Antidepressant compliance in a rural setting: Response.

Sir,

We thank Thakurdesai et al.[1] for their interest in our article.[2] We agree that recruitment was not naturalistic; this is because the study, and recruitment therein, was part of a larger study that we had described earlier.[3] In any case, no study procedure, from recruitment to assessment, can be considered naturalistic; it is the management, including management across follow-up, that is usually the naturalistic element.

Patient education and guidance was provided during the PACECAR study, but only to the extent that it would be provided anyway, as necessary, in primary care contexts such as that in which the study was conducted. In similar manner, interventions to reduce adverse effect burden were permitted to the extent that such would have occurred in primary care.

We agree that the presentation of the data in the paper falls short of the ideal. This is because the primary data, including the source data, were in the custody of an investigator who met with a fatal traffic accident. Recovery of the data has subsequently been problematic, and we had to make do with what we had.

We agree that different drugs were escalated at different rates to therapeutic or peak doses. However, in this regard, some drugs are naturally at an advantage. For example, escitalopram is easily started at 10 mg/day and escalated to 20 mg/day with no need for further escalation. In contrast, whereas escalation of tricyclic antidepressants is ideally in 25 mg steps, dispensing tablets of smaller strengths with instructions to escalate to appropriate doses could have been challenging in a primary care, rural setting in which patients are commonly seen once a month. We did, however, instruct patients to take half of a 75 mg tablet of the tricyclic in the initial week to improve tolerability; we were reluctant to instruct patients to raise the dose to two tablets in later weeks because we wished to first evaluate the patients before increasing the dose. In this regard, we were proven right because even with this slow-dose uptitration, nearly half of the patients who received tricyclics reported the experience of adverse events as an explanation for drop out.

In retrospect, duloxetine could have been started at 60 mg/day from the outset, as is generally recommended, or uptitrated to 60 mg/day after the first week, as advised for the tricyclics. However, the occurrence of gastrointestinal adverse effects in unsupervised patients could then have risked early drop out. In similar manner, perhaps sertraline could have been uptitrated to 100 mg/day after the first week and the maximum dose of the drug could have been set at 200 mg/day. Nevertheless, as the results showed, outcomes with duloxetine and sertraline were reasonable, despite the slow uptitration, and nonresponse was not an important reason for noncompliance or drop out.

Finally, we agree that the use of simple ratings instruments could have made decision-making and reporting more objective. However, such ratings does not occur in naturalistic practice even in large academic departments of psychiatry, let alone in primary care in rural settings.

In summary, whereas we agree with Thakurdesai et al.[1] with regard to almost all if not all of the issues raised in their letter, our study was conducted as it was in order to determine what works best in naturalistic practice in rural India.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.