  |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
26-September-2003, Vol 116 No 1182
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Adults’ perceptions of the causes and primary
prevention of common fatal cancers in New Zealand
Anthony Reeder and Judy Trevena
Cancer is the second leading cause of death in New Zealand,
after cardiovascular disease,1 and the leading
cause of death among those aged 35–64 years. Potentially modifiable
behavioural, social and environmental factors may be implicated in as many as
75% of cancers.2 Lung cancer, the most common
cause of cancer death in New Zealand,3 is
linked to tobacco smoke. There is ‘convincing evidence’ that
physical activity and vegetable consumption are protective against cancer of the
large bowel,4 the second leading cause of
cancer death. New Zealand melanoma rates are among the highest in the
world.5 Most melanomas are attributed to excess
sun exposure.6 New Zealand has higher cancer
incidence rates than Australia, overall (for both sexes), for colorectal cancer
(both sexes), and lung cancer and melanoma
(females), such that priority needs ‘to be given to efforts to discover
reasons for the high incidence of cancer in New Zealand, and to enhance
prevention, appropriate screening, and early
diagnosis.’7 Reports about public
perceptions of cancer causation and prevention have been disseminated
overseas,8–10 but in New Zealand most
reports remain unpublished.11–14 This is
despite their relevance for primary prevention and also potential provision of
insight into motivations that may affect the uptake of screening and
acceptability of diagnostic and treatment services. In New Zealand, research
into ‘better methods of preventing cancer, investigating the social and
behavioural factors that discourage people from seeking treatment’ and
other social issues is ‘unfortunately sparse’ and uncertainty
‘in prevention, socio-behaviour and epidemiology can only be answered by
specific New Zealand-based research.’15
The aim of this study is to help inform the New Zealand Cancer Control
Strategy.
MethodsA
national telephone survey was carried out by an experienced contractor (Phoenix
Research), August to September 2001, using a computer-assisted telephone
interviewing system. The study was approved by the Otago Ethics Committee
(reference number 00/03/10). A sample size of around 400 (20 years and older)
was estimated as sufficient to obtain point estimates of population frequencies
with no more than a 5% margin of error. Quotas were set to recruit proportions
consistent with the 1996 Census, by 20-year age bands, sex, self-reported Maori
ethnicity and geographic region. Telephone numbers were randomly selected from
directory listings. To achieve the desired proportion of Maori, further contacts
who identified as Maori were randomly selected from electoral rolls.
Trained interviewers asked to speak with the person in
the household, 20 years of age or older, who had most recently celebrated a
birthday. Contacts were advised that the questionnaire would take about 12
minutes to complete and that for quality control
reasons, including the need to ensure
consistency between interviewers, a supervisor would monitor a proportion of
interviews. Interviewers offered to call back if the timing was inconvenient. Up
to six, mostly evening, call-backs were made. All interviewers were female. In
some previous research certain questions were asked of women only by female
interviewers.8 The questionnaire included items
designed to explore perceptions of the causes and prevention of cancer, and to
provide demographic information. Questionnaire content drew on many
sources,8,11–14,16–18 and was
refined through consultation. Presentation of questions was dependent on sex:
items about breast and cervical cancer were asked only of women and questions
about prostate cancer asked only of men. Questions about melanoma, bowel and
lung cancer were addressed to all. Responses are reported in the order that the
questions were asked.
For fixed-response questions, interviewers read out
permitted answers and electronically recorded responses. For open-ended
questions, codes for categorising anticipated answers were provided to the
contractor. Other answers were recorded verbatim, coded by one researcher and
checked by the other. Where several verbatim responses were similar, a new code
was created to encompass them. For questions where multiple responses were
permitted, after each response, participants were asked ‘Anything
else?’ until either they could provide no further answers or a maximum of
10 replies was reached.
The terms used by respondents for cancer types were
categorised to correspond with those used in official cancer databases so that,
for example, the category ‘bowel cancer’ includes all mentions of
colorectal cancer and cancers of the colon and rectum, whereas stomach cancers
form a distinct and separate group. Little difficulty was experienced in
classifying responses in this way.
ResultsSample
characteristics Of 1565 attempts to perform interviews (Table 1), 1130
contacts were made and 689 were deemed eligible, according to population quotas.
Of these, 251 refused to participate, producing 438 completed interviews (64%
participation).
The age, sex and self-defined ethnicity distributions of the
438 respondents were similar to the 1996 Census population (Table 2).
Table 1. Attempted and completed interviews, by call
status and contact source
*Statistics New
Zealand19
Respondents
were asked to indicate all ethnic groups with which they identified; those
reporting more than one were coded according to the census hierarchy (Table 2).
If a person said ‘Maori and European’, their primary coding was
Maori, whereas if someone said ‘Asian and Pacific Island’ it was
Pacific Island. Of the 47 Maori participants, 11 were recruited through
supplementary Maori sampling procedures.
Compared with the 1996 Census population, the geographic
distribution of the sample included 6% more respondents from the Auckland region
and 5% more from the Waikato, but otherwise differed by no more than 3%. The
sample was better educated, with considerably fewer reporting no school
qualifications (11.6% vs 32.3%) and more reporting secondary (40.0% vs 23.7%) or
tertiary qualifications (45.7% vs 27.9%).20 The
sample also contained more in full-time employment (56.2% vs 47.6%) and fewer
unemployed (2.1% vs 4.2%).21
Female cancers In
response to the question ‘Which three cancers do you think cause the most
deaths among New Zealand women?’
most women were able to name three (55%), 31% two, and 11% one. Only 3%
(seven women) were unable to name any. The full results are presented in Table
3, where the cancer sites listed in italics were not specifically included in
survey questions.
Table 3. Selected causes of cancer deaths, ranked by
1998 New Zealand population statistics* and frequency of mention by sample, by
sex
*New Zealand Health Information
Service22;
†percentage of all cancer deaths
NMSC = non-melanoma skin cancer When asked ‘Do you know of anything that increases the
risk of getting breast cancer?’ 54% of women were unable to suggest
anything. The most commonly mentioned risk factors were a family history of
breast cancer (21%), and tobacco smoking (10%). The wide range of other replies
included: ‘not having (medical) checks’ (6%); hormone replacement
therapy and diet (5% each); and use of the contraceptive pill, not breast
feeding and high fat intake (4% each). No other factors were mentioned by more
than 2%. When asked ‘In what age group do you think a woman is most likely
to develop breast cancer?’ nearly half (47%) gave age ranges with upper
limits of 50 years or less. In addition, 13% said ‘any age’, 5%
40–60 years, and 2% said risk increased with age.
When asked ‘Do you know of anything that increases the
risk of getting cervical cancer?’
nearly half of the women (47%) could not name any risk factors. One woman
declined to answer. Most frequently mentioned were multiple sex partners (23%),
not having regular smear tests (9%) and a range of other factors including
having sex or a sexually transmissible disease (6% each), having genital wart
virus, specifically (4%), tobacco smoking (4%), early intercourse (3%), and
having a sex partner who has had several sex partners (2%). ‘Other’
responses (15% in total) included polyps, tampons, diet, genetics, the
contraceptive pill, IUD, fatty food and ‘uncircumcised
men’.
Male
cancers In response to the question ‘Which three cancers do you
think cause the most deaths among New Zealand men?’ most men were able to
name three (59%), 22% two, 10% one, and 8% could not name any (Table 3). When
asked, ‘Do you know of anything that increases the risk of getting
prostate cancer?’ most men (80%) could not suggest anything. The most
common risk factor mentioned was diet (8%) followed by ‘other’ (7%),
which included vasectomy, smoking, not drinking or urinating enough, and
cycling. Increasing age and a lack of regular medical checkups were mentioned by
3% each, and a family history of prostate cancer and being overweight or
inactive by 2%. When asked ‘In what age group do you think a man is most
likely to develop prostate cancer’, less than 2% said that likelihood
increases with increasing age. A substantial proportion (41%) mentioned age
ranges with upper limits of 50 years or less, 4% said that it was most likely
within the 40–60 age group, and 2% said the risk occurred at any
age.
Other
cancers Men and women were asked ‘What do you understand by the
term ‘melanoma’?’ Unprompted answers were coded according to a
hierarchy used for presenting earlier survey
results,16 with one answer per person. Overall,
64% were aware that melanoma is a type of skin cancer. An additional 11%
mentioned moles and freckles, 9% the sun or UV radiation, 5% cancer, and 3%
described melanoma as a ‘skin condition’. Only 3% said that they
didn’t know what it was. When asked,
‘Do you know of anything that increases the risk of getting
melanoma?’ only 2% were unable to describe any risk factors and most (84%)
mentioned exposure to excess solar UV radiation. Also mentioned were having skin
that burns easily (9%), moles (5%), unprotected sun exposure at an early age
(4%), sun-lamp use (3%), and a family history of melanoma (2%). Over half (55%)
replied in the affirmative to the question ‘Have you or anyone else
deliberately checked your skin for changes which could be melanoma or other skin
cancer in the last 12 months?’ Most of the checks were either performed by
a doctor/specialist (32%), self-examinations (16%) or carried out by family
members (7%). Few mentioned friends and partners (1%).
When asked ‘Do you know of anything that increases the
risk of getting bowel cancer?’ 51% were unable to describe any risk
factors. Most frequently mentioned were unspecified dietary factors (28%),
inadequate fibre (18%), excess fat (9%), and a family history of bowel cancer
(8%). Meat consumption (6%), lack of regular bowel movements (5%), alcohol (4%),
smoking (3%) and stress (2%) were other factors mentioned.
In response to the question ‘Do you know of anything
that increases the risk of getting lung
cancer?’ almost all (98%)
identified the relationship between tobacco smoking and lung cancer. Other
factors included asbestos (16%), exposure to workplace hazards (14%),
second-hand smoke and chemicals (12% each). Some other replies, no more than 3%
each, referred to dust, genetic factors, alcohol and non-tobacco smoke. To
assess perceived health gain from quitting, respondents were asked ‘How
much do you think that a regular smoker can reduce their risk of lung cancer
by
quitting smoking?’ Most considered that quitting would either
‘greatly’ (53%) or ‘moderately’ (25%) reduce lung cancer
risk, whereas only 7% thought that it would ‘completely eliminate’
risk, 8% that it would ‘slightly’ reduce risk and 1% not reduce risk
at all.
DiscussionThe
present study appears to be the only one of its kind published in New Zealand
recently, although a number of unpublished reports have been
produced.11–14,16
The 64% response rate obtained exceeds levels reported for earlier
population-based surveys where that information is
provided.11,12 Nevertheless, our findings tend
to confirm patterns identified previously, although there are difficulties
comparing results obtained from prompted and unprompted questionnaire items and
telephone and postal surveys. For example, although all surveys report very high
levels of awareness of the link between tobacco smoking and lung cancer, surveys
where prompts were provided recorded far greater awareness of other risk factors
for lung cancer. In one postal survey,11
exposure to asbestos was selected by 83% from a list of possible factors,
whereas in the present study asbestos was mentioned, unprompted, by only 16%.
Nevertheless, asbestos ranked second as a risk factor in both studies.
When women were asked to name the three most commonly fatal
cancers among New Zealand women, 45% named fewer than three. Those perceived as
most common included the three most frequent causes of cancer death, but in an
order probably influenced by screening programmes. Lung and bowel cancers each
cause around seven times more deaths among women than cervical cancer, and they
warrant increased attention. Lung cancer is mostly caused by tobacco smoke and
is readily preventable; bowel cancer is linked to
nutrition4 and there is ‘sufficient
evidence’ that colon (and breast) cancers are related to physical
inactivity and many are, therefore, also potentially
preventable.23 The far greater prominence
afforded breast cancer relative to lung and bowel cancers has been noted
elsewhere.24
Despite awareness of the cancers, about half the women were
unable to name any risk factors for cervical cancer and breast cancer. Although
there was moderate awareness that cervical cancer was related to sexually
transmissible infections, any increased publicity should support preventive
empowerment rather ‘victim blaming’. The prominence of ‘family
history’ as an explanation for breast cancer requires qualification, as
most cases have no such history. There was a lack of consensus about the age at
which women are most likely to develop breast
cancer. Nearly half gave ages less than
50 years old – the age at which the national breast screening programme in
New Zealand presently begins. Although this may reflect a possible greater
impact of cases among acquaintances that occurred at a younger age, women may
also have been influenced by a promotion featuring Lucy Lawless, an actress in
her 30s. This demonstrates the need to frame and target messages appropriately
in order to avoid unnecessarily raising anxiety. The dissemination of primary
prevention information (healthy nutrition and physical activity) among younger
women could complement the secondary prevention focus (breast screening) among
older age groups.
The cancers most commonly perceived as fatal by men in New
Zealand included the three most common, but in a different order. As among
women, lung and bowel cancer deserve greater prominence. There was limited
appreciation that prostate cancer mainly affected older age groups, with 40%
including ages younger than 50 years as the age of greatest risk. As mentioned
with respect to breast cancer among women, this may reflect the possible greater
impact of cases when cancer occurs at a young age. There is currently inadequate
evidence to support prostate screening, though GPs are likely to refer men over
50 years of age for tests.25 There was little
knowledge about causation, reflecting a lack of evidence of risk factors for
prostate cancer, although a recent New Zealand study found that vasectomy did
not increase risk.26
In addition to high awareness among both sexes of the link
between lung cancer and smoking, most respondents (85%) considered that quitting
would reduce risk of lung cancer. This promising result suggests that in New
Zealand health promotion campaigns about the risks of smoking and the benefits
of quitting have successfully raised public awareness about causation and the
efficacy of preventive strategies. Nevertheless, around 25% of New Zealanders
over the age of 15 years remain daily smokers and the prevalence of smoking has
not declined significantly in recent years.27
Clearly, high awareness of risk and preventive strategies is not sufficient to
further reduce smoking prevalence. The addictive nature of nicotine reinforces
the need to strengthen tobacco control efforts and maintain a broad approach
that goes beyond individual behaviour change and removes social and
environmental support for smoking. The creation of smoke-free workplaces and
public places is one such strategy. Tobacco control efforts could be funded
directly from tobacco taxes, which raise around $800 million a year; at present,
less than 10% of this amount is spent on tobacco control, despite smoking being
the leading cause of preventable, premature death. In Australia, the public
health benefits of tobacco control are well
appreciated,28 but the estimated amount
committed per death in 1998 was low (less than $500 compared with $419 619 for
road safety, $34 603 for cervical cancer, and $20 172 for breast cancer
programmes).29 A similar pattern is likely to
exist in New Zealand, but appears to be undocumented.
Our results confirm high levels of awareness that melanoma
is a skin cancer related to excess sun
exposure,11,13 yet there is little evidence of
recent improvement in adults’ sun protective attitudes and
behaviours.30 This may be due, at least partly,
to the widely held,13 and correct, perception
that melanoma is largely curable if treated early. In order to reduce the
frequency of risk behaviour there is a need to build stronger social and
environmental support for sun protection. For example, this can be achieved
through policies and practices that promote shade provision at recreational
facilities and the rescheduling of outdoor activities to times outside the hours
of highest solar UV risk. As is the case with smoking, focusing on individual
responsibility alone would prevent fewer cancers. The promotion of cosmetic
tanning by ‘health’ centres, despite evidence that it increases the
risk of melanoma and other skin cancers, is another case in point. Given doubts
about the effectiveness of voluntary standards, solaria could be included under
radiation protection legislation, presently under
review.31 Considering the preventability of
skin cancer and significant health service costs (conservatively estimated at
$30 million a year32) it is surprising that,
since the demise of the Public Health Commission in 1994, preventive efforts
have been left almost entirely to the Cancer Society and the Health Sponsorship
Council. Nevertheless, skin cancer has recently been identified as a cancer
prevention priority.15
Lack of unprompted awareness about the protective role of
physical activity in reducing bowel cancer
risk34 suggests that more publicity is needed,
especially given New Zealand’s high incidence rates, particularly among
women.7 With respect to causation, large
proportions in an earlier study selected items, such as poor diet and family
history,11 that were perceived to increase the
risk. In response to an unprompted question, however, 51% of our sample could
not name any risk factors. Nevertheless, the high ranking of nutritional factors
is common to New Zealand
studies.11–13,16
There was considerable variation in awareness about
different cancers, and knowledge about the relative frequency with which
specific cancer deaths occur among the population may be better than knowledge
about appropriate preventive strategies for most cancers. The proportions saying
that they did not know of anything that increased cancer risk ranged from 1% for
lung cancer (there was almost universal awareness of the causal link with
smoking) and 2% for melanoma (linked with excess sun exposure), to 51% for bowel
cancer and 80% of men for prostate cancer (reflecting that causation is not
scientifically well understood). It is perhaps more surprising that about half
of the women did not know of any risk factors for cervical cancer (46%) or
breast cancer (54%). Overall, there is
a need for clear, consistent, and coordinated messages that reflect current
evidence about risk and prevention. It is likely that the socioeconomically
advantaged nature of the study sample, in terms of education and employment when
compared with the 1996 Census population, has resulted in an underestimate of
this need. Furthermore, our study mostly focused on relatively high-profile
issues. Known key modifiable risk factors to be targeted should include exposure
to tobacco smoke (lung cancer), physical inactivity (bowel, breast, possibly
prostate and other cancers), obesity (bowel, oesophageal, and post-menopausal
breast cancer), inappropriate nutrition (oral, stomach and bowel cancer),
alcohol consumption (oral, pharynx, oesophagus, larynx and possibly breast
cancers), transmission of infectious diseases (liver, cervical and stomach
cancers), excess sun exposure (skin cancer), and occupational
exposures.15 Preventive activities need to go
beyond raising awareness and encouraging individual behaviour change to include
modification of the social and economic environment. The comprehensive New
Zealand Cancer Control Strategy provides opportunities for promoting policies
and practices that assist the development of health-promoting
behaviours.
Author
information: Anthony Reeder, Senior Research Fellow, Social &
Behavioural Research in Cancer Group, Department of Preventive & Social
Medicine; Judy Trevena, Lecturer, Department of Psychological Medicine,
University of Otago, Dunedin
Acknowledgements: We
thank Betsy Marshall for her comments on an earlier draft and the following for
their contributions to the development of the questionnaire: Wyn Barbezat, Dr
John Broughton, the Cancer Education Research Program (NSW) for permission to
adapt survey instruments, Dr Brian Cox, Chris Gold, Helen Darling, Dr Caroline
Horwath, Associate Professor Rob McGee, the Marlborough Health Trust, Betsy
Marshall, Associate Professor Charlotte Paul, Dr David Perez, Rose Richards, and
Carolyn Watts. Dr Reeder and the Social & Behavioural Research in Cancer
Group receive support from the Cancer Society of New Zealand Inc. and the
University of Otago. For this research, Dr Trevena received support from an
Otago University research grant to Dr Reeder and Associate Professor Rob McGee.
Allan Wyllie and Phoenix Research, Auckland, were responsible for sample
selection and survey administration.
Correspondence: Dr
Tony Reeder, Social & Behavioural Research in Cancer Group, Department of
Preventive & Social Medicine, University of Otago, P O Box 913, Dunedin.
Fax: (03) 479 7298; email: treeder@gandalf.otago.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|