CHAPTER 5: DISCUSSION
5.1 Data Analysis
Making a thorough comparison between the four difference cases allows us to clearly determine the relationship between service feature and performance. These analysis results support our original hypothesis about moderation fit. Next, we will discuss our findings from this research.
5-1 Data Analysis
In this section, we analyze the fit for the four cases shown in Table 5-1. Based on research hypotheses made in previous chapter, we evaluate the fits between the patient segmentation and service features of each case in order to test our hypothesis, as proposed in the previous chapter. In order to calculate the total fit scores of the four cases, the fit score is calculated by the percentage of hypotheses are met.. After the calculation, we will assess the fit and summarize all the fit scores of the four cases in Tables 5-1 and 5-2.
A1 perfectly fits the five service hypotheses. In this patient segmentation, its high disease emergency/variety fits with the four service features: ICT complexity, value focus, customization, supply and demand, and process complexity. In our hypotheses, patients with high disease urgency prefer lower device and process complexity in order to prevent the service to becoming too complicated or costly to fit their emergency needs. In the same sense, hard attributes like service availability or functionality would be more highly emphasized than soft attributes. In addition, patients with higher urgency usually corresponded to more highly fluctuating demands that had to be met with the same fluctuation of supply. Clearly, since the post discharge care (A1) division had five full fits that corresponded to its service features, this shows a 100% fit.
Next, we discuss the A2 division. In this division, the ICT complexity isn‟t fit for urgent disease conditions. Our hypotheses state that once disease variety gets higher, then involving too many complex devices can easily lead to costly and tedious processes and may confuse patients; therefore, this service category does not fit based on our hypotheses. For customization and process complexity categories, the division gets two moderate fits, because the process and service items offered in A2 are range between high and low, so we mark them as a moderate fit. Overall, cardiovascular disease care division (A2) has two service categories that fit fully and two with moderate fits; as a result, we calculated the total percentage of fit to be 60%.
In A3, the device complexity is in the middle and fits moderately according to its patient segmentation. From the point of view of value focus, soft attributes are more suitable for low disease urgency cases than high, a trend that fits the other three service
‧
categories as well. In total, A3 has two service categories that fully fit and three service categories that are moderate fits. Overall, the chronic disease care division (A3) gets a percentage of fit of 70%. The hospice care division (A4), which is almost the same as A1, gets four full fits and one moderate fit, resulting in a total fit of 90%. After a fit analysis was done for four cases, we organized our results into the cross-case analytical results table presented in Table 5-2. Next, we compared the service performance and analysis of fit to the four cases. In Figure 5-2., it shows the fit between patient segment and service features can contribute the service performance.
Urgency/Varity Fit or not Device Complexity
A1 High/High Fit Low
A2 High/Low Did not fit High
A3 Low/High Moderately fit Middle
A4 Low/Low Moderately fit Middle
Urgency Fit or not Hard/Soft
A1 High Fit Hard
A2 High Fit Hard
A3 Low Fit Soft
A4 Low Fit Soft
Urgency/Varity Fit or not Customization
A1 High/High Fit Low
A2 High/Low Moderately fit Medium
A3 Low/High Moderately fit Medium
A4 Low/Low Fit High
Urgency Fit or not Demand/Supply wave
A1 High Fit High
A2 High Fit High
A3 Low Fit Low
A4 Low Fit Low
Urgency/Varity Fit or not Process Complexity
A1 High/High Fit Simple
A2 High/Low Moderately Fit Medium
A3 Low/High Moderately Fit Medium
A4 Low/Low Fit Complicate
Table 5-1. Analysis results in A1-A4 divisions
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Division Number of Fits to features Fit score (%)
A1 5 fits 100
A2 2 fits, 2 moderately fits, one does not fit 60
A3 2 fits and 3 moderately fits 70
A4 4 fits and 1 moderately fits 90
Table 5-2. Analysis result of fit score in A1-A4 divisions
Figure 5-2. Cross-analysis of four cases
A1 A2 A3 A4
Device Complexity Value Focus Customization Supply wave Process Complexity Fit
Moderation Not Fit
Figure 5-1. Analysis result of fit score to five service features in four cases
0 20 40 60 80 100 120
A1 A2 A3 A4
Fit Score
rate of CASE Completionservice performance (rate of case completion) fit scores
‧
findings by reviewing and comparing the analysis results from the four cases with the fit scores and service performance shown in Figure 5-2.Finding 1: The fit between service features and patient preference positively affects service performance. If we compare the two key indexes of the A1 division in detail, we can clearly see that the A1 segment totally fits the five service features and also boasts the highest service performance of all four divisions, which indicates that it has the lowest rate of case withdraw and highest rate of case completion in the e-Health program.
This result supports our hypothesis proposed by previous chapter in that there was a fit between service features and patients‟ preference, both of which led to better quality service performance. The A1 division‟s policies, for example, maintain that free and minimally complex service is the key to delivering quality service to targeted patients, according to their specific preferences. The next best case was A4 case, which had a 90% fit and didn‟t have any withdrawn case records. Though the A4 division had the second-highest segmentation-service features fit, we still scored its performance to be equal to or slightly less than that of A1, due to the fact that the number of cases enrolled was smaller in A1 than in the other divisions (only 30 valid patients to be exact). If we look at the factor of device complexity in Figure 5-1, A4 gets a moderate fit for device complexity because the devices used (e.g. webcams) only facilitate the communication between the patient and case manager. “Caring and comfort are the values that we promise our patients, not devices.” A4 physician said. From this point of view, the service feature and value can be perfectly moderated by A4 patients‟ preference.
Finding 2: A2 can improve its performance by enhancing the stability of its complicated devices.
A2‟s performance is highly decreased by its unstable, complicated devices. For example, an electrocardiogram and hemadynamometer would not help in providing fist aid service for to any patient with a sudden heart failure or malady. Worse, the intricacy of the operation steps may actually confuse the user, causing them constant to make the complications worse, thus prompting a stream of customer complaints about the stability and accuracy of devices.