Nine modifiable risk factors for dementia

Dementia is the greatest challenge for health and social care in this century. What can individuals and societies do to prevent dementia ?  

The Lancet commission on Dementia reviewed this question in its report published this week in Lancet ( online , July 20). The word dementia has demeaning connotations and DSM has replaced it with the phrase ‘ major neuro cognitive disorders”. The present report uses the word dementia to cover all types of dementing disorders.

Dementia Decreasing? :  Some countries have noticed an age specific decrease in dementia prevalence/ incidence. This is attributed to an increase in education.  However, many societies would see increase in dementia due to increasing mid life rates of obesity and associated ill health.

We know that vascular / metabolic factors , diet, life style, education and mental health are  the key modifiable risk factors here. However, the greatest risk factor, age is unmodifiable.

Cognitive reserve: Those with more brain reserve can tolerate more neuropathology without much cognitive and functional decline. At population level, access to reserve enhancing factors ( like physical exercise, intellectual stimulation, leisure activities ) over  life span can reduce the risk of dementia.  Hypertension can reduce the cognitive reserve and thus increase the risk of dementia. Cognitive resilience in later life is associated with healthier life styles, higher education and improved socio economic status during early childhood.

Population Attributable Risk 

This review identified 9 modifiable risk factors. 35% of dementia is attributable to these factors. For example: We can see that if no one in above 65 age group is smoking, this will eliminate 5.5% of dementia in the population.

Risk factor Relative Risk (RR)  Pop. Attributable Factor (PAF)  (Weighted)
1.Less education 1.6 7.5%
2.Hypertension 1.6 2%
3.Obesity 1.6 0.8%
4.Hearing Loss 1.9 9.1%
5.Smoking 1.6 5.5%
6.Depression 1.9 4%
7.Physical inactivity 1.4 2.6%
8.Social isolation 1.6% 2.3%
9.Diabetes 1.5% 1.2%

Factors 2, 3 and 4 are Midlife ( 45-65) risk factors . Factors 5,6,7,8,9 are later life risk factors. To see the PAF in perspective, you may want to see how much reduction in incidence of dementia will happen if   apolipoprotein E4 allele is completely eliminated. This would bring  7% reduction at population level. If every one had good education (up to secondary level), this would bring the same population benefit.

One interesting finding is the risk due to hearing loss.The RR is highest for this factor. It is estimated that one third of all aged above 55 have some degree of hearing loss.  Microvascular pathology may be a confounding factor here. Whether hearing corrections reduce this risk is not known at this moment.

What are the evidence based  prevention strategies? 

1.Treating hypertension is shown to reduce the risk of dementia

2.NSAID and statins do not reduce the risk. HRT is also not recommended for this benefit.

3. Mediterranean Diet: Possible positive effect . Reduce cognitive ageing.

4. Cognitive interventions: Engaging in cognitively stimulating activities can improve cognition and reduce dementia

5. Physical exercise improves cognition, but evidence to reduce dementia is lacking

6. Not much is known about the effect  of social activities on risk reduction.

What about high risk group interventions?

The FINGER study was a mammoth undertaking to provide 4 intensive life style based interventions (diet, exercise, cognitive training, vascular management ) to 600 participants over 2 years. General cognition did improve , but there was no difference in memory when compared with controls. PreDIVA study ( Netherlands) showed that vascular risk factor management for elderly  did not reduce the dementia incidence over 6 year period.

The current evidence base (from clinical trails)  is unclear  regarding  specific interventions at population level that would reduce the dementia risk. Despite this, risk factor understanding  is pointing towards the need to implement  safe interventions that confer general  health benefits. Increasing education in early life, increasing physical activity and social engagement, reducing smoking, treating hypertension , depression, diabetes, and hearing impairment should be our priorities.

Summary of the article

Dementia prevention, intervention, and care.

Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, Samus Q, Schneider LS, Selbæk G, Teri L, Mukadam N.

Lancet. 2017 Jul 19. pii: S0140-6736(17)31363-6. doi: 10.1016/S0140-6736(17)31363-6. [Epub ahead of print] Review



institute for mind and brain ( inmind hospital ), Thrissur, Kerala.


Early deprivation and mental health outcomes? too late?

April 18, 2017

It is generally thought that early social and emotional deprivation have pernicious and persistent negative effects on mental health . Strong supporting evidence comes from animal studies (Mela V, 2015). Studying this in children is not at all easy. Observational studies have serious limitations. RCTs are impossible and unethical.

Sonuga-Barke and team from the brain behaviour lab at Southampton University, UK report the results of a unique study. The English and Romanian Adoptees study  is so unique in that it is a very rare natural experiment of early life severe deprivation.  165 children were adopted from depriving institutions in Romania in 1990s (immediately after the fall and execution of Nicolae Ceaușescu) to stable, supporting families in UK.  Deprivation was extreme and global. There were two subgroups in the adopted children, those who spent less than 6 months in an institution and those who spent more than 6 months ( up to 43 months).  The comparison group was of UK adoptees who did not experience any such deprivation. Over the years, the participants were comprehensively assessed at ages 6,11,15 and between 22-25 years. A wide range of behavioural , social and cognitive measures were studied.

Results: Children severely deprived up to 6 months and non deprived  UK adoptees had similarly low levels of problems. The group that suffered more than 6 months of deprivation had multiple problems:- autism spectrum disorder, disinhibited social engagement, inattention and overactivity, and this was seen in adult hood. Compared with UK adoptees group, they also had more  low educational achievement, unemployment, and mental health service use. Interestingly, the higher rates of cognitive impairment in high deprivation group ( at age 6 and 11) remitted to normal by adult age. Emotional problems showed a late on set pattern and an increase over time in  the prolonged deprivation group.

Disappearance of deprivation related cognitive problems  suggest neuroplasticity / remediation possibility even in prolonged early deprivation . Emotional problems emerged later and in young adulthood this was 3-4 times higher than other groups. This may be due to stress induced vulnerability that expresses only later. This could also be a direct developmental consequence of deprivation. Inappropriate social engagement with strangers , considered usually as a childhood limited feature, continued to manifest in to adulthood. The aberrant social communication and obsessive behaviours ( quasi autism) , also continued to adulthood.


Not much information is available on pre institutional risk factors.  However, It is unlikely that children who spent more time in institutions were exposed more to such factors than those who spent less time.


The problems seen in this deprived group, could they be adaptive behaviours? (i.e.  disinhibited social engagement could be adaptive where caregiver continuity is absent). The fact that emotional problems have emerged later, suggest that at least some  such behaviours/ outcomes are not adaptive ? A positive finding: A fifth of adoptees with longer period of deprivation were problem free after age 6, suggesting high degree of resilience as well. We don’t know what that constitutes in biological/ psychological terms from this study.

Earlier this group has reported very high prevalence of ADHD in high deprived group, this was 4 times higher in adolescence ( age 15) and 7 times higher in adult hood(22-25 yrs). The ADHD types was complex with distinctive features. It was equally common in men and women , was predominantly inattentive type.This was comorbid with features of Autistic SD, Conduct D and Emotional callousness. ( Kennedy et al 2016).


Early life years are crucial in adult mental health. Supportive , caring family environments can offset some of the effects of early social emotional deprivation. High risk for adult mental health problems ( emotional and social engagement related problems)  remain in those who were deprived for longer periods. One way to reduce the burden of  mental health problems is to ensure  safe and supportive early years for every child.

In the accompanying comment  Frank C Verhulst reminds us of how John Bowlby warned the UK Govt in 1939 ( in BMJ) about the practice of evacuating children ti unfamiliar families ( for safety reasons) during world war 2:: “Sir—The evacuation of small children between the ages of 2 and 5 introduces major psychological problems.”

Common sense? But then these separations are continuing to happen worldwide. Orphanages are still around us!


Summary of the article

Child-to-adult neurodevelopmental and mental health trajectories after early life deprivation: the young adultfollow-up of the longitudinal English and Romanian Adoptees study.

Sonuga-Barke EJ, Kennedy M, Kumsta R, Knights N, Golm D, Rutter M, Maughan B, Schlotz W, Kreppner J.Lancet. 2017 Feb 22. pii: S0140-6736(17)30045-4. doi: 10.1016/S0140-6736(17)30045-4. [Epub ahead of print]


Mela V, Diaz F, Borcel E, Argente J, Chowen JA, Viveros MP.
Long term hippocampal and cortical changes induced by maternal deprivation and neonatal leptin treatment in male and female rats. PLoS One 2015; 10: 1–25

Early severe institutional deprivation is associated with a persistent variant of adult attention-deficit/hyperactivity disorder: clinical presentation, developmental continuities and life circumstances in the English and RomanianAdoptees study.

Kennedy M, Kreppner J, Knights N, Kumsta R, Maughan B, Golm D, Rutter M, Schlotz W, Sonuga-Barke EJ. J Child Psychol Psychiatry. 2016 Oct;57(10):1113-25. doi: 10.1111/jcpp.12576.

Manoj Kumar MD FRCPsych, Inmind, Thrissur, Kerala.


Can we grade social cognition defects in schizophrenia?

05 10 2016

Social cognition defect is  a core feature of schizophrenia.This include defects in emotion processing, social perception, theory of mind  (mental state attribution) and attributional style/biases. These defects are seen in prodromes, first episode,   while the disease is in remission and in first degree relatives . These are more associated with negative symptoms .Only few studies have looked at the relationship between social cognition defects and functioning. Understanding the social cognition status might help to tailor interventions to suit individual needs.This may improve functional outcomes.

Rocca and colleagues from Italy report the results of a cluster analysis  of 809 individuals with schizophrenia living in the community across Italy. A matched control group was also studied. Symptom severity ( using PANSS), negative symptoms ( using Brief Neg symptoms scale), social cognition ( Emotional intelligence Test, Facial emotion identification test, awareness of social interference test) and functional status ( Specific Level of functioning scale & Validation of real life outcomes) were measured.

Cluster analysis revealed three clusters :  unimpaired, impaired, and very impaired social cognition.Proportions were :unimpaired (42%), impaired (50.4%) and very impaired (7.5%). Functioning decreased parallel to decrease in social cognition. Theory of mind tests  ( to know how the individual understands others mental state) were important for cluster definition.

The awareness of social interference test ( checking sarcasm and lying) was good enough to help grade the three clusters. Patients who were able to detect sarcasm were more likely to be in the unimpaired group. Unimpaired group had higher grasp of sarcasm. Those in impaired group were better able to detect lies than very impaired.

Why does this happen? comprehension of lies/deceit is acquired before sarcasm/ irony. sarcasm involves more complex inferential chains. so if the individual can detect sarcasm , probably he/she has unimpaired social cognition. If they cannot detect deceit/lie, they may have very impaired social cognition.

How do we take this in to practice? The social interference test use video based vignettes. Clinicians need their  own culturally relevant vignettes ( may be movie clips might help ) to test this.

Authors agree that the natural stability of these clusters were not tested in this study- this being a cross sectional study. However, this is the first study using best available tools to measure the relevant constructs.


Once we classify the defects in social cognition what do we do? Would various social skills training help? Would training in virtual settings benefit ? Would improving these measures thorough interventions translate to real benefits in outcomes?… We don’t know…yet.

Summary of article : Social cognition in people with schizophrenia: a cluster-analytic approach.Rocca P, Galderisi S, Rossi A, Bertolino A, Rucci P, Gibertoni D, Montemagni C, Sigaudo M, Mucci A, Bucci P, Acciavatti T, Aguglia E, Amore M, Bellomo A, De Ronchi D, Dell’Osso L, Di Fabio F, Girardi P, Goracci A, Marchesi C, Monteleone P, Niolu C, Pinna F, Roncone R, Sacchetti E, Santonastaso P, Zeppegno P, Maj M; Italian Network for Research on Psychoses.

Psychol Med. 2016 Oct;46(13):2717-29. doi: 10.1017/S0033291716001100.PMID:27649341

Social deprivation leads to psychosis?

05 08 2016

 Does incidence  schizophrenia vary between countries? Faris and  Dunham (1939) showed that disorganised neighbourhoods in Chicago had higher rates of schizophrenia sparking all sorts of explanations for this. WHO study in 1980’s found that incidence of  narrowly defined schizophrenia did not differ across countries, but  broadly defined illness did differ significantly across countries. In 2004, MCgrath et al summarised all relevant data from 161 studies and showed that there is a five fold difference across countries. It is also thought that there are possible significant regional differences  with in countries as well.

What would explain such geographical differences? It is possible that factors like social deprivation, population density, urbanicity, social capital and social fragmentation may have some roles to play. There are theoretical meanings and explanatory models attached to these factors.

Is social deprivation associated with higher incidence of psychosis ? Brian O’Donoghue, Eric Roche and Abbie Lane  looked at association between the level of social deprivation at the neighbourhood level and the incidence of psychotic disorders in this systematic review.They included 28 studies in the analysis.

Three types of places ( place of birth , place when high risk status was positive , place   when schizophrenia emerged)  and their association with illness status was explored  in these studies.Most studied ( 21 studies)  deprivation of the place where illness emerged.


Most studies (17/23)  show that there is an association between incidence of illness and the level of social deprivation at the time of presentation.

Social deprivation at place of birth is less studied keeping the ‘drift theory’ alive.  The largest study on this question ( place of birth ) did not support an association. However, a well designed Israel study ( Werner , 2007) showed that  individuals who later on develop schizophrenia were more likely to be born in deprived areas.

Social causation can occur ( if at all it is happening), through multiple, direct/ indirect channels ( weaker social cohesion, poorer support systems). Established risk factors ( family history ,cannabis use, traumatic experiences )  can be more prevalent in such areas. They may also have more residential mobility in childhood which is now shown to be another possible risk factor.We do not know how much  gene- environment interaction  may be explaining these observed associations. The  EU-GEI study  ( European Genetic Interaction Study), which is now established,  is likely to offer some insights .

How would we know more about environmental effects ?  This review suggest that more studies on UHR/ psychotic symptoms in general population are needed to  identify vulnerability periods and associated environmental risk factors.

Understanding these relationships have some practical , immediate relevance when we allocate resources , for example for early intervention services . Psychiatric Mapping Translated into Innovation for Care (PsyMaptic) is an example of a resource that predict the incidence in a geographical area to inform resource allocation. However, in countries where basic mental health data is lacking, these are distant dreams.

Summary of the article.

Neighbourhood level social deprivation and the risk of psychotic disorders: a systematic review.

O’Donoghue B, Roche E, Lane A.

Soc Psychiatry Psychiatr Epidemiol. 2016 Jul;51(7):941-50. doi: 10.1007/s00127-016-1233-4. Epub 2016 May 13. Review.


Youth with psychotic experiences have brain tissue loss

05 01 2016

Psychotic disorders like schizophrenia have long been established to have accompanying changes in brain structure and function. Hippocampal volume loss is the most significant and consistent of such changes. Medial temporal lobe seems to be the area where most changes occur. Volume reductions seen in unaffected relatives make us support the idea of heritable phenotypes.

Youth who are classified as high risk,  who later convert to psychosis ,also have  (North American Prodrome Longitudinal Study consortium , 2015 ) accelerated gray matter loss in frontal cortex compared with non- converters and healthy comparators. what about those who experience some psychotic spectrum experiences?

Do non-help seeking youth who experience psychotic spectrum symptoms have structural changes in brain? Transition to psychosis in this group will be lower than the high risk groups. Large imaging studies haven’t addressed this question so far.Philadelphia Neu- rodevelopmental Cohort (PNC) is trying to answer this question. Theodore D. Satterthwaite and team from University of Pennsylvania Perelman School of Medicine report the findings of their study.

The cohort comprises 1429 individuals ( mean age around 15) . PS symptoms were present in 408 participants. This was assessed using the GOASSESS interview ( combination of questions from K SADS, PRIME, Scale of prodromal symptoms).Analysis is based on the final sample of 391 youth with Psychotic spectrum  features (PS group) and 400 Typically Developing youth (TD group) 8 to 22 years old.


Youth with Psychotic spectrum  symptoms had reduced intra cranial volume as compared with typically developing youth. They showed marked reduction in gray matter volume. Difference was more in older participants. Volume loss was more marked in medial temporal lobe.Greater severity of PS symptoms was associated with volume reduction in bilateral medial temporal lobe.

It is important to note that this is  a community based study and participants were non help seeking. Still, the changes  mirrors that seen in both adults with clinically diagnosed psychotic disorders and youth at clinical risk. One could say that regions affected are part of the default mode network. (large functional network critical for internally directed attention, theory of mind, social cognition, and memory).


Adolescents with psychotic spectrum experiences have identifiable brain structural changes. These are similar to those observed in established psychotic disorders.imaging phenotypes are important in evaluating risk of conversion to psychosis. This may ,in future, help in early identification and intervention.

summary of the article :

Structural Brain Abnormalities in Youth With Psychosis Spectrum Symptoms. Satterthwaite TD, Wolf DH, Calkins ME, Vandekar SN, Erus G, Ruparel K, Roalf DR, Linn KA, Elliott MA, Moore TM, Hakonarson H, Shinohara RT, Davatzikos C, Gur RC, Gur RE.JAMA Psychiatry. 2016 May 1;73(5):515-24.


Do autism risk increase with parental age difference?


Population based studies have shown that older paternal age is linked to Autistic Spectrum Disorders (ASD).Risk is thought to increase with maternal age also.  Are these  ( ie paternal and maternal age) independent risk factors ? Is age difference between parents also important ?

Sandin et al address this important question using combined population based cohorts from various countries set up as International Collaboration for Autism Registry Epidemiology (ICARE). Children were followed from birth to reported diagnosis of ASD. The cohort  comprised of 5766794 births across Denmark, Sweden , Norway, Israel and Western Australia.0.54% of these children received a diagnosis of  ASD and 0.18% had  AD.


Independent effects: Increasing paternal age was associated with more ASD. Relative to fathers aged 20–29 years, fathers 50 years or older had a statistically significantly increased risk for offspring with ASD (RR=1.66 95% CI:1.49–1.85).  Maternal age showed a U shaped relationship. Relative to mothers aged 20–29 years, mothers younger than 20 years had a statistically significantly increased risk for offspring with ASD (RR=1.18 95% CI:1.08–1.29. After 30 years, increasing maternal age  increased risk of ASD.

Data also show evidence to combined effect as well. Risk was highest when both parents were older. Increasing age difference between parents also increased the risk. Lowest risk was shown by couples that generated the majority of births, specifically, 29–39-year-old fathers and 25–35-year-old mothers. Relative risk increased in all directions from this region as the parental age difference increased.

Limitations: We do not have information on confounding factors ( Socio economic status , parental psychiatric history, obstetric complications etc). For example:  Individuals who are shy and aloof  ( may be traits of ASD) ,can be be over represented in older parents)

Conclusions: Increasing paternal age as well as maternal age increase the risk of ASD.Increasing difference in parental age is also a risk factor.

Summary of the article:

Autism risk associated with parental age and with increasing difference in age between the parents.Sandin S, Schendel D, Magnusson P, Hultman C, Surén P, Susser E, Grønborg T, Gissler M, Gunnes N, Gross R, Henning M, Bresnahan M, Sourander A, Hornig M, Carter K, Francis R, Parner E, Leonard H, Rosanoff M, Stoltenberg C, Reichenberg A.Mol Psychiatry. 2016 May;21(5):693-700.


How much do Schizophrenia cost to the society?


Schizophrenia is one of the most disabling disorders. The cost of this illness to individuals and society is enormous. But do we really know how much it costs to the society ?

Cost estimates are calculated by multiplying cost per patient with the prevalence of the illness. Prevalence estimates vary widely. A systematic review calculated  (Saha et al 2005) life time prevalence  as 0.4% and the period prevalence (1–12 months) as 0.33%. cost calculation also involves employment rate of individuals with schizophrenia. This also varies widely. In western societies 12%- 39%  of those with schizophrenia are reported as employed. There are various uncertainties around this data as well.

Evensen et al from Oslo report a unique study that sheds light on the costs of schizophrenia.  This is a  population based study from Norway.  It used a top-down design, where data from national registers for the period January 1 to December 31, 2012 were used to  (aggregate figures of actual  resource consumtption) calculate costs. Direct (treatment and community care) and indirect (ie, social security payments and lost productivity) costs were  also taken in to account.Costs related to premature mortality, volunteer and family care, and the criminal justice system were not included.

They  used  a very comprehensive approach to get all information ( IP care, Op care, GP care, medication use, home care, social benefits ) making this one of the best addressing this issue.


The average 12 month prevalence of schizophrenia was 0.17% for the entire Norwegian population.This would be 0.22% for those aged above 18. Employment rate in general ppulation in teh age group 30-50 is  around 80% compared with around 10% in those with schizophrenia. The employment rate (full time and part time) among those of working age was 10.24%–10.5% of the men and 9.8% of the women.

Total national cost associated with schizophrenia is USD  890 million/ year.The average cost per individual with schizophrenia was USD 106,014.00 . One third of total societal cost is due to hospitalisation. 18 % of all hospital days are due to schizophrenia, and 14% of all OP visits are due to schizophrenia . Lost productivity cost accounted for 29% of the total cost.

Studies from UK and Sweden have showed lesser proportion of expenses attributed to IP care ( 8-16%). Norway has more number of beds than others ( 83 beds/ 100,000) compared to around 50 in UK) and this may be a factor.

This is a unique study which adds to the argument that poor  mental health is a big economic drain on communities world wide.

Comments: In  many countries  individuals are left to their own resources leading to impoverishment , debt and early death.  In countries like Norway, where society take responsibility for mental health services, greater effort and resources are allocated to improve population mental health, provide most effective service and to research best options. Mental health is a serious matter for such welfare societies.

Summary of the article

 Prevalence, Employment Rate, and Cost of Schizophrenia in a High-Income Welfare Society: A Population-Based Study Using Comprehensive Health and Welfare Registers.

Evensen S, Wisløff T, Lystad JU, Bull H, Ueland T, Falkum E. Schizophr Bull. 2016 Mar;42(2):476-83.