TL; DR – Loneliness and Social Isolation as a Risk Factor for Mortality

Holt-Lunstad, J., Smith, T., Baker, M., Harris, T. and Stephenson, D. (2015). Loneliness and Social Isolation as Risk Factors for Mortality. Perspectives on Psychological Science, 10(2), pp.227-237.



A meta-analysis of studies into the effect of social factors on mortality, with a focus on social isolation. This area has received less interest in research and literature than other factors such as smoking, obesity, and sedentary lifestyles. There have been some studies which suggest that social connections have an impact on mental health, emotional well-being, and physical health.



70 studies were included in the meta-analysis, with participants with a mean age of 66 years at initial contact and a mean length of 7 years as a follow-up. The total sample size was 3,407,134. 63% of the studies involve standard community samples, 37% involved patients with medical conditions.

Measures such as the Social Isolation Scale, the Social Network Index, and Loneliness Scales were used to give measurements on social connectedness and social isolation.

Studies where mortality was the result of suicide or accident, were excluded from this meta-analysis.



  • Social Isolation, loneliness, and living alone resulted in a higher likelihood of mortality, regardless of whether the measurement was objective or subjective.
  • The statistics reported for increased risk of death were:
    • 26% for reported loneliness
    • 29% for social isolation
    • 32% for living alone
  • Those who were social isolated, lonely, or living alone were more likely to be dead by the time of the follow-up than their counterparts, regardless of age or socioeconomic status.
  • Middle-aged adults were at greater risk of mortality when lonely or living alone than any other age group in the meta-analysis.
  • Fully adjusted models which accounted for health status, or studies which did not include physically ill individuals still found social isolation and loneliness to be predictive of mortality.



The meta-analysis could not confirm causality.

This meta-analysis could not identify any “threshold” for social isolation at which increased mortality risk occurred – given the complexity and individuality of the matter; it is unlikely this threshold could accurately be identified even in future research.

Most of the studies online looked at one of the three factors: social isolation, loneliness, or living alone. Therefore it was not possible to identify one of the three as a greater risk factor than the others.

91% of the studies involved people younger than 50 years of age – future research should include participants from a broader range of age groups.


Why is this relevant to autism? How is it relevant to practice?

The National Autistic Society surveys regularly find that autistic people report being lonely and isolated more than the general population. Some autistic people like to be on their own for long periods of time, others are expected to feel like that because of the stereotypes around autism.

Assumptions should not be made about what friendship or socialising means to an autistic person. Try to find out what they want out of friendship and see if that is possible, or if a compromise can be made.

Being involved in the community can help reduce isolation – too often people with profound or severe learning difficulties are not involved in their community at all.


TL;DR – Gender and Age Differences in the Core Triad of Impairments in Autism Spectrum Disorders

Reference: Van Wijngaarden-Cremers, P., van Eeten, E., Groen, W., Van Deurzen, P., Oosterling, I. and Van der Gaag, R. (2013). Gender and Age Differences in the Core Triad of Impairments in Autism Spectrum Disorders: A Systematic Review and Meta-analysis. Journal of Autism and Developmental Disorders, 44(3), pp.627-635.



A meta-analysis of studies into the effect of gender on autism. Current findings on whether there is a gender difference are inconsistent. To investigate further, the meta-analysis also looked at the effect of age in conjunction with gender.



The meta-analysis looked at 22 studies which reported on gender differences in social and/or communication deficits, and differences in repetitive and stereotyped behaviour. Due to the variation in the studies the following data was used:

  • For social deficits, there were 4,783 test scores from males and 1,277 test score from females
  • For communication, there were 2,781 scores from males and 992 for females
  • For repetitive and stereotyped behaviour there were 2,093 scores from males and 781 scores from females.



  • The meta-analysis found very few differences in symptom severity between males and female.
  • Symptom severity in communication and social behaviour was similar between genders, girls showed less restricted interests, behaviours and stereotypes than boys. This difference was not significant below the age of 6, only after.
  • There are three hypotheses for this difference:
    1. Females present with a different autism phenotype to males
    2. Female ASD patients with Intellectual Disability may be over-represented in the studies used, and stereotyped and repetitive behaviours are not ASD-specific – they also present in ID.
    3. Females with less severe social and communication deficits were misdiagnosed with other conditions and therefore not included in these studies.



May have been false-positive ASDs among patients with restricted interests and behaviours and stereotypes as there was an over identification of these symptoms as autistic traits when they are also common in some regularly developing children, and those with Intellectual Disability or who experienced deprivation.

Females with normal to high intelligence may have been missed from the meta-analysis.

Intellectual Disability as a confounder could not be controlled for with the data available.


What to take from this for practical use?

Girls and women may present differently in terms of social and communication difficulties.

Girls with normal or above intelligence may exhibit less repetitive behaviours, it may be more difficult to identify ASD traits.

Repetitive behaviours and interests alone are not ASD, they are equally common in other conditions and in some typically developing children.

Even if there are some presentation differences, women and girls with ASD do still have (and still require for diagnosis) social and communication difficulties, and repetitive interests, behaviours or thoughts.

TL;DR – Meeting the Common Core State Standards for Students with Autism

So many journal articles – once published – just sit on a shelf and never get read by the very people who might benefit from any practical information gained by the research. That’s without thinking about the additional barrier often posed by overly complicated language or terminology. Research needs to be done for a purpose, and that information needs to be disseminated. So this is my attempt at doing a bit of that.

Reference: Constable, S., Grossi, B., Moniz, A., & Ryan, L. (2013). ‘Meeting the Common Core State Standards for Students with Autism: The Challenge for Educators’, Teaching Exceptional Children, 45(3), 6-13.

Country of publication/research: USA

Topic: Adaptations which can help autistic students meet the Common Core State Standards. These are details about what students should know in English Language Arts and Mathematics by the end of K-12.

Summary of content:

  • The adjustments to the Elementary and Secondary Education Act (now called the Every Student Succeeds Act) means that many students with SEND in mainstream classrooms will be expected to meet the Common Core State Standards.
  • If teachers understand the three main psychological theories of autism (Theory of Mind, Executive Dysfunction, and Weak Central Coherence), they might be able to improve their teaching of these standards and personalise them to autistic students..
  • Includes brief discussions on how the three theories might cause difficulties in the area of English – with a focus on reading – and gives examples of students and difficulties involving these.
    • Theory of mind: May find it difficult to interpret the actions of others, may not understand the impact of own actions on others, may find it difficult to interpret thoughts, feelings and intentions of characters.
    • Executive dysfunction: May have difficulty in planning, organisation, initiation, working memory, inhibition or impulse control, time management, and developing and using new strategies.
    • Weak Central Coherence: May focus on a single aspect of something (a place, a topic, a piece of text, a picture, a documentary etc.) to the exclusion of other relevant details.
  • Gives examples of how these areas of difficulty might be supported:
    • Social Narratives (Social Stories and Comic Strip Conversations)
    • Naturalistic Interventions (teaching skills in the daily routine and using interests as motivation)
    • Peer-mediated instruction and intervention (using peers to explain and model behaviour and new skills)
    • Visual supports (such as a graphic organiser)
    • Prompting.

Biases or issues with method:

Rigid view of autism – makes sweeping generalisations such as autistic children “don’t understand how actions impact other people”.

Only considers autistic students who are already accessing the Curriculum with their mainstream peers.

What to take from this for practical use:

Social Narratives (google: Social Stories and Comic Strip Conversations) can help with developing Theory of Mind and understanding.

Visual Supports can and should be used to support executive dysfunction.

Don’t forget to ensure that peers are helping each other – overreliance on an adult is not of benefit to anyone.