Validation of the Single Assessment Numeric Evaluation (SANE) Score as an Outcome Measure by Comparison to the Revised Foot Function Index (rFFI)

Abstract

Background:

Patient-reported outcome measures serve as an invaluable tool in both the clinical and research setting to monitor a patient’s condition and efficacy of treatments over time. We aim to validate the Single Assessment Numeric Evaluation (SANE) score for disorders of the lower extremity using the revised–Foot Function Index (rFFI) as a reference. The SANE score is a 1-question survey that may improve efficiency of outcome data collection in the clinical setting.

Methods:

Patient age, sex, visit diagnosis by ICD-10 code, SANE score, and rFFI score were collected retrospectively from 218 initial patient encounters between January 2015 through July 2017. Patients were included if they were 18 years or older and were excluded if they had incomplete SANE or rFFI data. Results of the two scores were compared using the Pearson or Spearman correlation coefficients, with correlation defined as excellent (>0.7), excellent-good (0.61-0.7), good (0.4-0.6), or poor (0.2-0.39). Diagnoses were categorized into 9 subgroups that were analyzed, including forefoot, plantar fasciitis, arthritis, deformity, fracture, tendinitis, osteochondral defect (OCD) of the talus, acute soft tissue trauma, and “other.”

Results:

The SANE score had good correlation with the overall rFFI score (r = 0.51, P < .01). When comparing the SANE score to the rFFI subscores, there was good correlation with pain (r = 0.42, P < .01), good correlation with stiffness (r = 0.44, P < .01), poor correlation with activity (r = 0.36, P <.01), good correlation with difficulty (r = 0.52, P < .01), and poor correlation with social issues (r = 0.39, P < .01). Subanalysis showed an excellent-good correlation between SANE and rFFI score for forefoot pathology (r = 0.67, P < .01) and plantar fasciitis (r = 0.63, P < .02), good correlation for arthritis (r = 0.49, P < .04), deformity (r = 0.60, P < .01), fracture (r = 0.50, P < .01), and tendinitis (r = 0.47, P < .02), and no significant correlation for OCD of the talus (r = 0.56, P < .15) and acute soft tissue trauma (r = 0.19, P < .32).

Conclusion:

The SANE score demonstrates excellent-good correlation with the rFFI for specific pathology including the forefoot and plantar fasciitis and has limited correlation with other pathology.

Introduction

Patient-reported outcome measures (PROMs) serve as an invaluable tool in both the clinical and research setting to monitor a patient’s condition over time. ^{9,12,17,19,22,24} They provide objective, typically numerical, data that can be used to measure disease progression and response to treatment. As the health care system shifts from a “fee-for-service” to a “value-based” model, physicians will increasingly rely on outcome measures to demonstrate the value they provide for their patients. ²⁵ An ideal outcome measure would be validated for a given condition, utilize limited resources, and require little time investment to obtain.

There are numerous PROMs presently being utilized for foot and ankle pathology. In 2018, the American Orthopaedic Foot and Ankle Society (AOFAS) put out a position statement regarding PROMs reporting that the organization is unable to recommend a single PROM for universal use and that selection depends on both the question being asked and the population in question. Of the options available, AOFAS recommends considering the Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function Computerized Adaptive Test (PF CAT) or the Lower Extremity Computerized Adaptive Test (LE CAT) in combination with an instrument for pain evaluation. ¹⁶ The main limitation of PROMIS is its difficulty to integrate with existing electronic hearth record systems and the infrastructure required to do so. ¹⁴ Other PROMs continue to be used as a result of this limitation and are advantageous for their relative ease of utilization.

Single-item scales are frequently chosen for their ease of administration and interpretation at the cost of more expansive data collection. For example, the visual analog scale (VAS) is a single-item instrument used primarily for measuring quality of life and pain. It was shown to be as valid, reliable, and responsive in evaluating quality of life in patients with cancer postoperatively when compared to a multi-item scale, the Medical Outcomes Study Short Form–20. ^4,5 The VAS demonstrates the potential for single-item scales to adequately replace more complicated multi-item scales in certain circumstances, and they have most often been chosen for patients evaluating their own health status or chronic conditions. ⁵

The Single Assessment Numeric Evaluation (SANE) score is another single-item PROM most frequently used for disorders of the shoulder. The measure consists of a patient’s answer to the question “How would you rate your [extremity] today as a percentage of normal (0 to 100% scale, with 100% being normal)?” ²⁶ The SANE consists of a single question, which allows physicians to record outcomes data on patients in a timely fashion and easily integrate it into clinical workflow. The score has been validated and utilized for disorders of the shoulder. ^8,20 The SANE score has also been shown to have moderate to strong positive correlations with the modified Cincinnati Knee Rating System and International Knee Documentation Committee (IKDC) subjective knee surveys for patients after anterior cruciate ligament (ACL) reconstruction or knee arthroscopy procedures. ²¹ However, this outcome measure is less commonly used to evaluate pathology of the lower extremity.

The revised–Foot Function Index (rFFI) is a validated patient-reported outcome tool in foot and ankle pathologies. ^1,3,6,7,23 The rFFI contains 34 questions with subscales for difficulty, stiffness, pain, social issue, and activity and has been extensively used in clinical and research settings to evaluate patients with foot- and ankle-related pathology.

The purpose of this study was to validate the SANE score for disorders of the foot and ankle using the rFFI as a reference. Given its ease of use, the SANE score represents a practical outcome measure that could gain universal acceptance and may be useful in places where the PROMIS CAT cannot be integrated. We hypothesize that the SANE score will demonstrate good to excellent correlation with the rFFI with respect to disorders of the foot and ankle.

Methods

Patient data were retrospectively collected from patients between January 2015 and July 2017. Data were collected from all patients 18 years or older presenting for outpatient evaluation by 2 fellowship-trained foot and ankle surgeons at a single institution (V.S. and L.G.). The rFFI and SANE scores were collected by the department as a quality measure prior to the start of the study. The rFFI has been used historically at the institution for its ease of scoring. Scores were collected from patients prior to receiving treatment. A total of 352 patients had both rFFI and SANE scores collected. Patients were excluded if they were unable to complete the survey questionnaires for any reason or who had incompletely or incorrectly filled out the questionnaires, resulting in a total of 218 patients. Patient age, sex, established International Classification of Diseases, Tenth Revision (ICD-10), codes from the visit, SANE score, and rFFI score were obtained from a chart review. The diagnoses were further subcategorized using obtained ICD-10 codes into the following groups: deformity, fracture, acute soft tissue trauma, plantar fasciitis, arthritis, forefoot, tendinitis, OCD of the talus, and other. The categorization of pathology is further detailed in Table 1.

Table 1.

Details of Foot and Ankle Pathology Classification.

Pathology	Corresponding diagnoses and details	Percentage of patients with diagnosis (of 218)
Forefoot	Hallux valgus, hallux rigidus, metatarsalgia, Morton’s neuroma, hammertoe	24
Other	Ankle effusion, non-specific pain, bone spur, dry gangrene, hematoma, chronic ankle instability, bursitis, osteomyelitis, synovitis, os trigonum syndrome	11
Plantar fasciitis	Includes heel pain, fibroma	6
Deformity	Pes planus, cavus, coalition	8
Fracture	Fractures including tarsal, metatarsal and ankle	14
Arthritis	Ankle, hindfoot or midfoot	8
Tendinitis	Achilles, posterior tibial, or peroneal tendons	11
OCD of talus	OCD of talus	4
Acute soft tissue trauma	Achilles rupture, sprain	14

Statistical Analysis

Descriptive statistics including mean and standard deviation for continuous variables, and frequency and proportion for categorical variables, were calculated to characterize the study group. The construct validity of the SANE score for the foot and ankle extremity was assessed by comparison to the rFFI instrument. The relationship between the 2 instruments was evaluated with the Pearson correlation coefficients. Consistent with previous work, rho values of greater than 0.7 were considered excellent, 0.61 to 0.7 excellent to good, 0.4 to 0.6 good, and 0.2 to 0.3 poor. ¹⁸ Floor and ceiling effects are considered present when an instrument is not able to measure individuals whose abilities lie outside the measurement range (ie, above or below the lowest or highest possible score on an instrument). Ceiling and floor effects were measured by determining the percentage of subjects with the highest and lowest possible score, respectively. Consistent with previous studies, if more than 15% of individuals scored the lowest or highest possible total score on a questionnaire, floor or ceiling effects were considered present. The alpha level for all statistical analyses was set at less than 0.05.

An a priori power analysis was calculated. The sample size calculation was based off a rho of 0.4 as this value represents the lower end of “good” for the strength of the relationship between the rFFI and the SANE score. A sample size of 47 patients will provide 80% power to observe a rho value of 0.4 at an alpha level of 0.05.

Results

A total of 218 patients were included in the analysis. Mean age was 47 years (median age, 50; range, 18-92), and 68% were female. The percentage of patients presenting with each subcategory of diagnosis, from greatest to least, were as follows: 24% forefoot (52 patients), 14% fracture (31 patients), 14% acute soft tissue trauma (31 patients), 11% tendinitis (25 patients), 11% other (23 patients), 8% arthritis (18 patients), 8% deformity (17 patients), 6% plantar fasciitis (13 patients), and 4% OCD of the talus (8 patients).

Overall, the SANE instrument showed good correlation with the rFFI (r = 0.51, P < .01). The data and breakdown of subcategories comparing the SANE and rFFI scores are shown in Tables 1 and 2. Both the SANE and rFFI did not have substantial floor or ceiling effects. For the SANE, 1 patient reported a score of 0 (worst score, 0.5% floor effect) and 1 patient reported a score of 100 (best score, 0.5% ceiling effect). For the rFFI, 1 patient scored a 204 (worst score, 0.5% floor effect) and 1 patient scored 34 (best score, 0.5% ceiling effect).

Table 2.

Correlation Coefficients and Classification of the Correlation Between the rFFI Survey and SANE Score in the Assessment of Foot and Ankle Pathology.

	Pearson Correlation Coefficients	p value	Classification
Overall	-0.51	<0.001	Good
Difficulty	-0.52	<0.001	Good
Stiffness	-0.44	<0.001	Good
Pain	-0.42	<0.001	Good
Social Issue	-0.39	<0.001	Poor
Activity	-0.36	<0.001	Poor
Forefoot	-0.67	<0.001	Excellent-good
Other	-0.65	0.001	Excellent-good
Plantar fasciitis	-0.63	0.016	Excellent-good
Deformity	-0.60	0.010	Good
Fracture	-0.50	0.004	Good
Arthritis	-0.49	0.038	Good
Tendinitis	-0.47	0.017	Good
OCD of talus	-0.56	0.145	No significance
Soft tissue trauma	-0.19	0.319	No significance

Abbreviations: rFFI, Revised Foot Function Index; SANE, Single Assessment Numeric Evaluation.

Discussion

The SANE tool has been previously validated and used to gather outcomes data for a variety of clinical pathologies, including those of the shoulder, elbow, and knee. ^21,26,27 The major benefit of the SANE tool is its minimal demand of time and resources in the clinical setting. This study has found that the rFFI and SANE scores demonstrated good correlation overall (r = 0.51, P < .01) and suggests the SANE tool has the potential for expanded use for foot and ankle pathologies.

The subcategories of the rFFI varied in their correlation with the SANE score from good to poor, with difficulty, stiffness, and pain showing good correlation and social issue and activity showing poor correlation. The poor correlation with the social issue subcategory is potentially because the SANE score specifically asks about the extremity as a function of normal and not overall health. Social issues are therefore unlikely targeted in this question, and if they do influence the response, it is possible that the correlation is being confounded by other conditions. It is well documented that conditions such as depression and anxiety can be confounders when evaluating patient-reported outcomes related to health interventions. ¹³ On the other hand, the subcategories difficulty, stiffness, pain, and activity all are directly related to the function of the lower extremity with respect to normal and were shown to correlate well with the exception of activity. Activity did not show a good correlation to the rFFI, and we hypothesize that this result may be a product of the range of activities that different patients take part in.

In addition, when the correlation between the rFFI and SANE scores was examined based on presenting pathology, the correlation ranged from excellent-good to no significance. The pathologies without statistical significance were OCD of the talus, which had the smallest n of 4% (8 patients), and acute soft tissue trauma. Clinicians should be cognizant of what the patient is presenting with before selecting which survey to give as the SANE score appears more helpful for certain pathologies, including the acuity of the pathology. The data suggests additional research with a larger sample size is needed for the pathologic subcategories.

There are numerous PROMs presently being used for foot and ankle pathology, and in particular there has been a lot of focus in the literature regarding the PROMIS CAT scores. These CATs present items from an item bank and then adjust subsequent items presented based on previous answers. The algorithm limits the number of items required to evaluate a specific domain, such as physical function referenced above, while retaining the ability to adequately assess that domain.

The PF CAT was initially validated for patients with surgery for 6 common foot and ankle disorders and has been further validated to include ankle fractures. ^11,14 The instrument was compared directly to the Foot and Ankle Ability Measure–Activity of Daily Living subscale (FAAM_ADL) and the Foot Function Index 5-point verbal rating scale (FFI-5 pt) for reliability, responsiveness, and efficiency in patients undergoing elective and common foot and ankle procedures. The study concluded that the PF CAT was the best of the instruments in those categories and took only 44 seconds to administer compared to 179 seconds and 194 seconds for the FAAM_ADL and FFI-5 pt, respectively. ¹⁵ Although the CATs can be excellent measures when used, there is a significant amount of time, infrastructure, and expense required for their use. The SANE score represents a simple and reliable alternative clinically and for research purposes.

In the process of collecting data for this study, approximately 38% of all rFFI surveys were excluded from the study because they were incomplete or filled incorrectly. This may be related to the length of the survey. A study by Agel in 2012 showed similar findings when analyzing the effectiveness of surveys in measuring outcome data for spine surgery. They found that only 60% of surveys administered to evaluate pre- and postoperative spine surgery patients were completed to a satisfactory level, and only 300 of 1863 patients had 3 or more completed surveys to evaluate their course of treatment. ² These findings are also consistent with studies showing that surveys with greater length are undertaken with less frequency, and questions further along in a given survey are completed with faster, shorter, and more uniform answers. ¹⁰ The SANE tool aims to limit these barriers to data collection and help to more efficiently track patient outcomes.

There were limitations to the present study, including the retrospective nature of the study. Additionally the data collection was nonconsecutive because many patients had incomplete data. There was also heterogeneity of the foot and ankle pathology and low numbers in some of the subgroups when doing the subgroup analysis. It is difficult to determine whether having more complete or continuous data would affect our results. Having an increased number of patients in each subgroup would potentially show a greater strength of correlation in the subgroups that had poor or good correlation. Additionally, a high proportion (68%) of the patients were female, and therefore this may limit generalizability of the data to the male population.

Conclusion

The SANE score demonstrates good correlation with the rFFI overall (r = 0.51, P < .01) and represents a simple and reproducible PROM that can be used clinically and for research purposes. Furthermore, the SANE score has better ease of use in comparison to the rFFI as well as significantly less infrastructure, expense, and IT work required in comparison to the PROMIS scores. The presenting pathology and acuity of pathology should be considered when choosing between the rFFI and SANE tool for use because the SANE score may be of limited use with particular pathologies, including acute soft tissue trauma and OCD of the talus. Forefoot pathology and plantar fasciitis had the best correlation between SANE score and rFFI.