Ascertaining Design and Implementation Requirements for a Perioperative Neurocognitive Training Intervention for the Prevention of Persistent Pain After Surgery

Abstract

Background

Persistent postsurgical pain (PPSP) is a common complication that impacts quality of life, often necessitating long-term opioid treatment. Certain neurocognitive factors, including reduced performance on cognitive flexibility tasks, are associated with increased risk of PPSP. We examine the perceptions of surgical patients and clinicians with regard to perioperative pain management activities and needs; patient acceptance and use of a perioperative neurocognitive training intervention; and implementation feasibility.

Methods

We conducted both individual and focus group interviews with patients undergoing thoracic surgery and clinicians in an academic medical center. The Consolidated Framework for Intervention Research guided the development of interview questions related to the adoption and implementation of a neurocognitive intervention to mitigate PPSP. A thematic analysis was used to analyze the responses.

Results

Forty patients and 15 clinicians participated. Interviews revealed that there is minimal discussion between clinicians and patients about PPSP. Most participants were receptive to a neurocognitive intervention to prevent PPSP, if evidence demonstrating its effectiveness were available. Potential barriers to neurocognitive training program adoption included fatigue, cognitive overload, lack of familiarity with the technology used for delivering the intervention, and immediate postoperative pain and stress. Implementation facilitators would include patient education about the intervention, incentives for its use, and daily reminders.

Conclusion

The study identified several guiding principles for addressing patients’ and clinicians’ barriers to effectively implementing a neurocognitive training intervention to mitigate PPSP after surgery. To ensure the sustainability of neurocognitive interventions for preventing PPSP, such interventions would need to be adapted to meet patients’ and clinicians’ needs within the perioperative context.

Introduction

More than 300 million surgeries are performed across the world every year [1], and more than 10% of surgical patients experience persistent postsurgical pain (PPSP) [2, 3]. PPSP refers to pain at the surgery site lasting 3 or more months after surgery [4]. Many patients with PPSP eventually receive long-term opioid medications to treat their pain, increasing the risk of opioid misuse, addiction, and overdose [5]. With the global increase in surgeries [1], PPSP has become a significant public health burden, resulting in patients experiencing impaired functioning, poor quality of life, and limited ability to return to daily life and work [6].

Common risk factors associated with PPSP include presurgical chronic pain, younger age, female sex, anxiety, pain catastrophizing, and type of surgery [7–11]. Prior research studies have suggested that a maladaptive response of the central nervous system to prolonged noxious stimuli is a critical factor in the development of PPSP [12–17]. PPSP can result in maladaptive changes in the central nervous system, leading to long-term pathological neuroplasticity, pain sensitization, and chronic pain that is often challenging to treat [12–15, 17–20]. Although several underlying factors could contribute to this maladaptive response, impaired cognitive flexibility, measured as the adaptability of an individual’s response to changing environmental stimuli [21–23], represents a significant risk factor for PPSP. Poor preoperative performance on a cognitive flexibility task in patients undergoing thoracic surgery, mastectomy, or knee replacement was associated with a three-fold increased risk of clinically meaningful PPSP [24].

Deficits in cognitive flexibility can be mitigated via targeted neurocognitive training, leading to symptom improvement. Within the context of post-trauma [25], neurocognitive training increased patients’ cognitive flexibility and reduced post-traumatic stress disorder symptoms. A similar neurocognitive training approach has yet to be tested in the perioperative setting for PPSP prevention [26]—a practical next step, given the potential association between poor cognitive flexibility and increased risk of PPSP.

To further develop and test the efficacy of such neurocognitive training interventions for preventing PPSP, we first need to understand current patient management activities along the perioperative continuum. We lack studies examining stakeholder perspectives on perioperative management for the prevention of PPSP and the potential need for related interventions. To address this gap, we conducted a qualitative study to ascertain perceptions of both surgical patients and clinicians on 1) pain management activities and needs along the perioperative continuum; 2) acceptance and use of a neurocognitive training intervention to prevent PPSP; and 3) implementation of the neurocognitive training intervention along the perioperative continuum. Insights gained from this study can inform the development and implementation of effective neurocognitive interventions tailored to patient and clinician needs that are feasible and sustainable within perioperative care.

Methods

Study Setting and Participants

This qualitative study used interviews and focus groups of patients and clinicians from a large academic medical center and was conducted between August and December 2020. These qualitative methods allowed us to elicit information about participants’ opinions and their experiences in their own words [27]. We examined PPSP within the context of thoracic surgeries for this initial exploratory investigation, given the high incidence of PPSP after thoracic surgeries [2, 3]. Our participants included 1) adult patients who had either undergone thoracic surgery within the past month or those who were scheduled to undergo thoracic surgery in the following month; and 2) various groups of clinicians involved in the care of these patients, including thoracic surgeons, anesthesiologists, and intensivists. Washington University’s institutional review board approved the study, and verbal consents were obtained from all participants.

Data Collection

Patient age and sex were collected from the electronic health record. Patients participated in individual semistructured interviews, whereas clinicians participated in either focus group interviews or individual interviews, according to their convenience. Patient interview questions focused on pain management, PPSP needs and resources within the current perioperative pain management workflow, and the use of a neurocognitive training program. The neurocognitive training program is planned to be delivered on the “Lumosity” application developed by Lumos Labs, Inc. (San Francisco, CA) and includes flexibility, memory, attention, speed, and problem-solving. We introduced the Lumosity app and its features to our participants during the interviews and focus group interviews. Many participants were unfamiliar with the application; therefore, we simplified our description to “cognitive exercises” and “brain games.” Table  1 displays the structured patient questions and response choices.

Table 1.

Structured interview questionnaire

Patient Questions	Response Choices
Before surgery: Did you receive any specific guidance/treatment to prepare and manage pain symptoms for your current surgery?	Yes No
After surgery: Are you experiencing any pain after your surgery?	Yes No
Do you have or use any specific support/treatment to manage your pain symptoms after the surgery?	Yes No
Do you have prior experiences with cognitive exercises or apps to improve pain, memory or mood, or reduce anxiety?	Yes No
Do you routinely use a smartphone or a tablet device?	Yes No
If yes to the above, how comfortable are you with playing cognitive games on your phone?	A scale of 1–5, with 5 being the most comfortable
If you were offered to do some cognitive exercises (which is like playing computer games) on your smartphone or tablet device to prevent chronic pain after surgery, do you think you would be interested in trying it?	Yes No
Would you be willing to spend half an hour each day for 5–6 weeks (i.e., 2 weeks before surgery and 4–6 weeks after surgery) around surgery participating in cognitive training?	Yes No
Would you rather do the training in one session, or split into several short sessions a day?	One session Split Don't know

Clinician questions focused on understanding the perioperative pain management workflow, ascertaining experiences with and perspectives on cognitive interventions for preventing PPSP, and gaining insights on potential intervention use and implementation needs (i.e., frequency of sessions, mode of cognitive intervention).

Individual and focus group interview questions were informed by the Consolidated Framework for Intervention Research (CFIR) [28], a framework consisting of 39 constructs across five domains: intervention characteristics, inner setting, outer setting, characteristics of individuals, and implementation process [29]. Specific questions incorporated patient needs and resources, mode and frequency of intervention, implementation climate, factors influencing intervention adoption, and knowledge and beliefs about the neurocognitive training intervention. Interview guides were developed, reviewed, and iteratively modified by the study team—which included a researcher trained in qualitative methods (JA), with significant experience interviewing similar surgical patient populations—multiple times to ensure clarity of language and relevance to study objectives (see Supplementary Data Appendix 1 for the guide).

Patient interviews were scheduled 1–3 weeks before or after surgery and conducted via telephone. Clinician focus groups and interviews were conducted via the Health Insurance Portability and Accountability Act (HIPAA)–compliant university instance of the Zoom platform (Zoom Video Communications, San Jose, California). All interviews were audio-recorded and transcribed for analysis.

Data Analysis

Patient demographics and structured interview questions were summarized with descriptive statistics stratified by age (18–64 years or 65+ years), sex (male or female), and surgery status (before or after surgery). Analysis of quantitative data was conducted in Stata/SE version 16.0 (copyright 2019, StataCorp, LLC, College Station, TX, USA) [30].

The qualitative data from the narrative questions were analyzed through the use of a hybrid thematic analysis approach with inductive and deductive coding. Two researchers (JA and AM) read the interview and focus group transcripts multiple times to familiarize themselves with the data. Data were then reviewed systematically and openly coded line by line with initial data-driven codes to best characterize the data’s underlying semantics (i.e., inductive coding) [31, 32]. Additionally, a priori codes based on the CFIR constructs were applied to relevant interview data on intervention design and implementation characteristics (i.e., deductive coding). Transcripts were iteratively coded, and inconsistencies were discussed and modified until 100% consensus on final codes was reached.

Once codes were finalized, all transcripts were independently coded by JA and AM to identify areas of similarity and overlap within the data. See Table  2 for an example of thematic analysis. Themes were then formed on the basis of relationships between subthemes within and across transcripts, and any discrepancies were discussed between researchers until 100% consensus was achieved.

Table 2.

An example of thematic analysis

	Interview Transcript Text	Initial Open Code	Subthemes	Theme
C2, FG3	If you help us make up a nice laminated sheet [with facts and figures about the usefulness of the neurocognitive training intervention], maybe those couple of graphs … of the preliminary data … [The educational resources would show to patients] that we think that this may help [them], which is why we're studying it … If [patients] really understand why this [intervention] may benefit them then, they’re going to be more likely to do it and stick with it.	Visualization of facts and figures; Educational resource facilitator to patient buy-in	Facilitators to neurocognitive training intervention acceptability, use, and implementation	Factors affecting acceptability and use of neurocognitive training interventions to prevent PPSP
P23	Just make sure everything [about the neurocognitive training intervention is explained] in detail, so the patient will know [what they need to do for the intervention] and have an idea of what is going on, and what they'd be expecting [out of the intervention].	Educational resource facilitator to patient buy-in

C = clinician; P = patient; FG = focus group.

Results

Fifty-five participants (15 clinicians and 40 patients) participated in the study (Table  3). With regard to the reason for surgery, the majority of patients had a diagnosis of cancer (n = 22) or possible cancer (i.e., biopsy) (n = 7). For three participants, the exact details of surgery were lacking at the time of the interview. Across interviews and focus groups, several themes were identified: 1) the perioperative pain management workflow for thoracic surgery patients, 2) factors impacting thoracic surgery perioperative workflow and pain management, 3) factors affecting acceptability and use of neurocognitive training interventions to prevent PPSP, and 4) CFIR-based neurocognitive training intervention implementation factors. With each theme, we describe the main findings and provide supporting participant quotes. Additional quotes are provided in Supplementary Data Appendix 2.

Table 3.

Study participant characteristics

Characteristic	Value
Patients (n = 40)
Age, years mean (SD)	59.7 (15.4)
Sex, % (n)
Male	40.0 (16)
Female	60.0 (24)
Surgery status, % (n)
Before surgery	47.5 (19)
After surgery	52.5 (21)
Reason for surgery, % (n)
Cancer	55.0 (22)
Possible cancer	17.5 (7)
Non-cancer	12.5 (5)
Transplantation	7.5 (3)
Unknown at time of interview	7.5 (3)
Clinicians (n = 15)
Sex, % (n)
Male	80.0 (12)
Female	20.0 (3)
Profession, % (n)
Anesthesiologist/intensivist	73.3 (11)
Surgeon	26.7 (4)

Perioperative Pain Management Workflow for Thoracic Surgery Patients

The current perioperative pain management workflow comprises presurgical planning, surgery, postsurgical hospital stay, and postsurgical follow-up in outpatient settings (Figure  1).

Figure 1.

Perioperative pain management workflow.

First, patients met with surgeons at the surgical clinic; next, they went to the Center for Preoperative Assessment and Planning (CPAP) to prepare for surgery with advanced practice nurses and registered nurses, reviewing educational materials. On the day of surgery, patients and their surgical teams (primarily attending anesthesiologists) discussed anesthesia and postoperative care plans in the waiting room. After surgery, attending anesthesiologists, residents, fellows, advanced practice nurses, and registered nurses took care of patients in the post-anesthesia care unit or intensive care unit. Advanced practice nurses and registered nurses monitored patients on the surgical floor units until discharge. The surgery clinic scheduled follow-up appointments, and if postoperative pain was persistent, surgeons referred patients to pain specialists at outpatient pain clinics.

Factors Impacting Thoracic Surgery Perioperative Workflow and Pain Management

Lack of Discussion About Chronic Pain Management

Clinicians and patients felt that both parties heavily prioritized procedural topics, frequently skimming over perioperative pain management. Several participants felt that there was more attention given to the logistics of the surgery than the discussion around potential postoperative complications, as “some patients want to know as little as possible to get through the surgery. A lot of them are … freaked out … beforehand” (C2, FG1). One of the two patients who discussed perioperative anxiety spoke about being “freak[ed] out” about procedural mistakes like nerve damage from IV use (P10), while the other stated that they were anxious because they had to process their cancer diagnosis and potential treatment (P1).

The lack of chronic pain discussions significantly impacted clinician care plans and patient expectations about postoperative pain. Clinicians remarked that “[presurgical discussion of chronic pain and management] would be incredibly rare unless the patient brings it up” (C1, CPAP).

Conversely, all five patients who mentioned preoperative pain discussions stated that their clinicians did not discuss pain enough with them. This was a problem for those with preexisting chronic pain, who stated, “[the clinicians] didn’t say anything about this being very painful … and [having] chronic pain already [is a problem for me,] and they never discussed anything about it” (P5). They felt frustrated and poorly informed about the surgery and postoperative pain management, stating that “[they] didn’t know what to expect … afterward and it has been excruciating pain” (P35).

Another challenge across surgical cases was limited pain management assistance. Clinicians and patients agreed that the current workflow limited communication—particularly on pain medication (i.e., epidural analgesia, postoperative opioids)—because of “action and time pressures” (C3, FG2). All seven clinicians who responded felt that “the opportunity for [pain] counseling is limited with the [current] workflow” (C1, FG2).

Additionally, patients unanimously expressed wariness about using pain medication. Some patients avoided taking medication (P17), whereas others took medications but suspected that “[the medications] brought on [their] own set of problems” (P1).

Disjointed Care for Postoperative Pain

Furthermore, two patients listed disjointed postoperative pain care as a factor that impacted postoperative care quality. The six remaining patients interviewed after surgery explained that different types of clinicians, from nurses (P29) to the attending surgeon (P40), talked to each patient about pain medication and treatments, and most conversations happened within the postoperative period and at the time of discharge.

“I wasn’t confident that the pain management [team] was talking to the surgical team very well … I had an epidural … and the pain management person came in and upped the epidural … The next day, somebody from the other team said, ‘We are going to take it back down ….’ The next day, the pain management person came back in and said, ‘… We’re just going to take it out …’ It didn’t seem like they were navigating the same path” (P30).

Factors Affecting the Acceptability and Use of Neurocognitive Training Interventions to Prevent PPSP

Buy-in for a Postoperative Pain Management Intervention

Clinicians demonstrated significant buy-in for intervention acceptability, given that surgical patients revealed a strong need for postoperative pain interventions, stating they would likely “feel better with a care team and treatment plan specifically for chronic pain” (A2, FG1).

When asked whether they would recommend the intervention to patients, all six clinicians stated they “would absolutely recommend it to patients” (C2, FG4). One clinician felt strongly that other care team members should demonstrate strong support for such an intervention, “if patients don’t feel like something’s prioritized by the people they’re looking to for guidance, they often will not prioritize it themselves” (C2, FG4).

Of the 13 who were asked about buy-in, seven stated that “[they’d] be willing try anything as long as it helps [them] deal with pain” (P2) and that they “would have participated if it would have been an option” (P36). However, the other six patients were skeptical about the benefits of a neurocognitive training intervention to prevent pain, stating that, “When you’re in real pain, [a neurocognitive training intervention is] not really [convincing]” (P1).

Despite being skeptical, patients were willing to try the intervention.

“I think I would try it; I would try it with skepticism” (P28).

Clinicians further asked to see surrogate outcomes to assess the impact of a neurocognitive training intervention and determine its acceptability, with some suggesting that “it will be interesting not just to see what it does with pain, but maybe post-op cognitive dysfunction” (C2, CTICU). Other suggested outcomes included functional independence through the use of the Barthel Index (C1, CPAP).

Challenges to Neurocognitive Training Intervention Acceptability, Use, and Implementation

Several participants discussed challenges such as adherence, technical concerns, surgical variability, immediate postoperative pain and stress, monetary concerns, time and effort, and physical barriers.

Fatigue and cognitive overload were common concerns among clinicians and patients, who feared that focus on the intervention would be difficult with the complexity of the intervention and strict time commitment (P10). One clinician speculated that patients would “see a lot of fatiguability … over time” (C1, CPAP), leading many to stop using the intervention within a few weeks.

Technical concerns also arose among clinicians, who were worried that “not everyone has Wi-Fi and access to technology” (C2, CTICU). Some felt that issues such as application malfunctions would prevent some patients from utilizing the neurocognitive training intervention.

With specific procedures causing more pain and postoperative complications than others, the usefulness of the neurocognitive training intervention would vary.

“So there’s a lot of variability in their surgical path, which may make it … somewhat difficult … to isolate this intervention” (C1, CPAP).

One patient explained that if their surgery was highly complex or emergent, they would not be willing to engage in such an intervention before completing the procedure: “If they find out [my condition] is life-threatening, I think the last thing I’m going to want to do is to play your video games” (P1).

Furthermore, postoperative pain and stress would hinder patients in completing cognitive exercises. All eight clinicians who were asked thought that the intervention could be started within the hospital after surgery, stating that “doing these exercises in-house, when they’re in the hospital, is very helpful” (C1, FG2). However, six out of eight patients suggested presurgical training on how to use the application, but postdischarge application use.

Another common challenge for patients was allocating time and effort to the daily exercises. Four out of six patients stated that they led busy lives and would have difficulty fitting intervention use into their schedules.

“After a couple of weeks, I’d probably get back to work and stuff like that, so I probably won’t have … much time” (P23).

The remaining two patients stated they had “all the time in the world” (P27) for the neurocognitive training intervention, stating that “there is very little else [to do]” (P13). One other patient remarked that they were especially willing to complete the intervention, provided it was on a convenient application/program (P9).

Facilitators to Neurocognitive Training Intervention Acceptability, Use, and Implementation

Facilitators to intervention acceptability, use, and implementation included familiarity with cognitive interventions, education, resources to improve buy-in, development of an interesting and rewarding intervention, intervention reminder notifications, and hospital training or at-home technical support.

Upon discussing PPSP mitigation through cognitive interventions, eight out of the nine interviewed clinicians were unfamiliar with the concept, admitting, “[they] don’t know a lot about cognitive … training” (C1, FG2). Few had experience with cognitive interventions using “different applications [in] prior [studies] relating to cognitive dysfunction preoperatively or delirium” (C2, FG4).

Conversely, three patients had experience with cognitive interventions for mental health management (e.g., anxiety, mindfulness [P28]), which would facilitate the use and adoption of a neurocognitive training intervention.

“I have the Calm app … I think [the meditation] helps” (P34).

To encourage buy-in, both clinicians and patients suggested that providing education and resources would help motivate patients to use the intervention.

“Make sure everything is in detail so the patient will know, have an idea what is going on, what they [would] be expecting” (P23).

Furthermore, clinicians felt that “[intervention buy-in would] depend on the [supporting] evidence” (C1, FG3), and “if [provided] with some studies, [they] could go to patients about [the intervention]” with more confidence (C4, FG2).

In addition, clinicians would be more convinced of the usefulness of the intervention for PPSP mitigation if they were able to record metrics of patient intervention usage and provide real-time support to patients, stating that “[patients were] going to need some kind of touch base from the coordinator to follow them in real-time and [see] whether [patients were] doing [the intervention] or not” (C3, FG3).

Clinicians also suggested that patients would likely keep playing if the game provided monetary or in-game incentives (i.e., recording high scores).

“There might be some reward system that would enhance their attention to it, and then you’re measuring how much time you spent on it. And that could be a way of getting … them to spend more time on it” (C1, FG3).

To ensure that patients remembered to complete their daily exercises, clinicians suggested using a notification system for daily reminders (C2, FG4).

One patient agreed, stating they “wouldn’t mind a reminder—maybe, a text message” (P4). They felt a notification would not be intrusive and could further increase intervention adherence.

Lastly, clinicians suggested that hospital-provided training and family or caregiver technical support could potentially mitigate technical concerns. Clinicians felt that patients needed “someone to help troubleshoot … so that if [a problem] happens, they just don’t fall off because they can’t load it, or they need to download something else” (C2, CTICU). By “playing in the waiting room” (C2, FG1), patients would receive technical support and training in a hospital.

CFIR-Based Neurocognitive Training Intervention Implementation Factors

Table  4 presents patients’ responses to structured questions about implementing the intervention stratified by age, sex, and surgery status. The majority of patients reported that they would be interested in participating in a neurocognitive training intervention. Although the majority of females (79.2%) and males (87.5%) were interested in trying the program, fewer males (80.0%) than females (100%) reported that they would be willing to spend the time necessary to complete the program. Almost all females (95.8%) stated that they routinely used smartphones or tablets, compared with only 68.8% of males. Regardless of surgery status, the majority of both presurgery and postsurgery patients reported interest in trying the intervention. All postsurgery patients reported willingness to devote the time for training 5–6 weeks around surgery.

Table 4.

Patient responses stratified by age, sex, and surgery status

	Younger Adults (n = 22) 55%	Older Adults 65+ (n = 18) 45%	Female (n = 24) 60%	Male (n = 16) 40%	Before Surgery (n = 19) 47.5%	After Surgery (n = 21) 52.5%
Age, years, mean (SD)	50.0 (14.4)	71.5 (4.8)	58.0 (16.1)	62.2 (14.4)	60.9 (16.6)	58.4 (14.3)
Sex, % (n)
Female	68.2 (15)	50.0 (9)	–	–	47.5 (10)	73.7 (14)
Prior experiences with cognitive exercises or apps to improve pain, memory, mood, or reduce anxiety, % (n)
No	86.4 (19)	66.7 (12)	75.0 (18)	81.3 (13)	66.7 (14)	89.5 (17)
Yes	13.6 (3)	33.3 (6)	25.0 (6)	18.8 (3)	33.3 (7)	10.5 (2)
Routine use of a smartphone or a tablet device, % (n)
No	9.1 (2)	22.2 (4)	4.2 (1)	31.3 (5)	23.8 (5)	5.3 (1)
Yes	90.9 (20)	77.8 (14)	95.8 (23)	68.8 (11)	76.2 (16)	94.7 (18)
Level of comfort with playing games on either smartphone or tablet, % (n)
1	4.6 (1)	29.4 (5)	8.3 (2)	26.7 (4)	14.3 (3)	16.7 (3)
2	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
3	4.6 (1)	0 (0)	0 (0)	6.7 (1)	4.8 (1)	0 (0)
4	4.6 (1)	5.9 (1)	4.2 (1)	6.7 (1)	4.8 (1)	5.6 (1)
5	54.6 (12)	35.3 (6)	62.5 (15)	20.0 (3)	42.9 (9)	50.0 (9)
No response	31.8 (7)	29.4 (5)	25.0 (6)	40.0 (6)	33.3 (7)	27.8 (5)
Interest in trying a neurocognitive training intervention, % (n)
Don't know	4.6 (1)	5.6 (1)	4.2 (1)	6.3 (1)	9.5 (2)	0 (0)
No	13.6 (3)	11.1 (2)	16.7 (4)	6.3 (1)	14.3 (3)	10.5 (2)
Yes	81.8 (18)	83.3 (15)	79.2 (19)	87.5 (14)	76.2 (16)	89.5 (17)
Readiness for training for half an hour each day for 5–6 weeks around the surgery, % (n)
Don't know	10.5 (2)	6.7 (1)	0.0 (0)	20.0 (3)	17.7 (3)	0 (0)
No	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Yes	89.5 (17)	93.3 (14)	100.0 (19)	80.0 (12)	82.4 (14)	100.0 (17)
Readiness for training in one session, or split to several short sessions a day?, % (n)
Don't know	22.2 (4)	7.7 (1)	11.1 (2)	23.1 (3)	20.0 (3)	12.5 (2)
One session	33.3 (6)	46.2 (6)	27.8 (5)	53.9 (7)	46.7 (7)	31.3 (5)
Split	44.4 (8)	46.2 (6)	61.1 (11)	23.1 (3)	33.3 (5)	56.3 (9)

Table  5 presents responses to narrative questions, highlighting potential factors influencing the implementation of a neurocognitive training intervention, including mode of intervention, frequency of intervention sessions, the timing of the intervention, and delivery of the intervention.

Table 5.

CFIR-based implementation factors

Implementation Factor	Position (Number of Participants Agreeing)	Quotations
Mode of intervention	The mode of the intervention is accessible and feasible for most older patients and younger patients alike (8 clinicians, 23 patients)	“My own experience with my 70+ grandma and my mom, who would be … our surgical candidate ages, is they're on their phones as much or more than me” (C2, CTICU). “I use [a smartphone] every day … with work and stuff” (P12).
	The mode of the intervention is inaccessible to some older patients; some patients who have access to smartphones do not use them frequently or extensively to play games (2 clinicians, 13 patients)	“Some patients are elderly, and they may not have a smartphone on, and they may not work on a smartphone” (C4, FG2). “I wouldn't even know how to bring my games up on my phone. I've never done it before. I would probably just not know how to play a game on the phone. Even if somebody showed me, I probably wouldn't do it” (P12). “[I only use smartphones] when [I have to] answer a call or check email … [I] don't get on it to search anything” (P26).
Frequency of intervention sessions	One longer session is preferred (1 clinician, 9 patients)	“If you told [patients to do the intervention] twice a day, they might do it once. So … do it once” (C2, CTICU). “[It would be easier to] do it all at one time to get it over with” (P11).
	Multiple, shorter sessions would be preferred over one longer session each day (6 clinicians, 11 patients)	“I think having the flexibility to do it in two 15-minute sessions lets people fit it in while they drink their coffee. If anybody asked for me to do something for 30 minutes a day, in my experience, it would make it harder for me to fit them into my life” (C1, FG2). “Yeah, if you split it up two 15 minutes. I'm not available on certain days at certain times. I play golf on Thursday mornings, so I wouldn't be able to do them until the afternoon there, and sometimes I'm not available at different times of the day” (P17).
	A long program (5–6 weeks) would see decreased adherence across patients (3 clinicians, 2 patients)	“Half an hour a day routinely for five to six weeks seems like a lot, especially if you don't feel great. I think that is a pretty high goal” (C2, FG1). “I probably wouldn’t do it for 5–6 weeks; I'd probably do it for a week or two” (P35).
Timing of intervention	A preoperative intervention implementation point would be most helpful—during surgery scheduling (9 clinicians, 2 patients)	“A week before and maybe four weeks afterwards … would be reasonable and more people might be able to complete that” (C1, FG2). “I would try it if it was something I would do before the surgery. I would definitely do it before surgery, definitely and again, I would do it after surgery if I was getting gratification and it was actually distracting me … Yeah, I'd do it” (P30).
Delivery of intervention	The intervention should be introduced by the surgeons and explained more thoroughly within the CPAP by nurse practitioners and physician assistants	“It's something that I believe the surgeons should be bringing up with the patients. So probably providing the clinicians and probably not just the surgeons, but the nurse practitioners and [physician] assistants who are interacting with the patients” (C1, FG4). “Doctors and nurse assistants [were the ones most active in my preoperative discussions]” (P38).

Discussion

With PPSP becoming a significant public health issue, previous studies have highlighted the usefulness of understanding patient and clinician perspectives to develop, tailor, and implement PPSP interventions across populations (i.e., abdominal, thoracic, orthopedic, and general surgery) [33–39]. Patient participation is invaluable in decision-making processes, as individualized pain management plans based on patient goals, resources, and pain management strategies can effectively address their chronic pain needs [37, 39].

Our study suggests that discussions about the risk and management of persistent pain are rare during the perioperative period. Given the substantial rates of PPSP [2], there is a need for pain management discussions and preemptive interventions that can prevent persistent pain before it occurs. Our findings suggest that patients had limited understanding of postoperative pain and its management, despite some reporting a strong desire to receive this information and manage their expectations realistically. Furthermore, potential power dynamics between clinicians and their patients could influence how infrequently patients felt comfortable asking for more details. Given recent data suggesting that presurgical expectations can affect postsurgical rates of pain [40], it would be essential to develop personalized approaches for the optimal alignment of expectations. Many challenges were related to overall perioperative pain management, suggesting that optimizing pain education, communication, and coordination should be heavily prioritized.

Previous research has suggested that patients receive minimal information about pain preoperatively [34, 35]. One study revealed that nurses inaccurately reported lower patient pain, in part, because of poor communication with their patients [34]. Similarly, our clinician participants concurred that chronic pain discussions were uncommon and presurgical discussions primarily centered on the surgery itself. They noted a need for better postoperative pain management and considered the proposed neurocognitive training a low-cost, low-risk intervention that their colleagues and patients would accept. Clinicians cited potential barriers for patients, including technological learning ability (particularly with older patients) and time commitment (affecting adherence). They recommended in-hospital education and training and “real-time” support to overcome these barriers. Like patients, clinicians felt that intervention adherence would improve with reminder notifications and incentives for engagement. Findings across studies have demonstrated patients’ value in decisions about their pain management in the perioperative setting [41].

Although studies point to numerous PPSP risk factors, few interventions have attempted to prevent PPSP by mitigating modifiable risk factors. One such factor is poor presurgical cognitive flexibility, which could be improved through neurocognitive training interventions [25].

The majority of participants reported willingness to engage in a neurocognitive training intervention. Although they perceived several potential implementation barriers in addition to immediate postoperative pain and stress (e.g., fatigue, cognitive overload, technical challenges), most did not consider these barriers insurmountable. They also identified different facilitators to address these barriers and strengthen buy-in and adherence to intervention use and acceptability, such as providing educational materials and resources for patients about the intervention, including data about its impact on PPSP. Clinicians and researchers should involve patient partners in developing these materials and resources to ensure their clarity, practicality, and relevance for patients [42].

Our study has several limitations. First, this study was conducted at a single academic center, limiting generalizability beyond our sample and setting. Furthermore, participants were from a single surgical population (thoracic surgery), which is known to have higher rates of PPSP [2, 3]. Therefore, it is likely that other surgical populations have different experiences with pain and unique pain management needs. Likewise, within a specific type of surgical population, the indication for surgery (e.g., cancer vs. non-cancer, transplantation) could affect patients’ pain condition and their perspective on interventions for pain. Thus, this area would benefit from a further qualitative study with diverse surgical populations and diagnostic indications for surgery. Additionally, though we interviewed anesthesiologists and surgeons who care for patients at different time points throughout the perioperative period, we did not collect data from non-physicians. The perspectives of other clinician stakeholders, including nurses, physician assistants, nurse practitioners, and non-attending physician trainees, might differ from those of the physicians we interviewed and provide further insight into the feasibility of using this intervention with this population. However, we view the present study as a starting point that will inform future studies that are more inclusive of stakeholders to test the effective and seamless integration of a neurocognitive intervention as part of the larger perioperative pain management and prevention paradigm. The present study did not collect information on patients’ race/ethnicity, education, or employment. It is possible that these variables could influence the feasibility of this intervention and its reception among patients. Finally, we did not ask patients about several pain-related factors that could confound their perspectives on pain, including pain level, previous experience with pain, prior surgeries, and medications.

Despite these limitations, findings from the present study can inform the design of interventional programs to minimize the impact of PPSP on patients’ functioning and quality of life. As part of future research efforts, we intend to incorporate this feedback and engage patient partners to tailor a neurocognitive intervention to help prevent PPSP in patients undergoing thoracic surgeries.

Supplementary Data

Supplementary Data may be found online at http://painmedicine.oxfordjournals.org.

 

Funding sources: Research reported in this publication was supported by the Washington University Center for Perioperative Mental Health grant number P50 MH122351 from the National Institute of Mental Health of the National Institutes of Health (NIH). The authors received no financial support for the research, authorship, and/or publication of this article. The content is solely the responsibility of the authors and does not necessarily represent the official view of the NIH.

Conflicts of interest: The authors report no conflicts of interest.