The Burden of Small Fiber Neuropathy: Changes in Pain, Costs, and Quality of Life Before and After Diagnosis

ABSTRACT

Background and Aims

Small fiber neuropathy (SFN) is a peripheral neuropathy causing neuropathic pain, reduced quality of life (QoL), and high societal costs. Previous studies assessed these societal costs before a definitive diagnosis was established, leaving it unclear whether diagnostic confirmation affects costs, pain intensity, or QoL. This study assessed changes in these outcomes before and after diagnosis.

Methods

Patients referred to the tertiary SFN expertise center in the Netherlands completed questionnaires before diagnosis (at placement on the waiting list, approximately 7 months before consultation) and at 3 and/or 6 months after confirmation. Healthcare utilization, patient and family costs, and productivity losses were assessed using the iMTA Medical Consumption and Productivity Cost Questionnaires. QoL was measured with the EQ‐5D‐5L and pain intensity with a numeric rating scale. Missing data were handled using multiple imputation, and changes over time were analyzed using within‐subject comparisons and generalized linear mixed models.

Results

Eighty‐four patients completed both pre‐ and postdiagnosis questionnaires. After follow‐up, total healthcare costs, patient and family costs, productivity losses, pain intensity, and QoL remained unchanged. However, diagnostic confirmation of SFN was associated with fewer medical specialist visits, lower related costs, and improved health perception.

Interpretation

These findings suggest that diagnostic clarification alone is insufficient to reduce the overall burden of SFN, highlighting the need for more effective symptom management and supportive care strategies.

1. Introduction

Small fiber neuropathy (SFN) is a disorder affecting thinly myelinated Aδ and unmyelinated C fibers and typically presents with neuropathic pain and autonomic symptoms [1]. SFN is not uncommon, with a prevalence of at least 53 cases per 100 000 inhabitants [2]. In clinical practice, diagnosis is commonly based on the Besta criteria, which require a combination of two clinical signs, a reduced intraepidermal nerve fiber density (IENFD) in skin biopsy and/or an temperature threshold test (TTT) [3]. Additionally, large fiber involvement should be excluded with nerve conduction studies (NCS) [3]. SFN is associated with various potentially treatable conditions; however, a specific underlying condition can be identified in only approximately 50% of cases [4, 5]. When no underlying condition is present, treatment of SFN focuses on symptomatic relief of neuropathic pain, which often yields disappointing results [6]. SFN has a substantial impact on patients' quality of life (QoL), while anxiety and depression further interfere with daily functioning [7, 8].

The number of patients diagnosed with SFN is expected to increase due to increasing awareness and improved recognition, as reflected by a growing incidence over time [9]. This increase may lead to higher healthcare utilization and associated costs. A previous cost of illness study, conducted by our research group in the Netherlands, a Western European country, estimated the total annual societal costs incurred by patients with SFN at €147.7 million, with €3610 as the average per‐patient cost [10]. This study accounted for both direct healthcare expenses and indirect costs arising from productivity losses, as SFN symptoms often impair patients' ability to work through absenteeism or reduced performance [10]. Notably, severe pain was associated with increased costs and a decrease in QoL [10].

This previous cost‐of‐illness study focused on the period before diagnosis. During this time, patients with SFN frequently undergo a prolonged diagnostic trajectory with repeated consultations and time‐consuming testing, which contributes to high healthcare utilization [1, 10]. A possible explanation for this frequent healthcare use may be the drive to seek an underlying explanation for their persistent pain and reduced QoL. Conversely, establishing a definitive diagnosis may facilitate acceptance of symptoms, better specialized treatment, and potentially reduce healthcare visits and costs, as has been shown in other chronic pain diseases [11, 12]. Overall, SFN imposes a substantial burden on patients and society, affecting pain, QoL, healthcare utilization, and work productivity.

This study aims to assess the burden of SFN by examining potential changes in pain, QoL, healthcare use and productivity through comparison before and after receiving a definitive diagnosis, to better understand the impact of the diagnostic confirmation on patients' burden.

2. Methods

2.1. Patient Population

This study was conducted at the SFN expertise center of the Maastricht University Medical Center+ in Maastricht, the Netherlands. The SFN Center is a tertiary referral center for patients with symptoms suspected of SFN. When referred, patients undergo a standardized diagnostic evaluation during a 1‐day admission. During this day admission, the workup includes NCS, TTT, and a skin biopsy to determine IENFD. The diagnosis of SFN is based on the Besta criteria, requiring at least two among the following: (i) clinical signs of small fiber impairment (pinprick and thermal sensory loss, and/or allodynia, and/or hyperalgesia); (ii) abnormal warm and/or cold thresholds at the foot, as assessed by TTT; and (iii) reduced IENFD at the distal leg [3]. Additionally, extensive blood analyses including peripheral blood cell counts, liver function, glucose, creatinine, creatine kinase, angiotensin converting enzyme, thyroid stimulating hormone, vitamins, anti‐nuclear antibodies, anti‐neutrophil cytoplasmic antibodies, myeloma protein, soluble IL‐2 receptor, serologic testing for Borrelia, hepatitis B, and HIV and genetic testing DNA analyses are done to identify a possible underlying condition. SFN cases are classified as idiopathic when no underlying cause can be identified after this standardized diagnostic work‐up.

All patients referred to the SFN center between April 2017 and February 2020 were invited to participate in this study, since diagnosis was not yet confirmed. Given the aim of this study to compare outcomes before and after diagnosis, only patients with a confirmed SFN diagnosis who completed the waitlist questionnaire (approximately 7 months prior to consultation and diagnosis) and had at least one follow‐up questionnaire (3 or 6 months after diagnosis) were included.

2.2. Study Design and Data Collection

Patients were invited by email to fill in the online questionnaires, with a paper version available on request. Patients were invited at four timepoints: at placement on the waiting list, during their visit to the SFN center, and at 3 and 6 months after their visit.

At each timepoint, patients were asked about pain intensity and completed the 5‐level EuroQol 5D (EQ‐5D‐5L), the iMTA Medical Consumption Questionnaire (iMCQ), and the iMTA Productivity Cost Questionnaire (iPCQ) [13, 14]. Pain intensity was measured using a 11‐point numerical rating scale (NRS) with 0 (no pain) and 10 (worst pain imaginable) [15]. The EQ‐5D‐5L measures QoL and consists of five questions representing health‐related QoL dimensions, and an additional Visual Analog Scale (VAS), which ranges from 0 (worst) to 100 (best) [16]. The health‐related QoL dimensions are mobility, self‐care, usual activity, pain/discomfort, and anxiety/depression. The responses to these questions result in a state‐of‐health combination which was converted into EQ‐5D utility scores according to the Dutch tariff [17]. Possible scores range from −0.466 (worst) to 1.00 (best). The iMCQ was used to assess healthcare resource use as well as patient and family costs (e.g., travel expenses, informal care, out‐of‐pocket costs) [14]. The iMCQ includes questions on visits to health care providers, hospitalizations, medication use, and out‐of‐pocket costs. Patients were asked to only report resource use and costs related to neuropathic and autonomic complaints of SFN within a 3‐month recall period. In accordance with the Dutch guideline for cost research, the iPCQ was used to measure productivity loss, including absenteeism (paid and unpaid work), reduced work schedule, and presenteeism [13]. Presenteeism was estimated based on patients' ratings of the impact of pain on work productivity over the past week using a NRS, which served as a reference to estimate productivity loss over the entire 3‐month period. As part of the standard of care, additional questionnaires were administered only at baseline (the day of admission). To assess anxiety and depression, the Hospital Anxiety and Depression Scale (HADS) questionnaire was used [18]. The questionnaire consists of an anxiety and depression subscale with seven questions recorded on a 4‐point Likert scale each. Scores can range from 0 to 21 and higher scores indicate more symptoms of anxiety and depression [18].

Sociodemographic data (age, sex, and education), clinical characteristics (onset and duration of symptoms, medical history), and diagnostic tests (IENFD, TTT, NCS, and blood test results) were obtained from the electronic patient file. Medication use was assessed using the iMCQ. However, detailed information regarding treatment indication, initiation, discontinuation, and changes in pharmacological or nonpharmacological treatment regimens over time was not systematically recorded. Education was classified according to the educational levels used by Statistics Netherlands (CBS) [19]. Patients were not involved in the design, conduct, reporting, or dissemination plans of this study.

2.3. Cost Estimation

Healthcare and patient and family costs were estimated per time point by multiplying the frequency of use by the unit prices obtained from the Dutch guideline for costs research [20, 21]. In the absence of reference prices (e.g., visits to specific specialists or nonpharmacological treatments), tariffs from the Dutch healthcare Authority (Nederlandse Zorgautoriteit) were applied [22]. Medication costs were valued based on purchase prices and standard dispensing fees as published at www.medicijnkosten.nl from the Healthcare Institute of the Netherlands (Zorginstituut Nederland) [23]. If multiple options were available for the same active substance, the lowest‐priced alternative was used. For over‐the‐counter (OTC) medications, the mean price of the three lowest‐priced locally available options was used. Unit costs were adjusted to the reference year 2024 by means of index numbers obtained from Statistics Netherlands (Centraal bureau voor Statistiek).

Productivity losses of paid employment were estimated by multiplying missed work hours due to SFN‐related illness by the average hourly cost of productivity loss. These hourly costs are reported in the Dutch guideline for cost research and are based on average labor costs per employee [21]. Although the guideline recommends using the friction cost approach to estimate the productivity costs of paid employment, this approach was not applied because the study aimed to compare productivity losses before and after diagnosis. The friction cost approach limits productivity costs to the period required to replace a worker (the friction period), which could artificially limit follow‐up costs, potentially underestimating productivity losses at follow‐up relative to baseline [21]. Not including the friction period enables a straightforward comparison between baseline and follow‐up. Productivity losses from unpaid employment were valued using replacement costs for household care [20].

Total societal costs were calculated as the sum of healthcare, patient and family, and productivity costs.

2.4. Missing Data

Missing values were handled using multiple imputation. A total of 10 imputed datasets were generated using predictive mean matching. All statistical analyses were performed separately on each of the 10 imputed datasets and results were subsequently combined using Rubin's rules where applicable [24]. For nonparametric analyses (e.g., Wilcoxon signed‐rank tests), results were pooled using the median of the p‐values across the 10 imputed datasets.

2.5. Statistical Analysis

In these analyses, primary and secondary outcomes on the waitlist questionnaire were interpreted as outcomes prior to diagnosis, while the 3‐ and 6‐month follow‐ups represent outcomes after diagnosis. Descriptive statistics were used to summarize the demographic and clinical characteristics of the total population. Continuous variables were reported as means with standard deviations (SDs), while categorical data were presented as frequencies and percentages. Mean cost estimates per patient before diagnosis and follow‐up at 3 and 6 months were calculated. To account for the nonnormal distribution of cost data, nonparametric bootstrapping with 1000 replications was performed to determine the 95% confidence intervals (CIs) around mean cost estimates.

Pairwise comparisons to assess within‐subject differences in primary (cost categories) and secondary outcomes (pain and QoL) over time were performed between the before‐diagnosis timepoint and 3‐ and 6‐month follow‐up using parametric (e.g., paired sample t‐test) or nonparametric tests (e.g., Wilcoxon signed‐rank test), depending on the distribution of the data.

Uni‐ and multivariable‐adjusted analyses of both the primary and secondary outcomes were conducted using generalized linear mixed models (GLMMs). Multivariable‐adjusted models were adjusted for age and sex (Model 2), and Model 2 + pain intensity, anxiety, depression, retirement status, and QoL (Model 3).

All analyses were conducted on the multiply imputed datasets (see Section 2.4). A p‐value < 0.05 will be considered statistically significant. Analyses were performed using SPSS version 29.0.

3. Results

3.1. Study Population

Between April 2017 and February 2020, 468 referred patients completed at least one questionnaire. Of these, 370 were diagnosed with SFN during the standardized diagnostic work‐up. All SFN patients have received an invitation to participate. Of these, only 84 patients completed the before‐diagnosis questionnaire and either the 3‐month (n = 17), the 6‐month (n = 15), or both (n = 52) follow‐up questionnaires and were included in the analyses (Figure 1).

FIGURE 1.

Flowchart of the selection of patients with confirmed small fiber neuropathy (SFN) included in the analysis.

The demographic and clinical characteristics of these patients are represented in Table 1. The mean age was 54.5 years (SD 13.04) and 32.1% were male. Most patients (42.9%) had a high educational level, 42 patients (50.0%) were employed, of which 18 patients were on sick leave (21.4% of total group), 20.2% were unable to work, and 17.9% were retired.

TABLE 1.

Characteristics of the included SFN population.

	Total (n = 84)
Age (SD)	54.53 (13.04)
Sex, male (%)	27 (32.1%)
Pain NRS, mean (SD)	5.52 (2.60)
EQUATION 5D utility score	0.607 (0.191)
Anxiety (SD)	7.05 (4.03)
Depression (SD)	6.21 (3.80)
Level of education
Low, n (%)	21 (25.0%)
Middle, n (%)	27 (32.1%)
High, n (%)	36 (42.9%)
Paid work, yes (%)	42 (50.0%)
Of which on sick leave, yes (%)	18 (21.4%)
Unable to work, yes (%)	17 (20.2%)
Retired, yes (%)	15 (17.9%)
SFN symptom duration, years (SD)	8.77 (9.33)
SFN diagnosis based on, n (%)
TTT	23 (27.4%)
Skin biopsy	14 (16.6%)
Both	47 (56.0%)
Cause of SFN, n (%)
SCN9A variant	3 (3.6%)
Diabetes type 2	1 (1.2%)
Glucose intolerance	2 (2.4%)
Chemotherapy	3 (3.6%)
Sarcoidosis	2 (2.4%)
Sjögren	1 (1.2%)
Other autoimmune	1 (1.2%)
Vitamin B12	10 (11.9%)
MGUS	3 (3.6%)
Idiopathic	50 (59.5%)

Note: Percentages are calculated based on the total study population (n = 84).

Abbreviations: MGUS: monoclonal gammopathy of unknown significance, NRS: numerical rating scale, SD: standard deviation, SFN: small fiber neuropathy, TTT: temperature threshold testing.

The mean duration of SFN symptoms was 8.77 years and SFN diagnosis was confirmed by TTT (27.4%), skin biopsy (16.7%), or both (56%). In 59.5%, no underlying cause was identified. Before diagnosis the mean EQUATION 5D utility score of all patients was 0.607 (SD 0.191). No differences in demographic or clinical characteristics were observed at follow‐up.

3.2. Healthcare Costs

SFN‐related healthcare costs are presented in Table 2. Before diagnosis, 61.9% of patients visited their general practitioner (GP), with a mean of 2.01 visits, corresponding to mean costs of €66.96 per patient (95% CI €49.84–€87.07). GP‐related costs did not change significantly at follow‐up.

TABLE 2.

SFN‐related healthcare costs.

Unit	Unit costs in €	Before diagnosis (n = 84)		3 months after diagnosis (n = 69)		6 months after diagnosis (n = 67)
		Average contacts (CI)	Average costs in € (CI)	Average contacts (CI)	Average costs in € (CI)	Average contacts (CI)	Average costs in € (CI)
GP practice	33.12	2.01 (1.48–2.57)	66.96 (49.84–87.07)	1.80 (1.23–2.39)	60.30 (40.72–80.75)	1.63 (1.15–2.18)	54.09 (38.07–70.80)
GP out‐of‐hours	204.56	0.00 (0.00–0.00)	0.00 (0.00–0.00)	0.04 (0.00–0.12)	1.52 (0.00–4.06)**	0.03 (0.00–1.12)	3.37 (0.00–9.58)**
Emergency room	276.83	0.02 (0.00–0.06)	6.59 (0.00–16.48)	0.03 (0.00–0.07)	8.02 (0.00–20.06)*	0.01 (0.00–0.04)	4.13 (0.00–12.40)*
Medical specialists	128.76	3.08 (2.56–3.73)	397.01 (328.03–475.15)	1.86 (1.35–2.38)	238.86 (173.55–317.23)**	2.46 (1.73–3.19)	317.10 (228.69–411.26)*
Hospitalization days	691.01	0.04 (0.00–0.08)	24.68 (0.00–57.58)	0.12 (0.01–0.28)	81.30 (10.16–182.91)**	0.04 (0.00–0.12)	30.94 (0.00–82.25)*
Physical therapist	41.73	2.76 (1.66–4.05)	115.13 (69.90–168.42)	2.33 (1.38–2.39)	97.37 (55.64–144.54)	3.93 (2.54–5.61)	163.80 (105.26–232.89)
Psychologist	132.41	0.35 (0.11–0.65)	46.83 (14.53–82.35)	0.62 (0.17–1.13)	82.52 (21.16–159.27)	0.40 (0.18–0.66)	53.36 (21.74–90.91)
Other healthcare providers	Various	0.90 (0.43–1.50)	69.50 (30.76–118.73)	0.65 (0.32–1.04)	49.09 (22.33–84.20)	0.94 (0.45–1.57)	75.91 (28.80–131.14)
Help by home nursing organization	48.46	2.63 (0.42–5.63)	92.48 (14.65–197.07)	2.56 (0.46–5.78)	94.8 (16.30–224.21)	1.97 (0.53–4.05)	70.04 (9.59–149.14)*
Prescriped medication	Various	1.89 (1.41–2.40)	40.65 (27.55–54.95)	2.13 (1.48–2.87)	46.10 (30.49–66.22)	1.84 (1.27–2.40)	38.96 (27.38–52.39)
Total Healthcare costs	Various	—	856.17 (703.49–1034.11)	—	758.69 (521.88–1080.93)*	—	811.78 (623.61–1019.76)

Abbreviations: CI: confidence interval, GP: general practitioner.

Asterisks indicate statistically significant differences with baseline (*p < 0.05; **p < 0.001).

Before diagnosis, 88% of patient visited a medical specialist, with a mean of 3.08 contacts and mean costs of €397.01 per patient (95% CI €328.03–€475.15). The number of specialist contacts and the associated costs decreased significantly at both 3‐month (€238.86; p < 0.001) and 6‐month follow‐up (€317.10; p = 0.013).

Other healthcare utilization showed variable changes at follow‐up compared with before diagnosis. Statistically significant increases were observed at follow‐up for out‐of‐hours GP costs, emergency room costs, and hospitalization costs, whereas home nursing costs decreased significantly at 6‐month follow‐up (Table 2).

Total healthcare costs decreased significantly at 3‐month follow‐up (€758.69, 95% CI €521.88–€1080.93) compared with costs before diagnosis (€856.17, 95% CI €703.49–€1034.11). However, the difference at 6‐month follow‐up (€811.78, 95% CI €623.61–€1019.76) was not statistically significant from prediagnosis costs.

3.3. Patient and Family Costs

The total patient and family costs did not differ significantly across time points (Table 3). Within subcategories, several significant changes were observed. Informal care from family or friends increased significantly from €606.40 (95% CI €337.77–€940.64) before diagnosis to €695.50 (95% CI €376.70–€1054.34) at 6 months. Additionally, expenditures on medical devices were higher at 6 months (€317.21, 95% CI €27.26–€844.07) than before diagnosis (€174.97, 95% CI €16.14–€393.80). Travel expenses, OTC medication, and other out‐of‐pocket costs differed at 3 months, but not at 6 months.

TABLE 3.

SFN‐related patient and family costs.

Unit	Unit costs in €	Before diagnosis (n = 84)		3 months after diagnosis (n = 69)		6 months after diagnosis (n = 67)
		Average contacts (CI)	Average costs in € (CI)	Average contacts (CI)	Average costs in € (CI)	Average contacts (CI)	Average costs in € (CI)
Private paid domestic help	Various	1.51 (0.75–2.32)	34.37 (14.81–53.93)	1.74 (0.68–2.94)	46.96 (16.24–82.27)	3.22 (1.08–6.85)	43.81 (17.41–82.16)
Informal care hours by family or friends	20.17	30.06 (16.34–45.80)	606.4 (337.77–940.64)	43.68 (24.03–67.39)	881.15 (495.63–1330.10)	34.48 (20.31–51.94)	695.5 (376.70–1054.34)*
Travel expenses	Various	5.15 (3.76–6.79)	33.02 (16.93–56.34)	3.38 (2.38–4.55)	20.74 (8.99–38.45)*	5.43 (3.69–7.54)	23.07 (10.95–44.85)
Over the counter medication	Various	0.63 (0.44–0.87)	5.93 (3.42–9.14)	0.58 (0.38–0.83)	44.06 (4.53–122.49)*	0.61 (0.42–0.84)	6.91 (3.53–11.11)
Medical devices	Various	0.27 (0.16–0.38)	174.97 (16.14–393.80)	0.23 (0.14–0.32)	23.73 (9.86–40.98)	0.28 (0.18–0.40)	317.21 (27.26–844.07)*
Other costs	Various	—	157.21 (61.76–288.73)	—	356.64 (50.29–840.24)*	—	84.81 (11.87–201.77)
Total patient and family costs	—	—	967.98 (614.91–1327.98)	—	1346.40 (833.65–1983.70)	—	1143.83 (687.79–1728.37)

Abbreviation: CI: confidence interval.

Asterisks indicate statistically significant differences with baseline (*p < 0.05).

3.4. Productivity Loss Costs

Before diagnosis, 50% of patients were employed, working an average of 4.24 days per week (95% CI 3.82–4.63) with 6.6 h per day (95% CI 5.98–7.21) (Table 4). Of these, 43% were on sick leave with a mean of 117.97 h in 3 months, corresponding to mean absenteeism costs of €2430.60 (95% CI €1349.68–€3817.98).

TABLE 4.

SFN‐related productivity loss.

	Before diagnosis (n = 84)	3 months after diagnosis (n = 69)	6 months after diagnosis (n = 67)
Work hours per day; average (CI)	6.6 (5.98–7.21)	6.37 (5.58–7.12)	5.88 (5.04–6.78)
Work days per week; average (CI)	4.24 (3.82–4.63)	4.44 (4.00‐4.81)	4.54 (3.21–5.15)
Total sick days; average (CI)	17.72 (10.69–25.20)	12.63 (5.07–20.53)	13.73 (5.27–23.62)
Total sick hours; average (CI)	117.97 (65.86–179.16)	78.19 (32.50–133.29)	86.03 (28.06–150.62)
Absenteism costs in €; average (CI)	2430.6 (1349.68–3817.98)	1551.65 (558.15–2793.09)	1428.53 (416.73–2583.10)
Extent of pain affecting work productivity; average (CI)	4.43 (3.43–5.47)	4.6 (3.35–5.80)	4.78 (3.44–6.11)
Presenteism in hours; average (CI)	41.42 (24.49–59.39)	38.27 (20.34–61.65)	38.71 (17.97–63.31)
Presenteism costs in €; average (CI)	1772.47 (1083.42–2580.94)	1637.77 (890.88–2546.65)	1656.4 (753.08–2735.74)
Total hours limited in performing daily tasks; average (CI)	96.02 (60.99–138.64)	65.99 (42.94–90.74)	61.7 (38.47–87.85)
Productivity loss due to limited in performing daily tasks in €; average (CI)	1937.04 (1263.05–2748.47)	1331.08 (860.41–1818.42)	1244.67 (813.86–1790.20)
Total productivity loss in €; average (CI)	6067.23 (4332.41–8013.93)	4458.18 (3157.26–6041.52)	4329.61 (2887.20–5974.09)
Total Societal costs in €; average (CI)	7746.93 (5937.85–9621.26)	6563.27 (4923.47–8401.53)	6285.22 (4581.92–8360.44)

Abbreviation: CI: confidence interval.

Patients rated the impact of pain on their work productivity with a mean score of 4.43 on a NRS (95% CI 3.43–5.47), resulting in presenteeism costs of €1772.47 (95% CI €1083.42–€2580.94). At 3‐ and 6‐month follow‐up, absenteeism and presenteeism costs did not significantly differ from prediagnosis costs.

Productivity losses related to limitations in performing daily tasks accounted for €1937.04 (95% CI €1263.05–€2748.47) of total productivity loss before diagnosis. At 3‐ and 6‐month follow‐up, these costs were numerically lower but did not differ significantly from prediagnosis values.

Overall, total productivity loss costs also did not differ significantly between time points.

3.5. Societal Costs

Total societal costs amounted to €7746.93 (95% CI €5937.85–€9621.26) before diagnosis (Table 4). At 3‐month follow‐up, mean total societal costs were €6563.27 (95% CI €4923.47–€8401.53), while at 6‐month follow‐up mean costs were €6285.22 (95% CI €4581.92–€8360.44). Differences in total societal costs between prediagnosis and follow‐up time points were not statistically significant.

3.6. Secondary Outcomes

Pain intensity and QoL did not differ significantly at either follow‐up timepoint in pairwise comparisons (Table 5). Nevertheless, self‐reported health perception statistically improved from 50% of patients feeling good at baseline to 71.6% at 6‐month follow‐up.

TABLE 5.

Changes in pain, quality of life, and health perception before and after diagnosis of small fiber neuropathy.

	Before diagnosis (n = 84)	3 months (n = 69)	6 months (n = 67)
Pain NRS, mean (SD)	5.52 (2.60)	5.62 (2.56)	5.46 (2.58)
EQUATION 5D utility score	0.607 (0.191)	0.623 (0.177)	0.602 (0.194)
Health perception
No change, n (%)	15 (17.9%)	6 (8.7%)*	5 (7.5%)*
Better, n (%)	42 (50.0%)	47 (68.1%)*	48 (71.6%)*

Abbreviations: NRS: numerical rating scale, SD: standard deviation.

Asterisks indicate statistically significant differences with baseline (*p < 0.05).

3.7. Multivariable Analysis

Multivariable GLMM results for the main costs categories are shown in Table 6. Healthcare costs, patient and family costs, and total societal costs showed no differences over time in any of the models. In the unadjusted model, time showed a significant decreasing effect on productivity costs at both 3‐ and 6‐month follow‐up. After full adjustment (Model 3), productivity costs were significantly lower at 3‐month follow‐up (−€1747.98, 95% CI −€3258.81 to €237.15), but not at 6 months. For the secondary outcomes, pain intensity and QoL did not change over time in any of the models.

TABLE 6.

Multivariable analysis.

Variable	Societal costs		Healthcare costs		Patient family costs		Productivity costs		Pain intensity		EQ5D
	Mean β	95% CI	Mean β	95% CI	Mean β	95% CI	Mean β	95% CI	Mean β	95% CI	Mean β	95% CI
Model 1—Unadjusted
Time
3 months	−1482.36	(−3143.17 to 178.45)	−107.90	(−337.32 to 121.52)	334.46	(−230.14 to 899.06)	−1727.16	(−3234.92 to −219.41)*	0.07	(−0.53 to 0.66)	0.01	(−0.04 to 0.05)
6 months	−1746.87	(−3742.20 to 248.45)	−47.58	(−241.92 to 146.76)	135.21	(−363.81 to 634.22)	−1864.36	(−3635.37 to −93.36)*	−0.05	(−0.61 to 0.51)	‐ 0.01	(−0.06 to 0.04)
Model 2—Adjusted for age and sex
Time
3 months	−1485.25	(−3147.74 to 177.23)	−108.01	(−337.51 to 121.51)	333.20	(−231.55 to 897.95)	−1731.05	(−3242.90 to −219.20)*	0.07	(−0.53 to 0.66)	0.01	(−0.04 to −0.05)
6 months	−1740.11	(−3744.76 to 264.54)	−46.81	(−241.76 to 148.14)	133.57	(−365.25 to 632.40)	−1860.55	(−3634.58 to −86.52)*	−0.05	(−0.61 to 0.51)	−0.01	(−0.06 to 0.04)
Model 3—Adjusted for age, sex, anxiety, depression, retirement status, pain intensity and EQ5D.
Time
3 months	−1536.21	(−3196.47 to 124.06)	−117.27	(−760.89 to 525.83)	334.23	(−233.07 to 901.52)	−1747.98	(−3258.81 to −237.15)*	0.11	(−0.46 to 0.68)	0.01	(−0.03 to 0.05)
6 months	−1642.35	(−3595.64 to 310.94)	−45.90	(−241.22 to 149.42)	138.29	(−381.52 to 658.11)	−1741.65	(−3494.04 to 10.74)	−0.03	(−0.60 to 0.54)	−0.01	(−0.06 to 0.04)

Abbreviation: CI: confidence interval.

Asterisks indicate statistically significant effect estimates (*p < 0.05).

4. Discussion

In this study of 84 patients with confirmed SFN, total healthcare costs and patient and family costs did not change significantly after diagnosis. Similarly, pain intensity and QoL remained stable over the follow‐up period. A short‐term reduction in productivity losses was observed at 3 months but was not sustained at 6 months. Interestingly, diagnostic confirmation was associated with improved health perception and fewer specialist visits, resulting in lower specialist‐related costs at both 3‐ and 6‐month follow‐up.

Despite a reduction in specialist‐related costs, overall healthcare costs remained stable. While medical specialist costs represented the largest category, fluctuations were observed in other categories. Costs associated with GP out‐of‐hours visits and hospitalizations were significantly increased at follow‐up, but overall frequency of use was very low. This may reflect chance or short‐term fluctuations due to the limited observation period and sample size and may not accurately represent typical utilization patterns or sustained changes in healthcare use. Because these services are relatively costly per event, even infrequent use within a limited time window may have partially offset the reductions in specialist‐related costs, preventing a significant decrease in total healthcare costs.

A short‐term reduction was observed in productivity losses at 3 months but was not sustained at 6 months. This temporary reduction may reflect short‐term changes in work participation or coping following diagnosis. Similarly, pain intensity and QoL remained stable over the follow‐up period. This suggests that diagnostic confirmation alone did not directly reduce symptom burden or improve overall QoL. The lack of sustained improvement in productivity implies that initial acceptance following diagnostic confirmation may be outweighed over time by the ongoing impact of symptoms.

Although EQ‐5D scores remained unchanged, patients reported significantly better health perception after diagnosis. This suggests that diagnostic clarification may positively affect patients' subjective experience, potentially increasing acceptance of their earlier unexplained symptoms. This distinction between standardized QoL measures and subjective health perception highlights that diagnosis may provide psychological benefits even if physical symptoms remain. However, it is also possible that part of the observed improvement is influenced by the dichotomization of health perception, which could exaggerate the apparent magnitude of change, as minor shifts may move patients into a “better” category even if true differences in perception are minimal.

Corresponding to the improved health perception, the number of specialist visits decreased, leading to lower related costs. This observed decrease may reflect a reduction in diagnostic uncertainty, thus patients perceiving recognition of their complaints through an established diagnosis. Patients may have been less likely to seek multiple specialist consultations once a definitive SFN diagnosis was established. Diagnosing SFN is challenging, and patients often consult several medical specialists before a diagnosis is made. Early recognition and more widely available diagnostic tests may further help streamline healthcare resource use and prevent unnecessary diagnostic consultations, potentially reducing healthcare costs. Studies in other chronic pain conditions have shown lower healthcare utilization with higher pain acceptance [11, 12]. The observed decrease in specialist visits in this study aligns with this, showing that diagnostic clarification can reduce unnecessary consultations, even if symptom severity remains unchanged.

Nevertheless, the lack of significant improvement in societal costs, pain intensity, QoL, and productivity suggests that the burden of SFN persists even if there might be more acceptance of the disease. This pattern is consistent with other chronic pain conditions such as painful diabetic neuropathy, in which costs remain stable over time [25].

These findings suggest that diagnostic confirmation alone is insufficient to mitigate the chronic impact of SFN. Better symptom management, including pharmacological and nonpharmacological interventions, patient education, and support for coping strategies, may be needed to reduce the societal impact of the disease and improve patient outcomes.

4.1. Limitations

Several limitations should be considered. The longitudinal design allowed for the assessment of changes in outcomes before and after diagnosis. Only 84 of the 370 SFN patients completed both pre‐ and postdiagnosis questionnaires, which may result in selection bias. However, to ensure within‐subject comparisons, only patients with both pre‐ and postdiagnosis questionnaires were included and further potential bias was reduced by using multiple imputation. In addition, cost estimates are based on the Dutch healthcare system with a gatekeeping structure and specific reimbursement pathways. The tertiary referral setting may further influence the results as patients with higher educational levels were overrepresented, while secondary causes such as diabetes mellitus were underrepresented. These factors may limit the generalizability of the findings to more socioeconomically diverse populations and to other healthcare systems. Secondly, healthcare use, productivity losses, and medication consumption were self‐reported and subject to recall bias, although standardized instruments were used to minimize this risk. These standardized instruments also allowed for detailed cost calculations providing a comprehensive estimate of the societal burden of SFN, although their validity and reliability have not been specifically tested in SFN populations. Moreover, detailed information on treatment indication and changes in pharmacological or nonpharmacological management over time was not available. Therefore, potential effects of treatment modifications on observed outcomes could not be evaluated. Thirdly, the follow‐up duration was relatively short, and it remains unclear how costs may evolve over a longer period. Productivity losses returned to baseline levels at 6 months, suggesting that reductions may not be sustained. A similar pattern may apply to the specialist‐related costs, which could also return to prediagnosis levels if pain intensity and QoL remain unchanged. Lastly, due to the observational design, causal inferences cannot be made, and observed associations between diagnosis and outcomes should be interpreted with caution.

5. Conclusion

In this study of 84 patients with confirmed SFN, specialist‐related healthcare utilization decreased after diagnosis, while total costs, productivity losses, pain intensity, and QoL remained largely unchanged. Patients' self‐perceived health improved, likely reflecting reduced diagnostic uncertainty and increased acceptance. These findings suggest that, although diagnostic clarification could reduce unnecessary specialist visits, the overall societal burden of SFN persists, highlighting the need for more effective symptom management and support strategies.

Funding

This research was funded by the Prinses Beatrix Spierfonds (W.TR22‐01).

Ethics Statement

The study was approved by the Medical Ethics Committee of Maastricht UMC+ (15‐4‐004).

Consent

Informed consent of all patients was obtained before participating in the study, in accordance with the principles of the Declaration of Helsinki.

Conflicts of Interest

J.G.J.H. reports grants from the Prinses Beatrix Spierfonds (W.OK17‐09, W.TR22‐01 and W.OR24‐04), outside the submitted work. C.G.F. reports grants from the European Union's Horizon 2020 research, innovation program Marie Sklodowska‐Curie grant for PAIN‐Net, Molecule‐to‐man pain network (grant no. 721841), grants from Prinses Beatrix Spierfonds (W.OB18‐03), and others from steering committees for studies in small fiber neuropathy of Biogen, Vertex, Lilly, Olipass, and Sangamo, outside the submitted work. I.S.J.M. reports grants from GBS/CIDP Foundation International and FP7 EU program and Talecris Talents program, as well as others from participation in steering committees/advisory boards for the Talecris Immune Globulin Intravenous For Chronic Inflammatory Demyelinating Polyneuropathy Study, Commonwealth Serum Laboratories, Behring, Octapharma, LFB, Novartis, Union Chimique Belge, Johnson Johnson, Argenx, Octapharma, and JJ study on CIDP, outside the submitted work. The other authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.