Staging dementia based on caregiver reported patient symptoms: Implications from a latent class analysis.

BACKGROUND: Tailoring interventions to the needs of caregivers is an important feature of successful caregiver support programs. To improve cost-effectiveness, group tailoring based on the stage of dementia could be a good alternative. However, existing staging strategies mostly depend on trained professionals.
OBJECTIVE: This study aims to stage dementia based on caregiver reported symptoms of persons with dementia.
METHODS: Latent class analysis was used. The classes derived were then mapped with disease duration to define the stages. Logistic regression with receiver operating characteristic curve was used to generate the optimal cut-offs.
RESULTS: Latent class analysis suggested a 4-class solution, these four classes were named as early (25.9%), mild (25.2%), moderate (16.7%) and severe stage (32.3%). The stages based on the cut-offs generated achieved an overall accuracy of 90.8% compared to stages derived from latent class analysis.
CONCLUSION: The current study confirmed that caregiver reported patient symptoms could be used to classify persons with dementia into different stages. The new staging strategy is a good complement of existing dementia clinical assessment tools in terms of better supporting informal caregivers.

Introduction

Dementia is an umbrella term that includes multiple diseases affecting memory, cognitive abilities, and behavior that interfere significantly with the individuals’ ability to maintain their daily functioning [1]. In 2016, there were around 47 million persons with dementia (PWD) globally [2]. Alzheimer’s Disease International suggested that for people aged 60 and above, the incidence of dementia doubles with every 6.3-year increase in age [3]. Consequent to the population aging [3], there will be 2 billion people aged 60 years and above globally in 2050 [4]; the number of PWD is also expected to increase, to an estimated number of 65.7 million in 2030 and 115.4 million in 2050 [5]. An increasing trend was also noticed in the financial costs associated with dementia. According to a 2017 systematic review, the total economic cost of dementia had escalated from US$279.6 billion in 2000 to US$948 billion in 2016, with an annual growth rate of 15.94% [6]; As a result, dementia has become one of the most important public health priorities [4].

PWD usually require a high level of care, and oftentimes they are taken care of by family members or other informal caregivers [7-9]. Informal caregiving refers to the activities and experiences involved in providing help and assistance to relatives or friends who are unable to care for themselves [10]. This evokes positive feelings among caregivers, such as a sense of satisfaction from observing improvements in the care recipient, fulfilling of parental obligation, or simply from the act of caring for a loved one [11]. However, in many cases, caregivers taking care of PWD tend to view these experiences as enduringly uncertain, stressful, and frustrating [12, 13]. Such adverse experiences might lead to high caregiving burden [8], or depressive symptoms [14]. Poor health of the caregivers can affect the patient as well, as it has been shown to result in lower quality of care and earlier institutionalization [7]. Therefore, it is imperative to support informal caregivers and to help them through the caregiving process.

Tailoring interventions to the needs of caregivers of PWD is an important component of effective caregiver support programs [7]. Mittelman and colleagues [15] found that individually tailored counseling sessions specifically targeting individual caregiver needs successfully reduced caregiver depressive symptoms over two years. This is consistent with the findings from a previous systematic review [16], suggesting that individually tailored behavior management therapy successfully reduces caregiver distress and burden both in the short and long term. However, individually-tailored interventions usually cost much more than group-based programs, resulting in low cost-effectiveness. To address this issue, one solution could be to adopt a medium level of customization—identifying specific subgroups of people who share common features or challenges and then tailoring interventions based on their specific demands. From this perspective, stages of dementia could be a good reference point, and it is also widely used in fields such as tailored e-health interventions for caregivers of PWD [17].

Several scales are used to determine the stage of dementia, such as the Clinical Dementia Rating scale [18] and the Mini-Mental State Examination [19]. However, these scales often require input from trained professionals [20]. In order to better support caregivers of PWD, assessments based on patient symptoms will be more operational i.e., caregivers can easily identify the stage of the PWD they are caring for and then search for specific caregiving tips. There have been efforts on staging dementia based on symptom profiles of PWD tracked through a web-based tool [21]; however, this tool has a full list of 60 symptoms, which is quite complex. The staging process relies on artificial intelligence and it is not easy for the caregivers to navigate this process. Thus, there is a need to stage dementia more practically and in a simpler way.

The current study aims to 1) use latent class analysis (LCA) to identify the latent subtypes of PWD based on caregiver reported patient symptoms, and to map these subtypes with illness duration to stage dementia; 2) estimate the association between the subtypes and depressive symptoms among the caregivers; and 3) calculate the optimal cut-off values based on the assessment tools used to develop screening cutoff scores which can be used to classify different stages of PWD.

Materials and methods

Participants and procedures

Primary informal caregivers of PWD were recruited through convenience sampling from the outpatient clinics of a tertiary mental health hospital and a general hospital in Singapore. A local Voluntary Welfare Organization serving caregivers helped to advertise the study by displaying the study flyers so that potential participants could contact the study team. Participants had to meet the following eligibility criteria: 1) Singapore citizens or permanent residents, 2) aged 21 years or above, 3) able to read, write, and speak in English/Chinese/Malay, 4) self -identified as the primary informal caregiver of PWD (defined as family member or friend most involved in providing care or ensuring provision of care to the patients). Caregivers were excluded if they had difficulty understanding the consent (as a result we could not obtain informed consent) and if they failed to visit the PWD at least once on a weekly basis.

Data were collected from Jan 2017 to Dec 2018. Potential participants were mainly approached by the study team members or referred by our collaborating clinicians (two geriatric psychiatrists from the tertiary mental health hospital, and one from the general hospital). Eligible caregivers who were interested in the study were followed up by the study team member through an interviewer-administered questionnaire for data collection. On average, the length of the interview (on the scales used in the current study) took around 20 minutes. A total of 433 primary informal caregivers of PWD were approached, among which 282 agreed to participate, representing a response rate of 65%.

The study was approved by the National Healthcare Group Domain Specific Review Board in Singapore (reference number: 2016/00921). The study was conducted in accordance with the precepts of the Declaration of Helsinki. All participants signed the written informed consent prior to their enrolment, and appropriate measures were taken to ensure confidentiality and data privacy.

Measurements

In the current study, symptoms of PWD were characterized by functional status, memory, and behavior problems. Functional status was measured by the widely used Activities of Daily Living Scale (ADL) [22] and the Instrumental Activities of Daily Living Scale (IADL) [23]. Both scales have been validated in Singapore as a measure of disability [24-26]. The ADL has 6 items, covering patient disability in six basic self-care activities (i.e., bathing, dressing, toileting, transfer, continence and feeding). The IADL includes 8 items, and it measured eight other self-care activities (i.e., ability to use a telephone, shopping, food preparation, housekeeping, laundry, mode of transportation, responsibility for own medication, and ability to handle finances). During the interview, caregivers were required to report whether the PWD received any assistance in any of these activities at the time of recruitment. Dependence of PWD was defined as receiving assistance from caregivers, and this included those who might be able to undertake the activity but refused to do so. The number of ADL and IADL dependencies were summed up separately to indicate the self-care impairment scores for the PWD. The memory and behavior problems of PWD were measured by the Revised Memory and Behaviour Problems Checklist (RMBPC) [27]. The RMBPC has 24 items in total, and it has three components, namely memory (7 items), disruptive behavior (8 items), and depression (9 items) [27]. Following a previous study [28], only the 15 items on memory and disruptive behaviour were used in the current study. Caregivers were asked to report whether these problems occurred during the week prior to their participation in the current study. Therefore, items under the ADL (independent vs. needs assistance), the IADL (independent vs. needs assistance), and the RMBPC (yes vs. no) were all binary variables.

Depressive symptoms among caregivers were measured by the 20-item Centre for Epidemiological Study Scale (CES-D) [29]. This scale was shown to have good validity and utility in detecting depression among family caregivers of PWD in Singapore [30]. Individuals were asked to rate how often they experienced depressive symptoms over the past week on a 4-point Likert scale (ranging from 0–3, with ‘0’ representing ‘rarely or none of the time and ‘3’ representing ‘most or all of the time). This scale has been widely used in epidemiological studies [31], and its total score ranges from 0 to 60, with higher scores indicating more severe depressive symptoms. A CES-D score of 16 or higher indicates at risk for clinical depression [32].

Socio-demographic information including caregiver’s age, gender, ethnicity, marital status, education level, employment status and personal monthly income were collected. Caregiving related variables including caregiver’s relationship to the PWD, living arrangement, having a domestic helper or not, and PWD’s duration of being diagnosed with dementia were also collected.

Data analysis

Four steps of statistical analyses were used to ultimately develop the screening criteria of dementia stages based on caregiver reported patient symptoms (i.e. ADL, IADL, and RMBPC):

latent class analysis was conducted using PROC LCA in SAS 9.3 [33] with regards to the scales to measure caregiver reported patient symptoms (i.e. ADL, IADL and RMBPC). The Akaike Information Criteria (AIC) [34], the Bayesian Information Criteria (BIC) [35], the consistent Akaike Information Criteria (cAIC) [36] and the interpretability of competing solutions [37] were considered while selecting the model with the optimal number of latent classes. Low information criteria indicate better fitting. Similar to previous studies [38, 39], interpretability was considered when information criteria contradicted; and a model with latent prevalence less than or equal to 10% was considered as limited clinical relevance in the current study.

the latent subtypes were mapped based on PWD’s duration of diagnosis to identify if there is a clear trend of latent classes by time—stages in other words.

logistic regression analysis was used to explore the association between latent subtypes of symptoms of PWD and the depressive symptoms among the caregivers (depressive = 1 if CES-D >= 16).

logistic regression with receiver operating characteristic (ROC) curve was conducted to identify the optimal cut-off values based on the assessment tools used. Similar analytical strategies have been used in other studies [40-42]. The Youden’s J statistic (sensitivity + specificity -1) [43] was used in the cut-off selection.

Results

The mean age of the 282 participants was 55.7 years (standard deviation = 11.8), with three quarters of them being female (75.2%), and a majority being Chinese (83.0%). Around half of the primary caregivers were daughters of the PWD (55.3%), followed by sons (17.0%) and spousal caregivers (15.3%). Most of them were living with the PWD (70.2%), and more than half of them had a domestic helper (57.1%). The descriptive statistics are shown in Table 1.

Table 1

Descriptive statistics of the study participants (n = 282).

	Frequency	Percentage
Gender
Male	70	24.8
Female	212	75.2
Ethnicity
Chinese	234	83.0
Malay	29	10.3
Indian & others	19	6.7
Education level
Secondary or below (including N/O level)	120	42.6
A level, polytechnic and other diploma	73	25.9
Degree or above	89	31.6
Marital status
Never married	79	28.0
Ever married	203	72.0
Employment status
Unemployed/retired/housewife	121	42.9
Employed	161	57.1
Monthly Income
<SGD2,000 (USD 1,477)	56	19.9
SGD2,000 (USD 1,477)–SGD5,999 (USD 4,431)	78	27.7
SGD6,000 (USD 4,432) or above	42	14.9
Not applicable	106	37.6
Relationship to the PWD
Spouse	43	15.3
Son	48	17.0
Daughter	156	55.3
Others	35	12.4
Living Arrangement
Together with the PWD	198	70.2
Separately from the PWD	84	29.8
Have a domestic helper
Yes	161	57.1
No	121	42.9
	Mean	Standard deviation
Age	55.7	11.8

Note: SGD—Singapore dollars; PWD—persons with dementia

Comparisons of fit indices across solutions showed the AIC reached its lowest value in the seven-class solution, whereas the BIC and cAIC favored the four-class solution. Please refer to Table 2 for the model fit indices. In this case, solutions with classes from four to seven were all considered. However, the six-class and seven-class solutions both had a low prevalence for one of the classes (5.0% in six-class model; and 5.6% in seven-class model); and in the five-class model, one class was not readily distinguished from another class. As a result, also considering the interpretability, the four-class model was selected. Fig 1 plots the conditional probability of class based on PWD symptoms. After mapping the classes with duration of illness, these four mutually exclusive classes were sequenced following the order of class 4 –class 2 –class 1 –class 3. Class 4 (25.9%) was characterized as low dependence on ADL, low dependence on IADL and low memory and behavior problems (MBP); class 2 (25.2%)—low dependence on ADL, low dependence on IADL and high MBP; class 1 (16.7%)—high dependence on ADL, high dependence on IADL and high MBP; class 3 (32.3%)—high dependence on ADL, high dependence on IADL, low MBP. This is consistent with the disease progression of dementia, thus class 4 is named as ‘early stage’, class 2 –‘mild stage’, class 1 –‘moderate stage’, and class 3 –‘severe stage’. Caregivers who were taking care of PWD in the moderate (55.3%) and mild stages (45.1%) showed much higher prevalence of potential depression compared to those providing care to PWD in the early (31.5%) or severe stages (26.4%). Please refer to Table 3 for the details.

Table 2

Comparisons of model fit indices for fitted LCA models.

Number of classes	Log-likelihood	AIC	BIC	cAIC	Entropy
4	-3716.59	4494.69	4928.08	5047.08	0.89
5	-3654.86	4431.24	4973.88	5122.88	0.89
6	-3615.89	4413.28	5065.18	5244.18	0.92
7	-3577.82	4397.14	5158.3	5367.3	0.94

Note: AIC—the Akaike Information Criteria; BIC—the Bayesian Information Criteria; cAIC—the consistent Akaike Information Criteria

Fig 1

Conditional probability of class based on PWD symptoms (answer = yes).

Table 3

Statistics for different stages of dementia in the current sample.

	Average diagnosis duration	No of ADL	No of IADL	No of MBP	Prevalence of potential depression among caregivers
Early stage (class 4, n = 73)	38.2	0.6	4.2	5.1	31.5%
Mild stage (class 2, n = 71)	48.4	0.9	5.0	9.6	45.1%
Moderate stage (class 1, n = 47)	54.7	4.1	7.3	10.0	55.3%
Severe stage (class 3, n = 91)	65.7	4.3	7.2	4.7	26.4%

Note: ADL—Activities of Daily Living Scale; IADL—Instrumental Activities of Daily Living Scale; MBP—memory and behavior problems

The logistic regression results suggested that after controlling for confounders (i.e. all variables listed in Table 1), the prevalence of potential depression based on CES-D cutoff of 16 was still significantly different amongst caregivers who were taking care of PWD at different stages—early stage vs. moderate stage (odds ratio = 0.357, 95% CI 0.153–0.829); and severe stage vs. moderate stage (odds ratio = 0.287, 95% CI 0.130–0.632) of dementia. Please refer to Table 4 for the details.

Table 4

Logistic regression results of stage of PWD on potential depression among caregivers.

	Odds ratio	95% Confidence Interval		p-value
Early stage (class 4, n = 73)	0.357	0.153	0.829	0.017*
Mild stage (class 2, n = 71)	0.485	0.215	1.094	0.081
Moderate stage (class 1, n = 47)	Ref
Severe stage (class 3, n = 91)	0.287	0.13	0.632	0.002*

Note: This regression had controlled for caregiver socio-demographic characteristics (including age, gender, ethnicity, education level, marital status, employment status, and personal monthly income) and caregiving related variables (including caregiver’s relationship to the PWD, living arrangement, having a domestic helper or not, and caregiving duration);

*p<0.05.

Since the LCA suggested similar patterns of ADL and IADL across the stages—low ADL and low IADL for both early and mild stages; and high ADL and high IADL for both moderate and severe stages, to simplify the process of obtaining cut-offs values for staging, these two were added together to form a single new variable (i.e. functional dependence). As a result, the four stages derived from the LCA can be represented by a 2x2 matrix on level of functional dependence and MBP level: 1) early stage—low functional dependence & low MBP; 2) mild stage—low functional dependence & high MBP; 3) moderate stage—high functional dependence & high MBP; and 4) severe stage—high function dependence & low MBP. Since the pattern of functional dependence and MBP are inconsistent across the stages, two cut-off values were calculated based on functional dependence (i.e. ADL+IADL) and the RMBPC separately. Logistic regression with ROC analysis was conducted for stages showed high levels of functional dependence (=1 if PWD fell under moderate or severe stages; = 0 otherwise), with functional dependence being the predictor. Another logistic regression with ROC analysis was also conducted for stages showed high level of MBP (=1 if PWD fell under mild or moderate stages; = 0 otherwise), regressing on the RMPBC. The former logistic regression suggested that when the probability level fell between 0.08 and 0.56, the cut-off value of functional dependence show highest Youden’s J statistic (i.e. 0.993+0.951–1 = 0.944). This converted to a range of the cut-off values for functional dependence from 8.11 to 8.95. Since functional dependence is an integer, 9 was selected as its cut-off (i.e. sum of ADL and IADL). Similarly, for level of MBP, Youden’s J statistic (0.89+0.957–1 = 0.847) was highest when the probability level was between 0.28 and 0.82, which converted to a range of cut-off values between 7.01 and 7.97. As a result, 8 was selected as the cut-off for the RMBPC. In this case for this screening strategy, PWD will be allocated to early stage if their score on ADL+IADL < 9 and RMBPC < 8; mild stage if ADL+IADL < 9 and RMBPC > = 8; moderate stage if ADL+IADL > = 9 and RMBPC > = 8; and severe stage if ADL+IADL > = 9 and RMBPC < 8. The screening based on these two cut-off values achieved an overall accuracy of 90.8% as compared to the stages derived from the LCA. Refer to Table 5 for the numbers under each stage from the LCA and the screening strategy based on these two cut-off values.

Table 5

Comparison of number of participants under each stage.

		Latent class analysis staging
		Early	Mild	Moderate	Severe
Screening staging	Early	69	8	0	0
	Mild	2	58	1	0
	Moderate	0	3	43	5
	Severe	2	2	3	86

Discussion

The latent class analysis revealed a 4-class solution based on caregiver reported symptoms of PWD. These four classes were mutually exclusive and showed a clear trend reflecting PWD’s duration of diagnosis. This supports our hypothesis that staging dementia purely based on caregiver reported PWD symptoms is a viable approach. Unlike other diseases, primary informal caregivers of PWD are the most suitable persons to report PWDs’ symptoms since they are the family members or friends who are most involved in PWDs’ daily care. From this point of view, this new staging mechanism derived from the LCA is reliable.

The new staging strategy classified PWD into four stages, namely early stage, mild stage, moderate stage, and severe stage. At the very beginning of the disease (early stage), PWD show very minimum dependence in ADL, IADL, and have relatively fewer issues on memory and behavior problems. This stage usually lasts for a few years until the disease progresses to the mild stage. During the mild stage, PWD have more memory and behavior problems; however, their abilities for ADL and IADL remain similar as in the early stage. During this stage, PWD might cause caregiver stress as they are physically independent but mentally dependent. Gradually, PWD move to the moderate stage in which PWD’s dependence on others for their ADL and IADL increases, but their level of memory and behavior problems remains similar as at the mild stage. This stage is usually most stressful for caregivers since they need to take care of both the physical needs and the memory and behavior issues of the PWD. This could also be seen from the higher odds of potential depression identified in this group of caregivers. Lastly, PWD move to the severe stage, in which PWD are mostly bed-ridden, with high dependence on ADL and IADL, and caregivers are less troubled by PWDs’ memory and behavior problems.

Other than the staging from the LCA, the current study also provides cut-offs on functional dependence (i.e., 9 for ADL + IADL) and level of memory and behavior problems (i.e., 8 for RMBPC), which enable caregivers to quickly assess which stage the PWD they are caring for belong to. This screening strategy had an overall accuracy of 90.8% among the current population. Moreover, unlike the traditional classification which usually needs to be assessed by trained healthcare professionals [20], this approach only requires caregivers’ rating on PWD’s dependence on the ADL and the IADL, and presence of symptoms on the RMBPC. This makes it much more operational compared to traditional clinical assessments and as a result requires much less resources if implemented in practice.

Findings from the current study have significant implications, especially in the area of tailoring interventions for caregivers. Firstly, since this new staging strategy only relies on caregivers’ rating of PWD’s dependence on the ADL, IADL, and symptoms on the RMBPC, it can be applied across all diagnoses of dementia. Secondly, through this new strategy, caregivers can easily figure out the stage of the PWD they are taking caring of, which provides the foundation for the group tailoring of support towards the caregiver. Given that there were successful attempts of organizing caregiving considerations based on stages of Alzheimer’s disease [44, 45], and qualitative evidence on caregiving experiences revealed similar key themes and caregiving needs despite caring for PWD with different diagnosis [7, 9, 46, 47], we believe that caregiving consideration of other types of dementia could also be organized similarly. For example, caregivers taking care of PWD in the early stage mostly need to deal with issues such as role transition and acceptance, and advice on dealing with PWDs’ memory and behavior problems; while for PWD at mild stage, priority should be on tips for handling PWDs’ memory and behavior problems and the emerging functional dependence. Thus, service or support providers could tailor their training and support for caregivers based on the typical needs of each specific stage the PWD belongs to. Such customization has a huge potential in e-health/m-health [17], in the sense that service providers could organize all information or resources based on stage of PWD, so that caregivers could specify which stage of dementia their relative is in and then receive the most pertinent support without an overload of information, which can lead to better outcomes for both the providers and the PWD.

Strengths and limitations

To the best of our knowledge, this is the first study in Asia which had explored the possibility of staging dementia purely based on caregiver reported PWD symptoms. It is also the first study that aimed to provide the screening cut-offs for each stage based on the tools used to assess PWD symptoms. With these cut-offs, caregivers could easily determine the stage of the PWD they are caring for. Since this new strategy does not require any specific training, it has a huge potential to be widely and easily used. A previous study also staged dementia based on patient symptoms; however, it relied on artificial intelligence and failed to provide a practical and straightforward solution for caregivers to figure out the PWDs’ stage by themselves [21]. For future studies, researchers could focus on investigating the typical challenges faced by caregivers when they are taking care of PWD at specific stages so that tailored support could be provided to caregivers depending on PWDs’ stage.

The study findings should be viewed with the following limitations in mind. First of all, the study sample comprised only primary informal caregivers of PWD recruited through convenience sampling in Singapore, and they voluntarily chose to join the study. This might cause self-selection bias and might limit the applicability of the new staging strategy among caregivers and PWDs elsewhere. More studies on the generalizability of this new strategy are needed. Secondly, the data were collected through interviewer-administered questionnaire reported by caregivers, and this may have led to social desirability bias. The rating might be influenced by caregivers’ own mental health status as well. Thirdly, readers should bear in mind that the diagnosis duration could be shortened by potential late diagnosis of dementia, and the deterioration of PWD’s functional status could be caused by the comorbid conditions of PWD. Fourthly, we failed to collect the clinical diagnosis of the PWD, and as a result, are unable to provide it as a reference, Lastly, the LCA classified the PWD into four different classes based on patterns of caregivers reported patient symptoms; however, this classification might differ from those derived from clinical assessments. Since we failed to collect the clinical stage information, we were unable to provide such comparisons. In this case, it will be ideal for future studies to compare this staging strategy with other clinical assessment tools to cross-validate it. Nonetheless, these should not lessen the novelty, the operability, and the huge potential of this new staging strategy, and this new strategy is a good complement of existing clinical assessment tools in terms of better supporting informal dementia caregivers.

Conclusion

The current study confirmed that caregiver reported patient symptoms could be used to classify PWD into different stages. PWD under each stage had their own symptom patterns, with the mild and moderate stages showing more memory and behavior problems and, as a result posed more stress to their caregivers. Further analyses generated cut-off values on functional dependence (i.e., ADL + IADL) and level of memory and behavior problems (i.e., RMBPC) separately. And these two cut-off values, when used together, could assign 90.8% of the PWD into the right stages derived from the LCA. The new staging strategy is a good complement to existing dementia clinical assessment tools in terms of better supporting informal caregivers.

29 Nov 2019

PONE-D-19-30761

Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

PLOS ONE

Dear Dr Yuan,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

While both reviewers saw merit in the manuscript, a number of issues were raised.  Of note, the skewed nature of the sample population is not thoroughly discussed. The remaining queries from the reviewers are clearly expressed.

We would appreciate receiving your revised manuscript by Jan 13 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

César Leal-Costa, Ph. D

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please refer to any sample size calculations performed prior to participant recruitment. If these were not performed please justify the reasons. Please refer to our statistical reporting guidelines for assistance (https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting).

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript reports results of staging Dementia based on Caregiver Reported Patient Symptoms: implications from a Latent Class Analysis.

Generally, it aims to stage dementia based on caregiver reported symptoms of persons with dementia. Optimal cut-offs for the assessment tools used were also calculated. The paper is an interesting study that evaluates several specific questions. However, in my opinion there are several aspects should be revised to improve the explanatory power of the manuscript as noted below.

Important:

*** Do not use more than 5 abbreviations in the manuscript. After reducing the number of abbreviations, create a table that includes the abbreviations used.

*** Materials and Methods, Results and Discussion: to facilitate reading I recommend to the authors include a paragraph for each objective.

Abstract

Results: Include the main results and p-values if there was significance.

Conclusion: Delete last sentence and include it in the last part of Discusion.

Materials and Methods

a) Participants and Procedures. To create a paragraph entitled: Ethics approval and consent to participate (or similar). I recommend this text:

The protocol was approved by xxxxxx (city, country; with reference number xxxxxxxxxx). The study was developed in accordance with the precepts of the Declaration of Helsinki. All patients signed the corresponding written informed consent prior to their enrolment and appropriate measures were taken to ensure data privacy.

b) Measurements

1. Lines 120-122: It is very important to justify why were used 2 scales from 1963 and 1969 respectively to know functional status, memory and behavior problems. Please, explain the reasons why more recent scales were not used.

2. Line 132: To include reference for the checklist RMBPC.

3. Lines 139-140: To explain the reason to use the 20-item Centre for Epidemiological Study Scale from 1977. Please, explain the reason why more recent scale was not used.

4. Are the used scales validated in Singapur?

c) Data analysis

Modify this section: One paragraph for every step performed.

Results

Table 1:

a) No simple size calculations are referred to in this manuscript. Besides, the skewed nature of the simple populaton is not throughly discussed.

b) To diference between 1) mean and frequency 2) SD and percentage.

c) Include a column with p-values.

d) At the table botton include the meaning of SGD and PWD.

e) Tables 2, 3 and 5 include: a) p-values and 2) the meaning of the abreviations.

Discussion

a) First paragraph: clearly include which are the main findings.

b) I recommend a paragraph for every objective and compare the main results with previous studies.

c) Include a paragraph that includes the main strengths and weaknesses of the study.

Reviewer #2: Congratulations on the excellent work. Visiting the important role of informal caregivers is a challenge. I think your study helps to empower the figure of the caregiver of people with dementia.

However, some aspects can be improved when reporting your work in the manuscript.

* Abstract:

I recommend deleting the following text from the "objective" section: "Optimal cut-offs for the assessment tools used were also calculated. “The writing of a research objective should not include a sentence about a result or method that has been performed, but what is intended to be discovered with the study.

* The introduction section is well structured, with a good justification for the study and an adequate background.

* I understand that no probabilistic sampling has been performed and this should be explained both in the method section and in the limitations of the study in the discussion section.

* More specific information about the ethical issues of the study would be necessary. The ethical aspects are not adequately explained and are summarized in that it was approved by a committee.

* The results are adequately discussed in the manuscript. The limitations are exposed and that is a strength in scientific research. Congratulations

* The possible future lines of research have already been set out in the discussion section. It is not necessary to repeat them in the conclusion section, so you would delete that text from one section or the other.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Dec 2019

6th Dec 2019

MS ID: PONE-D-19-30761

MS title: Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

Dear reviewers,

Thank you very much for the review and also for the valuable comments and suggestions. Our point-by-point responses are provided below and in bold for easy reference, and highlighted with track changes in the revised manuscript.

We hope that these changes are satisfactory. Thank you very much again in advance!

Sincerely,

Corresponding author

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=http%3a%2f%2fwww.journals.plos.org%2fplosone%2fs%2ffile%3fid%3dwjVg%2fPLOSOne%5fformatting%5fsample%5fmain%5fbody.pdf&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-02f1250312cbd8cc3126c5bcb57b48446f45e8ae and https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=http%3a%2f%2fwww.journals.plos.org%2fplosone%2fs%2ffile%3fid%3dba62%2fPLOSOne%5fformatting%5fsample%5ftitle%5fauthors%5faffiliations.pdf&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-cf0dcc3759a9e88581514f7555151ea3ba18bb04

RESPONSE: The manuscript is formatted following PLOS ONE style template.

2. Please refer to any sample size calculations performed prior to participant recruitment. If these were not performed please justify the reasons. Please refer to our statistical reporting guidelines for assistance (https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=https%3a%2f%2fjournals.plos.org%2fplosone%2fs%2fsubmission%2dguidelines.%23loc%2dstatistical%2dreporting&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-bc31dbd4e458bf3b4e0e8dc40582663843766e1d).

RESPONSE: We did calculate the sample size prior to initiating the study. However, since the calculation was based on the main project outcome (i.e. prevalence of depression), the current study is an exploratory analysis based on the sample collected, which is different from the main project outcome. In this case, we thought it would be misleading if we include the sample size calculation (based on main project outcome) in the manuscript.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=http%3a%2f%2fjournals.plos.org%2fplosone%2fs%2fdata%2davailability%23loc%2dunacceptable%2ddata%2daccess%2drestrictions&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-60195c19acf64d4762b069957f2ef692d6956ca7.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=http%3a%2f%2fwww.bmj.com%2fcontent%2f340%2fbmj.c181.long&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-5482821ba37a53710fcbddbac167cae3ed426272 for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see https://imsva91-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=http%3a%2f%2fjournals.plos.org%2fplosone%2fs%2fdata%2davailability%23loc%2drecommended%2drepositories&umid=4913B737-987A-2E05-A0D2-C3CB71FD25E2&auth=6e3fe59570831a389716849e93b5d483c90c3fe4-1305d747dc89a01960ad75d3185e772a533fd7de.

We will update your Data Availability statement on your behalf to reflect the information you provide.

RESPONSE: The authors' funding agency or government law only permits sharing of human participant data with researchers with whom they have a written agreement. The restrictions have been imposed by our Institutional Review Board (IRB) and Institutional Committee (NHG Domain Specific Review Board and IMH Clinical Research Committee). Our IRB guidelines suggest that a Research Collaboration Agreement (RCA) be signed with collaborating parties. However, data sharing with clear research purposes are available upon request to the senior author. The contact details are included in the cover letter to the editor.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know ________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript reports results of staging Dementia based on Caregiver Reported Patient Symptoms: implications from a Latent Class Analysis.

Generally, it aims to stage dementia based on caregiver reported symptoms of persons with dementia. Optimal cut-offs for the assessment tools used were also calculated. The paper is an interesting study that evaluates several specific questions. However, in my opinion there are several aspects should be revised to improve the explanatory power of the manuscript as noted below.

RESPONSE: Thank you very much!

Important:

*** Do not use more than 5 abbreviations in the manuscript. After reducing the number of abbreviations, create a table that includes the abbreviations used.

RESPONSE: We would like to thank the reviewer for this comment. According to the 6th APA guideline, it suggested that abbreviations should be used when the term appears at least three times in the manuscript, and there are no hard lines regarding how many abbreviations are too many. We have revised the manuscript following this guideline, keeping those abbreviations that appears at least three times. And as the reviewer suggested, we have created a list of abbreviations in the end of the manuscript. Please refer to the revised manuscript for more details.

*** Materials and Methods, Results and Discussion: to facilitate reading I recommend to the authors include a paragraph for each objective.

RESPONSE: We are sorry that we are not quite sure about what exactly the reviewer is suggesting. In fact, the current manuscript is organized the same way as all other published articles in the journal. May we request that this point be further clarified with an example from any other articles?

Abstract

Results: Include the main results and p-values if there was significance.

RESPONSE: For the latent class analysis and the accuracy statistics in the results section of the abstract, both didn’t involve any statistical comparisons, thus there were no relevant p-values. We had made some necessary changes to the results section in the revisions to make it clearer.

Conclusion: Delete last sentence and include it in the last part of Discusion.

RESPONSE: The manuscript has been revised accordingly.

Materials and Methods

a) Participants and Procedures. To create a paragraph entitled: Ethics approval and consent to participate (or similar). I recommend this text:

The protocol was approved by xxxxxx (city, country; with reference number xxxxxxxxxx). The study was developed in accordance with the precepts of the Declaration of Helsinki. All patients signed the corresponding written informed consent prior to their enrolment and appropriate measures were taken to ensure data privacy.

RESPONSE: In fact, we did have a separate paragraph on the study ethics (i.e. the last paragraph of the ‘Participants and Procedures’ section. The ethics statement has been revised accordingly as suggested by the reviewer.

b) Measurements

1. Lines 120-122: It is very important to justify why were used 2 scales from 1963 and 1969 respectively to know functional status, memory and behavior problems. Please, explain the reasons why more recent scales were not used.

RESPONSE: We would like to apologize for not clarifying this in the original manuscript. There are several reasons: first of all, these two scales are both very classical, and there are plenty of articles demonstrating their reliability. Secondly, it has been validated in Singapore (ADL [1, 2] and IADL [3]), and also been used among caregivers in Singapore before, showed good internal reliability [4]. Lastly, it is short and easily administered among dementia caregivers who usually have an average age of 50 years and more.

2. Line 132: To include reference for the checklist RMBPC.

RESPONSE: In fact, the reference of RMBPC was included in the manuscript – in the next sentence while describing the checklist. To avoid confusion, we have added in the citation in line 132 (now line 135) as suggested by the reviewer.

3. Lines 139-140: To explain the reason to use the 20-item Centre for Epidemiological Study Scale from 1977. Please, explain the reason why more recent scale was not used.

RESPONSE: There are a few reasons: 1) according the American Psychological Association [5], the CES-D provides cut-off scores that aid to identify individuals at risk for clinical depression, with good sensitivity and specificity and high internal consistency [6], and it is sensitive to difference between caregivers and non-caregivers [7], and it can be used across wide age ranges [6]; 2) it has been validated in Singapore before [8]; 3) a previous Singapore study suggested that it has good validity and utility in detecting depression among family caregivers of PWD [9].

4. Are the used scales validated in Singapur?

RESPONSE: Please refer to our responses to comments (answers to b) measurement – point 1 and point 3)

c) Data analysis

Modify this section: One paragraph for every step performed.

RESPOSE: To make it clearer to the audience, we use numbered list to present the four steps of the analyses. Please refer to the revised manuscript for the details.

Results

Table 1:

a) No simple size calculations are referred to in this manuscript. Besides, the skewed nature of the simple populaton is not throughly discussed.

b) To diference between 1) mean and frequency 2) SD and percentage.

c) Include a column with p-values.

d) At the table botton include the meaning of SGD and PWD.

e) Tables 2, 3 and 5 include: a) p-values and 2) the meaning of the abreviations.

RESPONSE: a) We did calculate the sample size prior to initiating the study. However, since the calculation was based on the main project outcome (i.e. prevalence of depression), the current study is an exploratory analysis based on the sample collected, which is different from the main project outcome. In this case, we thought it would be misleading if we include the sample size calculation (based on main project outcome) in the manuscript.

Our study sample is informal dementia caregivers, which in nature would differ from the general population. However, this doesn’t necessarily suggest that our sample is skewed. In fact, we compared our sample characteristics with the sample characteristics of dementia caregivers from a previous Singapore Dementia Caregiver Profile study (SDCP) [10], and we found very similar statistics. For example, average age: current study vs. SDCP – 55.7 (11.8) vs. 52.6 (11.1); female proportion: current study vs. SDCP – 75.2% vs. 73.6; relationship with PWD: current study vs. SDCP – 15.3% vs. 18% (spouse), 72.3% vs. 69.2% (children), and 12.4% vs. 12.8% (others); Living with PWD: current study vs. SDCP – 70.2% vs. 79.2%; and having a domestic maid: current study vs. SDCP – 57.1% vs. 52.8%. But we also understand the self-selection bias might affect the generalizability of the study findings, thus this limitation was added.

b) Table 1 was re-formatted accordingly.

c) Table 1 includes only the descriptive statistics, p-value is not applicable here.

d) The meaning of SGD and PWD were added as suggested.

e) Table 2 is the latent class analysis results, table 3 and table 5 both included descriptive statistics, all these three tables don’t involve p-values.

As suggested by the reviewer, we have added in the meaning of all abbreviations as the notes of the tables.

Discussion

a) First paragraph: clearly include which are the main findings.

b) I recommend a paragraph for every objective and compare the main results with previous studies.

c) Include a paragraph that includes the main strengths and weaknesses of the study.

RESPONSE: a) Our main finding is that staging dementia based on caregivers reported is a viable approach which had been included in the first paragraph of the discussion section. However, we do feel the original first paragraph included too many details; thus, we have cut it into two paragraphs so that the new first paragraph is focused on the main study findings, and the following paragraph is focused on the details.

b) After the above revisions, the structure of the discussion section follows exactly what the reviewer suggested. In fact, this is the first study which uses such analytical strategies among informal dementia caregivers. There was one study which used artificial intelligence to stage dementia based on caregiver reported patient symptoms, although we had included it in the original manuscript, since the two use quite different methodologies, we felt it was not appropriate for us to directly compare them. Other than that, we cannot identify any other similar previous studies for us to compare.

c) We are sorry for the confusion. In fact, the last two paragraphs of the discussion section were the study strengths and limitations. In the revision, we have added a section named ‘Strengths and Limitations’ to avoid confusion.

Reviewer #2: Congratulations on the excellent work. Visiting the important role of informal caregivers is a challenge. I think your study helps to empower the figure of the caregiver of people with dementia.

However, some aspects can be improved when reporting your work in the manuscript.

RESPONSE: Thanks!

* Abstract:

I recommend deleting the following text from the "objective" section: "Optimal cut-offs for the assessment tools used were also calculated. “The writing of a research objective should not include a sentence about a result or method that has been performed, but what is intended to be discovered with the study.

RESPONSE: We would like to thank the reviewer for this comment, the abstract was revised accordingly.

* The introduction section is well structured, with a good justification for the study and an adequate background.

RESPONSE: Thanks!

* I understand that no probabilistic sampling has been performed and this should be explained both in the method section and in the limitations of the study in the discussion section.

RESPONSE: We are sorry for the negligence. We have added in some descriptions on the non-probabilistic sampling in the methods and study limitations sections, please refer to the revised manuscript for more details.

* More specific information about the ethical issues of the study would be necessary. The ethical aspects are not adequately explained and are summarized in that it was approved by a committee.

RESPONSE: We appreciate the reviewer for this comment. More details regarding the ethics were added as suggested.

* The results are adequately discussed in the manuscript. The limitations are exposed and that is a strength in scientific research. Congratulations

RESPONSE: Thanks!

* The possible future lines of research have already been set out in the discussion section. It is not necessary to repeat them in the conclusion section, so you would delete that text from one section or the other.

RESPONSE: As suggested by the reviewer, the future lines of research in the ‘Conclusion’ section was removed to avoid duplicity.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

References

1. Sien NY, Jung HJTSFP. Assessment of the six activities of daily living in adults. 2014;40(4):26-36.

2. Xie F, Li S-C, Roos EM, Fong K-Y, Lo N-N, Yeo S-J, et al. Cross-cultural adaptation and validation of Singapore English and Chinese versions of the Knee injury and Osteoarthritis Outcome Score (KOOS) in Asians with knee osteoarthritis in Singapore. 2006;14(11):1098-103.

3. Ng T-P, Niti M, Chiam P-C, Kua E-HJTJoGSABS, Sciences M. Physical and cognitive domains of the instrumental activities of daily living: validation in a multiethnic population of Asian older adults. 2006;61(7):726-35.

4. Østbye T, Malhotra R, Malhotra C, Arambepola C, Chan AJJoGSBPS, Sciences S. Does support from foreign domestic workers decrease the negative impact of informal caregiving? Results from Singapore survey on informal caregiving. 2013;68(4):609-21.

5. The American Psycholiglcal Association. Center for Epidemiological Studies-Depression [cited 2019 2 Dec]. Available from: https://www.apa.org/pi/about/publications/caregivers/practice-settings/assessment/tools/depression-scale.

6. Lewinsohn PM, Seeley JR, Roberts RE, Allen NBJP, aging. Center for Epidemiologic Studies Depression Scale (CES-D) as a screening instrument for depression among community-residing older adults. 1997;12(2):277.

7. Pinquart M, Sörensen SJP, aging. Differences between caregivers and noncaregivers in psychological health and physical health: a meta-analysis. 2003;18(2):250.

8. Stahl D, Sum CF, Lum SS, Liow PH, Chan YH, Verma S, et al. Screening for depressive symptoms: validation of the center for epidemiologic studies depression scale (CES-D) in a multiethnic group of patients with diabetes in Singapore. 2008;31(6):1118-9.

9. Ying J, Yap P, Gandhi M, Liew TMJD, disorders gc. Validity and utility of the center for epidemiological studies depression scale for detecting depression in family caregivers of persons with dementia. 2019;47(4-6):323-34.

10. Tew CW, Tan LF, Luo N, Ng WY, Yap PJD, Disorders GC. Why family caregivers choose to institutionalize a loved one with dementia: a Singapore perspective. 2010;30(6):509-16.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

24 Dec 2019

PONE-D-19-30761R1

Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

PLOS ONE

Dear Dr Yuan,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The revised manuscript is much improved and has addressed many of the concerns of the reviewers. However, some issues remain outstanding and warrant attention. Notably, Reviewer 1 raises the query about the table 1.

We would appreciate receiving your revised manuscript by Feb 07 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

César Leal-Costa, Ph. D

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: First of all I would like to thank the authors of this manuscript for having addressed the issues raised previously (I think the article has been greatly improved, in my opinion).

I just have to make two points:

1) Table 1: in order that the results can be compared to other studies I recommend converting the currency "Singapore dollars" to "Euros or American dollars".

2) Strengths and limitations: delete paragraph 1 and put it at the end of the Discussion.

Thank you,

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Dec 2019

27th Dec 2019

MS ID: PONE-D-19-30761R1

MS title: Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

Dear reviewers,

Thank you very much for the review and also for the valuable comments and suggestions. Our point-by-point responses are provided below and in bold for easy reference, and highlighted with track changes in the revised manuscript.

We hope that these changes are satisfactory. Thank you very much again in advance!

Sincerely,

Corresponding author

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

RESPONSE: Thanks!

________________________________________

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

________________________________________

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

________________________________________

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

________________________________________

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

________________________________________

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: First of all I would like to thank the authors of this manuscript for having addressed the issues raised previously (I think the article has been greatly improved, in my opinion).

RESPOSE: Thank you very much!

I just have to make two points:

1) Table 1: in order that the results can be compared to other studies I recommend converting the currency "Singapore dollars" to "Euros or American dollars".

RESPONSE: We would like to thank the reviewer for this comment. In the revision, we have added in the income level as converted into US dollars to enable easier comparisons with international studies.

2) Strengths and limitations: delete paragraph 1 and put it at the end of the Discussion.

RESPONSE: We would like to clarify – in fact that section is about strengths and limitations. The first paragraph is on strengths and the second paragraph is on limitations. In this case, it makes more sense for us to keep the current structure.

Thank you,

Reviewer #2: (No Response)

________________________________________

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Jan 2020

Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

PONE-D-19-30761R2

Dear Dr. Yuan,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

César Leal-Costa, Ph. D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

6 Jan 2020

PONE-D-19-30761R2

Staging Dementia based on Caregiver Reported Patient Symptoms: Implications from a Latent Class Analysis

Dear Dr. Yuan:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. César Leal-Costa

Academic Editor

PLOS ONE