Phase III – Mixed-Method Evaluation Study via a Randomized-Controlled Trial

Stage of adoption and impact of direct-mail communications with and without phone intervention on Chinese women’s cervical smear screening behavior

Stage of Adoption and Impact of Direct-Mail Communications with and without Phone Intervention on Chinese Women’s Cervical Smear Screening Behavior


Su-I Hou, DrPH, CPH, MCHES, RN

Abstract

Background:  The purposes of the study were to (1) assess the impact of direct-mail communications with and without phone intervention, and (2) examine the characteristics of women who were more likely to respond.  Methods: Women were recruited from female family members of inpatients admitted to one of the major teaching hospitals in Taiwan (n=424), and were randomly assigned into an intervention group, who received direct-mails and a phone follow-up, or a control group, who received placebo messages. Results: Logistic regression analysis showed that women in the intervention group (ORadj.=2.31) and contemplation stage (ORadj.=4.18) were more likely to receive a screening at the end of the program.  Among women in the intervention group, contemplators were 5.58 times more likely to receive a screening before the phone intervention (early adopters); and 40% of the screening adopters responded after the phone intervention (late adopters).  Late and early adopters were similar in their stage, age, and education. Conclusions: Stage and intervention are both significant predictors of screening adoption. The study provides justification for programs to target women in contemplation stage. It also suggests that the boost of a later phone intervention may be consequential for encouraging more women with similar demographics to take action.

Keywords: stage of adoption; intervention studies; evaluation; cervical smears; screening; Chinese

Note

This is an Accepted Manuscript of an article published by Preventive Medicine in 09/2005, available online: https://www.sciencedirect.com/science/article/pii/S0091743505001131?via%3Dihub

Introduction

Cervical cancer is the second most common cancer among women, accounting for an estimated 12% of all female cancers worldwide.  Some 80% of these cancers occur in developing countries, including Southeast Asia.  And it has become the leading cancer among women in those regions [1].  In the U.S., the widely available and utilized cervical smear screening (Pap smear) has contributed greatly to the 70% decrease in the death rate over the past 50 years [2].

However, the few existing studies conducted among Chinese women showed low screening utilization.  Hiatt and associates (2001) assessed breast and cervical cancer screenings among ethnic groups (including Chinese women) in the San Francisco Bay area in the U.S. Results from their baseline survey showed that cervical smear screening in the past 3 years was markedly lower among non-English-speaking Chinese women (24%), and only 40% of them reported having ever had a cervical smear [3].  Another study [4] using data from a population-based phone survey in Los Angeles indicated that Chinese women had one of the lowest cervical screening rates, compared with white women (56% vs. 81%).  The disparity for cervical smear screening rate between whites and Chinese was significant even after controlling for major factors such as demographics, socioeconomic and health status.  Similar low rate of screening has also been observed among Chinese women in Southeast Asia countries such as Taiwan.  In Taiwan, the National Health Insurance Plan (1996) provides free health care coverage for an annual cervical smear screening among women aged 30 and over, or younger if married.  Despite the universal coverage, the prevalence of screening utilization reported in the few existing studies was low, ranging from 58% to 70% [5-7].  The screening utilization among Chinese women is far from satisfactory.

Although the proven effectiveness of cervical smear screenings in reducing mortality from cervical cancer has led to various public health programs to encourage screening [2,8], few have targeted Chinese women [3,9-11].  Recently, there have been some community based intervention programs in the U.S. targeting Asian ethnic groups, such as Vietnamese-American [12-15], Cambodian Americans [16-17], and Korean Americans [18].  Interventions included mostly labor intensive strategies such as small group educational sessions provided by trained bicultural lay health outreach workers [3,9,12,14,16,], home visits [9,16], counseling on perceived barriers and logistic assistance [9,17].  Less labor intensive strategies have been tried, however, these intervention usually succeeded only in increased screening recognition, intention, or awareness, but not behavior [13,14].  Taylor and colleagues [9] used a direct-mail intervention in one of their study arms; and were able to show some increase in screening utilization (p=.03).

In summary, current knowledge is limited regarding (1) the types of less labor intensive strategies that can also be effective or sufficient to encourage screening behavior; and (2) the segments of the screening non-adherent women who are more likely to respond to such messages regarding cervical smear screening.

The present approach was guided by the Transtheoretical Model (TTM) [19].  The model suggests that the acquisition and maintenance of a health protective behavior is not an all-or-none phenomenon, but a gradual process [20].  In the TTM defines stages of adoption are derived from information on past and present screening behavior, and intention to get a screening in the future.  This framework has recently been applied to mammography [21-23]. It has also been applied to minority groups [21,24-25].  Tailored intervention using stage-matched materials have been shown to improve mammography use [23,26].  Few studies have tried to apply the model to cervical smear screening behavior.  Only one published study was found that described cervical smear screening behavior in Southeast Asian women [27], and none were found that utilized concepts from the TTM in intervention or evaluation studies.

The intervention used in this study consisted of two direct-mail communications, one per month, and a phone intervention at the third month, to encourage non-screening women to obtain cervical smear screenings.  These direct-mail communications reflected messages relevant to different stage of readiness.  Each participant in the intervention group received materials relevant to all stages, allowing them to respond to the materials most relevant to their own stage of adoption.  Finally, there was a phone intervention at the third month.  In this call, participants were able to make known whether they had had a cervical smear screening; if not, they were encouraged to have one.  The overall effectiveness of the intervention on cervical cancer smear screening completion rate at the end of the program, as well as changes of several screening related beliefs, before and after the intervention, were reported earlier [28]. Results from this control randomized trial showed that the overall intervention was effective in encouraging women to receive a screening.  The intervention message was guided by TTM, and central to that model is the notion of stage of adoption.  From both theoretical and practical perspectives, it makes sense to know which segments of women are more likely to respond to the intervention.  It is as important to know whether there is screening adoption, and also which categories of women are more likely to respond.  In addition to examine the moderating effects of stage on the overall intervention in predicting women’s cervical smear screening adoption, this study aimed to (1) assess the impact of direct-mail communications with and without phone intervention, and (2) examine the characteristics of women who were more likely to respond to the intervention.  What were the characteristics of women who responded at the end of the direct-mail communications, which occurred before the phone intervention (early adopters), compared with those who did not (non-early adopters)?  Among all of the women who received a screening at the end of the program, what were the characteristics of early adopters versus late adopters (responded after the phone intervention)?  This study examined the impact of adding a phone intervention to moving additional women to higher levels of motivation / readiness for screening adoption.  By answering these questions, public health educators can better make informed decision on where and how to devote and allocate prevention efforts to achieve cost-effective outcomes.

Methods

Sample and study design

This study used a true control group, pretest-posttest experiment design. Women were randomly assigned to either intervention or control groups.  This study was conducted at one of the major teaching hospitals in Taiwan.  Female family members of inpatients admitted to the hospital during fall 1999 were approached.  A brief intake survey was used to determine the women’s screening history and eligibility. The inclusion criteria for this intervention trial were women who had not had a cervical smear screening in the past 12 months (screening non-adherents) and who were over 30 years and old (or younger than 30 years if married).  Women who did not meet the above criteria or had undergone a hysterectomy or had been diagnosed with cervical cancer were excluded.  A total of 424 eligible women agreed to participated in the study , with 212 women in each group.  The recruitment process has been documented in detail elsewhere [28]. This research was conducted with the approval of Committee for the Protection of Human Subjects (CPHS) at The University of Texas Houston Health Science Center, School of Public Health (HSC-SPH-99-013).

Intervention Description

The three-month intervention was built on existing theories, empirical evidence, and insights gained from a pilot study conducted among the target population.  Several important health belief constructs suggested by existing theories, such as perceived benefits (pros), perceived barriers (cons), perceived risk, and perceived norms, were incorporated in the design of the direct-mail communication messages [19, 29-30].  These factors have also been found to have significant association with cervical smear screening behavior among Chinese women with similar characteristics [7].  Additional insights on salient cultural concerns gained from focus groups in the pilot study were also taken into consideration in the communication messages and evaluation measures [31].

Women in the intervention group received two direct-mail communications, one per month, and a phone intervention at the third month.  In the first month, each woman received a welcome letter addressed to her personally, an educational brochure which addressed several theory-based health beliefs related to cancer screening, other women’s screening experience along with tips to reduce embarrassment or anxiety, and a screening schedule with a health hotline number.  In the second month, each woman received a personalized screening invitation letter, role model stories with personal accounts from women in various stages of screening adoption, testimonials from survivors who found out about their cervical cancer early, a cervical cancer and screening fact sheet, and an updated screening schedule.  In the third month, program staff proactively contacted women in the intervention group regarding their screening adoption status, offered screening counseling, and, if desired, assisted with arranging appointment for those who had not yet taken action.

Women in the control group received three monthly newsletters on general health information.  Women in the control group did not receive a phone call as it might serve as a stimulus for screening adoption.  Therefore, women in the control group only received the set of outcome measures at the end.  Detailed accounts of the intervention development and program content have been described elsewhere [32].

Measurement

Outcome variable

The cervical smear completion rate was measured two times, first by self-reporting during the phone intervention among women in the intervention group, and again at the end of the intervention via a mailed follow-up survey sent to all women in the study.  At least six phone attempts were made to each woman in the intervention group. Program staffs were asked to make these calls at different times of a day (morning, afternoon, or evening) and different days of a week (weekday and/or weekend) in order to maximize reach among participants.  When the line was busy, staffs were advised to call the number a maximum of three times and three minutes apart before that phone attempt was assigned a disposition code [33].  The overall phone contact rate in the program was 63%.

Similarly, several attempts were made to increase the response rate to the follow-up survey.  At the end of three-month intervention, women in both intervention and control groups were mailed a posttest questionnaire with a pre-stamped envelope enclosed.  As a token of appreciation for their participation in the follow-up survey, women were informed that they would receive a small gift (a nail cutter set) after the completed survey was received.  A short notice was sent to women in both groups one week before the follow-up survey [34].  A cover letter personally addressed to each woman was attached to the survey to emphasize the importance of her participation.  Reminder calls were conducted three weeks afterwards to those who had not yet returned the survey.  A second questionnaire was mailed to those who indicated that the questionnaire was lost or misplaced.  Finally, for those who still did not respond, a third round follow-up questionnaire was sent along with the gift incentive to further encourage response.  The overall response rate to the follow-up mailed survey in the program was 58%.

Definition on Stage of Adoption

The stages of cervical smear screening adoption were developed and modified by using criteria from previous mammography studies [22, 35]:

  1. Precontemplation: never had a screening and no intention to have one in the next 12 months.
  2. Contemplation: no screening in the past year (include either women who have never had a screening or had screenings in the past) but has intention to have one in the next 12 months.
  3. Action / Maintenance: has had one cervical smear screening in the past year (i.e. on schedule) and intends to have one in the next 12 months.
  4. Relapse Risk: on schedule, but no intention to get another one in the future.
  5. Relapse: one or more cervical smear screening in the past, none during the past 12 months, and does not plan to get one in the next 12 months.

Since this study recruited only screening non-adherent women (those who reported no cervical smear screening in the past 12 months), the possible stages for women in the study were Precontemplation, Contemplation, or Relapse.  Women in Action / Maintenance, or Relapse Risk stages were not included in the screening intervention trial.

Definition on Early and Late Adopters

Early Adopters

Early adopters in this study referred to women who reported having received a cervical smear screening before program staff conducted the phone intervention. These were women who took action (received a cervical smear screening) before the phone intervention.

Non-Early Adopters

Non-Early adopters in this study referred to women who reported that they have not yetreceived a cervical smear screening when program staff conducted the phone intervention.  Non-early adopters could be either those who did not receive a screening at all at the end of the program, or those who received a screening after the phone intervention (i.e. late adopter, see definition below).

Late Adopters

Late adopters in this study refers to women who reported not having had received a cervical smear screening when program staff conducted the phone intervention, but reported receiving a cervical smear screening at the end of three-month program.  These were women who took action after the phone intervention.

Again, it should be noted that only women in the intervention group were contacted for phone intervention (n=212).  Therefore, women in the control group were not included in the analysis of early versus late adopters.  Women in the control group only received general information on health and without any phone contact to avoid the possibility of screening adoption due to phone contact. Both early and late adopters were only identifiable among those who were reached by the program staff through the phone intervention.  Late adopters were further narrowed down to those who also returned the mailed follow-up survey at the end of the program.

Data Analysis

Multiple logistic regression analysis was used to examine the moderating effects of stage on the overall intervention in predicting women’s cervical smear screening adoption.    Furthermore, a combination of stage and demographic variables (age and education) were used in the multiple logistic regression analysis to predict early versus non-early adopters among women in the intervention group.  Since almost everyone in the study was married, it made sense not to include marital status in the regression model.  Those who received a cervical smear screening by the end of the intervention were classified as either early or late adopters based upon when they received a screening (i.e. before or after the phone intervention).  Due to the limited size of these subgroups, Fisher’s exact test (instead of multiple logistic regression) was used to compare early and late adopters on stage of adoption, demographics and other variables.

Results

Demographics and Stages at Baseline

Women who participated in the intervention trial were mostly young (mean age of 34 years) and married (90%).  Forty percent of the women worked full time and twenty-eight percent of the women did not have a high school education.  Prior screening behavior was similar between these two groups (58%).  No statistical differences were found in demographics between women in the intervention and control groups [28].  Stages of adoption at baseline between women in the two groups were also similar (see Table 1).

Table 1: Stage of adoption among women in the intervention and control groups at baseline survey (N=424)

Notes:
a Precontemplation: Never had a screening and no intention to have one in the next 12 months.
b Contemplation: no cervical smear screening in the past year (could include either women who never had a screening or have had screenings in the past) but has intention to have one in the next 12 months.
c Relapse: one or more cervical smear screening in the past, none during the past 12 months, and does not plan to get one in the next 12 months.
Intervention Control All
Stages of Adoption (N=212) (N=212) (N=424) P-value
N % N % N % [Chi-Square Tests]
Precontemplation (PC) a 35 16.5% 48 22.6% 83 19.6%
Contemplation (C) b 137 64.6% 129 60.8% 266 62.7% .271
Relapse (R) c 40 18.9% 35 16.5% 75 17.7% [X(2)=2.61]
Total 212 100.0% 212 100.0% 424 100.0%

Stage of Adoption and Intervention on Screening Completion

Univariate logistic regression analysis showed that women in the intervention group were more likely to obtain a cervical smear screening than women in the control group (odds ratio [OR] =2.29; p=.002).  Stage of adoption was also a significant predictor of screening completion, with women in the contemplation stage more likely to receive a screening than women in the precontemplation stage (OR=4.41; p=.001).  The full model which included intervention condition, stage, as well as the interaction of intervention and stage were examined.  The treatment by stage interaction was not significant (p>.05), suggesting the effect of stage was not statistically different between the intervention and control group.  The model was then re-run without the interaction term [36].  The multiple logistic regression analysis, which included both stage and intervention as predictors, provided some discrimination on women’s screening adoption (correct classification rate was 66.8%).  Wald statistics indicated that both the intervention (adjusted OR=2.31; p=.003) and women’s stage had statistically significant coefficients.  Similarly, women in the contemplation stage (adjusted OR=4.18; p=.002) were more likely to obtain a screening than pre-contemplators.  No significant differences were found for screening adoption between women in relapse and precontemplation stages (Table 2).

Table 2: Stage of adoption and intervention on screening completion at follow-up (N=247)

Notes: -2 Log Likelihood=303.591; Hosmer-Lemeshow Goodness-of-Fit p>0.5;
Model X2 (3)=31.299, p<.001
a Reference category
Interaction term (Intervention * Stage) was not significant
95% C.I. for Odds Ratio
Wald df p-value Odds Ratio (OR)adj. Lower Upper
Intervention 9.139 1 .003 2.311 1.343 3.979
Stage 18.987 2 .000
Precontemplationa 1.000
Contemplation 9.835 1 .002 4.180 1.710 10.219
Relapse .034 1 .854 1.107 .375 3.267
Constant 15.976 1 .000 .166

Stage and Demographics among Early versus Non-Early Adopters

In order to investigate the characteristics of women who were more likely to receive screening without phone intervention, women in the intervention group were examined (n=212).  Among these women, 63% (n=134) were reached during the phone intervention at the third month of the program.  One in four of these women reported that they had already received a cervical smear screening (early adopters, n=33).  Univariate logistic regression analysis showed that stage was a significant predictor of early adoption, with women in the contemplation stage more likely to receive a screening without phone intervention (OR=5.31; p=.031).  The full model which included age, education, stage, as well as the interaction term was then examined.  Again, no statistical difference was found for the interaction term (p>.05), thus it was dropped from the model.  Results showed that the multiple logistic regression model, which included age, education, and stage as predictors, was statistically reliable in distinguishing between early versus non-early adopters (model classification rate= 75.4%).   Analyses showed that both age and education were similar between early versus non-early adopters (p>.05).  However, compared to pre-contemplators, contemplators were 5.58 times (p=.028) more likely to be early adopters (Table 3).

Table 3: Stage of adoption and demographics among early adoptersa and non-early adoptersb (N=134)

Notes:
-2 Log Likelihood=131.239; Hosmer-Lemeshow Goodness-of-Fit test p>.05;
Model X2 (4) = 10.752, p=.029
a Only women in the intervention group and were reached in phone intervention were included. The phone intervention contact rate was 63.2% (134/212). One in four (33/134) of these women were early adopters, who responded at the end of the direct-mail communications (i.e. before the phone intervention).
b Those women who reported having not yet received a cervical smear screening when program staff conducted the phone intervention were non-early adopters (n=134-33=101). Non-early adopters could be either those who did not receive a screening at all at the end of the program, or those who received a screening after the phone intervention (i.e. later adopter, Table 4).
c Reference category
95% C.I. for Odds Ratio
Wald df P-value Odds Ratio (OR)adj. Lower Upper
Age .075 1 .784 .910 .465 1.784
Education 1.074 1 .300 1.378 .752 2.526
Stage 6.888 2 .032
Precontemplationc 1.000
Contemplation 4.823 1 .028 5.579 1.203 25.867
Relapse .346 1 .556 1.770 .264 11.861
Constant 5.435 1 .020 .053

Early Adopters versus Late Adopters

Among women who received a screening at the end of the intervention (n=55), 60% were early adopters (received screenings prior to the phone intervention) and 40% were late adopters.   Fisher’s exact tests showed that there were no significant differences in age, education, or stage between these late adopters and early adopters.  For example, although younger women (less than 30 years) were more likely to report receiving screening prior to the phone intervention (33.3%) compared with women over 30 years old (18.2%), this difference was not statistically different. Similarly, although women with less education (without high school degree) were more likely to report receiving screening after phone intervention (late adopters) compared to women with a high school education (31.8% versus 18.2%), the analyses did not show statistical differences (Table 4).  Screening knowledge and various screening beliefs (perceived pros, cons, susceptibility, and norms) were also examined among early versus late screening adopters, but the results revealed no significant differences (data not shown).

Table 4: Stage of adoption and demographics among early versus late Adopters (N=55)

Notes:
a Women who were not reached in phone intervention were not included. The phone intervention contact rate was 63.2% (134/212). One in four (33/134) of these women were early adopters, who responded at the end of the direct-mail communications (i.e. before the phone intervention).
b Women who were either not reached in phone intervention or were lost to follow-up were not included. The follow-up mail survey return rate was 58% (123/212) among women in the intervention group. Late adopters were identified among those who reported having not received a screening before the phone intervention (n=101), but reported receiving a screening in the follow-up mailed survey at the end of the program (n=22).
c Fisher’s exact tests were used in these comparisons.
Early Adopters Late Adopters All P-value c
Variables N a % N b % N %
Stage of Adoption
PC or R Stage 5 15.2% 5 22.7% 10 18.2% .498
C Stage 28 84.8% 17 77.3% 45 81.8%
Age
less than 30 years 11 33.3% 4 18.2% 15 27.3% .354
30 years and older 22 66.7% 18 81.8% 40 72.7%
Education
less than high school 6 18.2 7 31.8 13 23.6 0.334
high school & above 27 81.8% 15 68.2% 42 76.4%
Total 33 60.0% 22 40.0% 55 100.0%

Discussion

The findings from this study show that the adjusted effects of stage (ORadj.= 4.18) and the overall intervention (ORadj.= 2.31) were both significant predictors of women’s screening adoption at the end of the intervention.  Among women in the intervention group, contemplators were 5.58 times more likely to be early adopters.  Late adopters consisted of 40% of all screening adopters.  Regardless whether women responded before or after the phone intervention, results did not show significant differences in stage, demographics, screening knowledge or beliefs.

As mentioned, one limitation of the study was that screening adoption among women in the control group was only assessed at the end of the intervention.  It was not possible to assess the effectiveness of the specific intervention components through comparisons.  The current study could not conclude that screening among early adopters in the intervention group was the direct result of the direct-mail communications.  Similarly, screening among late adopters could be attributed to factors other than the phone intervention or the cumulative impact of the overall intervention.  Evidence from several process evaluation indicators examined in the current study provided reasonable support that the reported screening adoption was a result of the corresponding intervention strategies or components.  Overall, almost all women in the intervention group reported (in the follow-up mailed survey) that they received the direct-mail communications (97.6%) and had read the intervention materials (95.9%).  More than 90% of the women found the information “very helpful” or “helpful”; and almost 80% felt the messages were very or somewhat relevant. The exploratory examination of the intervention strategies assumptions showed that early adopters were more likely than non-early adopters to rank the most important intervention component on their screening decisions to be those related to the direct-mail communications, such as the theory-based educational brochure (86% versus 59%; p=.03) and other women’s sharing or role model stories (77% versus 46%; p=.019).  On the other hand, late adopters (received screenings after phone call) were more likely to rank phone intervention as the most important factor on their screening decisions, compared with early adopters (89% versus 53%; p=.029).  Taken together, these process indicators suggest that the program was implemented well (activities executed) and that the assumptions underlying the intervention (participant exposure, method / strategy assumptions) were relevant.  Additional qualitative information collected through the open-ended questions in the follow-up survey also supported these intervention assumptions (data not shown).

Another constraint was that women in the intervention group received information relevant to all states of TTM as opposed to receiving information specifically matched their stage of change.  As a result, we do not know if women in the contemplation stage responded because they received information specifically targeted to their stage of readiness.  Nevertheless, results from the study showed that health communication messages which included stage relevant notions seem to be effective in promoting the adoption of cervical smear screening.  Such implementation strategy could in fact make future adoption and dissemination of the program easier, especially in the case when information on participants’ stages is not available or too costly to obtain.  A previous study suggests that women in earlier stages (precontemplation and contemplation) might need more education about cancer and screening, while women in later stages (action and maintenance) might require more support to make appointments rather than education [37].

The small sample size of screening adopters could possibly contribute to the non-significant findings for stage or demographic factors among early versus late adopters.  Current findings (although not statistically significant), however, showed some indication that the phone intervention encouraged women age 30 and older or a without high school educated to take action.  Larger studies are needed before any firm conclusion can be made.

To the best knowledge of the author, this is the first study to examine the moderating effect of stage on the intervention effectiveness to encourage screening non-adherent Chinese women to receive a cervical smear screening.  It is also the first study to investigate the impact of stage relevant direct-mail communications with and without phone intervention on women’s screening behavior. Stage was a significant predictor of early adoption.  Results also indicated that stage relevant materials seemed to be effective in encouraging screening, and the boost of a latter phone intervention may be consequential in terms of encouraging more women with similar characteristics to take action.  This study provides support for considering stage of adoption in future interventions directed screening behaviors.  It also provides justification for targeting women in the contemplation stage.  More studies are needed to identify effective strategies that encourage women in precontemplation or relapse stages to undergo cancer-related screenings.

Acknowledgments

This study was supported by Cheng-Ching Hospital in Taichung, Taiwan.  The authors gratefully acknowledge the assistance of the Department of Community Health and Department of Nursing in Chen-Ching Hospital. Special thanks give to Dr. Abraham Tesser, research professor at the Institute for Behavioral Research, and Dr. Dave Dejoy, professor of Health Promotion and Behavior at the University of Georgia, for their thoughtful reviews and valuable feedback on this manuscript.

Reference

Muir CS, Nectoux J. International patterns of cancer. In: Schottenfeld D, Fraumeni JF Jr., Editors. Cancer Epidemiology and Prevention (2ndEd.).  New York: Oxford University Press; 1996, p.156.

Centers for Disease Control and Prevention. The National Breast and cervical cancer early detection program; At-A-Glance, 1998. U.S. Department of Health and Human Services, p.1-6.

Hiatt RA, Pasick RJ, Stewart S, et al. Community-based cancer screening fro underserved women: Design and baseline findings from the breast and cervical cancer intervention study. Prev Med 2001;33:190-203.

Chen JY, Diamant AL, Kagawa-Singer M, Pourat N, Wold C.  Disaggregating data on Asian and Pacific Islander women to assess cancer screening. Am J Prev Med 2004; 27(2):139-45.

Wang PD, Lin RS. Socio-demographic factors of Pap smear screening in Taiwan. Public Health 1996;110(2):123-27.

Lee TF, Kuo HS, Chou BS. Factors of Pap screening behavior among women in King-Men, Taiwan. Public Health of Republic of China 1997;16(3):198-209.

Hou S, Fernandez ME, Baumler E, Parcel GS, Chen PH. (2003). Correlates of cervical cancer screening among women in Taiwan. Health Care Women Int 2003;24(5):384-98.

Marks JS, Lee NC, Lawson HW, Henson R, Bobo JK, Kaeser MK.  Implementing recommendations for the early detection of breast and cervical cancer among low-income women. Morb Mort Wkly Rep 2000;49:35-55.

Taylor VM, Hislop TG, Jackson JC, et al.  A randomized controlled trial of interventions to promote cervical cancer screening among Chinese women in North America. J Natl Cancer Inst 2002; 94(9):670-77.

Twinn S.  The evaluation of the effectiveness of health education interventions in clinical practice: a continuing methodological challenge. J Adv Nurs 2001;34(2):230-37.

Twinn S, Shiu AT, Holroyd E. Women’s knowledge about cervical cancer and cervical screening practice: a pilot study of Hong Kong Chinese women. Cancer Nurs 2002;25(5):377-84.

Bird J A, McPhee SJ, Ha NT, Le B, Davis T, Jenkins CN.  Opening pathways to cancer screening for Vietnamese-American women: lay health workers hold a key. Prev Med 1998;27(6):821-29.

Jenkins CN, McPhee SJ, Bird JA, et al. Effect of a media-led education campaign on breast and cervical cancer screening among Vietnamese-American women. Prev Med 1999;28(4): 395-406.

Lam TK, McPhee SJ, Mock J, et al. Encouraging Vietnamese-American women to obtain Pap tests through lay health worker outreach and media education. J Gen Intern Med 2003;18(7): 516-24.

McPhee SJ, Stewart S, Brock KC, Bird JA, Jenkins CN, Pham GQ.  Factors associated with breast and cervical cancer screening practices among Vietnamese American women. Cancer Detect Prev 1997; 21(6):510-521.

Jackson CC, Taylor VM, Chitnarong K, et al.  Development of a cervical cancer control intervention program for Cambodian American women. J Community Health 2000; 25(5):359-75.

Kelly AW, Fores CM, Wollan PC, et al. A program to increase breast and cervical cancer screening for Cambodian women in a midwestern community. Mayo Clin Proc 1996;71(5): 437-44.

Chen AM, Wismer BA, Lew R, et al. “Health is strength”: a research collaboration involving Korean Americans in Alameda County. Am J Prev Med 1997;13(6 Suppl):93-100.

Prochaska, J.O., Norcross, J.C., & DiClemente, C.C.  Changing for Good. (3rd ed.). New York: William Morrow and Company, Inc.; 1994.

Prochaska JO. Strong and weak principles for progression from Pre-contemplation to Action on the basis of twelve problem behaviors. Health Psychol 1994;13:47-51.

Rimer B, Conaway M, Lyna P, et al.  Cancer screening practices among women in a community health center population. Am J Prev Med 1996;12(5):351-57.

Rakowski W, Anderson MR, Stoddard AM, et al. Confirmatory analysis of opinions regarding the pros and cons of mammography.  Health Psychol 1997;16:433-41.

Rakowski W.  Increasing mammography among women aged 40-74 years by use of a stage-matched, tailored intervention. Prev Med 1998;27(5 pt 1):748-56.

Stoddard A, Rimer B, Lane D, et al.  Underusers of mammogram screening: stage of adoption in five US subpopulations. Prev Med 1998;27:478-87.

Pearlman DN, Rakowski W, Ehrich B, Clark MA.  Breast cancer screening practices among black, Hispanic, and white women: reassessing differences. Am J Prev Med 1996;12(5):327-37.

Lipkus I, Rimer BK, Halabi S, et al. Can tailored intervention increase mammography use among HMO women.  Am J Prev Med 2000;18(1):1-10.

Taylor VM, Jackson JC, Yasui Y, et al.  Pap Testing Stages of Adoption among Cambodian Immigrants. Asian Am Pacific Islander J Health 2000;8(1):58-68.

Hou S, Fernandez M, Baumler E, Parcel GS.  Effectiveness of an intervention to increase Pap test screening among Chinese women in Taiwan.  J Community Health 2002; 27(4):277-90.

Ajzen, I., & Fishbein, M. Understanding attitudes and predicting social behavior.  Englewood Cliffs, NJ: Prentice-Hall; 1980.

Rosenstock, I.M., & Krischt, J.P. The Health Belief Model and personal health behavior. Health Education Monographs 1974;2:470-473.

Hou, S. & Luh, W.  Psychometric properties of the Cervical Smear Belief Inventory (CSBI) for Chinese women.  Int J Behav Med 2005; 12(3):180-191.

Hou S, Fernandez ME, Parcel GS. Development of a cervical cancer educational program for Chinese women using intervention mapping. Health Promot Pract 2004;5(1):80-87.

Aday LA.  Designing and conducting health surveys: a comprehensive guide (second edition).  San Francisco: Jossey-Bass; 1996.

Salant P. & Dillman D.A. How to conduct your own survey.  New York: John Wiley & Sons, Inc.; 1994.

Chamot E, Charvet AI, Perneger TV. Predicting stages of adoption of mammography screening in a general population. Eur J Cancer 2001;37:1869-77.

Friedrich, Robert.  In Defense of Multiplicative Terms in Multiple Regression Equations. Am J Political Sci 1982; 26:797-833.

Kelaher M, Gillespie AG, Allotey P, et al.  The Transtheoretical Model and cervical screening: its application among culturally diverse communities in Queensland, Australia. Ethnic Health 1999;4(4):259-76.

License

Share This Book

Feedback/Errata

Comments are closed.