We have a new website go to gov.scot

The Scottish Executive Central Heating Programme: Assessing Impacts on Health

Listen

SECTION 4 THE CENTRAL HEATING PROGRAMME AND SPECIFIC OUTCOME AREAS

4.1 Introduction

This section provides results for the individual outcome measures examined by the evaluation. Measures are grouped by conceptual area (for example, Section 4.2 reports on outcomes relating to perceptions of warmth in the home). For each outcome area, the following information is provided:

1) A key message statement, providing a non-technical interpretation of the findings.

2) A summary, briefly defining the area and listing those outcome measures within it which were found to be significantly associated with the receipt of heating under the CHP.

3) For the measures listed in (2) - that is, those which exhibited a statistically significant relationship with the CHP - one or more illustrative charts demonstrating the size of the effect associated with the Programme.

4) One or more tables showing numerical results for the significant outcome measures.

5) Where appropriate, a list of the remaining outcome measures within the area reported on; that is, those measures found not to have a statistically significant association with the Programme.

The material in the following sections makes some use of a small number of technical statistical terms which may not be familiar to all readers. Two key terms ('odds ratio' and 'confidence interval') have previously been introduced in Section 1. One further statistical concept - that of the p value - is described in Section 4.2. Explanations of these, and of other statistical terms used in this Report, are featured in the G lossary of Statistical Terms.

4.2 Perceptions of warmth in the home

KEY MESSAGE: Central Heating Programme recipients perceived their homes to be warmer in winter, indicated that their heating was less likely to be a serious problem and reported that they were more satisfied with their heating overall, relative to those who were not part of the CHP.

SUMMARY: There were three measures included in this area, providing different perspectives on the adequacy of domestic heating. For all three, there was a clear impact of the CHP. The three measures were:

1) whether the respondent is kept sufficiently warm by her / his heating during the period October - March 7

2) whether inadequate heating in the respondent's home presents a serious problem

3) whether the respondent is satisfied with her / his heating.

CHARTS: Chart 4.2.1 shows the differences between respondents who were part of the CHP and those who were not; results are presented in the form of odds ratios. In this chart, the black rectangles represent the estimated odds ratios for the effect of the CHP, the horizontal lines show the 95% confidence intervals for the odds ratios, and the dashed vertical line marks the odds ratio of one which indicates no difference between the groups. Although all three results are shown on a single chart, the individual estimates should not be directly compared with each other, because they represent responses to different questions. Chart 4.2.2 shows the proportions (percentages) of respondents who report the condition ( e.g. being kept warm enough) at the end of the evaluation. It should be stressed that the differences between the two groups ( i.e. heating recipients and the comparison group) do not necessarily match those indicated by the odds ratios shown in Chart 4.2.1 and in Table 4.2.1. This is because the absolute values shown in Chart 4.2.2 represent the 'raw' or unadjusted values of the measures for each group at the evaluation endpoint, while the odds ratio estimates are adjusted for the effect of other factors (see Appendix B; Section B.1). The estimated odds ratios, incorporating adjustment, provide a better approximation to the true effect of the CHP. This principle holds for all the results provided in this section of the Report.

Chart 4.2.1 Differences in perceptions of warmth in the home (odds ratios)

image of Chart 4.2.1 Differences in perceptions of warmth in the home (odds ratios)

Chart 4.2.2 Perceptions of warmth in the home (comparison group versus recipients)

image of Chart 4.2.2 Perceptions of warmth in the home (comparison group versus recipients)

RESULTS: Table 4.2.1 gives the results numerically. The odds are for CHP recipients, relative to the comparison group. The column headed '95% CI' shows the 95% confidence interval for the odds ratio estimate; that headed ' p' gives the associated p value; and the column headed ' n' shows the number of responses on which the result was based. (The p value returned by a statistical test is an indication of how likely the result is to have arisen by chance, as distinct from being due to some real effect. The smaller the p value, the greater the likelihood that a real effect has been observed - see the Glossary of Statistical Terms.) With regard to the latter, the n value for the middle row of the table is markedly lower than those for the other rows; this is because the question on which the 'serious problem' result is based was presented only to a small subset of respondents. This situation occurs elsewhere in these results.

Table 4.2.1 Differences in perceptions of warmth in the home (odds ratios)

measure

odds ratio

95% CI

p

n

Whether kept warm enough during Oct.-Mar.

3.50

2.85 to 4.29

< 0.01

2,289

Whether heating a serious problem

0.48

0.29 to 0.81

< 0.01

228

Whether satisfied with heating

4.96

3.87 to 6.37

< 0.01

2,323

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP: None

4.3 Patterns of heating in the home

KEY MESSAGE: Those who received central heating under the CHP reported that in general more of the home was heated, and for longer, than was the case for the comparison group. The only exception to this pattern was that recipients reported heating the living room in cold weather for a shorter time than comparison households. This should not automatically be interpreted as a 'negative' or adverse finding, as it is possible to explain this result in a way which is consistent with a positive effect of the CHP. For example, it may be that recipients were more able to maintain uniformly comfortable levels of heating throughout the home, and therefore had less need to retain a concentration of heat in the dwelling's living room.

SUMMARY: This area included 10 measures representing various aspects of the duration and extent of domestic heating use in cold weather. A clear impact of the CHP was identified for all ten measures, which were:

1) whether more than half of the rooms in the home are permanently unheated in cold weather

2) whether more than half of the rooms in the home are permanently heated (24 hours per day) in cold weather

3) whether more than half of the rooms in the home are heated for nine hours per day or less in cold weather

4) the duration (hours per day) for which the dwelling's kitchen is heated in cold weather

5) the duration (hours per day) for which the dwelling's bathroom is heated in cold weather

6) the duration (hours per day) for which the dwelling's main living room is heated in cold weather

7) the duration (hours per day) for which the dwelling's hall is heated in cold weather

8) the duration (hours per day) for which the dwelling's main bedroom is heated in cold weather

9) the duration (hours per day) for which the dwelling's second bedroom (where extant) is heated in cold weather

10) the average duration (hours per day) for which rooms in the dwelling are heated in cold weather

CHARTS: Chart 4.3.1 illustrates, for measures (1) to (3) above, the differences between CHP recipients and comparison group respondents. Results are presented as odds ratios, and the format of the chart is identical to that of Chart 4.2.1. Chart 4.3.2 shows the results for the remaining measures listed above. In this chart, the small black rectangles represent the estimated differences in heating duration (hours per day) between recipients and comparison households, the horizontal lines show the 95% confidence intervals for the differences, and the dashed vertical line marks the value of zero which indicates no difference between the groups 8.

Chart 4.3.1 Differences in patterns of heating in the home (odds ratios)

image of Chart 4.3.1 Differences in patterns of heating in the home (odds ratios)

Chart 4.3.2 Differences in duration of heating in the home

image of Chart 4.3.2 Differences in duration of heating in the home

Chart 4.3.3 shows, for measures (1) to (3) above, the proportions (percentages) of respondents who report each condition ( e.g. maintaining more than half of the rooms in the home unheated) at the end of the evaluation. Finally, Chart 4.3.4 shows average values at the end of the evaluation for the remaining measures listed above.

Chart 4.3.3 Patterns of heating in the home (comparison group versus recipients)

image of Chart 4.3.3 Patterns of heating in the home (comparison group versus recipients)

Chart 4.3.4 Duration of heating different rooms in the home (comparison group versus recipients)

image of Chart 4.3.4 Duration of heating different rooms in the home (comparison group versus recipients)

RESULTS: Table 4.3.1 gives the results numerically. The first three rows show odds ratio estimates and associated values for measures (1) to (3) above. The format of these rows is identical to that used in Table 4.2.1. The remaining rows in Table 4.3.1 provide results for the other measures listed in the Summary above. These rows show, for each measure, the estimated average difference in heating duration (in hours per day). A value greater than zero indicates that central heating recipients maintain heating for longer than do comparison group respondents. Each estimate is followed by its 95% confidence interval, the p value associated with the estimate, and the number of responses on which the result is based.

Table 4.3.1 Differences in patterns of heating in the home

measure

odds ratio

95% CI

p

n

Whether more than half of rooms are permanently unheated in cold weather

0.22

0.16 to 0.29

< 0.01

2,149

Whether more than half of rooms are permanently heated in cold weather

1.28

1.04 to 1.58

<0.05

2,149

Whether more than half of rooms are heated for 9 hours per day or less in cold weather

0.79

0.68 to 0.91

< 0.01

2,149

measure

Estimate (hours / day)

95% CI

p

n

Duration (hours / day) of heating KITCHEN

1.11

0.51 to 1.71

< 0.01

2,213

Duration (hours / day) of heating BATHROOM

2.37

1.74 to 3.00

< 0.01

2,270

Duration (hours / day) of heating LIVING ROOM

-1.15

-1.70 to -0.60

< 0.01

2,249

Duration (hours / day) of heating HALL

1.38

0.75 to 2.02

< 0.01

2,203

Duration (hours / day) of heating MAIN BEDROOM

1.58

0.95 to 2.21

< 0.01

2,249

Duration (hours / day) of heating 2 nd BEDROOM

1.72

1.01 to 2.42

< 0.01

1,874

Average duration (hours / day) of heating ALL

1.12

0.60 to 1.64

< 0.01

2,149

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP: None

4.4 Condensation, dampness and mould

KEY MESSAGE: The prevalence of poor environmental conditions - specifically, the presence of condensation, dampness and / or mould - in individual rooms within the home was significantly lower for those who received heating under the CHP than for the comparison group. Recipients were also less likely than comparison respondents to avoid the use of rooms due to difficulty in heating them, or to problems of damp or condensation.

SUMMARY: A total of nine measures was included in this area, each representing some aspect of the extent to which the respondent's home was affected by adverse environmental conditions (specifically, problems such as dampness, mould and condensation). Definitions of the measures are given in Appendix B (Section B.4). A statistically significant impact of the CHP was observed for all nine measures, which were as follows:

1) whether any environmental problem (condensation / damp /mould) is present in the dwelling's kitchen

2) as above, for the bathroom

3) as above, for the main living room

4) as above, for the hall

5) as above, for the main bedroom

6) as above, for the second bedroom

7) whether use of any room(s) in the home is avoided due to difficulty in heating

8) whether environmental problems (condensation / damp / mould) cause serious difficulty

9) whether any room(s) in the home cannot be used due to problems of damp or condensation

CHARTS: Chart 4.4.1 shows the observed differences between CHP recipients and comparison households for measures (1) to (6) above. The results are presented as odds ratios with 95% confidence intervals, and the format of the chart is identical to that of Chart 4.2.1. Chart 4.4.2 shows the results for the remaining measures (7) to (9), again in the form of odds ratios. Chart 4.4.3 shows, for measures (1) to (6) above, the proportions (percentages) of respondents who report the condition ( e.g. experiencing environmental problems in the kitchen) at the end of the evaluation. Finally, Chart 4.4.4 shows the same information for the remaining measures listed above.

Chart 4.4.1 Differences in poor environmental conditions (odds ratios)

image of Chart 4.4.1 Differences in poor environmental conditions (odds ratios)

Chart 4.4.2 Differences in impact of poor environmental conditions (odds ratios)

image of Chart 4.4.2 Differences in impact of poor environmental conditions (odds ratios)

Chart 4.4.3 Poor environmental conditions (comparison group versus recipients)

image of Chart 4.4.3 Poor environmental conditions (comparison group versus recipients)

Chart 4.4.4 Impacts of poor environmental conditions (comparison group versus recipients)

image of Chart 4.4.4 Impacts of poor environmental conditions (comparison group versus recipients)

RESULTS: Table 4.4.1 presents the results numerically, in the form of odds ratios.

Table 4.4.1 Differences in poor environmental conditions and impact of poor environmental conditions (odds ratios)

Measure

odds ratio

95% CI

p

n

Whether any environmental problem(s) present in KITCHEN

0.49

0.39 to 0.61

< 0.01

2,291

Whether any environmental problem(s) present in BATHROOM

0.46

0.37 to 0.59

< 0.01

2,302

Whether any environmental problem(s) present in MAIN LIVING ROOM

0.59

0.46 to 0.75

< 0.01

2,305

Whether any environmental problem(s) present in HALL

0.56

0.40 to 0.79

< 0.01

2,271

Whether any environmental problem(s) present in MAIN BEDROOM

0.48

0.39 to 0.61

< 0.01

2,297

Whether any environmental problem(s) present in SECOND BEDROOM

0.44

0.34 to 0.57

< 0.01

1,939

Whether use of any room(s) in the home is avoided due to difficulty in heating

0.43

0.31 to 0.59

< 0.01

2,330

Whether environmental problems cause serious difficulty

0.52

0.31 to 0.86

<0.01

297

Whether any rooms in the home cannot be used due to problems of damp or condensation

0.39

0.15 to 1.00

<0.05

300

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP: None

4.5 Overall satisfaction with the home

KEY MESSAGE: Recipients of heating under the Programme expressed greater attachment to their homes (via strong disagreement with the concept of home as a place to "get away from") than the comparison group. However, this result was apparently contradicted by a second finding to the effect that recipients were less likely than comparison respondents not to wish to move home. While this is clearly paradoxical, the general measures of attachment to the home which were investigated are open to influence by many factors beyond the adequacy of the dwelling's heating ( e.g. the quality of the physical environment, noisy / hostile neighbours etc). It is arguably unwise to interpret these findings purely as indications of the impact of the CHP.

SUMMARY: Five measures were included in this area, providing different perspectives on respondents' overall levels of satisfaction and contentment with their homes. For one of these five measures - the perception of home as "a place I want to get away from" - recipients of heating reported a greater degree of attachment to the home than did those in the comparison group. For one further item, which asked whether respondents would move home if they were able to do so, recipients were significantly less likely than comparison respondents to provide a NO response. The remaining three measures did not show significant associations with the Programme.

CHARTS: Chart 4.5.1 shows results for the two measures which were significantly associated with the CHP. Results are expressed as odds ratios. The first item ("a place I want to get away from") is represented in analysis such that an odds ratio of greater than one indicates that recipients are more likely to express strong disagreement with the statement (that is, to display greater attachment to their homes) than comparison respondents. The definition of this measure is explained in Appendix B (Section B.5.2).

Chart 4.5.1 Differences in overall satisfaction with the home (odds ratios)

image of Chart 4.5.1 Differences in overall satisfaction with the home (odds ratios)

Chart 4.5.2 shows, for the two measures featured in Chart 4.5.1, the respective proportions of respondents who report each condition at the end of the evaluation.

Chart 4.5.2 Overall satisfaction with the home (comparison group versus recipients)

image of Chart 4.5.2 Overall satisfaction with the home (comparison group versus recipients)

RESULTS: Table 4.5.1 shows numeric results (odds ratios).

Table 4.5.1 Differences in overall satisfaction with the home (odds ratios)

measure

odds ratio

95% CI

P

n

Whether respondent indicates strong disagreement with the perception of home as "A place I want to get away from"

1.19

1.03 to 1.37

0.02

2,322

Whether respondent would not move home if able to do so.

0.83

0.69 to 0.99

0.04

2,207

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP: Three measures within the area did not exhibit significant associations with the receipt of central heating under the CHP. These measures were (see Appendix B, Section B.5 for full definitions):

1) the respondent's overall level of satisfaction with her/his home

2) the respondent's perception of home as "A place where I feel safe"

3) the respondent's perception of home as "A place where I feel at home"

Thus, there were no statistically significant differences between recipients and comparison group with respect to level of satisfaction with their home, or the perception of the home as place of safety and ease.

4.6 Drinking and smoking

KEY MESSAGE: Receipt of central heating under the Programme exerted no significant influence on respondents' drinking or smoking behaviour.

SUMMARY: This area consisted of two measures, representing respondents' recent consumption of alcohol and their current smoking status. Neither measure exhibited a significant association with the receipt of central heating under the CHP.

CHARTS: None shown, as no statistically significant results were found.

RESULTS: None shown.

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP:

1) whether the respondent has consumed an alcoholic drink within the last seven days

2) whether the respondent currently smokes cigarettes

4.7 Nature and extent of social contacts

KEY MESSAGE: Those receiving heating under the Programme were less likely than the comparison group to dissuade friends or relatives from staying overnight, or from visiting, due to poor housing conditions such as dampness or cold.

SUMMARY: There were four measures included in this area, representing aspects of respondents' recent contacts with friends and / or relatives. Of these measures, two were found to be associated with a significant impact of the CHP. These two measures were:

1) whether friends and / or relatives have been dissuaded from staying overnight within the past four weeks, due to poor housing conditions such as dampness or cold

2) whether friends and / or relatives have been dissuaded from visiting within the past four weeks, due to poor housing conditions such as dampness or cold.

CHARTS: Chart 4.7.1 shows results for measures (1) and (2) above, in the form of odds ratios. Chart 4.7.2 shows the respective proportions (percentages) of respondents who report each condition ( e.g. dissuading friends / relatives from staying overnight) at the end of the evaluation.

Chart 4.7.1 Differences in nature and extent of social contacts (odds ratios)

image of Chart 4.7.1 Differences in nature and extent of social contacts (odds ratios)

Chart 4.7.2 Nature and extent of social contacts (comparison group versus recipients)

image of Chart 4.7.2 Nature and extent of social contacts (comparison group versus recipients)

RESULTS: Table 4.7.1 shows numerical results, in the form of odds ratios.

Table 4.7.1 Differences in nature and extent of social contacts (odds ratios)

measure

odds ratio

95% CI

P

n

Whether friends / relatives dissuaded from staying overnight due to poor housing conditions

0.42

0.26 to 0.70

< 0.01

2,292

Whether friends / relatives dissuaded from visiting due to poor housing conditions

0.40

0.23 to 0.70

< 0.01

2,322

Despite the small numbers reporting these negative impacts of housing conditions, the differences between recipients and comparison group were statistically significant (see table 4.7.1).

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP:

1) number of times the respondent has gone out to visit family or friends in the past two weeks

2) number of times the respondent has been visited at home by family or friends in the past two weeks.

4.8 Perceived financial strain

KEY MESSAGE: Those who received heating under the CHP were less likely than the comparison group to report experiencing any degree of financial difficulty.

SUMMARY: This area consisted of a single measure, representing whether the respondent reported any degree of financial difficulty. A significant effect of the CHP was observed, recipients of heating under the Programme being less likely than the comparison group to indicate that they were experiencing financial difficulty.

CHARTS: Chart 4.8.1 shows the result (odds ratio) for the measure.

Chart 4.8.1 Difference in perceived financial strain (odds ratio)

image of Chart 4.8.1 Difference in perceived financial strain (odds ratio)

Chart 4.8.2 shows the respective proportions (percentages) of respondents who report the condition ( i.e. experiencing any degree of financial difficulty) at the end of the evaluation.

Chart 4.8.2 Perceived financial strain (comparison group versus recipients)

image of Chart 4.8.2 Perceived financial strain (comparison group versus recipients)

RESULTS: Table 4.8.1 provides numerical results for the single measure featured in the area.

Table 4.8.1 Difference in perceived financial strain (odds ratio)

measure

odds ratio

95% CI

p

n

Whether respondent reported any degree of financial difficulty

0.77

0.60 to 0.99

0.04

2,318

4.9 Specific symptoms and health conditions, and use of primary and secondary health services

KEY MESSAGE: Those who received heating under the CHP were less likely than the comparison group to report having received a first diagnosis of heart disease or of high blood pressure during the evaluation period, but were more likely to report a first diagnosis of nasal allergy. The finding that receipt of central heating was associated with a positive effect on two potentially serious classes of condition ( i.e. heart disease and high blood pressure) is potentially of considerable interest. However, fourteen further outcome measures related to specific symptoms and health conditions showed no significant association with the CHP. This indicates that the provision of central heating appeared to exert no discernible effect on respiratory conditions such as asthma and bronchitis.

Of a second subgroup of four outcome measures representing respondents' use of health services, none exhibited a significant association with the CHP. This is a potentially anomalous finding, in that significant between-group differences in the respective experience of heart disease and high blood pressure would be expected to co-occur with differences in the use of health services. This issue is not considered further here, but is discussed in the Summary of Findings ( Section 1) and in the Conclusions to this Report ( Section 5).

SUMMARY: This area consisted of 21 measures, covering two closely-related conceptual sub-areas. First, a group of 17 measures representing various aspects of the respondent's state of health (focussing especially on respiratory and cardiovascular health) was considered. Second, four further measures relating to different aspects of the respondent's use of health services were examined. These two sets of outcomes were treated as a single conceptual area for reporting purposes on the grounds that any significant observed effects of the CHP on specific symptoms and health conditions might reasonably be expected to generate consequent effects on the respondent's use of health services.

Of the 17 outcomes which related to specific symptoms and health conditions, three exhibited a significant effect of the CHP, namely:

1) whether the respondent reported having received a first diagnosis of heart disease during the evaluation period

2) whether the respondent reported having received a first diagnosis of high blood pressure during the evaluation period

3) whether the respondent reported having received a first diagnosis of a nasal allergy such as hayfever during the evaluation period.

Of the four measures which represented respondents' use of health services, none was found to be significantly associated with the receipt of central heating under the CHP.

CHARTS: Chart 4.9.1 shows results for the three measures which exhibited a significant effect of the CHP. Results are shown as odds ratios.

Chart 4.9.1 Differences in health conditions (odds ratios)

image of Chart 4.9.1 Differences in health conditions (odds ratios)

Chart 4.9.2 shows the proportions (percentages) of respondents who reported each condition ( e.g. having received a first diagnosis of heart disease) at the endpoint of the evaluation.

Chart 4.9.2 Health conditions (comparison group versus recipients)

image of Chart 4.9.2 Health conditions (comparison group versus recipients)

RESULTS: Table 4.9.1 provides numerical results for the three measures which showed significant associations with the receipt of heating under the CHP.

Table 4.9.1 Differences in health conditions (odds ratios)

measure

odds ratio

95% CI

p

n

Whether respondent reported a first diagnosis of heart disease during the evaluation period

0.69

0.52 to 0.91

0.01

1,928

Whether respondent reported a first diagnosis of high blood pressure during the evaluation period

0.77

0.61 to 0.97

0.02

1,340

Whether respondent reported a first diagnosis of nasal allergy during the evaluation period

1.52

1.05 to 2.20

0.03

2,136

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP:

(first sub-area: specific symptoms and health conditions)

1) number of reported episodes of cold / flu symptoms in the past six months

2) whether the respondent reported a first diagnosis of asthma during the evaluation period

3) whether the respondent reported a first diagnosis of chest problems such as chronic bronchitis or pulmonary disease during the evaluation period

4) whether the respondent reported a first diagnosis of eczema during the evaluation period

5) whether the respondent reported a first diagnosis of circulatory problems during the evaluation period

6) number of reported attacks of asthma in the past 12 months

7) whether the respondent has been woken by shortness of breath in the past 12 months

8) whether the respondent has been woken by tightness in the chest in the past 12 months

9) whether the respondent has experienced wheezing in the chest in the past 12 months

10) whether the respondent has experienced coughing or phlegm on most days for a minimum of three months a year and for at least two successive years

11) whether the respondent suffers from at least one respiratory health problem

12) whether the respondent has ever been advised by a health professional to change diet or lifestyle to reduce blood pressure or to avoid having high blood pressure

13) whether the respondent is currently taking any action in relation to diet or lifestyle because of concerns about blood pressure

14) whether the respondent currently has high blood pressure

(second sub-area: use of primary and secondary health services)

15) the number of times the respondent has seen or spoken to a GP or nurse, either at their practice or at home, over the past year

16) the number of times the respondent has visited a hospital outpatient clinic or day bed unit over the past year

17) the number of times the respondent has had an overnight stay in hospital over the past year

18) the number of times the respondent has visited a Casualty or Accident and Emergency Department for treatment over the past year

4.10 Self-reported health-related quality of life

KEY MESSAGE: A statistically significant positive effect of the CHP was observed for two of the individual dimensions of self-reported health and well-being which are measured by the SF-36 Health Survey [3]. However, the estimated effect sizes observed in the evaluation were small (around 2.5 points on a 100-point scale in each case). This is unlikely to be a clinically significant difference.

SUMMARY: This area consisted of nine measures representing various aspects of the respondent's self-reported health and wellbeing. All measures were drawn from the SF-36 Version 2 Health Survey, a 36 item self-administered questionnaire which measures health-related quality of life in the general population. Of the nine measures, two exhibited a significant effect of the CHP:

1) the respondent's score on the SF-36 physical functioning scale

2) the respondent's score on the SF-36 general health scale.

The remaining seven measures showed no statistically significant associations with the CHP.

CHARTS: Chart 4.10.1 shows results for the two measures listed above. In this chart, the small black rectangles represent the estimated differences in average values of the measures between recipients and comparison respondents. Both values are positive ( i.e. greater than zero), indicating that recipients recorded higher ('better') scores for these scales than did the comparison group 9. The horizontal lines show the 95% confidence intervals for the estimated differences, and the dashed vertical line marks the value of zero which indicates no difference between the two groups of respondents ( i.e. recipients and comparison households).

Chart 4.10.1 Differences in self-reported health-related quality of life (odds ratios)

image of Chart 4.10.1 Differences in self-reported health-related quality of life (odds ratios)

Chart 4.10.2 shows average values of the two measures at the endpoint of the evaluation.

Chart 4.10.2 Self-reported health-related quality of life (comparison group versus recipients)

image of Chart 4.10.2 Self-reported health-related quality of life (comparison group versus recipients)

RESULTS: Table 4.10.1 shows numerical results for the two measures which exhibited significant associations with the CHP. The column headed 'estimate' shows the average difference between recipients' scores on each scale and the scores for comparison group respondents.

Table 4.10.1 Differences in self-reported health-related quality of life (odds ratios)

measure

estimate

95% CI

p

N

Score on SF-36 Physical Functioning scale

2.51

0.62 to 4.40

< 0.01

2,171

Score on SF-36 General Health scale

2.57

0.87 to 4.27

< 0.01

2,314

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP:

1) the respondent's score on each of the following SF-36 scales: Role-Physical, Bodily Pain, Vitality, Social Functioning, Role-Emotional and Mental Health.

2) the respondent's score on the SF-36 Health Transition item

4.11 Long-standing illness or disability

KEY MESSAGE: Receipt of central heating under the Programme did not exert a significant influence on the likelihood of the respondent suffering from a long-standing illness, disability or health problem.

SUMMARY: This area consisted of a single measure, representing whether the respondent currently suffers from any long-standing illness, health problem or disability which limits her / his daily activities or the type of work which may be undertaken. This measure was found not to be significantly associated with the receipt of central heating under the CHP.

CHARTS: None shown, as no statistically significant result was found.

RESULTS: None shown.

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP: As outlined in Summary above.

4.12 Use of medications

KEY MESSAGE: The evaluation detected no significant association between the receipt of heating under the Programme and respondents' use of either prescribed or 'over the counter' medications.

SUMMARY: This area included two measures relating to the respondent's use of medications. Neither of these was found to be significantly associated with the receipt of central heating under the CHP.

CHARTS: None shown, as no statistically significant results were found.

RESULTS: None shown.

MEASURES NOT SHOWING A SIGNIFICANT ASSOCIATION WITH THE CHP:

1) whether the respondent is currently taking prescribed medications

2) whether the respondent is currently taking 'over the counter' medications

4.13 Subgroup analyses

A limited number of subgroup analyses were performed in order to assess whether the CHP operated differentially on specific subgroups of recipients. Differential effects were investigated for three key outcomes, namely: whether the respondent suffers from at least one respiratory health problem; the respondent's score on the SF-36 general health scale; and whether any environmental problem (damp, mould etc.) is present in the main living room and the main bedroom of the respondent's dwelling. In investigating differential effects, attention focussed on the possible influence of housing tenure, physical house type ( e.g. 'semi-detached'), gender and age. The investigation was performed in two stages.

Firstly, a revised series of statistical models was constructed in which specific subgroups of recipients ( e.g. those living in semi-detached homes) were compared to the entire control group. Comparison of the effects associated with each subgroup ( e.g. comparing recipients in semi-detached homes with recipients in tenements) provided an indication - though not formal statistical evidence - of whether receipt of heating operated uniformly across the specified subgroups of recipients on a particular outcome ( e.g.SF-36 general health). To provide formal confirmation of these initial insights, a number of additional analyses were performed in which the interactions between receipt of central heating and the factors of interest ( e.g. the age of the respondent) were estimated. This further stage was not performed for all possible combinations of outcomes and classifying factors (such as gender and tenure), but none of the results obtained can reasonably be interpreted as indicating that the effect of the Programme varied substantially across subgroups for the range of outcomes examined. Clearly, this conclusion cannot be definitive: even if all possible differential effects were formally tested, the number of individual statistical tests involved would be such as to introduce a real danger of 'false positive' results. However, it may be asserted with some confidence that the Programme does not appear to be have exerted meaningful differential effects on specific subgroups of those who received heating.