Background Hydroxychloroquine (HCQ), a 4-aminoquinolone antimalarial, is regarded as the oral therapy of choice for cutaneous and systemic lupus erythematosus (SLE). It is also licensed for rheumatoid arthritis (RA). Studies of HCQ-treated patients with SLE or RA have demonstrated a positive correlation between whole-blood HCQ levels and clinical response. Such studies have involved measuring whole-blood concentrations at any given time point after HCQ ingestion assuming that steady-state concentrations would undergo limited fluctuation over a daily interval because HCQ has a long half-life. This approach might not sufficiently take into account the potential intra-patient variation in HCQ blood levels that can occur over a 24-hour period. Such variation, if significant, could affect the credibility of any concentration-response relationship provided from these previous studies. Objectives The objectives of this report are to: (a) investigate the intra-patient variation in HCQ whole-blood levels and (b) suggest an optimum time for sampling patients for future studies. Methods Six patients were recruited with cutaneous lupus erythematosus who had each been on HCQ 200 mg twice daily for at least six months, so that they were at steady-state. Each patient was fasted overnight and had standardized meals and dosing schedule. Whole blood was sampled at seven time points over 24 hours. Whole-blood HCQ levels were measured with high-performance liquid chromatography using gradient elution, fluorimetric detection and chloroquine as an internal standard. The assay had a mean inter- and intra-day coefficient of variation of 10% and 5% respectively and a limit of detection of 5ng/ml. Results HCQ levels appeared to follow a biphasic pattern over the sampling period. Maximum levels were noted a median of four hours (range 2-6) after ingestion. Median intra-patient variation between trough and peak levels, 'Cmax' ((peak - trough)/trough × 100%), was 27% (range 8-150%). Conclusions This study demonstrated that whole-blood HCQ levels vary 27% (median, range 8-150%) within an individual over a 12-hour period. Drug levels might differ between individuals because of multiple factors, including variable adherence to medication. Measuring HCQ levels for assessment of drug adherence could be valuable in the 'real-world' clinical setting. This could be assessed by taking a blood sample at any time following HCQ ingestion. If patients were found to have very low or undetectable levels of HCQ, non-adherence to HCQ should be suspected.