Jump to comment:
- Page navigation anchor for Response to e-letter: Antibiotic consumption and time to recovery from uncomplicated urinary tract infectionResponse to e-letter: Antibiotic consumption and time to recovery from uncomplicated urinary tract infection
We thank Schmieding and colleagues for their e-letter, and here respond to their comments.
Only 11.4% of participants changed antibiotics, not the ‘one-quarter’ suggested by Schmieding and colleagues. We agree that our results cannot be generalised to patients who are prescribed more than one type of antibiotic for their UTI, but this is the minority of patients presenting with uncomplicated UTI in routine practice.
We agree that missing follow-up data may not be strictly missing at random. However, this study required accurate antibiotic consumption data and so excluded those where this was missing.
We agree that comorbid conditions, oral corticosteroids use, and prior duration of symptoms might have been useful predictors to include in a propensity score; unfortunately, such data was not available. However, POETIC included patients only if they had no UTI for 14 days prior recruitment. Analgesics and antimuscarinics may affect urinary symptoms and so are relevant to include and antifungals are relevant as fungal genital infection symptoms can be confused with UTI. We did include previous UTI treatments in the propensity score within the sensitivity analysis.
Both the main and sensitivity analyses estimated the average treatment effect for all units in the target population. We did not target the conditional treatment effect in the untreated.
Finally, it is not correct to say that 98.7% adhered to their treatment g...
Show MoreCompeting Interests: Nick Francis reports grants from European Union FP7, during the conduct of the study; personal fees, and non-financial support from Abbott and GSK, outside the submitted work. - Page navigation anchor for Antibiotic consumption and time to recovery from uncomplicated urinary tract infection: secondary analysis of observational data from a point-of-care test trialAntibiotic consumption and time to recovery from uncomplicated urinary tract infection: secondary analysis of observational data from a point-of-care test trial
We commend the authors for using propensity score matching models to control for important variables leading to confounding, but we believe that important methodological considerations should be addressed before these findings are used to guide clinical practice.
First, patients receiving multiple antibiotics successively (i.e., changed therapeutic regimen) were excluded, and it seems plausible that these patients may experience prolonged symptom durations. Thus, excluding them likely biases the effect estimates toward a greater beneficial effect of treatment. Despite accounting for one-quarter of the patients in the matched treatment group (i.e., 73 “changers” versus 333 patients), the authors describe the number of these “changers” as small in their limitations section. Similarly, the number of patients with incomplete data is larger than the control group (i.e., 116 total patients excluded for missing follow-up data versus 80 patients in the control group). Therefore, missing at random may not be a reasonable assumption.
Second, the list of measured variables included in the propensity score model is incomplete, as important factors guiding treatment choice and response are omitted, most notably comorbidities (e.g., diabetes), therapeutic agents (e.g., steroids), and the duration of symptoms prior to the office visit. In particular, we wonder why concurrent use of analgesic, antifungal, and antimuscarinic medications relevan...
Competing Interests: TK received outside of the submitted work, research grants from the Gemeinsamer Bundesausschuss (G-BA – Federal Joint Committee, Germany), and the Bundesministerium für Gesundheit (BMG – Federal Ministry of Health, Germany). He further has received personal compensation from Eli Lilly and Company, Teva Pharmaceuticals, TotalEnergies S.E., the BMJ, and Frontiers. MLS received outside of the submitted work, research grants from the Bundesministerium für Gesundheit (BMG – Federal Ministry of Health, Germany).