Patient-reported outcomes measured with and without dizziness associated with non-specific chronic neck pain: implications for primary care

Study design

A descriptive, cross-sectional design using non-probability convenience sampling was done. All participants were recruited from September 2017 until February 2018. Patients must have been diagnosed as NSCNP by their referring general practitioner from two primary healthcare centers in Madrid (Spain) and also had been referred to the Physical Therapy Unit. All subjects gave their written informed consent. This study was approved by the Southeast Local Research Committee of the Primary Health Care Management (Code 07/17). The study reporting followed the “Strengthening the Reporting of Observational studies in Epidemiology” (STROBE) guidelines (Von Elm et al., 2007).

Subjects

The study included two groups of patients with NSCNP: (1) those who perceived dizziness combined with NSCNP and (2) those who only had NSCNP. All of them were recruited with frequency matching pairing methods (Setia, 2016). The selection of the sample was carried out by a physical therapist with more than 13 years of clinical experience based on the clinical record and diagnoses provided by a primary care physician. Inclusion criteria for study participants consisted of several parameters: (1) referred to the Primary Health Physical Therapy Unit after complaining of neck pain; (2) having pain for at least three months or longer; (3) a pain intensity greater than 30 mm on a Visual Analogue Scale (VAS); (4) between 18 and 75 years old; and (5) (specific to study group) record of dizziness associated with their neck pain in the assessment date. The exclusion criteria consisted of several parameters: (1) patients who presented systemic illness (such as rheumatoid arthritis, cancer, and/or fibromyalgia); (2) history of cervical region traumas or surgeries; (3) presence of diagnosed radiculopathies or cervical myelopathies; or (4) presently undergoing treatment or having received any type of treatment during the last three months.

Data collection

A sociodemographic questionnaire asking about gender, age, height, weight, body mass index (BMI), civil state, education level, chronicity time, pain intensity, and the actual presence of perceived dizziness was administered to all study participants. The patient-reported outcome measurements were the HRQoL (primary outcome) in addition to pain intensity, disability, and kinesiophobia (secondary outcomes).

Primary outcome

Secondary outcomes

Sample size

The sample size was estimated using G*Power 3.1.9.2 for Mac OS X (G*Power© from University of Dusseldorf, Germany) in order to determine a sufficient sample size. The sample size calculation was based on the HRQoL Sum Score (severity index) of the EQ-5D-5L as the main outcome measurement (Herdman et al., 2011; Obradovic, Lal & Liedgens, 2013; Oppe et al., 2014), A Student’s t-test with two groups based on a medium effect size (d = 0.52) obtained from a pilot study with 10 subjects, five patients reported dizziness combined with NSCNP (34.00 ± 18.84 points) and five patients reported no dizziness with their NSCNP (26.00 ± 15.17 points), was used to obtain the sample size calculation with a statistical power of 80% using an α error of 0.05 and two tails (Faul et al., 2007). Based on the aforementioned assumptions, we estimated a sample size of at least 120 subjects, 60 patients for each group.

Statistical analysis

All data analyses were performed on SPSS for Mac OS X, Version 22.0 (SPSS Inc., Chicago IL) with an 95% confidence interval (CI). Statistically significant differences were considered P-values < 0.05. Descriptive statistics were generated for the sociodemographic data (such as age, BMI, chronicity time, pain intensity, and others). The Kolmogorov-Smirnoff test was performed and showed that the majority of the variables were normally distributed in the sample, so parametric tests were used because of the central limit theorem for samples >50 subjects per group (Mouri, 2013; Hazut et al., 2015). Continuous variables are presented as mean and standard deviation, and the categorical variables are presented as absolute numbers and relative frequency (percentages).

A chi-squared test was used for a comparison of the other qualitative sociodemographic variables and also for the five dimensions of the EQ-5D-5L comparison between groups. A Student’s t-test for independent samples was used to compare quantitative sociodemographic variables, the HRQoL Sum Score, EQ-VAS variables, neck pain disability, and kinesiophobia between groups. Effect sizes were calculated by Cohen’s d and divided into small effect size (d = 0.20 − 0.49), medium effect size (d = 0.50 − 0.79) and large effect size (d > 0.80) (Kelley & Preacher, 2012). Regarding both groups, univariate correlation analyses were performed by the Pearson’s correlation coefficient and categorized as weak (r = 0.00 − 0.40), moderate (r = 0.41 − 0.69) or strong (r = 0.70 − 1.00) correlations (Sedgwick, 2012).

Primary outcomes

The EQ-index analysis revealed a statistically significant difference between groups (P < 0.001) based on the Student’s t-test comparison with 0.77 ± 0.16 and 0.54 ± 0.21 (mean ± SD) for the non-dizziness and dizziness groups, respectively.

The chi-squared test compared the individual domains and levels of the EQ-5D-5L between groups and showed statistical differences (P < 0.05) in the five domains presenting higher levels for the dizziness group with respect to the non-dizziness group (Table 3).

Secondary outcomes

Regarding the Student’s t-test, a statistical difference (P < 0.05) was found between both groups with respect to the quantitative variables, showing greater neck pain disability and kinesiophobia values and also lower values of HRQoL variables in the dizziness group versus the non-dizziness group. The results of the comparison for the other quantitative variables are presented in Table 4.

Correlation analyses

The Pearson’s correlation analysis revealed statistically significant positive correlations in the dizziness group for the HRQoL Sum Score with the perceived disability (P < 0.001; r = 0.649), and moderate positive correlations with pain intensity and kinesiophobia. Also, the analyses revealed statistically significant correlations between all analyzed variables. The results are presented in Table 5.

Regarding the non-dizziness group, the Pearson’s correlation analyses showed statistically significant positive correlations between the HRQoL Sum Score and the perceived disability (P < 0.001; r = 0.725) and negative correlations with the self-perceived health status (P < 0.001; r =  − 0.558). The results are presented in Table 6.

HRQoL

EQ-index data showed worse outcomes in patients with combined dizziness and neck pain than in patients with only neck pain. On the other hand, after analyzing the EQ-5D-5L with an individual comparison between domains and levels, we also found worse results for patients with combined dizziness and neck pain than for those with only neck pain. Our results support prior studies addressing HRQoL in patients with NSCNP (Hoy et al., 2014; GBD 2015 Disease and Injury Incidence and Prevalence Collaborators, 2016) in addition to patients with dizziness (Wrisley et al., 2000; Weidt et al., 2014), but our findings add the negative HRQoL effects due to dizziness in conjunction with NSCNP. In addition, neck pain and dizziness are common conditions that frequently appear in patients who attend to primary healthcare services (Gureje et al., 1998; Wipperman, 2014). Thus, dizziness is a persistent condition that may be associated the persistent back pain that is often found in primary care environments (Gureje et al., 1998; Tschan et al., 2013; Wipperman, 2014; Iglebekk, Tjell & Borenstein, 2017). Consequently, this condition may produce a high economic burden to primary healthcare services (Sun et al., 2014; Mueller et al., 2014; Weidt et al., 2014).

Disability and fear of movement

The variables of perceived disability in neck pain and kinesiophobia demonstrated worse scores for patients with combined neck pain and dizziness as opposed to those whose only ailment was neck pain. Dizziness may predispose an individual with NSCNP to development of instability and misbalance in addition to an increase in disability levels (Wrisley et al., 2000; Reid & Rivett, 2005; Dickin, 2010; Alahmari et al., 2014; Weidt et al., 2014; Minguez-Zuazo et al., 2016; Grande-Alonso et al., 2018a), which may lead to a higher rate of falls, disability-related problems, and economic burden in primary care patients (Sun et al., 2014; Mueller et al., 2014; Weidt et al., 2014). Therefore, patients who suffered from dizziness associated to NSCNP showed greater fear of movement than patients with isolated NSCNP, which seemed to be in line with prior studies (Dickin, 2010; Alahmari et al., 2014; Minguez-Zuazo et al., 2016; Grande-Alonso et al., 2018a). This fact may be due to dizziness may produce a non-specific anomalous spatial sensation secondary to anomalous spine afferences which may alter visual, vestibular or somatosensory inputs (Minguez-Zuazo et al., 2016).

Regarding disability and kinesiophobia of our study, patients with NSCNP without cervicogenic dizziness presented a mean of 55 months of chronicity while patients with dizziness associated to NSCNP showed 43 months suffering from this condition. Despite there were not statistically significant differences between both groups, this fact could suggest greater differences in outcome measurements such as disability and fear of movement. Nevertheless, kinesiophobia scores showed a mean of 24 points in the non-dizziness group while patients with dizziness displayed a mean of 32 points. Regarding chronic pain, the minimum clinically important difference for the TSK-11 was set at 4.8 points and thus the difference of 8 points between both groups may be considered as clinically relevant (Woby et al., 2005; George, Valencia & Beneciuk, 2010; Hapidou et al., 2012). In addition, the NDI showed mild disability in the non-dizziness group while patients with dizziness presented moderate disability (Vernon, 2008; Kovacs et al., 2008; MacDermid et al., 2009). The values of the non-dizziness group could be low for disability and fear of movement compared to the chronicity time showed by these patients. Nevertheless, the presence of dizziness could influence a greater impairment of disability and kinesiophobia, even with less time of chronicity, due to a possible wider central sensitization process (Holle et al., 2015).

Outcome measurements correlations

Regarding correlation analyses of our study, all outcome measurements showed statistically significant correlations between them, except for the severity of health quality of life (HRQoL SumScore) and Kinesiophobia (TSK) in the non-dizziness group. Other previous studies have addressed the relationship between HRQoL and psychological outcome measurements associated with pain perpetuation, such as perceived disability and kinesiophobia. According to Kovacs et al. (2004) the association between these outcome measurements was analyzed in patients with chronic low back pain showing similar correlation findings in line with our study.

As mentioned above, in our study, kinesiophobia was the only outcome measure that did not show any statistically significant correlation with the quality of life severity index in the group of patients without dizziness. This finding could be due to kinesiophobia may be an outcome measure with less influence in subjects with neck pain who did not present associated dizziness. Indeed, kinesiophobia did not present statistically significant correlations with pain intensity or neck disability. Nevertheless, kinesiophobia did seem to be an key outcome measure in those patients with cervicogenic dizziness according to the correlations analysis in the group that presented dizziness and in the same line with previous studies evaluating postural control, psychological and disability variables in patients with cervicogenic dizziness and neck pain (Grande-Alonso et al., 2018b).

Implications for primary care

These results provide a better understanding of the combined effect of these two pathologies, demonstrating the great impact it has on the HRQoL. The area in which this study was conducted was primary care, which is of vital importance since it is the gateway to the public healthcare system in which chronicity and pathology of diseases can best be managed with a correct assessment, referral to specialists, and early intervention (Gureje et al., 1998; Wipperman, 2014). It is important for nurses and clinicians to evaluate all dimensions of the pathology since only by using a biopsychosocial approach can we understand the patient’s condition and intervene satisfactorily, thus avoiding the progression of high chronicity with its associated socio-sanitary costs given the high perceived disability that we report in this work (Valenzuela-Pascual et al., 2015).

Future studies

Due to the reported findings, new interventions should be administered to patients who have combined dizziness and NSCNP in primary care settings in order to increase patients’ HRQoL, functionality, and stability. Some treatments such as soft tissue manual therapy (Reid & Rivett, 2005; Reid et al., 2014b), joint mobilization (Reid et al., 2014b), therapeutic patient education, and exercise therapy (Minguez-Zuazo et al., 2016) form a multimodal approach that includes nursing, physical therapy, and other medical specialities. These treatments should be compared by means of randomized clinical trials in NSCNP patients with dizziness in primary healthcare services.

Key practice points

Greater impairment of HRQoL, functionality, and movement sense were shown in patients with NSCNP combined with dizziness compared to patients with isolated NSCNP. Healthcare actions in research, policy, management and/or education regarding primary care environments for minimizing/preventing HRQoL-associated impairment, disability, and kinesiophobia in patients with NSCNP should be taken, especially when it the NSCNP is combined with dizziness. Physicians and physical therapists can provide these questionnaires to patients with NSCNP and dizziness to detect HRQoL impairment, disability, and kinesiophobia in primary care settings and refer these patients to healthcare management specialists. NSCNP and dizziness management/education in primary care settings should be systematically included in multimodal treatment approaches for patients who suffer from both syndromes.

Limitations

With reference to the study limitations, there may be an inconsistent practice of data transcription. There may also be a limitation with respect to the sample population as patients were recruited from a primary care setting, and although in principle, the results were extrapolated to the entire population given the sample calculation, the conclusion may not be appropriate for certain populations. We also have a high chronicity in both groups within the sample although homogeneity between them is fulfilled. Associated with chronicity is the level of pain severity since one of the inclusion criteria was pain >3 on a VAS scale, which is considered severe/moderate pain. This result was not extrapolated to include patients with mild or no pain. In addition, motion sickness disability was not measured and may be recognized as a main limitation of the present study due to there was not specific outcome measurements related to dizziness. Finally, the present study only evaluated outcome measurements by self-registrations and this may be a limitation due to physical variables, such as cervical range of motion, would be useful in this type of patients in order to study their influence on quality of life and dizziness. Indeed, the TSK-11 total score was the only fear of movement outcome measurement due to this value may be considered as the most relevant index, nevertheless the comparison of the TSK-11 subscales such as Harm and Activity avoidance domains between both groups should be considered for future studies (Roelofs et al., 2004; Gómez-Pérez, López-Martínez & Ruiz-Párraga, 2011).