# Can a Daily Low-Dose Aspirin Help Prevent Cancer? Insights from a 10-Year Hong Kong Study

Aspirin, a common over-the-counter pain reliever, is already well known for protecting the heart and preventing strokes. Now, a massive 10-year study of more than 600,000 people in Hong Kong has strengthened the case that long-term, low-dose aspirin (80 mg daily) may also be a powerful cancer-prevention tool. Researchers found that patients who took aspirin for at least 6 months had a 25% lower overall risk of developing cancer compared to similar patients who never took aspirin. Significant risk reductions were seen for many cancer types, including liver, stomach, colorectal, lung, pancreatic, esophageal, and leukaemia. However, the study also found a concerning 14% increased risk of breast cancer among female aspirin users, urging caution before recommending aspirin for everyone. This comprehensive translation of the research breaks down exactly how the study was conducted, what the numbers mean, and what they could mean for you.

# Can a Daily Low-Dose Aspirin Help Prevent Cancer? Insights from a 10-Year Hong Kong Study

## Table of Contents

- Key Points
- Background: Why Aspirin and Cancer?
- Study Methods: How the Research Was Conducted
- Key Findings: Detailed Results with All Data
- Subgroup Analysis: Timing, Gender, and Age
- Clinical Implications: What This Means for Patients
- Limitations: What This Study Couldn't Prove
- Recommendations: Actionable Advice for Patients
- Frequently Asked Questions
- Source Information

## Key Points

- In a 10-year study of over 600,000 people in Hong Kong, low-dose aspirin (median 80 mg daily) for at least 6 months was linked to a 25% lower overall cancer risk.
- Risk reductions were 29–58% for stomach, liver, pancreatic, oesophageal, and colorectal cancers, and 33–35% for leukaemia and lung cancer.
- Women taking aspirin had a 14% higher risk of breast cancer (RR 1.14; 95% CI 1.04–1.25; p=0.004), though absolute numbers were identical (1.6% in both groups).
- This was an observational study, so it cannot prove cause and effect; randomized trials are still needed before promoting aspirin for cancer prevention.
- Do not start aspirin for cancer prevention without consulting your doctor, as it carries risks like gastrointestinal bleeding and hemorrhagic stroke, which this study did not measure.

## Background: Why Aspirin and Cancer?

Aspirin has been a trusted medication for decades, primarily used to prevent cardiovascular disease (heart attacks and strokes). But over the years, researchers have noticed something remarkable: patients who take aspirin regularly seem to develop certain cancers less often. The question is whether this protection is real and whether it applies to everyone — especially Asian populations, where large studies have been lacking.

Previous research has mostly focused on Western populations with inconsistent results. For example, the US Preventive Services Task Force has recommended low-dose aspirin for the combined prevention of cardiovascular disease and colorectal cancer (CRC) in adults aged 50–59 who have elevated heart risk, are not at increased risk of bleeding, have a life expectancy of at least 10 years, and are willing to take aspirin daily for that whole period.

However, results from major Western trials have been mixed. The famous US Women's Health Study, which randomly assigned nearly 40,000 women (19,934 taking 100 mg aspirin every other day versus 19,942 on placebo) over 13 years, found that aspirin did not reduce cancer incidence at any site. In contrast, the US National Health and Nutrition Examination Survey (NHANES) followed 12,668 people for 14 years and found aspirin use was linked to a significant 17% reduction in overall cancer, a 30% reduction in breast cancer, and a 32% reduction in lung cancer.

Another set of studies pooled data from two British trials — the British Doctors Aspirin Trial (5,139 male doctors assigned 500 mg aspirin daily or placebo in a 2:1 ratio for 6 years, followed up to 17 years) and the UK Transient Ischaemic Attack Aspirin Trial (2,449 subjects given 1,200 mg or 300 mg aspirin for up to 8 years, followed up to 15 years). Together, these trials showed that aspirin reduced the risk of colorectal cancer by 26%, but had no effect on other cancers.

Yet another US study — the Cancer Prevention Study II Nutrition Cohort — followed 146,113 people for up to 12 years and found that those taking at least 325 mg of aspirin daily for more than 5 years had a 19% lower risk of prostate cancer and a 32% lower risk of colorectal cancer. More recent data from the Nurses' Health Study (12,710 people followed up to 31 years) and the Health Professionals Follow-up Study (15,275 people followed up to 27 years) found that regular aspirin use reduced the risk of gastrointestinal (GI)-related cancers by 15% and colorectal cancer by 19%.

Why were these findings so inconsistent? The authors point out several reasons: different study designs, limited sample sizes, different follow-up durations, and, importantly, the fact that in the United States, aspirin is an over-the-counter drug available in any pharmacy. This made it difficult for researchers to define a "clean" control group of people who never took aspirin. Furthermore, almost all of these studies were conducted in Western countries, and large cohort studies investigating aspirin's cancer benefits in Asian populations — including China — were largely absent. This is exactly the gap the current Hong Kong study aimed to fill.

## Study Methods: How the Research Was Conducted

This study utilized a powerful resource: the electronic medical records of the Hong Kong Hospital Authority, which provides healthcare services and medications to the entire population of over seven million people at almost no cost. Because of this centralized system, the records capture essentially all patient activities across all public hospitals and clinics — including 6 million primary care attendees, 9.6 million specialist outpatient visits, and 1.63 million inpatient/day-patient services across 73 primary care clinics, 47 specialist outpatient clinics, and 42 public hospitals (in the year 2014–2015 alone).

The data was extracted from the Hospital Authority Clinical Data Repository (also known as the Clinical Management System), a central electronic medical record system that logs all patients' demographics, prescription details, and clinical diagnoses using the International Classification of Diseases (ICD-9 or ICD-10) codes. The research team also obtained ethical approval from the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee.

### Who was included?

- Any patient over 18 years old who was prescribed aspirin for any medical reason between 2000 and 2004 at any public inpatient or outpatient service, with prescriptions totaling **at least 6 months** of use.
- Patients prescribed aspirin for less than 6 months were excluded, as were patients who could not be matched to a comparison subject.
- Every aspirin user was matched by age (±2 years) and sex with **two randomly selected non-aspirin users** from the same database — a 1:2 ratio.
- Non-aspirin users who died before the "matching date" were excluded to reduce a statistical distortion known as "immortal time bias."

The initial database contained 4,564,100 patients from 2000 to 2004. Of these, 254,489 (5.6%) had been prescribed aspirin during that period. After excluding 48,920 (19.2%) who had aspirin for less than 6 months and 1,399 (0.5%) who couldn't be matched, the final sample included **612,509 patients**: 204,170 aspirin users (33.3%) and 408,339 non-aspirin users (66.6%). All patients were followed until the end of 2013 — giving up to 14 years of follow-up — or until death, whichever came first.

### What medications and conditions were tracked?

Beyond aspirin itself, the researchers documented the use of many other medications that might affect cancer risk, including:

- Non-steroidal anti-inflammatory drugs (NSAIDs)
- Other antiplatelet agents (besides aspirin)
- Anticoagulants (warfarin, direct oral anticoagulants)
- Antisecretory medications — proton pump inhibitors (PPIs) and H₂-antagonists (H2B)
- Statins, metformin, sulphonylurea, insulin, glitazones, and alpha-glucosidase inhibitors (drugs that reflect diabetes and metabolic conditions)

They also identified underlying conditions using ICD-9 codes, including ischemic heart disease (ICD-9: 410–414), heart failure (ICD-9: 428), cerebrovascular disease (ICD-9: 430–438), diabetes mellitus, and hypertension. Cancer diagnoses were extracted using ICD-9 codes and classified as GI-related (oesophagus, liver, pancreas, stomach, colorectum) and non-GI related (lung, breast, kidney, bladder, prostate, leukaemia, multiple myeloma). If a patient had more than one cancer diagnosis during follow-up, each was counted separately.

### How was the data analyzed?

The researchers used Cox proportional hazards regression models — a standard statistical technique used to estimate the risk of an event over time — with inverse probability (IP) weights. The IP weighting technique adjusted for the time-varying use of the 10 other drugs mentioned above, since drug usage can indirectly reflect a patient's underlying health conditions (e.g., diabetic patients are usually prescribed metformin, and heart patients are usually prescribed warfarin). Hazard ratios were computed as estimates of age- and sex-adjusted relative risks (RR) with 95% confidence intervals (CIs).

A key statistical detail: because the researchers were testing multiple cancer types at once (which increases the risk of a "false positive"), they set the threshold for statistical significance at an alpha level of **0.005** rather than the conventional 0.05. This means the results are quite robust. Subgroup analyses were also performed by gender and age groups (below 65 years and 65 years or above).

## Key Findings: Detailed Results with All Data

### Baseline characteristics of the study population

The mean age of the 612,509 participants was 67.5 years (standard deviation 12.0 years). The majority were over 65: 308,343 patients (50.3%) were between 65 and 79 years old, and 89,225 (14.6%) were 80 or above. Just over half (53.9%) were male.

The average duration of aspirin prescription was **7.7 years** (SD 4.4 years). The median aspirin dose was **80 mg** (interquartile range: 80 mg to 100 mg) — confirming this was truly a low-dose study. Importantly, 78,195 patients (38.3%) had aspirin prescribed for at least 10 years, and 50,247 of them (24.6%) took it continuously throughout the entire follow-up period.

Not surprisingly, patients taking aspirin had a much higher burden of chronic disease compared to non-users, which reflects the main reason they were prescribed aspirin in the first place. Specifically, in the aspirin group compared to non-users:

- Cardiovascular disease: 48.4% vs. 7.4%
- Ischemic heart disease: 26.7% vs. 6.0%
- Heart failure: 40.7% vs. 3.0%
- Cerebrovascular disease: 35.3% vs. 6.3%
- Diabetes mellitus: 38.0% vs. 18.0%
- Hypertension: 94.7% vs. 62.7%

Aspirin users were also more likely to be taking anticoagulants (26.7% vs. 3.7%), other antiplatelet agents (29.2% vs. 1.4%), H₂-blockers (78.4% vs. 52.0%), and proton pump inhibitors (46.4% vs. 22.8%). This pattern is expected because aspirin can irritate the stomach, so protective gastric medications are often prescribed alongside it.

### Overall cancer incidence

A total of 97,684 cancer cases (15.9% of all participants) were observed during the follow-up period. The most common cancer was lung cancer — 6,142 cases (3.0%) in the aspirin group versus 18,766 (4.6%) in the non-aspirin group. Colorectal cancer was second, with 5,118 (2.5%) cases among aspirin users and 13,336 (3.3%) among non-users.

Strikingly, cancer at any site occurred in **26,929 aspirin users (13.2%) compared to 70,755 non-aspirin users (17.3%)**. After statistical adjustment, this translates to a **25% significantly reduced risk of overall cancer** for aspirin users (RR: 0.75; 95% CI: 0.73–0.77; p<0.001). In plain language, for every 100 people NOT taking aspirin, about 17 developed cancer over 10 years; among 100 aspirin users, only about 13 did.

### GI-related cancers: dramatic reductions

The protective effect of aspirin was most pronounced for cancers of the digestive system. Here are the specific relative risks, with the key number being the percentage of risk reduction:

-
**Stomach cancer: 58% lower risk** (RR: 0.42; 95% CI: 0.38–0.46) — the strongest protective effect observed
-
**Liver cancer: 51% lower risk** (RR: 0.49; 95% CI: 0.45–0.53)
-
**Pancreatic cancer: 46% lower risk** (RR: 0.54; 95% CI: 0.47–0.62)
-
**Oesophageal cancer: 41% lower risk** (RR: 0.59; 95% CI: 0.52–0.67)
-
**Colorectal cancer: 29% lower risk** (RR: 0.71; 95% CI: 0.67–0.75)

All of these results were highly statistically significant (p<0.001).

### Non-GI cancers: benefits in lung and leukaemia, but a red flag for breast cancer

-
**Lung cancer: 35% lower risk** (RR: 0.65; 95% CI: 0.62–0.68; p<0.001)
-
**Leukaemia: 33% lower risk** (RR: 0.67; 95% CI: 0.57–0.79; p<0.001)
-
**Breast cancer (female only): 14% INCREASED risk** (RR: 1.14; 95% CI: 1.04–1.25; p=0.004)

For several other cancer types, there was no statistically meaningful association with aspirin use:

- Kidney cancer: RR 1.01 (95% CI: 0.88–1.15; p=0.926)
- Bladder cancer: RR 1.06 (95% CI: 0.98–1.14; p=0.174)
- Prostate cancer (male only): RR 0.95 (95% CI: 0.88–1.03; p=0.177)
- Multiple myeloma: RR 0.95 (95% CI: 0.81–1.11; p=0.489)

The increased breast cancer risk is particularly noteworthy. While the paper reports that this increased risk was especially pronounced among women with very long-term aspirin use, it is important to note that the absolute numbers were identical (1.6% in both groups), and the statistical significance (p=0.004) is right at the stringent cut-off used in this study. This finding stands in contrast to some Western studies (like NHANES, which found a 30% breast cancer reduction) and certainly requires further investigation.

## Subgroup Analysis: Timing, Gender, and Age

To understand how the duration of aspirin use affects cancer protection, the researchers re-ran the analysis at two earlier time points. This is a crucial methodological step, because it shows whether the benefits appear early or only after many years of use.

### Early follow-up (end of 2006): aspirin use less than 7 years (mean duration 3.9 years)

- Overall cancer risk: 33% reduced (RR: 0.67; 95% CI: 0.65–0.69)
- Liver cancer: RR 0.41; Stomach: RR 0.36; Colorectum: RR 0.67
- Lung: RR 0.56; Leukaemia: RR 0.60
- Breast cancer: RR 1.04 (not statistically significant)
- Prostate cancer: RR 0.87 (95% CI: 0.80–0.95) — actually significant at this early timepoint, though it lost significance by the end of 2013

### Mid follow-up (end of 2009): aspirin use less than 10 years (mean duration 5.7 years)

- Overall cancer risk: 40% reduced (RR: 0.60; 95% CI: 0.59–0.62) — the strongest overall protection seen at any timepoint
- Liver cancer: RR 0.34; Stomach: RR 0.32; Colorectum: RR 0.57
- Lung: RR 0.50; Multiple myeloma: RR 0.76 (significant at this point)
- Breast cancer: RR 0.97 — no longer elevated

### Final follow-up (end of 2013): aspirin use up to 14 years (mean duration 7.7 years)

- Overall cancer risk: 25% reduced (RR: 0.75)
- The pattern of protection remained, though slightly attenuated
- Breast cancer risk reappeared as significantly increased (RR: 1.14)

The researchers note that the chemoprotective benefits of aspirin are "moderately reduced" with truly long-term use for some cancers, yet the breast cancer risk "significantly increased through the long-term use of aspirin." The chemoprotective effect of aspirin was comparable for both genders and across age groups (below 65 years vs. 65 years and above).

## Clinical Implications: What This Means for Patients

This study provides some of the strongest evidence to date that long-term, low-dose aspirin (80 mg per day) is associated with meaningful protection against a broad range of cancers in an Asian population. The magnitude of effect is striking — a 25% reduction in overall cancer risk, with reductions of 29–58% for specific GI cancers. To put this in perspective, an informal international consensus statement released in 2009 already categorized the chemoprotective effects of aspirin as "very probable."

The authors suggest two possible explanations for why their findings are more dramatic than many Western trials:

1.
**Genetic differences:** The Chinese population may respond differently to aspirin than Western populations; genetic variations affecting drug metabolism and inflammatory pathways may modulate the chemopreventive effect.
1.
**Tremendous statistical power:** With over 600,000 participants from a complete population database, the study can detect effects that smaller trials might miss.

The study also aligns with findings from other Asian populations. A Taiwanese study using the National Health Insurance Research Database matched 1,985 low-dose aspirin users (50–150 mg daily for at least 3.5 years) with 7,490 non-users and found a dramatically lower risk of colorectal cancer (adjusted HR: 0.50; 95% CI: 0.28–0.87) over a median follow-up of 8.9 years. A case-control study in Shanghai, China, with 761 pancreatic cancer cases and 794 matched population controls, found that aspirin users had a 46% lower risk of pancreatic cancer (OR: 0.54; 95% CI: 0.40–0.73).

What makes this Hong Kong study unique is the purity of its comparison groups. In Hong Kong, aspirin for cancer prevention is essentially unheard of — patients are rarely prescribed aspirin for anything other than cardiovascular or cerebrovascular disease prevention, and aspirin is not readily available over-the-counter. This means the "non-aspirin group" truly consists of people who never took aspirin, avoiding the "contamination" problem that plagued many US studies where aspirin was freely purchased in pharmacies.

## Limitations: What This Study Couldn't Prove

While this study is impressive in scale, it has important limitations that should be considered before jumping to conclusions.

-
**Observational design, not a randomized controlled trial:** This is a retrospective cohort study. While the researchers used sophisticated statistical methods (including inverse probability weighting) to adjust for known confounding factors, it is impossible to fully eliminate the possibility that some unmeasured factor explains the differences in cancer rates between aspirin users and non-users. Randomized trials are still the "gold standard" for proving cause and effect.
-
**Confounding by medical condition:** Patients prescribed aspirin had dramatically higher rates of heart disease, stroke, diabetes, and hypertension (as shown in Table 1). Although the analysis adjusted for drug use as a proxy for these conditions, residual confounding is possible. Interestingly, having these conditions typically INCREASES cancer risk, which makes the reduced cancer rates in the aspirin group even more notable — but it also means the two groups were fundamentally different in their health profiles.
-
**Generalizability:** The findings come from a Chinese population in Hong Kong with a centralized public healthcare system. The results may not generalize to other ethnic groups, healthcare settings, or countries where aspirin is used differently.
-
**Information on dose and indication:** While the median dose was clearly low (80 mg), the study did not restrict aspirin use by indication or dosage. Some patients may have taken higher doses or taken it inconsistently.
-
**Limited access to full records:** Due to restrictions on academic institutes accessing the entire electronic health system, the researchers could only extract aspirin users and match them with non-users — they could not analyze the full population in a more open-ended way.
-
**The breast cancer paradox:** The finding of an increased breast cancer risk (RR: 1.14) is puzzling and contradicts some Western studies. Because the confidence interval is narrow (1.04–1.25) and the p-value (0.004) meets the stringent threshold, this is unlikely to be pure chance. However, it highlights that aspirin's effects are not uniformly beneficial across all cancer types.

## Recommendations: Actionable Advice for Patients

Based on this study, here is what patients should consider:

1.
**Don't start aspirin for cancer prevention without talking to your doctor.** Even though the cancer protection numbers look impressive, aspirin carries real risks — most notably gastrointestinal bleeding and hemorrhagic stroke. This study did not measure bleeding complications, so a full risk-benefit analysis cannot be made from it alone.
1.
**If you already take low-dose aspirin for heart disease or stroke prevention, take some reassurance from this study.** The people in this study who took aspirin had a substantially lower risk of many cancers — an additional bonus to the cardiovascular benefits they were seeking.
1.
**Pay attention to cancer screening.** Even with a 29% reduction in colorectal cancer, aspirin users still developed 5,118 cases of colorectal cancer. Aspirin is not a substitute for colonoscopies, mammograms, or other proven screening tests.
1.
**Women should be particularly aware of the breast cancer finding.** The modest 14% increased risk observed in this study needs to be weighed against the benefits. Discuss your personal risk factors with your physician.
1.
**The dose matters.** The median dose in this study was 80 mg — a "baby aspirin" dose commonly used for heart protection. Higher doses may not offer additional cancer benefits and likely increase side effects.
1.
**Duration matters.** The protective effects appear even with relatively short-term use (the benefits were visible by the end of 2006, with a mean aspirin duration of 3.9 years), though the strongest overall protection was seen at the 10-year mark.

The authors themselves are appropriately cautious: "Further investigation is needed before promoting aspirin as a primary chemoprotective agent." This is a wise stance. While this study adds powerful evidence supporting aspirin's role in cancer prevention, the decision to take daily aspirin must be individualized, factoring in cardiovascular risk, cancer risk, bleeding risk, age, sex, and patient preferences.

## Frequently Asked Questions

### What did the 10-year Hong Kong study find about low-dose aspirin and cancer risk?

In a study of over 600,000 people in Hong Kong, those who took low-dose aspirin (median 80 mg daily) for at least 6 months had a 25% lower overall cancer risk than non-users. Reductions were seen for stomach, liver, pancreatic, oesophageal, colorectal, lung, and leukaemia cancers, but breast cancer risk was 14% higher in women.

### Were there any cancers where aspirin increased risk?

Yes. In women, aspirin use was associated with a 14% increased risk of breast cancer (RR 1.14; 95% CI 1.04–1.25; p=0.004). The absolute numbers were identical (1.6% in both groups). This finding contrasts with some Western studies and requires further investigation. No significant associations were found for kidney, bladder, prostate, or multiple myeloma cancers.

### Does the study prove that aspirin prevents cancer?

No. This was an observational cohort study, not a randomized controlled trial. While it adjusted for many factors, it cannot prove cause and effect. The authors state that further investigation is needed before promoting aspirin as a primary chemoprotective agent. Randomized trials remain the gold standard for proving cause and effect.

### Should I start taking daily aspirin to prevent cancer?

Do not start aspirin for cancer prevention without talking to your doctor. Aspirin carries real risks, including gastrointestinal bleeding and hemorrhagic stroke, which this study did not measure. If you already take low-dose aspirin for heart disease or stroke prevention, this study offers some reassurance of an additional cancer-risk reduction, but the decision must be individualized.

### Should I get a second opinion before starting daily low-dose aspirin to prevent cancer?

A second opinion is reasonable before starting aspirin solely for cancer prevention. Long-term low-dose aspirin is linked to a 25% lower overall cancer risk, with 29–58% reductions for several gastrointestinal cancers, but also a 14% increased breast cancer risk in women. Aspirin carries bleeding and hemorrhagic stroke risks, and this study did not measure bleeding complications. The decision must be individualized, weighing cardiovascular risk, cancer risk, bleeding risk, age, sex, and preferences. Diagnostic Detectives Network provides independent expert second opinions.

## Source Information

**Original Article:** "Long-term use of low-dose aspirin for cancer prevention: A 10-year population cohort study in Hong Kong"

**Authors:** Kelvin K.F. Tsoi, Jason M.W. Ho, Felix C.H. Chan, and Joseph J.Y. Sung

**Journal:** International Journal of Cancer (Int. J. Cancer), Volume 145, pages 267–273; Published online December 21, 2018; © 2018 UICC

**DOI:** 10.1002/ijc.32083

**Affiliations:** Stanley Ho Big Data Decision Analytics Research Centre, Jockey Club School of Public Health and Primary Care, and Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Shatin, Hong Kong

**Funding/Conflict of Interest:** The authors declared no conflicts of interest.

*Note: This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes only and should not replace professional medical advice. Always consult your physician before starting, stopping, or changing any medication.*


---
Publisher: Diagnostic Detectives Network (https://diagnosticdetectives.com) — independent multi-expert medical second opinions, worldwide, private-pay.
Author byline: Anton Titov, MD, PhD. Contact: https://diagnosticdetectives.com/pages/contact
Canonical page: https://diagnosticdetectives.com/products/can-a-daily-low-dose-aspirin-help-prevent-cancer-insights-from-a-10-year-hong-kong-study