery period. The impact of specific changes in practice on long-term outcomes remains unclear, and our rapid review of quality indicators will enable rigorous studies of the implementation and impact of interventions to improve opiate stewardship in the perioperative period.

Algorithms and electronic systems

The screening of patients for potential opioid tolerance, future likelihood of PPOU, and patient-group-specific prescribing with limits on the type, dose, and duration of opioid prescription may be best undertaken with the use of algorithms and the development in machine learning (Ellis et al. 2019). Electronic records and prescribing (which are already well-embedded in primary care) are now used increasingly in hospital clinical practice and this may present a good opportunity to develop patient-or patient-group-specific guidelines for opioid prescribing with limits and alerts if there is deviation from agreed protocols.

Limitations

Limitations of this work include those relating to rapid review methodology. This is a relatively recently developed form of knowledge synthesis, and while valid (Garrity et al. 2021), is less comprehensive than a systematic review. Most of the studies included are retrospective cohort studies, and most originate using data from patients in a different healthcare systems (often from the USA). The characteristics of papers are reported (Supplementary Materials 2) but an assessment of risk of bias was not undertaken. This work has been done to drive improvement in outcomes for patients undergoing bowel cancer and this may limit its applicability to a wider perioperative population.

Conclusion and future work

The concept of ‘universal precautions’ have been suggested as being applicable to the prescribing and administration of opioids in the perioperative period (Lee et al. 2017; Macintyre et al. 2014) and encompass strategies at each stage of a patient’s perioperative journey to ensure that the lowest dosage, shortest acting opioids are used for the shortest possible time, while ensuring good analgesia and patient satisfaction. This will be used as an underpinning principle for our ongoing work.

This project forms part of the wider YCRBCIP program for use in the improvement of outcomes for patients with bowel cancer undergoing surgery. We have identified a set of quality indicators which may help to improve quality of care for patients undergoing major abdominal surgery for bowel cancer who receive perioperative opioids. We will now integrate the extracted quality indicators with clinician expertise and patient values to develop a more concise toolkit which providers in our region can use to benchmark and improve quality in the use of perioperative opioids for patients with bowel cancer.