Introduction

Needles are routinely used in various clinical settings [1] to enable the delivery of drugs, vaccines, and other substances into the body or for fluid and tissue extraction, [2] and billions of injections are administered annually worldwide according to the WHO [3]. Needle phobia is an understudied condition, [4] which is experienced by patients undergoing procedures such as venipunctures and blood donations as well as those with chronic conditions necessitating frequent injections [5–8]. Needle fear exists on a continuum of severity, and can lead to delayed therapy, treatment avoidance, and vaccine hesitancy [6, 7, 9–13].

A significant portion of literature regarding the prevalence and severity of needle-phobia focused on specific subpopulations (i.e., pregnant patients, children, travelers), [5, 14, 15] and the limited research in the general population reports a wide range of prevalence rates from 2.1% to 30%. While a US-based survey reported a remarkably low prevalence (2.1%) of blood-injection-injury phobia (BII; i.e., fear of seeing blood or getting an injection) in adults, [16] a more recent review of the literature showed a needle fear prevalence rate of 20‐30% in the 20–40 age group [8].

Multiple innovation-related (e.g., using smaller/thinner needles or autoinjectors) [17–21] and other strategies (e.g., educational and psychological interventions) [22, 23] to address needle phobia were presented in the literature. However, these assessments were generally limited to evaluating a single approach without comparing different approaches/strategies to address needle fear. Furthermore, patient preferences were not considered, which may provide essential input to developing tailored approaches that address patients’ needs [24]. It is crucial to engage with patients, and integrate their perspectives into the clinical practice and product life cycle of medical devices [24]. In fact, the FDA recognized the importance of patient perspectives and began to incorporate patient input into the medical device evaluation processes about a decade ago [24].

There is a need to investigate the prevalence and severity of needle phobia in a general adult population, and capture patient perspectives regarding strategies that can be utilized to address needle fear. This study aims to identify how commonly and to what extent needle phobia is experienced by a global adult population, its underlying reasons, impacts, and potential strategies to alleviate it.

Methods

Analyses and statistical models

The minimum sample required to achieve a 95% confidence interval and a 2.5% error margin was 1,573. Expecting a 20% drop-out, 2,000 participants was the enrollment target for the study.

Various statistical techniques were utilized to evaluate the data collected by the surveys. The questions were categorized into three different groups: (i) participant characteristics, (ii) alleviation strategies, and (iii) intensity, potential causes, and impacts. Participant characteristics consisted of binary and categorical features, while the other two groups consisted of multiple-choice, ranking, and Likert-scale questions. Counts and percentages were calculated for questions in all three groups, and statistical significances were determined for participant characteristics using chi-square tests [25]. Descriptive statistics (mean, standard deviation, interquartile range) were utilized for Likert-scale questions [26] to understand data distributions better.

Pre-processing was conducted as necessary to transform the dataset into the appropriate data formats for further analysis and modeling. Exploratory data analyses were performed to visualize general patterns present in the dataset. Data representing the participant characteristics were transformed from categorical to numeric values using binary and multi-class encoding (for binary and multi-class variables) and ordinal encoding (for ordinal variables) [27]. Descriptive statistics were acquired for all independent variables (i.e., participant characteristics), and the dependent variable (i.e., needle phobia), to understand their distributions. Correlations with respect to needle phobia were calculated for all variables. A heatmap with all variable correlations was used to visualize the relationships between the independent variables and the dependent variable (i.e., outcome), as well as the relationships among the independent variables (Fig 1). As a test for multicollinearity among the regressors, variance inflation factor (VIF) was calculated for all independent variables [28].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Correlation heat map.

https://doi.org/10.1371/journal.pone.0276814.g001

Statistical models were built using the transformed data and were optimized for accuracy in predicting needle phobia. The dataset was normalized into the range (0, 1) [29] and split into training and test sets [30]. Four algorithms were used to build initial models: Logistic Regression, K Nearest Neighbors, Decision Tree, and Random Forest [31]. Model evaluation was performed using a classification report for all initial models. Odds coefficients were calculated for the Logistic Regression model [32]. Feature importance was visualized for Logistic Regression, Decision Tree, and Random Forest models. Based on insights from exploratory data analyses and initial models, a final model was built, and Logistic Regression was selected as the algorithm for the final model. Numerous model combinations were tested, using grid search to exhaustively search for the best combination of parameter values and feature importance [33]. The final model was evaluated using odds coefficients for feature importance, 5-fold cross-validation, [34] confusion matrix, and classification report. The following equation was used for the final Logistic Regression model:

All statistical analyses were performed using Microsoft Excel and Python version 3.7.6. Python libraries used included: Pandas, Sklearn, Scipy, Matplotlib, and Seaborn.

Results

Prevalence of needle phobia and participant characteristics

A total of 2,098 participants completed the survey and 63.2% (n = 1,325) reported experiencing needle fear. Detailed participant characteristics are provided in Table 1. Most participants experiencing needle fear were female (n = 739; 55.8%; p = 0.004) and approximately half were in either 45–54 (n = 318; 24.0%) or 25–34 age groups (n = 292; 22.0%). The majority of participants with needle fear did not have family history (n = 897; 67.7%; p < 0.001) and had not worked as healthcare professionals (70.9%; n = 940; p < 0.001). Non-needle-related medical fears were experienced by 36.8% (n = 488) of participants with needle fear (p < 0.001) and 31.2% (n = 414; p < 0.001) had a condition that requires frequent injections or blood draws.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participant demographics and characteristics.

https://doi.org/10.1371/journal.pone.0276814.t001

Statistical methods, including regression models, were utilized to identify relationships between participant characteristics and needle phobia. Correlation values were low to moderate for the respective variables, ranging from approximately 0 to as high as 0.3 (Fig 1). The highest positively correlated independent variables with needle phobia were non-needle-related medical fears (correlation coefficient: 0.21), family history of needle phobia (0.18), and presence of a condition that requires frequent injection or blood draw (0.14). The highest negatively correlated independent variables with needle phobia were age group (-0.14), household income (-0.08), and the highest level of education (-0.06). VIF values for all independent variables were within an acceptable range (< 5), which indicated that multicollinearity was not an issue.

Results for the final model were summarized by odds coefficients for feature importance (Fig 2), confusion matrix (S1 Fig), and classification report (S2 Fig). The most important features, according to the Logistic Regression model, were non-needle-related medical fears (odds coefficient = 2.14), family history of needle phobia (1.67), and presence of a condition that requires frequent injections or blood draws (1.43). The model had 64% overall accuracy on the training data, and 66% precision, 87% recall, and 75% f1 score for the positive class. For the testing data, the model had 67% accuracy overall, with 71% precision, 86% recall, and 78% f1 score for the positive class.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Feature importance (odds coefficients) bar chart for final logistic regression model.

https://doi.org/10.1371/journal.pone.0276814.g002

Severity, underlying reasons, and impacts of needle phobia

A scale from 0 (no fear) to 10 (very strong/unreasonable fear or avoidance) was used to assess the intensity of needle fear experienced by the subjects. Participants with needle phobia provided an average rate of 5.7 (± 2.6) and a median [25th, 75th Percentiles] of 6.0 [4.0, 8.0] to describe the intensity of their fear before, during, and after a medical procedure/intervention.

Participant responses regarding the causes and impacts of needle phobia are reported in Table 2. General anxiety (n = 1,273; 96.1%) and pain (n = 1,266; 95.5%) before or during a medical procedure were the most common contributors to fear of needles. Pain was ranked by 37.1% of participants as the largest contributor to their needle fear.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Causes and impacts of needle fear.

https://doi.org/10.1371/journal.pone.0276814.t002

Potential impacts of needle phobia on the participant behavior were also evaluated and provided in Table 2. Of the participants experiencing needle phobia, 52.2% (n = 691) stated avoiding blood draws, followed by 49.0% for blood donations (n = 649), 33.1% for vaccinations (n = 439). Injections to treat a severe medical condition were the least likely to be avoided by the participants (18.3%; n = 242).

Strategies to alleviate needle phobia

Most participants (n = 988; 74.6%) had never sought help while 24.3% participants had seen a therapist either in person (n = 169; 12.8%) or remotely (n = 153; 11.5%). The majority of participants had shared their fear of needles with clinicians (nurses [n = 810; 61.1%]; physicians [n = 588; 44.4%]). The helpfulness of the providers was rated 4.9 (± 3.1) on average on a scale from 0 (unhelpful) to 10 (extremely helpful).

Participants experiencing needle phobia were also asked to indicate the strategies that would reduce their fear of needles and then rank their perceived effectiveness (Table 3). Non-invasive alternatives (n = 1,247; 94.1%) and smaller needles (n = 1,207; 91.1%) were most commonly identified as the device-related solutions to alleviate needle phobia. Smaller needles were ranked as the most helpful device to reduce needle fear by 32.6% of the participants, followed by non-invasive alternatives (26.4%) and autoinjectors (16.5%). Distractions (n = 1,220; 92.1%) and relaxation techniques (n = 1,215; 91.7%) were the most commonly selected non-device-related strategies, and distractions were identified as most helpful (35.9% of the participants experiencing needle phobia).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Device and non-device-based strategies that would reduce needle fear of participants.

https://doi.org/10.1371/journal.pone.0276814.t003

Discussion

Herein, we present a comprehensive survey-based evaluation of needle phobia in a large global adult population. We investigated the prevalence, underlying reasons, impacts, and potential management strategies for needle phobia. Our findings show that over two-thirds of the participants experience some level of needle fear, highlighting its widespread nature. This is significantly higher than the prevalence values reported in the literature for the general adult population (2.1% to 30%), [8, 16, 35, 36] which may partially be due to the variability in defining needle phobia.

Needle fear can be conceptualized on a continuum, from fear to more severe presentations and diagnoses of phobia, depending on the level of distress and impairment/interference [37]. In the Diagnostic and Statistical Manual of Mental Disorders (DSM), needle phobia was formally defined under the BII type of specific phobia [38]. However, the literature used various definitions/nomenclature interchangeably such as belonephobia (fear of needles and pins), trypanophobia (fear of injections), and aichmophobia (fear of sharp, pointed objects) [39]. In 2007, a large US-based survey evaluated the epidemiology of needle phobia using the DSM criteria and reported a strikingly low prevalence (2.1%) of BII phobia in the total adult population [16]. On the other hand, a review of the literature focusing on adults with different chronic diseases used a broader definition as discomfort, anxiety, fear, distress and/or phobia related to needles [22]. The study reported higher rates of needle fear, with prevalence ranging from 17% to 52% among adults with past or current experience of chemotherapy and from 25% to 47% among adults receiving peritoneal dialysis or hemodialysis [22].

While using a clear and universal definition of needle phobia can benefit clinical practice and research, any level of needle fear is important, considering its significant negative impacts on physical and mental well-being [6, 7, 9–13]. Adopting a broad definition may be more inclusive of patients experiencing needle fear in their daily lives, who may not fulfill the criteria for narrow definitions, and could still present risks of avoiding medical care [6, 7, 10–12]. Thus, the high prevalence reported in this study may be more representative of the burden needle phobia constitutes in the general adult population.

Our analyses revealed that non-needle-related medical fears and family history of needle phobia were the most important factors associated with needle phobia, which is consistent with the relevant literature. Individuals diagnosed with a specific phobia were previously shown to present with comorbid phobias [40]. Moreover, a familial tendency for needle phobia has been reported by Orenius et al., who stated that approximately 80% of patients with needle phobia report strong needle fear in a first-degree relative [39, 41].

Having a condition requiring frequent injections/blood draws was identified as another important factor in our regression model, which is particularly relevant considering distinct ways needle fear can alter overall well-being. First, patients with chronic or severe conditions may be more likely to develop needle phobia due to frequent and long-term exposure to needles as part of essential disease treatment. It was shown that the patients with frequent exposure to the medical system are more likely to have painful experiences and develop needle phobia [5, 17, 42]. For example, patients receiving hemodialysis typically require a minimum of 312 needle insertions per year, and needle fear was reported by 25–47% of adults receiving dialysis [22]. Second, needle phobia may lead to poor disease prognosis due to poor adherence to treatment. [10] Because these patients are likely suffering from chronic (e.g., kidney disease) or other severe conditions (e.g., cancer), they may be more vulnerable to needle phobia and its impacts on treatment adherence. Avoidance behavior may have a more significant impact on patients with conditions requiring frequent injections compared to healthier individuals [43]. In fact, literature reports adverse health outcomes and increased mortality in type II diabetes and hypertension patients with poor treatment adherence, which typically comprises frequent injections and blood draws [43].

Different potential reasons may be behind the emergence of needle phobia. While pain was identified as the largest contributor in our research, all the potential causes that we tested were selected by over 90% of participants experiencing needle phobia as contributors to their fear. This highlights the complex and multifactorial etiology behind needle phobia, which is consistent with the literature [10, 44]. Furthermore, it is not surprising that pain was the most critical reason considering a robust body of literature endorsing it as the most prominent underlying reason for needle phobia [5, 17, 42, 45].

As expected, needle fear was associated with widespread avoidance behavior. Most of the subjects with needle fear reported avoiding blood draws, and over one-third reported avoiding blood donations and vaccinations. Needle fear is a common barrier to initiating or adhering to medical treatments and vaccine hesitancy [9–13, 22]. An exploratory study in a tertiary care hospital reported that around half of diabetic patients delayed the start or avoided treatment due to the fear of needles and injections [10]. Moreover, in a school in Colombia, one-quarter of girls considered needle fear as a barrier to obtaining HPV vaccination [12]. Vaccine hesitancy is associated with significant public health implications: outbreaks of measles, mumps, rubella, and pertussis, were linked to under-vaccinated communities [46–48]. More recently, COVID-19 infection and death during the Delta-predominant period and Omicron emergence were higher among unvaccinated persons [49].

Although the treatment for severe conditions was less likely to be avoided (18.3% of participants with needle phobia), poor treatment adherence for these conditions may result in particularly consequential health outcomes [10]. Furthermore, patients experiencing needle phobia are likely more prone to become severely sick since avoiding treatment of a mild medical condition (low risk of morbidity) or vaccinations can lead to severe disease [10, 46].

While most participants have never sought outside help (i.e., therapy), they have shared their fear of needles with providers. The lack of perceived helpfulness of the providers (rated as 4.9 on a scale from 0 to 10) is concerning. Clinicians may benefit from education in identifying and addressing needle phobia [39].

Various strategies to reduce needle fear were presented in the literature and tested in this survey to understand patient preferences and perspectives regarding their comparative effectiveness. All strategies evaluated were considered effective by the participants (at least 86% of participants with needle phobia; Table 3), which is supported by studies that suggested the role of these approaches in alleviating needle fear [14, 50–52]. Smaller needles were identified as the most helpful device in reducing needle fear, followed by non-invasive alternatives. While there is no study directly evaluating the impact of these interventions on needle phobia, smaller needles and other device innovations were shown to minimize pain, which is closely linked with needle phobia. Three RCTs conducted in various population groups (i.e., adults who underwent arterial punctures, diabetic adults, pediatrics) showed that smaller/thinner needles could effectively decrease pain associated with injections [17–19]. The potential positive impacts of non-invasive alternatives on pain and patient satisfaction were also shown in RCTs [50, 53]. Moreover, distractions and medical equipment-related education were perceived as the most helpful non-device-related strategies by participants. While the literature is more limited relative to device-related strategies, previous studies suggested education and distraction-based techniques as potential components of needle phobia interventions. In a case study conducted in a chemotherapy outpatient unit, a combination of nursing interventions, including distractions, improved outcomes in a cancer patient with a reduction in the related anxiety and needle fear [51]. An observational study also demonstrated that multidisciplinary educational approaches during pregnancy significantly improved fear of self-injections [14]. Furthermore, a combination of psychoeducation and applied tension resulted in substantial BII fear reductions and a change in avoidance behavior in 70% of the patients undergoing blood draws [52].

Our study results highlight the extent of needle fear in a general adult population, different factors that contribute to needle fear associated with remarkable avoidance behavior, and the importance of adequate fear management considering potential consequences on the individuals’ well-being and public health. These findings also provide the basis for future studies to evaluate the clinical value of various mitigation strategies, including their impact on avoidance behavior and patient outcomes.

Conclusions

Our findings illustrate a strikingly high prevalence of needle phobia in a sizeable global adult population. Avoidance behavior is common among patients with needle fear and can negatively impact the well-being of individuals and their communities. Patients did not perceive clinician responses as helpful, highlighting a need to devise, evaluate and implement strategies to alleviate needle phobia, and improve the patient experience.

Supporting information

Acknowledgments

The authors would like to acknowledge the following from Boston Strategic Partners, Inc. for their contributions to the study: Dr. Halit Yapici and Dr. Nicholas Antaki for their research and editorial support, and Mr. Jonathan Lilley and Mr. Ian Weimer for statistical assistance.