Renting Poorer Housing: Ecological Relationships Between Tenure,Dwelling Condition,and Income and Housing-Sensitive Hospitalizations in a Developed Country

Abstract

Background

Previous research has shown two-way associations between rental tenure, poorer housing quality, and health outcomes, but little research has looked at relative housing contributions to health outcomes.

Aims

We investigated whether tenure and/or dwelling condition were associated with housing-sensitive hospitalizations and whether any association differed by income.

Method

Using a data set of housing characteristics matched to hospitalization records, rental tenure data, and income quintiles, we modeled differences in housing-sensitive hospitalization rates by ecological-level tenure and housing condition, controlling for age-group and mean temperatures.

Results

There were clear associations between income, tenure, and house condition, and winter-associated hospitalization risk. In the adjusted model, the largest risk differences were associated with neighborhoods with low income (risk ratio [RR] = 1.48) and high rental tenure (RR = 1.41). There was a nonsignificant difference for housing condition (RR = 1.04).

Discussion

Rental tenure and poor housing condition were risks for housing-sensitive hospitalization, but the association with income was stronger. Higher income households may be better able to offset quality and tenure-related health risks. This research illustrates the inverse housing law: Those most vulnerable, with most need for good-quality housing, are least likely to have it. Income inequity is inbuilt in tenure, quality, and health burden relationships.

Conclusion

These findings suggest that measures to address health inequities should include improvements to both tenure security and housing quality, particularly in low-income areas. However, policymakers aiming to reduce overall hospitalization rates should focus their efforts on reducing fuel poverty and improving the affordability of quality housing.

Keywords

hospitalizations housing housing condition housing quality income inequalities tenure

Introduction

Living in rental housing has been consistently associated with poorer physical and mental health (Clinch & Healy, 2000; Darlington-Pollock & Norman, 2017; Hales et al., 2010; Hiscock et al., 2003; Kavanagh et al., 2016; Macintyre et al., 1998; Macintyre et al., 2001; Macintyre et al., 2003; Muntaner et al., 1998; Pledger et al., 2019; Sundquist & Johansson, 1997; Turunen et al., 2017), though whether the association is causal or compositional is debated (Baker et al., 2013; Popham et al., 2015). If causal, the pathway through which the relationship acts is unclear. The most plausible pathways are direct, through dwelling age and condition (Baker et al., 2016; Gibson et al., 2011; Jacobs et al., 2009; Keall et al., 2010; Krieger & Higgins, 2002; Webb et al., 2013; Wilkinson, 1999), or indirect, through socioeconomic status (Davis et al., 2003; Salmond et al., 2007b).

Evidence of a clear relationship between dwelling tenure and condition is generally consistent but limited in quantity. Lane and Kinsey (1980) found lower satisfaction with housing characteristics among renters than home owners. Markham and Gilderbloom (1998) found that among the elderly, rental tenure was associated with housing inadequacy. Ellaway and Macintyre (1998) found that “renters were almost four times as likely to report problems with dampness and/or condensation . . . [as] owner-occupiers” (p. 145), similar to Pekkonen and Haverinen-Shaughnessy’s (2015) findings for moisture and mold damage in Finland. Stewart et al. (2002) found that North London residents aged 65 years and more were more likely to report their accommodation quality as “poor or very poor” if they lived in rental housing than if they owned their own home. Iwata and Yamaga (2008) found “that the quality of renter-occupied housing is lower than that of owner-occupied housing . . . for single-family housing in Japan” (p. 201). Howden-Chapman et al. (2011) found that a higher percentage of people living in rental housing had some or many housing problems compared with owner-occupiers, both in the 1985-1988 and the 2008-2009 phases of the Whitehall study. Tenants in Pearce et al.’s (2012) study had lower housing scores than owner-occupiers. However, López-Colás and Módenes (2014) argued that while owners had better residential conditions than other tenures in Norway, Germany, the United Kingdom, Spain, and Poland, more of the variation in Spanish housing quality was explained by house type than by tenure.

In New Zealand, the sole national-level housing survey is the Building Research Association of New Zealand (BRANZ) House Condition Survey. In 2010, this survey for the first time included rental properties among surveyed dwellings. Both the 2010 and the 2015 surveys found rental houses to generally be “in worse condition overall than owner-occupied houses” (Buckett et al., 2012, p. 3) with a higher incidence of components in poor or serious condition. In 2010, “nearly twice as many rented houses were in poor condition compared to owner occupied houses” (Buckett et al., 2012, p. 3), though the gap had narrowed a little in 2015 (Buckett et al., 2012; White et al., 2017). While these surveys were very detailed at the individual dwelling level, they were unfortunately limited by the need for tenant and landlord agreement to participate in the study, and by the need for the property to have a landline, which could potentially exclude most of the worst condition properties.

Few studies have examined the relative contributions of tenure and housing quality to health outcomes. Turunen et al. (2017) found that quality indicators such as mold odor, winter drafts, and older windows were associated with a greater risk for respiratory symptoms. However, while there were differences in health risk by tenure, the difference was not between rental tenants and owner-occupiers, leaving the relative contribution of quality and tenure unclear.

In general, most research has examined differences in housing stock condition between different locations, or different time periods (Moon & Stotsky, 1993), rather than between tenures, and where tenure is looked at, the surveys may exclude those with the worst condition of housing due to the need for consent. An ecological-level study is a simple way to overcome such difficulties.

In this New Zealand study, we aimed to further examine the interrelationships between rental tenure, housing quality, socioeconomic status, and health outcomes using available administrative data.

Method

We investigated whether there was any relationship between rental tenure, housing age or condition, income quintile, and housing-sensitive hospitalization rates, at an ecological (suburb) level, controlling for annual mean minimum temperature and 5-year age-group. We measured associations for all ages and for children aged under 5 years and adults aged 65 years and more.

Area Unit

The unit of measurement was census area unit (CAU), roughly equivalent to suburbs, with a typical population of 2,000 people. In 2006—the central date of the data used—New Zealand had 1,927 CAUs.

Dwelling Age and Condition

We used a data set of 2006 records for 1,084,282 of New Zealand’s estimated 1.4 million residential dwellings held by the property data and analytics company CoreLogic (Telfar Barnard et al., 2015). Records included census meshblocks, which were matched to CAUs. Building age information was available for 1,038,900 dwellings, and a condition rating of “Superior,” “Average,” or “Poor” for 1,023,879 residences, assigned by CoreLogic, based on visual inspection of the exterior of the dwelling. Residence condition was assigned a value of 3 for “Superior” (104,362 properties), 2 for “Average” (855,873 properties), and 1 for “Poor” (50,244 properties). While these units were technically nominal, they were treated as continuous for study purposes. Mean dwelling condition was calculated for each CAU and is reported as “housing condition” at the CAU level, or “dwelling condition” at the individual level.

Rental Proportion

Using StatsNZ 2006 census data, we established the proportion of households in each CAU that were rented. For each CAU, we calculated the “rental proportion,”—that is, the proportion of households renting in a CAU, as “Dwelling not owned by usual residents”/(“Total households in private occupied dwellings” − “Not elsewhere included”).

Health Data

We obtained hospital admission data from the New Zealand Ministry of Health for the period March 1, 2003, to February 28, 2009, 3 years either side of the March 2006 census. From these data, we included only acute hospitalizations for New Zealand residents, and excluded hospitalizations, such as transfers and birth-related admissions, which were not a new or adverse health event, to avoid double counting and noise.

Hospitalizations were aggregated by CAU and 5-year age-group into four categories: total hospitalizations; and three categories of “potentially avoidable hospitalizations” (PAH), also potentially related to housing, which we abbreviate to “housing-sensitive” hospitalizations. These three categories were as follows:

PAHHE: a subset of a set of PAH developed by a panel of child health experts to monitor PAH in New Zealand. This PAHHE subset identified hospitalizations avoidable if central and local government policies “ensured that families with children had access to high-quality housing and a safe physical environment” (Oliver et al., 2018, p. 328).

WIH: “Winter impact hospitalizations” is a set of conditions identified as having strong winter health impact. These were selected from a list of conditions with more than 100 hospitalizations per year over the period 2000 to 2006 and an Excess Winter Hospitalization Index of 1.10 or higher (Telfar Barnard, 2010). From this list were selected, for Māori, Pacific, and the total population, the 20 conditions with the highest absolute winter excess, with the addition of any other conditions with an Excess Winter Hospitalization Index greater than 2.00. These conditions are listed in the appendix.

MoH: a group of conditions, identified in Oliver et al. (2018), and used by the Ministry of Health to target their housing interventions. These conditions target “selected infectious PAH conditions thought to be associated with [acute rheumatic fever] and streptococcal infections, based on expert opinion” (Oliver et al., 2018, p. 328).

We also measured outcomes for “nonhousing-sensitive hospitalizations,” which were defined as hospitalizations that were in none of the categories above, nor in a broader list of potentially housing-sensitive hospitalizations related to crowding (Oliver et al., 2018).

Climate Data

Mean minimum temperature by CAU came from a data set created by New Zealand Landcare Research/Manāki Whenua. This data set was based on New Zealand Meteorological Service temperature data for 346 geographical points. “Estimates of the mean minimum temperature in July, the coldest month of winter, were derived from a surface fitted to monthly estimates of mean daily temperatures” (Leathwick et al., 2002, p. 6).

Population Data

2006 New Zealand Census population data aggregated by CAU and 5-year age-group were used as the exposure variable for hospitalization models.

Socioeconomic Factors

Socioeconomic factors heavily influence both access to quality housing and choice of tenure, so it was important to include socioeconomic measures in the model. The measure most commonly used in health research in New Zealand is the small-area measure of socioeconomic status NZDep (Salmond et al., 2007a). However, NZDep includes tenure as a variable in its construction. We therefore instead used quintile-stratified neighborhood income levels, as measured by the income Index of Multiple Deprivation based on income-assessed tax credits and social welfare payments (Exeter et al., 2017).

Statistical Methods

We used Pearson’s correlation tests to measure the amount of variation in rental proportion explained by meshblock average dwelling condition.

As there was high overdispersion in the hospitalization count data, we used negative binomial regression with robust standard errors. We first modeled statistical relationships between the two domains housing condition, rental proportion, and income quintile and hospitalizations, controlling for annual average minimum temperature and 5-year age-group; and then in the full model included the third-domain income quintile.

Ethics

Ethics consent for the use of anonymized routinely collected administrative health data was granted by the New Zealand Multi-Region Health and Disability Ethics Committee (MEC/06/09/106).

Results

For included properties, the mean rental proportion was 0.317 (95% confidence interval [CI] [0.317, 0.317]), meaning that the average percentage of rental households in included property CAUs was 31.7%. This estimate was a little lower than the 2006 census-sourced national rental proportion of 0.331 (33.1%).

The mean condition for included properties was 2.054 (95% CI [0.053, 0.054]). This figure was only marginally higher than the mean known condition of all properties of 2.052 (95% CI [0.052, 0.053]).

Hospitalization rates for indicator categories are shown in Table 1.

Table 1.

Hospitalization Rates by Category.

Hospitalization category	Hospitalization rate per 1,000 people per year
Hospitalization category	All ages	0-4 years	65+ years
WIH	26.78	80.54	95.21
PAHHE	16.21	70.44	31.55
MoH	5.93	27.67	18.65
Non-housing sensitive	63.01	45.83	138.98
Total hospitalizations	98.09	138.95	250.74

Note. WIH = Winter impact hospitalizations; PAHHE = subset of potentially avoidable hospitalization; MoH = Ministry of Health.

Relationships Between Variables

The data were noisy, but in general, the poorer the condition of a dwelling, the higher the proportion of rental properties in its surrounding CAU (R² = .0404): The CAU rental proportion was higher for poor-condition properties, and lower for superior-condition properties, than for average-condition properties.

The mean rental proportion was 0.26 for the 105,370 “Superior”-condition dwellings; 0.33 for the 864,285 “Average”-condition dwellings; and 0.36 for the 51,473 “Poor”-condition dwellings. The mean rental proportion was significantly higher than the “Average” baseline for “Poor”-condition dwellings (RR = 1.08, 95% CI [1.07, 1.10], p < .001), and significantly lower for “Superior”-condition dwellings (RR = 0.80, 95% CI [0.79, 0.81], p < .001), indicating that there is a strong relationship between dwelling condition and the proportion of dwellings in rental tenure in the surrounding meshblock.

Dwelling condition increased, and rental proportion decreased, as income quintile increased (Table 2).

Table 2.

Dwelling Condition, Rental Proportion, and Census Area Unit Income.

Income quintile	Mean dwelling condition	Mean rental proportion
1 (higher income)	2.10	0.27
2	2.05	0.29
3	2.01	0.29
4	2.01	0.33
5 (lower income)	1.92	0.44

These associations are best illustrated by showing the differences in population distribution across rental proportion and housing condition quintiles, for the highest and lowest income quintiles (Figure 1). At one end of the income spectrum, people living in the highest income quintile were much less likely to live in areas with a high proportion of rentals, or in areas with poor-condition housing. At the other end, the only low-income quintile people who lived in areas with best housing conditions were those in the highest rental areas; and less than half a percent of people in the lowest income quintile lived in areas with the lowest proportion of rentals.

Figure 1.

Study population distribution by housing condition and rental proportion quintiles, for the highest and lowest income quintiles.

Housing-Sensitive Hospitalizations

In the first model (Table 3), which included only housing condition and rental proportion, gradients by rental proportion were steeper than gradients for housing condition; and in the 65 years and over age-group there was no meaningful housing condition gradient in housing-sensitive hospitalizations. First model results for the 0-4 and the 65+ years age-groups have not been shown as they do not meaningfully alter the results.

Table 3.

All-Age Hospitalization Risk Ratios by Housing Condition and Rental Proportion.

Variable (quintile)	Winter	PAHHE	MoH	Nonhousing
Housing condition
1	1.06	1.09	1.13	1.06
2	1.04	1.04	1.00	1.08
3 (baseline)	1.00	1.00	1.00	1.00
4	1.00	0.99	0.98	0.99
5	0.89	0.88	0.84	0.98
Rental proportion
1	0.82	0.82	0.80	0.90
2	0.89	0.89	0.88	0.93
3 (baseline)	1.00	1.00	1.00	1.00
4	1.14	1.13	1.14	1.07
5	1.51	1.54	1.59	1.32

Notes. Housing condition: 1 = worse condition to 5 = better condition. rental quintile: 1 = fewer rentals to 5 = more rentals. Results control for age and annual average minimum temperatures. Values in bold are statistically significant (p < .05). PAHHE = subset of potentially avoidable hospitalization; MoH = Ministry of Health.

With income added into the model (Table 4), the housing condition gradient disappeared, the rental proportion gradient flattened, and the income gradient took primary importance, except in the 65+ years age-group, where the rental proportion gradient also remained.

Table 4.

All-Age Hospitalization Risk Ratios, Full Model.

Variable (quintiles)	Age-group
	All ages				0-4 years				65+ years
	Winter	PAHHE	MoH	Nonhousing	Winter	PAHHE	MoH	Nonhousing	Winter	PAHHE	MoH	Nonhousing
Housing condition
1	0.98	1.00	1.02	1.02	1.00	1.03	1.18	1.00	0.86	0.84	0.83	0.84
2	1.01	1.00	0.95	1.06	1.01	1.02	1.06	1.02	0.89	0.85	0.82	0.90
3 (baseline)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
4	1.06	1.05	1.06	1.02	1.04	1.04	1.10	1.02	1.06	1.02	1.02	1.04
5	1.04	1.02	1.02	1.06	1.09	1.09	1.12	1.10	1.08	1.05	1.03	1.16
Rental proportion
1	1.03	1.03	1.02	1.03	1.15	1.12	1.01	1.15	0.96	1.00	1.00	0.98
2	0.99	0.99	1.00	0.98	1.06	1.04	1.03	1.06	0.94	0.95	0.98	0.93
3 (baseline)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
4	1.09	1.08	1.07	1.05	0.97	0.97	1.00	0.96	1.16	1.17	1.16	1.15
5	1.41	1.43	1.42	1.30	1.11	1.13	1.22	0.98	1.60	1.72	1.74	1.52
Income
1	0.76	0.79	0.77	0.86	0.88	0.85	0.74	0.98	0.80	0.88	0.87	0.84
2	0.83	0.85	0.83	0.88	0.91	0.90	0.85	1.00	0.85	0.90	0.90	0.87
3 (baseline)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
4	1.26	1.28	1.29	1.17	1.27	1.29	1.32	1.16	1.16	1.17	1.18	1.11
5	1.48	1.54	1.70	1.20	1.61	1.66	1.92	1.28	1.19	1.21	1.24	1.02

Note. Housing condition: 1 = worse condition to 5 = better condition; rental: 1 = fewer rentals to 5 = more rentals; income: 1 = higher income to 5 = lower income. Results control for age and annual average minimum temperatures. Values in bold are statistically significant (p < .05). PAHHE = subset of potentially avoidable hospitalization; MoH = Ministry of Health.

Discussion

Housing-sensitive hospitalizations showed a strong gradient for income quintile, which swallowed most of the gradient for rental proportion, and all of the gradient for housing condition, except in adults aged more than 65 years, for whom the rental proportion gradient remained, and remained stronger than for income quintile.

We note that New Zealanders aged 65 years and over qualify for universal superannuation, which may mean they are less susceptible to the effects of income than those who have not yet received superannuation. At the same time, home ownership rates among older adults are much higher than for younger age-groups, so the proportion of rentals in a neighborhood is likely to be a weaker marker for individual home ownership among adults aged over 65 years than for younger ages.

Previous research has proposed the inverse housing law (Blane et al., 2000), under which those who are most vulnerable, and thus have most need for good housing, are the least likely to have it. The results of this study support that hypothesis in part, in that poor dwelling condition was associated with lower CAU income.

Housing condition, rental tenure, and income are well established in the broad range of health determinants, and health determinants are recognized as being interdependent, so the relationships between average housing condition, proportion of rentals, and income quintile are no real surprise. Nonetheless, they are an important starting point for considering housing and health relationships. Finding that rented housing is in poorer condition than owner-occupier housing complicates the association between rental tenure and health outcomes: Is ill health among tenants due to the set of social circumstances for which tenure acts as a marker, or is it a direct physiological response to suboptimal environmental conditions? If the main risk is the dwelling condition, does rental tenure still have a residual adverse health effect? And do either of these factors have a health effect beyond the known association between health inequities and income differences?

The division of burden matters because the interventions are different. If being in rental tenure is on its own a health risk, then interventions need to improve avenues to home ownership, and discover and address the factors that make rental tenure a health risk, whether they be lack of security of tenure, neighborhood insecurity, lack of sense of autonomy, or some other unidentified problem. However, if dwelling condition is the main driver of poorer health in rental tenure, then given the growing evidence that interventions to improve housing condition and/or energy efficiency can improve health (Howden-Chapman et al., 2007; Howden-Chapman et al., 2008; Jackson et al., 2011), interventions must improve the housing stock overall.

If the health burden is relatively evenly split between tenure and condition, then it may make sense to target interventions at the condition of rental dwellings in particular, while also looking to increase owner-occupation rates. Targeted interventions make sense because, on average, renters are of lower socioeconomic status than owner-occupiers and therefore generally have additional health risks compounding the risks from both their tenure and their dwelling condition, and because renters have the least control over the condition of their housing, in part because rental housing is not a standard competitive market (Gilderbloom et al., 2008).

If income differences explain tenure or housing quality differences in housing-sensitive health outcomes, interventions might also include economic instruments addressing housing affordability or fuel poverty.

Finally, where research measuring these health and health determinant interactions are at risk of including reverse causation or unmeasured confounders, analysis and solutions should look to holistic models that address income, tenure, and housing condition together.

To the extent that dwelling condition is the, or a, problem, legislative intervention may also be needed due to the ongoing problem of the split incentive between health and energy efficiency improvements to rental dwellings (de T’Serclaes & Jollands, 2007), and due to landlord reluctance to improve their rented properties even when subsidies are available (Saville-Smith, 2008).

This study’s findings for hospitalization suggest that income inequity is ingrained in the relationship among housing tenure, housing quality, and their respective contributions to the health burden. Both rental tenure and housing condition were implicated in poor health outcomes, but income inequalities were identified as an important underlying factor, most likely as the driver for people’s access to quality housing and home ownership: Housing quality and tenure matter, but people’s access to both is dictated by income. These findings also provide further evidence for income, housing quality, and tenure as health determinants. Following our argument above, the findings suggest that when considering interventions to improve housing-sensitive health outcomes, policymakers should target those on low incomes, improving the affordability of energy and quality housing and ensuring that affordable dwellings meet basic quality standards. Policymakers could also enhance the effectiveness of such interventions by considering their interactions with other social determinants of health, such as public housing supply, “education, employment, transport, child care, health systems, taxation, wages, benefit levels and job security” (World Health Organization, 2018, p. 11). Additional interventions are needed in these areas to ensure that people can afford healthy housing.

Limitations

CoreLogic ratings for overall dwelling condition are generally more favorable than standardized assessment-based ratings assigned by the BRANZ to the sample in relevant House Condition Surveys (BRANZ, 2005; Buckett et al., 2011). The CoreLogic condition variable is also crude, with only three categories for all houses. Therefore, the results from this study are likely to underestimate any association between rental proportion and dwelling condition. Nevertheless, although CoreLogic condition data for any individual house may be roughly measured, or out of date, on a national scale they still provide a useful measure.

Also, this study used only ecological data. It is possible that in any given CAU, the rentals are the better-quality dwellings. However, while we cannot say definitively that rental dwellings are in poorer condition than owner-occupier dwellings, it is a reasonable working assumption.

Conclusion

While housing-sensitive hospitalizations showed gradients for both housing condition and rental proportion, the rental proportion gradient was reduced, and the housing condition gradient eliminated, by adding income quintiles to the model. These findings suggest that while measures to address health inequities could usefully include improvements to both tenure security and housing quality, particularly in low-income areas, policymakers aiming to reduce housing-sensitive hospitalization rates should focus on reducing fuel poverty and improving the affordability of quality housing.

Footnotes

Appendix

Appendix.

Winter-Impact Hospitalizations.

ICD-10 code	Title
A08	Viral and other specified intestinal infections
A09	Other gastroenteritis and colitis of infectious and unspecified origin
A39	Meningococcal infection
B34	Viral infection of unspecified site
H66	Suppurative and unspecified otitis media
I21	Acute myocardial infarction
I50	Heart failure
J00	Acute nasopharyngitis
J05	Acute obstructive laryngitis (croup) and epiglottitis
J06	Acute upper-respiratory infections of multiple and unspecified sites
J10	Influenza due to identified influenza virus
J11	Influenza, virus not identified
J12	Viral pneumonia, not elsewhere classified
J13	Pneumonia due to Streptococcus pneumoniae
J14	Pneumonia due to Haemophilus influenzae
J18	Pneumonia, organism unspecified
J20	Acute bronchitis
J21	Acute bronchiolitis
J22	Unspecified acute lower-respiratory infection
J40	Bronchitis, not specified as acute or chronic
J44	Emphysema
J45	Asthma
J46	Status asthmaticus
J47	Bronchiectasis
R04	Hemorrhage from respiratory passages
R51	Headache
R55	Syncope and collapse
R56	Convulsions, not elsewhere classified
S73	Dislocation, sprain, and strain of joint and ligaments of hip
T68	Hypothermia

Note. ICD-10 = 10th revision of the International Classification of Diseases.

Acknowledgements

The authors acknowledge advice provided by CoreLogic New Zealand and data and advice provided by Dr. Daniel Exeter of the University of Auckland Index of Multiple Deprivation research team.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the New Zealand Health Research Council [Ref: 01/365].

ORCID iDs

Lucy Telfar Barnard

Philippa Howden-Chapman

Nevil Pierse

References

Baker

Bentley

Mason

(2013). The mental health effects of housing tenure: Causal or compositional? Urban Studies, 50(2), 426-442. https://doi.org/10.1177/0042098012446992

Baker

Lester

L. H.

Bentley

Beer

(2016). Poor housing quality: Prevalence and health effects AU—Baker, Emma. Journal of Prevention & Intervention in the Community, 44(4), 219-232. https://doi.org/10.1080/10852352.2016.1197714

Blane

Mitchell

Bartley

(2000). The “inverse housing law” and respiratory health. Journal of Epidemiology & Community Health, 54, 745-749. https://doi.org/10.1136/jech.54.10.745

Buckett

N. R.

Jones

M. S.

Marston

N. J.

(2012). BRANZ 2010 house condition survey: Condition comparison by tenure. BRANZ.

Buckett

N. R.

Marston

N. J.

Saville-Smith

Jowett

J. H.

Jones

M. S.

(2011). Preliminary BRANZ 2010 house condition survey report (2nd ed.). BRANZ.

Building Research Association of New Zealand. (2005). New Zealand 2005 House Condition Survey.

Clinch

J. P.

Healy

J. D.

(2000). Housing standards and excess winter mortality. Journal of Epidemiology & Community Health, 54, 719-720. https://doi.org/10.1136/jech.54.9.719

Darlington-Pollock

Norman

(2017). Examining ethnic inequalities in health and tenure in England: A repeated cross-sectional analysis. Health Place, 46, 82-90. https://doi.org/10.1016/j.healthplace.2017.04.011

Davis

Jenkin

Coope

(2003). New Zealand Socio-Economic Index 1996. An update and revision of the New Zealand Socio-Economic Index of Occupational Status. Statistics New Zealand, Wellington.

10.

de T’Serclaes

Jollands

. (2007). Mind the gap: Quantifying principal-agent problems in energy efficiency. International Energy Agency and the Organisation for Economic Cooperation and Development, Paris.

11.

Ellaway

Macintyre

(1998). Does housing tenure predict health in the UK because it exposes people to different levels of housing related hazards in the home or its surroundings? Health Place, 4(2), 141-150. https://doi.org/10.1016/S1353-8292(98)00006-9

12.

Exeter

D. J.

Zhao

Crengle

Lee

Browne

(2017). The New Zealand Indices of Multiple Deprivation (IMD): A new suite of indicators for social and health research in Aotearoa, New Zealand. PLOS One, 12(8), e0181260. https://doi.org/10.1371/journal.pone.0181260

13.

Gibson

Petticrew

Bambra

Sowden

A. J.

Wright

K. E.

Whitehead

(2011). Housing and health inequalities: A synthesis of systematic reviews of interventions aimed at different pathways linking housing and health. Health Place, 17(1), 175-184. https://doi.org/10.1016/j.healthplace.2010.09.011

14.

Gilderbloom

J. I.

Pan

Lehman

Roosa

S. A.

Appelbaum

R. P.

(2008). Why rents rise. In Gilderbloom

J. I.

(Ed.), Invisible city: Poverty, housing, and new urbanism (pp. 47-66). University of Texas Press.

15.

Hales

Blakely

Foster

R. H.

Baker

M. G.

Howden-Chapman

(2010). Seasonal patterns of mortality in relation to social factors. Journal of Epidemiology & Community Health. https://doi.org/10.1136/jech.2010.111864

16.

Hiscock

Macintyre

Kearns

Ellaway

(2003). Residents and residence: Factors predicting the health disadvantage of social renters compared to owner-occupiers. Journal of Social Issues, 59(3), 527-546. https://doi.org/10.1111/1540-4560.00076

17.

Howden-Chapman

P. L.

Chandola

Stafford

Marmot

(2011). The effect of housing on the mental health of older people: The impact of lifetime housing history in Whitehall II. BMC Public Health, 11, 682. https://doi.org/10.1186/1471-2458-11-682

18.

Howden-Chapman

P. L.

Matheson

Crane

Viggers

Cunningham

Blakely

Cunningham

Woodward

Saville-Smith

O’Dea

Kennedy

Baker

Waipara

Chapman

Davie

(2007). Effect of insulating existing houses on health inequality: Cluster randomized study in the community. British Medical Journal, 334, 460. https://doi.org/10.1136/bmj.39070.573032.80

19.

Howden-Chapman

P. L.

Pierse

Nicholls

Gillespie-Bennett

Viggers

Cunningham

Phipps

Boulic

Fjällström

Free

Chapman

Lloyd

Wickens

Shields

Baker

Cunningham

Woodward

Bullen

Crane

(2008). Effects of improved home heating on asthma in community dwelling children: Randomised controlled trial. British Medical Journal, 337, a1411. https://doi.org/10.1136/bmj.a1411

20.

Iwata

Yamaga

(2008). Rental externality, tenure security, and housing quality. Journal of Housing Economics, 17(3), 201-211. https://doi.org/10.1016/j.jhe.2008.06.002

21.

Jackson

Thornley

Woolston

Papa

Bernacchi

Moore

(2011). Reduced acute hospitalization with the healthy housing programme. Journal of Epidemiology & Community Health, 65(7), 588-593. https://doi.org/10.1136/jech.2009.107441

22.

Jacobs

D. E.

Wilson

Dixon

S. L.

Smith

Evens

(2009). The relationship of housing and population health: A 30-year retrospective analysis. Environmental Health Perspectives, 117(4), 597-604. https://doi.org/10.1289/ehp.0800086

23.

Kavanagh

A. M.

Aitken

Baker

LaMontagne

A. D.

Milner

Bentley

(2016). Housing tenure and affordability and mental health following disability acquisition in adulthood. Social Science & Medicine, 151, 225-232. https://doi:10.1016/j.socscimed.2016.01.010

24.

Keall

Baker

M. G.

Howden-Chapman

Cunningham

Ormandy

(2010). Assessing housing quality and its impact on health, safety and sustainability. Journal of Epidemiology & Community Health, 64(9), 765-771. https://doi.org/10.1136/jech.2009.100701

25.

Krieger

Higgins

(2002). Housing and health: Time again for public health action. American Journal of Public Health, 92(5), 758-768. https://doi.org/10.2105/AJPH.92.5.758

26.

Lane

Kinsey

(1980). Housing tenure status and housing satisfaction. Journal of Consumer Affairs, 14(2), 341-366. https://doi.org/10.1111/j.1745-6606.1980.tb00674.x

27.

Leathwick

Morgan

Wilson

Rutledge

McLeod

Johnston

(2002). Land environments of New Zealand: A technical guide. Ministry for the Environment. https://www.landcareresearch.co.nz/__data/assets/pdf_file/0020/21773/LENZ_Technical_Guide.pdf

28.

López-Colás

Módenes

J. A.

(2014). Sociodemographic heterogeneity in housing quality: Spain in European comparison. Boletín de la Asociación de Geógrafos Españoles N., 66, 505-508. https://bage.age-geografia.es/ojs/index.php/bage/article/view/1810

29.

Macintyre

Ellaway

Der

Ford

Hunt

(1998). Do housing tenure and car access predict health because they are simply markers of income or self-esteem? A Scottish study. Journal of Epidemiology & Community Health, 52, 657-664. https://doi.org/10.1136/jech.52.10.657

30.

Macintyre

Ellaway

Hiscock

Kearns

Der

McKay

(2003). What features of the home and the area might help to explain observed relationships between housing tenure and health? Evidence from the west of Scotland. Health & Place, 9(3), 207-218. https://doi.org/10.1016/S1353-8292(02)00040-0

31.

Macintyre

Hiscock

Kearns

(2001). Housing tenure and health inequalities: A three-dimensional perspective on people, homes and neighbourhoods. In Graham

(Ed.), Understanding health inequalities (pp. 129–142). Open University Press.

32.

Markham

J. P.

Gilderbloom

J. I.

(1998). Housing quality among the elderly: A decade of changes. International Journal of Ageing and Human Development, 46(1), 71-90. https://doi.org/10.2190/A6EQ-FY4C-F4CF-18G9

33.

Moon

C.-G.

Stotsky

J. G.

(1993). The effect of rent control on housing quality change: A longitudinal analysis. Journal of Political Economy, 101(6), 1114-1148. https://doi.org/10.1086/261917

34.

Muntaner

Eaton

W. W.

Diala

Kessler

R. C.

Sorlie

P. D.

(1998). Social class, assets, organizational control and the prevalence of common groups of psychiatric disorders. Social Science & Medicine, 47(12), 2043-2053. https://doi.org/10.1016/S0277-9536(98)00309-8

35.

Oliver

Foster

Kvalsvig

Williamson

D. A.

Baker

M. G.

Pierse

(2018). Risk of rehospitalisation and death for vulnerable New Zealand children. Archives of Disease in Childhood, 103(4), 327-334. https://doi.org/10.1136/archdischild-2017-312671

36.

Pearce

Abbas

Ferguson

Graham

Law

(2012). Does the home environment influence inequalities in unintentional injury in early childhood? Findings from the UK Millennium Cohort Study. Journal of Epidemiology & Community Health, 66(2), 181-188. https://doi.org/10.1136/jech.2011.139626

37.

Pekkonen

Haverinen-Shaughnessy

(2015). Housing satisfaction in Finland with regard to area, dwelling type and tenure status. Central European Journal of Public Health, 23(4), 314-320. https://doi.org/10.21101/cejph.a4080

38.

Pledger

McDonald

Dunn

Cumming

Saville-Smith

(2019). The health of older New Zealanders in relation to housing tenure: Analysis of pooled data from three consecutive, annual New Zealand Health surveys. Australian and New Zealand Journal of Public Health, 43(2), 182-189. Doi:10.1111/1753-6405.12875

39.

Popham

Williamson

Whitley

(2015). Is changing status through housing tenure associated with changes in mental health? Results from the British household panel survey. Journal of Epidemiology & Community Health, 69, 6-11. https://doi.org/10.1136/jech-2014-203990

40.

Salmond

Crampton

Atkinson

(2007a). NZDep2006 Index of Deprivation. Department of Public Health, University of Otago. http://www.wnmeds.ac.nz/academic/dph/research/NZDep/NZDep2006%20research%20report%2004%20September%202007.pdf

41.

Salmond

Crampton

Atkinson

(2007b). NZDep2006 Index of Deprivation user’s manual. Department of Public Health, University of Otago Wellington. http://www.wnmeds.ac.nz/academic/dph/research/NZDep/Instruction%20book%202006%20-%2004%20September%202007.pdf

42.

Saville-Smith

(2008). Appendix C: Landlords energy working paper: Preliminary analysis of the telephone survey data (Report EN-6570). Beacon Pathway.

43.

Stewart

Prince

Harwood

Whitley

Mann

(2002). Quality of accommodation and risk of depression in later life: An analysis of prospective data from the Gospel Oak Project. International Journal of Geriatric Psychiatry, 17(12), 1091-1098. https://doi.org/10.1002/gps.749

44.

Sundquist

Johansson

S. E.

(1997). Indicators of socio-economic position and their relation to mortality in Sweden. Social Science & Medicine, 45(12), 1757-1766. https://doi.org/10.1016/S0277-9536(97)00107-X

45.

Telfar Barnard

L. F

. (2010). Home truths and cool admissions: New Zealand housing attributes and excess winter hospitalisation [Unpublished doctoral dissertation]. University of Otago. https://ourarchive.otago.ac.nz/handle/10523/591

46.

Telfar Barnard

L. F.

Baker

M. G.

Hales

Howden-Chapman

. (2015). Novel use of three administrative datasets to establish a cohort for environmental health research. BMC Public Health, 15, 246. https://doi.org/10.1186/s12889-015-1580-1

47.

Turunen

Iso-Markku

Pekkonen

(2017). Statistical associations between housing quality and health among Finnish households with children: Results from two (repeated) national surveys. Science of the Total Environment, 574, 1580-1587. https://doi.org/10.1016/j.scitotenv.2016.08.194

48.

Webb

Blane

de Vries

(2013). Housing and respiratory health at older ages. Journal of Epidemiology & Community Health, 67(280-285). https://doi.org/10.1136/jech-2012-201458

49.

White

Jones

Cowan

Chun

(2017, April). BRANZ 2015 house condition survey: Comparison of house condition by tenure. BRANZ. https://www.branz.co.nz/pubs/research-reports/sr370/

50.

Wilkinson

(1999). Poor housing and ill health: A summary of research evidence. Housing Research Branch, The Scottish Office Central Research Unit. https://www2.gov.scot/resource/doc/156479/0042008.pdf

51.

World Health Organization. (2018). WHO housing and health guidelines.