Nordic gynecologists’ opinion on quality assessment registers
A questionnaire study and presentation of Nordic guidelines for quality assessment of gynecological surgery
Abstract
Background. The Board of the Nordic Federation of Societies of Obstetrics and Gynecology established a Quality Group for Gynecological Surgery connected with benign diseases with one member from each of the Nordic countries. The aim was to suggest guidelines for the quality assessment of such surgery.
Methods. A questionnaire study was carried out in all Nordic gynecological departments concerning quality assessment and the elaboration of guidelines.
Results. Most of the departments (80.4%) want to participate in establishing and maintaining a quality register. The majority of responders strongly agree that registers detect areas of satisfactory as well as unsatisfactory quality, and are necessary for comparison of treatment results among departments. Several departments are hesitant about joining a register if it will incur financial costs (59.5%) and/or extra work (47.6%). Most departments (71.4–94.6%) consider that all major gynecological surgery should be monitored. The departments agree that pre-, per-, and post hospital patient events should be included in any register. These results are incorporated into common general guidelines for quality assessment of gynecological surgery in the Nordic countries.
Conclusion. The gynecological departments in the Nordic countries agree on the necessity for and the main content of registers for quality assessment, but the interest drops when extra work is needed, and if the departments have to pay to participate, which emphasize the need for simple and low-cost registers. In accordance with the results of the Nordic questionnaire, common general quality guidelines for gynecological surgery are suggested.
Abbreviations
-
- NFOG
-
- Nordic Federation of Obstetrics and Gynecology.
Background
The Board of the Nordic Federation of Obstetrics and Gynecology (NFOG) established the NFOG Quality Group for Gynecologic Surgery with one member, experienced in quality assessment, from each of the Nordic countries. Its mandate from 2000 to 2002 was to suggest guidelines for quality assessment connected with benign gynecological surgery.
Patients and health authorities demand quality assessments of medical treatment. The medical profession has to take the initiative in introducing such quality assessments in order to improve quality of treatment (1–3). The medical outcome may differ considerably depending on both the physician and the department (4). In order to make analyzes and comparisons possible, comparable data collected in a uniform manner are needed (5–7). These have to be organized in registers, which will also serve as a resource for statistical analyzes.
Objective measurements and assessments are needed (8). Valid data collection pre- and postoperatively can often be done by the patient herself (9). This will make evaluation of indications possible, and reduce the physician's work in terms of data collection while avoiding physician bias (10,11). The board of the NFOG wants a few simple and reliable indices. In the analysis, an adequate number of preoperative parameters concerning health and symptoms are essential to avoid confounding analyzes due to patient selection (6). A dilemma in quality assessment is to set enough parameters without creating too much extra work for the participants. No quality assessment, however, exists without extra work.
In the Nordic countries there are national quality registers on benign disease. The Hysgo database in Denmark monitors hysteroscopic surgery, the Norwegian Urogynecologic Group database follows urinary incontinence surgery and in Sweden the Gynop register monitors the benign gynecological surgery concerning the uterus and adnexa as well as hysteroscopic surgery. The need and wish for quality registers in the Nordic departments have not been inquired, neither what they shall include. There has been no investigation into the need or desire for registers in the Nordic departments or into what the registers should include.
In order to elaborate practical and usable guidelines for quality assessment, we wanted to ask to what extent quality assessment is carried out and what level of quality assessment the departments of obstetrics and gynecology in the Nordic countries want.
Methods
A questionnaire concerning quality assessment of surgery was constructed, which inquired about the following: (i) volume of gynecological surgery connected with benign diseases; (ii) current status of quality assessments of this surgery; (iii) planned or setup quality assessment registers; (iv) parameters needed for quality assessment; (v) procedures and methods to be monitored; and (vi) willingness to participate in general (national) quality assessment registers, including acceptance of extra work and costs.
The questionnaire was constructed in English in order to pose identical questions in the five Nordic countries, and was validated for language comprehension and apprehension in three steps. The five representatives from participating countries sent the questionnaires to the head of every department of gynecology in the Nordic countries and collected the completed forms. The head of departments or in a few cases the person responsible for quality assessment answered the forms. At least two reminders were sent and answers were entered into a common database.
Based on the results from the questionnaire, common Nordic guidelines for quality assessment of gynecological surgery connected with benign diseases were elaborated.
Statistics
The statistical package SPSS 9.0 (SPSS Inc., Chicago, IL) was used for analyzes. Kruskal–Wallis and chi-square tests were performed. P < 0.05 was considered significant.
Results
The questionnaire was sent to a total of 185 departments of which 168 (90.8%) answered. The national figures are: Denmark 34 out of 36 (94.4%), Finland 40 out of 50 (80.0%), Sweden 48 of 53 (90.5%), Norway 43 of 43 (100%) and Iceland 3 out of 3 (100%).
In Denmark, 25 (73.5%) of the gynecological departments participate in the Hysgo database, of the Norwegian departments 29 (67.4%) register data in the NUGG database. The Swedish Gynop register monitors 25 (52.1%) of the Swedish departments.
Very few of the Nordic departments (7.7%) collect data in local registers: Finland 4 (10.0%), Denmark 4 (11.8%), Norway 2 (4.7%), Sweden 3 (6.3). Almost half (45.2%) of the Nordic departments do not participate in any register at all. In Iceland none of the three departments participate in any register, in Finland the figure is 36 (90.0%), in Denmark 5 (14.7%), in Norway 12 (27.9%) and in Sweden 20 (41.7%).
Several departments (54.8%) have some kind of in-patient data files which can include participation in quality registers, but also simple hospital discharge data on sex, age procedures and diagnosis, which are not useful for quality assessment, but can be used for surveillance of disease, changes in treatment procedures, etc. However, regular evaluation of results is only performed in 41.1% of the departments with no major differences among the countries.
Using a visual analog scale (VAS), the majority of responders agree that registers: (i) are necessary for comparison of treatment results among departments (bench marking) (Mean: 2.05); (ii) they improve quality in themselves (Mean: 2.59); (iii) act as a checklist for essential data to be collected (Mean: 2.28); and (iv) detect areas where quality improvement may be needed, as well as areas where quality is satisfactory (Mean: 2.35). (VAS: 0 = total agreement, 10 = total disagreement).
The results do not differ markedly among the countries.
A majority of the departments agree that uterine, adnexal, incontinence and prolapse surgery should be monitored, together with abdominal, vaginal, laparoscopic and hysteroscopic procedures (Table I). There are only small differences among the countries. However only 71.4% and 82.1% consider that adnexal surgery and hysteroscopic surgery, respectively, should be recorded.
Finland n = 40 % | Denmark n = 34 % | Norway n = 43 % | Sweden n = 48 % | Iceland n = 3 % | Total n = 168 % | |
---|---|---|---|---|---|---|
Uterine surgery | 85.0 | 94.1 | 88.4 | 93.8 | 100 | 90.4 |
Adnexal surgery | 70.0 | 67.6 | 62.7 | 81.3 | 100 | 71.4 |
Incontinence surgery | 97.5 | 85.2 | 97.6 | 95.8 | 100 | 94.6 |
Prolapse surgery | 92.5 | 82.4 | 83.7 | 91.7 | 100 | 88.1 |
Abdominal procedures | 82.5 | 88.2 | 88.4 | 93.8 | 100 | 88.7 |
Vaginal procedures | 97.5 | 88.2 | 90.7 | 93.8 | 100 | 92.9 |
Laparoscopic procedures | 95.0 | 85.2 | 93.0 | 93.8 | 100 | 92.3 |
Hysteroscopic procedures | 82.5 | 88.2 | 72.1 | 87.5 | 66.7 | 82.1 |
The departments agree that all the parameters health status, gynecologic status, preoperative treatment, indication for operation, operative procedures, postoperative diagnosis, medical complications, operative complications, recovery and patients satisfaction should be included (VAS: 0 = total agreement, 10 = total disagreement, range of means 1.2–2.3).
Several departments (72.6%) claim they can collect the operative indications, the procedures (84.5%), the postoperative diagnosis (83.9%), and the operative complications (80.4%) from their current registration procedures. Departments that do not participate in a register nevertheless consider that they can deal with the parameters on patient health status (38.2%), gynecological status (39.4%), preoperative treatment (38.2%), medical complications (61.8%), complications after discharge (38.2%) and patient satisfaction (27.6%) from their current system of registration. No significant differences were found between those departments participating in registers and those which did not with one exception – more departments participating in a register thought they could get data on patient satisfaction (p < 0.05).
Generally, most departments 135(80.4%) express an interest in participating in a quality register, want to have early reports (88.1%) and will send participants to annual meetings on quality maintenance/improvements of registries (64.3%)(Table II).
Finland n = 40 % | Denmark n = 34 % | Norway n = 43 % | Sweden n = 48 % | Iceland n = 3 % | Total n = 168 % | |
---|---|---|---|---|---|---|
Want to participate in registers? | 72.5 | 82.4 | 83.7 | 81.2 | 100.0 | 80.4 |
Want to participate even when extra work is needed? | 47.5 | 55.9 | 62.8 | 45.8 | 33.3 | 52.4 |
Want to participate in registers even if you have to pay? | 25.0 | 23.5 | 30.2 | 72.9 | 66.7 | 40.5 |
Want yearly reports? | 85.0 | 85.3 | 90.7 | 89.6 | 100.0 | 88.1 |
Want to send participants to meetings? | 42.5 | 82.4 | 58.1 | 75.0 | 66.7 | 64.3 |
However, the interest drops to 52.4%, when extra work is needed, and drops further to 40.5% if the departments have to pay to participate. The departments in Norway are more willing to accept extra work, while the Swedish departments are more willing to pay to participate in a register. Several departments (27.2%) do not give a clear answer to this question, commenting that it would depend on the price.
The departments were asked if they follow-up operated patients and if so to describe how this is done. Only 30.3% of the departments in the Nordic countries followed all or almost all of their patients. Denmark (5.8%) and Finland (25.0%) had the lowest level of follow-up compared with Norway (44.2%), Sweden (37.5%) and Iceland (66.7%). The follow-up was restricted to certain operations and groups of patients.
Discussion
Principal findings
The results of the first Nordic study concerning quality assessment of surgical treatment within benign gynecological disease reveal a general interest in quality work. More than 4/5 of the departments want to participate in a quality register, when established, almost all agree that quality registration is a tool for quality improvement, and in those Nordic countries with registers the participation rate is fairly high. However, the interest for quality assessment is dependent on an acceptable workload and low cost, which emphasize the need for simple not too expensive registers. These results and the experience from the present Nordic registers made it possible for the NFOG quality group to suggest the first common Nordic guidelines for quality assessment of benign gynecological surgery.
Strength and weakness of the study
The questionnaire was validated by discussion among the authors, tested on eight selected key people, one/two from each of the departments of the five authors and adjusted according to the results. It can be argued that this is a sub-optimal validation. The overall response rate for the questionnaires is high with very few non-responding departments. The difference in response rates among the countries might reflect the intensity of the reminders.
Implications of the study
The interpretation is that almost all departments want quality registration. However, there is a lack of both time and/or resources to carry it out. Some may wait until health authorities make registration compulsory. The majority of the departmentswants annual reports and will send delegates to yearly meetings. The level of acceptance of extra work in connection with the registration and the willingness to pay for the service a register can provide varies. This might reflect a departmental feeling that if they join quality registers, public authorities should pay for it. By legislating for quality assessment, health authorities will claim that it is an integral part of the department's duties and there will be no compensation in the form of extra funding. The medical profession and their administrators will have to find a solution to this problem.
Neither Iceland nor Finland has on-going national quality registers. Finland has previously registered laparoscopic hysterectomies over a 2-year period (12), all hysterectomies for 1 year (Finhyst study) (4) and tension-free vaginal tape operations for 3 years (13).
In the Nordic countries there is a low participation in local registers that might be used to carry out quality assessment. The registers have to make themselves attractive to potential participants. However, the extra work and costs connected with participation in a register counteract the general wish to participate. Double work, such as registration in a patient file and in a register, will reduce both participation and validity. Integration of registration in the patient file and the register is of vital importance, as data should only be registered once (14,15).
Half the departments record surgical data in data files and/or participate in registers, but only 41.1% analyze their data regularly (locally or centrally). The major part of the work for the participant in the register lies in registering the data. Stored but not analyzed data cannot be expected to increase motivation.
The departments' opinions about what should be registered cover all the main gynecological procedures as well as anatomical structures defined as gynecological. They are not able to make any priority, and may expect their organization or health authorities to make the priority.
In order to be able to compare results both on hospital and national levels patient selection, background factors (16) and type of preoperative treatment have to be recorded (17). It is therefore logical that all the relevant data, from the patient's preoperative consultation to follow-up after discharge, are digitized and registered as agreed on by the departments. This calls for a large numbers of parameters, almost a complete patient record. A lot of data can be collected from the patient, as it is in her interest to supply the physician with information (18). The physician's interest in registering data is less (10). Compromises have to be made between the amount of data needed and the workload involved in collecting it (19). Resources for analyzes must be provided. Most departments, including those not participating in any register, claim they can collect patient parameters even those not registered in the national coding system.
Thus, the departments regard it as both possible and practical to extract these data from their own patient records. However, if these data are not digitized in data files, it is almost impossible to access them in a useful and systematic way. The departments in fact, do not know what parameters they can obtain, and that valid assessment can only be achieved by prospective registration. This might imply that the response to the questionnaire suggesting that they are in favor of, and will participate in and wish for extensive quality registration might not be realistic when it comes to practice.
Organizational aspects were not included in the questionnaire, but all the present registers are dependent on groups of engaged gynecologists with special interest in quality assessment.
The Swedish Society of Obstetrics and Gynecology appoints the board of the Swedish Gyn-op register. Twice yearly representatives for participating departments are called to meetings, which serve as chief executive. The government and a minor fee from participating departments finance the register. The databases are owned by the county council but accessed only by the board of Gyn-op. The software of the NUGG database is developed by the company Urogyndata A/S, and the license for use is sold to the hospital departments. Their budget balances around 300.000 NOK. The collected data are owned by a non-profit organization NUGG. It consists of five self-established urogynecologists. This group invites all interested departments to participate in the NUGG register. The HYSGO-database is located at Glostrup Hospital and the yearly budget of around 300.000 DK is dependent on grants from the country council budget for databases. The collected data are accessed by the HYSGO board, which consist of self-established gynecologists approved by the Danish Society of Obstetrics and Gynecology.
To provide sufficient information quality registers should probably be divided into subspecialties regarding diseases or surgical methods both for audit and feedback. The registers can be totally separated and individual with different organizations, software, and registration forms such as the ongoing registers or those under construction in Denmark and Norway (NUGG, HYSGO and a Danish hysterectomy database). On the other hand, sub-specialized registers can also have a common organization, software and use identical basic parameters, but this requires more compromises and cooperation as in the Swedish Gynop register, containing uterine, adnexal, and hysteroscopic surgery.
There is no regular follow-up of surgically treated patients in 29% of the Nordic departments.
Classifying follow-up performance shows that 20% of the departments report that their intention is to have complete follow-up with another 10% following up most patients. The actual results and complications are not documented systematically and cannot therefore be reported by most, if any departments, unless they participate in a register.
The Finhyst study (4) together with the national reports from the Norwegian Urogynecologic Group database and the Swedish Gynop register show a good standard of treatment administrated, with no major differences among the participating departments nor overall results falling below or complication rates rising above those reported from other international studies. However, with systematic follow-up the departments can show both authorities and patients that they provide sufficient quality in their medical treatment. Comparable follow-up among departments will make it possible to identify areas needing improvement, as is also shown in the Finhyst study4 and in the national reports from the Norwegian Urogynecologic and Swedish Gynop register.
Suggested guidelines for quality assessment of gynecological surgery connected with benign diseases
There have been limited attempts to create a common framework for quality assessment of the treatment of gynecological patients (5,8).
Based on the results of the questionnaire study, a study of the relevant literature (1–24), experiences from the present registers, and the work of the NFOG quality group, a suggestion for common Nordic guidelines for quality assessment of gynecologic surgery connected with benign diseases is elaborated and presented (Table III).
Parameter | Comment |
---|---|
Preoperative examination | To ensure appropriateness of surgery, comparability of patient groups, to minimize bias and identify high-risk patients patients |
Patient symptoms; H, G, N | Why the patient sees the doctor. To avoid doctor's bias, the symptoms should be registered by thepatient. Findings at examination are secondary to patient's symptoms. Ex. fibromyoma is not a symptom |
Indication(s) for procedure H, G, N | All possible indications should be included. Selection of indication must be carried out in order to benchmark the departments |
Preoperative treatment G | Ex.: Is medical treatment tried before extirpation of the uterus to treat menorrhagia? |
Patient health status G | Maximum of 3–4 defined levels are suggested. Example: ASA system |
Gynecological status H, G | Ex.: Degree of genital prolapse (ICS classification) and uterine size |
Hospital stay | To compare treatment modalities |
Emergency/elective H,G | |
Surgical procedure H, G, N | Indicated by current coding systems. |
Complication(s), medical (H) G | Defined complications, indicated by current coding systems |
Complication(s), surgical H, G, N | Defined complications, indicated by current coding systems |
Follow-up after discharge | Main outcome measurement. Effect of treatment |
The minimum measurement is a validated questionnaire sent to all patients, and a high response | |
rate must be ensured | |
Change in condition H, G, N | Improvement of the condition, which took the patient to the doctor |
Patient satisfaction H, G, N | This correlates with other factors than treatment results: is the patient satisfied with information |
the hospital stay, how was she met and t,reated by nurses and doctors, etc. | |
Complications H, G, N | Were there no/minor/major-defined complications after discharge from hospital? |
Monitoring system | |
Quality register | |
Patient inclusion H, G, N | Prospective inclusion of patients is mandatory. Retrospective registration will create incomplete data |
Data collection H, G, N | Validated doctors' forms and patient questionnaires. Integration in patient file would be optimal |
Data analyzes H, G, N | All data to be analyzed at regular intervals |
Data quality H, G, N | ‘Data missing’ must be minimized and methods for validation of data must be established |
Feedback and audit system H, G, N | Results from processing of the data to be distributed in shortest possible time, providing feedback to the departments making an ‘unbroken circle’ of physician-register-physician. Improvement in quality based on the suggested monitoring system is dependent on establishment of an effectivenal level |
Quality improvement | Use of checklists will ensure collection of essential data for high quality treatment Registration in itself improves quality Benchmarking Study of the processed data will detect areas of adequate as well as inadequate quality Actions based on results to improve quality, which will be documented in later analyzes |
According to these guidelines all departments should be able to present exchangeable patient data covering: (i) preoperative examination, (ii) hospital stay, and (iii) follow-up after discharge. All the registers include the most important suggested parameters, and use the suggested monitoring system. Most vital is high data quality. Registration makes no sense without a prospective registration with complete and validated data with an effective feedback and audit system, for example annually meeting with discussion of results.
Main consequences from the questionnaire and experienced elsewhere (20–23) is that quality registration has to be simple and not too extensive. Essential is a few carefully selected key parameters (indicators). According to the present experience from the Nordic registers, registration of surgical procedures and a few selected essential outcome parameters is most important (Table III). These parameters are main complications – chance in condition and – satisfaction with outcome. Additionally a few preoperative parameters such as patient's symptoms and indications for procedures are needed. The parameters: patient's health status and preoperative treatment are only included in the Gynop register and suggested in the questionnaire. Moreover the NUGG register does not include gynecological status, medical complications, as well as emergency surgery (rarely performed in urogynecology). In all registers the current coding system for hospitals are used [the International Classification of Diseases, ICD-10 and the NOMESCO (Nordic Medical Statistic Committee) Classification of Surgical Procedures, NCSP]. In this way, data can be statistically processed and compared across national borders. Moreover the completeness of registration can be validated by comparison of the register data to the data in the hospital system. At present there has only been limited publication of the experiences from the Nordic registers (10,13,14), and the future will reveal the relevance of the present registration in improvement of quality assessment.
Conclusions
The result of the questionnaire study showed that the majority of the Nordic gynecological departments are in favor of systematic but simple quality assessment. Based on the study and present registers, general quality guidelines for surgeries connected with benign gynecological diseases are suggested.
Data should be exchangeable and comparable between departments and across national borders. Inter Nordic co-operation is of vital importance for the pooling of resources and experience, and for avoiding the repetition of mistakes.
Acknowledgments
The NFOG Board is recognized for providing financial support. Secretary Birgitta Renström, Norrland University Hospital in Umeå is thanked for the data work with the questionnaires. E.R. received financial support from the Directorate of Health and Social Welfare, Norway.