Journal of Neurology (2022) 269:3541–3549 https://doi.org/10.1007/s00415-021-10955-2 ORIGINAL COMMUNICATION Measuring the functional impact of cognitive impairment in Huntington’s disease Andrea Horta‑Barba1,2,3,4,5 · Saül Martínez‑Horta1,2,3,4,5 · Jesus Pérez‑Pérez1,2,3,4,5 · Frederic Sampedro1,3 · Arnau Puig‑Davi1,3 · Javier Pagonabarraga1,2,3,4 · Jaime Kulisevsky1,2,3,4,5 Received: 1 September 2021 / Revised: 28 December 2021 / Accepted: 29 December 2021 / Published online: 21 January 2022 © The Author(s) 2022 Abstract Background Patients with Huntington’s disease (HD) exhibit a variable predominance of cognitive, behavioral and motor symptoms. A specific instrument focusing on the impact of cognitive impairment in HD over functional capacity is lacking. Objective To address the need for a brief and specifically developed HD questionnaire able to capture functional aspects suspected to be sensitive to cognitive impairment. Methods We developed and validated the “Huntington’s Disease-Cognitive Functional Rating Scale” (HD-CFRS) in 78 symptomatic carriers of the Huntington’s disease mutation. We also administered the HD-CFRS to a knowledgeable informant to measure the level of agreement. To explore the association between HD-CFRS scores and participants’ cognitive status, we administered objective measures of cognition. Participants were classified as cognitively preserved (HD-NC), as having mild cognitive impairment (HD-MCI), or as having dementia (HD-Dem). Results The HD-CFRS showed concurrent validity and internal consistency in the three groups. HD carriers and informants in the HD-NC group obtained similar HD-CFRS scores. However, in patients with mild cognitive impairment and dementia, informers reported greater functional impairment than HD participants. The HD-CFRS total score showed strong correla- tions with measures assessing cognition. Conclusions These findings support the utility of the HD-CFRS as a brief and reliable instrument to measure functional defects associated with cognitive impairment in HD. We believe this questionnaire could be a useful tool both for clinical practice and research. Keywords Huntington’s disease · Cognition · Functional assessment · Functionality · Cognitive impairment Introduction Huntington’s disease (HD) is an autosomal-dominant monogenetic neurodegenerative disease caused by a cyto- sine–adenine–guanine (CAG) trinucleotide repeat expansion * Jaime Kulisevsky jkulisevsky@santpau.cat in the HTT gene [1]. Clinically, HD is characterized by a complex constellation of progressive motor abnormalities, 1 Movement Disorders Unit, Neurology Department, cognitive decline and neuropsychiatric symptoms starting Hospital de la Santa Creu i Sant Pau, Mas Casanovas 90, around mid-adulthood [2]. In association with the worsen- 08041 Barcelona, Spain ing of HD symptoms, all patients experience a progressive 2 Department of Medicine, Universitat Autònoma de Barcelona loss of functional independence [3]. The clinical stages of (U.A.B.), Barcelona, Spain HD are defined on the basis of the degree of independence 3 Institut d’Investigacions Biomèdiques-Sant Pau (IIB-Sant in functional capacity [4]. In HD, functionality is commonly Pau), Barcelona, Spain addressed using the Total Functional Capacity (TFC) score 4 Centro de Investigación en Red-Enfermedades and the Functional Assessment Scale (FAS) from the func- Neurodegenerativas (CIBERNED), Madrid, Spain tional assessment of the Unified Huntington’s Disease Rating 5 European Huntington’s Disease Network (EHDN), Ulm, Scale (UHDRS) [5]. The TFC rates the level of functional Germany 13 Vol.:(0123456789) 3542 Journal of Neurology (2022) 269:3541–3549 independence in five domains: occupation, finances, domes- considered to be affected by cognitive impairment but not tic chores, activities of daily living (ADLs), and care. Each by motor and neuropsychiatric symptoms. It is composed of of these domains is rated based on the patient’s capacity 12 items that determine the degree of difficulty in perform- to independently perform these activities. The FAS scale ing activities of daily living involving dual tasking, sustained is based on 25 yes/no questions and qualifies abilities to attention, planning and organization, problem solving, ver- independently perform ADLs. Given that both the TFC and bal expression, comprehension, temporal and spatial orienta- the FAS provide a global measure of patients’ functional tion, and memory. capacity, these instruments have been extensively used as The main aims of the present study were to study the psy- outcome measures and classification parameters in clinical chometric properties of the HD-CFRS so as to provide the trials and in other research settings [6, 7]. HD community with a specific screening method to capture Patients with HD present a variable predominance and the impact of cognitive changes on daily living functional- severity of cognitive, motor, and behavioral symptoms, all ity and an instrument of potential usefulness as a functional of which may contribute to functional capacity in a differ- outcome measure in interventional trials. ent way. However, the TFC and the FAS do not distinguish between the independent contribution of each of these domains, or those of mixed domains, to patients’ global Methods functionality. Further characterizing the influence of each domain on overall functioning may help routine clinical Participants practice and research because changes in functional inde- pendence may be driven differently by motor, cognitive, or We prospectively recruited 78 symptomatic, early-to-mild behavioral symptoms. Functional HD-specific instruments gene-mutation carriers (CAG > 39) from the HD outpatient that can be used to assess the contribution of each of these clinic in the Movement Disorders Unit at Hospital de la symptoms are needed. Santa Creu i Sant Pau in Barcelona. Each gene-mutation Progressive cognitive impairment is an essential feature carrier was accompanied by a knowledgeable informant (KI) of HD and the development of dementia is an inevitable con- such as their caregiver, partner, or family member. Exclusion sequence [8, 9]. Cognitive changes have an enormous impact criteria were presence of a neurological disorder other than on functional capacity in patients with neurodegenerative HD, history of head trauma, epilepsy, drug abuse, non-cor- disease [10–12] or other conditions involving damage to the rected visual problems, active major psychotic or delusional central nervous system damage [13]. Accordingly, instru- syndrome, and severe language difficulties. ments focusing on the impact of cognitive impairment on All patients and KI gave written informed consent. The functional capacity have been developed in clinical contexts study protocol was reviewed and approved by the local such as Parkinson’s disease, Alzheimer’s disease, and stroke research ethics committee at Hospital de la Santa Creu i [14–18]. Although functional assessment instruments in HD Sant Pau and the study was performed in compliance with include items that can be influenced by cognitive changes, the 1964 Declaration of Helsinki and latter amendments. none of the instruments currently available is able to specifi- cally isolate the impact of cognitive impairment on function- Procedure ality. It was recently suggested that the Neuro-Qol cognitive measure questionnaire has psychometric properties[19] that We recorded sociodemographic and clinical data, includ- are appropriate for HD and could reliably address cogni- ing age, education level, CAG repeat length, cognitive sta- tive status by assessing executive functions and general tus and severity of neuropsychiatric symptoms. Each study concerns [20, 21]. However, although it focuses on daily visit included the administration of a battery of cognitive, living, the Neuro-Qol is not exactly a cognitive-functional behavioral, motor and functional assessments. Motor status assessment instrument. It was not specifically developed for and disease severity were assessed by a neurologist. Dis- HD and it does not consider the influence of patients’ lack ease stage was determined according to the Shoulson and of awareness. Fahn staging for HD [4]. Motor symptoms were rated using To overcome the lack of HD-specific cognitive-functional the UHDRS total motor score (UHDRS-TMS). The dis- assessment instruments, we developed the “Huntington’s ease burden score (DBS) was calculated using the formula Disease-Cognitive Functional Rating Scale” (HD-CFRS). ([CAG—35.5] × age) [22]. Functionality was assessed with The approach we used was the one we used previously to the UHDRS Functional Assessment Scale (FAS), the Inde- develop a Parkinson’s disease-specific cognitive-functional pendence scale (IS) and the TFC. assessment scale [14]. Behavioral symptoms were assessed using the short form The HD-CFRS is a 5-min questionnaire specifically of the Problem Behavior Assessment for HD (PBA-s) [23]. developed to capture a wide range of functional aspects The PBA-s consists of a semi-structured interview designed 13 Journal of Neurology (2022) 269:3541–3549 3543 to cover several neuropsychiatric symptoms occurring in having major cognitive impairment in the range of dementia HD by assessing their severity and frequency over the past (HD-Dem). 4 weeks. The PBA-s was administered simultaneously to both the HD carrier and the KI. HD‑CFRS development and administration The screening measures used to assess global cognitive status were the Mini-Mental State Examination (MMSE) The HD-CFRS is a semi-structured interview designed to and the Parkinson’s Disease-Cognitive Rating Scale (PD- measure cognitive-related functional changes in HD. It com- CRS). The PD-CRS has shown to be a reliable instrument prises 12 items, scored in a Likert-like manner, that have to assess global cognition in HD, allowing patients to be been specifically developed to take into account the proto- classified into groups according to the severity of cognitive typical neuropsychological characteristics of HD [8, 9, 30]. impairment [24]. The MMSE has been tested as a cognitive For the item development, expert members of an HD spe- screening test for use in HD [25, 26], and despite, its limi- cialized unit met and generated items that were sufficiently tations in terms of sensitivity to cognitive changes in this sensitive to detect the most affected cognitive domains in disorder, it provides measurements that are not covered by HD. In addition, they examined the wording to make sure the PD-CRS and that may be relevant for the present study it was comprehensible. In a pre-testing stage, the interview (i.e., temporal and spatial orientation). As additional cogni- was then randomly administered to patients visiting the HD tive measures, we administered the Stroop color-naming test, unit. After receiving input from patients and informants, we the word-reading and interference tests, the phonetic verbal modified the scale slightly to make it more understandable, fluency test with letters F, A and S (FAS), the semantic ver- and decided to administer it as a semi-structured interview. bal fluency test (animals), parts A and B of the Trail Making The items comprising the HD-CFRS assess dual tasking, Test (TMT), and the Symbol Digit Modalities Test (SDMT). sustained attention, organization, problem solving, lan- The HD-CRFS was administered to all participants and guage, orientation, and memory in daily living scenarios. KI to measure the agreement between them. Because HD Scoring is based on the frequency of having or not hav- is associated with a significant pattern of lack of awareness ing some difficulty in performing the activities listed for about the severity and functional repercussion of symptoms each item (0 = none; 1 = some of the time; 2 = most of the [27, 28], we compared the HD-CRS total score provided by time; 8 = subject has never done the activity in the past). the KI to the total score provided by gene-mutation carriers. When answers are scored “8”, this score is replaced by the We also calculated the time it took to administer the HD- mean score obtained in all the other items. The scale was CFRS to both participants and KI. Global cognitive status, developed and administered in Spanish. For the purposes behavior and the HD-CFRS were assessed by a trained neu- of this manuscript, the HD-CFRS in Spanish was sent to an ropsychologist, expert in HD. external language review service for translation into English To explore the discriminative capacity of the HD-CFRS (Appendix). to classify patients according to global cognitive status, we used an approach based on the combination of the Clinical Statistical analysis Dementia Rating Scale (CDR) and the Independence Scale (IS) [24, 29]. The CDR is used to characterize cognitive Data are expressed as means ± standard deviations (SDs) and functional performance in memory, orientation, judg- for continuous variables and as percentages for categorical ment and problem solving, community affairs, home and variables. Group differences were analyzed using analyses hobbies, and personal care. A CDR rated as 0 means absent of variance (ANOVAs) and t tests for continuous variables, cognitive impairment, 0.5 indicates questionable/very mild the Mann–Whitney test for ordinal data, and X2 for categori- cognitive impairment, and ratings between 1 and 3 indicate cal variables. Normality analysis was carried out from fre- the presence of mild-to-severe cognitive deficits. The IS quency distribution and Kruskal–Wallis test was used for scale is based on 25 yes/no questions and qualifies the abil- asymmetric variables. Internal consistency was assessed ity to independently perform ADLs. Based on the answers using Cronbach’s α coefficients and intraclass correlation provided, a score of “independence” is computed, ranging coefficients (ICC, two-way mixed model and absolute agree- from 100% (no special care needed) to 10% (tube-fed, total ment) were used to examine agreement between patients bed care). A score above 80% means that the participant and caregivers regarding HD-CFRS scores. The associa- remains employed, does household chores, and manages tion between HD-CFRS scores and the various cognitive finances. Accordingly, patients with a CDR score of 0 and measures was assessed using partial correlation coefficients, IS score > 80% were classified as cognitively preserved (HD- controlling for the effects of age, education, CAG, UHDRS- NC), those with a CDR of 0.5 and IS score > 80% were clas- TMS, PBA-s and gender. Receiver-operating characteristic sified as having mild cognitive impairment (HD-MCI), and (ROC) curves were generated to determine the HD-CFRS those with a CDR ≥ 1 and IS score ≤ 80% were classified as cutoff scores that best discriminated between the three 13 3544 Journal of Neurology (2022) 269:3541–3549 cognitive groups. Sensitivity, specificity, and positive (PPV) Participants in the HD-NC were not significantly and negative predictive value (NPV) were calculated for the younger than HD-MCI [t(52) = 0.209; p = 0.835] or HD- HD-CFRS cutoff points. Dem [t(44) = − 0.921; p = 0.362] participants. Further- All statistical procedures were performed using SPSS more, HD-NC participants had a similar level of education software, and a p value < 0.05 was considered statistically to the HD-MCI [t(52) = 1.634; p = 0.108] group. However, significant. HD-NC had a significantly higher level of education than the HD-Dem group [t(44) = 2.516; p = 0.016]. As expected, functional capacity and motor values differed significantly between cognitive HD groups (see details in Table 1). Results Focusing on behavioral variables, we found significant differences between HD-NC and HD-MCI and HD-Dem Clinical and sociodemographic data groups regarding apathy severity (see details in Table 1). Regarding global cognitive status, HD-MCI participants The HD sample consisted of 78 symptomatic gene-mutation scored significantly lower than HD-NC in the PD-CRS carriers (59% females, mean age 53.1 ± 10.9 years, mean [t(52) = 5.141; p < 0.001] and in the MMSE [t(52) = 2.960; CAG repeat length = 43.4 ± 2.8, mean years of education p = 0.005]. HD-Dem participants scored significantly lower 11.2 ± 4.1). Twenty of the 78 were classified as HD-NC in the PD-CRS [t(57) = − 6.833; p < 0.001] and in the (mean age = 52.2 ± 9.5; mean CAG = 42.9 ± 2.4; mean years MMSE tests [t(57) = − 4.959; p < 0.001] than participants of education = 13.1 ± 3.7; mean TFC = 12.1 ± 1), 33 as HD- in HD-MCI group (see details in Table 1). MCI (mean age = 51.7 ± 9.2; mean CAG = 43.1 ± 2.3; mean Regarding the total score of the HD-CFRS, and as seen years of education = 11.2 ± 4.0; mean TFC = 10.6 ± 0.9) in Table 1, HD-MCI almost had a significantly higher score and 25 as HD-Dem (mean age = 55.6 ± 13.8; mean than HD-NC in the HD-CFRS total score [t(52) = − 1.991; CAG = 44.0 ± 3.4; mean years of education = 9.9 ± 4.3; mean p = 0.052] when the HD-CFRS was answered by the par- TFC = 7.0 ± 2.4). ticipant. When it was answered by KI, the differences Table 1  Clinical and sociodemographic characteristics of the sample HD (n = 78) HD-NC (n = 20) HD-MCI (n = 33) HD-Dem (n = 25) p a Age 53.1 ± 10.9 52.2 ± 9.5 51.7 ± 9.2 55.6 ± 13.8 0.835; b0.362; c0.203 Gender (f/m) 46/32 14/6 17/16 15/10 χ2 = a0.150; b0.352; c0.354 a Education 11.2 ± 4.1 13.1 ± 3.7 11.2 ± 4.0 9.9 ± 4.3 0.108; b0.016; c0.249 a CAG​ 43.4 ± 2.8 42.9 ± 2.4 43.1 ± 2.3 44.0 ± 3.4 0.806; b0.215; c0.198 DBSA 396 ± 103.7 375.5 ± 97.1 375.7 ± 88.3 436.6 ± 117.4 a 0.992; b0.073; c0.031 UHDRS-TMSB 38 ± 22.8 20.1 ± 16.3 34.6 ± 16.5 57.2 ± 21.4 a 0.004; b< 0.001; c< 0.001 TFCC 9.8 ± 2.6 12.1 ± 1 10.6 ± 0.9 7.0 ± 2.4 a < 0.001; b< 0.001; c< 0.001 a IS 79.6 ± 12.6 92.9 ± 8.3 80.7 ± 7.6 67.9 ± 8.7 < 0.001; b< 0.001; c< 0.001 a MMSE 24.9 ± 3.8 28.1 ± 1.4 25.8 ± 2.9 21.5 ± 3.5 0.005; b< 0.001; c< 0.001 a PD-CRS 73.5 ± 19.9 94.7 ± 11.1 75.7 ± 13.5 53.3 ± 9.1 < 0.001; b< 0.001; c< 0.001 a HD-CFRS 5.6 ± 4 3.6 ± 2.8 5.2 ± 3.0 7.8 ± 4.9 0.052; b0.001; c0.016 a HD-CFRS (KI) 10.2 ± 6.5 3.2 ± 2.5 9.1 ± 3.2 17.2 ± 4.6 < 0.001; b< 0.001; c< 0.001 PBA-s a Depression 3.4 ± 4.9 3.0 ± 3.3 3.8 ± 6.3 3.2 ± 3.9 0.625; b0.920; c0.655 a Irritability 3.5 ± 5.1 2.2 ± 3.2 4.7 ± 6.4 2.9 ± 4.1 0.120; b0.542; c0.235 a Apathy 4.6 ± 4.1 2.3 ± 3.0 4.4 ± 3.7 6.5 ± 4.3 0.061; b0.001; c0.049 a Psychosis 0.5 ± 1.9 0.4 ± 1 0.4 ± 1.3 0.6 ± 2.8 0.930; b0.757; c0.756 a Executive dysfunction 3.7 ± 4.4 2.6 ± 3.6 3.1 ± 4.8 5.4 ± 4.2 0.736; b0.027; c0.065 A Disease burden score B Unified Huntington’s disease rating scale-total motor score C Total functional capacity a HD-NC vs HD-MCI b HD-NC vs HD-Dem c HD-MCI vs HD-Dem 13 Journal of Neurology (2022) 269:3541–3549 3545 between HD-MCI and HD-NC groups were significant Correlations between HD‑CFRS and cognitive [t(52) = − 6.935; p < 0.001]. HD-Dem scored significantly measures higher than HD-MCI in the HD-CFRS total score corre- sponding to the participant [t(58) = 2.474; p = 0.016] and Partial bivariate correlation analysis was performed between when answered by KI [t(58) = 7.874; p < 0.001]. HD-CFRS and global cognitive measures. Responses pro- The average time needed for the researcher to complete vided by HD participants showed a significant associa- the HD-CFRS together with a semi-structured interview was tion between the HD-CRS and the total PD-CRS scores 8 ± 2 min for HD participants and 5 ± 1 min for KI. (r = − 0.336; p = 0.004), and the MMSE total score (r = − 0.245; p < 0.044). When answers were provided by Agreement between HD participants and KI the KI, strong correlations were found between the HD- in HD‑CFRS CFRS total score and the PD-CRS total score (r = − 0.753; p < 0.001), and the MMSE total score (r = − 0.722; Looking at the whole HD sample, we found significant dif- p < 0.001). Partial correlation coefficients controlling for age, ferences between answers provided by participants and KI education, CAG, UHDRS-TMS, PBA-s and gender showed [t(78) = − 7.232; p < 0.001]. Total scores in the HD-CFRS mild-to-strong correlations between the HD-CFRS answered were lower for HD participants than for KI. by KI and the PD-CRS total score (r = − 0.560; p < 0.001), Focusing on the HD-NC group, we found no signifi- and the MMSE total score (r = − 0.631; p < 0.001). When cant differences between answers provided by partici- covariables were added, no significant association was found pants and those by KI regarding the HD-CRS total score between the HD-CRS answered by HD participants and cog- [t(20) = − 0.972; p = 0.343]. The average ICC was 0.927 nitive measures. with a 95% confidence interval ranging from 0.82 to 0.971 Based on the findings regarding the discrepancies [F(1,19) = 13.735, p < 0.001]. Correlation analysis between between scores provided by KI and scores provided by gene- the HD-CFRS total score provided by participants and mutations carriers, we performed the following correlation KI showed a significantly positive association (r = 0.870; analyses using the HD-CFRS total score provided by KI. p < 0.001). These associations remained strongly signifi- In the HD-NC group, no significant association was cant when age, education, CAG, UHDRS-TMS, PBA-s and found between the HD-CFRS and disease stage (r = − 0.287; gender were controlled in the partial correlation analysis p = 0.233). In the HD-NC group, the association with the (r = 0.802; p = 0.002). TFC was not significant (r = − 0.393; p = 0.096). In the Focusing on the HD-MCI group, we found signifi- HD-MCI group, no association was found between the HD- cant differences when comparing HD-CFRS scores CFRS and the disease stage (r = − 0.201; p = 0.357) or with provided by participants and those provided by KI the TFC (r = 0.235; p = 0.281). In addition, in the HD-Dem, [t(33) = − 6.026; p < 0.001]. The average ICC was 0.463 significant associations were found between HD-CFRS and with a 95% confidence interval ranging from − 0.087 to the disease stage (r = 0.666; p = 0.004), and also between 0.735 [F(1,32) = 1.863, p = 0.058]. Correlation analysis HD-CFRS and the TFC (r = − 0.593; p = 0.012). between the HD-CFRS total score provided by partici- The HD-CFRS showed significant associations with sev- pants and KI did not show a significant positive association eral subscores for all the cognitive measures obtained. All (r = 0.302; p = 0.088). These associations were worse when these associations remained significant after controlling for age, education, CAG, UHDRS-TMS, PBA-s and gender the effects of age, education, CAG, UHDRS-TMS, PBA-s were controlled in the partial correlation analysis (r = 0.323; and gender. In the MMSE, associations were found between p = 0.107). the HD-CFRS total sore and orientation (r = − 0.532; Focusing on the HD-Dem group, we observed signifi- p < 0.001), attention (r = − 0.503; p < 0.001), recall cant differences when comparing HD-CFRS scores between (r = − 0.319; p < 0.01), writing (r = − 0.391; p = 0.005), gene-mutation carriers and KI [t(25) = − 7.875; p < 0.001]. and copy of a pentagon (r = − 0.458; p = 0.001). In the The average ICC was 0.354 with a 95% confidence interval PD-CRS, associations were found with immediate recall ranging from − 0.466 to 0.715 (F(1,24) = 1.548, p = 0.146). (r = − 0.264; p < 0.05), naming (r = − 0.290; p < 0.05), atten- Correlation analysis between the HD-CFRS total score tion (r = − 0.455; p = 0.001), working memory (r = − 0.530; provided by participants and KI did not show a significant p < 0.001), drawing of a clock (r = − 0.418; p < 0.005), copy positive association (r = 0.215; p = 0.301). These associa- of a clock (r = − 0.318; p < 0.05), delayed recall (r = − 0.264; tions were worse when age, education, CAG, UHDRS-TMS, p < 0.05), alternate fluency (r = − 0.472; p < 0.001), fronto- PBA-s and gender were controlled in the partial correlation subcortical score (r = − 0.521; p < 0.001), and posterior- analysis (r = − 0.041; p = 0.869). cortical score (r = − 0.350; p < 0.05). In the other cogni- tive measures, associations were found with the SDMT (r = − 0.485; p < 0.001), semantic fluency (r = − 0.416; 13 3546 Journal of Neurology (2022) 269:3541–3549 Fig. 1  Receiver-operating characteristic (ROC) curves illustrating the discriminative properties of the HD-CFRS (KI) Table 2  Accuracy measures for the screening of MCI and dementia using different HD-CFRS (KI) cutoff scores HD-CFRS cutoff for MCI Sensitivity Specificity 3.5/4.5 0.97 0.75 4.5/5.5 0.88 0.80 5.5/6.5 0.79 0.85 HD-CFRS cutoff for dementia Sensitivity Specificity 10.5/11.5 0.92 0.79 11.5/12.5 0.92 0.83 12.5/13.5 0.79 0.87 p = 0.002), Stroop color-naming (r = − 0.397; p < 0.05), Discussion Stroop word-reading (r = − 0.299; p < 0.05), Stroop inter- ference test (r = − 0.519; p = 0.001), TMT-A (r = 0.392; In the present study, we developed a brief HD-specific cog- p < 0.01), TMT-B (r = 0.516; p < 0.001) and verbal fluency nitive-functional assessment instrument and explored its (r = − 0.446; p = 0.001). main psychometric properties. We found it had good psycho- metric attributes and was sensitive to discriminate between Discriminative capacity of HD‑CFRS patients with varying degrees of cognitive impairment. The highly significant differences between scores rated by Discriminant ROC analysis showed that a HD-CFRS total a symptomatic HD participant with mild-to-severe cognitive score ≥ 5.5/6.5 was the optimal cutoff to discriminate impairment and scores rated by KI suggest that cognitive- between HD-NC and HD-MCI [sensitivity, 88%; speci- functional assessment in HD should focus on an interview ficity, 80%; PPV, 93%; NPV, 86%; area under the curve with a reliable informant. (AUC) = 0.929; 95% CI 0.859–0.999]. A HD-CFRS total The HD-CFRS appeared to capture the functional impact score ≥ 11.5/12.5 was the optimal cutoff to discriminate of cognitive defects in the HD population. Specific cutoff between HD-Dem and HD-MCI/NC [sensitivity, 92%; spec- scores were determined to detect patients with cognitive- ificity, 83%; PPV, 92%; NPV, 68%; area under the curve functional deficits in the range of mild cognitive impairment (AUC) = 0.944; 95% CI 0.899–0.989] (Fig. 1; Table 2). (HD-CFRS cutoff ≥ 5.5/6.5), and in the range of dementia (HD-CFRS cutoff ≥ 11.5/12.5). 13 Journal of Neurology (2022) 269:3541–3549 3547 Our results also confirmed a strong association Appendix between cognitive impairment and functional alterations in HD, and showed that these functional alterations can be appropriately measured with the HD-CFRS. We also Huntington’s disease cognitive functional rating scale (HD-CFRS) found correlations with several cognitive measures of sustained attention, cognitive flexibility, working mem- Score ory, dual tasking, and processing speed. These cognitive alterations have been characteristically associated with 1 Do you have difficulties 0 1 2 8 performing two tasks at the frontal-striatal damage typical of HD disease. The the same time? (such as associations found thus highlight the functional impact of chatting while walking, deficits at the levels of these processes. We also observed or listening to someone associations in items related to orientation, semantic while doing something else) integrity, writing, memory, and constructional praxis. 2 Do you have difficulties 0 1 2 8 This study has some limitations. First, the scale was not concentrating or paying developed together with an expert panel. It was guided by attention? (e.g., following the review of the literature in the field and input from a group the plot of a movie or a of experts working together in an HD unit. The purpose was book, or the thread of a conversation) to illustrate the usefulness of an instrument focused on the 3 Do you have difficulties 0 1 2 8 cognitive-functional aspects of HD. However, the scale will organizing what you have require revision; additional items should be explored, a 5-point to do throughout the day? or 7-point rating could be used for the rating scale format, and (and at the end of the day consensus by specialists should be reached. Therefore, as both you have not done all the things you intended to) Spanish and English versions of the HD-CFRS will require 4 Do you have difficulties 0 1 2 8 further validation they are not yet ready for use in clinical prac- planning or organiz- tice or research. Second, future studies with larger samples ing your vacations or are needed as the sample sizes after group stratification were get-togethers with your relatively small. Third, language and communication should family or friends? be assessed in greater detail. Fourth, a KI is needed. Finally, 5 Do you have difficulties 0 1 2 8 controlling your corre- longitudinal studies should be performed to demonstrate the spondence, doctor visits, ability of the HD-CFRS to detect functional changes over time. or times when to take Despite these shortcomings, the study has two main medication, etc.? strengths. Above all, it is the first attempt to develop an HD- 6 Do you have trouble 0 1 2 8 specific cognitive-functional rating scale. Furthermore, it solving unforeseen or unexpected problems? can be completed in a short amount of time, usually taking 7 Do you have a hard time 0 1 2 8 no more than 5 min. It is also of note that we found appropri- explaining what you want ate psychometric attributes in terms of discriminative capac- to say? ity and that we determined specific cutoffs to detect patients 8 Do you have difficulties 0 1 2 8 with cognitive-functional deficits in the range of mild cog- understanding what you nitive impairment and in the range of dementia. The HD- read: books, magazines, newspapers? CFRS could be considered for use both in clinical practice 9 Do you have a hard time 0 1 2 8 and for the design of interventional trials. Since functionality remembering what day is a primary endpoint in interventional trials and because it is? cognitive aspects of HD are a main target of these trials, 10 Do you have difficulties 0 1 2 8 an instrument able to measure specific cognitive-functional using public transport aspects of the disease is of major interest. because it is difficult for you to understand which line to take? 11 Do you have difficulties 0 1 2 8 Conclusions finding your way in unfa- miliar places? The HD-CFRS appears to be a valid and reliable instrument 12 Do you have a hard time 0 1 2 8 to measure cognitive-functional changes in HD. We believe remembering the things you have to do? this questionnaire can be useful both in clinical and research contexts. 13 3548 Journal of Neurology (2022) 269:3541–3549 3. Ross CA, Pantelyat A, Kogan J, Brandt J (2014) Determinants of Score functional disability in Huntington’s disease: role of cognitive and 0: Never or almost never Total motor dysfunction. Mov Disord 29(11):1351–1358 1: Some of the time score: 4. Shoulson I, Fahn S (1979) Huntington disease: clinical care ande- 2: Most of the time ___ valuation. Neurology 29(1):1–3 8: Subject has never done the activity in the past (0–24) 5. H. S. Group (1996) Unified Huntington’s Disease Rating Scale: reliability and consistency. Mov Disord 11(2):136–142 6. Salem L et al (2014) The most appropriate primary outcomes to Acknowledgements The authors wish to thank all the study partici- design clinical trials on Huntington’s disease: meta-analyses of pants and their families for taking part in the study, and Miriam Llorens cohort studies and randomized placebo-controlled trials. Fun- and Elisabet Tejero for their collaboration. dam Clin Pharmacol 28(6):700–710 7. Paulsen JS et al (2010) Challenges assessing clinical endpoints in early Huntington disease. Mov Disord 25(15):2595–2603 Author contributions AH-B: research project conception, organization, 8. Paulsen JS, Smith MM, Long JD (2013) Cognitive decline in execution, and design; formal analysis; writing—original draft. SM-H: prodromal Huntington disease: implications for clinical trials. research project conception, organization, execution, design; formal J Neurol Neurosurg Psychiatry 84(11):1233–1239 analysis; writing—original draft. JP-P: research project organization and 9. Martinez-Horta S et al (2020) Structural brain correlates of execution; writing—review and critique. FS: statistical analysis execution; dementia in Huntington’s disease. Neuroimage Clin 28:102415 writing—review and critique. AP-D: research project organization and 10. Atchison TB, Massman PJ, Doody RS (2007) Baseline cogni- execution; writing—review and critique. JP: research project organization tive function predicts rate of decline in basic-care abilities of and execution; writing—review and critique. JK: research project concep- individuals with dementia of the Alzheimer’s type. Arch Clin tion, organization, design; formal analysis; writing—original draft. Neuropsychol 22(1):99–107 11. Goldman JG et al (2018) Cognitive impairment in Parkinson’s Funding Open Access Funding provided by Universitat Autonoma de disease: a report from a multidisciplinary symposium on unmet Barcelona. This study was partially funded by a Spanish Government needs and future directions to maintain cognitive health. NPJ Grant (PI17/001885). Park Dis 4(1):1–11 12. Mioshi E, Hodges JR (2009) Rate of change of functional abili- Data availability Data supporting the findings in this study are avail- ties in frontotemporal dementia. Dement Geriatr Cogn Disord able on request from the corresponding author. Such data are not pub- 28(5):419–426 licly available due to privacy or ethical restrictions. 13. Tatemichi TK, Desmond DW, Stern Y, Paik M, Sano M, Bag- iella E (1994) Cognitive impairment after stroke: frequency, patterns, and relationship to functional abilities. J Neurol Neu- Declarations rosurg Psychiatry 57(2):202–207 14. Kulisevsky J et al (2013) Parkinsonism and related disorders Conflicts of interest Nothing to report. measuring functional impact of cognitive impairment: valida- tion of the Parkinson’ s Disease Cognitive Functional Rating Ethical standard The study was approved by the ethical committee Scale. Park Relat Disord 19(9):812–817 of Hospital de la Santa Creu i Sant Pau (Barcelona, Spain). Written 15. Morris JC (1997) Clinical dementia rating: a reliable and valid informed consent was obtained from all participants and the study was diagnostic and staging measure for dementia of the Alzheimer conducted in accordance with the ethical standards laid down in the type. Int Psychogeriatr 9(SUPPL. 1):173–176 1964 Declaration of Helsinki and its later amendments. 16. Cumming TB, Blomstrand C, Bernhardt J, Linden T (2010) The NIH stroke scale can establish cognitive function after stroke. Cerebrovasc Dis 30(1):7–14 Open Access This article is licensed under a Creative Commons Attri- 17. Kounti F, Tsolaki M, Kiosseoglou G (2006) Functional cogni- bution 4.0 International License, which permits use, sharing, adapta- tive assessment scale (FUCAS): a new scale to assess execu- tion, distribution and reproduction in any medium or format, as long tive cognitive function in daily life activities in patients with as you give appropriate credit to the original author(s) and the source, dementia and mild cognitive impairment. Hum Psychopharma- provide a link to the Creative Commons licence, and indicate if changes col 21(5):305–311 were made. The images or other third party material in this article are 18. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL included in the article's Creative Commons licence, unless indicated (1982) A new clinical scale for the staging of dementia. Br J otherwise in a credit line to the material. If material is not included in Psychiatry 140(6):566–572 the article's Creative Commons licence and your intended use is not 19. Carlozzi NE (2021) Psychometric properties and responsiveness permitted by statutory regulation or exceeds the permitted use, you will of Neuro-QoL Cognitive.pdf. Qual Life Res 29(5):1393–1403 need to obtain permission directly from the copyright holder. To view a 20. Cella D (2021) Neuro-QOL. Brief measures of health-related copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. quality of life for clinical research in neurology. Neurology 78(23):1860–1867 21. Lai JS et al (2018) Evaluating cognition in individuals with Huntington disease: Neuro-QoL cognitive functioning meas- References ures. Qual Life Res 27(3):811–822 22. Penney JB, Vonsattel JP, MacDonald ME, Gusella JF, Myers 1. Wa l ke r F O ( 2 0 0 7 ) H u n t i n g t o n’s d i s e a s e . L a n c et RH (1997) CAG repeat number governs the development rate of 369(9557):218–228 pathology in huntington’s disease. Ann Neurol 41(5):689–692 2. Ross CA, Tabrizi SJ (2011) Huntington’s disease: from molecular 23. Callaghan J et al (2015) Reliability and factor structure of the pathogenesis to clinical treatment. Lancet Neurol 10(1):83–98 short problem behaviors assessment for Huntington’s disease (PBA-s) in the TRACK-HD and REGISTRY studies. J Neu- ropsychiatry Clin Neurosci 27(1):59–64 13 Journal of Neurology (2022) 269:3541–3549 3549 24. Martinez-Horta S et al (2020) Utility of the Parkinson’s disease- awareness of cognitive, emotional, and functional abilities. J Clin Cognitive Rating Scale for the screening of global cognitive Exp Neuropsychol 29(4):365–376 status in Huntington’s disease. J Neurol 267(5):1527–1535 28. Sitek EJ, Thompson JC, Craufurd D, Snowden JS (2014) Una- 25. Ringkøbing SP, Larsen IU, Jørgensen K, Vinther-Jensen T, wareness of deficits in Huntington’s disease. J Huntingtons Dis Vogel A (2020) Cognitive screening tests in huntington gene 3(2):125–135 mutation carriers: examining the validity of the mini-mental 29. Peavy GM et al (2010) Cognitive and functional decline in state examination and the Montreal cognitive assessment. J Huntington’s disease: dementia criteria revisited. Mov Disord Huntingtons Dis 9(1):59–68 25(9):1163–1169 26. Landwehrmeyer GB et al (2017) Data analytics from Enroll-HD, 30. Julayanont P, McFarland NR, Heilman KM (2020) Mild cognitive a global clinical research platform for Huntington’s disease. Mov impairment and dementia in motor manifest Huntington’s disease: Disord Clin Pract 4(2):212–224 classification and prevalence. J Neurol Sci 408:116523 27. Hoth KF, Paulsen JS, Moser DJ, Tranel D, Clark LA, Bechara A (2007) Patients with Huntington’s disease have impaired 13