All analyses were performed using SPSS (version 19.0 for Windows, SPSS Inc., Chicago, IL, USA).
Study setting
For this study, data was obtained from IVF patients and healthcare professionals in 8 public IVF units in public hospitals in Israel. The patients in this sampling were infertile couples, 142 women and 62 men, who had undergone or were currently undergoing IVF treatment. Healthcare professionals were enlisted to ask patients for their cooperation. Out of 300 questionnaires distributed, 204 were filled in and returned, i.e., a response rate of 68 %.
The sample of healthcare professionals consisted of gynecologists, and fertility nurses from the same 8 public IVF units. Out of 24 questionnaires distributed, 19 were valid. Respondents were asked to answer all of the questions without exception: questionnaires that were not filled out in their entirety were disqualified, yielding a response rate of 79 %.
All the questionnaires were printed out and delivered manually, for both IVF patients and healthcare professionals.
Ethical approval was obtained in advance from the Ethics Committees (Helsinki committees) in each public hospital.
Written informed consent for participation in the study was obtained from each participant in the IVF patient group and from each participant in the group of healthcare professionals.
Procedure
The research instruments were questionnaires that were constructed for the study by Spiegel et al. see [37] in a three-stage process: (i) For the initial exploratory stage conducting in-depth interviews with eight fertility experts and 40 IVF patients, 30 women and ten men, to identify which items would be included in the research questionnaires; (ii) Pilot study of 40 IVF patients, five from each hospital; (iii) Main survey: Based on findings of the pilot study, the research questionnaires were revised and modified.
The same version of the research questionnaire was distributed to all the healthcare professionals (see also Aarts et al., [32]). In filling out the questionnaire, the professionals were asked to consider the fertility patients who were treated in their clinic.
This paper focuses on the following issues:
Evaluation of treatment
In order to evaluate patient satisfaction with IVF treatments and perceptions of it by professionals, this study was based on the research of Gerteis et al. [38] and on the Picker survey instruments that measure the patient’s experience of care in eight dimensions of patient-centeredness (www.pickerinstitute.org).
The following three major dimensions were tested:
-
1.
Coordination and integration of care: Professionalism of fertility clinic staff; attitude and sensitivity of fertility clinic staff and their relationship with patients; no personnel changes in the fertility clinic staff from beginning of treatment to the end; provision of consulting services and follow-up support – (medical, social and psychological factors).
-
2.
Information: Information on the chances of success (taking home a baby); information on prognosis, different treatment options, clinical aspects, and possible side effects of treatment; information about medical issues during pregnancy (multiple pregnancies, ectopic pregnancies, miscarriages, etc.); information about potential health problems of “test tube babies” - defects, prematurity; information on treatment costs.
-
3.
Access to care and Physical conditions: Geographical accessibility; physical conditions in the operating room - new/old medical equipment; physical conditions in the recovery room (number of beds, personal bedside cabinet, location of bathroom, privacy); physical conditions in the waiting room (new/old furniture, drinks available, reading material, newsletters, atmosphere); waiting times; standby time on the waiting list.
The patients’ Evaluation of treatment questions were presented on a 7-point Likert scale in which 1 represents ‘Completely dissatisfied’; 2 represents ‘Mostly dissatisfied’; 3 represents ‘Somewhat dissatisfied’; 4 represents ‘neither satisfied or dissatisfied’, 5 represents ‘Somewhat satisfied’; 6 represents ‘Mostly satisfied’ and 7 represents ‘Completely satisfied’.
In order to enable a comparison of the three major dimensions of satisfaction – of patients and fertility professionals - two methods were used for processing the data:
-
1.
Principal Component Analysis (PCA)
-
2.
Indices construction
Principal Component Analysis (PCA)
Principal Component Analysis (PCA) is the most widely used extraction method of component analysis and is most appropriate when the purpose is to reduce the number of items to a smaller number of representative components [39, 40].
The number of components to retain is determined by the criteria, which are that each PC explain at least 5 % (5 %-10 %) of the variance; the cumulative variance is at least 50 % .The literature varies on how much variance should be explained before the number of factors is sufficient. The majority suggest that 75–90 % of the variance should be accounted for [41, 42]. However, some indicate as little as 50 % of the variance explained is acceptable [43], and eigenvalues, which indicate the amount of variance explained by each component [42], are greater than one (Kaiser criterion) [44].
All the items relating to satisfaction with treatment were analyzed using PCA, and the analysis yielded three factors of satisfaction: (i) Human factor: satisfaction with coordination and integration of care; (ii) Information factor: satisfaction with information; (iii) Physical factor: satisfaction with access to care and physical conditions
Indices construction
Following Van Empel et al. [8], a sum score was calculated adding up the accompanying item scores. The dimension sum scores with diverse maxima were transformed into indices from 1.00 (worst possible) to 10.00 (best possible), using the same formula of Van Empel et al. ([8], p.144): “satisfaction index = 9* [(actual sum score - lowest possible sum score)/ (highest possible sum score - lowest possible sum score)] +1.”
Three satisfaction indices were defined here: (i) Human Satisfaction index (ii) Information Satisfaction index (iii) Physical Satisfaction index.
General psychological responses of the respondents
The experience of infertile couples is described in the literature as emotionally taxing [45–47]. The unique psychological factors of in-vitro fertilization (IVF) have been examined and assessed in order to discover whether psychological variables are correlated to patient satisfaction in these factors.
The psychological items were formulated as questions, such as ‘To what degree do you experience the following feelings at these times: guilt, success, etc.?’ Each item was analyzed individually and then graded on a five-point Likert scale in which: 1 represents ‘very slightly or not at all’, 2 ‘a little’, 3 ‘moderately’, 4 ‘a lot’ and 5 ‘extremely’
A Principal Component Analysis (PCA) of the psychological responses was conducted; this analysis yielded three psychological factors (i) Pessimism (ii) Activeness: active involvement in obtaining information and making decisions during treatment, taking initiative, and accepting full responsibility for the stages of treatment and results (iii) Shame.
Monetary evaluation of a treatment cycle – what is the maximum amount a respondent is willing to pay for a cycle of IVF treatment
The instrument chosen for economic evaluation of IVF treatment was the willingness to pay (WTP) [48].
The foremost economic theory in decision making by consumers posits that individuals try to maximize the utility of the goods and services they receive (subject to certain constraints). According to Lancaster [49, 50], the utility derived by each consumer from the characteristics of the good is different than the utility derived from the good as a whole.
Ryan [51] applied Lancaster’s utility approach to the field of health economics, using the contingent valuation methodology (CVM), which allows the assessment of a non-market good with a complex utility function. This assessment is made using a technique known as Willingness-to-Pay (WTP), where respondents are asked questions directly in a survey about their “Maximum WTP” – the maximum amount which they would be willing to pay for a service/product or an attribute of a service/product not available in a regular market, or non-priced goods and services. WTP is based on the assumption that “the maximum amount of money an individual is willing to pay for a commodity is an indicator of the value to him/her of that commodity” ([52], p. 182).
The respondents were asked to state ‘what is the amount of money they are willing to pay for one IVF treatment?’
The present study sought to check whether the dimensions of patient satisfaction are correlated with the willingness to pay for IVF treatment.
Demographic, socio-economic and health characteristics
Questions about socioeconomic position, number of children not from IVF, number of children from IVF, years of infertility, diagnosed infertility, and number of fertility treatments were derived from the baseline questionnaire.
Statistical analysis
Using a Pearson correlation, each of the three satisfaction factors were correlated with the demographic, socio-economic, health characteristics, psychological factors, and the WTP variable. P-values < 0.05 were considered statistically significant.
Gender differences in the satisfaction indices were assessed using a t-test. Another comparison was made between patients’ experiences and professionals’ perceptions of these experiences. The mean scores of patients and of professionals were compared using t-tests to detect any statistical differences. The group of professionals was taken as one group rather than broken down into physicians and nurses which would have made the group sizes too small. As for significance, P < 0.05 was considered statistically significant.
In order to assess the demographic, socio-economic, health characteristics, and psychological factor influence on the satisfaction indices, an Ordinary least squares (OLS) regression was used. As with OLS regression, F Value is the F-statistic signifying the Mean Square Model divided by the Mean Square Error. The F value should be with a p value (Pr > F) smaller than the standard criterion of 0.05.
R-Square is the proportion of variance in the dependent variable which can be explained by the independent variables. This is an overall measure of the strength of association and does not reflect the extent to which any particular independent variable is associated with the dependent variable.
In the social sciences, low R-squared values are common and expected. “Micro data on individuals, families, or households tend to have low R-squared because there is so much variation in individual behavior. Low R-squared do not necessarily mean that the model is poor” [53]; p 43. For example, Levitt [54] reports R-squared in the range of 0.06 and 0.37. In the present study, the acceptable R-squared were in the range of 0.04 and 0.1.
Adj R-Sq is a modification of the R-squared that penalizes the addition of external predictors to the model. In the social sciences, Adjusted R-squared is also used for a measure of effect size [55]: small effect 0.0196, medium effect 0.1300, and large effect 0.2600. Savage [56] reports adjusted R-squared in the range of 0.05 to 0.1. In the present study, the acceptable adjusted R-squared values were in the range of 0.03 and 0.1.
