Introduction

The Coronavirus (COVID-19) pandemic pushed healthcare systems to their limits and exposed their vulnerabilities [1, 2]. Due to the surge of COVID-19 patients, hospitals were forced to use their critical care resources more flexibly. Surgical capacity (e.g., personnel and operating rooms (OR)) was shifted to COVID-19 patients which led to the postponement of operations and a growing surgical backlog [3–5]. Prioritizing certain surgical patients became inevitable, triggering a discussion on which strategies to adopt and how to apply such strategies.

Since the outbreak of the COVID-19 pandemic, there has been ongoing research on criteria and methods for setting surgical priorities. Several models, frameworks, and guidelines have been developed to inform prioritization strategies, and optimize resource usage [6–11]. This also led to the development of a decision model within our hospital. This model quantifies the health loss due to delay of surgery, thereby providing an urgency measure for physicians [7].

Notably, the patients’ perspective was frequently neglected during the development of these COVID-19 response tools. Some understanding of their perspective on prioritization is known, but these findings originate from studies executed before the pandemic commenced, and it is unclear whether these factors have changed [12–15]. Studies which were published during the pandemic only evaluated patients’ perspectives on the ethical theories used to underpin prioritization strategies or addressed non-surgical resources such as intensive care unit (ICU) beds [16–18]. How to utilize these ethical theories in clinical practice in accordance with patients’ perspectives on surgical prioritization remains unclear.

The exclusion of patient inputs during COVID-19 policymaking can be partly justified by the acute impact of the pandemic. However, incorporating their perspective is important for three reasons. First, patients are a major stakeholder in healthcare, and their involvement contributes to high quality and sustainable policymaking [19, 20]. Second, it is reasonable to incorporate their perspective from a societal point of view since they pay for the resources allocated and are also subject to the prioritization policies. Third, it is valuable for healthcare professionals to know whether the current prioritization strategy accords with patients’ perspectives since this will likely contribute to effective communication strategies about the policy adopted.

The first, essential step towards involving patients in surgical priority setting is to gain a deep understanding of their perspectives. As such, this exploratory study had a two-fold aim: first, to identify which factors patients value in priority setting and, second, to assess patients’ perspectives on the use of a decision model. By pursuing these aims, this study contributes to the broader conversation on surgical prioritization during times of scarcity. Conducting this exploratory study is essential as a preliminary step to identify these factors for potential integration into quantitative methods, given the persistent challenges posed by personnel shortages and surgical scarcity in the coming years.

Materials and methods

Given the knowledge gap surrounding patients’ perspectives since the outbreak of the pandemic on surgical prioritization and the use of decision models, an exploratory qualitative study was conducted. This method was deliberately chosen since it is flexible and open-ended, and therefore appropriate for obtaining a rich and detailed understanding of patients’ perspectives. Ethical approval was obtained from the Medical Research Ethics Committee (reference number: MEC-2021-0679) of Erasmus MC. The study was prepared in accordance with the consolidated criteria for reporting qualitative research (COREQ) (see S1 File) [21].

Results

Details of the 15 participants are provided in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of the patient sample (n = 15).

https://doi.org/10.1371/journal.pone.0294026.t001

The interviews revealed three themes covering multiple aspects. The themes were categorized as: general attitudes towards surgical prioritization, patient-related factors, and the perspectives on a decision model for surgical prioritization.

Data saturation was reached after the analysis of ten interviews. Further analysis of the third batch of five interviews revealed no new themes or factors. We therefore concluded that it was not necessary to recruit any new participants, such that our final sample included just the 15 original participants.

Discussion

The main purpose of this study was to explore patients’ perspectives on surgical prioritization and identify which factors they value in priority setting. Furthermore, the patients’ perspectives on using a decision model during this process was assessed. We found that, while patients understand the complexity and difficulties of prioritization, they criticize the prioritization process itself. They want greater transparency and are concerned about the impact of subjectivity on decision-making. From the patients’ perspective, allocative decisions should be guided, albeit to varying degrees, by patient-related factors such as age, physical functioning, waiting time, expected gain in QoL and survival, and resource usage. Overall, patients support the use of a decision model in daily practice, but want physicians to make the final decision based on their clinical expertise. Patient involvement in the decision-making process was not deemed to be of added value but they would value an explanation of a model if one was being used.

With respect to allocative decisions, several factors were considered relevant to patients. Factors that were clinically oriented (e.g., age, physical functioning, expected gain in QoL and survival) were considered important and this finding supports previous research into the allocation of other scarce resources [22–24, 27, 30, 31]. Notably, behavioral factors were widely discussed in our interviews. It is known that the general public often supports a more blame-oriented approach, suggesting that the public believes that every patient has some degree of responsibility for their disease [26, 32]. This perspective may have been especially emphasized in our study since the interviews took place during a heated debate in the Netherlands on COVID-19 vaccination hesitancy. Similarly, other studies carried out during the pandemic show that Dutch people attach substantial value to behavioral factors and would penalize patients who do not adhere to the COVID-19 measures or are obese [33].

From the results, we conclude that the patients’ views on a decision model are not straightforward. Consistent with earlier evidence, patients prefer physicians to make the final decision [16, 18, 22]. This attitude accords with other studies evaluating patients’ views on other tools that could support prioritization decisions such as algorithms. The majority of patients feel that algorithms should not be used without the involvement of a physician and, in general, patients are ambivalent about such tools [34, 35]. As the application of algorithms in healthcare continues to evolve, ‘algorithm aversion’ has received considerable attention, with people appearing reluctant to let healthcare providers rely on such algorithms [36]. People tend to discount algorithmic decisions, even when they are proven to be superior, and instead favor human decision-makers. Along with individual preferences and the nature of the task to be executed, the “black-box character” of an algorithm (perceived complexity, lack of transparency, and inaccessible to people) has been advanced as an explanation for this attitude [36, 37]. These factors that contribute to ‘algorithm aversion’ could also play a significant role in the acceptance or rejection of decision models. In further development and adaption, this perception of a model being a “black box” needs to be addressed. Providing stakeholders with sufficient and understandable information on the model could possibly resolve some of this aversion.

This study adds to the broader discussion about surgical prioritization and how to operationalize this in times of scarcity. Given the complexity and novelty of this topic, there was a paucity of evidence on which factors patients value. Therefore, conducting this exploratory study was required as an initial step to shed some light on these factors. Subsequently, these can be integrated into quantitative methods to elicit preferences (e.g., discrete choice experiments). Considering the ongoing challenges due to personnel shortages, surgical scarcity will remain a bottleneck in the upcoming years and therefore this line of research is of great significance.

No research is free of limitations, and our findings should be interpreted in light of these. First, the results of this study might be influenced by selection bias. All the participants are members of the patient council or panel, which implicitly indicates that they want to be involved in the day-to-day affairs of the hospital. This could increase their desire for more explanation and communication about the prioritization dilemmas. Furthermore, the participants do not adequately reflect the diversity of the general population as they are all rather health literate. Finally, some of the participants could have heard about the decision model prior to this study since, in recent years, the model has received attention both inside the hospital as well as nationwide. This prior knowledge could have influenced their attitude, either positively or negatively, towards the model. Likewise, by showing them the model and vignettes during the interviews, participants were possibly primed to report factors already mentioned (e.g., survival and QoL). In general, we would like to stress the exploratory nature of our study. We purposively choose this research design to gain a deep understanding of patients’ perspectives. Though, the results obtained in our study do not provide an exhaustive overview of all possible perspectives that might exist but contribute to extend our understanding of patient perspectives

Despite these limitations, our findings have two major implications for clinical practice and research. First, this study provides an extensive overview of factors that patients consider important when reasoning about prioritization. These factors should be discussed with stakeholders (e.g., physicians, model developers, and policymakers) to see whether it is desirable and feasible to incorporate them in a decision model or other prioritization tools. Further studies should also evaluate how other stakeholders value these factors.

The second call for action addresses providing appropriate information to patients. The participants called for information on the prioritization decisions made. Although they do not want to be involved in the process itself, or want to have a formal say, they do want some explanation. Given the algorithm aversion, it would be beneficial for wider acceptance if we “opened the black box” and explained and educated patients on the prioritization decisions and the potential usage of a prioritization tool to guide these processes. However, it is not clear how to one should effect this information provision since some participants also expressed concerns (e.g., impersonal feelings). We would encourage further studies involving patients to develop a full understanding of their needs and preferences.

Conclusions

This study evaluated patients’ perspectives on surgical prioritization dilemmas and the use of a decision model to guide these decisions. Whilst a comprehensive set of factors was established, it also became evident that the support for the various factors differed among the participants. Therefore, it seems unfeasible to immediately incorporate all these factors in prioritization strategies or decision models. Instead, this study concludes that more research is needed to identify feasible avenues and seek consensus.

Supporting information

Acknowledgments

We would like to thank Kees Ahaus for his valuable feedback on an earlier manuscript version. We would also like to acknowledge Thomas Reindersma who participated in the pilot interview, and all members of the patients’ council and panel of Erasmus University Medical Center who participated in this study.