Reliability of DNMSQuest as a Screening Tool for Mood Disorders in Cervical Dystonia

ABSTRACT

Background

The high prevalence of mood disorders in cervical dystonia, often unaddressed in botulinum toxin clinics, is a major factor in impaired quality of life. There is a clear need for a brief screening method for identifying these disorders; the Dystonia non‐motor symptoms questionnaire (DNMSQuest) has been proposed as such.

Objective

We aimed to assess the practical utility of the DNMSQuest and compare it with validated rating scales for anxiety, depression and quality of life.

Methods

In 88 patients with cervical dystonia, we compared results from the DNMSQuest with mood rating scales [Beck Anxiety Inventory (BAI), Beck Depression Index (BDI‐II) and Hospital Anxiety and Depression Scale (HADS)], quality of life measures [European Quality of Life (EQOL) and European Quality of Life Visual Analogue Scale (EQOLVAS)] and with assessments of dystonia severity [Cervical Dystonia Impact Profile‐58 (CDIP58) and Toronto Western Rating Scale for Spasmodic Torticollis (TWSTRS)].

Results

Using a cut off score on the DNMSQuest of 5, we noted that DNMSQuest had a sensitivity of 85% for detecting anxiety and depression using the BAI and BDI‐II, and 76% and 78% for anxiety and depression respectively using the HADS. The DNMSQuest correlated strongly with BAI (ρ = 0.715), BDI‐II (ρ = 0.658), HADS‐Anxiety (ρ = 0.616), HADS‐Depression (ρ = 0.706), EQOL (ρ = 0.653) and CDIP‐58 (ρ = 0.665).

Conclusion

The DNMSQuest is a brief, sensitive and non‐specific instrument for identifying patients that warrant further review for anxiety and depression and can easily be implemented in a neurologist‐run botulinum toxin clinic.

The importance of non‐motor symptoms in cervical dystonia has been confirmed by a growing body of evidence.¹, ² Anxiety and depression, seen in nearly 40% of patients are important predictors of quality of life.² Deficits in memory and cognition are also seen.³ Thus, the entire spectrum of this disorder consists of the evident motor dystonic features as well as multiple clinically relevant, non‐motor symptoms.

Dystonia severity, mood symptoms, quality of life and benefit of botulinum toxin therapy, can all be assessed using validated rating scales. Cervical dystonia‐specific measures are commonly used: the Cervical Dystonia Impact Profile‐58 (CDIP‐58) has eight subscales to quantify motor symptoms, activity limitations and psychosocial features.⁴ Others, like the Tsui score, have shown poor correlation between objective findings and patient symptoms.⁵ Rating scales for analyzing non‐motor phenomena are more limited. The Structured Clinical Interview for DSM‐IV is the gold standard for assessing mood disorders⁶ but is not practical for use in a dystonia clinic. The Beck Anxiety Inventory (BAI), Beck Depression Index (second version) (BDI‐II) and Hospital Anxiety and Depression Scales (HADS‐A & HADS‐D), validated self‐reported questionnaires, are used more commonly, but are not dystonia specific.⁷, ⁸

The ideal assessment for non‐motor symptoms would be a short, comprehensive questionnaire with strong inter‐rater reliability. Recently, Klingelhoefer and colleagues developed the Dystonia Non‐Motor Symptoms Questionnaire (DNMSQuest) as a brief, “holistic,” self‐assessment instrument.

The DNMSQuest is a 14‐item questionnaire incorporating fatigue, sleep, sensory symptoms, autonomic symptoms, stigma, emotional well‐being and activities of daily living (ADLs).

The authors suggested that “DNMSQuest appears robust, reproducible, and valid in clinical practice showing a tangible impact of NMS on quality of life in CD. As there is no specific, comprehensive, validated tool to assess the burden of NMS in dystonia, the DNMSQuest can bridge this gap and could easily be integrated into clinical practice.”⁹

One of the difficulties in the latter statement is that there are already two recognized disease‐specific health‐related quality of life measures in existence (CDIP‐58 & CDQ‐24).¹⁰ Also, it appears that the main driver of impaired quality of life in the cervical dystonia is mood disorder.²

In this present study, we compared patient‐reported DNMSQuest scores with recognized validated measures of mood disorder, disease severity and quality of life. In particular, we wished to determine whether DNMSQuest can reliably detect mood disorder in this population.

Methods

Study Population

201 patients attending our clinic with adult onset idiopathic cervical dystonia (as diagnosed by two movement disorder specialists¹¹) were approached to complete all the following assessments: TWSTRS‐2 (severity), CDIP‐58, BAI, BDI‐II, Hospital Anxiety and Depression scales (HADS‐A & HADS‐D), DNMSQuest, European Quality of Life‐5D‐5L (EQOL‐5D‐5L) and EQOL visual analogue scale (EQOLVAS). These questionnaires were completed over a 3‐month period. All patients were treated with three monthly botulinum toxin. Assessments were performed within 1 week of receiving of botulinum toxin, a period at which the effect is minimal. Eighty‐eight patients returned all the questionnaires to us and were included for analysis. Informed consent was provided by all participants and the study was approved by the local Medical Research Ethics Committee.

Study Instruments

Motor Features

Severity of motor dystonic symptoms was assessed using (1): TWSTRS‐2 severity score by the neurologist and (2): CDIP‐58, head and neck domain, by the patient.

Mood Symptoms

The presence of depression and anxiety were determined using a cut‐off score of ≥14 on the BDI‐II, and ≥10 on the BAI (respectively).¹² A cut‐off score of ≥8 was used for anxiety in the HADS‐A, which has a specificity of 0.78 and a sensitivity of 0.9; a cut‐off score of ≥8 in HADS‐D has a specificity for depression of 0.79 and a sensitivity of 0.83.¹³

Quality of Life (QoL)

QoL was assessed using (1): Utility Values derived the European Quality of Life Questionnaire (EQOL‐5D‐5L), (2): the EQOL visual analogue score (EQOLVAS), and (3) the total CDIP‐58 score.

General non‐motor symptoms were assessed using the DNMSQuest. This is a 14‐item self‐completed questionnaire covering a range of NMS in patients with cervical dystonia during the past month. The response option is “yes” or “no” to each question. Completion time by patients is around 5 minutes. The DNMSQuest was used to provide a brief, overall impression of the prevalence and severity of non‐motor symptoms (Fig. 1).

FIG. 1.

Dsytonia non‐motor symptoms questionnaire for cervical dystonia, scored from 0–14. Adapted from Klingelhoefer and colleagues' initial validation study.⁹

Statistical Analysis

Normality was tested for using Shapiro–Wilk and Kolmogorov–Smirnov tests. We evaluated correlations between the DNMSQuest and BDI‐II, BAI, HADS‐A. Data were mainly not normally distributed so Spearman's Rank test was performed. Pearsonʼ rank testing on normally distributed data showed minimal difference in outcome. Correlation coefficients (ρ) were strong if value ≥0.6, intermediate if 0.3 ≥ ρ ≥ 0.59 and weak if ρ ≤ 0.29. Mann–Whitney test (M‐W) was used to compare the differences between two independent groups with dependent variables. The area under the curve of receiver operator characteristics (AUC) was used to assess optimal cut‐off values for DNMSQuest in diagnosing anxiety and depression using BAI, BDI and HADS. Cut‐offs expressed as fractions were rounded to the nearest whole number. A P value of <0.05 was considered to indicate statistical significance.

Results

Cohort Characteristics

The 88 cervical dystonia patients (27 men and 61 women) had a mean age of 59.8 years ±13.4 (range 33–86 years). Mean duration of disease was 15.7 years (Table 1).

TABLE 1.

Cohort characteristics and results of mood measures in the 88 patients with cervical dystonia

Eighty‐eight patients with cervical dystonia
Variable	Mean ± SE	Range
Age (yr)	59.8 ± 1.4	33–86
BAI	12 ± 1.3	0–51
BDI‐II	12.5 ± 1.15	0–40
HADS‐D	7 ± 0.48	0–16
HADS‐A	7 ± 0.62	0–21
HADS‐T	14 ± 1.03	0–37
DNMSQuest	5.6 ± 0.36	0–13

Abbreviations: BAI, Beck Anxiety Inventory; BDI‐II, Beck Depression Index; HADS‐D, Hospital Anxiety and Depression Score (depression); HADS‐Anx, Hospital Anxiety and Depression Scale (anxiety); DNMSQuest, Dystonia non‐motor symptoms questionnaire; SE, standard error.

DNMSQuest

The DNMSQuest mean score was 6 (± 3), range 0–13, (median 6; inter‐quartile range (IQR) 3–8). 80/88 (91%) patients answered “yes” to at least one of the DNMSQuest symptoms. 35/88 (40%) patients scored less than 5. Of these, 23% were diagnosed with depression using HADS‐D; 25% with anxiety using HADS‐A; 14% with depression using BDI and 17% with anxiety using BAI.

Motor Symptoms

The mean TWSTRS‐2‐ severity score and CDIP‐58 (head and neck domain) scores were 10.7 (SD 4.6) and 34.7 (SD 22) respectively. There was no significant difference in motor severity by the CDIP‐58 (head and neck) between male and female participants; P = 0.138, M‐W U = 659.6. Similarly, there was also no statistically significant difference between men and women in TWSTRS‐2 ‐severity; P = 0.98, M‐W U = 822. DNMSQuest had intermediate correlation with CDIP‐58 head and neck, ρ = 0.507; P < 0.01. There was no significant correlation between the DNMSQuest and TWSTRS‐2‐severity ρ = 0.170 (P = 0.55).

Mood Symptoms

43/61 women (70%) and 14/27 men (52%) (P < 0.01) met criteria for mood disorder on at least one of the assessment tools (BAI, BDI, HADS‐A, HADS‐D) (Table 2). Elevated (abnormal) scores were reported by 45% of patients using the BAI for anxiety and 40% for depression using the BDI‐II; 32% patients met criteria for both anxiety and depression using BAI and BDI‐II.

TABLE 2.

Number of patients and percentages by sex, meeting criteria for anxiety and depression using standard mood assessment tools

Assessment tool (total = 88)	Men (27)	Women (61)
BAI ≥ 10	10 (37%)	30 (49%)
BDI ≥ 14	11 (41%)	24 (39%)
HADS‐A ≥ 8	9 (33%)	30 (49%)
HADS‐D ≥ 8	9 (33%)	29 (48%)
HADS‐Total ≥ 16	10 (37%)	29 (48%)

Significantly more women than men reported high scores for depression using the HADS‐D (P = 0.034).

Abbreviations: BAI, Beck Anxiety Inventory; BDI‐II, Beck Depression Index; HADS‐D, Hospital Anxiety and Depression Scale (depression); HADS‐A, Hospital Anxiety and Depression Scale (anxiety).

Women had non‐significantly higher prevalence of anxiety (49%) than men on BAI (37%) and HADS‐A (women 49%; men 33%). The prevalence of depression using the BDI was similar between the sexes. Using HADS‐D, a significant sex‐difference was found; 9/27 (33%) men and 29/61 (48%) women reported abnormal scores (P = 0.034, M‐W U = 589.5).

Quality of Life

There was strong correlation between DNMSQuest and EQOL‐5D‐5L (Spearman's ρ = 0.653). Intermediate correlation was seen with EQOL‐VAS (ρ = 0.454). Both tests reached statistical significance. Similarly the DNMSQuest showed a significant negative correlation with health‐related quality of life using the EQoL Utility Index and a strong positive correlation with the disease‐specific HRQoL measure, the CDIP‐58.(ρ = 0.665; P < 0.001).

Correlations between DNMSQuest and Other Instruments

As can be seen from Table 3 there were highly significant correlations between the DNMSQuest and all the measures used to assess anxiety and depression (Table 3). There was no correlation between the physician–assessed measure of disease severity, TWSTRS‐2 severity. It is noteworthy that of the other TWSTRS2 measures, only the pain assessment correlated significantly with DNMSQuest.

TABLE 3.

Spearman's correlation scores, and where relevant, Pearson's correlation scores, are shown for tools assessing mood, quality of life and motor symptoms

Spearman's correlation (rho) with DNMSQuest screening measure
Measures	Correlation coefficient	95% CI	P value
BAI	0.715	0.490 ≥ 0.815	P < 0.001
BDI‐II	0.658	0.449 ≥ 0.786	P < 0.001
HADS‐A	0.616	0.451 ≥ 0.787	P < 0.001
HADS‐D	0.706	0.589 ≥ 0.880	P < 0.001
HADS‐T	0.698	0.555 ≥ 0.858	P < 0.001
EQOL Utility Index	−0.653	−0.852 ≥ −0.545	P < 0.001
CDIP‐58 total	0.665	0.559 ≥ 0.861	P < 0.001
CDIP58 Head and neck	0.543 (Pearsons)	0.363 ≥ 0.723	P < 0.01
EQOLVAS	−0.454	−0.661 ≥ −0.283	P < 0.001
TWSTRS‐Disability	0.511	0.281 ≥ 0.66	P = 0.055
TWSTRS‐Pain	0.511	0.372 ≥ 0.740	P < 0.001
TWSTRS‐Severity	0.205	−.005 ≥ 0.415	P = 0.55

Abbreviations: BAI, Beck Anxiety Inventory; BDI‐II, Beck Depression Index; HADS‐D, Hospital Anxiety and Depression Scale (depression); HADS‐A, Hospital Anxiety and Depression Scale (anxiety); EQOL, European Quality of Life; CDIP‐58, Cervical Dystonia Impact Profile‐58; EQOLVAS, European Quality Of Life Visual Analogue Scale; TWSTRS, Toronto Western Spasmodic Torticollis Scale; CI, confidence interval.

Receiver Operator Curves: DNMSQuest versus BDI‐II

The unadjusted AUC was 0.824 (95% CI 0.739–0.909) implying excellent discrimination. A cut‐off point of 5 provided a sensitivity of 0.857 and specificity of 0.566. This cut‐off score had a positive predictive value of 56.5% and negative predictive value of 85.7%. Results of AUC for BAI, HADS‐Dep and HADS‐Anx are provided in Table 4.

TABLE 4.

Area under curve of receiver operator characteristics (AUC) discriminatory values for DNMS and BAI, HADS‐A and HADS‐D

	AUC	Sensitivity	Specificity	PPV	NPV
DNMSQuest vs. BAI Optimal DNMS score = 5	0.858	85%	60%	64.2%	82.9%
DNMSQuest vs. HADS‐A Optimal DNMS score = 5	0.803	76.9%	53.1%	56.6%	74.3%
DNMSQuest vs. HADS‐D Optimal DNMS score = 5	0.845	78.9%	54%	56.6%	77.1%

Abbreviations: BAI, Beck Anxiety Inventory; HADS‐D, Hospital Anxiety and Depression Scale (depression); HADS‐A, Hospital Anxiety and Depression Scale (anxiety); PPV, positive predictive value; NPV, negative predictive value.

False Negative & False Positive DNMSQuest Scores

35/88 (40%) patients scored below 5 on the DNMSQuest. Of these, 12 patients met at least one criteria for diagnosis of anxiety or depression using HADS, BDI or BAI. Thus there was a false negative rate of 12/ 35 (34%). Of the 53 patients scoring 5 and above on the DNMSQuest, eight patients did not meet the threshold for mood disorder on any of the four instruments resulting in a false positive rate of 15%. Figure 2 shows spread of DNMSQuest results on patients who met criteria for mood disorders on the employed instruments. Table 5 shows DNMSQuest sensitivities and specificities for diagnosis of anxiety and depression.

FIG. 2.

DNMSQuest scores for patients who met criteria for mood disorder on one, one or more or on none of BAI, BDI‐II, HADS‐A and HADS‐D questionnaires, indicated as medians and interquartile ranges.

TABLE 5.

Sensitivities and specificities for a DNMSQuest score of above five for corresponding assessment tools for anxiety and depression

Tool	Sensitivity	Specificity
BAI ≥ 10	85%	60%
BDI‐II ≥ 14	85.7%	56.6%
HADS‐Anx ≥ 8	76.9%	53.1%
HADS‐Dep ≥ 8	78.9%	54%

Abbreviations: BAI, Beck Anxiety Inventory; BDI‐II, Beck Depression Index; HADS‐D, Hospital Anxiety and Depression Scale (depression); HADS‐A, Hospital Anxiety and Depression Scale (anxiety).

Discussion

The aim of this study was to evaluate the reliability of DNMSQuest as a simple screening assessment tool for mood disorder in cervical dystonia. The gold standard assessment for mood disorders would be a clinical diagnostic interview by a psychiatrist with expertise in the area. This is not feasible in a neurologist‐run dystonia service. Strong AUC values show that the DNMSQuest has reliable discriminatory qualities and may be a useful alternative as a screening tool in the dystonia clinic.

DNMSQuest correlates strongly with BAI, BDI, HADS‐Anx, HADS‐Dep and HADS‐Total (P < 0.001). For quality of life, correlation was strong for EQOL‐5D‐5L and intermediate for EQOLVAS. Correlations with motor rating scales (TWSTRs severity and CDIP58, head & neck domain) were less robust.

Our cohort of 88 patients have a high burden of mood disorders with nearly one third meeting criteria for both anxiety and depression. There was no significant difference between men and women. In a previous assessment of 196 patients with cervical dystonia we noted that 26% of women showed onset of mood symptoms prior to development of cervical dystonia. Women reporting a history of anxiety or depression at any stage, had a significantly earlier onset of cervical dystonia than women not reporting anxiety/depression.⁷ It appears likely that these mood symptoms are an intrinsic part of the disease process of adult‐onset idiopathic focal dystonia (AOIFD), rather than a consequence of living with a chronic illness. One study, comparing AOIFD to alopecia areata (a cosmetically disfiguring condition), suggests that high levels of anxiety and depression may not be secondary to altered physical appearance or chronic illness.¹⁴ Previous studies have shown that psychiatric features can persist despite improvements in motor severity, however it remains unclear if other non‐motor phenomena behave this way.¹⁵ In our analysis, the DNMSQuest did not correlate with motor severity as measured by the neurologist.

The mechanism behind the psychiatric features may relate to changes within a network involving the amygdala. Recently, by voxel based morphometry, patients with task specific dystonias showed enlarged amygdala volumes.¹⁶ The amygdala is crucial in emotional regulation and alterations in amygdala volume has been linked with major depressive disorder.¹⁷ Aberrant amygdala activity has been noted in patients with anxiety disorder.¹⁸ Apart from the mood disorders, the amygdala also provides rapid and automatic processing involved in social cognition. Damage to this area can impair the ability to recognize facial expressions.¹⁹ We have shown previously that patients with cervical dystonia have abnormalities in basic social cognition.³ A network involving the superior colliculus, amygdala and pulvinar nucleus has been shown to be active in processing emotional content, in particular, appraisal of faces and facial expressions.²⁰ We hypothesize that this network for “covert orienting of attention,” mediates the spectrum of motor and non‐motor symptoms of AOIFD.²¹

In terms of the utility of a brief disease specific assessment tool for detecting non‐motor symptoms, the DNMSQuest is useful. It has the advantage of being short and self‐reported. The majority of our assessments of DNMSQuest were carried out in under 5 minutes, while patients were in the waiting room. The high sensitivity in relation to other measures of mood disorder, make it advantageous for use as a screening tool. The only caveat would be that there was a relatively high false negative rate. Using other measures of anxiety and depression as the comparator; 34% of 35 patients, reporting a low DNMSQuest score (<5), also reported a significant mood disorder by these instruments. The false negative rate and lack of specificity should be noted and clinicians intending on using this in their service should be prepared to employ more specific methods of diagnosing mood disorders.

Our dystonia service has approximately 250 patients with AOIFD attending at 3‐monthly intervals for botulinum toxin therapy. Employing the DNMSQuest routinely on all patients at yearly intervals, would provide an accurate perspective on non‐motor symptoms and, in particular, anxiety and depression. Using a cut‐off value of 5, clinicians can then focus on these patients and spend more time, using more specific measures to assess psychiatric disorders.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the first draft, B. Review and Critique.

S.R.: 1B, 1C, 2A, 2B, 3A

I.N.: 1C, 3B

S.O.R.: 1C, 3B

M.H.: 1A, 2C, 3B

Disclosures

Ethical Compliance Statement

This study was approved by the St Vincent's University Hospital Medical Research and Ethics Committee. Informed consent was obtained from all study participants. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest

No specific funding was received for this work. The authors declare that there are no conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months

The authors declare that there are no additional disclosures to report.