PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 4, 2026BMC Health Services Research1 citationsOpen Access

Cost-utility analysis of cerebrospinal fluid versus blood biomarkers for early detection of Alzheimer’s disease and mild cognitive impairment in Thailand: a modeling study

PLPattara LeelahavarongSiriraj HospitalJPJuthamas PrawjaengSiriraj HospitalPAPakkanan AngkabSiriraj Hospital

Key Points

  • The aim is to evaluate the cost-utility of cerebrospinal fluid and blood-based biomarkers for early Alzheimer’s disease detection in high-risk older adults in Thailand.
  • Constructed a decision tree with Markov models
  • Used a societal perspective with a 1-year cycle over a lifetime horizon
  • Collected sensitivity and specificity data from Thai patients
  • Calculated lifetime costs, QALYs, and ICERs
  • CSF strategy yielded 0.011 additional life-years and 0.019 additional QALYs for an extra cost of THB2,589
  • ICER for CSF was THB132,961 per QALY; for blood, it was THB907,057 per QALY
  • CSF biomarker testing is cost-effective while blood-based testing is not
  • Reducing Simoa costs by 83% would enhance blood biomarker cost-effectiveness

Abstract

To evaluate the cost-utility of cerebrospinal fluid (CSF) and blood-based amyloid beta biomarkers for early detection of Alzheimer’s disease and mild cognitive impairment among older Thai adults at high risk of dementia. We constructed a decision tree with Markov models from a societal perspective, using a 1-year cycle over a lifetime horizon and applying a 3% annual discount. Amyloid beta (Aβ)1–42 was used for CSF, while Aβ40 and Aβ42 were used for blood. Sensitivity and specificity data were primarily derived from Thai patients at Siriraj Hospital, and nonpharmacological treatment efficacy was obtained from the FINGER study. Epidemiological data, transition probabilities, and costs were collected from the literature and Siriraj Hospital; direct nonmedical costs and utility values were also obtained from Siriraj. We calculated lifetime costs, quality-adjusted life years (QALYs), and incremental cost-effectiveness ratios (ICERs), evaluating cost-effectiveness against THB160,000 (USD4,370) per QALY. Parameter uncertainty was explored via sensitivity analyses. This is non-clinical trail study. Compared with no testing, the CSF strategy yielded 0.011 additional life-years and 0.019 additional QALYs for an extra cost of THB2,589 (USD71), resulting in an ICER of THB132,961 (USD3,632) per QALY. Blood-based testing provided 0.009 additional life-years and 0.017 QALYs for an extra cost of THB15,467 (USD422), leading to an ICER of THB907,057 (USD24,776) per QALY. CSF biomarker testing is cost-effective in Thailand, whereas blood-based biomarkers are not. Reducing the cost of Simoa (the technology used to measure blood biomarkers) by approximately 83% would improve the cost-effectiveness of blood-based biomarkers. Future research should enhance blood biomarker accuracy. This is the first study to assess the economic impact of using CSF or blood biomarkers for early Alzheimer’s disease detection in Thailand. The findings can inform coverage decisions under the Thai Universal Coverage Scheme (a government health insurance program). Sensitivity and specificity estimate for both biomarker tests were derived from Thai patients at Siriraj Hospital, and direct non-medical costs and utility values also came from this setting. Direct medical cost parameters were sourced from Thai literature, ensuring relevance to Thai healthcare. The reliance on noncountry-specific data for disease progression and mortality may limit generalizability. Further epidemiological studies in Thailand could refine these parameters and update the results. The limited sample size for blood biomarker measurements introduced variability in sensitivity and specificity values. Larger studies are needed to improve accuracy and reliability of these tests.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Leelahavarong et al. (2026) studied this question.

synapsesocial.com/papers/69d0adc2659487ece0fa4505https://doi.org/10.1186/s12913-026-14405-5
Ask AI
Helpful
Bookmark
Share
View Full Paper