The Use of a Respiratory Biofeedback Instrument in Managing Fear and Anxiety in Clinical Practice

Abstract

Introduction and aims

Fear and anxiety are commonly experienced across diverse healthcare settings globally, particularly in clinical practices where these emotions tend to be heightened due to the nature of the procedures. This single-blinded, randomized controlled trial aimed to investigate the effectiveness of a respiratory biofeedback instrument in reducing state and dental anxiety levels during dental extractions in adult patients.

Methods

The trial design and reporting adhered strictly to the CONSORT statement guidelines. Patients were randomly assigned to either the biofeedback group (n = 30) or the control group (n = 30). State anxiety levels were measured using pulse rate, respiratory rate, respiratory regularity, respiratory amplitude, Visual Analogue Scale (VAS) scores, and the State-Trait Anxiety Inventory (STAI). Dental anxiety was assessed using the Modified Dental Anxiety Scale (MDAS), and Dental Fear Survey (DFS).

Results

Results demonstrated that the biofeedback group had a significantly lower state anxiety level during dental extraction, as evidenced by lower respiratory rate (mean difference = –2.75 bpm, P = .03), higher respiratory regularity (5.63%, P = .035) and higher respiratory amplitude (48.83 units, P = .005). The biofeedback group also had a significantly lower STAI-State score after dental extraction (–2.04, P = .015), and a larger reduction in pulse rate from the time of extraction to after extraction (–3.61 bpm, P = .030). However, biofeedback implementation did not significantly affect dental anxiety levels, as measured by the MDAS and DFS.

Conclusions

This study highlights the potential benefits of employing on-site biofeedback instruments to alleviate anxiety during dental extractions in adult patients without the need for multiple training sessions.

Clinical relevance

Further research is needed to explore its impact on dental anxiety levels and investigate its applicability to a broader range of dental procedures.

Introduction

Dental anxiety is a common phenomenon that affects a significant proportion of individuals worldwide. The prevalence of dental anxiety varies across countries. According to a recent meta-analysis that pooled data from over 70,000 adults in 31 studies across 13 countries, the global prevalence of dental anxiety was found to be 15.3%.¹ Furthermore, another meta-analysis examining data from 21 countries estimated that dental anxiety affected 25.8% of children and 13.3% of adolescents.²

Recognizing the nature of dental anxiety is essential due to its significant impact on oral health and general well-being. Individuals who suffer from dental anxiety are more likely to exhibit reluctance to visit the dentist³ and tend to postpone or even skip dental appointments,⁴ resulting in an increased likelihood of oral health issues and deterioration of oral hygiene.^5,6 As a result, the decline in oral hygiene among these individuals becomes more apparent, further exacerbating their dental issues.^7,8 Persistent anxiety and stress related to dental care can lead to sleep disturbance, mood disorders, and even affect personal relationships,^9,10 all of which can hinder normal daily functioning and reduce overall life satisfaction.¹¹ Moreover, dental anxiety can impose a substantial economic burden on both individuals, the healthcare system and the society.¹² The longer a person delays dental treatment, the more complex and costly the required interventions will become. This not only places a financial strain on the individual, but also increases the demand for dental services, stretching the resources of an already overburdened healthcare system.

Dental anxiety management techniques involve the use of pharmacological and nonpharmacological interventions, or a combination of both. Pharmacological interventions such as oral sedatives, nitrous oxide, intravenous sedation, and local anesthesia13, 14, 15 play a significant role in alleviating dental anxiety. These interventions achieve their effects by inducing a state of relaxation and reducing pain perception, allowing individuals to undergo dental procedures with less stress and discomfort.^15,16 However, relying solely on pharmacological interventions may not effectively address the underlying causes of dental anxiety.

Meanwhile, there is a growing body of evidence supporting the effectiveness of non-pharmacological interventions in managing dental anxiety. One type of nonpharmacological intervention, known as the psychological approach, exemplified by cognitive behavioral therapy (CBT) and behavioral therapy (BT), has been proven effective in helping adults cope with mental distress during dental procedures.¹⁷ Other nonpharmacological interventions such as audiovisual distraction, music therapy, progressive muscle relaxation, hypnosis, and virtual reality have demonstrated efficacy in reducing anxiety in dental patients.¹⁸ Nonpharmacological interventions should be well-received by both adult patients and parents of pediatric patients due to the perceived potential medical risks associated with pharmacological interventions. The use of nonpharmacological treatments has been increasingly supported by recent systematic review,^18,19 specifically in reduce anxiety during 1 of the most anxiety-inducing dental treatment, third molar extraction.^18,20

Biofeedback is a non-invasive therapeutic technique that focuses on delivering and monitoring real-time feedback regarding an individual’s physiological process. This information is then disseminated visually or audibly, allowing the monitored individual to observe his own body’s reactions to specific situations or stimuli. The biofeedback approach aims to enhance self-regulation, leading to improved health outcomes.^21,22 By incorporating biofeedback data along with modifications in thoughts, emotions, and behaviors, individuals can achieve their desired physiological changes. With practice, these changes can become long-lasting and can be sustained without the continuous use of monitoring devices.²²

Biofeedback encompasses various techniques that monitor different physiological processes, including (1) electromyography (EMG) biofeedback, (2) thermal biofeedback, (3) electrodermal biofeedback, (4) cardiovascular biofeedback, (5) electroencephalography (EEG) biofeedback, and (6) respiratory biofeedback, among others.23, 24, 25, 26, 27 The clinical efficacy of biofeedback in treating various psychiatric disorders delineated in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) has been evaluated, including (1) neurodevelopmental disorders, particularly autism spectrum disorder (ASD)^28,29; (2) depressive disorders30, 31, 32; (3) schizophrenia spectrum³³; (4) anxiety disorders, such as panic disorder^34,35 and generalized anxiety disorder36, 37, 38; (5) obsessive-compulsive and related disorders³⁷; (6) trauma- and stressor-related disorders, such as posttraumatic stress disorder^39,40; and (7) feeding and eating disorders, such as anorexia nervosa.⁴¹

In recent years, the application of biofeedback therapy as an intervention for dental anxiety has also been investigated. 1 study examined the effectiveness of biofeedback training in reducing dental anxiety among children receiving dental restorations, and reported that biofeedback therapy led to lower levels of anxiety during the initial appointments.⁴² Another study compared the effectiveness of biofeedback therapy and audio-visual distraction in reducing children’s anxiety during local anesthesia administration, and reported that the heart rate of children in biofeedback group was significantly lower.⁴³ In a study conducted with adults, a respiratory rate biofeedback intervention was used to alleviate dental anxiety, and the results suggested a significant reduction in negative feelings regarding the overall injection experience.⁴⁴

While biofeedback therapy has shown significant results in reducing anxiety, most studies have focused primarily on children,^42,43 with only a limited investigation in adult population. Furthermore, most research employs biofeedback training sessions prior to dental treatment, exemplifying a typical technique that requires multiple training sessions and consistent practice to achieve improvements. This may not be feasible for patients who require immediate relief or have limited time resources. If a biofeedback device can be utilized on-site without multiple prior training sessions, it would be more cost-effective and applicable to more dental patients, particularly those receiving single-visit dental treatments such as tooth extractions. Among the aforementioned biofeedback types, it seems that respiratory biofeedback would be 1 of the most promising modalities to be tested on these patients, as it would be more intuitive for patients to control their breathing than to control their heart rate or brainwave. Moreover, a recent study has shown that a 5-minute introduction of respiratory biofeedback followed by a 25-minute biofeedback session could already reduce salivary cortisol levels without multiple prior training sessions.⁴⁵

To the best of our knowledge, no study has yet explored the utilization of a biofeedback instrument during dental extractions to reduce the anxiety level of patients. Therefore, the present study aimed to investigate the use of a respiratory biofeedback instrument as an adjunct to local anesthesia for reducing dental anxiety and general anxiety (i.e. state anxiety) levels in adult patients during dental extractions. We hypothesize that the (i) state anxiety level and (ii) dental anxiety level of patients using a biofeedback instrument during dental extraction will be lower than those of patients not using it.

Materials and methods

Study design

A single-blinded, parallel-grouped with balanced allocation ratio (1:1) prospective randomized controlled trial was conducted. The study investigated the efficacy of a respiratory rate biofeedback device on reducing dental anxiety in patients undergoing dental extraction under local anesthesia.

Participants

Patients presenting to the department of Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong were recruited for the study. Patients who were 18 years of age or above and required non-surgical dental extraction were included in the study. Patients were excluded if they (1) required tooth extraction under monitored anesthesia care (MAC) or general anesthesia (GA); (2) were illiterate; (3) had a history of epilepsy; (4) had known cardiomyopathy; (5) had dementia, mental retardation, mental illness, or similar conditions; (6) were currently undergoing any treatment for anxiety or phobia; (7) had previous experience of feeling ill when using electronic screens; (8) were under medication that had a sedative or drowsiness effect; (9) had intake of preoperative analgesic; or (10) had intake of anti-anxiolytic medication.

Setting

The randomized controlled trial was conducted in a clinical setting. Data were collected in the department of Oral and Maxillofacial Surgery, Faculty of Dentistry, University of Hong Kong. The recruiting period and data collection will start from October 2023 and end in June 2024. Data collection was completed at 1 time-point without the need for follow-up.

Randomization and blinding

Eligible patients with teeth that required extraction under local anesthesia were randomized to 1 of the 2 groups in the study. A research assistant of this study generated a randomization table by using computerized randomization. Eligible patients were assigned following a simple randomization procedure to 1 of the 2 groups: the biofeedback group (study group) and the non-biofeedback group (control group). The allocation sequences were concealed from both the surgeons and the patients in sequentially numbered, opaque and sealed envelopes kept by the research assistant. The research assistant opened the sealed envelope containing the allocation sequence after the study consent was obtained. The surgeon was blinded to the allocation sequence and randomization. The chance for each recruited patient randomizing into the biofeedback group and the non-biofeedback group was 50/50.

Interventions

After study consent was obtained, patients were instructed to complete a structured questionnaire containing a battery of measures to assess dental and general anxiety. Physiologic monitoring was started once the patient was positioned in the dental chair. Before dental extraction, patients in both groups were explained about the nature of the surgical procedure, potential risks and benefits. The dental extraction started after a written consent to the surgery had been obtained.

Patients in the biofeedback group were instructed to wear a belt with a respiration sensor around the chest (RESPeRATE^TM), and smart glasses with biofeedback information. They were instructed to read the biofeedback information displayed on the smart glasses. Biofeedback information was continuously presented with color symbols on the screen. The color of the symbols varied from time to time, depending on the real-time measured value from patients. Four colors of red, orange, yellow and green represent the breathing patterns of patients. A red symbol indicated a pattern of incoherent breathing, while a green symbol indicated a pattern of coherent and deep breathing. See Figure 1A for an example of the displayed biofeedback information.

Fig. 1.

The setup and equipment used in the biofeedback group. (A) An example of biofeedback information displayed through smart glasses. Red, extremely incoherent breathing pattern; Orange, very incoherent breathing pattern; Yellow, slightly incoherent breathing pattern; Green, coherent breathing pattern. (B) The eSense Respiration sensor unit. (C) Adjustable eSense Respiration stretch strap. (D) Wellue SleepU^TM monitor. (E) Smart glasses. (F) The setup and application of the biofeedback instrument during dental extraction.

Patients in the control group were similarly instructed to wear a belt with respiration sensors around the chest, but the smart glasses they wore provided no biofeedback information. The only difference between the 2 groups was the delivery versus no delivery of biofeedback information through the smart glasses.

The implementation of the biofeedback device was carried out during the entire extraction procedure. After dental extraction, patients were asked to complete the questionnaire once again. Physiological monitoring was ended after the completion of the questionnaire. In other words, various data was collected at 3 different time points, namely preoperative (T1), perioperative (T2), and postoperative (T3). The implementation process is shown in Figure 2.

Fig. 2.

The implementation process of a randomized controlled trial.

Outcome measures

Outcome variable in this study was dental anxiety and state anxiety. The independent variable was the implementation of the biofeedback device (i.e. with or without biofeedback information) throughout the dental extraction process.

Measurement tools

Patients were requested to complete a comprehensive, structured questionnaire encompassing various dimensions, including (1) dental anxiety; (2) state anxiety; (3) overall anxiety level before, during, and after dental extraction; and (4) demographic details. Dental anxiety was evaluated using a psychometric measurement, the Chinese version of the Modified Dental Anxiety Scale (MDAS)^46,47 and the Dental Fear Survey (DFS).^48,49 State anxiety was assessed through the psychometric measurement, the State-Trait Anxiety Inventory (STAI)50, 51, 52 and physiological indicators, including pulse rate (PR), oxygen saturation level (SpO₂), and respiratory rate (RR), respiratory amplitude and regularity. Overall anxiety level before, during and after tooth extraction was investigated using a self-reported 10-point Likert Scale. Surgical characteristics such as the number of teeth extracted, complexity of the surgery, and duration of the surgery were recorded.

Measurement of dental anxiety

Dental anxiety was evaluated using the MDAS⁴⁶ and its validated Chinese version.⁴⁷ The MDAS is a self-reported instrument comprising 5 hypothetical questions related to specific dental-related situations: (1) treatment tomorrow; (2) waiting room; (3) tooth drilled; (4) teeth scaled and polished; and (5) local anesthetic injection. Responses of each question rely on a 5-point Likert scale, ranging from “not anxious” to “extremely anxious.” The total MDAS score ranges from 5 to 25, with higher scores indicating a higher level of dental anxiety. A cut-off value of ≥19 has been empirically determined to represent high dental anxiety.^46,53 The MDAS has demonstrated high levels of internal consistency.^46,47,53

Dental Fear Survey (DFS). Dental anxiety was further assessed using the Dental Fear Survey⁴⁸ and its validated Chinese version⁴⁹. The DFS is a 20-item self-administered questionnaire that evaluates (1) avoidance behavior; (2) physiological response; and (3) specific dental-related stimulation or situation; and (4) one item concerning the overall level of fear. The response format employs a 5-point Likert scale ranging from either “never” to “nearly every time” or “not at all” to “very much.” The total DFS score ranges from 20 to 100, with higher scores indicating a higher level of dental fear. A cut-off value of 59 or above was suggested to represent high dental fear.⁵⁴ The DFS was found to have good reliability and validity.^48,49,54,55

Assessment of state anxiety

State-Trait Anxiety Inventory (STAI). State anxiety was assessed using the State-Trait Anxiety Inventory.50, 51, 52 The STAI is designed to evaluate an individual’s levels of state and trait. State anxiety in the scale refers to the temporary, situational feelings of anxiety and stress that an individual might experience in response to specific events or circumstances. It fluctuates over time and can vary in intensity depending on the situation. Trait anxiety in the scale represents an individual’s general tendency to experience anxiety and stress across various situations, reflecting a more stable and enduring aspect of an individual’s personality and predisposition to anxiety. The STAI consists of 40 self-reported items, with 20 items for measuring state anxiety, and the other 20 items for measuring trait anxiety. Response options are based on a 4-point Likert scale ranging from “not at all” to “very much so”. The total score for the STAI ranges from 20 to 80 for each scale, with higher scores indicating higher levels of state or trait anxiety.

Physiological measures. Vital signs, including pulse rate (PR) and oxygen saturation (SpO₂) were measured before, during and after dental extraction, with continuous monitoring throughout the process. Respiration rate (RR), respiratory amplitude, and regularity were measured during the extraction period.

Other subjective measures. A visual analog scale (VAS) was employed to assess overall anxiety levels before, during, and after dental extraction.

Devices and equipment

The respiratory response was measured using the Mindfield® eSense Respiration (Mindfield Biosystems Ltd, Gronau, Germany), a breathing sensor specifically designed for assessing respiration, as shown in Figure 1B and 1C. The strap was fastened around the abdomen. Once the strap was applied, the sensor unit was inserted under the strap, and the cables from the eSense Respiration were connected to the sensor’s push-button connectors. Subsequently, the other end of the eSense Respiration sensor was connected to the eSense Respiration program installed on the tablet. The function of the sensor was primarily to provide feedback and collect data regarding an individual’s breathing frequency, depth, and pattern. Icons shown in Figure 1A are used to display real-time breathing information, allowing patients to receive immediate feedback.

The real-time oxygen levels (SpO₂) and pulse rates were assessed utilizing the Wellue SleepU^TM monitor (Wellue, Shenzhen, China) (Figure 1D), a device designed for the continuous tracking of SpO₂ and pulse rate over time. The system incorporated a ring sensor that was placed on the finger, and the monitoring unit was worn on the wrist.

A pair of smart glasses (Epson Moverio BT-30C; Epson, Suwa, Japan) (Figure 1E) was used to display the biofeedback information from the Mindfield eSense Respiration program on a tablet. This allowed the individual to conveniently assess real-time data regarding their respiration patterns and make necessary adjustments accordingly during dental extraction. The setup and application of the biofeedback instrument during dental extraction in the biofeedback group are demonstrated in Figure 1F. The devices were not calibrated by the researchers. However, they were monitored on-site during data collection, and any instances of null data were immediately addressed.

Sample size calculation

The sample size calculation was conducted based on the effect size of respiratory interventions on anxiety reduction as reported in a published meta-analysis.⁵⁶ The effect size was quantified using Cohen’s d, which was found to be 0.678, indicating a moderate to large effect of the intervention on anxiety reduction. In order to achieve a statistical power of 0.80 at an alpha level of 0.05, a sample size calculation was performed using G*Power software. Considering a type 1 error of 5%, and 80% power, the calculated sample size by G*Power was determined to be 28 patients per group. Therefore, a total of 56 patients would be required for the study to be adequately powered to detect the effect of the respiratory biofeedback instrument on dental anxiety reduction.

Statistical analysis

The data collected was analyzed using the Statistical Package for Social Sciences (Version 26.0 for Windows; IBM Corporation, Armonk, NY). Descriptive statistics were calculated for all variables, including demographic factors, psychometric measures, and vital signs, to determine the mean, standard deviation (SD), and range. Independent samples t-tests were conducted to compare the means of dental anxiety and state anxiety (or psychometric and physiological measures) between the control group and the biofeedback group (inter-group) at various time points (i.e. T1, T2, and T3). Paired t-tests and repeated measures ANOVA were used to compare state anxiety at various time points (i.e. T1, T2, and T3) (within-group). A p-value of less than .05 is considered statistically significant. An independent sample t-test was conducted to compare the surgical characteristics between the 2 groups.

Results

Sample characteristics

Patients will be recruited from 1 January 2024 to 28 June 2024. Out of the 108 patients assessed for eligibility, 62 were randomized. Among the 62 enrolled, 32 were randomly assigned to the biofeedback group, while another 30 were assigned to the control group. Two patients in the biofeedback group were excluded because they closed their eyes while using the biofeedback instrument. In the end, 30 participants from each group were analyzed (Figure 3). Among the sample, 2-thirds was female (66.7%). The age range of the participants in this study ranged from 22 to 77 years. The young adult group (20-39 years) constituted the largest portion of the participants, accounting for 43.3% of the total. Following, the middle-aged group (40-59 years) constituted 35.0% of the sample. The older adult group, comprising those aged 60 years and above, represented the remaining 21.7% of the study population. The demographic characteristics of the study sample and potential confounding variables between the control group and the biofeedback group are shown in Table 1. The dental-related variables between the 2 groups showed no significant difference in the mean number of teeth extracted (P = .105), difficulty (P = .500) and duration of the extraction procedure (P = .204). The extractions were performed by Training Support Dental Officers in Oral and Maxillofacial Surgery, with a total of 4 dental officers participating in the study.

Fig. 3.

CONSORT flow diagram with the ‘Loss to follow-up’ section omitted, as it is not relevant to the study design.

Table 1.

Demographic characteristics of patients randomized into the biofeedback group and control group.

	Sample (n = 60)	Biofeedback group (n = 30)	Control group (n = 30)	p-value†
Age; mean (SD)	45.08 (16.19)	41.63 (15.83)	48.53 (16.06)	.099
Age; range	22-77	23-70	22-77
Gender; n (%)
Male	20 (33.3)	13 (43.3)	7 (23.3)	.100
Female	40 (66.7)	17 (56.7)	23 (76.7)
Level of education attended / completed; n (%)
Primary school	5 (8.3)	0 (0)	5 (16.7)	.023
Secondary school	26(43.3)	10 (33.3)	16 (53.3)
Certificate / Diploma	3 (5.0)	2 (6.7)	1 (3.3)
Associate / Sub-degree	2 (3.3)	1 (3.3)	1(3.3)
Bachelor degree	12 (20)	9 (30)	3 (10.0)
Master’s degree	5 (8.3)	5 (16.7)	0 (0.0)
Doctorate	2 (3.3)	1 (3.3)	1 (3.3)
Not disclosed	5 (8.3)	2 (6.7)	3 (10.0)
Economic status; n (%)
Employed	37 (61.7)	20 (66.7)	17 (56.7)	.322
Unemployed	0 (0)	0 (0)	0 (0.0)
Retired persons	11 (18.3)	4 (13.3)	7 (23.3)
Home-makers	7 (11.7)	2 (6.7)	5 (16.7)
Students	2 (3.3)	2 (6.7)	0 (0.0)
Not disclosed	3 (5.0)	2 (6.7)	1 (3.3)
Dental-related parameters; n (%)
Tooth location (I)
Posterior	59 (98.3)	30 (100)	29 (96.7)	.313
Anterior	0 (0)	0 (0)	0 (0)
Both	1 (1.7)	0 (0)	1 (3.3)
Tooth location (II)
Upper	25 (41.7)	13 (43.3)	12 (40.0)	.961
Lower	25 (41.7)	12 (40.0)	13 (43.3)
Both	10 (16.6)	5 (16.7)	5 (16.7)
Surgeon
A	11 (18.3)	7 (23.3)	4 (13.3)	.133
B	12 (20.0)	4 (13.3)	8 (26.7)
C	21 (35.0)	8 (26.7)	13 (43.3)
D	16 (27.7)	11 (36.7)	3 (16.7)
Duration of extraction procedure (mins)*	27.22 (1543.37)	28.53 (1588.28)	26 (1520.08)	.204
Difficulty of the extraction procedure (1-10 rating scale)*	5.0 (2.53)	5.0 (2.49)	5.0 (2.61)	.500
Number of teeth extracted*	1.38 (0.72)	1.27 (0.52)	1.50 (0.86)	.105

^†Comparison between the biofeedback group and the control group.

^⁎Reported in mean (SD).

Dental anxiety

Inter-group comparison. At T1, there was no significant inter-group difference in both MDAS and DFS scores. Likewise, at T3, the MDAS scores of the control group and the biofeedback group did not significantly differ from each other. The change in MDAS score from T1 to T3 for the control group and the study group was not statistically significant. These results suggest that the dental anxiety level did not significantly differ between the control group and the biofeedback group at either of the 2 time points or in the changes between them (Table 2).

Table 2.

Inter-group comparison of psychometric measures between control and biofeedback groups.

			T1	T3	T1-T3
		Group	Mean (SD)	Mean (SD)	Mean difference (SD)
Dental anxiety
MDAS		Control	11.17 (4.40)	10.70 (4.32)	–0.47 (2.89)
		Biofeedback	11.90 (3.81)	10.07 (4.43)	–1.83 (3.56)
		p-value†	0.246	0.289	0.051
DFS		Control	40.87 (14.90)	-	-
		Biofeedback	41.60 (14.01)	-	-
		p-value†	0.422	-	-
State anxiety
STAI-S		Control	50.57 (3.91)	51.17 (4.18)	0.60 (4.38)
		Biofeedback	49.20 (3.50)	49.13 (2.73)	–0.07 (3.27)
		p-value†	0.079	0.015	0.253
Trait anxiety
STAI-T		Control	51.30 (4.20)	-	-
		Biofeedback	51.67 (3.42)	-	-
		p-value†	0.356	-	-

T1, preoperative; T3, postoperative.

^†Between-group comparison.

Intra-group comparison. In the control group, the analysis revealed no significant differences in MDAS scores between T1 (M = 11.7, SD = 4.40) and T3 (M = 10.70, SD = 4.32), t(29) = 0.886, P = .192. In contrast, the biofeedback group demonstrated a significant decrease in MDAS scores from T1 (M = 11.90, SD = 3.81) to T3 (M = 10.07, SD = 4.43), t (29) = 2.898, P = .004.

State anxiety

STAI-S (Inter-group comparison). At T1, there was no significant difference in scores between the control group and the biofeedback group. At T3, the mean STAI-S score of the control group (M = 51.17, SD = 4.18) was significantly higher than the biofeedback group (M = 49.13, SD = 2.73); t(58) = 2.233, P = .015, with a medium effect size (Cohen’s d = 0.576). However, the change in STAI-S scores from T1 to T3 for the control group did not significantly differ from that of the study group. These results suggest that there was no significant difference in STAI-S scores between the control and biofeedback groups at T1 and in the changes between T1 and T3. At T3, the biofeedback group had a significantly lower state anxiety level than the control group with a medium effect size (Table 2).

STAI-S (Intra-group comparison). In the control group, there were no significant differences in STAI-S scores between T1 (M = 50.57, SD = 3.91) and T3 (M = 51.17, SD = 4.18); t(29) = –0.751, P = .229. Similarly, for the biofeedback group, the difference in STAI-S scores between T1 (M = 49.20, SD = 3.50) and T3 (M = 49.13, SD = 2.73); t (29) = 0.112, P = .456. These results indicate that there was no significant change in state anxiety level in both the control and biofeedback groups across the 2 time points.

Visual Analogue Scale (VAS) (Inter-group comparison). Results revealed no significant differences between the control and biofeedback groups in VAS scores at T1, T2, or in the changes from T1 to T2 (Table 3). However, significant differences emerged at T3. The VAS scores of the control group (M = 4.10, SD = 2.35) are significantly higher than that of the biofeedback group (M = 2.53, SD = 1.48); t(58) = 3.086, P = .002, with a large effect size (Cohen’s d = 0.797). Moreover, the changes from T2 to T3 in the biofeedback group (M = –2.20, SD = 2.48) were significantly larger than that of the control group (M = –1.00, SD = 2.08); t(58) = 2.027, with a medium effect size (Cohen’s d = 0.523). In other words, the biofeedback group exhibited a significantly lower state anxiety level at T3, and experienced a greater decrease in state anxiety level from T2 to T3 compared to the control group.

Table 3.

Inter-group comparison of state anxiety using the visual analog scale (VAS) between the control and biofeedback groups.

	T1	T2	T3	T1-T2		T2-T3
	Mean (SD)	Mean (SD)	Mean (SD)	Mean difference (SD)	Cohen’s d	Mean difference (SD)	Cohen’s d
Control	4.33 (2.63)	5.10 (2.95)	4.10 (2.35)	0.77 (1.29)	0.293	−1.00 (2.08)	0.523
Biofeedback	4.47 (1.94)	4.73 (2.56)	2.53 (1.48)	0.27 (1.82)		−2.20 (2.48)
p-value†	0.412	0.305	0.002	0.131		0.024

T1, preoperative; T2, perioperative; T3, postoperative.

^†Between-group comparison.

VAS (Intra-group comparison). For the control, group, the VAS scores differed significantly between 3 time points, F(1.710, 49.577) = 4.005, P = .030, ηp² = 0.121. Post hoc tests with Bonferroni correction revealed significant differences between T1 (M = 4.33, SD = 2.63) and T2 (M = 5.10, SD = 2.95), as well as between T2 and T3 (M = 4.10, SD = 2.35). For the biofeedback group, VAS scores also showed statistically significant differences across the 3 time points, F(1.789, 51.876) = 17.641, P < .001, ηp² = 0.378. Bonferroni-corrected post hoc tests indicated significant differences between T2 (M = 4.73, SD = 2.56) and T3 (M = 2.53, SD = 1.48), as well as between T1 (M = 4.47, SD = 1.94) and T3.

SpO₂. No significant group differences were found for SpO₂ data (Table 4).

Table 4.

Inter-group comparison of state anxiety using physiological measures between control and biofeedback groups.

	T1	T2	T3	T1-T2		T2-T3
Group	Mean (SD)	Mean (SD)	Mean (SD)	Mean difference (SD)	Cohen’s d	Mean difference (SD)	Cohen’s d
SpO2
Control	97.74 (2.93)	98.39 (0.82)	98.22 (1.04)	0.65 (2.97)	0.178	–0.17(0.50)	−0.106
Biofeedback	98.21 (0.88)	98.48 (0.72)	98.38 (0.84)	0.27 (0.50)		–0.11 (0.64)
p-value†	.201	.318	.263	.247		.342
PR
Control	85.53 (11.77)	89.94 (11.45)	86.33 (10.27)	4.41 (5.77)	1.11	–3.61 (5.17)	−0.49
Biofeedback	85.78 (9.47)	83.22 (10.19)	82.14 (10.60)	−2.55 (6.75)		–1.08 (5.08)
p-value†	.465	.010	.063	<.001		.030

PR, pulse rate; SpO2, peripheral oxygen saturation.

T1, preoperative; T2, perioperative; T3, postoperative.

^†Between-group comparison.

Pulse rate (Inter-group comparison). The analysis revealed significant differences between the 2 groups at T2, the changes from T1 to T2, and from T2 to T3. At T2, the pulse rate for the control group (M = 89.94, SD = 11.45) was significantly higher than the biofeedback group (M = 83.22, SD = 10.19); t(58) = 2.400, P = .010 (Table 4). In addition, the change in pulse rate from T1 to T2 was also significantly different between the control group (M = 4.41, SD = 5.77) and the biofeedback group (M = –2.55, SD = 6.75); t(58) = 4.292, P < .001. Similarly, the change in pulse rate from T2 to T3 showed a significant difference between the control group (M = -3.61, SD = 5.17) and the biofeedback group (M = –1.08, SD = 5.08); t(58) = –1.912, P = .030.

Pulse rate (Intra-group comparison). The analysis revealed significant differences in pulse rate over time within both groups. For the control group, the result showed a significant effect of time on pulse rate, F(1.783, 51.702) = 9.189, P < .001, ηp² = 0.241. Post hoc tests revealed significant differences in pulse rate between T1 and T2 (Mean diff = 4.41, P < .001), and between T2 and T3 (Mean diff = –3.61, P = .030). Similarly, for the biofeedback group, the results demonstrated a significant effect of time on pulse rate, F(1.586, 46.000) = 4.541, P = .023, ηp² = 0.135).

Respiratory rate, breathing regularity and respiratory amplitude (Intra group comparison). For respiratory rate, the control group (M = 22.09, SD = 5.40) was significantly higher than the biofeedback group (M = 19.34, SD = 5.48); t(58) = 1.958, P = .028. For regularity, the control group (M = 73.40, SD = 11.44) was significantly lower than the biofeedback group (M = 79.03, SD = 12.21); t(58) = –1.842, P = .035. Lastly, for respiratory amplitude, the control group (M = 70.95, SD = 54.36) was significantly lower than the biofeedback group (M = 114.78, SD = 72.16); t(58) = –2.657, P = .005. These findings suggest that the biofeedback instrument applied in the biofeedback group had a significant impact on respiratory rate, regularity, and respiratory amplitude compared to the control group. These results provide evidence of the effectiveness of the use of a biofeedback instrument in altering these respiratory parameters (Figure 4).

Fig. 4.

Comparison of respiratory rate (number of breaths per minute), regularity (0%-100%), and amplitude (arbitrary unit) between the control group and the biofeedback group.

*Significant changes were observed (P < .05).

Discussion

The present study has revealed several key findings regarding the impact of biofeedback on anxiety levels in patients undergoing dental procedures: (1) There was a significant decrease in state anxiety level in the biofeedback group compared to the control group; (2) The biofeedback intervention did not significantly affect dental anxiety level, as measured by the MDAS and DFS, when compared to the control group; (3) Using biofeedback techniques during the dental extraction process helped patients manage state anxiety.

The inability of using a biofeedback device to reduce dental anxiety may be explained by the different forms between state anxiety and dental anxiety. Dental anxiety, in contrast to state anxiety, represents a more generalized and enduring fear associated with dental procedures, often stemming from past negative experiences or deeply ingrained psychological factors,⁵⁷ which may be more resistant to change than state anxiety. Our results suggest that while biofeedback techniques can effectively reduce state anxiety during the actual procedure, they may not directly address the underlying dental anxiety that individuals may have. That might be the reason why physiological parameters and state anxiety as measured by STAI-S at T3 showed significant differences, but self-reported dental fear and anxiety did not. The complexity of dental anxiety, often interconnected with past dental experiences and individual psychological factors, may have a more complex etiology, involving a combination of genetic, environmental, and cognitive factors^57,58 that may not be adequately addressed by biofeedback alone, but require a more comprehensive approach for effective management. Future research could explore the potential benefits of integrating biofeedback with other interventions, such as cognitive-behavioral therapy or other psychological interventions in order to better tackle dental anxiety, as patients with state anxiety under dental-related situations and patients with dental anxiety or even dental phobia may not experience the same benefits. By addressing both state and dental anxiety, these integrated approaches could potentially lead to improved patient experiences and outcomes in dental care.

Dental anxiety is a complex phenomenon, and dental extractions are often considered 1 of the most anxiety-provoking procedures in dentistry.⁵⁹ The relationship between dental extractions and dental anxiety is multifaceted and can be influenced by various factors. Furthermore, dental anxiety is an intricate psychological phenomenon that continues to present challenges in the field of dentistry, often impeding oral healthcare delivery. Several psychometric scales such as the MDAS and DFS are currently used to evaluate and measure dental anxiety. However, these tools have their limitations.⁶⁰ While MDAS is praised for its simplicity, it only focuses on anticipatory anxiety and does not consider situational anxiety. However, the scope of both MDAS and DFS is limited as they primarily focus on common fears associated with dental procedures, potentially missing out on more subtle fears like loss of control. Neither tool probes into the specifics of what aspects of the dental visit are causing fear or anxiety. They might not adequately differentiate between fear of the dentist as a person, fear of specific dental procedures, or fear of the dental environment. This lack of specificity could limit their ability to accurately pinpoint the source of a patent’s dental fear and anxiety, potentially making it more challenging to develop effective interventions.

The shortcomings of traditional psychometric measures such as the MDAS and DFS are further pronounced when considering the dynamic nature of dental anxiety and its manifestation in contemporary society. Rapid social changes, including changes in social structure, living environment, and healthcare practices, have likely impacted the reasons individuals develop dental anxiety and the specific aspects they fear. Consequently, relying solely on traditional psychometric measures may not provide a comprehensive understanding of dental anxiety in the current context. More recently developed tools such as IDAF-4C+ were designed to evaluate multiple components of dental anxiety, including emotional, behavioral, physiological, and cognitive aspects.⁶¹ Notwithstanding, a single tool may be unable to assess all potential aspects of dental anxiety. For instance, the advent of new dental technologies may have reduced the prevalence of certain fears, such as those related to pain or invasive procedures. Simultaneously, increased accessibility to information and shared experiences on the internet may have introduced new dimensions to dental anxiety, such as concerns about treatment costs, privacy, or quality of care. Furthermore, as societies become more diverse, it is important to understand how different cultures can affect dental anxiety.⁶² This highlights the need for developing a culturally-sensitive assessment tool.

To address these limitations and better comprehend the evolving complexities of dental anxiety, researchers and practitioners should consider a multifaceted approach. There have been dozens of psychometric tools to measure dental anxiety in the scientific literature, each with its own strengths and weaknesses.⁶³ To further move on, efforts should be directed towards developing more comprehensive and contextually relevant psychometric measures that can capture the diverse dimensions of dental anxiety, including both anticipatory and situational aspects. Moreover, incorporating qualitative research methods, such as interviews and focus groups, can provide valuable insights into the individual experiences and underlying factors contributing to dental anxiety. By understanding patients’ unique perspectives, dental professionals can tailor their approach to alleviate anxiety and foster a more positive dental experience.

The baseline dental anxiety level in the current sample was 11.54 as measured by MDAS. It was comparable to the mean values obtained from prior cohorts of Hong Kong adults (11.94)⁵⁷ and Taiwan adults (10.5).⁴⁷ Notwithstanding, cross-cultural studies should be conducted to explore the influence of cultural factors on dental anxiety and inform the development of culturally appropriate interventions. Furthermore, longitudinal research should be undertaken to examine the impact of social changes on dental anxiety over time, providing a more comprehensive understanding of the phenomenon in contemporary society. By integrating these suggestions, the field of dentistry can move towards a more holistic and accurate understanding of dental anxiety, ultimately improving patient care and outcomes. Biofeedback differs from other distraction techniques, such as listening to music, watching videos, or virtual reality, which divert the attention of an individual away from the dental procedure. Those techniques do not involve self-regulation of anxiety, as they rely on external stimuli to manage anxiety levels. The primary advantage of using biofeedback instruments during dental extraction procedures is their non-invasive nature, which significantly reduces the discomfort and fear typically associated with invasive procedures. Furthermore, biofeedback is primarily a self-regulation technique that involves providing real-time information about an individual's physiological processes, such as heart rate, muscle tension, or skin temperature, which empowers patients by giving them control over their physiological responses, thereby reducing the sense of helplessness and promoting a more positive experience. By gaining insight into their physiological responses to stress, individuals can develop strategies to manage and reduce anxiety, making biofeedback a therapeutic approach that goes beyond simple distraction.

Though the mean difference of 2.75 bpm in respiratory rate between the biofeedback and control groups during dental extraction seems small, it was perceived to be clinically meaningful. Without using a biofeedback device, the control group showed a mean respiratory rate of 22.09 bpm, which exceeded the cut-off threshold of >20 bpm to be classified as tachypnea and increased the risk of hyperventilation-related complications.⁶⁴ In contrast, the biofeedback group had a mean respiratory rate of 19.34 bpm, which was slightly below the tachypnea threshold and could be considered within the normal range. Meanwhile, the improved respiratory regularity and deeper respiratory amplitude were consistent with the principle of having regular, deep breathing that could help individuals to reduce stress and become more relaxed,⁶⁵ though there might not be established numerical references to determine how clinically significant the changes reported in the current study were.

In light of the findings from this study, the use of biofeedback devices during dental extractions offers significant advantages, particularly for patients with high blood pressure and elderly adults who may be at an increased risk of complications during such procedures. The observed reduction in pulse rate among the biofeedback group suggests that these devices can effectively alleviate anxiety and stress associated with dental extractions, subsequently promoting a safer and more comfortable experience for patients. Moreover, the improved hemodynamic stability resulting from biofeedback intervention may enable a wider range of individuals, including those with pre-existing medical conditions, to undergo essential dental treatments on the day of their appointment, minimizing delays in care. In addition, the incorporation of biofeedback devices in dental practice has the potential to enhance patient satisfaction and treatment outcomes. Further research is warranted to optimize the implementation of the biofeedback instrument in various dental settings and to explore its long-term effects on patient well-being and oral health.

The multifaceted applicability and encouraging results of biofeedback should be considered with the understanding that its efficacy may differ based on individual factors and the disorder. Despite the promising benefits of biofeedback instruments, it is important to acknowledge potential challenges faced by certain patient populations, such as elderly individuals, who might struggle to understand and follow the instructions provided by these devices. Cognitive decline, sensory impairments, and unfamiliarity with the technology may hinder the efficacy of biofeedback intervention in this demographic. To address this concern, dental practitioners should consider implementing tailored, patient-centered approaches, such as simplified instructions, larger visual displays, and additional support from dental staff, to ensure the effective use of biofeedback devices among elderly patients. Further research is necessary to investigate the most effective strategies for overcoming these barriers and to assess the feasibility and efficacy of biofeedback intervention in diverse patient populations. Furthermore, biofeedback should not be regarded as a standalone treatment, but rather as an adjunctive method that supplements conventional medical care. By acknowledging the issue and pursuing ongoing research, biofeedback consistently acts as a valuable instrument for healthcare professionals, encouraging a more proactive approach for patients dealing with dental anxiety and maintaining their overall health and well-being.

There are several limitations that should be acknowledged. Firstly, some patients might experience difficulty in understanding how to use the biofeedback instruments, resulting in a need for additional support and assistance during the research process. This could potentially have an impact on the overall effectiveness of the intervention. Secondly, the lighting conditions in the dental clinic were not optimal, as the intensity of the light above the dental chair was sometimes very strong. The strong lighting caused 2 patients in the biofeedback group to close their eyes, which in turn led to the exclusion of their data from the analysis.

Future studies should broaden the scope of investigation to better understand the applicability and effectiveness of biofeedback interventions across various dental procedures. Specifically, future research should consider including a diverse range of dental procedures such as minor oral surgery, root canal treatment, dental implant surgery and orthodontic procedures, among others. By encompassing a wider range of dental settings, future studies will be able to provide a more comprehensive understanding of the utility of biofeedback in managing anxiety across different dental treatments. Furthermore, it would be beneficial for future studies to employ a comprehensive standardized dental anxiety assessment tool, in order to improve the reliability and validity of the results. Lastly, this study highlights the potential of a respiratory biofeedback device in alleviating dental anxiety, aligning with technology-driven innovations in patient care. Artificial intelligence (AI) applications in dentistry, such as real-time physiological monitoring,⁶⁶ similarly aim to enhance precision and precision and personalization in interventions. However, Tuygunov et al.⁶⁷ caution that ethical concerns such as data security and over reliance on technology apply equally to biofeedback and AI. The International Dental Federation’s emphasis on clinician oversight and patient-centered safeguards remains critical to ensure safe implementation. Ultimately, by addressing these research gaps, we can work towards developing more effective strategies for reducing anxiety and improving patient experience during dental treatments. Future studies should consider conducting a multivariate model for more robust analysis to reveal how different independent variables may influence the outcomes.

Conclusion

In conclusion, exploring biofeedback as a tool for managing anxiety in dental practices shows significant promise. Incorporating biofeedback instruments can enhance patient comfort and satisfaction during dental procedures such as extractions, although its impact on dental anxiety and phobia remains to be elucidated. Understanding dental anxiety is crucial for improving oral health and overall well-being. This study underscores the benefits and challenges of biofeedback in dental care and highlights the need for further research to test its efficacy to improve patient experiences and outcomes in other dental procedures.

Conflict of interest

The corresponding author, Professor Yiu Yan Leung, is an Associate Editor of International Dental Journal. He did not involve in the handling or peer review of this manuscript.