Pap smear screening at an urban Aboriginal health service: report of a practice audit and an evaluation of recruitment strategies

Transcription

1 Pap smear screening at an urban Aboriginal health service: report of a practice audit and an evaluation of recruitment strategies Abstract A culturally appropriate women s health service was established at an Aboriginal community-controlled health service in Darwin in An initial file audit found Jennifer M. Hunt and Gwenda Lawton Gless Danila Dilba Medical Service, Darwin, Northern Territory Judith A.Y. Straton Department of Public Health, University of Western Australia that 48% of included women had,ever been screened with a Pap smear and 37% of women were considered to have been adequately screened. The enhancement of opportunistic screening by file tagging had a modest effect on screening coverage over a 12-month period for women who attended the health service. The proportion of these women who were adequately screened increased from 43% to 48% and of those ever screened increased from 54% to 62%. A randomised trial of recruitment interventions including personal approach, letter and control groups was subsequently performed for women for whom Pap smears were overdue or not recorded. The impact of both interventions on the number of Pap smears performed was low, with 7% of women in the personal approach group, 2% of women in the letter group and no women in the control group having Pap smears during the three-month follow-up period. Low rates of abnormalities were observed for women having Pap smears over a twoyear period. The minimal effect of a formal reminder system and letters at this urban Aboriginal health service has resulted in a re-orientation of activities towards strengthening opportunistic screening and the continued promotion of Pap smears in a range of clinic and community settings. It is important to place Pap smear screening in the context of other social, economic and health priorities for Aboriginal women and health workers. (Aust NZ J Public Health 1998; 22: 720-5) Correspondence to: Dr Jenny Hunt, PO Box 1265, Darlinghurst, NSW jennymh@ bigpond.com A boriginal women have a mortality from cervical cancer of 6-10 times the rate for non-aboriginal women, and are more frequently diagnosed with invasive cervical cancer. However, little information is available about Pap smear screening coverage and recruitment strategies for Aboriginal ~ omen.~ ~ A few reports describe successful Pap smear recruitment programs resulting in high levels of screening coverage for groups of Aboriginal women living in remote community setting~.~.~ An early report of Pap smear screening found only 27% of Aboriginal women attending avictorian urban Aboriginal health service had ever had a Pap smear.6 Danila Dilba is an Aboriginal community-controlled medical service which provides primary health care services to the Aboriginal and Torres Strait Islander people of the Darwin urban area. Staff includes Aboriginal Health Workers, doctors and health educators. It is managed by an elected committee ofaboriginal community members. In late 1994, Danila Dilba was funded to implement the Gumileybirra Women s Health Project. The project s aim was to improve the health ofaboriginal and Torres Strait Islander women of the Darwin urban area by providing a range of accessible and culturally appropriate women s health services. As part of the project, a file audit of Pap smear screening coverage for all women attending Danila Dilba was performed. Subsequently, activities attempting to improve Pap smear coverage were undertaken. This report presents an evaluation of these activities. Methods The women s clinic In November 1994, a women s clinic was established at Danila Dilba, staffed by Aboriginal Health Workers and a female doctor, all working part time. After consultation with Aboriginal staff and clients of Danila Dilba, features were adopted to ensure the clinic was culturally appropriate and accessible. These included a screening wall to separate the women s clinic from the main clinic area, a policy of women only in the reception area and consulting room, an area for child care, and comfortable and welcoming furnishings including a mural by an Aboriginal artist telling a women s health promotion story. The clinic operated regularly one day per week, initially without an appointment system.an informal appointment system was adopted in February 1996 in an attempt to reduce long waiting times. A transport service was provided for women if needed. Women and their children were seen at the clinic for a range of women s and general health problems, as well as for antenatal care and Pap smear screening. For each consultation, records of whether the client was a woman or a child and the main reason for attendance at the clinic were collected. These statistics were collated using Access software. Pap smear screening was also available to women attending Danila Dilba s general clinic, and was usually conducted in the women s clinic consulting room. Aboriginal Health Workers spent time in the community talking to groups of women about women s health issues including Pap smear screening, and advertising the women s clinic. They selected materials felt to be culturally appropriate including a flip chart 720 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 1998 VOL. 22 NO. 6

2 Pap smear screening at an urban Aboriginal health service with pictorial representations of the Pap smear procedure.8 On average, Aboriginal Health Workers spent between half and one day per week in the community during the first year of the project, and 1-2 days per week in the second year. Pap smear screening was also promoted at Danila Dilba by having several in-service sessions for all female Aboriginal Health Workers. Complete file audit Files of all women aged 18 to 70 years who had ever attended Danila Dilba were identified in late Between April and June 1995 these files were reviewed. A woman was included if she had a Darwin address, had been seen at Danila Dilba more than twice in the past three years and had no record of hysterectomy. These criteria were used to try and define a more meaningful denominator for estimating screening coverage in a mobile population of women who use multiple providers for Pap smears and other services. For each included woman, dates and results of Pap smears were collected from files, and by telephoning other practitioners, laboratories and community clinics. Pap smear and general information gathered for entry onto an Access database. Analysis of the Pap smear database was performed in January 1997 using Access. Screening coverage was determined as at 1 July 1995 using all Pap smear information available at the time of analysis. Women were considered to have been adequately screened if they had a history of a Pap smear within three years if all Pap smears had been negative, or if they were not more than two months overdue with recommended management according to National Health and Medical Research Council (NHMRC) guidelines following Pap smear abnormalities. File tagging Files of all included women were tagged with red stickers indicating the date their next Pap smear was due. If no Pap smear screening history was available, stickers requesting information were used. Between April and June 1996, files of all included women were reviewed. Pap smear screening information on the database and file tags were updated. Whether or not women had attended Danila Dilba since the initial file audit was noted. For women who had attended when due for a Pap smear and not been screened, file notes were examined for possible reasons the test had not been performed. The impact of file tagging was assessed in January 1997 by comparing Pap smear screening coverage at 1 July 1996 with that at 1 July 1995, for women who had been seen in the clinic between the file audit and review. Women who had undergone hysterectomy during this period were excluded from the 1996 population. Pap smear recruitment intervention trial A woman was selected for the active Pap smear recruitment intervention trial if, at the time of the 1996 file review, she was not known to have moved out of the Darwin area, &d she was overdue for screening. A woman was classed as overdue if there was no Pap smear information recorded, if she had not had a Pap smear for at least three years after a normal Pap smear, or if she was more than one year overdue for screening following an abnormal Pap smear according to NHMRC guideline^.^ Women were randomly allocated to one of three groups by matching a list of the women s file numbers to a list of computer-generated random numbers designating the group number (I, 2 or 3).1 A list of women in the personal approach group was given to Aboriginal Health Workers to follow up, either in the community or in the clinic when women attended for other reasons. Women in the letter group were sent a simple reminder letter designed by Aboriginal Health Workers, indicating they were due for a Pap smear and giving details of the women s clinic. Letters were sent to the most recent address recorded in the woman s file. At the time of obtaining the address, it was noted whether an address change was recorded in the file during the previous three years. At Danila Dilba, an attempt is made to update addresses at every client visit. The control group received no intervention apart from the pre-existing file tags. The Aboriginal Health Workers did not know which women were in letter or control groups. At the end of a three month follow-up period, files were examined to determine attendance at the clinic and whether Pap smears had been done. The person reviewing the files (JH) was not aware of the women s group allocation, and was not involved in sending letters or contacting women in the personal approach group. Records were kept of whether letters had been returned to sender, and whether and how the Aboriginal Health Workers had contacted women. Informal discussions were held with Aboriginal Health Workers after the study to determine their experience of contacting women. The impact of the Pap smear recruitment interventions was assessed by determining the proportions of women in each group who were potentially or actually contacted, who attended Danila Dilba, and for whom Pap smears were discussed and done. Women in the letter group were considered to have been potentially contacted if the letter was not returned to sender. Pap smear results The prevalence of Pap smear abnormalities was determined by collating the results of Pap smears performed between 1 July 1994 and 30 June 1996 for all included women. For women who had more than one Pap smear during this time, only the first result was counted. Ethical approvals The Gumileybirra Women s Health Project was approved by Dada Dilba s management committee and by the Aboriginal Sub- Committee of the Joint Institutional Ethics Committee in the Northem Territory. The management committee gave specific approvals to undertake the randomised trial and to publish this report. Results Women s clinic Between November 1994 and September 1996, 853 consultations for women and 140 consultations for children were conducted at 90 women s clinic days. The main reasons for attendance for women were antenatal care (36% of consultations with women), gynaecological problems (30%). Well Women s checks (16%), and management of other medical problems (not specifically women s 1998 VOL. 22 NO. 6 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 72 1

3 Hunt et al. Table 1 : Pap smear screening status 1 July 1995 by age for included women. Number of Adequately Inadequately No record of any Age (years) women screeneda screened Pap smears No. %b NO. Yo NO. Yo ~~ Total Notes: (a) Adequately screened is defined as Pap smear within three years if a// Pap smears negative, or not more than two months overdue with the next management step according to NHMRC guidelines if history of abnormalities. (b) Percentages are rowpercentages for each age group. health, 14%). The remainder of consultations were for post-natal check-ups (3%) or had unrecorded reasons for attendance (2%). File audit There were 2,770 files of women aged between 18 and 70 years identified from Danila Dilba's file register. Following the file audit, 2,033 women were excluded. While records were not kept of specific reasons for exclusion, the vast majority were because women had not been seen at Danila Dilba for more than three years, or had attended twice or less. The age distributions for women included and excluded were similar to each other, and to that of all Aboriginal women in the Darwin urban area according to census estimates.'' Initially, 737 women were included on the Pap smear database. An additional four women were excluded at the time of analysis because they had died. Pap smear screening status at 1 July 1995 for included women is shown in Table 1. Overall, 47.8% of women had records of a Pap smear at some time, and 36.6% had records of being adequately screened. Women who were older were less likely to have records of having been ever screened with Pap smears and were less likely to be adequately screened. Pap smear abnormalities The prevalence of abnormalities for women who had Pap smears Table 2: Prevalence of Pap smear abnormalities for all women screened 1 July 1994 to 1 July 1996.a Pap smear result Number of women % Negative Inflammation Atypia less than CIN H PV CI N Unsatisfactory Total Note: (a) For women with more than one Pap smear during this period, only the result of the first taken is included. between 1 July 1994 and 1 July 1996 is shown intable 2. Only two women (1.1%) had reports of CIN of any grade, with 127 (70.9%) having completely normal Pap smears. Opportunistic Pap smear screening with file tags During the period between the file audit and the file review, 355 women (48.4% of included women) attended Danila Dilba at least once. Pap smear screening status at 1 July 1995 and 1 July 1996 for these women is compared in Table 3. There was a small, statistically non-significant increase in the proportion of women who were adequate11 screened with Pap smears during this period. For the majority of the 223 women who were due for a Pap smear when they attended Danila Dilba but did not have one, there were no reasons recorded in their file suggesting why Pap smears had not been performed (seetable 4). However, for 23% of women who had attended and not been screened, there was documentation that Pap smears had been discussed or offered. Pap smear recruitment intervention trial In the Pap smear recruitment intervention trial, 372 women were randomly allocated to personal approach, letter and control groups. Six women were subsequently found to have had a Pap smear before any interventions had been implemented and were excluded at the time of analysis. Aboriginal Health Workers contacted only 22 of the 1 19 women (18.5%) in the personal approach group. In discussions after the trial, they indicated that other work commitments and difficulty in locating women had prevented them contacting more women on the list. They had successfully contacted women by asking other women and Aboriginal Health Workers about their whereabouts, by approaching women when seen in the clinics or in the community and, on one occasion, by making a home visit. The Aboriginal Health Workers reported they did not feel comfortable approaching women not known to them outside a clinic setting specifically to discuss Pap smears. They felt.it was usually inappropriate to visit women at home for the sole purpose of telling them a routine Pap smear was due. Many Aboriginal women in the Darwin area can not be contacted by telephone. 722 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 1998 VOL. 22 NO. 6

4 Pap smear screening at an urban Aboriginal health service Table 3: Impact of opportunistic screening between 1 July 1995 and 1 July 1996 for women who attended the clinic. Pap smear screening status 1 July 1995 Number Yo 95YoCI 1 July 1996 Number Yo 95?'ocI Adequately screened Inadequately screened No record of any Pap smears Total Note: (a) Totals differ because four women had hysterectomies during the period between the two reviews of the files. For 37 of the 125 women in the letter group (29.6%) letters were undeliverable or returned to sender. For an additional 32 (25.6%) at least one address change was noted on the woman's file in the previous three years. There were no marked differences in attendance at Danila Dilba between the personal approach, letter and control groups. (Table 5) Of women who attended during the follow-up period, 46% in the personal approach group and 48% in the letter group had documentation of Pap smears having been discussed, compared to only 22% of women in the control group. The number of Pap smears taken for women in the trial was small. Only eight of the 119 women in the personal approach group (6.7%) had Pap smears performed during the follow-up period, compared to threg of the 125 women in the letter group (2.4%) and none of the 122 women in the control group. Discussion Promoting and performing Pap smears was one of a number of Gumileybirra Women's Health Project activities. Aboriginal women have expressed a preference for a holistic, generalist approach to women's health.'* This model was used to establish Danila Dilba's women's clinic and was felt by staff and women attending the clinic to be an important factor in its acceptability. The regular women's clinic quickly became busy and long waiting times may have discouraged women from attending for Pap smear screening. Pap smear screening coverage af Danila Dilba was lower than has been reported for populations of Aboriginal women from remote comm~nities.~,~ However, the method of estimating coverage may have contributed to this result. In Danila Dilba's urban setting, the population of women who would identify Danila Dilba as their Pap smear provider is difficult to define from practice records. In the Darwin region, many Aboriginal women are mobile, moving Table 4: Reasons for lack of Pap smear screening between 1 July 1995 and 1 July 1996 for all women who attended the clinic when due for a Pap smear. Reason for Pap smear not being done Number Yo No reason evident in file Documentation that Pap smears were discussed or offered Many serious medical problems 18 8.,... Preanant and inamrotxiate to perform Pap when seen 7 3 ~ ~ Reported Pap smear elsewhere but incomplete documentation 6 3 Mainly lived elsewhere 2 1 Total between Darwin and remote communities and maintaining addresses in both places. In addition, women attending Danila Dilba have access to, and use, multiple health care providers for Pap smears and other services, both in town and other communities. This makes it difficult to ascertain fully any Pap smears done. In remote communities, populations are more easily defined by health service staff and Pap smears done are usually performed by a single health care provider. Hence, both the numerator and denominator used to calculate screening coverage are more certain in remote communities than in urban settings such as Darwin. The trend of declining screening coverage with increasing age is similar to that observed elsewhere for both Aboriginal women and for all Australian women.4 The low rates of Pap smear abnormalities are similar to those reported for other populations of Aboriginal w~rnen.~-~.'~ Possible explanations are that Aboriginal women who are at lower risk for cervical cancer are selectively screened with Pap smears or that cervical cancer is a more rapidly progres- Table 5: Impact of Pap smear recruitment strategies 1 July 1996 to 1 October Group Personal approach Number Yo Letter Number YO Control Number YO Women in aroup Actually or potentially contacted a nla Attended Danila Dilba Pap smears discussed Pap smears performed Note: (a) Women were defined as being potentialy contacted for this group if the letter was not returned to sender VOL. 22 NO. 6 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 723

5 Hunt et al. sive disease for Aboriginal women. Currently, there is little evidence to support either theory. A study of screening histories of Aboriginal women who have been diagnosed with cervical cancer is needed to explore this question further. In general practice settings, file tagging has been shown to be an effective prompt to encourage women to have Pap ~mears.~. ~ In this study, women attending the clinic and therefore able to benefit from a file tag intervention. had only a modest increase in Pap smear screening coverage demonstrated in the year following file tagging. For almost a quarter of women seen in the clinic who were due for a Pap smear but not screened, there was documentation in the file of Pap smears having been discussed. These women declined Pap smears at the time seen and did not return for them at another time. This may have indicated reluctance by women to have the test, or other priorities identified by women as taking precedence. A qualitative study of the cervical screening experience of Aboriginal women in the Fitzroy Valley of WA concluded that Pap smear screening was perceived as of a lesser priority than many other activities.i5 In the context of the social and economic disadvantage and poor health which many Aboriginal women and their families suffer, this is understandable. The interventions assessed in the recruitment trial had a minimal impact on increasing documented Pap smear screening for this population. It is possible that Pap smears were performed elsewhere but not documented in Danila Dilba files. Other factors may have contributed to the small numbers of Pap smears performed during the follow-up period. A funding crisis for the project, resulting in staff turnover and instability, coincided with the intervention follow-up period. Aboriginal Health Workers had less time available to follow up women and had to reduce health promotion activities. The financial insecurity faced by many Aboriginal medical services is a frequent barrier to the smooth running of projects and services. This study provides limited evidence that a personal approach by Aboriginal Health Workers can encourage women to attend for Pap smear screening in an urban setting, provided the health workers are able to make contact with the women. Of the 22 women in the personal approach group who were actually contacted by the Aboriginal Health Workers, eight had a Pap smear performed. For other populations of women, reminder letters sent from general practitioners have increased Pap smear ~creening.~j~ In this study, letters appeared to prompt an increased discussion of Pap smears during subsequent clinic attendance. Of women sent letters, 15% had documented discussion of Pap smears, compared to 6% of controls. It is possible that this discussion would have influenced women s future decisions to have Pap smears. However, letters had little impact on women choosing to have Pap smears during the follow-up period, with only 2% of women in the letter group having Pap smears. One explanation for this result is that Aboriginal women are more mobile than other women and therefore letters are less likely to reach them. This is supported by evidence from this study by the large proportion of letters returned to sender and the changes of address recorded for almost half the women in the letter group in the previous three years. Another explanation is that Aboriginal women are less likely than other women to respond to a letter for social or cultural reasons. The Northern Territory Pap Smear Register began operation in March The response in the letter group of this study suggests that reminder letters generated by the register would be of limited value in encouraging Aboriginal women in Darwin to have Pap smears. For small and remote communities, it may be feasible and acceptable for women to be followed up from a list sent to the community clinic. However, in an urban setting, the higher numbers of women, their mobility and diversity, and the use of multiple service providers would make this follow-up process more difficult, and less likely to be effective or cost effective. The Gumileybirra Women s Health Project was funded as an ongoing women s health service from October 1996, and continues to provide clinical and community women s health services. The evaluation of Pap smear screening reported here stimulated much discussion among staff, helped staff improve the organisation of the clinic and provided a basis for the re-orientation of activities. The long waiting times have been addressed by opening the clinic for a second day each week and by further refining the appointment system. A number of Aboriginal Health Workers at Gumileybirra have chosen to undertake specialist training in women s health and now perfom Pap smears, giving women a choice of provider. Women s health staff continue to talk to and encourage women who have never been screened, or who are overdue to have a Pap smear, through discussions with groups of women in the community, and with individual women when they attend the women s and general clinics. Opportunistically updated file tags indicating Pap smear sfatus have continued to be used to prompt discussion. While active efforts continue to be made to ensure the follow up of women with abnormal Pap smears, Danila Dilba s Pap smear database has not been maintained. Gumileybirra Women s Clinic staff have not continued either to generate lists of women due for routine Pap smears or to send reminder letters. The experience gained from the evaluation reported here suggests that the most appropriate and feasible means of encouraging Pap smear screening for urban Aboriginal women is by opportunistically promoting and conducting Pap smears, and by using a holistic and culturally sensitive approach to providing women s health services. Acknowledgments The authors would like to acknowledge the significant contribution of staff of Danila Dilba and in particular all the Gumileybirra Women s Health Project staff who made this research possible. Danila Dilba s management and committee have generously supported Gumileybirra and the research projects. Thanks also to the Aboriginal and Torres Strait Islander women of Darwin who have made Gumileybirra a success. The Gumileybirra Women s Health Project was primarily funded by two project grants from the Divisions of General Practice Program, Department of Health and Family Services. Additional funding was provided by grants from the Women s Cancer Prevention Program of Temtory Health Services and Healthy Women Strong Families Program of the Department of Health and Family Services. 724 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 1998 VOL. 22 NO. 6

6 Pap smear screening at an urban Aboriginal health service Thanks to Dr Fiona MacDonald who provided helpful comments on an earlier draft of the manuscript and to the two reviewers for their useful comments. References I. Plant AJ, Condon JR, Durling G. Northern Territory health outcomes: Morbidit)) and mortulify Darwin: Northern Territory Department of Health and Community Services, Redman S, Barratt A. Towards a population-based screening program for cervical cancer. Aust J Public Health 1995; 19: Mak DB, Straton JAY. The Fitzroy Valley Pap smear register: cervical screening in a population of Australian Aboriginal women. Med J Aust 1993; 158: Straton JAY. Recruitment fur cervical screening: A review of the literature. Canberra: AGPS, Gilles MT, Crewe S. Granites IN, Coppola A A community-based cervical screening program in a remote Aboriginal community in the Northern Territory. Aust J Public Health 1995; 19: Guest CS, Mitchell H, Plant A. Cancer of the uterine cervix and screening of Aboriginal women. Aust N Z J Obstet Cvnecol 1990; 30: Access Version 2.0. Microsoft, Women s health business - the Pap test (flip chart). Darwin: Family Planning NT, National Health and Medical Research Council. Guidelines for the munagement of wumen with screen detected abnormalities. Canberra: AGPS, Dean AG, Dean JA. Burton AH, Dicker RC. Epilnfo [computer program]. Version 5.Ola. Atlanta: Centers for Disease Control, Castles 1. Census characteristics of the Northern Temtory: 199 I Census of population and housing. Australian Bureau of Statistics, Catalogue No.: Canberra: Commonwealth Government Printer, Carter B, Hussen E. Abbott L, et al. Borning: Pmere Laltyeke Anweme Ampe Mpwaretyeke: Congress Alukura by the Grandmothers Law - a report prepared by the Central Australian Aboriginal Congress. Australian Aboriginul Studies 1987; 1: Reath JS, Patel M, Moodie R. Cervical cytology in Central Australian Aboriginal women. Aust Fam Physician 1991; 20: Pritchard DA, Straton JAY, Hyndman J. Cervical screening in general practice. Aust J Public Health 1995; 19(2): Toussaint S. Mak D, Straton J. Marnin Business: Anthropological interpretations of cervical screening among Australian Aboriginal women. Aust J Primary Health Intetchange 1998; 4(2): In press VOL. 22 NO. 6 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 725