Associations of Patient Mood, Modulators of Quality of Life, and Pharmaceuticals with Amyotrophic Lateral Sclerosis Survival Duration

Abstract

Associations of modulators of quality of life (QoL) and survival duration are assessed in the fatal motor neuron disease, Amyotrophic Lateral Sclerosis. Major categories include clinical impression of mood (CIM); physical health; patient social support; and usage of interventions, pharmaceuticals, and supplements. Associations were assessed at p < 0.05 and p < 0.001 significance thresholds using applicable methods (Chi-square, t-test, ANOVA, logistical regression, random forests, Fisher's exact test) within a retrospective cohort of 1585 patients. Factors significantly correlated with positive (happy or normal) mood included family support and usage of bi-level positive airway pressure (Bi-PAP) and/or cough assist. Decline in physical factors like presence of dysphagia, drooling, general pain, and decrease in ALSFRS-R total score or forced vital capacity (FVC) significantly correlated with negative (depressed or anxious) mood (p < 0.05). Use of antidepressants or pain medications had no association with ALS patient mood (p > 0.05), but were significantly associated with increased survival (p < 0.05). Positive patient mood, Bi-PAP, cough assist, percutaneous endoscopic gastrostomy (PEG), and accompaniment to clinic visits associated with increased survival duration (p < 0.001). Of the 47 most prevalent pharmaceutical and supplement categories, 17 associated with significant survival duration increases ranging +4.5 to +16.5 months. Tricyclic antidepressants, non-opioids, muscle relaxants, and vitamin E had the highest associative increases in survival duration (p < 0.05). Random forests, which examined complex interactions, identified the following pharmaceuticals and supplements as most predictive to survival duration: Vitamin A, multivitamin, PEG supplements, alternative herbs, antihistamines, muscle relaxants, stimulant laxatives, and antispastics. Statins, metformin, and thiazide diuretics had insignificant associations with decreased survival.

Keywords: ALS; antidepressants; mood; off-label drugs; quality of life; supplements; survival.

Figure 1

The use of antidepressants does not improve Clinical Impression of Mood (CIM). CIM was classified at each clinic visit using a published binary assessment [20] based on visual and verbal ques, as detailed in the Methods. All antidepressant interventions in a patient’s chart were recorded on a per-visit and per-patient basis. The use of antidepressants (n = 458) did not have a significant effect on a patient’s mood compared to those who did not take antidepressants (n = 717) (p > 0.05, t-test).

Figure 2

The type of antidepressant does not have an effect on Clinical Impression of Mood (CIM). CIM was classified at each clinic visit using a published binary assessment [20] based on visual and verbal ques as detailed in the Methods. All antidepressant interventions in a patient’s chart were recorded on a per-visit and per-patient basis. Antidepressants were divided into four general categories; SSRIs (n = 348), tricyclic antidepressants (n = 55), SNRIs (n = 56), and general antidepressants. General antidepressants were excluded from this analysis due to the small sample size. The type of antidepressants did not have a significant effect on a patient’s mood compared to those who did not take antidepressants (p > 0.05, one-way ANOVA).

Figure 3

Correlation between the most common disease and intervention categories: “Most common” was defined as usage by more than 8.5% of the ALS cohort (see Methods). On the circular relationship plot, each 1-pixel line illustrates a patient connection. Gender (G) is replaced by “male” or “female”, onset type (O) is replaced with “bulbar” or “limb”, onset age (O. Age) is replaced with “≥55 years” and “<55 years”, disease survival duration (DD) is replaced with “longer” or “shorter” (than the average disease duration, which is equal to 3.67 years for this analysis). Other shorthand symbols are defined in detail in Table S2.

Figure 4

Pharmaceuticals & supplements and patient characteristics as predictors of ALS disease duration. Conditional random forests were used to determine how much each individual metric helped increase the area under the curve (AUC), the primary metric of model performance prediction. Relative importance is normalized and scaled for ease of visualization. Note that “feeding tube” represents nutritional supplements given via a percutaneous endoscopic gastrostomy (PEG).”C” denotes that the factor is a patient characteristic versus a medication, and “L” represents that the medication or supplement is more strongly associated with longer survival duration.