Change in access to prescribed medication following an episode of deliberate self-poisoning: a multilevel approach

Abstract

Objective: Patients with a history of deliberate self-poisoning (DSP) are prescribed a greater amount of medication than the general public. DSP is the most robust risk factor for repeat episodes of DSP and subsequent death by suicide, and one might therefore expect that access to prescribed medication would be reduced following an episode of DSP. However, it is unclear whether access to prescribed medication changes after an episode of DSP. The objectives of this study were to investigate changes in 1) overall, psychotropic, non-psychotropic and the psychotropic subgroup antidepressant prescribed medication availability in DSP patients following an episode of DSP, 2) prescribing of the medication ingested in the episode, and 3) potential effects of gender, age and repeater status on such change.

Methods: The design was longitudinal. We included 171 patients admitted for DSP between January 2006 and March 2007. Data on patients' prescriptions prior to admission were retrieved from The Norwegian Prescription Database. The outcome measure was the difference between medication load in the year following compared to the year prior to the DSP episode.

Results: There was a significant increase in total medication load following DSP, including both psychotropic and non-psychotropic medication. Antidepressant medication load remained stable. There was a tendency for access to drugs ingested in the episode to increase following the episode, albeit not significantly. Medication load increased with age across all medication groups irrespective of time period and gender.

Conclusions: The findings show that physicians do not curb prescribing to patients who have recently deliberately self-poisoned. Moreover, they highlight the need for cautious and judicious prescribing for these patients, in combination with psychological and social interventions.

Figure 1. One year medication load by age and gender.

Total medication load increased significantly in the year following the episode compared to beforehand (p<0.01). Figure 1A displays one year total medication load (DDD) for females by age before and after the episode, with the means (solid lines) and standard error envelopes (dashed lines). Females collected significantly more medication in the year after compared to the year prior to the episode (p<0.01), with the increase in DDD collected amounting to 21.2%. The pre-post difference for females gradually increased with age. There was no significant pre-post difference for males.

Figure 2. One year psychotropic load by age and gender.

Figure 2A displays one year psychotropic medication load before and after the episode for females with the means (solid lines) and standard error envelopes (dashed lines). Psychotropic medication load increased significantly in the year after compared to the year prior to the episode (p = 0.04). Psychotropic medication load increased with age irrespective of time period and gender (p<0.01), and peaked in the 50–70 age bracket. There was no significant difference between males' and females' psychotropic medication load (p = 0.27).

Figure 3. One year non-psychotropic load by age and gender.

Figure 3A displays one year non-psychotropic medication load before and after the episode for females, and Figure 3B displays the corresponding information for males. Solid lines represent the means with dashed lines showing standard error envelopes. There was a significant increase in non-psychotropic medication load in the year after the episode (p<0.01) compared to the year prior to the DSP episode (a 24.3% increase in DDD collected), but this increase was not dependent on gender (p = 0.53). Non-psychotropic medication load increased with age (p<0.01), and females collected more non-psychotropic overall irrespective of age and time period (p = 0.01).

Figure 4. DDD per quarter year before and year after the episode by medication group, cumulating quarterly within each year.

Comparison of year prior to and following the episode is potentially inaccurate as patients may be prescribed medication at different times in the two years, i.e., we may not be comparing the same time periods. We therefore investigated whether they were accumulating at the same rate in the two years. Figure 4 depicts the accumulated (total) DDD collected over the year prior to and the year following the episode, broken down into quarters and reported for non-psychotropic, psychotropic, and antidepressants within each quarter. The pace with which patients accumulated prescribed medication was steady over the two years, although the increase was faster over the first three months following the episode compared to the year preceding it.