Mild cognitive impairment sits in the shadow zone between normal aging and dementia. It involves objective decline in memory, attention, language or executive function that exceeds what would be expected for a person’s age, but that does not meet the criteria for dementia and does not substantially interfere with daily life. Because people with the condition face an elevated risk of progressing to dementia, identifying it early opens a window for clinical assessment, counselling, management of modifiable risk factors and structured follow-up before decline becomes functionally limiting. Globally, an estimated 15.6 percent of community-dwelling adults aged 50 and older are affected, rising to 19.7 percent when nursing home and hospital populations are included; among European adults aged 65 and over, prevalence has been estimated at 23.9 percent. Romania’s 28.5 percent figure places it at the upper end of reported values, in a country where the older population continues to grow.

Despite this burden, mild cognitive impairment remains frequently under-recognized in primary care, and not only in Romania. Its symptoms can be subtle, slow to progress, or dismissed by patients and families as ordinary aging. Without an organised programme for systematic early detection, many cases go unnoticed until the decline becomes visible or disabling. Time pressure, limited training and the absence of clear guidance compound the problem. General practitioners, who in Romania as in much of Europe are often the first and most regular point of contact for older adults, are well placed to notice gradual cognitive and functional changes over the course of long-term relationships with their patients. Whether that potential translates into actual detection, however, depends on their knowledge, their attitudes, their familiarity with brief assessment tools, and the practical conditions of their working day. International studies in China, India and several European countries have already shown that physicians may endorse early cognitive assessment in principle while reporting limited training, diagnostic uncertainty and inconsistent screening.

The new study measured these dimensions using a questionnaire adapted from a published knowledge, attitudes and practices instrument and tailored to the Romanian context with input from practicing family physicians. Knowledge was tested with eight items covering prevalence, risk factors, diagnostic criteria, referral pathways, screening tools, interventions, prognosis and conversion to dementia. Attitudes were probed with eleven Likert-scale items, and practices with eleven items spanning symptom detection, risk-factor assessment, screening use, referral and communication with patients and families. Completion varied across sections, with 416 physicians finishing the knowledge scale, 381 the attitude scale and 365 the practice scale, so the researchers reported denominators separately for each analysis. The sample mirrored the Romanian workforce in broad strokes: 82.8 percent of respondents were women, 70.4 percent were over 50 years old, 76.8 percent had more than 15 years of experience, and 61.5 percent practiced in urban areas.

The knowledge results were sobering. The mean knowledge score was just 33.20 out of 100, with a standard deviation of 16.37, and not a single respondent achieved a perfect score. Only two respondents answered seven items correctly, while twelve answered none at all. Just 6.5 percent correctly identified what is required to diagnose mild cognitive impairment according to the Petersen diagnostic criteria, 10.8 percent recognized the most widely used cognitive screening instrument, and only 20.7 percent could identify modifiable risk factors that affect cognitive function; knowledge of treatment was similarly weak, at 9.9 percent correct. Knowledge was strongest where clinicians might reasonably be expected to be confident: 84.9 percent correctly identified which doctors are needed for diagnosis, and 66.8 percent knew effective non-pharmacological interventions. Nearly half of respondents, 47.1 percent, fell into the lowest knowledge band, answering between 11 and 25 percent of items correctly, and no respondent reached the highest category.

Attitudes, by contrast, were strikingly positive. The mean attitude score was 73.26 on a scale where 60 represents a neutral response across all items. A remarkable 97.9 percent of general practitioners agreed that they should be involved in detecting, diagnosing and managing mild cognitive impairment, 91.4 percent believed that early recognition and management can delay progression to Alzheimer’s disease, and 97.4 percent said that all patients suspected of having the condition should undergo diagnostic evaluation. Some 82.4 percent saw more advantages than disadvantages in finding out whether a patient has the condition, 65.9 percent endorsed pharmacological treatment and 70.1 percent endorsed non-pharmacological approaches. Yet uncertainty lingered beneath the surface: 27.5 percent still considered mild cognitive impairment a normal part of aging rather than a disease, 66.9 percent worried that disclosing a diagnosis could cause stress and frustration for patients and their families, and 81.1 percent saw dementia as a drain on medical and social resources.

Reported practice fell between these poles, with a mean score of 73.01 but wide variation in individual behaviors. Only 29.0 percent of respondents said they always or usually used screening methods during patient visits, and fewer than half considered psychiatric symptoms as detection criteria, at 30.4 percent, or linked frailty to the condition, at 42.8 percent. More than half, 57 percent, relied on memory disorder as the detection criterion. The pattern suggests greater preparedness for reactive management than for proactive case identification. More than 60 percent of physicians reported routinely discussing a probable diagnosis with patients, at 61.4 percent, and with families, at 66 percent, asking about family history of Alzheimer’s disease, at 65.4 percent, consulting specialists for final diagnosis, at 66.3 percent, and providing non-pharmacological interventions, at 69.1 percent. In other words, once cognitive impairment is already suspected, Romanian family doctors generally know what to do next; the weak link is the systematic step of actively looking for it before concerns become obvious.

The statistical architecture of the study revealed why this gap persists. In multiple linear regression models estimated on 364 complete cases, higher practice scores were independently associated with more favorable attitudes, greater knowledge, previous attendance at condition-specific training and higher daily patient volume, while larger registered patient lists were associated with lower practice scores. Only 10.8 percent of respondents had ever received training specifically on mild cognitive impairment, even though 78.8 percent considered such training useful for family physicians. Intriguingly, training was not independently associated with knowledge after adjustment, suggesting that current educational offerings may encourage engagement without consolidating technical understanding. Even more unexpectedly, trained physicians reported slightly less favorable attitudes, an association that survived adjustment and that the authors interpret cautiously, noting that self-selection into training or greater awareness of the complexities of assessment could explain it, though the cross-sectional design cannot distinguish between these possibilities. In the knowledge model, higher educational attainment and higher practice scores were the significant predictors, while in the attitudes model, larger numbers of patients diagnosed with the condition were associated with more favorable views.

The workload findings add a subtle but important dimension. Higher daily patient volume was associated with better reported practice, plausibly because greater clinical exposure creates more opportunities to encounter cognitive concerns, while larger registered lists were associated with worse practice, plausibly because responsibility for a bigger population brings cumulative administrative and continuity-of-care burdens that crowd out time for structured assessment. The authors stress that these explanations remain hypothesis-generating. Notably, the three regression models explained only a modest share of variance, with adjusted R-squared values of 0.037 for knowledge, 0.074 for attitudes and 0.080 for practices, indicating that much of what shapes physicians’ behavior lies outside the variables measured, potentially including consultation duration, access to specialist assessment and reimbursement structures. Bivariate analyses found no statistically significant differences in any domain by gender, age group, years of experience, professional title or rural versus urban location, and the correlations between knowledge, attitudes and practice were positive but weak.

The authors frame their results as evidence of a public health gap between population need and health-system preparedness. Applying available prevalence estimates to Romania’s older population suggests that roughly one million older adults may be living with mild cognitive impairment, in a country with no organised national programme for systematic early detection. Closing the gap, they argue, will require more than awareness campaigns: targeted continuing education that goes beyond facts to practical assessment skills, clear and feasible guidance on whom to assess and with which validated tools, and organizational support that gives physicians the time and referral pathways to act on what they find. The study has limitations that the authors acknowledge candidly, including its cross-sectional design, self-reported measures vulnerable to social desirability bias, a convenience sample recruited through open online channels that precluded calculating a response rate, and a questionnaire not yet formally psychometrically validated in Romanian. The findings are associations, not causal effects, and the practice score blends reported behavior, intention and perceived importance. Even so, the core message resonates well beyond Romania’s borders: in aging health systems everywhere, the willingness to detect cognitive decline early is abundant, but the knowledge, tools and organizational capacity to do it consistently remain the missing ingredients.

Subject of Research: Health-system preparedness for early detection of mild cognitive impairment: primary-care evidence from a national study among general practitioners in Romania

Article Title: Health-system preparedness for early detection of mild cognitive impairment: primary-care evidence from a national study among general practitioners in Romania

Article References: Paraschiv, A.-C., Spiru, L., Stamate, A., & Mănuc, D. (2026). Health-system preparedness for early detection of mild cognitive impairment: primary-care evidence from a national study among general practitioners in Romania. Frontiers in Public Health, 14, Article 1969120. https://doi.org/10.3389/fpubh.2026.1969120

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1969120

Keywords: Health-system, preparedness, early, detection, mild, cognitive, impairment, primary-care, evidence, national, general, practitioners