What He Stopped Doing

October 20, 2026

His family feared dementia because he had stopped doing familiar tasks on his own. The evaluation needed to determine whether the 76-year-old could no longer do those tasks—or could still do them once someone helped him get started.

A fictional case, written for teaching. The patient, his family, and every finding in this article are invented.

Michelle, a hospital dietitian, brought a typed list to the first appointment in March, organized by month and dated the way she would chart a patient’s decline. In November her father had missed his granddaughter Lily’s eleventh birthday, the first he had missed in her life.

The December property tax was paid late, with a penalty, for the first time in forty years. In January he did not call his children for their tax forms, a call he had made every year since they started working. He bathed about once a week and no longer walked, and he spent most of the day reading the news on his phone.

The item that frightened her most went back ten months, to a family dinner in May, when her father accused her uncle Victor, her mother’s younger brother, of taking fifteen thousand dollars from the family. When Victor denied it and her mother said nothing, her father left the table, and he had said little to anyone in the family since.

“I think my dad has dementia,” she said. “I need to know before he does something none of us can undo.”

At 76 every item on the list justified her question, though the list hid a second one that families tend to fold into the first. A man can stop doing something he can still do, because starting it on his own has become hard or because it no longer means what it did. Either change can have neurological or psychiatric causes, alone or together, and the list could not say which.

The Man They Remember

Albert came to Los Angeles in 1979, at 29, after a war cost his family their home and most of their savings. Within a year he was running a lathe on the night shift at an aerospace supplier in the San Fernando Valley. He retired from the same company in 2018 as the quality-control supervisor whose signature released parts for aircraft.

At home he kept the accounts in a green clothbound ledger, the first bought in 1981, when the family’s savings fit on one line. Every January he wrote the year’s due dates and birthdays into its margins, and every Sunday evening he sat at the kitchen table and went through the week ahead. For twenty-five years he also prepared his children’s tax returns at that table.

His children describe a reserved father who kept his feelings to himself and could be counted on for anything that needed doing. He drove them to the airport before dawn, and every grandchild got a birthday card from him with a check inside. His care had always taken the form of tasks, so when the tasks stopped, his children took the silence to mean that he had stopped caring.

After retirement his mornings belonged to Ray, a machinist he had worked beside for thirty years. For five years they walked three miles at six every morning and argued about the news over coffee. Since Ray died in the fall of 2023, Albert has not walked their route. The mornings went first, his wife, Rose, said later, a year and a half before any dated item on Michelle’s list.

His Account

Albert agreed to the evaluation on condition that he would hear the results first and that no report would go to Victor. He had come, he said, because he wanted someone to state in writing that he was not crazy.

“Do you miss the work?” I asked.

“No.” He thought about it for longer than a one-word answer seemed to need. “Forty years is enough. My knees were finished.”

“And the people?”

“Ray.” He said the name and then spent several minutes on a fixture Ray had designed in 1994 that the shop still used.

“Do you still walk?”

“Not since Ray. I could walk alone. Why would I?”

Retirement had helped his knees, and the standing he lost with it stayed hidden while Ray filled his mornings.

“Who do you talk to now?” I asked.

“Talk about what? Ray and I disagreed about everything for thirty years. You can say anything to a man like that. You cannot say it to your children. They worry, and then they call each other.”

“Michelle thinks I do not care about the grandchildren,” he said later. “I care. On Sundays everybody talks at once, fast, in English, and my grandson Owen shows me a game on his tablet, and I do not know the game.”

Of Rose, he said that in forty-six years of marriage she had always been good to him, and that she slept in another room because he snored “like a truck.” She knew him, he said, which was not the same as talking, and they had never needed to talk about money because the money had always been straight. Then he added that she should not have done what she did.

“Victor asked me for fifteen thousand dollars, January of last year,” he said. “I said no. In February the fifteen thousand is gone from the savings. I asked Rose and she said it was taken care of, and I did not ask again. You tell me what happened.”

When you lose everything once, he told me, you write everything down, and if the money is straight, everybody is okay. He held money to be private and help to be for a man who could no longer manage, beliefs that came from his own history. Without heat, he said he worried that the evaluation was a step toward someone else signing his checks. Asked what he wanted, he said his family should stop watching him, and after a moment he added that he wanted the money back.

The Ledger

He had brought the ledger without being asked, and near the end of the hour he opened it to February of the year before and turned it toward me. A withdrawal of fifteen thousand dollars from the joint savings was circled twice in red, with a question mark and, added later, the letter V.

He wanted me to see the circle, and on the way to it I saw the margins, filled in his small block capitals on the first day of that year. In November, beside LILY 11 $50, there was no check mark, and in December, beside PROP TAX 1ST DUE 12/10, a check mark had been added later in a different pen, with PD 12/22 + PENALTY.

Both lapses were there as intentions, written in advance on the right page, and when I asked he gave Lily’s birthday and both tax deadlines at once. The margins for the new year were filled in as well, on New Year’s Day as in every year since 1981, with OWEN 9 $50 waiting beside August. What had stopped, sometime in the fall by Rose’s account, was the Sunday evening review that brought each entry forward in time to act on it. The pill organizer he filled at the same table every Sunday morning had survived, prompted each week by its own empty compartments. The evening review had no such cue, and the book now fell open at the circled page. Of the review he said only that he looked when he needed to look.

He kept the dates and could give them on request, yet they no longer came to him on their own when they called for action. He also drifted in conversation at loud dinners held in fast English, though not at home in their first language, and decades of shop noise made his hearing worth checking.

The weekly bath had its own history, since he had showered every evening for thirty-eight years to get the cutting oil off and, after retirement, every morning after walking with Ray. When the walks ended the shower went with them, helped by a knee that made the bathtub wall awkward, and with nothing else slipping, the answer was a grab bar.

The Testing

Albert chose to be tested in English, the language of his working life, and declined an interpreter. No normative sample I could identify represents a man schooled in one language who has worked in another since 29. I read his results qualitatively for his age and schooling, giving the most weight to the pattern within them.

Nothing in his effort or on the validity indicators gave me reason to doubt the results. He learned a word list a little slowly, then kept nearly all of it after a half-hour delay and recognized the words without false alarms. Attention, visual memory, reasoning, and visuospatial construction raised no concern for a man of his age and schooling. Processing speed was the one clear weakness, slow on both tasks that measured it, a symbol-coding task and a timed trail-making task that depend little on English but still reflect schooling. His hands kept pace on a simple tapping task, and set-shifting was slow but accurate, with most of the slowness already present in the simpler trail.

I also gave him two errands that no manual covers, as structured observations. He was to remind me at 11:30 to validate his parking, with a wall clock in view and his watch on his wrist, and he repeated the instruction back. He was also to ask about the nearest pharmacy when I handed him a red folder. At 11:30 I paused and let half a minute pass without a word from him, though at 11:40 he answered at once when I asked whether he had meant to tell me something. Twenty minutes later he asked about the pharmacy as the folder reached his hands. He had retained both intentions and missed only the one that depended on the clock.

His score on a depression scale was positive, and he did not accept the word when I first used it.

“I am not sad,” he said. “I am angry, and I do not sleep.”

We settled on his own word, heavy, for a mood that had sat on him most of the day since the summer, within two months of the dinner. He woke at four and ate half his dinners, and he had stopped going to the garage where he and his son, David, worked on cars. The news he read for hours was habit more than interest, he said, and he read it twice over, losing his place. By his account and Rose’s, he met criteria for major depressive disorder, recurrent, with a current episode of moderate severity. It was his second, after a milder episode in the winter of 2021 that had remitted on bupropion.

More than once that autumn he had said that the family would manage fine without him, and I asked him directly about death and suicide. Some nights, he said, he would not mind if he did not wake up, though he had never thought of ending his life or made an attempt. He had no plan, intent, or preparation, and he named his grandchildren as a reason to stay, along with the unsettled money, which I counted as real but temporary. With his consent, the pistol he had kept since a 1992 break-in went through a lawful transfer into a safe at David’s house that only David could open. David agreed in front of his father that it would not come back on request, and Albert agreed to tell David or me if the thought grew stronger.

On a loneliness scale he described both kinds of loneliness, the felt lack of a close attachment and the lack of a wider circle. Asked whom he could talk with about private matters, he said no one, and then, after a pause, David. The circle he missed was the shop, men who had known his work for decades and whose talk he could follow without effort. A senior center or volunteer group would offer company without that history, among strangers talking fast in his second language, and he said he would rather read the news.

His physician found treated hypertension and a normal neurological examination with unremarkable laboratory studies, and Albert drank nothing stronger than a Sunday beer. The MRI showed a handful of small white-matter spots, a mild burden for a man of 76 with longstanding high blood pressure. A home sleep study, ordered for the snoring behind the separate bedrooms that relatives had long read as a cold marriage, found moderate obstructive sleep apnea. The store-brand sleeping pill he had started the winter after Ray died was diphenhydramine, an anticholinergic drug older adults are advised to avoid (American Geriatrics Society Beers Criteria® Update Expert Panel, 2023).

Michelle, who saw him on Sundays in a loud room that often included Victor, rated his apathy far higher than Rose, who watched him sharpen her knives at home, and Albert rated himself as engaged. On a daily-function questionnaire, both women described a man who still drove, shopped, cooked, and paid his own bills without help, though the bills and family occasions now took effort. They also reported enough change to cross a dementia screen’s threshold, and Michelle called the accusation a suspicious belief. Like the ledger, those answers documented change, and none of them could explain it.

In my office he completed every task I set, but he started only what mattered to him, from the terms of the evaluation to the ledger page he wanted me to see. He asked nothing about the tests and let the clock run past 11:30, and at home, by every account, he started neither a walk nor the Sunday review. The new year’s entries complicated a simple account of lost initiative, since he could still start a familiar task with personal meaning on his own. What he could not reliably do was carry an intention forward when the moment to act depended on his noticing it. Initiation is a function of the brain as much as a matter of wanting, and depression, poor sleep, an anticholinergic drug, and small-vessel disease can each slow it. In Albert, performance under structure, self-started action, felt motivation, and social life had come apart, a divergence that fit each cause and pointed to no purely psychiatric or purely neurological one.

Reduced spontaneous initiation was therefore a finding, and it met the 2018 consensus criteria for apathy, which allow apathy to coexist with depression (Robert et al., 2018). Until a sustained remission, I deferred giving it a separate explanatory role, because its contribution could not yet be separated from the depression, the lost routines, the apnea, and the nightly drug.

What Rose Knew

With Albert’s consent I interviewed Rose alone, through a professional interpreter, because she preferred their first language and I did not want one of their children translating. She answered the memory questions patiently and the money questions with her eyes on the table.

“He is right,” she said. “I gave Victor the money.”

In February of the year before she had taken fifteen thousand dollars from their joint savings to keep Victor’s dry cleaner open, where she had worked the counter for twenty years. Two of Victor’s children still worked there, and since Albert had already refused her brother, she could not refuse him too. At the dinner she had said nothing, because speaking would have proved her husband right in front of her brother, and she was not ready for her children to know. When she asked whether the children had to know, I reminded her that what she said could appear in my report to Albert and that telling them was not mine to do.

Michelle’s alarm about the money was reasonable, since an accusation against a relative can be an early sign of a failing mind or of money leaving. Rose had the strongest reason to stay quiet, and she still gave the most accurate account of his days. The core of Albert’s belief, that money had gone to Victor with Rose’s help, was true and soundly inferred from Victor’s request, the withdrawal, and Rose’s evasion. Its forecast, that Victor would keep coming back, was a reasonable expectation he held open to doubt, and he accused no one else. Neither part of his belief, then, was a delusion or an overvalued idea.

Michelle had also asked whether David should take over her father’s finances, which mixed what had happened to the money with whether Albert could still manage it. Rose had moved the money from their joint savings without telling him, so his central suspicion had a factual basis. The bank statement he let me review showed the withdrawal on the date his ledger gave. Whether the transfer was lawful and proper between spouses was a legal question outside my evaluation, and Victor’s intentions were unknown. The undisclosed transfer, made after Albert had refused the request, gave reasonable suspicion of financial abuse of an older adult, which California requires psychologists to report (Cal. Welf. & Inst. Code § 15630, 2026). I made the report to Adult Protective Services, which would decide whether abuse had occurred, and I told Rose and Albert. On the second question the law presumes capacity, and a diagnosis does not by itself remove it (American Bar Association Commission on Law and Aging & American Psychological Association, 2008). Capacity is judged decision by decision, and his history held no poor decisions about spending or lending. The one departure from their usual practice had been Rose’s, made out of obligation to her brother. He reconciled a mock bank statement and caught the duplicate charge I had planted, which showed one skill under supervision and was no assessment of capacity. The help he needed was with timing, and a family’s disagreement with him was no reason to take his money out of his hands.

The Formulation

The evidence supported one explanation most strongly, a recurrent depressive episode that had begun within two months of the dinner in a man whose structure had been coming apart since Ray died. Michelle’s list had compressed a sequence into one decline, since his ledger kept running on time for two years after Ray’s death, and the Sunday review lapsed only after the depression began. With learning and retention intact and his affairs still his own, the evidence did not support a major neurocognitive disorder.

Most of the lapses on Michelle’s list had more than one cause, and I could not weigh them precisely. The birthday and the tax reflected the depression, the lost review, and his slowed thinking, along with a book that now opened onto the worst entry of his year. The January call he skipped was also a call to a family he had been avoiding since May.

Other findings stayed open, beginning with the slowing, which could belong to the depression, the apnea, or the drug. Prominent slowing with white-matter change is also the pattern DSM-5-TR associates with vascular disease (American Psychiatric Association, 2022), and his mild changes could neither confirm nor dismiss a vascular contribution. Mild neurocognitive disorder, a modest decline with independence kept through extra effort, could be neither established nor excluded while a depression that might account for the decline was still untreated. If unexplained behavioral change emerged or persisted after the depression and the family dispute were addressed, mild behavioral impairment, proposed as an early sign of dementia, would need a second look (Ismail et al., 2016).

His family’s deeper question was whether the withdrawal meant that dementia was on its way. Loneliness and late-life depression predict later dementia, though the loneliness association shrinks after adjustment for depression and largely disappeared in one large cohort (Brain et al., 2025; Luchetti et al., 2024; Shen et al., 2022). For Albert, that argued for treating both and watching his withdrawal, and treatment could not promise to lower his risk. The formulation was the most reasonable account of the evidence available in March, and it left some causes of his year unexplained.

Feedback

I met with Albert alone first, as he had asked, and told him that he stored new information well, although his thinking had slowed and starting things on time had become hard. He was depressed for the second time in his life, and it was treatable. About the money, I could confirm what had happened, though not what Victor intended.

After a long pause he said, “So I am not crazy.”

“You were right about the money, and you were also depressed,” I said.

He agreed that I could explain the findings to his family, with one exception.

“The money,” he said, “is for Rose to tell.”

At the family meeting Rose told her children about the money herself, first in their language and then in English. When Michelle asked whether this was dementia, I told her that nothing we had found in March showed dementia. His memory held and he still ran his own affairs, though some of them now took him more effort. Most of the problems on her list belonged to a depression and to a life whose structure had gone. His slowed thinking was the part we could not yet explain, which kept a milder disorder or an early neurological cause on the list for the retest. Her list had been accurate, and her one wrong inference rested on a fact kept from her.

She asked whether he had stopped caring about them, which was the fear beneath the list, and Albert answered before I could.

“Sundays are too fast,” he said. “Come on a Tuesday.”

His own goals were to get the money straight, to sleep, and to have his family stop watching him, and the plan was built around them first. The ledger stayed his, while property tax and utilities went to automatic payment and David began a monthly review of the accounts with him. He and Rose agreed in writing that neither would lend more than a thousand dollars without the other, and they agreed to see a couples therapist about the rest. He accepted help with timing and kept every decision, and every arrangement was his to end. Michelle would bring any new change to her father first, a condition he set himself.

He chose problem-solving therapy, an evidence-based treatment for late-life depression (Alexopoulos et al., 2021; Areán et al., 2010), over the options I described, because he liked a problem he could define. His physician restarted bupropion, which can raise blood pressure, and watched his pressure with the white-matter changes in mind (Bausch Health US, 2025). Albert also agreed to a hearing test for the trouble at loud dinners. CPAP was for the apnea and the daytime sleepiness, and I told him it would not fix the early waking or promise anything for his thinking. A phone alarm brought back the Sunday review, so the moment no longer depended on his noticing it, and a grab bar went in beside the tub.

He twice declined a role among peers, and I did not press it. David’s Saturday walks were the part of the plan I cared about most, because they gave Albert a confidant and a reason to leave the house in the same hour. Testing would be repeated in twelve months, and a fall, a change in gait or personality, a new loss of daily function, or a spreading suspicion would bring a neurological consultation sooner.

Six Months Later

When Albert came back in September, his sleep had improved first, within weeks of CPAP and the end of the diphenhydramine, and his energy had followed. His mood lifted more slowly, from nine to four on the depression scale, over months in which several treatments began at once, so no single change could claim it.

Most Saturdays he walked with David, and twice in August he called to suggest a walk himself, which David reported more proudly than the miles. His emotional loneliness eased with one confidant in his week, though the change on the scale was small. The scales showed no change in his social loneliness or his network. With Owen he went once to a volunteer bike-repair program and did not go back, because the young volunteers talked too fast. The hearing test had found a mild loss in both ears, and he wore the new aids in the car and left them in a drawer on Sundays. The aids made every voice at the table loud at once, he said, so a visit to reset them for noise was booked.

Outside routines that a person or an alarm started for him, he still started little, and his initiative lagged behind his mood. On a brief recheck with alternate forms, processing speed remained his clearest weakness. Without standardized change calculations I could not tell whether it had improved meaningfully. Because slowing often outlasts a late-life depression (Bhalla et al., 2006), the vascular question stood where it had in March. We talked again about a neurological consultation, and with his function stable and his mood better, he chose to wait for the March retest, with his physician’s agreement.

Daily function held steady, with no missed bills or doses and the monthly review with David kept, and with the grab bar in he bathed every few days. Adult Protective Services had met with Albert and Rose in April and closed its case. His caution about money had eased but not ended, and although Victor returned two thousand dollars in July, which Albert wrote beside the circle, he still checked the account every week.

He had brought the ledger again, although I had not asked, and opened it to August. Beside OWEN 9 $50 in the margin there was a check mark and a check number, entered on the right day after a Sunday alarm whose sound Owen had chosen.

His family had asked whether he could still remember, and for the most part he could. Like the ledger, he had kept the dates, and what had stopped was the starting, which in Albert had more than one cause. His care now supplied some of that starting from outside, through the Sunday alarm and David’s Saturday walks, and it treated his mood and sleep while his slowing was watched. The check mark beside Owen’s name came from that help, and whether something else had begun in his brain was the question the retest in March would take up.

Author’s Note: This article presents an entirely fictional clinical case developed for educational purposes. The characters, circumstances, assessment findings, and clinical outcome are invented.

Selected References

Alexopoulos, G. S., Raue, P. J., Banerjee, S., Marino, P., Renn, B. N., Solomonov, N., Adeagbo, A., Sirey, J. A., Hull, T. D., Kiosses, D. N., Mauer, E., & Areán, P. A. (2021). Comparing the streamlined psychotherapy “Engage” with problem-solving therapy in late-life major depression: A randomized clinical trial. Molecular Psychiatry, 26(9), 5180–5189. https://doi.org/10.1038/s41380-020-0832-3

American Bar Association Commission on Law and Aging & American Psychological Association. (2008). Assessment of older adults with diminished capacity: A handbook for psychologists. https://www.apa.org/pi/aging/programs/assessment/capacity-psychologist-handbook.pdf

American Geriatrics Society Beers Criteria® Update Expert Panel. (2023). American Geriatrics Society 2023 updated AGS Beers Criteria® for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society, 71(7), 2052–2081. https://doi.org/10.1111/jgs.18372

American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). https://doi.org/10.1176/appi.books.9780890425787

Areán, P. A., Raue, P., Mackin, R. S., Kanellopoulos, D., McCulloch, C., & Alexopoulos, G. S. (2010). Problem-solving therapy and supportive therapy in older adults with major depression and executive dysfunction. American Journal of Psychiatry, 167(11), 1391–1398. https://doi.org/10.1176/appi.ajp.2010.09091327

Bausch Health US. (2025). Wellbutrin XL (bupropion hydrochloride extended-release tablets) [Prescribing information; revised November 2025]. DailyMed, U.S. National Library of Medicine. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=a435da9d-f6e8-4ddc-897d-8cd2bf777b21

Bhalla, R. K., Butters, M. A., Mulsant, B. H., Begley, A. E., Zmuda, M. D., Schoderbek, B., Pollock, B. G., Reynolds, C. F., III, & Becker, J. T. (2006). Persistence of neuropsychologic deficits in the remitted state of late-life depression. American Journal of Geriatric Psychiatry, 14(5), 419–427. https://doi.org/10.1097/01.JGP.0000203130.45421.69

Brain, J., Alshahrani, M., Kafadar, A. H., Tang, E. Y. H., Burton, E., Greene, L., Turnbull, D. A., Myers, B. J., Naheed, A., Siervo, M., Tully, P. J., & Stephan, B. C. M. (2025). Temporal dynamics in the association between depression and dementia: An umbrella review and meta-analysis. eClinicalMedicine, 84, Article 103266. https://doi.org/10.1016/j.eclinm.2025.103266

Cal. Welf. & Inst. Code § 15630 (2026). https://codes.findlaw.com/ca/welfare-and-institutions-code/wic-sect-15630/

Ismail, Z., Smith, E. E., Geda, Y., Sultzer, D., Brodaty, H., Smith, G., Agüera-Ortiz, L., Sweet, R., Miller, D., Lyketsos, C. G., & ISTAART Neuropsychiatric Symptoms Professional Interest Area. (2016). Neuropsychiatric symptoms as early manifestations of emergent dementia: Provisional diagnostic criteria for mild behavioral impairment. Alzheimer’s & Dementia, 12(2), 195–202. https://doi.org/10.1016/j.jalz.2015.05.017

Luchetti, M., Aschwanden, D., Sesker, A. A., Zhu, X., O’Súilleabháin, P. S., Stephan, Y., Terracciano, A., & Sutin, A. R. (2024). A meta-analysis of loneliness and risk of dementia using longitudinal data from >600,000 individuals. Nature Mental Health, 2(11), 1350–1361. https://doi.org/10.1038/s44220-024-00328-9

Robert, P., Lanctôt, K. L., Agüera-Ortiz, L., Aalten, P., Brémond, F., Defrancesco, M., Hanon, C., David, R., Dubois, B., Dujardin, K., Husain, M., König, A., Lévy, R., Mantua, V., Meulien, D., Miller, D., Moebius, H. J., Rasmussen, J., Robert, G., . . . Manera, V. (2018). Is it time to revise the diagnostic criteria for apathy in brain disorders? The 2018 international consensus group. European Psychiatry, 54, 71–76. https://doi.org/10.1016/j.eurpsy.2018.07.008

Shen, C., Rolls, E. T., Cheng, W., Kang, J., Dong, G., Xie, C., Zhao, X.-M., Sahakian, B. J., & Feng, J. (2022). Associations of social isolation and loneliness with later dementia. Neurology, 99(2), e164–e175. https://doi.org/10.1212/WNL.0000000000200583

Clinician and Research Supplement

This supplement holds the technical material behind the article, for clinicians who want the evaluation in detail and for readers who want to check the evidence. Every patient-level entry describes an invented case, and every literature entry was checked against its source in the way S7 describes.

How to use each part

  • S1 sets out the normative, linguistic, and cultural limits and the timeline of the case, and it should be read before borrowing any interpretation from this case for a real patient with a similar history.
  • S2 lists the evaluation as administered, with the basis for interpreting each measure, the fictional result, and the point where interpretation stops, and Table S2a shows which interpretations rest on a calculation and which are illustrative.
  • S3 traces each hypothesis from referral through assessment to six months, names the finding that would move it next, and is followed by the follow-up interpretation and the note on the apathy criteria.
  • S4 groups the published estimates behind the evidence by the clinical question each answers, followed by notes on overlapping cohorts and causal inference.
  • S5 holds the collateral interview template and the worked example of the disputed transfer.
  • S6 gives the monitoring schedule with the fictional course, the triggers for earlier review, the risk assessment and firearm arrangements, and a guide to reading a response to treatment.
  • S7 and S8 cover methods, verification, limitations, and the full reference list.

S1. Normative, Linguistic, and Cultural Limits

Albert is a fictional man from an Asian country who completed eleven years of school and two years at a technical institute in his first language before a war displaced his family. After arriving in Los Angeles at 29 he took machining and English courses at a community college, worked thirty-eight years in English, and still reads the news in both languages. His country of origin is left unnamed on purpose, so that no interpretation in the case rests on a generalization about a national or ethnic group.

He chose English for testing and declined an interpreter, and the evaluation honored that choice while treating it as a source of uncertainty. Rose’s interview used a professional interpreter, since consensus guidance holds that family members should not ordinarily serve as interpreters (Nielsen et al., 2024). No published norms we could identify describe adults schooled in one language who have worked in another since early adulthood, so any comparison with U.S. norms for age and education would only approximate. Hearing was screened only in a quiet office, so a mild loss could also have lowered scores on the spoken tasks, and audiometry followed feedback. In this fictional case no normative comparison was calculated, so the performance descriptors are qualitative illustrations, and only the questionnaire and screening totals were scored, as Table S2a sets out.

Verbal measures in his second language, such as list learning and letter fluency, were read with caution, and picture naming was not interpreted against U.S. norms. The pattern within his own profile, such as intact retention beside slowed processing, carried more weight than any single comparison with a norm. Years of education can misstate the educational experience of older adults from diverse backgrounds (Manly et al., 2002), and his schooling was divided between two languages and two systems. Any education adjustment was therefore treated as approximate, and change on alternate forms at twelve months will carry more weight than any comparison with norms.

The questionnaires carry similar limits, beginning with loneliness bands that were set in Dutch adults aged 55 and older and that their authors describe as tentative (de Jong Gierveld & van Tilburg, 1999). The social-network cut-off may not apply to Chinese older adults (Chang et al., 2018), and single cut-offs on cognitive screens misclassify across education and culture (Wong et al., 2015). Research on loneliness in Asian American older adults is sparse, mostly cross-sectional, and drawn largely from Chinese and Korean American samples (Min et al., 2025). It can suggest questions, but it cannot describe a man whose country of origin is unnamed.

The Cultural Formulation Interview, with its informant version and its older-adult and immigrant modules, organized the questions about meaning (American Psychiatric Association [APA], n.d.). The answers were his own, including the view that money is private, that help is for a man who can no longer manage, and that care is shown through tasks. The evaluation treated those views as features of one man’s history and drew no diagnostic inference from his origin.

Table S1 gives the timeline of the case, against which every date, age, and interval in the article and this supplement was checked.

Table S1

Timeline of the Fictional Case

WhenEventAlbert’s age
1979Arrives in Los Angeles after a war; within a year, runs a lathe on the night shift at an aerospace supplier29
1980Marries Rose30
1981Buys the first green ledger and writes the year’s due dates and birthdays into its margins, as he will every New Year’s Day after30
About 1988Begins thirty years in the shop beside Ray38
1992Buys a pistol after a break-in42
2018Retires as quality-control supervisor; the morning walks with Ray begin68
Winter 2021First depressive episode, about three months; bupropion from February, with remission by spring71
December 2021Bupropion ends after ten months71
Fall 2023Ray dies; the walks, the morning showers, and the morning structure end, and weekly bathing begins73
Early 2024Nightly diphenhydramine begins, the winter after Ray died74
January 1, 2025Writes the 2025 margins74
January 2025Victor asks for $15,000 and Albert refuses; the last routine call to his children for their tax forms75
February 2025Rose transfers $15,000 from joint savings to Victor’s dry cleaner; Albert circles the withdrawal, asks Rose, and is told it is taken care of75
May 2025Family dinner: Albert accuses Victor, Victor denies it, and Rose says nothing75
July 2025The current depressive episode begins75
Fall 2025The Sunday evening review lapses; he says more than once that the family would manage without him75
November 2025Lily’s eleventh birthday missed75
December 2025Property tax paid late, with a penalty75
January 1, 2026Writes the 2026 margins, including OWEN 9 $50 for August75
January 2026Does not call his children for their tax forms76
February 2026Primary-care visit; screen of 24 of 30; referral76
March 2026Intake with Michelle; two testing sessions; Rose interviewed; report to Adult Protective Services; physician workup; pistol moved by lawful transfer to David’s safe; feedback76
April 2026Bupropion restarted; CPAP; diphenhydramine stopped; problem-solving therapy; Sunday alarm; automatic payments; lending agreement; hearing test ordered; grab bar installed; Adult Protective Services meets with Albert and Rose and closes its case76
May and June 2026Six-week and three-month reviews; sleep better first, mood more slowly76
July 2026Victor returns $2,000, which Albert enters beside the circle76
August 2026Owen’s ninth birthday check mailed on the day; two self-started calls to David76
September 2026Six-month visit; a neurological consultation is discussed again, and he chooses to wait for the retest with his physician’s agreement76
March 2027Twelve-month retest scheduled77

Note. Fictional case. Ages assume a birthday in early January 1950. The evaluation took place in March 2026, and every “last year” or “the year before” in the article counts back from that month.

S2. The Evaluation as Administered

Table S2

The Evaluation as Administered, With the Fictional Results

DomainMeasure as administeredBasis for interpretationResult in the fictional caseLimits in this case
Validity and screening
Performance validityOne standalone and two embedded indicators, chosen for his age and languagePublished cut-offs for each indicator, read within the AACN consensus framework (Sweet et al., 2021)Valid on each indicator, an illustrative result like the performance descriptors; no cut-off comparison was calculated for this fictional caseEmbedded indicators built on English span or word tasks carry extra uncertainty for a second-language speaker, and some indicators lose specificity in established dementia (Dean et al., 2009)
Global screen, before referralMontreal Cognitive Assessment, given in primary careThe initially recommended cut-off of 26 inflates false positives in older and less educated adults, and 23 lowers them (Carson et al., 2018)24 of 30, below 26 and above 23Single cut-offs misclassify across education and culture (Wong et al., 2015), and a screen cannot separate depression from early disease
Cognition (English; qualitative fictional descriptors, no normative calculation)
Attention and working memoryDigit span forward and backwardAge and education normsWithin expectationAdministered in his second language
Processing speedSymbol coding; Trail Making Test, Part AAge and education normsBelow expectation on both tasks, to a similar degree; the one clear weakness in the profile, on the tasks least dependent on EnglishA domain-level statement is supported here only because the two measures agreed. Depression, untreated sleep apnea, a nightly anticholinergic drug, and small-vessel disease can each lower speed, and the first three were present at baseline while the fourth could not be excluded. Low language demand does not make a timed task culture-free (Nielsen et al., 2024), so schooling in another system and limited practice with timed tests may also contribute, which is why change within his own scores at twelve months carries the weight
Motor speedFinger tapping, dominant and nondominant handsAge normsWithin expectation for both handsA simple test of hand speed. It makes peripheral slowing a less likely explanation for the two written speed tasks, which both need a pencil, and it does not exclude a contribution from fine motor control
Executive functionTrail Making Test, Part B; letter and category fluency; a verbal reasoning measureAge and education normsSet-shifting at the low end of expectation, slow and accurate with one self-corrected error, with most of the slowness already present in Part A; letter fluency in English at the low end of expectation; category fluency and reasoning within expectationNo domain-level summary is supported, because the results diverge and two of them have explanations outside executive function, the Part A slowing and his second language
Verbal learning and memoryWord-list learning with delayed free recall and recognitionAge and education normsLearning at the low end of expectation, with slow acquisition; delayed recall within expectation, with nearly all learned words retained; recognition within expectation, with no false-positive errorsSecond-language list learning may slow acquisition. Cue and recognition benefit cannot exclude early disease, and one combined cued-recall procedure separated MCI from depression with only 58% sensitivity (Dierckx et al., 2007)
Visual memoryFigure learning with delayed recallAge and education normsLearning and retention within expectationLower language demand than the word list
LanguageConversation, comprehension, and picture namingQualitativeFluent, with comprehension intact; naming not interpreted against U.S. normsNaming items are culturally and linguistically loaded
Visuospatial functionFigure copy; block constructionAge and education normsWithin expectationNone specific
Prospective memoryInformal embedded probe, given as a structured observation: a time-based errand (remind the examiner at 11:30 to validate parking, with a wall clock in view and his own watch, instruction repeated back correctly) and an event-based errand (ask about the nearest pharmacy when handed a red folder)No normsTime-based errand not carried out when the examiner paused at 11:30 for half a minute; the intention was retrieved at once when he was asked at 11:40 whether he had meant to say anything. Event-based errand carried out as the folder was handed overA structured clinical observation, not a measure and not evidence for any one diagnosis. In laboratory studies older adults do worse than younger adults on time-based and event-based tasks by similar margins, event-based tasks with high strategic demand show larger age effects than those supported by more automatic processes, and naturalistic studies favor older adults (Henry et al., 2004). A normed test would be needed to say whether his miss exceeded what age predicts
Everyday financial skillStructured reconciliation of a mock bank statement with a planted duplicate chargeDescriptive; no normsReconciled without error and found the duplicate chargeShows one skill under supervision, not initiation or decisions about lending, and is not a capacity assessment, which would examine knowledge, skills, and judgment for specific financial decisions along with values and risk (American Bar Association Commission on Law and Aging & American Psychological Association, 2008). Financial skills can decline early in MCI (Triebel et al., 2009)
Mood, loneliness, apathy, and grief
Depressive symptomsGDS-15Recommended cut-off of 5, with pooled sensitivity .89 and specificity .77 (Pocklington et al., 2016)9 of 15. Three endorsements (dropped activities, prefers to stay home, lacks energy) carry withdrawal and low-vigor content, a fourth (fears something bad will happen) overlaps with his worry about money, and five concern mood. He did not endorse the memory itemNo suicide item. Withdrawal and low-vigor content may overidentify depression in older adults (Adams, 2001)
Depressive episode, currentStructured diagnostic interview (mood and psychotic modules) with collateral from RoseDSM-5-TR criteria (APA, 2022)Major depressive disorder, recurrent, current episode moderate. Onset in July 2025, within two months of the May dinner, eight months at evaluation. Seven of nine symptoms: depressed mood most of the day; loss of interest, including the garage he had stopped visiting, with news reading continued by habit and without pleasure; early waking at four; reduced appetite nearly every day, with about nine pounds lost since the summer on his physician’s scale; fatigue, shared with the apnea and the drug; impaired concentration; and a recurrent passive wish not to wake up, acknowledged on direct inquiry. Six counted, with fatigue set aside. Severity moderate rather than mild because the functional cost, missed obligations and withdrawal from his family and the garage, went beyond the minor impairment of a mild episode, with one symptom over the five required; not severe, because the symptoms did not markedly interfere with his functioning, his distress was not unmanageable, and his instrumental independence was keptConducted in English at his choice. Apnea and diphenhydramine can contribute to fatigue, poor sleep, and poor concentration, and fatigue was set aside. Early waking and the concentration complaint were counted because both began with the mood change in July 2025, long after the snoring began and well after the drug was started, and waking at four is a typical depressive pattern. Symptoms are excluded only when clearly attributable to another medical condition, so the diagnosis rested on six symptoms, two of them possibly shared
Depressive episode, priorInterview with Albert and Rose; his physician’s note from 2021DSM-5-TR criteria (APA, 2022)Winter 2021, about three months. Criteria judged met retrospectively from his and Rose’s account of duration and impairment, with five symptoms documented in his physician’s note (low mood, loss of interest, insomnia, weight loss, fatigue). Bupropion from February to December 2021, with remission by spring 2021Retrospective. A primary-care note cannot by itself establish two weeks of near-daily symptoms or the impairment, so the recurrent specifier rests on the combined account
Suicide riskDirect inquiry after his remarks, with collateral from RoseClinical risk formulation (S6)Remarks that the family “would manage fine” without him. On direct inquiry, passive suicidal ideation, a recurrent wish not to wake up, without active ideation, plan, intent, preparation, or prior attempt. Protective factors his grandchildren and, more temporarily, the unsettled money. Firearm access and medication supply addressed (article and S6)The ninth depressive criterion requires recurrent thoughts of death, so the remarks alone would not have counted without his answer on direct inquiry. The remarks also express perceived burdensomeness, which S6 weighs as a risk factor
LonelinessDe Jong Gierveld Loneliness Scale, 11 itemsTotal-score and subscale bands in the manual’s updated web edition, which its authors describe as tentative and tied to culture and time; the social cut-off was chosen “arbitrarily” (de Jong Gierveld & van Tilburg, 1999)Emotional 4 of 6 (band 3 to 6, lonely); social 3 of 5 (band 3 to 5, lonely); total 7 (band 3 to 8, moderately lonely)Bands were set in Dutch adults aged 55 and older, and the manual says it is not known whether they classify individuals. The subscale bands appear in the updated web edition, and whether the printed 1999 edition carried them was not confirmed, so the article describes the subscales without thresholds
Social networkLubben Social Network Scale, 6 items (Lubben et al., 2006)Range 0 to 30; a cut-off below 12 is attributed to Lubben by Chang et al. (2018)Total 10, below that cut-off (family 8, friends 2). One relative, David, named for private matters after a pauseA self-reported estimate. The cut-off may not apply to Chinese older adults (Chang et al., 2018), which shows that a cut-off set in one population may not transfer to another
ApathyApathy Evaluation Scale, self and informant versions (Michelle; Rose), with clinician observationComparison across the self-rating and the two informant ratings (Marin et al., 1991); no cut-off appliedSelf lowest, Rose intermediate, Michelle highest by a wide margin. In the office, self-starting was selective, since he set the terms of the evaluation and brought and opened the ledger unasked but asked nothing about the tests and gave no reminder at 11:30. Full completion of examiner-started tasks, and intact emotional responsivenessTotals are not reported, because the meaning lies in the comparison and in each rater’s vantage point
BereavementInterview, with PG-13-R content as a guideDSM-5-TR prolonged grief criteria (Prigerson et al., 2021)The interview did not support prolonged grief disorder. Thinks of Ray when passing the park and avoids their route, without near-daily yearning or preoccupation; no PG-13-R total calculatedNone specific
TraumaInterview screen for current symptoms tied to the warPCL-5 only when current symptoms are reported (Blevins et al., 2015)No current intrusive or avoidance symptoms; PCL-5 not givenWar history kept open for later review
Collateral and daily function
Informant-reported changeAD8 (Michelle; Rose)Two or more endorsed items read as likely impairment (Galvin et al., 2005); pooled sensitivity .92 and specificity .64 at that cut-off (Hendry et al., 2019)Both reached the cut-off. Michelle 5 (judgment, hobbies, complicated finances, appointments, daily thinking or memory); Rose 2 (hobbies, appointments)Items on judgment, finances, and hobbies can register a money dispute, apathy, or depression as cognitive change
Instrumental functionFunctional Activities Questionnaire (Michelle; Rose)Range 0 to 30 (Pfeffer et al., 1982). The FAQ has substantial reliability and validity evidence, including separation of normal cognition, mild cognitive impairment, and dementia (González et al., 2022) and of mild cognitive impairment from very mild Alzheimer’s disease (Teng et al., 2010). The Hartford guide recommends a cut-off of 9 (Mayo, 2016) without establishing its fit for this clinical and cultural contextMichelle 4 (bills, tax records, hobbies, and remembering occasions, each rated as done with difficulty); Rose 2 (bills and remembering occasions). No item rated as needing assistanceThe elevations are functional observations of acquired inefficiency, not of dependence, and with the late tax and the lost tax routine they document real change. Neither total approaches 9, and the item-level changes and the difference between the two informants carry the interpretation more than any single threshold. Informants handle tasks he never did in different ways
Neuropsychiatric symptomsNPI-Q (Michelle)A positive item prompts inquiry (Kaufer et al., 2000)Depression, apathy, irritability, and delusions endorsed, the last for the belief about moneyThe delusions item was endorsed for a belief later shown to be largely accurate
Cultural formulationCultural Formulation Interview, core and informant versions, with the older-adult and immigrant modules (APA, n.d.)QualitativeMoney is private, help is for a man who can no longer manage, and care is shown through tasks; English preferred for testing, first language with RoseGenerates hypotheses about meaning, not scores
Medical
Medical reviewPhysician examination; Tier 1 laboratory studies; MRI (Dickerson et al., 2025); home sleep apnea test; medication reviewClinicalTreated hypertension; normal neurological examination; laboratory studies unremarkable, including thyroid function and vitamin B12; scattered white-matter changes, read as punctate foci in the deep white matter without early confluence and without infarcts, lacunes, or microbleeds, a mild burden for his age and blood pressure; moderate obstructive sleep apnea; diphenhydramine nightly since early 2024; bupropion from February to December 2021By the 2024 Canadian recommendations, beginning confluent or confluent white-matter change is sufficient to impair cognition in many people, the threshold is unclear and varies with reserve, and no lower threshold is set (Swartz et al., 2025). Punctate change therefore neither establishes nor excludes a vascular contribution, and it cannot settle the DSM-5-TR requirement that cerebrovascular disease be sufficient to account for the deficits (APA, 2022). Imaging also leaves neurodegenerative disease open. First-generation antihistamines, diphenhydramine included, are on the list of drugs older adults should avoid (American Geriatrics Society Beers Criteria® Update Expert Panel, 2023)
Substance useInterview with Albert and Rose; physician historyClinicalOne beer with Sunday dinner; no other alcohol, cannabis, or drug use; the sleeping pill was an over-the-counter diphenhydramine productSelf-report with collateral and no biomarker. The criteria for a depressive episode and for apathy both exclude an explanation by a substance (APA, 2022; Robert et al., 2018)
Hearing and visionScreening in a quiet office, with correction for vision; audiometry ordered after feedbackClinicalAdequate for testing in a quiet office, where he repeated instructions correctly. Rose described his drifting at loud, fast dinners, and he described everyone talking at onceScreened because hearing impairment is associated with psychotic symptoms (Linszen et al., 2016) and is listed among the modifiable risk factors for dementia (Livingston et al., 2024). A quiet-office screen cannot detect difficulty with speech in noise, which decades of shop noise make plausible, and a loss could have lowered the spoken tasks though not the two written speed tasks
DeliriumAttention across two sessions; collateral questions about fluctuationConfusion Assessment Method if attention fluctuates (Inouye et al., 1990)No fluctuation; the Confusion Assessment Method was not indicatedNone specific

Note. All results are invented for this fictional case. Performance results, including performance validity, are qualitative illustrations worded as within, at the low end of, or below expectation for his age and education, and the basis column names the normative reference a calculated interpretation would use. No cognitive raw scores, standard scores, or percentiles were calculated, and none should be inferred (Table S2a). A report on a real patient would apply the AACN uniform labels (Guilmette et al., 2020) to calculated scores. Questionnaire results are raw totals read against each instrument’s published convention, and named tests are examples. AACN = American Academy of Clinical Neuropsychology; APA = American Psychiatric Association; GDS-15 = Geriatric Depression Scale, 15-item version; MCI = mild cognitive impairment; NPI-Q = Neuropsychiatric Inventory Questionnaire; PCL-5 = PTSD Checklist for DSM-5; PG-13-R = Prolonged Grief-13-Revised.

Table S2a

Score and Normative Verification: What Was Calculated and What Is Illustrative

Measure or resultHow it is reportedStatusWhat a calculated version would requireVerified basis for the interpretation
Cognitive performance (fictional)
Performance validityValid on each indicatorIllustrative, like the performance descriptors; no cut-off comparison was calculatedPublished cut-offs for each named standalone and embedded indicatorThe AACN consensus on validity assessment is cited for the framework only (Sweet et al., 2021)
Montreal Cognitive Assessment, 24 of 30Raw total against two cut-offsCalculated: a raw total read against published cut-offsNothing furtherCut-offs of 26 and 23 verified in Carson et al. (2018); misclassification across education and culture in Wong et al. (2015)
Digit spanWithin expectationIllustrativeA scaled score from the manual’s age norms, with an education adjustment where a demographic-norm system provides oneNone; no normative table was consulted
Symbol codingBelow expectationIllustrativeA scaled score from the manual’s age norms, with an education adjustment where a demographic-norm system provides oneNone
Finger tappingWithin expectation for both handsIllustrativeA mean tapping rate for each hand against the manual’s age normsNone; the statement that it makes peripheral slowing a less likely explanation is a feature of the fictional profile
Trail Making Test, Parts A and BBelow expectation; low end of expectationIllustrativeDemographically corrected scores for each part from the Heaton norms (raw time to scaled score, then T score), whose ethnicity tables do not include his background. An A-to-B comparison would gauge how much baseline speed contributes to Part B without isolating set-shifting, since Part B adds alternation and letter-number sequencing that schooling in another language may affectNone; the statement that most of the Part B slowness was present in Part A is a feature of the fictional profile, not a calculation
Letter and category fluencyLow end of expectation; within expectationIllustrativeAge- and education-stratified norms; no published norm fits his bilingual schooling and work historyNone
Word-list learning, delayed recall, recognitionLow end; within expectation; within expectationIllustrativeAge norms for each index, with an education adjustment where a demographic-norm system provides one, and a recognition-discriminability scoreNone
Visual memory, figure copy, block construction, verbal reasoningWithin expectationIllustrativeAge norms from each manual, with an education adjustment where a demographic-norm system provides oneNone
Prospective-memory errandsCarried out or notStructured observation; not scorableA normed prospective-memory testHenry et al. (2004), abstract level, normal aging
Mock bank statementReconciled; duplicate charge foundStructured observation; not scorableA standardized financial-capacity instrument and a decision-specific interviewThe ABA and APA handbook for the framework (American Bar Association Commission on Law and Aging & American Psychological Association, 2008)
Six-month speed recheck, alternate formsStill his clearest weakness; whether it improved could not be determinedIllustrative; no reliable-change calculationAlternate-form reliability and the standard error of measurement for each task at a six-month interval, with a practice-effect adjustmentNone; the finding is reported as qualitative persistence of a relative weakness, and a score can improve and still sit below expectation
Questionnaires and screens (fictional raw totals)
GDS-15: 9 at baseline; 7, 5, and 4 at follow-upRaw totals against the cut-off of 5Calculated against a verified cut-offNothing furtherPocklington et al. (2016) for the cut-off; item content verified; Adams (2001) for the withdrawal and low-vigor content
De Jong Gierveld, 11 items: emotional 4 of 6, social 3 of 5, total 7; emotional 2 of 6 and total 5 at six monthsRaw subtotals against the manual bandsCalculated against published bands that their authors call tentativeNothing furtherTotal-score and subscale bands verified in the manual’s updated web edition (de Jong Gierveld & van Tilburg, 1999). The six-month emotional change is two items on a six-item subscale, and the total of 5 stays in the same moderately lonely band as the baseline total of 7, so the change is descriptive
Lubben Social Network Scale, 6 items: 10 (family 8, friends 2) at baseline and 10 at six monthsRaw total against a cut-off below 12Calculated against a cut-off known only from a secondary attributionA primary citation for the cut-offScale verified (Lubben et al., 2006); the cut-off is attributed to Lubben by Chang et al. (2018)
Apathy Evaluation Scale, self and two informantsRelative ordering onlyIllustrative; totals not reportedTotals compared with published meansScale verified (Marin et al., 1991)
PG-13-RNot scored; the interview did not support the disorderInterview-basedA scored item set against the published thresholdCriteria verified (Prigerson et al., 2021)
AD8: Michelle 5, Rose 2Raw totals against the cut-off of 2Calculated against a verified cut-offNothing furtherGalvin et al. (2005); Hendry et al. (2019)
FAQ: Michelle 4, Rose 2Raw totals from items scored 0 to 3Calculated against the published range; the guide’s cut-off of 9 used only as a reference pointA cut-off validated for his clinical and cultural contextPfeffer et al. (1982); reliability and validity in González et al. (2022), at the abstract level; Teng et al. (2010), at the bibliographic level; the cut-off from Mayo (2016)
NPI-QItems endorsedCalculated as presenceNothing furtherKaufer et al. (2000)

Note. Fictional case. A calculated profile would require identified instruments, normative sources, demographic adjustments, and independently checked conversions, and none was constructed for this case, since invented conversions were ruled out. The table shows what each measure would require. “Within expectation”, “low end of expectation”, and “below expectation” are the only performance descriptors used, and they map to no score. AACN = American Academy of Clinical Neuropsychology; GDS-15 = Geriatric Depression Scale, 15-item version; FAQ = Functional Activities Questionnaire; NPI-Q = Neuropsychiatric Inventory Questionnaire; PG-13-R = Prolonged Grief-13-Revised.

S3. Weighted Differential

Table S3

Weighted Differential, From Referral to Six Months

HypothesisAt referralAfter assessmentAt six monthsWhat would move it next
Major depressive episodePlausible: withdrawal, early waking, and a past response to bupropionMost likely primary contributor: seven of nine criteria symptoms for eight months (six counted, with fatigue, shared with the apnea and the drug, set aside), with depressed mood and loss of interest among them and passive suicidal ideation acknowledged on direct inquiry; onset within two months of the May dinner; moderate severity; second episode, the 2021 episode judged retrospectively to have met criteria and remitted by spring 2021. Four of nine GDS-15 endorsements nonspecificSupported, not proved: the GDS-15 fell from 9 to 4 while bupropion, problem-solving therapy, CPAP, diphenhydramine withdrawal, and the walks all began, so the improvement cannot be assigned to any one changeRelapse, or initiation and speed that stay low after a sustained remission
Reduced initiation, possible apathy syndromePlausible: little self-started activityReduced initiation observable. Criteria A to C of the 2018 consensus met, in two of three dimensions with emotional responsiveness intact; criterion D did not bar the diagnosis, since the excluded factors present (the lost routines, possible hearing loss, and nightly diphenhydramine) did not clearly explain all of the reduction and depression is not an exclusion (Robert et al., 2018). A separate syndrome not diagnosed for now, by clinical judgment, because its share could not be separated from the depressive, environmental, and drug contributions. Initiation selectively preserved for affectively charged goals, which excludes neither apathy nor executive change, and a familiar, meaningful annual ritual intact. The note on the apathy criteria gives the full reasoningPartly improved with external cues and a companion; self-starting still limited outside cued routines, and initiative lagged behind mood. Early persistence is expected, since apathy in late-life depression is common and changed little with escitalopram in an uncontrolled study (Yuen et al., 2015)Persistence after a sustained remission favors a separate syndrome or a neurological cause, and the judgment against a separate diagnosis would then be revisited
Sleep apnea and anticholinergic burdenNot consideredProbable contributors to fatigue and slowing: moderate obstructive sleep apnea; nightly diphenhydramine since early 2024. Weighed as contributors to the depressive episode rather than its cause, since the snoring was decades old and the mood change followed the dinnerSleep and energy improved first. Processing speed was still his clearest weakness after CPAP and diphenhydramine withdrawal, so their weight as sole causes of the slowing fell, provided CPAP use is confirmed. That does not by itself raise the weight of a vascular causeCPAP adherence; residual daytime sleepiness
Hearing lossNot consideredPossible contributor to the drifting at loud, fast dinners and to some spoken-task scores; heard well in a quiet office; cannot explain the two written speed tasks. Audiometry orderedMild loss in both ears; aids worn in the car and left in a drawer on Sundays. Whether the Sunday difficulty eases with aid use is not yet knownAid use at dinners, and any change in how he follows speech there
Emotional and social lonelinessAssumed by some relatives and placed in the marriageContributor: emotional and social loneliness in the tentative lonely ranges; Lubben total below the conventional cut-off, with the shortfall among friends; no regular confidant, though he named David after a pause; he missed both a confidant and the shop’s circle of peersEmotional loneliness 2 of 6, out of the subscale’s lonely band, with the total of 5 still in the same moderately lonely band as 7, a two-item change on a six-item subscale whose bands are tentative, so the weekly confiding contact with David carries the finding more than the score does; social loneliness 3 of 5 and the Lubben total unchanged at 10 after one peer program tried once. The two dimensions moved apart, and the questionnaire change cannot be assigned to one interventionA confidant and a circle of peers that change little
Amnestic neurocognitive disorder (for example, Alzheimer’s disease)Leading concern: age, forgetfulness, missed dates, an abnormal screenLow: learning, retention, and recognition intact; each lapse had a mechanismLow: no new lapsesStorage-pattern decline or new functional loss on retesting
Mild neurocognitive disorder, any etiologyRaised by the screenNeither established nor excluded: one domain clearly weak; instrumental independence kept with compensatory effort, which the functional criterion permits; the exclusion for a better explanation by another mental disorder could not be applied while the depression was untreated, and the apnea and the anticholinergic drug were reversible medical contributors to treat before any etiology was assigned (APA, 2022)Unchanged: speed still weak, independence stable, remission not yet sustainedTwelve-month retest with reliable-change methods after a sustained remission
Vascular cognitive impairmentNot raisedPossible and unresolved: below-expectation speed on two tasks, low-end set-shifting largely explained by speed, punctate white-matter change without confluence, treated hypertension, and two depressive episodes beginning after 70 (late-life depression is more strongly associated with vascular dementia than with Alzheimer’s disease; Diniz et al., 2013). By the 2024 Canadian recommendations the diagnosis requires cognitive impairment attributable to cerebrovascular injury, and his punctate change had not reached the beginning-confluent change held sufficient to impair cognition in many people, with no lower threshold named (Swartz et al., 2025). DSM-5-TR asks that cerebrovascular disease be considered sufficient to account for the deficits, and its text treats a single stroke or extensive white-matter disease as generally sufficient for the mild disorder, which his scattered punctate change is not (APA, 2022)Unchanged: persisting slowing after sleep and medication changes narrows the explanations without naming one, and slowing often persists after late-life depression remits (Bhalla et al., 2006). The weight was not raised on this evidence. Neurological consultation discussed again and deferred by his choice to the retest, with his physician’s agreementDecline beyond reliable-change limits at twelve months, new gait change or falls, or progression of white-matter change on repeat imaging
Other neurodegenerative syndromes (frontotemporal, Lewy body)Raised by apathy, anger, and the new suspicionLow: no disinhibition, loss of empathy, hallucinations, or parkinsonism, and no dream enactment reported, though with separate bedrooms no bed partner could have observed itLowAny of those features
Mild behavioral impairmentPossible: later-life change in motivation and thought content (Ismail et al., 2016)Not assigned: a depressive episode better explains the changeNot assigned; suspicion diminishedNew or persisting unexplained behavioral change, sustained for six months or more, after the depression and the family dispute have been addressed
Delusional or other psychotic disorderRaised by the family’s accountNot supported: core belief true and soundly inferred from Victor’s request, the withdrawal, and Rose’s evasion; its forecast a reasonable expectation held open to doubt, so not an overvalued idea either; no accusation extended to anyone else, and his worry that the evaluation could put his checks in other hands matched a question Michelle had actually raised; no hallucinations. Conviction and extension are dimensions of delusional experience (Kendler et al., 1983)Not supported: suspicion diminishedFixed conviction, spread, or hallucinations. Persistent late-onset psychotic symptoms have predicted dementia in research cohorts (Ismail et al., 2023; Stafford et al., 2023)
Medication effectsBupropion raised, since its label lists paranoia, delusions, and confusion among possible reactions (Bausch Health US, 2025)Bupropion excluded by timing; diphenhydramine a probable contributor to slowing, not to the beliefDiphenhydramine stopped; bupropion restarted without new symptoms, with blood pressure watched because the label warns that bupropion can raise it (Bausch Health US, 2025)New symptoms after a medication change
DeliriumNot raisedNot supported: attention steady; no fluctuation at homeNot supportedFluctuating attention or an abrupt change
Financial exploitationDismissed by the family. Past-year financial abuse affects an estimated 6.8% of older adults in community studies (Yon et al., 2017)A real, undisclosed transfer from joint savings by Rose, so Albert’s central suspicion had a factual basis. Whether it was lawful and proper between spouses is a legal question outside the evaluation, and California’s community-property and joint-account rules bear on it (Cal. Fam. Code § 1100, 2025; Cal. Prob. Code § 5301, 2025; Cal. Prob. Code § 5305, 2025). Reported to Adult Protective Services, because California makes psychologists mandated reporters of suspected financial abuse of an elder, which includes taking or helping another take an elder’s property for a wrongful use (Cal. Welf. & Inst. Code § 15610.30, 2026; Cal. Welf. & Inst. Code § 15630, 2026); the report gave the facts and left any finding to the agency. Exploitation not established; Victor’s intent unknown. Whether Victor exerted undue influence on Rose was not assessed, and since diminished capacity is not a necessary component of undue influence, a finding about her capacity would not have settled it (American Bar Association Commission on Law and Aging & American Psychological Association, 2008)Adult Protective Services met with Albert and Rose in April and closed its case; $2,000 returned; no new requests; weekly account checks continueNew requests, pressure for signatures, or unexplained withdrawals
Self-neglectRaised by weekly bathingNot supported: no consequences; a lost routine and a physical barrier identifiedNot supportedFurther weight loss, missed medications, unpaid bills, or an unsafe home
Prolonged grief disorderNot raisedCriteria not met (Prigerson et al., 2021)Not applicableNear-daily yearning or preoccupation with impairment

Note. Weights are clinical judgments in a fictional case, not probabilities, and hypotheses are not mutually exclusive. “At referral” gives what the family’s account and the screen suggested before assessment. A sustained remission, here and in S6, means at least six months without significant depressive symptoms. No formal financial-capacity assessment was undertaken, because no transaction was proposed and no one sought authority over his finances. The family’s question was answered from the presumption of capacity, his history, and his preserved everyday financial skill, and a decision-specific assessment would precede any transfer of control. APA = American Psychiatric Association; GDS-15 = Geriatric Depression Scale, 15-item version.

Follow-up interpretation at six months

The six-month visit supports the formulation but cannot prove its etiology, because many changes began within weeks of one another. CPAP, diphenhydramine withdrawal, bupropion, problem-solving therapy, the Saturday walks, and the Sunday alarm all started in the same weeks as the family disclosure, couples therapy, and the report to Adult Protective Services. Victor’s return of two thousand dollars in July may also have eased his mood.

The order in which sleep, energy, and mood improved is consistent with several of these changes, and the strands of the follow-up moved separately. Sleep improved within weeks and mood over months, and emotional loneliness fell below the subscale’s lonely band after weekly confiding walks with David began. That was a two-item change on a six-item subscale, and the total stayed in the same moderately lonely band. Social loneliness, the Lubben total, and self-started activity changed little, and processing speed stayed below expectation on alternate forms. Instrumental independence held with the supports in place, and his financial vigilance diminished but did not end.

The speed finding is reported as qualitative persistence of a relative weakness, because two testings six months apart on alternate forms cannot show that a score is stable. Each score carries measurement error, alternate forms are not perfectly equivalent, and practice usually raises a retest score, so without comparable scores the recheck cannot show whether he improved. A score can also improve and still sit below expectation, so the persistence of the weakness does not rule out recovery. No reliable-change calculation was made, since it would require the alternate-form reliability and standard error of each task at that interval.

Slowing is also nonspecific, and in one cohort it was among the deficits most often present a year after late-life depression remitted (Bhalla et al., 2006). Its persistence after the apnea and the drug were treated therefore narrows the explanations, and it does not name one. The vascular weight was left where it stood, and the twelve-month retest, with the same norms and a reliable-change method, is the point at which the question can be asked properly.

Note on the apathy criteria

The 2018 international criteria apply across brain disorders, naming major and mild neurocognitive disorders, schizophrenia, and major depression among the conditions in which apathy occurs (Robert et al., 2018). They require a quantitative reduction in goal-directed activity compared with the person’s previous level, reported by the patient or observed by others. The reduction must appear in at least two of three dimensions, behavior and cognition, emotion, and social interaction, for at least four weeks and most of the time. It must also cause clinically significant impairment, and under criterion D it must not be explained exclusively by physical or motor disability, reduced consciousness, a substance, or major changes in the person’s environment. The criteria give blindness and loss of hearing as examples of physical disability and a medication as an example of a substance. Depression is not among the exclusions, so the criteria allow apathy to be identified alongside a depressive disorder.

The 2021 criteria apply only to people who already meet criteria for a neurocognitive disorder (Miller et al., 2021). They require at least one symptom in at least two of three dimensions, diminished initiative, diminished interest, and diminished emotional expression or responsiveness, as a change from usual behavior. The symptoms must persist or recur for at least four weeks, cause clinically significant impairment, and not be explained exclusively by psychiatric illness, intellectual, physical, or motor disability, a change in consciousness, or a substance. The 2021 set did not apply to Albert, because no neurocognitive disorder was established.

The formulation keeps apart several statements about Albert that are easy to run together. Reduced initiation was observable in the informant accounts, at home, and selectively in the office, and it is a finding. Criteria A through C of the 2018 set were met, with reductions in the behavior and cognition dimension and the social interaction dimension and with emotional responsiveness intact. The reduction had lasted well over four weeks, was present most of the time, and impaired his functioning.

His activity fell after retirement, the death of the friend who structured his mornings, the loss of every routine built on that friendship, and a rupture inside his family. A possible hearing loss and a nightly anticholinergic drug were also present, and both belong to criterion D’s exclusions. The depressive episode, which the criteria do not exclude, explained still more of the reduction, so criterion D, which bars the diagnosis only when the excluded factors explain the reduction entirely, did not bar it. By the letter of the 2018 criteria, then, apathy could have been diagnosed alongside the depression.

The decision not to diagnose a separate apathy syndrome was a clinical judgment and not a result the criteria produce. It rests on Marin’s definition of apathy, as summarized by Lanctôt et al. (2023), as a loss of motivation not attributable to emotional distress, cognitive impairment, or reduced consciousness. His loss of drive could not be separated from the distress of the depressive episode, and the share left after the depressive, environmental, and drug contributions could not be measured. The working formulation, that depressive and environmental contributors explained his reduced goal-directed behavior, was therefore held with a residual of uncertainty, and it is not a finding that apathy is absent.

The distinction between apathy and depression did not rest on treatment response, because the two differ in their phenomenology. Marin’s definition excludes motivation loss that distress explains, and it does not require the absence of distress, since the two frequently co-occur in neurocognitive disorders (Lanctôt et al., 2023). In those disorders, people with depression are more often pessimistic, anxious, ruminative, possibly suicidal, and troubled by vegetative symptoms (Lanctôt et al., 2023). Apathy without depression tends instead toward passive, compliant behavior without those features. Albert’s withdrawal came with worry, anger, early waking, weight loss, and a passive wish not to wake up, and his emotional responsiveness was intact. Those features placed his reduced initiation inside the depressive episode for the time being.

His initiation was also selective, which the office showed as plainly as the home did. He set the terms of the evaluation, brought the ledger, opened it to the disputed page, and spoke unasked about Ray. He started no walk, no call, and no Sunday review, and in the office he asked nothing about the tests and gave no reminder at 11:30. Initiation is itself a neurobehavioral function, which is why Levy and Dubois (2006) defined apathy as a quantitative reduction of self-generated voluntary and purposeful behavior. At its severe end they described an auto-activation deficit, in which self-initiated action is lost while externally driven behavior is relatively spared.

Albert’s pattern was milder and differently shaped, since the actions he still generated were the ones with the strongest affective pull, and preserved initiation for those goals excludes neither apathy nor executive change. Treatment response adds corroboration over time, because initiation that stays low after a sustained remission would need another explanation, and the judgment against a separate diagnosis would then be revisited.

S4. Published Estimates Behind the Evidence

Table S4

Published Estimates Behind the Evidence Cited, by Clinical Question

SourceDesign, population, and sizeEstimate (95% CI)Heterogeneity, follow-up, and limits
Does loneliness or isolation point toward dementia?
Luchetti et al. (2024)Meta-analysis of new cohort analyses and published longitudinal data; cohorts from North America, Europe, Asia, and Oceania; All-cause dementia: 21 samples, N = 608,561All-cause dementia HR 1.306 (1.197–1.426); Alzheimer’s disease 1.393 (1.290–1.504; k = 5); vascular dementia 1.735 (1.483–2.029; k = 3)I² = 87.04%. Models adjusted for depressive symptoms, social isolation, and/or modifiable risk factors: HR 1.189 (1.101–1.285; k = 18). Follow-up length did not moderate the estimate (p = .586). Heterogeneity fell when the largest sample, from the UK Biobank (n > 450,000), was excluded. Observational; the abstract’s causal wording outruns the design; does not show that reducing loneliness lowers risk
Shen et al. (2022)UK Biobank prospective cohort, mean baseline age 57; N = 462,619; 4,998 dementia cases over a mean 11.7 yearsLoneliness, fully adjusted: HR 1.04 (0.94–1.16), with 75% of the association attributable to depressive symptoms. Social isolation: HR 1.26 (1.15–1.37), independent of loneliness and depressionFull adjustment included depression. Mostly a midlife baseline; observational; the same cohort supplied the largest sample in Luchetti et al. (2024)
Sutin et al. (2020)Health and Retirement Study, adults aged 50 and older; N = 12,030; 1,104 dementia cases over up to 10 yearsPer point of loneliness: HR 1.40 (1.26–1.56) with demographics; 1.19 (1.05–1.33) after adding depressive symptomsAt least 6 years of follow-up: HR 1.35 (1.16–1.58). Excluding baseline cognitive impairment: HR 1.44 (1.23–1.69). Dementia classified by a cognitive-test cut-off; sensitivity analyses reduce but do not remove reverse causation
Livingston et al. (2024)Lancet Commission consensus reportFor social contact, seven of eight studies in one review had under four years of follow-up, “making reverse causation likely”Population attributable fractions assume causality. States that existing intervention studies are too small and too short to show any effect on dementia incidence
Coyle and Dugan (2012)Health and Retirement Study, 2006 and 2008; n = 11,825Loneliness and social isolation correlated at r = .201Cross-sectional. Association only; supports measuring the two separately
Is loneliness feeding the depression?
Mann et al. (2022)Systematic review and meta-analysis of longitudinal studies; six of seven pooled studies in older adults; 7 studies (8 cohorts)Depression onset, often lonely versus not: adjusted OR 2.33 (1.62–3.34)I² = 64%; random effects. Three studies suggested reciprocal effects; does not show that loneliness causes depression
Do apathy and late-life depression predict dementia?
van Dalen et al. (2018)Systematic review and meta-analysis of prospective cohorts without dementia, mostly memory-clinic samples with mild cognitive impairment; 16 studies, N = 7,365Dementia RR 1.81 (1.32–2.50; 12 studies); HR 2.39 (1.27–4.51; 7 studies)I² = 76% (RR) and 90% (HR); the association weakened with longer follow-up, age, and cognitive impairment. Clinical samples; not generalizable to community patients; consistent with a prodrome
Brain et al. (2025)Umbrella review of 9 systematic reviews with a meta-analysis of hazard ratios; Late life: 18 studies, n = 901,762. Midlife: 7 studiesLate-life depression HR 1.95 (1.68–2.26); midlife 1.56 (1.12–2.18)I² = 77.5% (late life) and 97.5% (midlife); most included reviews rated low or critically low quality. The authors leave open whether the late-life association reflects a risk factor or early neurodegeneration
Singh-Manoux et al. (2017)Whitehall II occupational cohort, recruited at ages 35 to 55; N = 10,189; 322 dementia cases over 28 yearsSymptoms in 1985: HR 1.21 (0.95–1.54). Symptoms in 2003: HR 1.72 (1.21–2.44)Depressive symptoms measured nine times. Supports a prodrome or shared causes
Bhalla et al. (2006)Longitudinal cohort; nondemented late-life depression, assessed while depressed and one year later; 56 patients and 40 comparison participants45% cognitively impaired one year after remission; 23% of those cognitively normal while depressed became impaired; visuospatial ability, processing speed, and delayed memory most often impairedOne-year follow-up. Small sample; no dementia outcome
Diniz et al. (2013)Systematic review and meta-analysis of community-based cohorts; baseline age over 50; 23 studies; all-cause dementia N = 49,612Pooled risk: all-cause dementia 1.85 (1.67–2.04); Alzheimer’s disease 1.65 (1.42–1.92); vascular dementia 2.52 (1.77–3.59); vascular risk higher than Alzheimer’s risk in the five studies reporting both (p = .03)Low heterogeneity for all-cause and vascular dementia; follow-up 1 to 17 years. Pooled risk combines hazard and odds ratios, and the authors advise caution; depression defined by scale cut-offs
Lanctôt et al. (2023)Narrative review; neurocognitive disordersApathy and depression co-occur in an estimated 14% to 38%Review-level estimate. Applies to neurocognitive disorders
Yuen et al. (2015)Uncontrolled open-label escitalopram treatment; nondemented older adults with nonpsychotic major depression; N = 7138% had significant apathy at baseline; change in apathy with treatment was significant but smallTwelve weeks of escitalopram 10 mg daily. No control arm for apathy change
Do medical contributors and memory type matter?
Leng et al. (2017)Systematic review and meta-analysis; sleep-disordered breathing; studies with mean age 40 or older; 14 studies; 4,288,419 participantsCognitive impairment, prospective studies: RR 1.26 (1.05–1.50). Executive function, cross-sectional: slightly worse. No association with global cognition or memorySignificant heterogeneity. Association; does not show that treatment restores cognition
Swartz et al. (2025)Canadian Stroke Best Practice Recommendations, vascular cognitive impairment module, 7th edition (2024); consensus guidelineVascular cognitive impairment is “a range of new or worsening cognitive deficits” attributed to or accelerated by cerebrovascular injury; beginning confluent or confluent subcortical white-matter hyperintensity on the Fazekas scale “is sufficient to cause clinical cognitive impairment in many individuals”The threshold of vascular damage “is not clear, and will likely vary between patients due to differing levels of cognitive reserve”; depression can worsen vascular cognitive impairment, and depression in the context of stroke can mimic it; screening for sleep and mood disorders is recommended. A guideline, not an estimate; read in the web edition of the module, with the citation confirmed through Crossref
Henry et al. (2004)Meta-analysis of prospective memory and aging; k and N not extractedLaboratory age effects: time-based r = −.39 and event-based r = −.34; within event-based tasks, r = −.40 for high and −.14 for low strategic demandNaturalistic studies favored older adults (r = .35 event-based, .52 time-based). Abstract-level verification; concerns normal aging, not disease
Is late-life suspiciousness a warning sign?
Linszen et al. (2016)Review and meta-analyses of hearing impairment and psychosis; mixed ages; n = 227,005 to 250,470 per outcomeDelusions OR 1.55 (1.36–1.78); psychotic symptoms 2.23 (1.83–2.72)Heterogeneity not extracted. Association; not specific to older adults
Ismail et al. (2023)Research-center cohort (NACC); normal cognition or MCI at baseline; 66 with psychosis meeting MBI criteria, 72.7% with MCIIncident dementia HR 3.76 (2.53–5.58)Convenience sample; most affected participants already had MCI
Stafford et al. (2023)Swedish register cohort; very-late-onset schizophrenia-like psychosis; 15,409 cases; 154,090 comparatorsDementia HR 4.22 (4.05–4.41); after a two-year washout, 2.22 (2.10–2.36)Excess risk strongest in the first year. Register diagnoses; early dementia misread as psychosis contributes to the signal
Yon et al. (2017)Meta-analysis of community prevalence studies; adults aged 60 and older; 52 studies from 28 countriesFinancial abuse, past year, 6.8% (5.0–9.2)Prevalence varied with sample size, income classification, and data collection method. A prevalence, not a probability for one family
What helps, and what can it not promise?
Orgeta et al. (2017)Systematic review and meta-analysis of RCTs; stand-alone analysis in community-dwelling older adults with depression or subthreshold symptoms; Stand-alone analysis: 5 studies, n = 175Behavioral activation versus usual care: SMD −0.72 (−1.04 to −0.41) at 4 to 12 weeksI² = 0%; GRADE moderate. Small trials; short-term symptom change; no cognitive outcome
Areán et al. (2010)RCT; adults aged 60 and older with major depression and executive dysfunction; n = 221Problem-solving therapy exceeded supportive therapy in severity, response, and remission (remission 45.6% versus 27.8% at week 12)Twelve weekly sessions. Single trial; participants were selected for executive dysfunction, which Albert’s testing did not establish
Alexopoulos et al. (2021)Non-inferiority RCT; adults aged 60 and older with major depression, treated by community social workers; n = 249The streamlined therapy Engage was non-inferior to problem-solving therapy, which the abstract calls evidence-basedNine weekly sessions; blind raters; Engage therapists needed about 30% less training time. Problem-solving therapy was the comparator; the abstract did not describe selection for executive dysfunction; the full text was not read
Shekelle et al. (2024)Systematic review and meta-analysis; community-living older adults; 60 studies (36 RCTs)Group-based treatment SMD −0.25 (−0.42 to −0.08) in the Results, −0.27 (−0.48 to −0.08) in the abstract; internet training −0.22 (−0.30 to −0.14)Moderate certainty; insufficient evidence for telephone or individual contact. Small effects; durability unknown
Clegg et al. (2026)Cochrane review of RCTs; adults with depression, all ages; 57 trials against control (2,189 participants)Exercise SMD −0.67 (−0.82 to −0.52), low certainty; trials with adequate allocation concealment, intention-to-treat analysis, and blinded assessment −0.46 (−0.88 to −0.04; 7 trials)Long-term effect very low certainty. Not specific to older adults; not cited in the article
Lamb et al. (2018)Pragmatic RCT; mild to moderate dementia; n = 494Cognitive scores at 12 months slightly favored usual care: adjusted difference −1.4 (−2.6 to −0.2)Four months of supervised exercise. Exercise did not slow decline in established dementia
Do the screening measures discriminate?
Pocklington et al. (2016)Diagnostic accuracy meta-analysis of brief GDS versions; 32 studies across versionsGDS-15 at a cut-off of 5: sensitivity .89 (.80–.94), specificity .77 (.65–.86)Possible selective reporting of cut-offs. Screening accuracy, not diagnosis
Hendry et al. (2019)Cochrane diagnostic accuracy review of the AD8; 9 studies with data; 4,045 peopleCut-off 2: sensitivity .92 (.86–.96), specificity .64 (.39–.82)Significant heterogeneity and risk of bias. “Not a diagnostic test in itself”
González et al. (2022)Psychometric evaluation in the National Alzheimer’s Coordinating Center Uniform Data Set; N = 27,916Reliability moderate to strong, lower within diagnostic subgroups; correlated with neurocognitive measures; classified impairment stages accurately; one factorAbstract-level verification; most sensitive to mild functional difficulty; research-center sample
Dierckx et al. (2007)Cross-sectional diagnostic study; 40 MCI, 35 Alzheimer’s disease, 46 depression, 52 controlsCombined Visual Association Test and Memory Impairment Screen-plus score: MCI versus depression sensitivity 58%, specificity 85%Cut-off optimized by ROC analysis in the same sample, so accuracy is likely optimistic. Specific to that combined procedure, cut-off, and sample

Note. Values are reproduced as printed in the full text or abstract reviewed; S7 describes how each was checked. CI = confidence interval; HR = hazard ratio; OR = odds ratio; RR = risk ratio; SMD = standardized mean difference; RCT = randomized controlled trial; MCI = mild cognitive impairment; MBI = mild behavioral impairment; NACC = National Alzheimer’s Coordinating Center; GDS = Geriatric Depression Scale; ROC = receiver operating characteristic. None of these estimates is an individual’s absolute risk.

Notes on independence and causal inference

Loneliness, social isolation, social contact, and social participation are different exposures, beginning with loneliness, a subjective discrepancy between the relationships a person wants and the relationships the person has (de Jong Gierveld & van Tilburg, 1999). Emotional and social loneliness are separable dimensions of it (van Baarsen et al., 2001), while isolation is the objective state of having few relationships or little contact (National Academies of Sciences, Engineering, and Medicine, 2020). Loneliness and isolation correlated at only r = .201 in a large U.S. sample (Coyle & Dugan, 2012). One meta-analysis reported separate dementia estimates for low social participation (RR 1.41), less frequent contact (1.57), and more loneliness (1.58; Kuiper et al., 2015). A finding about contact frequency therefore does not describe the consequences of emotional loneliness.

The sources are not independent confirmations of one another, because the same cohorts recur across them. The largest sample in Luchetti et al. (2024), whose exclusion reduced the heterogeneity, came from the UK Biobank, the cohort that Shen et al. (2022) analyzed. The Health and Retirement Study is also among the data sources of Luchetti et al. (2024), and it supplies the cohort analyzed by Sutin et al. (2020) and by Coyle and Dugan (2012). It also appears among the cohorts listed by Mann et al. (2022), although whether it entered Mann’s pooled estimate was not confirmed. The Lancet Commission’s statement on loneliness draws on three earlier reviews, including Kuiper et al. (2015) and Lara et al. (2019), whose included cohorts were not compared here (Livingston et al., 2024). Agreement among these sources is better read as repeated observation of overlapping populations than as independent replication.

Depression may be a confounder of the loneliness association or a step on its path. In the UK Biobank, depressive symptoms accounted for most of it (Shen et al., 2022), while in the Health and Retirement Study it persisted after adjustment (Sutin et al., 2020). Because the UK Biobank also supplies the largest sample in Luchetti et al. (2024), the contrast between Shen’s adjusted estimate and Luchetti’s pooled adjusted estimate is not a contrast between independent data. Luchetti’s adjusted estimate of 1.19 also pools models adjusted for depressive symptoms, social isolation, or other modifiable risk factors, so it does not reflect one consistent adjustment for depression. Observational data cannot decide between these readings, and an adjusted estimate is not a measure of a causal effect.

Reverse causation and prodrome remain live explanations, since social activity declines during the preclinical phase of dementia (Livingston et al., 2024) and the apathy association weakened with longer follow-up (van Dalen et al., 2018). The authors of the umbrella review left open whether late-life depression is a risk factor or an early stage (Brain et al., 2025). In Whitehall II, later-life symptoms predicted dementia while midlife symptoms did not (Singh-Manoux et al., 2017), and the sensitivity analyses in Sutin et al. (2020) reduce this concern for loneliness but do not remove it.

Heterogeneity was large in the central estimates, with I² of 87% for loneliness, 76% for apathy, and 77.5% for late-life depression. The populations also differ, from memory-clinic patients in van Dalen et al. (2018) to a midlife baseline in Shen et al. (2022) and adults of all ages in Clegg et al. (2026). Prediction is not intervention, since none of the intervention trials measured dementia and the Commission judged existing social-intervention studies too small and too short to show an effect on incidence (Livingston et al., 2024). In people with established dementia, four months of supervised exercise did not slow cognitive decline (Lamb et al., 2018).

S5. Collateral Interview Framework

Interview each informant separately where possible, after the patient has agreed to the contact and to what may be shared. Ask for dated, specific examples, and code each answer by its source so that observation, report, and interpretation stay distinct. The worked example after the template shows how one discrepancy in this case was investigated.

Informant: ______________     Relationship: ______________

Years known: ______________     Date of interview: ______________

Interviewer: ______________

Contact with the patient (in person or by phone, and how often): ______________

Language and interpreter, if any: ______________

Consent confirmed by the patient (yes or no): ______________

Informant’s own stake in the matters discussed (for example, a party to the financial concern, or related to someone who is): ______________

Source codes for each answer: O = observed directly; R = reported by the patient; H = heard from someone else; I = informant’s interpretation

1. Purpose, consent, and limits (with the patient, before any informant contact)

  • What do you understand the evaluation to be for, and what worries you about it?
  • Whom do you agree I may contact, what may be shared with them, and who receives the results?
  • Do you want a professional interpreter, and is there anyone you do not want involved?
  • Explain the limits of confidentiality, including the duty to report suspected abuse of an older or dependent adult, and confirm that the patient understands them.

2. The person before the change

  • How would you have described the patient before the change, in sociability, talkativeness, and the ways they showed care?
  • Which roles did the patient hold in the family, such as keeping accounts, fixing things, or organizing gatherings?
  • Has the patient always been private about feelings, or is that new?

3. Onset, course, and lifelong habits

  • What did you notice first, and when, and what was happening in the patient’s life at that time?
  • Which behaviors are lifelong, such as misplacing keys, and which are new?

4. Stopped tasks

  • For each task the patient stopped, could they no longer do it, did they no longer start it, or did they no longer want to?
  • Once someone else starts an activity, does the patient join in and finish it?

5. Memory failures and their consequences

  • Describe the last time the patient forgot something that mattered, and what happened as a result.
  • Did the patient know the date or the task when asked afterward?

6. Mood, interest, and responsiveness

  • Does the patient seem sad, empty, irritable, or indifferent, and do they enjoy anything once it begins?
  • Has the patient said that life is not worth living or that others would be better off without them, and are there firearms, stored medications, or other lethal means at home?

7. Money and possible exploitation

  • Who manages bills, banking, and lending now, and who did before? Have there been missed payments, unusual transactions, or requests for loans?
  • What do you know firsthand about any money dealings within the family?

8. Suspicious beliefs

  • What exactly does the patient believe, what do they cite as evidence, and when did it start?
  • How firmly is the belief held, and has it spread to other people or topics (Kendler et al., 1983)?
  • Has the patient acted on the belief or threatened to?

9. Medications, sleep, and health

  • Which prescriptions, sleep aids, or other substances does the patient use, and when did each start, change, or stop?
  • Is there snoring, pauses in breathing, daytime sleepiness, acting out of dreams, falls, or a change in walking, and does anyone share the bedroom who could see it?

10. Connection and setting

  • Who does the patient talk with, how often, and who usually starts the contact? In which settings is the patient most and least engaged?
  • Has anyone important to the patient died, moved away, or withdrawn?

11. Values and wishes

  • What gives the patient a sense of being useful or respected, and what kind of contact do they seem to want?

Closing questions

  • Is there anything you would hesitate to say in front of other family members?
  • What would change your own view of what is happening?

Reconciliation log (complete after all informants)

  • For each disagreement, enter the topic, each informant’s account, each informant’s opportunity to observe, possible sources of bias, the plan for investigating it, and whether the evidence was sufficient to reconcile it.

Reporting check (complete before feedback)

  • If any account gives reasonable suspicion of abuse, including financial abuse, of an older or dependent adult, make the report the law requires, tell the patient, and document the facts reported and the reasoning.

Table S5

Worked Example: Investigating the Disputed Transfer

SourceOpportunity to observeStake in the questionWhat the source reportedHow it was checkedWeight given
AlbertKnew of Victor’s January request; kept the ledgerThe belief is his, and he feared losing control of the financesMoney left the account for Victor with Rose’s help; Victor would keep coming backLedger entry; Rose’s interview; the bank statement, reviewed with his consentCore belief accurate; forecast unconfirmed
MichelleOne evening a week in a crowded room, often with Victor presentWants peace in the family; fears dementia; close to her uncleHer father accused Victor without basisHer information came from Victor’s denial and her mother’s silenceObservations accurate; inference about the money wrong because her information was incomplete
RoseDaily contact; made the transferThe strongest reason to keep the transfer quietShe took $15,000 from joint savings in February 2025 to keep Victor’s dry cleaner open, without telling AlbertSeparate interview with a professional interpreter; matched the ledger entry and the bank statementDecisive on the fact of the transfer
DavidMonthly visits, mostly one-on-one in his garage until the summer of 2025Would take on financial tasks if askedHis father stopped coming out to the garage that summer and was quieter, though he still talked about cars when David raised them and talked more easily alone than at Sunday dinnersTelephone interview; consistent with Albert’s account and Rose’sCorroborates the loss of interest and the context-dependence of engagement
VictorNot interviewed, because Albert withheld consentReceived the moneyAt the dinner, said he had never taken anything from AlbertNot applicableA statement that was literally accurate and misled the family

Note. Fictional case. Collateral contact followed the patient’s consent. At the start, Albert and each informant were told that what they said could appear in the report and that suspected abuse of an older adult would be reported as California law requires.

S6. Longitudinal Monitoring Framework

The formulation at feedback is provisional by design, so the schedule below keeps it under review, and the triggers that follow can reopen it at any point.

Table S6

Monitoring Schedule, With the Fictional Course

Time pointClinical actionsMeasuresDecision ruleCourse in the fictional case
BaselineFeedback to Albert, then to the family with his consent; safety plan, including the firearm transfer and the medication supply; report to Adult Protective Services; referrals for therapy, prescribing (bupropion to continue well beyond remission this time), sleep medicine, and audiometry; financial safeguards agreed; no driving when drowsy until sleep improves; grab bar beside the tubFull evaluation (Table S2)Formulation documented with weights (Table S3)Completed; pistol moved to his son’s house; Michelle to bring new changes to her father first
6 to 8 weeksReview therapy, CPAP use, and the diphenhydramine taper; prescriber reviewGDS-15; sleep and CPAP useEscalate at once for suicidal thinking, worsening self-care, or new psychotic featuresSleep better; diphenhydramine stopped; GDS-15 of 7
3 monthsFamily session; review the Saturday walks and the Sunday alarmGDS-15; AES self and informant; FAQApathy that persists while mood improves is expected and is rechecked; apathy that outlasts a sustained remission prompts reformulationGDS-15 of 5; Michelle’s AES rating lower; self-starting still limited
6 monthsBrief cognitive recheck with alternate forms of the two speed tasksGDS-15; DJG-11; LSNS-6; AES; FAQ; processing speedSlowing that persists after sleep and medication changes narrows its explanations and keeps a vascular contribution open; it does not by itself raise that weight, and a qualitative recheck cannot show stabilityGDS-15 of 4; emotional loneliness 2 of 6, social 3 of 5; Lubben total 10, unchanged; speed still his clearest weakness, change not determinable; two self-started walks; no missed bills or doses; bathing every few days; $2,000 returned; Adult Protective Services case closed in April; mild hearing loss in both ears, aids worn in the car and not on Sundays and booked for resetting; couples therapy begun; neurological consultation deferred by his choice
12 monthsFull retest with alternate forms and the same norms, with reliable-change comparisonFull evaluation; AD8 and FAQ from the same informantsDecline beyond reliable-change limits or new functional loss reopens neurodegenerative hypotheses and prompts neurology referral if none has been made; the 2018 apathy criteria are reapplied if initiation remains low after a sustained remissionScheduled for March 2027
OngoingAnnual review, or sooner when a trigger below appearsCore measuresThe formulation stays open, and a neurological consultation can be requested at any point when cognitive, behavioral, functional, or neurological concerns justify itNot applicable

Note. Abbreviations as in Table S2; AES = Apathy Evaluation Scale; DJG-11 = De Jong Gierveld Loneliness Scale, 11 items; FAQ = Functional Activities Questionnaire; LSNS-6 = Lubben Social Network Scale, 6 items. Intervals are defaults and should follow clinical change. Baseline evaluation March 2026; six-month visit September 2026.

Triggers that reopen the formulation at any time

Any of these calls for review before the next scheduled visit, and the neurological ones call for a neurological consultation without waiting for the twelve-month retest.

  • New or worsening forgetfulness with practical consequences, or loss of an instrumental activity such as finances, medications, or driving
  • Fluctuating attention, acute confusion, or altered alertness
  • Hallucinations, misidentifications, or a suspicious belief that becomes fixed or spreads to new people or themes
  • Personality change, disinhibition, loss of empathy, or compulsive behavior
  • Parkinsonism, falls, a change in gait, or acting out of dreams
  • Further weight loss, missed medications, unpaid bills, an unsafe home, or other signs of self-neglect
  • Suicidal thoughts, statements of hopelessness or of being a burden, or renewed access to firearms or other lethal means
  • New requests for money, pressure for signatures, or unexplained withdrawals

Risk assessment and firearm access

Albert’s remarks that the family would manage fine without him led to direct questions about death and suicide. He acknowledged passive suicidal ideation, a recurrent wish not to wake up, without active ideation, plan, intent, preparation, or prior attempt. That was weighed against the factors that raise risk in an older man, among them the depressive episode, insomnia, loneliness, the death of his closest friend in 2023, family conflict, and access to a firearm. His remark about the family was also weighed as perceived burdensomeness, which the interpersonal theory of suicide links to the desire for death (Van Orden et al., 2010).

Against those factors stood the absence of any plan, intent, preparation, or earlier attempt, and his grandchildren as a reason to live. The unsettled money also mattered to him, though a reason tied to an open dispute can end when the dispute does, so it was not counted as a durable protection. His medications were reviewed for quantity and storage, the diphenhydramine on hand was cut to what the taper needed, and bupropion was dispensed in limited quantities at first. The plan was made with him, and he agreed to tell David or the examiner if the thought grew stronger or began to include a plan.

A firearm at a relative’s home is only as restricted as the relative keeps it, so the arrangement specified storage, access, and return. The pistol was locked unloaded in a safe that only David could open, and David agreed in front of his father that it would not come back on request. California treats a transfer from a parent to an adult child as an intrafamilial transfer exempt from the licensed-dealer requirement, and a return follows the same rules (California Department of Justice, n.d.). The department’s current reporting page requires the recipient to hold a valid firearm safety certificate or a qualifying exemption (California Department of Justice, 2024). The transfer must be reported within thirty days, and the report can be filed online. Those requirements change, so a clinician should confirm the current rules with the Department of Justice before relying on this summary. Risk was reassessed at each visit, and the triggers above include renewed access to firearms.

Reading the response

  • Here and in S3, a sustained remission means at least six months without significant depressive symptoms.
  • Improvement after treatment supports the formulation but does not prove it, because time, regression toward the mean, and simultaneous changes can also produce recovery.
  • Apathy that persists while mood improves is expected, since apathy in late-life depression is common and changed little with an antidepressant (Yuen et al., 2015), and antidepressants are not thought to address it (Lanctôt et al., 2023). Early persistence therefore settles nothing, and only apathy that outlasts a sustained remission reopens the question.
  • Slowing that persists after sleep apnea treatment and anticholinergic withdrawal narrows its explanations and keeps a vascular contribution open, though it does not raise that weight. Decline beyond reliable-change limits at twelve months prompts neurological referral if none has been made, and nothing in this schedule delays an earlier consultation.
  • A mild behavioral impairment checklist becomes useful only if behavioral change persists after the depression remits and the family dispute is addressed (Ismail et al., 2017).
  • Twelve months of stable testing and function lowers concern for neurodegeneration but does not eliminate it.

S7. Methods, Verification, and Limitations

This is a fictional case-based narrative review, in which the case, its history, its dialogue, its test results, and its outcome were invented for teaching. The literature was drawn from existing systematic reviews, meta-analyses, consensus criteria, guidelines, and selected primary studies. No original meta-analysis, systematic search, or pooling was performed, and sources were chosen for their bearing on the decisions in the case.

The fictional results were constructed to be internally coherent, with one clear weakness in processing speed on two concordant tasks, three results at the low end of expectation, and intact retention and recognition. They are reported as qualitative illustrations, worded as within, at the low end of, or below expectation, because no normative calculation was performed and no standard score or percentile should be inferred. Table S2a sets out what that choice means for each measure and what a calculated profile would require. Questionnaire totals are raw scores read against each instrument’s published convention, and the prospective-memory errands and the mock bank statement are structured observations without norms. The chronology was set out as a master timeline, given in Table S1, and every date, age, interval, and medication exposure in the article and the supplement was checked against it.

Every quantitative claim was checked against the source cited for it, and where a value was seen only as an attribution in a secondary source, the text says so. Luchetti et al. (2024) were read in the PubMed Central author manuscript and Leng et al. (2017) at the publisher. Robert et al. (2018) were read in the Cambridge Core full text, with criterion D read in full. The capacity handbook was read at the American Psychological Association’s site, and four of its statements were checked by page. The first three hold that the law starts from a presumption of capacity, that a disorder does not by itself mean capacity is impaired, and that capacity depends on the decision in question. The fourth holds that diminished capacity is not a necessary component of undue influence (p. 21).

The 2024 Canadian recommendations on vascular cognitive impairment were read in the module’s web edition at the Canadian Stroke Best Practices site, and the citation was confirmed through Crossref (Swartz et al., 2025). Levy and Dubois (2006) were read at the publisher’s site, and Henry et al. (2004) were checked at the abstract level. The Beers entry for first-generation antihistamines was checked in the American Geriatrics Society pocket card reproduced by a university geriatrics program, and its citation was confirmed through the bibliographic listing. The other estimates in Table S4 were checked against full texts, publisher pages, or abstracts, and the entry for Van Orden et al. (2010) was confirmed through Crossref.

California Family Code section 1100, Probate Code sections 5301 and 5305, and Welfare and Institutions Code sections 15610.30 and 15630 were read in public reproductions of the codes, not on the legislature’s own site. The Department of Justice guidance on intrafamilial firearm transfers was read at its site, and its firearm reporting page, last modified in October 2024, was read for the current certificate and reporting requirements. The de Jong Gierveld and van Tilburg manual was read in its updated web edition, which states both the total-score bands and the emotional and social subscale bands. The printed 1999 edition was not available to confirm whether it carried the subscale bands. The FAQ’s reliability and validity were checked in the abstract of González et al. (2022), and Teng et al. (2010) were confirmed through their bibliographic entry without the abstract.

The DSM-5-TR provisions the article and supplement rely on were checked in the DSM-5-TR text itself. They include the major depressive disorder criteria, the recurrence rule, and the severity specifiers, and the major and mild neurocognitive disorder criteria. They also include the vascular criteria, with the requirement that cerebrovascular disease be sufficient to account for the deficits, and the prolonged grief disorder criteria. The volume does not reprint the glossary of technical terms that DSM-5 carried, so the definition of delusions was checked in its chapter on psychotic disorders. The overvalued-idea definition follows the DSM-5 glossary as reproduced in secondary sources.

Most longitudinal evidence comes from Western cohorts, few instruments have been validated in the languages an older immigrant may speak, and nearly all of the risk evidence is observational. The case cannot show base rates, and its sequence of improvement is a teaching construction, not evidence that the same sequence would follow in a real patient.

S8. References

Adams, K. B. (2001). Depressive symptoms, depletion, or developmental change? Withdrawal, apathy, and lack of vigor in the Geriatric Depression Scale. The Gerontologist, 41(6), 768–777. https://doi.org/10.1093/geront/41.6.768

Alexopoulos, G. S., Raue, P. J., Banerjee, S., Marino, P., Renn, B. N., Solomonov, N., Adeagbo, A., Sirey, J. A., Hull, T. D., Kiosses, D. N., Mauer, E., & Areán, P. A. (2021). Comparing the streamlined psychotherapy “Engage” with problem-solving therapy in late-life major depression: A randomized clinical trial. Molecular Psychiatry, 26(9), 5180–5189. https://doi.org/10.1038/s41380-020-0832-3

American Bar Association Commission on Law and Aging & American Psychological Association. (2008). Assessment of older adults with diminished capacity: A handbook for psychologists. https://www.apa.org/pi/aging/programs/assessment/capacity-psychologist-handbook.pdf

American Geriatrics Society Beers Criteria® Update Expert Panel. (2023). American Geriatrics Society 2023 updated AGS Beers Criteria® for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society, 71(7), 2052–2081. https://doi.org/10.1111/jgs.18372

American Psychiatric Association. (n.d.). DSM-5-TR online assessment measures: Cultural Formulation Interview, informant version, and supplementary modules. Retrieved October 11, 2026, from https://www.psychiatry.org/psychiatrists/practice/dsm/educational-resources/assessment-measures

American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). https://doi.org/10.1176/appi.books.9780890425787

Areán, P. A., Raue, P., Mackin, R. S., Kanellopoulos, D., McCulloch, C., & Alexopoulos, G. S. (2010). Problem-solving therapy and supportive therapy in older adults with major depression and executive dysfunction. American Journal of Psychiatry, 167(11), 1391–1398. https://doi.org/10.1176/appi.ajp.2010.09091327

Bausch Health US. (2025). Wellbutrin XL (bupropion hydrochloride extended-release tablets) [Prescribing information; revised November 2025]. DailyMed, U.S. National Library of Medicine. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=a435da9d-f6e8-4ddc-897d-8cd2bf777b21

Bhalla, R. K., Butters, M. A., Mulsant, B. H., Begley, A. E., Zmuda, M. D., Schoderbek, B., Pollock, B. G., Reynolds, C. F., III, & Becker, J. T. (2006). Persistence of neuropsychologic deficits in the remitted state of late-life depression. American Journal of Geriatric Psychiatry, 14(5), 419–427. https://doi.org/10.1097/01.JGP.0000203130.45421.69

Blevins, C. A., Weathers, F. W., Davis, M. T., Witte, T. K., & Domino, J. L. (2015). The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. Journal of Traumatic Stress, 28(6), 489–498. https://doi.org/10.1002/jts.22059

Brain, J., Alshahrani, M., Kafadar, A. H., Tang, E. Y. H., Burton, E., Greene, L., Turnbull, D. A., Myers, B. J., Naheed, A., Siervo, M., Tully, P. J., & Stephan, B. C. M. (2025). Temporal dynamics in the association between depression and dementia: An umbrella review and meta-analysis. eClinicalMedicine, 84, Article 103266. https://doi.org/10.1016/j.eclinm.2025.103266

Cal. Fam. Code § 1100 (2025). https://california.public.law/codes/family_code_section_1100

Cal. Prob. Code § 5301 (2025). https://california.public.law/codes/probate_code_section_5301

Cal. Prob. Code § 5305 (2025). https://law.justia.com/codes/california/code-prob/division-5/part-2/chapter-3/section-5305/

Cal. Welf. & Inst. Code § 15610.30 (2026). https://codes.findlaw.com/ca/welfare-and-institutions-code/wic-sect-15610-30/

Cal. Welf. & Inst. Code § 15630 (2026). https://codes.findlaw.com/ca/welfare-and-institutions-code/wic-sect-15630/

California Department of Justice. (n.d.). Firearms frequently asked questions. Office of the Attorney General, Bureau of Firearms. https://oag.ca.gov/firearms/pubfaqs

California Department of Justice. (2024). Firearms reporting & law enforcement release application. Office of the Attorney General. https://oag.ca.gov/node/60090

Carson, N., Leach, L., & Murphy, K. J. (2018). A re-examination of Montreal Cognitive Assessment (MoCA) cutoff scores. International Journal of Geriatric Psychiatry, 33(2), 379–388. https://doi.org/10.1002/gps.4756

Chang, Q., Sha, F., Chan, C. H., & Yip, P. S. F. (2018). Validation of an abbreviated version of the Lubben Social Network Scale (“LSNS-6”) and its associations with suicidality among older adults in China. PLOS ONE, 13(8), Article e0201612. https://doi.org/10.1371/journal.pone.0201612

Clegg, A., Hill, J. E., Mullin, D. S., Harris, C., Smith, C., Lightbody, C. E., Dwan, K., Cooney, G. M., Mead, G. E., & Watkins, C. L. (2026). Exercise for depression. Cochrane Database of Systematic Reviews, 2026(1), Article CD004366. https://doi.org/10.1002/14651858.CD004366.pub7

Coyle, C. E., & Dugan, E. (2012). Social isolation, loneliness and health among older adults. Journal of Aging and Health, 24(8), 1346–1363. https://doi.org/10.1177/0898264312460275

Dean, A. C., Victor, T. L., Boone, K. B., Philpott, L. M., & Hess, R. A. (2009). Dementia and effort test performance. The Clinical Neuropsychologist, 23(1), 133–152. https://doi.org/10.1080/13854040701819050

de Jong Gierveld, J., & van Tilburg, T. (1999). Manual of the loneliness scale [Updated web edition, 2022]. Vrije Universiteit Amsterdam, Department of Social Research Methodology. https://home.fsw.vu.nl/TG.van.Tilburg/manual_loneliness_scale_1999.html

Dickerson, B. C., Atri, A., Clevenger, C., Karlawish, J., Knopman, D. S., Lin, P.-J., Norman, M. S., Onyike, C. U., Sano, M., Scanland, S., & Carrillo, M. C. (2025). The Alzheimer’s Association clinical practice guideline for the diagnostic evaluation, testing, counseling, and disclosure of suspected Alzheimer’s disease and related disorders (DETeCD-ADRD): Executive summary of recommendations for specialty care. Alzheimer’s & Dementia, 21(1), Article e14337. https://doi.org/10.1002/alz.14337

Dierckx, E., Engelborghs, S., De Raedt, R., De Deyn, P. P., & Ponjaert-Kristoffersen, I. (2007). Differentiation between mild cognitive impairment, Alzheimer’s disease and depression by means of cued recall. Psychological Medicine, 37(5), 747–755. https://doi.org/10.1017/S003329170600955X

Diniz, B. S., Butters, M. A., Albert, S. M., Dew, M. A., & Reynolds, C. F., III. (2013). Late-life depression and risk of vascular dementia and Alzheimer’s disease: Systematic review and meta-analysis of community-based cohort studies. British Journal of Psychiatry, 202(5), 329–335. https://doi.org/10.1192/bjp.bp.112.118307

Galvin, J. E., Roe, C. M., Powlishta, K. K., Coats, M. A., Muich, S. J., Grant, E., Miller, J. P., Storandt, M., & Morris, J. C. (2005). The AD8: A brief informant interview to detect dementia. Neurology, 65(4), 559–564. https://doi.org/10.1212/01.wnl.0000172958.95282.2a

González, D. A., Gonzales, M. M., Resch, Z. J., Sullivan, A. C., & Soble, J. R. (2022). Comprehensive evaluation of the Functional Activities Questionnaire (FAQ) and its reliability and validity. Assessment, 29(4), 748–763. https://doi.org/10.1177/1073191121991215

Guilmette, T. J., Sweet, J. J., Hebben, N., Koltai, D., Mahone, E. M., Spiegler, B. J., Stucky, K., Westerveld, M., & Conference Participants. (2020). American Academy of Clinical Neuropsychology consensus conference statement on uniform labeling of performance test scores. The Clinical Neuropsychologist, 34(3), 437–453. https://doi.org/10.1080/13854046.2020.1722244

Hendry, K., Green, C., McShane, R., Noel-Storr, A. H., Stott, D. J., Anwer, S., Sutton, A. J., Burton, J. K., & Quinn, T. J. (2019). AD-8 for detection of dementia across a variety of healthcare settings. Cochrane Database of Systematic Reviews, 2019(3), Article CD011121. https://doi.org/10.1002/14651858.CD011121.pub2

Henry, J. D., MacLeod, M. S., Phillips, L. H., & Crawford, J. R. (2004). A meta-analytic review of prospective memory and aging. Psychology and Aging, 19(1), 27–39. https://doi.org/10.1037/0882-7974.19.1.27

Inouye, S. K., van Dyck, C. H., Alessi, C. A., Balkin, S., Siegal, A. P., & Horwitz, R. I. (1990). Clarifying confusion: The Confusion Assessment Method. A new method for detection of delirium. Annals of Internal Medicine, 113(12), 941–948. https://doi.org/10.7326/0003-4819-113-12-941

Ismail, Z., Agüera-Ortiz, L., Brodaty, H., Cieslak, A., Cummings, J., Fischer, C. E., Gauthier, S., Geda, Y. E., Herrmann, N., Kanji, J., Lanctôt, K. L., Miller, D. S., Mortby, M. E., Onyike, C. U., Rosenberg, P. B., Smith, E. E., Smith, G. S., Sultzer, D. L., Lyketsos, C., & NPS Professional Interest Area of the International Society to Advance Alzheimer’s Research and Treatment. (2017). The Mild Behavioral Impairment Checklist (MBI-C): A rating scale for neuropsychiatric symptoms in pre-dementia populations. Journal of Alzheimer’s Disease, 56(3), 929–938. https://doi.org/10.3233/JAD-160979

Ismail, Z., Ghahremani, M., Munir, M. A., Fischer, C. E., Smith, E. E., & Creese, B. (2023). A longitudinal study of late-life psychosis and incident dementia and the potential effects of race and cognition. Nature Mental Health, 1(4), 273–283. https://doi.org/10.1038/s44220-023-00043-x

Ismail, Z., Smith, E. E., Geda, Y., Sultzer, D., Brodaty, H., Smith, G., Agüera-Ortiz, L., Sweet, R., Miller, D., Lyketsos, C. G., & ISTAART Neuropsychiatric Symptoms Professional Interest Area. (2016). Neuropsychiatric symptoms as early manifestations of emergent dementia: Provisional diagnostic criteria for mild behavioral impairment. Alzheimer’s & Dementia, 12(2), 195–202. https://doi.org/10.1016/j.jalz.2015.05.017

Kaufer, D. I., Cummings, J. L., Ketchel, P., Smith, V., MacMillan, A., Shelley, T., Lopez, O. L., & DeKosky, S. T. (2000). Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. The Journal of Neuropsychiatry and Clinical Neurosciences, 12(2), 233–239. https://doi.org/10.1176/jnp.12.2.233

Kendler, K. S., Glazer, W. M., & Morgenstern, H. (1983). Dimensions of delusional experience. American Journal of Psychiatry, 140(4), 466–469. https://doi.org/10.1176/ajp.140.4.466

Kuiper, J. S., Zuidersma, M., Oude Voshaar, R. C., Zuidema, S. U., van den Heuvel, E. R., Stolk, R. P., & Smidt, N. (2015). Social relationships and risk of dementia: A systematic review and meta-analysis of longitudinal cohort studies. Ageing Research Reviews, 22, 39–57. https://doi.org/10.1016/j.arr.2015.04.006

Lamb, S. E., Sheehan, B., Atherton, N., Nichols, V., Collins, H., Mistry, D., Dosanjh, S., Slowther, A. M., Khan, I., Petrou, S., & Lall, R. (2018). Dementia And Physical Activity (DAPA) trial of moderate to high intensity exercise training for people with dementia: Randomised controlled trial. BMJ, 361, Article k1675. https://doi.org/10.1136/bmj.k1675

Lanctôt, K. L., Ismail, Z., Bawa, K. K., Cummings, J. L., Husain, M., Mortby, M. E., & Robert, P. (2023). Distinguishing apathy from depression: A review differentiating the behavioral, neuroanatomic, and treatment-related aspects of apathy from depression in neurocognitive disorders. International Journal of Geriatric Psychiatry, 38(2), Article e5882. https://doi.org/10.1002/gps.5882

Lara, E., Martín-María, N., De la Torre-Luque, A., Koyanagi, A., Vancampfort, D., Izquierdo, A., & Miret, M. (2019). Does loneliness contribute to mild cognitive impairment and dementia? A systematic review and meta-analysis of longitudinal studies. Ageing Research Reviews, 52, 7–16. https://doi.org/10.1016/j.arr.2019.03.002

Leng, Y., McEvoy, C. T., Allen, I. E., & Yaffe, K. (2017). Association of sleep-disordered breathing with cognitive function and risk of cognitive impairment: A systematic review and meta-analysis. JAMA Neurology, 74(10), 1237–1245. https://doi.org/10.1001/jamaneurol.2017.2180

Levy, R., & Dubois, B. (2006). Apathy and the functional anatomy of the prefrontal cortex–basal ganglia circuits. Cerebral Cortex, 16(7), 916–928. https://doi.org/10.1093/cercor/bhj043

Linszen, M. M. J., Brouwer, R. M., Heringa, S. M., & Sommer, I. E. (2016). Increased risk of psychosis in patients with hearing impairment: Review and meta-analyses. Neuroscience & Biobehavioral Reviews, 62, 1–20. https://doi.org/10.1016/j.neubiorev.2015.12.012

Livingston, G., Huntley, J., Liu, K. Y., Costafreda, S. G., Selbæk, G., Alladi, S., Ames, D., Banerjee, S., Burns, A., Brayne, C., Fox, N. C., Ferri, C. P., Gitlin, L. N., Howard, R., Kales, H. C., Kivimäki, M., Larson, E. B., Nakasujja, N., Rockwood, K., . . . Mukadam, N. (2024). Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. The Lancet, 404(10452), 572–628. https://doi.org/10.1016/S0140-6736(24)01296-0

Lubben, J., Blozik, E., Gillmann, G., Iliffe, S., von Renteln Kruse, W., Beck, J. C., & Stuck, A. E. (2006). Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. The Gerontologist, 46(4), 503–513. https://doi.org/10.1093/geront/46.4.503

Luchetti, M., Aschwanden, D., Sesker, A. A., Zhu, X., O’Súilleabháin, P. S., Stephan, Y., Terracciano, A., & Sutin, A. R. (2024). A meta-analysis of loneliness and risk of dementia using longitudinal data from >600,000 individuals. Nature Mental Health, 2(11), 1350–1361. https://doi.org/10.1038/s44220-024-00328-9

Manly, J. J., Jacobs, D. M., Touradji, P., Small, S. A., & Stern, Y. (2002). Reading level attenuates differences in neuropsychological test performance between African American and White elders. Journal of the International Neuropsychological Society, 8(3), 341–348. https://doi.org/10.1017/S1355617702813157

Mann, F., Wang, J., Pearce, E., Ma, R., Schlief, M., Lloyd-Evans, B., Ikhtabi, S., & Johnson, S. (2022). Loneliness and the onset of new mental health problems in the general population. Social Psychiatry and Psychiatric Epidemiology, 57(11), 2161–2178. https://doi.org/10.1007/s00127-022-02261-7

Marin, R. S., Biedrzycki, R. C., & Firinciogullari, S. (1991). Reliability and validity of the Apathy Evaluation Scale. Psychiatry Research, 38(2), 143–162. https://doi.org/10.1016/0165-1781(91)90040-V

Mayo, A. M. (2016). Use of the Functional Activities Questionnaire in older adults with dementia (Try This: Best Practices in Nursing Care to Older Adults with Dementia, Issue D13). Hartford Institute for Geriatric Nursing, NYU Rory Meyers College of Nursing. https://alz.org/getmedia/73ea01f0-690b-4ff3-b909-4f0fa9a022ef/functional-activities-questionnaire.pdf

Miller, D. S., Robert, P., Ereshefsky, L., Adler, L., Bateman, D., Cummings, J., DeKosky, S. T., Fischer, C. E., Husain, M., Ismail, Z., Jaeger, J., Lerner, A. J., Li, A., Lyketsos, C. G., Manera, V., Mintzer, J., Moebius, H. J., Mortby, M., Meulien, D., . . . Lanctôt, K. L. (2021). Diagnostic criteria for apathy in neurocognitive disorders. Alzheimer’s & Dementia, 17(12), 1892–1904. https://doi.org/10.1002/alz.12358

Min, D., Ðoàn, L. N., Chong, S. K., Kaki, A., Kim, H., & Kwon, S. C. (2025). Social isolation and loneliness in Asian American older adults: A scoping review. The Gerontologist, 65(7), Article gnaf139. https://doi.org/10.1093/geront/gnaf139

National Academies of Sciences, Engineering, and Medicine. (2020). Social isolation and loneliness in older adults: Opportunities for the health care system. The National Academies Press. https://doi.org/10.17226/25663

Nielsen, T. R., Franzen, S., Watermeyer, T., Jiang, J., Calia, C., Kjærgaard, D., Bothe, S., & Mukadam, N. (2024). Interpreter-mediated neuropsychological assessment: Clinical considerations and recommendations from the European Consortium on Cross-Cultural Neuropsychology (ECCroN). The Clinical Neuropsychologist, 38(8), 1775–1805. https://doi.org/10.1080/13854046.2024.2335113

Orgeta, V., Brede, J., & Livingston, G. (2017). Behavioural activation for depression in older people: Systematic review and meta-analysis. British Journal of Psychiatry, 211(5), 274–279. https://doi.org/10.1192/bjp.bp.117.205021

Pfeffer, R. I., Kurosaki, T. T., Harrah, C. H., Jr., Chance, J. M., & Filos, S. (1982). Measurement of functional activities in older adults in the community. Journal of Gerontology, 37(3), 323–329. https://doi.org/10.1093/geronj/37.3.323

Pocklington, C., Gilbody, S., Manea, L., & McMillan, D. (2016). The diagnostic accuracy of brief versions of the Geriatric Depression Scale: A systematic review and meta-analysis. International Journal of Geriatric Psychiatry, 31(8), 837–857. https://doi.org/10.1002/gps.4407

Prigerson, H. G., Boelen, P. A., Xu, J., Smith, K. V., & Maciejewski, P. K. (2021). Validation of the new DSM-5-TR criteria for prolonged grief disorder and the PG-13-Revised (PG-13-R) scale. World Psychiatry, 20(1), 96–106. https://doi.org/10.1002/wps.20823

Robert, P., Lanctôt, K. L., Agüera-Ortiz, L., Aalten, P., Brémond, F., Defrancesco, M., Hanon, C., David, R., Dubois, B., Dujardin, K., Husain, M., König, A., Lévy, R., Mantua, V., Meulien, D., Miller, D., Moebius, H. J., Rasmussen, J., Robert, G., . . . Manera, V. (2018). Is it time to revise the diagnostic criteria for apathy in brain disorders? The 2018 international consensus group. European Psychiatry, 54, 71–76. https://doi.org/10.1016/j.eurpsy.2018.07.008

Shekelle, P. G., Miake-Lye, I. M., Begashaw, M. M., Booth, M. S., Myers, B., Lowery, N., & Shrank, W. H. (2024). Interventions to reduce loneliness in community-living older adults: A systematic review and meta-analysis. Journal of General Internal Medicine, 39(6), 1015–1028. https://doi.org/10.1007/s11606-023-08517-5

Shen, C., Rolls, E. T., Cheng, W., Kang, J., Dong, G., Xie, C., Zhao, X.-M., Sahakian, B. J., & Feng, J. (2022). Associations of social isolation and loneliness with later dementia. Neurology, 99(2), e164–e175. https://doi.org/10.1212/WNL.0000000000200583

Singh-Manoux, A., Dugravot, A., Fournier, A., Abell, J., Ebmeier, K., Kivimäki, M., & Sabia, S. (2017). Trajectories of depressive symptoms before diagnosis of dementia: A 28-year follow-up study. JAMA Psychiatry, 74(7), 712–718. https://doi.org/10.1001/jamapsychiatry.2017.0660

Stafford, J., Dykxhoorn, J., Sommerlad, A., Dalman, C., Kirkbride, J. B., & Howard, R. (2023). Association between risk of dementia and very late-onset schizophrenia-like psychosis: A Swedish population-based cohort study. Psychological Medicine, 53(3), 750–758. https://doi.org/10.1017/S0033291721002099

Sutin, A. R., Stephan, Y., Luchetti, M., & Terracciano, A. (2020). Loneliness and risk of dementia. The Journals of Gerontology: Series B, 75(7), 1414–1422. https://doi.org/10.1093/geronb/gby112

Swartz, R. H., Longman, R. S., Lindsay, M. P., Lund, R., Ganesh, A., Eskes, G. A., Austin, M., Bechard, L. E., Bhangu, J., Bruto, V. C., Carter, S., Chow, N., Deschaintre, Y., Fedorchuk, K., Fellows, L., Foley, N., Greer, L.-A., Lee, D. S., Leonard, C., . . . Canadian Stroke Best Practice Recommendations Advisory Committee, in collaboration with the Canadian Stroke Consortium, Canadian Neurological Sciences Federation, and CanStroke Recovery Trials Platform. (2025). Canadian Stroke Best Practice Recommendations: Vascular cognitive impairment, 7th edition practice guidelines update, 2024. Alzheimer’s & Dementia, 21(1), Article e14324. https://doi.org/10.1002/alz.14324

Sweet, J. J., Heilbronner, R. L., Morgan, J. E., Larrabee, G. J., Rohling, M. L., Boone, K. B., Kirkwood, M. W., Schroeder, R. W., Suhr, J. A., & Conference Participants. (2021). American Academy of Clinical Neuropsychology (AACN) 2021 consensus statement on validity assessment: Update of the 2009 AACN consensus conference statement on neuropsychological assessment of effort, response bias, and malingering. The Clinical Neuropsychologist, 35(6), 1053–1106. https://doi.org/10.1080/13854046.2021.1896036

Teng, E., Becker, B. W., Woo, E., Knopman, D. S., Cummings, J. L., & Lu, P. H. (2010). Utility of the Functional Activities Questionnaire for distinguishing mild cognitive impairment from very mild Alzheimer disease. Alzheimer Disease & Associated Disorders, 24(4), 348–353. https://doi.org/10.1097/WAD.0b013e3181e2fc84

Triebel, K. L., Martin, R., Griffith, H. R., Marceaux, J., Okonkwo, O. C., Harrell, L., Clark, D., Brockington, J., Bartolucci, A., & Marson, D. C. (2009). Declining financial capacity in mild cognitive impairment: A 1-year longitudinal study. Neurology, 73(12), 928–934. https://doi.org/10.1212/WNL.0b013e3181b87971

van Baarsen, B., Snijders, T. A. B., Smit, J. H., & van Duijn, M. A. J. (2001). Lonely but not alone: Emotional isolation and social isolation as two distinct dimensions of loneliness in older people. Educational and Psychological Measurement, 61(1), 119–135. https://doi.org/10.1177/00131640121971103

van Dalen, J. W., van Wanrooij, L. L., Moll van Charante, E. P., Brayne, C., van Gool, W. A., & Richard, E. (2018). Association of apathy with risk of incident dementia: A systematic review and meta-analysis. JAMA Psychiatry, 75(10), 1012–1021. https://doi.org/10.1001/jamapsychiatry.2018.1877

Van Orden, K. A., Witte, T. K., Cukrowicz, K. C., Braithwaite, S. R., Selby, E. A., & Joiner, T. E. (2010). The interpersonal theory of suicide. Psychological Review, 117(2), 575–600. https://doi.org/10.1037/a0018697

Wong, A., Law, L. S. N., Liu, W., Wang, Z., Lo, E. S. K., Lau, A., Wong, L. K. S., & Mok, V. C. T. (2015). Montreal Cognitive Assessment: One cutoff never fits all. Stroke, 46(12), 3547–3550. https://doi.org/10.1161/STROKEAHA.115.011226

Yon, Y., Mikton, C. R., Gassoumis, Z. D., & Wilber, K. H. (2017). Elder abuse prevalence in community settings: A systematic review and meta-analysis. The Lancet Global Health, 5(2), e147–e156. https://doi.org/10.1016/S2214-109X(17)30006-2

Yuen, G. S., Bhutani, S., Lucas, B. J., Gunning, F. M., AbdelMalak, B., Seirup, J. K., Klimstra, S. A., & Alexopoulos, G. S. (2015). Apathy in late-life depression: Common, persistent, and disabling. American Journal of Geriatric Psychiatry, 23(5), 488–494. https://doi.org/10.1016/j.jagp.2014.06.005

You Might Also Like