SEARCH

SEARCH BY CITATION

Keywords:

  • Gynaecological surgery;
  • hysterectomy;
  • laparoscopic adnexal surgery;
  • prediction model;
  • predictor;
  • return to work

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Objective

To measure the impact of the level of invasiveness of gynaecological procedures on time to full Return to Work (RTW) and to identify the most important preoperative sociodemographic, medical and work-related factors that predict the risk of prolonged sick leave.

Design

Prospective cohort study.

Setting

Dutch university hospital.

Population

A total of 148 women aged 18–65 years scheduled for gynaecological surgery for benign indications.

Methods

A questionnaire regarding the surgical procedure as well as perioperative and postoperative complications was completed by the attending resident at baseline and 6 weeks after surgery. All other outcome measures were assessed using self-reported patient questionnaires at baseline and 12 weeks post-surgery. The follow-up period was extended up to 1 year after surgery in women failing to return to work. Surgical procedures were categorised into diagnostic, minor, intermediate and major surgery.

Main outcome measures

Time to RTW and important predictors for prolonged sick leave after surgery.

Results

Median time to RTW was 7 days (interquartile range [IQR] 5–14) for diagnostic surgery, 14 days (IQR 9–28) for minor surgery, 60 days (IQR 28–101) for intermediate surgery and 69 days (IQR 56–135) for major surgery. Multivariable analysis showed a strongest predictive value of RTW 1 year after surgery for level of invasiveness of surgery (minor surgery hazard ratio [HR] 0.51, 95% CI 0.32–0.81; intermediate surgery HR 0.20, 95% CI 0.12–0.34; major surgery HR 0.09, 95% CI 0.06–0.16), RTW expectations before surgery (HR 0.55, 95% CI 0.36–0.84), and preoperative functional status (HR 1.09, 95% CI 1.04–1.13). A prediction model regarding the probability of prolonged sick leave at 6 weeks was developed, with a sensitivity of 89% and a specificity of 86%.

Conclusions

RTW often takes a long time, especially after intermediate and major surgery. This study reveals important predictors for prolonged sick leave and provides a prediction model for the risk of sick leave extending 6 weeks after benign gynaecological surgery in the Netherlands.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Return to work (RTW) and full recovery after benign gynaecological surgery often takes a long time, irrespective of the introduction of minimally invasive surgery and other improvements in perioperative care aiming to reduce morbidity and to enhance recovery.[1-3] Delays in recovery and time to RTW reduce Quality of Life in postoperative women and generate unnecessary yet substantial costs for society through lost working hours, physician consultation and increased use of medication.[4-6] Patients with a delay in time to RTW after gynaecological surgery reported pain/discomfort, anxiety, depression and infections as important delaying factors.[7] In contrast, recovery and RTW time were shorter when women received clear uniform recommendations at discharge or when the woman had been provided with a time to RTW advice.[8, 9] However, little is known about preoperative personal and work-related factors as predictors for delayed RTW in women who undergo gynaecological surgery. Knowledge about factors that predict prolonged sick leave provides opportunities to identify high-risk women. These women could receive preventive or therapeutic treatments for the factors that can be influenced, e.g. counselling of RTW expectations and workplace adaptations. Anticipating important general factors for prolonged sick leave, and improvement of perioperative care could be realised for all gynaecological patients.

The first aim of this study was to measure the impact of the level of invasiveness of gynaecological procedures on time to full RTW. The second aim was to identify most important preoperative sociodemographic, medical and work-related factors that might predict the risk of prolonged sick leave after gynaecological surgery.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Study design and participants

This prospective observational cohort study was conducted in the VU University Medical Centre in the Netherlands between July 2008 and December 2010. Patients were recruited for participation in the study when scheduled for elective gynaecological surgery for benign indications. Recruitment took place by sending these women an invitation letter on behalf of their gynaecologist, together with an information package consisting of: (1) a patient information letter about the study, (2) a leaflet about participating in scientific research in general, (3) an informed consent form, and (4) a prepaid envelope. By sending back the signed informed consent form to the researchers, the woman indicated that she was interested in participating in the study after which the researcher contacted her to evaluate whether she was eligible to participate in the study. Women aged between 18 and 65 years undergoing gynaecological surgery (abdominal, vaginal, laparoscopic) on benign indication and employed for at least 8 hours per week (paid or unpaid) including housewives, were included in the study. Exclusion criteria were: (1) (suspicion of) malignancy; (2) (ectopic) pregnancy; (3) deep infiltrating endometriosis (because of more frequent sick leave before surgery, resulting in other RTW expectations and health status at baseline; (4) acute, ambulatory or only hysteroscopic surgery, (5) working temporarily for an employment agency (because of the increased risk of not completing the follow-up period of 1 year with the same employment) and (6) not able to understand or complete the questionnaires written in the Dutch language.

Data collection

For each woman, a questionnaire on type of procedure, perioperative complications and postoperative complications was filled out by the attending resident at baseline and 6 weeks after surgery. All other outcome measures were assessed using self-report questionnaires and were taken at baseline (1 or 2 days before surgery) and 12 weeks after surgery. When a woman had not completed the questionnaire within 1 week after schedule, she received a reminder by email or by post. If no response followed, she was reminded by a telephone call. RTW was measured 12 weeks after surgery and if the woman had not RTW fully 12 weeks after surgery, the researchers approached her by telephone at 6, 9 and 12 months after surgery to investigate the time to first full RTW. The follow-up period regarding time to RTW was at maximum 1 year after surgery, or ended when first full RTW was reached.

Outcome measure

The primary outcome measure in this study is sick leave duration until first full RTW, defined as duration of sick leave in calendar days from the day of surgery until the actual day of full RTW in own work or in other work with equal earnings. Prolonged sick leave was defined as no RTW at 6 weeks after surgery, based on expert recommendations.[10-12]

Potential prognostic factors

Based on a literature search in Pubmed (English, no other limitations) about prognostic factors regarding RTW in a wide variety of surgical patients and clinical experience of the researchers, potential prognostic factors were determined. These factors were divided into five different categories; sociodemographic factors, medical factors, work-related factors, patient expectations for time to RTW and health status. Each category consisted of the following factors:

  • 1.
    Sociodemographic factor:
    1. Age (years).[13-15]
    2. Living condition (alone or with family).
    3. Children or partner in need of care.
  • 2.
    Medical factors:
    1. Type of surgery: according to the extent of surgery, classification into four groups took place: diagnostic surgery, minor surgery, intermediate surgery, major surgery. Precise assignment per group is presented in Table 1 and was based on previous research.[2]
    2. Major surgical complications: complications during or related to the surgery, defined as enlargement of the wound with more than 8 cm or re-surgery within 6 weeks after initial surgery (major complications).
  • 3.
    Work-related factors:
    1. Employment: salaried or voluntary (unpaid) and salaried or self-employed.
    2. Physical workload: light to moderate or heavy.[14, 16, 18]
    3. Work hours per week: <20 per week, or 20 hours and more per week.[15]
    4. Job satisfaction: (very) unsatisfied or (very) satisfied. This prognostic factor was scored using a numerical visual analogue scale (VAS, range 0–10). Score 0–5 was defined as (very) unsatisfied and 6–10 as (very) satisfied.[17]
    In the Netherlands, all women with paid work have rights to claim an occupation sick pay scheme in the Netherlands, so this factor was not considered as a relevant predictor.
  • 4.
    Patients' expectations for time to RTW after surgery[13, 18-21]:
    1. Expectation of the woman regarding time till first full RTW after surgery. Expectations were classified into ‘low’ and ‘normal’ expectations, based on detailed multidisciplinary guidelines on time to RTW, which were developed by an expert panel of gynaecologists, occupational physicians and general practitioners through a modified Delphi consensus method.[10] Low expectations on time to RTW were defined as longer than 1, 2, 4 and 6 weeks for respectively diagnostic, minor surgery, intermediate surgery and major surgery. A shorter or equal expected time to full RTW was regarded as ‘normal’.
  • 5.
    Health status[22]:
    1. Physical and mental health status assessed according to the Short-form health survey (SF-36).[23, 24] Considering the wide range of comorbidities and limited incidence in our relatively small population, comorbidities were considered to be most carefully covered by the physical health status of the SF-36.
    2. Functional status measured by a validated Recovery Index (RI) questionnaire.[2, 25] Before surgery, only five questions of this questionnaire are relevant (RI-baseline, see Appendix S1).
Table 1. Classification of surgery according to the level of invasiveness
Diagnostic surgery
Diagnostic laparoscopy
Minor surgery
Laparoscopic
Adnexal surgery
Abdominal cerclage
Vaginal uterine artery occlusion
Intermediate surgery
Laparoscopic
Removal of a cervix (after previous laparoscopic assisted supracervical hysterectomy)
Hysterectomy
Myomectomy
Sacrocolpopexy
Vaginal prolapse surgery (colporraphia, vaginal sacrospinous fixation, Manchester Fothergill)
Vaginal hysterectomy
Major surgery
Laparotomic
Adnexal surgery
Hysterectomy
Myomectomy

Statistical analyses

Data entry was performed using Microsoft Office Access 2003. All data entries were visually double-checked by two independent research assistants. Assignment of type of surgery took place according to the actually performed treatment. Statistical analyses were performed using SPSS statistical package (SPSS 20, Armonk, NY, USA) and R. The influence on overall survival for time to RTW was determined by calculating the Kaplan–Meier estimate and comparisons between curves were performed by the log-rank test. P < 0.05 was considered significant.

Developing prediction model

Stochastic regression imputation was performed to estimate missing scores. Cox proportional hazard models were used to analyse the effect of each potential prognostic factor on RTW over the whole follow-up period, uni- and multivariably.[26] The cumulative probability of prolonged sick leave at 6 weeks was chosen as the primary goal to develop the prediction model. To construct the prediction model, the balance between the number of prognostic factors and the number of RTW events in the model was considered, which is recommended not to be lower than 10–15 events per factor.[27] Therefore only the most relevant predictive factors were added to the Cox proportional hazard model. Subsequently, backward regression was applied and predictors were removed from the model when the P-value was >0.05. To examine whether the outcomes were influenced by the regression imputation, an additional backward regression was applied on the data without regression imputation for the factors with more than 5% missing.

The regression coefficients from this final model were used to obtain the risk of sick leave extending 6 weeks after surgery. This probability was calculated by using the baseline probability of RTW for an individual woman with a follow up of 6 weeks and the regression coefficients obtained from the Cox model after backward selection.[28] The regression coefficients express the effect of the predictors on RTW, while the purpose of the model was to express the risk of prolonged sick leave. Therefore, the sign of the regression coefficients was reversed (e.g. positive became negative). To make our prediction tool suitable for clinical use, each coefficient was divided by the lowest value of a continuous predictor and transformed to a round number of risk scores, reflecting the relative weight in the prediction of prolonged sick leave. The total risk score for each individual woman was determined by multiplying the risk scores by the value of each predictor and summing them up. Next, we divided the women into four equal-sized groups based on these risk scores, ranging from low to high. Risk score intervals were calculated with corresponding 6 weeks predicted probabilities of prolonged sick leave.

We compared the mean predicted probability as estimated by the model of prolonged sick leave of each group to the actual observed probability of the group by the Kaplan–Meier method. These predicted and observed probabilities of prolonged sick leave at 6 weeks were also plotted to assess calibration (i.e. agreement between predicted and observed probabilities of prolonged sick leave at 6 weeks). To test the discriminative ability of our model, the concordance statistic was determined, which is equal to the area under the curve. Explained variation—the amount of variance between women that can be explained by the predictive factors—was calculated with the Pseudo R2 technique.[29]

To adjust our prediction model for the fact that it was developed and tested in the same population, which causes over-optimism of the predictors in the model, bootstrapping techniques were applied (250 bootstrap samples).[26] With this technique, a shrinkage factor, to adjust the predictors for this over-optimism of its predictive ability, was calculated. Sensitivity, specificity as well as the positive and negative predictive values of this model were calculated for the same cut-off scores used to delineate the risk score categories.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

A total of 157 women met the inclusion criteria and were approached to participate in the study, of which 148 were willing to participate and were included in this study. Figure 1 presents the patient flow throughout this trial. Follow-up time ranged from 12 to 52 weeks after surgery.

image

Figure 1. Patient flow.

Download figure to PowerPoint

Patient characteristics and loss to follow up

The questionnaires filled out by the attending resident and the baseline questionnaires filled out by the women were available for all patients. Table 2 presents the baseline characteristics of participating women, represented per surgical category and reporting the amount of missing data per characteristic. Data regarding the primary outcome measure, time to full RTW after surgery, were available for 145 (98%) women, and were censored for two women due to the follow-up period of 1 year.

Table 2. Baseline characteristics and primary outcome measure of participating patients
 Diagnostic surgery (n = 40)Minor surgery (n = 36)Intermediate surgery (n = 35)Major surgery (n = 37)Overall (n = 148)Missing n (%)
  1. *One case was censored at 365 days due to end of follow-up period.

  2. **One case was censored at 365 days due to end of follow-up period.

Sociodemographic factors
Age (years), mean (SD)35.5 (4.1)38.8 (8.3)46.3 (7.7)43.3 (7.0)40.8 (8.0)0
Medical factors
Surgical category, n (%)40 (27.0)36 (24.3)35 (23.6)37 (25.0) 0
Surgical complications, n (%)0 (0)0 (0)3 (8.6)3 (8.1)6 (4.1)0
Work-related factors
Employment, n (%)
Salaried38 (95.0)32 (88.9)26 (74.3)31 (83.8)127 (85.8)1 (0.6)
Unpaid2 (5.0)4 (11.1)9 (25.7)6 (16.2)20 (13.5)
Physical workload, n (%)
Light and moderate37 (92.5)31 (86.1)27 (77.1)31 (83.8)126 (85.1)2 (1.4)
Heavy3 (7.5)5 (13.9)7 (20.0)5 (13.5)20 (13.5)
Work hours per week, n (%)
<20 hours per week4 (10.0)7 (19.4)6 (17.1)5 (13.5)22 (14.9)6 (4.1)
≥20 hours per week36 (90.0)28 (77.8)26 (74.3)30 (81.1)120 (81.1)
Satisfaction with job, n (%)
(Very) unsatisfied4 (10.0)2 (5.6)3 (8.6)4 (10.8)13 (8.8)13 (8.8)
(Very) satisfied35 (87.5)32 (88.9)27 (77.1)28 (75.7)122 (82.4)
Women's expectations
RTW expectation (days) before surgery, n (%)
Low expectation3 (7.5)7 (19.4)16 (45.7)8 (21.6)34 (23.0)27 (18.2)
Normal expectation34 (85.0)24 (66.7)9 (25.7)20 (54.1)87 (58.8)
Health status
RAND-36 baseline, mean (SD)
Physical health337 (79.3)310 (74.6)252 (80.3)244 (104.0)287 (93.8)7 (4.7)
Mental health318 (73.0)291 (74.7)248 (105.0)233 (97.1)273 (93.8)
Recovery index total baseline score (functional status), mean (SD)20 (4.6)20 (3.8)18 (4.6)17 (5.5)19 (4.9)4 (2.7)
Full RTW (days) after surgery
Median (IQR)7 (5–14)14 (9–28)60 (28–101)69 (56–135)28 (10–69)3 (2.0)
Mean (SD)13.1 (17.2)21.2 (18.8)78.4 (72.6)*103.8 (80.4)*53.4 (66.7)**

Return to work

Figure 2 presents the Kaplan–Meier curves for all four surgical categories. Median times to RTW were 7 days for diagnostic surgery (interquartile range [IQR] 5–14), 14 days for minor surgery (IQR 9–28), 60 days for intermediate surgery (IQR 28–101) and 69 days for major surgery (IQR 56–135). The difference between the curves was significant (log rank test: < 0.001). Six weeks after surgery, 87 (59%) women had returned to work.

image

Figure 2. Kaplan–Meier survival curves, presented per type of surgery. Number of days represent days of sick leave after surgery until RTW. In both the intermediate and the major surgery groups three patients were censored at 182 days because they had not yet RTW.

Download figure to PowerPoint

Prediction model

Table 3 presents the results of the univariable and multivariable Cox regression analysis of the 11 most relevant predictive factors. After applying backward regression analyses, surgical category, RTW expectation (days) before surgery and total baseline score of the RI were the strongest predictors for RTW 1 year after surgery. The results were not influenced by the regression imputation of the two factors with the highest number of missing (RTW expectation and satisfaction with job). The coefficients from the multivariable analysis are presented after shrinkage. Per predictor, the explained variance and the transformation of the predictor into risk scores of sick leave more than 6 weeks after surgery can be found in Table 3. With these risk scores, the risk of sick leave more than 6 weeks after surgery can be calculated for each individual. This score can vary between −25 and +31. The higher the score, the higher the risk of prolonged sick leave. Each categorical predictor that is not relevant for a particular woman should be multiplied by zero and each predictor that is relevant should be multiplied by one. For the baseline recovery index, the score itself can be used. The weight of all positively scored predictors needs to be added up to form the total risk score. For example, a woman who underwent minor surgery, with a normal RTW expectation and with a total score of the baseline recovery index of 18, had a total risk score of 1*8 (minor surgery) + score 0*7 (normal RTW expectation) + −1*18 = −10. The predicted risk of sick leave extending 6 weeks after surgery for this woman is 16% (Table 4).

Table 3. Univariable and multivariable analysis of RTW after 1 year
PredictorsUnivariable analysisMultivariable analysisRisk score***
HR*95% CIP-valueHR*95% CIP-valueCoefficient**
  1. *Hazard ratio (HR) > 1 means a shorter time until RTW and HR < 1 means a longer time until RTW.

  2. **Values are regression coefficients after correction for overoptimism using a shrinkage factor of 0.91.

  3. ***Risk score of prolonged sick leave of more than 6 weeks determined with the regression coefficients after shrinkage.

Sociodemographic factors
Age (≥45 years)0.730.50–1.040.08     
Medical factors
Surgical category
Diagnostic laparoscopy0Reference category  Reference category 
Minor surgery0.550.35–0.860.090.510.32–0.81<0.01−0.618
Intermediate surgery0.140.08–0.23<0.010.200.12–0.34<0.001−1.4620
Major surgery0.100.06–0.17<0.010.090.06–0.16<0.001−2.1429
Surgical complications0.470.21–1.070.07     
Work-related factors
Employment (unpaid)0.580.35–0.930.02     
Physical workload (heavy)0.830.52–1.330.43     
Work hours per week (<20)0.670.43–1.040.77     
Satisfaction with job ([very] unsatisfied)0.650.39–1.090.10     
Patients expectations
RTW expectation (days) before surgery (low expectation)0.450.32–0.64<0.010.550.36–0.84<0.01−0.557
Health status at baseline
Mental health1.001.00–1.01<0.01     
Physical health1.011.00–1.01<0.01     
Recovery index total score at baseline1.111.06–1.15<0.0011.091.04–1.13<0.0010.07−1*Baseline score
Table 4. Risk of sick leave 6 weeks after surgery according to risk categories
Risk categoryTotal risk scoren (%)RTW at 6 weeks, n (%)Prolonged sick leave (no RTW at 6 weeks), n (%)Observed risk of prolonged sick leave (Kaplan–Meier estimate) %Predicted probability of prolonged sick leave %
1−25 to −1641 (28)38 (93)3 (7)52
2−15 to −333 (22)28 (85)5 (15)616
3−2 to 938 (26)9 (24)29 (76)6856
410 to 3136 (24)2 (6)34 (94)8680
OverallTotal148 (100)77 (52)71 (48)3841

Evaluation of the model

In the multivariable analysis, the selected factors for the prediction model—surgical category (47.5%), RTW expectation before surgery (13.9%) and baseline score of the RI (18.4%)—together explained 57.5% of the variation in RTW, which is good. The area under the curve was 0.67, representing a satisfactory discrimination. Figure 3 presents the calibration plot, which shows that the agreement between predicted and observed probabilities of sick leave at 6 weeks after surgery was good.

image

Figure 3. Calibration plot showing the observed (Kaplan–Meier estimate) versus the predicted probability for sick leave 6 weeks after surgery. The dotted line represents the perfect calibration.

Download figure to PowerPoint

The total predicted probability of sick leave at 6 weeks was 38% which reasonably matched the observed risk (Kaplan–Meier estimate) of 41% (see Table 4). For example in this population, a woman with a risk score more than 10, has 80% predicted chance of sick leave more than 6 weeks.

Based on Table 5, a score of ≥−2 is chosen as threshold value for high risk of prolonged sick leave. Regarding the group of women with prolonged sick leave, 89% scored ≥−2 points and was therefore correctly identified (sensitivity) and only 11% of the cases would be missed using this cut-off value. Of the women with normal sick leave, 86% scored <−2 points and was correctly classified (specificity). In addition, the positive predictive value of the prediction rule at the score level of ≥−2 is 85%. This means that in this group of high-risk women, additional care for prolonged sick leave is justified in 85% of the cases because they will actually develop prolonged sick leave. The negative predictive value of women with a risk score <−2 is 89%, indicating that when this threshold is chosen, 89% of these women will correctly be classified as low risk of prolonged sick leave.

Table 5. Prognostic test characteristics for sick leave 6 weeks after surgery
Cut-off total scoren (%)Sensitivity (%)Specificity (%)Positive predictive value (%)Negative predictive value (%)
≥−25148 (100)1000480
≥−15107 (70)96496493
≥−274 (50)89868589
≥1036 (24)48979467

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Main findings

This study showed that RTW after benign gynaecological surgery, especially after intermediate and major surgery, takes a long time (median more than 8 weeks).

The level of invasiveness of the surgery, expectation about time of RTW and functional status measured by the baseline score of the RI questionnaire were identified as the strongest predictors for the risk of prolonged sick leave. These three factors together explained 58% of the variation in time to RTW between the women in this study.

Strengths and weaknesses

Few studies have evaluated time to RTW after gynaecological surgery as the primary outcome measure. Because of the wide variety of surgeries studied, all surgical levels of invasiveness in benign gynaecological surgery are represented in this study and RTW between various surgical categories may be compared. In addition, this is the first study that developed a prediction model for prolonged sick leave after gynaecological surgery, considering a wide range of sociodemographic, medical and work-related factors. Relevant predictors were predefined based on a literature search. This prospective cohort study is of high quality owing to only 2% loss to follow up on the primary outcome measure and in total only 1% of the RTW data are censored due to the follow up period of 1 year.

A limitation of this study is that it was performed in a university hospital, which may have caused a relatively wide range in days until full RTW due to a high percentage of complex pathology. This may result in a reduced external validity. Another limitation is the fact that the primary outcome might be susceptible to recall bias, because sick leave duration was self-reported by women. Furthermore, in this study the first full day of RTW is taken as outcome measure regarding time to RTW, which does not take into account recurrences of sick leave and therefore is probably an underestimation of work-loss days.[30] A limitation, common to all prognostic studies, is the possibility of omitting an important predictor. This study may also be underpowered to detect other predictive factors for RTW. The questions of the RI-baseline questionnaire were selected from the RI-10/RI-6 questionnaire,[2, 25] based on their appropriateness to be answered before surgery. Therefore, no information is available about the validity and reliability of this selection of five questions. A final limitation is that the generalisability of the prediction model has not yet been evaluated by external validation in another population of gynaecological patients, which is necessary before clinical application.[31] However, this study is currently underway. Furthermore, dissimilarities in healthcare systems, legislative and insurance systems, reintegration policies and RTW expectations between countries may decrease the relevance of the prediction model outside the Netherlands.

Interpretation

Time to RTW after surgery

Patients undergoing diagnostic surgery, RTW 1 week sooner than women undergoing minor (laparoscopic) surgery. RTW after minor surgery in our study is about 1 week slower than reported in Danish, Korean and Japanese studies of comparable types of surgery.[9, 32, 33] A wide range in days until time to RTW was seen in women who received intermediate or major surgery, which is also found in other studies reporting on these types of surgeries.[34] Almost all studies outside the Netherlands and UK report an earlier RTW after intermediate and major surgery of at least several weeks,[35-38] but in other Dutch and English studies the period of sick leave is comparable to our results.[7, 39-41] Part of the explanation for the differences in sickness absence between countries might be the result of dissimilarities in healthcare systems, legislative and insurance systems, reintegration policies and RTW expectations.[42, 43] Patients undergoing major surgery RTW much later than expected, only one had RTW at 6 weeks after surgery, while 6 weeks is generally considered as a normal recovery period for full RTW by gynaecologists.[10-12] An explanation might be an extended recommended sick leave period by occupational physicians, which was about 2 weeks longer than the advice given by gynaecologists for several types of hysterectomies in two Danish and Dutch prospective cohort studies.[3, 44]

Predictive factors of prolonged sick leave and the prediction model

In this study, less invasive surgery was associated with a lower chance of sick leave lasting more than 6 weeks. This result corresponds with several studies that showed a quicker RTW after laparoscopic hysterectomy compared with laparotomic hysterectomy.[36, 45] Another important predictor for prolonged sick leave turned out to be baseline expectations regarding time to RTW after surgery. This is in line with a number of other studies that identified this factor as an important predictor for time to RTW in a wide variety of patients.[13, 18-21] The total baseline score of the RI turned out to be a stronger predictor for prolonged sick leave than the RAND-36 for mental or physical health. This finding is in line with previous research, which showed that functional status 2 weeks after surgery, was more closely related to prolonged sick leave than the type of surgery.[2] In contradistinction to other studies, age and work-related factors had little association with prolonged sick leave in our group of women.[13-15, 18, 17] However, limited variation of age and work-related factors in this cohort could not be ruled out as an explanation for this finding. Surgical complications were not found as a strong predictor of prolonged sick leave either, which may be because our definition was restricted to major complications to make sure that only complications affecting time until full RTW were included in this model.

The selected predictors together explained 58% of the variance between individuals regarding prolonged sick leave. The other 42% of unexplained variance is partly caused by predictors that were not measured in this study or were not strong enough to survive the backward selection process, but a large part of the unexplained variance in sick leave will always remain unexplained due to individual random causes of prolonged sick leave.

Implications for practice

Increased risk of prolonged sick leave with a higher level of surgical invasiveness underlines the importance of minimal invasive surgery regarding a faster RTW. Despite more expensive instrumentation, longer duration of surgery time and higher training costs due to longer learning curves, many minimal access surgical procedures are cost-effective as a result of both shorter hospitalisation and a reduction of sick leave after surgery.[46] However, as expectation of time to RTW before surgery also appeared to be an important predictor for prolonged sick leave, it is assumed that even more advantage of minimal invasive surgery regarding a faster RTW could be reached when RTW expectations are optimised. In the present situation, detailed recommendations on resumption of activities are mostly not provided,[47-50] show substantial variability when present[12, 51-53] and are often not specified per surgical technique.[44, 53] Several studies have shown that uniform convalescence recommendations regarding return to normal and work activities in a variety of surgical operations reduced sick leave by several weeks.[4, 51, 54, 55] As a result of this study, and considering that RTW expectations seem a relatively easily adapted factor, it seems advisable to implement guidelines regarding RTW recommendations after gynaecological surgery. The effect of guidelines and tailored recommendations regarding RTW is currently being studied.[56] The final predictor for prolonged sick leave found in this study was the RI-baseline score. This questionnaire with only 5 Likert scale questions requires minimal effort for women to fill out and could therefore easily be used as a supplement to the other two predictors to evaluate the risk of prolonged sick leave before surgery.

The practical applicability of the risk scores of the prediction model depends, besides the results of the external validation, on the importance attached to the threshold of 6 weeks of sick leave after surgery. If no RTW at 6 weeks after gynaecological surgery is considered as prolonged sick leave and the cut-off point of the risk score of <−2 is taken, the negative predictive value is high, thereby preventing treatment in women with a low risk of prolonged sick leave. It should be noted that there is no consensus of what implies prolonged sick leave and the consequences can be looked at from both psychological and economical perspectives; delayed RTW after surgery reduces Quality of Life in postoperative women and generates unnecessary yet substantial costs for society.[4, 6] Furthermore, considering the great contribution of the surgical category to the total risk score of prolonged sick leave, it seems reasonable to develop separate prediction models for every surgical category. However, this study did not include enough women to develop prediction models for separate groups of women, but could be used as a basis for new research to develop these models.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

The results of this study provide an opportunity to identify women with a high risk of prolonged sick leave and to anticipate this by giving them additional care. Furthermore, this study underlines the relevance of s woman's expectations for time to RTW, which emphasises the importance of preoperative counselling and guidelines regarding RTW recommendations after surgery.

Disclosure of interests

AVN, JRA, MDL, MWH, WvM, HAMB and JAFH have no conflicts of interest or financial ties to disclose.

Contribution to authorship

All authors made substantial contributions to this study and manuscript. AVN set up this research project, performed the data collection, analysed the data and wrote the paper. JAH, HAB, JRA and WME set up this research project and participated in writing the manuscript, MDL performed a part of the data collection, MWH analysed the data and participated in writing the paper. All approved this version to be published.

Details of ethics approval

This study design and procedures were approved by the Medical ethics Committee of the VU University Medical Centre (number 2005/94).

Funding

No funding was received.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

The authors would like to thank all women for filling out the questionnaires.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • 1
    Kehlet H. Fast-track colorectal surgery. Lancet 2008;371:7913.
  • 2
    Brolmann HA, Vonk Noordegraaf A, Bruinvels DJ, de Vet RH, Dirksz AA, Huirne JA. Can prolonged sick leave after gynecologic surgery be predicted? An observational study in the Netherlands. Surg Endosc 2009;23:223741.
  • 3
    Johansen P, Al-Khafagi SK, Thostesen LM, Lauszus FF, Rasmussen KL. Analysis of need for sick leave after hysterectomy. Ugeskr Laeger 2008;170:14658.
  • 4
    Bisgaard T, Klarskov B, Rosenberg J, Kehlet H. Factors determining convalescence after uncomplicated laparoscopic cholecystectomy. Arch Surg 2001;136:91721.
  • 5
    Henderson M, Glozier N, Holland EK. Long term sickness absence. BMJ 2005;330:8023.
  • 6
    Waddell G, Burton AK. Is Work Good for Your Health and Well-Being?. London: The Stationary Office, 2006.
  • 7
    Clayton M, Verow P. A retrospective study of return to work following surgery. Occup Med (Lond) 2007;57:52531.
  • 8
    Clayton M, Verow P. Advice given to patients about return to work and driving following surgery. Occup Med (Lond) 2007;57:48891.
  • 9
    Kikuchi I, Takeuchi H, Shimanuki H, Kitade M, Kumakiri J, Kuroda K, et al. Questionnaire analysis of recovery of activities of daily living after laparoscopic surgery. J Minim Invasive Gynecol 2008;15:169.
  • 10
    Vonk Noordegraaf A, Huirne JA, Brolmann HA, van Mechelen W, Anema JR. Multidisciplinary convalescence recommendations after gynaecological surgery: a modified Delphi method among experts. BJOG 2011;118:155767.
  • 11
    Naidu M, Sultan AH, Thakar R. Convalescence advice following gynaecological surgery. J Obstet Gynaecol 2012;32:5569.
  • 12
    Moller C, Ottesen M, Kehlet H, Ottesen BS. Convalescence recommendations after hysterectomy. A study of opinions among Danish physicians. Ugeskr Laeger 2001;163:70437.
  • 13
    Tolver MA, Strandfelt P, Forsberg G, Hjorne FP, Rosenberg J, Bisgaard T. Determinants of a short convalescence after laparoscopic transabdominal preperitoneal inguinal hernia repair. Surgery 2012;151:55663.
  • 14
    Steenstra IA, Verbeek JH, Heymans MW, Bongers PM. Prognostic factors for duration of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occup Environ Med 2005;62:85160.
  • 15
    Donders N, Bos J, van der Velden K, van der Gulden J. Age differences in the associations between sick leave and aspects of health, psychosocial workload and family life: a cross-sectional study. BMJ Open 2012;2.
  • 16
    Fan JK, McLeod CB, Koehoorn M. Sociodemographic, clinical, and work characteristics associated with return-to-work outcomes following surgery for work-related knee injury. Scand J Work Environ Health 2010;36:3328.
  • 17
    Heymans MW, Anema JR, van Buuren S, Knol DL, van Mechelen W, de Vet HCW. Return to work in a cohort of low back pain patients: development and validation of a clinical prediction rule. J Occup Rehabil 2009;19:15565.
  • 18
    Reme SE, Hagen EM, Eriksen HR. Expectations, perceptions, and physiotherapy predict prolonged sick leave in subacute low back pain. BMC Musculoskelet Disord 2009;10:139.
  • 19
    Kuijer W, Groothoff JW, Brouwer S, Geertzen JHB, Dijkstra PU. Prediction of sickness absence in patients with chronic low back pain: a systematic review. J Occup Rehabil 2006;16:43967.
  • 20
    Heymans MW, de Vet HCW, Knol DL, Bongers PM, Koes BW, van Mechelen W. Workers' beliefs and expectations affect return to work over 12 months. J Occup Rehabil 2006;16:68595.
  • 21
    Turner JA, Franklin G, Fulton-Kehoe D, Sheppard L, Wickizer TM, Wu R, et al. Worker recovery expectations and fear-avoidance predict work disability in a population-based workers' compensation back pain sample. Spine (Phila, PA 1976) 2006;31:6829.
  • 22
    Clay FJ, Newstead SV, Watson WL, Ozanne-Smith J, McClure RJ. Bio-psychosocial determinants of time lost from work following non life threatening acute orthopaedic trauma. BMC Musculoskelet Disord 2010;11:6.
  • 23
    Aaronson NK, Muller M, Cohen PD, Essink-Bot ML, Fekkes M, Sanderman R, et al. Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations. J Clin Epidemiol 1998;51:105568.
  • 24
    Ware JEJ, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:47383.
  • 25
    Kluivers KB, Hendriks JCM, Mol BWJ, Bongers MY, Vierhout ME, Brolmann HAM, et al. Clinimetric properties of 3 instruments measuring postoperative recovery in a gynecologic surgical population. Surgery 2008;144:1221.
  • 26
    Harrell FE. Regression Modeling Strategies with Applications to Linear Models, Logistic Regression, and Survival Analysis. New York: Springer, 2001.
  • 27
    Steyerberg EW. Clinical Prediction Models. A Practical Approach to Development, Validation, and Updating. Rotterdam: Springer, 2009.
  • 28
    Heymans MW, Ford JJ, McMeeken JM, Chan A, de Vet HCW, van Mechelen W. Exploring the contribution of patient-reported and clinician based variables for the prediction of low back work status. J Occup Rehabil 2007;17:38397.
  • 29
    Kent JT, O'Quigley J. Measures of dependence for censored survival data. Biometrika 1988;75:52534.
  • 30
    Baldwin ML, Butler RJ, Johnson WG, Cote P. Self-reported severity measures as predictors of return-to-work outcomes in occupational back pain. J Occup Rehabil 2007;17:683700.
  • 31
    Steyerberg EW, Moons KGM, van der Windt DA, Hayden JA, Perel P, Schroter S, et al. Prognosis Research Strategy (PROGRESS) 3: prognostic model research. PLoS Med 2013;10:e1001381.
  • 32
    Cho YJ, Kim ML, Lee SY, Lee HS, Kim JM, Joo KY. Laparoendoscopic single-site surgery (LESS) versus conventional laparoscopic surgery for adnexal preservation: a randomized controlled study. Int J Womens Health 2012;4:8591.
  • 33
    Jakobsen DH, Callesen T, Schouenborg L, Nielsen D, Kehlet H. Convalescence after laparoscopic sterilization. Ambul Surg 2003;10:959.
  • 34
    Miskry T, Magos A. Randomized, prospective, double-blind comparison of abdominal and vaginal hysterectomy in women without uterovaginal prolapse. Acta Obstet Gynecol Scand 2003;82:3518.
  • 35
    Wattiez A, Cohen SB, Selvaggi L. Laparoscopic hysterectomy. Curr Opin Obstet Gynecol 2002;14:41722.
  • 36
    Persson P, Kjolhede P. Factors associated with postoperative recovery after laparoscopic and abdominal hysterectomy. Eur J Obstet Gynecol Reprod Biol 2008;140:10813.
  • 37
    Brummer THI, Jalkanen J, Fraser J, Heikkinen AM, Kauko M, Makinen J, et al. FINHYST 2006–national prospective 1-year survey of 5,279 hysterectomies. Hum Reprod 2009;24:251522.
  • 38
    Hwang JL, Seow KM, Tsai YL, Huang LW, Hsieh BC, Lee C. Comparative study of vaginal, laparoscopically assisted vaginal and abdominal hysterectomies for uterine myoma larger than 6 cm in diameter or uterus weighing at least 450 g: a prospective randomized study. Acta Obstet Gynecol Scand 2002;81:11328.
  • 39
    Dietz V, van der Vaart CH, van der Graaf Y, Heintz P, Schraffordt Koops SE. One-year follow-up after sacrospinous hysteropexy and vaginal hysterectomy for uterine descent: a randomized study. Int Urogynecol J 2010;21:20916.
  • 40
    Sculpher M, Manca A, Abbott J, Fountain J, Mason S, Garry R. Cost effectiveness analysis of laparoscopic hysterectomy compared with standard hysterectomy: results from a randomised trial. BMJ 2004;328:134.
  • 41
    Hehenkamp WJ, Volkers NA, Birnie E, Reekers JA, Ankum WM. Pain and return to daily activities after uterine artery embolization and hysterectomy in the treatment of symptomatic uterine fibroids: results from the randomized EMMY trial. Cardiovasc Intervent Radiol 2006;29:17987.
  • 42
    Anema JR, Schellart AJM, Cassidy JD, Loisel P, Veerman TJ, van der Beek AJ. Can cross country differences in return-to-work after chronic occupational back pain be explained? An exploratory analysis on disability policies in a six country cohort study. J Occup Rehabil 2009;19:41926.
  • 43
    Loisel P, Durand MJ, Berthelette D, Vezina N, Baril R, Gagnon D. Disability prevention—new paradigm for the management of occupational back pain. Dis Manag Health Outcomes 2001;9:35160.
  • 44
    Dirksz AA, Bruinvels DJ, de Vet HCW, Brolmann HA. Werkhervatting na gynaecologische ingrepen; pilot study. 2008.
  • 45
    Sarmini OR, Lefholz K, Froeschke HP. A comparison of laparoscopic supracervical hysterectomy and total abdominal hysterectomy outcomes. J Minim Invasive Gynecol 2005;12:1214.
  • 46
    Roumm AR, Pizzi L, Goldfarb NI, Cohn H. Minimally invasive: minimally reimbursed? An examination of six laparoscopic surgical procedures. Surg Innov 2005;12:26187.
  • 47
    Anema JR, Van Der Giezen AM, Buijs PC, Van Mechelen W. Ineffective disability management by doctors is an obstacle for return-to-work: a cohort study on low back pain patients sicklisted for 3–4 months. Occup Environ Med 2002;59:72933.
  • 48
    Merrill RN, Pransky G, Hathaway J, Scott D. Illness and the workplace: a study of physicians and employers. J Fam Pract 1990;31:558.
  • 49
    Weevers HJ, van der Beek AJ, Anema JR, van der WG, van MW. Work-related disease in general practice: a systematic review. Fam Pract 2005;22:197204.
  • 50
    Lofgren A, Hagberg J, Alexanderson K. What physicians want to learn about sickness certification: analyses of questionnaire data from 4019 physicians. BMC Public Health 2010;10:61.
  • 51
    Ottesen M, Sorensen M, Kehlet H, Ottesen B. Short convalescence after vaginal prolapse surgery. Acta Obstet Gynecol Scand 2003;82:35966.
  • 52
    Ottesen M, Moller C, Kehlet H, Ottesen B. Substantial variability in postoperative treatment, and convalescence recommendations following vaginal repair. A nationwide questionnaire study. Acta Obstet Gynecol Scand 2001;80:10628.
  • 53
    van Well AM, van Ijsseldijk AL, Bonjer HJ, Vrijland WW, Brouwer KJ, de Boer K. Onnodig thuis; ziekteverzuim na liesbreukoperatie. Med Contact 2002;18:6969.
  • 54
    Callesen T, Klarskov B, Bech K, Kehlet H. Short convalescence after inguinal herniorrhaphy with standardised recommendations: duration and reasons for delayed return to work. Eur J Surg 1999;165:23641.
  • 55
    Bay-Nielsen M, Thomsen H, Andersen FH, Bendix JH, Sorensen OK, Skovgaard N, et al. Convalescence after inguinal herniorrhaphy. Br J Surg 2004;91:3627.
  • 56
    Vonk Noordegraaf A, Huirne J, Brolmann H, Emanuel M, van Kesteren P, Kleiverda G, et al. Effectiveness of a multidisciplinary care program on recovery and return to work of patients after gynaecological surgery; design of a randomized controlled trial. BMC Health Serv Res 2012;12:29.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
bjo12494-sup-0001-AppendixS1.pdfapplication/PDF52KAppendix S1. Baseline questionnaire of the recovery index.
bjo12494-sup-0002-AppendixS2.pdfapplication/PDF90KAppendix S2. Calculation of risk score of sick leave more than 6 weeks.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.