Sixty-eight percent of the residents never speak to their families. Not rarely, not on holidays. Never.
That figure comes from a survey of 175 elderly people living in three old age homes in and around Dhaka, Bangladesh, conducted by Rubaiya Binte Kabir of Dhaka Medical College and Hospital with Md. Allama Faysal and Shamsun Nahar of State University of Bangladesh. The team interviewed residents face to face at Probin Nibash in Agargaon, Priyojon Nibash on Darus Salam Road, and the Old Rehabilitation Center in Gazipur. Alongside questions about their backgrounds, health and daily lives, each participant completed the DASS-21, a widely used 21-item questionnaire that scores depression, anxiety and stress separately.
The scores were bleak. Among all participants, 80.57 percent fell somewhere on the scale from mild to extremely severe depression. Sixty-eight percent registered anxiety, and 70.29 percent registered stress. These are screening results rather than clinical diagnoses, a distinction worth holding onto, but the proportions are high enough that the reverse question becomes the striking one: in these three homes, fewer than one resident in five scored clear of depressive symptoms.
Not the food, not the beds
What makes the picture unusual is what the residents said about the institutions themselves. They were, by and large, satisfied. Ninety-nine percent reported satisfaction with social activities inside the home. Nearly the same share, 98.86 percent, felt safe. Caregiver services drew 96.57 percent satisfaction, physical exercise 93.43 percent, recreation and the funeral arrangements 83.43 percent each, medical care 80 percent. Food scored lowest of the categories the team asked about, and even that reached 61.14 percent.
So the distress does not track the obvious things. It tracks contact. When the authors looked for what predicted higher scores, less communication with family came out significantly associated with all three: depression, anxiety and stress. Female residents scored significantly higher on depression and anxiety than male residents. And length of stay mattered in a specific way: residents who had been in a home for one to five years showed significantly elevated depression and stress. For depression and stress, the associations with communication and length of stay reached a p value below 0.001, meaning a result this strong would be unlikely to arise by chance alone; the gender and communication links to anxiety cleared the looser threshold of 0.05.
The residents' circumstances fill in the rest. The average age was 75.31 years, with a standard deviation of 7.72. Just over half, 52 percent, were women. More than nine in ten, 90.3 percent, were widowed. Eighty percent had one or two children. Fully 80.6 percent said there was nobody at home who could have taken care of them. These were not, on the whole, people who had chosen institutional life over a waiting family.
Physical health added its own weight. More than half, 58.3 percent, lived with a chronic illness. Among those, the authors recorded diabetes in 47.06 percent, hypertension in 16.67 percent and heart disease in 7.84 percent.
The residents were also, by Bangladeshi standards, an educated group. Graduates made up the largest single category at 38.3 percent. Half, 50.3 percent, had worked non-government jobs; 40 percent had been housewives.
Why it matters
Bangladesh's elderly population is growing, and the authors open with the reason this study exists at all: the cultural expectation that families care for their oldest members is loosening, and institutionalization is filling the gap. Old age homes are becoming a normal destination rather than a last resort.
The practical lesson here is narrow but useful. If distress in these homes were driven mainly by conditions inside them, the fix would be budgets: better food, better medicine, more staff. The residents' own satisfaction ratings suggest those conditions are largely acceptable to them. What was missing was the phone call, the visit, the sense of still belonging to someone outside the gates. That is not something an institution can procure.
The limits should temper any grand reading. The three homes were selected by purposive sampling, which means the researchers chose them deliberately rather than drawing a random national sample, so the numbers describe these institutions and not Bangladesh's old age homes as a group. The study is cross-sectional: it photographs a moment and cannot show whether isolation caused the depression, whether depressed residents withdrew from contact, or whether the same family rupture produced both. And it is a preprint, posted to medRxiv in July 2026 and not yet through peer review.
Still, the authors have documented something specific. They received ethics approval from the Bangladesh Medical Research Council, obtained written consent from every participant, and have made their dataset available. In a country where research on institutionalized elders is thin, 175 careful interviews are worth having on the record.