Abstract
Objectives
In 2009, an on-site diagnostic colposcopy clinic was established within a large, urban sexual health clinic to enhance follow-up of abnormal cervical cancer screening among vulnerable women, including those who are uninsured. This service model uses a family physician colposcopist and a colposcopy nurse who provides pre-visit counselling, telephone reminders, patient-tailored ongoing support during the diagnostic process, and tracking of missed appointments. This study examined whether this program was associated with improved adherence to the first colposcopy visit after an abnormal Papanicolau smear among a high needs population.
Methods
We conducted a retrospective chart review of women referred for colposcopy between January 2007 and September 2010, and examined non-adherence before (pre-group) and after (post-group) establishment of the on-site program. Univariable and multivariable logistic regression was used to examine patient and clinical factors associated with non-adherence.
Results
Six hundred eighty-five women were referred during the study period, with 302 in the pre-group and 383 in the post-group. Non-adherence to the first colposcopy visit fell from 13% to 4% after institution of the on-site service. Pre-group status, cervical screening performed at an abortion-related visit (as opposed to a contraception or cervical screening visit), parity ≥ 1 and younger age were all associated with non-adherence in the multivariable analysis.
Conclusion
An on-site colposcopy service that incorporated multifaceted, client-tailored support throughout the diagnostic process significantly reduced non-attendance for an initial colposcopy visit in an urban sexual health clinic population. Broader adoption of this model could improve effectiveness of cervical cancer screening programs. However, future research should determine which specific elements of the model are more important in influencing adherence rates.
Résumé
Objectifs
En 2009, une clinique de colposcopie diagnostique sur place a été mise sur pied au sein d’une importante clinique de santé sexuelle en milieu urbain, en vue d’améliorer le suivi des résultats anormaux de dépistage du cancer du col utérin obtenus par des femmes vulnérables, y compris les femmes n’étant pas couvertes par le régime d’assurance-santé. Ce modèle de service fait appel à un médecin de famille-colposcopiste et à une infirmière spécialisée en colposcopie qui offre des services de counseling préconsultation, des rappels téléphoniques et un soutien continu et adapté à chaque patiente au cours du processus diagnostique, en plus d’effectuer le suivi des rendez-vous manqués. Cette étude s’est penchée sur la question de savoir si ce programme était associé à une amélioration de l’observance du premier rendezvous en colposcopie à la suite de l’obtention d’un résultat anormal de frottis de Papanicolaou, au sein d’une population présentant des besoins élevés.
Méthodes
Nous avons mené une analyse rétrospective des dossiers des femmes orientées en colposcopie entre janvier 2007 et septembre 2010, et examiné la non-observance avant (prégroupe) et après (post-groupe) la mise sur pied du programme sur place. Une régression logistique univariée et multivariée a été utilisée pour examiner les facteurs cliniques et liés aux patientes qui étaient associés à la non-observance.
Résultats
Six cent quatre-vingt-cinq femmes ont été orientées en colposcopie au cours de la période d’étude (302 au sein du pré-groupe et 383 au sein du post-groupe). La non-observance de la première consultation en colposcopie est passée de 13 % à 4 % à la suite de la mise sur pied du service sur place. Le fait de faire partie du pré-groupe, le dépistage cervical mené dans le cadre d’une consultation liée à un avortement (par opposition à une consultation liée à la contraception ou au dépistage cervical), la parité ≥ 1 et le jeune âge ont tous été associés à la nonobservance dans le cadre de l’analyse multivariée.
Conclusion
La mise sur pied d’un service de colposcopie sur place ayant incorporé un soutien diversifié et adapté aux clientes tout au long du processus diagnostique a entraîné une baisse importante de la non-observance de la consultation initiale en colposcopie, au sein d’une population desservie par une clinique de santé sexuelle en milieu urbain. L’adoption à grande échelle de ce modèle pourrait améliorer l’efficacité des programmes de dépistage du cancer du col utérin. Toutefois, d’autres recherches devraient être menées pour identifier les éléments du modèle qui influencent le plus les taux d’observance.
Key Words
INTRODUCTION
Colposcopy follow-up after abnormal Papanicolaou (Pap) screening is an important component of cancer screening programs. However, non-adherence, defined as failure to attend for the recommended colposcopy, is common, with rates varying from 0.4% to 47.3%.
1.
Non-adherence to follow-up could allow progression of premalignant lesions to cancer.2.
, 3.
Furthermore, missed appointments result in inefficient use of scarce health care resources and increase health care costs.4.
Risk factors for non-adherence include younger age; lack of knowledge about cervical screening and colposcopy; lower grade abnormalities; lack of child care, time, or transportation; and high anxiety.1.
, 5.
, 6.
, 7.
There is conflicting evidence about the effects of socioeconomic and insurance status.5.
The impact of on-site versus off-site provision of colposcopy seems to be variable; non-adherence rates in a high-risk Canadian inner city population fell from 17.2% to 1.3% when an on-site service was established,8.
whereas an American study of on-site colposcopy in family practice did not find a difference.9.
To improve adherence, researchers and clinical practice guidelines advise that colposcopy programs address barriers to care, such as cost, an unfamiliar site, long appointment wait times, and lack of support for women’s needs.
1.
, 8.
, 10.
, 11.
, 12.
They also cite the importance of providing effective, culturally competent counselling and information-sharing to reduce anxiety among women who have abnormal cervical cytology results, and highlight that such communication should be standard patient care.12.
, 13.
To address some of the barriers to adherence, we established an on-site diagnostic colposcopy service within a sexual health clinic in downtown Toronto, Ontario. This clinic has 20 000 annual patient visits for contraception, abortion, Pap testing, treatment of sexually transmitted infections, and other sexual and reproductive health concerns. Approximately 1800 Pap smears are performed annually. Although hospital-affiliated, the clinic is situated outside the hospital in a nearby office building. It serves a diverse population of mainly younger women, including many who have high needs, such as those without health insurance, newcomers, street-involved women, and women of low socioeconomic status. Uninsured women receive care without charge.
The program model is as follows: the colposcopy service uses a family physician colposcopist and a dedicated program nurse. The nurse makes initial telephone contact with each woman who requires colposcopy and provides information about her Pap smear abnormality and about colposcopy. During this call, the nurse also schedules a colposcopy appointment, and ascertains whether individual concerns, such as high anxiety, language barriers, or difficulties with child care require specific supports. Further counselling and support are then tailored to the individual woman’s needs. For example, a woman may bring a support person to accompany her in the examination room, translators are available, and extra appointment time is allotted for highly anxious women. Patients receive a telephone reminder in the week before the scheduled appointment, and those who miss their appointment are tracked to reschedule. The colposcopy nurse is available to provide ongoing support throughout the whole diagnostic process. The on-site service collaborates with the affiliated gynaecologist-led hospital-based service, if treatment with loop excision or laser is required. Once treatment has been provided, patients are referred back to the on-site service for ongoing surveillance (Figure).
The objective of this study was to determine whether this model of care improved adherence to the initial colposcopy visit after an abnormal Pap smear and to examine factors related to non-adherence.
METHODS
Patients were referred to the on-site service starting on January 1, 2009. We conducted a retrospective chart review of women attending the sexual health clinic who were referred for colposcopy between January 1, 2007, and September 30, 2010. This allowed accrual for 24 months in the pre-group (January 1, 2007, to December 31, 2008) and 21 months in the post-group (January 1, 2009, to September 30, 2010). Before implementation of this program, women needing colposcopy were referred off-site; almost always they were referred to a nearby hospital or to office-based services, unless they lived far away or already had a colposcopist. After the clinic was established, almost all women were referred to the on-site clinic, unless this was inconvenient or they already had a colposcopist. Excluded from the study were women in the pre-group who made their own referrals through their family physician or gynaecologist, and those in the post-group who were referred off-site.
Bay Centre for Birth Control Care Model
We collected data on age, parity, gravidity, country of birth, health insurance status, purpose of the clinic visit when Pap screening was performed (for abortion, contraception, sexually transmitted infection testing/treatment, or Pap testing based on previous recommendation or [self] referral), referral date, colposcopy findings, referral cytology, and history of previous Pap smear abnormality (by previous cytology report or clinic note).
Patients were classified as adherent to the recommended initial colposcopy appointment if there was a consultation note or chart notation indicating a colposcopy visit within six months and one day of the referral date. If the referral was to the hospital-affiliated clinic, and there was no consultation note in the chart, the hospital clinical encounter record was searched for attendance during the indicated time frame. Adherence was classified as “unknown” if these methods could not yield a definitive answer. All other patients were classified as non-adherent.
There is no standard definition of non-adherence for colposcopy care. Many women who initially fail to attend appointments eventually do so.
14.
Considering that some delay in attendance is reasonable, some studies have suggested defining non-adherence as failure to attend an appointment within six months of the scheduled appointment, rationalizing that “a longer delay interval would have had implications for disease progression.”14.
, 15.
For this reason, we chose a time frame of six months and one day, after a review of the literature1.
, 14.
, 15.
and discussion with expert colleagues.Univariable analysis of patient and clinical factors associated with non-adherence was performed using an unadjusted logistic regression model. Tests were 2-sided and a P < 0.05 was considered significant. This was followed by multivariable logistic regression, adjusted for pre- and post-group status and factors found to be significant at a level of P = 0.2 in the preliminary analysis. When variables were found to be collinear, the variable that made the most clinical sense was included in the model.
This study was approved by the Women’s College Hospital Research Ethics Board.
RESULTS
In the 45-month period between 2007 and 2010, 685 women were referred for colposcopy, with 302 in the pre-group and 383 in the post-group. Overall, group characteristics were similar: women were young, with a mean age of 26 years; approximately one third were foreign-born; and approximately 10% were uninsured. Approximately 16% of referrals were for high-grade cytological abnormalities. The only significant differences between the two groups were the type of visit associated with Pap screening and parity, with more nulliparous women in the post-group (Table 1).
Table 1Demographic and clinical characteristics of patients referred for colposcopy before (n = 302) and after (n = 383) introduction of the on-site model of care
| Variable | Pre-group n (%) | Post-group n (%) | P |
|---|---|---|---|
| Mean age in years ± SD | 26.3 ± 5.0 | 26.4 ± 5.1 | 0.81 |
| Has provincial health insurance | 263 (87) | 328 (86) | 0.58 |
| Foreign-born | 105 (35) | 136 (36) | 0.78 |
| Nulliparous | 257 (85) | 333 (87) | 0.49 |
| Nulligravid | 152 (50) | 228 (60) | 0.016 |
| Pap screening associated with | < 0.001 | ||
| Contraception | 110 (36) | 166 (44) | |
| Pap screening visit | 120 (40) | 122 (32) | |
| Abortion | 55 (18) | 40 (10) | |
| STI screening / treatment | 17 (6) | 53 (14) | |
| History of previous Pap smear abnormality | 209 (69) | 293 (76) | 0.11 |
| Severity of cytological abnormality | 0.72 | ||
| Low grade | 229 (76) | 300 (78) | |
| High grade | 53 (18) | 59 (15) | |
| Other | 20 (7) | 24 (6) |
AGC: atypical glandular cells; ASC-H: atypical squamous cells cannot exclude high-grade squamous intraepithelial lesion; ASCUS: atypical squamous cells of undetermined significance; HPV: human papillomavirus HSIL: high-grade squamous intraepithelial lesion; LSIL: low-grade squamous intraepithelial lesion; STI: sexually transmitted infection
* Normal, ASCUS-reactive, ASCUS-HPV, and LSIL
† ASC-H, HSIL, and AGC
On univariable analysis, non-adherence to initial colposcopy was 13% in the pre-group, compared with 4% in the post-group (P < 0.001). Parity and gravidity, lack of insurance, and Pap screening performed during an abortion visit were also significantly associated with non-adherence (Table 2). Neither age nor severity of the cytological abnormality was associated with adherence, although age was within the predetermined significance level for the multivariable analysis (P = 0.18). In the adjusted multivariable regression analysis (Table 2), pre- or post-group status, Pap screening associated with an abortion, parity, and age were significant. Women in the post-group were less than one third as likely as those in the pre-group to be non-adherent (OR 0.30; 95% CI 0.16 to 0.57, P < 0.001), and women screened at an abortion visit were almost three times more likely to be non-adherent than women referred after a Pap screening-specific visit (OR 2.9; 95% CI 1.39 to 6.05, P = 0.002). Younger women and those who had had one or more deliveries were also more likely to be non-adherent.
Table 2Logistic regression of factors associated with non-adherence to first colposcopy visit (n = 685)
| Factors | Univariable | Multivariable | ||
|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | |
| Group (post-group vs. pre-group) | 0.30 (0.17 to 0.54) | < 0.001 | 0.30 (0.16 to 0.57) | < 0.001 |
| Age | 0.96 (0.91 to 1.02) | 0.18 | 0.93 (0.87 to 0.99) | 0.023 |
| Insurance type (No OHIP vs. OHIP) | 2.27 (1.19 to 4.35) | 0.01 | 1.97 (0.96 to 4.04) | 0.06 |
| Pap screening associated with: (Reference: Pap screening visit) | ||||
| Abortion | 4.10 (2.06 to 8.15) | < 0.001 | 2.90 (1.39 to 6.05) | 0.002 |
| Contraception | 0.73 (0.35 to 1.51) | 0.09 | 0.90 (0.43 to 1.91) | 0.27 |
| STI screening/treatment | 0.59 (0.17 to 2.08) | 0.14 | 0.86 (0.24 to 3.12) | 0.45 |
| Parity (0 vs. ≥ 1) | 0.27 (0.15 to 0.49) | < 0.001 | 0.34 (0.16 to 0.71) | 0.004 |
| Gravidity (0 vs. ≥ 1) | 0.29 (0.16 to 0.52) | < 0.001 | Not included in the multivariable analysis | |
| Foreign-born (yes vs. no) | 1.30 (0.75 to 2.28) | 0.35 | ||
| History of previous Pap smear abnormality (yes vs. no) | 0.81 (0.44 to 1.52) | 0.57 | ||
| Severity of cytological abnormality (high risk vs. low risk) | 0.72 (0.34 to 1.50) | 0.37 | ||
OHIP: Ontario Health Insurance Plan; STI: sexually transmitted infection
Adherence status could not be determined in 4% of women in the pre-group and 0% in the post-group. A sensitivity analysis showed no change in the overall results when these women were classified as adherent or non-adherent.
DISCUSSION
Many strategies have been employed to reduce non-adherence to colposcopy, with varying degrees of success.
1.
, 5.
, 11.
Our multi-faceted, on-site model of care that incorporates pre-procedure nurse telephone counselling tailored to patient needs (appointment reminders; tracking of patients who missed appointments; and individualized supports, such as allowing accompaniment in the procedure room, translation, and extra accommodation for highly anxious patients) reduced non-adherence by two thirds. Our non-adherence rate of 4% at six months is similar to the 3% rate of default for a first colposcopy assessment found in one systematic review,1.
and well below the 15% often used as the maximum acceptable standard.12.
While 4% is low, it suggests that there are a few women, who, despite intensive efforts, will be very difficult to reach and may need different approaches to ensure necessary follow-up.Our model addresses a number of barriers to adherence. Anxiety and lack of knowledge about colposcopy are common among women, and both factors have been shown to be contributors to non-adherence.
10.
, 16.
Our pre-procedure telephone counselling sought to increase knowledge and allay anxiety by providing information about the procedure, answering questions and addressing concerns, and assessing patients’ need for further information or psychological and emotional support throughout the diagnostic process. Women could contact the colposcopy nurse for further information or support if they chose, and highly anxious women often had multiple telephone encounters with the nurse. Simply forgetting about the colposcopy appointment has also been shown to be a cause for non-attendance.15.
, 17.
Our telephone appointment reminder addressed this. One study of a simple pre-visit telephone appointment reminder found that women who received the reminder were five times more likely to attend their initial colposcopy appointment, with 9% failing to attend.18.
The impact of having on-site colposcopy services has not been well-described. Theoretically, familiar clinic surroundings could reduce anxiety, but further research is needed to assess this effect.In addition to pre- or post-group status, predictors of non-adherence included younger age, parity, the type of visit at which the Pap smear was performed, and insurance status. Unlike most studies, we did not find an association between the grade of severity of the referral Pap smear and adherence. However, as in other studies, younger women and parous women were more likely to be non-adherent.
19.
, 20.
, 21.
, 22.
Child care responsibilities, which have been identified as barriers to keeping appointments, could explain poorer adherence in parous women.15.
Our service permitted women to bring their children with them to their appointment if they were unable to make alternative arrangements; however, this did not eliminate the effect of parity on adherence.Our finding of poorer adherence among women having Pap screening associated with an abortion visit is not surprising. There is some evidence that women experiencing stressful events may be less likely to prioritize follow-up for Pap smear abnormalities.
23.
In our clinic, cervical screening is offered opportunistically to women who are attending for abortion services. Compared with women who come specifically for a Pap test or have one as a routine part of a contraception visit, women who are screened at an abortion visit may not see Pap screening as a priority and may also have less knowledge of its purpose and the implications of an abnormality. Opportunistic screening is often advocated for women in hard-to-reach populations,24.
but our results suggest that women who undergo Pap screening opportunistically may require additional support to ensure follow-up if an abnormality is detected.Lastly, non-adherence was more common in uninsured women, but this was only significant in the univariable analysis. Since patient costs were eliminated for the uninsured in the post-group, this finding was expected. However, the trend for uninsured women to be almost twice as likely to be non-adherent remained in the multivariable analysis. The number of uninsured women in this study was small (39 and 55 in the pre- and post-groups, respectively), and may have been inadequate to detect a statistically significant difference. It is possible that uninsured women may experience additional barriers unrelated to cost.
This study has a number of strengths. Many studies that have examined non-adherence used the rate of “no-show” for appointments as their outcome measure.
1.
This is likely to have little significance for actual patient outcomes, as many women eventually do keep their appointments. Instead, we have used patients as our denominator and a six-month time interval from referral to try and capture delays that are more likely to represent loss to follow-up, or delayed follow-up that could have health implications for individual women. As well, although changes in practice or populations over time are a concern in pre- and post-intervention studies, there were no changes in screening guidelines or referral algorithms during our study period, and therefore our pre-group and post-group populations were likely very similar.There are several limitations that should be acknowledged. Findings in this sexual health clinic population may not be broadly generalizable. Lower needs populations may not require as much intensive support, although the underlying concept of providing knowledge and support tailored to women’s needs is likely to be useful in most populations and can be adapted to local needs. A second limitation relates to wait times. Shorter wait times have been associated with better adherence rates.
25.
, 26.
We could not reliably ascertain wait times to the first colposcopy booking in the pre-group, and it is possible that the improved post-group adherence rate was in part due to shorter wait times associated with the establishment of this new service. Likewise, other patient factors such as socioeconomic status, need for translation services, or distance travelled for colposcopy could also be influential and were not measured. However, given the similarity in the pre- and post-groups, we think it is unlikely that these factors affected our primary outcome. Since most women in the pre-group were referred to the colposcopy services located close to the sexual health clinic, travel distance would have been similar for both groups.A potential limitation of the program model relates to its resource-intensiveness. Some of the supports in this model, such as pre-visit counselling and patient tracking after missed appointments, are time-consuming and therefore may not be practical in some lower resourced practice settings. In our situation, we had institutional support for a half-time nurse dedicated to the one half-day per week colposcopy clinic. However, there are likely some efficiencies resulting from this initial support. For example, women receiving pre-visit education and support do not require extensive counselling at the time of colposcopy, thus increasing the clinic’s efficiency and capacity. Other components of the model, such as telephone appointment reminders (which have been shown to reduce missed appointments
18.
) do not require significant resources and are already standard practice in many clinical settings. Finally, although this model of care was associated with much improved adherence, we do not know with certainty whether it was the whole model or specific components of it that affected adherence.CONCLUSION
An on-site colposcopy service using a multi-faceted model of support for patients significantly reduced non-attendance for an initial colposcopy visit following an abnormal Pap smear in this urban sexual health clinic population. Although it is resource intensive, broader adoption of this model of care, or some of its components, has the potential to improve the effectiveness of cervical screening programs by reducing non-attendance rates. Although this model works in its current setting, future research should determine whether it is applicable in the broader population, and which specific elements of the model are most important in influencing adherence rates.
ACKNOWLEDGEMENTS
This project was supported through Echo: Improving Women’s Health in Ontario, and the Innovation Fund from the Alternate Funding Plan of the Academic Health Sciences Centres of Ontario. We also wish to acknowledge the support of the Department of Family and Community Medicine, University of Toronto, Clinician Investigator Award (Sheila Dunn), and the contributions of Dr Michael Shier in the development of the colposcopy service, and Melini Gupta and Susan Hum in the production of this manuscript.
REFERENCES
- Is default from colposcopy a problem, and if so what can we do? A systematic review of the literature.Br J Gen Pract. 1999; 49: 223-229
- Problems of follow-up for abnormal cervical smears: discussion paper.J R Soc Med. 1990; 83: 94-95
- Cervical cancer in women with comprehensive health care access: attributable factors in the screening process.J Natl Cancer Inst. 2005; 97: 675
- Can default rates in colposcopy really be reduced?.BJOG. 2008; 115: 403-408
- Understanding barriers for adherence to follow-up care for abnormal Pap tests.J Womens Health. 2007; 16: 311-330
- Association of knowledge, anxiety, and fear with adherence to follow up for colposcopy.J Low Genit Tract Dis. 2002; 6: 17-22
- Barriers to follow-up of an abnormal Pap smear in Latina women referred for colposcopy.J Gen Intern Med. 2010; 25: 1198-1204
- Access to colposcopy services for high-risk Canadian women: can we do better?.Can J Public Health. 2004; 95: 346-351
- On-site colposcopy services in a family practice residency clinic: impact on physician test-ordering behavior, patient compliance, and practice revenue generation.J Am Board Fam Pract. 1997; 10: 259-264
- Telephone counseling improves adherence to colposcopy among lower-income minority women.J Clin Oncol. 1992; 10: 330-333
- Adherence to care plan in women with abnormal Papanicolaou smears: a review of barriers and interventions.J Am Board Fam Pract. 2001; 14: 123-130
- Colposcopy and Programme Management: Guidelines for the NHS Cervical Screening Programme Second Edition NHSCSP Publication No. 20.NHS Cancer Screening Programmes, Sheffield, UK2010
- Roy M; SOGC HPV Consensus Guidelines Committee. Canadian Consensus Guidelines on Human Papillomavirus. Society of Obstetricians and Gynecologists of Canada Clinical Practice Guideline no. 196, August 2007.J Obstet Gynaecol Can. 2007; 29: S55-S56
- An investigation into the default rate at the Fife colposcopy clinic: implications for target setting.J Public Health Med. 1995; 17: 65-69
- Enhancing adherence following abnormal pap smears among low-income minority women: a preventive telephone counseling strategy.J Nat Cancer Institute. 1997; 89: 730-738
- Measurement of colposcopy-associated distress using the psychosocial effects of having an abnormal Pap smear-questionnaire in a Latina Population.J Low Genit Tract Dis. 2004; 8: 25-32
- No-show behavior in a county hospital colposcopy clinic.Am J Gynecol Health. 1990; 4: 15-21
- Preclinic appointment telephone contact: an effective intervention for colposcopy clinic nonattendance.J Low Genit Tract Dis. 2007; 11: 35-38
- Improving compliance among women with abnormal Papanicolaou smears.Obstet Gynecol. 1995; 86: 353
- Abnormal Pap smear follow-up in a high-risk population.Cancer Epidemiol Biomarkers Prev. 2001; 10: 1015
- Inadequate follow-up for abnormal pap smears in an urban population.J Nat Med Assoc. 2003; 95: 825-832
- The effectiveness of a community outreach intervention to improve follow-up among underserved women at highest risk for cervical cancer.Prev Med. 2005; 41: 741
- What predicts adherence to follow-up recommendations for abnormal Pap tests among older women?.Gynecol Oncol. 2007; 105: 74-80
- Cervical cancer: screening hard-to-reach groups.CMAJ. 1997; 157: 543-545
- Referrals to hospital by general practitioners: a study of compliance and communication.BMJ. 1988; 297: 1246-1248
- An audit of colposcopy appointment processes in women with abnormal cervical cytology.Cytopathology. 2011; 22: 184-188
Article info
Publication history
Accepted:
May 15,
2013
Received:
January 8,
2013
Footnotes
Competing Interests: None declared.
Identification
Copyright
© 2013 Society of Obstetricians and Gynaecologists of Canada. Published by Elsevier Inc. All rights reserved.
30863-X&id=fx1.jpg){kind=link}