Summary
Data from a large, population-based cohort of 94 759 patients with incident gout were analysed, followed for >10 years in a universal healthcare setting. The study included patients whose condition was treated and those where it was untreated and allowed a longitudinal evaluation of SU and made it possible to identify factors associated with optimal management. Comorbidities were highly prevalent (hypertension 65.15% [61 732/94 759], dyslipidaemia 50.58% [47 932/94 759], type 2 diabetes 24.19% [22 919/94 759]), and 62.70% (59 416/94 759) of patients received ULT. Overall, SU control was poor, with <13% achieving sustained target levels. Better control was associated with early ULT initiation, higher allopurinol doses, febuxostat, longer treatment duration, and better adherence, whereas male sex, obesity, alcohol abuse, and CKD were linked to poorer control.
Strengths and limitations
One of the main strengths of this study is that it represents, to the authors’ knowledge, one of the largest population-based cohorts of patients with gout to date, integrating clinical, laboratory, and dispensing data from primary care, where over 90% of patients are managed. The SIDIAP database ensures broad representativeness (covering >75% of the population) and data completeness.21 Another strength of the current study is the longitudinal analysis of SU control — unlike prior cross-sectional reports11,13,14,17,22 — that provides a realistic assessment of disease control over time thus recognising gout as a chronic disease, and linkage to pharmacy-dispensing data (ATC codes), allowing reliable evaluation of ULT adherence and dosing. Unlike most studies, patients who were not treated with ULT were not excluded, avoiding selection bias and improving real-world validity. Another strength is that both single and multivariable analyses were conducted to evaluate factors associated with good SU control, and the results were very similar, which reinforces the robustness of these findings.
As with all electronical medical record-based studies, misclassification of gout diagnosis is possible, although prior validation supports data reliability.19 Large primary care databases may also not capture the nuances of individual treatment decisions. Other limitations include lack of data on gout flares, quality of life, use of diuretics or other drugs that may influence SU levels, and mortality outcomes. Another limitation is the absence of ethnicity data, as Spanish legislation does not permit their routine collection in health records. Although total follow-up spanned 12 years, the median follow-up (5 years) may still underestimate long-term management dynamics. Nonetheless, this time exceeds that of most published cohorts.
Comparison with existing literature
The current cohort’s demographic and clinical characteristics are consistent with previous studies from the Mediterranean region.22–24 Compared with non-European cohorts, variability emerges: US studies report similar or higher comorbidity and obesity rates,12,14,25 whereas Asian cohorts describe younger populations with less obesity but with persistent cardiometabolic comorbidities.11,26,27 Differences likely reflect healthcare structures, regional patterns, and study designs, as some previous cohorts included only patients who received treatment or specific subpopulations (such as Veterans Affairs in the US, or treated-only cohorts in the UK).16,28 The data from this study and the published previously one reinforce that cardiovascular comorbidities are central to gout management and should guide individualised, comorbidity-focused strategies.29
Only 62.70% (59 416/94 759) of patients in the current cohort received ULT, and adherence was poor. Similar underutilisation has been reported elsewhere, with treatment rates ranging from 27% to 66% in other population-based cohorts,12,30 and 47% in Veterans Affairs patients.31 These rates have not improved substantially in the past decade.32 In Spain, as in most other countries, gout management remains outside chronic disease protocols and most family physicians report limited familiarity with guideline recommendations.33 Moreover, 15.92% (9459/59 416) of patients had a single ULT dispensation, suggesting persistent misperceptions of gout as an acute condition rather than a chronic disease.34 Differences in guideline recommendations may contribute to this variability. The American College of Physicians (ACP) advises against starting ULT after a first flare or infrequent flares,6 whereas ACR, EULAR, the Spanish clinical practice guideline, and NICE recommend a treat-to-target strategy.7–9,35 The ACP position, however, reflected the limited long-term evidence available then, rather than evidence against ULT efficacy.
The proportion of patients achieving SU <6 mg/dL (360 µmol/L) was markedly low and stable over time, consistent with prior evidence showing suboptimal control globally.17,23,36 The proportion in the current study (<13% with good control) is comparable with, or slightly lower than, other real-world studies, highlighting that long-term control remains inadequate despite guideline recommendations. Another important finding is the lack of progress in both ULT initiation and achievement of the therapeutic target. Although some positive changes have been reported in countries with similar healthcare systems, these remain largely insufficient.13 Unlike in some countries, SU control in the current cohort was not influenced by diagnosis year, follow-up duration, or socioeconomic status, aligning with results from universal-access health systems.17,36
The current results align with previous studies identifying pharmacological factors such as higher allopurinol doses, febuxostat use, treatment duration, and adherence as key determinants of control.11,14,17 In the current cohort, febuxostat use may be associated with better control, not only because allopurinol is often prescribed at low doses, but also because of higher adherence. The authors of the current study hypothesised that this improved adherence with febuxostat could be related to the fact that it may be prescribed by physicians with a greater interest in gout and to its use as a second-line therapy in patients with more severe disease, who may therefore be more aware of, and engaged in, their treatment.
Non-pharmacological predictors of poor control — male sex, obesity, alcohol abuse, hypertension, and CKD — were also consistent with the literature.11,14,17,36 Interestingly, CKD stage V was not associated with poorer control, possibly because of dialysis or renal transplant bias.37 In contrast to the STOP GOUT trial,38 the current study found that comorbidities influenced SU control, likely reflecting differences between real-world primary care populations and tightly managed clinical trial cohorts. The multivariable analysis of pharmacological factors reinforces the findings of the univariable analysis. In the current study the multivariable model was restricted to adherence and early treatment initiation because these are modifiable factors and relevant to clinical practice. The remaining variables (age, sex, and comorbidities) act as non-modifiable confounders, and their effects were already demonstrated in the univariable analysis. Early initiation of ULT was associated with improved SU control, reinforcing the need to complement the ‘start-low, go-slow’ ULT therapy principle39 with ‘and begin soon’. Although it is plausible that patients who start treatment earlier achieve better control, in the current study the difference in initiation time between the good-control and suboptimal control groups was relatively small (25.57 versus 69.40 days; a difference of 43.83 days). Given that such a modest temporal difference was associated with more than a threefold increase in the likelihood of achieving the target outcome, this suggests that additional, unmeasured factors — such as patient or physician attitudes, treatment expectations, or clinical decision-making patterns — may play an important role, as also illustrated by qualitative studies and expert opinion.40–42
Additionally, recent evidence indicates that initiating ULT during a flare does not worsen the duration or intensity of the flare.43 This raises the question of whether the optimal timing for ULT initiation might be during the flare itself. The authors of the current study believe that this approach could help patients better understand their disease by associating clinical improvement of the flare with ULT, while also aligning the prescription pattern more closely with that of other chronic diseases. Some of the similarities observed when comparing the current results with those from other countries with universal healthcare coverage reinforce the current findings and suggest that a strategy of early initiation of ULT may be particularly effective in this context.13,17
