Preconsultation Anxiety and Pain during Blood Collection: Impact of Virtual Reality in Children with Cancer – Randomized Clinical Trial

We included patients aged between 7 and 16 years with their first scheduled outpatient appointment. Patients with visual/auditory deficits, sensitivity to light, nausea or vomiting induced by visual stimuli, with tumors that prevent the use of the glasses, epilepsy, impeditive cognitive deficit, untreated anxiety or mental disorders, and organic cerebral syndrome with impeditive deficits were excluded according to the recommendations of the manufacturer and validation of the team.

The researchers evaluated pain in two moments to identify the need for intervention: during the invitation and in the precolection questionnaire. Patients with pain that required intervention were referred to the Center of Outpatient Complications for appropriate treatment.

Participants were recruited through convenience, non-probability sampling, as only patients attending their first consultation were eligible and recruitment depended on the hospital's flow. To estimate feasibility, we calculated the average number of first consultations within the age range of our inclusion criteria. The sample size was based on previous VR studies, considering 80% power, a 5% significance level, and effect sizes reported in the literature, resulting in an estimated requirement of approximately 80 participants.

The characterization questionnaire was created by the authors. To evaluate pain and anxiety before and after the collection, semi-structured questionnaires were also developed, based on Manzini's interview in social research theory,10 on literature instruments that address anxiety and fear in contexts such as those involving VR, and on studies on anxiety before dental and nursing procedures.11 12

The development of the semi-structured interview questionnaires was evaluated by a panel of health professionals composed of 4 nurses and 1 physician; the instruments were pilot-tested for clarity and feasibility. For the measurement of puncture pain, the numerical verbal scale was selected, because its use is a standard at our institution.13

In the present study, we analyzed quantitative variables—age of the child and pain scores (0–10) before and during venipuncture—and qualitative variables, classified as nominal or ordinal.

Nominal variables: sex, family history of cancer, previous diagnosis, prior blood collection and related pain, pain before and during the procedure, anxiety regarding the medical consultation, and perception after the procedure. In the intervention group, additional items assessed prior knowledge of VR, participants' opinions about the headset, and their sensations and perceived performance during the VR game. and a comparison between collection with and without the glasses. The companions reported perceived pain in the child, usual reactions during blood draws, and evaluated the care provided; in the intervention group, they also commented on the headset's usefulness and benefits.

Ordinal variables: companion's level of schooling and kinship to the child, and the child's emotional state while waiting and when called for the procedure (calm, anxious, a little scared/uncomfortable, afraid, or very frightened).

The questionnaires followed the sequence of the encounter:

At the first appointment, the pediatric patients followed the hospital's standard flow: medical record opening, laboratory tests, and nursing screening. During screening, the researcher invited eligible patients, obtained consent, and performed randomization through REDCap (Vanderbilt University). Blood collection was performed in an equipped office. The intervention group used VR glasses before the procedure, while the control group followed the conventional protocol. Patients and companions filled out the pre- and postprocedure questionnaires, and a nurse trained in blood collection and VR assisted the process.

The analysis of the sample involved frequency tables for qualitative variables and measures of central tendency and dispersion for quantitative variables. To compare the pain between groups, the Student's t-test or the Mann-Whitney test was used, according to the normality of the data evaluated by the Shapiro-Wilk test. The associations among categorical variables were analyzed through the Fisher's exact test or the Chi-squared test. The data were recorded in REDCap, version 11.1.18, and analyzed in the R (R Foundation for Statistical Computing) software, version 4.5.0, with a significance level of 5%. The answers to the open question were independently examined by two researchers, who grouped similar passages into descriptive thematic categories.14

For the intervention, Mirage Solo VR glasses (Lenovo Group Limited) were used. The Peruvian company FeelsGood adapted the device by creating a hospital-specific game and blocking access to websites and other tools, allowing only this customized game to be used. When adjusted to the child's face, the game starts: the child uses eye movements to hit objects, with no sudden movements, scary scenarios, or penalties for missed objects. The game includes multiple immersive scenarios (sea, forest, sky, beach, camping), which change automatically.

During blood collection, the professional places the glasses 2 to 5 minutes before the puncture while preparing materials and performing antisepsis. At puncture time, a button intensifies visual and auditory stimuli, such as animated animals or characters, to distract the child. The procedure can be repeated, if necessary, as the game only ends when the glasses are removed. The glasses are hospital-safe, disinfectable, and the game can be played with headphones, using eye movements to interact with objects in the immersive scenarios.

The project was approved by the institutional Ethics Committee in December 2022 (under opinion no. 5.790.508), in accordance with resolution no. 466/2012 of the Brazilian National Health Council.15 All patients signed free and informed consent forms, received a copy, and had confidentiality assured.

Table 3 presents the results of the postcollection questionnaires.

During the blood collection itself, 10 participants in the control group reported no pain, compared with 9 in the intervention group. Fig. 2 illustrates these findings.

No association was observed between sex and the presence of pain (controls: p = 1; intervention group: p = 0.68), previous diagnosis (controls: p = 1; intervention group: p = 0.38), or participant age (controls: p = 0.476; intervention group: p = 0.989). Likewise, pain before the procedure was not associated with pain during collection (controls: p = 1; intervention group: p = 0.723).

It is important to note that the sample size was smaller than originally planned due to logistical limitations. The reduced number of participants may have lowered the statistical power of the analyses, potentially masking clinically-relevant differences between the groups.

During the precollection questionnaire, there was no significant relationship between the participants' age and the presence of anxiety in either group (controls: p = 0.969; intervention group: p = 0.827). In the postcollection questionnaire, age had a significant association with anxiety in the control group (p = 0.049), but not in the intervention group (p = 0.573).

The sex of the participants had no significant influence on the presence of anxiety, either before or after collection, as evidenced by high p-values in both groups (precollection: p = 0.66 for the controls, and p = 0.64 for the intervention group; postcollection: p = 1.00 for the controls, and p = 0.68 for the intervention group). Similarly, there was no association between the presence of anxiety and the diagnostic history of the participants, with non-significant results in the pre- (controls: p = 0.405; intervention group: p = 1.00) and postcollection questionnaires (controls: p = 0.65; intervention group: p = 0.413).

In Table 2, the comparison of preconsultation anxiety between the groups was not significant (p = 0.737), indicating that the groups were equivalent at baseline. Neither was the comparison of postcollection anxiety between the groups significant (Table 3; p = 0.137). However, when comparing pre- and postcollection anxiety between the groups, there was a statistically significant reduction in anxiety levels exclusively in the intervention group (p = 0.026), while the control group remained stable (p = 1.00). This difference is illustrated in Fig. 3.

The anxiety of waiting was evaluated in two moments: during the waiting to be called and at the moment when the nurse called them. The waiting time for blood collection could vary according to the number of research participants in the day and the patient's call order for the previous evaluation. The waiting time varied from 20 minutes to 1 hour. There was no significant association regarding feelings in either group, with a p-value of 0.964 while they were waiting, and of 0.902 and when they were called. Figs. 4 5 show a comparison of the feelings in these two moments between the two groups.

Patients in the intervention group were asked about adverse reactions to the use of VR glasses, such as nausea, vomiting, or headache, and 20 patients (87%) answered that they did not feel anything, while 3 (13%) reported having one of these symptoms.

Patients' responses to the intervention evaluation questionnaire showed statistical significance, indicating that they perceived a positive difference when using the VR headset compared with not using it. The question was asked so that the participants could compare this experience to previous experiences in other institutions. Moreover, when asked about their perception of the use of glasses, 100% stated they enjoyed the VR glasses, none found the game difficult, 91,3% said they think that they played well, and, when asked if they think it makes any difference to collect blood with VR, 82,6% said “yes”.

The companions' perception of pain during blood collection did not differ significantly between the groups. When asked if they thought their child/adolescent experienced pain, 9 companions in the control group and 4 in the intervention group answered “yes” (p = 0.189).

In the intervention group, 22 companions (95.7%) reported that the VR glasses were beneficial to the child/adolescent. In the descriptive thematic analysis of these responses, we identified 3 main categories: 11 companions reported that VR promoted calmness or reduced nervousness, 5 noted it provided distraction during collection, and 7 indicated that the tool “did well” or reduced the pain perceived.