Breast cancer worry in further examination of mammography screening – a systematic review

Abstract

Aims:  In this systematic review, we explored which factors are associated with breast cancer (BC) worry in further examination (FE) of mammography screening and for how long women experience BC worry associated with FE.

Methods:  Medline, Cochrane, Cinahl and the International Network of Agencies for Health Technology Assessment databases were used in the search. In addition to this, reference lists of the selected articles were inspected. The subject headings and keywords searched included breast screening, mammography screening, BC worry, distress or anxiety, additional test, further examination and recall. These were used separately and in combination. The databases gave 598 references. From the selected titles, 83 abstracts were read; from these, 23 publications were selected for full-text review. On the basis of full texts, 15 articles were chosen.

Findings:  There were very few studies about the associations between FE and BC worry. In those found, the amount of measured variables was limited. BC worry in FE of breast screening was associated with several sociodemographic and psychosocial factors of recalled women, as well as those related to the FE process. About half of the recalled women were quite or very anxious in the prediagnostic phase.

Conclusions:  The review gave implications for interventions in the FE process of mammography screening. These interventions should predominantly be made in the preclinical phase and should differ according to the type of recall visit and characteristics of recalled women. In communicating the diagnosis to the women, the time from the examination process to diagnosis should be as short as possible, and in addition to physicians being present, other health care professionals should also be there to ensure the necessary emotional and practical support a woman needs in such a situation.

Introduction

Breast cancer (BC) aetiology is complex. It is associated with several genetic and environmental factors such as sociodemographic factors, family and personal history (1, 2). Lifestyle factors such as exercise, weight reduction and stress reduction as well as nutritional factors like low-fat diet, sufficient folate intake and reduced alcohol intake have been associated with the decreased risk of BC (3, 4). Also chemoprevention has decreased the onset of some BC types (4). Because of this complexity, we need multiple means to prevent BC or its regeneration. One of the most efficient ways is early detection.

Mammography screening has nowadays established its position as a natural part of women’s lives, latest after the age of 50, especially in Europe and also in Western countries in general. The European Community has recommended screening every 2–3 years for women aged 50–69 (5). It reduces BC mortality about 15%. The benefits of mammographic screening measured by lives saved are greater in absolute terms than the harm caused in terms of overdiagnosis. Between 2 and 2.5 lives are saved for every overdiagnosed case (6, 7). However, there is no clear consensus about its economic and psychological benefits (8).

The debate about psychological consequences and effects of breast screening on the quality of life is still continuing (9, 10). In considering screening tests, it must be remembered that they carry a chance of misclassification of disease and false-positive test results that lead to further interventions. The process following recalls affects women emotionally and decreases their well-being and quality of life for weeks and even months (10). For most women who have immediate negative screening test, breast screening does not create anxiety (9, 11, 12). However, for those recalled, it causes extra distress (11, 13, 14). Studies have reported anxiety, distress and intrusive thoughts especially among women with an abnormal mammogram. Also, impaired working capacity, reduced self-esteem and self-confidence as well as negative effects on women’s social relationships have been reported (15–17). Previous breast screening experiences, fear of pain, worry about BC and other psychological consequences of screening including concerns related to further examinations have had positive (15, 18) and negative (19–21) effects on the uptake of breast self-examination and on screening attendance. These factors may also affect women’s quality of life especially for those who have been recalled (22). Although general distress might be short-lived, worries last longer than a few months (6, 23). It has also been found that social well-being is associated with recurrence of BC as well as mortality. Because of this, social support is especially important for those women with the previous history of BC in recall situation (24).

So far, we are aware of the psychological consequences of breast screening (16, 17) and false-positive findings (23, 25). Major part of the worry associated with breast screening consists of worry about the result of the mammogram (26, 27). However, in most studies, BC worry has not been separated from other types of worries, for example distress and worry about pain caused by mammography examination associated with initial breast screening and further examinations. Both BC-specific and general distress measures have been used. The studies have been mainly retrospective or longitudinal without prescreening measures (28, 29). However, also some prospective studies have been carried out (23, 30).

In evaluating the studies about BC worry associated with mammography screening, it should be taken into account whether the examination is offered for all eligible women or whether the screening is conducted outside organized programmes. The setting of the screening programme as well as invitation system may have an impact on the psychological consequences of the examination process. For example, in the United States and Australia, women take part in mammography screening on the basis of recommendation of health care professionals or encouragement by media campaigns (31–33). This kind of ‘opportunistic screening’ favours women who are spontaneous or have certain specific reasons to participate in the programme. Those programmes attract especially women with BC worries and those with the family history of BC. In Scandinavian population–based screening programmes, all the women included in the programmes are called for screening by personal invitation letter. Best predictors of attendance in these programmes have been previous attendance, sociodemographic factors and disposition to use other health services (34, 35).

The time from an abnormal screening mammogram to definitive diagnosis is called preclinical phase (29). This preclinical phase can be divided into (i) the time interval from announcement of an abnormal finding to the time of further examination and (ii) the time interval from further examination to the definitive diagnosis. The time from an announcement of an abnormal finding to further examination is very stressful (9, 13), and some interventions such as educational interventions, discussion before the examination and information leaflets have been tested to increase screening uptake and reduce adverse effects during this time (36–38). However, the results of these studies have been inconclusive.

Objective of the study

The purpose of this review is to determine:

• •
   Which are the factors associated with BC worry in further examination of mammography screening?

• •
   For how long do women experience BC worry associated with further examination?

Evidence review strategy

Because the results of breast screening cause anxiety (9, 13) and the results of the effectiveness of interventions developed to prevent it (36–38) have been inconclusive, it was necessary to collect and evaluate the research evidence about the topic by means of systematic review. Systematic review is an effective and valid method of evaluating and summarizing research evidence (39). The following electronic databases were used: Journals@Ovid, Ovid MEDLINE(R), Ovid MEDLINE(R) Daily Update, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, Cochrane Methodology Register (Methods Studies) and The International Network of Agencies for Health Technology Assessment database and Cinahl. In addition to this, reference lists of the selected articles were inspected. The subject headings and keywords searched included breast screening, mammography screening, BC worry, distress or anxiety, additional test, further examination and recall. These were used separately and in combination.

Inclusion criteria of the studies and Health Technology Assessments (HTA) were that they were (i) written in English; (ii) published during the years 2000–2010; (iii) prospective or longitudinal studies or controlled trials; (iv) studies in which setting was population-based or large local mammography screening programmes; they included (v) prescreening or pre-further examination measures. The last inclusion criterion was (vi) good quality quantitative or qualitative studies. The quality was assessed according to the modified version of strobe v.4 checklist (40).

Exclusion criteria of the studies were (i) written in a language other than English; (ii) performed in other than population-based or larger local mammography screening context; (iii) no mentioning of BC worry, distress or anxiety; and (iv) studies that concentrated on women who had previous BC.

When formulating systematic review questions, PICO model is often used. It helps to define relevant population groups, interventions, comparators and outcomes of interest. It also helps to define inclusion and exclusion criteria, which are essential for identifying relevant literature (39).

The PICO

• Population: Women called for further examination of breast screening.

• Intervention: Interventions associated with reducing BC worry prior to further examination.

• Control and comparison: Women called for further examination compared to women with normal findings.

• Outcomes of Interest: Factors associated with BC worry in further examination and duration of BC worry.

The year 2000 was chosen as year limit because the idea was to have the latest knowledge about women’s experience about factors associated with BC worry, so that the knowledge produced in this review will serve as a basis for developing future interventions to reduce the worry.

The first author of the article performed searches on the databases between 3 and 14 June 2010. Then the researchers looked independently through all the titles and selected relevant titles for abstract search. Also, abstracts were inspected independently. Among the abstracts, relevant articles for full-text review were chosen after a consensus discussion on the 18 June 2010. In this discussion, exclusion criteria were amended so that it was decided to exclude studies about genetic testing, those not specific to mammography screening and those concentrating mainly on the effectiveness of breast screening. Articles included in previous systematic reviews were also included in cases where the main focus of systematic review did not reveal essential points from the viewpoint of this systematic review, and the last phase references of the selected papers were carefully checked up.

A qualitative synthesis was carried out because the settings of the studies as well as age groups of samples of women in the studies varied too much for a proper quantitative or qualitative meta-analysis. Narrative summary was drawn upon the data extraction, with an emphasis on textual summary of study characteristics as well as data relevant to the specified outcomes (39). Results chapters of the selected studies were inspected by looking for similarities and differences in the population groups, settings and results of the selected studies. Because of small number of relevant studies found, and because of their heterogeneity, proper qualitative analysis and forming categories of the results were not relevant. For these reasons, the summary of the results was made by describing the findings and characteristics of the studies selected as data for this review.

Results

Cochrane Central Register of Controlled Trials, Systematic reviews, Method studies and HTA gave 328 hits, Cinahl 55 and Medline 215, totalling 598 titles. These included duplicates in different databases, which were reviewed separately. From the selected titles, 83 abstracts were read; from these, 23 publications were selected for full-text review. On the basis of full texts, 15 articles were chosen. Two of them were found from Cochrane Central Register of Controlled Trials, six from Cinahl, five from Ovid database and two from the reference lists of the articles found from the databases. Articles were evaluated by their quality according to a modified version of the strobe v.4 checklist (40). Evaluation of the selected articles is presented in Table 1.

Table 1. Critical assessment of the studies

	Assessment criteria of the studies
	1	2	3	4	5	6	7	8	9
1: Study background and theoretical framework are clearly defined. 2: Purpose, aim and research questions are clearly defined. 3: The design is clearly stated. 4: The setting is clearly described. 5: Independent, dependent, confounders. 6: Data sources and analysis methods are clearly described. 7: Describes any efforts to address potential sources of bias 8: Answers the research questions logically. 9: Discusses the study’s limitations and generalizability. **: Satisfies assessment criteria. *: Partly satisfies assessment criteria. _: Hardly or not at all satisfies assessment criteria. x: Assessment criteria do not apply.
Aro et al. (23)	**	**	**	**	**	**	**	**	**
Brett and Austoker (42)	**	**	**	**	**	**	*	**	*
Ekberg et al. (43)	**	**	**	**	**	**	**	**	**
Lampic et al. (41)	**	**	**	**	**	**	**	**	*
Meystre-Agustoni et al. (45)	**	**	**	**	*	**	**	**	**
Decker and Harrison (49)	**	**	**	**	**	**	**	**	**
Sandin et al. (48)	**	**	**	**	**	**	*	**	*
Absetz et al. (44)	**	**	**	**	**	**	**	**	**
Barton et al. (46)	*	**	**	**	**	**	**	**	**
Heckman et al. (50)	**	*	*	*	**	**	*	**	**
Brett et al. (12)	**	**	x	x	x	**	**	**	**
Watson et al. (52)	*	**	x	x	x	*	*	**	_
Pineault (47)	**	**	**	**	**	**	*	**	**
Tyndel et al. (51)	*	**	**	**	**	**	*	**	**
Brain et al. (30)	**	**	**	*	**	**	**	**	*

Description of the included studies

Eight of the selected 15 studies fulfilled at least eight of the nine quality criteria totally. There were two systematic reviews to which all the assessment criteria did not apply. Most deficiencies in the quality of the studies were in clear assessment of the potential sources of bias of the studies (Table 1).

Twelve of the studies were European, two were from Canada and one from the USA. Europe was represented in the study selection from the north to the southern Europe. Seven of the studies were about population-based screening programmes in which all eligible women received a personal invitation to the programme (23, 41–47). One study was performed in a local screening programme with a personal invitation letter (48). Two studies were in the setting of opportunistic screening where women were recommended screening by health care professionals, and they needed doctor’s referral for it (49, 50). In two studies among women with family history of BC, health care professionals offered screening programme (27, 51) (Table 2).

Table 2. Characteristics of reviewed studies

References	Country	Objective of the study	(i) Study design and (ii) screening setting	Population/sample and control/comparison group	Measures of breast cancer (BC) worry and anxiety	Duration of worry and factors associated with BC worry in FE
BSE, breast self examination; BS, breast screening; FE, further examination; STAI, State Trait Anxiety Inventory; PCQ, Psychological Consequences Inventory; HADS, Hospital Anxiety and Depression Scale; IES, Impact of Events Scale; BSI, Brief Symptom Inventory; BCAI, Breast Cancer Anxiety Inventory; CWS-R, Revised Cancer Worries Scale.
Aro et al. (23)	Finland	To examine psychological distress in a mammography screening process as a consequence of screening after adjusting for background, personality and prescreening distress	(i) Longitudinal study: measurements were 1 month before screening invitation with postscreening after 2 and 12 months. (ii) Population-based screening with the invitation letter including examination time to all eligible women	There were three groups: normal findings (n = 1407), false-positive findings (n = 492) and referents from outside the screening programme (n = 1718) totalling 3617. Women were aged 48–49 years	STAI, single items to measure BC worry	In the univariate analyses, the false-positive group reported more intrusive thinking and worry about BC, increased frequency of BSE, heightened perceived risk and more symptoms both 2 and 12 months postscreening than women with normal findings or referents. This also applied to the multivariate analysis although multivariate effect of distress was not statistically significant in 12 months postscreening
Brett and Austoker (42)	UK	To study the effect of the previous experience of breast screening on adverse psychological consequences reported by false-positive women	(i) Cohort study: data collections 1, 5 and 11 months after BS1 and 3 years after that just before BS2. (ii) Population-based screening with the invitation letter including examination date to all eligible women	Data collections 1, 5 and 11 years after BS1 (n = 661). Fourth data collection just before BS2 (n = 505). Response rate 77%	PCQ	Women called for FE experienced adverse psychological consequences including BC worry 1.28–2.07 as often as those with negative screening results 1 month postscreening. Three years after the first screen, for 1 month before their next breast screening, recalled women still experienced adverse effects. Associated with general BS anxiety including BC anxiety were false-positive result in a previous BS 3 years ago, experiences of previous BS and FE (level of information and communication, service and fear of radiation)
Ekberg et al. (43)	Norway	To measure: (i) The level of psychological distress (anxiety and depression) among women on the day of recall mammography in an official screening programme and 4 weeks after the diagnosis was given. (ii) The level of psychological distress according to age. (iii) The level of psychological distress according to the oncological diagnosis. (iv) The satisfaction with the programme and willingness to recommend other women to undergo screening	Longitudinal study: (i) Psychological distress was measured among women on the day of recall after mammography in an official screening programme and 4 weeks after the diagnosis was given. (ii) Population-based screening with the invitation letter including examination date	HADS and a questionnaire to assess reactions to the examination were completed by 213/220 (97%) women aged 50–69 years. The patients were classified into the following groups according to the medical diagnoses: (i) Healthy after mammography, (ii) Healthy after triple diagnosis (iii) Healthy after biopsy (iv) Hospitalized with cancer after triple diagnosis (v) Hospitalized with cancer after biopsy	Norwegian version of HADS	Recall after mammography is associated with transiently increased levels of anxiety and depression in women without cancer. For those with cancer, the psychological distress continues. The age of women was not associated with their anxiety level
Lambic et al. (41)	Sweden	To investigate the psychological consequences of further investigation after BC screening	(i) Longitudinal study including control and comparison groups. Data collections to recalled group (comparison group = G1) were prior to visit, a few days, 3 and 12 months after the FE. Data collection to normal finding screening group (control group = G2) 3 and 12 months postscreening. (ii) Population-based screening with the invitation letter including examination date	Study participants included 509 women recalled because of suspicious findings on screening mammograms, and a matched control group of 285 women with normal mammograms. All the recalled answered the questionnaires prior to visit, 85% of them at the data collection few days and 3 months after FE and 82% of them 12 months after FE. Moreover, all of the normal screening group completed their questionnaires at the first data collection and 82% of them at the 12-month follow-up	HADS that includes seven items measuring anxiety and seven items measuring depression	46% of the women reported borderline or clinically significant anxiety prior to the recall visit. A few days after the visit, anxiety and depression had decreased significantly in women informed about normal or benign results at the recall clinic, while reported distress remained at relatively high levels in women referred to surgical biopsy. The type of FE and information received at the recall visit were associated with the anxiety perceived during the FE process
Meystre-Agustoni et al. (45)	Switzerland	To follow up anxiety in a cohort of women screened for BC	(i) Longitudinal study: data collection just before the screening, after the examination before the result or FE, for those with negative screening result 2 weeks after results, for both groups 8 weeks after DG. (ii) Population-based screening with an invitation letter	Swiss women aged 50–70 years (n = 924, final response rate 93.7%) participating to a pilot screening programme	PCQ, a single item to measure BC anxiety	Anxiety measured with BCA was highly correlated with PCQ. Among those screening negative, anxiety at screening was very low and remained so during the screening process. Among those screening false positive, anxiety was significantly higher 8 weeks after having received a negative diagnosis. Just before initial screening: low education and younger age. After the FE, before the result or FE: ever had a PAP smear, ever had a breast disease and the anxiety level at the initial screening. Two weeks after the results (those with negative answer after initial screening): low education and anxiety level at screening. Eight weeks after results: low education, anxiety level at screening and false-positive result at FE
Decker and Harrison (49)	Canada	To evaluate a plain-language guideline sent to women with abnormal screening results who attended the Manitoba Breast Screening Programme (MBSP)	(i) Interventional study: a plain-language guideline was mailed with a result letter to 258 randomly chosen women who had abnormal mammograms and/or abnormal clinical breast examinations. Four weeks later, a satisfaction questionnaire was mailed to these women (cases) as well as to 254 randomly chosen women with abnormal results who were not sent a guideline (controls). All cases were interviewed by telephone 3 weeks after the questionnaire was mailed. (ii) Women need doctor’s referral for screening. Also in the case of an abnormal finding, the family physician is expected to coordinate further assessment of abnormality	Comparison group (G1) 258 women, control group (G2) n = 254 women. A total of 345 patient satisfaction questionnaires (67%) were returned, and 47% of the cases completed the telephone interview. The women were 50 and older	Breast Screening Satisfaction Questionnaire. Telephone interview about the MBSP guideline readability, usefulness and effect on women’s anxiety level including two 5-point Lickert scale statements about the anxiety provoking effect of the guideline	A plain-language guideline was useful for most women who had abnormal screening results, although it did not alter the women’s satisfaction with the MBSP. The guideline did increase anxiety for 26% of the women
Sandin et al. (48)	Spain	To examine anxiety, depressive mood, depression and somatic symptoms associated with a second-stage (FE) screening for BC	(i) Quasi experimental. Interviews were conducted pre- and postmammogram. (ii) Navarra breast screening programme where all eligible women got an invitation	Participants were total sample of 1195 Spanish 45–65 years old women attending for routine BC screening who were recalled for further mammogram, and women who were not recalled	STAI	Women attending the second-stage screening (FE) exhibited significantly higher levels of anticipatory anxiety and depressive mood before the mammogram than women attending for routine screening. This emotional impact was not relevant 2 days after the mammogram
Absetz et al. (44)	Finland	To examine whether the psychological impact of organized mammography screening is influenced by women’s preexisting experience with BC and perceived susceptibility (PS) to the disease	(i) Comparative study: Data were collected with postal questionnaires 1-month before screening invitation and 2 and 12 months after screening. (ii) Population-based screening with the invitation letter including examination time to all eligible women	From a target population of 16 886, a random sample of women with a normal screening finding (n = 1407) and women who were recalled for FE (n = 535). Response rate was 63% at baseline; 86, and 80% of the baseline participants responded to the follow-ups	STAI, single items to measure BC worry	Women with BC experience had higher risk perception already before screening invitation; after screening they were also more distressed. Women with high perceived susceptibility (PS) were more distressed than women with low PS also at preinvitation. The distress was not alleviated by screening, but instead remained even after normal mammograms. Experience and PS did not influence responses to different screening findings. Of the finding groups, false positives experienced most adverse effects: their risk perception increased and they reported most postscreening BC-specific concerns even 12 months after screening. In false-positive group there was an increase in BSE frequency
Barton et al. (46)	UK	To compare the effects of both an immediate reading of mammograms and of an educational intervention that taught skills to cope with anxiety for women whose mammograms were normal or abnormal	(i) Interventional study using control and comparison groups: radiology intervention only (immediate read-ing of mammograms), educational intervention only (teaching coping skills), both interventions or neither intervention. Data collection 3 weeks and 3 months after mammogram. (ii) Population based screening with the invitation letter including examination date	n = 8543 women aged 39 years and older. Response rates at 3 weeks interview was 84% and for 3 months 72%	IES	Immediate reading of screening mammograms, but not an educational intervention targeting coping skills, was associated with less anxiety among women with false-positive mammograms 3 weeks after mammography
Heckman et al. (50)	USA	To characterize women’s levels of emotional distress associated with mammography screening and relationships between women’s coping and psychosocial adjustment	(i) Longitudinal study: data collection 1 day after receiving mammogram screening, after notification of screening result that necessitated additional testing and after being informed of their BC free status. (ii) Screening with doctor’s referral	Total sample of 98 women receiving annual BS. All in the sample received questionable screening result that needed FE, finally proved to have false-positive finding. Average age 52.5 years (range 30–80)	STAI	Women reported a significant increase in anxiety following notification of the need to return for FE, the reduction in anxiety level was associated with coping strategies, especially with avoidance coping
Brett et al. (12)	UK	To assess the extent of and factors associated with the adverse psychological consequences of mammographic screening	Systematic review	Medline, Cinhal, EMBASE and Psycinfo. Search limited to 1982–2003. Hand searching to NHS Breast Cancer Searching literature Updates 1996–2003	In the studies included: PCQ, STAI and HADS	Women who have FEs experience significant anxiety in the short term and possibly in the long term. The nature and extent of the FE that women are exposed to during mammographic screening determines the intensity of the psychological impact. BC worry was not separated from other psychological consequences
Watson et al. (52)	UK	Assess psychological impact of mammographic screening on women with family history of BC	A systematic review	Medline, Cinhal, EMBASE and Psycinfo. Search limited to 1982–2003. Hand searching to NHS Breast Cancer Searching literature Updates 1996–2003	In the studies included: IES, BSI	Family history of BC was not associated with high levels of anxiety in initial BS or FE
Pineault (47)	Canada	To describe anxiety experienced by participants in a BC screening programme who have received an abnormal screening mammography result and are waiting for further testing and diagnosis and to identify the social support needed during this period	(i) Exploratory, descriptive. Data collection by mailed questionnaire in prediagnostic phase 3 weeks after learning their abnormal screening mammogram result. (ii) Population-based screening. omen were called with invitation letter with the necessary phone numbers	Nonprobability sample of 631/951 asymptomatic women (response rate 66%), aged 50–69, who had abnormal screening mammogram results in the 2 months prior to the survey and who spoke or read French or English	PCQ; BCAI	51% of the participants were quite or very anxious at every stage of prediagnostic phase. 75% expressed their feelings to family and friends whose support was comforting but did not diminish participants’ anxiety. Instead support offered by health care professionals reduced anxiety. Younger participants and those having an abnormal mammogram for the first time were more anxious than others. Also those women to whom the abnormal screening mammogram was communicated by physician instead of other health care professional and those whose waiting period was prolonged were more anxious than the others
Tyndel et al. (51)	UK	Investigate positive and negative psychological effects of BS	(i) Longitudinal study: data collection 1 month before, 1 and 6 months after BS. (ii) Younger than normal screening aged women with moderate or high BC risk were offered BS and counselling by health professional	Women under 50 with a family history of BC (n = 1286/4206). The group comprised of women who received all-clear result (n = 1174) and of those recalled for additional test before all-clear result (n = 112). Response rate at T1 was 62%, at T2 it was 80% and at T3 it was 73%	CWS-R, PCQ	Women invited to FE experienced more cancer worry than those with negative findings, but the worry was reduced in 6 months time. Associated with cancer worry were previous mammogram and high perceived risk for BC
Brain et al. (30)	UK	Investigate prescreening factors associated with BC specific distress	Longitudinal study: (i) data collection 1 month before, 1 and 6 months after BS. (ii) Younger than normal screening aged women with moderate or high BC risk were offered BS and counselling by health professional	Women under 50 with a family history of BC (n = 1286/4206). Final response rate was 31%	CWS-R	Having been recalled for FE in previous screening was associated with BC specific distress in the following screening rounds

Most of the studies used general anxiety measures, which included statements to measure BC worry such as State Trait Anxiety Inventory (STAI), Psychological Consequences Inventory (PCQ), Hospital Anxiety and Depression Scale (HADS) and Brief Symptom Inventory (BSI). Also, BC–specific anxiety measures were used. Those were Impact of Events Scale (IES), Breast Cancer Anxiety Inventory (BCAI) and Revised Cancer Worries Scale (CWS-R). In nine studies, worry associated with BC was measured for the first time latest at the day of the recall. In a study of Brett and Austoker (42), anxiety at the second breast screening round was compared with the anxiety at the first round among women having false-positive finding. Two studies were interventional: one with postintervention follow-up in which they used control and comparison groups (46, 49). One study was descriptive with data collection 3 weeks after the women had received their abnormal breast screening result (47). The sample sizes varied from 98 (50) to 3617 (23), and response rates from 31% (30) to 97% (43). The studies of Sandin et al. (48) and Heckman et al. (50) used total samples. Age distribution of the women participating in these studies was between 30 and 80 (Table 2).

Only in five of the studies, factors associated with psychological consequences of further examination, including BC worry, were mentioned as the central outcome of the study (41, 43, 46–48). Seven papers were aimed at studying psychological consequences of breast screening in general including those called for further examinations (12, 23, 30, 44, 45, 50, 51). One study concentrated on the impact of previous experiences of breast screening on adverse psychological consequences among false-positive women (42); one evaluated a guideline sent to women with abnormal screening results (49); and one systematic review assessed psychological impact of mammographic screening on women with family history of BC (52). However, all the studies reported findings on either factors associated with BC worry in further examination or duration of BC worry associated with further examination (Table 2).

Factors associated with BC worry in further examination

The anxiety level before the breast screening was associated with the anxiety level after hearing about abnormal finding and need for further examination (12, 44, 45), and with previous screening experiences (30, 42, 51). These are so-called prescreening factors. Also, coping strategies (49), low education, ever having a Pap smear (45) and having an abnormal screening result for the first time (47) were associated with BC worry in further examination. Results of the associations between age and BC worry varied. According to Pineault (47), younger age was positively correlated with higher BC worry. In a study of Ekberg et al. (43), no association between age and BC worry was found. Family history of BC was not associated with higher levels of anxiety in initial breast screening or further examination. In postdiagnosis measures, those with false-positive findings reported higher anxiety levels than other groups (Table 2).

The type of further examination and information received at the recall visit were associated with the anxiety perceived during the whole preclinical phase (41). Women who were advised to have more intensive follow-up after having an abnormal mammogram had higher levels of distress than did women with less intensive follow-up recommendations (46). Support offered by health care professionals, hearing about the abnormal screening result from other health care professionals than from physician, as well as short waiting time between the recall and further examination were associated with lower anxiety levels of women (46, 47). However, using plain language guideline, where further examination process was described, educational intervention or teaching coping skills had no significant impact on the level of BC anxiety (46, 49). Women whose further examination was performed at the primary screening site experienced less distress than those whose further examination was performed at the laboratory or outpatient clinic (43).

Duration of the BC worry associated with further examination

The selected studies reported that women got anxious when they got an invitation to further examination (12, 43, 47, 50, 52). They experienced adverse psychological consequences including BC worry 1.28–2.07 times as often as those with negative screening results 1 month postscreening (42). Of those recalled, 51% were quite or very anxious at every prediagnostic phase of the screening process (47). About half of them reported borderline or clinically significant anxiety prior to the recall visit (41, 46, 47). Studies stated that those with negative findings were relieved already a few days after the further examination (41, 48) or after a few weeks (50). Some studies reported high levels of anxiety for false-positive findings for several months up to 2–1 year (23, 44–46, 51). Especially high levels of anxiety were reported by women referred to surgical biopsy (41, 42, 47), and for those women with cancer diagnosis, depression and anxiety were present 3 weeks after diagnosis (43).

Discussion

Factors associated with BC worry in further examination

First of all, it must be stated that there were very few studies about the associations between further examination and BC worry and practically no studies with the main focus on factors associated with BC worry in preclinical phase. In the five studies we found, there were a limited number of independent variables. The topic should be studied with more comprehensive number of variables using prospective design and multivariate methods. There were only few interventional studies about the topic between 2000 and 2010 (46, 49). Results of these (46, 49) were as inconclusive as the results of earlier interventional studies (46, 47). Only shortening the time between informing about the abnormal finding to further examination and announcement of the final results seemed to be efficient in reducing BC worry (46, 47). There were implications that communication at the recall visit was associated with the anxiety perceived during the whole preclinical phase (41).

This article concentrates on women who do not know yet whether they have BC or not. They are those who are in between the diagnostic tests and diagnosis in the screening process. The studies found in this systematic review give implications for developing more effective interventions reducing BC worry in further examination of breast screening. The screening organization as well as primary and specialized care organizations which women visit should take into account the type of recall visit the recalled women come for and the type of information they are offered during the visit (41, 42). In the type of information offered, women with low level of education (45), those with first abnormal finding (47), those needing intensive follow-up and those with negative previous screening experience (30, 42, 51) should be treated as a special target group (46).

The interventions for recalled women should be targeted especially to the preclinical phase (47). The recalled women should be offered enough and right type of support by the health care professionals (46). According to the studies of Lampic et al. (41) and Pineault (47), the information about the abnormal screening result and examination procedure should be given rather by other health care professionals than physicians. Studies about the associations of lifestyle factors show that they may have some association with BC risk (3, 4, 53, 54). Some information about lifestyle and nutritional factors associated with BC should be available for women at the preclinical phase by nurses and nutritional experts to promote general well-being. Especially information that you may affect your own health, such as reducing alcohol consumption, reducing carbohydrate intake if necessary, supporting fatty acids intake and taking care of sufficient folate intake, is an important factor in strengthening the physical health (3). In the preclinical phase, the screening staff should be aware of different types of coping strategies women use, and they should support their coping (49) although, in a study of Barton et al. (46), intervention targeted to coping skills was not very effective in reducing distress. However, the perception of the controllability of health fosters breast screening compliance (34).

According to Greco et al. (55), social support is something that best help women to cope with breast screening recall situation. This should be taken seriously in breast screening clinics and hospitals where further examinations take place, because social well-being seems to have impact even on the BC treatment outcome (24). According to Pineault (47), most anxiety reducing was social support given by health care professionals. In prediagnostic phase, women who have no other health problems and who are referred to less invasive examinations such as extra mammograms or fine needle biopsies meet usually only breast screening staff including radiographer, radiologist and assisting personnel. For those women, right kind social support given by this group is the most important and in most cases, sufficient. However, women who are referred to primary of specialized care hospital for core biopsy or surgical biopsy meet also other health care staff such as nurses and surgeon of the operating hospital. According to Lampic et al. (41), BC worry is associated with the invasiveness of the further examination. Women with invasive further examination should receive social support given by multiprofessional health care team including breast screening staff, for example radiographer and/or radiologist and nurse taking part clinical care at the operating department. Depending on the clients situation, social worker and next of kin of the patient could also be included. This team should meet the patient before the invasive further examination and after she has had her diagnosis to ensure her optimal well-being and follow-up and treatment if needed. With diagnosed BC patients, there are encouraging results on the benefits of case management (56).

Duration of the BC worry associated with further examination

Findings on the duration of BC worry are mixed; However, it is clear that when women get an invitation to further examination, most of them get anxious (12, 30, 44, 45, 51). There are studies that state that the increased level of BC worry might have even positive effects on the uptake of breast self-examination and screening attendance (15, 18). However, the studies are inconclusive. According to some other studies, this worry associated with breast screening has negative impact on screening and breast self-examination uptake (19–22).

Most important variables associated with the duration of BC worry seem to be (i) time from the recall to further examination and to diagnosis (46, 47); (ii) the type of further examination: only mammogram, fine needle aspiration, clinical examination or surgical biopsy (41, 42, 47); and (iii) final finding (23, 43). According to this review, screening setting or women’s family history had no association with the duration of BC worry or screening uptake. BC worry is a negative side effect of breast screening that decreases the quality of life of women taking part in breast screening (10), and it implies that interventions to reduce unnecessary worry should be developed.

Potential bias and limitations of this systematic review

The search was limited to years 2000–2010. The limit of 10 years was chosen because mammography screening settings and protocols as well as women’s views about health issues change during decades. The search was limited to English-language studies, which could favour a specific perspective on the matter under study. The ‘grey literature’ such as master theses was not searched because we only wanted to include the highest quality studies. However, the strategy used might have excluded some studies with a relevant focus or topic.

The first and the second author made the literature searches independently having gone through literature search education during their education. The first author has previous experience on systematic reviews. This has positive and negative impact on the search. On the one hand, the authors were able to focus and orient the search during the search process, which made the review process flexible. On the other hand, professional librarians might have come up with some up-to-date methods to refine the search process.

Only in five of the selected studies, the main focus of the study was BC worry associated with further examination. In addition to this, the amount of independent factors tested was very limited, and measures of anxiety varied from solitary statements to general anxiety scales to BC worry specific scales. This may set some limitations to conclusions made on the basis of the review summary. However, according to some studies (26, 27), the major part of the worry associated with breast screening consists of worry about the result of the mammogram.

The study settings varied from the stabilized population-based screenings to opportunistic screening programmes. Also the geographic spread of the selected studies was quite wide. Despite this, the results on the factors associated with BC worry did not vary according to screening setting. These issues speak for the unique nature of process that may be more general to various types of screening settings and cultural contexts than previously hypothesized.

Conclusions

This review gave implications for interventions in the further examination process of mammography screening. These interventions should predominantly be targeted to the preclinical phase and should differ according to the type of recall visit and characteristics of recalled women. In communicating the diagnosis to the recalled women, the time from the examination process to diagnosis should be as short as possible, and in addition to physicians, other health care professionals should also be present in the situation to ensure the emotional and practical support the women need in the situation.

Until we completely know the aetiology of BC, early detection and screening as a part of it seems to be the best way. In the medicine and caring sciences, our challenge is to research the aetiology of BC and to develop new ways of intervening the factors associated with the onset of BC. We also need to develop the best possible screening system and multiprofessional patient-centred service chain to avoid unnecessary worry and suffering and to promote compliance and screening uptake.

Author contributions

All the authors were responsible for review strategy of the systematic review. Authors Metsälä and Pajukari made the literature search and reviewed the articles. All the authors took part in drafting of manuscript, Eija Metsälä (former E. Grönroos) being mainly responsible for it. Professor Arja R. Aro had the primary responsibility for the critical revisions and supervision of the article.