Depressive Symptoms and Their Impact on Quality of Life in Parkinson’s Disease: An Exploratory Network Analysis Approach

Ci a ion: Heim ich, K.G.; Mendo , S.; Schönenbe g, A.; San os-Ga cía, D.; Mi , P.; COPPADIS S udy G oup; P ell, T. Dep essi e Symp oms and Thei Impac on Quali y o Li e in Pa kinson’s Disease: An Explo a o y Ne wo k Analysis App oach. J. Clin. Med. 2023,12, 4616. h ps:// doi.o g/10.3390/jcm12144616 Academic Edi o s: Alejand o Rod íguez-Moline o and Jussi Sipilä Recei ed: 19 June 2023 Re ised: 4 July 2023 Accep ed: 8 July 2023 Published: 11 July 2023 Copy igh : © 2023 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion (CC BY) license (h ps:// c ea i ecommons.o g/licenses/by/ 4.0/). Jou nal o Clinical Medicine A icle Dep essi e Symp oms and Thei Impac on Quali y o Li e in Pa kinson’s Disease: An Explo a o y Ne wo k Analysis App oach Kons an in G. Heim ich 1,*,† , Sa ah Mendo 1,† , Aline Schönenbe g 2, Diego San os-Ga cía3, Pablo Mi 4,5, COPPADIS S udy G oup 6,‡ and Tino P ell 2 1Depa men o Neu ology, Jena Uni e si y Hospi al, Am Klinikum 1, 07747 Jena, Ge many 2Depa men o Ge ia ics, Halle Uni e si y Hospi al, E ns -G ube-S aße 40, 06120 Halle, Ge many 3Depa men o Neu ology, CHUAC (Complejo Hospi ala io Uni e si a io de A Co uña), c/As Xubias 84, 15006 A Co uña, Spain 4Unidad de T as o nos del Mo imien o, Se icio de Neu ología y Neu o isiología Clínica, Ins i u o de Biomedicina de Se illa, Hospi al Uni e si a io Vi gen del Rocío/CSIC/Uni e sidad de Se illa, 41013 Se ille, Spain 5Cen o de In es igación Biomédica en Red Sob e En e medades Neu odegene a i as (CIBERNED), 28031 Mad id, Spain 6Fundación Española de Ayuda a la In es igación en En e medades Neu odegene a i as y/o de O igen Gené ico, Calle An onio J de Suc e 1A, 15179 Olei os, Spain *Co espondence: [email p o ec ed] †These au ho s con ibu ed equally o his s udy. ‡Membe ship o he COPPADIS S udy G oup is p o ided in he Appendix A. Abs ac : The clinical p esen a ion o Pa kinson’s disease (PD) is o en domina ed by dep essi e symp oms, which can signi ican ly impac he pa ien s’ quali y o li e (QoL). Howe e , i is no clea how hese dep essi e symp oms a e in e connec ed, o i some symp oms a e mo e in luen ial in a ec ing QoL. In he Coho o Pa ien s wi h Pa kinson’s Disease in Spain (COPPADIS) s udy, 686 pa ien s wi h PD we e analyzed using ne wo k analyses. The pa ien s comple ed he Beck Dep ession In en o y II (BDI-II) and p o ided hei o e all QoL (EUROHIS-QOL) a he beginning o he s udy. The s udy used cen ali y measu es such as Expec ed In luence and B idge Expec ed In luence o iden i y dep essi e symp oms ha had he g ea es impac on o e all QoL. The esul s o explo a o y ne wo k analyses indica e ha he BDI-II i ems ela ed o loss o ene gy,pas ailu e, and i edness o a igue ha e he g ea es impac on o e all QoL as measu ed by he EUROHIS-QOL 8-i em index. The loss o ene gy and i edness o a igue BDI-II i ems a e also s ongly associa ed wi h a numbe o di e en EUROHIS-QOL i ems, acco ding o B idge Expec ed In luences. Fo indi iduals su e ing om PD, ne wo k analysis can aid in iden i ying signi ican non-mo o symp oms ha impac hei QoL, hus pa ing he way o po en ial imp o emen s. Keywo ds: Pa kinson’s disease; quali y o li e; BDI-II; dep ession; a igue; ne wo k analysis 1. In oduc ion Pa kinson’s disease (PD) is one o he mos common p og essi e neu odegene a i e diso de s and cha ac e ized by mo o and a ious non-mo o symp oms [ 1 ]. Dep ession is a c ucial ac o ha de e mines he quali y o li e (QoL) in indi iduals wi h PD, and i is a pa icula ly signi ican non-mo o symp om [ 2 – 5 ]. The o e all QoL and heal h- ela ed QoL a e c i ical ou comes o heal hca e, and hey a e impo an p edic o s o mo bidi y and mo ali y [ 6 , 7 ]. The e o e, iden i ying and ea ing dep ession is essen ial o main ain QoL in indi iduals wi h PD. Howe e , jus a small numbe o indi iduals wi h PD who epo dep essi e symp oms ul ill he diagnos ic c i e ia o a majo dep essi e diso de as de ined by he Diagnos ic and S a is ical Manual o Men al Diso de s (DSM) [ 8 , 9 ]. J. Clin. Med. 2023,12, 4616. h ps://doi.o g/10.3390/jcm12144616 h ps://www.mdpi.com/jou nal/jcm J. Clin. Med. 2023,12, 4616 2 o 13 None heless, e en sub h eshold dep essi e symp oms can nega i ely a ec he QoL o indi iduals wi h PD [ 10 ]. To p o ide ailo ed ea men o indi iduals wi h PD and dep essi e symp oms, i is impo an o de e mine which dep essi e symp oms ha e he g ea es impac on gene al QoL and di e en domains o QoL. To add ess his inqui y, one mus con on a ious me hodological challenges. Measu - ing and diagnosing dep ession in PD p esen s di icul ies since he e a e mul iple dis inc measu es a ailable. Diagnosis and hus p e alence may a y depending on he de ini ion o dep essi e diso de s, e.g., majo dep ession acco ding o DSM c i e ia o dep essi e episode acco ding o he In e na ional Classi ica ion o Diseases (ICD), as well as a ious sel - epo o clinician- a ed psychia ic symp om a ing scales [ 9 , 11 ]. In addi ion, he e is signi ican o e lap be ween dep ession and o he non-mo o symp oms o PD [ 7 , 12 , 13 ]. Addi ionally, di e en dep essi e symp oms may ha e he same o opposi e e ec on QoL, making i challenging o de e mine he one-di ec ional e ec o hese symp oms. Consequen ly, con en ional s a is ical me hods may no cap u e he in ica e in e play be ween dep essi e symp oms and QoL measu es. This p esen s an exci ing scien i ic challenge ha equi es ca e ul conside a ion. The me hod o ne wo k analysis is a p omising way o model in e ac ions be ween a la ge numbe o a iables, which is pa icula ly impo an o he s udy o men al heal h p oblems [ 14 ]. Unlike classical eg ession modeling, which educes he s uc u e o a iables o hei sha ed in o ma ion, ne wo k analysis di ec ly es ima es he associa ions be ween all a iables [ 15 , 16 ]. This explo a o y app oach allows o he isualiza ion o he ela ionships be ween mul iple a iables wi hou he assump ion o a di ec ion o e ec s. In his s udy, we implemen ed ne wo k analysis o e eal he complex in e ac i e ela ionship be ween dep essi e symp oms and QoL in people wi h PD (PwPD). 2. Ma e ials and Me hods 2.1. S udy Design and Pa icipan s We used da a om he na ional, mul icen e , and longi udinal Coho o Pa ien s wi h Pa kinson’s Disease in Spain (COPPADIS) s udy [ 17 ]. Non-demen ed PwPD be ween 30 and 75 yea s we e ec ui ed om Janua y 2016 o No embe 2017. Mo e de ailed in o ma ion ega ding he s udy design, con en , and exclusion c i e ia can be ound in he COPPADIS s udy p o ocol [ 17 ]. We selec ed PwPD o whom measu es o dep ession and QoL we e a ailable a baseline (N = 686). 2.2. Ex ac ed Va iables Fo measu emen o dep essi e symp oms, he Beck Dep ession In en o y II (BDI-II) was used. The BDI-II consis s o 21 i ems (each a ed on a 4-poin Like scale anging om ze o o h ee). The summed o al sco e anges om 0 o 63 poin s, wi h highe alues indica ing mo e dep essi e symp oms [ 18 ]. In ou ne wo k analysis, we e ained om classi ying pa ien s in o wo ca ego ies o ha ing o no ha ing dep ession. We el ha e en he p esence o subsynd omal dep ession o some dep essi e symp oms could impac he QoL. Howe e , o desc ip i e s a is ics, we elied on he es ablished BDI h eshold o iden i y pa ien s wi h dep ession. QoL was assessed wi h he EUROHIS-QOL [ 19 ]. The EUROHIS-QOL was de i ed om he Wo ld Heal h O ganiza ion Quali y o Li e assessmen [ 20 ]. I consis s o eigh i ems gi en on a 5-poin Like scale and co e s psychological, physical, social, and en i- onmen al domains o QoL [ 19 ]: How would you a e you quali y o li e? (QOL); How sa is ied a e you wi h you heal h? (HEA); Do you ha e enough ene gy o e e yday li e? (ENE); How sa is ied a e you wi h you abili y o pe o m you daily li ing ac i i ies? (ACT); How sa is ied a e you wi h you sel ? (YOU); How sa is ied a e you wi h you pe sonal ela ionships? (REL); Ha e you enough money o mee you needs? (MON); How sa is ied a e you wi h he condi ions o you li ing place? (LIV). An index o o e all QoL (EUROHIS-QOL 8-i em index) is calcula ed by summa ion o he sco es o e e y i em, whe eby highe alues indica e a be e QoL [21]. J. Clin. Med. 2023,12, 4616 3 o 13 Fu he mo e, he ollowing a iables we e ob ained o desc ibe he coho : age, sex, Hoehn and Yah s age [ 22 ], Uni ied Pa kinson’s Disease Ra ing Scale (UPDRS) pa s III and IV [ 23 ], o al sco es o he Non-Mo o Symp oms Scale in Pa kinson’s disease (NMSS) [ 24 ], and he Mini-Men al S a e Examina ion (MMSE) [25]. 2.3. S a is ical Analyses Fo s a is ical analyses, we used R ( e sion 4.2.1, R Founda ion o S a is ical Com- pu ing, Vienna, Aus ia), SPSS (IBM SPSS S a is ics 27, IBM, A monk, NY, USA), and JASP ( e sion 0.15, JASP Team, Ams e dam, The Ne he lands). Fo he cha ac e iza ion o he coho , desc ip i e s a is ics we e applied. Da a we e es ed o no mali y by using he Shapi o–Wilk es . Fo non-no mally dis ibu ed da a, he median and in e qua ile ange we e de e mined. The le el o s a is ical signi icance o all es s was se a p< 0.05 ( wo- ailed). Explo a o y ne wo k analyses we e conduc ed o explo e he associa ions be ween he 21 BDI-II i ems and he EUROHIS-QOL. In a ne wo k, he whole complex in e ac ing sys em be ween a ious symp oms is used o unde s and hei connec ions. A egula iza ion echnique was used o p e en o e i ing he s uc u e o he ne wo k [ 26 ], called he ex ended Bayesian in o ma ion c i e ion (EBIC) [ 27 , 28 ] wi h he leas absolu e sh inkage and selec ion ope a o (LASSO) [ 29 ]. Due o he o dinal s uc u e o he da a, polycho ic co ela ions we e es ima ed [ 30 ]. The uning pa ame e o EBICglasso was se o 0.5 o allow mo e sensi i e and speci ic ne wo k analysis. The i ems o he ques ionnai es a e displayed by he nodes o he ne wo k and posi ioned by he F uch e man–Reingold algo i hm [ 31 ]. The connec ions be ween nodes a e called edges and hei hickness indica es he s eng h o he co ela ions. Co ela ions we e classi ied as low (| |= 0.1), mode a e (| | = 0.3), o s ong (| | = 0.5) [ 32 ]. Blue edges e e o a posi i e co ela ion, and ed edges o a nega i e one. To desc ibe he ne wo k, he cen ali y measu e Expec ed In luence was de e mined by ela i e alues. The Expec ed In luence o a node is de ined as he sum o he absolu e edge weigh s ha a e connec ed o ha node, aking in o accoun posi i e and nega i e edges [ 33 ]. Addi ionally, he cen ali y measu e B idge Expec ed In luence (1-s ep) was de e mined using ela i e alues. B idge Expec ed In luence e e s o he sum o he alue o all edges ha exis be ween a node o one communi y (i.e., a dep essi e symp om) and all nodes o ano he communi y (i.e., he eigh i ems o he EUROHIS-QOL) [34]. Mo eo e , we de e mined nodewise p edic abili y o de e mine how well a gi en node o a ne wo k (i.e., he EUROHIS-QOL 8-i em index) is p edic able by all nodes di ec ly connec ed o i (i.e., associa ed dep essi e symp oms) [ 35 ]. The ob ained explained a iance R 2 can be be ween 0 and 1, and alues ≥ 0.13 a e de ined as mode a e and ≥ 0.26 as high [ 32 ]. Ne wo k s abili y was assessed by means o he co ela ion s abili y (CS) coe icien (numbe o case-d opping boo s aps = 1000). The CS coe icien quan i ies he p opo ion o cases which can be omi ed o s ill main ain a co ela ion wi h he o iginal cen ali y measu e ha is a minimum 0.7 in a leas 95% o samples [ 16 ]. The CS coe icien should be in gene al abo e 0.25 and p e e able abo e 0.5 [16]. 3. Resul s 3.1. Desc ip i e Analyses Desc ip i e s a is ics o he s udy popula ion a e shown in Table 1. O he 686 PwPD, 274 (39.9%) we e emale and 412 (60.1%) we e male. Pa ien s had a median age o 64 yea s (IQR = 57–70 yea s), and a median du a ion o he disease o i e yea s (IQR = 2–8 yea s). Mos pa ien s had a disease s age wi h bila e al in ol emen (Hoehn and Yah s age ≥ 2), and mode a e mo o impai men (median UPDRS III: 21 poin s; IQR = 14–30). F equen ly, hey expe ienced one mo o complica ion (median UPDRS IV: 1 poin ; IQR = 0–3). Pa ien s p esen ed non-mo o symp oms acco ding o he NMSS wi h a median o al sco e o 35 poin s (IQR = 19–61). The median BDI-II o al sco e was 7 poin s (IQR = 3–13). Using J. Clin. Med. 2023,12, 4616 4 o 13 he p e iously desc ibed BDI-II cu -o s, 16.2% (N = 111) had dep ession, 27.0% (N = 185) had sub h eshold dep ession, and 56.9% (N = 390) had no dep ession [36]. Table 1. Desc ip i e s a is ics (N = 686). Va iable Age 64 (57–70) Disease du a ion 5 (2–8) HY o 2 (2–2) UPDRS III o 21 (14–30) UPDRS IV o 1 (0–3) NMSS, o al sco e 35 (19–61) MMSE, o al sco e 30 (29–30) BDI-II, o al sco e 7 (3–13) 1. Sadness 0 (0–1) 2. Pessimism 0 (0–1) 3. Pas ailu e 0 (0–0) 4. Loss o pleasu e 0 (0–1) 5. Guil y eelings 0 (0–0) 6. Punishmen eelings 0 (0–0) 7. Sel -dislike 0 (0–0) 8. Sel -c i icalness 0 (0–1) 9. Suicidal hough s o wishes 0 (0–0) 10. C ying 0 (0–1) 11. Agi a ion 0 (0–1) 12. Loss o in e es 0 (0–1) 13. Indecisi eness 0 (0–1) 14. Wo hlessness 0 (0–1) 15. Loss o ene gy 1 (0–1) 16. Changes in sleeping pa e n 1 (0–2) 17. I i abili y 0 (0–1) 18. Changes in appe i e 0 (0–1) 19. Concen a ion di icul y 1 (0–1) 20. Ti edness o a igue 1 (0–1) 21. Loss o in e es in sex 0 (0–1) EUROHIS-QOL 8-i em index 31 (28–33) 1. QOL, quali y 4 (3–4) 2. HEA, heal h 3 (3–4) 3. ENE, ene gy 4 (3–4) 4. ACT, ac i i ies 4 (3–4) 5. YOU, you sel 4 (3–4) 6. REL, ela ionships 4 (4–4) 7. MON, money 4 (3–4) 8. LIV, li ing 4 (4–5) Values a e gi en as he medians and in e qua ile anges. BDI-II: e ised Beck Dep ession In en o y; EUROHIS- QOL: Eu opean Union Heal h In e iew Su ey o Quali y O Li e (QOL: How would you a e you quali y o li e?; HEA: How sa is ied a e you wi h you heal h?; ENE: Do you ha e enough ene gy o e e yday li e?; ACT: How sa is ied a e you wi h you abili y o pe o m you daily li ing ac i i ies?; YOU: How sa is ied a e you wi h you sel ?; REL: How sa is ied a e you wi h you pe sonal ela ionships?; MON: Ha e you enough money o mee you needs?; LIV: How sa is ied a e you wi h he condi ions o you li ing place?); HY: Hoehn and Yah s age; MMSE: Mini-Men al S a e Examina ion; N: numbe o pa icipan s; NMSS: Non-Mo o Symp oms Scale in Pa kinson’s Disease; UPDRS: Uni ied Pa kinson’s Disease Ra ing Scale. 3.2. Ne wo k Analyses To explo e he links be ween dep essi e symp oms (21 BDI-II i ems) and EUROHIS- QOL, ne wo k analyses we e conduc ed. The s udy examined wo models: he i s model included he EUROHIS-QOL 8-i em index, while he second model included all eigh i ems o he EUROHIS-QOL. J. Clin. Med. 2023,12, 4616 5 o 13 3.2.1. Ne wo k Model 1: Associa ion be ween BDI-II I ems and O e all QoL (EUROHIS-QOL 8-I em Index) The ne wo k plo o model 1 is shown in Figu e 1. The blue nodes display he i ems o he BDI-II (b1–b21), and he o ange node displays he EUROHIS-QOL 8-i em index (QOL8) as a measu e o o e all QoL. Figu e 1. Ne wo k s uc u e BDI-II i ems and EUROHIS-QOL 8-i em index (model 1). The blue nodes display he i ems o he BDI-II (b1–b21), and he o ange node displays he summed EUROHIS-QOL 8-i em index (QOL8). The edges display he co ela ions be ween he nodes. Blue edges ep esen posi i e associa ions, and ed edges ep esen nega i e associa ions. The hickness o he edges indica e how s ong hese connec ions a e. BDI-II: e ised Beck Dep ession In en o y (b1: Sadness; b2: Pessimism; b3: Pas ailu e; b4: Loss o pleasu e; b5: Guil y eelings; b6: Punishmen eelings; b7: Sel -dislike; b8: Sel -c i icalness; b9: Suicidal hough s o wishes; b10: C ying; b11: Agi a ion; b12: Loss o in e es ; b13: Indecisi eness; b14: Wo hlessness; b15: Loss o ene gy; b16: Changes in sleeping pa e n; b17: I i abili y; b18: Changes in appe i e; b19: Concen a ion di icul y; b20: Ti ed- ness o a igue; b21: Loss o in e es in sex); EUROHIS-QOL: Eu opean Union Heal h In e iew Su ey o Quali y O Li e. On a global le el, he ne wo k was well-connec ed (137 o 231 non-ze o edges) wi hou isola ed nodes. The BDI-II and QOL8 nodes had nume ous in e ac ions, p ima ily nega i e ones (indica ed by ed edges). This means ha dep essi e symp oms we e linked o poo e QoL. The s onges nega i e connec ions we e obse ed be ween QOL8–loss o ene gy (b15), QOL8–pas ailu e (b3), and QOL8– i edness o a igue (b20), wi h edge weigh s de ailed in Table S1. Expec ed In luence was de e mined o each node (see Figu e 2, and Table S1). A high Expec ed In luence means ha changing he alue o his node can ha e a apid e ec on o he nodes wi hin he ne wo k. He e, QOL8 had he s onges nega i e Expec ed In luence. Acco dingly, QOL8 had he highes nega i e inpu weigh s om o he nodes ha a e di ec ly connec ed. The 13 nodes ha we e di ec ly connec ed o QOL8 (sadness, pessimism,pas ailu e,loss o pleasu e,sel -dislike,suicidal hough s o wishes,c ying,agi a ion, loss o in e es ,indecisi eness,loss o ene gy,changes in appe i e, i edness o a igue) explained 42.8% o QOL8 a iance, as e ealed by nodewise p edic abili y analysis (see Table S1). J. Clin. Med. 2023,12, 4616 6 o 13 Figu e 2. Expec ed In luence BDI-II i ems and EUROHIS-QOL 8-i em index (model 1). Expec ed In luence cen ali y measu es o he i ems o he BDI-II (b1–b21) and he EUROHIS-QOL 8-i em index (QOL8) a e gi en in ela i e alues. BDI-II: e ised Beck Dep ession In en o y (b1: Sadness; b2: Pessimism; b3: Pas ailu e; b4: Loss o pleasu e; b5: Guil y eelings; b6: Punishmen eelings; b7: Sel -dislike; b8: Sel -c i icalness; b9: Suicidal hough s o wishes; b10: C ying; b11: Agi a ion; b12: Loss o in e es ; b13: Indecisi eness; b14: Wo hlessness; b15: Loss o ene gy; b16: Changes in sleeping pa e n; b17: I i abili y; b18: Changes in appe i e; b19: Concen a ion di icul y; b20: Ti ed- ness o a igue; b21: Loss o in e es in sex); EUROHIS-QOL: Eu opean Union Heal h In e iew Su ey o Quali y O Li e. Acco ding o he case-d opping boo s apped p ocedu e, he ne wo k can be consid- e ed s able as he CS coe icien o Expec ed In luence emained high (CS(co = 0.7) = 0.67) (see Figu e S1). 3.2.2. Ne wo k Model 2: Associa ion be ween BDI-II I ems and he Eigh EUROHIS-QOL I ems In he second ne wo k, we analyzed he associa ion be ween BDI-II i ems and he dis inc EUROHIS-QOL i ems. The ne wo k plo o model 2 is shown in Figu e 3. The blue nodes ep esen he BDI-II i ems (b1–b21) and he o ange nodes ep esen he eigh i ems o he EUROHIS-QOL. Again, ne wo k analysis demons a ed a well-connec ed ne wo k (183 o 406 non-ze o edges) wi hou isola ed nodes. One can dis inguish wo pa s e lec ing he wo di e en ques ionnai es. Howe e , he e we e many in e ac ions be ween he nodes o he BDI-II and he EUROHIS-QOL. These in e ac ions we e p ima ily nega i e, as depic ed by he ed edges. This implies ha a lowe in ensi y o dep essi e symp oms is associa ed wi h a highe QoL. The s onges nega i e connec ion be ween he wo pa s was obse ed be- ween “ i edness o a igue” and “Do you ha e enough ene gy o e e yday li e?” (b20 and ENE), which is no su p ising as bo h a e simila cons uc s. Howe e , iden i ying he de- p essi e symp oms ha a e linked wi h all eigh i ems on he EUROHIS-QOL ques ionnai e is a ma e o g ea in e es . This is why B idge Expec ed In luences we e de e mined. The B idge Expec ed In luence cen ali y measu es a e shown in Figu e 4(and also abula ed in Table S2). The nodes b15 (loss o ene gy) and b20 ( i edness o a igue) we e ound o ha e he highes B idge Expec ed In luences, sugges ing ha hey a e he mos s ongly J. Clin. Med. 2023,12, 4616 7 o 13 associa ed dep essi e symp oms wi h all eigh i ems on he EUROHIS-QOL ques ionnai e. The eby, node b15 is associa ed in pa icula wi h ACT (How sa is ied a e you wi h you abili y o pe o m you daily li ing ac i i ies?) (edge weigh : − 0.102). This means ha eeling loss o ene gy has g ea es in luence on he abili y o pe o m daily li ing ac i i ies among he QoL domains. Mo eo e , node b20 showed an associa ion wi h ENE (Do you ha e enough ene gy o e e yday li e?) o he quali y-o -li e domain (edge weigh : − 0.236). Respec i ely, pa ien s who epo ed i edness o a igue had less ene gy o e e yday li e. As e ealed by he ne wo k plo o model 2 (Figu e 3), hese wo dep essi e symp oms wi h he highes B idge Expec ed In luence (loss o ene gy and i edness o a igue) we e closely connec ed o each o he . Loss o pleasu e (b4) was he hi d in luen ial dep essi e symp om, which is especially associa ed wi h QOL (How would you a e you quali y o li e?; edge weigh : − 0.063) and REL (How sa is ied a e you wi h you pe sonal ela ionships?; edge weigh : −0.053). Figu e 3. Ne wo k s uc u e BDI-II i ems and EUROHIS-QOL i ems (model 2). The blue nodes display he i ems o he BDI-II (b1–b21), and he o ange nodes display he eigh i ems o he EUROHIS-QOL. The edges display he co ela ions be ween he nodes., Blue edges ep esen posi i e associa ions, and ed edges ep esen nega i e associa ions. The hickness o he edges indica e how s ong hese connec ions a e. BDI-II: e ised Beck Dep ession In en o y (b1: Sadness; b2: Pessimism; b3: Pas ailu e; b4: Loss o pleasu e; b5: Guil y eelings; b6: Punishmen eelings; b7: Sel -dislike; b8: Sel -c i icalness; b9: Suicidal hough s o wishes; b10: C ying; b11: Agi a ion; b12: Loss o in e es ; b13: Indecisi eness; b14: Wo hlessness; b15: Loss o ene gy; b16: Changes in sleeping pa e n; b17: I i abili y; b18: Changes in appe i e; b19: Concen a ion di icul y; b20: Ti edness o a igue; b21: Loss o in e es in sex); EUROHIS-QOL: Eu opean Union Heal h In e iew Su ey o Quali y O Li e (QOL: How would you a e you quali y o li e?; HEA: How sa is ied a e you wi h you heal h?; ENE: Do you ha e enough ene gy o e e yday li e?; ACT: How sa is ied a e you wi h you abili y o pe o m you daily li ing ac i i ies?; YOU: How sa is ied a e you wi h you sel ?; REL: How sa is ied a e you wi h you pe sonal ela ionships?; MON: Ha e you enough money o mee you needs?; LIV: How sa is ied a e you wi h he condi ions o you li ing place?). J. Clin. Med. 2023,12, 4616 8 o 13 Figu e 4. B idge Expec ed In luence BDI-II i ems and EUROHIS-QOL i ems (model 2). B idge Expec ed In luence cen ali y measu es o he i ems o he BDI-II (b1–b21) and he eigh i ems o he EUROHIS- QOL a e gi en in ela i e alues. BDI-II: e ised Beck Dep ession In en o y (b1: Sadness; b2: Pessimism; b3: Pas ailu e; b4: Loss o pleasu e; b5: Guil y eelings; b6: Punishmen eelings; b7: Sel -dislike; b8: Sel -c i icalness; b9: Suicidal hough s o wishes; b10: C ying; b11: Agi a ion; b12: Loss o in e - es ; b13: Indecisi eness; b14: Wo hlessness; b15: Loss o ene gy; b16: Changes in sleeping pa e n; b17: I i abili y; b18: Changes in appe i e; b19: Concen a ion di icul y; b20: Ti edness o a igue; b21: Loss o in e es in sex); EUROHIS-QOL: Eu opean Union Heal h In e iew Su ey o Quali y O Li e (QOL: How would you a e you quali y o li e?; HEA: How sa is ied a e you wi h you heal h?; ENE: Do you ha e enough ene gy o e e yday li e?; ACT: How sa is ied a e you wi h you abili y o pe o m you daily li ing ac i i ies?; YOU: How sa is ied a e you wi h you sel ?; REL: How sa is ied a e you wi h you pe sonal ela ionships?; MON: Ha e you enough money o mee you needs?; LIV: How sa is ied a e you wi h he condi ions o you li ing place?). Acco ding o he case-d opping boo s apped p ocedu e, he ne wo k can be conside ed s able as he CS coe icien o B idge Expec ed In luence emained high ( CS(co = 0.7) = 0.60 ) (see Figu e S2). 4. Discussion Ou esea ch u ilized ne wo k analysis o unco e in ica e ela ionships be ween dep essi e symp oms and QoL in PwPD. Ou indings con i med ha he e is a s ong co ela ion be ween he 21 i ems o he BDI-II and EUROHIS-QOL, whe he we conside ed he EUROHIS-QOL 8-i em index (model 1) o all eigh i ems oge he (model 2). These co ela ions a e la gely nega i e, indica ing ha lowe le els o dep essi e symp oms a e associa ed wi h highe QoL. Speci ically, he wo BDI-II i ems ela ed o loss o ene gy (b15) and i edness o a igue (b20) had he mos signi ican impac on QoL. Fu he mo e, ou ne wo k analysis e ealed ha hese wo BDI-II i ems (b15 and b20) we e closely linked o each o he . J. Clin. Med. 2023,12, 4616 9 o 13 When e alua ing he impac o dep essi e symp oms on QoL using he EUROHIS- QOL 8-i em index, ou ne wo k analysis (model 1) ound ha eelings o loss o ene gy (b15), pas ailu e (b3), and i edness o a igue (b20) we e in luen ial symp oms. Conside ing he impac o dep essi e symp oms on all eigh i ems o he EUROHIS-QOL (model 2), ne wo k analysis e ealed nega i e associa ions be ween QoL and especially loss o ene gy (b15) and i edness o a igue (b20). Taking in o accoun he complex in e ac ions o all 21 BDI-II i ems and he eigh QoL measu es, ne wo k analysis demons a ed ha bo h symp oms a e closely connec ed o each o he . In addi ion, loss o pleasu e (b4) was iden i ied as he hi d in luen ial dep essi e symp om as e ealed by B idge Expec ed In luence. In summa y, he mos in luen ial dep essi e symp oms a ec ing bo h he EUROHIS- QOL 8-i em index and all eigh EUROHIS-QOL i ems oge he we e loss o ene gy (b15) and eelings o i edness o a igue (b20). O e all, his s udy sheds ligh on he impo ance o conside ing he complex in e play be ween dep essi e symp oms and QoL measu es. Fa igue is gene ally desc ibed as an all-encompassing eeling o i edness, dec eased ene gy, and o en a sense o comple e deple ion. I should no be con used wi h symp oms o dep ession, such as eelings o wo hlessness, despai , o hopelessness, al hough i can be a sign o dep ession. Addi ionally, i is no equi alen o limb weakness o any isible sign o physical weakness [ 37 ]. The de ini ion indica es ha loss o ene gy may be pa o he clinical pic u e o a igue. Howe e , hese symp oms a e conside ed sepa a e componen s o he BDI-II since no all indi iduals wi h loss o ene gy expe ience es ablished a igue. Ne e heless, ou esea ch has e ealed ha bo h aspec s a e closely linked o one ano he . One o he mos p e alen and disabling non-mo o symp oms in PwPD is a igue [ 38 ], which can mani es e en in he ea ly s ages o he disease [ 39 ] and o en pe sis s o wo sens o e ime [ 40 , 41 ]. This can lead o dec eased pa icipa ion in social and ec ea ional ac i i- ies [ 42 , 43 ], nega i ely impac ing he pa ien s’ QoL [ 38 , 44 ]. This was also demons a ed in his s udy, in which bo h BDI-II i ems (loss o ene gy and i edness o a igue) had a s ong in luence on QoL. The e o e, iden i ica ion and ea men o a igue in PD appea s o be p omising o imp o e pa ien s’ QoL. Howe e , he e a e cu en ly no speci ic guidelines o managing PD- ela ed a igue, and esea ch has no p o ided enough e idence o suppo he use o pha maceu ical o non-pha maceu ical ea men s [ 45 , 46 ]. The p ima y and essen ial s ep in managing a igue in PD is o educa e pa ien s and hei amilies abou i s common occu ence. Ano he BDI-II i em ha has g ea in luence on QoL was pas ailu e (b3). The ex- pe ience o pas ailu es can a ec QoL [ 47 , 48 ], leading o nega i e psychological ou - comes [ 49 , 50 ]. Indi iduals who ha e expe ienced se backs, ailu es, o nega i e li e e en s such as di o ce o job loss, may a ibu e blame o hemsel es and expe ience shame due o hei inabili y o mee hei own o o he s’ expec a ions in a ious aspec s o li e [ 51 , 52 ]. These eelings o shame can also con ibu e o a lowe QoL and inc eased suscep ibili y o psychopa hological symp oms [53,54]. Mo eo e , he p esence o loss o pleasu e (b4), also known as anhedonia, is o cen al impo ance wi hin he ne wo k. Anhedonia is seen as he inabili y o eel pleasu e om ac- i i ies o expe iences ha a e no mally enjoyable o ewa ding [ 55 ]. The e o e, anhedonia can ha e a signi ican impac on QoL [56,57]. Using ne wo k analyses, ou p e ious s udy e ealed he impac o dep essi e symp- oms in PwPD [ 58 , 59 ]. I was demons a ed ha in pa icula eelings o sadness a e impo an wi hin he complex in e ac ing ne wo k o non-mo o symp oms in PD, as as- sessed by he NMSS [ 58 ]. Fu he mo e, a igue was iden i ied as non-mo o symp om ha is mos s ongly associa ed wi h heal h- ela ed QoL, as assessed by he Pa kinson’s Disease Ques ionnai e 39 (PDQ-39) summa y index and in pa icula he mobili y and ac i i ies o daily li ing subscales o he PDQ-39 [ 59 ]. These indings demons a e ha dep ession is a he e ogenous cons uc , and dep essi e symp oms o en o e lap wi h o he non-mo o symp oms in PD [60]. Ou s udy has shown ha he loss o ene gy and cons an a igue a e c ucial ac o s in he ne wo k o dep essi e symp oms and QoL measu es. These symp oms a e ca ego ized