Patient-Reported Outcomes in Onychomycosis: A Review of Psychometrically Evaluated Instruments in Assessing Treatment Effectiveness

Onychomycosis is the most frequent nail disorder, accounting for 50-60% of nail pathology [1], with toenail disease being about 10 times more common than fingernail disease [2]. Onychomycosis affects over 4% of people in North America and Europe and imposes a substantial economic burden [1,3]. As a disease with mostly benign clinical features, the impact of onychomycosis on patients' quality of life (QOL) is often underestimated. Nail abnormalities can lead to pain, discomfort, limited dexterity and mobility in daily activities [4], and onychomycosis may also be complicated by skin and soft tissue infections [5]. Psychosocial consequences such as embarrassment, decreased self-esteem, anxiety, and depression are often more common in onychomycosis patients than physical sequelae [4,6].

Patient-reported outcomes (PRO) refer to patients' assessment of their health status or impression of a treatment. Such measurements provide quantifiable data regarding disease impact apart from objective clinical measures. PROs can highlight the most bothersome aspects of onychomycosis for patients and identify targets for intervention [7], thus allowing for the customization of care around patients' goals, treatment optimization, and monitoring of disease activity during therapy [8]. PROs are increasingly becoming an outcome parameter in clinical studies as well as in health-care management and reimbursement [8]. General health questionnaires have limited utility as they are unable to reliably identify problems experienced by dermatologic patients [9]. Instead, disease-specific PRO instruments have been created, validated, and used for onychomycosis.

This review will summarize the current tools used to assess QOL and responsiveness to treatment in onychomycosis patients. First, the 5 current psychometrically validated PRO instruments specific for onychomycosis will be reviewed with a brief discussion of nonvalidated or nonspecific metrics. Then, the use of PROs in evaluating onychomycosis treatment will be discussed.

We searched PubMed, Embase, CINAHL, and PsycINFO databases for all published, English language, clinical studies using PRO instruments for onychomycosis through December 31, 2016. Search terms included onychomycosis, patient-reported outcomes, and quality of life. All articles using any PRO instrument to evaluate QOL of onychomycosis patients were included, regardless of study design. While some studies evaluated QOL in onychomycosis patients using nonvalidated or nonspecific tools, we emphasized psychometrically evaluated PRO instruments in our discussion. Articles that studied nail or foot pathology in general and failed to separate onychomycosis from other pathology were excluded. Bibliographies were searched for additional studies that met inclusion and exclusion criteria. Of 168 articles, 25 unique studies with an aggregate of 12,821 patients met the inclusion criteria. Quality of evidence was graded on a scale of 1-5 per the Oxford Center for Evidence-Based Medicine levels of evidence. Because of the poor quality of evidence, we were unable to complete a quantitative review and thus present our findings qualitatively.

Of 168 results, 25 unique studies met the inclusion criteria. Eight studies describe baseline characteristics of onychomycosis patients using PRO instruments [10,11,12,13,14,15,16,17], while 8 other studies use nonvalidated instruments or instruments nonspecific to onychomycosis (Table 1) [7,18,19,20,21,22,23,24,25]. Ten studies used PRO instruments to evaluate onychomycosis treatments (Table 2) [26,27,28,29,30,31,32,33,34,35].

Data collected with PRO instruments are not meaningful without extensive psychometric evaluation; the aspects of this validation are summarized in Table 3. To date, 5 onychomycosis-specific PRO instruments have been validated and are described in Table 4. Overall, all 5 instruments fulfill psychometric validation. Earlier questionnaires contained a generic core of general physical and mental health questions [10,12,15], whereas later instruments consist only of disease-specific items [16,17], which address symptoms, physical functioning, and psychosocial impact particular to onychomycosis. Multiple nonvalidated PRO instruments have been reported, wherein rationale for item selection and psychometric evaluation were absent [18,19,20,21,22,23].

Lubeck et al. [10,18] produced the first instrument measuring QOL in onychomycosis patients, the Onychomycosis QOL Questionnaire, which was later revised and validated in patients with toenail onychomycosis. In the validation study for the Lubeck instrument [10], patients with clinical improvement demonstrated statistically significant change in most disease-specific scales over time, whereas no significant change was noted in clinically stable patients, reflecting responsiveness of disease-specific measures.

Turner and Testa [15] later created the Onychomycosis Disease-Specific Questionnaire (ODSQ), which derives questions from the Lubeck instrument and has been validated for both toenail and fingernail disease. The OnyCOE-t™ instrument is another questionnaire developed and modified from the Lubeck instrument [16]. When comparing OnyCOE-t™ scores after treatment with baseline, the clinically improved group of subjects had statistically significant change in all domains including symptom frequency and bothersomeness, physical activities problems, appearance problems, and stigma. However, the clinically unimproved group also had significant changes for 5 out of 7 domains. Uniquely, clinical meaningfulness was evaluated by estimating the minimally clinical important difference (MCID), i.e. how much a scale needs to change to reflect a corresponding clinical change. The authors found that an 8.5-point change on their 100-point scale corresponded to a 25% improvement in nail clearance. The MCID allows physicians to correlate clinical findings with PRO data. Weaknesses include lack of variability or test-retest evaluation in addition to the possibility of an overly sensitive response when no clinical change has occurred.

Drake et al. [12] introduced the International Onychomycosis Questionnaire, which has been evaluated in multiple languages and validated in a Polish cross-sectional study for both toenail and fingernail disease [13]. The toenail-specific scale was also validated in a Serbian cross-sectional study [14].

The fifth validated PRO instrument is the NailQoL, which consists of the previously validated Skindex-29 (a dermatological QOL measure) and 10 questions from nonvalidated disease-specific questionnaires [17]. Subjects with complete cure of the target toenail and all 10 toenails reported significantly improved PRO scores (p ≤ 0.02).

Finally, Warshaw et al. [30] developed a scale assessing treatment ease and convenience for a cross-sectional study, although a comprehensive onychomycosis instrument was not developed. Warshaw's instrument was valid and responsive with acceptable internal consistency but did not evaluate test-retest reliability.

Ten unique studies used PROs to evaluate treatment in onychomycosis patients, none of which included fingernail disease (Table 2). Patient QOL was examined in 6 studies of oral antifungals, 3 studies of topical antifungals, and 1 study of laser therapy. Only 3 studies used a validated, onychomycosis-specific PRO instrument, the OnyCOE-t™ [31,33,34]. Four studies used instruments derived from previously validated questionnaires [27,28,29,30], 1 study used a validated instrument for podiatry-related QOL [32], and 2 studies used instruments of unclear origins [26,35]. There were 2,224 patients in 4 randomized controlled trials, 1,825 patients in 5 case series, and 142 patients in 1 cross-sectional study.

As systemic antifungals are the standard onychomycosis treatment, most QOL data have been collected from onychomycosis patients treated with oral medications. Potter et al. [31] validated the OnyCOE-t™ in a prospective, randomized, multicenter trial comparing terbinafine alone with the combination of terbinafine and debridement. Both groups experienced QOL improvement. From baseline, debridement significantly improved treatment satisfaction (p = 0.0077) and symptom frequency (p = 0.0395). Debridement had the most impact in the final week (week 12) of terbinafine treatment; afterwards, the terbinafine-only group had equivocal QOL improvement, suggesting that debridement only had a short-term effect when terbinafine had yet to adequately penetrate the nail plate. While debridement did not cure onychomycosis, it did improve comfort and appearance, thus enhancing QOL.

Multiple studies have used aspects of Lubeck's questionnaire. Stier et al. [27,28] compared oral antifungals to local therapy in 2 case series. In their cost effectiveness study, QOL was assessed with the Toenail Symptom Index, an instrument using 2 reliable, valid, and responsive scales from Lubeck's questionnaire [10]. Oral medications such as terbinafine, itraconazole, and fluconazole were significantly superior to local treatment such as topical treatments or debridement (p = 0.003); patients treated with oral medications also had greater direct medical expenses with an additional cost of USD 304-491 for each case improved over 9 months, although extrapolation showed eventual cost equivalence [28]. The patient satisfaction study by Stier et al. [27] demonstrated that treatment satisfaction was correlated with nail condition, although the content of the questionnaire was unclear. Warshaw et al. [30] used the treatment satisfaction scale from Lubeck's questionnaire in their study of terbinafine versus pulsed-dose itraconazole in a multicenter, cross-sectional study with 142 patients. Subjects reported greater ease and satisfaction with terbinafine (p = 0.008 and p = 0.003, respectively).

Using an instrument modified from Drake's international questionnaire, Firooz et al. [29] evaluated the impact of itraconazole pulse therapy on the QOL of 20 toenail onychomycosis patients and reported that mean QOL scores significantly improved with treatment. However, with this incomplete instrument modified from Drake's validated tool, no rationale was provided regarding the items chosen for the questionnaire; thus, the reliability and validity of this instrument are in question.

Finally, the earliest use of QOL metrics in onychomycosis treatment was Pollak and Billstein's evaluation of terbinafine efficacy using a large, multicenter, open-label study of 1,534 toenail onychomycosis patients [26]. Patients reported overall satisfaction with the treatment results and improved symptom severity. The details of the PRO instrument were not included.

PRO analyses of topical treatments are limited, likely in part because numerous studies have demonstrated the clinical superiority of oral medications [36,37]. Additionally, until recently, ciclopirox was the only topical drug approved for onychomycosis treatment in the US [36]. With the approval of efinaconazole and tavaborole in 2014, more QOL studies are warranted given the high cost and nonsuperiority of these medications to oral antifungals [37].

Tosti and Elewski [33] used the OnyCOE-t™ to evaluate the impact of topical efinaconazole versus vehicle on PROs in a 48-week, randomized, multicenter, double-blind study with 1,655 toenail onychomycosis patients. The efinaconazole group had significantly greater mean QOL scores across all domains (p ≤ 0.002) and significantly higher treatment satisfaction scores (p < 0.001). The improvement was especially marked for the 47% of subjects who were considered clinically improved. Notably, even subjects in the efinaconazole group who failed to demonstrate clinical improvement had better scores for symptom frequency, appearance, and overall problem as compared to baseline, suggesting that even the smallest change in nail appearance can improve QOL. The change in percent nail affected correlated with change in mean domain scores.

Patients with HIV are at higher risk for onychomycosis and yet are an understudied group. The OnyCOE-t™ was also used in a 48-week prospective study of Vicks VapoRub for treating toenail onychomycosis in 20 HIV-positive subjects [34]. This treatment is a non-FDA-approved modality containing eucalyptus oil, a natural home remedy for onychomycosis with little-to-no evidence of clinical utility. After 24 weeks of treatment, fewer patients reported embarrassment and pain/discomfort, and 94% of patients reported treatment satisfaction. Limitations of this study included small sample size, poor follow-up, and poor adherence. Additionally, PRO data were not analyzed for statistical significance.

Malay et al. [32] used the Bristol Foot Score, a validated, podiatry-specific PRO instrument to evaluate QOL in 55 toenail onychomycosis patients who underwent either debridement alone or debridement with adjunct ciclopirox treatment in a multicenter randomized controlled trial. Debridement alone did not result in any mycological cure, while the addition of ciclopirox resulted in a cure rate of 76.7%. Both groups had improved QOL scores (p < 0.001), but the addition of a topical lacquer improved QOL more than debridement alone (p = 0.002).

Given that achieving cure for onychomycosis requires prolonged treatment, investigations have turned towards alternative modalities to improve clinical response. Laser, light, and combination therapy provide modalities that free patients from the burden on daily medication adherence and drug-drug interactions [38]. Although there have been hundreds of recent studies examining lasers to treat onychomycosis, PRO analyses are limited. Ortiz et al. [35] investigated the efficacy of the 1,320-nm neodymium:yttrium aluminum garnet laser in a 24-week, randomized, placebo-controlled study of 10 subjects with bilateral great toe onychomycosis. Overall, no significant improvement was observed in treated versus untreated nails. A questionnaire was used to evaluate PRO, although its contents were not identified. Patients reported less self-consciousness and embarrassment about treated toenails despite no clinical difference. Limitations of this study included small sample size, concomitant treatment with a topical antifungal, and possible therapeutic effect of the sham treatment.

Although onychomycosis has traditionally been viewed as a cosmetic problem, PRO analyses have demonstrated the substantial morbidity of this disease [5,7,10,18,19,20,21,22]. This review identified 5 psychometrically evaluated instruments that can evaluate patient QOL: Lubeck's [10] and Drake's [12] instruments as well as the ODSQ [15], OnyCOE-t™ [16], and NailQoL [17]. Only the OnyCOE-t™ has been used in its entirety [31,33,34]. In practice, patient responses in these QOL instruments have paralleled clinical change, which lends to their credibility. Additionally, PRO instruments allow practitioners to codify and translate subjective patient experiences into quantitative data.

While onychomycosis is not life-threatening, in studies of the effect of onychomycosis on QOL, 92% of patients reported negative psychosocial or physical impairments [22], and 70% of patients considered onychomycosis to be at least a moderate problem [23]. Compared to controls, onychomycosis patients had significantly poorer QOL in general health perception, bodily pain, mental health, social functioning, health concern [18,19], physical appearance [13,14,17,18,19,25], and physical functioning, such as limitations with standing and working with one's fingers [12,14,18,19,20,21,23]. Onychomycosis patients were more self-conscious [17,18,19,22,25] and embarrassed [12,13,17,18,19,21,23], feared spread of transmission to other nails or other people [12,13,14,22,25], and experienced pain [12,13,17,18,19,20,21,22,23]. In using the validated, dermatology-specific instrument Skindex-29, onychomycosis QOL impairment was worse than other nail disorders [7], comparable to nonmelanoma skin cancer and benign growths [17], and better than eczema and body psoriasis [17]. This morbidity is not always readily evident. Pain associated with onychomycosis is often underestimated by physicians [12], and although clinicians may not appreciate any clinical change after treatment, patients report improved QOL [33].

To assess patient QOL, 5 PRO instruments have been validated for onychomycosis. Drake's questionnaire is the only instrument validated in multiple languages but also the only one without data on responsiveness, thus limiting its use in monitoring disease activity. Drake's questionnaire and the ODSQ are the only instruments validated for fingernail disease, although the ODSQ does not explicitly list fingernail-specific items. The validation studies for the ODSQ, OnyCOE-t™, and NailQoL established diagnosis by mycological criteria; the validity of Lubeck's and Drake's instruments may be questionable as patients could have had nonfungal nail disease. Though earlier validated questionnaires included general health measures, the OnyCOE-t™ and NailQoL assess only onychomycosis-specific items. A generic core allows comparisons across interventions and diseases as it uses previously validated measures to evaluate parameters of general health distress, psychosocial well-being, and physical functioning [7,17,24]. However, this core also has low sensitivity to group differences and clinical changes [10,12,15,17,18,19]. It also lengthens questionnaires, making clinical implementation less feasible [9], which may explain why Lubeck's and Drake's questionnaires have not been utilized in their entirety though selections from each have been appropriated [27,28,29,30]. To date, only the OnyCOE-t™ has been used in full [31,33,34]. The OnyCOE-t™'s popularity may be secondary to its shorter length and defined MCID, which allows clinicians to correlate PRO data with clinical change.

The majority of PRO data for treatment assessment parallel clinical data: oral antifungals have higher cure rates and also larger QOL changes when compared to topicals[26,27,28]; terbinafine is superior for clinical improvement and treatment satisfaction when compared to pulsed-dose itraconazole [30]. However, with new topical antifungal medications and increasing interest in laser and light therapy, PROs are important to understand patient priorities. In a 1998 study, although 79% of patients asked about treatment for onychomycosis, 26% reported that their providers discouraged medication use [19]. Even if insurance would not cover the cost, 97% of patients were willing to pay for medications with an 80% cure rate; 57% were still willing to pay even if the cure rate dropped to 35% [23]. This is particularly relevant given the recent introduction of expensive topical therapies for onychomycosis. Efinaconazole has been reported to result in improvement in only 42.6% of patients, and yet has been reported to improve QOL domains even in patients without appreciable clinical improvement, which further emphasizes that patient QOL cannot be deduced from clinical exam alone [33]. PRO instruments are essential for optimizing treatment to maximize compliance, treatment satisfaction, and QOL.

It should be noted that some studies are limited by their study design. For example, Ortiz et al. [35] showed that patients undergoing laser treatment had QOL improvement without significant clinical change, suggesting that laser therapy can be worthwhile for onychomycosis patients; however, the validity of the conclusion is limited by a small sample size and concurrent antifungal use. Additionally, the effect of disease recurrence on patient QOL and willingness to pay was not evaluated. Given such limitations, study results should be extrapolated with caution, particularly when patients may face exorbitant expenses for a therapy that may not be clinically effective or offer long-term, sustained results.

It is worth noting that the reviewed literature does not address the differences between subjects presenting with a specific complaint of onychomycosis versus those in which the disease is an incidental, and sometimes unconcerning, finding. Research subjects and patients seeing a specialist for onychomycosis are self-selecting and may disproportionately reflect QOL impairment. Given that, it is difficult to quantify the extent of QOL impairment, or perhaps even to firmly establish that onychomycosis is the cause of QOL impairment without comparison to a non-self-selecting control group with incidental onychomycosis. It is possible that QOL impairment may not be a reflection of onychomycosis per se, but more a marker of a threshold of QOL impairment necessary for patients to present to a dermatologist.

Future research should include psychometric evaluation of these instruments in special populations, fingernail disease, and between patients self-selecting for onychomycosis and controls with onychomycosis as incidental findings. For example, HIV and diabetes mellitus are risk factors for onychomycosis and may modify disease type and course, but most PRO trials exclude patients with these comorbidities. Additionally, neither the OnyCOE-t™ nor the NailQoL instrument is validated for fingernail disease despite the fact that fingernail involvement has been shown to increase patient disease burden [17,23].

As medical practice increasingly emphasizes patient-centered care and shared decision-making, QOL assessments are becoming increasingly important. Through PRO instruments, it is apparent that onychomycosis is not only a cosmetic hindrance but also a burdensome medical disorder that patients want and are willing to treat, at times even for limited or no clinically apparent benefit. Five PRO questionnaires have been validated for onychomycosis, and they have been infrequently used in clinical studies examining treatment efficacy. Future research should utilize these instruments to generate further insight into patients' assessment of their health status and of new therapies.

The authors have no conflicts of interest to declare. There were no funding sources for this project.