Background
Unhealthy lifestyle behaviors are major contributors to the global cancer burden and generate substantial economic costs. Although lifestyle interventions are effective in reducing cancer risk, robust economic evaluations to support policy and resource-allocation decisions, particularly in middle-income settings, remain limited.
Objectives
To evaluate the cost-effectiveness of a mobile health (mHealth)-based intervention for primary cancer prevention among adults in China.
Methods
We conducted a quasi-experimental pre-post intervention study in Henan Province, China, starting in 2022. Participants completed a comprehensive risk assessment covering 19 cancer types and 17 lifestyle-related risk factor modules. Personalized risk feedback and behavior change recommendations were delivered through the Smart Health Management Digital Platform for Primary Cancer Prevention (SmartHMDP-PCP). Lifestyle changes were assessed after 6-12 months. Adherence to cancer prevention guidelines was quantified using modified World Cancer Research Fund (WCRF)/America Institute of Cancer Research (AICR) score incorporating smoking status, a validated predictor of lung, colorectal, and breast cancer risk.
Cost-effectiveness was assessed using sex-specific Markov models with five health states: healthy, lung cancer, colorectal cancer, breast cancer (women only), and death (Figure 1). A hypothetical control group assuming no lifestyle change was used for comparison. For each sex, 10,000 individuals were simulated from the mean baseline age to age 85. Intervention costs, including platform implementation subsidies, and participant and investigator time, were derived from real-world program data. To adopt a conservative assumption, intervention effects were assumed to persist for one year only, after which cancer risks reverted to baseline. Other model parameters (transition probabilities, cancer treatment costs, utilities, relative risks, discounting, and willingness-to-pay thresholds) were sourced from published literature. One-way deterministic and Monte Carlo probabilistic sensitivity analyses were conducted to assess the uncertainty of the incremental cost-effectiveness ratios (ICERs).
Results
By 2024, 36,344 participants completed baseline assessments, and 29,554 completed follow-up evaluations. The mean increase in WCRF/AICR score was 0.743. Markov simulations initiated at age 38 estimated gains of 3.77 quality-adjusted life years (QALYs) per 10,000 men and 12.19 QALYs per 10,000 women, compared with the hypothetical control group. For men, the intervention incurred an incremental cost of USD 44,700, corresponding to an ICER of USD 11,844.95 per QALY gained, below the willingness-to-pay threshold of USD 13,445, defined by China's 2024 GDP per capita. For women, the intervention was cost-saving, with an ICER of negative USD 3,035.27 per QALY. Sensitivity analyses identified effect duration, relative risk estimates linking WCRF/AICR scores to cancer outcomes, discount rates, cancer treatment costs, and implementation costs as key sources of uncertainty.
Conclusion
This study demonstrates that a scalable mHealth-based lifestyle intervention for primary cancer prevention is cost-effective, and potentially cost-saving, under conservative assumptions. By achieving measurable improvements in cancer-related behaviors at relatively low cost, such interventions offer a compelling strategy to strengthen primary cancer prevention and support value-based cancer control policies, particularly in resource-constrained and middle-income settings.
Markov Model Schematic