Study design
This study was a simple randomized controlled trial that randomly assigned individual participants to the intervention and control groups and measured outcomes before and after the intervention. The trial was approved by the Research Ethics Committee of the Faculty of Medicine at the University of Tsukuba (No.1216) and registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN000029894). We also obtained official permission from Tsukuba City to conduct the trial at the City’s health centers. This article complies with the Consolidated Standards of Reporting Trials 2010 guideline [18, 19].
Setting
We conducted this study in Tsukuba City, Ibaraki Prefecture, a semi-urban city located 50 km northeast of Tokyo, with a population of 237,000. In Japan, cancer screening is conducted as a health promotion service based on the Health Promotion Act. The local government introduced a population-based cervical cancer screening system in 1983. In 2001, the Research Group for Cancer Screening Guidelines commissioned by the Japanese government recommended the use of cytology-based screening for cervical cancer [5]. In Ibaraki Prefecture, women aged 20–69 years are invited to undergo cervical cancer screening every two years through a letter with a subsidized or free coupon. Those who undergo cervical cancer screening are informed of the results through a letter. In case of a positive result, the cytology result based on the Bethesda System (TBS; indicating the cytological stage) is included.
Participants
The participants were women aged 20–69 years who were about to undergo cervical cancer screening at health centers in Tsukuba City from July 2017 to February 2018. Those who were able to read and write Japanese were eligible to participate in the study. We selected two health centers where a large number of women underwent cervical cancer screening. The participants were recruited by health centers’ public health nurses at the reception of screening venues. They explained the aim of the study in written materials and provided written informed consent if they agreed to participate.
The sample size required for this study was 814 participants (407 each for the intervention and control groups) to ensure an 80% power and 5% significance level, and with an assumption that the prevalence of psychological distress among women who received a positive screening result was 50%, based on our previous findings, and that the intervention could reduce it to 40% [6].
Intervention
The provision of information through a leaflet was the intervention in this study. The leaflet contained basic information about cervical cancer and its screening, the advantages and disadvantages of screening (including information on false-positives, false-negatives, and overdiagnosis), purpose and validity of the screening/further examination, meaning and causes of a positive screening result, explanation of TBS [20], effectiveness of treatment, importance of undergoing further examination, and where to undergo further examination and how to make a reservation for it. We explicitly stated that a positive screening result does not indicate a cancer diagnosis.
To develop the leaflet, we first reviewed current screening result notification-related practices in Ibaraki Prefecture as well as the leaflets regarding cervical cancer screening developed by the National Cancer Center in the United States and the National Health Service in the United Kingdom [15]. Second, we conducted an individual interview and focus group discussion to understand exactly what information should be provided to women who undergo cervical cancer screening. For the interview, we invited 10 women aged 20–69 who visited the hospital to undergo further examination for cervical cancer in order to draft the leaflet. We then conducted two focus group discussions with a total of 12 women to make the necessary revisions of the drafted leaflet. For the focus group discussions, we recruited women aged 20–69 from the general population through advertisements. Finally, we checked the appropriateness of the leaflet with health professionals, including gynecologists.
Procedure and randomization
The participants received self-administered questionnaires that consisted of baseline questions, a hypothetical notification letter of cervical cancer screening, the leaflet (only for the intervention group), and follow-up questions. A hypothetical notification letter indicated a positive or negative result. In the case of a positive result, the letter indicated a hypothetical cytological stage, one of the five TBS-based grades, or no stage. Thus, there were six patterns of positive results, plus a negative result.
In baseline questions, the participants rated their psychological distress. They were then instructed to read the hypothetical notification letter as if they were being informed of the actual screening results. Those in the intervention group read the leaflet after the notification letter. Finally, all the participants rated their psychological distress again in the follow-up questions.
The leaflet was randomly included in the questionnaires that were consecutively handed to the participants as they came to health centers for the screening. The questionnaires were handed only to those who agreed to participate. This was the procedure of random allocation of the participants to the intervention and control groups. Since the intervention (leaflet) was known to the participants but not to the authors or public health nurses who handed the questionnaires, the trial was single-blinded.
The participants were requested to answer and return the questionnaire while waiting to undergo screening. As an ethical consideration, all the participants were provided with the leaflet when they returned the questionnaire. They were also allowed to consult with public health nurses during the screening or later by phone or e-mail, in cases where they became too worried about the screening results because of this study and they wished to consult. The participants received actual screening results from health centers by mail about two weeks after the screening.
Measurement
The primary outcome was cancer-related psychological distress. Psychological distress was measured using the Cancer Worry Scale [21, 22]. The Japanese version of this scale has been demonstrated to be valid in the context of breast cancer-related worry. In this study, we replaced “breast cancer” with “cervical cancer.” The scale consists of six items, five of which are measured on a four-point Likert scale, and one measured on a five-point Likert scale. The total score ranges from 6 to 25, with a score of 15 or higher indicating a significantly high degree of psychological distress. The secondary outcome was the intention to undergo further examination for cervical cancer, which was determined with a yes/no question.
The covariates measured included demographic characteristics, mental health, perceived health competence, and attitudes toward cervical cancer screening. Mental health was assessed using the Japanese version of the K6 scale, which was designed to measure depression and anxiety [23, 24]. The K6 scale consists of six items rated on a five-point Likert scale. The total score ranges from 0 to 24, with higher scores indicating higher degrees of depression and anxiety. Perceived health competence was measured using a Japanese version of the modified Perceived Health Competence Scale [25]. This tool, which measures self-efficacy regarding general health-related behaviors [26], contains eight items measured on a five-point Likert scale. The total score ranges from 8 to 40, with higher scores indicating higher degrees of self-efficacy. Attitudes toward cervical cancer screening were measured using a modified version of the 16-item Attitude toward Breast Cancer Screening Scale [21]. The items were measured on a five-point Likert scale. In this study, we replaced “breast cancer” in the original scale with “cervical cancer.” This scale includes four factors: barriers toward screening, lack of perceived importance of screening, perceptions of screening, and subjective norms related to screening. The score for each factor was calculated. The total score ranges from 4 to 20, with lower scores indicating better attitudes toward cervical cancer screening. The Japanese versions of the three scales mentioned above have been validated [21, 23, 25].
Statistical analysis
First, we compared the participants’ baseline characteristics between the intervention and control groups, and then calculated the proportion of psychological distress (i.e., Cancer Worry Scale score ≥ 15) among the participants before and after the intervention in the intervention and control groups by positive and negative screening results. To take into account psychological distress at baseline, the proportion of psychological distress after the intervention was also calculated specifically among participants without psychological distress before the intervention. Finally, we estimated the effect of the intervention with odds ratios (ORs) and 95% confidence intervals (CIs) by positive (each cytological grade) and negative screening results. All analyses were based on intention-to-treat.