Technical efficiency evaluation of colorectal cancer care for older patients in Dutch hospitals.

BACKGROUND: Preoperative colorectal cancer care pathways for older patients show considerable practice variation between Dutch hospitals due to differences in interpretation and implementation of guideline-based recommendations. This study aims to report this practice variation in preoperative care between Dutch hospitals in terms of technical efficiency and identifying associated factors.
METHODS: Data on preoperative involvement of geriatricians, physical therapists and dieticians and the clinicians' judgement on prehabilitation implementation were collected using quality indicators and questionnaires among colorectal cancer surgeons and specialized nurses. These data were combined with registry-based data on postoperative outcomes obtained from the Dutch Surgical Colorectal Audit for patients aged ≥75 years. A two-stage data envelopment analysis (DEA) approach was used to calculate bias-corrected DEA technical efficiency scores, reflecting the extent to which a hospital invests in multidisciplinary preoperative care (input) in relation to postoperative outcomes (output). In the second stage, hospital care characteristics were used in a bootstrap truncated regression to explain variations in measured efficiency scores.
RESULTS: Data of 25 Dutch hospitals were analyzed. There was relevant practice variation in bias-corrected technical efficiency scores (ranging from 0.416 to 0.968) regarding preoperative colorectal cancer surgery. The average efficiency score of hospitals was significantly different from the efficient frontier (p = <0.001). After case-mix correction, higher technical efficiency was associated with larger practice size (p = <0.001), surgery performed in a general hospital versus a university hospital (p = <0.001) and implementation of prehabilitation (p = <0.001).
CONCLUSION: This study showed considerable variation in technical efficiency of preoperative colorectal cancer care for older patients as provided by Dutch hospitals. In addition to higher technical efficiency in high-volume hospitals and general hospitals, offering a care pathway that includes prehabilitation was positively related to technical efficiency of hospitals offering colorectal cancer care.

Introduction

With nearly 60% of patients aged above 70 years and more than 35% of patients aged 75 years and over, colorectal cancer is predominantly a disease of older adults [1]. As multimorbidity is commonly prevalent from the age of 70 years and multimorbidity is associated with more postoperative complications, older patients are more prone to postoperative complications and mortality [1-3].

Regardless of age or number of co-morbid conditions, surgery is the cornerstone for curative treatment in patients with colorectal cancer [4]. To maximize treatment outcomes, it is therefore important to optimize resilience in this population to withstand colorectal surgery as a stressful event. Recognition of risk factors by a multidisciplinary team approach could help identify patients at high-risk for developing postoperative complications [5]. Especially in frail older patients a comprehensive geriatric assessment (CGA), a multidimension evaluation to identify medical, psychosocial, and functional limitations of a frail older patient, is helpful to make a tailored treatment plan taken into account goals and wishes of individual patients [6].

After preoperative risk stratification, employing interventions to prepare a patient for surgery (also called “prehabilitation”) including physical therapy and nutritional assessment, could counteracting the complication risks by enhancing resilience and functional capacity [7]. Prehabilitation can be unimodal, focusing solely on for instance exercise, or multimodal including physical exercise, nutrition assessment as well as psychological stress reduction [8]. Prehabilitation has shown promising results, especially for patients at greatest risk of poor postoperative outcomes [9, 10]. However, other studies did not show significant improvement of postoperative outcomes after prehabilitation in colorectal cancer surgery [11, 12].

Currently, Dutch guidelines recommend screening on frailty and geriatric assessment in case of frailty in colorectal cancer patients aged ≥ 70 years. However, quality indicators of the Dutch Health and Youth Care Inspectorate (part of the Ministry of Health, Welfare and Sport) show that screening on frailty and geriatric assessment in case of frailty is not yet completely implemented in all Dutch hospitals [13]. On the other hand, despite level II evidence implied that multimodal prehabilitation improves postoperative outcomes [14], prehabilitation is still restricted to research settings in the Netherlands because data supporting efficiency is contradictory [15, 16]. Meanwhile, several forms of prehabilitation programs have been started in a number of Dutch hospitals [17]. Therefore, colorectal care pathways show considerable practice variation between Dutch hospitals.

This study will detail this practice variation, in terms of technical efficiency, in preoperative colorectal cancer care for older patients in Dutch hospitals. Technical efficiency is defined as the extent to which a hospital invests in multidisciplinary preoperative care in relation to its outputs in terms of postoperative complications. Further, this study will identify factors associated with the variation in the technical efficiency of this preoperative care.

Materials and methods

Data collection

This observational study was based on retrospective data of perioperative care given to patients with colorectal cancer of 75 years and above in Dutch hospitals. Hospital data for the year 2017 and 2018 were used. In total, 56 Dutch hospitals with an active practice of colorectal cancer surgery (about 70% of all hospitals), were approached to participate in this study.

This study combined three data sources (S1 Table) to investigate the variation in technical efficiency and its potential drivers: 1. Data on preoperative involvement of physical therapists and dieticians were collected using questionnaires (S1 File) among colorectal cancer surgeons and specialized nurses. They were asked if physical therapists and dieticians were involved in the preoperative period (yes/ no/ by indication). Additionally, clinicians’ judgement on prehabilitation implementation (yes/ no/ by indication) was asked. 2. Data on preoperative involvement of geriatricians were collected using the quality indicators published by the Health and Youth Care Inspectorate. These quality indicators represent, on hospital level, the percentage of frail older patients (≥70 years) which is assessed by a geriatrician [13]. It was assumed that this percentage was a representative reflection of geriatric involvement in patients aged 75 years or over. 3. Data on postoperative complications obtained from the Dutch Surgical Colorectal Audit [18]. Only postoperative data of patients with elective surgery were included, as multidisciplinary preoperative interventions are not applicable in case of urgency or emergency surgery. Patients with hyperthermic intraperitoneal chemotherapy (HIPEC) or intraoperative radiation therapy (IORT) were excluded for this analysis.

The multiple datasets were merged to make a dataset that registered all hospital-related health services delivered to these patients.

The institutional review board (IRB) CMO Region Arnhem-Nijmegen, NL55712.091.16 (file number 2018–4163) advised that this study doesn’t fall within the remit of the Medical Research Involving Human Subjects Act (WMO).

Outcomes

The primary outcome variables of the study were the scores of technical efficiency for each individual hospital and the comparisons between this scores. The technical efficiency score was defined as the extent to which a hospital invests in multidisciplinary preoperative care (input) in relation to its outputs in terms of postoperative complications.

The secondary outcome variable was the relationship between hospital technical efficiency and quality performance and the factors affecting this relationship.

Statistical analysis

The analysis on this collapsed dataset consisted of a two-stage data envelopment approach (DEA). DEA is a non-parametric technique based on linear programming that allows for the construction of the most efficient production frontier based on the inputs and outputs of the decision-making units (DMUs: these are the hospitals delivering colorectal cancer surgery care). In other words, this technical efficiency frontier reflects the graphical line that can be constructed when connecting the DMUs that use the least amount of inputs to produce one unit of output (input-oriented DEA) or that produces the most amount of outputs with one unit of input (output-oriented DEA).

The relative technical (in)efficiency, the difference between the DEA score and the efficient frontier is calculated in the first stage of the DEA by comparing its inputs and outputs for each DMU in relation to the rest of the DMUs, i.e. hospitals. In this study input was defined as the average costs of geriatrician, physical therapist and dietician involvement with a patient who is scheduled for colorectal cancer surgery in a period between setting operation indication and admission to the hospital because of tumor resection. Output was defined as the percentages of severe complications in each hospital. A severe complication was defined as a complication within 90 days after resection with serious consequences: leading to mortality, a surgical reintervention (operative or percutaneous), a postoperative hospital stay of at least 14 days or readmission. As lower values of severe complications represent better quality of care, and DEA usually assumes that more outputs contribute to higher technical efficiency, the percentages of no severe complications were used in the DEA analysis.

To explain differences in technical efficiency scores, the second stage of the DEA comprised a bootstrapped truncated regression analysis where estimated technical efficiency scores were regressed on a set of preselected case-mix adjusting and explanatory variables. Preselected case-mix variables were ASA score, tumor stage and tumor localization. Selected explanatory (independent) variables were hospital volume, hospital teaching status and clinicians’ judgement on the implementation of prehabilitation. The complete approach is described in more detail in S2 File.

Data on population and hospital level are presented as mean (standard deviation [SD]) or number (%) when indicated. Comparisons between means and between DEA scores were done using Student’s t-test for numerical variables. Comparisons between categorical variables were done using Chi-Square test. A P-value of <0.05 was considered statistically significant.

The entire analysis, i.e. Simar & Wilson approach, was carried out using STATA version 15.1.

Results

In total 56 of the 79 hospitals (71%) conducting colorectal cancer surgery were approached. 25 out of this 56 hospitals (45%) provided sufficient information (meaning at least one questionnaire completed by surgeon or specialized nurse and available data on postoperative complications) and were taken into analysis (S2–S4 Tables). In these 25 hospitals a total of 2470 elective colorectal cancer patients of 75 years and older underwent surgery. These patients comprised 39% of the total group of patients (n = 6349) who were treated in the Netherlands during this period.Mean age of patients in the participating hospitals ranged from 78 to 81 years (Table 1).

Table 1

Descriptive data on population level.

	Patients (n = 2470) from hospitals included (n = 25)				Patients (n = 3879) from hospitals not included (n = 31)	Total (n = 6349)
	University hospitals (n = 148)	General hospitals (n = 2322)	Total (n = 2470)	P-value
Hospital size a	30 (10)	116 (42)	99 (52)	<0.001	75 (42)	82 (46)
Age	79 (4)	80 (4)	80(4)	0.001	80 (4)	80 (4)
Sex				0.034
Male	91 (61%)	1219 (52%)	1310 (53%)		2073 (53%)	3383 (53%)
Female	57 (39%)	1103 (48%)	1160 (47%)		1806 (47%)	2966 (47%)
ASA ≥ 3	58 (39%)	1040 (45%)	1098 (44%)	0.184	1583 (41%)	2681 (42%)
Cancer type				0.011
Colon	95 (64%)	1712 (74%)	1807 (73%)		2895 (75%)	4702 (74%)
Rectum	53 (36%)	610 (26%)	663 (27%)		984 (25%)	1647 (26%)
Tumor stage IV	21 (14%)	120 (5%)	141 (6%)	<0.001	186 (5%)	327 (5%)
Severe complications	41 (28%)	552 (24%)	593 (24%)	0.278	952 (25%)	1545 (24%)

Data are n (%) or mean (SD).

Average number of treated patients in each hospital in 2017–2018.

The aggregated descriptive data on hospital level used to calculate uncorrected and bias-corrected (bootstrapped) DEA scores is shown in Table 2.

Table 2

Hospital characteristics representing input and output for DEA score.

Hospital number	Hospital size (n)	University hospitala	Prehabilitationb	ASA ≥ 3 (%)	Colon cancer (%)	Tumor stage IV (%)	Psychical therapistb	Dieticianb	Geriatricianc	Severe complications (%)	Uncorrected DEA scored	Bias-corrected DEA scoree
1	14	Y	I	4 (29%)	8 (57%)	2 (14%)	I	I	1	4 (29%)	.962	.904
2	27	Y	I	15 (56%)	15 (56%)	6 (22%)	I	I	2	8 (30%)	.822	.789
3	31	Y	N	14 (45%)	17 (55%)	7 (23%)	I	I	3	12 (39%)	.678	.623
4	35	Y	N	10 (29%)	21 (60%)	3 (9%)	I	I	3	8 (23%)	.852	.784
5	41	Y	N	15 (37%)	29 (71%)	3 (7%)	I	I	3	9 (22%)	.842	.774
6	41	N	Y	24 (59%)	30 (73%)	2 (5%)	Y	Y	2	8 (20%)	.757	.700
7	46	N	N	7 (15%)	32 (70%)	1 (2%)	Y	Y	3	15 (33%)	.443	.416
8	64	N	N	29 (45%)	47 (73%)	5 (8%)	Y	Y	2	26 (41%)	.590	.548
9	70	N	Y	49 (70%)	46 (66%)	6 (9%)	Y	N	1	18 (26%)	1	.768
10	88	N	I	33 (38%)	54 (61%)	4 (5%)	I	I	1	24 (27%)	1	.914
11	98	N	I	35 (36%)	74 (76%)	11 (11%)	I	I	3	20 (20%)	.859	.775
12	99	N	Y	39 (39%)	70 (71%)	4 (4%)	N	Y	2	20 (20%)	1	.923
13	107	N	Y	56 (52%)	81 (76%)	2 (2%)	Y	I	1	22 (21%)	1	.901
14	112	N	I	49 (44%)	90 (80%)	2 (2%)	I	I	2	20 (18%)	.984	.954
15	116	N	I	59 (51%)	83 (72%)	8 (7%)	I	Y	2	28 (24%)	.841	.774
16	121	N	I	56 (46%)	78 (65%)	9 (7%)	Y	I	3	23 (19%)	.798	.719
17	124	N	N	56 (45%)	86 (70%)	4 (3%)	N	I	3	41 (33%)	.843	.778
18	127	N	Y	56 (44%)	91 (72%)	5 (4%)	N	I	2	36 (28%)	.902	.829
19	135	N	Y	54 (40%)	103 (76%)	7 (5%)	I	I	1	34 (25%)	.988	.945
20	136	N	N	61 (45%)	99 (73%)	5 (4%)	N	I	1	28 (21%)	1	.880
21	140	N	N	77 (55%)	101 (72%)	2 (1%)	N	N	3	37 (26%)	1	.717
22	145	N	Y	74 (51%)	94 (65%)	8 (6%)	Y	Y	1	42 (29%)	.857	.810
23	163	N	I	45 (28%)	105 (64%)	16 (10%)	I	I	2	26 (16%)	1	.968
24	187	N	N	50 (27%)	131 (70%)	8 (4%)	I	I	3	40 (21%)	.848	.785
25	203	N	I	121 (60%)	128 (63%)	8 (4%)	I	I	1	44 (22%)	.988	.963

a N = no, Y = yes.

b N = no, I = by indication, Y = yes.

c 1 = involvement in < 15% of patients, 2 = involvement in 15–25% of patients, 3 = involvement in >25% of patients.

d Classical non-parametric DEA analysis.

e Bootstrapped (semi-parametric) DEA analysis.

The classical DEA analysis showed that 7 hospitals (28%) were lying on the technical efficiency frontier (uncorrected DEA score = 1) and served therefore as benchmarks for the other hospitals (Table 2). The large distribution of technical efficiency scores (0.443–1), based on both variation in postoperative outcomes (output) as well as preoperative involvement of physical therapists, dieticians and geriatricians (input), is depicted in Fig 1.

Fig 1

Distribution of hospital uncorrected (non-parametric) DEA scores in proportion to the percentage of severe complications (output).

Mean bias-corrected DEA score from the bootstrapped DEA analysis was 0.798 (range 0.416 to 0.968), and differed significantly from the technical efficiency frontier (p = <0.001, 95%CI -0.257, -0.147).

Table 3 shows the results of the truncated regression analysis. After case-mix correction the higher bias-corrected technical efficiency was significantly associated with larger practice size (β = 0.003, p = <0.001, 95%CI 0.002,0.003) and surgery performed in a general hospital versus a university hospital (β = 0.402, p = <0.001, 95%CI 0.234,0.557). Additionally, clinicians’ judgement on the implementation level of prehabilitation was positively associated with hospital preoperative colorectal cancer pathway technical efficiency. Implementation of prehabilitation by indication or as usual care both showed better technical efficiency than no prehabilitation at all (respectively, β = 0.209, p = <0.001, 95%CI 0.127,0.292 and β = 0.187, p = <0.001, 95%CI 0.094,0.282).

Table 3

Results of truncated regression analysis.

	Coefficient (β)a	P-value	95% confidence interval
Tumor stage IV	-0.012	0.037	-0.023,-0.002
ASA ≥ 3	0.000	0.947	-0.003,0.003
Tumor localization, colon	0.004	0.262	-0.003,0.012
Practice size	0.003	<0.001	0.002,0.003
General hospital	0.402	<0.001	0.234,0.557
Prehabilitation (reference no prehabilitation)
By indication	0.209	<0.001	0.127,0.292
Yes	0.187	<0.001	0.094,0.282

a Positive coefficient represents increase in efficiency score.

Discussion

This study studied technical efficiency of preoperative colorectal cancer care, based on preoperative involvement of physical therapists, dieticians and geriatricians as input, and severe postoperative complications as output. Across Dutch hospitals, considerable technical inefficiencies were shown in this study, as only 7 of the 25 hospitals were on the efficient frontier. In regression analysis, practice size, being a general hospital and implementation of prehabilitation (both by indication or as usual care) were significantly positively associated with technical efficiency.

As far as we know, this is the first study that analyzed the influence of both hospital volume and hospital teaching status on technical efficiency scores in preoperative colorectal cancer care for older patients. Previous studies already described the association between hospital volume and postoperative outcomes in colorectal cancer care as such, with reduced postoperative mortality in high volume hospitals [19-21]. Kolfschoten et al. found that this association between hospital volume and postoperative outcomes was mostly explained by high-volume hospitals treating patients with less comorbid diseases, lower ASA-classification and less often in an urgent or acute setting [22]. In this study only elective surgery was included and this study corrected for ASA-classification, therefore it is likely that other factors play a role in the variance in technical efficiency. An explanation could be that high-volume hospitals use more frequently selection criteria to identify only those patients that require multidisciplinary preoperative care. This hypothesis is supported by the fact that the hospital with the lowest bias-corrected DEA score was a relatively low-volume general hospital, with involvement of both physical therapist and dietician on a regular basis and involvement of a geriatrician in a relatively high number of patients. In contrast, in the hospitals with the highest technical efficiency both physical therapist and dietician were only involved by indication and the percentage of patients in which a geriatrician was involved was nearly a third in comparison with the hospital with the lowest bias-corrected DEA score.

The association between hospital teaching status and serious complication rates, with higher serious complication rates in university hospitals compared to general hospitals, was previously described by van Groningen et al. [23]. They also found that hospital volume could not explain these differences and that there was a considerable hospital variation. This hospital variation is also seen in the present study, with one of the five university hospitals represented in the top ten of most technical efficient hospitals, while another of the university hospitals had nearly the lowest bias-corrected DEA score. As this study did only correct for ASA-score and tumor stage, it is possible that the association between technical efficiency and hospital teaching status is based on differences in case mix such as location of the tumor, type of resection or gender [24]. It is likely that more complex operations, with higher postoperative complications rates, will take place in university care settings, which might also biased the results [22].

The newly found association between technical efficiency and the clinicians’ judgement on the implementation of prehabilitation may be explained by the beneficial effects of prehabilitation on postoperative outcomes as shown in the literature [25]. Additionally, it is likely that implementation of prehabilitation in a care pathway ensures that preoperative care is more structured, with probably better selection criteria of patients that need multidisciplinary preoperative care. Though, the association between clinicians’ judgement on prehabilitation can also be reversed; whereby hospitals with a higher technical efficiency are more aware of (the importance of) local care pathways and therefore gave a positive answer to the question whether or not prehabilitation was implemented.

This study is the first that investigates practice variation between Dutch hospitals in terms of technical efficiency in preoperative colorectal cancer care.

This study has several limitations. Firstly, only 25 of the 79 Dutch hospitals conducting colorectal cancer surgery were included. However, the patients treated in the included hospitals were a representative patient population. Above, quantitative data on involvement of physical therapists and dieticians was not available and this study was based on subjective clinicians’ judgement. In some cases, there was discrepancy between the answers of the surgeon and the specialized nurse inside the hospital. To deal with these incongruities it was chosen to select the highest value of involvement, as it was assumed that the highest value came from consultants ordering the physical therapists or dieticians involvement. Additionally, assumptions were made about the degree of involvement of the physical therapist and dietician in the participating hospitals, because it was not possible to quantify this for each participating hospital. It was also assumed that data on preoperative involvement of geriatricians in patients aged 70 years or over, collected using the quality indicators published by the Health and Youth Care Inspectorate, was a representative reflection of geriatric involvement in patients aged 75 years or over. It is possible that for some hospitals an over- or underestimation of the real involvement was made based on this assumptions. Next, this study did not correct for all potential case-mix factors in the regression analysis. A pre-specified set of case-mix variables was chosen based on relevancy, availability and clinical consensus. However, as mentioned in previous studies, variables as gender and comorbidity levels are associated with postoperative outcomes too [22, 26]. Nevertheless, it is questionable whether or not these case-mix factors influences the clinicians’ judgement on prehabilitation implementation.

In conclusion, this study showed high technical efficiency variation in colorectal cancer care for older patients between Dutch hospitals. In addition to higher technical efficiency in high-volume hospitals and general hospitals, a care pathway including prehabilitation seemed to be positively related to technical efficiency meaning that hospitals that implemented prehabilitation, by indication or as usual care, are technical efficiency benchmarks for the those hospitals that did not. Further prospective research should therefore focus on cost-effectiveness of prehabilitation and selection criteria for patients that benefit the most from prehabilitation.

Questionnaire.

(DOCX)

Click here for additional data file.

Detailed statistical analysis.

(DOCX)

Click here for additional data file.

Overview of datasources.

(DOCX)

Click here for additional data file.

Questionnaire responses.

(DOCX)

Click here for additional data file.

Detailed hospital characteristics.

(DOCX)

Click here for additional data file.

Detailed surgical procedures.

(DOCX)

Click here for additional data file.

23 Jun 2021

PONE-D-20-38041

Technical efficiency evaluation of colorectal cancer care for older patients in Dutch hospitals

PLOS ONE

Dear Dr. Zonneveld-Heil,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Aug 07 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Alaa El-Hussuna

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information."

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 3 in your text; if accepted, production will need this reference to link the reader to the Table.

5. One of the noted authors is a group or consortium [PRECOLO consortium]. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper looks at the impact of prehabilitation for colorectal cancer therapy on outcomes for Colorectal cancer across Dutch hospitals through the involvement of geriatricians, physiotherapy and dieticians. Administrative data from the Dutch Surgical Colorectal Audit was used. A 2-stage data envelopment analysis (DEA) approach was used. 25 hospital were included between 2017-2018. Findings were that higher technical efficiency was associated with larger practices, non-university hospitals, and the use of prehabilitation. This is an interesting and important study highlighting the need for good preoperative therapy for patients undergoing surgery for colorectal cancer. However, the study does not account for a multitude of factors that do into a successful patient outcome.

Major Issues

• Were colon and rectal cancer treatment done differently? Were rectal cancers which required neoadjuvant treatment also considered as separate. Neoadjuvant has major implications for prehabilitation as the lead time to surgery is longer. Similarly, these patients may be more deconditioned heading into their operation.

• I am still not entirely sure what prehabilitation is beyond a physiotherapy consults and a geriatric consults. It seems that consideration was not made for the type of prehabilitation but rather whether a visit had occurred. The true measure of the impact of prehabilitation is difficult to discern and as such, conclusions drawn from the study are challenging to interpret. Can you please provide an overview as to what prehabilitation includes?

• I suspect that there is a heterogeneity of surgical interventions. For example, older patients may be offered limited resections for colorectal cancer with less aggressive lymphadenectomies or even transanal excisions (TEM). Were attempts made to correct for this?

• If outcomes were complications, was any consideration given for accounting for CCI scores?

• Can you explain what the ”technical efficiency frontier” is?

Minor Issues

• Can the authors clarify which patients are offered the frailty and comprehensive geriatric assessments preoperatively. Is it age over age 70?

• In stage 1 of the methods, can you please clarify what “by indication” means and how does the division of physiotherapy become 5% vs Dietician was 40%

Reviewer #2: Thank you for inviting me to review this interesting study. The authors collected data from different sources to investigate the technical efficiency in elderly patients with colorectal cancer in Dutch hospitals. They found considerable variation in technical efficiency amongst hospitals and reported that high-volume hospitals and general hospitals offering a care pathway that includes prehabilitation had better postoperative outcome. I have several comments on the manuscript as outlines below.

Introduction

“On the other hand, as there is still a lack of data supporting efficiency, prehabilitation is generally restricted to research settings”, this is not entirely true as level II evidence implied that multimodal prehabilitation improves outcomes after surgery for abdominal cancer in general (Front. Surg., 19 March 2021 | https://doi.org/10.3389/fsurg.2021.628848).

Methods

• Please state whether the data used in the study were prospectively collected.

• Consider adding a diagram entailing three data sources used in the study.

• Regarding data source#1, how was the questionnaires sent to colorectal surgeons and nurses? How many surgeons and how many nurses were invited and how many accepted the invitation? This needs to be clarified in the results section later.

• You mention postoperative complications as the study output, however; you do not include how this outcome was defined and graded (e.g. Clavien-Dindo classification). Also, were all complications included as equal, as simple wound infection was counted equivalent to major anastomotic leak or pulmonary embolism?

• You defined technical efficiency in the introduction; this definition needs to be more clarified and moved to the methods section since this is the primary study outcome.

• Add a section detailing the primary and secondary study outcomes

• How was missing data dealt with? This needs to be addressed.

• The statistical analysis section is too long and needs to be written in a more concise manner.

Results

• Table 1 illustrates the detail of the cohort used in the study and compared University and General hospitals. It would be helpful to add the p value for each comparison to clarify whether the differences were significant.

• The coefficients of the regression analysis should be added to the text, just before the 95% confidence interval.

• Since you used a regression analysis, please report its accuracy and discriminatory ability by reporting the area under the curve.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Sameh Hany Emile

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Aug 2021

Dear Editor-in-Chief,

Enclosed please find our revised manuscript PONE-D-20-38041 “Technical efficiency evaluation of colorectal cancer care for older patients in Dutch hospitals”. We thank the reviewers for their comments, and below we describe how we have addressed each one.

Journal Requirements:

Based on the journal requirements, we have made the following adjustments:

(1) Additional information regarding questionnaire

We have developed a survey and included S2 File, containing an overview of the questionnaire in Dutch and English.

(2) Data avalaibility

We have included the answers to the questionnaires and more detailed hospital characteristics, including surgical procedures, in S2 - S4 Tables.

The publicly available quality indicators published by the Dutch Health and Youth Care Inspectorate are available from the website of the Dutch Health and Youth Care Inspectorate: https://www.dhd.nl/producten-diensten/omniq/Paginas/Databestanden-Basisset-MSZ.aspx. Because this data is publicly available and contains identifying information, it is not possible to link this data directly to our data (website is added to S1 Table).

Data on postoperative complications are obtained from the Dutch Surgical Colorectal Audit and are limited available on request against payment: https://dica.nl/dcra/onderzoek (website is added to S1 Table).

(3) Reference Table 3

The reference has been added.

(4) PRECOLO consortium

We added the affiliations of the individual authors.

Lead author for this group: M.G.M. Olde Rikkert, Marcel.OldeRikkert@radboudumc.nl

Reviewer #1:

(1) Were colon and rectal cancer treatment done differently? Were rectal cancers which required neoadjuvant treatment also considered as separate. Neoadjuvant has major implications for prehabilitation as the lead time to surgery is longer. Similarly, these patients may be more deconditioned heading into their operation.

We understand the reviewer’s concern regarding the differences between treatment for colon and rectal cancer and the implications of neoadjuvant treatment for both prehabilitation and postoperative outcomes. To take into account the possible differences in colon versus rectum cancer treatment and associated postoperative outcomes, we selected tumor localisation as case-mix factor to adjust for in the bootstrapped truncated regression analysis (second stage of the analysis). As a result, independent of tumor localisation, the association between technical efficiency and the three explanatory variables (practice size, general versus academic hospital and implementation of prehabilitation) persists.

We have not considered to correct for neoadjuvant treatment. We performed a restriction in the number of case-mix factors due to limited power to do so, because of the low number of eligible patients in some of the participating hospitals. However, we included S3 and S4 Tables with additional information on i.a. neoadjuvant treatment.

(2) I am still not entirely sure what prehabilitation is beyond a physiotherapy consults and a geriatric consults. It seems that consideration was not made for the type of prehabilitation but rather whether a visit had occurred. The true measure of the impact of prehabilitation is difficult to discern and as such, conclusions drawn from the study are challenging to interpret. Can you please provide an overview as to what prehabilitation includes?

In this study we asked the clinicians’ judgement on prehabilitation implementation (explanatory variable second stage of the analysis) separately from the actual involvement of the physiotherapist, dietician and geriatrician (input for production function DEA, stage one). This difference becomes clear in table 1: hospitals that involve the physiotherapist/dietician/geriatrician in the preoperative phase have not always implemented prehabilitation and vice versa.

When asking whether or not prehabilitation was applied in the hospital, no criteria of prehabilitation and what it exactly should entails were given. So it is purely the clinicians’ judgement if prehabilitation is implemented or not.

(3) I suspect that there is a heterogeneity of surgical interventions. For example, older patients may be offered limited resections for colorectal cancer with less aggressive lymphadenectomies or even transanal excisions (TEM). Were attempts made to correct for this?

We agree that there is a heterogeneity of surgical interventions and that the surgery type is associated with postoperative complications. However, we performed a restriction in the number of case-mix factors due to the low number of eligible patients in some of the participating hospitals and have therefore chosen to only correct for tumor stage and not for resection type.

Based on this comment, we included a S4 Table with detailed information on surgical procedures.

(4) If outcomes were complications, was any consideration given for accounting for CCI scores?

As previously mentioned, we performed a restriction in the number of case-mix factors due to the low number of eligible patients in some of the participating hospitals. Because the Charlson Comorbidity Index is only a constellation of diseases and age without taking into consideration the severity of a disease, we have chosen to correct for ASA instead of CCI. Moreoevr, ASA classification is very common to be used pre and post-surgical procedures, which allows for comparisons with other studies. This selection is also supported by literature, showing that, although CCI had a similar predictive value for 30-day mortality and prolonged length of stay after colorectal cancer surgery, the only predictive comorbidity measure for the occurrence of post-operative surgical complications was ASA score.(1)

In S3 Table we also included information on the CCI.

(5) Can you explain what the ”technical efficiency frontier” is?

We have added a brief explanation of the “technical efficiency frontier” in the statistical analysis section:

DEA is a non-parametric technique based on linear programming that allows for the construction of the most efficient production frontier based on the inputs and outputs of the decision-making units (DMUs: these are the hospitals delivering colorectal cancer surgery care). In other words, this technical efficiency frontier reflects the graphical line that can be constructed when connecting the DMUs that use the least amount of inputs to produce one unit of output (input-oriented DEA) or that produces the most amount of outputs with one unit of input (output-oriented DEA). The relative technical (in)efficiency, the difference between the DEA score and the efficient frontier is calculated by comparing its inputs and outputs for each DMU in relation to the rest of the DMUs, i.e. hospitals. DEA was chosen in this study because it can deal with multiple inputs and outputs and needs no assumptions about the distribution between outputs and inputs

(6) Can the authors clarify which patients are offered the frailty and comprehensive geriatric assessments preoperatively. Is it age over age 70?

We have made the following clarification in the introduction section:

Currently, Dutch guidelines recommend screening on frailty and geriatric assessment in case of frailty in colorectal cancer patients aged ≥70 years.

(7) In stage 1 of the methods, can you please clarify what “by indication” means and how does the division of physiotherapy become 5% vs Dietician was 40%

We have made the following clarification in the method section:

To indicate involvement, a surgeon and/or a specialized nurse of each participating hospital indicated involvement of physical therapists and dieticians by yes (100%), no (0%) or by indication (only in a selection of patients for whom involvement was thought to be useful). Because it was impossible to quantify the level of involvement in each individual hospital, the involvement ‘by indication’ was quantified based on clinical data of the Radboud University Medical Center (Nijmegen, the Netherlands). Between 2017-2018 the physical therapist was involved in 5% of the cases in the Radboud University Medical Center and the dietician in 40% of the cases.

Please note:

- Based to comment (10) of reviewer 2, we enclosed this information in S2 File.

- The use of figures of the Radboud University Medical Center for other DMU’s, is mentioned as a limitation in the discussion.

Reviewer #2:

(1) “On the other hand, as there is still a lack of data supporting efficiency, prehabilitation is generally restricted to research settings”, this is not entirely true as level II evidence implied that multimodal prehabilitation improves outcomes after surgery for abdominal cancer in general (Front. Surg., 19 March 2021 | https://doi.org/10.3389/fsurg.2021.628848).

This has been rephrased as:

On the other hand, despite level II evidence implied that multimodal prehabilitation improves postoperative outcomes(2), prehabilitation is still restricted to research settings in the Netherlands because data supporting efficiency is contradictory.

(2) Please state whether the data used in the study were prospectively collected.

We have made the following clarification in the method section:

This observational study was based on retrospective data of perioperative care given to patients with colorectal cancer of 75 years and above in Dutch hospitals.

(3) Consider adding a diagram entailing three data sources used in the study.

We have taken on board this suggestion and added S1 Table.

(4) Regarding data source#1, how was the questionnaires sent to colorectal surgeons and nurses?

Surveys were sent by e-mail via Castors EDC.(3)

This information is also included in S1 Table.

(5) How many surgeons and how many nurses were invited and how many accepted the invitation? This needs to be clarified in the results section later.

We agree that it is important to be transparent about this and therefore we have added this information to the result section by rephrasing the first sentence of this section:

In total 56 of the 79 hospitals (71%) conducting colorectal cancer surgery were approached. 25 out of this 56 hospitals (45%) provided sufficient information (meaning at least one questionnaire completed by surgeon or specialized nurse and available data on postoperative complications) and were taken into analysis.

(6) You mention postoperative complications as the study output, however; you do not include how this outcome was defined and graded (e.g. Clavien-Dindo classification). Also, were all complications included as equal, as simple wound infection was counted equivalent to major anastomotic leak or pulmonary embolism?

The sentence: “A severe complication was defined as a complication within 90 days after resection with serious consequences: leading to mortality, a surgical reintervention (operative or percutaneous), a postoperative hospital stay of at least 14 days or readmission.” may have caused some confusion. Let us clarify this. The Dutch Surgical Colorectal Audit collects both information on the nature as well as the severity of postoperative complications. As mentioned by the reviewer, not all complications can be counted equivalent. To be able to calculate an overall outcome measure, independent from the nature of the complications, we used only information on the severity of complications. As described in the manuscript, only complications with serious consequences (leading to mortality, a surgical reintervention (operative or percutaneous), a postoperative hospital stay of at least 14 days) or readmission were registered. A simple wound infection was therefore not taken into analysis.

(7) You defined technical efficiency in the introduction; this definition needs to be more clarified and moved to the methods section since this is the primary study outcome.

As written in our response to comment (5) of reviewer 1, we have added a brief explanation of the “technical efficiency frontier” in the statistical analysis section.

(8) Add a section detailing the primary and secondary study outcomes

We added a section detailing the outcomes.

The primary outcome variable of the study was the score of technical efficiency for each individual hospital and the comparisons between this scores. The technical efficiency score was defined as the extent to which a hospital invests in multidisciplinary preoperative care (input) in relation to its outputs in terms of postoperative complications.

The secondary outcome variable was the relationship between hospital technical efficiency and quality performance and the factors affecting this relationship.

(9) How was missing data dealt with? This needs to be addressed.

There was only missing data for the questionnaires (S2 Table). If only the surgeon or the specialized nurse completed the questionnaire after multiple reminders by e-mail, only one questionnaire was used for analysis. This is addressed as limitation in the discussion.

(10) The statistical analysis section is too long and needs to be written in a more concise manner

We rephrased the statistical analysis section and added S2 File for additional information on the statistical analysis.

(11) Table 1 illustrates the detail of the cohort used in the study and compared University and General hospitals. It would be helpful to add the p value for each comparison to clarify whether the differences were significant.

We have included p-values for each comparison between University and General hospitals in table 1 and explained the statistical analysis in the method section.

(12) The coefficients of the regression analysis should be added to the text, just before the 95% confidence interval.

We agree and have changed this in the result section.

(13) Since you used a regression analysis, please report its accuracy and discriminatory ability by reporting the area under the curve.

As this is not a study on diagnostic performance we cannot add accuracy, discriminatory ability nor AUC. We did report a secondary multivariable analysis to identify relevant drivers of efficiency, but these cannot be considered diagnostic variables.

We hope to have addressed your comments satisfactorily and thank you once more for reviewing our manuscript.

Yours sincerely,

Thea Zonneveld-Heil, MSc

References

1. Dekker JWT, Gooiker GA, van der Geest LGM, Kolfschoten NE, Struikmans H, Putter H, et al. Use of different comorbidity scores for risk-adjustment in the evaluation of quality of colorectal cancer surgery: Does it matter? European Journal of Surgical Oncology (EJSO). 2012;38(11):1071-8.

2. Waterland JL, McCourt O, Edbrooke L, Granger CL, Ismail H, Riedel B, et al. Efficacy of Prehabilitation Including Exercise on Postoperative Outcomes Following Abdominal Cancer Surgery: A Systematic Review and Meta-Analysis. Frontiers in Surgery. 2021;8(55).

3. Castor EDC. Castor Electronic Data Capture 2019 [27 Aug. 2019]. Available from: https://castoredc.com.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Nov 2021

Technical efficiency evaluation of colorectal cancer care for older patients in Dutch hospitals

PONE-D-20-38041R1

Dear Dr. Zonneveld-Hell

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Alaa El-Hussuna

Academic Editor

PLOS ONE

10 Dec 2021

PONE-D-20-38041R1

Technical efficiency evaluation of colorectal cancer care for older patients in Dutch hospitals

Dear Dr. Heil:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Alaa El-Hussuna

Academic Editor

PLOS ONE