Recent Pap Tests among Canadian Women: Is Depression a Barrier to Cervical Cancer Screening?

Transcription

1 JOURNAL OF WOMEN S HEALTH Volume 17, Number 7, 2008 Mary Ann Liebert, Inc. DOI: /jwh Recent Pap Tests among Canadian Women: Is Depression a Barrier to Cervical Cancer Screening? Angela Kaida, M.Sc., 1 Ian Colman, Ph.D., 2 and Patricia A. Janssen, Ph.D. 1,3 Abstract Background: Previous studies have shown that depression is associated with both lower use of preventive cancer screening programs and lower probability of cancer survival. Given the increasing incidence of depression among Canadian women, this study sought to determine if recent Pap testing varies by the presence of depression. Methods: This population-based study used cross-sectional, self-reported data from the Canadian Community Health Survey (CCHS) Cycle 3.1 (2005) to estimate the association between depression and recent Pap testing within the previous 3 years among 2351 Canadian women without hysterectomy aged years. Results: After adjustment for confounders, depressed women had nonsignificantly increased odds of a recent Pap test (OR 1.08, 95% CI 0.95, 1.29); however, age was an important effect modifier of this relationship. Young depressed women (18 29 and years) were significantly more likely to report a recent Pap test (AOR 1.78, 95% CI 1.37, 2.31, and AOR 1.47, 95% CI 1.00, 2.15, respectively), whereas middle-aged depressed women (40 49 and years) were significantly less likely to report a recent Pap test (AOR 0.76, 95% CI 0.58, 0.98, and AOR 0.68, 95% CI 0.50, 0.93, respectively) compared with their nondepressed counterparts. No significant relationship was detected for the oldest age group (60 69 years). Conclusions: To our knowledge, these results are the first to demonstrate an interaction effect of age on the association between depression and recent Pap testing. Longitudinal studies should be conducted to explore the role of age as an effect modifier of this relationship and to inform policy and programming aimed at improving rates of cervical cancer screening across all age groups. Introduction LTHOUGH THE INCIDENCE of cervical cancer in Canada has Adeclined dramatically over the last 30 years, 1 it remains the second most common type of cancer among women under 50 years of age. 2 Regular Papanicolaou (Pap) test screening constitutes the single most important determinant of declines in cervical cancer incidence and related mortality. 1 4 Although cervical cancer is almost entirely preventable with regular Pap test screening and such screening is covered under Canada s publicly funded universal access healthcare system, 5 more than 50% of eligible women report not having been screened within the previous 3 years. 1 Numerous studies have sought to identify characteristics associated with lower Pap testing rates to better target screening promotion initiatives. Women who are not screened regularly tend to be older ( 50 years of age), 6,7 immigrants, 6,8 of Aboriginal descent 6 or nonwhite ethnicity, 9 have lower education and socioeconomic status, 6 8 and live in rural areas. 8 In addition, studies have reported on the independent influence of comorbid illnesses, 10 risky health behaviors, 11 and provider characteristics 12 on cancer screening rates. An emerging body of evidence indicates that depression may also be associated with lower use of preventive cancer screening programs. Most of this research has focused on breast cancer screening and reveals that depressed women are less likely to self-report mammography, 13 to attend breast cancer screening clinics, 14,15 and to receive mammography 16 than their nondepressed counterparts. These stud- 1 Department of Health Care and Epidemiology, Faculty of Medicine, University of British Columbia (UBC), Vancouver, British Columbia, Canada. 2 Department of Psychiatry, University of Cambridge, Cambridge, U.K. 3 Child and Family Research Institute, Vancouver, British Columbia, Çanada. A.K. is supported by the Canadian Institutes of Health Research (CIHR) and the Michael Smith Foundation for Health Research (MSFHR). 1175

2 1176 ies, however, suffer from limitations such as small sample sizes, 15 highly homogeneous study samples, 16 enrollment restricted to older women, 13,14,16 and nonstandardized measures of depression. 14,15 The only identified study of cervical cancer screening reported that women with a high burden of depressive symptoms at baseline are less likely to undergo Pap testing in the subsequent year of followup, although the trend was not significant. 13 However, this study restricted enrollment to women aged rather than the full age range for whom cervical cancer screening is indicated. Given the dearth and limitations of studies on this topic and the increasing incidence of depression among women across Canada, 17 it is important to define the association between depression and Pap test screening among eligible women across the general population. The purpose of this study was to determine if the prevalence of recent Pap testing varied according to the presence of depression among women aged 18 69, using populationbased data from the Canadian Community Health Survey (CCHS) Cycle 3.1. We hypothesized that depressed women would be less likely to have undergone cervical cancer screening within the previous 3 years. Materials and Methods Data source Data for this study were obtained from the CCHS Cycle 3.1. The CCHS is a nationally representative cross-sectional survey that collects data on health status, healthcare utilization, and determinants of health every 2 years. The target population includes individuals aged 12 years living in private dwellings in Canada s 10 provinces and 3 territories. Individuals are excluded if they live in institutions, on First Nations reserves, on government-owned land, or in certain remote regions. Thus, the CCHS represents approximately 98% of the Canadian population aged The CCHS Cycle 3.1 employed a complex multistage sampling strategy to randomly select households in 125 health regions. Individuals living in each selected household were randomly chosen to participate in the survey. 18 Data collection occurred between January and December A total of 132,947 valid interviews were completed. Approximately half of the interviews were conducted in person, and the other half were conducted by telephone. 18 Further details on the methodology of the CCHS Cycle 3.1 are reported elsewhere. 18 Study sample In accordance with current Canadian cervical cancer screening guidelines, we restricted the study sample to women aged years, without hysterectomy. 7,19 22 The sample was further restricted to women from six provinces (Nova Scotia, Prince Edward Island, Quebec, Saskatchewan, Alberta, and British Columbia) that included the optional depression module in the CCHS Cycle 3.1. Consequently, our final study sample included 25,351 women. With 25,351 women, we had sufficient power (type I error 0.05/type II error=0.20) to detect a difference in the prevalence of recent Pap testing by depression status equal to , which corresponds to an odds ratio (OR) of KAIDA ET AL. Measurement of recent Pap test The outcome variable, recent Pap testing, was assessed through the use of two survey questions. All respondents were asked: Have you ever had a Pap smear test? If the answer was Yes, the respondent was then asked When was the last time you had a Pap smear test? Responses to these questions were dichotomized into Had a Pap smear test within the previous 3 years and Did not have a Pap smear test in the previous 3 years. Pap testing every 3 years is a commonly accepted screening interval in Canada. 7,19 21 Measurement of depression Data on the prevalence of depression were collected using a Short-Form scale of items from the Composite International Diagnostic Interview (CIDI-SF). The CIDI-SF is a widely used structured diagnostic instrument designed to produce diagnoses according to the definitions and criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) and the ICD-10 Classification of Mental and Behavioral Disorders (ICD-10). 24 Scores on the CIDI-SF scale range from 0 to 8, with higher scores indicating higher levels of depression. 24 Individuals with a score of 5 are considered to have at least a 90% probability of being diagnosed as having experienced a major depressive episode in the past 12 months. 24,25 Thus, we defined depressed individuals as those with a CIDI-SF score of 5 and nondepressed individuals as those with a score of 5. Covariates Variables known to be associated with Pap testing and depression were included in the analysis to provide an adjusted estimate of the association. Covariates included age group, highest level of education attained, annual household income, marital status, visible minority status, whether the individual has a regular doctor, and number of consultations with a family physician/general practitioner (GP) in the previous year. We grouped women by age a priori by selecting the age categories commonly used to report Pap smear testing and cervical cancer rates by the Canadian Cancer Agency. 1 The list and grouping of all covariates are shown in Table 1. Statistical analysis The prevalence of recent Pap testing and depression for Canadian women aged years (without hysterectomy) was computed. Univariate analyses were performed to assess the relationship between recent Pap testing and depression, as well as survey covariates. Differences in recent Pap testing between groups are reported using Pearson s chisquare test statistic (for categorical covariates) or Student s independent t test (for continuous variables). The association between recent Pap testing and depression is reported using a crude OR with a 95% confidence interval (95% CI). Multivariate logistic regression was used to measure the presence and strength of the association between depression and the likelihood of recent Pap testing while controlling for the effect of covariates. After testing for colinearity (using Spearman s rho, ) 26 and interaction, 27 all covariates tested in the univariate analysis were included in the final model to obtain adjusted ORs (and 95% CIs). All statistical tests were two-sided and considered significant at Data were analyzed using SAS version 9.1 for Windows (SAS Institute, Cary, NC).

3 DEPRESSION AND RECENT PAP TESTING 1177 TABLE 1. BASELINE CHARACTERISTICS AND UNIVARIATE ASSOCIATIONS WITH RECENT PAP TESTING (WITHIN PREVIOUS 3 YEARS) AMONG CANADIAN WOMEN AGED YEARS (WITHOUT HYSTERECTOMY): CCHS a CYCLE 3.1, 2005 b,c Recent Pap test Frequency % Yes % No p Characteristic (n 25,351) (%) (n 19,376) (n 5,975) value c Explanatory variable Depression 0.02 No 23, Yes 1, Sociodemographic variables Age group, years , , , , , Education High school 3, High school graduate 4, Some post-secondary 2, Post-secondary graduate 15, Household income No or $15,000 1, $15,000 $29,999 2, $30,000 $49,999 4, $50,000 $79,999 6, $80,000 6, Marital status Married/common law 16, Widowed/separated/divorced 2, Single, never married 5, Ethnicity White 21, Visible minority 3, Healthcare utilization variables Regular family doctor No 3, Yes 21, Mean number of consultations 3.21(4.3) 3.64(4.6) 2.75(4.3) with family doctor/gp in previous year (SD) a CCHS, Canadian Community Health Survey. b All Not stated responses were removed from this table. In general, with the exception of Household income, these responses comprised 1% of the responses in each characteristic. However, 15% of respondents did not state their household income. c Values may not add up because of rounding. d Differences between groups are reported using Pearson s chi-square test statistic (for categorical variables) or Student s independent t test (for continuous variables). e SD, standard deviation. The frequency weights provided by Statistics Canada for CCHS Cycle 3.1 were applied to generate point estimates that adjust for differences in individual and household sampling probability and nonresponse. 18 The frequency weights were rescaled to yield standardized probability weights with a mean of 1 and a sum equal to the sample size. The standardized probability weight variable was calculated by dividing the frequency weight for each individual by the mean weight of the analytical sample. The probability weights were applied to the analysis to derive more appropriate estimates of the variance of the study sample and, thereby, to provide more precise confidence intervals. Because the stratification and clustering of the sample s design are still not completely taken into account, the variance estimates calculated in this way may still be underestimates. 18,28 Results The large majority of women aged years without hysterectomy reported having a Pap test (76%) within the previous 3 years. Baseline characteristics Distributions of baseline covariates are presented in Table 1. Approximately 7% of women in this sample were charac-

4 1178 KAIDA ET AL. terized as having depression. Most women were married or common law (65%), had graduated from postsecondary school (62%), and were white (86%). The vast majority had a regular family doctor (85%), and the mean number of consultations with a family doctor or GP in the last year was 3.2 (SD 4.3). Univariate analyses In the unadjusted analyses, all the measured baseline characteristics were significantly associated with recent Pap testing (Tables 1 and 2). Depressed women were significantly more likely to report a recent Pap test (OR 1.14, 95% CI 1.02, 1.28). Women who reported a recent Pap test were also more likely to be between 30 and 59 years of age, have a higher education and household income, be currently married, and be of white ethnicity. They were also more likely to have a regular family doctor and reported a higher mean number of GP visits in the previous year. TABLE 2. Adjusted analyses After adjusting for potential confounders and established covariates shown in Table 1, depression was no longer significantly associated with recent Pap testing, although the direction of the effect remained the same (adjusted odds ratio [AOR] 1.08; 95% CI 0.95, 1.24) (Table 2). Whereas all other measured baseline characteristics maintained significant associations, having a regular family doctor remained most strongly associated with reporting a recent Pap smear (AOR 2.49, 95% CI 2.28, 2.72). In the unadjusted model, the odds of recent Pap testing were higher for older women, with the exception of women aged years, compared with women in the youngest age group. In the adjusted model, however, the odds of recent Pap testing were lower for older women compared with women in the youngest age group. We thus performed an analysis stratified by age group to test for possible effect modification of the association between depression and re- UNADJUSTED AND ADJUSTED ORS AND 95% CI OF VARIABLES ASSOCIATED WITH RECENT PAP TESTING (WITHIN PREVIOUS 3 YEARS) AMONG CANADIAN WOMEN AGED YEARS (WITHOUT HYSTERECTOMY): CCHS CYCLE 3.1, 2005 a Unadjusted OR Adjusted OR OR 95% CI AOR 95% CI Explanatory variable Depression No Ref Ref Ref Ref Yes , , 1.24 Sociodemographic variables Age group, years Ref Ref Ref Ref , , , , , , , , 0.59 Education High school Ref Ref Ref Ref High school graduate , , 2.02 Some post-secondary , , 1.87 Post-secondary graduate , , 2.62 Household income No or $15,000 Ref Ref Ref Ref $15,000 $29, , , 1.30 $30,000 $49, , , 1.42 $50,000 $79, , , 1.56 $80, , , 2.00 Marital status Married/common law Ref Ref Ref Ref Widowed/separated/divorced , , 0.99 Single, never married , , 0.57 Ethnicity White Ref Ref Ref Ref Visible minority , , 055 Healthcare utilization variables Regular family doctor No Ref Ref Ref Ref Yes , , 2.72 Mean number of consultations , , 1.08 with family doctor/gp in previous year (per additional consultation) a CCHS, Canadian Community Health Survey; OR, odds ratio; AOR, adjusted odds ratio; Ref, reference category.

5 DEPRESSION AND RECENT PAP TESTING 1179 cent Pap testing. As shown in Figure 1, age group was an effect modifier of this relationship (p ). Depressed young women (aged and years) were significantly more likely to report recent Pap testing (AOR 1.78, 95% CI 1.37, 2.31, and AOR 1.47, 95% CI 1.00, 2.15, respectively). However, depressed middle-aged women (aged and years) were significantly less likely to report a recent Pap test compared with their nondepressed counterparts (AOR 0.76, 95% CI 0.58, 0.98, and AOR 0.68, 95% CI 0.50, 0.93, respectively). Among seniors (aged 60 69) there was no relationship between depression and recent Pap testing. Discussion In contrast to our initial hypothesis, we found that depressed Canadian women were slightly more likely to report a recent Pap test compared with nondepressed women, although the relationship was not statistically significant after adjusting for known confounders. We found, however, that age modified the effect of depression on Pap testing. Whereas depressed younger women (aged years) were more likely to have a recent Pap smear, depressed middleaged women (40 59 years) were less likely to have a recent Pap smear compared with their nondepressed counterparts. No such relationships were detected among senior women (60 69 years). Our findings for middle-aged women are consistent with those from a recent American study that enrolled women aged years and reported a 12% decrease (nonsignificant) in Pap testing among depressed women. 13 These findings were similar despite different study and healthcare system contexts, namely, socialized medicare available in Canada vs. mainly privatized health services available to American women. These findings are also concordant with a large study of military veterans that reported individuals with psychiatric disorders were less likely to receive a range of preventive services in primary care, including cervical cancer screening. 29 Although older adults with depression are more likely to worry about their physical health, they are less likely to be motivated to use health services to ease these concerns. 30 It is less clear why depressed young women (aged years) were more likely to report recent Pap testing. It may be the case that when young women access the healthcare system for mental health concerns, they are also encouraged to undergo routine preventive screening programs, including cervical cancer screening. This may be particularly so given that mental and sexual/reproductive health concerns constitute two of the most common reasons why young women visit GPs. 31,32 It is also possible that depressed young women go to GPs with somatic symptoms rather than psychiatric symptoms, 33 making it less likely that the physician will diagnose the mental illness. 34 Consequently, physicians may be more likely to focus on physical health, including preventive health, in these situations. Such explanations, however, do not fully account for why these effects would differ according to the age of the patient. The role of age as an effect modifier of the relationship between Pap smear testing and depression may be a function FIG. 1. Adjusted odds ratios (AOR) and 95% confidence intervals (95% CI) of the association between depression and recent Pap testing among Canadian women (without hysterectomy) stratified by age group, Canadian Community Health Survey Cycle 3.1, 2005.

6 1180 of broader sexual and reproductive behaviors or physician counseling practices. Consistent with national screening guidelines, women who have a history of sexually transmitted infections, have recently been pregnant, or have had a prior abnormal Pap smear may be more likely to be advised to have a Pap smear, and younger women in our study were more likely than older women to report these behaviors (data not shown). Further, competing screening demands may play a role. Among women over the age of 50, other cancer screening measures (including mammography and colonoscopy), counseling around menopause, and other chronic disease prevention may take priority. Thus, in light of these competing screening demands, physicians may be less likely to encourage older women to undergo cervical cancer screening. In these circumstances, depression, potentially through pathways of self-neglect 35,36 and low perceived self-efficacy to affect health outcomes, 37 may lead to poorer cervical cancer screening-seeking behaviors, as shown here. As depression is more common among middleaged women in Canada, 17 the observed small effect indicates that a considerable number of middle-aged women are not receiving regular cervical cancer screening. Consistent with other studies, numerous sociodemographic and healthcare utilization variables were also strongly associated with the likelihood of cervical cancer screening. Importantly, having a regular family doctor was the strongest independent predictor of a recent Pap smear. The limitations of this study must be acknowledged. First, the cross-sectional nature of this analysis precludes us from determining causality between the explanatory variable and the outcome. Although reverse causality (i.e., recent Pap testing influencing depression) is considered unlikely, it is possible that some women who have an abnormal Pap test may experience anxiety and depression of a serious enough nature to be detected through the CIDI- SF. 38 Second, there is potential for recall bias, as women were asked to self-report their Pap smear history. Although it was not possible to verify reports with medical records, self-report for Pap testing has been shown to be a reasonable proxy of actual screening history, particularly over a 3-year recall period. 39 A third potential limitation relates to the different assessment periods for Pap testing and depression; the CCHS determined Pap screening in the last 3 years, whereas depression was assessed within the last 1 year. Thus, women who had a Pap test 2 3 years ago may have had a different depression status at that time. Finally, we assumed eligibility for screening based on being a female aged 18 69, without hysterectomy. The Canadian guidelines for cervical cancer screening recommend Pap testing beginning at onset of sexual activity 7,19 21 ; however, we did not distinguish nonsexually active women from our sample. Given high reported rates of ever having had sexual intercourse among Canadian women, 25 it is unlikely that failing to control for this eligibility criterion for screening would have biased our results. These findings are strengthened by our use of a nationally representative, population-based survey sample. The large sample size allowed for adjustment of known and important confounders while maintaining sufficient power to detect associations. In addition, we used a well-validated measure of depression (i.e., the CIDI-SF) and a stringent definition of depression in the analyses (i.e., CIDI-SF score of 5). 24 Finally, our data are from Canada, where cervical cancer screening and mental healthcare are provided through a publicly funded, universal access healthcare system whereby access to these services should not, in theory, be mediated by an ability to pay. 5 Conclusions To our knowledge, these results are the first to demonstrate an interaction effect of age on the association between depression and recent Pap testing. Longitudinal studies should be conducted to explore the role of age as an effect modifier of this relationship and to inform policy and programming aimed at improving rates of cervical cancer screening among women of all ages. Acknowledgments We thank Dr. Mieke Koehoorn for providing helpful feedback and guidance on earlier drafts of the manuscript. Disclosure Statement No competing financial interests exist. KAIDA ET AL. References 1. Canadian Cancer Society/National Cancer Institute of Canada. Canadian Cancer Statistics Toronto, Canada: Canadian Cancer Society/National Cancer Institute of Canada, ISSN Canadian Cancer Society (Ontario Division). Insight on cancer: News and information on cervical cancer. Toronto, Canada: Canadian Cancer Society, Peto J, Gilham C, Fletcher O, et al. The cervical cancer epidemic that screening has prevented in the U.K. Lancet 2004;364: Quinn M, Babb P, Jones J, et al. Effect of screening on incidence of and mortality from cancer of cervix in England: Evaluation based on routinely collected statistics. BMJ 1999;318: Health Canada. Canada s health care system, Available at Accessed May 9, Ontario Cervical Screening Program. Program report, Toronto, Canada: Cancer Care Ontario, Stuart G, Taylor G, Bancej CM, et al. Report of the 2003 pan- Canadian forum on cervical cancer prevention and control. J Obstet Gynaecol Can 2004;26: Goel V. Factors associated with cervical cancer screening: Results from the Ontario health survey. Can J Public Health 1994;85: Quan H, Fong A, De Coster C, et al. Variation in health services utilization among ethnic populations. Can Med Assoc J 2006;174: Wee CC, McCarthy EP, Davis RB, et al. Screening for cervical and breast cancer: Is obesity an unrecognized barrier to preventive care? Ann Intern Med 2000;132: Rakowski W, Clark MA, Ehrich B. Smoking and cancer screening for women ages 42 75: Associations in the national health interview surveys. Prev Med 1999;29: Haggstrom DA, Phillips KA, Liang SY, et al. Variation in screening mammography and Papanicolaou smear by primary care physician specialty and gatekeeper plan (United States). Cancer Causes Control 2004;15:

7 DEPRESSION AND RECENT PAP TESTING Pirraglia PA, Sanyal P, Singer DE, et al. Depressive symptom burden as a barrier to screening for breast and cervical cancers. J Womens Health 2004;13: Aro AR, de Koning HJ, Absetz P, et al. Two distinct groups of non-attenders in an organized mammography screening program. Breast Cancer Res Treat 2001;70: Burton MV, Warren R, Price D, et al. Psychological predictors of attendance at annual breast screening examinations. Br J Cancer 1998;77: Carney CP, Jones LE. The influence of type and severity of mental illness on receipt of screening mammography. J Gen Intern Med 2006;21: Patten SB, Wang JL, Williams JV, et al. Descriptive epidemiology of major depression in Canada. Can J Psychiatry 2006;51: Statistics Canada. Canadian Community Health Survey (CCHS) Cycle 3.1 (2005): Public use microdata file (PUMF) user guide. Ottawa, Canada: Statistics Canada, McLachlin CM, Mai V, Murphy J, et al. Cervical screening: A clinical practice guideline. Toronto, Canada: Cancer Care Ontario, Parboosingh EJ, Anderson G, Clarke EA, et al. Cervical cancer screening: Are the 1989 recommendations still valid? National workshop on screening for cancer of the cervix. Can Med Assoc J 1996;154: Morrison BJ. Screening for cervical cancer. Ottawa, Canada: Health Canada, Sirovich BE, Welch HG. Cervical cancer screening among women without a cervix. JAMA 2004;291: Hsieh FY, Bloch DA, Larsen MD. A simple method of sample size calculation for linear and logistic regression. Stat Med 1998;17: Kessler R, Andrews G, Mroczek D, et al. The World Health Organization Composite International Diagnostic Interview Short-Form (CIDI-SF). Int J Methods Psychiatr Res 1998;7: Statistics Canada. Canadian Community Health Survey (CCHS) Cycle 3.1 (2005): Public use micro data file (PUMF): Integrated derived variable (DV) and grouped Variable Specifications. Ottawa, Canada: Statistics Canada, Pagano M, Gauvreau K. Principles of biostatistics, 2nd ed. Pacific Grove, CA: Duxbury, Van Ness PH, Allore HG. Using SAS to investigate effect modification. Paper SUGI 31. Statistics and data Analysis. Available at www2.sas.com/proceedings/ sugi31/ pdf Accessed October 10, Binder DA, Roberts GR. Design-based and model-based methods for estimating model parameters. In: Chambers R, Skinner C, eds. Analysis of survey data. Chichester: Wiley, 2003: Druss BG, Rosenheck RA, Desai MM, Perlin JB. Quality of preventive medical care for patients with mental disorders. Med Care 2002;40: Levinson CM, Druss BG. Health beliefs and depression in a group of elderly high utilizers of medical services. Gen Hosp Psychiatry 2005;27: Mustard CA, Kaufert P, Kozyrskyj A, et al. Sex differences in the use of health care services. N Engl J Med 1998;338: Statistics Canada. How healthy are Canadians report. Ottawa, Canada: Statistics Canada, Report No XPE. 33. Starkes JM, Poulin CC, Kisely SR. Unmet need for the treatment of depression in Atlantic Canada. Can J Psychiatry 2005;50: Kirmayer LJ, Robbins JM, Dworkind M, et al. Somatization and the recognition of depression and anxiety in primary care. Am J Psychiatry 1993;150: Abrams RC, Lachs M, McAvay G, et al. Predictors of selfneglect in community-dwelling elders. Am J Psychiatry 2002;159: Pavlou MP, Lachs MS. Could self-neglect in older adults be a geriatric syndrome? J Am Geriatr Soc 2006;54: Hogenmiller JR, Atwood JR, Lindsey AM, Johnson DR, Hertzog M, Scott JC Jr. Self-efficacy scale for Pap smear screening participation in sheltered women. Nurs Res 2007; 56: Gray NM, Sharp L, Cotton SC, et al. Psychological effects of a low-grade abnormal cervical smear test result: Anxiety and associated factors. Br J Cancer 2006;94: Caplan LS, McQueen DV, Qualters JR, et al. Validity of women s self-reports of cancer screening test utilization in a managed care population. Cancer Epidemiol Biomarkers Prev 2003;12: Address reprint requests to: Angela Kaida, M.Sc. Department of Healthcare and Epidemiology Faculty of Medicine University of British Columbia (UBC) 5804 Fairview Avenue Vancouver, British Columbia V6T 123 Canada akaida@interchange.ubc.ca

8