The research team, led by Raden M. Febriyanti of Universitas Padjadjaran, mined the 2023 Indonesian Health Survey, a nationally representative dataset collected by the Ministry of Health. From it, they isolated 2,703 adults aged fifteen and older who reported a physician diagnosis of diabetes mellitus, hypertension, or both, and who had complete records for the exposures, biomarkers, and covariates the analysis required. This was not a small convenience sample of herbal enthusiasts; it was a statistically weighted cross-section of the entire archipelago, designed so that the findings could be projected onto the national population of chronically ill adults.

The headline number is arresting. Among these diagnosed patients, 1,309 individuals reported using traditional health services during the previous twelve months, which translates to a weighted prevalence of 53 percent. In other words, more than one in every two Indonesian adults with diabetes or hypertension had sought out some form of traditional care within a single year. Herbal preparations dominated the menu, used by 76.2 percent of those who turned to traditional services, while manual therapies such as massage and related hands-on practices were used by 52.9 percent. The two categories overlap, meaning many users combined swallowing botanical remedies with submitting their bodies to manipulation.

Who was most likely to reach for these services? The sociodemographic patterning turned out to be one of the most interesting dimensions of the study. After adjusting for confounders, urban residents showed higher odds of traditional health service use than their rural counterparts, with an adjusted odds ratio of 1.28 and a 95 percent confidence interval of 1.05 to 1.56. That finding cuts against the common assumption that traditional medicine is primarily a rural phenomenon sustained by distance from clinics. Self-employed respondents were also significantly more likely to be users, with an adjusted odds ratio of 1.49, as were people working in agriculture or fisheries, whose adjusted odds ratio of 1.44 suggests that occupational culture and perhaps traditional knowledge transmission within these sectors play a role.

Methodologically, the study took considerable care to avoid the statistical traps that plague cross-sectional research. The analysts applied survey-weighted multinomial and binary logistic regression, explicitly incorporating stratification, clustering, and sampling weights to reflect the complex design of the national survey. Recognizing that testing multiple biomarkers invites false positives, they controlled the false discovery rate using the Benjamini-Hochberg procedure, a standard guard against mistaking statistical noise for signal. They then ran a propensity-score weighting analysis as a sensitivity check, attempting to balance the characteristics of users and non-users as if the exposure had been assigned at random. It is a rigorous architecture for an observational question, and it makes the null results that follow harder to dismiss.

And those null results are the study’s second headline. Across every outcome examined, HbA1c, total cholesterol, LDL cholesterol, HDL cholesterol, systolic blood pressure, and diastolic blood pressure, the researchers detected no association with traditional health service use. No modality-specific analysis, separating herbal users from manual therapy users, produced a signal that survived correction for multiple testing. Nor did stratifying by pharmacotherapy status, comparing patients who were taking conventional medications against those who were not, reveal any hidden relationship. For a practice used by half the chronic disease population, the laboratory values of its users look, on average, indistinguishable from those of non-users.

The authors are careful, and rightly so, about what this null finding does and does not mean. The survey captured whether people used traditional health services, but not why, not what they took, not in what dose, not how often, and not for how long. Information on indication, product, frequency, duration, and treatment timing was simply unavailable. A person who sips a mild ginger infusion once a month and a person who takes concentrated herbal antidiabetic preparations daily are lumped into the same exposure category. Without dose-response data, a genuine biological effect could be diluted into invisibility, or a harmful one could hide behind population averaging. The findings, the team emphasizes, describe population-level patterns rather than evidence of benefit or harm.

There is also the matter of temporality. A cross-sectional snapshot cannot tell you whether biomarker differences preceded or followed traditional service use, and the study’s own design rules out causal inference entirely. What it can do, and does, is establish that traditional care is not a fringe behavior among Indonesia’s chronically ill. It is mainstream, socially patterned, and concentrated in identifiable groups, which matters enormously for public health planning. If clinicians assume their diabetic and hypertensive patients are not using herbal products, they are wrong half the time, and that assumption carries real risks, from herb-drug interactions to delayed care-seeking, that this study cannot measure but its prevalence data make impossible to ignore.

The urban skew in the results deserves particular attention. Higher use in cities may reflect the commercialization of traditional medicine, with jamu products now packaged, branded, and sold through modern retail channels, alongside urban wellness culture that embraces complementary therapies. Meanwhile, the elevated odds among agricultural and fisheries workers hint at communities where traditional knowledge remains embedded in daily life and occupational networks. Self-employment, too, may grant the schedule flexibility that clinic visits do not. These patterns sketch a picture of traditional health services as an accessible, culturally familiar supplement to formal care rather than a substitute born of desperation or lack of access.

For a country where diabetes and hypertension are rising relentlessly and health system resources are stretched across seventeen thousand islands, the study offers a pragmatic message. Traditional health service use is too common to ignore and too clinically silent, at least at the biomarker level, to celebrate or condemn. The next generation of research, the authors’ caveats imply, needs to capture what people actually consume, in what quantities and for how long, before anyone can say whether Indonesia’s beloved jamu is helping, harming, or simply accompanying its millions of chronic disease patients on their long road of daily management. Until then, the safest reading is that half the country is experimenting, and the blood tests, for now, are keeping their secrets.

Subject of Research: Traditional health service use and cardiometabolic biomarker profiles among Indonesian adults with diabetes or hypertension

Article Title: Traditional health service use and cardiometabolic biomarker profiles among Indonesian adults with diabetes mellitus or hypertension: a national cross-sectional study

Article References: Febriyanti, R. M., Irawan, A. A., Pangestu, D. S., Alfian, S. D., Levita, J., & Diantini, A. (2026). Traditional health service use and cardiometabolic biomarker profiles among Indonesian adults with diabetes mellitus or hypertension: a national cross-sectional study. BMC Complementary Medicine and Therapies. https://doi.org/10.1186/s12906-026-05558-1

Image Credits: AI Generated

DOI: 10.1186/s12906-026-05558-1

Keywords: traditional medicine, complementary medicine, diabetes mellitus, hypertension, Indonesia, cardiometabolic biomarkers, HbA1c, cholesterol, blood pressure, herbal medicine, cross-sectional study, public health