Factors contributing to decisions for giving treatment and referrals to patients with cognitive complaints in chat-based telemedicine services in Indonesia

Abstract

Introduction: Chat-based telemedicine has become a popular method for consulting cognitive complaints in Indonesia, especially during the COVID-19 pandemic. This study explores patient characteristics and their correlation with physicians’ decisions on treatment and referral. Methods: A retrospective cross-sectional study analyzed secondary data from patient chat databases on Indonesian telemedicine platforms between March 2020 and December 2021. Bivariate and multivariable analysis were conducted. Results: Out of 1,016 patients (0.4% of 2,199,527 users), most were females under 40. Clinicians were more likely to provide pharmacological treatment for patients aged ≥60, females, or those with forgetfulness complaints or a history of cognitive disorders (p < 0.01). Referral likelihood increased when caregivers filed complaints or prior diagnoses existed (p < 0.01). Discussion: Clinicians’ decisions in telemedicine were influenced by age, gender, complaint type, and caregiver involvement. Further research is needed to improve telemedicine for cognitive care.

Keywords

cognitive complaints forgetfulness chat telemedicine

Background

Cognitive function is a broad term for the mental processes involved in acquiring knowledge, manipulating information, and reasoning. Cognitive function includes the domains of perception, memory, learning, attention, decision-making, and language skills.¹ The decline in cognitive function in the advanced stages can be diagnosed as dementia according to the Diagnostic and Statistical Manual of Mental Disorders V (DSM-V). However, 50-67% of dementia conditions were undiagnosed in patients aged over 65 years^2–4 With the increasing prevalence of dementia, intervention in the form of early detection and appropriate and immediate follow-up is needed. Early detection can be predicted by analyzing the patient’s cognitive complaints record, especially complaints of forgetfulness or memory disorders which are the most common cognitive complaints of dementia patients (54%).⁵

Several studies have explored the use of telemedicine for diagnosing patients, including those with cognitive impairment. It can support history taking, cognitive and language assessment, and follow-up management.^2,6 Telemedicine delivers healthcare through electronic communication, including phone calls, video conferencing, web platforms, and chat applications. In Indonesia, most chat-based telemedicine services rely on text messaging for history taking, with video calls added when needed for further assessment.⁷

Telemedical service chat-based with internet and smartphone access allow patients to consult with doctors or other paramedics without having to meet in person.⁸ Telemedicine helps overcome distance and geographic barriers, improving access to healthcare without the need for face-to-face contact. It offers a wide range of services and can expand care to large populations. However, limitations remain, including differing expectations between patients and clinicians, restricted physical examination capabilities, and concerns about data security and confidentiality.^7,9

During the COVID-19 pandemic in Indonesia, many memory and high-function clinics suspended face-to-face services, leading to a decline in consultations for cognitively impaired patients and a risk of worsening illness. Telemedicine emerged as an effective alternative, enabling diagnosis, monitoring, management guidance, and referral for patients with memory complaints despite the limitations of in-person visits. To date, no research has examined the use of telemedicine for cognitive function in Indonesia, making this study the first of its kind. Globally, there is also a lack of research analyzing chat-based telemedicine for cognitively impaired patients using big data approaches. Such research is increasingly important, particularly during and after the COVID-19 pandemic, when health services have had to adapt to chat-based telemedicine to maintain care access. Telemedicine can capture outpatient complaints, support early detection, and provide timely management for patients with cognitive concerns.⁶ This study describes the characteristics of patients with cognitive complaints who used chat-based telemedicine services in Indonesia and the factors that affect the physicians’ treatment and referral decisions.

Methods

This was a retrospective cross-sectional study using secondary data from the chat databases of four Indonesian telemedicine providers (Halodoc, Alodokter, Good Doctor, and Milvik) (Figure 1). Formal permission was obtained from each provider, and ethical approval was granted by the Health Research Ethics Committee.

Figure 1.

Flowchart of study procedures for data collection and analysis.

Chat-based teleconsultations begin when a patient or caregiver selects a consultation topic (e.g., forgetfulness) and is connected to a general practitioner (GP) or neurologist. The clinician conducts structured text-based history taking, exploring symptom onset, severity, daily impact, and red-flag features. When appropriate, the GP provides preliminary advice, prescribes treatment in accordance with regulations, or issues a referral to a neurologist or hospital.

Data were collected by total sampling from March 2020 to October 2021. For this study, the term “patients” refers to adult users (≥18 years) or their caregivers who submitted cognitive complaints which includes forgetfulness, memory loss, difficulty concentrating, word-finding difficulty, and related symptoms. Cases with incomplete demographic data or interrupted consultations were excluded.

Data retrieval

Secondary data were obtained by searching the chat database of each participating telemedicine provider using keywords such as “forgetfulness,” “memory loss,” “difficulty concentrating,” “difficulty focusing,” “difficulty getting words out,” “getting lost easily,” “difficulty understanding conversation,” and “difficulty finding objects.” All retrieved conversations were reviewed, and data were collected from records that met the inclusion criteria.

The independent variables, coded as categorical data, included: age (<40, 40–59, ≥60 years), sex (male or female), previous diagnosis of cognitive impairment (yes or no), complainant (patient or caregiver), responder type (general practitioner, neurologist, other specialist), patient’s cognitive complaint (forgetfulness/memory loss vs other symptoms), and whether an early cognitive screening test was performed (yes or no). The dependent variables were treatment decision and referral decision.

Data analysis

Data analysis was performed using IBM SPSS Statistics Version 20. Univariate analysis was used to describe patient characteristics as frequencies and percentages. Bivariate analysis was conducted to examine the association between patient characteristics and the two outcome variables (treatment decision and referral decision). Chi-square tests were used for categorical variables. When the expected cell count in the contingency table was too low to meet the assumptions of the Chi-square test, we used Fisher’s exact test instead. Variables with p < 0.25 in bivariate analysis were included in a multivariable logistic regression model to identify independent predictors. Results from logistic regression were reported as odds ratios (ORs) with 95% confidence intervals (CI), and p < 0.05 was considered statistically significant.

Result

Demographic data

From 1,801,996 consultations, 1,016 (4 per 10,000) involved cognitive complaints. Most patients were <40 years old (52.6%) and female (62.5%) (Table 1). Complaints were mainly submitted by caregivers (54.3%), and only 9.4% had a prior diagnosis of cognitive impairment. General practitioners were the most frequent responders (58.9%), followed by neurologists (17.5%). Most patients did not receive treatment (70.1%), whereas 81.1% were referred to a specific specialist or hospital (Table 2), suggesting chat-based telemedicine mainly served as a triage tool.

Table 1.

Patient characteristics.

Patient characteristics		N	%
Age	<40 years	534	52.6
	40-59 years	173	17.0
	≥60 years	309	30.4
Sex	Male	381	37.5
Sex	Female	635	62.5
Complainant	Patient themselves	464	45.7
Complainant	Caregiver	552	54.3
Previous diagnosis of cognitive impairment	Yes	96	9.4
Previous diagnosis of cognitive impairment	No	920	90.6
Responder	General practitioner	598	58.9
	Neurologist	178	17.5
	Other specialist	240	23.6
Disease complaint	Forgetfulness/memory loss	784	73.6
Disease complaint	Other cognitive symptoms	268	26.4
Early detection test for cognitive impairment	Yes	1	0.1
Early detection test for cognitive impairment	No	1015	99.9

Table 2.

Patient consultation outcome.

Consultation outcome		N	%
Treatment	No treatment	707	70.1
	Given pharmacological treatment	301	29.9
	Given pharmacological and non-pharmacological treatment		17.3
Cognitive impairment counseling	Yes	834	82.7
Cognitive impairment counseling	No	174	17.2
Referral decision	No referralact	201	19.9
	Referral to neurologist	321	31.8
	Referral to psychiatrist	138	13.7
	Referral to internist/geriatric consultant	10	1
	Referral to minimum two specialist	14	1.4
	Referral to hospital	324	32.1

Bivariate analysis

Bivariate analysis (Table 3) showed that treatment was significantly associated with age ≥60 years (p < 0.001), female sex (OR = 1.62, 95% CI 1.22–2.16, p < 0.01), caregiver-submitted complaints (OR = 4.72, 95% CI 3.55–6.49, p < 0.01), neurologist or specialist responder (p < 0.01), and forgetfulness/memory loss (OR = 2.02, 95% CI 1.51–2.71, p < 0.01). Intergroup comparisons (Table 4) confirmed that neurologists were most likely to provide treatment.

Table 3.

Bivariate analysis of patient characteristic on treatment decision.

Patient characteristics		Treatment decision		p value	Odds ratio (95% CI)
Patient characteristics		No (N (%))	Yes (N (%))	p value	Odds ratio (95% CI)
Age	<40 years	471 (88.2)	63 (11.8)	<0.01	NA
	40-59 years	122 (70.5)	51 (29.5)
	≥60 years	118 (38.2)	191 (61.8)
Sex	Male	290 (76.1)	91 (23.9)	<0.01	1.62 (1.22-2.16)
Sex	Female	421 (66.3)	214 (33.7)
Complainant	Patient themselves	399 (86)	65 (14)	<0.01	4.722 (3.55-6.49)
Complainant	Caregiver	312 (56.5)	240 (43.5)
Previous diagnosis of cognitive impairment	Yes	64 (66.7)	32 (33.3)	0.457	0.84 (0.54-1.32)
Previous diagnosis of cognitive impairment	No	647 (70.3)	273 (29.7)
Responder	General practitioner	561 (93.8)	37 (6.2)	<0,001	NA
	Neurologist	38 (21.3)	140 (78.7)
	Other specialist	112 (46.7)	128 (53.3)
Disease complaint	Forgetfulness/memory loss	554 (74.1)	194 (25.9)	<0.001	2.019 (1.51-2.71)
Disease complaint	Other cognitive symptoms	157 (58.6)	111 (41.4)
Early detection test for cognitive impairment	Yes	0 (0)	1 (100)	0.300^a	NA
Early detection test for cognitive impairment	No	711 (70)	305 (30)

^aFisher’s Test.

Bold values indicate statistically significant associations at p < 0.05.

Table 4.

Bivariate analysis between groups of patient age and respondents on treatment decision.

Patient characteristics		Treatment decision		p value	Odds ratio (95% CI)
Patient characteristics		No (N (%))	Yes (N (%	p value	Odds ratio (95% CI)
Patient age
40-59 years old vs ≥ 60 years	40-59 years old	122 (70.5)	51 (29.5)	<0.001	3.87 (2.60-5.77)
40-59 years old vs ≥ 60 years	≥60 years old	118 (38.2)	191 (61.8)	<0.001	3.87 (2.60-5.77)
<40 years old vs 40-59 years old	<40 years old	471 (88.2)	63 (11.8)	<0.001	3.13 (2.06-4.75)
<40 years old vs 40-59 years old	40-59 years old	122 (70.5)	51 (29.5)	<0.001	3.13 (2.06-4.75)
<40 years old vs ≥ 60 years old	<40 years old	471 (88.2)	63 (11.8)	<0.001	12,10 (8.54-17.16)
<40 years old vs ≥ 60 years old	≥60 years old	118 (38.2)	191 (61.8)	<0.001	12,10 (8.54-17.16)
Responder
Neurologist vs other specialist	Neurologist	38 (21.3)	140 (78.7)	<0.001	0.310 (0.20-0.48)
Neurologist vs other specialist	Other specialist	112 (46.7)	128 (53.3)	<0.001	0.310 (0.20-0.48)
General practitioner vs neurologist	General practitioner	561 (93.8)	37 (6.2)	<0.001	55.861 (34.25-91.10)
General practitioner vs neurologist	Neurologist	38 (21.3)	140 (78.7)	<0.001	55.861 (34.25-91.10)
General practitioner vs other specialist	General practitioner	561 (93.8)	37 (6.2)	<0.001	17.33 (11.41-26.33)
General practitioner vs other specialist	Other specialist	112 (46.7)	128 (53.3)	<0.001	17.33 (11.41-26.33)

Bold values indicate statistically significant associations at p < 0.05.

For referral decisions (Table 5), age and responder type were significant (p < 0.001). Patients ≥60 years were more likely to be referred than those <40 years (OR = 0.529, p < 0.001) or 40–59 years (OR = 0.585, p = 0.022). Intergroup analysis (Table 6) showed the highest referral rates among general practitioners.

Table 5.

Bivariate analysis of patient characteristic on referral decision.

Patient characteristics		Referral decision		P value	Odds ratio (95% CI)
Patient characteristics		No (N (%))	Yes (N (%))	P value	Odds ratio (95% CI)
Age	<40 years	88 (16.5)	446 (83.5)	0.001	NA
	40-64 years	31 (17.9)	142 (82.1)
	≥60 years	84 (27.2)	225 (72.8)
Sex	Male	77 (20.2)	304 (79.8)	0.887	1.02 (0.75-1.41)
Sex	Female	126 (19.8)	509 (80.2)
Complainant	Patient themselves	84 (18.1)	380 (81.9)	0.170	0.80 (0.60-1.10)
Complainant	Caregiver	119 (21.6)	433 (78.4)
Previous diagnosis of cognitive impairment	Yes	43 (44.8)	53 (55.2)	<0.001	3.85 (2.49-5.97)
Previous diagnosis of cognitive impairment	No	160 (17.4)	760 (82.6)
Responder	General practitioner	74 (12.4)	524 (87.6)	<0.001	N/A
	Neurologist	66 (37.1)	112 (62.9)
	Other specialist	63 (26.3)	177 (73.8)
Disease complaint	Forgetfulness/memory loss	149 (19.9)	599 (80.1)	0.936	0.99 (0.70-1.40)
Disease complaint	Other cognitive symptoms	54 (20.1)	214 (79.9)
Early detection test for cognitive impairment	Yes	0 (0)	1 (100)	1^a	N/A
Early detection test for cognitive impairment	No	203 (20)	813 (80)

^aFisher’s Test.

Bold values indicate statistically significant associations at p < 0.05.

Table 6.

Bivariate analysis between groups of patient age and respondents on referral decision.

Patient characteristics		Referral decision		p value	Odds ratio (95% CI)
Patient characteristics		No (N (%))	Yes (N (%))	p value	Odds ratio (95% CI)
Patient age
40-59 years old vs ≥ 60 years	40-59 years old	31 (17.9)	142 (82.1)	0.022	0.59 (0.37-0.93)
40-59 years old vs ≥ 60 years	≥60 years old	84 (27.2)	225 (72.8)	0.022	0.59 (0.37-0.93)
<40 years old vs 40-59 years old	<40 years old	88 (16.5)	446 (83.5)	0.660	0.904 (0.58-1.42)
<40 years old vs 40-59 years old	40-59 years old	31 (17.9)	142 (82.1)	0.660	0.904 (0.58-1.42)
<40 years old vs ≥ 60 years old	<40 years old	88 (16.5)	446 (83.5)	<0.001	0.53 (0.38-0.74)
<40 years old vs ≥ 60 years old	≥60 years old	84 (27.2)	225 (72.8)	<0.001	0.53 (0.38-0.74)
Responder
Neurologist vs other specialist	Neurologist	66 (37.1)	112 (62.9)	0.018	1.66 (1.09-2.52)
Neurologist vs other specialist	Other specialist	63 (26.3)	177 (73.7)	0.018	1.66 (1.09-2.52)
General practitioner vs neurologist	General practitioner	74 (12.4)	524 (87.6)	<0.001	0.24 (0.16-0.35)
General practitioner vs neurologist	Neurologist	66 (37.1)	112 (62.9)	<0.001	0.24 (0.16-0.35)
General practitioner vs other specialist	General practitioner	74 (12.4)	524 (87.6)	<0001	0.40 (0.27-0.58)
General practitioner vs other specialist	Other specialist	63 (26.3)	177 (73.7)	<0001	0.40 (0.27-0.58)

Bold values indicate statistically significant associations at p < 0.05.

Multivariable analysis

Independent predictors of treatment (Table 7) included age ≥60 years (aOR = 3.32, p < 0.01), female sex (aOR = 1.87, p < 0.01), neurologist (aOR = 47.51, p < 0.01) or other specialist responder (aOR = 9.88, p < 0.01), complaints other than forgetfulness/memory loss (aOR = 1.94, p < 0.01), and no prior diagnosis of cognitive impairment (aOR = 2.01, p = 0.02).

Table 7.

Multivariate analysis on treatment decision.

Variable	B	SE	p value	Adjusted OR	CI 95%
Age
<40 years old	Ref
40-59 years old	0.57	0.34	0.10	1.77	0.90-3.46
≥60 years old	1.20	0.37	< 0.01	3.32	1.62-6.79
Sex
male	Ref
female	0.62	0.20	<0.01	1.867	1.27- 2.75
Responder
General practitioner	Ref
Neurologist	3.86	0.30	<0.01	47.51	26.59- 84.88
Other specialist	2.29	0.27	<0.01	9.88	5.78-16.88
Complainant
Patient themselves	Ref
caregiver	−0.57	0.34	0.09	0.56	0.29-1.09
Disease complaint
Forgetfulness/memory loss	Ref
Other cognitive symptoms	0.66	0.20	< 0.01	1.94	1.29-2.91
Previous diagnosis of cognitive impairment
Yes	Ref
No	0.70	0.31	0.02	2.01	1.10-3.68

Bold values indicate statistically significant associations at p < 0.05.

Referral decisions (Table 8) were more likely with a prior diagnosis (aOR = 3.96, p < 0.01) and caregiver-submitted complaints (aOR = 2.26, p < 0.01) but less likely when consultations were handled by neurologists (aOR = 0.17, p < 0.01) or other specialists (aOR = 0.27, p < 0.01).

Table 8.

Multivariate analysis on referral decision.

Variable	B	SE	p value	Adjusted OR	CI 95%
Previous diagnosis of cognitive impairment
Yes	Ref
No	1.38	0.24	<0.01	3.96	2.47-6.35
Complainant
Patient themselves	Ref
caregiver	0.82	0.21	<0.01	2.26	1.49-3.43
Responder
General practitioner	Ref
Neurologist	−1.78	0.24	<0.01	0.17	0.11-0.27
Other specialist	−1.32	0.23	<0.01	0.27	0.17-0.42

Bold values indicate statistically significant associations at p < 0.05.

Discussion

In this study, 1,016 consultations with cognitive complaints were identified from 1,801,996 total consultations, representing a prevalence of 4 per 10,000. This figure is lower than reported in a systematic review, which found that the global prevalence of cognitive disorders ranges from 5.1% to 41% and the incidence from 22 to 76.8 per 1,000 person-years.¹⁰ These results differ from the global epidemiological data where the prevalence of dementia is 6.4%.¹¹ In addition, WHO also notes that there are 50 million people in the world who have dementia with 60% living in low to middle-income countries.¹² The low proportion of cognitive complaints observed in this study may reflect limited familiarity among Indonesian adults with using chat-based telemedicine to report memory concerns. Cultural perceptions that forgetfulness is a normal part of aging may also lead to underreporting and delayed medical attention. Among younger patients, forgetfulness is often attributed to depression rather than cognitive impairment, which may further inhibit diagnosis. Limited caregiver awareness can contribute to delayed recognition and reporting of cognitive problems. These findings are important because cognitive complaints can signal serious underlying conditions such as dementia, psychiatric disorders, or delirium.^13–15

This study found that most patients with cognitive complaints were under 40 years of age, followed by those aged ≥60 years. This distribution differs from WHO data, which report that 5–8% of the general population aged ≥60 years have dementia at any given time.¹² Data from Central Norway also show an incidence of early-onset dementia of 14.8 per 100,000 person-years, suggesting possible underrepresentation of older patients in our sample.¹⁶ This may reflect limited technology adoption among older adults, a finding similarly reported in U.S.-based telemedicine studies.¹⁷

Female patients accounted for 1.7 times more consultations than males. This contrasts with previous studies on early-onset dementia, which report little difference between sexes.^18,19 However, our finding is consistent with Hale’s study, which reported a slightly higher prevalence of cognitive impairment in females compared with males (21.2% vs 21.0%) and a higher prevalence of dementia in females (7.0% vs 5.4%).²⁰ This finding is also consistent with global data showing that dementia is more than twice as common in females as in males. In a recent study across 12 European populations, the prevalence of dementia ranged from 5.9% to 9.4%.¹¹ The higher proportion of females reporting cognitive complaints through telemedicine may reflect greater awareness and utilization of telemedicine services among women, a trend also observed in studies from the United States.¹⁷

Over half of the complaints (54.3%) were submitted by caregivers, while 45.7% were submitted by patients themselves. Caregiver involvement is likely higher because patients with cognitive impairment may have difficulty using chat-based telemedicine platforms and are often more familiar with hospital-based services. Additionally, patients may have limited awareness of their deficits or provide minimal information about their symptoms. In those with subjective cognitive complaints, coexisting depression or anxiety may further influence the accuracy and reliability of the reported complaints.^21,22 Cognitive complaints submitted by patients without cognitive impairment are often more accurate. Moreover, individuals with moderate to severe cognitive impairment may be unable to participate meaningfully in chat-based telemedicine consultations, limiting the quality of information obtained.

In this study, 9.4% of patients had previously been diagnosed with a cognitive disorder by a neurologist before their telemedicine consultation. The rise in telemedicine use for cognitive complaints during the COVID-19 pandemic likely reflects the greater vulnerability of cognitively impaired patients, especially older adults, who face higher risks of severe illness, ICU admission, and death when infected with COVID-19.^23–25 This concern may make caregivers hesitant to bring patients for in-person consultations. In such cases, application-based telemedicine that combines mobile messaging, video, and chat features can help address the needs of patients and caregivers. Internet-based tools can also provide additional support through electronic reminders, cognitive stimulation activities, and daily task management.^23,24 Beyond being cost-effective, remote telemedicine assessments can help protect cognitively impaired patients and their families from the risk of COVID-19 exposure in hospital settings.^23,26 Telemedicine has also supported the implementation of public health strategies during the COVID-19 pandemic.²⁷

Of the 1,016 patients, 73.6% reported forgetfulness or memory loss. Patients aged ≥60 years were the second largest age group with cognitive complaints, consistent with a descriptive study by Mesiano and Samino (2010) of 56 older adults (>60 years) in Jakarta, which reported that 59% experienced cognitive impairment, with forgetfulness being the most common complaint (54%).⁵ This also aligns with community-based studies reporting memory complaints prevalence of 25% to 50% in older adults.²⁸

In this study, 58.9% of cognitive complaints were managed by general practitioners, while only 17.5% were handled by neurologists. This reflects the triage process in chat-based telemedicine, where patients are typically directed to consult a general practitioner first, particularly when they are unsure which specialist to choose. Based on the consultation, the general practitioner may then refer the patient to a neurologist or other specialist as appropriate.

Most patients did not receive treatment (70.1%), either pharmacological or non-pharmacological. Bivariate analysis showed significant associations between treatment provision and age, sex, complainant, prior diagnosis of cognitive impairment, responder type, and type of complaint. In the multivariable model, sex, responder type, complaint type, and prior diagnosis remained significant predictors. Neurologists and other specialists were more likely to provide treatment compared with GPs, consistent with their role in definitive management. General practitioners tended to focus on initial recognition and referral, which may be due to limited familiarity with cognitive disorder therapies and regulatory restrictions on prescribing psychiatric medications via chat-based telemedicine.^29,30

Sex may have influenced treatment decisions, as most telemedicine users in this study were female, making women the largest group to receive treatment. Patients with a prior diagnosis were also more likely to receive treatment, likely because physicians continued established therapy. Clinicians may have been more confident treating cognitive complaints other than forgetfulness or memory loss, whereas the high volume of memory-related complaints may have prompted more referrals rather than direct treatment.^{13–15,30,31}

Referral decisions were common (81.1%), most often to neurologists (31.8%). Age was a significant predictor, with patients ≥60 years more likely to be referred, consistent with the higher dementia prevalence in this age group.^12,32 In multivariable analysis, prior diagnosis, responder type, and complainant were independent predictors of referral. Referrals were more frequently given to patients with prior diagnosis of cognitive impairment, often after being assessed by a general practitioner. A clear diagnosis facilitates referral for further examination or follow-up therapy to a neurologist, in line with dementia management practices in Indonesia.³⁰ Patients treated by a neurologist were less likely to receive a referral, as neurologists can manage and monitor non-emergency cases directly through telemedicine.³³ Complaints submitted by family members were more likely to result in referral, as they were generally clearer and more detailed, giving physicians greater confidence to make referral decisions.^21,22

The high proportion of referrals observed in this study aligns with the role of telemedicine as an initial point of consultation, focusing on prevention, early detection, and timely diagnosis before definitive treatment. Telemedicine is particularly valuable for elderly patients with cognitive complaints, allowing neurologist consultations and follow-up therapy, both pharmacological and non-pharmacological, without the need for face-to-face visits. This approach is especially beneficial during the COVID-19 pandemic and for patients in rural areas with limited access to neurologists, thereby improving continuity of care.^34,35

During the study period, cognitive assessment relied on history taking and patient- or caregiver-reported complaints, as no validated telemedicine-adapted MoCA (t-MoCA) was available in Indonesia. Objective cognitive screening was not systematically performed, which may introduce variability in clinical judgment and limit comparability between cases. Although some clinicians advised MMSE or MoCA testing during offline follow-up visits, this was not consistently performed or recorded. Future telemedicine workflows should integrate validated screening tools with clear threshold values to ensure standardized and reproducible cognitive assessments. Objective screening using validated thresholds (MMSE <24/30, MoCA <24/30, AD8 ≥2) would allow for a more standardized and reproducible assessment.³⁶

MMSE and MoCA examinations can be performed using telemedicine services. Research shows that MMSE can be carried out with the help of collaborators, and patient responses are recorded between remote nurses and direct nurses so that the results can be compared. Meanwhile, visual items are obtained with photos or fax copies.^37,38 For MoCA testing, video teleconferencing is preferable as it allows patients to interact visually with the physician and view test materials in real time. Physicians also benefit by observing facial expressions, assessing visuospatial functions, and confirming patient comprehension during the examination.^39–42 Even with videoconferencing, factors such as video and audio quality, camera positioning, and internet connectivity must be optimized to ensure accurate and reliable assessment.^40,42,43 High inter-rater reliability has been demonstrated between face-to-face and telemedicine-administered MoCA, supporting its use in remote settings.³³

Limitations

This study is the first to characterize cognitive complaints in Indonesian telemedicine services using chat-based user data, and as such, there was limited literature available for direct comparison. This study has several limitations. First, the cross-sectional design of this study precludes causal inference. Second, the data were derived from chat transcripts, and detailed clinical diagnoses were not available for most participants, limiting the ability to stratify by specific cognitive disorders. Third, age was recorded in broad categories (<40, 40–59, ≥60 years), and raw data were not retained outside the telemedicine providers’ secure servers, preventing reanalysis with finer age bands, which may have captured greater variability in technology use and cognitive risk among older adults. Fourth, our data were derived from chat transcripts, detailed clinical diagnoses were not available for most participants. As a result, the findings should be interpreted as reflecting patterns of telemedicine use and decision-making for cognitive complaints rather than outcomes in a specific, confirmed diagnostic group.

Additionally, keyword search results were categorized on-site at the telemedicine providers’ offices, and raw data could not be taken off-site. As a result, we were unable to retain the exact number of hits for each individual keyword. Because most retrieved complaints were related to forgetfulness or memory loss, other cognitive complaints were grouped together, limiting our ability to explore the diversity of less common complaints. Finally, because data collection occurred during the COVID-19 pandemic, utilization patterns may not fully reflect post-pandemic telemedicine trends.

Conclusion and future directions

This study is the first to describe cognitive complaints in Indonesian chat-based telemedicine services using large-scale user data. This study revealed that treatment decisions were more likely when patients were aged ≥60 years, female, had complaints other than forgetfulness/memory loss, no prior diagnosis of cognitive impairment, and were consulted by a neurologist or other specialist rather than a general practitioner. Referral decisions were more likely when complaints were submitted by caregivers and when patients already had a prior diagnosis of cognitive impairment, but were less likely when consultations were handled by neurologists or other specialists (who often completed management in the same session). Together, these findings show that both patient characteristics and provider type strongly influenced whether patients received treatment or referral through telemedicine.

Chat-based telemedicine plays an important role in early detection and triage for cognitive complaints and may improve care access, particularly for older patients and those in underserved areas. To optimize its use, telemedicine platforms should integrate standardized diagnostic pathways, validated cognitive screening tools (MMSE, MoCA, t-MoCA, AD8), and structured referral workflows. Training healthcare providers to recognize cognitive deficits and deliver individualized interventions is crucial for improving care quality. Future research should also examine post-pandemic data and explore more granular age categories to better capture variability among older adults.

Footnotes

ORCID iD

Pukovisa Prawiroharjo

Ethical approval

This research has been approved by the Ethics Committee of the Faculty of Medicine, the University of Indonesia with the number-letter: KET-669/UN.2F1/ETIK/PPM.00.02/2021 and conducted in accordance with the Helsinki Declaration. Researchers were not involved with the patients or the telemedical service providers directly.

Inform consent

As researchers were not directly involved with patients or telemedical service providers and the study utilized anonymized secondary data, informed consent was not required.

Author contributions

Pukovisa Prawiroharjo, Irma Widyasari, and Iskandar Purba Geraldi conceptualized the study. Yetty Ramli and Adre Mayza conducted data analysis. Diatri Nari Lastri prepared the initial draft. Dinda Nisrina helps conceptualize and writing of the final manuscript. All authors reviewed and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The author affirms that the journal is committed to facilitating openness, transparency, and reproducibility of research.*

References

Kiely

. Cognitive function. Encyclopedia of Quality of Life and Well-Being Research. Springer, 2014. p. 974–977.

Tso

Farinpour

Chui

, et al. A multidisciplinary model of dementia care in an underserved retirement community, made possible by telemedicine. Front Neurol 2016; 7: 225.

Zygouris

Tsolaki

. Computerized cognitive testing for older adults: a review. Am J Alzheimers Dis Other Demen 2015; 30(1): 13–28.

Alzheimer’s Association . 2022 Alzheimer’s disease facts and figures. Alzheimer’s Dement 2022; 18(4): 700–789.

Mesiano

. Samino. Pola gangguan fungsi organ dan fungsi kognitif pada usia lanjut. Neuron 2010; 27(3): 1–5.

Carotenuto

Rea

Traini

, et al. Cognitive assessment of patients with Alzheimer’s disease by telemedicine: pilot study. JMIR Ment Health 2018; 5(2): e31.

Prawiroharjo

Pratama

Librianty

. Layanan telemedis di Indonesia: keniscayaan, risiko, dan batasan etika. J Etik Ked Ind 2019; 3(1): 1–7.

Haleem

Javaid

Singh

, et al. Telemedicine for healthcare: capabilities, features, barriers, and applications. Sens Int 2021; 2: 100117.

Prawiroharjo

Sundoro

Hartanto

, et al. Tinjauan etik layanan konsultasi daring dan kunjungan rumah berbasis aplikasi. J Etik Ked Ind 2019; 3(2): 37–45.

10.

Pais

Ruano

P Carvalho

, et al. Global cognitive impairment prevalence and incidence in community-dwelling older Adults—A systematic review. Geriatrics 2020; 5(4): 84.

11.

van der Flier

Scheltens

. Epidemiology and risk factors of dementia. J Neurol Neurosurg Psychiatry 2005; 76(Suppl 5): v2–v7.

12.

World Health Organization . Dementia fact sheet. World Health Organization, 2021.

13.

Parveen

Haunch

Kerry

, et al. Initial evaluation of a university dementia awareness initiative. J Ment Health Train Educ Pract 2018; 13(5): 264–272.

14.

Draper

Cations

White

, et al. Time to diagnosis in young-onset dementia and its determinants: the INSPIRED study. Int J Geriatr Psychiatr 2016; 31(11): 1217–1224.

15.

Yoon

Ooi

How

. Approach to the forgetful patient. Singap Med J 2018; 59(3): 121–125.

16.

Kvello-Alme

Bråthen

White

, et al. Incidence of young onset dementia in central Norway: a population-based study. J Alzheimers Dis 2020; 75(3): 697–704.

17.

Lam

Shi

, et al. Assessing telemedicine unreadiness among older adults in the United States during the COVID-19 pandemic. JAMA Intern Med 2020; 180(10): 1389–1391.

18.

Chiari

Vinceti

Adani

, et al. Epidemiology of early onset dementia and its clinical presentations in the province of Modena, Italy. Alzheimer's Dement 2021; 17(1): 81–88.

19.

Kelley

Boeve

Josephs

. Young-onset dementia: demographic and etiologic characteristics of 235 patients. Arch Neurol 2008; 65(11): 1502–1508.

20.

Hale

Schneider

Gampe

, et al. Trends in the risk of cognitive impairment in the United States, 1996–2014. Epidemiology 2020; 31(5): 745–754.

21.

Numbers

Crawford

Kochan

, et al. Participant and informant memory-specific cognitive complaints predict future decline and incident dementia: findings from the Sydney memory and ageing study. PLoS One 2020; 15(5): e0232961.

22.

Jiménez-Huete

Del Barrio

Riva

, et al. Subjective evaluation of mood and cognitive functions in a general neurology clinic: patients versus informants. J Clin Neurol 2017; 13(3): 259–266.

23.

Goodman-Casanova

Dura-Perez

Guzman-Parra

, et al. Telehealth home support during COVID-19 confinement for community-dwelling older adults with mild cognitive impairment or mild dementia: survey study. J Med Internet Res 2020; 22(5): e19434.

24.

Cuffaro

Di Lorenzo

Bonavita

, et al. Dementia care and COVID-19 pandemic: a necessary digital revolution. Neurol Sci 2020; 41(8): 1977–1979.

25.

Weiss

Malik

Santos

, et al. Telehealth for the cognitively impaired older adult and their caregivers: lessons from a coordinated approach. Neurodegener Dis Manag 2021; 11(1): 83–89.

26.

Dorsey

Glidden

Holloway

, et al. Teleneurology and Mobile technologies: the future of neurological care. Nat Rev Neurol 2018; 14(5): 285–297.

27.

Migliaccio

Bouzigues

. Dementia and COVID-19 lockdown: more than a double blow for patients and caregivers. ADR 2020; 4(1): 231–235.

28.

Jonker

Geerlings

Schmand

. Are memory complaints predictive for dementia? A review of clinical and population-based studies. Int J Geriatr Psychiatr 2000; 15(11): 983–991.

29.

Wibowo

. Pemanfaatan teknologi informasi dalam rangka pencegahan penyebaran COVID-19. Kementerian Kesehatan Republik Indonesia, 2020.

30.

Perhimpunan Dokter Spesialis Saraf Indonesia . Panduan praktik klinik. Perhimpunan Dokter Spesialis Saraf Indonesia, 2015.

31.

Bature

Guinn

Pang

, et al. Signs and symptoms preceding the diagnosis of Alzheimer’s disease: a systematic scoping review of literature from 1937 to 2016. BMJ Open 2017; 7(8): e015746.

32.

Prince

Bryce

Albanese

, et al. The global prevalence of dementia: a systematic review and metaanalysis. Alzheimer's Dement 2013; 9(1): 63–75.e2.

33.

DeYoung

Shenal

. The reliability of the Montreal cognitive assessment using telehealth in a rural setting with veterans. J Telemed Telecare 2019; 25(4): 197–203.

34.

Bossen

Kim

Steinhoff

, et al. Emerging roles for telemedicine and smart technologies in dementia care. SHTT 2015; 5: 49–57.

35.

Wechsler

. Advantages and limitations of teleneurology. JAMA Neurol 2015; 72(3): 349–354.

36.

de Oliveira

Brucki

SMD

. Computerized neurocognitive test (CNT) in mild cognitive impairment and Alzheimer’s disease. Dement neuropsychol 2014; 8(2): 112–116.

37.

Ciemins

Holloway

Coon

, et al. Telemedicine and the mini-mental state examination: assessment from a distance. Telemed J e Health 2009; 15(5): 476–478.

38.

McEachern

Kirk

Morgan

, et al. Reliability of the MMSE administered in-person and by telehealth. Can J Neurol Sci 2008; 35(5): 643–646.

39.

Iiboshi

Yoshida

Yamaoka

, et al. A validation study of the remotely administered Montreal cognitive assessment tool in the elderly Japanese population. Telemed J e Health 2020; 26(7): 920–928.

40.

Phillips

Chertkow

Pichora-Fuller

, et al. Special issues on using the Montreal cognitive assessment for telemedicine assessment during COVID-19. J Am Geriatr Soc 2020; 68(5): 942–944.

41.

Geddes

O’Connell

Fisk

, et al. Remote cognitive and behavioral assessment: report of the Alzheimer society of Canada task force on dementia care best practices for COVID‐19. Alzheimer's Dement 2020; 12(1): e12111.

42.

Chapman

Cadilhac

Gardner

, et al. Comparing face-to-face and videoconference completion of the Montreal cognitive assessment (MoCA) in community-based survivors of stroke. J Telemed Telecare 2021; 27(8): 484–492.

43.

Brearly

Shura

Martindale

, et al. Neuropsychological test administration by videoconference: a systematic review and meta-analysis. Neuropsychol Rev 2017; 27(2): 174–186.