Understanding population trends and process UPTAP: Inaugural Conference

1
Understanding population trends and process
(UPTAP) Inaugural Conference

• 30 March 2006
• Claudia Thomas
• Centre for Paediatric Epidemiology
  Biostatistics
• Institute of Child Health
• University College London

2
Mid-Career fellowship

• Start date May 2007
• Duration 2 years
• Supervisor
• Professor Chris Power (ICH)
• Collaborators
• Professor Heather Joshi, Centre for Longitudinal
  Studies, Institute of Education, London
• Professor Stephen Stansfeld, Queen Marys School
  of Medicine and Dentistry, London

3
How does employment affect cardiovascular risk? A
life-course approach in the 1958 cohort

• AIM
• To understand the role of labour market
  participation as a process that leads to social
  inequalities in cardiovascular disease.

4
Background employment and health

• Structure and organisation of paid employment has
  adverse consequences for various health outcomes
• Lack of job security associated with poor
  self-reported health, chronic disease and
  psychological distress (Whitehall II).
• (Virtanen et al. JECH, 2002, 56569 Ferrie et
  al. BMJ, 2001, 322647 Ferrie et al. BMJ, 1995,
  3111264)
• Unemployment related to mortality and
  psychological distress
• (Thomas et al. JECH 2005, 59243 Pensola et al.
  Soc Sci Med, 2004, 582149 Murphy Athanasou, J
  Occ Org Psych, 1999, 7283)
• Combination of work and motherhood (role
  overload) has adverse consequences for health.
• (Weatherall et al. Soc Sci Med, 1994, 38285)

5
Background employment and CVD

• Most studies have looked at psycho-social affects
  of work on cardiovascular disease
• Job strain, job demands, decision latitude
• (Agardh et al. Diabetes Care, 2003, 26719 Kuper
  et al. JECH, 2003, 57147)
• Less is known about aspects of employment on
  cardiovascular disease
• Evidence for relationship with lower
  socio-economic position
• (Lawlor et al. Am J Pub Hlth, 2005, 9591
  Feldman Steptoe, Hlth Psych, 2003, 22398)
• Relationship with shift-work
• (Knutsson Boggild, Rev Environ Hlth, 2000,
  15359 Boggild Knutsson, Scand J Work
  Environ Hlth, 1999, 2585)

6
Objectives

• To study the direct relationships between
  employment characteristics and adult
  cardiovascular risk markers.
• To study the indirect pathways diet, physical
  activity, weight gain.
• Take into account social processes occurring
  earlier in life, such as education, that
  determine how individuals arrive in their
  occupational destinations.
• Understand how the more complex patterns of
  labour market participation experienced by women,
  such as, the combined roles of work and
  motherhood, are related to cardiovascular disease.

7
Methods

• Dataset 1958 British birth cohort
• Outcomes cardiovascular markers measured at age
  45 years (BP, HbA1c, triglycerides, cholesterol,
  waist/hip circumference, BMI)
• Main exposures employment information back to
  age 16y
• Mediators health behaviours in adulthood
  (smoking, alcohol, diet, physical activity)
  adiposity throughout the lifecourse
• Other covariates/confounders SEP in childhood
  and adulthood, early environment (birth weight,
  health in childhood/adolescence) home
  circumstances in adulthood, education, region of
  residence.

8
Methods Analyses

• Direct relationship between labour market
  participation and cardiovascular measures
• Cross-sectional and longitudinal associations
• Whether observed associations are explained by
  earlier life factors (neonatal circumstances,
  education, childhood health, parents SEP)
• Indirect associations between labour market
  participation and cardiovascular measures
• Mediators weight change, health behaviours
• Statistical methods linear and logistic
  regression multilevel modelling structural
  equations (pathway analysis)

9
1958 British birth cohort (NCDS)

• Perinatal Mortality Survey
• All live births one week in March 1958
• England, Scotland and Wales
• 17000
• Followed up
• age 7y (1965)
• 11y (1969)
• 16y (1974)
• 23y (1981)
• 33y (1991)
• 42y (2000)
• 45y (2003) biomedical survey

10
1958 cohort biomedical survey

• At age 45 years
• MRC Health of the Public grant (Prof C Power,
  Prof D Strachan)
• Nurse interviews
• Physical measurements, e.g. height, weight, BP
• Blood collection
• Nurse administered questionnaire (CAPI)
• Self completed CAPI section sensitive
  information

11
1958 cohort biomedical sample
920 immigrants added NCDS1-NCDS3
6489 cases not in biomedical sample 1196
ineligible - dead 1236 ineligible - living
outside GB 1041 permanent refusals 31 NCDS6 proxy
respondents 2985 not issued for other reasons
93 ineligible cases (28 dead, 65 living outside
GB) 1804 refusals (includes 1 case whose data was
withdrawn after interview) 697 non-contacts 98
other unproductives
Liz Fuller, 2006
12
1958 cohort employment patterns

• Historical context
• Cohort entered labour market in mid-1970s to
  early 1980s
• Unemployment had started to rise, subsequent
  periods of boom and recession
• Womens participation had also increased
  part-time, low level and low-paying jobs
• During cohort members working lives,
  reorganisation and restructuring of work eg
  increased computer use

13
1958 cohort economic activity
Source Changing Britain, Changing Lives. E.
Ferri, J. Bynner, M. Wadsworth, 2003
14
1958 cohort employment characteristics

• Occupation
• Grade (Registrar general)
• Hours of work
• At 33y, 50 of male employees compared to 11 of
  female employees worked more than 40 hours per
  week
• Night work done more frequently by men than women
• Paid versus self-employment
• Approximately 14 self-employed (more men than
  women)

15
1958 cohort work and health
Key publications on labour participation and
health in the cohort to date Llena-Nozal et al.
The effect of work on mental health does
occupation matter? Health Economics, 2004, 13
1045-1062 Power et al. Childhood and adulthood
risk factors for socio-economic differentials in
psychological distress evidence from the 1958
British birth cohort. Social Science Medicine,
2002, 55 1989-2004 Matthews Power.
Socio-economic gradients in psychological
distress a focus on women, social roles and
work-home characteristics. Social Science
Medicine, 2002, 54 799-810 Matthews et al.
Gender, work roles and psychosocial work
characteristics as determinants of health. Social
Science Medicine, 1998, 46 1417-1424 Montgomer
y et al. Health and social precursors of
unemployment in young men in Great Britain. JECH,
1996, 50 415-422 Joshi, et al. Employment after
childbearing and womens subsequent labour force
participation Evidence from the 1958 birth
cohort. J Population Economics, 1996, 9325-348
16
Acknowledgements

• ESRC UPTAP secondary data analysis initiative for
  provision of Mid-Career Fellowship
• MRC funding of biomedical survey
• Centre for Longitudinal Studies management of
  the1958 cohort co-ordination of follow-up
  surveys data collection and provision of data.