08 Sep Doctors are people too
Although doctors understand the risks associated with opioids only too well, some researchers believe that the daily pressures that these professionals experience could interfere with their decision making.
Doctors dealing with a high number of complex cases each day are likely to experience financial pressure, time pressure, and cognitive pressure.
As the authors of a study published in JAMA explain, The concept that time pressure can drive physician decision making is long-standing, but few studies have examined the question in detail.
In particular, the researchers wanted to investigate the timing of appointments. They hypothesized that doctors might prescribe opioids more often in appointments that take place toward the end of the day.
The findings of a previous study that looked at the same effect in relation to antibiotic overprescribing — another concerning issue — supported this theory. The researchers concluded that doctors were more likely to prescribe antibiotics as their shift progressed.
To investigate, the scientists took data from 678,319 primary care appointments.
All of the patients had visited the doctor to discuss new painful conditions, which the researchers grouped into five categories. These were: headache, back pain, joint diseases, other musculoskeletal conditions including fibromyalgia, and other pain syndromes.
None of the patients had received an opioid prescription in the past 12 months.
For their analysis, the researchers grouped the appointments into sets of three, up to a maximum of 21 appointments per day. They collated information on the lateness of appointments in 10- minute increments, for instance, 0–9 minutes late or 10–19 minutes late.
Overall, doctors prescribed opioids during 4.7% of the appointments.
As expected, when the team dived into the timing and lateness of the appointments, there was an effect. Although, as the authors note, the effect size was only moderate, it was significant. The authors write:
We observed increasing rates of opioid prescribing as appointments progressed through the day and as they ran behind schedule.
In the first three appointments of the day, doctors prescribed opioids 4% of the time. In the 19th–21st appointments, 5.3% of appointments resulted in an opioid prescription. In relative terms, this is a 33% increase between the first and last appointments of the day.
When they assessed the lateness of the appointments, there was a similar but smaller effect. Of the appointments that were running 0–9 minutes late, 4.4% ended with an opioid prescription. In comparison, appointments that were at least 1 hour late resulted in an opioid prescription in 5.2% of cases, which is a 17% relative increase.
The scientists also analyzed both prescriptions for nonsteroidal anti-inflammatory drugs, which are a nonopioid alternative for pain relief, and referrals to physical therapy, which is a delayed treatment for pain.
In both cases, there were no significant differences across the day or in relation to lateness.
Although the size of the effect that this study reported was only moderate, the authors believe that a change in prescribing the behavior of this magnitude could have meaningful relevance for national trends in opioid use.
However, they also note that the associations that they measured are much smaller than the variation that exists among doctors, hospitals, and geographical regions of the U.S.
As an example, a CDC report on opioids explains that average per capita amounts prescribed in the top prescribing counties [were] approximately six times the amounts prescribed in the lowest prescribing counties in 2015."
Although the researchers behind the current study cannot know precisely what influences a doctor’s decisions, they believe that as the day progresses, the pressures that doctors are under mean that they are less likely to make effortful decisions, such as denying opioid therapy.
The study does have certain limitations. First and foremost, it is observational, so it is not possible to interpret the results as causal. Also, although the researchers had access to a large quantity of data, these may not be generalizable to the U.S. population at large.
It is also worth noting that the researchers did not have access to information about pain severity and how effective earlier treatments had been.