Variability in article processing charges of open-access publishing across medical specialties

Open-access (OA) journals provide public access to research but often impose a financial burden on researchers through article processing charges (APCs). The connection between APCs and indicators of journal prestige, such as journal impact factor (JIF), and their variation across medical specialties remains unclear. This study investigates how publication costs relate to journal prestige across diverse medical fields.

Data from 1,117 hybrid and OA journals across 29 medical specialties were compiled from Journal Citation Reports and journal websites. Pearson correlation coefficients between APCs and journal characteristics (JIF, journal citation indicator, citation counts, and percentage of OA publications) were calculated. Linear regression was used to assess the impact of these journal characteristics on APC variance. Bartlett's test for homogeneity of variances was performed to evaluate differences in APC variance across specialties and publishing models. Journal counts by country of origin and corresponding median APCs were also analyzed.

Significant variation in APCs was observed across specialties, with hematology/oncology journals having the highest median APC ($4,690) and primary care journals the lowest ($2,690). Hybrid journals had higher median APCs than OA journals ($4,248 vs. $2,909, p < 0.001). JIF and the proportion of OA publications accounted for only 14.1% of the APC variance in Q1 journals. Weak positive correlations were found between APC and both JIF (r = 0.38) and citation counts (r = 0.38), and a weak negative correlation between APC and the proportion of OA publications (r = -0.28).

Across all medical specialties, hybrid journals have higher APCs than fully open-access journals. Although APCs vary within specialties, differences across fields are less pronounced. The weak correlations between APC and journal prestige metrics suggest that factors such as JIF, citation counts, and the proportion of OA publications account for only a small fraction of APC variability.