K08016; Online publication date 12 February 2009
Received 30 June 2008; accepted 28 November 2008

Kōtuitui: New Zealand Journal of Social Sciences Online, 2009, Vol. 4: 55–70
1177–083X/09/0401–0055  © The Royal Society of New Zealand 2009

PDF file of entire paper: Print-quality (782K)

Kōtuitui

New Zealand Journal of Social Sciences Online

The health status of New Zealand workers: an analysis of the New Zealand Health Survey 2002/03

Megan Pledger¹

Jacqueline Cumming¹

Janet McDonald¹

Michelle Poland^1,2

¹Health Services Research Centre, School of Government
Victoria University of Wellington
PO Box 600
Wellington 6140, New Zealand

²Motu Economic and Public Policy Research Trust
PO Box 24390
Wellington 6142, New Zealand

Abstract The relationship between occupation and health is examined using the New Zealand Health Survey 2002/03. SF-36 domain scores, health service use, and health risk factors were analysed using regression or logistic regression in two ways: (1) in an unadjusted model with occupation alone; or (2) in an adjusted model with occupation, ethnicity, age, sex, and the interactions of age crossed with sex. Significant differences were found between occupations for these SF-36 domains: Role Limitation-Physical; Bodily Pain; and Role Limitation-Emotional in the adjusted model. Significant differences were found between occupations in the adjusted models for some health service use but these tended to be for non-publicly subsidised services. Significant differences were found between occupations for health risk factors, with the proportion of people using drugs and alcohol hazardously tending to be highest in Trade Workers and Plant and Machinery Operators.

Keywords occupation; health; New Zealand

INTRODUCTION

Significant attention is being paid internationally to the role that social and economic factors play in determining well-being. It is well recognised that the determinants of health include social, cultural, and economic factors such as social cohesion, ethnicity, income, employment, education, and housing (National Advisory Committee on Health and Disability 1998). Howden-Chapman & Cram (1998) illustrate this in a model of well-being shown in Fig. 1.

 

Fig. 1 Model of the social and economic determinants of health (from Howden-Chapman & Cram 1998)

 

Aspects of employment make important contributions to health status in this model, through structural features such as low unemployment and safe working conditions with high job control; through sufficient disposable income to afford safe working conditions with high job control; and through psychological coherence from positive future prospects and perceived control.

Much has been written about associations between employment, unemployment, and health, providing information about the relationship between structural features and health status; about the mediating role that particular occupations may have on health; and more recently about associations between types of employment and well-being, with a particular emphasis on job control. However, very little material exists which examines associations between occupation and health using national population surveys. The purpose of this paper is to examine these relationships in New Zealand, using a national population survey—the New Zealand Health Survey 2002/03. The focus of the paper is on differences across occupation groups in New Zealand in health status, health services use, and risk factors.

The paper firstly provides background to the research undertaken here, then sets out details on the data source used in this research and the methods used to analyse the data, before reporting on the findings from the research. The paper ends with a summary of conclusions from the research. Throughout the paper, we talk about the relationship between work and well-being, the latter as short-hand for health and well-being.

BACKGROUND

Research into the relationship between work and well-being has tended to focus on four key sets of issues: (1) relationships between unemployment and health and between changes in employment rates and health over economic cycles; (2) relationships between employment and health; (3) relationships between different types of employment status (e.g., full-time, part-time, or casual) and health; and (4) relationships between particular occupations and health.

This paper falls into the fourth category of research on work and well-being issues. Our focus is on identifying the relationship between occupation and general health status. Therefore, the first key issue identified above—unemployment and health—falls outside the scope of this paper, as do particular occupational diseases such as asbestosis, silicosis, and occupational injury. However, literature on the other three issues (employment, types of employment status, and occupation) is of particular relevance to our research.

Much literature has focused on exploring the relevance of employment to socio-economic differences in health. Work is seen to be socially graded: those with low educational achievements generally work in less favourable jobs (Siegrist & Theorell 2006). More recent literature has focused on exploring psycho-social work environments: in particular, the impact on well-being of psycho-social demands on people working in particular jobs, the control they may have over their work, and the social support they have at work; and in relation to dependency, strategic choice, and over-commitment which may lead to high levels of stress (Siegrist & Theorell 2006).

The literature also suggests that simply being in employment may not be sufficient to lead to improvements in personal health. A systematic review and meta-analysis of 485 studies of job satisfaction and health published since 1970 found low levels of job satisfaction were strongly related to burn-out, poor self-esteem, depression and anxiety, and modestly correlated with subjective physical illness (Faragher et al. 2005). Job satisfaction levels are thus suggested to be an important factor influencing workers’ health.

A number of studies have also begun to explore the possible effects on well-being of different forms of work, such as shift work, being on-call, temporary, and part-time employment. A review of 16 studies of on-call work found it deceases the quality and quantity of workers’ sleep, and may increase stress and decrease mental well-being (Nicol & Botterill 2004). A meta-analysis of 72 studies of job insecurity found this was associated with negative effects on employees’ health, job attitudes, organisational attitudes, and work-related behaviour (Sverke et al. 2002). A review of 27 studies of temporary employment and various health outcomes suggested a relationship between temporary employment and increased psychological morbidity, a higher risk of occupational injuries and lower sickness absence rates than permanent employment (Virtanen et al. 2005). In Australia, Bohle et al. (2004) found that casual shift-workers reported poor sleep, irregular exercise, and poor eating habits. They reported less control over their hours than the full-time employees. Other Australian research has also found that sleeping problems are associated with shift-work for nursing home caregivers (Takahashi et al. 2008), hospital nurses (Dorrian et al. 2008), and freight-haul train drivers (Darwent et al. 2008). Pisarski et al. (2008) developed a model to explain these relationships. They found that control over workload and task autonomy was important in mediating the negative health impacts of hospital nurses employed in shift-work. Team identity, team climate, and supervisor support also all had a significant impact on the health of nurses doing shift-work.

Of particular relevance to our research are studies examining general health status by occupation. Using data from the United States National Health Interview Survey, Caban et al. (2005) found differences in rates of obesity across 41 occupations. Between 1997 and 2002, they found higher rates of self-reported obesity in male motor vehicle operators, police and fire-fighters, and other transportation and material-moving equipment operators; and for females, among motor-vehicle operators, other protective services workers, material-moving equipment operators, and cleaning and building service workers. Lower rates of obesity among women were found among those employed in the health-diagnosing occupations, among architects and surveyors, and among those in the construction trades (Caban et al. 2005).

Lee et al. (2006) used the same survey to explore the relationship between occupation and health across 206 separate occupations. Well-being was measured using responses to questions relating to restricted activity days, work days lost due to illness or injury, doctor visits and hospital stays, broad measures of health status, and acute and chronic conditions. The study found social workers to be the most disadvantaged occupational group, followed by inspectors, testers and graders; postal clerks (though not mail carriers); psychologists; grinding/ abrading/ buffing/ polishing machine operators; nursing aids, orderlies and attendants; specified mechanics and repairers; inspectors/compliance officers (except construction); correctional institutional officers; and licensed practical nurses. Dentists were the least disadvantaged group, followed by pilots, other healthcare professionals, pharmacists, dietitians, driver-sales workers, farm workers, apparel sales workers, sales counter clerks, and tool and die makers.

Turning to New Zealand literature, only a small number of New Zealand studies relating to work and health generally were found.

The Dunedin Multidisciplinary Health and Development Study investigated the relationship between work stress and depression or anxiety among participants at age 32 (Melchior et al. 2007). Adjusting for socio-economic position and negative affectivity or a history of psychiatric disorder prior to entering the labour market, participants exposed to high psychological work demands (such as excessive workload and extreme time pressures) had about a two-fold risk of major depressive disorder or generalised anxiety disorder compared to those with low job demands (Melchior et al. 2007).

In the New Zealand Blood Donors Health Study, 15,687 blood donors who were currently in paid employment were recruited by the Northern Regional Blood Service over 18 months to complete a self-administered questionnaire (Fransen et al. 2006). Working rotating shifts, with or without nights, was significantly associated with work injury, although working permanent nights was not. Work injury was also highly associated with occupation (heavy manual industries and service/sales) and working more than 40 hours per week.

Of most relevance to our research, significant work has been undertaken in New Zealand in developing a New Zealand Socio-economic Index (NZSEI), based on occupation, with the following key groupings:

Level 1 (High): Senior management; higher professional

Level 2: Middle management and professionals

Level 3: Technicians and associate professionals

Level 4: Trades workers; many agricultural occupations

Level 5: Plant/machine operators; some clerical and service

Level 6 (Low): Labourers; unskilled service and clerical workers.

Validation of the index using smoking rates and self-assessed health status produced the expected relationships. Lower NZSEI groups were found to be associated with higher rates of smoking and higher rates of poor health status (except for Level 6 occupations which have better health than Level 5 occupations). No clear relationship was found between the NZSEI groups and the percentage of people visiting a general practitioner in the last year (Davis et al. 1999, 2004).

In summary, unemployment can have a negative effect on health and happiness, while the work environment and the way work is organised can also affect well-being. Few studies have, however, assessed how different types of work may impact on well-being using population surveys, which is the focus of this study.

METHODS

In this paper, we examine relationships between occupation and health in New Zealand, using data from a population health survey, the New Zealand Health Survey 2002/03. Occupation has traditionally been a key measure of socio-economic status, as it is seen to be related to other key determinants of well-being, such as income, education, and social status; it is relatively easily measured and it is linked to the economic, social, and political structure of society (Howden-Chapman & Cram 1998). However, it is less useful for considering the well-being of those not working—that is, those who are unemployed, homemakers, those caring for children or other dependents, those unable to work due to health or other problems, and those who are retired. The use of individual occupation is also considered problematic for some groups who see family/whānau outcomes as more important than individual outcomes (Howden-Chapman & Cram 1998), and where social classifications based on occupation have little meaning, for example, as in traditional Māori society (Durie 1985).

For this study, confidentialised, unit record data from the 2002/03 New Zealand Health Survey were supplied by the Ministry of Health. This dataset contains 12,529 respondents, aged 15 years and older, who were living in a private dwelling in New Zealand (Ministry of Health 2004). The survey over-sampled Māori, Pacific, and Asian peoples and used a complex method of sampling; however, the survey has been weighted to produce a representative sample. Estimates produced by these weights form unbiased estimates of population values. The dataset also includes a set of 100 replicate weights which were created using the delete-a-group method (Ministry of Health 2004). Each of these weights creates an estimate. The variance of these 100 estimates around the unbiased estimate gives the sampling variance of the unbiased estimate. For the purpose of this paper, Sudaan(Research Triangle Institute 2004) was used to do these calculations.

In the survey, respondents were asked whether they had worked in a job or in a family business or farm in the last seven days. If they had, or would have if they had not been absent for some reason, they were asked how many jobs they had (1 or 2+), how many hours they worked in their primary job and all their secondary jobs, and their current occupation. Of the 12,529 people surveyed, 6894 were classified as workers. The occupations were coded into 10 types which included armed forces. The latter group contained few respondents and they were re-coded into administrators/manager category as any armed forces personnel found at a residence were assumed to be administrators and managers rather than active military personnel.

The final nine occupation classifications therefore are:

Administrator/Managers

Professionals

Technicians and Associated Professionals

Clerks

Service and Sales Workers

Agriculture and Fishery Workers

Trade Workers

Plant and Machinery Operators

Labourers and Unskilled Workers

In the survey, respondents’ health status was assessed using version 1 of the Australian and New Zealand version of the SF-36 questionnaire (Ministry of Health 2004). The questionnaire consists of 36 items that are answered by self-report. Each item is given a score according to the answer given. Thirty-five of the items are grouped together to form eight domains. The scores from each item in that domain are put together to give a summary score for that domain. The domains are listed in Table 1 with the interpretation of the domain and score, as well as the abbreviated contents of the items that make up that domain score. A domain score lies between 0 and 100, with 100 representing the highest level of well-being in each domain. The Physical Health Summary Score (PHSS) and the Mental Health Summary Score (MHSS) put together the domains scores to provide an overall indication of physical health and mental health, respectively.

A 36th question, Health Transition, does not belong to a particular domain but measures whether the respondents health is better than a year ago (and given a low score) or whether it is worse (and given a high score).

Respondents were also asked about the main health services they used in the last year (e.g., general practitioner, nurse, complementary and alternative medical practitioners; and, if they visited that service, how many visits they had made in the last year. The main health services and the ones applicable to work, where there were sufficient respondents, are reported on here. The respondents were also asked about their medical insurance status.

There were a series of questions about risk factors and the respondent’s height, weight, and waist were measured. The respondents were also asked basic demographic questions such as age, sex, ethnicity, income, education, and any government benefits they received.

ANALYSIS

The characteristics of the occupational groups are explored and presented in Table 2. Modelling is used to test for significant differences between the occupational groups. The models used in the analyses are (1) regressions, if the dependent variable is continuous or count data, or (2) logistic regressions if the data are dichotomous. While it is more appropriate to model count data using log linear models, such parameters are modelled on the log scale whereas the greater interest is in parameters from the natural scale (i.e., means rather than log means). The dataset is large, so that the Central Limit Theorem applies, which gives the appropriate distributional tests of hypotheses.

We first investigate the relationship between health status, as measured by the SF-36 domains, and occupation. Two models are run using the SF-36 domains as the dependent variable in both models. The first model (referred to as the unadjusted model) has only one explanatory variable, which is occupation. The second model (referred to as the adjusted model) is the same as the unadjusted model, however, it also includes age, sex, ethnicity, and the interaction of age and sex as confounders. The unadjusted model reports on differences in health status by occupation without taking into account demographic composition. The adjusted model reports on health status by occupation after controlling for differences in demographic composition.

We then proceed to use the same models to assess the relationship between health service use and occupation and between health risk factors and occupation. For each set of analyses, a table of results is presented, displaying the unadjusted mean (in the case of SF-36 scores and counts of health service use in the last year) or proportion of respondents (in the case of health service use and health risk factors) from the unadjusted model for each occupation. The table also displays the p-value for the factor "occupation" in both the unadjusted and adjusted models. If the p-value is <0.05 in the F-test for occupation, then it is considered that there is a significant difference between at least two of the occupations.

RESULTS

A summary of respondents’ characteristics by occupation is displayed in Table 2. Service & Sales Workers (33%) and Labourers & Unskilled Workers (27%) have the largest percentage of people in the 15–24 age group, while those working in Agriculture & Fishery (7%) have the largest percentage in the 65+ age group. Women are more likely to make up the bulk of the Clerks (86%) and Service & Sales Workers (63%), while men make up the bulk of Trade Workers (90%) or Plant & Machinery Operators (82%). Māori and Pacific people are over-represented in Trade Workers (18 and 8%, respectively) or Plant & Machinery Operators (19 and 11%, respectively), while European New Zealanders/Others are under-represented in these groups (70 and 66%, respectively).

Twelve percent of people currently employed as Labourers & Unskilled Workers had received the unemployment benefit in the previous year and 6% had been on the sickness or invalids benefit.

Health status

Significant differences were found in average health status by occupational group in both models for the SF-36 domains of Physical Functioning, Bodily Pain, General Health, and Role Limitation-Emotional. The results are displayed in Table 3. In all domains except for health transition, a score of 100 represents perfect health. For the purposes of reporting on the occupations with the highest and lowest scores, all occupations scoring within 0.5 points of another are grouped together.

For Physical Functioning (ability to perform physical activities), Clerks had the worst health score (91.3), while Trade Workers, Technicians & Associated Professionals, and Administrators/Managers had the best health score (94.5, 94.2 and 94.1, respectively).

For Bodily Pain, Agriculture & Fishery Workers, Labourers & Unskilled Workers, and Trade Workers had the worst health scores (74.5, 74.7 and 74.9, respectively), while Technicians & Associated Professionals had the best health score (79.5).

For General Health, Labourers & Unskilled Workers, Clerks, Service & Sales Workers, and Plant & Machinery Operators had the worst health scores (75.9, 76.4, 76.5 and 76.8, respectively), while Administrators/Managers, and Technicians & Associated Professionals had the best health scores (79.9 and 79.6, respectively).

For Role Limitation-Emotional (problems performing day-to-day activities because of emotional problems), Technicians & Associated Professionals and Service & Sales Workers had the worst health scores (90.1 and 90.6), while Plant & Machinery Operators and Trade Workers had the best health scores (95.6 and 95.1).

For Role Limitation-Physical (problems performing day-to-day activities because of physical problems), the effect of occupation was not significant in the unadjusted model but was significant in the adjusted model. The variable that influenced this was sex. Using the adjusted model, we found that the occupation with the best health score was Clerks (90.5) and the effect was quite pronounced. This effect is unobserved in the unadjusted model as clerks are primarily women and women have a worse average health score in this area than men.

For the SF-36 domains of Vitality, Mental Health, and Health Transition, occupation was significant in the unadjusted model but not significant in the adjusted model. The reason for this was differences in frequencies in the demographic characteristics between occupation groups. For Vitality, gender induced the change of significance of occupation, with male respondents reporting stronger vitality than females. For Mental Health, both the age variable and the gender variable were important confounders. Those in the older age group had better mental health than those in the younger age group, and males reported better mental health than females. For Health Transition, age induced the change, with older respondents more likely to report that their health is worse now compared to a year ago.

For Social Function, there is no significant difference between the mean domain scores for each occupation level in either the unadjusted or adjusted model.

Health service use

An additional way of thinking about the relationship between work and the health of the people who work is to look at their use of health services, although the findings are likely to be affected by the cost of accessing services. In 2002/03, those with lower incomes or who used a large number of primary health care services were eligible for government subsidies for primary health care services (around half of the New Zealand population), while public hospital services were (and still are) available free-of-charge to all New Zealanders (Cumming & Gribben 2007).

The health services assessed were general practitioners (GPs), medical specialists, nurses, "complementary or alternative practitioners", physiotherapists, and opticians. Results are displayed in Table 4. Service use was assessed in terms of whether the respondent had used the service in the past 12 months and how many times they had used the service in the last 12 months.

There was no significant difference in use of a GP, medical specialist, nurse, or physiotherapist by occupation. However, Plant & Machinery Operators, Labourers & Unskilled Workers, and Trade Workers were less likely to have visited a physiotherapist in the last 12 months. Similarly, the same occupations along with Agriculture & Fisheries were significantly less likely to have visited an optician in the last 12 months.

Health risk factors

Unadjusted and adjusted models were used to analyse whether different occupational groups had different risk factors. Risk factors assessed were high blood pressure, high blood cholesterol, obesity, waist measurement, eating three or more vegetables a day, eating two or more pieces of fruit per day, physical activity, regular smoking, use of marijuana over the last year, and alcohol consumption.

The risks factor variables of current regular smoking, last year marijuana use, abstaining from alcohol, hazardous consumption of alcohol, low levels of physical activity, and low levels of fruit consumption all show a significant occupation effect in both the adjusted models and unadjusted models. Across these risk factors, the occupation groups having the highest proportions of people engaging in risky behaviours are Plant & Machinery Operators, Trade Workers, and Labourers & Unskilled Workers.

Being underweight, normal weight, overweight, or obese was classified according to people’s Body Mass Index (BMI) value. Plant & Machinery Operators (70%) and Administrators/Managers (64%) were most likely to be in the "overweight or obese" category, but Administrators/Managers were most likely to be classified as obese (27%). Plant & Machinery Operators, being active workers, may be being incorrectly classified as overweight when they are really over-muscled, as BMI is known to misclassify people with higher levels of muscle mass. The top 10% waist measurement was calculated for males and females and compared across occupational groups. There was no significant occupation effect for this variable.

DISCUSSION

There are significant differences in health status across the occupational groups even after taking into account the different age, sex, and ethnic composition of the occupational groups.

Technicians & Associated Professionals generally had the best self-reported health. They had the best health score for the Bodily Pain and General Health domains, and the second best score for the Physical Functioning domain. However, they had the worst Role Limitation-Emotional score. This category is likely to include many occupations that have an emotional component to the work, such as social workers and teachers. Work that involves displays of emotion has been found to create stress and have a negative impact on health (Johnson et al. 2005).

Labourers & Unskilled Workers generally had the worst self-reported health. They had the worst health score for General Health as well as the second worst health scores for Physical Functioning, Role Limitation-Physical, Bodily Pain, and Role Limitation-Emotional. However, this group also had the highest proportion of people who had received an unemployment benefit or sickness or invalids benefit in the last year. Part of their lower score could be attributed to those on these benefits rather than the nature of the job itself. However, other research has also found that manual workers tend to have poorer health than those in other occupations (Lahelma et al. 2005). Furthermore, unskilled work is likely to be associated with other factors known to have a negative impact on health such as low income and low educational qualifications (Howden-Chapman & Tobias 2000).

Of particular interest is the occupational relationship with Physical Functioning and Role Limitations-Physical. Physical Functioning measures an individual’s ability to do physical activities, while Role Limitation-Physical measures the extent to which physical problems limit work or other day-to-day activities. Clerks had the worst health score for Physical Functioning, however, they had the second best health score for Role Limitation-Physical. One explanation for this apparent contradiction is that the Clerk occupation may have less physical demands than the other occupations. If workers with poor physical health choose to work in this occupation for this reason, it would explain why Clerks scored poorly on Physical Functioning but highly on Role Limitation-Physical.

Meanwhile, Service & Sales Workers and Labourers & Unskilled Workers, who both had the second worst Physical Functioning health scores, also had the second and third worst health scores for Role Limitation-Physical. It is likely that these two occupations have greater physical requirements than Clerks. As such, physical health problems found within these occupations also correspond with physical limitation problems.

Our results support the findings in New Zealand of Davis et al. (2004) who compared General Health Status with the NZSEI. Furthermore, the value of analysing the SF-36 using the individual domains is shown in this study, where differences in health domains emphasise the heterogeneity between occupations.

While occupation was significantly associated with some of the SF-36 health domains, there was no significant difference for Mental Health, Social Functioning, or Vitality. This finding is supported by research in Finland, which also found significant differences in physical health by occupation, but no significant differences for mental health measures. However, other international research using different socio-economic status measures such as income or education has had mixed results (Lahelma et al. 2005).

Overall, few differences in the use of key health services were found to be associated with occupation. This is also similar to the result found by Davis et al. (2004). The main difference we find is that the health services not subsidised by the government tend to be used more by people in the occupations associated with higher personal income. This finding implies that health services are generally accessible to a wide variety of New Zealand workers.

Most health risks differed by occupation. These include Body Mass Index (but not top 10% of waist measurements), eating two or more pieces of fruit per day (but not eating three or more vegetables per day), exercise, smoking, marijuana use, and alcohol consumption.

The risk of licit and illicit drug use is highest amongst Trade Workers and Plant & Machinery Operators. Drug use might alter perceptions and mask mental health problems. This may explain the very high Mental Health Summary Score of these workers (results not displayed). Davis et al. (2004) also found an occupational gradient for smoking using NZSEI. Other research has found that smoking is highly correlated with socio-economic status (Howden-Chapman & Tobias 2000).

Nutritional risk is highest in Administrators/Managers. This differs from the findings by Caban et al. (2005) in the United States, who reported that motor vehicle operators were the most likely to be classified as obese. However, it is interesting to note that Plant and Machinery Operators in New Zealand were the most likely to be classified as "overweight or obese".

There are a number of limitations to this study. This is a cross-sectional study and therefore is unable to determine the direction of causality. The study does not control for socio-economic variables other than occupation. Income and education are also known to be associated with health (Howden-Chapman & Tobias 2000). Some of our findings will therefore reflect differences in socio-economic status, as well as differences in occupation. This study also does not differentiate between types of work, such as permanent employees and temporary, casual, or on-call workers. Research has shown that health status can vary by type of work (Virtanen et al. 2005). If some occupations are made up of a greater share of permanent workers then this may have influenced differences in health by occupation.

CONCLUSION

There are significant differences in the health of people in different occupation groups even when age, sex, and ethnicity are taken into account. This is reflected in their SF-36 domain scores, and their risk behaviour. However, health services use does not generally differ by occupation.

In aiming to improve the health of the New Zealand population (King 2000), our findings provide an indication of key occupations where particular interventions may lead to improvements in health. Further examination of the reasons why these differences might occur in New Zealand would provide us with more information to assess the plausibility of introducing interventions to improve health in key occupations in New Zealand.

Table 1 SF-36 domains and abbreviated item content.

Code	Domain	Low score interpretation	High score interpretation	Item	Item abbreviated contents
PF	Physical Functioning	Limited a lot in performing all physical activities, including self-care, due to health	Performs all types of physical activities, including the most vigorous, without limitations due to health	PF1 PF2 PF3 PF4 PF5 PF6 PF7 PF8 PF9 PF10	Vigorous activities, such as running, lifting heavy objects Moderate activities, such as vacuuming, bowling Lifting or carrying groceries Climbing several flights of stairs Climbing one flight of stairs Bending, kneeling, stooping Walking more than1 km Walking half a kilometre Walking 100 m Bathing or dressing yourself
RP	Role Limitation-Physical	Limited a lot in work or other daily activities as a result of physical health	No problems with work or other daily activities as a result of physical health	RP1 RP2 RP3 RP4	Cut down the amount of time spent on work or other activities Accomplished less than would like to Limited in the kind of work or other activities Difficulty performing work or other activities
BP	Bodily Pain	Very severe and extremely limiting bodily pain	No pain or limitations due to pain	BP1 BP2	Intensity of bodily pain Extent pain interfered with normal work
GH	General Health Perceptions	Evaluates own health as poor and believes it is likely to get worse	Evaluates own health as excellent	GH1 GH2 GH3 GH4 GH5	Is your health excellent, very good, good, fair or poor I seem to get sick a little easier than other people I am as health as anybody I know I expect my health to get worse My health is excellent
VT	Vitality	Feels tired and worn out all of the time	Feels full of energy all of the time	VT1 VT2 VT3 VT4	Feel full of life Have a lot of energy Feel worn out Feel tired
SF	Social Functioning	Extreme and frequent interference with normal social activities due to physical or emotional problems	Performs normal social activities without interference due to physical or emotional problems	SF1 SF2	Extent health problems interfered with normal social activities Frequent health problems interfered with social activities
RE	Role Limitation-Emotional	Problems with work or other daily activities as a result of emotional problems	No problems with work or other daily activities as a result of emotional problems	RE1 RE2 RE3	Cut down the amount of time spent on work or other activities Accomplished less than would like to Didn’t do work or other activities as carefully as usual
MH	Mental Health	Has feelings of nervousness and depression all of the time	Feels peaceful, happy and calm all of the time	MH1 MH2 MH3 MH4 MH5	Been a very nervous person Felt so down in the dumps that nothing could cheer you up Felt calm and peaceful Felt down Been a happy person

Notes: Adapted with permission by Public Health Intelligence, Ministry of Health (Ministry of Health 2004, 2008).

Table 2 Demographic profile of each occupational group.

		Type of work
	Administrator/ Managers	Professionals	Technicians & Associated Professionals	Clerks	Service & Sales Workers	Agriculture & Fisheries Workers	Trade Workers	Plant & Machinery Operators	Labourers & Unskilled Workers	All workers	All
Sample size (n)	939	1212	389	538	1255	689	629	435	808	6894	12529
Age group (column %)
15–24	6	7	12	14	33	14	19	6	27	16	17
25–34	20	27	24	21	19	18	21	22	20	21	18
35–44	31	31	29	24	19	24	25	30	23	26	21
45–54	24	22	22	28	19	24	23	26	13	22	17
55–64	16	11	10	12	9	13	12	13	12	12	12
65+	3	3	3	2	2	7	1	3	5	3	15
Sex (column %)
Female	47	54	43	86	63	32	10	18	48	47	48
Male	53	46	57	14	37	68	90	82	52	53	52
Ethnicity (column %)
Maori	7	6	7	6	10	12	13	18	19	10	11
Pacific people1	2	2	3	2	5	–	3	8	11	4	4
Asian1	3	6	6	5	6	–	3	4	4	4	6
European New	88	86	85	87	79	86	81	70	66	82	79
Zealander / other
Highest educational qualifications (column %)
None	10	1	4	12	20	29	14	27	39	16	22
Secondary School	28	8	23	48	41	33	19	32	35	28	29
Trade/professional	27	30	38	26	23	24	60	35	20	31	27
University	35	61	35	14	16	15	7	5	6	25	21
Personal yearly income (column %)
loss to $10k	5	6	9	12	26	13	7	5	30	13	21
$10–$20K	6	8	13	24	20	11	8	9	26	14	21
$20–$40k	27	27	31	47	30	30	45	50	27	33	25
$40k+	57	54	42	11	15	33	33	29	8	33	23
Refused/don’t know	5	5	5	7	9	13	8	7	9	7	9
Personal yearly income (av.)	$54k	$48k	$41k	$26k	$23k	$41k	$37k	$37k	$18k	$37k	$29k
Government income support in the last year (%)
Unemployment benefit	1	1	2	2	6	4	5	3	12	4	6
Sickness or invalids benefit1	–	1	–	–	2	2	2	–	6	2	5
Personal medical insurance (column %)
Yes	63	55	50	55	39	48	41	43	25	47	39
No	37	45	49	45	59	51	58	57	72	52	60
Don’t know/refused1	–	–	–	–	2	–	–	–	4	1	1

1Results in cells which contain fewer than 10 people are not displayed.

Table 3 SF-36 unadjusted average domain scores by occupation group.

	Type of work
	Administrator/ Managers	Professionals	Technicians & Associated Professionals	Clerks	Service & Sales Workers	Agriculture & Fishery Workers	Trade Workers	Plant &Machinery Operators	Labourers & Unskilled Workers	p-value for occupation	p-value for occupation in adjusted model
Sample size (n)	939	1212	389	538	1255	689	529	435	808
SF-36 component scores
Physical  Functioning	93.0	94.1	94.2	91.3	92.4	93.0	94.5	93.3	92.4	0.0001	0.0165
Role Limitation-  Physical	87.2	86.6	87.7	88.3	85.6	81.3	86.7	89.5	84.5	0.0934	0.0346
Bodily Pain	78.0	78.2	79.5	78.7	76.4	74.5	74.9	78.1	74.7	0.0019	0.0005
General Health	79.9	79.0	79.6	76.4	76.5	78.7	78.4	76.8	75.9	0.0105	0.0473
Vitality	67.8	67.2	68.5	62.8	65.1	66.6	68.7	68.0	66.1	0.0016	0.0947
Social Functioning	92.5	92.8	91.0	92.3	91.6	91.2	93.0	94.6	91.0	0.2283	0.4441
Role Limitation-  Emotional	93.4	91.9	90.1	94.3	90.6	91.7	95.1	95.6	91.7	0.0023	0.0154
Mental Health	84.6	84.7	84.9	83.3	82.1	85.5	85.7	85.7	82.6	0.0000	0.4804
Health Transition	43.9	43.7	44.1	43.0	42.1	44.3	43.4	47.1	41.2	0.0112	0.5698

Table 4 Health service use and risk factors by occupation.

	Type of work
	Administrator/ Managers	Professionals	Technicians & Associated Professionals	Clerks	Service & Sales Workers	Agriculture & Fishery Workers	Trade Workers	Plant & Machinery Operators	Labourers & Unskilled Workers	p-value for occupation	p-value for occupation in adjusted model
Sample size (n)	939	1212	389	538	1255	689	529	435	808
Health service use1
Visit GP in the last year (%)	80	81	83	81	79	74	75	76	72	0.0100	0.0989
No. of visits to the GP in the last  year (n)	3.1	3.0	3.2	3.4	3.8	3.4	3.0	3.3	3.4	0.0600	0.3788
Needed to go but didn’t2 (%)	11	11	12	13	14	11	11	13	10	0.6964	0.7010
Visit specialist in the last year (%)	31	32	27	33	29	27	25	20	24	0.0040	0.1810
No. of visits to specialist in last year  (n)	2.5	2.8	2.8	3.0	2.8	2.2	2.6	2.4	2.5	0.3999	0.7974
Visit nurse in the last year (%)	41	45	39	43	42	41	37	40	35	0.3191	0.2263
No. of visits to nurse in last year (n)	2.75	2.89	2.87	2.89	3.29	3.84	3.40	4.52	3.20	0.3324	0.5146
Visit complementary or alternative  medical practitioner in last year (%)	36	28	28	32	27	25	19	18	18	0.0000	0.0001
No. of visits to a complementary or  alternative medical practitioner  in last year (n)	9.9	10.3	9.9	8.2	9.2	6.7	8.3	7.4	7.7	0.4707	0.6209
Visit physiotherapist in last year (%)	25	25	28	22	26	24	31	30	26	0.7103	0.9748
No. of visits to physiotherapists in  last year (n)	9.10	6.50	5.94	8.64	8.47	10.70	6.72	9.80	5.53	0.0035	0.0155
Visit optician in last year (%)	34	34	35	36	25	20	20	20	20	0.0000	0.0000
No. of visits to optician in last year  (n)	1.5	1.4	1.3	1.5	1.5	1.4	1.2	1.6	1.4	0.1445	0.5753
Risk factors (%)
Have high blood pressure	14	12	10	11	14	13	11	12	14	0.1609	0.5231
Have high cholesterol2	18	15	13	13	10	14	12	13	10	0.0016	0.0882
Are overweight or obese	64	54	52	49	53	63	61	70	55	0.0000	0.0357
Are obese	27	19	18	18	21	21	16	21	22	0.0091	0.0189
In top 10% of waist measures3	12	8	8	9	11	11	8	11	14	0.0973	0.2123
Vegetables, 3+ per day	70	70	63	73	66	76	65	71	64	0.0096	0.2040
Fruit, 2+ per day	49	61	52	66	53	52	48	40	46	0.0000	0.0000
Zero, one or two days of physical  activity	29	28	23	33	24	8	17	24	22	0.0000	0.0000
Are current, regular smokers	20	11	19	22	25	27	25	40	33	0.0000	0.0000
Last year marijuana use4	11	10	16	12	19	13	23	22	23	0.0000	0.0154
Last year alcohol abstention	8	10	9	9	13	9	6	7	16	0.0001	0.0336
Last year hazardous drinking5	15	11	20	12	21	21	33	34	29	0.0000	0.0000
1The number of visits are the average for those that attend the service.
2The two youngest age groups are combined.
3There is a separate top 10% measure for men and women.
4The three oldest age groups are combined.
5The two oldest age groups are combined.

ACKNOWLEDGMENTS

We thank Public Health Intelligence (now Health and Disability Intelligence) in the Ministry of Health for supplying the New Zealand Health Survey 2002/03 data for this research and we gratefully acknowledge all those who participated in the survey. We also thank the two anonymous referees for their comments on an earlier draft of this paper.

REFERENCES

Bohle P, Quinlan M, Kennedy D, Williamson A 2004. Working hours, work-life conflict and health in precarious and ‘permanent’ employment. Rev. Saude Publica 38(Supl.): 19–25.

Caban AJ, Lee DJ, Fleming LE, Gomez-Marin O, LeBlanc WG, Pitman T 2005. Obesity in US workers: The National Health Interview Survey, 1986 to 2002. American Journal of Public Health 95(9): 1614–1622. http://eurpub.oxfordjournals.org/cgi/reprint/16/3/294.

Cumming J, Gribben B 2007. Evaluation of the Primary Health Care Strategy: Practice Data Analysis 2001–2005. Wellington, Health Services Research Centre.

Darwent D, Lamond N, Dawson D 2008. The sleep and performance of train drivers during an extended freight-haul operation. Applied Ergonomics 39(5): 614–622.

Davis P, McLeod K, Ransom M, Ongley P, Pearce N, Howden-Chapman P 1999. The New Zealand Socioeconomic Index: developing and validating an occupationally-derived indicator of socio-economic status. Australian and New Zealand Journal of Public Health 23(1): 27–33.

Davis P, Jenkin G, Coope P, Blakely T, Sporle A, Kiro C 2004. The New Zealand Socioeconomic Index of Occupational Status: methodological revision and imputation for missing data. Australian and New Zealand Journal of Public Health 28(2): 113–119.

Dorrian J, Tolley C, Lamond N, van den Heuvel C, Pincombe J, Rogers AE, Drew D 2008. Sleep and errors in a group of Australian hospital nurses at work and during the commute. Applied Ergonomics 39(5): 605–613.

Durie MH 1985. A Māori perspective of health. Social Science and Medicine 20(5): 483–486. http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%235925%231985%23999799994%23324301%23FLP%23&_cdi=5925&_pubType=J&_auth=y&_acct=C000053190&_version=1&_urlVersion=0&_userid=1495406&md5=68798a2c0bc1ce402467b105d7f4ea69

Faragher E, Cass M, Cooper C 2005. The relationship between job satisfaction and health: a meta-analysis. Occupational and Environmental Medicine 62(2): 105–112. http://oem.bmj.com/cgi/reprint/62/2/105.

Fransen M, Wilsmore B, Winstanley J, Woodward M, Grunstein R, Ameratunga S, Norton R 2006. Shift work and work injury in the New Zealand Blood Donors’ Health Study. Occupational and Environmental Medicine 63(5): 352–358. http://oem.bmj.com/cgi/reprint/63/5/352.

Howden-Chapman P, Cram F 1998. Social, economic and cultural determinants of health. Wellington, National Advisory Committee on Health and Disability.

Howden-Chapman P, Tobias M ed. 2000. Social inequalities in health. Wellington, Ministry of Health.

Johnson S, Cooper C, Cartwright S, Donald I, Talor P, Millet C 2005. The experience of work-related stress across occupations. Journal of Managerial Psychology 20(2): 178–187.

King A 2000. The New Zealand health strategy. Wellington, Ministry of Health.

Lahelma E, Martikainen P, Rahkonen O, Roos E, Saastamoinen P 2005. Occupational class inequalities across key domains of health: results from the Helsinki Health Study. European Journal of Public Health 15(5): 504–510.

Lee DJ, Fleming LE, Gomez-Marin O, LeBlanc WG, Arheart KL, Caban AJ, Christ SL, Chung-Bridges K, Pitman T 2006. Morbidity ranking of U.S. workers employed in 206 occupations: The National Health Interview Survey (NHIS) 1986–1994. Journal of Occupational and Environmental Medicine 48(2): 117–134. http://www.joem.org/pt/re/joem/toc.00043764–200602000–00000.htm;jsessionid=LybWkG0ZCTvQpnKV92dRXTFnmvbnyzkBFQS3vSTfB97tBprx9DbB!–732617930!181195629!8091!–1.

Melchior M, Caspi A, Milne BJ, Danese A, Poulton R, Moffitt TE 2007. Work stress precipitates depression and anxiety in young, working women and men. Psychological Medicine 37(8): 1119–1129. http://journals.cambridge.org/action/displayIssue?jid=PSM&volumeId=37&issueId=08.

Ministry of Health 2004. A portrait of health: key results of the 2002/03 New Zealand Health Survey. Wellington, Ministry of Health.

Ministry of Health 2008. A portrait of health: key results of the 2006/07 New Zealand Health Survey. Wellington, Ministry of Health.

National Advisory Committee on Health and Disability 1998. The social, cultural and economic determinants of health in New Zealand: action to improve health. Wellington, National Advisory Committee on Health and Disability.

Nicol A-M, Botterill JS 2004. On-call work and health: a review. Environmental Health: A Global Access Science Source. http://www.ehjournal.net/content/3/1/15

Pisarski A, Lawrence SA, Bohle P, Brook C 2008. Organizational influences on the work life conflict and health of shiftworkers. Applied Ergonomics 39(5): 580–588.

Research Triangle Institute 2004. SUDAAN Language Manual, Release 9.0. Research Triangle Park, NC, Research Triangle Institute.

Siegrist J, Theorell T 2006. Socio-economic position and health: the role of work and employment. In: Siegrist J, Marmot M ed. Social inequalities in health: new evidence and policy implications. Oxford, Oxford University Press. Pp. 73–100.

Sverke M, Hellgren J, Naswall K 2002. No security: a meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology 7(3): 242–264.

Takahashi M, Iwakiri K, Sotoyama M, Higuchi S, Kiguchi M, Hirata H, Hisanaga N, Kitahara T, Taoda K, Nishiyama K 2008. Work schedule differences in sleep problems of nursing home caregivers. Applied Ergonomics 39(5): 597–604.

Virtanen M, Kivimaki M, Joensuu M, Virtanen P, Elovainio M, Vahtera J 2005. Temporary employment and health: a review. International Journal of Epidemiology 34(3): 610–622. http://ije.oxfordjournals.org/cgi/reprint/34/3/610.

---

K08016; Online publication date 12 February 2009
Received 30 June 2008; accepted 28 November 2008

Kōtuitui: New Zealand Journal of Social Sciences Online, 2009, Vol. 4: 55–70
1177–083X/09/0401–0055  © The Royal Society of New Zealand 2009

PDF file of entire paper: Print-quality (782K)

This year's abstracts | Journal home page | All abstracts | Publishing home page