Increased mortality rate in ADHD: effect of age when diagnosed and comorbidity

Introduction

A study just published online in the prestigious medical journal Lancet (Dalsgaard et al, 2015) has for first time obtained substantive evidence of an increased mortality rate in persons with attention deficit hyperactivity disorder (ADHD). The study by Dalsgaard and his colleagues at Aarhus University in Denmark and Yale University in the States also provides evidence that the increase in death rate is caused mostly by coexisting conditions such as antisocial and addictive behaviour, in particular the latter.  In addition, the researchers found that if the diagnosis of ADHD is delayed until one reaches adulthood the mortality rate is increased.

Although the death rate in individuals with ADHD is approximately double what it is in individuals without ADHD, it is important not to over dramatize this result by realizing that the death rate in the non-ADHD population is relatively low and so doubling this low rate still results in a relatively low absolute number of deaths.  But at the same time, from a public health point of view, the increased mortality rate in ADHD is surprisingly large and along with the other conclusions reached in this study have a number of significant implications for public health policy and how mental health professionals deal with ADHD.

Brief description of how the study was carried out

The Dalsgaard study employed an exceptionally large sample size of 1.92 million individuals, 32,061 of whom had received a diagnosis of ADHD.  Such a large number of participants was made possible by the fact that all people living in Denmark or by law required to register in the Danish Civil registration System and the government keeps a huge database of medical information, including ADHD status. From this database the researchers tracked all children born between 1981 and 2011, with a maximum span of 32 years. As well the researchers had access to the Danish Cause of Death Register which enabled them to compare rates of mortality for those with and without ADHD.

The authors calculate mortality in two main ways: mortality rate per 10,000 person-years and mortality rate ratios (MRR) calculated by dividing the mortality rate for ADHD individuals by that for non-ADHD individuals. Hence an MRR of 2 means the morality rate is twice that of the control group.  In the statistical analyses they also, where appropriate adjusted for various possible confounding factors eg gender, the existence of comorbid conditions.

Results and implications

Mortality increased in ADHD

An analysis of the raw data, indicated that the "all-cause" mortality rate per 10,000 person-years among the control group of non-ADHD individuals was 2.21 but increased to 5.85 in the ADHD sample, being an increase by a factor of 2.64.  Even after statistical adjustment for various possible factors (e.g. gender, calendar year, maternal age etc.) the MRR was 2.07 and this result was highly significant statistically (p<.0001).  Hence the mortality in ADHD is greater than the mortality in non-ADHD sample by a whopping 107%.

Interpretation of the increased rate of mortality

This study with its huge sample sizes for the first time presents conclusive evidence that having ADHD puts an individual at increased risk of death  and that the increase is not minor but by approximately two fold. In this regard, it is relevant to quote Dr. Soren Dalsgaard the leading author of this study. "I did expect to see an increase in mortality with ADHD, but it was a surprise to see the difference was this large, and I think most clinicians will also be surprised by the magnitude of increased mortality."  However he went on to say "It is important to reassure patients with ADHD and their families that although ADHD is associated with increased mortality, the absolute risk of death was quite low."  

Dalsgaard also is quoted as saying that the study results indicate that "ADHD … is not as benign a disorder as some may believe." (Medscape Medical News.)  And, as Faraone (2015) has written “For too long, the validity of ADHD as a medical disorder has been challenged.” (Faraone, 2005).  This new and very substantive study with its reliable results should lay such challenges to rest.  It should also signal to public health policy makers that ADHD not only is a legitimate psychiatric condition but it is one with potentially lethal consequences.

"ADHD can no longer be dismissed as a benign disorder unworthy of serious attention by mental health professionals and public health officials."

Causes of increased mortality

Having shown that the mortality rate in those with ADHD is at least twice that in those without this condition, Dalsgaard and his colleagues went on to investigate possible three possible causes of the increase:

1. time elapsed between diagnosis and death
2. the age of individual when the diagnosis was made
3. comorbidity which is the presence in addition to ADHD of conditions frequently associated with ADHD.

Time elapsed between diagnosis and death

The researchers found the crude MRR within the first year of diagnosis, was 2.68; at 1-2 years, 2.45; 3-4 years, 3.75;, 5-9 years, 2.01;  and 10 years and greater after ADHD diagnosis, it was 1.90. The authors concluded "The risk seemed to be highest within the first 5 years of diagnosis of ADHD, but even after more than 10 years of diagnosis, the mortality rate was still almost doubled (1.90)."  Since one might expect treatment to be given after diagnosis, one might also expect the mortality rate to decrease due to the beneficial effects of treatment. While it does,  the effect did not reach statistical significance. This is potentially a very significant factor and deserves further research.

Age when the diagnosis was made

The MRR was 1.86 for diagnosis at age 0-6 years; 1.58 for diagnosis at 6-17; and highest for adults being 4.25 with the increases in MRR being statistically significant for at all ages groupings.  

Evidence supporting early diagnosis

These results suggest that diagnosis be made as soon as possible in order to avoid the increased mortality found when diagnosis delayed until adulthood.

Evidence supporting increased help is required for adults with ADHD with a focus on comorbidity

The high MRR of 4.25 found in adults represents a 2.69 fold increase over the MRR found in adolescence. This indicates an urgent need to provide help to newly diagnosed adults with ADHD

Comorbidity

The next question the Dalsgaard group asked was whether the observed increase in death rate was due to the presence of ADHD disorder itself or due to any other psychiatric disorders that persons with ADHD in general are known to have at a higher rate than in persons without ADHD. The researchers were fortunate in that the Danish records recorded, not only the occurrence of ADHD, but also a variety of other psychiatric problems including the two anti-social disorders of oppositionally defiant disorder and conduct disorder as well as substance use disorder.  The presence of this information allowed the researchers to test statistically for the effect of various combinations of these disorder as listed below: 

• For cases of ADHD without any recorded comorbidity (called hereafter “pure” ADHD), the MRR was 1.50 and was highly statistically significant. This indicated that even without coexisting antisocial or addiction disorder the rate of mortality was still increased by 50%. 
• For cases of ADHD with coexisting oppositionally defiant disorder or conduct disorder, the MRR increased to 2.17.
• For cases of ADHD with coexisting substance use disorder, the MRR increased to 5.63.
• For cases of ADHD with an antisocial disorder and a substance use disorder, the MRR increased to a shocking 8.29.

Does ADHD alone without comorbidity cause increased mortality rate?

As mentioned above, the researchers found an increased MRR of 1.50 even in their sample of "pure" ADHD. However in estimating the MRR for "pure" ADHD, only oppositional defiant behaviour, conduct disorder, and substance use disorder were used as comorbid conditions.  Yet there is evidence of a long list of other possible comorbid conditions in ADHD. One of these is depression which of course carries increased risk of death due to suicide. 

Consequently their condition of "pure" ADHD likely contained some persons with coexisting depression and this might have increased the MRR.  Other possible causes of the increased MRR in "pure" ADHD is that the presence of bipolar disorder or a biological/genetic predisposition to ADHD which causes general medical problems leading to death. 

More likely is the possibility that the stresses which having ADHD causes, eg attentional problems and impulsivity, lead to general health problems and morbidity. Therefore more research is required to determine if "pure" ADHD causes increased mortality.

Substance use disorder and ADHD

By far, the comorbidity causing the greatest increase in mortality rate was substance use disorder with an MRR of 5.63 which is 2.59 time greater than the MRR associated with the antisocial disorders. Moreover, the potential lethality of these two main comorbid conditions somehow are additive resulting in the shocking MRR of 8.29 in the sample having ADHD, antisocial disorder and substance use disorder.

These results provide strong support for attempting to routinely assess for these comorbid conditions whenever ADHD itself if being assessed.  However, because of the stigmatization associated with antisocial and addictive behaviours, great care and sensitivity must be exercised in how and when this is carried out.  A motivational interview approach and psychoeducation about comorbidity effects might facilitate the clients' participation in the further assessment.

Discussion

Limitations of this study

Faraone (2015) writes "Although no study can be definitive, this one comes close".  However, although the sample size of 1.92 individuals was huge and unprecedented, still only 107 individuals with ADHD were found to have died during the 32 year follow up period of this study.  So it is important to realize that all the conclusions about mortality rates in this study are based on this final, relatively small, subsample. 

General implications of this study

The first and the core conclusion of this study by Dalsgaard and his colleagues is that individuals with ADHD die approximately twice as frequently as individual without ADHD.  Therefore the study provides sobering but strong evidence that ADHD can no longer be dismissed as a benign disorder unworthy of serious attention by mental health professionals and public health officials. These research results are relatively simple and clear, yet speak to a need for a reappraisal of the seriousness of ADHD.

A second main conclusion is that the increased rate of mortality in ADHD seems mainly due to comorbid conditions, such as oppositionally defiant behaviour, conduct disorder, and substance use disorder.  Fortunately there are proven therapeutic techniques eg cognitive behaviour therapy, relapse prevention etc which can help many people to recover from these comorbid conditions, thus reducing the mortality rate in ADHD.

Summarizing, this study has provided substantive evidence that ADHD not only is a legitimate psychiatric condition but it is one with potentially serious consequences if not assessed early in life and treated.  In particular those with anti-social behaviour problems and addictions should have access to a vast option of treatment facilities.

References

Dalsgaard, S., Øtergaard, S. D., Leckman, J. F., Mortensen, P. B., & Pedersen, M. G. (2015). Mortality in children, adolescents, and adults with attention deficit hyperactivity disorder: a nationwide cohort study. The Lancet.

Faraone, S. V. (2005). The scientific foundation for understanding attention deficit hyperactivity disorder as a valid psychiatric disorder. European Child and Adolescent Psychiatry, 14: 1-10.

Faraone, S. V. (2015). Attention deficit hyperactivity disorder and premature death. The Lancet.

Copyright © 2015  Brian S. Scott