The Impact of Individualizing Sodium Bicarbonate Supplementation Strategies on World-Class Rowing Performance

The researchers compared two different timing strategies for taking sodium bicarbonate (NaHCO₃) before a 2,000-meter rowing time trial:

**Consensus timing (CON):** Athletes took 0.3 grams of NaHCO₃ per kilogram of body weight, then started their time trial exactly 60 minutes later. This matches the minimum recommendation from the International Olympic Committee (IOC) consensus statement.

**Individualized peak timing (IP):** Athletes first completed a separate "loading profile" session where they took the same dose (0.3 g/kg) and had blood samples taken every 10 minutes to determine exactly when their blood bicarbonate levels peaked. Then, on a different day, they timed their time trial to start at that individual peak time (which ranged from 30 to 180 minutes across athletes).

The comparator was the CON condition (standard timing). There was no placebo group. The primary outcome was 2,000-meter rowing ergometer time-trial performance (in seconds). Secondary outcomes included blood bicarbonate levels, gastrointestinal distress scores, and acid-base balance measures.

Twenty-three (n = 23) elite male rowers from two research centers (Canadian Sport Institute Pacific and New South Wales Institute of Sport, Australia). The group included:

4 lightweight rowers (body weight 73.6 ± 2.1 kg)

19 open-class rowers (body weight 93.6 ± 5.8 kg)

13 Olympic or World Championship team members

One rowing ergometer world record holder

All athletes could complete a 2,000-meter ergometer time trial at or below 6 minutes 20 seconds (personal bests ranged from 5 minutes 39 seconds for open-weight to 6 minutes 14 seconds for lightweight)

15 of 23 had previous experience with NaHCO₃ supplementation

Average training experience: ~9 ± 3 years of competitive rowing

Average of 3 × 2,000-meter time trial tests per year over the 12 months prior to the study

**Blood bicarbonate concentration ([HCO₃⁻]):** Measured via finger-prick capillary blood samples analyzed on an ABL 80 Flex blood gas analyzer (Radiometer, Copenhagen). Samples were taken in duplicate every 10 minutes during the loading profile and before/after warm-up during experimental trials.

**Time-to-peak bicarbonate:** Defined as the final measurement time point before a decline in [HCO₃⁻] greater than the technical error of measurement (0.6 mmol·L⁻¹). Determined during the individualized loading profile session.

**Performance:** 2,000-meter rowing ergometer time trial (Concept 2 ergometer). Time recorded in seconds.

**Gastrointestinal (GI) distress:** Self-reported using a visual analog scale (0–10, where 0 = no symptoms and 10 = severe symptoms). Assessed upon arrival and throughout the sampling period.

**Hydration status:** Urine specific gravity (USG) measured at the start of each trial using a refractometer (Atago PAL-10S, Tokyo).

**Body weight:** Measured upon arrival, before and after toilet visits, and at the end of sampling (Avery Berkel Model HL120 scale).

**Study design:** This was a randomized, crossover, single-blind (athlete-blinded deception) study. Each athlete completed three sessions: (1) an individualized NaHCO₃ loading profile to determine their personal time-to-peak bicarbonate, (2) a consensus timing (CON) experimental trial, and (3) an individualized peak (IP) experimental trial. The order of the two experimental trials was randomized.

**Randomization:** The two experimental trials (CON vs. IP) were randomly assigned. The paper does not specify the randomization method (e.g., computer-generated random numbers, coin flip).

**Blinding:** Athletes were "single-blinded" using a deception strategy. They were told the study was investigating "the effect of different ingestion-to-exercise time intervals on performance" but were not told which condition was hypothesized to be superior. This is a form of partial blinding—athletes knew they were taking NaHCO₃ (not a placebo), but they did not know which timing was expected to work better.

**Washout period:** Trials were separated by >5 days and <14 days. This is adequate to avoid carryover effects since NaHCO₃ is cleared from the body within hours.

**Duration:** Each trial session lasted approximately 3–4 hours (including baseline measures, ingestion period, waiting time, warm-up, and the time trial itself). The entire study per athlete spanned 2–4 weeks (loading profile plus two experimental trials with washout periods).

**Dietary controls:** Athletes replicated their typical 24-hour pre-competition nutrition practices (macronutrient composition, volume, and timing) before all three trials. This is a strength because it mimics real-world conditions, but it also means diet was not strictly standardized across athletes.

**NaHCO₃ dosing:** 0.3 g·kg⁻¹ body mass, consumed as gelatin capsules (1,000 mg per capsule) over a 30-minute period (one-third at 0, 15, and 30 minutes) with a standardized snack (1.5 g·kg⁻¹ carbohydrate) and 10 ml·kg⁻¹ fluid. Capsules were third-party batch tested for prohibited substances.

**Warm-up:** Athletes followed their own individualized pre-race warm-up routines (not standardized), which is another ecological validity feature.

**Statistical approach:** Paired t-tests or equivalent non-parametric tests were used to compare CON vs. IP conditions. Effect sizes were reported as Cohen's d. Confidence intervals (95% CI) were provided for blood bicarbonate differences. The primary analysis compared performance times between conditions.

**What this design can and cannot prove:**

This crossover design can prove that, within the same athletes, the IP timing strategy produced different performance outcomes compared to CON timing. Because each athlete served as their own control, individual variability in response to NaHCO₃ is accounted for. The randomization of trial order controls for order effects (e.g., learning or fatigue).

However, this design **cannot** prove that NaHCO₃ itself improves performance—there was no placebo condition. Both trials involved NaHCO₃ ingestion, so the comparison is between two active timing strategies, not between supplement and no supplement. The design also cannot determine whether the performance improvement was due to the timing of peak bicarbonate or to other factors (e.g., athletes may have performed better simply because they had more time to prepare or because of expectation effects, despite the deception). The lack of a placebo control means we cannot rule out that both conditions improved performance relative to no supplement, or that neither did.

**Major methodological weaknesses:**

No placebo condition (cannot assess absolute efficacy of NaHCO₃)

Single-blind only (researchers knew the conditions)

Small sample size (n = 23) for detecting small effects

No power analysis reported

Diet was replicated but not strictly controlled (athletes self-reported)

Warm-up was not standardized

The deception blinding may not have been fully effective (athletes with prior NaHCO₃ experience may have had expectations)

**Primary outcome: Performance time**

IP condition: 367.0 ± 10.5 seconds (6 minutes 7.0 seconds)

CON condition: 369.0 ± 10.3 seconds (6 minutes 9.0 seconds)

Mean difference: 2.0 seconds faster in IP condition

p = 0.007

Cohen's d = 0.15 (small effect size)

**Blood bicarbonate levels**

At pre warm-up: IP had significantly higher [HCO₃⁻] than CON (mean difference +2.9 ± 0.4 mmol·L⁻¹; 95% CI 2.0 to 3.8 mmol·L⁻¹; p = 0.02; d = 1.08)

Immediately pre time trial (post warm-up): No significant difference between conditions (CON ~+5.5 mmol·L⁻¹ above baseline; IP ~+6.0 mmol·L⁻¹ above baseline)

This suggests the performance benefit occurred despite similar blood alkalinity at the start of exercise

**Time-to-peak bicarbonate (from loading profile)**

Individual times ranged from 30 to 180 minutes post-ingestion

This confirms high inter-individual variability in absorption kinetics

**Gastrointestinal distress**

No significant differences in GI distress scores between CON and IP conditions

The paper does not report specific GI distress numbers or p-values in the abstract

**Hydration status**

Urine specific gravity was similar between trials (no significant differences reported)

The 2-second improvement on a ~6-minute effort represents a **0.54% performance improvement**. In competitive rowing, this is meaningful—at the elite level, races are often decided by fractions of a second. However, the effect size (d = 0.15) is considered "small" by conventional standards. To put it in perspective:

Over 2,000 meters, 2 seconds corresponds to roughly 5–7 meters of boat length advantage

This is approximately the same magnitude of benefit seen with caffeine supplementation in endurance events

The effect is smaller than what meta-analyses of NaHCO₃ vs. placebo typically report (which average ~1.5–2% improvement)

The blood bicarbonate data are puzzling: despite a clear difference at pre warm-up (+2.9 mmol·L⁻¹), this difference disappeared by the time the race started. This suggests the performance benefit may not have been due to greater blood alkalinity during the race itself, but perhaps to some other factor (e.g., better pacing strategy, psychological readiness, or residual effects from the different timing).

**Acknowledged by authors:**

Small sample size (n = 23) limits generalizability

No placebo condition (cannot separate NaHCO₃ effect from timing effect)

Single-blind design only

The mechanism for the performance improvement remains unclear (blood alkalinity was similar at race start)

Athletes were all male and elite-level, limiting applicability to other populations

The study was conducted on ergometers, not on-water, which may not fully replicate competitive conditions

**Additional critical observations:**

**No power analysis:** The authors do not report a priori power calculations, raising questions about whether the study was adequately powered to detect the small effect observed

**Multiple comparisons:** Several outcomes were tested (performance, blood measures, GI distress) without correction for multiple comparisons, increasing the risk of Type I error

**Deception effectiveness:** 15 of 23 athletes had prior NaHCO₃ experience and may have had expectations about timing effects, potentially compromising blinding

**Dietary replication:** Athletes self-reported their 24-hour diet replication, which is less reliable than researcher-controlled feeding

**Warm-up variability:** Individualized warm-ups introduce uncontrolled variance that could affect performance independently of the supplement timing

**No female athletes:** Results may not generalize to female rowers, who have different physiology and may respond differently to NaHCO₃

**Funding source:** The paper does not declare specific funding, but several authors are affiliated with national sport institutes, which may have interest in positive results

**Single dose only:** Only one dose (0.3 g/kg) was tested; results may differ at other doses

**No assessment of long-term effects:** Only acute effects were measured; chronic use of NaHCO₃ may have different outcomes

For someone running their own n=1 experiment:

**What to test:**

Compare two timing strategies for sodium bicarbonate supplementation before a high-intensity effort lasting 1–10 minutes (e.g., 2,000-meter row, 1,000-meter kayak, 400-meter swim, 5 km cycling time trial, or repeated sprints)

Strategy A: Take 0.3 g/kg NaHCO₃ exactly 60 minutes before your event

Strategy B: First determine your individual time-to-peak bicarbonate by taking the same dose and measuring blood bicarbonate every 10–15 minutes (using a portable blood gas analyzer or lactate meter that measures bicarbonate), then time your event to start at your personal peak

**Minimum meaningful duration:**

Each trial session: ~3–4 hours (including preparation, ingestion, waiting, warm-up, and the event)

Total experiment: At least 3 sessions (1 loading profile + 2 experimental trials) over 2–4 weeks

Allow >5 days between trials to avoid carryover effects

For a simpler n=1 test without blood sampling: Try 3–4 different timing windows (e.g., 45, 60, 90, and 120 minutes before exercise) on separate days and see which feels best and produces the best performance

**What to measure (specific metrics):**

**Primary:** Performance time (seconds) for your chosen event. Use the same equipment, course, and conditions each time.

**Secondary:** Perceived exertion (RPE 1–10 scale), gastrointestinal distress (0–10 scale for nausea, bloating, cramping), and subjective feeling of readiness

**Optional:** Blood bicarbonate if you have access to a portable analyzer (e.g., Abbott i-STAT or similar)

**Confounders to track:** Sleep quality (hours and subjective rating), hydration status (urine color or specific gravity), pre-exercise nutrition (timing and composition of last meal), caffeine and other supplement use, time of day, menstrual phase (if female), and stress levels

**Key confounds to control for:**

**Diet:** Replicate your pre-competition meal exactly before each trial (same foods, same timing, same portion sizes)

**Hydration:** Drink the same amount of fluid before each trial

**Warm-up:** Use exactly the same warm-up protocol each time

**Time of day:** Conduct all trials at the same time of day (circadian rhythms affect performance and bicarbonate metabolism)

**Caffeine and other supplements:** Keep these consistent or avoid them entirely during the testing period

**Sleep:** Aim for the same sleep duration and quality the night before each trial

**Familiarization:** Complete at least 2–3 practice time trials before starting the experiment to minimize learning effects

**Blinding:** If possible, have someone else prepare your supplement so you don't know which timing condition you're in (use opaque capsules and identical protocols)

**What a positive result would look like:**

A consistent performance improvement of 0.5–2% (e.g., 1–4 seconds on a 6-minute effort) when using individualized timing compared to standard 60-minute timing

The improvement should be reproducible across at least 2–3 trials per condition

You should also experience acceptable GI distress (≤3 on a 0–10 scale) and feel that the timing strategy is practical for competition

If you don't have blood testing capability, a positive result would be simply feeling better and performing better at a specific timing window (e.g., 90 minutes vs. 60 minutes)

Note: The effect is small and may not be noticeable in a single trial—you may need 3–5 trials per condition to detect a reliable difference

**Important caveats for your n=1 experiment:**

NaHCO₃ can cause significant GI distress (nausea, bloating, diarrhea) in some people. Start with a lower dose (0.2 g/kg) to test tolerance before moving to 0.3 g/kg.

The capsules should be taken with food (1.5 g/kg carbohydrate) and plenty of water (10 ml/kg) to reduce GI issues

Individual time-to-peak can vary from 30 to 180 minutes, so if you don't test your own profile, try multiple timing windows

This supplement is not banned by WADA, but always check current regulations

The effect is small—don't expect dramatic improvements. The 2-second gain in