I ordered my first quarterly blood panel because I was writing my book and wanted to understand my own baseline before recommending anything to patients that I hadn’t applied to myself. That was straightforward enough. What I didn’t anticipate was how interesting the longitudinal data would become — not the absolute values, which were mostly within range, but the patterns. What moved, and when. What correlated with what.
I ran the full panel in January, April, July, and October. Each time, I also noted training load, sleep quality, significant life events, dietary changes, and anything clinically notable. By the end of the year, I had a dataset that was, by any measure, the most detailed picture of my own health I’d ever had.
The headline: several things I thought were stable were not. And one marker that I’d assumed was fine required intervention.
What I tested — and why
The panel I ran quarterly was deliberately broader than a standard screen. It included:
Full blood count, metabolic panel, and liver function. HbA1c and fasting insulin — to track glucose metabolism at the precision end rather than waiting for flag-level results. Full lipid panel including ApoB and Lp(a) — because total cholesterol and LDL alone are inadequate risk stratifiers, particularly in perimenopausal women. High-sensitivity CRP and IL-6 as inflammatory markers. Complete hormone panel: oestradiol, progesterone, testosterone, DHEA-S, SHBG, FSH, LH. Thyroid panel: TSH, Free T4, Free T3, and thyroid antibodies. Vitamin D, magnesium (RBC, not serum), zinc, B12, folate, ferritin. Omega-3 index.
This is not a small list. The cost was real — approximately SGD 600–900 per quarter. I’ve made peace with this as a professional expense and a personal commitment. Most people can achieve meaningful monitoring with a simpler panel; I was specifically interested in the detailed longitudinal picture.
What I found
I’ll start with what didn’t change: my HbA1c stayed stable throughout the year, my full blood count remained consistent, and my liver function was unremarkable. These were reassuring but uninteresting.
The more informative findings:
My hsCRP varied significantly across the four time points. In April — following a heavy training block for Hyrox and a period of inadequate sleep — it was 2.1 mg/L. This is above the low-risk threshold of 1mg/L and approaching the moderate-risk zone. In October, following a deliberate reduction in training intensity, improved sleep, and a month of increased omega-3 supplementation, it had fallen to 0.6 mg/L. This pattern confirmed for me personally what I tell patients clinically: CRP is not a static marker. It reflects lifestyle conditions in real time, and the conditions that elevate it are modifiable.
My vitamin D was low in both January and April despite living in Singapore with year-round sun. I was supplementing — 2000 IU daily — and it wasn’t enough. I increased to 5000 IU with vitamin K2 co-supplementation in May and by July my level had normalised. This is a common finding even in Singapore: darker skin phototypes need more sun exposure or higher supplementation doses to maintain adequate levels, and our air-conditioned indoor lifestyles mean many of us are getting less sun contact than we assume.
My testosterone was low-normal in January and declined further in April. This coincided with the high training load period — overtraining is associated with HPA axis disruption and reduced androgen output. I reduced training intensity, prioritised recovery, and by October my levels had recovered meaningfully. This validated a pattern I’d suspected clinically but hadn’t observed in myself directly.
The marker that required intervention: my Lp(a). Lipoprotein(a) is a genetically determined cardiovascular risk factor that doesn’t respond to statins or lifestyle and is rarely included in standard lipid panels. Mine was elevated — above 75 nmol/L, which is in the high-risk category [1]. It had been there all year, consistent across all four panels. This is genetic. It will not change with diet or exercise. It requires cardiovascular risk management that takes it into account — and I now have that conversation with my cardiologist rather than assuming my otherwise reasonable lipid profile means I’m at low cardiovascular risk.
What changed in my practice
Having data on myself changed how I communicate with patients. Not dramatically — I already ran regular panels on my patients in longevity consultations. But having lived through the experience of watching my own numbers move in response to specific changes gave me a more nuanced ability to explain what the data means in real time.
I now include quarterly blood monitoring as a specific recommendation for patients in the perimenopausal and early postmenopausal transition — not just the annual screen that standard primary care offers. The rate of change during this period is too fast for annual monitoring to capture meaningfully. Hormones shift quarter to quarter. CRP responds to life events. Glucose metabolism can deteriorate between annual screens in ways that are fully reversible if caught early.
In my practice at SW1 Clinic, the patients who engage with this level of monitoring consistently make better decisions than those who don’t — not because they’re more motivated, but because they have actual data to respond to rather than vague awareness that something should change.
The Asian context
Singapore has genuinely excellent laboratory access. Most major labs offer comprehensive panels that can be self-requested or ordered by any physician. The cost is accessible for most professional-income earners. There is no infrastructural barrier to this level of monitoring.
The barrier is awareness and cultural normalisation of proactive health monitoring. In many Asian families, blood tests are done when you’re sick, not to prevent sickness. The concept of monitoring for optimisation — rather than investigation — requires a shift in health culture that I think is beginning to happen, but slowly.
Particularly for women in their 40s, who are often the last to prioritise their own health, the simple act of having data shifts the conversation from vague concern to specific action. Data is permission to take yourself seriously.
What you can actually do
You don’t need to run the panel I ran to get meaningful information. A starting point that I’d recommend for any woman over 40: full blood count, metabolic panel, HbA1c, fasting insulin, full lipids including ApoB and Lp(a), hsCRP, full hormone panel, vitamin D, ferritin, thyroid function.
Run it at least annually. If you’re in perimenopause, twice yearly gives you considerably more useful data.
Track the trends, not just the individual data points. A result that has moved from one end of the normal range to the other over three years is informative even if each individual value technically sits within range.
And find a physician who can interpret the results with clinical nuance rather than simply flagging red and green. Reference ranges are population averages. Your optimal range may be different from the population average, particularly for hormones.
The most useful thing I learned from a year of testing my own blood is that I was making assumptions about my health that the data didn’t fully support.
That’s a useful thing to learn at 51 rather than 65.
References
[1] Kronenberg, F. (2016). Human genetics and the causal role of lipoprotein(a) for various diseases. Cardiovascular Drugs and Therapy, 30(1), 87–100. https://doi.org/10.1007/s10557-016-6648-3 [VERIFY — confirm before publishing]
[2] Mora, S., & Ridker, P. M. (2006). Justification for the use of statins in primary prevention: An intervention trial evaluating rosuvastatin (JUPITER) — can C-reactive protein be used to target statin therapy? The American Journal of Cardiology, 97(2), 33–41. [VERIFY — confirm before publishing]
[3] Hollis, B. W., & Wagner, C. L. (2013). Clinical review: The role of the parent compound vitamin D with respect to metabolism and function. Journal of Clinical Endocrinology & Metabolism, 98(12), 4619–4628. https://doi.org/10.1210/jc.2013-2503 [VERIFY — confirm before publishing]
