A Validation Study of PROMIS Sleep Disturbance (PROMIS-SD) and Sleep Related Impairment (PROMIS-SRI) Item Banks in Individuals with Idiopathic Parkinson’s Disease and Matched Controls

Abstract

Background:

Parkinson’s disease (PD) is associated with sleep disturbance (SD) and sleep-related impairment (SRI). Validation of self-report measures of these problems is needed in PD. The Patient-Reported Outcomes Measurement Information System (PROMIS) includes tools that assess these problems (PROMIS-SD and PROMIS-SRI, respectively).

Objective:

This study aimed to further validate these measures in individuals with PD and matched controls.

Methods:

Individuals with early-stage PD (n=50) and matched controls (n=48) completed measures of SD including the PROMIS-SD, Pittsburgh Sleep Quality Index (PSQI), and Insomnia Severity Index (ISI). They also completed measures of daytime impairment including the PROMIS-SRI, Epworth Sleepiness Scale, State-Trait Anxiety Inventory, Beck Depression Inventory 2nd edition, and Parkinson’s Disease Questionnaire-39. Internal consistency for the PROMIS measures were assessed using Cronbach’s α coefficient and item-total correlations in the total sample. Convergent and divergent validity of the PROMIS item banks were assessed using Spearman correlations.

Results:

The PROMIS item banks had excellent internal consistency (α>0.94). Supporting convergent validity, the PROMIS-SD had strong correlations with other measures of SD (ρ>0.68, for PSQI and ISI) and the PROMIS-SRI had moderate to strong correlations with all measures of daytime impairment (ρ=0.41–0.72). Supporting divergent validity within the PD group, the PROMIS-SD correlated more strongly with SRI than with the Parkinson’s Disease Questionnaire total score, a metric of PD related impairment.

Conclusion:

In middle-aged and older adults, with and without early-stage PD, the PROMIS-SD and PROMIS-SRI are reliable and valid measures of SD and SRI, respectively.

Keywords

Parkinson’s disease sleep validation study patient-reported outcome measures

INTRODUCTION

Parkinson’s disease (PD) is a debilitating neurodegenerative disorder that affects 700,000 individuals in the United States and at least 6.1 million individuals worldwide [1]. While problems with physical mobility are the hallmark symptoms of PD, sleep disturbance and sleep-related impairment are common, affecting about 60% of individuals with PD [2 –5]. Individuals with PD often experience reduced total sleep time, fragmented sleep, and lower sleep quality [3 , 7]. Sleep disturbances are rated within the top-ten most bothersome complaints by individuals in both the early and advanced stages of PD [8]. Self-report sleep measures are quick, simple, and cost-effective ways to measure sleep disturbances. However, few self-report measures of sleep disturbance have been validated specifically in PD [9] and no study has validated a sleep-related impairment measure in this population. Establishing valid self-report measures of sleep disturbances and sleep-related impairment in PD is critical for precision assessment and treatment of these problems.

The Patient-Reported Outcomes Measurement Information System (PROMIS) initiative developed various measures, including item banks for sleep disturbance (PROMIS-SD) and sleep-related impairment (PROMIS-SRI), to provide a national resource of publicly available, precise, reliable, and efficient measurements of self-reported patient symptoms and outcomes [10]. The PROMIS sleep item banks or short forms have been validated in several populations, including adults in the general population [11], pediatric populations [12], adults with atopic dermatitis [13], and spinal surgery patients [14]. Few studies have examined the validity of the PROMIS-SD and the PROMIS-SRI in individuals with PD or aging samples. The psychometric properties of a short form of the PROMIS-SD were examined in a sample of older adults, wherein the reliability, construct validity, and predictive validity were shown to be acceptable and strong [15]. In a sample of individuals with PD and their caregivers, the reliability and feasibility of using various PROMIS measures, including the PROMIS-SD and PROMIS-SRI item banks, to track changes in health over time were examined. The PROMIS sleep item banks showed acceptable reliability and feasibility [16]. Further studies examining the divergent and convergent validities of the PROMIS-SD and PROMIS-SRI are necessary to establish the validity of these item banks of sleep disturbance and sleep-related impairment in individuals with PD.

The purpose of this study was to further validate the PROMIS sleep measures by providing evidence for their internal consistency, convergent validity, and divergent validity in a sample of individuals with PD and neurologically healthy matched controls (N=98). We hypothesized that the PROMIS sleep measures would be comparably reliable and valid measures of sleep disturbance and sleep-related imp-airment in individuals with PD as in the matched controls. Moreover, we predicted that the PROMIS-SD would show strong divergent validity by correla-ting more strongly with the PROMIS-SRI than with Parkinson’s related impairment assessed by the total score on the Parkinson’s Disease Questionnaire (PDQ-39). Validating the PROMIS-SD and PROMIS-SRI in individuals with PD and aging adults will establish their usefulness and utility in these populations.

METHODS

Participants

This paper is a secondary analysis of data collected from a previously published study where the methods are explained in detail [17]. Participants ages 53–80 included 50 patients with idiopathic PD and 48 neurologically healthy, matched controls (MC) who were matched for age, sex, race, and education. Participants with PD, who had a clinical diagnosis of idiopathic PD as defined by the United Kingdom PD Society Brain Bank criteria [18], were recruited during routine visits or from an IRB-approved clinical research protocol. Controls were recruited through various methods including patient spouses, newspaper advertisements, local aging communities, fliers in the community, and mailed fliers. The demographic characteristics of this sample have been previously published [17]. In brief, participants were non-dementing, middle-aged to older adults (M=67 years old, SD=7 years), 23% female, 96% Caucasian, 88% married, 23% employed, and well-educated (M=16 years). Sample characteristics are summarized in Table 1.

Table 1

Comparison of sample characteristics between patients with Parkinson’s disease (PD) and matched controls (MC)

Characteristic	Total sample	MC	PD	df	t/z/χ ²	p-value
	(N=98)	(n=48)	(n=50)
Age^a, years	67.5(7.3)	67.5(7.4)	67.4 (7.2)	96	<0.01	0.946
Education^a, years	16[14,19]	16.5[14.5,19]	16[13,18]	–	1.04	0.301
Female gender^a, n	23 (23%)	11 (23%)	12 (24%)	1	0.02	0.899
Caucasian^a, n	94 (96%)	46 (96%)	48 (96%)	2	<0.01	0.999
Married, n	86 (88%)	43 (90%)	43 (86%)	1	3.63	0.604
Employed, n	23 (23%)	13 (27%)	10 (20%)	2	1.51	0.470
PROMIS-SD, total score	49[41,59]	44.5[40,57.5]	52[45,61]	–	–1.98	0.0478^*
PROMIS-SRI, total score	27[21,36]	23[18,28.5]	33[26,39]	–	–4.58	<0.001^**
PDQ–39^b, total score	–	–	32[17.5,47.5]	–	–	–
BDI-II^a, total score	5[2,9]	3[1,4.5]	8[5,11]	–	–5.08	<0.001^**
STAI-state, total score	27.5[23,38]	24[20.5,30]	33[24,43]	–	–3.51	0.004^**
STAI-trait, total score	30[24,38]	28[23,31]	34.5[27,42]	–	–3.09	0.002^**
ISI^a, total score	5[2,11]	4[1,7.5]	6.5[3,12]	–	–2.51	0.012^*
MMSE score^a	29[28,30]	29[28,30]	29[28,30]	–	0.46	0.647
Hoehn-Yahr stage	–	–	1.8(1.1)	–	–	–
Disease duration, years	–	–	7.8(4.9)	–	–	–

Note. M(SD), Median[Interquartile Range]; PROMIS-SD, Patient-Reported Outcomes Measurement Information System - Sleep Disturbance; PROMIS-SRI, Patient-Reported Outcomes Measurement Information System - Sleep-Related Impairment; PDQ-39, Parkinson’s Disease Questionnaire-39; BDI-II, Beck Depression Inventory-II; STAI-State and STAI-Trait, State-Trait Anxiety Inventory; ISI, Insomnia Severity Index; MMSE, Mini-Mental Status Exam. ^aThese sample characteristics have been reported previously in another paper [17]. ^bPDQ-39 was solely administered to the PD group. ^*p<0.05; ^**p<0.01.

Participants completed screening, a clinical sleep interview, and various questionnaires related to sleep and mood in a clinical testing room at the University of Florida or in participants’ homes. Participants were informed of the risk of fatigue associated with the multiple tests and questionnaires. They were allowed breaks as needed. The participants scored in the normal range on the MMSE, a cognitive screener, with no difference between the groups. Participants completed the study while on their medications and rarely required assistance with competing the research protocol. All patients with PD were assessed in the ON state (i.e., while medicated) and were stabilized on the medications for at least three months before participating. A modified clinical sleep interview was used to assess sleep history, psychosocial functioning, medical conditions, medications, and mental health. This study focused on measures pertinent to the present analyses.

Measures

PROMIS sleep measures

The Patient-Reported Outcomes Measurement Information System –Sleep Disturbance (PROMIS-SD) is an item bank consisting of 27 items that assess sleep quality and sleep disturbance. There are various short forms of the PROMIS-SD including an 8-item short form [19]. For the main analyses of this study, we used the full item bank of 27 items. The Sleep-Related Impairment (PROMIS-SRI) is an item bank of 16 items that assess daytime impairment related to sleep. There are various short forms of the PROMIS-SRI including an 8-item short form [19]. For the main analyses of this study, we used the full 16 items.

Other measures of sleep disturbance

The Insomnia Severity Index (ISI) is a widely used, validated, 7-item questionnaire that assesses sleep quality and insomnia severity over the previous 2 weeks [20]. The ISI is a continuous measure with scores ranging between 0 and 28.

The Pittsburgh Sleep Quality Index (PSQI) is an 18-item, self-report questionnaire used to measure sleep quality across several dimensions including sleep disturbance and total sleep time within the past month. The PSQI is a reliable and valid measure of sleep disturbance, including in samples of older adults [21].

Measures of daytime impairment

The Epworth Sleepiness Scale is an 8-item questionnaire that provides an index of the severity of daytime sleepiness. Participants are asked to rate the likelihood of dozing in eight different situations. This measure is widely used in PD [22] and has been validated in older adult samples [23].

The State-Trait Anxiety Inventory is a 40-item scale that assesses current (state) and enduring (trait) levels of anxiety. This measure is well-validated in healthy, clinical, and older adult populations [24].

The Beck Depression Inventory 2nd edition is a 21-item self-report inventory that measures depression symptom severity. The BDI-II is widely used to assess depression in older adult samples. A Movement Disorders Society task force recommended the BDI-II as a valid continuous measure of depression severity in PD [25].

The Parkinson Disease Questionnaire-39 (PDQ-39) is a validated 39-item self-report questionnaire that assesses Parkinson’s disease-related impairment by asking: “Due to having Parkinson’s disease, how often during the last month have you... ” had difficulty with specific dimensions of daytime functioning including activities of daily living, cognition, communication, depression, functional mobility, quality of life, and interpersonal functioning [26]. The total score of all 39 items from the PDQ-39 was used as a measure of Parkinson’s disease-related impairment for the analyses of this study. The PDQ-39 measure was only collected from the PD group.

Statistical analyses

Data were analyzed using Stata SE 16. Internal consistency was assessed for the PROMIS-SD and PROMIS-SRI item banks and the 8-item short forms using Cronbach’s α coefficient in the total sample. Cronbach’s alpha values <0.70 indicates inadequate internal consistency while values >0.95 indicates redundant items, with values in between indicating adequate reliability. Correlations of each item with the total score (item-total correlations) were examined in the total sample and for each group. Cor-relation values >0.20 were considered satisfactory. The groups’ item-total correlations were compared using 2-sample Fisher’s z tests. Convergent validity of the PROMIS-SD was examined using Spearman’s rho coefficients conducted between the PROMIS-SD, PSQI, and ISI in the total sample and in both groups. Wald’s multivariate test for equality of correlation matrices was used to compare the correlations of the PROMIS-SD and PROMIS-SRI with the other measures by group. Spearman’s rho was also used to correlate the PROMIS-SRI with other measures of daytime impairment (i.e., ESS, BDI-II, and STAI) in the total sample and each group separately. In addition, Spearman’s rho was used to correlate the PROMIS-SRI with the PDQ-39 within the PD group only. Divergent validity was examined by comparing how much variation in Parkinson’s disease-related impairment (PDI) and SRI could be explained by SD. Two simple linear regressions were computed: one with SD predicting SRI and another with SD predicting PDI. The resulting correlation coefficients were transformed using Fisher’s z and compared using a 2-sample Fisher’s z test.

RESULTS

Internal consistency

The PROMIS-SD and PROMIS-SRI item banks showed good internal consistency in the total sample (Cronbach α = 0.94, for both). The PROMIS-SD and PROMIS-SRI 8-item short forms also indicated good internal consistency (Cronbach α = 0.87 and α = 0.91, respectively). Based on Fisher’s z 2-sample test, which transforms the sampling distribution of correlation coefficients to normal distributions and tests for significant differences [27], the item-total correlations did not differ between groups for the PROMIS-SD and PROMIS-SRI (z=0.30, p-value= 0.76; z=1.12, p-value= 0.264, respectively). All SD items except for item 70 “I felt sad at bedtime” (ρ =0.19) correlated satisfactorily (values >0.20) with the total score in the total sample. Items 109 “My sleep quality was very poor, poor, fair, good, very good”, 90 “I had trouble sleeping”, and 115 “I was satisfied with my sleep” correlated the highest with the total score (ρ =0.77–0.81).

Convergent validity

Table 2 shows the results of Spearman correlation analyses between PROMIS-SD and PROMIS-SRI total scores with other measures of sleep disturbance and daytime impairments. The PROMIS-SD and PROMIS-SRI were highly correlated with each other (r_s=0.69, p-value <0.01). Suggesting convergent validity, the PROMIS-SD was strongly correlated with the other measures of sleep disturbance, namely the PSQI and ISI. The PROMIS-SRI was significantly correlated with all other measures of daytime impairment (i.e., ESS, BDI-II, STAI-State and Trait, and PDQ-39) (p-value <0.01). From Wald’s χ² test which evaluated the equality of the correlation matrices, there were no significant differences between groups in the PROMIS-SD, PROMIS-SRI, PSQI, BDI-II, STAI-state, or STAI-trait total-score correlations (χ²₍₁₅₎ =20.86, p-value= 0.141).

Table 2

Correlations between the total scores of the PROMIS-SD and PROMIS-SRI and other measures in the total sample, individuals with Parkinson’s disease (PD), and matched controls (MC)

	PROMIS-SD			PROMIS-SRI - Sleep-Related Impairment
	Total sample	PD	MC	Total sample	PD	MC
	(N=98)	(n=50)	(n=48)	(N=98)	(n=50)	(n=48)
PSQI global score	0.68^*	0.73^*	0.63^*	0.56^*	0.55^*	0.51^*
ISI, total score	0.73^*	0.77^*	0.69^*	0.61^*	0.64^*	0.53^*
PDQ-39^a, total score	–	0.50*	–	–	0.68^*	–
BDI-II, total score	0.52^*	0.46^*	0.40^*	0.72^*	0.71^*	0.49^*
STAI-State, total score	0.57^*	0.41^*	0.66^*	0.61^*	0.61^*	0.48^*
STAI-Trait, total score	0.56^*	0.45^*	0.57^*	0.64^*	0.60^*	0.62^*
ESS, total score	0.21	0.29	0.05	0.41^*	0.43^*	0.22

Note. PROMIS-SD, Patient-Reported Outcomes Measurement Information System - Sleep Disturbance; PROMIS-SRI, Patient-Reported Outcomes Measurement Information System - Sleep-Related Impairment; PSQI, Pittsburgh Sleep Quality Index; PDQ-39, Parkinson’s Disease Questionnaire-39; BDI-II, Beck Depression Inventory-II; STAI-State and STAI-Trait, State-Trait Anxiety Inventory. In a sample that included patients with Parkinson’s disease and matched controls, we found strong associations between the PROMIS-SD and other measures of sleep disturbance (PSQI and ISI) but only weak-moderate associations with measures of daytime impairment (PDQ-39, BDI-II, STAI, and ESS). Moderate to strong associations between PROMIS-SRI with all measures of sleep disturbance and daytime impairment were also found. ^aThe PDQ-39 was only administered to patients with PD, thus the sample for these correlations was the PD group (n=50). ^*p<0.01.

Divergent validity

Two simple linear regressions were conducted to evaluate the percent of variance in SRI and PDI that sleep disturbance predicted within the PD group. Sleep disturbance predicted 63% of the variance in sleep-related impairment and 28% of the variance in PDI. Using Fisher’s z transformations, the percentage of variances explained in SRI and PDI by SD were significantly different (z=2.39, p-value= 0.017).

DISCUSSION

This study sought to validate the PROMIS-SD and PROMIS-SRI in a sample of individuals with PD and age-matched controls. The PROMIS-SD and -SRI item banks and short forms had excellent internal consistency in the total sample. Demonstrating convergent validity, the PROMIS-SD correlated strongly with other widely used measures of sleep disturbance and the PROMIS-SRI moderately correlated with daytime impairments common to both PD and sleep disturbance. In PD, the PROMIS-SD was a better predictor of SRI than PDI, demonstrating its divergent validity. Thus, the PROMIS-SD and the PROMIS-SRI were found to be valid measures of sleep disturbance and sleep-related impairment in both individuals with PD and age-matched controls. Other studies have also shown the PROMIS-SD to be a reliable measure and that it is significantly associated with negative health outcomes [13, 14]. This suggests that the PROMIS-SD item bank is a valid measure of the construct of sleep disturbance. This research contributes divergent and convergent validation of the PROMIS-SD and preliminary evidence for the convergent validity of the PROMIS-SRI, building on previous evidence that these measures are useful and valid in individuals with PD and in aging adults.

To our knowledge, this study is the first to provide preliminary evidence of the convergent validity of the PROMIS-SRI and begins to answer the call for further studies of the convergent validity of the PROMIS-SRI [11]. In other samples, the PROMIS-SRI was significantly higher in patients with Temporomandibular disorder, a disorder that is associated with impaired sleep and daytime impairment due to sleep [28]. Furthermore, the PROMIS-SRI has been shown to be a significant predictor of fatigue in patients with epilepsy [29]. Perceived stress and fatigue are both highly associated with sleep impairment [30, 31]. Taken together, these observations suggest that the PROMIS-SRI measures the construct of daytime sleep-related impairment. However, few studies have examined the construct, convergent, and divergent validities of the PROMIS-SD and -SRI in PD or other aging adult samples. Further research is necessary to fully establish the validity of these sleep measures in these populations.

A limitation of the current study was the relatively small sample size. A larger sample would have provided opportunities to conduct factor analyses to strengthen the evidence for the construct validity of the PROMI sleep measures. Future studies could provide additional information about the validation of these measures in larger samples that include a wide range of PD severity and against polysomnography and actigraphy. Nevertheless, our sample size was sufficient to provide evidence for other forms of validity and reliability. The PD participants in this sample were in earlier stages of the disease which may have contributed to a less severe presentation of symptoms; therefore, these results may not generalize to individuals with a more severe presentation of PD. The study utilized self-report measures and did not validate these measures against other types of sleep measurements (e.g., polysomnography, actigraphy).

In conclusion, we found that the PROMIS sleep measures are reliable and valid measures of sleep disturbance and sleep-related impairment in individuals with PD and in an age-matched control group of middle-aged and older adults. Additionally, we showed evidence for their convergent and divergent validity. Because sleep disturbance and sleep-related impairment are common symptoms of PD, having valid and reliable measures of these constructs is critical for their identification and consequent treatment in PD.

CONFLICT OF INTEREST

There are no conflicts of interest to report.

Footnotes

ACKNOWLEDGMENTS

This work was supported by the Ethel Moore Alzheimer’s Program, National Institutes of Health (grant/award number T32 NS082168); National Institute on Aging (grant/award numbers R01 AG054370, R21 AG057200, T32 AG020499); National Institute of Mental Health (grant/award number R03 MH109336); National Institute of Child Health and Human Development (grant/award number R01 HD091658); National Institute of Nursing Research (grant/award number R01 NR014810); State of Flo-rida Education, and National Institute of Neurological Disorders and Stroke (grant/award numbers R01 NS082386, R01 NS096008, UH3 NS095553).

References

Collaborators GBDPsD (2018) Global, regional, and nat-ional burden of Parkinson’s disease, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 17, 939–953.

Chahine

, Amara

, Videnovic

(2017) A systematic review of the literature on disorders of sleep and wakefulness in Parkinson’s disease from 2005 to 2015. Sleep Med Rev 35, 33–50.

Tandberg

, Larsen

, Karlsen

(1998) A community-based study of sleep disorders in patients with Parkinson’s disease. Mov Disord 13, 895–899.

Shin

, Pohlig

, Habermann

(2017) Self-reported symptoms of Parkinson’s Disease by Sex and Disease Duration. West J Nurs Res 39, 1412–1428.

Karlsen

, Larsen

, Tandberg

, Maeland

(1999) Influence of clinical and demographic variables on quality of life in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 66, 431–435.

Bergonzi

, Chiurulla

, Gambi

, Mennuni

, Pinto

(1975) L-dopa plus dopa-decarboxylase inhibitor. Sleep organization in Parkinson’s syndrome before and after treatment. Acta Neurol Belg 75, 5–10.

Lysen

, Darweesh

SKL

, Ikram

, Luik

, Ikram

(2019) Sleep and risk of parkinsonism and Parkinson’s disease: A population-based study. Brain 142, 2013–2022.

Politis

, Wu

, Molloy

, P

, Chaudhuri

, Piccini

(2010) Parkinson’s disease symptoms: The patient’s perspective. Mov Disord 25, 1646–1651.

Zea-Sevilla

, Martinez-Martin

(2014) Rating scales and questionnaires for assessment of sleep disorders in Parkinson’s disease: What they inform about? J Neural Transm (Vienna) 121(Suppl 1), S33–40.

10.

Cella

, Riley

, Stone

, Rothrock

, Reeve

, Yount

, Amtmann

, Bode

, Buysse

, Choi

, Cook

, Devellis

, DeWalt

, Fries

, Gershon

, Hahn

, Lai

, Pilkonis

, Revicki

, Rose

, Weinfurt

, Hays

(2010) The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. J Clin Epidemiol 63, 1179–1194.

11.

Buysse

, Yu

, Moul

, Germain

, Stover

, Dodds

, Johnston

, Shablesky-Cade

, Pilkonis

(2010) Development and validation of patient-reported outcome measures for sleep disturbance and sleep-related impairments. Sleep 33, 781–792.

12.

Forrest

, Meltzer

, Marcus

, de la Motte

, Kratchman

, Buysse

, Pilkonis

, Becker

, Bevans

(2018) Development and validation of the PROMIS Pediatric Sleep Disturbance and Sleep-Related Impairment item banks. Sleep 41, doi: 10.1093/sleep/zsy054

13.

Lei

, Yousaf

, Janmohamed

, Vakharia

, Chopra

, Sacotte

, Patel

, Singam

, Immaneni

, Kantor

, Hsu

, Silverberg

(2020) Validation of Patient-Reported Outcomes Information System Sleep Disturbance and Sleep-Related Impairment in adults with atopic dermatitis. Br J Dermatol 183, 875–882.

14.

Purvis

, Andreou

, Neuman

, Riley

3rd , Skolasky

(2017) Concurrent validity and responsiveness of PROMIS health domains among patients presenting for anterior cervical spine surgery. Spine (Phila Pa 1976) 42, E1357–E1365.

15.

Full

, Malhotra

, Crist

, Moran

, Kerr

(2019) Assessing psychometric properties of the PROMIS Sleep Disturbance Scale in older adults in independent-living and continuing care retirement communities. Sleep Health 5, 18–22.

16.

Shin

, Pohlig

, Habermann

(2018) Feasibility of using PROMIS((R)) in individuals with advanced Parkinson’s disease and their caregivers. Res Gerontol Nurs 11, 129–136.

17.

Kay

, Tanner

, Bowers

(2018) Sleep disturbances and depression severity in patients with Parkinson’s disease. Brain Behav 8, e00967.

18.

Hughes

, Daniel

, Ben-Shlomo

, Lees

(2002) The accuracy of diagnosis of parkinsonian syndromes in a specialist movement disorder service. Brain 125, 861–870.

19.

, Buysse

, Germain

, Moul

, Stover

, Dodds

, Johnston

, Pilkonis

(2011) Development of short forms from the PROMIS sleep disturbance and Sleep-Related Impairment item banks. Behav Sleep Med 10, 6–24.

20.

Bastien

, Vallieres

, Morin

(2001) Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med 2, 297–307.

21.

Mollayeva

, Thurairajah

, Burton

, Mollayeva

, Shapiro

, Colantonio

(2016) The Pittsburgh sleep quality index as a screening tool for sleep dysfunction in clinical and non-clinical samples: A systematic review and meta-analysis. Sleep Med Rev 25, 52–73.

22.

Kumar

, Bhatia

, Behari

(2003) Excessive daytime sleepiness in Parkinson’s disease as assessed by Epworth Sleepiness Scale (ESS). Sleep Med 4, 339–342.

23.

Spira

, Beaudreau

, Stone

, Kezirian

, Lui

, Redline

, Ancoli-Israel

, Ensrud

, Stewart

, Osteoporotic Fractures in Men S (2012) Reliability and validity of the Pittsburgh Sleep Quality Index and the Epworth Sleepiness Scale in older men. J Gerontol A Biol Sci Med Sci 67, 433–439.

24.

Stanley

, Beck

, Zebb

(1996) Psychometric properties of four anxiety measures in older adults. Behav Res Ther 34, 827–838.

25.

Schrag

, Barone

, Brown

, Leentjens

, McDonald

, Starkstein

, Weintraub

, Poewe

, Rascol

, Sampaio

, Stebbins

, Goetz

(2007) Depression rating scales in Parkinson’s disease: Critique and recommendations. Mov Disord 22, 1077–1092.

26.

Peto

, Jenkinson

, Fitzpatrick

(1998) PDQ-39: A review of the development, validation and application of a Parkinson’s disease quality of life questionnaire and its associated measures. J Neurol 245(Suppl 1), S10–14.

27.

Fisher

(1915) Frequency distribution of the values of the correlation coefficient in samples of an indefinitely large population. Biometrika 10, 507–521.

28.

Boggero

, Schneider

2nd , Thomas

, Nahman-Averbuch

, King

(2019) Associations of self-report and actigraphy sleep measures with experimental pain outcomes in patients with temporomandibular disorder and healthy controls. J Psychosom Res 123, 109730.

29.

Kwon

, Park

(2016) Interictal fatigue and its predictors in epilepsy patients: A case-control study. Seizure 34, 48–53.

30.

Buysse

, Thompson

, Scott

, Franzen

, Germain

, Hall

, Moul

, Nofzinger

, Kupfer

(2007) Daytime symptoms in primary insomnia: A prospective analysis using ecological momentary assessment. Sleep Med 8, 198–208.

31.

Drake

, Pillai

, Roth

(2014) Stress and sleep reactivity: A prospective investigation of the stress-diathesis model of insomnia. Sleep 37, 1295–1304.