Foot & Ankle International - July 2018 - 769

769

Anderson et al
differences in pre- to posttest probabilities range from
33.3% to 43.2%, suggesting large shifts in probabilities.
Additionally, the posttest probabilities are all higher than
83.3%, suggesting that the preoperative thresholds yield
higher than 80% probabilities that a patient will or will
not achieve MCID improvement on the basis of preoperative data. If these thresholds were applied to the present
data set, 10 procedures (16.4%) that did not produce
MCID improvement might have been avoided and may
have saved health care dollars allocated to surgery (Table
5, Figure 2). In total, the χ2 analysis confirms that proportions of patients based on previous thresholds lead to
meaningful groups beyond what can be expected from
chance (Table 5). This analysis shows that if used, clinically preoperative PROMIS PF data may influence 45.8%
of patient encounters.
There were several important limitations to this study.
First, multiple disparate diagnoses were included. As a validation study, it was important to match the inclusion criteria
of Ho et al,9 but there may be more optimal PROMIS PF or
PI thresholds for each particular diagnosis.9,14
Second, the inclusion criteria allowed for patients with
6-month follow-up. Patients doing poorly following surgery from delayed union, persistent swelling, infection, or
other complications may be overrepresented and skew the
change scores down compared with other studies.14
Prospective follow-up data would eliminate bias from
patients who did not attend follow-up for various reasons
(eg, a smooth recovery). Longer term follow-up also may
demonstrate differences, with some patients continuing to
experience improvement.
Third, a larger sample size would increase the precision
of the threshold values. The moderate correlation (r = 0.63;
Figure 2) between preoperative PROMIS PF scores and
change in preoperative to postoperative PROMIS PF scores
makes it necessary to increase the sample size to more accurately define optimal thresholds.
Fourth, MCID was defined by using the distribution
method, specifically, 0.5 standard deviations. Additional
studies are being performed to correlate this method with an
anchor-based method.
Fifth, this study appropriately classified approximately
40% of patients as benefiting or failing to benefit from surgery. The goal of any test is to have high sensitivity and
specificity to identify the trait (or disease) of interest in a
population. Although this study did validate the prior
threshold values (nearly identical), it did not aid in increasing the value of these PROMIS PF findings. However, correctly identifying even 40% of patients is a great aid to
patients, clinicians, and the health care system.
Finally, applicability of the threshold values to unique
populations remains unknown. For example, elite athletes
likely have a higher threshold for preoperative PF or may
respond favorably to less than 0.5 standard deviations of

MCID given the thin margin of function that can affect performance in athletes. Caution should be used when applying these findings to unique patient populations.

Conclusions
This study replicated the ability of PROMIS PF and PI
scores to discriminate aggregate improvement after foot
and ankle surgery. In addition, this validation study supports the use of preoperative PROMIS PF but not PROMIS
PI for clinical decision making with patients with high or
low PROMIS PF t scores and considering foot and ankle
surgery. Foot and ankle surgeons should consider including
PROMIS PF assessment as part of their preoperative
workup when engaging patients in clinical decisions regarding foot and ankle surgery.
Acknowledgments
We thank Jerry Bounsanga, BS, and Yushan Gu, BS, at the
University of Utah Department of Orthopaedics Quality Outcomes
Research and Assessment for their assistance in data preparation,
cleaning, and processing.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this article. ICMJE forms for all authors are available online.

Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
project was partly funded by the National Institute of Arthritis and
Musculoskeletal and Skin Diseases of the National Institutes of
Health under award number U01AR067138. The content is solely
the responsibility of the authors and does not necessarily represent
the official views of the National Institutes of Health.

References
1. Basch E. Patient-reported outcomes-harnessing patients'
voices to improve clinical care. N Engl J Med. 2017;376(2):
105-108.
2. Beckenkamp PR, Lin CW, Chagpar S, et al. Prognosis of
physical function following ankle fracture: a systematic
review with meta-analysis. J Orthop Sports Phys Ther.
2014;44(11):841-851, B2.
3. Beckmann JT, Hung M, Bounsanga J, et al. Psychometric
evaluation of the PROMIS physical function computerized adaptive test in comparison to the American Shoulder
and Elbow Surgeons score and simple shoulder test in
patients with rotator cuff disease. J Shoulder Elbow Surg.
2015;24(12):1961-1967.
4. Berliner JL, Brodke DJ, Chan V, SooHoo NF, Bozic KJ. John
Charnley Award: preoperative patient-reported outcome measures predict clinically meaningful improvement in function
after THA. Clin Orthop Relat Res. 2016;474(2):321-329.

Table of Contents for the Digital Edition of Foot & Ankle International - July 2018