A SUPER-powered WP Engine Atlas Test Site

Joe&#039;s Atlas Experimentation Site

Significant delays in the diagnosis of <a href="https://alsnewstoday.com/what-is-als/">amyotrophic lateral sclerosis</a> (ALS) were revealed by a new large-scale analysis of real-world patient records using artificial intelligence (AI).
In fact, the median time from the first symptom to an ALS diagnosis was 11 months for these patients.
Similar delays were seen regardless of the location of symptom onset — either the limbs or the bulbar region that&#8217;s involved in speech and swallowing. But researchers noted that having shortness of breath, called dyspnea, as the first reported symptom led to further delays in diagnosis and neurologist referral.
&#8220;Dyspnea was fairly common and when it was present, a longer delay until visiting the specialist (neurologist) and until diagnosis were observed,&#8221; the team wrote, adding, &#8220;We cannot exclude that this symptom, so common in other [diseases], could have gone unnoticed or could have been attributed to other causes when appearing in very early stages of the disease.&#8221;
The study, “<a href="https://www.nature.com/articles/s41598-023-27863-2">Symptoms timeline and outcomes in amyotrophic lateral sclerosis using artificial intelligence</a>,” was published in the journal <a href="https://www.nature.com/srep/">Scientific Reports</a>.
[recommended-reading id=79108]
<h3>Delayed diagnosis found for ALS patients in real world</h3>
Real-world data describing ALS clinical characteristics, disease management, outcomes, and treatment responses can be found in patient electronic health records, called EHRs for short.
In recent years, computer-based AI and machine learning tools have been applied to EHRs stored in patient registries to extract and analyze real-world clinical information on large numbers of patients. This strategy can be especially helpful for rare diseases such as ALS, according to researchers.
“From a clinical standpoint, diseases with low prevalence are best understood using population-based registries with available follow-up information across large numbers of patients,” the investigators wrote.
To that end, a team based at the University Hospital of Albacete, in Spain, applied AI tools to describe the clinical profile, symptom onset, diagnosis, and relevant outcomes of ALS patients in that country. They used EHR data held in the Spanish SESCAM Healthcare Network registry.
The AI tool — called natural language processing or NLP — enabled computers to understand written text in medical records, then extract the relevant information.
Over the five-year study period, 250 patients were diagnosed with ALS, of whom 61.6% were male. The patients had a mean age of 64.7.
Among them, 159 (64%) had spinal ALS, which first affects the limbs, and 91 (36%) had bulbar ALS, marked by initial symptoms of speaking and swallowing difficulties. The most common co-existing medical conditions were high blood pressure and an imbalance of blood fats.
Data spanning at least one year before the <a href="https://alsnewstoday.com/als-diagnosis/">ALS diagnosis</a> was available for 208 patients. The most visited healthcare services before diagnosis were primary care, for 88.4%, and treatment in an emergency room, for 38.4%. Neurology was the most visited specialist service (38.8%).
The most common symptom up until diagnosis was weakness, followed by shortness of breath, speaking difficulties, swallowing problems, and muscle twitches.
Overall, the median diagnostic delay, or the time from the onset of symptoms to an ALS diagnosis, was 11 months. Similar delays were seen in both spinal and bulbar ALS subgroups.
[quote align=&#8221;center&#8221;]Our results point to &#8230; an overall diagnostic delay of 11 months before the first mention of ALS in patients’ [health records].[/quote]
Additionally, data showed that people who experienced shortness of breath as their first symptom had longer diagnostic delays, of a median of 12 months. Likewise, with shortness of breath as the first symptom, there was a longer delay in neurologist referral and subsequent diagnosis.
&#8220;The patient journey seems to play a role in [diagnostic] delay, with early referral to a neurologist linked to shorter timing to diagnosis,&#8221; the researchers wrote, noting that &#8220;ALS patients are frequently referred to other hospital departments before the neurologist.&#8221;
Compared with the period prior to a diagnosis, there was a two-to-threefold increase in speech problems (15.6% vs. 31.2%), swallowing difficulties (14.0% vs. 34.0%), and muscle twitches (13.6% vs. 37.2%) at the time of diagnosis.
During the median follow-up of 25 months after diagnosis, 52% of participants underwent a feeding tube insertion, while 64% received non-invasive breathing support and 16.4% had a breathing tube insertion, known as a tracheostomy. These procedures all occurred more frequently in the bulbar ALS group.
A total of 87.6% of patients were treated with riluzole (sold as <a href="https://alsnewstoday.com/als-treatment/rilutek-riluzole/">Rilutek</a>, <a href="https://alsnewstoday.com/tiglutik-riluzole-oral-suspension/">Tiglutik</a>, and <a href="https://alsnewstoday.com/exservan-riluzole-oral-film/">Exservan</a>).
Over a median of 19 months from ALS diagnosis, 87 patients (34.8%) died. When tracheostomy also was considered as a mortality outcome, that number increased to 110 (44.0%). The likelihood of feeding tube insertion and needing non-invasive ventilation was significantly higher for those with bulbar ALS.
These findings “highlight the diagnostic delay in ALS and revealed differences in the clinical characteristics and occurrence of major disease-specific events across ALS subtypes,” the researchers wrote.
&#8220;Our results point to &#8230; an overall diagnostic delay of 11 months before the first mention of ALS in patients’ EHRs,&#8221; the team concluded.

Tag: real-world patient data

Delayed diagnosis in ALS found in new AI analysis of patient records