- Email: [email protected]
Are cervical cancer screening rates different for women with schizophrenia? A Manitoba population-based study
Schizophrenia Research 113 (2009) 101–106
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Are cervical cancer screening rates different for women with schizophrenia? A Manitoba population-based study Patricia J. Martens a,⁎, Harvey Max Chochinov b, Heather J. Prior a, Randall Fransoo a, Elaine Burland a, The Need To Know Team 1 a b
Department of Community Health Sciences, University of Manitoba, Manitoba Centre for Health Policy, 408-727 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 3P5 Department of Psychiatry, University of Manitoba, CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 0V9
a r t i c l e
i n f o
Article history: Received 28 January 2009 Received in revised form 9 April 2009 Accepted 14 April 2009 Available online 6 May 2009 Keywords: Schizophrenia Cervical cancer screening Continuity of care Screening uptake Manitoba Centre for Health Policy Administrative data
a b s t r a c t Context: Barriers to cervical cancer screening (Pap tests) may exist for women experiencing schizophrenia. Design: This study analyzed healthcare records of all women in the province of Manitoba, Canada to: (a) compare cervical cancer screening rates of women with and without schizophrenia; and (b) determine factors associated with screening uptake. Setting: This study took place in Manitoba, Canada, utilizing anonymized universal administrative data in the Population Health Research Data Repository at the Manitoba Centre for Health Policy. Participants: All females aged 18–69 living in Manitoba December 31, 2002, excluding those diagnosed with invasive or in situ cervical cancer in the study period or previous 5 years. Main outcome: To determine factors associated with Papanicolaou (Pap) test uptake (1+ Pap test in 3 years, 2001/02–2003/04), logistic regression modeling included: diagnosis of schizophrenia, age, region, average household income, continuity of care (COC), presence of major physical comorbidity. Good COC was defined as at least 50% of ambulatory physician visits from the same general/family practitioner within two years. Results: Women with schizophrenia (n = 3220) were less likely to have a Pap test (58.8% vs. 67.8%, p b .0001) compared to all other women (n = 335 294). In the logistic regression, a diagnosis of schizophrenia (aOR = 0.70, 95% CI 0.65–0.75); aged 50+, and living in a low-income area or the North decreased likelihood; good continuity of care (aOR 1.88, 95% CI 1.85–1.91) and greater physical comorbidity (1.21, 95% CI 1.04–1.41) increased likelihood. Conclusion: Women with schizophrenia are less likely to receive appropriate cervical cancer screening. Since good continuity of care by primary care physicians may mitigate this, psychiatrists should consider assisting in ensuring screening uptake. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Invasive cervical cancer is considered preventable when screening and treatment of pre-invasive lesions are timely and ⁎ Corresponding author. Manitoba Centre for Health Policy, 408-727 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 3P5. Tel.: +1 204 789 3791; fax: +1 204 789 3910. E-mail address: [email protected] (P.J. Martens). 1 A collaboration of the Regional Health Authorities of Manitoba, Manitoba Health & Healthy Living, and the Manitoba Centre for Health Policy, directed by P.J. Martens and co-directed by R. Fransoo. 0920-9964/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2009.04.015
appropriate. A Canadian meta-analysis found that 53.8% (95% CI 43.6–66.3) of cervical cancer patients had inadequate screening, and a further 41.5% (95% CI 35.4–48.7) had no history of screening (Spence et al., 2007). Thus, appropriate screening for cervical cancer is considered the most important attributable factor in reducing cervical cancer incidence and deaths. Even with ad-hoc (non-organized) screening, incidence of cervical cancer in Canada has decreased by 50% and death rates by 60% since 1977 (Canadian Cancer Society, 2006). According to the consensus document, “Health Canada Programmatic Guidelines for Screening for Cancer of the
102
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
Cervix in Canada”, women 18 years and older should be screened initially with two screening tests one year apart. If these are negative, re-screening should be done every three years thereafter until age 69 years (Alberta Medical Association, 2006; Lofters et al., 2007). The National Population Health Survey (NPHS) in 1994/95 (Lee et al., 1998; Snider and Beauvais, 1998; Lotocki, 2000) reported that 68% of Canadian women indicated having a Papanicolaou (Pap) test within the past three years. Predictors of not having a Pap test were: being older, being single, a resident of Quebec, being an immigrant to Canada, and being of lower educational and income levels; other risk factors that have been identified include living in a rural area with high physician turnover and having a mother tongue other than English or French (Lotocki, 2000). The 2005 Canadian Community Health Survey (Statistics Canada, 2005) found a screening rate of 72.8% for Canadian women aged 18–69 years, with the province of Manitoba at 75.1%. The literature has been relatively silent on gaps in cervical cancer screening rates and appropriate strategies to address these disparities for women experiencing severe mental illness such as schizophrenia. Researchers have postulated that disparities in health for those experiencing schizophrenia may potentially arise from lower rates of seeking health care (Copeland et al., 2006; Fleischhacker et al., 2008; Mateen et al., 2008), general unhealthy lifestyles (Kirkpatrick et al., 2008), lack of instrumental and emotional support (Kilbourne et al., 2007) or lack of awareness, reticence or difficulty amongst people experiencing schizophrenia in conveying their physical problems to doctors. Moreover, negative experiences with health care providers can cause poor treatment compliance (Brown et al., 2000; Vandiver, 2007), particularly when patients perceive their caregivers hold stigmatized attitudes about mental illness (Wrigley et al., 2005), or when they themselves have internalized these attitudes and stereotypes, seeing themselves unworthy of treatment (Tal et al., 2007). Knowing that barriers to preventive care may exist in Canada for these women, the purpose of this study was to: (a) compare the cervical cancer screening rates of women diagnosed with schizophrenia to all other Manitobans; and (b) determine the unique factors associated with screening uptake. 2. Methods The province of Manitoba is centrally-located within Canada, with a total provincial population of approximately 1.2 million. This study used anonymized administrative claims data from the Population Health Research Data Repository (herein referred to as the Repository), housed at the Manitoba Centre for Health Policy (MCHP) at the University of Manitoba. The Repository data files contain information for all residents of Manitoba through the universal healthcare system, including hospital claims (records of hospital admissions); medical claims (records of visits to physicians outside of those occurring to a hospital in-patient); home care (records of the use of provincial home care services); personal care homes (records of the use of nursing homes); registry files (records of the time a person is registered as a resident of Manitoba, as well as their age, sex, and area of residence); vital statistics (records of births and deaths); and pharma-
ceutical claims (pharmaceutical use from the Drug Program Information Network). Although anonymized, all files are linkable at the person-level through the use of an encrypted personal health number and “crosswalks” between databases. All tabulations and statistical analysis were done using SAS version 9.1.3 (SAS Institute Inc, Cary NC, USA). Ethical approval was obtained from the University of Manitoba's Faculty of Medicine Human Research Ethics Board, and the provincial Health Information Privacy Committee. Cervical cancer screening was defined by the presence of a physician tariff code for a Pap test or a pathology or laboratory tariff code for a Pap test in the medical claims data. To calculate the proportion receiving ‘appropriate’ cervical cancer screening, the numerator included all woman age 18–69 that had at least one Pap test in a three-year period, calculated for fiscal years 2001/02–2003/04. For the numerator, age was calculated at the time of physician visit for a Pap test. The denominator included all females aged 18 to 69 in Manitoba as of December 31, 2002 and excluded women diagnosed with invasive cervical cancer or cervical cancer in situ in the study period or the previous five years. Women experiencing a diagnosis of schizophrenia in hospital separations or medical claims data within the study period or within the previous 9 years were included in the “schizophrenia” group calculation, with all other women in the “all other Manitobans” group. Provincially, there were 3220 woman in the ‘schizophrenia group’ and 335,294 in the ‘all other Manitobans’ group. Region of residence within the province was assigned based on the first record for the three-year period, grouped into five categories: North (three health regions of Burntwood, Churchill and Nor-Man, above the 53rd parallel of latitude); Mid (three health regions of Parkland, Interlake and North Eastman, approximately between the 51st and 53rd parallel); South (three health regions of Assiniboine, Central and South Eastman, between the 51st parallel and the 49th parallel at the Canada–US border); and two urban health regions — Brandon (Manitoba's 2nd largest city) and Winnipeg (the provincial capital). Within Winnipeg, the city was subdivided into three groupings based on previous analyses at MCHP (Martens et al., 2008, 2002): Winnipeg Most Healthy, Winnipeg Average Health and Winnipeg Least Healthy. These subdivisions were based upon the overall population health status of the areas as indicated by the surrogate measure of premature mortality rate (PMR), or age- and sexadjusted death before the age of 75. The number of women in the schizophrenia group and the all other Manitobans group are as follows: North n = 128 and 18,021; Mid n = 312 and 41,181; South n = 320 and 60,172; Brandon n = 148 and 14,385; Winnipeg overall n = 2312 and 201,535; Winnipeg Most Healthy sub-region n = 839 and 103,838; Winnipeg Average Health sub-region n = 741 and 60,776; Winnipeg Least Healthy sub-region n = 732 and 36,921. Negative binominal regression modelling was used to estimate and compare age-adjusted rates of cervical cancer screening for women with schizophrenia vs. the ‘all other Manitobans’ group, by region and provincially. The model controlled for age (both a linear and quadratic age effect were included), region (Manitoba was the reference), group (reference was no schizophrenia diagnosis), and a region by group interaction effect. In a further analysis using logistic regression, the probability of a woman receiving a Pap test in
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
the three-year period was also predicted, to control for possible confounders in the comparison of groups. Covariates in the logistic regression model included: (a) region of residence, including the North, Mid and South rural areas, the city of Brandon, city of Winnipeg (effect coding employed to allow the entire province of Manitoba to be the reference); (b) age, grouped as: 18–29, 30–39, 40–49 (reference group), 50–59, 60–69; (c) average enumeration area household income from the 2001 Canadian Census (used as a continuous variable); (d) a dichotomous variable for continuity of care from a general/family practitioner (GP/FP), with the reference being no continuity of care; and (e) a dichotomous variable for the presence of any major physical Aggregated Diagnostic Groups (ADGs) in 2002/03 as a proxy measure for any severe physical illnesses (reference was no major physical ADGs). A woman was considered to have good continuity of primary physician care if at least 50% of her ambulatory physician visits within the two-year period of 2001/02– 2002/03 were to the same GP/FP. Two logistic regression models were analyzed: (1) including a dichotomous variable for schizophrenia diagnosis in hospital separations or medical claims data in study period or the previous 9 years (reference was no schizophrenia diagnosis); and (2) only including those having a schizophrenia diagnosis and excluding the ‘all other Manitobans’ group. 3. Results Table 1 shows a comparison of demographics for the two comparison groups in the study (women diagnosed with schizophrenia compared to all other women in Manitoba). The three-year cervical cancer screening rates were 58.8% for women diagnosed with schizophrenia compared with 67.8% for ‘all other Manitobans’ (p b .0001). A similar rate differential between 9% and 11% lower for women who were diagnosed with schizophrenia was persistent regionally in Winnipeg (60.7% vs. 70.8%, p b .0001), the South (53.8% vs. Table 1 Demographics and health indicator rates of women aged 18–69 included in this study, comparing those diagnosed with schizophrenia to all other women.
Mean age in years (SD) Mean household income of neighbourhood (SD) Region of residence Urban areas Brandon Winnipeg Rural areas South Mid North Percent receiving “Good Continuity of Primary Care” Percent having at least one major physical ADG Percent having at least one Pap test in three years
Women diagnosed with schizophrenia (n = 3220)
All other women (n = 335,294)
45.4 (13.0) $44,247 ($20,461)
40.4 (13.8) $53,082 ($23,482)
4.60% 71.80%
4.29% 60.11%
9.69% 3.98% 9.94% 66.46%
12.28% 5.37% 17.95% 63.69%
4.60%
4.29%
71.80%
60.11%
103
Table 2 Probability of cervical cancer screening (Pap test) within a three-year period (2001/02–2003/04), comparing women with a schizophrenia diagnosis vs. no diagnosis. Variable (reference)
Odds ratio Probability (95% confidence limits)
Schizophrenia diagnosis (no diagnosis is the reference) Region (Manitoba is the reference) Brandon Winnipeg Mid North South Age (40–49 years is the reference) 18–29 years 30–39 years 50–59 years 60–69 years Average household income (per $10,000) Continuity of care Physical comorbidity as measured by major ADGs
0.70 (0.65–0.75)
p b .0001
1.62 (1.57–1.67) 1.23 (1.21–1.25) 1.07 (1.05–1.09) 0.46 (0.44–0.47) 1.03 (1.01–1.05)
p b .0001 p b .0001 p b .0001 p b .0001 p b .002
0.97 (0.95–0.99) 1.15 (1.12–1.18) 0.82 (0.80–0.84) 0.51 (0.50–0.52) 1.110 (1.106–1.114)
p b .004 p b .0001 p b .0001 p b .0001 p b .0001
1.88 (1.85–1.91) 1.30 (1.28–1.33)
p b .0001 p b .0001
63.8%, p b .04) and Mid (54.2% vs. 65.1%, p b .03) regions. Brandon showed a similar differential (64.3% vs. 73.5%) that was not statistically significant, likely due to small numbers in this region. The two groups in the North experienced the lowest screening rates in the province, and were also similar to each other (41.4% vs. 43.8%, p = .68, NS). Within the capital city of Winnipeg, there was also evidence of a differential in two of the three sub-regions, with patterns similar to the rest of the province when comparing women in the ‘schizophrenia’ group to women in the ‘all other Manitobans’ group: Winnipeg Most Healthy (64.6% vs. 74.9%, p b .002); Winnipeg Average Health (60.3% vs. 69.1%, p b .008); Winnipeg Least Healthy (57.1% vs. 62.9%, p = .06, NS). In the logistic regression modeling with all women (see Table 2), the probability of having a Pap test within the threeyear period was lower for women diagnosed with schizophrenia (adjusted Odds Ratio = 0.70, 95% CI 0.65–0.75, p b .0001) after adjusting for region, age, average household income, continuity of care, and physical comorbidities. Within this model, women were more likely to receive a Pap test if they lived in Brandon, Winnipeg, Mid, and South, and less likely in the North (see Table 2). Compared to women aged 40–49 years, women aged 30–39 were more likely to receive a Pap test, but all other women (ages 18–29, 50–59, 60–69) were less likely. The higher the average household income and the greater the physical comorbidity level, the greater the likelihood of receiving a Pap test. Continuity of care was also associated with a higher likelihood of receiving Pap tests, after controlling for all other factors. A within-group logistic regression including only women diagnosed with schizophrenia (see Table 3) found that women living in Brandon and Winnipeg were more likely to receive a Pap test within the three-year period compared to the Manitoba within-group rate; those living in the North were less likely, and those in Mid and South rural areas were similar to the provincial rate. Women diagnosed with schizophrenia and in the older age brackets (50–59, 60–69) were less likely than woman age 40–49 to receive the appropriate screening,
104
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
Table 3 Probability of cervical cancer screening (Pap test) within a three-year period (2001/02–2003/04), for women with a schizophrenia diagnosis only. Variable (reference) Region (Manitoba is the reference) Brandon Winnipeg Mid North South Age (40–49 years is the reference) 18–29 years 30–39 years 50–59 years 60–69 years Average household income (per $10,000) Continuity of care Physical comorbidity as measured by major ADGs
Odds ratio Probability (95% confidence limits) 1.57 (1.18–2.10) 1.24 (1.09–1.43) 0.98 (0.79–1.21) 0.55 (0.41–0.74) 0.95 (0.77–1.17)
p b .003 p b .002 p = .84, NS p b .0002 p = .64, NS
1.20 (0.94–1.54) 1.00 (0.81–1.25) 0.70 (0.57–0.86) 0.41 (0.33–0.51) 1.07 (1.04–1.12)
p = .14, NS p = .98, NS p b .0006 p b .0001 p b .0002
1.23 (1.05–1.43) 1.21 (1.04–1.41)
p b .02 p b .02
while women age 18–29 and 30–39 were similar to women age 40–49. Women with major physical comorbidities, with higher average household income, and receiving good continuity of care were more likely to have a Pap test. Converting the logistic regression odds ratios into probabilities, for a woman living in an area with an average household income of $40,000, the approximate Pap test rates would be as follows, comparing women with a diagnosis of schizophrenia to women who do not: 60% vs. 68% in Winnipeg (p b .0001); 65% vs. 72% in Brandon (p b .0001); 55% vs. 63% in the South (p b .002); 56% vs. 64% in Mid (p b .0001); 33% vs. 42% in the North (p b .0001). Fig. 1 illustrates the effect of having a schizophrenia diagnosis, having good/not good continuity of care, and average household income on the probability of receiving the Pap test for cervical cancer screening within the three-year period. This figure gives actual probabilities (as derived from the logistic regression odds ratios in Table 2), with all four lines significantly different from each other. As income increases, so does the probability of a Pap test. As well,
those receiving good continuity of care also have the highest screening rates. However, having good continuity of care is associated with a higher rate of screening, no matter what the income level or whether a woman has a diagnosis of schizophrenia. For example, the overall Manitoba provincial probability of receiving a Pap test within three years for a woman living in an area with an average household income of approximately $40,000 per year is 48.3% (95% CI 46.4–50.1) if she has a diagnosis of schizophrenia and lack of good continuity of care, 62.7% (95% CI 60.9–64.4) if she has a diagnosis of schizophrenia and good continuity of care; 56.7% (95% CI 56.1–57.2) with no diagnosis and lack of good continuity of care and 70.6% (95% CI 70.2–71.1) with no diagnosis and good continuity of care. 4. Discussion Manitoba women living with a diagnosis of schizophrenia had substantially lower rates of receiving cervical cancer screening during the three-year period of this study (2001/02–2003/04) compared to women with no diagnosis of schizophrenia, at 58.8% vs. 67.8% provincially. Despite the variability by region of Manitoba, this gap remained persistent at around 9–11%. Factors influencing the uptake of Pap tests were surprisingly similar between women with schizophrenia and the ‘all other Manitobans’ group. Living in an urban area (Brandon or Winnipeg), and in an area with a higher average household income, were both associated with higher screening rates. The largest gap in regional effects was seen for the North, where rates were extremely low in both the age-adjusted rates, as well as in the regression modeling which controlled for income, physical comorbidity and continuity of care. This leads one to speculate that poorer access to health care services (culturally appropriate, community-based services offered through public health or through physician clinics) may be problematic for women in rural areas. Recent efforts started in 2008 by the Manitoba Government and CancerCare Manitoba to initiate a provincewide strategy to increase cervical cancer screening uptake through notification letters and through more accessible services may result in less regional rate inequities, although this will need
Fig. 1. Probability of a Pap test within three years (2001/02–2003/04) according to household income, for women aged 18–69 with and without a schizophrenia diagnosis, and with and without good continuity of care (COC).
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
to be studied over the next few years to determine the outcomes. Literature reviews (Black et al., 2000, 2002) found that letters of invitation produced a 40% increase in screening compared to a control group. In a Cochrane Systematic Review, Forbes et al. (2002) found that intervening through invitation letters continues to show the largest effect in increasing screening rates, with only limited effects shown for educational interventions. Thus the Manitoba approach reflects this literature. However, care must be taken to ensure that these strategies also promote better screening rates for those with schizophrenia. There was also a prominent age effect, with women aged 50–69 showing consistently lower rates, in the low 40% range. This age effect held true across women with schizophrenia as well as those without. This shows the need for concerted effort by physicians and other health care providers to be aware of the need for Pap testing in this vulnerable age group. A somewhat non-intuitive result is that women experiencing a higher level of physical comorbidity also had higher screening rates, with the effect persisting for those experiencing schizophrenia. This may be an indication that more attention is paid to screening when other illnesses are present, or that there are more possible opportunities for physician/patient interactions. Although many of the demographic factors are nonmodifiable, there is the potential for continuity of care to be modifiable through health system level interventions. In this study, good continuity of care was defined as seeing the same family physician for at least 50% of a person's care over a twoyear period. This is presumably a surrogate for more coordinated patient care, although within the limitations of this study, psychiatric care was not examined. Good continuity of care was strongly associated with better cervical cancer screening rates for women with and without schizophrenia. Continuity of care appears to be associated with a 14% higher rate of Pap test screening for both groups (see Fig. 1) and across the income spectrum. Because many women diagnosed with schizophrenia may be treated by psychiatrists, it is important for those physicians to take an active role in recommending and monitoring women's uptake of cervical cancer screening tests. In countries without a universal healthcare system, the disparities may be even greater for those who may not be able to afford healthcare services. This may underscore the importance of providing Pap tests by alternative, publicly funded healthcare providers or primary healthcare teams. Continuity of care, although defined in this study as care by the same physician, is a much broader concept which should include the use of primary care teams with multi-disciplinary providers. This team approach is a noteworthy practice both in Canada and throughout the world, and may overcome income barriers if provided through public health systems. Limitations of this study include the use of administrative healthcare data, with the fact that people who do not contact the healthcare or mental health system will not receive a diagnostic coding of schizophrenia. However, this study went back 10 years to retrieve possible diagnoses. As well, the overall population prevalence of schizophrenia at 1.2% (Martens et al., 2004) using the Manitoba data is similar to reported prevalence from survey sources. The fact that women living with schizophrenia, women living in rural areas and low-income areas, and women who are 50 or older have lower likelihood of having a Pap test
105
points to access issues. Strategies that combine letters of invitation with easily accessible and culturally appropriate health care provision is, according to the literature, the most effective means of increasing cervical cancer screening rates. Particular strategies to reach women experiencing severe mental illness such as schizophrenia may also need to include participation of mental health specialists as part of the primary health care team approach to ensure uptake of appropriate preventive services. Role of funding source None. Contributors Patricia J. Martens, Harvey Chochinov, Randy Fransoo, and the Need To Know Team designed the study and wrote the protocol. Elaine Burland, Patricia J. Martens, and Harvey Chochinov managed the literature searches and analyses. Heather J. Prior undertook the statistical analysis, and Patricia J. Martens wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript. Conflict of interest There are no conflicts of interest. Acknowledgements This work was supported as part of a research project completed in 2008 called, “What Works? A First Look at Evaluating Manitoba's Regional Health Programs and Policies at the Population Level”, one of several projects undertaken each year by the Manitoba Centre for Health Policy (MCHP) under contract to Manitoba Health & Healthy Living. The results and conclusions are those of the authors and no official endorsement by Manitoba Health & Healthy Living was intended or should be inferred. Dr. Harvey Chochinov holds a Canada Research Chair in Palliative Care and received funding for this study from the Canadian Institutes of Health Research. Dr. Patricia Martens would like to acknowledge funding from Canadian Institutes of Health Research (CIHR) which supports her research endeavours through her CIHR/PHAC Applied Public Health Chair, her former CIHR New Investigators' Award (2003–2008), CIHR Community Alliances for Health Research grant (2001–2006), and CIHR KT Award for Regional Impact (2005), all of which helped support The Need to Know Team. Dr. Chochinov would like to acknowledge the support of the Advisory Committee for the Study on Addressing Disability in Cancer Care. The authors are indebted to Decision Support Services of Manitoba Health & Healthy Living, and the Office of Vital Statistics in the Agency of Consumer and Corporate Affairs for the provision of data. The authors also thank Theresa Daniuk and Eileen Bell for research support.
References Alberta Medical Association. Screening for cervical cancer: revised. 2006. Accessed August 7, 2008 at http://www.cancerboard.ab.ca/accsp/pdf/ accsp_guide_cervical_cancer_06-03-07.pdf. Black, M., Yamada, J., Mann, V., Cava, M., Micucci, S., 2000. Effectiveness of strategies to increase cervical cancer screening: a systematic review of the literature 1989–1999. Toronto, Ontario: Public Health Branch, Ontario Ministry of Health — Report of the Effective Public Health Practice Project (EPHPP) Steering Committee. Black, M., Yamada, J., Mann, V., 2002. A systematic literature review of the effectiveness of community-based strategies to increase cervical cancer screening. Can. J. Public Health 93, 386–393. Brown, S., Inskip, H., Barraclough, B., 2000. Causes of the excess mortality of schizophrenia. Br. J. Psychiatry 177, 212–217 Sep. Canadian Cancer Society. Cancer screening in Canada not realizing full potential, 2006. Accessed August 7, 2008 at http://www.cancer.ca/ccs/ internet/mediareleaselist/0,,3543_434465_943594571_langId-en,00. html. Copeland, L.A., Zeber, J.E., Rosenheck, R.A., Miller, A.L., 2006. Unforeseen inpatient mortality among veterans with schizophrenia. Med. Care 44, 110–116. Fleischhacker, W.W., Cetkovich-Bakmas, M., De Hert, M., Hennekens, C.H., Lambert, M., Leucht, S., Maj, M., McIntyre, R.S., Naber, D., Newcomer, J.W., Olfson, M., Osby, U., Sartorius, N., Lieberman, J.A., 2008. Comorbid
106
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
somatic illnesses in patients with severe mental disorders: clinical, policy, and research challenges. J. Clin. Psychiatry 69 (4), 514–519 Apr. Forbes, C., Jepson, R., Martin-Hirsch, P., 2002. Interventions targeted at women to encourage the uptake of cervical screening. Cochrane Database Syst. Rev. (3), CD002834. Kilbourne, A.M., McCarthy, J.F., Edward, P.P., Welsh, D., Frederic, C.B., 2007. Social support among veterans with serious mental illness. Soc. Psychiatry Psychiatr. Epidemiol. 42, 639–646. Kirkpatrick, B., Messias, E., Harvey, P.D., Fernandez-Egea, E., Bowie, C.R., 2008. Is schizophrenia a syndrome of accelerated aging? Schizophr. Bull. 34 (6), 1024–1032. Lee, J., Parsons, G.F., Gentleman, J.F., 1998. Falling short of Pap test guidelines. Health Rep. 10 (1), 9–19 Accessed August 7, 2008 at http://www.statcan. ca/english/studies/82-003/feature/hrar1998010001s2a01.pdf. Lofters, A., Glazier, R., Agha, M., Creatore, M., Moineddin, R., 2007. Inadequacy of cervical cancer screening among urban recent immigrants: a populationbased study of physician and laboratory claims in Toronto, Canada. Prev. Med. 44, 536–542. Lotocki, R.J., 2000. The Pap smear: guidelines for screening and follow-up. Can. J. CME Accessed August 7, 2008 at http://www.cancercare.mb.ca/ cancercare_resources/MCCSP/pdfs/professionals_article_1.pdf. Martens, P.J., Frohlich, N., Carriere, K., Derksen, S., Brownell, M., 2002. Embedding child health within framework of regional health: population health status and sociodemographic indicators. Can. J. Public Health 93 (Supplement 2), S15–S20. Martens, P., Fransoo, R., McKeen, N., The Need to Know Team, Burland, E., Jebamani, L., Burchill, C., DeCoster, C., Ekuma, O., Prior, H., Chateau, D., Robinson, R., Metge, C., 2008. Patterns of Regional Mental Illness Disorder Diagnoses and Service Use in Manitoba: A Population-Based Study. Manitoba Centre for Health Policy, Winnipeg, Manitoba. Accessed April 7,
2008 at http://mchp-appserv.cpe.umanitoba.ca/reference/mental.health. pdf. Martens, P., Fransoo, R., The Need to Know Team, Burland, E., Prior, H., Burchill, C., Romphf, L., Chateau, D., Bailly, A., Ouelette, C., 2008. What Works? A First Look at Evaluating Manitoba's Regional Health Programs and Policies at the Population Level. Manitoba Centre for Health Policy, Winnipeg, Manitoba. Accessed August 7, 2008 at http://mchpappserv. cpe.umanitoba.ca/reference/fullwwreport.pdf. Mateen, F.J., Jatoi, A., Lineberry, T.W., Aranguren, D., Creagan, E.T., Croghan, G.A., Jett, J.R., Marks, R.S., Molina, J.R., Richardson, R.L., 2008. Do patients with schizophrenia receive state-of-the-art lung cancer therapy? A brief report. Psychooncology 17 (7), 721–725 Jul. Snider, J.A., Beauvais, J.E., 1998. Pap smear utilization in Canada: estimates after adjusting the eligible population for hysterectomy status. Chronic Dis. Can. 19 (1), 19–24. Spence, A.R., Goggin, P., Franco, E.L., 2007. Process of care failures in invasive cervical cancer: systematic review and meta-analysis. Prev. Med. 45 (2–3), 93–106 Aug–Sep. Statistics Canada. Canadian Community Health Survey (CCHS 3.1), 2005. CANSIM Table 105-0442. Accessed September 25, 2007 at http://www. statcan.ca. Tal, A., Roe, D., Corrigan, P.W., 2007. Mental illness stigma in the Israeli context: deliberations and suggestions. Int. J. Soc. Psychiatry 53 (6), 547–563. Vandiver, V., 2007. Health promotion as brief treatment: strategies for women with co-morbid health and mental health conditions. Brief Treatment Crisis Interv. 7, 161–175. Wrigley, S., Jackson, H., Judd, F., Komiti, A., 2005. Role of stigma and attitudes toward help-seeking from a general practitioner for mental health problems in a rural town. Aust. N. Z. J. Psychiatry 39 (6), 514–521 Jun.
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Are cervical cancer screening rates different for women with schizophrenia? A Manitoba population-based study Patricia J. Martens a,⁎, Harvey Max Chochinov b, Heather J. Prior a, Randall Fransoo a, Elaine Burland a, The Need To Know Team 1 a b
Department of Community Health Sciences, University of Manitoba, Manitoba Centre for Health Policy, 408-727 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 3P5 Department of Psychiatry, University of Manitoba, CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 0V9
a r t i c l e
i n f o
Article history: Received 28 January 2009 Received in revised form 9 April 2009 Accepted 14 April 2009 Available online 6 May 2009 Keywords: Schizophrenia Cervical cancer screening Continuity of care Screening uptake Manitoba Centre for Health Policy Administrative data
a b s t r a c t Context: Barriers to cervical cancer screening (Pap tests) may exist for women experiencing schizophrenia. Design: This study analyzed healthcare records of all women in the province of Manitoba, Canada to: (a) compare cervical cancer screening rates of women with and without schizophrenia; and (b) determine factors associated with screening uptake. Setting: This study took place in Manitoba, Canada, utilizing anonymized universal administrative data in the Population Health Research Data Repository at the Manitoba Centre for Health Policy. Participants: All females aged 18–69 living in Manitoba December 31, 2002, excluding those diagnosed with invasive or in situ cervical cancer in the study period or previous 5 years. Main outcome: To determine factors associated with Papanicolaou (Pap) test uptake (1+ Pap test in 3 years, 2001/02–2003/04), logistic regression modeling included: diagnosis of schizophrenia, age, region, average household income, continuity of care (COC), presence of major physical comorbidity. Good COC was defined as at least 50% of ambulatory physician visits from the same general/family practitioner within two years. Results: Women with schizophrenia (n = 3220) were less likely to have a Pap test (58.8% vs. 67.8%, p b .0001) compared to all other women (n = 335 294). In the logistic regression, a diagnosis of schizophrenia (aOR = 0.70, 95% CI 0.65–0.75); aged 50+, and living in a low-income area or the North decreased likelihood; good continuity of care (aOR 1.88, 95% CI 1.85–1.91) and greater physical comorbidity (1.21, 95% CI 1.04–1.41) increased likelihood. Conclusion: Women with schizophrenia are less likely to receive appropriate cervical cancer screening. Since good continuity of care by primary care physicians may mitigate this, psychiatrists should consider assisting in ensuring screening uptake. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Invasive cervical cancer is considered preventable when screening and treatment of pre-invasive lesions are timely and ⁎ Corresponding author. Manitoba Centre for Health Policy, 408-727 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 3P5. Tel.: +1 204 789 3791; fax: +1 204 789 3910. E-mail address: [email protected] (P.J. Martens). 1 A collaboration of the Regional Health Authorities of Manitoba, Manitoba Health & Healthy Living, and the Manitoba Centre for Health Policy, directed by P.J. Martens and co-directed by R. Fransoo. 0920-9964/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2009.04.015
appropriate. A Canadian meta-analysis found that 53.8% (95% CI 43.6–66.3) of cervical cancer patients had inadequate screening, and a further 41.5% (95% CI 35.4–48.7) had no history of screening (Spence et al., 2007). Thus, appropriate screening for cervical cancer is considered the most important attributable factor in reducing cervical cancer incidence and deaths. Even with ad-hoc (non-organized) screening, incidence of cervical cancer in Canada has decreased by 50% and death rates by 60% since 1977 (Canadian Cancer Society, 2006). According to the consensus document, “Health Canada Programmatic Guidelines for Screening for Cancer of the
102
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
Cervix in Canada”, women 18 years and older should be screened initially with two screening tests one year apart. If these are negative, re-screening should be done every three years thereafter until age 69 years (Alberta Medical Association, 2006; Lofters et al., 2007). The National Population Health Survey (NPHS) in 1994/95 (Lee et al., 1998; Snider and Beauvais, 1998; Lotocki, 2000) reported that 68% of Canadian women indicated having a Papanicolaou (Pap) test within the past three years. Predictors of not having a Pap test were: being older, being single, a resident of Quebec, being an immigrant to Canada, and being of lower educational and income levels; other risk factors that have been identified include living in a rural area with high physician turnover and having a mother tongue other than English or French (Lotocki, 2000). The 2005 Canadian Community Health Survey (Statistics Canada, 2005) found a screening rate of 72.8% for Canadian women aged 18–69 years, with the province of Manitoba at 75.1%. The literature has been relatively silent on gaps in cervical cancer screening rates and appropriate strategies to address these disparities for women experiencing severe mental illness such as schizophrenia. Researchers have postulated that disparities in health for those experiencing schizophrenia may potentially arise from lower rates of seeking health care (Copeland et al., 2006; Fleischhacker et al., 2008; Mateen et al., 2008), general unhealthy lifestyles (Kirkpatrick et al., 2008), lack of instrumental and emotional support (Kilbourne et al., 2007) or lack of awareness, reticence or difficulty amongst people experiencing schizophrenia in conveying their physical problems to doctors. Moreover, negative experiences with health care providers can cause poor treatment compliance (Brown et al., 2000; Vandiver, 2007), particularly when patients perceive their caregivers hold stigmatized attitudes about mental illness (Wrigley et al., 2005), or when they themselves have internalized these attitudes and stereotypes, seeing themselves unworthy of treatment (Tal et al., 2007). Knowing that barriers to preventive care may exist in Canada for these women, the purpose of this study was to: (a) compare the cervical cancer screening rates of women diagnosed with schizophrenia to all other Manitobans; and (b) determine the unique factors associated with screening uptake. 2. Methods The province of Manitoba is centrally-located within Canada, with a total provincial population of approximately 1.2 million. This study used anonymized administrative claims data from the Population Health Research Data Repository (herein referred to as the Repository), housed at the Manitoba Centre for Health Policy (MCHP) at the University of Manitoba. The Repository data files contain information for all residents of Manitoba through the universal healthcare system, including hospital claims (records of hospital admissions); medical claims (records of visits to physicians outside of those occurring to a hospital in-patient); home care (records of the use of provincial home care services); personal care homes (records of the use of nursing homes); registry files (records of the time a person is registered as a resident of Manitoba, as well as their age, sex, and area of residence); vital statistics (records of births and deaths); and pharma-
ceutical claims (pharmaceutical use from the Drug Program Information Network). Although anonymized, all files are linkable at the person-level through the use of an encrypted personal health number and “crosswalks” between databases. All tabulations and statistical analysis were done using SAS version 9.1.3 (SAS Institute Inc, Cary NC, USA). Ethical approval was obtained from the University of Manitoba's Faculty of Medicine Human Research Ethics Board, and the provincial Health Information Privacy Committee. Cervical cancer screening was defined by the presence of a physician tariff code for a Pap test or a pathology or laboratory tariff code for a Pap test in the medical claims data. To calculate the proportion receiving ‘appropriate’ cervical cancer screening, the numerator included all woman age 18–69 that had at least one Pap test in a three-year period, calculated for fiscal years 2001/02–2003/04. For the numerator, age was calculated at the time of physician visit for a Pap test. The denominator included all females aged 18 to 69 in Manitoba as of December 31, 2002 and excluded women diagnosed with invasive cervical cancer or cervical cancer in situ in the study period or the previous five years. Women experiencing a diagnosis of schizophrenia in hospital separations or medical claims data within the study period or within the previous 9 years were included in the “schizophrenia” group calculation, with all other women in the “all other Manitobans” group. Provincially, there were 3220 woman in the ‘schizophrenia group’ and 335,294 in the ‘all other Manitobans’ group. Region of residence within the province was assigned based on the first record for the three-year period, grouped into five categories: North (three health regions of Burntwood, Churchill and Nor-Man, above the 53rd parallel of latitude); Mid (three health regions of Parkland, Interlake and North Eastman, approximately between the 51st and 53rd parallel); South (three health regions of Assiniboine, Central and South Eastman, between the 51st parallel and the 49th parallel at the Canada–US border); and two urban health regions — Brandon (Manitoba's 2nd largest city) and Winnipeg (the provincial capital). Within Winnipeg, the city was subdivided into three groupings based on previous analyses at MCHP (Martens et al., 2008, 2002): Winnipeg Most Healthy, Winnipeg Average Health and Winnipeg Least Healthy. These subdivisions were based upon the overall population health status of the areas as indicated by the surrogate measure of premature mortality rate (PMR), or age- and sexadjusted death before the age of 75. The number of women in the schizophrenia group and the all other Manitobans group are as follows: North n = 128 and 18,021; Mid n = 312 and 41,181; South n = 320 and 60,172; Brandon n = 148 and 14,385; Winnipeg overall n = 2312 and 201,535; Winnipeg Most Healthy sub-region n = 839 and 103,838; Winnipeg Average Health sub-region n = 741 and 60,776; Winnipeg Least Healthy sub-region n = 732 and 36,921. Negative binominal regression modelling was used to estimate and compare age-adjusted rates of cervical cancer screening for women with schizophrenia vs. the ‘all other Manitobans’ group, by region and provincially. The model controlled for age (both a linear and quadratic age effect were included), region (Manitoba was the reference), group (reference was no schizophrenia diagnosis), and a region by group interaction effect. In a further analysis using logistic regression, the probability of a woman receiving a Pap test in
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
the three-year period was also predicted, to control for possible confounders in the comparison of groups. Covariates in the logistic regression model included: (a) region of residence, including the North, Mid and South rural areas, the city of Brandon, city of Winnipeg (effect coding employed to allow the entire province of Manitoba to be the reference); (b) age, grouped as: 18–29, 30–39, 40–49 (reference group), 50–59, 60–69; (c) average enumeration area household income from the 2001 Canadian Census (used as a continuous variable); (d) a dichotomous variable for continuity of care from a general/family practitioner (GP/FP), with the reference being no continuity of care; and (e) a dichotomous variable for the presence of any major physical Aggregated Diagnostic Groups (ADGs) in 2002/03 as a proxy measure for any severe physical illnesses (reference was no major physical ADGs). A woman was considered to have good continuity of primary physician care if at least 50% of her ambulatory physician visits within the two-year period of 2001/02– 2002/03 were to the same GP/FP. Two logistic regression models were analyzed: (1) including a dichotomous variable for schizophrenia diagnosis in hospital separations or medical claims data in study period or the previous 9 years (reference was no schizophrenia diagnosis); and (2) only including those having a schizophrenia diagnosis and excluding the ‘all other Manitobans’ group. 3. Results Table 1 shows a comparison of demographics for the two comparison groups in the study (women diagnosed with schizophrenia compared to all other women in Manitoba). The three-year cervical cancer screening rates were 58.8% for women diagnosed with schizophrenia compared with 67.8% for ‘all other Manitobans’ (p b .0001). A similar rate differential between 9% and 11% lower for women who were diagnosed with schizophrenia was persistent regionally in Winnipeg (60.7% vs. 70.8%, p b .0001), the South (53.8% vs. Table 1 Demographics and health indicator rates of women aged 18–69 included in this study, comparing those diagnosed with schizophrenia to all other women.
Mean age in years (SD) Mean household income of neighbourhood (SD) Region of residence Urban areas Brandon Winnipeg Rural areas South Mid North Percent receiving “Good Continuity of Primary Care” Percent having at least one major physical ADG Percent having at least one Pap test in three years
Women diagnosed with schizophrenia (n = 3220)
All other women (n = 335,294)
45.4 (13.0) $44,247 ($20,461)
40.4 (13.8) $53,082 ($23,482)
4.60% 71.80%
4.29% 60.11%
9.69% 3.98% 9.94% 66.46%
12.28% 5.37% 17.95% 63.69%
4.60%
4.29%
71.80%
60.11%
103
Table 2 Probability of cervical cancer screening (Pap test) within a three-year period (2001/02–2003/04), comparing women with a schizophrenia diagnosis vs. no diagnosis. Variable (reference)
Odds ratio Probability (95% confidence limits)
Schizophrenia diagnosis (no diagnosis is the reference) Region (Manitoba is the reference) Brandon Winnipeg Mid North South Age (40–49 years is the reference) 18–29 years 30–39 years 50–59 years 60–69 years Average household income (per $10,000) Continuity of care Physical comorbidity as measured by major ADGs
0.70 (0.65–0.75)
p b .0001
1.62 (1.57–1.67) 1.23 (1.21–1.25) 1.07 (1.05–1.09) 0.46 (0.44–0.47) 1.03 (1.01–1.05)
p b .0001 p b .0001 p b .0001 p b .0001 p b .002
0.97 (0.95–0.99) 1.15 (1.12–1.18) 0.82 (0.80–0.84) 0.51 (0.50–0.52) 1.110 (1.106–1.114)
p b .004 p b .0001 p b .0001 p b .0001 p b .0001
1.88 (1.85–1.91) 1.30 (1.28–1.33)
p b .0001 p b .0001
63.8%, p b .04) and Mid (54.2% vs. 65.1%, p b .03) regions. Brandon showed a similar differential (64.3% vs. 73.5%) that was not statistically significant, likely due to small numbers in this region. The two groups in the North experienced the lowest screening rates in the province, and were also similar to each other (41.4% vs. 43.8%, p = .68, NS). Within the capital city of Winnipeg, there was also evidence of a differential in two of the three sub-regions, with patterns similar to the rest of the province when comparing women in the ‘schizophrenia’ group to women in the ‘all other Manitobans’ group: Winnipeg Most Healthy (64.6% vs. 74.9%, p b .002); Winnipeg Average Health (60.3% vs. 69.1%, p b .008); Winnipeg Least Healthy (57.1% vs. 62.9%, p = .06, NS). In the logistic regression modeling with all women (see Table 2), the probability of having a Pap test within the threeyear period was lower for women diagnosed with schizophrenia (adjusted Odds Ratio = 0.70, 95% CI 0.65–0.75, p b .0001) after adjusting for region, age, average household income, continuity of care, and physical comorbidities. Within this model, women were more likely to receive a Pap test if they lived in Brandon, Winnipeg, Mid, and South, and less likely in the North (see Table 2). Compared to women aged 40–49 years, women aged 30–39 were more likely to receive a Pap test, but all other women (ages 18–29, 50–59, 60–69) were less likely. The higher the average household income and the greater the physical comorbidity level, the greater the likelihood of receiving a Pap test. Continuity of care was also associated with a higher likelihood of receiving Pap tests, after controlling for all other factors. A within-group logistic regression including only women diagnosed with schizophrenia (see Table 3) found that women living in Brandon and Winnipeg were more likely to receive a Pap test within the three-year period compared to the Manitoba within-group rate; those living in the North were less likely, and those in Mid and South rural areas were similar to the provincial rate. Women diagnosed with schizophrenia and in the older age brackets (50–59, 60–69) were less likely than woman age 40–49 to receive the appropriate screening,
104
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
Table 3 Probability of cervical cancer screening (Pap test) within a three-year period (2001/02–2003/04), for women with a schizophrenia diagnosis only. Variable (reference) Region (Manitoba is the reference) Brandon Winnipeg Mid North South Age (40–49 years is the reference) 18–29 years 30–39 years 50–59 years 60–69 years Average household income (per $10,000) Continuity of care Physical comorbidity as measured by major ADGs
Odds ratio Probability (95% confidence limits) 1.57 (1.18–2.10) 1.24 (1.09–1.43) 0.98 (0.79–1.21) 0.55 (0.41–0.74) 0.95 (0.77–1.17)
p b .003 p b .002 p = .84, NS p b .0002 p = .64, NS
1.20 (0.94–1.54) 1.00 (0.81–1.25) 0.70 (0.57–0.86) 0.41 (0.33–0.51) 1.07 (1.04–1.12)
p = .14, NS p = .98, NS p b .0006 p b .0001 p b .0002
1.23 (1.05–1.43) 1.21 (1.04–1.41)
p b .02 p b .02
while women age 18–29 and 30–39 were similar to women age 40–49. Women with major physical comorbidities, with higher average household income, and receiving good continuity of care were more likely to have a Pap test. Converting the logistic regression odds ratios into probabilities, for a woman living in an area with an average household income of $40,000, the approximate Pap test rates would be as follows, comparing women with a diagnosis of schizophrenia to women who do not: 60% vs. 68% in Winnipeg (p b .0001); 65% vs. 72% in Brandon (p b .0001); 55% vs. 63% in the South (p b .002); 56% vs. 64% in Mid (p b .0001); 33% vs. 42% in the North (p b .0001). Fig. 1 illustrates the effect of having a schizophrenia diagnosis, having good/not good continuity of care, and average household income on the probability of receiving the Pap test for cervical cancer screening within the three-year period. This figure gives actual probabilities (as derived from the logistic regression odds ratios in Table 2), with all four lines significantly different from each other. As income increases, so does the probability of a Pap test. As well,
those receiving good continuity of care also have the highest screening rates. However, having good continuity of care is associated with a higher rate of screening, no matter what the income level or whether a woman has a diagnosis of schizophrenia. For example, the overall Manitoba provincial probability of receiving a Pap test within three years for a woman living in an area with an average household income of approximately $40,000 per year is 48.3% (95% CI 46.4–50.1) if she has a diagnosis of schizophrenia and lack of good continuity of care, 62.7% (95% CI 60.9–64.4) if she has a diagnosis of schizophrenia and good continuity of care; 56.7% (95% CI 56.1–57.2) with no diagnosis and lack of good continuity of care and 70.6% (95% CI 70.2–71.1) with no diagnosis and good continuity of care. 4. Discussion Manitoba women living with a diagnosis of schizophrenia had substantially lower rates of receiving cervical cancer screening during the three-year period of this study (2001/02–2003/04) compared to women with no diagnosis of schizophrenia, at 58.8% vs. 67.8% provincially. Despite the variability by region of Manitoba, this gap remained persistent at around 9–11%. Factors influencing the uptake of Pap tests were surprisingly similar between women with schizophrenia and the ‘all other Manitobans’ group. Living in an urban area (Brandon or Winnipeg), and in an area with a higher average household income, were both associated with higher screening rates. The largest gap in regional effects was seen for the North, where rates were extremely low in both the age-adjusted rates, as well as in the regression modeling which controlled for income, physical comorbidity and continuity of care. This leads one to speculate that poorer access to health care services (culturally appropriate, community-based services offered through public health or through physician clinics) may be problematic for women in rural areas. Recent efforts started in 2008 by the Manitoba Government and CancerCare Manitoba to initiate a provincewide strategy to increase cervical cancer screening uptake through notification letters and through more accessible services may result in less regional rate inequities, although this will need
Fig. 1. Probability of a Pap test within three years (2001/02–2003/04) according to household income, for women aged 18–69 with and without a schizophrenia diagnosis, and with and without good continuity of care (COC).
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
to be studied over the next few years to determine the outcomes. Literature reviews (Black et al., 2000, 2002) found that letters of invitation produced a 40% increase in screening compared to a control group. In a Cochrane Systematic Review, Forbes et al. (2002) found that intervening through invitation letters continues to show the largest effect in increasing screening rates, with only limited effects shown for educational interventions. Thus the Manitoba approach reflects this literature. However, care must be taken to ensure that these strategies also promote better screening rates for those with schizophrenia. There was also a prominent age effect, with women aged 50–69 showing consistently lower rates, in the low 40% range. This age effect held true across women with schizophrenia as well as those without. This shows the need for concerted effort by physicians and other health care providers to be aware of the need for Pap testing in this vulnerable age group. A somewhat non-intuitive result is that women experiencing a higher level of physical comorbidity also had higher screening rates, with the effect persisting for those experiencing schizophrenia. This may be an indication that more attention is paid to screening when other illnesses are present, or that there are more possible opportunities for physician/patient interactions. Although many of the demographic factors are nonmodifiable, there is the potential for continuity of care to be modifiable through health system level interventions. In this study, good continuity of care was defined as seeing the same family physician for at least 50% of a person's care over a twoyear period. This is presumably a surrogate for more coordinated patient care, although within the limitations of this study, psychiatric care was not examined. Good continuity of care was strongly associated with better cervical cancer screening rates for women with and without schizophrenia. Continuity of care appears to be associated with a 14% higher rate of Pap test screening for both groups (see Fig. 1) and across the income spectrum. Because many women diagnosed with schizophrenia may be treated by psychiatrists, it is important for those physicians to take an active role in recommending and monitoring women's uptake of cervical cancer screening tests. In countries without a universal healthcare system, the disparities may be even greater for those who may not be able to afford healthcare services. This may underscore the importance of providing Pap tests by alternative, publicly funded healthcare providers or primary healthcare teams. Continuity of care, although defined in this study as care by the same physician, is a much broader concept which should include the use of primary care teams with multi-disciplinary providers. This team approach is a noteworthy practice both in Canada and throughout the world, and may overcome income barriers if provided through public health systems. Limitations of this study include the use of administrative healthcare data, with the fact that people who do not contact the healthcare or mental health system will not receive a diagnostic coding of schizophrenia. However, this study went back 10 years to retrieve possible diagnoses. As well, the overall population prevalence of schizophrenia at 1.2% (Martens et al., 2004) using the Manitoba data is similar to reported prevalence from survey sources. The fact that women living with schizophrenia, women living in rural areas and low-income areas, and women who are 50 or older have lower likelihood of having a Pap test
105
points to access issues. Strategies that combine letters of invitation with easily accessible and culturally appropriate health care provision is, according to the literature, the most effective means of increasing cervical cancer screening rates. Particular strategies to reach women experiencing severe mental illness such as schizophrenia may also need to include participation of mental health specialists as part of the primary health care team approach to ensure uptake of appropriate preventive services. Role of funding source None. Contributors Patricia J. Martens, Harvey Chochinov, Randy Fransoo, and the Need To Know Team designed the study and wrote the protocol. Elaine Burland, Patricia J. Martens, and Harvey Chochinov managed the literature searches and analyses. Heather J. Prior undertook the statistical analysis, and Patricia J. Martens wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript. Conflict of interest There are no conflicts of interest. Acknowledgements This work was supported as part of a research project completed in 2008 called, “What Works? A First Look at Evaluating Manitoba's Regional Health Programs and Policies at the Population Level”, one of several projects undertaken each year by the Manitoba Centre for Health Policy (MCHP) under contract to Manitoba Health & Healthy Living. The results and conclusions are those of the authors and no official endorsement by Manitoba Health & Healthy Living was intended or should be inferred. Dr. Harvey Chochinov holds a Canada Research Chair in Palliative Care and received funding for this study from the Canadian Institutes of Health Research. Dr. Patricia Martens would like to acknowledge funding from Canadian Institutes of Health Research (CIHR) which supports her research endeavours through her CIHR/PHAC Applied Public Health Chair, her former CIHR New Investigators' Award (2003–2008), CIHR Community Alliances for Health Research grant (2001–2006), and CIHR KT Award for Regional Impact (2005), all of which helped support The Need to Know Team. Dr. Chochinov would like to acknowledge the support of the Advisory Committee for the Study on Addressing Disability in Cancer Care. The authors are indebted to Decision Support Services of Manitoba Health & Healthy Living, and the Office of Vital Statistics in the Agency of Consumer and Corporate Affairs for the provision of data. The authors also thank Theresa Daniuk and Eileen Bell for research support.
References Alberta Medical Association. Screening for cervical cancer: revised. 2006. Accessed August 7, 2008 at http://www.cancerboard.ab.ca/accsp/pdf/ accsp_guide_cervical_cancer_06-03-07.pdf. Black, M., Yamada, J., Mann, V., Cava, M., Micucci, S., 2000. Effectiveness of strategies to increase cervical cancer screening: a systematic review of the literature 1989–1999. Toronto, Ontario: Public Health Branch, Ontario Ministry of Health — Report of the Effective Public Health Practice Project (EPHPP) Steering Committee. Black, M., Yamada, J., Mann, V., 2002. A systematic literature review of the effectiveness of community-based strategies to increase cervical cancer screening. Can. J. Public Health 93, 386–393. Brown, S., Inskip, H., Barraclough, B., 2000. Causes of the excess mortality of schizophrenia. Br. J. Psychiatry 177, 212–217 Sep. Canadian Cancer Society. Cancer screening in Canada not realizing full potential, 2006. Accessed August 7, 2008 at http://www.cancer.ca/ccs/ internet/mediareleaselist/0,,3543_434465_943594571_langId-en,00. html. Copeland, L.A., Zeber, J.E., Rosenheck, R.A., Miller, A.L., 2006. Unforeseen inpatient mortality among veterans with schizophrenia. Med. Care 44, 110–116. Fleischhacker, W.W., Cetkovich-Bakmas, M., De Hert, M., Hennekens, C.H., Lambert, M., Leucht, S., Maj, M., McIntyre, R.S., Naber, D., Newcomer, J.W., Olfson, M., Osby, U., Sartorius, N., Lieberman, J.A., 2008. Comorbid
106
P.J. Martens et al. / Schizophrenia Research 113 (2009) 101–106
somatic illnesses in patients with severe mental disorders: clinical, policy, and research challenges. J. Clin. Psychiatry 69 (4), 514–519 Apr. Forbes, C., Jepson, R., Martin-Hirsch, P., 2002. Interventions targeted at women to encourage the uptake of cervical screening. Cochrane Database Syst. Rev. (3), CD002834. Kilbourne, A.M., McCarthy, J.F., Edward, P.P., Welsh, D., Frederic, C.B., 2007. Social support among veterans with serious mental illness. Soc. Psychiatry Psychiatr. Epidemiol. 42, 639–646. Kirkpatrick, B., Messias, E., Harvey, P.D., Fernandez-Egea, E., Bowie, C.R., 2008. Is schizophrenia a syndrome of accelerated aging? Schizophr. Bull. 34 (6), 1024–1032. Lee, J., Parsons, G.F., Gentleman, J.F., 1998. Falling short of Pap test guidelines. Health Rep. 10 (1), 9–19 Accessed August 7, 2008 at http://www.statcan. ca/english/studies/82-003/feature/hrar1998010001s2a01.pdf. Lofters, A., Glazier, R., Agha, M., Creatore, M., Moineddin, R., 2007. Inadequacy of cervical cancer screening among urban recent immigrants: a populationbased study of physician and laboratory claims in Toronto, Canada. Prev. Med. 44, 536–542. Lotocki, R.J., 2000. The Pap smear: guidelines for screening and follow-up. Can. J. CME Accessed August 7, 2008 at http://www.cancercare.mb.ca/ cancercare_resources/MCCSP/pdfs/professionals_article_1.pdf. Martens, P.J., Frohlich, N., Carriere, K., Derksen, S., Brownell, M., 2002. Embedding child health within framework of regional health: population health status and sociodemographic indicators. Can. J. Public Health 93 (Supplement 2), S15–S20. Martens, P., Fransoo, R., McKeen, N., The Need to Know Team, Burland, E., Jebamani, L., Burchill, C., DeCoster, C., Ekuma, O., Prior, H., Chateau, D., Robinson, R., Metge, C., 2008. Patterns of Regional Mental Illness Disorder Diagnoses and Service Use in Manitoba: A Population-Based Study. Manitoba Centre for Health Policy, Winnipeg, Manitoba. Accessed April 7,
2008 at http://mchp-appserv.cpe.umanitoba.ca/reference/mental.health. pdf. Martens, P., Fransoo, R., The Need to Know Team, Burland, E., Prior, H., Burchill, C., Romphf, L., Chateau, D., Bailly, A., Ouelette, C., 2008. What Works? A First Look at Evaluating Manitoba's Regional Health Programs and Policies at the Population Level. Manitoba Centre for Health Policy, Winnipeg, Manitoba. Accessed August 7, 2008 at http://mchpappserv. cpe.umanitoba.ca/reference/fullwwreport.pdf. Mateen, F.J., Jatoi, A., Lineberry, T.W., Aranguren, D., Creagan, E.T., Croghan, G.A., Jett, J.R., Marks, R.S., Molina, J.R., Richardson, R.L., 2008. Do patients with schizophrenia receive state-of-the-art lung cancer therapy? A brief report. Psychooncology 17 (7), 721–725 Jul. Snider, J.A., Beauvais, J.E., 1998. Pap smear utilization in Canada: estimates after adjusting the eligible population for hysterectomy status. Chronic Dis. Can. 19 (1), 19–24. Spence, A.R., Goggin, P., Franco, E.L., 2007. Process of care failures in invasive cervical cancer: systematic review and meta-analysis. Prev. Med. 45 (2–3), 93–106 Aug–Sep. Statistics Canada. Canadian Community Health Survey (CCHS 3.1), 2005. CANSIM Table 105-0442. Accessed September 25, 2007 at http://www. statcan.ca. Tal, A., Roe, D., Corrigan, P.W., 2007. Mental illness stigma in the Israeli context: deliberations and suggestions. Int. J. Soc. Psychiatry 53 (6), 547–563. Vandiver, V., 2007. Health promotion as brief treatment: strategies for women with co-morbid health and mental health conditions. Brief Treatment Crisis Interv. 7, 161–175. Wrigley, S., Jackson, H., Judd, F., Komiti, A., 2005. Role of stigma and attitudes toward help-seeking from a general practitioner for mental health problems in a rural town. Aust. N. Z. J. Psychiatry 39 (6), 514–521 Jun.