Dialogue support for spine surgery – based on Swespine, the Swedish Spine Register

Basic information
Back-specific information

Subgroup

Diagnosis group

Operated levels

1 2 3 4 5+

Clinical department type (only length of stay)

Sociodemographics

Age

Sociodemographics

Age

Gender

Unemployed

Disability pension

Retirement pension

Health profile

Smoker

Previous spine surgery

Quality of life (EQ-5D)

Comorbidity*

Results

Proportion With Successful Outcome = Pain Completely Resolved Or Significantly Improved After One Year (Green And Light Green)

DESCRIPTION

Proportion satisfied patients

The proportion of satisfied patients has been calculated at: 84 %
for patients with the selected profile.

Question in Swespine:

What is your attitude to the outcome of your spine surgery?

Definition:

Satisfied patient: satisfied
Dissatisfied patient: unsure or dissatisfied.

Number of observations:

22 522 surgeries
The number of observations refers to the total number of surgeries that serve as the basis for the prediction model and is not specific for selected patient type.

Variables in model:

Age
Comorbidity
Disability pension
Duration of pain in back
Duration of pain in legs
Functional impairment ODI
Gender
Operated levels
Preoperative pain (NPS) – back
Preoperative pain (NPS) – legs
Previous spine surgery
Quality of life (EQ-5D)
Retirement pension
Smoker
Unemployed
Walking distance

Proportion with successful outcome

(pain completely resolved or greatly improved)

Proportion of patients with a successful outcome is calculated to be: 84 %
for patients with the selected profile.

Question in Swespine:

How would you rate your leg pain/sciatica today compared with prior to surgery?

Definition:

Completely resolved

greatly improved

Number of observations:

21 687 surgeries
The number of observations refers to the total number of surgeries that serve as the basis for the prediction model and is not specific for selected patient type.

Variables in model:

Age
Comorbidity
Disability pension
Duration of pain in back
Duration of pain in legs
Functional impairment ODI
Gender
Operated levels
Preoperative pain (NPS) – back
Previous spine surgery
Smoker
Unemployed
Walking distance

Expected length of stay

Average length of stay is : 1.4 days
for patients with the selected profile.

Question in Swespine:

Date of Surgery, Date of Discharge

Definition:

Date of Discharge minus Date of Surgery = number of days

Number of observations:

29 742 surgeries
The number of observations refers to the total number of surgeries that serve as the basis for the prediction model and is not specific for selected patient type.

Variables in model:

Age
Clinical department type
Comorbidity
Disability pension
Duration of pain in back
Functional impairment ODI
Gender
Operated levels
Preoperative pain (NPS) – back
Preoperative pain (NPS) – legs
Previous spine surgery
Quality of life (EQ-5D)
Smoker
Unemployed
Year of surgery

This tool consists of prediction models for two health outcomes and one resource measure, for four different diagnosis groups. The data was derived from patients operated throughout Sweden, as reported to the Swespine quality assessment register during the period 2007-2020.

The models were designed as follows:

16 variables describing the patient's circumstances and health profile prior to surgery were selected in collaboration with experts.
For each health outcome a variable selection method (Stepwise selection using AIC) was used with the goal of choosing the simplest model with the best predictability.
Patients for whom information was missing for one or more variables were excluded (“listwise deletion”).

Description of the models: The models calculate the average health outcome for patients with different combinations of patient characteristics, based on historical correlations between different patient characteristics and health outcomes. The body of data is comprehensive, which allows the correlation to be estimated with relatively high reliability. However, individual variation still occurs due to factors (known and unknown) that could not be included in the models.

The variable selection method was used to find the model with the best predictability (which is done by excluding certain patient characteristics). This is necessary since a model with many variables that do not affect the outcome, impairs the ability of the model to predict outcomes for new patients. It is important to note that it is not possible to draw conclusions about what variables causally affect outcomes solely from the regression models in the tool. It cannot be ruled out that an unknown variable accounts for the causal effect. Furthermore, the variable selection method can select and exclude certain patient characteristics that correlate strongly with each other, since only one is needed to predict the outcome.

Reference

Dialogue support for spine surgery – based on Swespine, the Swedish Spine Register

Results

Proportion satisfied patients

Proportion with successful outcome

Expected length of stay

Proportion With Successful Outcome = Pain Completely Resolved Or Significantly Improved After One Year (Green And Light Green)

DESCRIPTION

Proportion satisfied patients

Proportion with successful outcome

(pain completely resolved or greatly improved)

Expected length of stay

Dialogue support for spine surgery – based on Swespine, the Swedish Spine Register

Description

Details

Delete Scenario

Save Scenario

Dialogue support for spine surgery – based on Swespine, the Swedish Spine Register

Reference