Chapter 5: Household Crowding and Health
Dr Michael Baker, Senior Lecturer, Wellington School of Medicine and Health
Sciences1 and Public Health Physician, Institute of Environmental Science and
Research Limited; Dr Rosemary Goodyear, Analyst, Statistics New Zealand; Dr
Philippa Howden-Chapman, Associate Professor, Wellington School of Medicine
and Health Sciences.2
Introduction
This chapter reviews some of the evidence that demonstrates a link between household
crowding and health. It also presents descriptive data on the distribution of three important
infectious diseases in New Zealand and explores how this is related to levels of household
crowding at the neighbourhood level.
Measuring crowding
This report defines household crowding as a situation where one or more additional bedrooms
are required to meet the sleeping needs of the household. This standard was first used in
Canada so is sometimes called the Canadian National Occupancy Standard (CNOS).3
Statistics New Zealand has adopted this standard in recent reports4 as have other agencies,
including the Australian Bureau of Statistics5 and Housing New Zealand Corporation
(HNZC).
Household crowding, as measured by a deficit of one or more bedrooms, was relatively
uncommon in New Zealand in 2001, accounting for 5.1 percent (65,091) of households. Only
1.2 percent (15,453) required two or more bedrooms. These census findings are consistent
with results of the 1996/7 New Zealand Health Survey, which found that 7.8 percent of those
aged 15 to 64 lived in overcrowded conditions (based on needing one or more additional
bedrooms).6 The 2001 Census also identified that levels of crowding had decreased over time
in most parts of New Zealand, except Auckland, where crowding increased numerically but
declined proportionately.
Crowding is not evenly distributed across the population. As the analysis in this report has
shown, it is highly concentrated in some population groups (chapter 3) and geographic areas
(chapter 4). Low-income families with children, households containing Mäori and Pacific
people and households with recent migrants are more likely to experience crowding. There is
1. The Wellington School of Medicine has extablished He Kainga Oranga, the Housing and Health Research
Programme, which aims to contribute knowledge and robust research to inform policy to improve housing and
health in New Zealand.
2. The authors would like to acknowledge the assistance of the following people: Jasminka Milosevic,
Wellington School of Medicine and Health Sciences, for assistance with the literature review of the effects of
household crowding on mental health and psychological well being; Tony Blakely, Wellington School of
Medicine and Health Sciences, for helpful comments on the methodology; Mark Wheldon and Alex Bayley,
Analysts, Statistics New Zealand, for helping with SAS programs and compiling the maps, respectively;
Robyn Bishop and Brenda Colville of Geoframes, SNZ, for the extra geocoding; Peter O’Brien, SNZ for
supplying population data; and The Institute of Environmental Science and Research Limited (ESR), supplier
of the notification data, which was collected on behalf of the Ministry of Health.
3. Canadian Ministry of Housing (1991).
4. Statistics New Zealand (1998).
5. Australian Bureau of Statistics (1996).
6. Howden-Chapman (2000) 133-146.
58
Household Crowding and Health
also evidence that levels of crowding have risen for some groups. While overall crowding
levels reduced between 1986 and 1996, the gaps between crowding among Mäori and Pacific
families and other groups increased.
These findings are also validated by research carried out by the Mäori Women’s Housing
Research Project, which examined the housing conditions of Mäori women in Gisborne/East
Coast, South Auckland and Christchurch.7 It found that crowding, both temporary and
permanent, and varying degrees of homelessness were common among the families studied.
Health Effects of Household Crowding
There is a relatively small amount of robust evidence showing a link between household
crowding and disease. Most of this evidence relates to infectious disease and mental health.
Infectious Diseases
Much of the thinking about household crowding has focused on its potential role as a risk
factor for infectious diseases, particularly those with respiratory spread. New Zealand is
experiencing a severe and prolonged meningococcal disease epidemic8 and sustained high
rates of locally acquired tuberculosis,9 rheumatic fever,10 and childhood pneumonia.11 Rates
of hospitalisations for infectious diseases generally rose markedly during the 1990s.12
In New Zealand, the link between household crowding and illness has been convincingly
demonstrated for meningococcal disease. A case-control study conducted from 1997 to 1999
compared 202 cases of meningococcal disease in Auckland children with a control group of
313 children.13 Risk of disease was strongly associated with household crowding, as measured
by the number of adolescent and adult (10 years or older) household members per room. This
association would mean a doubling of risk with the addition of two adolescents or adults to a
six-room house. The association applied equally to all ethnic groups and there was no
threshold effect. Household crowding is also highly likely to contribute to other infectious
diseases, although, except for meningococcal disease, this link has not been thoroughly
investigated in New Zealand.14
The association between crowding and meningococcal disease has been shown in institutional
settings as far back as World War One15 and more recently for children in the UK,16 Ireland,17
and South Africa.18 Some studies have found the association with crowded housing to be
overwhelmed by a stronger association with passive smoking.19
Acute rheumatic fever has also long been associated with crowded living conditions: in the
UK in the 1930s,20 the US in the 1950s,21 and more recently in New Zealand.22 One of the few
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Mäori Women’s Housing Research Project Report (1994).
Baker et al (2001) S13-S19.
Carr et al (2001).
Thornley et al (2001) 41-44.
Grant (1999) 345-347.
Mills et al (2002) 254-257.
Baker et al (2001) 983-990.
McNicholas et al (2000) 453-454.
Glover (1920) 133-165.
Stanwell-Smith et al (1994) 315-328.
Grein et al (2001) 435-441.
Moodley et al (1999) 56-59.
Fisher et al (1997) 979-983, Kriz et al (2000) 117-121.
Perry et al (1937) 154-158.
Gray et al (1952) 400-412.
Lennon et al (1988) 396-398.
59
What is the extent of crowding in New Zealand?
studies to investigate this disease using robust epidemiological methods was carried out in
Yugoslavia in 1982 and found a significantly increased risk associated with home crowding.23
Tuberculosis has been linked, at the neighbourhood level, to crowding as a risk factor in New
York,24 London25 and Birmingham,26 and tuberculosis death rates to levels of overcrowding of
local authorities across England and Wales.27
Several well-designed studies have found that children living in crowded households are
vulnerable to a range of respiratory infections. Case-control studies carried out in the United
States have found household crowding28 and large household size29 to be risk factors for
Haemophilus influenzae infection, which causes a range of bacterial infections in children
similar to meningococcal diseases. Respiratory syncytial virus, which is the most important
cause of pneumonia and croup in children worldwide, has been found to be linked with
crowded homes in Norway30 and Alaska.31 A Brazilian study found that hospitalisation with
pneumonia in children less than two years of age was associated with household size of more
than six people.32 And a case-control study in Israel found that the risk of chronic ear
infection was associated with larger families and higher crowding levels.33
The link between respiratory infectious diseases and crowded living conditions is highly
plausible given that crowding increases contact between people who are susceptible and those
who are infected or carrying the organism.34 Crowding may also increase exposure to
infectious co-factors,35 or exposure to tobacco smoke,36 further increasing the risk of
respiratory disease.
Household crowding also increases the risk of infectious diseases transmitted by other means.
This includes enteric diseases such as hepatitis A;37 Helicobacter pylori infection, which is an
important cause of stomach ulceration and cancer38 and acute diarrhoeal illness in pre-term
infants.39 There is also some evidence for blood-borne diseases that may be transmitted by
direct physical contact. A study that tested blood specimens from Kawerau children showed
the risk of hepatitis B infection rose with increasing household size.40 There is also evidence
that household crowding may increase the risk of conjunctivitis in outbreak situations,41 and
head lice.42
A small amount of research has also identified an association between household crowding
and non-infectious outcomes, including pedestrian injuries in children,43 burns in children,44
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
Adanja et al (1988) 329-335.
Drucker et al (1994) 1482-1485.
Mangtani et al (1995) 963-966.
Hawker et al (1999) 1031-1034.
Elender et al (1998) 673-681.
Jafari et al (1999) 363-368.
Vadheim et al (1992) 221-235.
Carlsen et al (1987) 153-160.
Bulkow et al (2002) 210-216.
Victora et al (1994) 997-985.
Fliss et al (1991) 895-899.
Munford et al (1974) 1275-1279.
Openshaw (1991) 935-936.
Fisher et al (1997) 979-983.
Redlinge et al (1997) 1715-1717.
McCallion et al (1996) 18-21, Fall et al (1997) 310-314, Webb et al (1994) 750-753.
Emond et al (1997) 228-239.
Milne et al (1987) 84-90.
Patriarca et al (1983) 1283-1289.
Sarov et al (1988) 176-179.
Agran et al (1998) 188-193.
Delgado et al (2002) 38-41.
60
Household Crowding and Health
and, at a neighbourhood level, household size and mortality from myocardial infarction.45 A
New Zealand cot death study found that while the risk of cot death increased with increasing
household density, this effect lost statistical significance when other risk factors were
included in the analysis.46
Mental Health and Psychological Well-being
It is widely assumed that crowding represents a threat to mental health, though the evidence
base for this is less than for the physical effects of crowding. The range of mental health
outcomes extends from psychological symptoms such as anxiety, irritableness, sleeping
difficulties and depression to well-defined mental illness. In such studies it is difficult to
separate the effects of crowding from other aspects of housing conditions, such as noise, cold,
and dampness,47 living in high rise buildings48 and run-down neighbourhoods, and lack of
control over these conditions.49
There is little evidence from New Zealand about the relationship between mental health and
crowding. Mental illness or stress symptoms were cited in a New Zealand study of lowincome residents from Auckland and Christchurch as health problems they perceived as being
caused or exacerbated by housing conditions; however, they did not emerge as significant
predictors of mental health.50 A recent study of Pacific families living in New Zealand found
an association between cold and damp housing and postnatal depression in new mothers.51
The Mäori Women’s Housing Project found that Mäori women in rural and urban areas
considered that crowding was associated with stressful situations and family violence.52
Internationally, some studies have been published linking crowding and poor mental health
outcomes. One large cross-sectional study in Chicago found a strong association between
crowding and poor mental health.53 Ecological research in London found that crowding was
associated with psychiatric re-admissions.54
There have been only a few methodologically sound studies that have measured the effects of
crowding and housing quality on psychological distress.55 A study in Thailand found a link
between perceived lack of privacy, psychological distress and self-reported health.56
Situations where families double up in households were associated with stress in Jakarta57 and
among Hong Kong residents living in multi-storey housing.58
The relationship between crowding and mental health may not be consistent or linear.59 There
is some evidence for increased psychological symptoms for both low-density and highdensity housing.60 This relationship between housing and mental health is particularly
noticeable in people with mental illness.61 Poor mental health may affect people’s ability to
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
Wong et al (2001) 159-163.
Schluter et al (1997) 243-246.
Martin et al (1987) 1125-1127.
Ineichen et al (1974) 369-374.
Evans et al (2000) 526-530.
Smith et al (1992) 2-10.
Butler et al (2003).
Mäori Women’s Housing Research Project Report (1994).
Gove et al (1979) 59-80.
Hwang et al (1999).
Evans (2000) 526-530.
Fuller et al (1993) 1417-1428.
Clauson-Kaas et al (1997).
Mitchell (1976).
Freeman et al (1998) 147-174.
Wilkinson (1999) 297-330, Gabe et al (1987) 667-679.
Kearns (1995) 5-11.
61
What is the extent of crowding in New Zealand?
access and maintain adequate housing, while poor housing may exacerbate mental conditions.
People who applied for re-housing on the grounds of mental ill health showed improvement in
mental health symptoms.62
Crowding is more likely to have a detrimental effect on the psychological well-being of those
who are unable to find ways to reduce their stress, such as solo parents, the elderly, children
and those whose network of social support has been disrupted, such as immigrants with a
language barrier.63 Psychological distress due to crowding is mitigated by the individual’s
ability to exert power in the context of the household. More powerful members of a household
can minimise negative effects by their greater ability to organise space and activities at
home.64
There is some evidence that children may be particularly vulnerable to the effects of
household crowding. A New York study of four-year-old children found that children living
in crowded homes had more severe behavioural and cognitive development problems.65 A
household survey in West Belfast found an association between crowding and psychological
distress among children.66 A cross-sectional study of children aged 10–12 years in India found
poorer academic achievement and conflict among those living in more crowded situations.67
The need for privacy as a prerequisite for achieving comfort and better mental health may be
ethnically and culturally determined.68 A study by Evans found that Asian and Latino cultures
reported less psychological stress than white Americans when living in overcrowded homes.
However, at the biological level, as measured both by skin conductance and by task
performance, all three cultural groups showed a similar stress response. The authors
concluded that there are no differences between different ethnic groups in response to
crowding.69
A link between crowding and psychological distress and poor mental health seems
biologically plausible. Crowding is associated with information and stimulus overload, fewer
behaviour choices, increased unpredictability in the behaviour of others and anxiety triggered
by invasion of personal space.70 As an adaptation to over-stimulation from a crowded
immediate environment, people may withdraw from social interaction with members of their
household and community.71
Crowding and Infectious Diseases – Relationship at the Small Area Level
Method
The relationship between household crowding and health was investigated by exploring the
distribution of infectious disease in relation to levels of household crowding. This analysis
used the crowding quintile of the area unit (AU) as the marker of exposure to household
crowding. The diseases chosen for analysis were those where New Zealand has unusually
high rates, and household crowding is considered to be a potential risk factor. These diseases
were meningococcal disease, acute rheumatic fever and tuberculosis.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
Elton, et al (1986) 221-227.
Freeman et al (1998) 147-174.
Gove et al (1979) 59-80.
Maxwell (1996) 494-511.
Blackman et al (1989) 1-26.
Evans et al (1998) 1514-1523.
Evans et al (2000) 204-210.
Evans et al (2000) 204-210.
Freeman (1984) 197-225.
Evans et al (1989) 994-999.
62
Household Crowding and Health
Disease rates and maps were compiled using a combination of Ministry of Health notification
data (collected by the Institute of Environmental Science and Research (ESR) Ltd) and census
data. The data included all notified cases of meningococcal disease, acute rheumatic fever and
tuberculosis in New Zealand from 1998–2002.
Area units were assigned crowding quintiles according to the CNOS; that is, they were based
on the proportion of households with a deficit of one or more bedrooms (as described in the
glossary). The level of household crowding varied from 0–1.75 percent in quintile 1, 1.76–
2.66 percent in quintile 2, 2.67–4.04 percent in quintile 3, 4.05–6.90 percent in quintile 4 and
reached 6.91–43.54 percent in quintile 5.
Disease rates by crowding quintile were calculated using 2001 Census data. The number of
cases in each quintile and each sub-population (by area, ethnicity and age) were counted for
the five-year period. The ethnicities used were Mäori, European and Pacific peoples. Statistics
New Zealand used the census data to calculate the denominator population of each quintile
and also the size of each sub-population (by area, age, and ethnic group). These figures were
combined with disease incidence expressed as an average annual rate per 100,000 population.
Population characteristics and levels of crowding varied markedly across New Zealand. The
age and ethnic structures of each quintile were different, with quintile 5, for example, having
a younger age structure. Almost half of Auckland’s usually resident population (who lived in
an area unit that had been assigned a crowding quintile) was included in quintile 5, compared
with just 0.8 percent of the South Island’s usually resident population.
This analysis was based on 2,558 cases of meningococcal disease, 1,837 cases of tuberculosis
and 460 cases of acute rheumatic fever that occurred between 1998 and 2002. The analysis of
disease rates by ethnicity was confined to people of Mäori, Pacific peoples and European
ethnicity. Cases were excluded from this analysis if they could not be assigned to an area unit
or if the area unit they lived in was too small to be assigned a crowding quintile (fewer than
50 households). In total, 76 (2.8 percent) of the meningococcal disease cases, 87 (4.5 percent)
of the tuberculosis cases and 48 (9.5 percent) of the acute rheumatic fever cases were
excluded.
Meningococcal Disease
The following maps (figures 5.1 and 5.2) show the distribution of cases of meningococcal
disease in the Auckland and East Cape areas. These cases have been superimposed on the
maps where area units are shaded according to their crowding quintile. Cases appear clustered
in area units with higher levels of crowding.
63
What is the extent of crowding in New Zealand?
Figure 5.1
Cases of Meningococcal Disease by Area Unit
For Auckland territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
64
Household Crowding and Health
Figure 5.2
Cases of Meningococcal Disease
By census area unit for East Cape territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
65
What is the extent of crowding in New Zealand?
A more meaningful analysis is to look at rates of disease by crowding quintile. For
meningococcal disease, the incidence increased from 6.7 cases per 100,000 in the least
crowded quintile to 27.8 cases per 100,000 in the most crowded quintile.
To understand this relationship better, it is useful to look at disease rates by age, ethnicity and
area. As figure 5.3 shows, rates of meningococcal disease were highest in children under 15
years of age, with particularly high rates among those under five years. Within each of these
age groups, rates increased markedly for those living in the more crowded quintiles.
Comparing the most crowded quintile with the least crowded, there was a four-fold increase
in rates for children under five years of age, and a five-fold increase for children aged 5–14
years.
Figure 5.3
Rates of Meningococcal Disease per 100,000 Population
By age group and crowding quintile
Annual average for 1998–2002
180
Meningococcal disease rate per 100,000 population
Quintile 1
160
Quintile 2
140
Quintile 3
Quintile 4
120
Quintile 5
100
80
60
40
20
0
0-4 years
5-14 years
15-29 years
30-59 years
60+ years
Age group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by ethnicity showed that rates were higher for Mäori and Pacific people compared
with Europeans (Figure 5.4). Again, rates were markedly higher among those living in the
most crowded areas.
66
Household Crowding and Health
Figure 5.4
Rates of Meningococcal Disease per 100,000 Population
for Children (0–14 Years)
By ethnic group and crowding quintile
Annual average for 1998–2002
140
Meningococcal disease rate per 100,000 population
Quintile 1
Quintile 2
120
Quintile 3
Quintile 4
100
Quintile 5
80
60
40
20
0
European
Mäori
Pacific
Ethnic group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by geographic area (Auckland, rest of North Island, and South Island) again
showed that rates were markedly higher among those living in the most crowded quintiles,
regardless of region (Figure 5.5).
67
What is the extent of crowding in New Zealand?
Figure 5.5
Rates of Meningococcal Disease per 100,000 Population
for Children (0–14 Years)
By geographic area and crowding quintile
Annual average for 1998–2002
100
Meningococcal disease rate per 100,000 population
Quintile 1
90
Quintile 2
Quintile 3
80
Quintile 4
Quintile 5
70
60
50
40
30
20
10
0
Auckland
Rest of North Island
South Island
Area
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Acute Rheumatic Fever (ARF)
The following maps (Figures 5.6 and 5.7) show the distribution of cases of acute rheumatic
fever (ARF) in the Auckland and East Cape areas. These cases have been superimposed on
maps where the area units are shaded according to their crowding quintile. As with
meningococcal disease, cases of ARF appear clustered in area units with higher levels of
crowding.
68
Household Crowding and Health
Figure 5.6
Cases of Acute Rheumatic Fever by Area Unit
For Auckland territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
69
What is the extent of crowding in New Zealand?
Figure 5.7
Cases of Acute Rheumatic Fever by Area Unit
For East Cape territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
70
Household Crowding and Health
Analysing rates of disease by crowding quintile showed that ARF incidence increased from
0.2 cases per 100,000 in the least crowded quintile to 7.8 cases per 100,000 in the most
crowded quintile.
As Figure 5.8 shows, rates of ARF are highest in children 5–14 years of age, with lower rates
among those 15–29 years, and very few cases in other age groups. Within these two age
groups, rates increase markedly for those living in the more crowded quintiles. Comparing the
most crowded quintile with the least crowded, rates increased by 32-fold for children aged 514 years and by a similar amount for people aged 15–29 years.
Figure 5.8
Rates of Acute Rheumatic Fever per 100,000 Population
By age group and crowding quintile
Annual average for 1998–2002
35
Rheumatic fever rate per 100,000 population
Quintile 1
Quintile 2
30
Quintile 3
Quintile 4
25
Quintile 5
20
15
10
5
0
0 - 4 years
5 - 14 years
15 - 29 years
30 - 59 years
60+ years
Age Group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by ethnicity (Figure 5.9) shows that rates are higher for Mäori and Pacific people,
compared with Europeans. Again, rates are markedly higher among those living in the most
crowded quintile.
71
What is the extent of crowding in New Zealand?
Figure 5.9
Rates of Acute Rheumatic Fever per 100,000 Population
for Children (0–14 Years)
By ethnic group and crowding quintile
Annual average for 1998–2002
35
Rheumatic fever rate per 100,000 population aged 0-14 years
Quintile 1
Quintile 2
30
Quintile 3
Quintile 4
25
Quintile 5
20
15
10
5
0
European
Mäori
Pacific
Ethnic group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by geographic area (Auckland, rest of North Island, and South Island) again shows
that rates of ARF are markedly higher among those living in the most crowded quintile,
regardless of region (Figure 5.10).
72
Household Crowding and Health
Figure 5.10
Rates of Acute Rheumatic Fever per 100,000 Population
for Children (0–14 Years)
By geographic area and crowding quintile
Annual average for 1998–2002
25
Rheumatic fever rate per 100,000 population aged 0-14
Quintile 1
Quintile 2
Quintile 3
Quintile 4
20
Quintile 5
15
10
5
0
Auckland
Rest of North Island
South Island
Area
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Tuberculosis
The following maps (Figures 5.11 and 5.12) show the distribution of cases of tuberculosis in
the Auckland and East Cape areas. These cases have been superimposed on maps showing
levels of crowding at the area unit level. Cases appear clustered in area units with higher
levels of crowding.
73
What is the extent of crowding in New Zealand?
Figure 5.11
Cases of Tuberculosis by Area Unit
For Auckland territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
74
Household Crowding and Health
Figure 5.12
Cases of Tuberculosis by Area Unit
For East Cape territorial authorities
Total for 1998–2002
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
75
What is the extent of crowding in New Zealand?
Analysing rates of disease by crowding quintile showed that tuberculosis incidence increased
from 2.4 cases per 100,000 in the least crowded quintile to 18.7 cases per 100,000 in the most
crowded quintile.
As Figure 5.13 shows, rates of tuberculosis are highest in those over 60 years, with a second
peak in those aged 15–29 years. The association with crowding quintile was apparent across
all age groups but was most marked among the young. Comparing the most crowded quintile
with the least crowded, rates for those aged 0–4 years increased 25-fold, and rates for those
aged over 60 years increased five-fold.
Figure 5.13
Rates of Tuberculosis per 100,000 Population
By age group and crowding quintile
Annual average for 1998–2002
40
Disease rate per 100,000 population
Quintile 1
Quintile 2
35
Quintile 3
Quintile 4
Quintile 5
30
25
20
15
10
5
0
0-4 years
5-14 years
15-29 years
30-59 years
60+ years
Age group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by ethnicity (Figure 5.14) showed that rates were higher for Mäori and Pacific
people compared with Europeans. Again, rates were markedly higher among those living in
the most crowded quintile.
76
Household Crowding and Health
Figure 5.14
Rates of Tuberculosis per 100,000 Population
By ethnic group and crowding quintile
Annual average for 1998–2002
35
Disease rate per 100,000 population
Quintile 1
Quintile 2
30
Quintile 3
Quintile 4
Quintile 5
25
20
15
10
5
0
European
Mäori
Pacific Peoples
Ethnic group
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 1998–
2002.
Analysing by geographic area (Auckland, rest of North Island, and South Island) again shows
that rates of tuberculosis are markedly higher among people living in the most crowded
quintiles regardless of region (Figure 5.15).
77
What is the extent of crowding in New Zealand?
Figure 5.15
Rates of Tuberculosis per 100,000 Population
By geographic area and crowding quintile
Annual average for 1998–2002
30
Disease rate per 100,000 population
Quintile 1
Quintile 2
Quintile 3
25
Quintile 4
Quintile 5
20
15
10
5
0
Auckland Tas
Rest of North Island
South Island
Area
Source: Statistics New Zealand, Census of Population and Dwellings, 2001, and ESR, Notification Data, 19982002.
78
Household Crowding and Health
Establishing the relationship between crowding and infectious disease
These New Zealand findings are consistent with overseas ‘ecological’ studies that have found
higher rates of infectious diseases in areas with higher proportions of crowded households.
This association has been seen most consistently for tuberculosis,72 but also for acute
rheumatic fever,73 meningococcal and pneumococcal meningitis74 and meningococcal
disease.75 Crowding is often measured as part of a composite deprivation score. Many disease
outcomes are found to be associated with deprivation measured in this way. In the UK, for
example, meningococcal disease,76 tuberculosis,77 and hospital admissions for infectious
intestinal diseases78 are significantly higher in more deprived areas, as measured by their
Townsend scores. Townsend scores are a composite deprivation score based on the proportion
who are unemployed, do not own a car, and do not own their own home, as well as those
living in houses that have more than one person to a room.
Such analyses have a number of limitations in terms of being able to establish a causal link
between observed levels of household crowding and disease. The following table shows the
characteristics of crowded households compared with those defined as not crowded, using the
Canadian National Occupancy Standard. It also compares the characteristics of the least and
most crowded area unit quintiles.
72. Drucker et al (1994) 1482-1485, Mangtani et al (1995) 963-966, Elender et al (1998) 673-681, Hawker (1999)
1031-1034.
73. Gordis et al (1969) 655-666.
74. Jones et al (1997) 794-795.
75. Fone et al (2003) 53-58.
76. Stuart et al (2002) 327-328.
77. Spence et al (1993) 759-761.
78. Olowokure et al (1999) 807-808.
79
What is the extent of crowding in New Zealand?
Figure 5.16
Comparisons of Household Classified as Crowded and Not Crowded and those
in the Most Crowded Quintile and Least Crowded Quintile of Area Units
By selected households and individual characteristics
1
For 2001
Characteristic
Not crowded
households
(CNOS)
Total
Crowded
households
(CNOS)
Least crowded
CAU quintile
Most crowded
CAU quintile
1,344,267
1,220,916
65,088
207,456
285,312
27.1%
29.7%
12.3%
2.2%
28.7%
28.9%
29.7%
11.2%
1.3%
28.9%
1.3%
38.9%
33.4%
18.4%
8.0%
32.8%
32.7%
8.1%
0.9%
25.5%
20.1%
29.1%
17.5%
4.9%
28.4%
Family size
Average number of people in household
Median number of people in household
2.67
2
2.54
2
5.35
5
2.57
2
3.01
3
House size
Average number of rooms
Average number of bedrooms
Median number of rooms
Median number of bedrooms
6.15
3.05
6
3
6.20
3.07
6
3
5.42
2.76
5
3
6.58
3.22
6
3
5.74
2.94
6
3
0.46
0.91
5.1%
1.2%
0.43
0.85
0
0
1.05
2.01
100.0%
23.7%
0.40
0.82
1.3%
0.1%
0.56
1.06
12.6%
3.8%
Age structure
Average age
Median age
Average number of children <15 years per household
Median number of children <15 years per household
Households with children < 15 years %
34.9
34
0.62
0
32.9%
36.0
36
0.6
0
30.7%
23.7
20
2.0
2
79.8%
37.5
38
0.6
0
30.1%
31.4
30
0.8
0
40.0%
Ethnicity of household
European %
Maori %
Pacific %
Asian %
Other %
85.5%
15.0%
5.2%
6.2%
0.8%
89.8%
13.9%
4.0%
5.7%
0.7%
52.9%
40.8%
27.7%
16.7%
2.3%
95.2%
7.1%
1.1%
3.2%
0.4%
66.4%
27.3%
15.5%
10.3%
1.2%
Origins
Born overseas %
Of people born overseas, arrived in NZ in previous 5 years %
19.5%
27.5%
18.3%
24.8%
28.5%
43.9%
15.0%
22.8%
25.9%
31.7%
Education
No educational qualifications %
27.4%
26.5%
37.4%
24.6%
32.7%
Tenure
Own house %
Rent house %
67.8%
32.2%
69.6%
30.4%
37.6%
62.4%
78.0%
22.0%
55.6%
44.4%
Number of households
Household structure
Couple only (with or without others) %
Couple with children (with or without others) %
One parent with child(ren) (with or without others) %
Two or more family household %
Other household composition %
People per room and room deficit
Average people per room
Average people per bedroom
Short 1 or more bedrooms %
Short 2 or more bedrooms %
80
Household Crowding and Health
Figure 5.16
continued
Characteristic
Income
Average JEAH
Median JEAH
Receipt of income from Government benefit (DPB, Community
Wage, Invalids) but excluding superannution and student
allowance %
Resources
Households with one or more motor vehicles %
Households with one or more telephones %
Total
Not crowded
households
(CNOS)
Crowded
households
(CNOS)
Least crowded
CAU quintile
Most crowded
CAU quintile
45,431
34,689
46,182
36,319
30,375
25,398
49,400
41,544
41,043
34,420
24.5%
22.9%
64.9%
16.5%
34.6%
89.9%
96.3%
90.4%
96.8%
85.2%
87.7%
94.0%
98.4%
84.9%
92.5%
Home heating
Combustion (one or more of mains gas, bottled gas, wood, coal) %
71.0%
71.8%
59.7%
76.7%
62.5%
Bottled gas %
Never use any form of heating in dwelling %
28.3%
2.7%
28.4%
2.4%
28.1%
7.8%
24.9%
0.9%
29.3%
5.7%
Smoking (based on 1996 census)
Any smoker in household (based on 1996 census) %
Proportion of adults in household who smoke %
32.9%
0.24
32.1%
0.23
59.3%
0.39
26.6%
0.18
41.3%
0.31
1. Smoking data is based on 1996 census.
Source: Statistics New Zealand, Census of Population and Dwellings, 1996 and 2001.
These tabulated data illustrate the two most important limitations in the analysis presented in
this chapter:
• Confounding effects – Crowding is highly associated with other measures of socioeconomic deprivation such as low income, unemployment, low education level and fewer
material resources such as cars and telephones. Other risk factors for respiratory disease,
notably active and passive smoking, are also higher in such areas. Children living in a
crowded house are almost twice as likely to have a smoker in the household as children in
a house that is not crowded. These other risk factors could potentially explain (confound)
some of the observed association between living in a residence in a more crowded
neighbourhood and disease rates.
• Ecological effects – The area unit where a household is located is a less specific measure
of exposure to crowding than household level census data. The average number of people
per room is 1.05 for crowded houses, compared with 0.56 for houses in the most crowded
quintile of area units. This same ecological effect is seen for all of the exposures listed in
the above table. For example, 59.3 percent of households classified as crowded contain a
smoker, compared with 41.3 percent of households in the most crowded area unit quintile.
Assigning average area unit values to every household in that area unit will therefore
provide a fairly imprecise measure of exposure.
Because of these limitations, it is not possible to ascribe causality to the observed associations
between crowding and higher disease rates. However, these data do provide supporting
evidence that is consistent with the results of more rigorous investigations of the link between
household crowding and infectious disease.
Conclusions and the Need for Further Research
Despite the apparent weight of evidence, a recent New Zealand review of the effects of
crowding on health concluded that “The debate about the relationship between crowding and
health is long standing and inconclusive”.79 That review suggested that more research is
needed to resolve this question. Researchers at the Wellington School of Medicine and Health
Sciences conclude that there is sufficient evidence to support interventions to reduce
79. Gray (2001).
81
What is the extent of crowding in New Zealand?
household crowding in New Zealand, at least as a strategy to reduce the risk of respiratory
infectious diseases. Evaluating the impact of such interventions may also provide more robust
evidence to support a causal relationship between housing conditions and health outcomes.
Partly based on evidence linking household crowding and meningococcal disease, Housing
New Zealand Corporation (HNZC) has implemented its Healthy Housing Programme to
reduce levels of crowding in its properties.80 HNZC is also collaborating with the Housing
and Health Research Programme at the Wellington School of Medicine and Health Sciences
on a cohort study of HNZC housing applicants and tenants. This study aims to investigate the
relationship between levels of household crowding and hospitalisation for a wide range of
health conditions. It will also assess the impact of the move from waiting list to being rehoused. Such research could add to the very scant literature on the health effects of housing
improvements.81
80. Bullen et al (2003).
81. Thomson et al (2001) 187-190.
82
Household Crowding and Health
References
Adanja B, Vlajinac H, Jarebinski M (1988). “Socioeconomic factors in the etiology of
rheumatic fever", Journal of Hygiene Epidemiology Microbiology and Immunology, 32: 329335.
Agran P F, Winn D G, Anderson C L, Del Valle C (1998). “Family, social, and cultural
factors in pedestrian injuries among Hispanic children", Injury Prevention, 4: 188-193.
Australian Bureau of Statistics (1996). 1994 Australian Housing Survey, Housing
Characteristics, Costs and Conditions, Canberra: Australian Government Publishing Service.
Baker M, McNicholas A, Garrett N, et al (2000). "Household crowding a major risk factor for
epidemic meningococcal disease in Auckland children", Pediatric Infectious Disease Journal,
19: 983-990.
Baker M G, Martin D R, Kieft C E M, et al (2001). “A 10-year serogroup B meningococcal
disease epidemic in New Zealand: Descriptive epidemiology, 1991-2000”, Journal of
Paediatrics and Child Health, 37, S13-S19.
Blackman T, Evason E, Melaugh M, et al (1989). “Housing and health: A case study of two
areas in West Belfast”, Journal of Social Policy, 18: 1-26.
Bulkow L R, Singleton R J, Karron R A, et al (2002). “Risk factors for severe respiratory
syncytial virus infection among Alaska native children”, Pediatrics, 109: 210-216.
Bullen C and Woolston J (2003). “Realising healthcare outcomes through cross-agency
collaboration: The Healthy Housing Development Pilot Programme” , in Second Managing,
Monitoring and Measuring Outcomes in Health Conference, Auckland.
Butler S, Williams M, Tukuitonga C, et al (2003). “Problems with damp and cold housing
among Pacific families in New Zealand”, New Zealand Medical Journal, 116: U494.
Canadian Ministry of Housing Corporation (1991). Core Housing Need in Canada, Canadian
Government Print, Ottawa.
Carr J, Baker M, Kieft C, et al (2001). “The epidemiology of tuberculosis in New Zealand
1995-1999”, unpublished report, Institute of Environmental Science and Research Ltd,
Porirua.
Carlsen K H, Larsen S, Bjerve O, et al (1987). “Acute bronchiolitis: predisposing factors and
characterization of infants at risk”, Pediatric Pulmonology, 3: 153-60.
Clauson-Kaas J, Dzikus A, Surjadi C, et al (1997). Crowding and Health in Low-income
Settlements. United Nations Centre for Human Settlements (Habitat), Avebury, England.
Delgado J, Ramirez-Cardich M E, Gilman R H, et al (2002). “Risk factors for burns in
children: Crowding, poverty, and poor maternal education”, Injury Prevention, 8: 38-41.
Drucker E, Alcabes P, Bosworth W, et al (1994), “Childhood tuberculosis in the Bronx, New
York” Lancet, 343: 1482-1485.
Elender F, Bentham G and Langford I (1998). “Tuberculosis mortality in England and Wales
during 1982-1992: Its association with poverty, ethnicity and AIDS”, Social Science and
Medicine, 46: 673-681.
Fischer M, Hedberg K, Cardosi P, et al (1997). “Tobacco smoke as a risk factor for
meningococcal disease”, Pediatric Infectious Disease Journal, 16: 979-983.
83
What is the extent of crowding in New Zealand?
Elton P J and Packer J M (1986). “A prospective randomised trial of the value of rehousing on
the grounds of mental ill-health”, Journal of Chronic Diseases, 39: 221-227.
Emond A M, Howat P, Evans J A, et al (1997). “The effects of housing on the health of
preterm infants”, Paediatric and Perinatal Epidemiology, 11: 228-239.
Evans G, Lepore S J, Shejwal B R, et al (1998). “Chronic residential crowding and children's
well-being: An ecological perspective”, Child Development, 69: 1514-1523.
Evans G W, Chan H Y E, Wells N M, et al (2000). “Housing quality and mental health”,
Journal of Consulting and Clinical Psychology, 58: 526-530.
Evans G W, Lepore S J and Allen K M (2000). “Cross-cultural differences in tolerance for
crowding: Fact or fiction?”, Journal of Personality and Social Psychology, 79: 204-210.
Evans G W, Palsane M N, Lepore S J, et al (1989). “Residential density and psychological
health: The mediating effects of social support”, Journal of Personality and Social
Psychology, 57: 994-999.
Fall C H, Goggin P M, Hawtin P, et al (1997). “Growth in infancy, infant feeding, childhood
living conditions, and Helicobacter pylori infection at age 70”, Archives of Disease in
Childhood, 77: 310-314.
Fliss D M, Shoham I, Leiberman A, et al (1991). “Chronic suppurative otitis media without
cholesteatoma in children in southern Israel: Incidence and risk factors”, Pediatric Infectious
Disease Journal, 10: 895-899.
Fone D L, Harries J M, Lester N, and Nehaul L (2003). “Meningococcal disease and social
deprivation: A small area geographical study in Gwent, UK”, Epidemiology and Infection,
130: 53-58.
Freeman H (1984). “Housing” in Freeman H (ed), Mental Health and the Environment,
Churchill Livingston, London.
Freeman H and Stansfeld S (1998). “Psychosocial effects of urban environments, noise, and
crowding” in Lundberg A (ed), The Environment and Mental Health: A Guide for Clinicians,
Lawrence Erlbaum Associates, Mahway, New Jersey.
Fuller T D, Edwards J N, Sermsri S, et al (1993). “Housing, stress, and physical well-being:
evidence from Thailand”, Social Science and Medicine, 36: 1417-1428.
Gabe J and Williams P (1987). “Women, housing, and mental health”, International Journal
of Health Services, 17: 667-679.
Glover J (1920). Observations on the Meningococcus Carrier Rate and their Application to
the Prevention of Cerebrospinal Fever, Medical Research Council Special Report vol 50.
Medical Research Council, London.
Gordis L, Lilienfeld A and Rodriguez R (1969). “Studies in the epidemiology and
preventability of rheumatic fever. II. Socio-economic factors and the incidence of acute
attacks”, Journal of Chronic Diseases, 21: 655-666.
Gove W R, Hughes M and Galle O R (1979). “Overcrowding in the home: An empirical
investigation of its possible pathological consequences” , American Sociological Review, 44:
59-80.
Grant C C (1999). “Pneumonia in children: Becoming harder to ignore”, New Zealand
Medical Journal, 112: 345-347.
84
Household Crowding and Health
Gray A (2001). Definitions of Crowding and the Effects of Crowding on Health: A Literature
Review, Ministry of Social Policy, Wellington.
Gray F G, Quinn R W, Quinn J P (1952). “A long term survey of rheumatic and non
rheumatic families; with particular reference to environment and heredity”, American Journal
of Medicine, 13: 400-412.
Grein T, O'Flanagan D (2001). “Day-care and meningococcal disease in young children",
Epidemiology and Infection, 127: 435-441.
Hawker J I, Bakhshi S S, Ali S, et al (1999). “Ecological analysis of ethnic differences in
relation between tuberculosis and poverty", British Medical Journal, 319: 1031-1034.
Howden-Chapman P and Wilson N (2000). “Housing and health", in Howden-Chapman P
and Tobias M (eds), Social Inequalities and Health: New Zealand 1999, Ministry of Health:
Wellington.
Hwang S, Fuller-Thompson E, Hulchanski D, et al (1999). Housing and Population Health: A
Review of the Literature, Canada Mortgage and Housing Corporation, Toronto.
Ineichen B and Hooper D (1974). “Wives’ mental health and children’s behaviour problems
in contrasting residential areas”, Social Science and Medicine, 8: 369-374.
Jafari H S, Adams W G, Robinson K A, et al (1999). “Efficacy of Haemophilus influenzae
type b conjugate vaccines and persistence of disease in disadvantaged populations. The
Haemophilus Influenzae Study Group", American Journal of Public Health, 89: 364-368.
Jones I R, Urwin G, Feldman R A, et al (1997). “Social deprivation and bacterial meningitis
in north east Thames region: Three year study using small area statistics”, British Medical
Journal, 314: 794-795.
Kearns R A (1995). “Worried sick about housing: Extending the debate on housing and
health", Community Mental Health in New Zealand, 9: 5-11.
Kriz P, Bobak M and Kriz B (2000). “Parental smoking, socioeconomic factors, and risk of
invasive meningococcal disease in children: A population based case-control study” , Archives
of Disease in Childhood, 83: 117-121.
Lennon D, Martin D, Wong E, et al (1988). “Longitudinal study of poststreptococcal disease
in Auckland: Rheumatic fever, glomerulonephritis, epidemiology and M typing 1981-86” ,
New Zealand Medical Journal, 101: 396-398.
McCallion W A, Murray L J, Bailie A G, et al (1996). “Helicobacter pylori infection in
children: Relation with current household living conditions", Gut, 39: 18-21.
McNicholas A, Lennon D, Crampton P, et al (2000). “Overcrowding and infectious diseases –
when will we learn the lessons of our past?”, New Zealand Medical Journal, 113: 453-454.
Mangtani P, Jolley D J, Watson J M, et al (1995). “Socioeconomic deprivation and
notification rates for tuberculosis in London during 1982-91”, British Medical Journal, 310:
963-966.
Mäori Women's Housing Research Project (1994). '...For the Sake of Decent Shelter...' Mäori
Women's Housing Research Project Report, Wellington.
Martin C J, Platt S D, Hunt S M (1987). “Housing conditions and ill health”, British Medical
Journal, 294: 1125-1127.
85
What is the extent of crowding in New Zealand?
Maxwell L E (1996). “Multiple effects of home and day care crowding”, Environment and
Behaviour, 28: 494-511.
Mills C F, Tobias M, Baker M (2002). “A re-appraisal of the burden of infectious disease in
New Zealand: Aggregate estimates of morbidity and mortality”, New Zealand Medical
Journal, 115: 254-257.
Milne A, Allwood G K, Moyes C D, Pearce N E and Newell K (1987). “A
seroepidemiological study of the prevalence of hepatitis B infections in a hyperendemic New
Zealand community”, International Journal of Epidemiology, 16: 84-90.
Mitchell R (1976). “Cultural and health influences on building, housing and community
standards: Cost implications for the human habitat”, Human Ecology, 4, 297-330.
Moodley J R, Coetzee N and Hussey G (1999). “Risk factors for meningococcal disease in
Cape Town”, South African Medical Journal, 89: 56-59.
Munford R S, Taunay A, De Morais J S, et al (1974). “Spread of meningococcal infection
within households”, Lancet, 1: 1275-1279.
Olowokure B, Hawker J, Weinberg J, et al (1999). “Deprivation and hospital admission for
infectious intestinal diseases”, Lancet, 353: 807-808.
Openshaw P (1991). “When we sneeze does the immune system catch a cold?”, British
Medical Journal, 303: 935-936.
Patel P, Mendall M A, Khulusi S, et al (1994). “Helicobacter pylori infection in childhood:
Risk factors and effect on growth”, British Medical Journal, 309: 1119-1123.
Patriarca P A, Onorato I M, Sklar V E, et al (1983). “Acute hemorrhagic conjunctivitis
Investigation of a large-scale community outbreak in Dade County, Florida", Journal of the
American Medical Association, 249: 1283-1289.
Perry C B and Roberts J A F (1937). “Study of the variability and incidence of rheumatic
heart disease within the city of Bristol", British Medical Journal, supplement: 154-158.
Redlinger T, O'Rourke K and Van Derslice J (1997). “Hepatitis A among schoolchildren in a
US-Mexico border community", American Journal of Public Health, 87: 1715-1717.
Sarov B, Neumann L, Herman Y, et al (1988). “Evaluation of an intervention program for
head lice infestation in school children", Pediatric Infectious Disease Journal, 7: 176-179.
Schluter P J, Ford R P, Mitchell E A, et al (1997). “Housing and sudden infant death
syndrome. The New Zealand Cot Death Study Group", New Zealand Medical Journal, 110:
243-246.
Smith C J, Kearns R A, Abbott M W (1992). “A tale of two cities: The experience of housing
problems in Auckland and Christchurch", New Zealand Geographer, 48: 2-10.
Spence D P, Hotchkiss J, Williams C S, et al (1993). “Tuberculosis and poverty", British
Medical Journal, 307: 759-761.
Stanwell-Smith R E, Stuart J M, Hughes A O, et al (1994). “Smoking, the environment and
meningococcal disease: A case control study", Epidemiology & Infection, 112: 315-328.
Statistics New Zealand (1998). New Zealand Now, Housing, Statistics New Zealand,
Wellington.
86
Household Crowding and Health
Stuart J, Middleton N and Gunnell D (2002). “Socioeconomic inequality and meningococcal
disease", Communicable Disease and Public Health, 5: 327-328.
Thomson H, Petticrew M and Morrison D (2001). “Health effects of housing improvement:
Systematic review of intervention studies", British Medical Journal, 323: 187-190.
Thornley C, McNicholas A, Baker F, et al (2001). “Rheumatic fever registers in New
Zealand", New Zealand Public Health Report, 8: 41-44.
Vadheim C M, Greenberg D P, Bordenave N, et al (1992). “Risk factors for invasive
Haemophilus influenzae type b in Los Angeles County children 18-60 months of age",
American Journal of Epidemiology, 136: 221-235.
Victora C G, Fuchs S C, Flores J A, et al (1994). “Risk factors for pneumonia among children
in a Brazilian metropolitan area", Pediatrics, 93: 977-85.
Webb P M, Knight T, Greaves S, et al (1994). “Relation between infection with Helicobacter
pylori and living conditions in childhood: evidence for person to person transmission in early
life", British Medical Journal, 308: 750-753.
Wilkinson D (1999). “Poor housing and ill health: A summary of the evidence", Central
Research Unit, The Scottish Office.
Wong Y K, Dawkins K D, and Ward M E (2001). “The association between deaths from
myocardial infarction and household size in England and Wales", Journal of Cardiovascular
Risk, 8: 159-163.
87