ADHD meds assoc with decreased mortality

Although I just sent out a blog on cardiovascular problems with ADHD medications, another article just came out finding that all-cause mortality was actually significantly lower with ADHD stimulants (see  ADHD meds dec all-cause mortality JAMA2024, or doi:10.1001/jama.2024.0851)

 

Details:

-- 148,578 individuals aged 6 through 64 with ADHD were assessed in an observational nationwide cohort study; data were from 2007-2019, no ADHD medication dispensed prior to the diagnosis, and follow-up was until January 2021

-- 84,204 (56.7%) initiated an ADHD medication and 64,296 (43.3%) did not

-- median age 17 yo, 41% female, education level: primary or lower secondary 17%/upper secondary 51%/postsecondary or postgraduate 32%

-- accidental injuries 60%, depression 19%, nonalcohol drug use disorder 9%, suicide attempt history 8%, alcohol use disorder 8%, anxiety disorder 8%, personality disorders 4%, bipolar 3%, autism spectrum disorder 3%, eating disorder 2%, schizophrenia 1.6%, conduct disorder 1.5%, intellectual disability 1.5%

-- cardiovascular disease 3%, epilepsy 2%, diabetes 0.7%, dyslipidemia 0.4%, tobacco use disorder 0.4% , hypertension not mentioned

 

-- main outcomes: all-cause mortality within 2 years of ADHD diagnosis, as well as natural-cause mortality (e.g. physical conditions) and unnatural-cause mortality (e.g. unintentional injuries, suicide, and accidental poisonings). 5-year event rates were also assessed.

 

Results:

-- results below were adjusted for age, sex, highest level of education, parents’ highest level of education, number of outpatient visits for psychiatric and nonpsychiatric reasons, hospitalizations for psychiatric and nonpsychiatric reasons, diagnosis of psychiatric disorders (anxiety disorder, autism spectrum disorder, bipolar disorder, conduct disorder, depressive disorder, eating disorder, intellectual disability, personality disorder, schizophrenia, alcohol use disorder, tobacco use disorder, and drug use disorder), physical diseases (cardiovascular disease, epilepsy, type II diabetes, and hyperlipidemia), suicide attempts, external injuries or trauma, and dispensation of other psychotropic meds (antipsychotics, anxiolytics, hypnotics and sedatives, antidepressants, antiepileptic meds, anti-addiction meds, and opioids)

 

-- 632 individuals died during the 2-year follow-up, and 1402 died during the 5-year follow-up

 

-- 2-yr all-cause mortality risk in those initiating treatment versus those not initiating treatment:

    -- on medications: 598 deaths (210,054 person-years), incidence rate 19.3 per 10,000 person-years, 2-year risk per 10,000 individuals 39.1 (33.8-45.4)

    -- not on medications:  731 deaths, incidence rate 23.8 per 10,000 person-years, 2-year risk per 10,000 individuals 48.1 (42.5-54.5)

        -- two-year risk difference: -8.9 (-17.3 to -0.6)

        -- 23% decrease with meds, adjusted hazard ratio 0.79 (0.70-0.88)

 

-- 2-yr natural-cause mortality risk in those initiating treatment versus those not initiating treatment:

    -- on medications: 203 deaths, incidence rate 6.6 per 10,000 person-years, 2-year risk per 10,000 individuals 13.1 (10.0-17.3)

    -- not on medications:  226 deaths, incidence rate 7.4 per 10,000 person-years, 2-year risk 14.7 per 10,000 individuals (11.9-18.2)

        -- two-year risk difference: -1.6 (-6.4 to +3.2)

        -- 14% decrease, adjusted hazard ratio 0.86 (0.71-1.05), not statistically significant (though strong trend)

 

-- 2-yr unnatural-cause mortality risk in those initiating treatment versus those not initiating treatment:

    -- on medications: 395 deaths, incidence rate 12.7 per 10,000 person-years, 2-year risk per 10,000 individuals 25.9 (21.8-30.8)

    -- not on medications:  505 deaths, incidence rate 16.4 per 10,000 person-years, 2-year risk per 10,000 individuals 33.3 (28.5-35.8)

        -- two-year risk difference: -7.4 (-14.2 to -0.5)

        -- 25% decrease with meds, adjusted hazard ratio 0.75 (0.66-0.86)

 

    -- Of the unnatural causes, accidental poisonings were statistically significant (there were small trends favoring medications for suicide and trends favoring no medications for accidental injuries)

        -- 2-yr accidental poisonings mortality risk in those initiating treatment versus those not initiating treatment:

            -- on medications: 92 deaths, incidence rate 3.0 per 10,000 person-years, 2-year risk per 10,000 individuals 6.0 (4.2-8.7)

            -- not on medications:  183 deaths, incidence rate 6.0 Per 10,000 person-years, 2-year risk per 10,000 individuals 12.1 (0.36-6.0)

        -- 2-year risk difference: -6.0 (-9.8 To -2.3)

        -- 53% decrease with meds, adjusted hazard ratio 0.47 (0.36-0.60)

 

-- analysis stratified by age (age 6 to 24 versus age 25 to 64):

    -- all-cause mortality: 140 versus 458 deaths; 26% versus 18% decreased risk, statistically significant for both age groups

    -- natural-cause mortality: 20 versus 178 deaths; 37% versus 13% decreased risk, neither age group statistically significant

    -- unnatural-causes mortality: 120 versus 279 deaths; 24% versus 21% decreased risk, statistically significant for both groups

 

-- analysis stratified by sex (female versus male):

    -- all-cause mortality:  171 versus 434 deaths; 14% versus 23% decreased risk, statistically significant for both males and females

    -- natural-cause mortality: 54 versus 154 deaths; not statistically significant in males, significant 36% decrease in females

    -- unnatural-cause mortality: 117 versus 280 deaths; not statistically significant females, significant 32% decrease in males

 

-- 92% of patients on meds started with stimulants, but there was no difference in all-cause mortality rate in those who started on stimulants versus non-stimulants

 

-- analysis of the 5-year mortality risk, comparing those initiating treatment versus those not initiating treatment:

    -- all-cause mortality (1549 events versus 1483 events): strong trend though not statistically significant difference, with adjusted hazard ratio 0.94 (0.88-1.01)

    -- natural-cause mortality (583 events versus 512 events): no statistical difference

    -- unnatural-cause mortality (966 events versus 971 events): adjusted HR 0.89 (0.81- 0.97), 11% decreased risk

        -- accidental injuries (64 events versus 85 events, adjusted HR 0.71 (0.51-0.98), 29% decreased risk

        -- accidental poisoning (323 events versus 386 events): a adjusted HR 0.74 ( 0.64-0.86), 26% decreased risk

        -- no significant difference in suicide rates

 

Commentary:

-- ADHD is remarkably common, affecting 5.9% of youths and 2.5% of adults worldwide, per the 2021 World Federation of ADHD International Consensus Statement

-- in the US, the prevalence of ADHD is approximately 9.8% among children and adolescents, and 4% among adults

    -- these numbers are likely very much

- underreported

-- ADHD itself is associated with a 2-fold increased risk of premature death, likely in part because of the wide range of psychiatric and physical comorbidities. This study did a great job in adjusting for these comorbidities in their statistical analysis (see above for the large array of comorbidities included in their adjusted model) 

 

-- one of the important points about ADHD is that 50-90% of patients have significant psych comorbidities (more so in adults):

    -- children commonly have learning disorders, language disorders, autism spectrum disorders, developmental coordination disorders, sleep disorders, tics, depression/suicidality, anxiety, substance use, oppositional defiant disorder, and conduct disorders

    -- adults most commonly have substance use disorder, mood disorder, anxiety, and antisocial personality disorder

        – and, many of these psych comorbidities themselves are associated with increased death rates, as are the medical comorbidities

-- however, there is concern about cardiovascular safety, as noted in https://gmodestmedblogs.blogspot.com/2024/02/adhd-medications-and-cardiovasc-effects.html

    -- however, the major finding in that study was increased hypertension associated with medications, as well as arrhythmias but to a lesser degree than hypertension.

          -- hypertension is however supposed to be carefully monitored in patients on ADHD medications, which should (hopefully) mitigate much of the hypertension-related cardiovascular effects, though of the 27 million people with hypertension in the US in 2023, 45% were not appropriately treated (https://www.cdc.gov/bloodpressure/facts.htm)

          -- in Sweden (in this study), there is more aggressive control of blood pressure (on the order of 25% with uncontrolled hypertension in 2023), and thereby likely to decrease cardiovascular outcomes there more so than in the US

 

-- in the large population of people with ADHD in this Swedish study, initiation of ADHD medication was associated with a lower mortality 2 years after diagnosis, especially for unnatural-cause mortality

    -- it is likely that these benefits were related largely to decreasing the impulsivity and improving the decision-making in those on ADHD therapy

-- other studies have also found that ADHD meds are associated with lower risk of injuries and traumas, motor vehicle accidents, suicidality, substance use disorder, and criminality (see ADHD med effects on behav BioPsych2019 in dropbox, or doi.org/10.1016/j.biopsych.2019.04.009); this is consistent with the findings in this Swedish study where “unnatural” causes of death were decreased substantially with ADHD meds

-- stimulants may also contribute to lower rates of smoking in adults, improvements in lifestyle, self-regulation, and executive function; all of these might improve “natural” mortality (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2782603/ ).

-- this study also found that both the younger and older groups had significant mortality benefits from medications as well as females vs males (though females were more likely to have benefits in the natural causes of mortality and males with the unnatural causes)

    – not sure why the difference here. would have been useful to have a breakdown of some of the specifics: was there a difference in those on ADHD meds in terms of continued smoking? or hypertension? or diabetes? or hyperlipidemia? any of these could easily be related to "natural" mortality

-- prior studies on ADHD and mortality have had mixed results, with none of them dealing with adult ADHD. This is the first study including adults in the analysis

-- the difference in outcomes between younger versus older patients (with many more events in the older group) is not really surprisingly in a 2-year or 5-year assessment, and this difference might limit the statistical validity of the study in younger individuals (versus a much longer study). Perhaps more information might be also gleaned if there were other age breakdowns, such as adding a 24-50yo age bracket?

-- other studies have found that females with ADHD have had higher rates of depression, sleep disorder, atrial fibrillation, and asthma than men

 

Limitations:

-- this study only provides 2-year and 5-year follow-ups, and the lack of long-term follow-ups could undermine the generalized applicability of their results

-- as an observational study, one can only make conclusions about the association between ADHD meds and mortality (and not causality), given the likelihood of unmeasured confounders

-- it is hard to disentangle how much of the death rate is related to ADHD itself versus is comorbidities

    -- as noted above, there is significant mortality associated with many of the psychiatric comorbidities themselves (depression, bipolar, anxiety, etc).  we do not know if these were treated, or how. there might well also be a selection bias here: those started on ADHD meds who had psych comorbidities may have been more likely to get meds for these comorbidities than in those not treated for the ADHD. If so, that would affect the final analysis of this study

-- though 92% of the patients were on stimulants, we do not know which ones, whether they changed meds during the course of the study, what dose of meds were used, or how often the meds were actually taken. All of this information would have been useful in interpreting their results

-- there is no information on the prevalence of hypertension in the overall group, nor the level of hypertension, nor the medications used, nor the control, nor the difference in those on stimulants vs not….   And, per the prior blog, it seems likely that hypertension is associated with meds (stimulants and nonstimulants) and is the likely association with later cardiovascular mortality (see https://gmodestmedblogs.blogspot.com/2024/02/adhd-medications-and-cardiovasc-effects.html )

-- there is no information about non-drug-related therapies, including psychotherapy, social engagement, and support. these non-drug interventions are very important. were there different outcomes with effective non-drug interventions either alone or with meds vs just meds?

 

So,

-- a useful large data-mining study finding that treatment of ADHD with medications decreases all-cause (and more specifically “unnatural” causes) of mortality

    -- in women, their results suggested that there was a decreased in “natural causes” only. But there were pretty few “unnatural” causes as compared to men.

-- it is abundantly clear that meds do improve the quality of life pretty dramatically in those with ADHD of any age, an important factor supporting our finding patients with ADHD and treating them pretty aggressively

-- but, that being said, there is a very important role for non-pharmacologic treatments, both as a single modality in some patients, or as an addition to meds

-- this Swedish study and the one in the last blog on ADHD should also be a reminder to follow patients with ADHD on meds closely for hypertension and treat them pretty aggressively (again, with nonpharmacologic and pharmacologic means as appropriate)

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list