Defining Patient-relevant Thresholds and Change Scores for the HOOS JR and KOOS JR Anchored on the Patient-acceptable Symptom State Question

Abstract

Background

When evaluating the results of clinical research studies, readers need to know that patients perceive effect sizes, not p values. Knowing the minimum clinically important difference (MCID) and the patient-acceptable symptom state (PASS) threshold for patient-reported outcome measures helps us to ascertain whether our interventions result in improvements that are large enough for patients to care about, and whether our treatments alleviate patient symptoms sufficiently. Prior studies have developed the MCID and PASS threshold for the Hip Disability and Osteoarthritis Outcome Score for Joint Replacement (HOOS JR) and Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR) anchored on satisfaction with surgery, but to our knowledge, neither the MCID nor the PASS thresholds for these instruments anchored on a single-item PASS question have been described.

Questions/purposes

(1) What are the MCID (defined here as the HOOS/KOOS JR change score associated with achieving PASS) and PASS threshold for the HOOS JR and KOOS JR anchored on patient responses to the single-item PASS instrument? (2) How do patient demographic factors such as age, gender, and BMI correlate with MCID and PASS thresholds using the single-item PASS instrument?

Methods

Between July 2020 and September 2021, a total of 10,970 patients underwent one primary unilateral THA or TKA and completed at least one of the three surveys (preoperative HOOS or KOOS JR, 1-year postoperative HOOS or KOOS JR, and 1-year postoperative single-item anchor) at one large, academic medical center. Of those, only patients with data for all three surveys were eligible, leaving 13% (1465 total; 783 THAs and 682 TKAs) for analysis. Despite this low percentage, the overall sample size was large, and there was little difference between completers and noncompleters in terms of demographics or baseline patient-reported outcome measure scores. Patients undergoing bilateral total joint arthroplasty or revision total joint arthroplasty and those without all three surveys at 1 year of follow-up were excluded. A receiver operating characteristic curve analysis, leveraging a 1-year, single-item PASS (that is, “Do you consider that your current state is satisfactory?” with possible answers of “yes” or “no”) as the anchor was then used to establish the MCID and PASS thresholds among the 783 included patients who underwent primary unilateral THA and 682 patients who underwent primary unilateral TKA. We also explored the associations of age at the time of surgery (younger than 65 years or 65 years and older), gender (men or women), BMI (< 30 or ≥ 30 kg/m²), and baseline Patient-Reported Outcome Measure Information System-10 physical and mental component scores (< 50 or ≥ 50) for each of the MCID and PASS thresholds through stratified analyses.

Results

For the HOOS JR, the MCID associated with the PASS was 23 (95% CI 18 to 31), with an area under the receiver operating characteristic curve of 0.75, and the PASS threshold was 81 (95% CI 77 to 85), with an area under the receiver operating characteristic curve of 0.81. For the KOOS JR, the MCID was 16 (95% CI 14 to 18), with an area under the receiver operating characteristic curve of 0.75, and the PASS threshold was 71 (95% CI 66 to 73) with an area under the receiver operating characteristic curve of 0.84. Stratified analyses indicated higher change scores and PASS threshold for younger men undergoing THA and higher PASS thresholds for older women undergoing TKA.

Conclusion

Here, we demonstrated the utility of a single patient-centered anchor question, raising the question as to whether simply collecting a postoperative PASS is an easier way to measure success than collecting preoperative and postoperative patient-reported outcome measures and then calculating MCIDs and the substantial clinical benefit.

Level of Evidence

Level III, therapeutic study.

Introduction

Patient-reported outcome measures (PROMs) are commonly used to evaluate the preoperative and postoperative pain and functional status of patients undergoing procedures such as THA and TKA [1, 25, 26, 36, 46]. Given that raw numeric values from these metrics are challenging to interpret [2, 11, 16, 32, 40], constructs to frame PROM scores in the context of improvement over time have been developed. Broadly speaking, PROM metrics can be characterized as either change-based metrics (such as the minimum clinically important difference [MCID] and the substantial clinical benefit), which focus on improvement [12, 13, 29, 34], or threshold metrics, which focus on the final outcome (such as the patient-acceptable symptom state [PASS] threshold) [10, 21, 33]. The MCID reflects the minimum change in PROM scores that patients can perceive as a change in their health [29]. The PASS threshold is a PROM score at which patients report that their symptoms are acceptable and is closely associated with patient satisfaction [47]. These metrics are complementary; the MCID emphasizes whether an individual has improved after a certain therapy, whereas the PASS threshold emphasizes whether the achieved outcome is acceptable from the patient’s perspective [24, 34, 38, 44]. The MCID and PASS thresholds allow for the interpretation of PROMs in the context of a given treatment, and they may fulfill a variety of roles: as clinically meaningful benchmarks, patient-centered outcomes, and guides for surgeons to contextualize a patient’s symptom state. These standards can aid clinicians in understanding care outcomes and can be used to measure quality, allowing for a more comprehensive assessment of value informed by outcomes that matter to patients [19, 41].

Although there is no consensus on the best way to use these metrics to determine the efficacy of an orthopaedic intervention [35, 42], there is growing reliance on the notion that “it’s good to feel better but it’s better to feel good” [43]. In contrast to other commonly used PROMs that rely on change in scores and hence must be collected at two timepoints, the PASS single-item survey instrument (not to be confused with the PASS threshold) identifies patients who feel good enough regardless of the amount of improvement they achieve, and it requires administration at only one timepoint.

Different methods have been proposed to define meaningful PROM change scores or thresholds. These methods are generally grouped into anchor-based or distribution-based approaches [8]. In anchor-based approaches, PROM scores are compared either at a single timepoint (differences between different groups at a single timepoint) or longitudinally (differences within a single group over time), with the use of a patient assessment anchor to evaluate the perceived importance of an observed change. On the other hand, distribution-based approaches rely only on statistical criteria from the PROM instrument (such as standard deviations or standard errors of measurement) to estimate clinical relevance [48]. Although an anchor-based approach is generally preferred because it is more patient-centric and not based on arbitrary statistical criteria, both methods have limitations. Anchor-based approaches have an element of subjectivity based on the choice of the anchor, which is intrinsically a subjective assessment (such as patients report feeling “no better,” “a little better,” or “a lot better”). The validity and reliability of anchor-based change scores or thresholds depend on the validity of the anchor itself. Distribution-based approaches, despite their simplicity to use, can only identify a minimal detectable change. In other words, it is a change that is unlikely to be attributable to random measurement error but is not inherently linked to clinical importance. They only define the minimum value below which a change in pain score is not because of measurement error.

Developed specifically for THA and TKA, the Hip Injury and Osteoarthritis Outcome Score, Joint Replacement (HOOS JR) and Knee Injury and Osteoarthritis Outcome Score, Joint Replacement (KOOS JR) are frequently administered to inform clinical decisions and as measurements of pain and function for clinical studies. Additionally, these PROMs are currently used or are proposed for use in several Centers for Medicare and Medicaid Services quality measures, one of which uses an anchor-based PASS threshold for TKA [4]. However, prior studies have produced inconsistent PASS thresholds for the HOOS JR and KOOS JR [4, 21], limiting the clinical utility of both instruments in defining success. It is necessary to better define the PROM threshold at which a patient achieves a PASS to better inform patients, physicians, and payers about surgical outcomes. The purpose of this study was to report the MCIDs for the KOOS JR and HOOS JR based on calculations from the single-item PASS survey instrument, and to similarly define the PASS thresholds for these instruments using the same anchor. Importantly, the MCIDs presented here could more objectively be labeled as the HOOS or KOOS JR change score associated with achieving a PASS—although for consistency and reconciliation with prior evidence, we use the label “MCID.”

In this study, we asked: (1) What are the MCID (defined here as the HOOS/KOOS JR change score associated with achieving PASS) and PASS threshold for the HOOS JR and KOOS JR anchored on patient responses to the single-item PASS instrument? (2) How do patient demographic factors such as age, gender, and BMI correlate with MCID and PASS thresholds using the single-item PASS instrument?

Patients and Methods

Study Design and Setting

We conducted a retrospective study between July 2020 and September 2021 from the largest longitudinally maintained institutional database in the United States on patients undergoing unilateral primary total joint arthroplasty (TJA) using PROM data (HOOS JR, KOOS JR, and single-item PASS) collected preoperatively and 1 year postoperatively.

The HOOS JR was developed from the original long version of the HOOS survey using Rasch analysis [28]. The HOOS JR contains six items from the original HOOS survey that are coded from 0 to 4, none to extreme. The HOOS JR score is determined by summing the raw response (range 0 to 24) and then converting it to an interval score using a validated conversion table. The interval score ranges from 0 to 100, where 0 represents total hip disability and 100 represents perfect hip health.

Similarly, the KOOS JR was developed from the original long version of the KOOS survey using Rasch analysis [27]. The KOOS JR contains seven items from the original KOOS survey that are coded from 0 to 4, none to extreme. The KOOS JR score is determined by summing the raw response (range 0 to 28) and then converting it to an interval score using a validated conversion table. The interval score ranges from 0 to 100, where 0 represents total knee disability and 100 represents perfect knee health.

The single-item PASS survey question, which is used as an anchor, reads: “Do you consider that your current state is satisfactory?” The possible answers were either “yes” or “no.”

Patients

In our database, between July 2020 and September 2021, a total of 10,970 patients underwent one primary unilateral THA or TKA and completed at least one of the three surveys (preoperative HOOS or KOOS JR, 1-year postoperative HOOS or KOOS JR, and 1-year postoperative single-item anchor). The individual collection proportions for each of the surveys over this period were 84% for preoperative HOOS and KOOS JR, 55% for 1-year postoperative HOOS and KOOS JR, and 29% for the 1-year postoperative single-item PASS anchor. Including only patients who completed all three surveys, we were left with 13% (1465 total; 783 THAs and 682 TKAs) for analysis. Despite this low percentage, the overall sample size is large, and there was little difference in terms of demographics or baseline PROM scores between completers and noncompleters, whose relatively robust collection proportions are reported. No imputation methods were applied. We excluded patients who underwent more than one TJA during the study period, including those who underwent revision TJA, contralateral TJA, and simultaneous bilateral TJA, to avoid including patients whose subsequent surgical procedures could alter their 1-year PROM scores.

Before surgical consultations for TKA and THA, patients were sent an email or text message requesting that they complete several PROMs including the HOOS JR or KOOS JR. Patients who did not complete these surveys in advance of their visit could complete them at their visit. Starting at 11 months after the surgery, patients received an email and had 30 days to complete several PROMs, including the HOOS or KOOS JR and the single-item PASS survey question, before their anniversary. A reminder was sent 7 days before the 365-day mark. Patients also had an opportunity to complete these PROMs at their 1-year follow-up visit. Hence, the window for completion of 1-year postoperative PROMs was 335 to 395 days after surgery.

Descriptive Data

As noted, 1465 patients who underwent elective primary unilateral TJA (783 THAs and 682 TKAs) were included. Among patients with complete data, the mean age of the THA and TKA cohorts was 69 ± 9 years and 70 ± 8 years, respectively, and the mean BMI was 28 ± 6 kg/m² and 30 ± 6 kg/m², respectively. A total of 58% (847) of patients were women (Table 1). We further note summary statistics for preoperative and postoperative HOOS JR and KOOS JR scores (Table 2).

Table 1.

Baseline characteristics of included patients undergoing THA or TKA (n = 1465)

Baseline characteristic	THA (n = 783)	TKA (n = 682)
Age in years	69 ± 9	70 ± 8
BMI in kg/m2	28 ± 6	30 ± 6
Gender, men	42 (329)	42 (289)
PROMIS-10 PCS preoperative	42 ± 7	42 ± 7
PROMIS-10 MCS preoperative	51 ± 9	51 ± 8

Data presented as mean ± SD or % (n). PROMIS = Patient-Reported Outcome Measurement Information System; PCS = physical component score; MCS = mental component score.

Table 2.

Median and IQR for preoperative and postoperative PROMs among patients who underwent THA or TKA

	Median (IQR)
THAa
Preoperative HOOS JR	53 (47 to 65)
Postoperative HOOS JR	92 (81 to 100)
TKAb
Preoperative KOOS JR	53 (45 to 59)
Postoperative KOOS JR	76 (66 to 85)

^aThe HOOS JR is scored from 0 to 100, where 0 represents total hip disability and 100 represents perfect hip health.

^aThe KOOS JR is scored from 0 to 100, where 0 represents total knee disability and 100 represents perfect knee health.

Primary and Secondary Study Outcomes

Our primary study goal was to calculate the MCID and PASS thresholds for the HOOS JR and KOOS JR anchored on patient responses to the single-item PASS instrument. The HOOS and KOOS JR scores were treated as continuous variables, and the answers to the single-item PASS survey question were treated as a binary variable. The MCID and PASS threshold were established using receiver operating characteristic curves (ROCs) to find the cutoff that best discriminated between patients answering “yes” to the PASS question and those answering “no.” The identified cutoff was determined to be the value at which the sum of sensitivity and specificity - 1 (Youden index) was maximized [21]. The area under the ROC curve estimates and their associated 95% confidence intervals and p values are also provided. The AUC estimate can be interpreted as the probability that a randomly chosen satisfied-patient score will have a higher score than a randomly chosen dissatisfied-patient score. An AUC between 0.7 and 0.8 is considered acceptable and 0.8 and 0.9 is considered excellent [7].

We quantified CIs for the reported thresholds using a described percentile bootstrapping method [21]. We randomly drew, with replacement, a sample equal in size to the overall sample and repeated this 1000 times. The resulting MCID and PASS threshold results from each draw were then sorted in ascending order, and the 25th and 975th results are reported as the 95% CI. Occasionally, because the underlying HOOS JR and KOOS JR scores were noncontinuous and many patients reached an acceptable state, the CIs collapsed. These cases, in which one or more of the CI bounds was equal to the point estimate itself, demonstrate instances in which the underlying variability in our MCID and PASS estimates was less than the resolution in the HOOS JR and KOOS JR instruments. As such, we also report the full discrete distribution of the 1000 bootstrapped iterations.

Our secondary study goal was to elucidate how demographic factors such as age, gender, BMI, and baseline functional scores correlate with MCID and PASS thresholds using the single-item PASS instrument. To this end, we stratified our cohort by age at surgery (younger than 65 years or 65 years or older; relevant to Medicare coverage status), gender (men or women), BMI (< 30 or ≥ 30 kg/m²), baseline Patient-Reported Outcome Measure Information System-10 (PROMIS) physical component score (PCS; < 50 or ≥ 50), and baseline PROMIS-10 mental component score (MCS; < 50 or ≥ 50). These analyses were repeated independently for each stratification, and the MCID and PASS thresholds for each subgroup are similarly reported.

Ethical Approval

Ethical approval for this study was obtained from the institutional review board of the Hospital for Special Surgery, New York, NY, USA.

Results

MCID and PASS Thresholds for the HOOS JR and KOOS JR

The MCIDs for the HOOS JR and KOOS JR were 23 (95% CI 18 to 31) and 16 (95% CI 14 to 18), respectively, both with an area under the ROC (AUROC) of 0.75. The PASS thresholds for the HOOS JR and KOOS JR were 81 (95% CI 77 to 85) and 71 (95% CI 66 to 73), respectively, with AUROCs of 0.81 and 0.84, respectively. A total of 88% (689 of 783) of our patients achieved the PASS at 1 year after THA, while 79% (539 of 682) achieved the PASS at 1 year after TKA.

Covariate Analyses of MCID and PASS Thresholds for the HOOS JR and KOOS JR

Controlling for the individual variables above, we demonstrated that point estimates of MCIDs for HOOS JR were higher in younger men with a BMI greater than 30 kg/m² and with preoperative PROMIS MCS and PCS of less than 50 (Table 3). However, for the KOOS JR, we demonstrated that MCIDs were equal across tested covariates (Table 4). Using a similar approach to the PASS thresholds, we found that point estimates of PASS thresholds for the HOOS JR were higher in younger men with a BMI of less than 30 kg/m² and with a preoperative PROMIS MCS or PCS of 50 or more (Table 5). This contrasted with the KOOS JR, where point estimates of PASS thresholds were higher in older women with a BMI of less than 30 kg/m² and those with a preoperative PROMIS PCS of 50 or more (Table 6).

Table 3.

HOOS JR^a MCID, overall and stratified by covariates

	Sample Size	MCID (95% CI)	AUROCb
Overall	783	23 (18 to 31)	0.75
Age younger than 65 years	227	30 (16 to 33)	0.89
Age 65 and older	556	23 (15 to 26)	0.71
BMI < 30 kg/m2	511	19 (18 to 31)	0.76
BMI ≥ 30 kg/m2	271	28 (12 to 39)	0.75
Men	329	23 (18 to 35)	0.69
Women	454	21 (15 to 31)	0.79
Preoperative PROMIS-10 PCS < 50c	646	23 (18 to 34)	0.78
Preoperative PROMIS-10 PCS ≥ 50	133	19 (0 to 19)	0.65
Preoperative PROMIS-10 MCS < 50c	354	31 (18 to 36)	0.79
Preoperative PROMIS-10 MCS ≥ 50	425	22 (15 to 26)	0.70

^aThe HOOS JR is scored from 0 to 100, where 0 represents total hip disability and 100 represents perfect hip health.

^bAUROC between 0.7 and 0.8 is considered acceptable, and between 0.8 and 0.9 is considered excellent.

^cA score of 50 represents the mean of the general population, and higher scores indicate better physical or mental health. AUROC = area under the receiver operating curve; PROMIS = Patient-Reported Outcome Measurement Information System; PCS = physical component score; MCS = mental component score.

Table 4.

KOOS JR^a MCID, overall and stratified by covariates

	Sample size	MCID (95% CI)	AUROCb
Overall	682	16 (14 to 18)	0.75
Age younger than 65 years	162	16 (12 to 24)	0.74
Age older than 65 years	520	16 (9 to 25)	0.75
BMI < 30 kg/m2	363	16 (7 to 18)	0.75
BMI ≥ 30 kg/m2	318	16 (14 to 25)	0.75
Men	289	17 (9 to 21)	0.81
Women	393	16 (14 to 26)	0.71
Preoperative PROMIS-10 PCS < 50c	573	16 (14 to 22)	0.76
Preoperative PROMIS-10 PCS ≥ 50	99	-7 (-9 to 22)	0.74
Preoperative PROMIS-10 MCS < 50c	290	16 (10 to 25)	0.72
Preoperative PROMIS-10 MCS ≥ 50	382	15 (8 to 24)	0.79

^aThe KOOS JR is scored from 0 to 100, where 0 represents total knee disability and 100 represents perfect knee health.

^bAUROC between 0.7 and 0.8 is considered acceptable, and between 0.8 and 0.9 is considered excellent.

^cA score of 50 represents the mean of the general population, and higher scores indicate better physical or mental health. AUROC = area under the receiver operating curve; PROMIS = Patient-Reported Outcome Measurement Information System; PCS = physical component score; MCS = mental component score.

Table 5.

PASS thresholds for the HOOS JR^a overall and stratified by covariates

	Sample size	PASS threshold (95% CI)	AUROCb
Overall	783	81 (77 to 85)	0.81
Age younger than 65 years	227	85 (77 to 92)	0.90
Age older than 65 years	556	81 (77 to 85)	0.78
BMI < 30 kg/m2	511	81 (77 to 85)	0.84
BMI ≥ 30 kg/m2	271	77 (65 to 92)	0.76
Men	329	85 (77 to 92)	0.75
Women	454	81 (74 to 85)	0.84
Preoperative PROMIS-10 PCS < 50	646	81 (77 to 85)	0.81
Preoperative PROMIS-10 PCS ≥ 50	133	85 (70 to 92)	0.72
Preoperative PROMIS-10 MCS < 50	354	77 (77 to 85)	0.80
Preoperative PROMIS-10 MCS ≥ 50	425	85 (77 to 85)	0.80

^aThe HOOS JR is scored from 0 to 100, where 0 represents total hip disability and 100 represents perfect hip health.

^bAUROC between 0.7 and 0.8 is considered acceptable, and between 0.8 and 0.9 is considered excellent. AUROC = area under the receiver operating curve; PROMIS = Patient-Reported Outcome Measurement Information System; PCS = physical component score; MCS = mental component score.

Table 6.

PASS thresholds for the KOOS JR^a overall and stratified by covariates

	Sample size	PASS threshold (95% CI)	AUROCb
Overall	682	71 (66 to 73)	0.84
Age younger than 65 years	162	64 (64 to 66)	0.87
Age older than 65 years	520	71 (66 to 73)	0.83
BMI < 30 kg/m2	363	71 (66 to 71)	0.85
BMI ≥ 30 kg/m2	318	68 (64 to 73)	0.82
Men	289	68 (66 to 71)	0.87
Women	393	71 (64 to 76)	0.81
Preoperative PROMIS-10 PCS < 50	573	66 (64 to 71)	0.83
Preoperative PROMIS-10 PCS ≥ 50	99	76 (55 to 76)	0.93
Preoperative PROMIS-10 MCS < 50	290	68 (66 to 73)	0.79
Preoperative PROMIS-10 MCS ≥ 50	382	68 (64 to 71)	0.87

^aThe KOOS JR is scored from 0 to 100, where 0 represents total knee disability and 100 represents perfect knee health.

^bAUROC between 0.7 and 0.8 is considered acceptable, and between 0.8 and 0.9 is considered excellent. AUROC = area under the receiver operating curve; PROMIS = Patient-Reported Outcome Measurement Information System; PCS = physical component score; MCS = mental component score.

Discussion

The MCIDs and PASS thresholds anchored on the single-item PASS questionnaire allow for a clinical interpretation of PROMs from the patient’s perspective of the success of a given treatment. No studies of which we are aware have established the PASS threshold for the HOOS JR, and prior studies seeking to establish an anchor-based PASS threshold for the KOOS JR have reported inconsistent thresholds, thereby limiting the utility of both instruments to patients, providers, and payers. Here, we report the MCIDs for the HOOS JR and KOOS JR as 23 and 16, respectively, and the PASS thresholds for the HOOS JR and KOOS JR as 81 and 71, respectively. Interpreting these values, we demonstrated that the single-item PASS performs well in relation to previously established MCIDs for the HOOS JR and KOOS JR, raising the question as to whether simply collecting a postoperative PASS is an easier way to measure success than collecting both preoperative and postoperative PROMs and then calculating MCIDs.

Limitations

First, our single-institution design with 1-year follow-up may be a potential limitation to generalizability to other populations or follow-up periods, particularly because the MCID and PASS may depend on time and population [6, 20, 21]. However, our sample is the largest study of the single-item PASS anchor (“Do you consider that your current state is satisfactory?”), and the values reported herein provide a consistent and interpretable result when linking HOOS and KOOS JR scores to the single-item PASS anchor.

Second, we acknowledge that the use of a binary, single-timepoint PASS survey may not capture all the nuances that comprise patient satisfaction. However, the single-item PASS has been validated [9, 24, 43], and this study contributes to what is known about this PROM (and PROMs more generally) by removing the subjectivity of defining anchor thresholds inherent in other multiresponse anchors. In other work performed by our team to define anchor-based MCIDs and substantial clinical benefit [21, 29], we needed to make decisions on where the cutoff would be across the anchors (for example, should the cutoff to define the MCID be minimal improvement or moderate improvement?). Although these decisions were informed by the data, there was still some degree of judgment or subjectivity on the part of the investigators. The binary nature of the PASS removes this subjectivity.

In addition, the proportion of patients who completed the PASS survey was low (29%). This stems from the relatively recent implementation of routine collection of this measure at our institution. Despite this low percentage, the overall sample size was large, and there was little difference between completers and noncompleters in terms of demographics or baseline PROM scores, for which we have a high proportion of patients who completed the surveys (84% preoperative and 55% at 1 year postoperatively). As such, we present our findings with confidence and encourage a consensus at the registry level about the routine collection of anchors that are most convenient, widely collected, and clinically interpretable. Doing so would aid in the ongoing effort to define the best way to measure quality that is valid, accurate, reliable, and meaningful. Moreover, our study excluded patients who underwent a revision or additional TJA procedure during the follow-up period. Although this criterion was important given that additional surgery might influence follow-up PROMs or responses to the anchor questions, it may not reflect the clinical reality of additional or revision procedures. Lastly, given that the aim of this study was to establish MCID and PASS thresholds for patients undergoing TJA, we did not establish individualized MCID and PASS thresholds that accounted for the various patient-specific factors or baselines that may affect MCID and PASS [3, 5, 17, 18, 20, 22, 36, 39, 45]. How these variables interact to influence the MCID and PASS is not well understood; future studies are needed to elucidate individualized MCID and PASS thresholds patients undergoing TJA.

Discussion of Key Findings

Our reported values for the MCID for the HOOS JR and KOOS JR were 23 and 16, respectively. Our reported values for the PASS threshold for the HOOS JR and KOOS JR were 81 and 71, respectively. Similar to reported thresholds [4, 21], our results suggest a higher MCID and PASS threshold for the HOOS JR than for the KOOS JR, consistent with the finding that patients undergoing THA are often more satisfied with their outcome (PASS) and have greater changes in PROM scores than patients undergoing TKA [3, 30, 49].

Furthermore, the range of our dataset for the KOOS JR (see Table 2) suggests minimal floor and ceiling effects for calculating the MCID and PASS thresholds. There appears to be a ceiling for the HOOS JR, but it is well above the calculated PASS threshold. The spread across the 25th and 75th percentiles also suggests our calculated PASS threshold could be useful clinically, because it might help to differentiate patients who have a range of HOOS and KOOS JR scores. In this context, our gold-standard thresholds also provide a reliable value on which to base meaningful clinical outcomes moving forward. With an increasing trend in policy toward value-based care that relies on the accurate identification of success, we report values for the single-item PASS that can be incorporated into existing Centers for Medicare and Medicaid Services measures to validate value-based care.

The MCIDs anchored to the single-item PASS instrument for the HOOS JR and KOOS JR reported here are higher than many previously reported MCIDs, and they even exceed the substantial clinical benefit reported for the HOOS JR (Table 7) [29]. The fact that our presented change scores are larger than those of most prior reports [16, 23, 29] may reflect our limited sample size because collection of the single-item PASS instrument was not yet standard practice at the time of this study. However, in obtaining an MCID that exceeds prior MCIDs and even surpasses the reported substantial clinical benefit, we highlight the relatively arbitrary nature of various PROM metrics. Theoretically, achieving PASS should mean a patient achieves perceptible change (MCID); that is, he or she must be able to notice a difference from baseline to claim a newly satisfactory state regarding his or her health. However, as currently defined, this is not necessarily true; the MCID reflects a change that depends on multiple-timepoint measurements, as opposed to PASS, which emphasizes a single-timepoint evaluation of symptomatic state. Taken further, a patient's perception of pain, stiffness, dysfunction, and what is acceptable may recalibrate after surgery, such that postoperative assessments are not measuring the same things as the preoperative assessments. Accordingly, the MCID presented here should more objectively be labeled as the HOOS or KOOS JR change score associated with achieving a PASS—although for consistency and reconciliation with prior evidence, we use the label “MCID.”

Table 7.

Comparison of studies presenting anchor-based change scores for HOOS JR and KOOS JR

Study	Change score	Attributed category	Mean follow-up	Anchor item and response	Anchor instrument
HOOS JR
Hung et al. [15]	18	MCID	3 to 6 months	Compared to your FIRST EVALUATION at the University Orthopaedic Center: How would you describe your physical function now?”	UOC satisfaction surveya
				2- or 3-point change
Lyman et al. [29]	18	MCID	2 years	“How much did your surgery improve your quality of life?”	HSS satisfaction surveyb
				Moderate improvement
Lyman et al. [29]	22	SCB	2 years	“How much did your surgery improve your quality of life?”	HSS satisfaction surveyb
				Great improvement OR more improvement than I dreamed possible
Kuo et al. [23]	18	MCID	1 year	SAPS items: (1) satisfaction with results of surgery, (2) satisfaction with results of surgery for improving pain, (3) satisfaction with results of surgery for improving ability to do home or yard work, and (4) satisfaction with results of surgery for improving ability to do recreational activities	SAPSc
				All SAPS items > 50 (4-point Likert scale)
Guenthner et al. [14]	35	MCID		“In general, would you say your quality of life is:” (5-point Likert scale)	PROMIS-10 QOLd
				1- or 2-point increase
Current study	23	MCID	1 year	“Do you consider that your current state is satisfactory?”	PASS
				Yes
KOOS JR
Hung et al. [15]	15	MCID	3-6 months	Compared to your FIRST EVALUATION at the University Orthopaedic Center: How would you describe your physical function now?”	UOC satisfaction surveya
				2- or 3-point change
Lyman et al. [29]	14	MCID	2 years	“How much did your surgery improve your quality of life?”	HSS satisfaction surveyb
				Moderate improvement
Lyman et al. [29]	20	SCB	2 years	“How much did your surgery improve your quality of life?”	HSS satisfaction surveyb
				Great improvement OR more improvement than I dreamed possible
Kuo et al. [23]	21	MCID	1 year	SAPS items: (1) satisfaction with results of surgery, (2) satisfaction with results of surgery for improving pain, (3) satisfaction with results of surgery for improving ability to do home or yard work, and (4) satisfaction with results of surgery for improving ability to do recreational activities	SAPSc
				All SAPS items > 50 (4-point Likert scale)c
Only et al. [35]	29	MCID	1 year	“In general, would you say your quality of life is:” (5-point Likert scale)	PROMIS-10 QOLd
				1- or 2-point increase
Current study	16	MCID	1 year	“Do you consider that your current state is satisfactory?”	PASS
				Yes

^aFor the UOC (University Orthopaedic Center) satisfaction survey, no change equates to a 0 value; the negative ratings are from -3 to -1 and positive ratings are from 1 to 3.

^bIn this QOL survey, answers include “more improvement than I ever dreamed possible,” “great improvement,” “moderate improvement,” “a little improvement,” or “no improvement.”

^cThe SAPS (Self-administered patient satisfaction scale) instrument is scored on a 4-point Likert scale, with responses consisting of very satisfied (100 points), somewhat satisfied (75 points), somewhat dissatisfied (50 points), and very dissatisfied (25 points). The SAPS total score is the unweighted mean of the individual items, ranging from 25 to 100.

^dThe PROMIS-10 (Patient-Reported Outcome Measurement Information System) QOL (quality of life) anchor questions are rated on a 5-point Likert scale (poor, fair, good, very good, or excellent), where the difference between increasing responses is 1 point. MCID = minimum clinically important difference; SCB = substantial clinical benefit; HSS = Hospital for Special Surgery.

The single-item PASS threshold for the HOOS JR is higher than a reported PASS threshold (81 versus 77 [21]). The single-item PASS threshold for the KOOS JR calculated here agrees with the reported crosswalk-derived PASS (71) [4], lending validity to our construct, and is higher than a reported PASS threshold (64) [21]. The similarity of our KOOS JR PASS threshold to the crosswalk-derived KOOS JR threshold, and the differences between previous HOOS JR and KOOS JR MCIDs and our MCID and PASS thresholds, may be explained by considering when patients were queried postoperatively. The MCID and PASS in the current study were defined at 1 year after surgery, and the crosswalk-derived PASS was determined 9 to 15 months postoperatively. However, both prior MCID and the PASS thresholds were determined at 2 years postoperatively.

Alternatively, the difference between our derived values and previously derived values for both the HOOS JR and KOOS JR thresholds could reflect the limitation of a single quality-of-life question rather than a global impression of satisfaction [31]. For example, Kunze et al. [21] used a quality-of-life anchor question for the HOOS JR and KOOS JR [29] with a Likert response scale and reported derived PASS scores for the HOOS JR and KOOS JR as 77 and 64, respectively, which differ substantially from our calculated values.

On the other hand, the MN Community Measurement developed a PASS score for the KOOS JR that was derived from the Oxford Knee Score [20] by using an external crosswalk between the Oxford Knee Score and KOOS JR [4]. That study reported a crosswalk-derived PASS score for the KOOS JR that was more than 71. The similarity of this value to our calculated value could provide credence to the importance of the timing of evaluating the PASS threshold but may also highlight the importance of considering pain and function as components of a patient’s overall level of satisfaction, rather than simply quality of life.

Conclusion

We demonstrated the utility of a single anchor question that focuses on outcome, mitigates concerns of patient recall, is simple to respond to unlike prior PASS thresholds (yes or no), avoids a choice of cutoff (Likert scale), and is of equal utility to prior, more elaborate questionnaires. Unlike prior studies, we define more than just a perceptible change of state, but a change that is meaningful in producing an acceptable symptom state using the gold-standard single-item PASS anchor. In doing so, we suggest that routine collection of a single postoperative PASS may be an easier way to measure success than the routine collection of preoperative and postoperative PROMs.